Caribbean Basins. Volume 4. Sedimentary Basins of The World

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.
SEDIMENTARY BASINS OF THE WORLD

An Introduction to the Series
Etymology reveals much about the essence of a word. Science in German, Wissenschaft, is
the art of observing whereas science in Chinese, koxue, is the study of classifying. Scientific
observations, with the help of modern equipments, have made leaps and bounds in our century, but
taxonomy seems irrelevant.
Classification can be science. The rise of the natural sciences in Europe could be traced back
to Carl Linnaeus in 1750 when he used criteria of mutual exclusiveness to establish the taxonomy
of living organisms. Unfortunately, this prerequisite in dividing and subdividing is not always
appreciated, and a common practice in geology has been to "classify" basins through reference
to incidental attributes. So, we have coastal basins, back-arc basins, extensionally rifted basins,
successor basins, deep-sea basins, flysch basins, etc. These qualifications describe the geography,
tectonic setting, principal stresses, orogenic chronology, depositional environment or sedimentary
association of a basin, but these all could be different aspects of one and the same.
"A basin is a basin is a basin", paraphrasing Gertrude Stein. Even though no two basins are
exactly alike, subsidence is the common denominator of all. A systematic classification of basins
depends upon recognition of the mutually exclusive causes of subsidence.
Forty years ago, when I first went to study in the United States, I was fascinated by the debate
whether basin subsidence is isostatically induced by sedimentary load. Later, in the 1950's, I
began to realize that depressions on Earth are underlain by thin crust and came to the conclusion
that subsidence could be the surface manifestation of an endogenetic process of crustal thinning,
in response to the Airy isostasy. Still later, in the early 1960's, after geophysical studies had
revealed the heterogeneity of the Earth's mantle, subsidence could be related to mantle cooling and
corresponding density change, in response to the Pratt isostasy. Meanwhile, we all concede that the
weight of a sedimentary pile filling a subsiding depression will induce isostatic subsidence. The
three different mechanisms of isostatic adjustment are, however, not mutually exclusive, and they
may have operated concurrently. To classify basins on the basis of the various modes of isostasy
can thus not be systematic. But, as indicated by gravity studies, not all sedimentary basins are
isostatically adjusted, and not all subsidence is isostatic. For a first-order division we could thus
recognize two classes of basins, those which have subsided isostatically and those which have not.
Crustal thinning is a prerequisite to initiate Airy isostasy. Thinning can be induced by two
different systems of principal stresses, with the least principal stress either horizontal or vertical.
The former leads to the genesis of rifted basins under horizontal extension, and the latter results in
pull-apart basins in transform or strike-slip fault zones. The orientation of principal stresses could
thus be the criterion for second-order distinctions.
Rift basins are commonly present in the continental interior. After the appearance of oceanic
crust between separated continents, the rifts become narrow oceanic gulfs (like the Red Sea).
Eventually the loci of subsidence are shifted to passive margins where coastal plains are underlain
by thick basinal sediments. Rifted basins may also form on an active margin where a segment
of continent is torn apart from the mainland to form islands arcs; those are back-arc basins. The
position with respect to plate margins can thus be used as the criterion for third-order distinctions.
Subsidence induced by horizontal compression, the third of the three possible configurations
of principal stress, is not isostatic. Plate-tectonics theory relates the origin of trenches to the
underthrusting of ocean lithosphere on an active plate margin and the origin of foreland basins
to the underthrusting within the continental lithosphere. These two major basins of compressional
origin are thus also distinguished by the third-order criterion concerning their position with respect
to plate margins.

VI SEDIMENTARY BASINS OF THE WORLD m AN INTRODUCTION TO THE SERIES
When we first planned the series of the Sedimentary Basins of the World, we intended to adopt a
genetic classification. Basins are subdivided into the three sets of criteria discussed above:
1. Isostasticallyadjusted basins
1.1. Extensionalbasins
1.1.1. Rift basins in the continental interior
1.1.2. Narrowoceanic gulf basins
1.1.3. Basins of deposition on passive margins
1.1.4. Rifted basins on active margins or back-arc basins
1.2. Transcurrent(transform or strike-slip) pull-apart basins
2. Isostasticallynot adjusted basins
2.1. Compressionalbasins
2.1.1. Forelandbasins (in a continental interior)
2.1.2. Oceanic trenches (on a continental margin)
With this scheme in mind, the editors of Elsevier and I made up a list of the volumes for the
projected series, and we started our search for volume editors. Our first priority, taking into account
current demand, was to bring out a volume on China. We were concerned that the Chinese basins
could not be fitted into our scheme, because we were told that they represent a special group of
unclassifiable basins on "paraplatforms". However, as I became personally involved in researches
on the geology of China, I came to the conclusion that the Chinese basins were not "unclassifiable".
Rifted basins, back-arc basins, pull-apart basins, foreland basins, etc., exist in China, as they are
present elsewhere in the world. The Chinese basins of different origins do share a common history
in geologic evolution, and they are united by their geography. It would be illogical to discuss the
various Chinese basins in separate volumes. Yet if we are to include all of them in one, we have to
designate that volume by their unifying geography.
When we started to work on our second volume on rifts, we were still trying to keep our
genetic scheme, although we were resigned to make a single exception for the Chinese opus. We
were thinking of the East African Rift Valleys, and the emphasis was on Africa. As chance would
have it, I just happened to accept a consulting contract on Africa. After a year of working on the
assignment, I realized that basins on that continent are as diverse in origin as those in China; there
are foreland basins, pull-apart basins, as well as rift basins in Africa.
About this time, our choice of the editor for a volume on rifted basins, Professor R.C. Selley,
brought up another issue. He pointed out to me the impracticability of putting out volumes on the
sole basis of their postulated origins. The purpose of the series is to provide information on the
geology of sedimentary basins of the world in order to help a novice to start a project. Commonly,
the one who seeks information knows the geographic extent of his interests, but not necessarily
the genesis of his targets. Taking, for example, the case of a person who is to start an exploration
venture in some region, how should he know if he is to study a monograph on pull-apart basins
or one on foreland basins. He knows, of course, if the location is in China or in Africa, and could
consult an opus on Sedimentary Basins of China, or that on Africa accordingly.
The arguments by Professor Selley finally convinced me to change our scheme. The criterion
of dividing the volumes will have to be geographical. In addition to those on China, Africa, and
the Caribbean, volumes on sedimentary basins of Australia, South America, and the Soviet Union
(Russian Platform and Siberia) are planned.
Geographical groupings are satisfactory if the basins of various origins in a region share some
common heritage, but to throw all heterogeneous entities into one big pot could be disconcerting.
To produce, for example, a volume on the Sedimentary Basins of Europe to include all those in the
Russian Platform, under the North Sea, and in the Prealps may make a good lexicon, but not an
opus harmonized by a unifying theme. This consideration led us to the decision to place priority
on certain natural boundaries, so that each volume would sustain a certain coherence in geology
as well as in geography. The Cenozoic basins in the Tethyan orogenic belt, for example, are to
be grouped under the title of Foreland Basins of the Alpine-Mediterranean Region. The pull-apart
and back-arc basins on the shores of the Pacific will be included in two or more volumes of the
Sedimentary Basins of the Circum-Pacific Region.
There are, of course, other sedimentary basins of the Earth, especially those of the Near East,
North America and Antarctic, which should be included in the series if the coverage is to be
complete. On the other hand, we shall also evaluate the demand of the profession for such volumes
as the initial volumes of the series successively appear during the next decade.

SEDIMENTARY BASINS OF THE WORLD - - AN INTRODUCTION TO THE SERIES VII
It is my hope that the series of volumes would not be compared to philately albums; there should
be unity in style and in substance. Yet the accumulation of geological information has reached such
immense proportions since Eduard Suess wrote his monumental work Das Antlitz der Erde that no
single person could ever hope to master the geology of the world. The series of our volumes will,
therefore, have to be collective efforts. Coordinations by volume editors are indispensable. My job,
as the series editor, is to further enhance the unity and harmony of the whole. We have, however, to
accept the fact that each article of a volume may "speak" a different dialect, and each volume of the
series may "speak" a different language. Perfect consistency can only be achieved if a person has
the time or the capacity to translate all those hundreds of articles in more than a dozen volumes into
one universal script. This is not possible, and the practical alternative to the ideal is, therefore, to
leave each author or group of co-authors a maximum freedom in their style of presentation and in
their interpretations of geology. The articles are to be accepted as expressions of the present state of
understanding of an area by leading geologists working in that area. They may or may not represent
the understanding of the volume editor, or that of the series editor.
Through my experiences in editing the first volume on Chinese basins, I appreciate the potential
dangers of such freedom of expressions; a lack of precision in semantics could lead to grave
misunderstandings. I felt impotent when I saw basic terms, like orogeny, platform, shelf sediments,
intracratonic basins, etc., defined, in certain communities of our profession, on a basis distinctly
different from that adopted by modem students of geology. The misconceptions in some instances
are so deeply rooted, that nothing short of a rewrite could save the situation. Yet neither the volume
nor the series editor could completely revise all the articles. To avoid complete chaos, I plan,
therefore, to write a summary, at the end of each volume (or at least some), in my style, and to
interpret the geology on the basis of my understanding. Such summaries may contain an overdose
of personal opinions and may involve interpretative errors by a single geologist, but they should,
at least, be consistent, and may eliminate misconceptions caused by the divergent meaning of the
same words as they are used in various "dialects" and "languages".
The preceding pages were written in January, 1988 for the first volume of the series on Chinese
basins. The reviewers of the Chinese volume gave me encouragement that I, as the series editor,
should continue my role as an interpreter of different cultures. I have, therefore, taken the initiative
to write the last chapter of the second volume m A Distant View of the South Pacific Geology. The
volume edited by Peter Ballance and my summary are both written in English. There is little need
for translation. Nevertheless, New Zealanders speak English with their local accent, and the same
"words" are pronounced differently by one who speaks English with a strong Chinese and Swiss
accent. It is not surprising that Peter, a dear old friend, found it difficult to "agree entirely" with me.
On the other hand, he was tolerant enough not to protest too strongly, and I could have my Distant
View for the reference of other distant readers.
In reviewing the articles in the second volume of our series, I became more convinced than
ever of the wisdom of Selley's advice that the basins should be grouped regionally. I have devoted
most of my professional career with a process-oriented approach. When I edited a book on
Mountain-Building Processes, the emphasis was on processes. Instead of regional syntheses, I
adopted an analytical approach to look into the different processes involved in mountain-building,
sedimentary, magmatic, deformational, and metamorphic. In editing this series on Sedimentary
Basins, there is no better alternative than regional synthesis. Not only basins in China have diverse
origins, those in the South Pacific are equally diverse. Yet they all seem to belong to the same
set of diverse basins. If the geology of the South Pacific seems very different from that of China,
the apparent distinction can be attributed to the fact that they have advanced to different stages of
tectonic evolution.
Four years have gone by since the publication of the second volume of the series of The
Sedimentary Basins of the World. On the eve of the publication of the third volume, I am
encouraged to find that the series will not be philately albums; they will not be collections of
random observations. I saw the parallelism in the pattern of orogenic deformations in China and
in South Pacific, and I could see the same pattern in Africa. There is a difference in the stage of
the tectonic evolution. China, as a part of Eurasia has undergone a billion years of amalgamation.
The South Pacific is still in an earlier phase of accretion. Africa has gone a long away since its
separation of Pangaea, and it is being pushed toward Eurasia to its ultimate destiny of a place in a
supercontinent. The figures and the colorations of the mosaic pieces are different, but they are all to
be pieced together for a unifying theory of global tectonics.

VIII S E D I M E N T A RBASINS
Y OF THE WORLD - - AN INTRODUCTION TO THE SERIES
With the publication of the Caribbean volume, the Sedimentary Basins of the World series
has come of age. The "philately albums" have accumulated such a wealth of information, and
the series is on its way to becoming an encyclopaedia of sedimentary basins. More significant
is the fact that geology could shed its label of provincialism. In addition to basins formed by
ocean-continent interactions, the Caribbean volume provided a basis for interpreting geology on
the basis of assuming ocean-ocean plate-interactions. The perspectives have caused me to think
again about the genesis of some Chinese basins. Were not the Tarim and Qaidam basins fragments
of an oceanic plateau, like the Colombia and Venezuela basins, that survived as rigid blocks during
the Phanerozoic deformations?
We originally hoped that the series would be completed in 10 or 15 years. With the publication
of this fourth volume, we see the light at the end of the tunnel, even though another decade will
have to elapse before we complete our task. The two volumes on sedimentary basins of the former
Soviet Union are being edited. We are pleased to persuade Andrew Miall to edit the volume of
North American basins. Other volumes being planned are South American basins, Mediterranean
and Tethyan basins, Southwestern Asian basins, West Pacific basins, Australian and Antarctic
basins.
In order to fulfill our dream of producing an opus comparable to Suess's The Face of the Earth,
two more series are required: The Mountains of the World and The Oceans of the World. Elsevier
might see the wisdom to initiate these ambitious projects, even though I myself may be too ancient
to see their completion.
Zurich, Switzerland, January, 1999 KENNETH J. HSU

Series Editor
Sedimentary Basins of the World

Dedication
This volume on Caribbean sedimentary basins is dedicated to Professor Willem A. van den Bold,
a pioneer in geologic and paleontologic studies of the circum-Caribbean basins.
Early years. 'Wim' van den Bold was born in 1917 and raised in Amsterdam. In 1942, he began
his Ph.D. studies at the Geological Institute of the University of the Utrecht under the supervision
of Professor L.M.R. Rutten, a well known structural geologist and tectonicist. The proposed topic
of his Ph.D. study was a study of Cretaceous and Tertiary ostracodes contained within samples from
Cuba that had been collected by Rutten and other geologists from the University of Utrecht as part
of their ongoing study of the regional geology and tectonics of that island. Within one year of start-
ing this study, Wim was forced into hiding in Utrecht to avoid conscription by Nazi authorities for
forced labor in the German war effort. When it appeared that the Nazis had given up their search for
him, Wim retrieved his microscope and literature from the university and resumed his Ph.D. study in
his house. To compensate for the loss of electricity in 1944, Wim was able to illuminate his micro-
scope by devising a set of mirrors and lenses to direct sunlight through a small opening in his attic.
His persistence paid off and in 1945 Wim completed his Ph.D. study, 'Contribution to the Study
of Ostracoda, with Special Emphasis to the Tertiary and Cretaceous Microfauna of the Caribbean
Region'. In addition to his original Cuban samples from Rutten, his study was expanded to include
sample material from Bonaire, British Honduras (now Belize) and Guatemala. His dissertation
results showed that the vertical ranges of many species of ostracodes in various sedimentary basins
were limited enough for stratigraphic correlations and that the ability of many ostracoda to adapt to
W.A. van den Bold

X DEDICATION
changing water conditions made them ideal indicators for reconstructing marine vs. brackish-water
environments.
Shell years. Wim joined Royal Dutch Shell in 1946 to participate in the post-war resurgence of
international oil exploration. His first assignments took him to Venezuela, Colombia and Trinidad
where he contributed to a wider industry-sponsored stratigraphic correlation effort that led to the
modem Tertiary biostratigraphic age correlation schemes in use today. The tectonic inversion of
deep- and shallow-water basins related late Tertiary migration of the Caribbean plate provided
Wim and his colleagues thick onland exposures for identifying and correlating planktic and benthic
formaninifera. Wells drilled during oil exploration led to rapid increases in the understanding of
the exposed geology of the basin edges and the subsurface geology of basins like the Magdalena
basin of Colombia, the Maracaibo and Eastern Venezuela basins of Venezuela, and the Southern
basin of Trinidad. In his initial assignments for Shell in Venezuela and Colombia, Wim further
developed the use of ostracodes as stratigraphic markers in exploration wells, but, after his transfer
in 1950 to Trinidad, he participated with Hans Bolli of Trinidad Leaseholds in development of the
first planktic formaniniferal zonation of that island. During this period, Hans G. Kugler of Trinidad
Leaseholds and later the University of Basel provided him access to ostracode material from
the tectonically active Northern and Southern basins of Trinidad and underlying 'passive margin'
section that provided him the basis of many significant papers on the Cenozoic biostratigraphy and
regional geology of Trinidad in the late 1950's and early 1960's.
Years at Louisiana State University. In 1958, Wim was invited to join the Department of
Geology of Louisiana State University (LSU) by Henry V. Howe, a leading ostracode researcher
and founding chairman of that department. His hiring of Wire insured the continuation of ostracode
studies at LSU along with the continued growth of the geology department. During his early years
at LSU, Wim continued his studies of the biostratigraphy of the Trinidad area with seminal papers
on the ages and environments of the Brasso Formation (1958), the Eocene and Oligocene section
(1960), and the Upper Miocene and Pliocene section (1963) of Trinidad. His teaching duties in the
small department included structural geology and igneous and metamorphic petrology along, with
a long-running tenure as instructor for the department's summer field camp in the Front Range of
Colorado. Students, impressed by his ability to toil in the field for long days without water or rest,
nicknamed him 'the Camel'.
During the mid-to-late 1960's, Wire expanded his scale of observation of Caribbean basins
to include field-based studies in Guatemala, Panama, Jamaica, Puerto Rico, Antigua, Anguilla,
St. Martin, St. Croix, Costa Rica, Nicaragua, the Lesser Antilles, and Mexico. He systematically
described the ostracode faunas in these areas and, where possible, related these faunas to
better known biostratigraphic frameworks based on planktic foraminifera. The resulting regional
correlations laid the framework for looking at the Caribbean region as a large-scale tectonic and
sedimentary system rather than as a mosaic of individual and unrelated basins. In 1971, Wire
returned to Cuba, the main topic of his Ph.D. study more than 30 years before. Through the
invitation of Cuban geologists, Wim began a field-based study of the Neogene basins of the
Matanzas and Santiago areas. In returning to Cuba at the height of the Cold War and the nadir of
Cuba-US relations, Wim became one of the first US-based earth scientists to 'break the ice' and
begin collaborative work with Cuban geologists.
From the late 1960's to the present, graduate student enrollment at LSU increased and Wim
supervised the research of many masters and doctoral students working on ostracodes and
foraminifera. These student projects adapted Wire's tenets of establishing the basic geology and
stratigraphy, determining the microfauna in a systematic way and cross-comparing ostracode and
planktic foraminiferal faunas, and using ostracodes for stratigraphic correlation and assessment
of marine vs. brackish conditions. Some of these students have gone on to become influential
researchers and teachers in industry and academia, where they have in turn passed his time-proven
methods of basinal studies on to a third generation of researchers. Follow-up studies to some of his
work in this volume include Gill et al. for St. Croix, Babb and Mann for Trinidad, and Mann et al.
for the Dominican Republic.
Retirement years. In 1990, Wim retired with his wife Nettie to Baarn, Holland, near Utrecht.
His children are now living both in the U.S. and Holland. He is currently a Professor Emeritus

DEDICATION XI
at his alma mater, the University of Utrecht, where he is completing a monograph on Caribbean
ostracodes. His mailing address there is:
Dr. Willem A. van den Bold
Rijksuniversiteit te Utrecht
Instituut voor Aardwetenschappen
Budapestlaan 4, Postbus 80.021
3508 TA Utrecht, Netherlands
Dedication. We dedicate this volume to Wim van den Bold in acknowledgement of his tireless
pursuit of his research goals and for his unselfish sharing of his knowledge of Caribbean basins and
microfauna with us and his other students and colleagues. We dedicate this volume to him as not
only a mark of respect, but also one of affection.
Acknowledgement. We thank Hans van den Bold for the photograph and some of the information
in this dedication. This dedication was submitted by Peter E McLaughlin, Jr. (Exxon Exploration
Company) and Barun K. Sen Gupta (Louisiana State University).
Selected dissertations supervised by Willem A. van den Bold at Louisiana State University
(1971-1989)
George Esker, 1968, Biostratigraphy of the Cretaceous-Tertiary Boundary in the East Texas
Embayment Based on Planktonic Foraminifera.
Paul Steineck, 1973, Paleoecologic and Systematic Analysis of Foraminifera from the Eocene-
Miocene Montpelier and Lower Coastal Groups, Jamaica, West Indies.
Burton Bordine, 1974, Neogene Biostratigraphy and Paleoenvironments, Lower Magdalena Basin,
Colombia.
Walter Kessinger, 1974, Stratigraphic Distribution of the Ostracoda of Comanche (Cretaceous)
Series of North Texas.
Gilbert Taylor, 1975, The Geology of the Limon Basin of Costa Rica.
Patricia Fithian, 1980, Distribution and Taxonomy of the Ostracoda of the Paria/Trinidad/Orinoco
Shelf.
Maria-Luisa Machain Castillo, 1985, Pliocene Ostracoda of the Saline Basin, Veracruz, Mexico.
Peter E McLaughlin, 1989, Neogene Basin Evolution in the Southwestern Dominican Republic: A
Foraminiferal Study (co-supervised by B.K. Sen Gupta).
SELECTED PUBLICATION LIST W.A. VAN DEN BOLD
Bold, W.A. van den, 1946. Contribution to the study of Ostracoda with special reference to the
Tertiary and Cretaceous microfauna of the Caribbean region. Doctoral thesis, Utrecht Univ.,
175 pp. Reprint Antiq. Junk, 1970, with addition of a list of changes in genetic and specific
assignments of species, and collection numbers of holo- and paratypes.
Bold, W.A. van den, 1950. Miocene Ostracoda from Venezuela. Journal of Paleontology 24 (1),
76-88.
Bold, W.A. van den, 1957. Ostracoda from the Paleocene of Trinidad. Micropaleontology 3 (1),
1-18.
Bold, W.A. van den, 1957. Oligo-Miocene Ostracoda from southern Trinidad. Micropaleontology
3 (3), 231-254.
Bold, W.A. van den, 1958. Ambocythere, a new genus of Ostracoda. Annals and Magazine of
Natural History, ser. 12, 10, 801-813.
Bold, W.A. van den, 1958. Ostracoda of the Brasso Formation of Trinidad. Micropaleontology 4
(4), 391-418.
Bold, W.A. van den, 1960. Eocene and Oligocene ostracoda of Trinidad. Micropaleontology 6 (2),
145-196.
Bold, W.A. van den, 1960. The genus Eucyitheridea Bronstein (Crustacea, Ostracoda) with a
redescription of the type species. Annals and Magazine of Natural History, ser. 13, 4 (41),
283-303.
Bold, W.A. van den, 1961. Some new Ostracoda of the Caribbean Tertiary. Koninkl. Nederlandse

XII DEDICATION
Akad. Wetensch. Proc., ser. B., 64 (5), 627-639.

Bold, W.A. van den, 1963. Anomalous hinge structure in a new species of Cytherelloidea.
Micropaleontology 9 (1), 75-78.
Bold, W.A. van den, 1963. Ostracods and Tertiary stratigraphy of Guatemala. Bulletin of the
American Association of Petroleum Geologists 47 (4), 696-698.
Bold, W.A. van den, 1963. The ostracode genus Orionina and its species. Journal of Paleontology
37 (1), 33-50.
Bold, W.A. van den, 1963. Upper Miocene and Pliocene Ostracoda of Trinidad. Micropaleontology
9 (4), 361-424.
Bold, W.A. van den, 1964. Ostracoden aus der Oberkreide von Abu Rawash, Aegypten. Palaeonto-
graphica, Abteilung A: Palaeozoologie-Stratigraphie 123; 111-135.
Bold, W.A. van den, 1965. Middle Tertiary Ostracoda from northwestern Puerto Rico. Micropale-
ontology 11 (4), 381-414.
Bold, W.A. van den, 1965. Pseudoceratina, a new genus of Ostracoda from the Caribbean. Koninkl.
Nederlandse Akad. Wetensch. Proc., ser. B, 68 (3), 160-164.
Bold, W.A. van den, 1965. New species of the ostracod genus Ambocythere. Annals and Magazine
of Natural History, ser. 13, 8 (1), 1-18.
Bold, W.A. van den, 1966. Les ostracodes du Neogene du Gabon. Revue de l'Institut Frawais du
Petrole 21 (2), 155-176.
Bold, W.A. van den, 1966. Miocene and Pliocene Ostracoda from northeastern Venezuela. Koninkl.
Nederlandse Akad. Wetensch. Verh., Affi. Natuurk., ser. 1, 23 (3), 43 pp.
Bold, W.A. van den, 1966. Ostracoda from Col6n Harbour, Panama. Caribbean Journal of Science
6, 43-64.
Bold, W.A. van den, 1966. Ostracoda from the Antigua Formation (Oligocene, Lesser Antilles).
Journal of Paleontology 40 (5), 1233-1236.
Bold, W.A. van den, 1966. Ostracoda of the Poz6n section, Falc6n, Venezuela. Journal of
Paleontology 40 (1), 177-185.
Bold, W.A. van den, 1966. Ostracode zones in Caribbean Miocene. Bulletin of the American
Association of Petroleum Geologists 50 (5), 1029-1031.
Bold, W.A. van den, 1966. Upper Miocene Ostracoda from the Tubara Formation (northern
Colombia). Micropaleontology 12 (3), 360-364.
Bold, W.A. van den, 1967. Miocene Ostracoda from Costa Rica. Micropaleontology 13 (1), 7586.
Bold, W.A. van den, 1967. Ostracoda of the Gatun Formation, Panama. Micropaleontology 13 (3),
306-318.
Bold, W.A. van den, 1968. Ostracoda of the Yaque Group (Neogene) of the northern Dominican
Republic. Bulletins of American Paleontology 54 (239), 106 pp.
Bold, W.A. van den, 1969. Messinella, a new genus of Ostracoda in the Caribbean Cenozoic.
Micropaleontology 15 (4), 397-400.
Bold, W.A. van den, 1969. Neogene Ostracoda from southern Puerto Rico. Caribbean Journal of
Science 9 (3-4), 117-125.
Bold, W.A. van den, 1970. Ostracoda of the lower and middle Miocene of St. Croix, St. Martin and
Anguilla. Caribbean Journal of Science 10 (1-2), 35-61.
Bold, W.A. van den, 1970. The genus Costa (Ostracoda) in the upper Cenozoic of the Caribbean
region. Micropaleontology 16 (1), 61-75.
Bold, W.A. van den, 1971. Ostracoda of the Coastal Group of formations of Jamaica. Transactions
Gulf Coast Association of Geological Societies 21,325-348.
Bold, W.A. van den, 1971. Ostracode associations, salinity and depth of deposition in the Neogene
of the Caribbean region. In: Paleoecology of ostracodes. Centre de Recherches de Pau (Societe
Nationale des Petroles d'Aquitaine), Bulletin 5 (5), 449-460.
Bold, W.A. van den, 1972. Ostracodos del post-Eoceno de Venezuela y regiones vecinas. In:
Congreso Geologico Venezolano, 4th, Memoria, Vol. 2. Boletfn de Geologfa Publicaci6n
Especial 5, 999-1063.
Bold, W.A. van den, 1973. Distribution of Ostracoda in the Oligocene and Lower and Middle
Miocene of Cuba. Caribbean Journal of Science 13 (34), 145-159.
Bold, W.A. van den, 1973. Nota geologica; notas sobre los ostracodos de la Formaci6n Punta
Gavilan. Boletfn de Geologfa (Caracas) 12 (22), 333-335.
Bold, W.A. van den, 1973. Ostracoda of the La Boca Formation, Panama Canal Zone. Micropale-

DEDICATION XIII
ontology 18 (4), 410-442.

Bold, W.A. van den, 1974. Neogene of Central Haiti. AAPG Bulletin 58 (3), 533-539.
Bold, W.A. van den, 1974. Ornate Bairdidae in the Caribbean. In: Bold, W.A. van den (Ed.),
Geoscience and Man 6, Ostracoda, the Henry V. Howe memorial volume. Louisiana State
University, School of Geoscience, Baton Rouge pp. 29-40.
Bold, W.A. van den, 1974. Taxonomic status of Cardobairdia (Van den Bold, 1960) and
Abyssocypris n. gen.; two deepwater ostracode genera of the Caribbean Tertiary. In: Bold,
W.A. van den (Ed.), Geoscience and Man 6, Ostracoda, the Henry V. Howe memorial volume.
Louisiana State University, School of Geoscience, Baton Rouge pp. 65-79.
Bold, W.A. van den, 1975. Distribution of the Radimella confragosa group (Ostracoda, Hemi-
cytherinae) in the late Neogene of the Caribbean. Journal of Paleontology 49 (4), 692-701.
Bold, W.A. van den, 1975. Neogene biostratigraphy (Ostracoda) of southern Hispaniola. Bulletins
of American Paleontology 66 (286), 549-639.
Bold, W.A. van den, 1975. Ostracodes from the late Neogene of Cuba. Bulletins of American
Paleontology 68 (289), 119-167.
Bold, W.A. van den, 1975. Remarks on ostracode-biostratigraphy of the late and middle Tertiary of
southwest Puerto Rico. Caribbean Journal of Science 15 (1-2), 31-40.
Bold, W.A. van den, 1976. Distribution of species of the tribe Cyprideidini (Ostracoda, Cytheridei-
dae) in the Neogene of the Caribbean. Micropaleontology 22 (1), 1-43.
Bold, W.A. van den, 1976. Ostracode correlation of brackish-water beds in the Caribbean Neogene.
Transactions of the Caribbean Geological Conference VII, Saint Francois, Guadeloupe, June
30-July 12, 1974, pp. 169-175.
Bold, W.A. van den, 1980. Notes on the distribution of some mid-Tertiary ostracodes of Puerto
Rico. In: Llinas, C.R., Gil, N., Seaward, M., Tavares, I., and Snow, W. (Eds.), Transactions of the
9th Caribbean geological conference, Santo Domingo, Dominican Republic, Aug. 16-20, 1980,
Vol. 1, pp. 225-230.
Bold, W.A. van den, 1981. Distribution of Ostracoda in the Neogene of central Haiti. Bulletins of
American Paleontology 79 (312), 136 pp.
Bold, W.A. van den, 1983. Shallow-marine biostratigraphic zonation in the Caribbean post Eocene.
Proc. 8th Int. Symp. Ostracoda, 400-416.
Bold, W.A. van den, 1985. Heinia, a new genus of Ostracoda from the Gulf of Mexico and the
Caribbean. Journal of Paleontology 59 (1), 1-7.
Bold, W.A. van den, 1986. Distribution of Ostracoda at the Eocene-Oligocene boundary in
deep (Barbados) and shallow-marine environment (Gulf of Mexico). In: Pomerol, C., and
Premoli-Silva, I. (Eds.), Terminal Eocene events. Elsevier, Amsterdam, pp. 259-263.
Bold, W.A. van den, 1986. Fresh and brackish water Ostracoda from the Neogene of northern
Venezuela. Tulane Studies in Geology and Paleontology 19 (34), 141-157.
Bold, W.A. van den, 1988. Neogene paleontology in the northern Dominican Republic; 7, The
subclass Ostracoda (Arthropoda). Bulletins of American Paleontology 94 (329), 105 pp.
Bold, W.A. van den, 1988. Ostracoda of Alacran Reef, Campeche Shelf, Mexico. Tulane Studies in
Geology and Paleontology 21 (34), 143-155.
Bold, W.A. van den, 1989. Ostracoda of the Montezuma Formation, Pliocene, Nicoya Peninsula,
Costa Rica. Tulane Studies in Geology and Paleontology 22 (1-2), 61-64.
Bold, W.A. van den, 1990. Late Holocene Ostracoda in and around Lake Enriquillo, Dominican
Republic. In: Larue, D.K., and Draper, G. (Eds.), Transactions of the 12th Caribbean geological
conference, St. Croix, United States Virgin Islands, Aug. 7-11, 1989, pp. 163-189
Bold, W.A. van den, 1990. Short review of the Ostracoda of the Montpelier and Coastal Groups of
Jamaica. In: Larue, D.K., and Draper, G. (Eds.), Transactions of the 12th Caribbean geological
conference, St. Croix, United States Virgin Islands, Aug. 7-11, 1989, pp. 99-103.
Bold, W.A. van den, 1990. Stratigraphical distribution of fresh and brackish water Ostracoda in the
late Neogene of Hispaniola. In: Whatley, R.C., and Maybury, C. (Eds.), Ostracoda and global
events m Tenth International Symposium on Ostracoda. Chapman and Hall, pp. 221-232.
Bold, W.A. van den, 1990. Note on the stratigraphy of the Bellad6re and Comendador structures,
Neogene of Haiti and the Dominican Republic. Transactions of the 10th Caribbean geological
conference (Cartagena, 1983), pp. 238-242.
McLaughlin, EE, Bold, W.A. van den, and Mann, E, 1991. Geology of the Azua and Enriquillo
basins, Dominican Republic; 1, Neogene lithofacies, biostratigraphy, biofacies, and paleogeogra-

XIV DEDICATION
phy. In: Mann, E, Draper, G., and Lewis, L.E (Eds.), Geologic and tectonic development of the
North America-Caribbean Plate boundary in Hispaniola. Geological Society of America Special
Paper 262, 337-366.
McLaughlin, EE, Gill, I.E, and Bold, W.A. van den, 1995. Biostratigraphy, paleoenvironments and
stratigraphic evolution of the Neogene of St. Croix, U.S. Virgin Islands. Micropaleontology 41
(4), pp. 293-320.

CARIBBEAN SEDIMENTARY BASINS
Preface
The purpose of this volume is to provide a collection of original studies of Caribbean basins
that were mainly carried out in the early and mid-1990's. It is hoped that this collection of studies
will become a valuable source of information for future students of Caribbean basins and will help
them better focus their research. I solicited original contributions crafted with a regional tectonic
scope rather than more localized geologic reviews mainly because there have been several recent
and comprehensive volumes on Caribbean geology. These volumes include:
9 Geological Society of America, Decade of North American Geology, volume H, on the geology
and geophysics of the Caribbean region, edited by Gabriel Dengo and James E. Case (1990). This
benchmark volume contains a wealth of information on Caribbean geology and basins arranged in a
comprehensive area by area manner.
9 Geological Society of America Special Paper 295 on the on- and offshore geology and geophysics
of southern Central America in Panama and Costa Rica, edited by myself (1995) (fig. 1B). This
volume contains a complete summary of basin studies in southwestern Caribbean through the early
1990's.
9 American Association of Petroleum Geologists, Petroleum Basins of South America, edited
by A.J. Tankard, R. Su~irez and H.J. Welsink (1995). This award-winning volume provides
comprehensive coverage of basins in the northern and northwestern areas of South America.
Given these objectives and existing Caribbean volumes, the scope and emphases of this volume
include the following:
9 A focus on sedimentary basins in the region of the Caribbean Sea and its margins. The common
tectonic events discussed in the chapters of this volume include the rifting and passive margin
history of North and South America that led to the formation of the Caribbean region, the entry of
an exotic, Pacific-derived Great Arc of the Caribbean at the leading edge of the Caribbean oceanic
plateau, the terminal collision of the arc and plateau with the passive margins fringing North and
South America, and subsequent strike-slip and accretionary tectonics that affected the arc-continent
collision zones (Fig. 1).
9 An emphasis on new subsurface and potential field data (well logs, shallow and deep penetration
seismic reflection, ship-based and satellite-based gravity and magnetics, and aeromagnetics). These
data are leading to rapid breakthroughs in our mainly outcrop-based understanding of Caribbean
tectonic history and Caribbean tectonic processes. The papers in this volume attempt to synthesize
these subsurface and potential field data at a regional or basin-wide scale to facilitate tectonic
interpretations.
9 Inclusion of new biostratigraphic and structural data on the highly deformed and sometimes
metamorphosed 'bits and pieces' of Caribbean sedimentary basins that are important for tectonic
reconstructions. These types of data, which vary from biostratigraphic dating studies to structural
and isotopic dating studies, are needed in this region where many of the major tectonic events
occurred in the Cenozoic and have strongly overprinted the existing Cretaceous and older rocks.
9 A regional plate tectonic introduction to Caribbean basins in the form of two introductory
chapters by myself and Mtiller et al. These chapters attempt to place Caribbean basins into a
wider regional context using revised quantitative plate reconstructions. This regional context helps
to better understand their often complex tectonic settings, short-lived subsidence mechanisms, and
later inversion histories. The last systematic effort at a quantitative Caribbean plate model was by
Ross and Scotese (1988) and was not able to incorporate the recent advances in seafloor mapping
using satellite-based gravity measurements that have now become available (Sandwell and Smith,
1997).
The volume is divided into six parts. Part 1 consists of the two tectonic setting papers by myself
and Mtiller et al.

XVI SEDIMENTARY BASINS OF THE CARIBBEAN - - PREFACE
Fig. 1. (A) Political subdivision of the Caribbean region. (B) Locations of chapters in this volume (numbers refer to chapter
number). Chapters 1 and 2 cover the entire area shown and Chapter 3 focusses on the central part of Mexico that is not
shown on this map. Diagonally lined area represents the Pacific-derived Caribbean oceanic plateau province, Great Arc of
the Caribbean and area of back-arc basins formed during eastward and northeastward migration of the Great Arc.
In Chapter 1, I use Geosat gravity data from Sandwell and Smith (1997) to illustrate the
large-scale tectonic features of the Caribbean with emphasis on Cenozoic sedimentary basins
visible on the gravity maps. I classify these basins using a simplified basin classification scheme.
I present a series of thirteen plate reconstructions based on the North A m e r i c a - S o u t h America
motions used with permission from the Mtiller et al. study and use these reconstructions to place
the major basin-forming events into a plate tectonic context. These tectonic events support the idea
of Pindell and Barrett (1990) and previous workers that the Caribbean is a Pacific-derived oceanic
plateau that diachronously collided with the passive margins of North and South America during

SEDIMENTARY BASINS OF THE CARIBBEAN m PREFACE XVII
Cenozoic time. To conclude this overview, I suggest possible avenues of future research based on
this compilation and the new results presented in the other chapters of this volume.
In Chapter 2, Mfiller et al. compute plate motions and uncertainties of motions for the
Mesozoic-Cenozoic interactions of North and South America using magnetic anomaly and satellite
gravity-based fracture zone identifications from the Atlantic Ocean. This refined data set supports
previous plate models for the Caribbean by showing the two Americas drifted apart until about
71 Ma (Campanian) and then underwent a steady convergence across the area of the present-day
Caribbean Sea from about 55 Ma (Eocene) to the present. The authors propose that their much
better defined Cenozoic convergent phase is responsible for shaping many of the convergent zones
and basins along the east-west-trending northern and southern margins of the Caribbean plate.
Part 2 consists of five chapters mainly focussed on basins overlying the North America plate
and recording its rifting from South America in Late Jurassic to Cretaceous time. Because the
structures related to this rift plane in Caribbean tectonic history are largely covered by water and
Cretaceous and younger carbonate platform rocks, three of these chapters rely heavily on either
well and seismic reflection data or on outcrop data from deformed sections now exposed onland.
Chapter 3 by Marton and Buffler uses seismic reflection data and the results of DSDP Leg 77
to define major tectono-stratigraphic sequences in the area of the southeastern Gulf of Mexico
(Fig. 1). This area is critical to understanding the early rift history of the Gulf of Mexico and
North and South America opening because it is adjacent to, but not overprinted by, the Paleo-
gene thrust event of Cuba. These data indicate that the Yucatan Peninsula underwent a major
counter-clockwise rotation from Oxfordian to late Berriasian time which produced a series of rifts
in the southeastern Gulf of Mexico. As rifting ceased, and the rift blocks subsided Early Cretaceous
carbonate platforms formed atop the rift blocks. These workers attempt to correlate the Mesozoic
offshore stratigraphic record of the southeastern Gulf of Mexico with onshore data described by
Pszcz6~kowski in chapter 4 from western Cuba.
Chapter 4 by Pszcz6~kowski chronicles the major stratigraphic and tectonic events known
from the rifted and passive margin of North America now exposed by Paleogene thrusting and
subsequent Cenozoic faulting in western Cuba (Fig. 1). The major tectono-stratigraphic events
included a rift event (Lower Jurassic-?Callovian-early Oxfordian), a passive margin subsidence
event (?Callovian/middle Oxfordian-Santonian) and the beginning of the arc-continental collision
phase (Campanian-Paleocene).
Chapter 5 by Pessagno et al. present new stratigraphic, age, and paleobathymetric data from the
area of Mexico southwest of the northwest-striking Walper megashear, a major terrane boundary
that bisects Mexico and thought to have been active during the early rift period between North and
South America. Faunal (radiolaria and megafossils) and paleomagnetic data indicate that the area
southwest of the megashear was transported from higher paleolatitudes during the Oxfordian and
reached lower paleolatitudes by the Berriasian. The authors propose that terranes southwest of the
megashear share similar stratigraphic signatures with the Jurassic and Early Cretaceous rocks of
western Cuba that are described in detail in Chapter 4 by Pszcz6~kowski (Fig. 1). The authors argue
that the Mexican and Cuban rocks may have formed a once continuous San Pedro del Gallo terrane
prior to disruption of the Cuban part of the terrane by Paleogene collision of the Caribbean arc.
Chapter 6 by Scott and Finch presents new stratigraphic and paleontologic data from a
Berriasian-Albian carbonate platform formed above the Chortis block in Honduras (Fig. 1). The
Chortis block provides a critical for clues to the paleoposition of the Caribbean because the Chortis
block is the only subaerial and continental part of the present-day Caribbean plate. At the time of
Mesozoic rifting between North and South America, the Chortis block probably formed a southern
extension of the continental area of southern Mexico. The ages and faunas reported by Scott
and Finch support the idea that by Aptian and Albian time the Chortis block had established a
biogeographic connection to the Caribbean.
In Chapter 7, Masaferro and Eberli use multi-channel seismic reflection lines to reveal the nature
and evolution of the Great Bahama bank, a Late Jurassic-Recent carbonate platform formed during
the early rifting between North and South America (Fig. 1). Their data show Early Jurassic rifts
and extensional structures overlain by a ca. 5-kin-thick carbonate platform. The carbonate platform
shows strong fault control that is interpreted as an effect of the collision between the Caribbean arc
and the Great Bahama bank in Late Cretaceous-Middle Eocene time. Fault activity largely ended
with the collision in Late Eocene time and fault scarps and tectonic depressions were masked by
infilling of the highly productive carbonate platform.

XVIII SEDIMENTARY BASINS OF THE CARIBBEAN m PREFACE
Part 3 consists of six chapters that focus on smaller, usually heavily faulted and onshore
Cenozoic basins of the northern Caribbean that formed in response to arc, collisional, and
strike-slip activity between the evolving North America-Caribbean plate boundary.
In Chapter 8, Av6 Lallemant and Gordon present new structural and isotopic data from Roatan
Island, which is one of the Bay Islands off the coast of northern Honduras (Fig. 1). Structures
in metamorphic and sedimentary rocks exposed on these islands include older, ductile structures
formed at metamorphic conditions during the Late Cretaceous to early Tertiary left-oblique collision
between the Chortis and Maya (southern Mexico) blocks. Most brittle faults are younger and formed
after the Late Eocene or Early Oligocene exhumation of the metamorphic rocks. Younger structures
indicate transtension related to North America-Caribbean strike-slip motion along the Cayman
trough. The Bay Islands form the northern flank of the offshore Tela sedimentary basin. These data
indicate a transtensional origin for this basin in late Neogene time.
In Chapter 9, Manton and Manton present new stratigraphic and age data from Miocene marine
sedimentary rocks from the southern margin of the Tela basin in the fault-bounded, coastal range
of northern Honduras (Fig. 1). These data suggest that this range is a deformed turbiditic basin of
Miocene age that has been inverted along strike-slip faults related to the North America-Caribbean
plate boundary. Deformed sedimentary rocks of this inverted range may have once been continuous
with undeformed sedimentary rocks of the Tela basin to the north.
In Chapter 10, Montgomery and Pessagno present new identifications of radiolaria and
foraminifera from deformed, deep basinal rocks faulted against Caribbean Cretaceous arc rocks
in Jamaica and Hispaniola (Dominican Republic) (Fig. 1). The authors interpret these rocks as
accreted crust of the proto-Caribbean and Atlantic seafloor that was subducted to the west and
southwest beneath the Caribbean arc. Pacific-derived fragments subducted to the east and northeast
beneath the Caribbean are also present in these areas and suggest a complex subduction polarity
reversal in the Caribbean arc sometime during the Cretaceous.
In Chapter 11, de Zoeten and Mann present new stratigraphic, paleocurrent, and sandstone
petrographic data from Paleocene to Early Pliocene basinal and carbonate platform rocks of
northern Hispaniola (Dominican Republic) (Fig. 1). We use this sedimentary record to interpret
three major tectonic phases that affected this segment of the North America-Caribbean plate
boundary: (1) an Early to Middle Eocene collisional event between the Caribbean arc and the
Bahama carbonate platform; (2) a period of strike-slip faulting and basin formation from Late
Eocene to Early Miocene; and (3) a period of transpressional uplift and folding from Late Miocene
to Early Pliocene.
In Chapter 12, Mann et al. present new outcrop, well, and seismic reflection data from the
Enriquillo basin of Hispaniola (Dominican Republic) and document the existence of two distinct
facies of Early Pliocene evaporites which formed in the center and edges of the tectonically active
ramp basin within the North America-Caribbean strike-slip zone. Analysis of seismic reflection
data tied to an exploration well in the center of the basin showed that the basin center deposit is
a ca. 1500-m-thick halite and gypsum deposit whose depocenter was controlled by reverse faults
that crosscut the center of the basin. Three occurrences of basin-edge, tidal flat gypsum deposits are
interpreted as having been deposited during Early Pliocene eustatic falls in sea level.
In Chapter 13, Gill et al. use stratigraphic and paleontologic data from both wells and outcrops
to document the evolution of the Neogene Kingshill basin on St. Croix (U.S. Virgin Islands)
within the North America-Caribbean strike-slip plate boundary. These data reveal a deep-water,
pre-basinal section of Oligocene to early Middle Miocene age which underwent transtensional
faulting no later than late Middle Miocene to form the rift-like basin. They relate this transtensional
event to oblique, left-lateral opening between the islands of St. Croix and Puerto Rico. Near the
end of the Middle Miocene, the occurrence of shallow-water limestone indicates an uplift of the
basin-bounding blocks. This uplift continued through the Pliocene when St. Croix acquired its
approximate present-day land area.
Part 4 consists of two chapters that focus on Cenozoic basins related to the Lesser Antilles arc
system of the eastern Caribbean.
In Chapter 14, Huyghe et al. use seismic reflection profiles and sidescan sonar images to review
the tectonic control and sedimentary patterns of late Neogene piggyback basins of the accretionary
wedge formed in Cretaceous?-Cenozoic time in front of the east-facing Lesser Antilles island arc.
These basins provide important modem examples of piggyback basins that are now preserved in
older more deformed areas of the circum-Caribbean like Cuba and Venezuela. Their data show two

SEDIMENTARY BASINS OF THE CARIBBEAN - - PREFACE XIX
main stages in the history of these piggyback basins: (1) rapid tilting of the basin in response to the
growth of a fault-bend fold at one edge of the basin; (2) reactivation of the bounding thrust surface
by folding or sediment diapirism.
In Chapter 15, Bird et al. integrate gravity, seismic reflection and refraction data with their
previously published magnetic data from the Grenada back-arc basin to produce a model for its
east-west opening during Paleogene time. Disruption of basin gravity and magnetic trends in the
northern part of the basin is attributed to localized tectonic shortening effects.
Part 5 consists of three chapters on the Jurassic-Recent sedimentary basins of the eastern
Venezuela and Trinidad area of the southeastern Caribbean. These basins reflect both the Jurassic-
Cretaceous rifting and passive margin history of separation between the North and South America
plates as well as a much younger phase of Oligocene to Recent transpression between the
eastward-migrating Lesser Antilles arc and accretionary wedge and the South American continent.
In Chapter 16, di Croce et al. use seismic reflection data tied to wells in the eastern Venezuela
and Trinidad areas to propose four distinct tectono-stratigraphic phases for this part of the South
America margin: (1) an ill-defined Paleozoic/pre-Jurassic pre-rift phase; (2) a Jurassic syn-rift phase
related to the separation of North and South America; (3) a Cretaceous to Oligocene passive margin
phase related to the thermal subsidence of the South American rifted margin; and (4) a Neogene
foredeep phase related to the oblique collision between the Lesser Antilles arc and accretionary
prism and the South American continent. The passive margin period of phase 2 from 131 Ma to
30 Ma is subdivided into five second-order transgressive-regressive cycles interpreted as eustatic
sea-level fluctuations superimposed on the thermally subsiding margin. The Oligocene and younger
history of the margin and its major sequences is dominated by the formation of a collisional
foredeep basin produced by the oblique collisional event and rapid basin infilling in response to
uplift and erosion of the Andes.
In Chapter 17, Flinch et al. use seismic reflection data and wells from eastern Venezuela, the
Gulf of Paria and Trinidad to reveal the presence of a submarine late Neogene pull-apart basin
beneath the shallow Gulf of Paria west of Trinidad (Fig. 1). The pull-apart formed at a stepover
between the fight-lateral Casanay-Arima fault zone to the north and the Warm Springs fault zone
to the south. These data also reveal northward thrusting coeval with the pull-apart formation that is
interpreted as a passive roof duplex formed above deeper southward-directed thrusts.
In Chapter 18, Babb and Mann integrate seismic reflection and well data from the Northern
and Gulf of Paria basins of Trinidad with existing outcrop and map information to identify three
Neogene deformational phases related to initial Late Miocene-Early Pliocene movement along the
E1 Pilar fault zone, formation of the Gulf of Paria pull-apart basin between the E1 Pilar and Warm
Springs-Central Range fault zones, and increasing movement on the Warm Springs-Central Range
fault zones. We relate the southward expansion of the late Neogene deformational zone in Trinidad
to the presence of oceanic crust southeast of Trinidad that allows for the southeastward migration
of strike-slip fault-bounded blocks in that direction.
Part 6 consists of three chapters containing revealing new information on the Cretaceous
Caribbean igneous plateau that forms the basement of the central Caribbean Sea beneath the
Venezuelan and Colombian basins (Fig. 1). All three chapters make use of deep penetrating
multi-channel seismic data to show the stratigraphic and structural features of this oceanic plateau
province which was previously known from isolated outcrops in the circum-Caribbean and from
widely spaced DSDP sites in the Venezuelan and Colombian basins.
In Chapter 19, Diebold and Driscoll use these deep-penetrating seismic lines to reveal previously
unseen structures within the entire thickness of the Cretaceous Caribbean oceanic plateau basaltic
crust east of the Beata Ridge. The lower basaltic section consists of submarine basaltic volcanoes
with dipping flanks that appear to have maintained their primary sense of dip. Flows can be
followed for distances of 20 to 100 km and appear to have been erupted on an existing section of
rough ocean floor of normal thickness. The upper basaltic sequence, whose top defines the well
known B" reflector, also contains widespread flows but is more heterogeneous than the lower unit
and fills morphological and extensional lows in the top of the lower sequence. Rifting and flexural
uplift have faulted both sequences along the Beata Ridge and large escarpments. The age relations
between the lower sequence and the upper sequence are not known since only the top of the upper
sequence (B" reflector) has been drilled.
In Chapter 20, Driscoll and Diebold use the same seismic reflection data set to document the
history of rift features in the upper basaltic section and the sedimentary and tectonic history of the

XX SEDIMENTARY BASINS OF THE CARIBBEAN m PREFACE
section overlying the basaltic basement described by them in Chapter 19. Extensional deformation
accompanying the formation of the rifted upper basaltic section of the Cretaceous plateau led to the
formation of the large escarpments along the Hess escarpment and Beata Ridge. The sedimentary
section between the B" and A" reflectors was probably derived from the uplift of ranges in northern
South America, is correlated to DSDP and ODP sites, and is constrained to be younger than
Senonian (~88 Ma) and older than Middle Eocene (~50 Ma). This section exhibits only minor
Neogene faulting along the Beata Ridge but does exhibit reworking and drifts related to bottom
currents.
Chapter 21 by Mauffret and Leroy uses an extensive seismic reflection, gravity, and magnetic
data set that is centered on the Beata Ridge. Regional mapping reveals post-Early Miocene
strike-slip and compressional features along the eastern edge of the Beata Ridge that formed by
a northeast-southwest-shortening event. These young deformational features increase from south
to north along the trend of the Beata Ridge. The authors propose that these faults are related
to shortening between the subduction zone along the northern margin of South America and the
subduction and strike-slip zone in Hispaniola and the Muertos trench.
Additional color illustrations are available at the Sedimentary Basins of the Caribbean site on
the Internet ~. The supplementary material includes color versions of selected figures from chapters
1, 2, 11, 12, 14 and 17 and additional figures for chapters 5, 7 and 15.
P. MANN
(Editor)
REFERENCES
Dengo, G. and Case, J.E. (Eds.), 1990. The Caribbean Region. The Geology of North America, Vol. H, Geological Society
of America, Boulder, Colorado, 528 pp. with attached volume of map enclosures.
Mann, E (Ed.), 1995. Geologic and Tectonic Development of the Caribbean Plate Boundary in Southern Central America.
Geological Society of America Special Paper 295, 349 pp.
Pindell, J.L. and Barrett, S.E, 1990. Geologic evolution of the Caribbean region. In: G. Dengo and J.E. Case (Editors),
The Caribbean Region. The Geology of North America, Vol. H, Geological Society of America, Boulder, Colorado, pp.
405-432.
Ross, M.I. and Scotese, C.R., 1988. A hierarchical tectonic model of the Gulf of Mexico and Caribbean region.
Tectonophysics, 155: 139-168.
Sandwell, D.T. and Smith, W.H.E, 1997. Marine gravity anomaly from Geostat and ERS-1 satellite altimetry. Journal of
Geophysical Research, 102: 10,039-10,054.
Tankard, A.J., Sufirez Soruco, R. and Welsink, H.J., 1995. Petroleum Basins of South America. American Association of
Petroleum Geologists Memoir 62, 792 pp.
1http://www.elsevier.nl/locate/caribas/with mirror sites: http://www.elsevier.com/locate/caribas/and

http://www.elsevier.jp/locate/caribas/

Reviewers
I would like to thank the following people for volunteering their time to review the papers in this
volume.
Lewis Abrams Frederick Hutson

James A. Austin, Jr. Keith James
Hans G. Av6 Lallemant James N. Kellogg
Nathan Bangs John E Lewis
Charles D. B lome Fernando Martinez
Burke Burkart Gyorgy L. Marton
Kevin C. Burke Kristian Meisling
Eric Calais Henry E. Mullins
Millard E Coffin Ian Norton
Daniel M. Davis Steven Pierce
James E Dolan James L. Pindell
Thomas W. Donnelly Walter Pitman
Grenville Draper Andrzej Pszcz6tkowski
Robert A. Duncan Edward Robinson
Paul A. Dunn Robert Rogers
Peter Emmet Amos Salvador
Lee Gerhard Kathryn M. Scanlon
Jan Golonka Charlotte B. Schreiber
Mark B. Gordon Robert E. Sheridan
Nancy R. Grindlay Norm Silberling
Chistoph E. Heubeck Carol Telemaque
Albert C. Hine Elazar Uchupi
Troy L. Holcombe
R MANN
(Editor)

List of Contributors* *
W. ALl 17 J.E CASEY 15

Trinmar Department of Geosciences
Point Fortin University of Houston
Trinidad and Tobago Houston, TX 77204-5503, USA
G. AVI~ LALLEMANT 8 Y. DENIAUD 14

Department of Geology and Geophysics, MS-126 Ddpartement de G6ologie et Oc6anographie m
Rice University URA 197
Houston, TX 77005-1892, USA Avenue des Facult6s
33405 Talance, France
S.E. BABB 18
Institute for Geophysics V. DE LISA 17
University of Texas at Austin Petroleos de Venezuela
4412 Spicewood Springs Road, Bldg. 600 Edificio PDVE&P
Austin, TX 78759, USA La Estancia
Chauo, Caracas, Venezuela
A.W. BALLY 16
Department of Geology and Geophysics, MS-126 R. DE ZOETEN 11
Rice University Unocal Thailand
Houston, TX 77005-1892, USA Central Plaza Office Bldg.
2993 Phaholyothin Road
D.E. BIRD 15 Bangkok 10990, Thailand
Bird Geophysical rdezoeten @unocal.com
16903 Clan Macintosh
Houston, TX 77084, USA J. D I CROCE 16
dbird@idt.net
J. DIEBOLD 19,20
R.T. BUFFLER 3 Lamont-Doherty Earth Observatory of Columbia
Institute for Geophysics University
University of Texas at Austin Palisades, NY 10964-8000, USA
4412 Spicewood Springs Road Bldg. 600 johnd @lamont.ldgo.columbia.edu
Austin, TX 78754, USA
dick@utig.ig.utexas.edu N. DRISCOLL 19,20
Geology and Geophysics
S.C. CANDE 2 Woods Hole Oceanographic Institution
Scripps Institution of Oceanography, UCSD Woods Hole, MA 02543, USA
9500 Gilman Drive ndriscoll @whoi.edu
La Jolla, CA 92093-0215, USA
cande @gauss.ucsd.edu G.R EBERLI 7
Comparative Sedimentology Laboratory
A. CANTU-CHAPA 5 Rosentiel School of Marine and
Department of Geology Atmospheric Science
Florida International University 4600 Rickenbacker Causeway
University Park Miami, FL 33149, USA
Miama, FL 33199, USA gerberli @rsmas.miami.edu
* Superior ciphers refer to the chapter number.

t E-mail address of first authors included when available.

XXIV LIST OF CONTRIBUTORS
J-C. FAUGERES 14 G. HERNANDEZ a7

D6partement de G6ologie et Oc6anographie Petroleos de Venezuela
URA 197 Edificio PDVE&P
Avenue des Facult6s La Estancia
33405 Talance, France Chauo, Caracas, Venezuela
R.C. FINCH 6 K.J. HSO 22

Department of Earth Sciences Geologisches Institut
Box 5062 ETH-Ziirich
Tennessee Technological University ZUrich, Switzerland
Cookeville, TN 38505, USA
rcf7332.tntech.edu D.K. HUBBARD 13
Virgin Islands Marine Advisors
J.E FLINCH 7 5046 Cotton Valley Rd
Departamento de Geologia Christiansted, St. Croix 00820
Lagoven
Caracas, Venezuela D.M. HULL 5
Programs in Geosciences
Present address:
University of Texas at Dallas
Exploration and Production Total
EO. Box 830688
Paris la Defense, France
Richardson, TX 77083, USA
joan.flinch @total.com
E HUYGHE 14
I. GILL 13
D6partement de G6ologie et Oc6anographie m
Department of Geology
URA 197
University of Puerto Rico, Mayaguez
Avenue des Facultds
E O. Box 5000
33405 Talance, France
Mayaguez, Puerto Rico 00681-5000
i-gill @rumac.upr.clu.edu Present address:
Laboratoire de G6odynamique des Chaines Alpines
E. GONTHIER 14 et VPRESA, 15 rue Maurice Gignoux
D6partement de G6ologie et Oc6anographie m 38031 Grenoble cedex, France
URA 197 huyghe @uj f-grenoble.fr
Avenue des Facult6s
33405 Talance, France M. KELLDORF s
Programs in Geosciences
M.B. GORDON 8 University of Texas at Dallas
Department of Geology and Geophysics EO. Box 830688
MS-126, Rice University Richardson, TX 77083, USA
Houston, TX 77005-1892, USA
M.E. LAMAR 12
Present address:
Department of Geological Sciences
GX Technology
University of Texas at Austin
5847 San Filipe, Suite 3500
Austin, TX 78713, USA
Houston, TX 77057, USA
mgordon@ gtx.com Present address:
Scott and White Memorial Hospital
R. GRIBOULARD 14 Department of Obstetrics and Gynecology
Ddpartement de G6ologie et Oc6anographie 2401 South 31st Street
URA 197 Temple, TX 76508, USA
Avenue des Facult6s
33405 Talance, France S.R. LAWRENCE 12
Exploration Consultants Ltd.
S.A. HALL 15 Highlands Farm, Greys Road
Department of Geosciences Henley-on-Thames, Oxon, RG9 4PR UK
University of Houston s.lawrence @ecgc.com
Houston, TX 77204-5503, USA

Chapter 1
Caribbean Sedimentary Basins" Classification and Tectonic

Setting from Jurassic to Present
P. MANN
INTRODUCTION Plate rates
The purpose of this introductory chapter is to Rates of relative plate motion as predicted by
describe the active tectonic setting of the Caribbean, the Nuvel-lA plate motion model of DeMets et
its major crustal provinces, and to provide a sim- al. (1994) are relatively slow (11-13 mm/year) be-
ple classification for sedimentary basins in the tween the Americas and the Caribbean plate but
Caribbean region. In addition to this background much faster (59-74 mm/year) between the Co-
information on Caribbean basins, I provide a series cos, Nazca and Caribbean plates (Fig. 1). Recent
of thirteen quantitative plate reconstructions based GPS-based studies of the relative motion between
on the revised plate model of Mtiller et al. (Chapter the North America and Caribbean plates in the
2). These reconstructions serve to place individual northeastern Caribbean by Dixon et al. (1998) have
basins into a better tectonic framework. shown that the actual North America-Caribbean
rate of east-west strike-slip motion may be twice
as fast as predicted by the Nuvel-lA plate motion
ACTIVE TECTONIC SETTING OF CARIBBEAN model.
SEDIMENTARY BASINS
Earthquake studies
Major plates
There have been many first-motion studies on
The distribution of recorded earthquakes, active large Caribbean earthquakes over the past 25 years
calc-alkaline volcanoes, and spreading ridges de- and I have compiled a representative group from
fines five rigid plates in the Caribbean region: the Harvard focal mechanism catalogue on Fig. 1.
North America, South America, Caribbean, and Because large earthquakes occur more frequently in
Nazca (Molnar and Sykes, 1969; Mann et al., 1990) subduction zone settings, many more focal mech-
(Fig. 1). Geologic and seismic studies indicate that anisms are available from the Lesser Antilles and
the Caribbean plate is moving eastward relative to Middle America arcs than for the northern and
the Americas, and this movement is accommodated southern dominantly strike-slip plate boundaries. In
by left-lateral strike-slip faults along its bound- general, subduction zone earthquakes are character-
ary with the North America plate, and right-lateral ized by shallow thrust faulting with the auxiliary
strike-slip faults along its boundary with the South plane striking approximately parallel to the trend
America plate. Oceanic lithosphere of the North and of the trench and dipping steeply away from the
South America plates is consumed along the east- arc and with the fault plane striking subparallel
ern edge of the Caribbean at the Lesser Antilles to the arc trend and dipping gently beneath the
subduction zone. Oceanic lithosphere of the Cocos arc. First-motion studies from the strike-slip plate
and Nazca plates is consumed along the western boundaries are consistent with shallow focus left-
and southwestern edge of the plate at the Middle lateral fault displacements along the northern plate
America subduction zone (Fig. 1). boundary zone and right-lateral displacements along
the southern plate boundary zone (Fig. 1).
Caribbean Basins. Sedimentary Basins of the World, 4 edited by E Mann (Series Editor: K.J. Hsti), pp. 3-31.
9 1999 Elsevier Science B.V., Amsterdam. All rights reserved.

4 E MANN
Fig. 1. Earthquake focal mechanisms and plate motions relative to a fixed Caribbean plate based on the NUVEL-1A global plate motion
model of DeMets et al. (1994). Numbers give rate of plate motion in mm/year and arrows give directions of plate motion. Focal
mechanisms are color-coded according to depth: red mechanisms are from earthquakes from 0 to 75 km in depth; blue mechanisms
are from earthquake from 75 to 150 km in depth; and green mechanisms are > 150 km in depth. Basemap is the Geosat gravity
map of the Caribbean compiled by Sandwell and Smith (1997). In accordance with plate motion model predictions, earthquake focal
mechanisms show predominantly left-lateral motion along the northern edge of the Caribbean plate, predominantly right-lateral motion
along the southeastern edge of the plate, predominantly thrust motion at the eastern and western ends of the plate, and predominately
northeast-directed right-lateral motion associated with the displacement of the Maracaibo block in the southwestern Caribbean. The
Maracaibo block is a continental-arc fragment of northwestern South America that is escaping to the north and northeast as a response to
the late Neogene collision of the Panama arc and oblique subduction of the northern Nazca plate.
First-motion studies along with geologic and MAJOR CRUSTAL PROVINCES OF THE CARIBBEAN
GPS-based geodetic studies have shown that the REGION
northwestern corner of South America (Maracaibo
block of Mann and Burke, 1990) is being displaced The Caribbean region consists of a rim of
northward and northwestward along the B o c o n 6 - Cretaceous-Recent arc terranes and associated
eastern Andean right-lateral strike-slip fault system back-arc basins molded about a sub-circular core
in Colombia and western Venezuela (McCann and consisting of a continental fragment in the western
Pennington, 1990; Kellogg and Vega, 1995) (Fig. 1). Caribbean (Chortfs block) and an oceanic plateau
This displacement appears to be a consequence of province beneath the central and eastern Caribbean
late Neogene collision of the Panama arc with north- (cf. Case et al., 1990, for a comprehensive review of
western South America (Mann and Burke, 1990). all Caribbean crustal provinces) (Fig. 2).
Chortis block
S u b d u c t e d slabs
The Chortfs block of northern Central America
Inclined subducted slabs extend to depths of 150 consists of well-dated Mesozoic and Cenozoic for-
km under most o f the land areas adjacent to the mations which unconformably overlie a continental
Lesser Antilles and Middle America arcs (McCann basement of poorly dated, metamorphic rocks of
and Pennington, 1990; Dewey and Sufirez, 1991) Paleozoic and possible Precambrian age (Gordon,
(Fig. 1). Subducted slabs are also present beneath 1991) (Fig. 2). Seismic velocities confirm that the
much of the northwestern comer of South America Chortfs block is continental (Case et al., 1990) but
(van der Hilst and Mann, 1994). the isotopic ages of metamorphic protolith rocks

C A R I B B E A N S E D I M E N T A R Y BASINS" C L A S S I F I C A T I O N A N D T E C T O N I C S E T T I N G 5
Fig. 2. Major crustal provinces of the Caribbean: Precambrian-Paleozoic Chortfs block; Cretaceous oceanic plateau of the central
Caribbean; Early Cretaceous-Recent Great Arc of the Caribbean; and passive margins of North and South America. Red lines indicate
active plate boundaries and white lines are magnetic anomaly and fracture zone trends from Coffin et al. (1992). Key to abbreviations: YB
-----Yucatan basin; GB = Grenada basin.
exposed in Honduras have not been established. and 6 km below sea level, and where it is overlain
Limited isotopic age dates indicate a late Paleo- by 2 km of sediments, it might be expected to lie
zoic metamorphic event around 300 Ma and show about 1 km deeper. The shallow depth of most of
that these rocks are pre-Mesozoic (Gordon, 1991). the Caribbean ocean floor, averaging 1-2 km less
The dominant basement rock types are phyllitic than predicted, is commonly attributed to the rapid
and graphitic schists which can contain interlayers and widespread emplacement of basaltic flows and
of metaconglomerate and quartzite. The overlying sills to form an' immense oceanic plateau during
Mesozoic stratigraphy consists of Middle Jurassic Santonian time (~88 Ma) (Burke, 1988; Donnelly et
through Early Cretaceous clastic rocks overlain by al., 1990; Sinton et al., 1997).
Aptian-Albian shallow marine limestone (Scott and Deformation of the Caribbean plate edges has led
Finch, Chapter 6). The stratigraphic record indicates to exposure of the edges of the Caribbean oceanic
that the Chortfs block was a region of minimal plateau in Costa Rica (Sinton et al., 1997), Panama
tectonic activity during the Mesozoic but was sub- (Bowland and Rosencrantz, 1988), southern His-
ject to regional folding and faulting events in the paniola (Sen et al., 1988), Colombia, and northern
Late Cretaceous and Cenozoic (Av6 Lallemant and Venezuela (Kerr et al., 1997). The thickness and
Gordon, Chapter 8; Manton and Manton, Chap- geochemistry of the Caribbean oceanic plateau is
ter 9). similar to that of western Pacific oceanic plateaus
including the Manihiki and Ontong Java (Bowland
Caribbean oceanic plateau and Rosencrantz, 1988; Kerr et al., 1997).
Diebold and Driscoll (Chapter 19) and Driscoll
The central and eastern Caribbean is underlain and Diebold (Chapter 20) show that Caribbean
by an oceanic plateau with a 12-15 km thickness oceanic plateau volcanism was a two-phase event.
that is intermediate between oceans and continents They suggest that the Santonian age sampled from
over most of its area (Case et al., 1990; Diebold and circum-Caribbean outcrops and from DSDP and
Driscoll, Chapter 19) (Fig. 2). Caribbean ocean floor ODP cores in the Colombian and Venezuelan basins
made in Cretaceous or earlier times would normally (Donnelly et al., 1990) dates only the second smaller
have subsided as it aged to depths of between 5 phase of plateau formation.

6 R MANN
Great Arc of the Caribbean Antilles island arc (Bouysse, 1988). Bird et al.
(Chapter 15) show that the direction of opening
Arc rocks of a Cretaceous-Eocene island-arc chain of the Grenada basin was approximately east-west
are found in a semi-continuous belt from Cuba to the rather than north-south as proposed by Pindell and
north coast of South America (Fig. 2). The northern, Barrett (1990). Heubeck et al. (1991) proposed that
or Greater Antilles, segment of the arc from Cuba to a narrow belt of now-inverted, Paleogene basinal
the Virgin Islands east of Puerto Rico has been in- rocks exposed on the island of Hispaniola may rep-
active since its collision with the Bahamas Platform resent the continuation of the Grenada and Yucatfin
in Late Paleocene to earliest Oligocene time. Ma- back-arc basins in this area (Fig. 2).
jor pulses of collision were broadly diachronous and
occurred in Late Paleocene/earliest Eocene time in Deformation of Caribbean crust and basin
western Cuba (Bralower and Iturralde-Vinent, 1997; formation
Gordon et al., 1997), Early to Middle Eocene in
central Cuba (Hempton and Barros, 1993), Middle Plate tectonics within collages of continental and
Eocene to Recent in Hispaniola (Mann et al., 1991) island-arc lithosphere like the Caribbean is well rec-
and Late Eocene to Early Oligocene in Puerto Rico ognized to be more complicated than that in the
(Dolan et al., 1991). Arc rocks of Early to Late Cre- oceans because of the existence of many older faults
taceous age are found in a continuous belt along the which act as lines of weakness and because silica-
Aves Ridge, the remnant arc produced by rifting of rich and feldspar-rich rocks of continents and island
the Grenada back-arc basin, and the Leeward Antilles arcs deform more easily at low temperatures than
along the northern coast of South America (Av6 Lalle- do oceanic basalts (Fig. 2). These facts explain the
mant, 1997) (Fig. 2). Although the lack of reliable iso- broad (~200-250 km) zones of plate-edge seismic-
topic ages does not allow recognition of di~chroneity ity and late Neogene deformation along all margins
in the age of the arc, the arc is adjacent to a west of the Caribbean plate as well as the diffuse zones
to east-younging fold-thrust belt along the northern of seismicity and active faulting within the Chortfs
margin of South America (Av6 Lallemant, 1997). block suggestive of large-scale, internal plate de-
Because all arc segments initiated during the formation (Manton, 1987; Gordon and Muehlberger,
Early Cretaceous and exhibit lithologic and geo- 1994) (Fig. 1). This complex crustal and active plate
chemical similarities, several groups of workers setting leads to basin subsidence in response to a vari-
have interpreted circum-Caribbean island-arc rocks ety of subsidence mechanisms as well as basin inver-
as a continuous volcanic arc chain that ringed sions related to abruptly changing tectonic settings.
the Cretaceous Caribbean oceanic plateau (Mal-
fait and Dinkelman, 1972; Pindell and Barrett,
1990) (Fig. 2). This apparently continuous vol- PREVIOUS CLASSIFICATIONS AND REGIONAL
canic chain has been called the 'Great Arc of the STUDIES OF CARIBBEAN SEDIMENTARY BASINS
Caribbean' (Burke, 1988), the 'Mesozoic Caribbean
Arc' (Bouysse, 1988), and the 'Proto-Antillean Arc' There have been many previous classifications
(Donnelly et al., 1990). In this chapter, I adopt the and regional studies of Caribbean sedimentary basins
term 'Great Arc' for three reasons: (1) brevity; (2) within a plate-tectonic framework. Gonzalez de
the age of the arc in the Greater Antilles ranges into Juana et al. (1980) carried out a thorough com-
the Cenozoic and therefore is not always restricted to pilation on sedimentary basins in Venezuela for the
the Mesozoic; and (3) the extent of island-arc rocks Venezuelan oil industry. Burke et al. (1984) com-
related to the arc may extend far beyond the present piled information on Mesozoic rifts in the Caribbean
geographic area of the Greater and Lesser Antilles. and Gulf of Mexico related to the breakup of North
and South America along with post-Eocene strike-
Back-arc basins associated with the Great Arc slip basins from the southern and northern margins
of the Caribbean. Ladd and Buffler (1985) compiled
Paleogene back-arc basins are present along most data on trench and forearc basins of the Middle
of the length of the Great Arc of the Caribbean America arc and Speed and Westbrook (1984) com-
(Fig. 2). Marine heat-flow measurements and depth piled data on forearc and intra-arc basins of the
to basement calculations using marine seismic pro- Lesser Antilles arc as part of data syntheses spon-
files from the Yucatan basin suggest that it formed sored by the Ocean Drilling Program. Burkart and
during a brief period of northeasterly extension Self (1985) and Manton (1987) compiled data on
between Paleocene and Early Eocene time (Rosen- active rift basins of the Chortfs block. Eva et al.
crantz, 1990). Similar calculations in the Grenada (1989) compiled information from mainly onland
back-arc basin indicate that it formed during the rift, arc and strike-slip basins in Venezuela, Trinidad
Paleocene hiatus in arc activity along the Lesser and the Leeward Antilles. Holcombe et al. (1990)

CARIBBEAN SEDIMENTARY BASINS: CLASSIFICATION AND TECTONIC SETTING 7
and Ladd et al. (1990) compiled information on faults; (3) fault-angle depressions parallel to a sin-
submarine basins from the plate interior and ac- gle strike-slip fault trace; (4) fault-flank depressions
tive plate margins, respectively, as part of the GSA between transverse secondary folds or normal faults;
Decade of North American Geology volume on the and (5) ramp or 'push-down' basins between reverse
Caribbean. St6phan et al. (1990) presented fourteen or thrust faults related to strike-slip movement (Cob-
reconstructions of the Caribbean with superimposed bold et al., 1993) (Fig. 3A). All of these late Neo-
paleogeographic information for the period from the gene basin types typically mark zones of strike-slip-
Jurassic to the present-day. Dolan et al. (1991) com- related tectonic subsidence and are found offshore or
piled information and presented a tectonic synthesis in topographically low onland depressions or valleys.
of onland Paleogene sedimentary basins of His- Areas of most rapid tectonic uplift in both the
paniola and Puerto Rico. Pindell (1995) compiled northern and southern Caribbean region are often lo-
information on rifts related to North America-South calized on restraining bend strike-slip fault segments
America breakup and along with foreland basins or 'push-ups' related to shortening at a discon-
related to the diachronous collision of the Great Arc tinuity or 'step' along a throughgoing strike-slip
of the Caribbean and the passive margins of North fault. Because these bends are usually the sites of
and South America. rapid, long-term (>5 m.y.) uplift, the bends typi-
cally form deeply eroded mountainous areas that are
typically structural domes exposing Cretaceous and
BASIN CLASSIFICATION USED IN THIS OVERVIEW older basement rocks.
Fig. 3 summarizes the nomenclature I use in this Island-arc basins

chapter to classify Caribbean sedimentary basins.
Four main types of basins are recognized that are Island-arc basins include trench-fill basins, fore-
associated with strike-slip, island-arc, collisional and arc basins, intra-arc basins bounded by highs or
rift environments. volcanoes within the volcanic arc, and back-arc
basins (Fig. 3B). The most prominent examples of
Strike-slip basins island-arc basins in the Caribbean include back-arc
basins of the active Middle America arc (Median-
Using nomenclature developed by geologists in Nicaraguan, Mann et al., 1990), the extinct Greater
California and New Zealand, I classify Caribbean Antilles segment of the Great Arc (Yucat~in basin,
strike-slip basins into five basin types based on their Rosencrantz, 1990), and the active Lesser Antilles
bounding fault structure: (1) pull-apart basins pro- arc (Grenada basin, Bird et al., Chapter 15). Is-
duced by extension at a discontinuity or 'step' along land-arc basins of the Great Arc are commonly
a section of a strike-slip fault; (2) fault-wedge basins deformed and require careful mapping to delineate
occurring at intersections of bifurcating strike-slip their extent and internal facies.
Fig. 3. Basin classification nomenclature used in this chapter to classify Caribbean sedimentary basins shown on Figs. 5-10. (A)
Strike-slip basin types. (B) Island-arc basin types. (C) Collisional basin types. (D) Rift basin types.

8 R MANN
Collisional basins (1997) make an excellent tool for the study and clas-
sification of Caribbean sedimentary basins. On these
This basin type includes foreland or foredeep maps, free-air gravity highs marked by the yellow
basins, which are by far the most extensive and and orange colors correspond to seafloor highs that
thickest of all the basin types shown on Fig. 3. In the include active volcanic arcs, remnant volcanic arcs,
Caribbean, these basins mark the flexure of the conti- uplifted oceanic plateau crust, peripheral bulges in
nental or thinned crust of the North and South Amer- flexed Mesozoic oceanic crust of the Atlantic Ocean
ica plates beneath the overriding thrust sheets of the and Cenozoic crust of the Pacific Ocean and car-
Great Arc of the Caribbean. Piggyback basins bonate platforms and isolated banks. Fracture zones
which can also form in non-collisional accretionary in oceanic crust are expressed as fine lineaments
prism settings like the Barbados Ridge complex in traceable over distances up to several hundred kilo-
front of the Lesser Antilles arc (Huyghe et al., Chap- meters. Trenches at subduction zones, sedimentary
ter 14) m form and are filled while being carried on accretionary wedges, and major sedimentary basins
moving thrust sheets (Ori and Friend, 1984). of the types shown on Fig. 3 are marked by large
free-air gravity lows.
Rift basins I have divided the Caribbean gravity data set into
six sub-areas that allow better resolution of basins in
This basin type includes full-grabens or rifts and the individual areas (Fig. 4). These areas include: (1)
half-grabens or rifts. Full-graben means a graben basins associated with the Middle America trench,
bounded on both sides by normal faults while half- arc, and back-arc in the western Caribbean (Cen-
graben means a graben bounded only on one side tral America) (Fig. 5); (2) basins associated with
by a normal fault. The full and half types can occur the North America-Caribbean plate boundary in the
singly or together and be linked by transverse strike- northern Caribbean (Fig. 6); (3) basins associated
slip faults called transfer faults. The best examples with the Lesser Antilles trench, arc, and back-arc in
of uninverted rifts in the Caribbean are Jurassic rifts the eastern Caribbean (Fig. 7); (4) basins associated
related to the breakup of North and South America with the South America-Caribbean plate bound-
found in the southeastern Gulf of Mexico (Marton ary in the southern Caribbean (Fig. 8); (5) basins
and Buffler, Chapter 3) and Paleogene rifts related to associated with the Panama arc-South America col-
the early formation of the Cayman trough pull-apart lisional zone in the southwestern Caribbean (Panama
basin (Leroy et al., 1996). and Costa Rica) (Fig. 9); and (6) basins associated
with the central Caribbean plate (Nicaraguan Rise,
Inverted basins Colombian basin, and Venezuelan basin) (Fig. 10).
Because free-air gravity is a close approximation
Structural inversion of basins means that the of seafloor bathymetry, only those submarine Ceno-
basin-controlling extensional faults reversed their zoic basins with prominent morphologic expression
movement because of convergent tectonics and the can be distinguished on these maps. Older deformed
basin was turned inside out to form a present-day basins from previous tectonic phases might be ex-
mountain range (Williams et al., 1989). Because of pressed as a gravity high or intermediate gravity
the continuing activity along its margin there are value. For this reason, the offshore basins classified
many examples of inverted Caribbean sedimentary in this study are generally Cenozoic basins gener-
basins that include inverted Jurassic rifts in north- ated during the more recent phases of Caribbean
western South America (Lugo and Mann, 1995), strike-slip and subduction tectonics. For this reason,
inverted intra-arc basins of the Greater Antilles seg- these maps should not be considered as complete
ment of the Great Arc (Mann and Burke, 1990; compilations of all Caribbean sedimentary basins.
Dolan et al., 1991), and inverted forearc basins of
the Middle America arc (Kolarsky et al., 1995a).
Inverted basins provide valuable insights into the BASINS AND MAJOR TECTONIC FEATURES
early stratigraphic history of basins provided that ASSOCIATED WITH THE MIDDLE AMERICA TRENCH,
their sediments can be well dated and their structural ARC, AND BACK-ARC (WESTERN CARIBBEAN)
overprint can be removed.
Rifts associated with the f r a g m e n t a t i o n of the
western Chortis block
USE OF GRAVITY MAPS TO ILLUSTRATE CARIBBEAN
SEDIMENTARY BASINS Seven approximately north-south-striking Neo-
gene rifts and half-rifts are present in the northwest-
Gridded, 2-min, satellite-derived free-air gravity ern corner of the Caribbean plate (Chortfs block)
data compiled and described by Sandwell and Smith between the Median back-arc basin and the North

Fig. 4. Map showing locations of regional gravity maps of Sandwell and Smith (1997) used in this chapter in Figs. 5-10 to illustrate
regional tectonic features.
America-Caribbean strike-slip zone in Guatemala, Inverted forearc basin rocks adjacent to the
Honduras and E1 Salvador (Nos. 1-7 in Fig. 5). subducting Cocos Ridge
Little is known about of the exact age for the onset
of rifting and the thickness of the sedimentary fill of Shallow subduction of the Cocos Ridge in Late
these rifts. Several authors have attributed the trans- Miocene to Recent time has inverted a marine fore-
verse nature of the faulting to internal deformation arc basin of Oligocene and Miocene age between
of the Caribbean plate as it moves eastward relative the arc and trench (Kolarsky et al., 1995a) (No. 13)
to the North America Plate along concave south- along with trench-slope facies on the Pacific penin-
ward, left-lateral strike-slip faults of the Motagua- sulas of Panama and Costa Rica (Corrigan et al.,
Polochfc system (No. 8 in Fig. 5) (Plafker, 1976; 1990; Collins et al., 1995) (No. 14). This inverted
Burkart and Self, 1985). basin is collinear with the undisturbed, offshore
Sandino forearc basin to the north along the margin
of Nicaragua and E1 Salvador (No. 16).
Median-Nicaraguan back-arc basin
Cocos Ridge
This late Neogene back-arc basin forms a promi-
nent, 800-km-long structural depression parallel to
The Cocos Ridge (No. 15) is a hotspot trace of
the Middle America volcanic arc and trench (No. 9).
the Galapagos hotspot that stands 2 to 2.5 km higher
This basin is most prominent in Nicaragua where
than the surrounding seafloor and is presently sub-
it is occupied by two large lakes (Managua and
ducting beneath the southern Middle America trench
Nicaragua) (No. 10). Late Quaternary arc volca-
(Kolarsky et al., 1995a). Collins et al. (1995) pro-
noes occur at the edges, in the center and adjacent
pose on the basis of detailed biostratigraphic work
to the basin. Back-arc basin sedimentary rocks of
that the Cocos Ridge contacted the Middle AmeriCa
Oligocene to Neogene age are inverted along the
trench about 3.6 Ma and inverted the back-arc area
southeastern extension of the back-arc basin in Costa
of Costa Rica by 1.6 Ma.
Rica (No. 11). This localized back-arc basin inver-
sion is Late Miocene to Recent in age and is related
to the shallow subduction of the Cocos Ridge (Ko- Panama fracture zone
larsky et al., 1995a) (No. 15). The Median back-arc
basin to the north (No. 12) is a less distinctive and This fault (No. 17) is a right-lateral transform
linear basinal feature than the Nicaraguan basin to fault that separates oceanic crust of the Cocos and
the south. Nazca plates.

10 P. M A N N
Fig. 5. Geosat free-air gravity map of marine areas associated with the Middle America trench, arc and back-arc in the western Caribbean
(Central America). Gravity highs are shown by darker colors and lows are shown by lighter colors. Numbers identify Cenozoic basins
that are described in the text.
Fig. 6. Geosat free-air gravity map of marine areas associated with the North America-Caribbean plate boundary in the northern
Caribbean. Gravity highs are shown by darker colors and lows are shown by lighter colors. Numbers identify Cenozoic basins that are
described in the text.

C A R I B B E A N S E D I M E N T A R Y BASINS" C L A S S I F I C A T I O N A N D T E C T O N I C S E T T I N G 11
Fig. 7. Geosat free-air gravity map of marine areas associated with the Lesser Antilles trench, arc and back-arc in the eastern Caribbean.
Gravity highs are shown by darker colors and lows are shown by lighter colors. Numbers identify Cenozoic basins that are described in
the text.
Fig. 8. Geosat free-air gravity map of marine areas associated with the South America-Caribbean plate boundary in the southern
Caribbean. Gravity highs are shown by darker colors and lows are shown by lighter colors. Numbers identify Cenozoic basins that are
described in the text.

12 E MANN
north from rough Cocos lithosphere created at the

Galapagos rift system to the south (Protti et al.,
1995). Forearc morphology adjacent to the Gala-
pagos seafloor and the Cocos Ridge is tectonically
eroded by the higher standing and rougher seafloor
southeast of the rough-smooth boundary (von Huene
et al., 1995; Kolarsky et al., 1995a).
BASINS AND MAJOR TECTONIC FEATURES

ASSOCIATED WITH THE NORTH
AMERICA-CARIBBEAN PLATE BOUNDARY
(NORTHERN CARIBBEAN)
North America-Caribbean foreland basin and

active plate boundary in Central America, Cuba,
Hispaniola, and the Puerto Rico trench
A semi-continuous foreland basin recording the

collision between the Great Arc of the Caribbean
and the passive margin of North America can be
traced from the Sepur foreland basin of northern
Central America (No. 1 in Fig. 6), along the eastern
edge of the Yucatan Peninsula (Rosencrantz, 1990;
Lara, 1993; No. 2), along the northern (Denny et al.,
1994; No. 3) and northeastern coasts of Cuba (Ball
et al., 1985; No. 4), along the northwestern (Dillon
et al., 1992; No. 5) and northeastern (Dolan et al.,
1998; No. 6) coasts of Hispaniola, and in the Puerto
Fig. 9. Geosat free-air gravity map of marine areas associ- Rico trench (Masson and Scanlon, 1991; Grindlay
ated with the Panama arc-South America collisional zone in et al., 1997; No. 7). The age of the foreland basin
the southwestern western Caribbean (Panama and Costa Rica). is diachronous with Late Cretaceous thrust-related
Gravity highs are shown by darker colors and lows are shown subsidence in northern Central America (Rosenfeld,
by lighter colors. Numbers identify Cenozoic basins that are
described in the text. 1990), Paleocene-Early Eocene subsidence in west-
ern Cuba (Bralower and Iturralde-Vinent, 1997),
Early to Middle Eocene in central Cuba (Hemp-
Rough-smooth boundary of the Cocos plate ton and Barros, 1993), Middle Eocene to Recent
in Hispaniola (Mann et al., 1991) and Late Eocene
This boundary (No. 18) separates a smooth Cocos to Early Oligocene in Puerto Rico (Dolan et al.,
lithosphere created at the East Pacific Rise to the 1991).
Fig. 10. Geosat free-air gravity map of marine areas associated with the central Caribbean plate (Nicaraguan Rise, Colombian basin,
Beata Ridge, Venezuelan basin). Gravity highs are shown by darker colors and lows are shown by lighter colors. Numbers identify
Cenozoic basins that are described in the text.

Yucat~in back-arc basin Muertos trough and 'forearc' basin
The Yucatfin basin formed in Paleocene time be- The Muertos trench (No. 18) accommodates
hind the Cuban segment of the Great Arc of the northward underthrusting of the Caribbean oceanic
Caribbean as it moved to the north and north- plateau of the Venezuelan basin beneath Hispaniola
east prior to its collision with the Bahama Plat- (Ladd et al., 1990; Dolan et al., 1998). A forearc-
form (Rosencrantz, 1990). The basin exhibits three type basin has formed on the overriding plate south
sub-basins. The West Yucatan basin (No. 8) is an of Hispaniola (Ladd et al., 1990; No. 19) but is not
oceanic-floored pull-apart basin formed along left- associated with a volcanic arc probably because the
lateral faults bounding the Yucatan Peninsula. The angle of subduction of the Caribbean plate is too low
Central Yucatan basin (No. 9) and Cayman Rise (No. to generate wellling.
10) are basins formed on stretched arc or continen-
tal crust thinned in a back-arc setting. The Cayman
Ridge (No. 11) south of the Cayman Rise could be Basins associated with the Anegada fault zone
considered a remnant arc although it has been strongly
overprinted by strike-slip faulting related to the active This fault (No. 20) has been interpreted by Jany et
plate boundary in the Cayman trough (No. 12). al. (1990) as an active right-lateral fault bounding the
eastern edge of a Puerto Rico-Hispaniola microplate
(Jany et al., 1990). Two right-steps along the fault
Cayman trough are interpreted as pull-apart basins that formed by
right-lateral motion in Late Miocene to Recent time.
The Cayman trough (No. 12) is an 1100-km-long Gill et al. (Chapter 13) propose that motion along
pull-apart basin that began its protracted history of the Anegada fault zone is left-lateral, rather than
oceanic spreading at a 100-kin-long spreading ridge right-lateral on the basis of detailed stratigraphic
during the Early Eocene (Rosencrantz et al., 1988). studies on St. Croix (U.S. Virgin Islands) to the
The eastern end of the trough (No. 14) is marked by south of the fault.
half-grabens of Paleocene-Eocene age (Leroy et al.,
1996) that may be coeval with an inverted Paleocene-
Eocene graben in Jamaica (Mann and Burke, 1990). BASINS AND MAJOR TECTONIC FEATURES
ASSOCIATED WITH THE LESSER ANTILLES TRENCH,
ARC, AND BACK-ARC(EASTERN CARIBBEAN)
Basins and inverted basins associated with the Aves Ridge remnant arc
southern edge of the Cayman trough
The Aves Ridge (No. 1 in Fig. 7) is a remnant
Late Neogene basin formation, restraining bend arc formed when Paleogene east-west opening of
uplifts, and inverted Paleogene rifts in this area the Grenada back-arc basin separated the ridge from
are linked to left-lateral strike-slip movements along the Lesser Antilles arc (Bouysse, 1988; Bird et al.,
the Enriquillo-Plantain Garden fault zone (No. 13). Chapter 15). Dredge hauls and marine geophysics
Some basins like the Tela of northern Honduras indicate that the ridge formed part of the Late
(No. 14) (Av6 Lallemant and Gordon, Chapter 8; Cretaceous Great Arc that ceased activity by the
Manton and Manton, Chapter 9) are not clearly time of back-arc opening of the Grenada basin in
linked to a specific fault zone and instead appear Early Paleogene time.
to be part of broad structural borderland within the
broad strike-slip plate boundary zone.
Grenada back-arc basin
Convergent strike-slip basins of the Hispaniola The Grenada back-arc basin (No. 2) with an
restraining bend average water depth of 2-3 km, contains 2 km
(north) to 9 km (south) of Cenozoic sediment derived
Convergence of the eastward-moving Caribbean from both the erosion of South America and the
plate relative to the southeastern extension of the Lesser Antilles arc (Bouysse, 1988). Opening of the
Bahama Platform has led to localized convergence basin was in an east-west direction (Bird et al.,
and topographic uplift in Hispaniola (Mann et al., 1993). The gravity low of the Grenada basin can be
1995). Three late Neogene basins in Hispaniola traced along much of the margin of northern South
(Nos. 15, 16, 17) are the thrust-bounded ramp type America where it is oriented east-west, is narrower,
(Mann et al., Chapter 12) (Fig. 3A). and parallel to collisional structures of the margin.

1 4 E MANN
Lesser Antilles volcanic arc Caribbean plate. The north-south lithospheric trace
and the east-west E1 Pilar fault zone of Trinidad and
The Lesser Antilles volcanic arc (No. 3) initi- northern Venezuela (No. 13) can be seen to form a
ated in Early Cretaceous and has remained active to continuous and curving lineament.
the present (Speed and Westbrook, 1984; Bouysse,
1988). Its position at the leading edge of the east-
ward-moving Caribbean plate means that it may Barbados Ridge accretionary complex and
be far-traveled and probably originated somewhere deformation front
in the eastern Pacific Ocean (Pindell and Barrett,
1990). Underthrusting of Jurassic-Cretaceous age This accretionary wedge (No. 7) between 100
oceanic crust of the Atlantic Ocean beneath the and 300 km wide and from 0.5 (toe of slope)
Lesser Antilles results in line of active calc-alkaline to 20 km (above lithospheric trace) thick consists
volcanoes forming the volcanic arc. The gravity high of the mainly clastic fluvial and pelagic sediments
of the volcanic arc can be traced to the southwest offscraped from the downgoing Atlantic ocean floor
along the northern margin of South America and to (Ladd et al., 1990). Southward widening of the
the northeast through the Virgin Islands and Puerto complex reflects the fluvial addition of material
Rico. from the Orinoco delta area south of Trinidad (No.
6). The deformation front is marked by the most
Kallinago basin eastward thrust fault juxtaposing the Barbados Ridge
accretionary complex with undeformed seafloor of
This basin (No. 4) forms an intra-arc basin with the Atlantic Ocean. This front is roughly east-west
the northern Lesser Antilles volcanic arc and formed and irregular in the area to the east of Trinidad.
in the Late Miocene by the westward migration of Piggyback basins (Fig. 3C) and shale diapirs derived
the volcanic line from the islands on its eastern from muds in the prodelta area of the Orinoco River
flank (Limestone Caribbees). The Kallinago basin is are common in this area (Huyghe et al., Chapter 14).
not a typical rift basin related to back-arc spreading
because the volcanic line jumped westward away
from the trench and not toward the trench as found BASINS AND MAJOR TECTONIC FEATURES
in most back-arc basins. McCann and Pennington ASSOCIATED WITH THE SOUTH
(1990) attributed this unusual behavior related to the AMERICA-CARIBBEAN PLATE BOUNDARY
subduction of the Barracuda fracture zone ridge (No. (SOUTHERN CARIBBEAN)
9) which lowered the angle of subduction and caused
the volcanic arc to migrate westward. The Aves Ridge (No. 1 in Fig. 8), the Grenada
back-arc basin (No. 2), the Lesser Antilles volcanic
Tobago trough arc (No. 3), and the Tobago trough (No. 4), which
extend into this area, are described above and are
This 10-km-thick, Miocene to Recent basin (No. also shown on Fig. 7.
5) is bounded on its eastern edge by back-thrusts
within the accretionary wedge of the Lesser Antilles
arc (Barbados Ridge complex, No. 7) (Speed et al., Guyana passive margin of South America
1989). The Tobago trough extends to the west along
the northern margin of South America. This Cretaceous-Recent margin (No. 5 in Fig. 8)
formed by rifting and strike-slip of the Africa plate
Lithospheric trace of the Lesser Antilles past the South America Plate in earliest Cretaceous
subduction zone time (Pindell and Barrett, 1990). The margin projects
into a buried passive margin buried beneath the Gulf
The contact or lithospheric trace between crys- of Paria west of Trinidad and may control the
talline rocks of the Lesser Antilles arc and down- location of strike-slip faults in this area (Babb and
going oceanic crust of the Atlantic Ocean (No. 6) Mann, Chapter 18).
lies at a depth of about 20 km north of Trinidad but
is marked by the line of demarcation between faint
northeast gravity trends on strike with Atlantic frac- Orinoco delta
ture zone trends and prominent north-south trends
of the Lesser Antilles arc. The island of Tobago The Orinoco River drains a large area of the
with a Cretaceous to Eocene record of arc activity is northeastern South American continent and forms
located just to the west of this contact and is there- one of the major shelf-margin deltas in the world
fore the most eastward outcrop of arc rocks of the (No. 6).

Eastern Venezuelan and Maracaibo basins BASINS AND MAJOR TECTONIC FEATURES
ASSOCIATED WITH THE PANAMA
This basin is a major foreland basin marked by ARC-SOUTHWESTERN CARIBBEAN COLLISIONAL
a 150 mGal gravity low formed by oblique con- ZONE (SOUTHWESTERNCARIBBEAN)
vergence between the South American continent
(Guyana Shield) and the Caribbean arc system in C6baco basin complex
Oligocene and Miocene time (di Croce et al., Chap-
ter 16; Flinch et al., Chapter 17). The basin is This submarine basin complex in a shelf setting
subdivided into two sub-basins which increase in (No. 1 in Fig. 9) is an active pull-apart formed at a
age from east (Maturfn, Oligocene to Recent, No. left-step in the Azuero-Son~i fault zone of western
7) to west (Gu~irico, Eocene to Pliocene, No. 8). Panama (Kolarsky et al., 1995b).
The western extension of the foreland basins is rep-
resented by the Maracaibo basin (Late Paleocene- Tonosi basin
Eocene, No. 9) (Lugo and Mann, 1995). The change
in strike of the older, western part of the fore- This now folded and uplifted Oligocene-Miocene
land basin (Maracaibo, No. 9) is related to late turbiditic basin (No. 2) appears to have been a
Neogene northward displacement of the Maracaibo forearc basin that has now become inverted as a
block along the fight-lateral Bocon6 fault (No. 13) result of strike-slip movements along the oblique-
and the left-lateral Santa Marta-Bucaramanga fault slip margin of southwestern Panama (Kolarsky et al.,
(No. 14). 1995b).
Riffs of the Canal area

Accreted rocks of the South America passive
margin and Great Arc of the Caribbean These late Neogene basins (No. 3) formed as a
consequence of diffuse east-west extension within
These fold-thrust belts (No. 10) contain mixtures this topographically lowest part of the Panama Isth-
of arc and margin-related lithologies formed during mus. These basins may have formed as a response
the oblique Cenozoic collision of the arc and the to bending of the isthmus of Panama following its
passive margin (Av6 Lallemant, 1997). collision with northwestern South America in Late
Miocene to Early Pliocene time (Mann and Kolarsky,
Cariaco pull-apart basin on the El Pilar fault 1995).
zone
San Bias 'forearc' basin
The Cariaco pull-apart basin (No. l l) is a
This basin (No. 4) has formed in a 'forearc'
1400-m-deep, closed depression in the Venezuelan
setting above the accreted North Panama deformed
shelf at a 35-km south-step between the right-lateral
belt in late Neogene times (Reed and Silver, 1995).
E1 Mor6n and E1 Pilar strike-slip faults. Schubert
(1984) estimated that 70 km of right-lateral motion Bayano-Chucanqu6 basin
on the faults was necessary to produce the basin over
the last 2 million years. This basin (No. 5) has formed in a large syncline
formed in response to the bending and strike-slip
South Caribbean marginal fault deformation of the Isthmus of Panama following its
collision with the South America margin (Mann and
This fault (No. 12) forms the deformation front Kolarsky, 1995).
of a large accretionary wedge formed by the un-
derthrusting of the Caribbean oceanic plateau and Sambfi basin
its overlying sedimentary cover beneath the South
American continent (Ladd et al., 1990). This basin (No. 6) appears to be a pull-apart basin
formed at a left-step in the left-lateral Samb6 fault
zone.
Maracaibo block
Pearl Islands basin
This triangular-shaped block of continental crust
is bounded to the east by the Bocon6 right-lateral This Middle Miocene-Pleistocene basin (No. 7)
fault (No. 13) and to the east by the left-lateral Santa formed as a small foreland basin in front of east-dip-
Marta-Bucaramanga fault (Mann et al., 1990) (No. ping reverse faults of the East Panama deformed belt
14). (Mann and Kolarsky, 1995).

1 6 R MANN
Colombian accretionary complex and forearc Beata Ridge

basin
The Beata Ridge (No. 4) is marked by a trian-
This margin (No. 8) developed in response to gular-shaped uplift of oceanic plateau crust at the
eastward subduction of oceanic crust of the Nazca place where the Caribbean sea is narrowest, between
plate beneath Colombia (Westbrook et al., 1995). northern South America and Hispaniola. Driscoll
and Diebold (Chapter 20) interpret the uplift as a
Atrato-San Juan basin mainly relict extensional fault block related to the
formation of the Cretaceous oceanic plateau while
This basin (No. 9) formed along the approximate Mauffret and Leroy (Chapter 21) emphasize its late
suture zone between the Panama arc and the South Neogene uplift history on thrust faults and its role
American continent (Bueno Salazar, 1989; Kellogg as major tectonic boundary between the Colombian
and Vega, 1995). and Venezuelan basins.
Venezuelan basin
BASINS AND MAJOR TECTONIC FEATURES
ASSOCIATED WITH THE CENTRAL CARIBBEAN The low-relief Venezuelan basin (No. 7) is a
PLATE rectangular area of oceanic plateau crust bounded on
the north by the Muertos trench, on the south by
Basins and carbonate banks of the Nicaraguan the South Caribbean marginal fault, on the west by
Rise the Beata Ridge, and on the east by the Aves Ridge,
the remnant volcanic arc of the Lesser Antilles. A
The Nicaraguan Rise is a broad submarine swell prominent east-west arch trends parallel to the long
underlain by island arc and continental crust of the axis of the basin and may be a regional flexure of the
Chortfs block that extends from northern Central Caribbean plate produced by its ongoing subduction
America to Jamaica and is bounded on the north by at the Muertos trough to the north and the South
the Cayman trough and on the south by the Hess Caribbean marginal fault (No. 6) to the south.
Escarpment (No. 1 in Fig. 10). Carbonate banks
of Cenozoic age occupy structural highs formed
by poorly understood faults with a predominantly TECTONIC EVOLUTION OF THE CARIBBEANPLATE
northeast strike. Holcombe et al. (1990) interpreted AND ITS SEDIMENTARYBASINS
the northeast faults as a set of left-lateral strike-slip
faults bounding a set or more northward-striking rift Two models for Caribbean evolution
basins associated in one case (San Andres trough)
with Quaternary basaltic volcanism present on the There are two contrasting models for the plate-
island of San Andres (No. 2). The linear Hess Es- tectonic evolution of the Caribbean. The first model
carpment (No. 3) has been interpreted by Mann et most recently put forward by Frisch et al. (1992) pro-
al. (1990) as a possible Neogene strike-slip feature poses that the Caribbean region formed during the
whereas others like Driscoll and Diebold (Chapter period of 130 Ma to 80 Ma as South America moved
20) have interpreted it as Cretaceous normal fault southeast away from North America (Fig. l lA).
linked to the Beata Ridge (No. 4) and to the forma- Igneous upwelling in the space that formed be-
tion of the Caribbean oceanic plateau (Fig. 2). tween the two continents is thought to have pro-
duced the anomalously thick oceanic plateau crust
Colombian basin of the Caribbean and Central America (Kerr et al.,
1997; Diebold and Driscoll, Chapter 19; Driscoll and
The low-relief Colombian basin (No. 5 in Fig. 10) Diebold, Chapter 20). This model recognizes some
is underlain by the Cretaceous Caribbean oceanic strike-slip motion along the northern and southern
plateau (Bowland and Rosencrantz, 1988) and is margins of the plate but does not view these offsets
bounded to the north by the Hess Escarpment, to as large enough to restore the Caribbean to a position
the south by the South Caribbean margin fault (Ladd in the eastern Pacific.
et al., 1990; No. 6), and to the east by the Beata An alternative school of thought and adopted in
Ridge (No. 4). van der Hilst and Mann (1994) used the reconstructions shown in this review was first
tomographic data to show that the oceanic plateau formulated by Wilson (1966) and later elaborated
crust of the Colombian basin is underthrust at the by Malfait and Dinkelman (1972), Ross and Scotese
South Caribbean front fault (No. 6) by a distance (1988), Pindell and Barrett (1990), and others. This
of several hundred kilometers beneath the South mobilistic view is that the Caribbean was originally
America margin. an area of eastern Pacific Ocean floor and oceanic

Fig. 11. Two possible origins for the Caribbean. (A) The Caribbean oceanic plateau forms by the separation of North and South America
during the period 130 to 80 Ma (Frisch et al., 1992). Crosses indicate areas of continental crust. The numbers give positions of the
northern margin of South America according to Pindell and Barrett (1990). (B) The Caribbean oceanic plateau forms as normal Pacific
oceanic crust drifts over the Galapagos hotspot, is thickened in the middle and Late Cretaceous, and passes into the gap between North
and South America. The numbers give positions of the leading edge of the Caribbean arc system and oceanic plateau through time,
according to Pindell and Barrett (1990). Note that the positions of continental masses allow 'no free face' for arc migration. The 'free
face' of the Atlantic Ocean acts to channel the arc in an eastward direction.
plateau that has been rafted behind the eastward- Sinton et al., 1997) (Fig. l i B ) . The passage of
moving Great Arc of the Caribbean of Burke (1988) this area of crust from a Pacific realm to an At-
(Fig. 11B). This area of Pacific normal ocean crust lantic one is recorded by the diachronous history
appears to have been modified and thickened into of collisions b e t w e e n the Great Arc at the lead-
the present-day Caribbean oceanic plateau province ing edge of the plateau and the passive margins
in the Cretaceous w h e n the crust drifted over the of North and South A m e r i c a (Pindell and Barrett,
Galapagos hotspot (Duncan and Hargraves, 1984; 1990). These collisions c o m m e n c e in Late Creta-

18 R MANN
ceous time in northern Central America and north- Relative motion path of South America relative
western South America and continue through to to North America
the present-day in the northeastern and southeastern
Caribbean (Fig. 11B). The path shows South America moving away
from North America during the Jurassic through the
Using North A m e r i c a - S o u t h America motion to Late Cretaceous, a process that led to the formation
infer Caribbean tectonics of ocean floor on the sites of the Caribbean and
Gulf of Mexico. The orientation of Mesozoic graben
Because magnetic anomalies and fracture zones is generally perpendicular to this direction except
nearly as old as the times of separation between in regions affected by the independent rotation of
North and South America have been mapped in the the Yucatan block (Marton and Buffler, Chapter 3).
Atlantic Ocean, the motions of North and South While the age of most circum-Caribbean rifts is
America with respect to Africa can be fully de- confined to the Jurassic, the path shows continued
scribed using the vectorial closure condition required separation of the Americas up through the Maas-
by a three-plate system (Pindell and Barrett, 1990; trichtian. Numerous geological studies such as those
MUller et al., Chapter 2). Improved maps of At- by Pessagno et al. (Chapter 5), Marton and Buffler
lantic fracture zones using Geosat gravity data by (Chapter 3), Masaferro and Eberli (Chapter 7), Scott
Mtfller et al. (Chapter 2) has allowed them to more and Finch (Chapter 6), and di Croce et al. (Chapter
precisely reconstruct the motion history of the two 16) show that the Cretaceous was a time of carbonate
Americas. A summary of the Jurassic to recent path passive margin formation atop these early rift struc-
of two points on northern South America relative to tures. These bank margins probably fronted large
a fixed North America using the data of MUller et expanses of Jurassic and Cretaceous ocean crust that
al. (Chapter 2) is shown on Fig. 12. This diagram formed following the separation of the two plates in
illustrates the steadily widening space between the Late Jurassic-earliest Cretaceous time.
two Americas that was presumably filled by oceanic The behavior of the Great Arc of the Caribbean
crust of Jurassic and Early Cretaceous age. This ex- in response to the convergence or strike-slip motion
panse of crust known as the proto-Caribbean Ocean of North and South America during the period of
is thought to have been consumed by the Great Arc 71 Ma to the present-day cannot be predicted with
of the Caribbean from the Late Cretaceous to Recent accuracy given that these larger plate motions will
time. Remnants of the proto-Caribbean Ocean are only be indirectly manifested across multiple strike-
preserved only as small fragments within rocks of slip and subduction Caribbean boundaries. Mann et
the Great Arc (Montgomery and Pessagno, Chap- al. (1995) and Gordon et al. (1997) propose that
ter 10). the trend and direction of the Great Arc during the
The vector diagram in Fig. 12 can be used to Late Cretaceous and Cenozoic is governed mainly
make general inferences about the regional deforma- the direction of a free face, or area of subductable
tional style of the intervening Caribbean plate. For oceanic crust in front of the moving arc. For ex-
example, from the Late Jurassic to Maastrichtian, ample, the presence of the Bahamas Platform led to
one would expect a generally divergent tectonic arc collision and reorientation of the arc to subduct
style to pervade much of this region and from Atlantic oceanic crust in a more eastward direction.
Maastrichtian to the Present one would expect a MUller et al. (Chapter 2) propose that 200-300 km of
convergent or strike-slip style to be present (Fig. 12). post-Early Miocene north-south convergence across
However, this direct dependence of Caribbean de- the Caribbean plate may have led to significant un-
formational style on the relative motion of North derthrusting of the Caribbean plate beneath North
and South America assumes that motion is taken and South America and may have modified existing
up along a single plate boundary, such as dur- sedimentary basins.
ing Jurassic separation. Studies including Marton
and Buffler (Chapter 3) show that even the young
North America-South America Plate boundary dur- MAIN PHASES OF CARIBBEAN BASIN DEVELOPMENT
ing Jurassic time was multi-branched and involved WITHIN THE FRAMEWORK OF NORTH
the motion of an intervening microcontinent, the AMERICA-SOUTH AMERICA RELATIVE MOTION
Yucat~in block. As the gap between the Americas HISTORY
widened, later Cretaceous-Cenozoic relative plate
motions acted across at least two plate boundaries Using the second plate-tectonic model shown
(northern and southern Caribbean arc or strike-slip in Fig. liB, five main phases of basin evolution
boundaries). Therefore, the North America-South can be predicted for the margins of the North and
America motions shown on Fig. 12 are only indi- South America plates. These phases include pre-rift
rectly manifested in Caribbean deformation. phase, Late Jurassic rift phase, Cretaceous passive

Fig. 12. Relative plate motion vectors of three points of northern South America with respect to a fixed North America based on data
presented by M~iller et al. in Chapter 2 of this volume. The position of South America with respect to North America provides a
framework in which to base key events in Caribbean evolution such as the entry and diachronous collision of the Great Arc of the
Caribbean. Points in millions of years along the vectors correspond to the ages of plate reconstructions give in Figs. 13-25 of this chapter.
margin phase, Late Cretaceous-Recent arc-passive South America shown in Fig. 13. These provinces in-
margin collisional phase, and late Cenozoic strike- clude: (1) Pan-African crustal age province of Africa
slip phase. I subdivide thirteen plate reconstructions and Brazil; (2) Grenville crustal age province of
of the Caribbean based on the plate parameters of North and South America that includes a possi-
Miiller et al. (Chapter 2) into these four phases. ble continuation through the Oaxaca area of south-
Several Cretaceous and Cenozoic structural phases ern Mexico and the Chortfs block (Renne et al.,
characterize the overriding Great Arc and the adjacent 1989; Hutson et al., 1998); and (3) Guyana Shield
oceanic plateau and Chortfs block. For example, the of pre-Grenville age (> 1.2 Ga) of northern South
oceanic plateau undergoes a two-phase Cretaceous America. The Yucat~in block fills the central part
volcanic and stretching event (Driscoll and Diebold, of the Gulf of Mexico and is presumably underlain
Chapter 20), the Great Arc may have experienced a by a prong of the Appalachian-Marathon-Ouachita
subduction polarity reversal (L6bron and Perfit, 1994; orogenic belt (Marton and Buffler, Chapter 3).
Draper et al., 1996; Kerr et al., 1997; Montgomery
and Pessagno, Chapter 10) and the Chortfs block Late Jurassic rift phase
experienced a Late Cretaceous folding and faulting
event (Scott and Finch, Chapter 6). Rifts of Late Jurassic age in the Caribbean form
part of a band of rifts associated with the early open-
Pre-rift phase ing of the central and northern Atlantic that crudely
follow orogenic grains from the North Atlantic to
Plate reconstructions such as those by Pindell and Guyana. By Oxfordian time, rifts are active along
Barrett (1990), Marton and Buffler (Chapter 3) and the northern Gulf of Mexico, the southeastern Gulf
Pszcz6tkowski (Chapter 4) leave no space for the of Mexico (Marton and Buffler, Chapter 3), the Ba-
Caribbean-Gulf of Mexico region in its present posi- hama Platform (Masaferro and Eberli, Chapter 7),
tion when South America is closed up against North the northeastern margin of South America (di Croce
America to reform western Pangea in pre-Late Juras- et al., Chapter 16), and the northwestern margin of
sic time (Fig. 13). Prior to rifting of the Americas South America (Eva et al., 1989; Lugo and Mann,
in the Middle Jurassic, three crustal age provinces are 1995) (Fig. 14). Rifts which extend southward along
present in the future area of rifting between North and the western margin of South America may be related

20 P. M A N N
Fig. 13. Reconstruction of the Caribbean region at 180 Ma (Bajocian). Key to abbreviations: M S M = Mohave-Sonora megashear; TMVB
-- Trans-Mexican volcanic belt; E A F Z -- eastern Andean fault zone.
Fig. 14. Reconstruction of the Caribbean region at 156 Ma (Oxfordian, magnetic anomaly M29). Gray areas represent oceanic crust of
normal thickness. Stippled areas indicate rifted areas. Key to abbreviations: P C - proto-Caribbean oceanic crust (dark line represents
speculative position of spreading ridge)" N B F Z -- northern Bahamas fracture zone; M S M -- Mohave-Sonora megashear; T M V B =
Trans-Mexican volcanic belt; E A F Z = eastern Andean fault zone.

Fig. 15. Reconstruction of the Caribbean region at 145 Ma (Tithonian, magnetic anomaly M19). Dots in Atlantic Ocean represent
magnetic anomaly and fracture zone picks by Mt~ller et al. (Chapter 2) based on interpretation of Geosat gravity images. Key to
abbreviations: PC = proto-Caribbean oceanic crust (dark line represents speculative position of spreading ridge); M S M = Mohave-
Sonora megashear; TMVB = Trans-Mexican volcanic belt; EAFZ = eastern Andean fault zone.
to back-arc rifting produced by subduction at that the Chortfs block occupied a southern extension of
margin. The widening gap between North and South Precambrian and Paleozoic orogenic belts in Mex-
America was presumably occupied by oceanic crust ico and was subsequently displaced eastwards by
generated at a proto-Caribbean spreading ridge. This strike-slip faults in the Cenozoic (Av6 Lallemant and
early oceanic corridor between the Atlantic and Pa- Gordon, Chapter 8; Manton and Manton, Chapter 9)
cific widens through continued rifting and oceanic (Fig. 16).
spreading into the Tithonian (Fig. 15).
Late Cretaceous-Recent arc-passive margin
Cretaceous passive margin phase collisional phase
By earliest Cretaceous time, rifting had ceased, By Late Cretaceous time, the Great Arc of the
the Yucatan block had rotated to its present-day Caribbean and its adjacent oceanic plateau province
position, and a post-rift passive margin section com- was colliding with the passive margin of north-
posed mainly of carbonate rocks had blanketed the western South America and the southern margin of
rift topography in the southeastern Gulf of Mex- northern Central America (Pindell and Barrett, 1990;
ico (Marton and Buffler, Chapter 3), the Bahamas Kerr et al., 1997) (Fig. 17). In northwestern South
Platform (Masaferro and Eberli, Chapter 7), the America, extensive areas of the plateau and arc rocks
northeastern margin of South America (di Croce et accreted to the continental cratonic rocks of north-
al., Chapter 16; Babb and Mann, Chapter 18), and western South America (Kerr et al., 1997). In these
the northwestern margin of South America (Lugo areas, ages of rocks of the Great Arc extend back to
and Mann, 1995). These passive margins enjoyed the Early Cretaceous but ages of the oceanic plateau
open ocean circulation and probably fronted a proto- are generally confined to the Santonian (Kerr et al.,
Caribbean oceanic basin that was several hundred 1997; Sinton et al., 1997). Diebold and Driscoll
kilometers wide (Fig. 16). It is interesting to note (Chapter 19) and Driscoll and Diebold (Chapter 20)
that the Chortfs block experienced a similar rift and present evidence that the oceanic plateau eruption
passive margin history to the above intra-Caribbean event was a two-phase event with the Santonian
margins (Scott and Finch, Chapter 6). Presumably event probably corresponding to the younger event.

22 P. M A N N
Fig. 16. Reconstruction of the Caribbean region at 118 Ma (Aptian, magnetic anomaly C34n). Key to abbreviations: M S M = Mohave-
Sonora megashear; TMVB = Trans-Mexican volcanic belt; EAFZ -- eastern Andean fault zone.
Fig. 17. Reconstruction of the Caribbean region at 83 Ma (Campanian, magnetic anomaly C34n). Key to abbreviation: EAFZ = eastern
Andean fault zone.

By Maastrichtian time, subsidence of the Sepur north and the Colombian trench to the south (Wor-
foreland basin was ending as the Chortfs block was tel and Cloetingh, 1981) (Figs. 22 and 23). By Late
sutured to the area of southern Mexico and the Yu- Miocene, localized convergence between the east-
catan block (Fig. 18) and the Great Arc was migrating ward-moving Caribbean plate and the southeastern
to the northeast towards its eventual collision with the extension of the Bahama Platform led to thrusting and
Bahama Platform in the Late Paleocene and Eocene topographic uplift in Hispaniola (Mann et al., Chap-
(Fig. 19). In Paleocene time, the end of the arc moved ter 12; de Zoeten et al., Chapter 11) (Fig. 24). Along
along a complex strike-slip zone at the eastern edge the southeastern margin of the plate, tectonic activity
of the Yucatan Peninsula (Lara, 1993) and opened involved the Trinidad area (Babb and Mann, Chapter
the Yucatan back-arc basin in its wake (Rosencrantz, 18; di Croce et al., Chapter 16; Flinch et al., Chapter
1990) (Fig. 19). In northwestern South America, a 17) (Fig. 24). By Late Pliocene, the margins of the
Maastrichtian foreland basin associated with the ac- Caribbean had reached its present-day configuration
cretion of oceanic plateau material widened and be- (Fig. 25). Two important tectonic and paleoceano-
gan to affect the area of western Venezuela (Pindell graphic events of this time was closure of the Panama
and Barrett, 1990) (Fig. 18). These foreland basin de- seaway by collision of the Panama arc against north-
posits will later become overprinted by the effects of western South America (Kellogg and Vega, 1995:
the late Neogene collision of the Panama arc with Mann et al., 1995) (Fig. 25) and the collision of the
northwestern South America. Cocos Ridge with southern Central America by about
By Early Eocene, arc-continent collision was 1.6 Ma. The Panama arc collisional event may have
complete in western Cuba and collision proceeded accelerated the northwestward expulsion of the Mara-
in a diachronous manner along the edge of the caibo block into the Caribbean.
Bahamas Platform (Gordon et al., 1997; Masaferro
and Eberli, Chapter 7) (Figs. 19 and 20). This
diachronous collision accompanied transfer of mi- FUTURE WORK ON CARIBBEAN SEDIMENTARY
croplates from the Caribbean plate to the North BASINS
America Plate in a clockwise fashion as forward
progress of the Great Arc was halted by its collision To conclude this review, I would like to leave
with the Bahamas Platform (Mann et al., 1995). the reader with some large-scale Caribbean tectonic
A thin foreland basin formed between the collision problems that could be addressed by future studies
zone and the Bahamas carbonate platform (Hemp- of Caribbean sedimentary basins.
ton and Barros, 1993). Similarly, in northern South
America, collision ended in Eocene time in the Lake Pacific vs. in situ origin of the Caribbean
Maracaibo area of western Venezuela and proceeded
in a diachronous manner eastward along the northern The problem of the origin of the Caribbean is
margin of South America in Middle to Late Eocene by no means solved despite the Pacific-origin ap-
time (Fig. 20). Initiation of oceanic spreading in the proach that I have followed in this introduction and
Cayman trough in Middle Eocene time may be the in the tectonic reconstructions. There are several
result of a change in the direction of the Great Arc problem areas for the Caribbean origin problem.
from a northeastward to an eastward direction to First, North America-South America relative plate
move around the salient formed by the southeast- motion history (Mtiller et al., Chapter 2) (Fig. 12)
ern Bahama Platform (Mann et al., 1995). By Late only indirectly bears on the position of this inter-
Oligocene, the zone of active collision is in the pre- vening Caribbean plate and Great Arc through time.
sent-day area of Puerto Rico on the northern plate Second, paleomagnetic studies in the Caribbean are
boundary and eastern Venezuela on the southern handicapped by several factors: (1) the problem
boundary (Fig. 21). of distinguishing large-scale plate-tectonic rotation
from local structural rotation about vertical axes and
Late Cenozoic strike-slip phase apparent tectonic rotation (cf. MacDonald, 1980, for
a discussion of paleomagnetic data from the Chortfs
The Miocene to Recent period of Caribbean his- block); (2) the inability of paleomagnetism to ad-
tory corresponds to its strike-slip phase since by this dress longitudinal changes in plate position of the
time the arc-continent collisional zones have length- type assumed for an eastward-moving Caribbean
ened and converted into long strike-slip faults along Great Arc and oceanic plateau in Cenozoic time;
the northern and southern edges of the Caribbean and (3) large error limits on existing data (Gose,
plate. During the Middle Miocene, the Cocos and 1985). And, third, studies of individual strike-slip
Nazca plates ruptured along the Galapagos rift prob- offsets are problematic because, as shown on the re-
ably as a response to simultaneous subduction in two constructions, these faults form somewhat late in the
directions beneath the Middle America arc to the Caribbean tectonic history and therefore represent

24 P. M A N N
Fig. 18. Reconstruction of the Caribbean region at 71 Ma (Maastrichtian, magnetic anomaly C32n.2n). Key to abbreviation: EAFZ --
eastern Andean fault zone.
Fig. 19. Reconstruction of the Caribbean region at 55.9 Ma (Early Eocene, magnetic anomaly C25n). Key to abbreviations: YB = Yucat~in
back-arc basin; G B -- Grenada back-arc basin; M B = Maracaibo foreland basin.

CARIBBEAN SEDIMENTARY BASINS" CLASSIFICATION AND TECTONIC SETTING 25
Fig. 20. Reconstruction of the Caribbean region at 41.3 Ma (Middle Eocene, magnetic anomaly C19n). Key to abbreviations: MB =
Maracaibo foreland basin; E A F Z = eastern Andean fault zone.
Fig. 21. Reconstruction of the Caribbean region at 25.5 Ma (Late Oligocene, magnetic anomaly C7An). Key to abbreviation: GB =
Gufirico foreland basin.

26 P. M A N N
Fig. 22. Reconstruction of the Caribbean region at 15.1 Ma (Middle Miocene, magnetic anomaly C32n.2n). Key to abbreviations: G B =
Gu~rico foreland basin; M B = Maturfn foreland basin.
Fig. 23. Reconstruction of the Caribbean region at 11.5 Ma (latest Middle Miocene, magnetic anomaly C5r.2n). Key to abbreviations: G B
-- Gu(trico foreland basin; M B = Maturfn foreland basin.

CARIBBEAN SEDIMENTARY BASINS" CLASSIFICATION AND TECTONIC SETTING 27
Fig. 24. Reconstruction of the Caribbean region at 9.2 Ma (Late Miocene, magnetic anomaly C4Ar.2n). Key to abbreviations: GB --
Gufirico foreland basin; M B = Maturfn foreland basin.
Fig. 25. Reconstruction of the Caribbean region at 3.1 Ma (Late Pliocene, magnetic anomaly C2An.2n). Key to abbreviation: MB =
Maturfn foreland basin.

28 P. MANN
only a small part of the total Caribbean displace- ing and interbedded sedimentary rocks could provide
ment. important clues to the origin of the ophiolites and
There are several promising new approaches for their paleolatitudes through time.
study of the origin of the Caribbean plate that can
augment traditional plate reconstruction and paleo-
Triggering of back-arc basins formation
magnetic methods. Paleoenvironmental studies such
as those by Pessagno et al. (Chapter 5) attempt
Depth to basement calculations and heat-flow
to define changes in the paleolatitude of terranes
measurements for the Yucat~in (Rosencrantz, 1990)
using macro- and micropaleontologic data. Mont-
and Grenada back-arc basins (Bird et al., Chapter 15)
gomery and Pessagno (Chapter 10) point out key
indicate that both basins formed rapidly over a short
indicator rocks, such as red cherts, that can be used
time interval in the Paleogene. Further geophysical
to distinguish a Pacific vs. Atlantic environment of
deposition. Finally, geochemical and high-resolution work and deep-sea drilling is needed to confirm
this history and understand why this basins opened
dating of igneous and metamorphic rocks of rocks of
rapidly and then became dormant despite continued
the Great Arc and oceanic plateau (e.g., Sinton et al.,
subduction beneath the Great Arc.
1997) or the grains in sedimentary rocks from the
allochthonous areas (e.g., Hutson et al., 1998) allows
better constraints on plate reconstructions. Diachronous arc-continent collision and
termination of arc activity
Age and environments of the Caribbean oceanic
plateau The timing of this event summarized on Fig. 11B
could be improved through careful biostratigraphic
Diebold and Driscoll (Chapter 19) and Driscoll and stratigraphic studies as done by Bralower and
and Diebold (Chapter 20) present data showing Iturralde-Vinent (1997) in Cuba and several of the
a two-stage Cretaceous evolution of the plateau papers in this volume.
and suggest that existing DSDP and ODP dated
drill samples from the plateau may constrain only
Amount of allochthoneity of Caribbean arcs
the later, smaller plateau-building event. Further
outcrop studies and deep ocean drilling are needed
The amount of overthrusting of the Great Arc
to constrain this hypothesis.
over the passive margins of North and South Amer-
ica is not well understood because deep seismic data
Polarity reversal of the Caribbean arc
has not been attempted over the arc-continent col-
lision zones. If the arc is far-traveled on a predom-
Montgomery and Pessagno (Chapter 10) note two
inantly unmetamorphosed passive margin sequence,
types of accreted sedimentary material in the Great
potential hydrocarbon deposits may exist at depth
Arc of the Caribbean: Pacific-derived material ac-
in areas where crystalline rocks are present at the
creted when the arc was west or southwest-facing
surface.
and Atlantic-type material accreted when the arc was
east or northeastward-facing as it is today. Structural
and stratigraphic outcrop studies are needed to con- Driving forces of Caribbean plate motion
firm the existence and age of the proposed Early
Cretaceous arc polarity reversal discussed by these Mann et al. (1995) and Mann (1996) proposed
authors and previous workers like L6bron and Perfit that the Caribbean plate is driven as a response to
(1994) and Draper et al. (1996). dense oceanic slabs sinking beneath the Great Arc
at the leading edge of the plate. The direction of
Origin of Caribbean ophiolites the arc movement is therefore always oriented in the
direction of oceanic crust or the 'free face'. How-
Extensive ophiolites were obducted during col- ever, MUller et al. (Chapter 2) have noted that the
lision of the Great Arc with the passive margins Caribbean plate remains fixed in a mantle reference
of North and South America in northern Central frame since Middle Eocene time. For a station-
America, the Greater Antilles, and northern South ary Caribbean plate, North America-South America
America. Gealey (1980) proposed that these ophio- post-Eocene north-south convergence rather than the
lites represent the basement of the forearc basement presence of an oceanic free face may be the dom-
of the Great Arc. Other possible origins for the inant plate-driving force affecting the Caribbean.
ophiolites include proto-Caribbean oceanic crust in- GPS-based geodetic studies spanning the Caribbean
volved in the arc-continent collision and Caribbean plate could be used to test these differing dynamic
oceanic plateau crust (Kerr et al., 1997). The overly- scenarios.

C A R I B B E A N SEDIMENTARY BASINS: C L A S S I F I C A T I O N AND T E C T O N I C SETTING 29
Nature and driving forces of internal Caribbean Burkart, B. and Self, S., 1985. Extension and rotation of crustal
plate deformation blocks in northern Central America and effect on the volcanic
arc. Geology, 13: 22-26.
Burke, K., 1988. Tectonic evolution of the Caribbean. Annu. Rev.
Diebold and Driscoll (Chapter 19) and Driscoll Earth Planet. Sci., 16: 201-230.
and Diebold (Chapter 10) propose that internal de- Burke, K., Cooper, C., Dewey, J.E, Mann, E and Pindell,
formation of the Caribbean plate in the Colombian J.L., 1984. Caribbean tectonics and relative plate motions.
and Venezuelan basins and along the Beata Ridge In: W.E. Bonini, R.B. Hargraves and R. Shagam (Editors),
and Hess Escarpment is a response to divergent The Caribbean-South American Plate Boundary and Regional
Tectonics. Geol. Soc. Am. Mem., 162:31-63.
deformation associated with the formation of the
Case, J.E., MacDonald, W.D. and Fox, EJ., 1990. Caribbean
Cretaceous Caribbean oceanic plateau. In their view, crustal provinces: seismic and gravity data. In: G. Dengo
modern escarpments on the seafloor are largely relict and J.E. Case (Editors), The Caribbean Region. The Geology
features that lack significant neotectonic deforma- of North America, Vol. H, Geological Society of America,
tion. In contrast, Mauffret and Leroy (Chapter 21) Boulder, CO, pp. 15-36.
propose that the scarps in this region reflect internal Cobbold, ER., Davy, E, Gapais, D., Rossello, E.A., Sady-
bakasov, E., Thomas, J.C., Tondji Biyo, J.J. and de Urreiztieta,
disruption of the Caribbean plate along the line of
M., 1993. Sedimentary basins and crustal thickening. Sedi-
the Beata Ridge. Faulting reflects mainly shorten- ment. Geol., 86: 77-89.
ing in this intra-plate zone of deformation. Contin- Coffin, M.E, Gahagan, L.M., Lawver, L.A., Lee, T.-Y. and
ued geophysical studies, reexamination of existing Rosencrantz, E., 1992. Atlas of Mesozoic/Cenozoic recon-
data, and GPS-based geodetic studies spanning the structions (200 Ma to Present Day), PLATES Progress Report
No. 1-0192, University of Texas, Institute for Geophysics,
Caribbean plate are needed to distinguish these two
Technical Rep. 122, 49 pp.
ideas. Collins, L.S., Coates, A.G., Jackson, J.B.C. and Obando, J.A.,
1995. Timing and rates of emergence of the Lfmon and Bocas
del Toro basins: Caribbean effects of Cocos Ridge subduction?
ACKNOWLEDGEMENTS In: E Mann (Editor), Geologic and Tectonic Development of
the Caribbean Plate Boundary in Southern Central America.
Geol. Soc. Am. Spec. Pap., 295: 263-289.
I would like to thank Dietmar MUller for pro-
Corrigan, J.D., Mann, E, Ingle, J.C., Jr., 1990. Forearc response
viding the plate information to create the recon- to subduction of the Cocos Ridge, Panama-Costa Rica. Geol.
structions in Figs. 13-25 and Lisa Gahagan and the Soc. Am. Bull., 102: 628-652.
UTIG PLATES project for creating the reconstruc- DeMets, C., Gordon, R., Argus, D. and Stein, S., 1994. Effect
tions. UTIG contribution 1422. of recent revisions to the geomagnetic reversal time scale on
estimates of current plate motions. Geophys. Res. Lett., 21:
2191-2194.
Denny, W.M., III, Austin, J.A., Jr. and Buffler, R.T., 1994. Seis-
REFERENCES mic stratigraphy and geologic history of middle Cretaceous
through Cenozoic rocks, southern Straits of Florida. Am. As-
Av6 Lallemant, H.G., 1997. Transpression, displacement par- soc. Pet. Geol. Bull., 78: 461-487.
titioning, and exhumation in the eastern Caribbean/South Dewey, J.W. and Su(trez, G., 1991. Seismotectonics of Middle
American plate boundary zone. Tectonics, 16: 272-289. America. In: D.B. Slemmons, M.D. Zoback and D.D. Black-
Ball, M.M., Martin, R.G., Bock, W.D., Sylvester, R.E., Bowles, well (Editors), Neotectonics of North America. Decade Map
R.M., Taylor, D., Coward, E.L., Dodd, J.E. and Gilbert, L., Volume I, Geological Society of America, Boulder, CO, pp.
1985. Seismic structure and stratigraphy of the northern edge 309-321.
of the Bahamian-Cuban collision zone. Am. Assoc. Pet. Geol. Dillon, W., Austin, J.A., Jr., Scanlon, K.M., Edgar, N.T. and
Bull., 69: 1275-1294. Parson, L.M., 1992. Structure and development of the insular
Bird, D.E., Hall, S.A., Casey, J.E and Millegan, ES., 1993. margin north of western Hispaniola: A tectonic accretionary
Interpretation of magnetic anomalies over the Grenada basin. wedge on the Caribbean plate boundary. Marine and Petroleum
Tectonics, 12: 1267-1279. Geology, 9: 70-88.
Bouysse, E, 1988. Opening of the Grenada back-arc basin and Dixon, T.H., Farina, E, DeMets, C., Jansma, E, Mann, E and
evolution of the Caribbean plate during the Mesozoic and Calais, E., 1998. Relative motion between the Caribbean and
early Paleogene. Tectonophysics, 149: 121-143. North American plates and related plate boundary zone defor-
Bowland, C.L. and Rosencrantz, E., 1988. Upper crustal structure mation from a decade of GPS observations. J. Geophys. Res.
of the western Colombian basin, Caribbean Sea. Geol. Soc. 15,157-15,182.
Am. Bull., 100: 534-546. Dolan, J.F., Mann, E, de Zoeten, R., Heubeck, C. and Shiroma,
Bralower, T.J. and Iturralde-Vinent, M.A., 1997. Micropaleon- J., 1991. Sedimentologic, stratigraphic, and tectonic synthe-
tological dating of the collision between the North American sis of Eocene-Miocene sedimentary basins, Hispaniola and
plate and the Greater Antilles arc in western Cuba. Palaios, 12: Puerto Rico. In: E Mann, G. Draper and J.F. Lewis (Editors),
133-150. Geologic and Tectonic Development of the North America-
Bueno Salazar, R., 1989. Hydrocarbon exploration and potential Caribbean Plate Boundary in Hispaniola. Geol. Soc. Am.
of the Pacific coastal basin of Colombia. In: G.E. Ericksen, Spec. Pap., 262: 217-263.
M.T. Pinochet and J.A. Reinemund (Editors), Geology of Dolan, J.F., Mullins, H.T. and Wald, D.J., 1998. Active tectonics
the Andes and its Relation to Hydrocarbon and Mineral Re- of the north-central Caribbean: oblique collision, strain parti-
sources. Earth Science Series, 11, Circum-Pacific Council for tioning, and opposing subducted slabs. In: J.E Dolan and E
Energy and Mineral Resources, Houston, TX, pp. 335-343. Mann (Editors), Active Strike-Slip and Collisional Tectonics

30 E MANN
of the Northern Caribbean Plate. Geol. Soc. Am. Spec. Pap., terpretation of combined Seabeam, Gloria, and seismic data
326, 1-61. for Anegada Passage (Virgin Islands, North Caribbean). Mar.
Donnelly, T.W., Beets, D., Carr, M.J., Jackson, T., Klaver, G., Geophys. Res., 12: 173-196.
Lewis, J., Maury, R., Schellekens, H., Smith, A.L., Wadge, Kellogg, J.N. and Vega, V., 1995. Tectonic development of
G. and Westercamp, D., 1990. History and tectonic setting of Panama, Costa Rica, and the Colombian Andes: constraints
Caribbean magmatism. In: G. Dengo and J.E. Case (Editors), from Global Positioning System geodetic studies and gravity.
The Caribbean Region. The Geology of North America, Vol. In: E Mann (Editor), Geologic and Tectonic Development of
H, Geological Society of America, Boulder, CO, pp. 339-374. the Caribbean Plate Boundary in Southern Central America.
Draper, G., Guti6rrez, G. and Lewis, J.E, 1996. Thrust emplace- Geol. Soc. Am. Spec. Pap., 295: 75-90.
ment of the Hispaniola peridotite belt: orogenic expression of Kerr, A.C., Tarney, J., Marriner, G.E, Nivia, A. and Saunders,
the mid-Cretaceous Caribbean arc polarity reversal? Geology, A.D., 1997. The Caribbean-Colombian Cretaceous igneous
24: 1143-1146. plateau: the internal anatomy of an oceanic plateau. In: J.J.
Duncan, R.A. and Hargraves, R.B., 1984. Plate tectonic evolution Mahoney and M.E Coffin (Editors), Large Igneous Provinces:
of the Caribbean region in the mantle reference frame. In: Continental, Oceanic, and Planetary Flood Volcanism. Am.
W.E. Bonini, R.B. Hargraves and R. Shagam, R. (Editors), Geophys. Union, Geophys. Monogr., 100: 123-144.
The Caribbean-South American Plate Boundary and Regional Kolarsky, R.A., Mann, E and Montero, W., 1995a. Island arc re-
Tectonics. Geol. Soc. Am. Mem., 162:81-93. sponse to shallow subduction of the Cocos Ridge, Costa Rica.
Eva, A.N., Burke, K., Mann, E and Wadge, G., 1989. Four-phase In: E Mann (Editor), Geologic and Tectonic Development of
tectonostratigraphic development of the southern Caribbean. the Caribbean Plate Boundary in Southern Central America.
Mar. Pet. Geol., 6:9-21. Geol. Soc. Am. Spec. Pap., 295: 235-262.
Frisch, W., Meschede, W. and Sick, M., 1992. Origin of the Kolarsky, R.A., Mann, E and Monechi, S., 1995b. Stratigraphic
Central American ophiolites: evidence from paleomagnetic development of southwestern Panama as determined from in-
results. Geol. Soc. Am. Bull., 104:1301-1314. tegration of marine seismic data and onshore geology. In: E
Gealey, W.K., 1980. Ophiolite obduction mechanism. In: Ophi- Mann (Editor), Geologic and Tectonic Development of the
olites, Proc. Int. Ophiolite Symp., Cyprus Geological Survey Caribbean Plate Boundary in Southern Central America. Geol.
Department, Nicosia, pp. 228-243. Soc. Am. Spec. Pap., 295: 159-200.
Gonzalez de Juana, C., Arozena, J. and Picard-Cadillat, X., 1980. Ladd, J.W. and Buffler, R.T. (Editors), 1985. Middle America
Geologfa de Venezuela y de sus cuencas petrolfferas. Foninves trench off western Central America. Atlas 7, Ocean Margin
Ediciones, Caracas (2 volumes), 1051 pp. Drilling Program, Regional Atlas Series, Marine Science In-
Gordon, M.B., 1991. The Chortfs block is a continental, ternational, Woods Hole, MA.
pre-Mesozoic terrane. In: D.K. Larue and G. Draper, G. (Ed- Ladd, J.W., Holcombe, T.L., Westbrook, G.K., Edgar, N.T., 1990.
itors), Transactions of the 12th Caribbean Geological Con- Caribbean marine geology: active margins of the plate bound-
ference, St. Croix, U.S. Virgin Islands. Miami Geological ary. In: G. Dengo and J.E. Case (Editors), The Caribbean
Society, Miami, FL, pp. 505-512. Region. The Geology of North America, Vol. H, Geological
Gordon, M.B. and Muehlberger, W.R., 1994. Rotation of the Society of America, Boulder, CO, pp. 261-290.
Chortfs block causes dextral slip along the Guayape fault. Lara, M.E., 1993. Divergent wrench faulting in the Belize South-
Tectonics, 13: 858-872. ern Lagoon: Implications for Tertiary Caribbean plate move-
Gordon, M.B., Mann, E, Cficeres, D. and Flores, R., 1997. ments and Quaternary reef distribution. Am. Assoc. Pet. Geol.
Cenozoic tectonic history of the North American-Caribbean Bull., 77: 1041-1063.
plate boundary zone in Cuba. J. Geophys. Res., 102: 10,055- L6bron, M.C. and Perfit, M.R., 1994. Petrochemistry and tec-
10,082. tonic significance of Cretaceous island-arc rocks, Cordillera
Gose, W.A., 1985. Paleomagnetic results from Honduras and Oriental, Dominican Republic. Tectonophysics, 229: 69-100.
their bearing on Caribbean tectonics. Tectonics, 4: 565-585. Leroy, S., Mercier de Ldpinay, B., Mauffret, A. and Pubellier,
Grindlay, N.R., Mann, E and Dolan, J.E, 1997. Researchers M., 1996. Structural and tectonic evolution of the eastern
investigate submarine faults north of Puerto Rico. Eos, Trans. Cayman trough (Caribbean Sea) from seismic reflection data.
Am. Geophys. Union, 78: 404. Am. Assoc. Pet. Geol. Bull., 80: 222-247.
Hempton, M.R. and Barros, J.A., 1993. Mesozoic stratigraphy of Lugo, J. and Mann, E, 1995. Jurassic-Eocene tectonic evolution
Cuba: depositional architecture of a southeast-facing continen- of Maracaibo basin, Venezuela. In: A.J. Tankard, R. Sufirez
tal margin. In: J.L. Pindell and R.E Perkins, R.E (Editors), Soruco and H.J. Welsink (Editors), Petroleum Basins of South
Mesozoic and Early Cenozoic Development of the Gulf of America. Am. Assoc. Pet. Geol. Mem. 62: 699-725.
Mexico and Caribbean Region: A Context for Hydrocarbon MacDonald, W.D., 1980. Net tectonic rotation, apparent tectonic
Exploration. Gulf Coast Section, Society of Economic Paleon- rotation, and the structural tilt correction in paleomagnetic
tologists and Mineralogists, Houston, TX, pp. 193-209. studies. J. Geophys. Res., 85: 3659-3669.
Heubeck, C., Mann, E, Dolan, J. and Monechi, S., 1991. Di- Malfait, B.T. and Dinkelman, M.G., 1972. Circum-Caribbean
achronous uplift and recycling of sedimentary basins during tectonic and igneous activity and the evolution of the
Cenozoic tectonic transpression, northeastern Caribbean plate. Caribbean plate. Geol. Soc. Am. Bull., 83: 251-272.
Sediment. Geol., 70: 1-32. Mann, E, 1996. Model for the formation of large, transtensional
Holcombe, T.L., Ladd, J.W., Westbrook, G.K., Edgar, N.T. and basins in zones of tectonic escape. Geology, 25: 211-214.
Bowland, C.L., 1990. Caribbean marine geology: ridges and Mann, E and Burke, K., 1990. Transverse intra-arc rifting:
basins of the plate interior. In: G. Dengo and J.E. Case (Edi- Palaeogene Wagwater belt, Jamaica. Mar. Pet. Geol., 7: 410-
tors), The Caribbean Region. The Geology of North America, 427.
Vol. H, Geological Society of America, Boulder, CO, pp. 231- Mann, E and Kolarsky, R.A., 1995. East Panama deformed
260. belt: structure, age, and neotectonic significance. In: E Mann
Hutson, E, Mann, E and Renne, E, 1998. 4~ dating (Editor), Geologic and Tectonic Development of the Caribbean
of single muscovite grains in Jurassic siliciclastic rocks (San Plate Boundary in Southern Central America. Geol. Soc. Am.
Cayetano Formation): constraints on the paleoposition of west- Spec. Pap., 295:111-130.
ern Cuba. Geology, 26: 83-86. Mann, E, Schubert, C. and Burke, K., 1990. Review of
Jany, I., Scanlon, K.M and Mauffret, A., 1990. Geological in- Caribbean neotectonics. In: G. Dengo and J.E. Case (Editors),

C A R I B B E A N SEDIMENTARY BASINS: CLASSIFICATION AND TECTONIC SETTING 31
The Caribbean Region. The Geology of North America, Vol. Rosencrantz, E., Ross, M.I. and Sclater, J.G., 1988. Age and
H, Geological Society of America, Boulder, CO, pp. 291-306. spreading history of the Cayman Trough as determined from
Mann, E, Draper, G. and Lewis, J.E, 1991. An overview of the depth, heat flow, and magnetic anomalies. J. Geophys. Res.,
geologic and tectonic development of Hispaniola. In: E Mann, 93: 2141-2157.
G. Draper and J.E Lewis (Editors), Geologic and Tectonic Rosenfeld, J.H., 1990. Sedimentary rocks of the Santa Cruz
Development of the North America-Caribbean Plate Boundary o p h i o l i t e - a proto-Caribbean history. Trans. 12th Caribbean
in Hispaniola. Geol. Soc. Am. Spec. Pap., 262: 1-28. Geol. Conf., U.S. Virgin Islands, pp. 513-519.
Mann, E, Taylor, E W., Edwards, R.L. and Ku, T.L., 1995. Ross, M.I. and Scotese, C.R., 1988. A hierarchical tectonic
Actively evolving microplate formation by oblique collision model of the Gulf of Mexico and Caribbean region.
and sideways motion along strike-slip faults: an example from Tectonophysics, 155: 139-168.
the northeastern Caribbean plate margin. Tectonophysics, 246: Sandwell, D.T. and Smith, W.H.E, 1997. Marine gravity
1-69. anomaly from Geosat and ERS-1 satellite altimetry. J.
Manton, W.I., 1987. Tectonic interpretation of the morphology Geophys. Res., 102: 10,039-10,054.
of Honduras. Tectonics, 6:633-651. Schubert, C., 1984. Basin formation along the Bocon6-Mor6n-
Masson, D.G. and Scanlon, K.M., 1991. The neotectonic setting E1 Pilar fault system, Venezuela. J. Geophys. Res., 89:
of Puerto Rico. Geol. Soc. Am. Bull., 103: 144-154. 5711-5718.
McCann, W.R. and Pennington, W.D., 1990. Seismicity, large Sen, G., Hickey-Vargas, R., Waggoner, D.G. and Maurrasse, F.,
earthquakes, and the margin of the Caribbean plate. In: G. 1988. Geochemistry of basalts from the Dumisseau Formation,
Dengo and J.E. Case (Editors), The Caribbean Region. The southern Haiti: implications for the origin of the Caribbean
Geology of North America, Vol. H, Geological Society of Sea crust. Earth Planet. Sci. Lett., 87: 423-437.
America, Boulder, CO, pp. 291-306. Sinton, C.W., Duncan, R.A. and Denyer, E, 1997. Nicoya
Molnar, E and Sykes, L., 1969. Tectonics of the Caribbean Peninsula, Costa Rica: a single suite of Caribbean oceanic
and Middle America regions from focal mechanisms and plateau magmas. J. Geophys. Res., 102: 15,507-15,520.
seismicity. Geol. Soc. Am. Bull., 80:1639-1684. Speed, R.C. and Westbrook, G.K. (Editors), 1984. Lesser
Ori, G.G. and Friend, EE, 1984. Sedimentary basins formed Antilles arc and adjacent terranes. Atlas 10, Ocean Margin
and carried piggyback on active thrust sheets. Geology, 12: Drilling Program, Regional Atlas Series, Marine Science
475-478. International, Woods Hole, MA.
Pindell, J.L., 1995. Circum-Caribbean sedimentary basin devel- Speed, R.C., Torrini, R., Jr. and Smith, EL., 1989. Tectonic
opment and timing of hydrocarbon maturation as a function evolution of the Tobago trough forearc basin. J. Geophys.
of Caribbean plate tectonic evolution. In: R.L. Miller, G. Res., 94:2913-2936.
Escalante, J.A. Reinemund and M.J. Bergin (Editors), Energy St6phan, J.-F., Mercier de L6pinay, B., Calais, E., Beck, C.,
and Mineral Potential of the Central American-Caribbean Tardy, M., Carfantan, J.-C., Thery, J.-M., Olivet, J.-L.,
Region. Earth Science Series, 16, Circum-Pacific Council for Bouysse, E, Tournon, J., Vila, J.-M., Mauffret, A., Bourgois,
Energy and Mineral Resources. Springer, Berlin, pp. 47-56. J., Blanchet, R. and Dercourt, J., 1990. Rdconstitutions
Pindell, J.L. and Barrett, S.E, 1990. Geologic evolution of the gdodynamiques du domaine cara'fbe du Lias h l'Actuel. Bull.
Caribbean region. In: G. Dengo and J.E. Case (Editors), The Geol. Soc. Fr., 8: 915-919.
Caribbean Region. The Geology of North America, Vol. H, The Van der Hilst, R.D. and Mann, E, 1994. Tectonic implications
Geological Society of America, Boulder, CO, pp. 405-432. of tomographic images of subducted lithosphere beneath
PlaNer, G., 1976. Tectonic aspects of the Guatemala earthquake northwestern South America. Geology, 22:451-454.
of 4 February 1976. Science, 193: 1201-1208. Von Huene, R. and 16 others, 1995. Morphotectonics of the
Protti, M., Gtiendel, E and McNally, K., 1995. Correlation Pacific convergent margin of Costa Rica. In: E Mann (Editor),
between the age of the subducting Cocos plate and the Geologic and Tectonic Development of the Caribbean Plate
geometry of the Wadati-Benioff zone under Nicaragua and Boundary in Southern Central America. Geol. Soc. Am. Spec.
Costa Rica. In: E Mann (Editor), Geologic and Tectonic Pap., 295: 291-307.
Development of the Caribbean Plate Boundary in Southern Westbrook, G.K., Hardy, N.C. and Heath, R.E, 1995. Structure
Central America. Geol. Soc. Am. Spec. Pap., 295: 309-326. and tectonics of the Panama-Nazca plate boundary. In: E
Reed, D.L. and Silver, E.A., 1995. Sediment dispersal and Mann (Editor), Geologic and Tectonic Development of the
accretionary growth of the North Panama deformed belt. In: Caribbean Plate Boundary in Southern Central America. Geol.
E Mann (Editor), Geologic and Tectonic Development of the Soc. Am. Spec. Pap., 295: 91-109.
Caribbean Plate Boundary in Southern Central America. Geol. Williams, G.D., Powell, C.M. and Cooper, M.A., 1989. Geom-
Soc. Am. Spec. Pap., 295: 213-223. etry and kinematics of inversion tectonics. In: M.A. Cooper
Renne, E, Mattinson, J.M., Hatten, C.W., Somin, M., Onstott, and G.D. Williams (Editors), Inversion Tectonics. Geol. Soc.
T.C., Millan, G. and Linares, E., 1989. 4~ and U-Pb London Spec. Publ., 44: 3-15.
evidence for Late Proterozoic (Grenville-age) continental Wilson, J.T., 1966. Are the structures of the Caribbean and
crust in north-central Cuba and regional tectonic implications. Scotia arcs analogous to ice rafting? Earth Planet. Sci. Lett., 1:
Precambrian Res., 42: 325-341. 335-338.
Rosencrantz, E., 1990. Structure and tectonics of the Yucatan Wortel, R. and Cloetingh, S., 1981. On the origin of the
basin, Caribbean Sea, as determined from seismic reflection Cocos-Nazca spreading center. Geology, 9: 425-430.
studies. Tectonics, 9:1037-1059.

Chapter 2
New Constraints on the Late Cretaceous/Tertiary Plate Tectonic

Evolution of the Caribbean
R. D I E T M A R MULLER, JEAN-YVES ROYER, STEVEN C. C A N D E , W A L T E R R. R O E S T

a n d S. M A S C H E N K O V
We review the plate tectonic evolution of the Caribbean area based on a revised model for the opening of the central
North Atlantic and the South Atlantic, as well as based on an updated model of the motion of the Americas relative to the
Atlantic-Indian hotspot reference frame. We focus on post-83 Ma reconstructions, for which we have combined a set of new
magnetic anomaly data in the central North Atlantic between the Kane and Atlantis fracture zones with existing magnetic anomaly
data in the central North and South Atlantic oceans and fracture zone identifications from a dense gravity grid from satellite
altimetry to compute North America-South America plate motions and their uncertainties. Our results suggest that slow sinistral
transtension/strike-slip between the two Americas at rates roughly between 3 and 5 mm/year lasted until chron 25 (55.9 Ma).
Subsequently, our model results in northeast-southwest-oriented convergence until chron 18 (38.4 Ma) at rates ranging between
3.7 4- 1.3 and 6.5 + 1.5 mm/year from 65~ to 85~ respectively. This first convergent phase correlates with a Paleocene-Lower
Eocene calc-alkaline magmatic stage in the West Indies, which is thought to be related to northward subduction of Caribbean
crust during this time. Relatively slow convergence until chron 8 at rates from 1.2 + 0.9 to 3.6 + 2.1 mm/year from 65~
to 85~ respectively, is followed by a drastic increase in convergence velocity. After chron 8 (25.8 Ma), probably at the
Oligocene-Miocene boundary, this accelerated convergence resulted in 92 • 22 km convergence from chron 8 to 6, 127 + 25 km
from chron 6 to 5, and 72 + 17 km from chron 5 to the present measured at 85~ near the North Panama Deformed Belt at
convergence rates averaging 9.6 4- 3.1 and 9.6 4- 2.1 mm/year from chron 8 to 6 and chron 6 to 5, respectively, slowing down
to 5.2 • 1.3 mm/year after chron 5. Neogene convergence measured at the eastern Muertos Trough, at 17.5~ 65~ is 41 4- 18
km from chron 8 to 6, 58 • 25 km from chron 6 to 5, and 22 4- 17 km from chron 5 to present day, at rates between 4.4 + 1.7
and 1.6 4- 1.0 ram/year. These well-resolved differential plate motions clearly show an east-west gradient in plate convergence
in the Neogene, correlating well with geological observations. We suggest that the Early Miocene onset of underthrusting of the
Caribbean oceanic crust below the South American borderland in the Colombian and Venezuelan basins, the onset of subduction
in the Muertos Trough, and folding and thrust faulting at the Beata Ridge and the Bahamas, and the breakup of the main part
of the Caribbean plate into the Venezuelan and Colombian plates, separated by the Beata Ridge acting as a compressional plate
boundary (Mauffret and Leroy, Chapter 21) may all be related to the accelerated convergence between the two Americas.
The main differences with previous analyses are that (1) our model results in substantial variations in convergence rates
between the two Americas after chron 25 (55.9 Ma), (2) we have computed uncertainties for our North America-South America
plate flow lines, and (3) we show Tertiary Caribbean plate reconstructions in an Atlantic-Indian hotspot reference system. Our
absolute plate motion model suggests that the Caribbean plate has been nearly stationary since chron 18 (38.4 Ma). The east-west
gradient in convergence between the Americas in the Neogene has not resulted in substantial eastward motion of the Caribbean
plate, but rather contributed to causing its breakup into the Colombian and Venezuelan plates along the Beata Ridge where
east-west-oriented compressional stresses are taken up. Our model also suggests that the eastward escape of the Caribbean
plate in a mantle reference frame ceased when seafloor spreading started in the Cayman Trough, if the current interpretation of
magnetic anomalies in the Cayman Trough is not grossly in error. Our model suggests that the opening of the Cayman Trough
was accomplished by westward motion of the North American plate relative to a stationary Caribbean plate in a mantle reference
system. This implies that subsequent North America-Caribbean and South America-Caribbean tectonic processes were no longer
dominated by Cocos-Caribbean and Nazca-Caribbean plate interactions, as the latter had ceased to drive the Caribbean plate
eastwards. We conclude that the west-northwestward motion of South America relative to a trapped, stationary Caribbean plate
caused oblique collision along the passive margin of eastern Venezuela in the Neogene.
Caribbean Basins. Sedimentary Basins of the World, 4 edited by E Mann (Series Editor: K.J. Hsii), pp. 33-59.

34 R.D. MfJLLER et al.
INTRODUCTION has accumulated from these and many other regional

studies pertaining to Caribbean tectonic history. The
Many decades of research on deciphering the tec- purpose of this paper is not a comprehensive review
tonic and sedimentary history of the Caribbean area of Caribbean tectonic evolution. Hence the reader
(Fig. 1) have resulted in a fairly well understood will not find tables of syntheses of all tectonic events
tectonic framework of its evolution. The plate tec- that may have occurred during Caribbean tectonic
tonic history between the two Americas has been history, or all models that have been put forward to
reconstructed based on regional geophysical and explain them. This information has been thoroughly
geological data and its implications for Caribbean reviewed by Pindell and Barrett (1990). Here we
geology and have been evaluated in syntheses in- rather focus on extracting information from recently
cluding Pindell et al. (1988), Ross and Scotese declassified dense satellite altimetry data that al-
(1988), Pindell and Barrett (1990) and Stdphan et al. low us to map the structure of the ocean floor in
(1990). Our analysis builds on the knowledge that much more detail than previously possible and to
Fig. 1. Seafloor spreading isochrons in the Atlantic and eastern Pacific oceans from M~illeret al. (1997). Light gray shades correspond to
young ocean floor ages and dark grays to old ages. The bold frame outlines the Caribbean area shown in Fig. 2.

NEW CONSTRAINTS ON THE PLATE TECTONIC EVOLUTION OF THE CARIBBEAN 35
Fig. 2. Main tectonic elements of the Caribbean area. Isochrons shown in the Cayman Trough are from Rosencrantz et al. (1988).
The cross-hatched area centered on the Beata Ridge indicates the plate boundary between the Venezuelan and Colombian microplates.
Regional names which are not tectonic elements are shown in italics.
better constrain past plate motions along consuming America plate motions after chron 34 (83 Ma) it
or transform plate boundaries such as the bound- is particularly important to obtain estimates of the
aries between the North American, South American, uncertainties for the relative plate motion vectors
and Caribbean plates (Fig. 2). We utilize these data between the two Americas in order to evaluate the
jointly with new and existing magnetic anomaly resolution of plate motions models.
data to investigate the Late Cretaceous and Tertiary We have combined a set of new magnetic
plate kinematic framework of the Caribbean region, anomaly data in the central North Atlantic between
including the computation of uncertainties for our the Kane and Atlantis fracture zones, the 'Canary-
plate reconstructions. Bahamas Transect' (Maschenkov and Pogrebitsky,
Before the equatorial Atlantic between Africa 1992), with existing magnetic anomaly data in the
and South America started opening at about chron central North and South Atlantic oceans as well as
M-0 time (120 Ma) (Pindell and Dewey, 1982; with Seasat and Geosat satellite altimetry data to
Mascle et al., 1988), the Caribbean tectonic frame- create a self-consistent data set for the two ocean
work was largely dependent on North America- basins. The finite motion poles and their uncertain-
Africa plate motions, which were identical to North ties were estimated for 15 times from chron 34 to
America-South America motion vectors prior to the present using an inversion method developed
the opening of the South Atlantic. After initial by Chang (1987, 1988), Chang et al. (1990), and
opening of the equatorial Atlantic, plate boundaries Royer and Chang (1991), which allows a simple
between the two Americas were affected by North parameterization of the rotation uncertainties along
America-South America relative plate motions re- a plate circuit path. The resolution of the estimates
sulting from the difference vectors between North for North America-South America plate motions
America-Africa and South America-Africa seafloor differs through time and is largely dependent on the
spreading, as first computed by Ladd (1976). Re- velocity of their relative motions, i.e. faster plate
construction of the Caribbean tectonic frame for motions are better resolved than slower motions.
this period requires closure of the North America- Even though the first-order features of our model are
Africa-South America plate circuit by using mag- similar to Pindell et al.'s (1988) model, our results
netic anomaly and fracture zone date sets that are differ in detail, especially in the late Tertiary, and we
standardized with respect to time. Because of the stress the evaluation of uncertainties of plate motion
relatively slow velocity of North America-South vectors. In particular, our conclusions differ from

36 R.D. MULLER et al.
Pindell et al. (1988) in that we suggest that North A comprehensive analysis of magnetic anomaly
America-South America plate motions may have data in the South Atlantic was carried out by Cande
had substantial effects on the structural development et al. (1988). In order to create a self-consistent set
of the Caribbean area after chron 34 (83 Ma), espe- of magnetic anomaly identifications for closing the
cially during a period of rapid plate convergence in North America-Africa-South America circuit, we
the Neogene. use Cande et al.'s (1988) magnetic anomaly cross-
ings in the South Atlantic (Fig. 3) and identify the
same magnetic chrons as Cande et al. (1988) in
DATA the central North Atlantic. All magnetic anomaly
identifications correspond to the young end of nor-
Magnetic anomaly data mal-polarity intervals, except for anomaly 33o. The
phase shift angles were determined from paleomag-
The central North Atlantic is probably the ocean netic poles for North America from Harrison and
basin best covered by magnetic anomaly data. The Lindh (1982) and from the IGRF90 reference field.
three most comprehensive individual data sets are the We use the young end of the following normal-polar-
northeast-southwest-trending Kroonvlag data (Col- ity intervals according to the Cande and Kent (1995)
lette et al., 1984), the trans-Atlantic Geotraverse magnetic reversal time scale: chron 5 (9.74 Ma), 6
(TAG) data set, which comprises a number of (19.05 Ma), 8 (25.82 Ma), 13 (33.06 Ma), 18 (38.43
long east-west-oriented lines (Rona, 1980), and Ma), 21 (46.26 Ma), 24 (52.36 Ma), 25 (55.90 Ma),
the Canary-Bahamas Transect (Maschenkov and 30 (65.58 Ma), and 32 (71.59 Ma), 33o (79.08 Ma),
Pogrebitsky, 1992), which comprises a dense set of and 34 (83.00 Ma).
survey lines between the Atlantis and Kane fracture
zones from the Mid-Atlantic Ridge to about mag- Fracture zone data
netic anomaly 13 (Fig. 3). These data sets are sup-
plemented by a large number of other geophysical Fracture zones represent important information
surveys (see Klitgord and Schouten, 1986, for previ- constraining plate motions and can be used in con-
ous compilation), resulting in dense data coverage. cert with magnetic anomaly data for computing
Fig. 3. New dense magnetic anomaly data in the Canary-Bahamas Transect area north of the Kane Fracture Zone (Maschenkov and
Pogrebitsky, 1992), combined with data from other sources. Our magnetic anomaly identifications are shown as triangles (C5y, C18y,
C24y), squares (C6y), upside-down triangles (C8y, C21y), and circles (C13y).

finite rotations. Geosat, Seasat and ERS-1 altimetry gravity data (Sandwell and Smith, 1997). For the
data provide a unique data set to uniformly map central North Atlantic reconstructions we use the
the height of the sea surface, whose short-wave- Atlantis, Northern and Kane fracture zones (Fig. 4).
length topography reflects uncompensated basement When fracture zone offsets change from medium to
topography, such as that related to fracture zones. small, as has happened in the case of the Northern
Miiller et al. (1991) demonstrated that there is an Fracture Zone, whose offset diminished from 80 km
excellent correlation between the geoid anomaly and at chron 25 to 20 km at chron 13, they may become
the basement structure of the Kane Fracture Zone unstable and commence to migrate along the ridge,
in the central North Atlantic. They used geoid data producing V-shaped patterns. We used only those
from Geosat and subsatellite basement topography portions of fracture zones that appear to follow flow
profiles of the Kane Fracture Zone to show that the lines, i.e. have not migrated along the ridge axis. The
average horizontal mismatch between geoid low and location of the Kane Fracture Zone is constrained
the axis of the basement trough, as mapped by Tu- by Tucholke and Schouten's (1988) compilation of
cholke and Schouten (1988), is 5 km. The results of basement structure.
this comparative study represent 'ground truth' for Numerous fracture zones from the equatorial At-
the use of satellite altimetry data for accurately map- lantic to the southern South Atlantic record plate
ping slowly slipping Atlantic-type fracture zones. flow lines of seafloor spreading in the South At-
Following the Kane Fracture Zone study, Mtiller lantic. Shaw and Cande (1990) pointed out that
and Roest (1992) identified a number of small- and the northernmost fracture zones in this spreading
medium-offset fracture zones from the along-track system, i.e. the Marathon, Mercurius, Doldrums,
Geosat and Seasat gravity data by picking the center and Four-North fracture zones in the equatorial At-
of the gravity troughs corresponding to the deepest lantic (Fig. 2), put important constraints on South
portion of the central fracture valleys. We re-identi- America-Africa plate motions due to their proxim-
fied these fracture zones from dense satellite-derived ity to the finite rotation poles. However, because no
Fig. 4. Gravity anomalies from satellite altimetry from Sandwell and Smith (1997) and interpreted and rotated magnetic anomaly and
fracture zone identifications in the central North Atlantic. The unrotated magnetic and fracture zone identifications are identified by the
following symbols: triangle (C5, 9.74 Ma; C18, 38.43 Ma; C30, 65.58 Ma); square (C6, 19.05 Ma, C21; 46.26 Ma; C32, 71.59 Ma),
upside down triangle (C8, 19.05 Ma; C24, 52.36 Ma; C33o, 79.08 Ma), circle (C13, 33.06 Ma; C25, 55.90 Ma; C34, 83 Ma). All rotated
data points are marked by crosses. Paleoridge or transform segments as defined by magnetic anomaly or fracture zone identifications,
approximated as great circles in the inversion method used here, are denoted by alternating small and large symbols.

38 R.D. M U L L E R et al.
Fig. 5. Gravity anomalies from satellite altimetry from Sandwell and Smith (1997) and interpreted and rotated magnetic anomaly and
fracture zone identifications in the South Atlantic. The symbols used for plotting magnetic and fracture zone identifications follow the
same convention as in Fig. 4.
magnetic anomaly data have been identified in this of incorrect fracture zone segments for constraining
area, we cannot resolve which portion of a fracture a given reconstruction would skew our results. We
zone is relevant for a particular age, and whether also avoid large-offset fracture zones, as they are
these fracture zones reflect South America-Africa not reliable indicators of plate motion changes over
spreading for their entire length. We find that by short geological time spans.
using fracture zones south of 8~ only, we obtain
very similar rotations compared with reconstructions
in which equatorial fracture zones are included. RECONSTRUCTION METHOD
Accordingly, our South Atlantic reconstructions are
constrained by the Bodo Verde, Martin Vaz, Rfo The finite plate motion poles and their uncer-
Grande, and some unnamed fracture zones (Fig. 5). tainties were estimated using an inversion method
Both in the central North Atlantic and the South developed by Chang (1987, 1988), Chang et al.
Atlantic only those fracture zones were used whose (1990), and Royer and Chang (1991), based on the
offsets through time are constrained by magnetic criterion of fit by Hellinger (1981). The uncertainties
anomaly data. This is important, because the use of a rotation are expressed as a covariance matrix,

which is conceptually equivalent to the 'partial un- a detailed analysis of the dispersion of magnetic
certainty rotations' described by Stock and Molnar anomaly-5 crossings in the Indian Ocean by Royer
(1983). In this method magnetic anomaly and frac- et al. (1997) we assigned 1 - a nominal uncertainties
ture zone data are both regarded as points on two of 4 km to the magnetic anomaly crossings and of
conjugate isochrons, which consist of great circle 5 km to fracture zone crossings following Mfiller
segments. The best fit reconstruction is computed et al.'s (1991) analysis. The uncertainties assigned
by minimizing the sum of the misfits of conju- to the data (6-) are related to their true unknown
gate sets of magnetic anomaly and fracture zone estimates (or) by the quality factor:
data points with respect to individual great circle
segments. Consequently, both the resulting best-fit- 2
ting rotations, as well as the sum of the misfits, O"

depend critically on correctly identifying conjugate
data points that belong to a common isochron seg- Although ~c is unknown, the method developed by
ment. In practice, the application of Hellinger's Royer and Chang (1991) allows to estimate s from
criterion of fit poses no problem, because rotations the misfit, the geometry of the plate boundary and
have been published for most plates describing their the number of data:
Late Cretaceous/Tertiary history of motion. Hence,
a starting rotation can be used for an initial re- N-2s-3
construction to identify conjugate isochron segments
and data points.
A possible disadvantage of applying Hellinger's where N is the number of points, s the number of
(1981) criterion of fit to both magnetic and fracture great circle segments, and r the total weighted misfit.
zone data is that fracture zones cannot necessarily Note that N - 2s - 3 corresponds to the number of
be expected to fit as well as magnetic anomaly data. degrees of freedom.
Although all portions of a fracture zone have at some Thus the parameter ~ indicates whether the as-
time been the location of a transform fault and part signed uncertainties are correct (s ~ 1), underes-
of an isochron, fracture zone morphology becomes timated (~ << 1) or overestimated (~ >> 1). Our
overprinted successively at the transform-ridge in- reconstructions indicate that our nominal uncertain-
tersections during changes in spreading direction. ties are generally slightly overestimated. For the
The amplitude of this effect is expected to increase central Atlantic, ~ ranges from 1.07 to 3.08 and all
as a function of transform length. For this reason we degrees of freedom are larger than 50. This means
do not use fracture zones with offsets of more than that the average true uncertainties range from 2.3
about 150 km. Shaw and Cande (1990) recognized (= 4/3,,/-3-,~.08) to 3.9 (= 4/~/-f.07) km for the mag-
this problem and suggested an inversion method that netic crossings, and 2.8 to 4.8 km for the fracture
incorporates fracture zones by minimizing the misfit zone crossings (Table 1). For the South Atlantic,
of fracture zone data with respect to plate flow lines. ranges from 0.67 to 1.74 with degrees of freedom
The benefits of this model were expected to be a between 28 and 42. Thus the uncertainties should
utilization of the 'integral constraints' of fracture lie between 3.0 and 4.9 km for the magnetic data,
zones, i.e. their continuity, as well as the possibility and 3.8 to 6.1 for the fracture zone data (Table 2).
of allowing for consistent asymmetries of fracture Several factors may contribute to this discrepancy in
zone limbs on conjugate plate flanks. Shaw and the dispersion of the magnetic crossings: (i) there
Cande (1990) implemented this method by minimiz- are much less data in the South Atlantic, hence we
ing the misfits of fracture zone data to symmetric are probably combining data from different spread-
flow lines. However, forcing fracture zones to fit ing corridors into individual segments; (ii) there are
symmetric flow lines may obscure distinct changes much less recent (i.e. GPS-navigated) cruises in the
in spreading direction recorded in fracture zones South Atlantic than in the Central Atlantic. With-
bounded by two asymmetrically spreading corridors. out further investigating this question, we decided
Even though seafloor spreading appears to be re- to keep an uncertainty of 4 km for the magnetic
markably symmetric in the long run, accretion of crossings. Our inversions also suggest that the dis-
ocean floor through time periods of short 'stages' persion of the fracture zone data is generally better
resolvable by magnetic anomaly data can be quite than 5 km. However, since fracture zones are not the
asymmetric. optimal records for plate reconstructions, and since
A critical aspect in our reconstruction method is we do not know how well gravity troughs relate to
the correct assessment of the uncertainties in the the actual location of (paleo-)transform segments,
location of the data that will propagate into the we choose to remain conservative and use the Mtiller
uncertainties on the rotation parameters (location et al. (1991) result of an uncertainty of 5 km for
of the rotation pole and rotation angle). Following fracture zones.

40 R.D. M U L L E R et al.
Table 1
North America-Africa finite rotations
Chron Age Latitude Longitude Angle r t~ df N s 6 mag 6 fz

(Ma) (+~ (+~ (o) (km) (km -1) (km) (km)
5 9.7 80.98 22.82 2.478 76.3 2.46 188 271 40 2.5 3.2
6 19.0 80.89 23.28 5.244 62.3 1.96 122 191 33 2.9 3.6
8 25.8 79.34 28.56 7.042 39.9 2.53 101 140 18 2.5 3.1
13 33.1 75.99 5.98 9.767 79.2 1.19 94 159 31 3.7 4.6
18 38.4 74.54 0.19 11.918 36.9 1.65 61 94 15 3.1 3.9
21 46.3 74.23 -5.01 15.106 42.9 1.19 51 96 21 3.7 4.6
24 52.4 77.34 -1.61 16.963 17.8 3.08 55 94 18 2.3 2.8
25 55.9 80.64 6.57 17.895 51.7 1.26 65 110 21 3.6 4.5
30 65.6 82.74 2.93 20.84 80.6 1.07 86 135 23 3.9 4.8
32 71.6 81.35 -8.32 22.753 66.9 1.33 89 142 25 3.5 4.3
33o 79.1 78.64 -18.16 26.981 55.2 1.85 102 155 25 2.9 3.7
34 83.0 76.81 -20.59 29.506 52.1 1.82 95 144 23 3 3.7
Parameters are: r -- sum of misfits; N -- number of data points" s = number of great circle segments; df = degrees of freedom; t~ = df/r
(see text for discussion). The parameters ~ and the misfit r in this table are calculated with nominal uncertainties of 4 km for the magnetic
anomaly crossings and 5 km for the fracture zone crossings. The 'true' data uncertainties O'mag and 6fz are related to their unknown
estimates (o-) (i.e. 4 and 5 km for magnetic and fracture zone data, respectively) by the quality factor x -- (6-/o-) 2.
Table 2
South America-Africa finite rotations
Chron Age Latitude Longitude Angle r ~ df N s 6-mag 6 fz

(Ma) (+~ (+~ (o) (km) (km -1) (km) (km)
5 9.7 62.05 -40.59 3.18 39.3 0.71 28 57 13 4.7 5.9

6 19.0 58.77 -37.32 7.049 26.3 1.10 29 58 13 3.8 4.8
8 25.8 57.59 -36.27 9.962 24.6 0.85 21 52 14 4.3 5.4
13 33.1 56.17 -33.64 13.41 52.4 0.67 35 64 13 4.9 6.1
18 38.4 57.10 -33.00 15.912 18.1 1.44 26 57 14 3.3 4.2
21 46.3 56.95 -31.15 19.107 47.1 0.89 42 79 17 4.2 5.3
24 52.4 58.89 -31.18 21.38 28.7 0.87 25 52 12 4.3 5.4
25 55.9 61.35 -32.21 22.273 27.7 1.01 28 53 11 4.0 5.0
30 65.6 63.88 -33.61 24.755 24.2 1.74 42 73 14 3.0 3.8
32 71.6 63.41 -33.38 26.573 24.1 1.16 28 53 11 3.7 4.6
33o 79.1 62.92 -34.36 30.992 20.7 1.74 36 65 13 3.0 3.8
34 83.0 61.88 -34.26 33.512 41.2 0.97 40 75 16 4.1 5.1
For explanation of parameters, see Table 1.
RESULTS Table 3
North America-Africa covariance matrices
North A m e r i c a - A f r i c a and South
A m e r i c a - A f r i c a finite rotations Chron a b c d e f
5 4.721 -4.047 2.792 4.354 -2.897 2.064
F i n i t e r o t a t i o n s for N o r t h A m e r i c a - A f r i c a and 6 6.134 -4.971 3.561 5.202 -3.586 2.679
South America-Africa plate motions were computed 8 17.122 -17.129 12.390 19.550 -14.054 10.432
13 8.318 -8.466 5.972 10.323 -7.144 5.207
for 12 t i m e s f r o m c h r o n 34 to the p r e s e n t . T h e r e s u l t -
18 19.815 -20.774 14.248 24.125 -16.397 11.551
i n g finite r o t a t i o n s , s t a t i s t i c a l p a r a m e t e r s , a n d c o v a r i -
21 18.001 -20.328 13.636 26.773 -17.973 12.551
a n c e m a t r i c e s are l i s t e d in T a b l e s 1 - 4 . T h e 9 5 % c o n - 24 26.464 -31.305 20.687 40.473 -26.665 17.992
f i d e n c e r e g i o n s are 3 - d i m e n s i o n a l e l l i p s o i d s in lati- 25 11.855 -13.570 8.641 18.742 -11.775 7.871
tude, l o n g i t u d e , a n d r o t a t i o n a n g l e s p a c e . In o r d e r to 30 9.118 -10.652 6.636 15.770 -9.831 6.448
32 12.322 -16.202 9.702 24.153 -14.573 9.103
r e p r e s e n t t h e m o n a m a p , the e l l i p s o i d s are p r o j e c t e d
33r 4.085 -4.022 2.339 5.730 -3.475 2.382
o n t o the l a t i t u d e - l o n g i t u d e s p h e r e (cf. R o y e r a n d
34 6.197 -4.831 2.468 6.957 -4.290 3.069
C h a n g , 1991). T h e p r o j e c t e d u n c e r t a i n t y e l l i p s e s in-
c l u d e u n c e r t a i n t i e s in b o t h l a t i t u d e , l o n g i t u d e , a n g l e , Covariance matrices can be reconstructed in the following way:
b u t o n e m u s t k e e p in m i n d that the true size o f the
r o t a t i o n u n c e r t a i n t i e s (i.e. e l l i p s o i d s ) , w h i c h are de-
s c r i b e d b y the c o v a r i a n c e m a t r i c e s , m i g h t n o t b e re-

Table 4 Cande's (1990) model and our model for the South
South America-Africa covariance matrices Atlantic (Fig. 5) is the anomaly-13 reconstruction,
which results in a distinct cusp in our pole path,
Chron a b c d e f while there is a less accentuated cusp in their path.
6 27.203 -7.073 -20.420 3.179 5.487 16.685 One way to qualitatively test whether or not such a
5 20.913 -4.353 -13.079 1.974 2.804 9.422 cusp is supported by the data is to construct plate
8 63.296 -20.659 -35.876 8.134 11.603 23.341
flow lines and evaluate their fit to fracture zones.
13 9.525 -2.385 -5.374 2.117 1.676 3.765
18 31.698 -12.098 -18.390 5.766 7.124 12.671
For South Atlantic spreading this is done best in
21 11.165 -3.955 -6.408 2.546 2.525 4.334 the equatorial Atlantic, because the fracture zones
24 25.100 -11.021 -11.714 6.070 5.352 6.722 here are the best recorders of changes in spread-
25 25.115 -11.568 -13.323 7.341 6.681 8.144 ing direction due to their proximity to the finite
30 15.017 -6.988 -8.168 5.207 4.322 5.358
motion poles. However, plotting plate flow lines in
32 30.163 -13.978 -16.671 9.696 8.326 10.185
33r 9.249 -2.900 -4.696 3.493 2.513 3.347
this area raises the problem of knowing where the
34 8.711 -4.332 -4.971 4.014 3.230 3.649 South America-North American plate was located
through time. Plotting flow lines both for North
For reconstruction of covariance matrices see Table 3.
America-Africa and South America-Africa for the
same fracture zone allows us to address this prob-
flected well by their 2-D projection onto the sphere. lem. We plot South Atlantic plate flow lines derived
For instance, in the case of the South Atlantic, the from our model in Fig. 7a with the gridded gravity
2-D uncertainty in the location of the chron-5 best- anomaly field computed from Seasat, Geosat, and
fitting pole appears to be at least 10 times larger than ERS-1 data (Sandwell and Smith, 1997). All flow
the 2-D uncertainty in the location of the chron-21 lines are constructed using both ridge-transform in-
rotation pole, whereas the volume of the 95% con- tersections as seed points. The area encompassed
fidence region for chron-5 rotation is only about 3 by the resulting dual flow lines approximates the
times larger than for the chron-21 rotation (7838 amount of transpression or transtension that oc-
versus 2773 km 3, respectively). Conversely, the 95% curred during changes in spreading direction, assum-
uncertainty volume for chron 8 (19,705 km 3) is 2.5 ing symmetric plate accretion. The Marathon and
larger than for chron 5, whereas its 2-D projection is Mercurius fracture zones do not fit the computed
about 3 times smaller than for chron 5. flow lines well. However, the Four-North and Dol-
In Fig. 6 we compare our results with the drums fracture zones show a good fit to our flow
North America-Africa pole path from Klitgord and lines. The latter observation confirms the validity
Schouten (1986) and the South America-Africa pole of our South Atlantic plate motion model, whereas
path from Shaw and Cande (1990). The main differ- the former indicates that the North-South American
ence for the central North Atlantic is that our path plate boundary may be located in the vicinity of the
shows a sharp bend at anomaly-8 time (25.8 Ma), Marathon and Mercurius fracture zones, resulting in
resulting in a pole path for chron 5 to chron 8 that deviations from South America-Africa flow lines.
is distinctly different from the chron-21 to chron-8 In comparison, we show central North Atlantic plate
path, whereas Klitgord and Schouten's (1986) path flow lines in the equatorial Atlantic Ocean in Fig. 7b.
is continuous between chrons 21 and 5. The con- The Marathon Fracture Zone flow line clearly does
sequences of this difference for Neogene North not match the gravity anomaly expression of this
America-South American plate motions are dis- fracture zone. The post-chron 6 flow line of the Mer-
cussed in subsequent sections. curius Fracture Zone fits its eastern limb better than
Our South Atlantic pole path is generally similar the South America-Africa flow line (compare with
to Shaw and Cande's (1990) path, but less smooth. Fig. 7a), but not its western limb. Its is clear that
The similarity reflects the fact that both models are this fracture zone would not be useful to constrain
based on very similar data sets, whereas the dif- either North America-Africa nor South America-
ferences reflect properties of the different inversion Africa plate motions, since it appears to have been
methods used to compute finite rotations as well affected by plate boundary processes. We can draw
as new fracture zone identifications. The 'integral the conclusion that applying Hellinger's criterion of
constraints' of fracture zone continuity used in Shaw fit jointly to the magnetic anomaly data and fracture
and Cande's (1990) method is probably the main zone data from continuous fracture zones, which
reason for the smoothness of their pole path. How- follow plate flow lines, and reconstructing these data
ever, the use of symmetric flow lines to evaluate independently for each isochron, results in plate
the fit of fracture zone data might smooth out real models that produce smooth continuous flow lines,
cusps in a finite pole path that is computed from even though this property is not utilized or imposed
reconstructing data from individual isochrons inde- by the model inversion, as it is in Shaw and Cande's
pendently. The main difference between Shaw and (1990) method.

R.D. MULLER et al.
Fig. 6. Finite rotation poles and 95% confidence ellipses for North America-Africa (upper right) and South America-Africa (lower left)
plate motions for 12 reconstruction times from chron 34 (83 Ma) to the present. As a comparison we show the North America-Africa
pole path from Klitgord and Schouten (1986) and the South America-Africa pole path from Shaw and Cande (1990) (gray triangles).
Note the sharp cusp in our North America-Africa pole path at chron 8 (25.8 Ma), resulting in a pole path for chron 5 to chron 8 that
is distinctly different from Klitgord and Schouten's (1986) path. The differences between ours and both Klitgord and Schouten's (1986)
and Shaw and Cande's (1990) models reflect a much improved accuracy in locating fracture zones based on a dense gravity anomaly grid
from satellite altimetry (Sandwell and Smith, 1997), and some new magnetic anomaly data.
North America-South America finite rotations 18 to 8; (ii) an important northward migration of

the rotation poles from chron 34 to chron 18; (iii)
Tables 5 - 8 list the rotation parameters for the followed by a southward migration until chron 6.
North A m e r i c a - S o u t h America relative motions, For this reason we computed only 8 stage rotations
for each chron resulting from the product of the (Table 6) to describe the main episodes of relative
South A m e r i c a - A f r i c a rotation with the correspond- motion between the two American plates. The North
ing Africa-North America rotation (Table 9). The A m e r i c a - S o u t h America rotation stage poles always
resulting pole path (Fig. 8) shows: (i) very stable lie outside the Caribbean plate (Table 6); for the ages
poles of motion from chron 25 to 21, and from chron younger than chron 25 (55.9 Ma), they lie within or

Fig. 7. (a) Gravity anomalies and South Atlantic symmetric plate flow lines in the Equatorial Atlantic Ocean. Flow lines are composed of
segments for stages bounded by our reconstruction times (see Fig. 4). All flow lines have been constructed using both ridge-transform
intersections as seed points. The area encompassed by the resulting dual flow lines approximates the amount of transpression or
transtension that occurred during changes in spreading direction, assuming symmetric plate accretion. The Marathon and Mercurius
fracture zones do not fit the computed flow lines well. However, the Four-North and Doldrums fracture zones show a good fit to our
flow lines. The latter observation confirms the validity of our South Atlantic plate motion model, whereas the former indicates that
the North-South American plate boundary may be located in the vicinity of the Marathon and Mercurius fracture zones, resulting in
deviations from South America-Africa flow lines. (b) Gravity anomalies and central North Atlantic symmetric plate flow lines in the
Equatorial Atlantic Ocean. The Marathon Fracture Zone flow line clearly does not match the gravity anomaly expression of this fracture
zone. The post-chron-6 flow line of the Mercurius Fracture Zone fits its eastern limb better than the South America-Africa flow line
(compare with Fig. 7a), but not its western limb. It is is clear that this fracture zone would not be useful to constrain either North
America-Africa nor South America-Africa plate motions, since it appears to have been influenced by plate boundary processes.
in the vicinity of the North A m e r i c a - S o u t h A m e r i c a C a y m a n Trough (Rosencrantz et al., 1988; Mauffret

plate boundary, implying a different sense of motion and Leroy, Chapter 21). The magnetic anomalies
along strike of this plate boundary. on older portions of the C a y m a n Trough are less
straightforward to identify, and the chronology of
North America-South America plate motions in both pre- and post-chron 6 seafloor spreading here
the Caribbean area is still the subject of debate. For these reasons
we first analyze the North A m e r i c a - S o u t h America
Ideally we would like to be able to compute North plate kinematic framework, as it is independent on
A m e r i c a - C a r i b b e a n and South A m e r i c a - C a r i b b e a n knowledge of the spreading history in the C a y m a n
plate motions and compare the m o d e l e d motion Trough, and compare the results with geological and
vectors with m a p p e d plate boundary deformation. geophysical data from the northern and southern
Accurate estimates for North A m e r i c a - C a r i b b e a n Caribbean plate boundaries.
plate motions are available for times after chron 6 Fig. 9 shows North A m e r i c a - S o u t h A m e r i c a n
(19.0 Ma), as recorded by seafloor spreading in the plate motion through time, illustrated by the suc-

Table 5 95% confidence region is plotted about the young

North America-South America finite rotations ends of the relative motion vectors, i.e. the ellipse
about points at chron 33o reflect the uncertainty for
Chron Age Latitude Longitude Angle df the chron 3 4 - 3 3 o stage rotation. A simultaneous
(Ma) (+~ (+~ (o)
confidence region represents the area (on the surface
5 9.7 15.06 -56.59 1.408 216 of the Earth) in which a point on a given plate
6 19.0 14.76 -52.43 3.433 151 may have been located with equal likelihood for a
8 25.8 17.52 -53.52 5.146 122
13 33.1 16.99 -53.13 6.067 129 particular reconstruction time, with 95% confidence.
18 38.4 18.12 -54.60 6.498 87 The simultaneous 95% confidence regions increase
21 46.3 13.61 -53.09 7.11 93 in size with increasing distance from the stage pole
24 52.4 13.95 -52.33 8.188 80 of motion; in the Caribbean area they increase in
25 55.9 13.87 -53.04 8.777 93 size correspondingly from east to west (Fig. 9).
30 65.6 11.25 -53.89 9.009 128
32 71.6 9.58 -53.74 8.947 117 The reader may note that even though the pro-
33o 73.6 9.03 -56.57 9.12 138 jected confidence ellipsoids of our finite rotation poles
34 83.0 7.31 -58.00 9.332 135 for North A m e r i c a - S o u t h American plate motion for
chrons 18-6 show large overlaps (Fig. 8), the simul-
taneous 95% confidence regions for the three rotated
Table 6 points shown show only little overlap. This may ap-
North America-South America stage rotations (North America pear puzzling, but only reflects the imperfect nature of
reference frame) the 3-dimensional finite rotation pole error ellipsoid
projections onto a spherical surface on Fig. 8, as dis-
Stage (Chron a-b) Age Latitude Longitude Angle
(Ma) (+~ (+~ (o) cussed before. In other words, this result shows that
the covariance matrices and uncertainty ellipsoids for
0- 5 9.7 15.06 -56.59 1.408 these reconstructions are different enough to distin-
5- 6 9.3 14.42 -49.56 2.03
6- 8 6.8 23.08 -55.54 1.727 guish different phases of convergence in the late Ter-
8-18 12.6 20.57 -58.63 1.358 tiary outside of the 95% error bounds.
18-25 17.5 1.76 -49.85 2.355 We cannot resolve very slow relative motions
25-30 9.7 -43.69 -80.57 0.484 between chrons 25 and 24; some overlap between
30-33o 8.0 -31.75 -124.85 0.557 the confidence regions for chrons 32 and 30 and
33o-34 9.4 -32.74 -102.67 0.417
chrons 18 and 13 is also observed. However, all
other stage vectors are well resolved. Consequently,
we compute a new set of 8 stage rotation poles
cessive motion of three points attached to the North (Table 6) and corresponding relative motion vec-
American plate with respect to the South American tors, including only stages for which we can resolve
plate for eleven stages from chron 34 (83 Ma) to the relative North A m e r i c a - S o u t h America plate mo-
present. The paths of these points through time as tions (Fig. 10). The relative motion vectors in this
shown on Fig. 9 correspond to conventional plate set of stages can be divided into four age groups:
flow lines illustrating relative North A m e r i c a - S o u t h (1) slow sinistral strike-slip/transtension between
America plate motion. The only difference here is chrons 34 and 25 (~2.8 • 0.8-4.8 • 1.1 m m / y e a r
that for each stage rotation vector a simultaneous at 85~ (2) northeast-southwest-oriented conver-
Table 7
Covariance matrices for North America-South America finite rotations
Chron a b c d e f
5 25.693 -8.424 -10.306 6.305 -0.060 11.451
6 33.509 -12.099 -16.901 8.283 1.988 19.289
8 81.193 -38.062 -23.479 27.207 -2.057 33.476
13 18.284 -11.008 0.625 12.113 -5.265 8.859
18 52.534 -33.276 -4.194 29.357 -8.715 23.734
21 30.632 -24.772 7.601 27.959 -14.851 16.779
24 54.006 -43.018 9.682 44.281 -20.378 24.534
25 38.048 -25.517 -4.414 25.021 -4.705 16.000
30 25.161 -18.055 -1.233 19.886 -5.158 11.870
32 44.118 -30.783 -6.327 32.094 -5.741 19.408
330 13.746 -7.130 -2.336 8.844 -0.753 5.696
34 15.362 -9.517 -2.616 10.472 -0.732 6.765
See parameter legend in Table 3.

N E W C O N S T R A I N T S ON THE PLATE T E C T O N I C E V O L U T I O N OF THE C A R I B B E A N 45
Fig. 8. Finite rotation poles for 12 reconstruction times for North America-South America relative plate motions since chron 34 (83 Ma).
These poles have been computed by combining the finite rotations and their 95% confidence regions shown in Fig. 6. Note the large
uncertainty for the chron-5 (9.7 Ma) reconstruction. This partly reflects that the South Atlantic reconstruction for this time is not well
constrained. It also reflects that the uncertainty of finite rotations increases with decreasing finite rotation angle.
Table 8
Volumes of uncertainty ellipsoids for North-South America finite rotation poles
Chron Latitude Longitude Angle df Vol. 95% Vol. 1 - o-

5 15.06 -56.59 1.408 216 14,548 650
6 14.76 -52.43 3.433 151 22,448 993
8 17.52 -53.52 5.146 122 62,456 2741
13 16.99 -53.13 6.067 129 16,470 724
18 18.12 -54.6 6.498 87 46,085 1988
21 13.61 -53.09 7.11 93 28,396 1230
24 13.95 -52.33 8.188 80 48,212 2069
25 13.87 -53.04 8.777 93 38,850 1682
30 11.25 -53.89 9.009 128 26,246 1154
33o 9.03 -56.57 9.12 138 13,898 613
34 7.31 -58 9.332 135 14,696 647

46 R.D. MOLLER et al.
Fig. 9. North America-South America plate motion through time, illustrated by the successive motion of three points attached to the
North American plate with respect to the South American plate for eleven stages from chron 34 (83 Ma) to the present. The paths of these
points through time as shown on Fig. 9 correspond to conventional plate flow lines illustrating relative North America-South America
plate motion. The only difference here is that for each stage rotation vector a simultaneous 95% confidence region is plotted about the
young ends of the relative motion vectors, i.e. the ellipse about points at chron 33o reflect the uncertainty for the chron 34-33o stage ro-
tation. A simultaneous confidence region represents the area (on the surface of the Earth) in which a point on a given plate may have been
located with equal likelihood for a particular reconstruction time, with 95% confidence. This figure demonstrates that we cannot resolve
very slow relative motions between chrons 25 and 24; some overlap between the confidence regions for chrons 32 and 30 and chrons
18 and 13 is also observed. However, all other stage vectors are well resolved. Consequently, we compute a new set of relative motion
vectors, including only stages for which we can resolve relative North America-South America plate motions well, shown in Fig. 10.
gence from chron 25 to 18 (6.5 4- 1.5 m m / y e a r at Pindell et al.'s (1988) (Fig. 11). The general shape of
85~ (3) slow compressional motion from chron both models is similar, reflecting the similarity of the
18 to 8 (3.6 4-2.1 m m / y e a r at 85~ and (4) fast magnetic anomaly data sets used to constrain both
n o r t h - s o u t h - o r i e n t e d convergence from chron 8 to 6 models. The main difference between the models is
(9.6 + 3.1 m m / y e a r for chrons 8 - 6 and 9.6 4- 2.1 that Pindell et al.'s (1988) model implies relatively
m m / y e a r for chrons 6 - 5 at 85~ followed by constant convergence between the Americas of rates
a deceleration in n o r t h - s o u t h - o r i e n t e d convergence b e t w e e n 6 and 4 m m / y e a r since chron 21 (46.3 Ma).
post-chron 5 (5.2 4- 1.3 m m / y e a r at 85~ Our model results in substantial variations in conver-
In order to directly compare our model with Pin- gence rates since chron 25 (55.9 Ma), as d o c u m e n t e d
dell et al.'s (1988) results, we compute N o r t h - S o u t h in Fig. 10 and Table 10. In particular our model re-
A m e r i c a relative motions for the same 7 stages as solves the onset of fast convergence after chron 8 at
a speed of nearly 10 m m / y e a r , c o m p a r e d with less
than 4 m m / y e a r from chrons 18 to 8 measured at
Table 9 85~ Without identifying magnetic anomaly 8, two
Volumes of uncertainty ellipsoids for North America-Africa stages of slow (chron 13-8) and fast (chron 8 - 6 )
stage rotation poles convergence would be averaged.
Stage Latitude Longitude Angle Vol. 95% Vol. 1 - cr

5- 6 14.42 -49.56 2.03 3 6 , 3 1 7 1662 DISCUSSION
6- 8 23.08 -55.54 1.727 116,330 5324
8-18 20.57 -58.63 1.358 149,039 6821
Lesser Antilles to Mid-Atlantic Ridge
18-25 1.76 -49.85 2 . 3 5 5 117,073 5358
25-30 -43.69 -80.57 0 . 4 8 4 87,663 4012
30-33o -31.75 -124.85 0 . 5 5 7 5 6 , 3 6 8 2580 The North A m e r i c a - S o u t h A m e r i c a plate bound-
33o-34 -32.74 -102.67 0 . 4 1 7 3 9 , 2 9 4 1798 ary east of the Caribbean region is characterized by

Table 10
Uncertainties on motion vectors (3-D 95% confidence limits)
Chron Time span Rate of motion Motion vectors Azimuth rmaj rmin Om~
(Ma) (mm/year) (km) (o from N) (km) (km) (o from N)
Longitude: 48~
0- 5 9.7 1.6 • 0.8 22 • 10 -2 27 6 53
5- 6 9.3 0.6 4- 0.5 7 • 7 -34 42 10 56
6- 8 6.8 3.6 • 4.0 34 • 38 48 55 12 57
8-18 12.6 1.6 4- 1.9 29 + 33 30 54 12 58
18-25 17.5 2.4 • 0.9 60 • 23 -82 41 12 60
25-30 9.7 3.6 • 1.0 49 4- 14 -64 34 11 62
30-33o 8.0 5.4 4- 0.7 61 • 8 -33 27 11 60
33o-34 9.4 3.3 • 0.6 43 4- 8 -42 24 12 60
Longitude: 52~
0- 5 9.7 0.7 -t- 0.7 9 • 10 -9 27 7 49
5- 6 9.3 1.0 • 2.0 13 4- 25 -159 43 l0 53
6- 8 6.8 2.7 • 3.9 25 • 37 72 56 13 53
8-18 12.6 1.0 • 2.1 18 • 38 44 55 15 54
18-25 17.5 2.6 • 1.1 64 • 28 -102 42 13 56
25-30 9.7 3.5 • 1.1 48 • 14 -67 35 12 58
30-33o 8.0 5.4 • 0.8 61 • 8 -35 28 12 57
33o-34 9.4 3.2 • 0.6 42 • 8 -45 24 12 56
Longitude: 58~
0- 5 9.7 0.3 • 1.3 4 -4- 18 -154 28 7 46
5- 6 9.3 2.4 -t- 1.8 32 + 23 - 170 45 11 49
6- 8 6.8 2.6 + 2.2 24 -4- 21 108 58 15 49
8-18 12.6 0.7 + 1.8 11 + 33 84 57 17 50
18-25 17.5 3.0 -t- 1.2 74 • 30 -119 44 16 51
25-30 9.7 3.5 + 1.1 48 i 14 -71 36 13 54
30-33o 8.0 5.3 • 0.8 61 -t- 9 -37 28 13 52
33o-34 9.4 3.1 • 0.6 42 • 8 -48 25 12 52
Longitude: 65~
0- 5 9.7 1.6 + 1.0 22 + 13 172 29 8 41
5- 6 9.3 4.4 -4- 1.7 58 4- 22 178 46 13 44
6- 8 6.8 4.2 + 1.4 40 4- 13 136 60 19 44
8-18 12.6 1.2 + 0.9 22 • 15 138 59 22 43
18-25 17.5 3.7 + 1.3 91 4-32 -139 45 20 44
25-30 9.7 3.4 4- 1.1 47 • 14 -76 38 16 46
30-33o 8.0 5.2 i 0.9 59 • 10 -39 30 14 45
33o-34 9.4 3.0 + 0.7 40 + 9 -52 26 13 45
Longitude: 75~
0- 5 9.7 3.5 • 1.2 47 -t- 16 174 30 l0 35
5- 6 9.3 7.1 • 1.9 94 • 25 177 48 15 39
6- 8 6.8 6.8 + 2.1 65 • 20 154 63 22 37
8-18 12.6 2.4 -t- 1.5 42 -t- 26 161 64 26 34
18-25 17.5 5.1 -+- 1.4 127 4-34 -151 49 23 33
25-30 9.7 3.4 4- 1.1 47 + 15 -85 41 19 35
30-33o 8.0 5.0 + 1.0 57 -+- 11 -45 32 16 35
33o-34 9.4 2.9 + 0.7 38 i 9 -60 29 14 36
Longitude: 85~
0- 5 9.7 5.2 + 1.3 72 -t- 18 174 31 11 30
5- 6 9.3 9.6 • 2.1 127 • 27 176 49 17 34
6- 8 6.8 9.6 + 3.1 92 + 30 161 66 25 31
8-18 12.6 3.6 + 2.1 65 • 36 167 68 29 27
18-25 17.5 6.5 + 1.5 162 + 38 - 159 54 26 24
25-30 9.7 3.5 • 1.1 47 • 15 -93 45 21 26
30-33o 8.0 4.8 -t- 1.1 55 4- 12 -51 35 17 26
33o-34 9.4 2.8 • 0.8 37 + 10 -70 31 15 28
The parameters rmaj and rmin are the semi-major and semi-minor axes of the ellipse of confidence (95% level). The variable 0maj is the
azimuth of the semi-major axis.

Fig. 10. Gravity anomalies in the Caribbean area and North America-South America plate motion through time of three points attached
to the North American plate with respect to the South American plate for eight stages from chron 34 (83 Ma) to the present. The
relative motion vectors can be divided into four groups: (1) slow sinistral strike-slip/transtension between chrons 34 and 25 (55.9 Ma)
(~2.8 4-0.8-4.8 + 1.1 mm/year at 85~ (2) northeast-southwest-oriented convergence from chron 25 to 18 (38.4 Ma) (6.5 -+- 1.5
mm/year at 85~ (3) slow motion from chron 18 to 8 (25.8 Ma) (3.6-t-2.1 mm/year at 85~ and (4) fast north-south-oriented
convergence from chron 8 to 6 (19.0 Ma) (9.6 4- 3.1 mm/year for chrons 8-6 and 9.6 4- 2.1 mm/year for chrons 6-5 at 85~ followed
by a deceleration in north-south-oriented convergence post-chron 5 (9.7 Ma) (5.2 4- 1.3 mm/year at 85~ See Table 6 for a complete
list of stage motion vectors. Note acceleration in convergence at chron 8.
a number of anomalous ridges and troughs. The Bar- the Barracuda-Tiburon ridge area started at chron 6.
racuda and Tiburon ridges east of the Lesser Antilles We model 32 4-23 km of convergence from chron 6 to
Arc (Fig. 12) both exhibit unusually large Bouguer 5, but virtually no relative motion after chron 5 ( 4 + 18
gravity anomalies (up to ~135 mGal). The eastern km), which is much too small to be resolved, by our
continuation of the plate boundary is expressed in model. The post-chron 8 relative motion in the Royal
the Researcher Ridge and Royal Trough (Fig. 12). Trough area can only be resolved for post-chron 5
The Royal Trough exhibits en-dchelon-shaped tec- time. Here our model implies 22 + 10 km of exten-
tonic fabric and fresh basalts on a basement charac- sion for the last 10 million years.
terized by spreading center type faulting identified Our plate model supports the suggestion that the
from GLORIA data, whereas the Researcher Ridge present topographic and gravity expression of the
has a large magnetic anomaly. Both features are Tiburon and Barracuda ridges may have resulted pri-
interpreted as extensional structures (Collette et al., marily from Neogene plate convergence (cf. Mtiller
1984; Roest and Collette, 1986). and Smith, 1993). In particular, the strong Moho up-
For the area at the Tiburon and Barracuda ridges, lift at the Tiburon Ridge, where the crust is modeled
our plate model results in a total of 151 -+- 31 km to be 1.5-2 km thick with a Moho uplifted more
left-lateral transtension from chron 34 (83 Ma) to 25 than 4 km (Mt~ller and Smith, 1993), is extremely
(55.9 Ma). A phase of slow transpression is predicted unstable. Without compressive forces, oceanic crust
for chron 25 (55.9 Ma) to 18 (46.3 Ma), which is not as thin as 1.5-2 km would subside and form a de-
well resolved, followed by extremely slow motion be- pression, rather than a ridge. The plate model used
tween chrons 18 and 6, during which time it is within by Miiller and Smith (1993) did not allow them
the computed 95% confidence areas for the entire part to discriminate when in the Tertiary North-South
of the plate boundary east of the Lesser Antilles sub- America convergence in this area and the uplift of
duction zone (Fig. 12). The plate boundary area east the two ridges might have been initiated. However,
of 56~ was located quite close to the stage poles of they argued that Middle Eocene-Late Oligocene tur-
motion; the resulting vectors of relative motion are so bidites on the slope of the Tiburon Ridge, which is
small that they cannot be resolved. Convergence in now located 800 m above the abyssal plain, may

Fig. 11. North America-South America plate motion through time of three points attached to the North American plate with respect to the
South American plate computed for the same seven stages as Pindell et al.'s (1988) model, shown by triangles. The general shape of both
models is similar, reflecting the similarity of the magnetic anomaly data sets used to constrain both models. The main differences between
the models are as follows. (1) Pindell et al.'s (1988) model implies sinistral strike-slip between North and South America from chron 34 to
21, followed by convergence. Our model implies strike-slip until chron 25, followed by convergence. (2) Pindell et al.'s (1988) model im-
plies relative constant convergence between the Americas of rates of 4-6 ram/year since chron 21. Our model results in substantial vari-
ations in convergence rates from chron 25, as documented in Fig. 10 and Table 6. In particular we resolve the onset of fast convergence
after chron 8 of nearly 10 ram/year, compared with less than 4 ram/year from chrons 18 to 8 measured at 85~ This figure also shows
the estimated motion of the Caribbean plate in a hotspot reference frame since chron 18 (38.4 Ma) (open circles); solid circles show the
present-day position of the two points attached to the Caribbean plate used as starting points for these flow lines. See text for discussion.
indicate that most of its uplift occurred in post- alleviate the need for upslope turbidite deposition on
Oligocene times, a conclusion strongly supported by the rise in the Middle E o c e n e - L a t e Oligocene, as
the plate model presented here. suggested by Dolan et al. (1989, 1990). Instead, the
This idea contrasts the interpretation of Dolan present elevation of the turbidites on the slope of the
et al. (1989, 1990), who suggested that the present rise would reflect post-depositional uplift.
topography of the Tiburon Ridge had existed prior to
deposition of the turbidites, which would have been Middle American Trench to Lesser Antilles
emplaced by upslope deposition from the abyssal
plain onto the rise. Dolan et al. (1989, 1990) argue Our results suggest that slow sinistral transten-
that the Tiburon Rise has existed as a bathymetri- sion/strike-slip between the two Americas lasted
cally shallow feature since the Late Cretaceous. An from chron 34 (83 Ma) until chron 25 (55.9 Ma)
extensive discussion of this question can be found in (~2.8 + 0.8-4.8 + 1.1 m m / y e a r at 85~ How-
Miiller et al. (1993) and Mtiller and Smith (1993). ever, we cannot attribute any particular Caribbean
We find that there is little evidence that would show tectonic events to this phase of slow sinistral
conclusively that the present bathymetric elevation motion. Our model results in roughly northeast-
of the Tiburon Rise has existed since the Late Cre- southwest-oriented slow convergence from chron
taceous. In contrast, the crustal structural modeling 25 (55.9 Ma) to chron 18 (38.4 Ma) (6.5 + 1.5
carried out by Mtiller and Smith (1993), together m m / y e a r at 85~ This time period of convergence
with the plate model presented here, indicates that includes a calc-alkaline magmatic stage in the West
much of the unusually shallow Moho topography Indies, dated as P a l e o c e n e - L o w e r Eocene (Perfit
and crustal uplift of the Tiburon Rise is a result of the and Heezen, 1978), which is thought to be related to
onset of North A m e r i c a - S o u t h America convergence southward subduction of proto-Caribbean crust dur-
at chron 6 (19.0 Ma). A Neogene formation of the ing this time. North-South America relative motion
present topography of the Tiburon Rise would also was characterized by relatively slow transpression

Fig. 12. North America-South America plate motion through time of three points attached to the North American plate with respect to
the South American plate for eight stages from chron 34 (83 Ma) to the present and their simultaneous 95% confidence regions. Only
the Barracuda Ridge area has been affected by North-South American plate motions as old as chron 34 (83 Ma). The ocean crust to the
east becomes successively younger. The oldest relative motion vectors plotted correspond roughly to the age of the ocean crust south of
the Fifteen-Twenty Fracture Zone. Relative plate motion in this area is not well resolved, except for oblique transtension in the Barracuda
Ridge area from chron 34 to 25, and post-chron-6 north-south compression, post-chron-6 extension in the Royal Trough area. Color
version at http://www.elsevier, nl/locate/caribas/
until chron 8 (25.8 Ma) (3.6-+-2.1 mm/year at drastically in the Early to Middle Miocene. The east-
85~ This phase of slow relative motion between ern portion of the North America-Caribbean plate
the two Americas correlates with a period of tectonic boundary displays an accretionary wedge along the
quiescence along some parts of the Caribbean mar- Muertos Trough (Case et al., 1984). Based on seis-
gins during the Oligocene-Early Miocene that was mological evidence Byrne et al. (1985) showed that
characterized by subsiding basins, unconformably the Muertos Trough is an active structure and sug-
overlying the previously deformed belts (Calais et gested it to be the site of subduction. North-south
al., 1989). However, at the same time, collision has convergence is accommodated by oceanic crust un-
occurred off Venezuela (Mann et al., 1995). derthrusting the Greater Antilles, or by folding and
Our model is different from Pindell et al.'s (1988) thrust faulting only, where the crust is thicker, as
plate model for North-South America plate motions, at the Beata Ridge and the Bahamas (Ladd et al.,
in that it results in a drastic increase in convergence 1990). Correlation of seismic reflection data from
velocity subsequent to chron 8 (25.8 Ma) (9.6 -+- 3.1 the turbidite fill in the Muertos Trough with the
mm/year for chrons 8-6 and 9.6 + 2.1 mm/year Venezuelan Basin indicates a Neogene, or possibly
for chrons 6-5 at 85~ compare also Figs. 10 late Neogene age for the initiation of underthrusting
and 11). A slowdown in convergence after chron 5 (Ladd et al., 1990). In the Early Miocene, the Peralta
resulted in a post-chron-5 convergence rate drop to and Rio Ocoa sediment groups on Hispaniola were
5.2 4- 1.3 mm/year at 85~ The modeled Neogene deformed in a southwest verging fold-and-thrust belt
convergence results in 92-4-30 km convergence from (Heubeck et al., 1991), supporting evidence for the
chron 8 to 6, 127 -1-27 km from chron 6 to 5, and onset of Early Miocene convergence within this part
72 4- 17 km from chron 5 to the present near the of the Caribbean-North American plate boundary.
North Panama Deformed Belt at 85~ (Fig. 10). Tectonic events at this time have also been mapped
The Neogene convergence measured south of the in the Dominican Republic, in Haiti, and on Cuba
Muertos Trough at 65~ is 40 4- 13 km from chron (Calais et al., 1992).
8 to 6, 58 4- 22 km from chron 6 to 5, and 22 4- 13 The Caribbean-South American plate boundary
km from chron 5 to present day. comprises a wide and complicated plate boundary
Geological data from the circum-Caribbean plate zone. It starts at the deformation front of the subduc-
boundaries indicate that the tectonic regime changed tion zone, includes various dextral strike-slip faults

of northern South America (e.g. Bocono, Oca, E1 Mao, 1997). They found an increase of differential
Pilar) and continues to the south as a fold-thrust north-south convergence from east to west from
belt (e.g. Ladd et al., 1984). Biju-Duval et al. (1984) about 1 mm/year at the Barracuda ridge to about
analyzed multichannel seismic reflection data in the 9 mm/year at 85~ It is interesting to note that
Venezuelan Basin and concluded that the present their modeled present-day convergence rate of 9
configuration of the margin, i.e. underthrusting of mm/year at 85~ is similar to the rates we calculate
the Caribbean oceanic crust below the South Amer- for chron 8-5 time, and significantly faster than our
ican borderland, developed in the Early or Middle post-chron-5 convergence estimate at 85~ of 5.2 -+-
Miocene. They also realized that north-south short- 1.3 mm/year. It is not clear whether this difference
ening observed at the North Venezuelan margin reflects a recent acceleration in North America-
may be the result of regional North-South America South America plate convergence, or whether it is
convergence. Recently, a comprehensive analysis of related to problems in our anomaly-5 reconstruction.
new and existingseismic data from the Beata Ridge The latter may well be the case, since the South
and adjacent areas by Mauffret and Leroy (Chapter Atlantic reconstruction for this time is not well
21) has shown that the Beata Ridge, a Cretaceous constrained; it is based on a small number of data
plateau, is bounded to the east by compressive struc- points only, and Table 2 shows that we have slightly
tures reactivated by right-lateral strike-slip, and by underestimated the uncertainties of both magnetic
normal faults to the west. Uplift of the ridge in- and fracture zone identifications for this data set. In
creases from south to north, and is estimated to have contrast, the anomaly-5 reconstruction in the central
started in the Early Miocene (23 Ma), resulting in North Atlantic is extremely well constrained.
a total shortening between 170 km and 240 km as We propose that the post-chron-8 convergence
a function of latitude (Mauffret and Leroy, Chapter between North and South America has also played a
21). They interpret the Beata Ridge as a compres- substantial role in the Neogene Panama Arc collision
sional plate boundary, resulting from overthrusting and subsequent arc deformation to an S-shaped
of the Colombian microplate onto the Venezuelan pattern. In the collision area, the computed north-
microplate. The implied clockwise rotation of the south convergence between the Americas resulted in
Colombian microplate and convergence between the a total of 291 -+- 75 km of north-south convergence.
latter and the Venezuelan microplate are consis- During the Middle to Late Miocene the onset of the
tent with differential convergence between the North collision of the Costa Rica-Panama Arc with the
and South American plates, increasing from east to western Cordillera of northwestern South America
west, thereby 'squeezing' the Colombian plate out (Wadge and Burke, 1983; Eva et al., 1989; Mann
to the east, as suggested by Burke et al. (1978). et al., 1990) started forming the North Panama
Mauffret and Leroy (Chapter 21) suggest that the Deformed Belt.
observed deformation may also be caused by the The tectonic events outlined above are all slightly
buoyancy of the Cocos plate, which is subducting younger (mostly Early Miocene) than the onset of
under the Caribbean plate (England and Wortel, North-South America convergence predicted by our
1980; Meijer, 1992). However, Central America ex- model (~26 Ma, Late Oligocene). This may reflect
periences extension in the back arc in the north an artifact of our model. Our plate model lacks
where the subducting Cocos plate is older than in resolution between anomalies 8 (25.8 Ma) and 6
the south, whereas a younger Cocos plate in the (19.0 Ma), because it is not straightforward to iden-
south causes shortening. Alternatively, compression tify the magnetic anomalies between 6 and 8 with
in Costa Rica, as expressed by the April 22, 1991 confidence in a slowly spreading tectonic regime. It
Costa Rica earthquake (Plafker and Ward, 1992), is possible that the Early Miocene tectonic events
has been attributed to the subduction of an aseismic in the Caribbean correspond to a global change in
ridge (Adamek et al., 1987). Mann and Burke (1984) plate motions. Evidence for this idea comes from
suggested that the Beata Ridge may be the conse- a detailed survey of the Pitman Fracture Zone in
quence of northward motion of the Maracaibo block, the South Pacific that shows a distinct change in
a tectonic block of South America. Recent work has spreading direction at chron 6c (Cande et al., 1995)
confirmed a north- to northeast-directed motion of which represents the Oligocene-Miocene boundary
these blocks relative to the Caribbean plate (Ego et (23.8 Ma). It is virtually impossible to identify this
al., 1995). In summary it is unclear what the role of magnetic anomaly in the slowly spreading central
the Cocos plate may be in terms of contributing to North Atlantic and South Atlantic oceans. There-
compression at the Beata Ridge. fore, we consider it possible that convergence started
An east-west gradient in convergence between at the Oligocene-Miocene boundary, 2 m.y. later
the Americas is also supported by a recent analysis than predicted by our plate model. Pindell et al.'s
of present-day relative plate motions between North (1988) model results in an acceleration in conver-
and South America based on GPS data (Dixon and gence at chron 6 (20 Ma in the DNAG timescale

52 R.D. MfJLLER et al.
(Kent and Gradstein, 1986), 19 Ma in the timescale initial contact of a Caribbean plate originating from
used here (Cande and Kent, 1995), with extremely the Pacific has been revised to late Campanian-
slow convergence from chron 13 to chron 6. This Maastrichtian, when syn-orogenic sedimentation and
demonstrates that a model not constrained by any northward verging folding, thrusting and obduction
magnetic anomaly identification between anomaly 6 of ophiolites have occurred at the southern margin
and 13 (a time interval 14 m.y. long) results in an ap- of the Yucat~in Peninsula in Guatemala (Rosenfeld,
parent acceleration in convergence at 19 Ma, about 5 1990). The arguments in favor of an allochthonous
m.y. later than the Oligocene-Miocene boundary. nature of the Caribbean oceanic crust have been
reviewed comprehensively by Pindell and Barrett
Plate motions relative to the mantle (1990).
In the Paleocene the Yucat~in Basin opened along
We put Caribbean plate motions into an abso- a left-lateral strike-slip fault (Rosencrantz, 1990)
lute hotspot reference flame based on Atlantic- and the prograding arc started colliding with the
Indian ocean hotspot tracks (Mtiller et al., 1993) for Bahamas Platform in western Cuba (Bralower et
understanding cause and effects of plate motions be- al., 1993). For the time after the Middle Eocene
tween the Americas and the Caribbean plate(s). Ross we have an estimate for North America-Caribbean
and Scotese (1988) used a paleomagnetic reference plate motions from the spreading history in the
flame for their model (which cannot resolve longitu- Cayman Trough (Rosencrantz et al., 1988), even
dinal motions of plates), and correspondingly do not though spreading here may not reflect the total
show a geographic flame on their reconstructions. North America-Caribbean motion (Rosencrantz and
Pindell et al. (1988) used the absolute plate motion Mann, 1991). Burke et al. (1980) suggested that
model by Engebretson (1982) and Engebretson et cumulative offsets of strike-slip faults on Jamaica
al. (1985) for the Pacific based on hotspot tracks to suggest a minimum rate of offset of 4 mm/year,
calculate Pacific-Caribbean relative motions, which in addition to an average of 16 mm/year of total
have likely exerted controls on Caribbean tectonic opening in the Cayman Trough. We implemented
evolution in the Mesozoic and early Tertiary (Pindell this suggestion in our rotation model, similar to Ross
et al., 1988), but not necessarily in the Neogene. In and Scotese (1988), by allowing for 4-6 mm/year of
any case, Pacific absolute plate motions are only of left-lateral strike-slip between Jamaica and southern
limited use to constrain absolute motions of plates Hispaniola. In Fig. 11 we show the resulting path of
bordering the Atlantic Ocean, as our knowledge on two points attached to the Caribbean plate relative
closing plate circuits crossing the boundary between to the mantle (without considering relative motion
East and West Antarctica is still inadequate (Molnar between the Colombian and Venezuelan microplates
and Stock, 1987; Cande et al., 1995). through time, which we cannot reconstruct). The two
We use the relative plate motion model for the absolute plate motion paths in Fig. 11 as well as the
central North Atlantic and South Atlantic presented plate reconstructions in Fig. 13b-d show that the
here, a revised relative plate motion model for the Caribbean plate has been virtually stationary with
Caribbean area, largely based on the tectonic ele- respect to the mantle at least since the onset of
ments and plate hierarchy from Ross and Scotese seafloor spreading in the Cayman Trough. The er-
(1988), and the model for motion of plates in the rors from combining the 'absolute' and relative plate
Atlantic and Indian Ocean Hemisphere relative to motions models involved in this calculation are prob-
major hotspots (Mtiller et al., 1993). The combined ably larger than the total length of the path shown.
rotation model has been adapted to the Cande and The Caribbean plate could have only maintained a
Kent (1995) timescale for post-chron-34 (83 Ma) substantial eastward component of motion if either
times and the Gradstein et al. (1994) timescale for the Cayman Trough opened much later and faster
earlier times. than presently assumed, and/or if there has been
The absolute plate motion model by MiJller et much faster strike-slip between the Caribbean plate
al. (1993) is based on jointly fitting dated hotspot and Jamaica than suggested by Burke et al. (1980).
tracks on the Australian, Indian, African, and North North America's and South America's plate mo-
and South American plates relative to present-day tions in the mantle reference flame are both char-
hotspots assumed fixed in the mantle. Therefore this acterized by relatively fast westward motion, with a
model is better constrained than a model solely small component of convergence added at chron 13
based on the hotspot tracks of one plate. due to a clockwise change in South American plate
Pindell et al. (1988) suggested that most of the motion, and even faster convergence after chron 8
total opening by seafloor spreading between the due to a counterclockwise change in North America
two Americas was accomplished some time between absolute plate motions. In contrast, the Caribbean
100 and 90 Ma, when the Caribbean plate started plate appears to have been virtually stationary in a
entering from the west. However, the age of the mantle reference flame at least since chron 18. This

result agrees with an idea put forward by Sykes et stationary Caribbean plate. The differences between
al. (1982), who noted that only a small fraction of North America-Africa and South America-Africa
its perimeter is attached to a subducting slab. Even plate motions, as described here, resulted in changes
though the Caribbean slab under the South American in relative motion between the two Americas whose
plate has been shown to be longer than previously effects are clearly seen in the tectonic develop-
thought, the forces assumed to be most important ment along the northern and southern margin of
for driving plates, namely ridge push, slab pull and the Caribbean area. Since the Caribbean plate does
trench suction (this force acts to draw plates together not appear to have moved substantially relative to
at a trench; Elsasser, 1971) must be relatively small. the mantle during the Neogene, there are no major
If they were not, then the Caribbean plate would not tectonic processes which can be attributed to the
rest in a mantle reference frame, as found by our eastward 'escape' of the Caribbean plate during this
plate kinematic analysis. time (e.g. Mann, 1997). In particular for the time
Our result is also in accordance with the anal- since chron 6 (19 Ma), for which Caribbean-North
ysis by Gripp and Gordon (1990) of present-day America relative motion is better constrained than
Caribbean plate motions with respect to the hotspots. for earlier times, we find that the Caribbean plate
Their analysis, based on motion of the Pacific plate was virtually fixed relative to the mantle.
relative to its underlying hotspots, and the NUVEL-1 This observation suggests that accelerated con-
relative motion model by DeMets et al. (1990) re- vergence post-chron 6c (23.8 Ma) at the Oligocene-
sults in roughly west-southwest-oriented motion of Miocene boundary reduced the space within the
the Caribbean plate. However, their motion vectors eastward 'escaping' arc could operate such that
do not differ significantly from zero. Caribbean plate motion relative to the mantle ceased.
It must be concluded that tectonic plate bound- It follows that most deformation at the northern and
ary processes between the Caribbean plate and southern Caribbean plate boundaries in the Miocene
the Americas are entirely driven by relatively and younger was entirely governed by changes in ab-
fast, mostly westward motion of North and South solute plate motion of the North American and South
America. The resulting differential motion between American plates, and the resulting motion relative
North and South America affects a stationary to the Caribbean plate. While strike-slip along the
Caribbean plate trapped between two larger plates northern and southern Caribbean margins continued
by edge-driven plate tectonic interactions, equiva- since the east-west component of absolute plate mo-
lent to some small plates in the Middle East (e.g. tion of the Americas was far larger than the north-
Arabia/Anatolian plate; McKenzie, 1972). south components, the magnitude of the latter in-
Sykes et al. (1982) recognized this possibility, creased, probably at the Oligocene-Miocene bound-
but suggested alternatively that the Caribbean plate ary, resulting in convergence between the two Amer-
may be forced to move eastward in response to icas. The rate of convergence increased from east to
the gradient in convergence rate between North and west, resulting in an eastward-directed 'squeeze' on
South America, increasing from east to west, as also the Caribbean plate which caused its breakup along
found by our analysis. In contrast, the plate motion the Beata Ridge, where east-west-oriented compres-
paths plotted in Fig. 11 suggest that the eastward sional stresses are absorbed (Mauffret and Leroy,
motion of the Caribbean plate with respect to the Chapter 21).
two Americas is entirely due to westward motion of Mann et al. (1995) show a model for the for-
the latter two plates with respect to the mantle, and mation of Caribbean microplates in six stages from
that the east-west convergence gradient quoted by the Maastrichtian to present-day in a fixed South
Sykes et al. (1982), which has been constrained to American framework. Their figures show that the
post-chron-8 (25.8 Ma) times by our model (prob- Caribbean plate has moved eastward by about 800
ably post-chron 6c as discussed above), may not km since the mid-Oligocene (relative to South Amer-
have resulted in substantial eastward motion of the ica). Mann et al. (1995) reason that collision ceased
Caribbean plate with respect to the mantle. Rather, in the Middle Eocene in central Cuba since the
the east-west gradient in post-chron-6c convergence arc could advance no further to the north-northeast
may have contributed to causing east-west compres- above the Bahamas Platform, and that this event
sion at the Beata Ridge, as described in Mauffret and rotated Caribbean plate motion clockwise in a more
Leroy (Chapter 21). easterly direction. They favor the 'tectonic escape'
Our combination of relative and absolute plate mechanism proposed by Burke and Seng6r (1986),
motions indicates that throughout the Tertiary tec- which results in the motion of a colliding plate to-
tonic processes at the northern and southern bound- wards the remaining 'free face', e.g. an island arc.
ary of the Caribbean plate were governed by the In case of the Caribbean in Middle Eocene times,
relatively fast westward motion of both the North the remaining free face would have been towards the
and South American plates with respect to a nearly east, i.e. the Lesser Antilles Arc. This argument is

America
t:J,'
20 ~
15 ~ Caribbean plate
stationary
%
10 ~
(a) America
Chron 18, 38.4 3 crr~/a
Early Eocene
America d"
20 ~
15 ~ Caribbean plate
stationary
10 ~
South
America
(b) 2.15 cnga
Chron 8, 25.8 Ma
Late Oligocene
Platform
20 ~
Gonave
Venezuela plate
15 ~
stationary
Colombia plate
10 ~
Venezuela
South
America
Cocos
(C) Plate 1.3 Cm/a
Chron 5, 9.7 Ma
Late Miocene
00~
_90 ~ _85 ~ _80 ~ -75 ~ .70 ~ -65* -60*

plausible. However, if we put Caribbean plate recon- formation of the oceanic crust at the Barracuda and
structions in the Atlantic-Indian hotspot framework Tiburon ridges. MUller and Smith (1993) inverted
(Fig. 13), it appears that the eastward motion of Bouguer anomalies for crustal layer structure, and
the Caribbean plate had ceased at Middle Eocene found that the Moho is uplifted 2 - 4 km over short
times. wavelengths ( ~ 7 0 km) at the Barracuda and Tiburon
it follows that the apparent continuing apparent ridges, implying large anelastic strains and an un-
'escape' of an arc system as described by Roy- stable density distribution. Together with the plate
den (1993), e.g. for the Scotia Arc between South model presented here, these results indicate that
America and the Antarctic plate, may not neces- much of the unusually shallow Moho topography
sarily involve the absolute motion of a small plate and crustal uplift of the Tiburon Rise is a result
(e.g. the Scotia Sea plate) relative to the mantle. A of North A m e r i c a - S o u t h America convergence af-
retreating subduction boundary may be initiated by ter chron 6 (19.0 Ma). This model is in contrast
the change in polarity of a subduction system, as in with Dolan et al.'s (1989, 1990) suggestion that the
the case of the proto-Caribbean, but the Caribbean present topography of the Tiburon Rise has existed
plate never reached the 'open ocean' as in the case since the Late Cretaceous.
of the Scotia Sea (Royden, 1993), since its east- Our results suggest that slow sinistral transten-
ward migration was inhibited by boundary forces to sion/strike-slip between the two Americas lasted
the north and south, due to progressive convergence from chron 34 (83 Ma) until chron 25 (55.9 Ma),
between the two Americas. As a result, Caribbean followed by roughly northeast-southwest-oriented
absolute plate motion stopped. Subsequently, all rel- convergence until chron 18 (38.4 Ma). This first
ative motion observed between the Americas and the convergent phase correlates with a P a l e o c e n e - E a r l y
Caribbean plate and associated tectonic elements has Eocene calc-alkaline magmatic stage in the Greater
been caused by the absolute plate motion of North Antilles, which is thought to be related to south-
America and South America relative to a stationary ward subduction of proto-Caribbean crust during
Caribbean plate. this time. Relatively slow transpression until chron 8
is followed by a drastic increase in convergence ve-
locity. Subsequent to chron 8 (25.8 Ma), probably at
CONCLUSIONS the O l i g o c e n e - M i o c e n e boundary, fast convergence
resulted in 92 4- 22 km convergence from chron 8 to
New gravity anomaly data from satellite altime- 6, 127 • 25 km from chron 6 to 5, and 72 4- 17 km
try and new magnetic data allow us to construct from chron 5 to the present measured at 11~ 85~
a modified plate model for plate motions between near the North Panama Deformed Belt. The Neo-
the two Americas, and calculate its uncertainties. gene convergence measured at the eastern Muertos
For the N o r t h - S o u t h America plate boundary area Trough, at 17.5~ 65~ is 41 + 18 km from chron 8
east of the Lesser Antilles Arc our results are in to 6, 58 4- 25 km from chron 6 to 5, and 22 4- 17 km
good agreement with the observed strong plate de- from chron 5 to present day. The modeled conver-
Fig. 13. Plate reconstructions of the Caribbean area in an Atlantic-Indian mantle reference system for chrons 18 (38.4 Ma), 8 (25.8 Ma),
and 5 (9.7 Ma). See Fig. 2 for labels of tectonic elements. Bold arrows show North American and South American plate motion relative
to the mantle for the time intervals from 38.4 to 25.8 Ma (a), 25.8 to 9.7 Ma (b), and 9.7 Ma to present day (c). The eastward escape
of the Caribbean plate had ceased when the opening of the Cayman Trough started (a). This time corresponds to the collision in central
Cuba which prevented a further advance of the Caribbean plate to the north-northeast above the Bahamas Platform. Only if the Cayman
Trough would have opened later and spread considerably faster than interpreted by Rosencrantz et al. (1988), and/or if contemporaneous
strike-slip between Jamaica and the Caribbean plate was considerably faster than suggested by Burke et al. (1980), would the Caribbean
plate have maintained any eastward-directed motion relative to the mantle after the Middle Eocene. (b, c) Relatively slow convergence
between the Americas from chron 18 (38.4 Ma) to chron 8 (25.8 Ma) was followed by rapid convergence after chron 8, probably
starting at the Oligocene-Miocene boundary, averaging 9.6 mm/year until chron 5 (9.7 Ma), slowing down to 5.2 mm/year after chron
5. Accelerated convergence was caused by a counterclockwise change in the absolute plate motion direction of North America. As a
result, the total area available for the western Caribbean plate at 85~ was reduced by at least "~230 km in north-south direction in the
last 25 m.y. We suggest that about half of the north-south extent (500 km) of the Maracaibo slab (subducted Caribbean oceanic plateau
crust) under the South American continent (van der Hilst and Mann, 1994) may have resulted from post-Oligocene South America-
Caribbean convergence. At least since chron 18 (38.4 Ma) Cocos plate-Caribbean interactions have not resulted in any substantial
motion of the Caribbean plate relative to the mantle. Continuing oblique collision along the passive margin of eastern Venezuela (Algar
and Pindell, 1993) must be attributed to the west-northwestward motion of South America relative to the mantle and relative to a
stationary Venezuelan plate, rather than to continuing eastward movement of the Caribbean plate. Equivalently, the Miocene and younger
transpression observed in Hispaniola (Heubeck and Mann, 1991) due to collision of arc rocks with the Bahamas Platform is the result
of continuing westward motion of the North American plate (and the Bahamas Platform) relative to a stationary Venezuelan plate in a
mantle reference frame, rather than a continuing eastward 'escape' of the Caribbean plate. The Gonave microplate has been transferred
from the Venezuelan plate to North America in the Pliocene (Mann et al., 1995), leaving the rest of the Venezuelan plate behind.

56 R.D. MCILLER et al.
gence may correspond to the Early Miocene onset of ative to a stationary Caribbean plate as shown be-
underthrusting of the Caribbean oceanic crust below fore.
the South American borderland in the Colombian Our plate model shows well resolved north-south
and Venezuelan basins, the onset of subduction in convergence between the Americas during the Neo-
the Muertos Trough, and folding and thrust fault- gene, and we argue that this plate convergence is
ing at the Beata Ridge and the Bahamas, and the likely the main cause for the formation of those de-
breakup of the main part of the Caribbean plate formed belts which cannot be attributed to east-west
into the Venezuelan and Colombian plates, separated convergence as described above. We also suggest
by the Beata Ridge acting as a convergent plate that the fast Neogene plate convergence between
boundary (Mauffret and Leroy, Chapter 21). The North and South America contributed to the Late
east-west shortening between the latter two plates Miocene onset of the collision of the Costa Rica-
may reflect the differential convergence between the Panama Arc with the western Cordillera of South
two Americas, increasing from east to west. America (Wadge and Burke, 1983; Eva et al., 1989;
The main differences with Pindell et al.'s (1988) Mann et al., 1990; Mann and Corrigan, 1990). One
model are the following. of the main tectonic events affecting the Caribbean
(1) Pindell et al.'s (1988) model implies relatively plate in the Neogene has been its breakup into
constant convergence between the Americas of rates the Venezuelan and Colombian plates (see Mauf-
of about 5 mm/year or less since chron 21, with fret and Leroy, Chapter 21). The breakup may have
the exception of faster convergence between chron been caused the observed east-west gradient in
6 and 5. Our model results in substantial variations convergence between the Americas, the subduction
in convergence rates from chron 25, as documented of the buoyant Cocos plate under the Caribbean
in Fig. 10 and Table 10. In particular, we resolve plate (Mauffret and Leroy, Chapter 21), or the north-
an initial phase of fast convergence between chron northeastward motion of the Maracaibo block, which
8 (25.8 ma) and chron 6 (19.0 Ma) of nearly 10 in turn may be related to differential North-South
mm/year, compared with less than 4 mm/year from America convergence.
chrons 18 to 8 measured at 85~ We suggest that By combining Atlantic-Indian hotspots as a ref-
most of this convergence occurred after chron 6c erence frame with revised North America-African
(23.8 Ma), which corresponds to the Oligocene- and South America-African relative plate motions,
Miocene boundary, a plate reorganization in the and with a revised plate model for the Caribbean
South Pacific (Cande et al., 1995), and the formation area we are able to show the following.
of the present-day deformed belts north and south (1) The eastward escape of the Caribbean plate
of the Caribbean area. Without identifying magnetic appears to have ceased when the opening of the
anomaly 8, two stages of slow (chrons 13-8) and Cayman Trough started. This time corresponds to
fast (chrons 8-6) convergence are averaged. the collision in central Cuba which prevented a
(2) We have computed uncertainties for our North further advance of the Caribbean plate to the north-
America-South American plate flow lines. Uncer- northeast above the Bahamas Platform. Only if the
tainty ellipses for rotated data points are especially Cayman Trough would have opened later and spread
helpful to evaluate whether or not we can resolve considerably faster than interpreted by Rosencrantz
relatively slow phases of relative motion. et al. (1988), and/or if contemporaneous strike-
(3) We have put Caribbean plate reconstructions slip between Jamaica and the Caribbean plate was
into the Atlantic-Indian hotspot reference system considerably faster than suggested by Burke et al.
from Mtiller et al. (1993). This allows us to evaluate (1980) would the Caribbean plate have maintained
causes and effects of relative plate motions in the any eastward-directed motion relative to the mantle
Caribbean area. after the Middle Eocene. If we rely only on the
St6phan et al. (1986) put forward the hypothesis interpreted post-chron-6 (19 Ma) spreading history
that both the northern and the southern Caribbean of the Cayman Trough by Rosencrantz et al. (1988),
deformed belts are the result of the bending of which is better constrained than its previous open-
the Caribbean continental frame related to east- ing, then the Caribbean plate is still found to have
west shortening. East-west shortening appears to been without any substantial motion relative to the
have a variety of different causes. In the Panama mantle subsequent to chron 6 within the errors of
area, east-west shortening is related to Nazca-South absolute plate motion models.
America convergence in that direction and colli- (2) It is not the case that North America-South
sion of an east-west-oriented arc with a north- America plate motions had only minor effects on the
south-oriented margin (Mann and Corrigan, 1990; development of the Caribbean region after the Cam-
Wadge and Burke, 1983). In Hispaniola, northeast- panian, as suggested by Pindell et al. (1988). After
southwest shortening is related to the interaction the breakup of the Caribbean plate into the Venezue-
of the westward-moving North American plate rel- lan and Colombian plates, the eastward driving force

of the latter plate may have still been derived from plate came to a halt in the Atlantic-Indian mantle
interactions with the Cocos plate. However, even reference system due to a progressive reduction in
if this is so, our model suggests that post-chron-8 space between the two Americas for the arc to move
(25.8 Ma) differential motion between the Americas eastward and due to its collision with the Bahamas
has resulted in a total of 291 + 64 km in conver- Platform. We show that the most severe reduction in
gence at 85~ near the North Panama Deformed space started at the Oligocene-Miocene boundary,
Belt. In other words, the total area available for the resulting in the gradual formation of many of today's
Colombian plate at 85~ was reduced by at least tectonic elements and sedimentary basins in the
~230 km in north-south direction in the last 25 Caribbean area.
m.y. Surely this reduction in space had a profound
influence on the Colombian plate margin, and has
contributed to convergence along the North Panama ACKNOWLEDGEMENTS
Deformed Belt. van der Hilst and Mann (1994)
show that the subducted Maracaibo slab underly- The contents of this paper have been clarified
ing northwestern South America extends up to 500 substantially by Jim Pindell's comprehensive review
km from the Caribbean-South America boundary to of an early draft. Given that we are no experts
the south. The Maracaibo slab corresponds to sub- on Caribbean geology, our discussion and the in-
ducted Caribbean oceanic plateau crust. Our results terpretation of our results has benefited from many
suggest that about half of the north-south extent discussions with Eric Calais and Alain Mauffret,
of the Maracaibo slab under the South American as well as from thorough reviews by Jan Golonka,
continent may have resulted from Miocene and later Mark Gordon, and Ian Norton. We thank Paul Mann
South America-Caribbean convergence, if most of for his encouragement to make a contribution to this
the North America-South America convergence was volume. JYR acknowledges support from the Centre
taken up at this boundary. Nationale de la Recherche Scientifique (CNRS) that
(3) The Caribbean plate has been trapped be- enabled his visit at the University of Sydney, and
tween two larger plates, and has been subject to from the Plan Nationale de Teledetection Spatiale.
edge-driven plate tectonic interactions since then. It UMR 6526 G6osciences Azur contribution 139, Ge-
follows that the main control on North America- ological Survey of Canada contribution 1997027.
Caribbean and South America-Caribbean plate in-
teractions has not originated from Cocos plate-
Caribbean interactions, as these interactions have not REFERENCES
resulted in any substantial motion of the Caribbean
plate relative to the mantle. Continuing oblique col- Adamek, S., Tajima, E and Wiens, D.A., 1987. Seismic rupture
lision along the passive margin of eastern Venezuela associated with subduction of the Cocos Ridge. Tectonics, 6:
757-774.
as reported by Algar and Pindell (1993) must be at-
Algar, S.T. and Pindell, J.L., 1993. Structure and deformation
tributed to the west-northwestward motion of South history of the Northern Range of Trinidad and adjacent areas.
America relative to the mantle and relative to a sta- Tectonics, 12: 814-829.
tionary Venezuelan plate, rather than to continuing Biju-Duval, B., Mascle, A., Rosales, H. and Young, G., 1984.
eastward movement of the Caribbean plate. Equiv- Episutural Oligo-Miocene basins along the North Venezuelan
margin. In: J.S. Watkins and C.L. (Editors), Drake Studies in
alently, the Miocene and younger transpression ob-
Continental Margin Geology. Am. Assoc. Pet. Geol. Mem.,
served in Hispaniola (Heubeck and Mann, 1991) due 34: 347-358.
to collision of arc rocks with the Bahamas Platform Bralower, T.J., Hutson, E, Mann, E, Iturralde-Vinent, M., Renne,
is the result of continuing westward motion of the R and Sliter, W.V., 1993. Tectonics of oblique arc-continent
North American plate (and the Bahamas Platform) collision in western Cuba, 1. Stratigraphic constraints. Eos,
Trans. Am. Geophys. Union, 74: 546.
relative to an approximately stationary Venezuelan
Burke, K. and Seng6r, C., 1986. Tectonic escape in the evolu-
plate in a mantle reference frame, rather than a tion of the continental crust. In: Barazangi, M. and Brown,
continuing eastward 'escape' of the Caribbean plate. L. (Eds.), Reflection Seismology: The Continental Crust. Geo-
By the same token, the Gonave microplate, which dyn. Ser. 14, Am. Geophys. Un., Washington, D.C., pp. 41-
has been detached from the Caribbean microplate in 53.
Burke, K., Fox, R and Seng6r, A.M.C., 1978. Buoyant ocean
the Pliocene and accreted to North America has not
floor and the evolution of the Caribbean. J. Geophys. Res., 83:
been 'left behind' (Mann et al., 1995), but has rather 3949-3954.
been transferred from a stationary Venezuelan plate Burke, K., Grippi, J. and Seng6r, A.M.C., 1980. Neogene
to North America, leaving the rest of the Venezuelan structures in Jamaica and the tectonic style of the northern
plate behind. Caribbean plate boundary zone. J. Geol., 88: 375-386.
Byrne, D.B., Suarez, G. and McCann, W.R., 1985. Muertos
(4) The Tertiary tectonic history of the Caribbean
Trench subduction microplatetectonics in the northern
plate can be described by 'tectonic escape' up to Caribbean? Nature, 317: 420-421.
the Middle Eocene. Subsequently the Caribbean Calais, E., St6phan, J.-F., Beck, Ch., Carfantan, J.-Ch., Tardy, M.,

58 R.D. MLILLER et al.
Thery, J.-M., Olivet, J.-L., Bouysse, Ph., Mercier de L6pinay, relative to the hotspots incorporating the NUVEL-1 global
B., Tournon, J., Vila, J.-M., Mauffret, A., Blanchet, R., Bour- plate motion model. Geophys. Res. Lett., 17:1109-1112.
gois, J. and Dercourt, J., 1989. l~volution pal6og6ographique Gradstein, EM., Agterberg, EP., Ogg, J.G., Hardenbol, J., van
et structurale du domaine cara'/be du Lias ~t l'Actuel: 14 6tapes Veen, P., Thierry, J. and Huang, Z., 1994. A Mesozoic time
pour 3 grandes p6riodes. C. R. Acad. Sci. Paris, 309: 1437- scale. J. Geophys. Res., 99:24,051- 24,074.
1444. Harrison, C.G.A. and Lindh, T., 1982. A polar wandering curve
Calais, E., B6thoux, N. and Mercier de L6pinay, B., 1992. From for North America during the Mesozoic and Cenozoic. J.
transcurrent faulting to frontal subduction: a seismotectonic Geophys. Res., 87:1903-1920.
study of the Northern Caribbean plate boundary from Cuba to Hellinger, S.J., 1981. The uncertainties in finite rotations in plate
Puerto Rico. Tectonics, 11: 114-123. tectonics. J. Geophys. Res., 86: 9312-9318.
Cande, S.C. and Kent, D.V., 1995. Revised calibration of the Heubeck, C. and Mann, P., 1991. Geologic evaluation of plate
geomagnetic time scale for the late Cretaceous and Cenozoic. kinematic models for the North American-Caribbean plate
J. Geophys. Res., 100: 6093-6098. boundary zone. Tectonophysics, 191: 1-26.
Cande, S.C., LaBrecque, J.L. and Haxby, W.E, 1988. Plate Heubeck, C., Mann, P., Dolan, J. and Monechi, S., 1991. Di-
kinematics of the South Atlantic: chron 34 to present. J. achronous uplift and recycling of sedimentary basins during
Geophys. Res., 93: 13,479-13,492. Cenozoic tectonic transpression, northeastern Caribbean plate
Cande, S.C., Raymond, C.A., Haxby, W.E and Haxby, W.E, margin. Sediment. Geol., 70: 1-32.
1995. Geophysics of the Pitman Fracture Zone and Pacific- Kent, D.V. and Gradstein, EM., 1986. A Jurassic to Recent
Antarctic plate motions during the Cenozoic. Science, 270: chronology. In: P.R. Vogt and B.E. Tucholke (Editors), The
947-953. Western North Atlantic Region. The Geology of North Amer-
Case, J.E., Holcombe, T.L. and Martin, R.G., 1984. Map of ica, Vol. M, Geological Society of America, Boulder, CO, pp.
geological provinces in the Caribbean region. In: W.E. Bonini, 45-5O.
R.B. Hargraves and R. Shagam (Editors), The Caribbean- Klitgord, K.D. and Schouten, H., 1986. Plate kinematics of
South America Plate Boundary and Regional Tectonics. Geol. the central North Atlantic. In: P.R. Vogt and B.E. Tucholke
Soc. Am. Mem., 162: 1-30.
(Editors), The Western North Atlantic Region. The Geology
Chang, T., 1987. On the statistical properties of estimated rota-
of North America, Vol. M, Geological Society of America,
tions. J. Geophys. Res., 92:6319-6329.
Boulder, CO, pp. 351-404.
Chang, T., 1988. On reconstructing tectonic plate motion from
Ladd, J.W., 1976. Relative motion of South America with respect
ship track crossing. J. Am. Stat. Assoc., 83:1178-1183.
to North America and Caribbean tectonics. Geol. Soc. Am.
Chang, T., Stock, J. and Molnar, P., 1990. The rotation group in
Bull., 87: 969-976.
plate tectonics and the representation of uncertainties of plate
Ladd, J.W., Truchan, M., Talwani, M., Stoffa, P.L., Buhl,
reconstructions. Geophys. J. Int., 101: 649-661.
P., Houtz, R., Mauffret, A. and Westbrook, G., 1984. In:
Collette, B.J., Slootweg, A.P., Verhoef, J. and Roest, W.R., 1984.
W.E. Bonini, R.B. Hargraves and R. Shagam (Editors), The
Geophysical investigations of the floor of the Atlantic Ocean
Caribbean-South America Plate Boundary and Regional Tec-
between 10~ and 38~ (Kroonvlag-project). Proc. Kon. Ned.
tonics. Geol. Soc. Am. Mem., 162: 153-160.
Akad. Wet., Ser. B, 87: 1-76.
Ladd, J.W., Holcombe, T.L., Westbrook, G.K. and Edgar, N.T.,
DeMets, C., Gordon, R.G., Argus, D.E and Stein, S., 1990.
1990. Caribbean marine geology; active margins of the
Current plate motions. Geophys. J. Int., 101: 425-478.
Dixon, T.H. and Mao, A., 1997. A GPS estimate of relative plate boundary. In: G. Dengo and J.E. Case (Editors), The
motion between North and South America. Geophys. Res. Caribbean Region. The Geology of North America, Geologi-
Lett., 24: 535-538. cal Society of America, Vol. H, Boulder, CO, pp. 261-290.
Dolan, J.E, Beck, Ch. and Ogawa, Y., 1989. Upslope deposition Mann, P., 1997. Model for the formation of large, transtensional
of extremely distal turbidites: an example from the Tiburon basins in zones of tectonic escape. Geology, 25:211-214.
Rise, west-central Atlantic. Geology, 17: 990-994. Mann, P. and Burke, K., 1984. Neotectonics of the Caribbean.
Dolan, J.E, Beck, Ch., Ogawa, Y. and Klaus, A., 1990. Eocene- Rev. Geophys. Space Phys., 22: 309-362.
Oligocene sedimentation in the Tiburon Rise/ODP Leg 110 Mann, P. and Corrigan, J., 1990. Model for late Neogene defor-
area: an example of significant upslope flow of distal turbidity mation in Panama. Geology, 18: 558-562.
currents. Proc. ODE Sci. Results, 110: 47-62. Mann, P., Schubert, C. and Burke, K., 1990. Review of Caribbean
Ego, E, Sebrier, M. and Yepes, H., 1995. Is the Cauca-Patia and neotectonics. In: G. Dengo and J.E. Case (Editors), The
Romeral fault system left or right lateral? Geophys. Res. Lett., Caribbean Region. The Geology of North America, Vol. H,
22: 33-36. Geological Society of America, Boulder, CO, pp. 307-338.
Elsasser, W.M., 1971. Sea-floor spreading as thermal convection. Mann, P., Taylor, E, Edwards, L. and Ku, T., 1995. Actively
J. Geophys. Res., 76:1101-1112. evolving microplate formation by oblique collision and side-
Engebretson, D., 1982. Model presentation for displacements be- ways motion along strike-slip faults: an example from the
tween western North America, eastern Eurasia, and adjacent northeastern Caribbean plate margin. Tectonophysics, 246: 1-
oceanic plates for the past 180 million years. In: Circum- 69.
Pacific energy and mineral resources conference, AAPG Bull. Maschenkov, S. and Pogrebitsky, Y., 1992. Preliminary results of
66: 966, Canary-Bahamas Project. Eos, Trans. Am. Geophys. Union,
Engebretson, D.C., Cox, A. and Gordon, R.G., 1985. Relative 73: 393-397.
motions between oceanic and continental plates in the Pacific Mascle, J., Blarez, E. and Marinho, M., 1988. The shallow
Basin. Geol. Soc. Am. Spec. Pap., 206, 59 pp. structures of the Guinea and Ivory Coast-Ghana transform
England, E and Wortel, R., 1980. Some consequences of the margins: their bearing on the Equatorial Atlantic Mesozoic
subduction of young slab. Earth Planet. Sci. Lett., 47: 403- evolution. Tectonophysics, 155: 193-210.
415. McKenzie, D., 1972. Active tectonics of the Mediterranean re-
Eva, A.N., Burke, K., Mann, E and Wadge, G., 1989. Four-phase gion. Geophys. J. R. Astron. Soc., 30: 109-185.
tectonostratigraphic development of the southern Caribbean. Meijer, P.T., 1992. The dynamics of motion of the South America
Mar. Pet. Geol., 6:9-21. plate. J. Geophys. Res., 97: 11,915-11,931.
Gripp, A.E. and Gordon, R.G., 1990. Current plate velocities Molnar, P. and Stock, J., 1987. Relative motions of hotspots in

N E W C O N S T R A I N T S ON T H E PLATE T E C T O N I C E V O L U T I O N OF T H E C A R I B B E A N 59
the Pacific, Atlantic and Indian Oceans since late Cretaceous Rosenfeld, J.H., 1990. Sedimentary rocks of the Santa Cruz
time. Nature, 327: 587-591. o p h i o l i t e - a proto-Caribbean history. Trans. 12th Caribbean
Mtiller, R.D. and Roest, W.R., 1992. Fracture zones in the North Geol. Conf., U.S. Virgin Islands, pp. 513-519.
Atlantic from combined Geosat and Seasat data. J. Geophys. Ross, M.I. and Scotese, C.R., 1988. A hierarchical tectonic
Res., 97: 3337-3350. model of the Gulf of Mexico and Caribbean region. Tectono-
Mfiller, R.D. and Smith, W.H.F., 1993. Deformation of the physics, 155: 139-168.
oceanic crust between the North America and South Amer- Royden, L.H., 1993. Evolution of retreating subduction bound-
ica plates. J. Geophys. Res., 93: 8275-8292. aries formed during continental collision. Tectonics, 12: 629-
Mtiller, R.D., Sandwell, D.T., Tucholke, B.E., Sclater, J.G. and 638.
Shaw, ER., 1991. Depth to basement and geoid expression of Royer, J.-Y. and Chang, T., 1991. Evidence for relative motions
the Kane Fracture Zone: a comparison. Mar. Geophys. Res., between the Indian and Australian plates during the last 20
13: 105-129. Myr from plate tectonic reconstructions: implications for the
Mfiller, R.D., Royer, J.-Y. and Lawver, L.A., 1993. Revised plate deformation of the Indo-Australian plate. J. Geophys. Res., 96:
motions relative to the hotspots from combined Atlantic and 11,779-11,802.
Indian Ocean hotspot tracks. Geology, 21: 275-278. Royer, J.-Y., Gordon, R.G., DeMets, C. and Vogt, ER., 1997.
Mfiller, R.D., Roest, W.R., Royer, J.-Y., Gahagan, L.M. and New limits on the motion between India and Australia since
Sclater, J.G., 1997. Digital isochrons of the world's ocean chron 5 (11 Ma) and implications for lithospheric deformation
floor. J. Geophys. Res., 102:3211-3214. in the equatorial Indian Ocean. Geophys. J. Int., 129: 41-74.
Perfit, M.R. and Heezen, B.C., 1978. The geology and evolution Sandwell, D.T. and Smith, W.H.E, 1997. Marine gravity anomaly
of the Cayman Trench. Geol. Soc. Am. Bull., 89:1155-1174. from Geosat and ERS-1 satellite altimetry. J. Geophys. Res.,
Pindell, J.L. and Barrett, S.F., 1990. Geological evolution of the 102:10,039-10,054.
Caribbean region; a plate tectonic perspective. In: G. Dengo Shaw, ER. and Cande, S.C., 1990. High-resolution inversion
and J.E. Case (Editors), The Caribbean Region. The Geology for South Atlantic plate kinematics using joint altimeter and
of North America, Geological Society of America, Vol. H, magnetic anomaly data. J. Geophys. Res., 95: 2625-2644.
Boulder, CO, pp. 405-432. Stdphan, J.F., Blanchet, R. and Mercier de L6pinay, B.,
Pindell, J.L. and Dewey, J.F., 1982. Permo-Triassic reconstruc- 1986. Northern and southern Caribbean festoons (Panama,
tion of western Pangea and the evolution of the Gulf of Colombian-Venezuela and Hispaniola-Puerto Rico), inter-
Mexico/Caribbean region. Tectonics, 1: 179-212. preted as pseudosubdivisions induced by the east-west short-
Pindell, J.L., Cande, S.C., Pitman, W.C., III, Rowley, D.B., ening of the peri-Caribbean continental frame. In: Wezel-
Dewey, J.F., LaBrecque, J. and Haxby, W., 1988. A plate-kine- Forese, C. (Ed.), The Origin of Arcs. Developments in Geo-
matic framework for models of Caribbean evolution. Tectono- tectonics, vol. 21, Elsevier, Amsterdam, pp. 401-422.
physics, 155: 121-138. Stdphan, J.F., Mercier De Ldpinay, B., Calais, E., Tardy,
PlaNer, G. and Ward, S.N., 1992. Thrust faulting and tectonic M., Beck, Ch., Carfantan, J.-Ch., Olivet, J.-L., Vila, J.-M.,
uplift along the April 1922, 1991 Costa Rica earthquake. Bouysse, Ph., Mauffret, A., Bourgois, J., Thery, J.-M.,
Tectonics, 11: 709-718. Tournon, J., Blanchet, R. and Dercourt, J., 1990. Paleogeo-
Roest, W.R. and Collette, B.J., 1986. The Fifteen Twenty Frac- dynamic maps of the Caribbean: 14 steps from Lias to Present.
ture Zone and the North American-South America plate Bull. Soc. G6ol. Fr., 8: 915-919.
boundary. J. Geol. Soc. London, 143: 833-843. Stock, J. and Molnar, E, 1983. Some geometrical aspects of
Rona, EA., 1980. The central North Atlantic Ocean Basin and uncertainties in combined plate reconstructions. Geology, 11:
continental margins: geology, geophysics, geochemistry, and 697-701.
resources, including the trans-Atlantic geotraverse (TAG). Na- Sykes, L.R., McCann, W.R. and Kafka, A.L., 1982. Motion
tional Oceanic and Atmospheric Administration/NOAA Atlas of Caribbean plate during last 7 million years and implica-
3. tions for earlier Cenozoic movements. J. Geophys. Res., 87:
Rosencrantz, E., 1990. Structure and tectonics of the Yucatan 10,656-10,676.
basin, Caribbean Sea, as determined from seismic reflection Tucholke, B.E. and Schouten, H., 1988. Kane fracture zone. Mar.
studies. Tectonics, 9:1037-1059. Geophys. Res., 10: 1-39.
Rosencrantz, E. and Mann, E, 1991. SeaMARC II mapping Van der Hilst, R. and Mann, E, 1994. Tectonic implications of
of transform faults in the Cayman trough, Caribbean Sea. tomographic images of subducted lithosphere beneath north-
Geology, 19:2141-2157. western South America. Geology, 22:451-454.
Rosencrantz, E., Ross, M.I. and Sclater, J.G., 1988. Age and Wadge, G. and Burke, K., 1983. Neogene Caribbean plate ro-
spreading history of the Cayman trough as determined from tation and associated tectonic evolution. Tectonics, 2: 633-
depth, heat flow, and magnetic anomalies. J. Geophys. Res., 643.
93: 2141-2157.

Chapter 3
Jurassic-Early Cretaceous Tectono-Paleogeographic Evolution of

the Southeastern Gulf of Mexico Basin
G Y O R G Y L. M A R T O N and R I C H A R D T. B U F F L E R
A new opening model for the Gulf of Mexico basin provides a framework in which the Jurassic-Early Cretaceous
tectono-paleogeographic evolution of the southeastern Gulf of Mexico and surrounding regions can be discussed. A detailed
analysis of available seismic data and the results of DSDP Leg 77 define four major tectono-stratigraphic sequences bounded by
major unconformity surfaces: crystalline basement, Paleozoic(?) pre-rift rocks, a Late Jurassic syn-rift sequence, and an Early
Cretaceous post-rift sequence. The pre-rift rocks are interpreted to represent a pre-Mesozoic (Late Paleozoic?) sedimentary cycle.
The Late Jurassic syn-rift sequence in the central continental domain of the southeastern Gulf of Mexico occurs in grabens or
half-grabens and is interpreted to consist of two units, a lower non-marine unit overlain by a marine carbonate unit consisting
of carbonate buildups (platforms) and adjacent deeper marine sediments. Jurassic rocks are absent over high-standing blocks as
well as the adjacent Yucatfin and Florida blocks. The Lower Cretaceous post-rift sequence drapes the entire area and consists
of deep-water carbonate sediments in the central basin flanked by shallow-water platforms on the adjacent Yucatan and Florida
blocks.
Six tectono-paleogeographic maps covering the eastern Gulf of Mexico and northwestern Cuba (palinspastically restored to the
southeastern margin of Yucatan) document the evolution of the area. During the late Middle Jurassic (Callovian) the southeastern
Gulf was a bridge between Yucatan and Florida, separating an area of widespread extension and salt deposition to the north
in the Gulf of Mexico from another area of extension and clastic sedimentation to the south between Yucatan and northern
South America. By Oxfordian time Yucatfin had rotated 11~ counter-clockwise and major continental rifting and non-marine
sedimentation in rift basins had begun all along the southeastern Gulf. To the north salt deposition had ceased and a major marine
transgression culminated in deposition of Smackover carbonates. South of Yucatan shallow-water carbonate sedimentation also
prevailed. During Kimmeridgian, Tithonian and into earliest Cretaceous time, rifting continued in the southeastern Gulf as Yucatan
continued to rotate counter-clockwise. As the basin subsided a marine seaway, characterized by shallow-water carbonate platforms
on high-standing blocks, became established, connecting the Gulf of Mexico with the proto-Caribbean. In the northeastern Gulf
the mixed clastic/carbonate Haynesville and Cotton Valley sequences were deposited, while to the south of Yucatan shallow-water
carbonate sedimentation gave way to deeper water sedimentation as the margin subsided. In late Berriasian spreading ceased in the
Gulf of Mexico, Yucatan reached its present-day position, and rifting stopped in the southeastern Gulf. Carbonate platforms atop
rift blocks drowned as the basin subsided and sea level rose, and the southeastern Gulf became the deep-water seaway that it is
today. The marine transgression reached the Yucatan and Florida blocks, where extensive carbonate platforms became established
and flourished throughout the Early Cretaceous. To the south of Yucatan, deep-water pelagic sedimentation continued throughout
the Early Cretaceous.
INTRODUCTION It was during the L a t e Jurassic rift stage that the

m a i n tectonic e l e m e n t s f o r m e d , w h i c h c o n t r o l l e d
T h e s o u t h e a s t e r n G u l f of M e x i c o (Fig. 1) has s u b s e q u e n t C r e t a c e o u s and C e n o z o i c s e d i m e n t a t i o n
b e e n a s e a w a y c o n n e c t i n g the G u l f of M e x i c o with patterns and p a l e o c e a n o g r a p h i c conditions.
the C a r i b b e a n since the L a t e Jurassic. It has had T h e p r i m a r y p u r p o s e of this p a p e r is to s u m m a -
a long and c o m p l e x g e o l o g i c a l history b e g i n n i n g rize the Jurassic t h r o u g h E a r l y C r e t a c e o u s tectono-
with a Jurassic t h r o u g h E a r l y C r e t a c e o u s passive p a l e o g e o g r a p h i c e v o l u t i o n of the s o u t h e a s t e r n G u l f
m a r g i n e v o l u t i o n that can be tied directly to the of M e x i c o b a s e d on an analysis of r e g i o n a l m u l t i f o l d
b r e a k u p of w e s t e r n P a n g e a . This early e v o l u t i o n is s e i s m i c data and D S D P drilling results. T h e s e data
c h a r a c t e r i z e d by pre-rift, rift, and post-rift stages. h a v e b e e n u s e d to define four m a j o r tectono-strati-

64 G.L. MARTON and R.T. BUFFLER
graphic sequences bounded by major unconformity PHYSIOGRAPHIC/GEOLOGIC SETTING

surfaces: crystalline basement, Paleozoic(?) pre-rift
rocks, a Late Jurassic rift sequence, and an Early The southeastern Gulf of Mexico is a deep sea-
Cretaceous post-rift sequence. Emphasis here will be way at the intersection of the northern Yucat~in
on the Late Jurassic rift sequence, as new interpre- Straits and the western Straits of Florida (Fig. 1)
tations of these rocks are presented. This important (Marton and Buffler, 1994; Marton, 1995). It lies
Late Jurassic sequence has no tectonic equivalent just north of Cuba in between the steep Campeche
around the Gulf of Mexico and Caribbean basin. and Florida escarpments, which reflect the paleoto-
The Early Cretaceous rocks have been the subject of pography of Lower Cretaceous carbonate margins.
earlier studies, and they will be discussed as appro- The northern part of the area is extremely fiat, an ex-
priate to complete the picture. The evolution of the pression of the distal Mississippi Fan turbidite plain
southeastern Gulf of Mexico area is summarized us- (Florida Plain). Most of this flat area is underlain
ing a set of regional tectono-paleogeographic maps by Mesozoic oceanic crust (Marton, 1995). To the
that includes the eastern Gulf of Mexico as well as south the seafloor rises above the turbidite plain to
northwestern Cuba. a broad area dissected by several prominent, north-
The details of this history, however, cannot be west-trending erosional channels (Fig. 1). The major
told without first considering the regional tectonic channel bifurcates to the south, with one arm extend-
setting and the tectonic events that have influenced ing into the Yucat~in Straits and one arm extending
the region. A revised tectonic model for the Gulf into the Straits of Florida.
of Mexico basin as well as the southeastern Gulf Seismic data provide evidence that this relatively
has been set forth in earlier works by the au- higher standing area is underlain by more or less
thors, and it is reviewed briefly herein. The reader extended continental crust (Schlager et al., 1984;
is referred to these references for additional back- Marton, 1995). Cenozoic and Upper Cretaceous sed-
ground and details (Marton and Buffler, 1994; Mar- iments are thin or absent over the central part of
ton, 1995). this platform. This overall shallow bathymetric high
Fig. 1. Map of the southeastern Gulf of Mexico study area showing major physiographic features. Also shown are the tectonostratigraphic
provinces in the Cordillera de Guaniguanico of western Cuba (1 = Sierra de los Organos; 2 = Sierra del Rosario meridional; 3 = Sierra
del Rosario septentrional). Water depth in meters.

EVOLUTION OF THE SOUTHEASTERN GULF OF MEXICO BASIN 65
and the linear northwest-trending features along the ton, 1995). Late Jurassic rifling characterized by
northern margin are believed to be an expression of generally east-west extension occurred contempo-
the underlying shallow basement configuration and raneously with seafloor spreading and ocean crust
Jurassic tectonic elements of the region (Schlager formation to the north in the central Gulf of Mexico.
et al., 1984; Marton, 1995). The prominent reen- When oceanic crust formation stopped, i.e., Yucatan
trant in the Campeche Escarpment (Catoche Tongue) reached its present-day position, rifting in the south-
also reflects an underlying Jurassic graben structure eastern Gulf of Mexico between Yucatan and Florida
(Shaub, 1983). The steep margin along the north also ceased. This important event is marked by a
coast of Cuba is the result of the collision between a prominent post-rift unconformity in the southeastern
Cretaceous volcanic arc system and North America, Gulf of Mexico. Dating of this unconformity, based
which culminated in Early to Middle Eocene time on the results of DSDP drilling and interpretation
(Angstadt, 1983; Angstadt et al., 1985). Conspicu- of seismic data, is established as earliest Cretaceous.
ous in the area are several large knolls that probably The regional-scale model, therefore, is vital for ex-
represent continental basement highs covered with plaining the tectonic processes responsible for the
thin sediments (Catoche Knoll) or basement highs formation of the southeastern Gulf of Mexico, and
capped with Lower Cretaceous platforms or atolls it provides the tectonic framework for discussing the
(Jordan and Pinar del Rfo) (Bryant et al., 1969; paleogeographic evolution of the region.
Schlager et al., 1984; Marton, 1995) (Fig. 1).
REFLECTION SEISMIC DATA

REGIONAL TECTONIC FRAMEWORK
The seismic data for the regional study of the
The authors' view of the Jurassic development entire eastern Gulf of Mexico consist of 6380 km of
of the Gulf of Mexico basin has been discussed at 12-24 fold marine seismic lines collected between
length elsewhere (Marton and Buffler, 1994; Mar- 1977 and 1983 (Marton, 1995). A map showing
ton, 1995). The preferred opening model is a new these data is presented here as Fig. 3. A more
two-stage model constrained by a refined definition detailed map showing the subset of seismic data
of oceanic crust and by the known kinematic frame- used in the southeastern Gulf plus the location of
work of the large continental blocks surrounding the figures is included in Fig. 4. The earliest data set
Gulf of Mexico basin (North American plate and was collected in 1977-78 and was designed to
Afro-South American plate) (Fig. 2). During the explore the Catoche Tongue as well as produce a
Late Triassic-Early Jurassic(?) to late Middle Juras- regional overview of the southeastern and eastern
sic (Callovian) syn-rift stage, the relatively stable Yu- Gulf (Catoche Grid (CATG) and Gulf Tectonics
catan block translated southeastward along a major (GT) lines; Fig. 3). The second period of data
transform zone in eastern Mexico (Fig. 2A,B). This acquisition during the first half of 1980 resulted
motion accommodated a large amount of extension in in a higher-quality data set and provided better
the area of the future northern Gulf. At the same time resolution (Straits of Florida (SF) lines; Fig. 3).
the Florida-Bahamas block extended also in a south- These lines were designed to collect a comparatively
east direction to form a series of basins and arches, dense grid over the central part of the area to prepare
partly accommodated by a postulated major shear the drilling sites for DSDP Leg 77, and after the
zone, the Bahamas Fracture Zone (BFZ) (Fig. 2A,B). drilling, to provide the data base for further seismic
A rotation pole for the Yucatan block in the stratigraphic studies and to extrapolate regionally
southeastern Gulf of Mexico (23.18~ 84.24~ is the drilling results (Fig. 4). In a third period of
proposed for the late Middle Jurassic (Callovian) data acquisition in 1983 the Mississippi Canyon
to earliest Cretaceous (Berriasian) drifting stage (MC) lines were collected, while a fourth set of
(Fig. 2B-D). Around this pole the Yucatan block data (GULFREX-Gulf) were provided by Chevron
rotated about 42 ~ counter-clockwise out from the Corporation as paper copies only. The seismic data
northern Gulf to accommodate the newly formed coverage is uneven, with the highest concentration
oceanic crust in the basin. being in the southeastern Gulf (Figs. 3 and 4).
The implications of this regional-scale model bear The University of Texas seismic data base was
directly on the Jurassic evolution of the southeast- originally processed at the University of Texas In-
ern part of the basin. The inferred opening motion stitute for Geophysics (then Marine Science Institute)
resulted in a southward propagating rift/spreading immediately following the acquisition of the data, and
center in the eastern Gulf of Mexico (Fig. 2). Just it is available on field tapes (in demultiplexed form),
north of the rotation pole, a tectonic setting oc- on final stack tapes and on films at high vertical exag-
curred, which has no counterpart elsewhere in the geration. In this form the data were used to study the
Gulf of Mexico (Marton and Buffler, 1994; Mar- Cretaceous and younger sediments, which occur in


E V O L U T I O N OF THE S O U T H E A S T E R N G U L F OF M E X I C O BASIN 67
Fig. 3. Location of available seismic data in the eastern Gulf of Mexico. M C = Mississippi Canyon; SF = Straits of Florida; G T =
Gulf tectonics; CATG -- Catoche Grid; Gulfrex = Chevron data. Also shown are the location of two studies used for construction of
tectono-paleogeographic maps presented in Fig. 14 (Dobson, 1990; DeBalko, 1991).
the s h a l l o w e r , t e c t o n i c a l l y u n d i s t u r b e d p a r t o f the T w o d i f f e r e n t a p p r o a c h e s w e r e u s e d as p a r t o f
s e i s m i c s e c t i o n s . H o w e v e r , d u e to the i n c r e a s i n g o v e r - this s t u d y to i m p r o v e the q u a l i t y o f the a v a i l a b l e
b u r d e n a n d t e c t o n i c c o m p l e x i t i e s , the o l d e r s e c t i o n s e i s m i c d a t a a n d to i m p r o v e r e s o l u t i o n at d e p t h : (A)
( J u r a s s i c a n d b a s e m e n t ) is n o t s u f f i c i e n t l y r e s o l v e d . a c o m p l e t e r e p r o c e s s i n g o f 1070 k m o f s e i s m i c d a t a
Fig. 2. Four-stage evolution of the Gulf of Mexico basin and adjacent areas: (A) Early Jurassic; (B) 166 Ma (Callovian)" (C) 160 Ma
(Oxfordian); (D) 140 Ma (Berriasian) (see text for discussion, and Marton and Buffler, 1994 and Marton, 1995, for more details). A =
Africa; A F B = Appalachian foldbelt; B F -- La Babia fault; B F Z = Bahamas fracture zone; BP = Blake plateau; CA -- Central Atlantic;
DP = Demarara plateau; G M -- Gulf of Mexico; GP = Guyana plateau; L U = Llano uplift; M F B -- Marathon folded belt; M g r =
Mid-Gulf ridge; M S M = Mojave-Sonora megashear; OFB = Ouachita foldbelt; P C = proto-Caribbean; SA = South America; S M F --
San Marcos fault; T M V = Trans-Mexican volcanic belt; URB -- undifferentiated rift basins; YUC -- Yucatan block; W M T = western
main transform. Light and dark stippled areas in northern Gulf are large-scale highs and basins, respectively. Solid lines are major faults
and foldbelts. Dashed lines outline areas of Late Triassic-Early Jurassic sediments and volcanics. Hachured pattern outlines basement
terranes in Mexico. Gray line outlines distribution of late Middle Jurassic salt (Louann and equivalent). P in southeastern Gulf is rotation
pole for the Yucatan block with degrees of rotation.

Fig. 4. Seismic grid used in the study of the southeastern Gulf of Mexico, including locations of Figs. 5 (SF-15), 6 (GT3-75), 7 (SF-2),
8 (SF-6), 9 (SF-7), 10 (SF-13), 11 (SF-17), and 12 (SF-11), and DSDP Leg 77 drill sites 535, 536, 537, 538A, and 540. P is location of
pole of rotation for Late Jurassic opening of the Gulf of Mexico basin (after Marton and Buffler, 1994; Marton, 1995).
that represent key NW-SE- and NE-SW-oriented The results at each site are discussed in detail in
sections and form a comprehensive grid over the Buffler et al. (1984) and are described briefly below,
southeastern Gulf, and (B) time migration of 3200 with emphasis on observations directly bearing on
km of data, the remaining stacked seismic data in the early Mesozoic evolution of the area.
the southeastern Gulf. This reprocessed data set
includes all of the CATG, GT and SF lines south of Sites 535/540 (basin sites)
approximately 25~ (Figs. 3 and 4). The remaining
2100 km of seismic data (MC lines and GT lines Sites 535 and 540 were drilled to sample the thick
north of approximately 25~ were not reprocessed Mesozoic and Cenozoic basin fill sediments in the
or migrated, but they also were used in interpretation eastern part of the study area (Fig. 4). The combined
and mapping. holes at these two sites sampled an almost complete
Lower Cretaceous deep-water carbonate section, pro-
viding the first view of Early Cretaceous sedimenta-
DSDP LEG 77 RESULTS tion in the deep part of the Gulf of Mexico basin.
In December/January 1980-1981, the R/V Glo- Site 535

mar Challenger drilled at five primary sites in the Site 535 bottomed in Berriasian age rocks, leav-
southeastern Gulf of Mexico (Fig. 4) (Sites 535, ing unsampled the thick sedimentary section below
536, 537, 538, 540: Buffler et al., 1984). The lo- observed in the seismic data (Fig. 5). This unsam-
cations of the sites were based on an analysis of pied sedimentary section is regionally extensive and
the seismic data discussed above. Holes 536, 537, is interpreted to comprise the Jurassic rocks (J)
538A were drilled on high-standing basement blocks that form the major focus of this study (Schlager
(basement sites), while Sites 535 and 540 were de- et al., 1984; Marton, 1995). At the bottom of Site
signed to sample the Mesozoic-Cenozoic basin fill 535, 100 m of condensed upper Berriasian to lower
(basin sites). The geologic and tectonic setting of Valanginian alternations of massive bioturbated to
each of these sites are shown schematically on three poorly laminated limestone and marly limestone
cross sections based on seismic lines (Figs. 4-7). with numerous hardgrounds were recovered. They

were interpreted to be deposited in at least bathyal dolomitic marls, muddy dolomites, and arkosic sand
depths. The lower 25 m section (upper Berriasian) represent a transition to marine conditions. Brack-
is characterized by a low (5%) porosity, a high (4.7 ish-water ostracods, plant and coal fragments in
km/s) sonic velocity, and a 2.57 g/cm 3 bulk density this unit were interpreted as deposition in a shal-
(Buffer et al., 1984). This is in contrast to the 17% low-marine or estuarine environment. The lower to
porosity, 3.4 km/s sonic velocity, and 2.43 g/cm 3 middle Berriasian transgressive sequence is followed
density measured in the overlying Valanginian lime- by a 58 m upper Berriasian to Valanginian skele-
stones. The abrupt change in acoustic properties tal limestone, grainstone, and wackestone sequence
results in a high-continuity, high-amplitude reflector, with abundant shallow-water fragments. These rocks
which is correlatable throughout most of the mapped were interpreted as in-situ deposits, related to a deep
area. Because of the limited resolution of the seismic carbonate platform. Alternatively, these sediments
(30-60 m at best) the horizon here is referred to as may had been reworked from a shallow-water set-
'top Berriasian' or TB (Fig. 5). It occurs approxi- ting and redeposited in a deep-water environment
mately at the Jurassic-Cretaceous boundary, and it (Buffler et al., 1984; Schlager et al., 1984). This
is used as the main mapping horizon in this study. sequence is capped by a thin sequence of Cretaceous
An overlying unit in Hole 535 consists of about (Aptian) to Cenozoic (Lower Pliocene) nannofossil
460 m of lower Valanginian to Cenomanian(?) lam- ooze with interbedded chert, mud, and volcanic ash,
inated, variably bioturbated, frequently organic-rich representing a typical pelagic environment (Fig. 6).
limestones and marly limestones, with increasing Based on the recovered abbreviated section at Site
sand-size, shallow-water skeletal debris towards the 537, it can be concluded that higher standing parts
top of the sequence (Fig. 5). of the Yucatan terrace, and probably the eastern Yu-
catfin block, were not transgressed until the very end
Site 540 of the Jurassic-earliest Cretaceous. Fault movement
Site 540 was drilled in a higher physiographic separating this block from the sediment source areas
setting (Fig. 5) and penetrated 460 m of Cre- of the Berriasian fluvial deposit, cannot be older
taceous (middle Albian to Cenomanian section), than late Berriasian to early Valanginian. Ongoing
mostly pelagic sediments with fine-grained input and tectonic activity (rifting) at this time is compatible
skeletal debris from the adjacent shallow platform with a model which suggests that in the southeastern
to the east (Florida Escarpment). Truncating the Gulf of Mexico, rifting occurred contemporaneously
Lower Cretaceous section is the widespread Middle with Late Jurassic-earliest Cretaceous oceanic crust
Cretaceous Sequence Boundary (MCSB), which at formation in the central Gulf of Mexico.
Site 540 is characterized by a thin Late Cretaceous
(Turonian to lower Paleocene) condensed interval Site 536
consisting of carbonate debris flows and turbidites. Site 536 was drilled at the base of the Campeche
Escarpment on the northern end of the Yucatfin ter-
Sites 537/536/538 (basement sites) race (Figs. 4 and 6). At the base of the hole, a
25-m-thick dolomite sequence of unknown age was
Site 537 drilled. The depositional setting of this section was
Site was drilled on the crest of a small knoll, established as a shallow-water platform, based on
about 25 km north of the Campeche Escarpment at observed algal laminations and possible desiccation
the entrance of the Catoche Tongue (Figs. 4 and 6). cracks. This dolomite is devoid of diagnostic fossils
At total depth 19 m of phyllite was recovered with and is characterized by a very tight fabric (1% poros-
an age of about 500 Ma (Pan-African), based on ity) and a very high sonic velocity (over 6 kin/s).
4~ measurements (Dallmeyer, 1984). Above A 87Sr/86Sr age for this rock can be interpreted
the metamorphic basement, quartz porphyry frag- as either Middle Jurassic-Early Cretaceous or late
ments were cored, indicating acid or silicic igneous Paleozoic (Permian) (Testarmata and Gose, 1984).
activity in the area prior to and perhaps during However, the accompanying paleomagnetic study
early Berriasian time. These fragments were inter- from the same work supported the Middle Jurassic-
preted to represent an in-situ igneous body (sill or Early Cretaceous age (Testarmata and Gose, 1984).
dike), or clasts from a nearby source (Buffler et al., Based on the physical characteristics of the rocks
1984). Above the porphyry is a 50-m-thick, early- and their tectonic setting, an interpretation that the
middle Berriasian transgressive sequence, starting enigmatic dolomite is pre-Middle Jurassic (possibly
with coarse-grained, poorly sorted arkosic sandstone, late Paleozoic) in age is preferred here. Seismic
which was interpreted to be deposited in a fluvial- evidence suggest that the Upper Jurassic section
alluvial or arid estuarine setting (Buffler et al., 1984). pinches out and is not present on the Yucatfin ter-
Thin bentonite beds represent contemporaneous vol- race (Figs. 6 and 8), which eliminates the Upper
canism. The overlying unit, consisting of mixed Jurassic-Early Cretaceous ages. Marine post-Per-

O

9
;Z
Fig. 5. Location of DSDP Sites 535 and 540 along seismic line SF-15. See Fig. 4 for location. (A) Uninterpreted seismic line. (B) Depth converted line drawing interpretation of seismic line with
simplified stratigraphic columns of Sites 535 and 540. J -- Jurassic; TB -- top Berriasian post-rift unconformity; L K = Lower Cretaceous" M C S B = Mid-Cretaceous sequence boundary; U K - C = Upper
Cretaceous-Cenozoic" VE = vertical exaggeration. (C) Stratigraphic summary of earliest Cretaceous rocks at DSDP Site 535 showing location of TB. Env. = depositional environment. Modified from
Buffler et al. (1984).
7~
,<
O
0
Z
0
~H
Ul
0
~H
U:
>.
,.H
2:
D
0
>.
r.~
:Z

Fig. 6. Location of DSDP Sites 536 and 537 along seismic line GT3-75C. See Fig. 4 for location. (A) Uninterpreted seismic line. (B) Depth converted line drawing interpretation of seismic line with
=
simplified stratigraphic columns of Sites 536 and 537. J Jurassic" TB -- top Berriasian post-rift unconformity; L K -- Lower Cretaceous; M C S B = Mid-Cretaceous sequence boundary; U K - C = Upper
Cretaceous-Cenozoic. (C) Stratigraphic summary of earliest Cretaceous rocks at DSDP Sites 536 and 537 overlying basement rocks. Env. = depositional environment. Modified from Buffler et al. (1993).
--.3
t',3

9
~Z
7z
Fig. 7. Location of DSDP Site 538 along seismic line SF-2. See Fig. 4 for location. (A) Uninterpreted seismic line. (B) Depth converted line drawing interpretation of seismic line with simplified
stratigraphic columns of Site 538. J = Jurassic; TB -- top Berriasian post-rift unconformity; L K = Lower Cretaceous; M C S B -- Mid-Cretaceous sequence boundary; U K - C = Upper Cretaceous-Cenozoic.
(C) Stratigraphic summary of earliest Cretaceous rocks at DSDP Site 538 overlying basement rocks. Env. = depositional environment. Modified from Buffler et al. (1993).
7z
Fig. 8. Seismic line SF-6B along northern edge of Yucatfin terrace. See Fig. 4 for location. (A) Uninterpreted. (B) Interpreted. J =
Jurassic; T B = top Berriasian post-rift unconformity; L K = Lower Cretaceous; M C S B ---=Mid-Cretaceous sequence boundary; U K - C =
Upper Cretaceous-Cenozoic. Vertical exaggeration about 10 at sea bottom.
mian to pre-uppermost Middle Jurassic sediments have been interpreted from seismic as a pre-rift sec-
are unknown from the Gulf of Mexico proper (Sal- tion in the basins and sometimes on the top of tilted
vador, 1991), also making it improbable that the age blocks (Fig. 6), indicating that deposition of this unit
of the dolomite is Early or Middle Jurassic. predates faulting in the southeastern Gulf of Mexico.
The possible equivalent of this early Mesozo- The dolomite sequence is overlain by 108 m of
ic(?)/Paleozoic(?) unit extends southward beneath Aptian to Albian porous, coarse-grained, skeletal,
the Yucatfin terrace and may attain several kilometers lime-packstone-grainstone with neritic material and
in thickness (Figs. 6 and 8). Equivalent rocks in the fine-grained radiolarian intercalations (Fig. 6). This
deep central domain (e.g., near Site 535/540, Fig. 5) Lower Cretaceous sequence was interpreted as talus

deposits at the foot of the Campeche Bank(Fig. 4). quences and tying them into the drill sites. Angstadt
In this association the shallow-water component rep- (1983), Angstadt et al. (1983) and (1985) conducted
resents redeposited platform sediments shed into a a detailed study of the overlying Late Cretaceous-
deep-water environment, and the fine-grained com- Cenozoic rocks, mapped four main units, and docu-
ponent represents ongoing pelagic sedimentation. mented development of an early Cenozoic foredeep
along the north flank of the Cuban orogen, followed
Site 5 3 8 a by current-controlled late Cenozoic pelagic sedimen-
Site 538A was drilled on the top of a large tation in the western Straits of Florida. Other seismic
knoll (Catoche Knoll) that rises high above the studies in the area include a description of the Catoche
Florida Plain (Figs. 4 and 7). In Hole 538A at to- Tongue (Shaub, 1983).
tal depth, 64 m of basement rocks were recovered, The studies by Angstadt (1983) and Phair (1984)
which consist of mylonitic gneiss and amphibolite. were incorporated into a paper included in the DSDP
4~ measurements again suggest a 500 Ma Leg 77 volume that made a preliminary interpreta-
(Pan-African) age (Dallmeyer, 1984). These crys- tion of the seismic stratigraphy, structural setting,
talline rocks are intruded by several generations and geologic history of the entire area (Schlager
of diabase dikes with 190-160 Ma (Lower-Middle et al., 1984). Schlager et al. (1984) established a
Jurassic) 4~ intrusion ages. This rock as- seismic stratigraphic framework, which included an
semblage was interpreted as rifted continental crust inferred Late Triassic-Early Jurassic unit (TJ), two
or 'transitional' crust (Dallmeyer, 1984; Schlager inferred Jurassic units (J1 and J2), as well as four
et al., 1984), an interpretation strongly supported Early Cretaceous units (EK 1-4).
by the general high-standing, block-faulted tectonic
setting (Marton, 1995) (Fig. 7). Basement is cov-
ered by 67 m of Early Cretaceous (lower(?)-upper TECTONO-STRATIGRAPHIC FRAMEWORK
Berriasian to Valanginian) skeletal-oolitic-oncolitic
limestones. There is a gradational change between In the present study of the southeastern Gulf
the more oncolitic (deeper-water) facies at the base of Mexico, the main purpose of seismic interpre-
towards the more oolitic (shallow-water) facies at tation is to delineate the distribution and reveal
the top. Based on similarities between Sites 537 the tectonic and depositional setting of the Jurassic
and 538A, Schlager et al. (1984) suggested that the sediments. The Lower Cretaceous sequences were
limestones may have been deposited in-place on an studied earlier, and they are only discussed briefly to
extensive carbonate platform, rather than on an iso- complete the picture. Major bounding surfaces (un-
lated block. Alternatively, it is possible that gradual conformities) previously defined in earlier studies
shallowing of this earliest Cretaceous section records have been re-correlated, including the Mid-Creta-
uplift(?) of the block (Catoche Knoll) from 50-200 ceous Sequence Boundary (MCSB, formerly MCU)
m to as little as a few meters (Marton, 1995). Such and the late (top) Berriasian horizon (here desig-
uplift could be explained by fault-block rotation nated TB) (Marton, 1995). In addition, two other
(footwall uplift) during tectonic activity (rifting) in horizons, the top of the pre-rift section and the top of
the southeastern Gulf of Mexico during the earliest acoustic basement (crystalline basement), have been
Cretaceous final phase of rifting and seafloor spread- defined (Marton, 1995). All four of these surfaces
ing. This latter interpretation would imply deposi- have been interpreted where possible and mapped
tion on an already isolated block. The Berriasian to across the entire eastern Gulf of Mexico as well as
Valanginian limestones are capped by 210 m of A1- in the southeastern Gulf (Marton, 1995).
bian to Upper Pliocene pelagic sediments, including These four regional unconformity surfaces have
chalk, foraminiferal-nannofossil ooze, radiolarian been used to define four major tectono-stratigraphic
mudstone, and chert, indicating low sedimentation sequences in the study area: (1) crystalline basement
rates, frequent non-deposition/erosion on the finally (rifted continental crust and oceanic crust); (2) Pa-
submerged basement high (Fig. 7). leozoic(?) pre-rift rocks; (3) a Late Jurassic syn-rift
sequence (J); and (4) a Lower Cretaceous post-rift
sequence (LK). Each of these units are shown on the
PREVIOUS SEISMIC STRATIGRAPHIC STUDIES regional sections presented above (Figs. 5-8). The
seismic character and additional details and interpre-
Following Leg 77, several seismic stratigraphic tations of each unit, along with their relationships to
studies were undertaken in the study area, designed tectonic development of the south-central part of the
to integrate the results of the drilling with exist- study area, are illustrated using four additional seis-
ing seismic data. Phair and Buffler (1983) and Phair mic lines oriented east-west, starting in the north and
(1984) described in detail the Lower Cretaceous sec- proceeding to the south (Figs. 9-12). The location of
tion by subdividing it into four main depositional se- these lines is shown in Fig. 4 and also on a structure

Fig. 9. Uninterpreted and interpreted seismic line SF-7. J = Jurassic; TB = top Berriasian post-rift unconformity; LK = Lower
Cretaceous; MCSB = Mid-Cretaceous sequence boundary; U K - C = Upper Cretaceous-Cenozoic. Vertical exaggeration about 10 at sea
bottom. See Fig. 4 for location.
map of the pre-Jurassic surfaces (top pre-rift or top depositional style change in a north to south direction,
acoustic (crystalline) basement) in this area (Fig. 13). (2) major rift-related features, such as half-grabens,
These sections and the map show (1) how faulting and grabens, major faults, fault blocks, and possible ac-

Fig. 10. Uninterpreted and interpreted seismic line SF-13. J - Jurassic; T B -- top Berriasian post-rift unconformity; L K - - Lower
Cretaceous; M C S B -- Mid-Cretaceous sequence boundary; U K - C = Upper Cretaceous-Cenozoic. Vertical exaggeration about 10 at sea
commodation zones, as well as (3) major depositional Crystalline basement

features, such as non-marine alluvial fans/fan deltas,
deep marine beds and carbonate buildups. Each of the Rifted continental crystalline basement (transi-
four tectono-stratigraphic sequences and their bound- tional crust), probably composed mainly of crys-
ing surfaces are discussed in more detail below. talline metamorphic rocks intruded by mafic rocks,

Fig. l l. Uninterpreted and interpreted seismic line SF-17. J = Jurassic; TB = top Berriasian post-rift unconformity; LK = Lower
Cretaceous; MCSB = Mid-Cretaceous sequence boundary; U K - C -- Upper Cretaceous-Cenozoic. Vertical exaggeration about 10 at sea
as drilled at Sites 537 and 538, is inferred to under- based on gravity, magnetic, and seismic data (Hall
lie the entire higher-standing central and south- and Najmuddin, 1994; Marton and Buffler, 1994;
ern parts of the study area (Schlager et al., Marton, 1995).
1984; Marton and Buffer, 1994; Marton, 1995) The top of crystalline basement throughout the
(Figs. 6 and 7). To the north beneath the Florida study area is interpreted to be the top acoustic
Plain, basement is believed to be oceanic crust basement on the seismic data. Acoustic basement
(Marton, 1995) (Fig. 6). The boundary between is characterized by chaotic, incoherent reflections,
oceanic and continental crust has been well defined which distinguishes it from the overlying sedimen-

Fig. 12. Uninterpreted and interpreted seismic line SF-11. J

- Jurassic; TB : top Berriasian post-rift unconformity; LK -- Lower
-
Cretaceous; MCSB -- Mid-Cretaceous sequence boundary; U K - C = Upper Cretaceous-Cenozoic. Vertical exaggeration about 10 at sea
tary sections characterized by more organized reflec- Paleozoic(?) pre-rift rocks

tions (Figs. 5-12). Most of the seismic lines show
good examples of this boundary. However, the exact The pre-rift rocks are defined on the seismic
location of this boundary on some of the seismic data as the section overlying acoustic basement
data is not well imaged, due to greater depth and/or and underlying interpreted Late Jurassic syn-rift
lack of acoustic impedance change along this bound- rocks filling grabens and half-grabens, or in some
ary. The boundaries illustrated on all the seismic cases younger Lower Cretaceous rocks filling sag
examples shown here, therefore, represent only the basins where the Jurassic syn-rift section is missing
best interpretations possible and are not unequivocal (Figs. 5-12). The top of this pre-rift section (or
everywhere. the top of basement when this section is missing),

Fig. 13. Structure map of the pre-Jurassic surfaces (top Paleozoic(?) pre-rift section or top acoustic (crystalline) basement) (pre-rift
surface) in the southeastern Gulf of Mexico, showing: (a) distribution of Upper Jurassic carbonate platforms (PL 1-3), (b) major faults
(heavy black lines), and (c) basement highs and lows. Graben morphologyin the central part of the area is shown for lines corresponding
to Figs. 5 (SF-15), 9 (SF-7), 10 (SF-13), 11 (SF-17), 12 (SF-11). Contours are in meters. White areas indicate where Jurassic inferred to
be thin or absent. HG = half grabens; AZ = possible accommodationzone.
which also defines the base of the syn-rift section, is the block between CDP's 1300 and 1400 on Line SF
defined here as the pre-rift surface. A structure map 7 (Fig. 9), suggesting long-term uplift and erosion
of this surface in the central study area shows the of the block. The pre-rift section on Line SF-13
regional rift morphology (Fig. 13). (Fig. 10) has been offset along a large west-dipping
The pre-rift section is characterized by organized normal fault bounding a half-graben on the east.
but generally low-amplitude reflections. It is inter- On Line SF-17 (Fig. 11) it is down-dropped into a
preted here to be an extensive sedimentary unit that full graben. The section is thinner and truncated on
occurs throughout most of the area with a fairly uni- the eastern up-thrown block, possibly due to erosion
form thickness (Figs. 5-12). In half-grabens (e.g., during or following rifting.
Line SF 7, Fig. 9) the pre-rift section is observed Beneath the shallow Yucatan terrace area to the
as uniform packages of sub-parallel reflections on southwest where the interpreted Jurassic rocks are
top of rotated fault blocks. Here this unit is distinct absent (Fig. 13), the pre-rift section is flat-lying and
from the reflection-free basement below and from is very thick (estimated over 2 km) (Figs. 6, 8 and
the more diverging and onlapping Late Jurassic syn- 12). Along the northern edge of the terrace, the
rift section above. These relationships indicate that unit is composed of two separate sub-units separated
these rocks have rotated with the underlying fault by a prominent angular unconformity, indicating a
blocks and were deposited prior to the Late Jurassic complex tectonic history (Fig. 6). On Line SF-11
rifting. In some places, the unit apparently is absent (Fig. 12) the thick pre-rift section along the eastern
over high-standing fault blocks. This can be seen on edge of the Yucatfin terrace is down-dropped to the

east along a half-graben, while the western margin The Late Jurassic syn-rift section in the sub-
shows a series of smaller antithetic faults. basins often can be divided into two units based on a
As argued above under the discussion of Site distinctive vertical change in internal reflection char-
536, a tentative correlation of these rocks with the acter or seismic facies. These units are interpreted to
dolomite drilled at the bottom of the hole (Fig. 6) be a non-marine unit below and a marine unit above.
suggests a possible late Paleozoic age for these
pre-rift rocks. The nearest Paleozoic sedimentary Nonmarine(?) unit
rocks of late Paleozoic age occur only in outcrop The lower unit occurs in a stratigraphically deeper
and in the subsurface to the south in Belize and position lying directly above the pre-rift surface. It
Guatemala (Santa Rosa group) (e.g., Bateson, 1972) represents the initial rocks deposited in the rift
and in the Appalachian foldbelt to the north (e.g., basins. The seismic response of this unit is more
Thomas et al., 1989). discontinuous, and of more variable amplitude, com-
pared with that of the overlying younger Jurassic
Late Jurassic syn-rift sequence sediments. This is best illustrated on Lines SF-7 and
SF-13 (Figs. 9 and 10). The wedge-shaped char-
Distribution acter of this lower unit in half-grabens adjacent to
A more detailed study of the Late Jurassic syn-rift fault scarps indicates its syn-rift origin. This vertical
sequence (J) throughout the study area was one of change in facies is also observed in the eastern part
the primary focuses of this study, as these rocks of the graben shown on Line SF 17 (Fig. 11). The
have not been examined in any detail. The distribu- lower unit, however, changes seismic facies laterally
tion and tectonic setting of the Late Jurassic rocks to a more continuous facies to the west, suggesting
(J) is illustrated on all the seismic lines (Figs. 5- a lateral change in depositional setting. The subdi-
12). A tectonic control on sedimentation is indi- vision into the two units is not as obvious on Line
cated, with presumed Jurassic sediments confined to SF-11 (Fig. 12), although the section does become
depocenters in half-graben or graben settings. The more discontinuous along the faulted flanks of the
characteristic wedge shape of the Jurassic sequence, half-graben. Based on the seismic character and the
i.e., the thickening against the border faults and tectonic setting, these lower rocks are tentatively
the progressive onlap onto the pre-rift section away interpreted to have been deposited in a non-marine,
from the fault, is diagnostic of syn-sedimentary rift-basin setting, possibly as footwall-sourced allu-
faulting (syn-rift setting). On some high standing vial fan deposits (Figs. 9-11) and/or hanging wall
blocks the sequence onlaps and is very thin or absent alluvial cones or dip slope fans (Fig. 12). Alter-
(e.g., Figs. 7-9 and 13), suggesting that many of the natively, they may represent fan-deltas deposited in
blocks remained high through the end of the Jurassic. lakes (i.e., represented by the seismic facies change
The generalized structure map on the pre-rift from discontinuous to continuous on Line SF-17;
surface (Fig. 13) also shows the tectonic setting Fig. 11). The overall depositional setting envisaged
of the Jurassic rift section in the south-central part here is a series of non-marine rift basins with interior
of the study area, including major basement blocks, drainage systems.
major faults and graben morphology. The map shows
the areas where Jurassic rocks are inferred to be Marine unit
missing. This includes the extensive lack of Jurassic The upper unit contains two laterally equivalent
over the Yucat~in terrace to the southwest (Fig. 13), seismic facies. Areas in the rift basin are charac-
as well as at Sites 536, 537 and 538, which suggests terized by more continuous reflections (Figs. 9-12),
that Jurassic sediments were not deposited across the while adjacent basement highs are often capped by
Yucat~in block to the west. chaotic, discontinuous, or reflection-free seismic fa-
The Jurassic section is separated from the overly- cies (Figs. 10 and 11). The continuous seismic facies
ing Lower Cretaceous section by the prominent Late in the basins is almost identical in seismic character
Berriasian unconformity (TB). This horizon was with that of the overlying deep marine Lower Creta-
drilled at Site 535 (Fig. 5; see also earlier discussion ceous section (Figs. 9-12), which has been sampled
of Site 535). At the drilling site it is seismically at DSDP Sites 535 and 540 (Fig. 5). These rocks,
a strong event which can be correlated through- therefore, are tentatively interpreted to also represent
out the area. It is interpreted to separate Jurassic sedimentation in a marine setting.
sediments, characterized by rapidly changing thick- The adjacent reflection-free facies overlying low-
nesses, from the overlying Cretaceous sediments, er-standing basement highs is characteristic of and is
characterized by more uniform distribution (Figs. 5- interpreted here to represent contemporaneous car-
12). This horizon is commonly onlapped by Lower bonate platforms or buildups, possibly with central
Cretaceous sediments, indicating its unconformable reef cores (Figs. 10 and 11). Above the reflection-
nature (Figs. 5-12). free zones drape can be observed, which represents

differential compaction above the reef tops (Figs. 10 horst and features, related to significant N E - S W
and 11). Further, between the reef more continuous extension. The dominant sense of faulting is to the
reflections are interpreted as the acoustic response southwest, except for a pair of northeastward-dip-
of lagoonal sediments (CDP 550, Fig. 10). The ping large faults on the northeastern end of the
more continuous reflectors adjacent to the banks line (CDP 1400). On Line SF-13 (Fig. 10) only
are interpreted as off-bank, more uniformly de- one half-graben can be identified with a southwest-
posited, deeper marine sediments (Figs. 10 and 11). ward-dipping basin bounding fault. This fault was
They probably consist of deeper-water shales and/or reactivated in the Early Cretaceous, probably due
pelagic-hemipelagic carbonates. to differential compaction and/or due to mild tec-
The distribution of the interpreted carbonate tonic rejuvenations. Jurassic faulting is also observed
buildups on basement highs is shown on the tectonic around CDP 1000. These faults became inactive by
map (Fig. 13). One large interpreted reef complex the end of the Jurassic, as they do not cut the Top
occurs on a northwest-trending elongated basement Berriasian surface (TB).
high (horst), which is bounded by a major fault on Further south on Line SF-17 (Fig. 11) syn-rift
the southwest side (Figs. 10 and 13). The area of this Jurassic sedimentation in the basin southwest of
bank appears to have become more restricted, with CDP 1000 was controlled by two faults dipping in
the margin stepping back through time (Fig. 10). opposite directions, resulting in a full-graben setting.
The bank appears to have finally drowned out near Thickness of the Jurassic section exceeds 0.5 s (1200
the end of the Jurassic (TB; Fig. 10). m). On the upthrown footwall, just northeast of CDP
Another Jurassic carbonate buildup was inter- 1000, the Jurassic is thin or absent. This indicates
preted on the northeast side of a tilted footwall block that even in the deep central part of the seaway large
shown on Line SF-17 at about CDP 500 (Figs. 11 fault blocks may have remained at or above sea level
and 13). The location of this carbonate edifice is throughout the Jurassic rifting.
somewhat different from the previously discussed The southernmost line shown here (SF-11,
bank, since it was established on a tectonically con- Fig. 12) shows a pair of southwestward-dipping ma-
trolled east-dipping slope, rather than on the top of jor faults controlling Jurassic sedimentation in the
a block. Syn-rift tilting of the block resulted in a half-graben between CDPs 500 and 1000. A thick,
buildup which faces to the northeast. The margin of tectonically disrupted pre-rift section is particularly
the platform is characterized by prograding clino- well defined with numerous, relatively small-scale
forms (CDP 550). Accelerated tilting and/or rising faults interpreted around CDP 500. These faults ap-
sea level then resulted in the drowning of the plat- parently were active only in the early phase of rifting
form toward the end of the Berriasian (TB). Lower as they cut only the pre-rift section and the lower-
Cretaceous onlap onto the tilted syn-rift section is most part of the Jurassic section. It can be observed
particularly well expressed on this section (Fig. 11). that Jurassic sediments which fill the half-graben
Based on the occurrence of the interpreted car- thin toward the southwest and pinch out against the
bonate buildups, and based on the widespread occur- Yucatan terrace (Fig. 12). Southwestward thinning
rence of laterally equivalent reflections with marine and pinchout of the Jurassic section against a tilted
characteristics, it can be postulated that the upper block is also observed on the northeastern part of
part of the Jurassic section was deposited in a shal- the section (CDP 1000-1200). The Jurassic is thin
low- to moderately deep-marine environment with or absent over the tilted block, and either was never
a dominantly carbonate facies. The age of the un- deposited or was later eroded. Between CDP 1000
derlying transition from non-marine to marine and and 1500 very well-defined onlap of Cretaceous beds
the timing of the establishment of a marine seaway defines the angular unconformity between the tilted
through the southeastern Gulf is not known, but it is Jurassic section and the more horizontal overlying
inferred to have taken place during the middle part Lower Cretaceous section (TB). Faults which extend
of the Late Jurassic. A Kimmeridgian age is sug- into the Lower Cretaceous section (about CDP 1200)
gested, which corresponds to the age of a possible again represent minor tectonic rejuvenations and/or
similar shallow- to deep-water setting that occurs to differential compaction.
the south within the Mesozoic section of western It has been recognized that half-grabens are
Cuba (Iturralde-Vinent, 1994). This is also a time of major tectonic elements in continental rifts (e.g.,
worldwide rise in sea level (Haq et al., 1988). Rosendahl, 1987). In the seismic examples shown
here it was documented that Jurassic sediments were
Tectonic setting deposited in both half-graben and full-graben set-
Interpreted faults indicate that deposition dur- tings in the central part of a presumed Jurassic
ing the Jurassic was controlled by faulting, which seaway (Figs. 9-13). In these examples the half-
resulted in typical graben and half-graben filling grabens change their sense of asymmetry along the
geometries. Line SF-7 (Fig. 9) shows impressive northwest-trending axis of the rift system (Fig. 13).

The transitional zones or overlapping zones between cially well expressed in the sub-basin around CDP
contrasting polarity rift segments are known as trans- 1000 in Fig. 9. Here thickness variations within the
fer or accommodation zones (e.g., Morley, 1988). A Lower Cretaceous section are not directly controlled
possible example of an accommodation zone may by the major faults. In general, the Lower Cretaceous
occur in the vicinity of Line SF-17 (Figs. 11 and section fills the remnants of the rift basins and forms
13), where there is an overlap between the main a widespread blanket. This geometry and the lack of
faults forming the opposing half-grabens. Here the major basement-involved faulting are characteristic
Jurassic occurs in a graben setting and thickening for sag basins in a post-rift setting.
of the Jurassic against major faults can not be ob- The Lower Cretaceous sequence is capped by
served (Fig. 11). In contrast, to the north and south the Middle Cretaceous Sequence Boundary (MCSB)
on Lines SF-15 (Fig. 5) and SF-11 (Fig. 12), the (e.g., Schlager et al., 1984; Buffer, 1991). The
Jurassic section thickens against major faults, half- MCSB always coincides with a high-amplitude re-
graben morphology and syn-depositional faulting is flector or reflection package. Its erosional character
interpreted (Fig. 13). Another accommodation zone is well expressed on many seismic lines, where it
probably occurs at the overlap of fault zones between truncates underlying Lower Cretaceous sediments
Lines SF 13 and SF-15 (Fig. 13). (Figs. 5-12). It is onlapped by younger Upper Cre-
The tectonic setting described above is generally taceous to Cenozoic sediments
characteristic of the syn-rift period in the evolution
of passive margins. This setting in the southeast-
em Gulf, however, is different, in that the rifting TECTONO-PALEOGEOGRAPHIC EVOLUTION OF THE
is contemporaneous with seafloor spreading in the SOUTHEASTERN GULF OF MEXICO BASIN AND
central Gulf to the north, rather than occurring prior SURROUNDING REGIONS
to seafloor spreading (Fig. 2). Faults in the area
generally do not cut the late Berriasian surface The Jurassic through Early Cretaceous tectono-
(TB), indicating the cessation of rifting near the paleogeographic evolution of the southeastern Gulf
Berriasian-Valanginian boundary. This is used as of Mexico is outlined in this section. These re-
evidence for the time of abandonment of seafloor constructions are based on the constraints drawn
spreading and oceanic crust formation in the north, from the regional scale model (Fig. 2), interpretation
as proposed in the regional model of Gulf evolution of the seismic data (Figs. 3-12), available DSDP
by Marton and Buffler (1994) and Marton (1995) well control (Figs. 4-8), plus the additional inter-
(Fig. 2). pretations presented by Marton and Buffer (1994)
and Marton (1995), particularly interpretations of
L o w e r C r e t a c e o u s post-rift s e q u e n c e (LK) regional geopotential data and structural and isopach
maps. To visualize the main events along with the
The Lower Cretaceous post-rift sequence (LK) already published and discussed kinematics of the
forms a widespread drape over the entire study area. eastern Gulf, a series of tectono-paleogeographic
It is relatively thin over the western and central part maps are presented (Fig. 14). Although paleogeo-
of the study area, but becomes thicker to the north graphic data are extremely scarce, an attempt is
over oceanic crust and particularly to the east toward made to tie the tectonic events to the seismically
the Florida Escarpment (Figs. 5-12). The geology identified rocks and infer, at least at first order, tec-
of this sequence has been studied in detail by Phair tonic styles, and depositional styles, as well as the
(1984), who divided the unit into four seismic se- paleo-extent of seas.
quences. He mapped the thickness, seismic facies, To place the southeastern Gulf of Mexico in a
and interpreted the depositional history of these se- more regional context, the Jurassic-Lower Creta-
quences using the DSDP sites as control. Results of ceous settings of the adjacent northeastern Gulf as
this study were also reported by Phair and Buffler well as western Cuba are included on the maps. Data
(1983) and Schlager et al. (1984). from the northeastern Gulf are taken from seismic
These studies concluded that the primary source stratigraphic studies of Lower Cretaceous rocks of
for the mainly hemipelagic deep-sea carbonates the offshore northeastern Gulf (Corso, 1987), Juras-
making up these sequences was sediment shed from sic rocks of the offshore northeastern Gulf (Dobson,
the Florida Escarpment to the east. These units thin 1990), and Mesozoic rocks along the Florida Es-
and onlap to the west onto basement highs in the carpment in the deep eastern Gulf (DeBalko, 1991)
south-central part of the area (e.g., Figs. 5, 6, 11 (Fig. 3). Additional information from these studies
and 12), and they also thicken and onlap onto the are included in Corso et al. (1989), DeBalko and
underlying Jurassic section in residual sag basins Buffler (1992), Buffer et al. (1993), and Dobson and
overlying the Jurassic rift grabens and half-grabens Buffler (1997). These studies document for the first
(Figs. 6, 8 and 9). This latter relationship is espe- time the evolution of Jurassic through Early Creta-

ceous shelf margins and provide a paleogeographic tion of evaporites (mainly salt) in a restricted Gulf
context for this part of the Gulf basin. of Mexico basin (Louann-Campeche salt) (Salvador,
Mesozoic rocks also occur throughout western 1991) (Fig. 14A). An important result of the regional
Cuba just to the south of the study area in the model (Fig. 2) (Marton and Buffler, 1994; Marton,
Cordillera de Guaniguanico region, and they provide 1995) was that oceanic crust formation began in the
the most comprehensive Mesozoic section of the Gulf of Mexico in late Middle Jurassic (Callovian)
northern proto-Caribbean passive margin (Fig. 1). time. In the northeastern Gulf of Mexico salt oc-
They occur as north-northwestward-thrust tectonic curs above an interpreted 'breakup unconformity',
slices contained in three main tectonostratigraphic which indicates that although salt deposition may
units: (1) the Sierra de los Organos; (2) Sierra del have started during the very end of the rifting stage
Rosario meridional (southern); (3) Sierra del Rosario in the central Gulf, it probably continued into the
septentrional (northern) (Fig. 1) (Pszcz6lkowski, early spreading stage around the periphery of the
1987; Iturralde-Vinent, 1994). Although the exact basin and was deposited on the initial oceanic crust
palinspastic reconstruction of these rocks can only in the center of the basin (Marton and Buffler, 1994;
be inferred, they have been interpreted by Iturralde- Marton, 1995).
Vinent (1994) to have originated along the southern The southeastern Gulf of Mexico formed a quasi-
continental margin of Yucat~.n, and then later in- continuous bridge between Yucatfin and Florida
corporated into the Cuban foldbelt during a Late separating the extending areas to the north (Gulf
Cretaceous to early Cenozoic orogenic event. This of Mexico) and to the south (proto-Caribbean)
palinspastic reconstruction is adopted for the pur- (Fig. 14A). Although Lower to Middle Jurassic
poses of the paleogeographic maps presented here dikes in DSDP Hole 538A provide some evidence
(Fig. 14). for the onset of extension, rifting during this period
There are significant differences among the three in the southeastern Gulf was subordinate to rifting in
regions (northeastern Gulf, southeastern Gulf, and the Late Jurassic. The only well-defined east-north-
western Cuba) shown on the maps. The northeast- east-trending extensional feature (corresponding to
ern Gulf stratigraphy represents sedimentation along the general northwest-southeast extension) is the
a Late Jurassic to Early Cretaceous passive mar- Catoche Tongue (Fig. 14A). This half-graben, based
gin which faces the eastern Gulf of Mexico ocean on its orientation, may represent an Early to Middle
basin. The southeastern Gulf stratigraphy represents Jurassic rifting episode (Shaub, 1983). The amount
an active Late Jurassic to earliest Cretaceous con- of early to late Middle Jurassic extension in the
tinental rift, which becomes a subsiding passive southeastern Gulf was possibly equivalent to the ad-
margin only in the Early Cretaceous. The western jacent Yucatfin block (fl = 1.25) or Florida block
Cuban section is interpreted to represent a portion (fl = 1.4) as discussed by Marton (1995). The
of a south-facing Mesozoic passive margin of the emergent position of the future seaway (continental
western proto-Caribbean located further to the south, bridge) is supported by the lack of preserved rocks
and it probably only had very broad similarities to of this age in the DSDP wells and in the area as
the former two areas. inferred from the seismic data. Further, distribution
of thick Callovian to Oxfordian salt also suggests
Late Triassic(?) to late Middle Jurassic (Fig. 14A) that marine waters did not reach the southeastern
Gulf of Mexico area (Fig. 14A).
During the Late Triassic(?) to late Middle Jurassic During the Late Triassic(?) to late Middle Jurassic
(Callovian), the Gulf of Mexico region accommo- period north of 27~ in the northeastern Gulf large-
dated a large amount of northwest-southeast ex- scale basins and structural highs formed, including
tension related to the incipient breakup of western the Apalachicola basin, Southern platform, and the
Pangea (Salvador, 1991; Marton and Buffler, 1994; Tampa embayment (Fig. 14A). Distribution of the
Marton, 1995; Fig. 2A,B). In general, this extension late Middle Jurassic salt reflects the areas of more
occurred above sea-level and was concentrated be- extreme extension, which subsided below sea level
tween the relatively unextended Yucatfin block and before the termination of rifting.
the southern margin of North America. Another lo- Farther south, along the southern margin of Yu-
cus of extension was located between Yucatfin and catfin, the Lower/Middle Jurassic(?) to Oxfordian
the northern margin of South America (Fig. 2A,B). terrigenous San Cayetano Formation was deposited
Marine rocks of this age have not been reported from (Fig. 14A) (Pszcz6tkowski, 1987). Maximum thick-
the Gulf of Mexico basin proper, which indicate a ness of the formation reaches and may exceed 3000
long period of extension in a continental rift setting m in the Sierra de los Organos of present-day west-
(Salvador, 1991). The end of this rifting phase is ern Cuba (Fig. 1). The Yucatfin block has been inter-
best characterized by subsidence below sea level in preted as the source area of these clastic sediments
the Callovian, resulting in the widespread deposi- (Pszcz6tkowski, 1987). The well-bedded shales and

4~
C3
b"

9
Z
Fig. 14. Maps showing interpreted Late Jurassic tectono-paleogeographic reconstructions of the eastern Gulf of Mexico and northwestern Cuba (palinspastically restored to southeastern margin of Yucat~in
block) for: (A) Callovian, (B) Oxfordian, (C) Kimmeridgian, (D) Tithonian, (E) Berriasian-Valanginian, and (F) Early Cretaceous. Gray line around Yucat~in and Florida block is present location of
<
O
t"
-]
0
Z
0
>.
-]
E"
0
0
>.

Fig. 14 (continued). Lower Cretaceous platform margin for reference. Dashed heavy lines indicate inferred landward limit of seas. Pole of rotation in southeastern Gulf is shown with degrees of rotation of
the Yucatfin block. Other lines and patterns on maps are as annotated. See text for discussion of each map.
...-:,
~ ,...~
.,..~

sandstones of the San Cayetano sediments repre- South of Yucatfin in western Cuba the middle
sent deposition in a deltaic and local shallow-marine Oxfordian shallow-water sediments of the Jagua
setting (e.g., Haczewski, 1987). In the Sierra del Formation (in the Sierra de los Organos, Fig. 1) and
Rosario (present-day western Cuba, Fig. 1) thick- the Francisco Formation (in the Sierra del Rosario,
ness of this unit never exceeds 1000 m, and facies Fig. 1) consist of shales and limestones similar to
represent a more distal setting for this segment of the the Smackover Formation. A thick unit of Oxfordian
Cordillera de Guaniguanico (Haczewski, 1987). The to early Kimmeridgian tholeiitic basalts, interbedded
San Cayetano sediments may represent deposition in with shale and limestones, in the northern Sierra del
an extensional rift basin, which formed as Yucatfin Rosario has been interpreted as the result of con-
and South America rifted apart. This is corroborated tinental margin volcanism (Iturralde-Vinent, 1994,
by the occurrences of basic dikes and sills in the San 1996).
Cayetano Formation (Iturralde-Vinent, 1994, 1996).
Kimmeridgian (Fig. 14C)
Oxfordian (Fig. 14B)
During Kimmeridgian time the southeastern Gulf
By Oxfordian time the Yucatan block had roof Mexico seaway became wider as seafloor spread-
tated 11~ counter-clockwise, a narrow oceanic tract ing continued, the Yucatfin block rotated 18~ coun-
had propagated into the northernmost segment of ter-clockwise, and the tip of the spreading center
the southeastern Gulf of Mexico, and salt deposi- propagated farther south (Fig. 14C). It is possible
tion was completed in the Gulf of Mexico basin that by this time a marine connection between the
(Fig. 2C, Fig. 14B). The identified salt in the south- Gulf of Mexico and the proto-Caribbean had become
eastern Gulf extends only slightly southward of the established, and the continental bridge had subsided
tip of the oceanic crust and is confined more to the below paleosea-level. Although drilling data are not
east, suggesting asymmetrical rifting and subsidence available, it is postulated that the seismically iden-
along the Florida margin. Major continental rifting tified carbonate buildups in the central part of the
had begun all along the southeastern Gulf, but the seaway (Figs. 10 and 11) started to grow during this
area remained at or just above sea level, prevent- time. Several of the large blocks in the central part
ing deposition of thick salt. The older part of the of the seaway probably remained above sea-level, as
wedge-shaped Jurassic syn-rift units, which occur well as the whole Yucatfin block and the Yucatfin
along the marginal faults bounding syn-rift basins terrace (Figs. 13 and 14C).
(e.g., Figs. 9 and 10), probably represent this con- In the northeastern Gulf of Mexico the carbon-
tinental (non-marine) syn-rift period, with the local ate-dominated Smackover Formation gave way to
deposition of alluvial fans or fan deltas building into the mixed clastic/carbonate sequence of the Kim-
small bodies of fresh water (Fig. 14B). meridgian Haynesville Formation (Salvador, 1987)
In the northeastern Gulf of Mexico, above the (Fig. 14C). Based on seismic stratigraphy and lim-
Louann salt, the early to middle Oxfordian basal ited (up-dip) well information, Dobson (1990) in-
clastics of the Norphlet Formation were deposited, terpreted clastic(?) shelf margin progradation in
which include an alluvial-fluvial-eolian system bor- the Tampa embayment and a pronounced carbon-
dering the Oxfordian sea (Salvador, 1987; Dobson, ate shelf margin in the outer Apalachicola basin
1990; Dobson and Buffler, 1997). Pronounced trans- (Gilmer equivalent) (Fig. 14C). Deep sea fans con-
gression of the Oxfordian sea is documented by tinued to be deposited in the deep basin adjacent
the deposition of the carbonate-dominated Smack- to the Tampa embayment/Sarasota arch (DeBalko,
over Formation (Fig. 14B). Shallow-water carbon- 1991).
ates were deposited on a monocline or in a ramp In the late Oxfordian to Kimmeridgian in the
setting, which deepened towards the rapidly sub- Sierra de los Organos segment of western Cuba
siding basin center. On local highs basin margin (Fig. 1) extensive carbonate platform(s) evolved
carbonate buildups (composed of corals, sponges (Pszcz6lkowski, 1987) (Fig. 14C). The thickness
and stromatolites) were established (Dobson, 1990) of the thick-bedded shallow-water platform sedi-
(Fig. 14B). These sediments can be regarded as the ments represented by the San Vicente Member of
foundation of the newly established passive margin the Guasasa Formation reaches 650 m. In the Sierra
facing the spreading Gulf of Mexico. West of the del Rosario segment (Fig. 1) the equivalent section
Tampa embayment and the Sarasota arch, in a deep- reaches only a few tens of meters. Kimmeridgian
marine (slope) setting, DeBalko (1991) has identified sediments in the Sierra del Rosario represent rel-
an Oxfordian section more than 1000 m thick. This atively slow sedimentation, but they do not have
may represent deep-sea fan sediments deposited in characteristics of typical pelagic deposits. They may
a rapidly subsiding basin next to the spreading instead represent a transition zone between the shal-
center. low-water platform of the Sierra de los Organos and

a deeper basinal area to the south (Iturralde-Vinent, commenced in the entire area. Late Jurassic car-
1994) (Fig. 14C). bonate buildups in the central part of the seaway
were drowned and marine transgression submerged
Tithonian (Fig. 14D) previously emergent blocks. Early to middle Berri-
asian transgression in the tectonically active area
By Tithonian time the tip of oceanic crust had near block 'BL' (Fig. 14E) is well documented at
propagated into the northern segment of the south- DSDP Site 537 (Fig. 6). Fault activity at this time
eastern Gulf of Mexico, leaving behind rapidly sub- or slightly later finally separated the block from Yu-
siding passive margins of increasing length on both cat~in, which had to be the source area for the drilled
sides of the seaway (Fig. 14D). Rifting had reached fluvial to nearshore sediments. The Berriasian to
a more advanced stage in the southern segment, Valanginian transgression also probably reached the
as the Yucat~in block rotated 25 ~ counter-clock- area of the Yucat~in terrace and possibly the entire
wise around the rotation pole. By Kimmeridgian Yucat~in block. Ephemeral shallow-water carbonate
or Tithonian time a full-fledged seaway may have platforms were established on the Catoche Knoll
evolved between the opening Gulf of Mexico and and on 'BL', represented by recovered late Berri-
the proto-Caribbean. This is based on the seismically asian to early Valanginian shallow-water limestones
identified carbonate buildups that are present in the at Site 537 (Fig. 6). Deep-sea conditions in the late
axial part of the seaway and the equivalent off-bank Berriasian in the central part of the seaway was
reflectors with marine characteristics (Figs. 10 and documented at DSDP Site 535 (Fig. 5).
11). Although no direct evidence is available, it is In the northeastern Gulf of Mexico, Cotton Val-
possible that drowning and step-back of these small ley sedimentation continued into the Berriasian
carbonate platforms began in the Tithonian (Figs. 10 (Fig. l lE). Close to the future Lower Cretaceous
and 11). Several of the large blocks still remained platform margin, the seismically and lithologically
above sea-level, including the Yucat~in block and the well defined Knowles ramp was mapped by Dobson
Yucat~in terrace (Figs. 13 and 14D). (1990) based on drilling and seismic stratigraphy. The
Tithonian sedimentation in the northeastern Gulf exact age of the transgression onto the higher standing
of Mexico is characterized by deposition of the thick parts of the Sarasota arch and south Florida platform
fluvial-deltaic sequence of the Cotton Valley clastics is not well established. In south Florida, in the type
that prograded across the Apalachicola basin and into well (Bass Collier Country 12-2), Applegate et al.
the Tampa embayment (Dobson, 1990) (Fig. 14D). (1981) described the latest Jurassic(?) to earliest Cre-
Deep-sea fans influenced by southward currents were taceous Wood River Formation overlying Early Juras-
interpreted in the deep basin adjacent to the Tampa sic (189 Ma radiometric age) rhyolite porphyry. Based
embayment/Sarasota arch (DeBalko, 1991). on palynological studies Salvador (1987) stated that
In western Cuba shallow-water sedimentation only the basal sandstone- and shale-dominated 30-50
ceased in the Tithonian (Pszcz6tkowski, 1987). Slow m of the section is Tithonian. Consequently, the upper
sedimentation and deepening environments charac- (600 m) dolomite and evaporite section of the Wood
terizes the E1 Americano Member of the Guasasa River Formation represents earliest Cretaceous (and
Formation in the Sierra de los Organos (Fig. 1). younger) platform deposits.
Similarities between this formation and the La Zarza In western Cuba in the Berriasian and Valan-
Member (upper part) of the Artemisa Formation ginian, pelagic sedimentation prevailed in the en-
in the Sierra del Rosario (Fig. 1) indicate uniform tire Cordillera de Guaniguanico region (Fig. 1),
pelagic deposition throughout the entire Cordillera including radiolarian ooze deposits and cherts
de Guaniguanico region (Pszcz6ikowski, 1987) (Pszcz6ikowski, 1987) (Fig. 14E). Dramatically re-
(Fig. 14D). This deep-water environment in the duced sedimentation rates and very deep-water fa-
Guaniguanico suggests a transgression towards the cies in western Cuba indicate gradual deepening
north onto the southern margin of the Yucat~in block and starved basin conditions south of the southern
(Fig. 14D). Continued extension along this margin Yucatfin margin in the proto-Caribbean. In contrast,
is suggested by coeval basaltic sills and dikes in the limestones of the Tumbadero and Tumbitas Mem-
Sierra del Rosario sections (Iturralde-Vinent, 1996). bers of the Guasasa Formation in the Sierra de los
Organos, and the Sumidero Member of the Artemisa
Berriasian-Valanginian (Fig. 14E) Formation in the Sierra del Rosario, do not contain
any terrigenous material (Fig. 14E).
In the late Berriasian spreading ceased in the Gulf
of Mexico and the Yucat~in block reached its present Early Cretaceous (Aptian-Albian) (Fig. 14F)
position (42~ rotation) (Fig. 2D, Fig. 14E). Conse-
quently, rifting also ceased throughout the southeast- During Early Cretaceous time in the southeastern
ern Gulf of Mexico and rapid (thermal) subsidence Gulf of Mexico passive margin conditions prevailed.

This period is represented by the deposition of m thick) is not related to subsequent extension, it
thick sections of platform carbonates on the Yu- is referred to as the 'pre-rift' section. Its age is not
catfin and Florida blocks (Fig. 14F). Steep margins clearly documented in the available DSDP wells.
formed which separated the platform areas from the Based on the interpreted structural setting and pos-
deep basins. Present-day relief along the Campeche sible correlation to DSDP Site 536, it is suggested
Florida escarpments reaches and exceeds 3000 m. In that this pre-rift unit represents a Pre-mesozoic (late
the axial trough deep-sea pelagic and hemipelagic Paleozoic?) sedimentary cycle.
sedimentation occurred, as interpreted at Sites 535 (2) Based on the available seismic data, the Late
and 540. Exceptions are Pinar del Rfo Knoll and Jurassic syn-rift section shows a complex basin-fill
Jordan Knoll, on top of which shallow-water car- geometry in the continental domain of the south-
bonate platforms were able to keep up with the eastern Gulf of Mexico. The syn-rift nature of the
fast subsidence (Fig. 14F). In Albian-Cenomanian Jurassic section is interpreted based on its character-
time these carbonate platforms were progressively istic wedge-shaped occurrence in half-grabens and
drowned, and the platform margins stepped back its progressive onlap onto the pre-rift section. In
to shallower positions towards Yucatfin and Florida. general, Jurassic is interpreted to be missing on the
The southeastern Gulf of Mexico became a sedi- high-standing Yucatfin and Florida blocks, as well
ment-starved deep seaway, the floor of which was as on large-scale basement highs in the central part
scoured at various times by deep-sea currents during of the southeastern Gulf of Mexico and the Yucatfin
the mid- and Late Cretaceous. terrace to the southwest.
(3) Syn-rift sedimentation has been interpreted
to evolve from a non-marine to a normal marine
CONCLUSIONS setting during the latest Middle Jurassic to earliest
Cretaceous period. Lack of salt deposition and the
The regional opening model (Fig. 2) (Marton non-marine character of seismic reflectors in the
and Buffler, 1994; Marton, 1995) provides a frame- deep half-grabens indicate that in the Callovian to
work in which the Mesozoic tectono-stratigraphic Oxfordian(?) in the southern and central segments
evolution of the southeastern Gulf of Mexico can of the southeastern Gulf of Mexico rifting occurred
be discussed. This involved a southward-propagat- in a continental setting. The southeastern Gulf of
ing spreading center/rift as Yucatfin rotated coun- Mexico during this time formed a continental bridge
ter-clockwise relative to Florida during the Late between Yucatfin and Florida. Seismically identified
Jurassic about a nearby pole in the southern part carbonate buildups and the more marine character of
of the study area. The continental domain of the seismic reflectors in the upper part of the Jurassic
southeastern Gulf of Mexico, based on interpre- section suggest that the later rifting-stage (Kim-
tation of reflection seismic data, is characterized meridgian(?) to late Berriasian) in the southeastern
by variable, 3500 to 7000 m basement depths, Gulf of Mexico occurred in a marine setting. Al-
and by a continental rift morphology. The domi- though no drilling data are available, it is suggested
nant tectonic elements in this area are northwest- that in Kimmeridgian to Tithonian time a marine
ward-trending horst and features, corresponding to seaway opened between the Gulf of Mexico and the
the main northeast-southwest maximum extensional proto-Caribbean as the southeastern Gulf of Mexico
strain. A detailed analysis of the available geophys- subsided below sea level. In late Berriasian time
ical data as well as the review of the DSDP Leg spreading in the Gulf of Mexico ceased and the
77 results in the southeastern Gulf of Mexico pro- southeastern Gulf of Mexico rift-system aborted.
vide important information to decipher details for (4) Onset of rapid thermal subsidence in the ear-
the development of this Late Jurassic-Early Creta- liest Cretaceous resulted in the final submergence
ceous seaway. Based on this analysis, four major of several of the large fault blocks as well as the
tectono-stratigraphic sequences bounded by major Yucatfin and Florida platforms. Long-lasting carbon-
unconformity surfaces have been defined: crystalline ate platforms became established on rift shoulders
basement, Paleozoic( ?) pre-rift rocks, a Late Juras- (Yucatfin and Florida blocks), while deep-sea sedi-
sic rift sequence, and an Early Cretaceous post-rift mentation prevailed in the axial part of the south-
sequence. Major conclusions of the tectono-paleo- eastern Gulf of Mexico, as indicated by the results of
geographic evolution of the area are summarized the DSDP Leg 77. Along the constructive carbonate
below. margins of Yucatan and Florida, thicknesses of the
(1) The oldest sedimentary unit in the southeast- Lower Cretaceous platform carbonates reached and
ern Gulf of Mexico has been delineated based on exceeded 1500 m. In the adjacent basinal setting, de-
the primary criterion that its thickness variations are pending of the availability of off-platform material, a
not correlated with offset along Jurassic extensional 500-m to 1500-m-thick carbonate-dominated section
faults. Because deposition of this unit (up to 2000 accumulated in the Early Cretaceous.

ACKNOWLEDGEMENTS Dallmeyer, R.D., 1984. Ar40/Ar39 ages from a pre-Mesozoic

crystalline basement penetrated at holes 537 and 538A of
S u p p o r t for this study was in part p r o v i d e d by the the Deep Sea Drilling Project Leg 77, southeastern Gulf of
Mexico: tectonic implications. Init. Rep. DSDP, 77: 497-506.
N a t i o n a l S c i e n c e F o u n d a t i o n grant O C E - 9 0 2 0 6 7 3 ,
DeBalko, D.A., 1991. Seismic Stratigraphy and Geologic History
Texas H i g h e r E d u c a t i o n C o o r d i n a t i n g B o a r d Ad- of Upper Middle Jurassic through Lower Cretaceous Rocks,
vanced Research program, generous fellowships Deep Eastern Gulf of Mexico. M.A. Thesis, The University of
g r a n t e d by T h e U n i v e r s i t y of Texas D e p a r t m e n t Texas at Austin, Austin, Texas, 143 pp.
of G e o l o g y and Institute for G e o p h y s i c s and the So- DeBalko, D.A. and Buffer, R.T., 1992. Seismic stratigraphy and
geologic history of Middle Jurassic through Lower Cretaceous
ciety of E x p l o r a t i o n G e o p h y s i c i s t s . T h e authors are
rocks, deep eastern Gulf of Mexico. Trans. Gulf Coast Assoc.
p a r t i c u l a r l y thankful to Paul M a n n for his support, Geol. Soc., 42: 89-105.
s u g g e s t i o n s and e n c o u r a g e m e n t during the prepa- Dobson, L.M., 1990. Seismic Stratigraphy and Geologic History
ration of the m a n u s c r i p t . R e v i e w o f this p a p e r by of Jurassic Rocks, Northeastern Gulf of Mexico. M.S. Thesis,
J a m i e Austin, A1 H i n e and A n d r z e j P s z c z 6 l k o w s k i The University of Texas at Austin, Austin, Texas, 165 pp.
Dobson, L.M. and Buffler, R.T., 1997. Seismic stratigraphy and
is d e e p l y a p p r e c i a t e d . T h e i r c o m m e n t s and profes-
geological history of Jurassic rocks, northeastern Gulf of Mex-
sional insights significantly i m p r o v e d the quality of ico. Am. Assoc. Pet. Geol. Bull., 81: 100-120.
the m a n u s c r i p t . This is T h e U n i v e r s i t y of Texas Haczewski, G., 1987. Sedimentological reconnaissance of the
Institute for G e o p h y s i c s C o n t r i b u t i o n 1339. San Cayetano Formation: an accumulative continental margin
in the Jurassic of western Cuba. In: A. Pszcz6tkowski, K.
Piotrowski, A. De la Torre, R. Myczynski and G. Haczewski
(Editors), Contribucion a la Geologia de las Provincia Pinar
REFERENCES del Rio (Contributions to the Geology of Pinar del Rio
Province). Editorial Cientffco-T6cnica, Ciudad de la Habana,
Angstadt, D.M., 1983. Seismic Stratigraphy and Geologic His- La Habana, pp. 228-247.
tory of the Southeastern Gulf of Mexico/southwestern Straits Hall, S.A. and Najmuddin, I.J., 1994, Constraints on the tec-
of Florida. M.A. Thesis, The University of Texas at Austin, tonic development of the eastern Gulf of Mexico provided by
Austin, Texas, 206 pp. magnetic anomaly data. J. Geophys. Res., 99: 7161-7175.
Angstadt, D.M., Austin, J.A., Jr. and Buffler, R.T., 1983. Deep- Haq, B.U., Hardenbol, J. and Vail, ER., 1988, Mesozoic and
sea erosional unconformity in the southeastern Gulf of Mex- Cenozoic chronostratigraphy and cycles of sea-level change.
ico. Geology, 11: 215-218. In: C.K. Wilgus, B.S. Hastings, C.G.St. Kendall, H.W. Posa-
Angstadt, D.M., Austin, J.A., Jr. and Buffler, R.T., 1985. Seismic mentier, C.A. Ross and J.C. Van Wagoner (Editors), Sea-Level
stratigraphy and geologic history of the southeastern Gulf Changes: An Integrated Approach. Soc. Econ. Paleontol. Min-
of Mexico-southwestern Straits of Florida. Am. Assoc. Pet. eral. Spec. Pap., 42: 40-45.
Geol. Bull., 69: 977-995. Iturralde-Vinent, M.A., 1994. Cuban geology: a new plate-tec-
Applegate, A.V., Winston, G.O. and Palacas, J.G., 1981. Subdi- tonic synthesis. J. Pet. Geol., 17: 39-70.
vision and regional stratigraphy of the pre-Punta Gorda rocks Iturralde-Vinent, M.A. (Editor), 1996. Ofiolitas y Arcos Volcani-
(lowermost Cretaceous-Jurassic?) in south Florida. Gulf Coast cos de Cuba. Project 364, Caribbean Ophiolites and Volcanic
Assoc. Geol. Soc. Suppl. Trans., 31: 447-453. Arcs, Special Contribution No. 1, Miami, Fla., 254 pp.
Bateson, J.H., 1972. New interpretation of geology of Maya Marton, G.L., 1995. Jurassic Evolution of the Southeastern Gulf
Mountains, British Honduras. Am. Assoc. Pet. Geol. Bull., 56: of Mexico. Ph.D. Dissertation, The University of Texas at
956-963. Austin, Austin, Texas, 276 pp.
Bryant, W., Meyerhoff, A.A., Brown, N., Furrer, M., Pyle, T. Marton, G. and Buffer, R.T., 1994. Jurassic reconstruction of the
and Antoine, J., 1969, Escarpments, reef trends, and diapiric Gulf of Mexico basin. Int. Geol. Rev., 36: 545-586.
structures, eastern Gulf of Mexico. Am. Assoc. Pet. Geol. Morley, C.K., 1988. Variable extension in Lake Tanganyika.
Bull., 53: 2506-2542. Tectonics, 7: 785-801.
Buffler, R.T., 1991. Seismic stratigraphy of the deep Gulf of Phair, R.L., 1984. Seismic Stratigraphy of the Lower Cretaceous
Mexico basin and adjacent margins. In: A. Salvador (Editor), Rocks in the Southwestern Florida Straits, Southeastern Gulf
The Gulf of Mexico Basin. The Geology of North America, J, of Mexico. M.A. Thesis, The University of Texas at Austin,
Geological Society of America, Boulder, Colo., pp. 353-387. Austin, Texas, 319 pp.
Buffer, R.T., Schlager, W. and Shipboard Scientific Party, 1984. Phair, R.L. and Buffler, R.T., 1983, Pre-Middle Cretaceous ge-
Initial Reports of the Deep Sea Drilling Project, 77. U.S. ologic history of the deep southeastern Gulf of Mexico. In:
Government Printing Office, Washington, D.C., 747 pp. A.W. Bally (Editor), Seismic Expression of Structural Styles
Buffer, R.T., Dobson, L.M. and DeBalko, D.A., 1993. Middle m A Picture and Work Atlas. Am. Assoc. Pet. Geol., Stud.
Jurassic through Early Cretaceous evolution of the northeast- Geol., 15 (2): 2.2.3-141.
ern Gulf of Mexico basin. In: J.L. Pindell and B.B. Perkins Pszcz6tkowski, A., 1987. Paleogeography and paleotectonic evo-
(Editors), Mesozoic and Early Cenozoic Development of the lution of Cuba and adjoining areas during the Jurassic-Early
Gulf of Mexico and Caribbean Region. Gulf Coast Section, Cretaceous. Ann. Soc. Geol. Pol., 57: 127-142.
Society of Economic Paleontologists and Mineralogists, pp. Rosendahl, B.R., 1987. Architecture of continental rifts with
33-50. special reference to east Africa. Annu. Rev. Earth Planet. Sci.,
Corso, W., 1987. Development of the Early Cretaceous North- 15: 445-503.
west Florida Carbonate Platform. Ph.D. dissertation, The Uni- Salvador, A., 1987. Late Triassic-Jurassic paleogeography and
versity of Texas at Austin, Austin, Texas, 136 pp. origin of Gulf of Mexico Basin. Am. Assoc. Pet. Geol. Bull.,
Corso, W., Buffer, R.T. and Austin, J.A., 1989. Erosion of the 71: 419-451.
southern Florida escarpment. In: A. Bally (Editor), Atlas of Salvador, A., 1991. Triassic-Jurassic. In: A. Salvador (Editor),
Seismic Stratigraphy. Am. Assoc. Pet. Geol. Stud. Geol., 27 The Gulf of Mexico Basin. The Geology of North America J,
(2): 149-157. Geological Society of America, Boulder, Colo., 131-180.

E V O L U T I O N OF THE S O U T H E A S T E R N G U L F OF MEXICO BASIN 91
Schlager, W., Buffler, R.T., Angstadt, D., Phair, R.L., 1984. southeastern Gulf of Mexico. Init. Rep. DSDE 77: 525-530.
Geologic history of the southeastern Gulf of Mexico. Init. Thomas, W.A., Chowns, T.M., Daniels, D.L., Neathery, T.L.,
Rep. DSDP, 77: 715-738. Glover, L. and Gleason, R.J., 1989. The subsurface Appa-
Shaub, EJ., 1983. Origin of the Catoche Tongue. In: A.W. Bally lachian-Ouachita orogen beneath the Atlantic and Gulf
(Editor), Seismic Expression of Structural Styles - - A Picture Coastal Plains. In: Hatcher, R.D. Jr., Thomas, W.A. and
and Work Atlas. Am. Assoc. Pet. Geol., Stud. Geol., 15 (2): Viele, G.W. (Editors), The Appalachian-Ouachita Orogen in
2.2.3-129. the United States. The geology of North America F-2, Geolog-
Testarmata, M.M. and Gose, W.A., 1984. A paleomagnetic eval- ical Society of America, Boulder, Colo., pp. 445-458.
uation of the age of the dolomite from site 536, Leg 77,

Chapter 4
The Exposed Passive Margin of North America in Western Cuba
ANDRZEJ PSZCZOLKOWSKI
The Mesozoic successions of western Cuba, now exposed in the Guaniguanico terrane, were deposited to the east of the
present NE Yucatan coast. The evolution of these passive margin successions encompasses the syn-rift stage (Early Jurassic-
?Callovian/early Oxfordian), drift stage (?Callovian/middle Oxfordian-Santonian), and the beginning of the active margin stage
(Campanian-Paleocene). Prior to the middle Oxfordian, the San Cayetano basin was located in an originally narrow rift zone
formed between Yucatan and South America. The onset of shallow-water carbonate sedimentation in the Sierra de los Organos
and Cangre belts occurred in the late Oxfordian or earliest Kimmeridgian. Drowning of a carbonate bank, or platform, in
the early Tithonian resulted in a considerable uniformity of facies in all belts of the Guaniguanico terrane, expressed by
widespread occurrence of ammonite-bearing limestones and radiolarian microfacies, especially in the upper Tithonian deposits.
Pelagic limestones accumulated during the Berriasian and Valanginian, while siliciturbidites occurred in the Northern Rosario,
La Esperanza and Placetas belts of western and central Cuba during the Valanginian-Barremian. These belts belonged to a
deep-water sector of the basin that extended between the Yucatan and Bahamas platforms. During the Aptian-Albian, siliceous
deposition extended across the entire deeper part of the northwestern proto-Caribbean basin. Pelagic carbonate sedimentation
resumed in the Cenomanian. Origin of the regional late Turonian (or Coniacian)-Santonian hiatus in the deep-water, pelagic
sequence of the northwestern proto-Caribbean basin was probably related to paleoceanographic conditions that existed during Late
Cretaceous times. These conditions were associated with paleogeographic changes in the southern part of the proto-Caribbean
basin, when the Nicaraguan Rise-Greater Antilles Arc partially closed the connection with the Pacific.
During the Campanian, abundant volcaniclastic detritus appeared in the upper Moreno Formation of the Northern Rosario belt.
The Bahfa Honda segment of the volcanic arc was located east of the Yucatan block margin and south of the Moreno depocenter.
This arc could be the westernmost part of the Greater Antilles Arc (GAA). Unlike previous interpretations, at the end of the
Cretaceous a more southerly position of the extinct volcanic arc is inferred from the paleotectonic reconstruction and lithology of
the late Maastrichtian deposits. During the Late Paleocene, clastic deposition occurred in a foreland basin setting, in front of a
thrust belt along the southern side of the remnant proto-Caribbean Sea.
INTRODUCTION and in the Sierra de Escambray (Fig. 1) are sim-

ilar to Mesozoic successions of the Guaniguanico
The Jurassic to Paleocene sedimentary succes- terrane (Khudoley and Meyerhoff, 1971; Mill~n and
sions of the passive margins of North America Myczyfiski, 1978). Stratigraphic and lithologic sim-
are exposed in Cuba (Fig. 1). The Mesozoic plat- ilarities existing between the Guaniguanico, Pinos
form and/or slope deposits crop out in the northern and Escambray terranes (Fig. 1) clearly suggest their
part of central Cuba. These deposits, traditionally paleogeographic proximity prior to the Late Creta-
linked to the Bahamas platform (Meyerhoff and ceous and Paleogene tectonic events (Pszcz6ikowski,
Hatten, 1974; Pardo, 1975) occur also in the Matan- 1981; Iturralde-Vinent, 1994).
zas Province (Pszcz6~kowski, 1986b) and in east- Studies of the Pinar del Rio geology were carried
ern Cuba (Iturralde-Vinent, 1996). In western Cuba out by oil companies before 1959, which resulted in
(Fig. 2) the Jurassic to Paleocene rocks occur in many advances in understanding of stratigraphy and
the Guaniguanico tectonostratigraphic unit (terrane). tectonics of this area (Hatten, 1957, 1967; Rigassi-
These rocks are considered to belong originally to Studer, 1963; Meyerhoff, in Khudoley and Mey-
the eastern margin and slope of the Yucatan platform erhoff, 1971; Pardo, 1975). Mapping and research
(Iturralde-Vinent, 1994, 1996). Metamorphic rocks carried out in western Cuba during the past 26 years
exposed in the Isla de la Juventud (Isle of Pines) has resulted in publication of many papers, includ-

94 A. PSZCZ()LKOWSKI
SGM ~FLORIPA .A.Np.BAHA.M.AS!:

A N Igl ~ U ~ / ~ , ~_ ~"ql'qlNNNNNNN'qI'qlN'
jKIj~_ I i ~ ,../ ~) "~11-1~ . _ NINIR"q"
HAVANA
Ill,, K'~N e \.d, f
Camagoey
Province
200 km -~.~.,
Fig. 1. (A) Map of Cuba showing the location of selected geological structures and deep wells. 1 = terranes of passive margin origin
exposed in western and south-central Cuba: GU -- Guaniguanico (stratigraphic terrane), P = Pinos (metamorphic terrane in the Isla
de la Juventud, or Isle of Pines), E = Escambray (metamorphic terrane in the Sierra de Escambray); 2 - the Placetas and Camajuanf
belts in north-central Cuba (Kimmeridgian?/Tithonian to Maastrichtian slope successions) and the Asunci6n metamorphic massif (AN)
in eastern Cuba; 3 - the Remedios belt (Cretaceous platform succession)" 4 = well location sites (shown as encircled numbers: 1 -
Martfn Mesa 1 (situated in the Martfn Mesa tectonic window), 2 = Pinar 1, 3 = Guanahacabibes, 4 = Los Arroyos 1), BH = Bahfa
Honda terrane (ophiolite and Cretaceous volcanic arc), Gh = Guanahacabibes Peninsula. (B) Schematic map showing the location of the
main terranes and belts in western and central Cuba and adjacent areas (partly after Rosencrantz, 1996 and Case et al., 1984, 1990): a
= Yucat;in platform; b = Florida and Bahamas platforms; c = Camajuanf and Placetas belts (undivided) in north-central Cuba (CA &
PS)" d -- Yucatfin basin; e = Cayman ridge; f = Camagtiey trench; GU = Guaniguanico terrane in western Cuba; BH -- Bahia Honda
terrane (ophiolite and Cretaceous volcanic arc); P -- Pinos terrane; E = Escambray terrane; RS & CC = Remedios and Cayo Coco belts
(undivided) in north-central Cuba (shallow-water and pelagic carbonates); SGM = southeastern Gulf of Mexico. Arrows indicate relative
movement along faults and barbed continuous lines denote major thrusts.
ing overviews of Cuban geology (Lewis and Draper, western Cuba. Two types of terranes may be distin-
1990; Iturralde-Vinent, 1994, 1996), and geological guished in western and south-central Cuba, namely
maps. The present paper focuses on the evolution of the stratigraphic and metamorphic terranes. How-
the Jurassic to Early Paleocene passive margin suc- ever, in western Cuba both types of terranes are not
cessions now exposed in the Guaniguanico terrane completely separated, as the Guaniguanico terrane
of western Cuba, before their Paleogene tectonic includes also metamorphic rocks (the Cangre belt).
deformation. The term 'Guaniguanico terrane' was introduced
by Iturralde-Vinent (1994). This author (Iturralde-
Vinent, 1994, 1996) proposed a generalized tectonic
TECTONIC SETTING scheme of the Pinar del Rio Province and placed the
Guaniguanico terrane among the southwestern Cuban
In this paper, a tectonostratigraphic terrane is de- terranes (with the Pinos and Escambray terranes). In
fined following the criteria of Howell et al. (1985), his opinion, the Guaniguanico terrane is composed
adopted in some recent Caribbean geological studies of five juxtaposed belts: Los Organos, Rosario South,
(for example, Mann et al., 1991). Also in Cuba some Rosario North, Quifiones and Felicidades.
geological structures have been characterized as 'ter- The Guaniguanico terrane (Figs. 1 and 2) is lo-
ranes' (Lewis and Draper, 1990; Pszcz6tkowski, cated mainly in the Pinar del Rio Province but
1990; Piotrowska, 1993; Iturralde-Vinent, 1994, its eastern extremity reaches the Havana Province.
1996; etc.), although the overall, generally accepted The Pinar fault forms the southern boundary of
scheme of Cuban tectonostratigraphic terranes is still this tectonostratigraphic unit. The eastern part of
to be achieved. It was not the aim of this paper to the northern boundary of the Guanigianico ter-
propose such a scheme. Rather, the concept of terrane is defined by the tectonic contact with the
ranes is merely used herein to explain a Mesozoic Bahfa Honda terrane and the Guajaibdn-Sierra Azul
evolution of a passive margin successions exposed in unit (Fig. 2B and C). To the west, the northern

THE EXPOSED PASSIVE MARGIN OF NORTH AMERICA IN WESTERN CUBA 95
84000 ' 83o30 ' 83c,00 '

I
TERRAN
I I ,co 0A
o
APS
0 S
LM
22030 '
/Y 0
Pinar del R{o
o P i n a r del R~o
0L , 25 km
jv /
/
Mantua (Y LP
La Co[oma <
N g ~ /7 0 20km
- I I l
84o00' 83030 .
Fig. 2. Location maps (A, C) and tectonic map (B) of the Guaniguanico terrane in the Pinar del Rfo and Havana provinces of western
Cuba. (A) Location of the area shown in (B). (B) Tectonic map of the Guaniguanico terrane (partly simplified, based mainly on data
taken from: Pszczdtkowski et al., 1975" Pszcz6tkowski, 1977, 1978, 1994b; Piotrowska, 1978; Martfnez and V~izquez, 1987); tectonic
units of the Sierra de los Organos belt: VP = Valle de Pons, I = Infierno, G = Sierra de Guane and Paso Real, V = Vifiales, PG = Pico
Grande, SG = Sierra de la Gtiira, A = Anc6n, APS = Alturas de Pizarras del Sur; CB = metamorphosed tectonic units of the Cangre
belt; tectonic units of the Southern Rosario belt: Z = La Zarza, T = Taco Taco, C = Caimito, CP = Cinco Pesos, LT = Los Tumbos, NP
= Niceto P6rez, M = Mameyal, LB = Los Bermejales, PU = Loma del Puerto, LP = La Paloma, LM = Loma del Muerto, J = gabbro
and serpentinite of the Jagua massif in the southwestern part of the Alturas de Pizarras del Norte; tectonic units of the Northern Rosario
belt: B V = Bel6n Vigoa, NO = Naranjo, D = Dolores, LS = La Serafina, CE = Cangre, CH = Sierra Chiquita, QS = Quifiones; GA
= Guajaib6n-Sierra Azul tectonic unit; N - Q = Neogene and Quaternary deposits south of the Cordillera de Guaniguanico; barbed lines
denote thrusts. (C) Location map of the Guaniguanico terrane belts: SO = Sierra de los Organos belt, CB = Cangre belt, SR = Southern
Rosario belt (in the Sierra del Rosario between Soroa and La Palma, and in the Alturas de Pizarras del Norte between La Palma, Mantua
and Guane), NR = Northern Rosario belt, LE = La Esperanza belt, GA = Guajaib6n-Sierra Azul belt; arrows indicate sense of the
movement along the Pinar fault.
boundary of the Guaniguanico terrane is located zoic successions of the northern Sierra del Rosario
in the southeastern Gulf of Mexico, north of the tectonic units and the La Esperanza belt (Fig. 2)
Pinar del Rio Province (Fig. 1B). To the southwest, are equivalents (Pszcz6tkowski, 1982, 1994a; Ro-
the Guaniguanico terrane is covered by Neogene- driguez, 1987). Consequently, the Esperanza belt is
Quaternary deposits. The wells drilled in the Gua- considered here as a continuation of the Northern
nahacabibes Peninsula revealed the metamorphosed Rosario belt.
rocks of the Guaniguanico terrane (Cangre belt) be- In the present paper, the Guaniguanico terrane
neath the Oligocene and Miocene rocks, about 55 is subdivided into the following tectonostratigraphic
km to the southwest of Guane (Fig. 1A). According belts (from south to north): Cangre (CB), Sierra de
to Rosencrantz (1996) the Guaniguanico terrane con- los Organos (SO), Southern Rosario (SR), Northern
tinues to the southwest as a fault-bounded, wedge- Rosario belt (NR) and the Guajaib6n-Sierra Azul
shaped block occurring between the Yucatfin basin (GA) (Fig. 2C). The stratigraphic successions of the
and the Yucatfin borderland (see also Fig. 1B). If this Cangre and Sierra de los Organos belts are simi-
interpretation is correct, the Guaniguanico terrane lar (Pszcz61kowski, 1985); however, the Cangre belt
may be about 400 km long. consists of metamorphic (mainly metasedimentary)
The thrust nappes of the Guaniguanico terrane rocks. The metamorphic Cangre belt consists of
consist of Jurassic to Paleogene rocks (Hatten, 1957; three tectonic units: Mestanza, Cerro de Cabras and
Pszcz6lkowski, 1971) and were formed during the Pino Solo (Piotrowska, 1978). These units occur in
Early Eocene (Pszcz6tkowski, 1977, 1994b). The the southeastern part of the Guaniguanico terrane,
Eocene tectonic deformation affected north-central as a narrow tectonic belt along the Pinar fault. The
Cuba as well. Rigassi-Studer (1963) distinguished Sierra de los Organos belt comprises the tectonic
the Sierra de los Organos and Sierra del Rosario units of (1) the Mogote zone and (2) Alturas de
as two distinct stratigraphic successions. The Meso- Pizarras del Sur (APS in Fig. 2B). In general, the SO

96 A. PSZCZOLKOWSKI
and CB represent the Jurassic platform that subsided

in the Tithonian and remained submerged in deep-
water pelagic conditions during the Cretaceous and
Paleocene.
The Rosario belts occur in the eastern and north-
ern parts of the Guaniguanico terrane (Fig. 2). The
Southern Rosario belt (SR) extends from Soroa to
Mantua and Guane. The 1:250,000 scale geological
map of Pszcz6lkowski et al. (1975), published as a
part of the geological map of Cuba (Puscharovsky,
1988), and the 1:50,000 scale map (Martfnez and
V~izquez, 1987) revealed that the southern Sierra del
Rosario tectonic units continue to the west, between
La Palma and Mantua (Fig. 2B and C). In fact, a
distinction between the SR and SO Jurassic lithology
is difficult in this area. Near Mantua, the Jurassic
(pre-upper Oxfordian) formations display features
characteristic for the Sierra de los Organos belt; to
the northeast these formations gradually change their
facies features reaching SR characteristics in the Mi-
nas de Matahambre-La Palma area. The Northern
Rosario belt (NR) occurs north of the SR (in the
Sierra del Rosario), and has its counterpart in the
Esperanza belt to the west and southwest.
The Jurassic to Paleogene rocks exposed in
the Martin Mesa tectonic window in the Havana
Province (located around the Martin Mesa 1 well,
but too small to be shown in Fig. 1) are similar to
Fig. 3. Lithologic column of the Martin Mesa 1 well, located
those known in the Guaniguanico terrane. The lime- northeast of Cayajabos, in the Havana Province (data from Se-
stones, shales and sandstones of Early Cretaceous gura Soto et al., 1985, simplified): 1 -- pelagic limestones, with
age drilled there in the Martin Mesa 1 well (Fig. 3) intercalations of sandstone and shale; 2 -- detrital limestones;
are equivalents of rocks occurring in the Northern 3 = sandstones, shales and marls; 4 --- tectonic contacts. The
Lower Cretaceous rocks and Campanian-Maastrichtian deposits
Rosario belt (Fig. 4). The Jurassic-Cretaceous and
(the Cacarajfcara Formation?) are equivalents of the formations
Paleocene rocks of this belt (NR) differ from those exposed in the Northern Rosario belt of the Guaniguanico terrane
occurring in the SO and SR to the south. In general, (see Fig. 2B).
the Rosario belts are interpreted as the continental
margin slope and (partly) adjacent basin floor.
The Guajaib6n-Sierra Azul belt 1 (GA) occurs as If so, the GA unit may represent a separate terrane
a single (and narrow) tectonic unit exposed between not related to the Guaniguanico tectonostratigraphic
the Northern Rosario belt and the Bahia Honda belts.
terrane (Fig. 2). The GA consists mainly of the The Paleogene thrusting changed the original
Cretaceous shallow-water carbonates (Pardo, 1975), relative positions of the belts and tectonic units
named the Guajaib6n Formation (Herrera, 1961; of the Guaniguanico terrane. Contrasting opinions
Pszcz6lkowski, 1978, 1982). Contrary to earlier in- have been expressed on the problem of the pre-tec-
terpretations (Pardo, 1975; Pszcz6lkowski, 1982) the tonic restoration of the Guaniguanico belts and units
pre-tectonic location of the Guajaib6n-Sierra Azul (Hatten, 1957; Rigassi-Studer, 1963; Pszcz6tkowski,
belt was probably to the south, but not necessar- 1978; Mossakovskiy and de Albear, 1979; Pi-
ily atop the deep-water Lower Cretaceous deposits otrowska, 1993; and others). In the present paper,
of the Northern Rosario belt as suggested by Itur- the author accepts the idea, that during the Paleo-
ralde-Vinent (1994, 1996). Rather, this belt could gene thrusting the relative positions of the belts and
be situated somewhere at the Yucat~in block edge. tectonic units were completely reversed (Iturralde-
Vinent, 1994, 1996). According to this interpretation,
the ophiolites and the Cretaceous volcanic arc rocks
1The original name of this belt (Cacarajfcara Belt Pardo, of the Bahia Honda composite terrane are the struc-
1975) cannot be maintained because of the existence of the
Cacarajfcara Formation established earlier (Hatten, 1957) in turally highest belts of the Pinar del Rio Province
western Cuba. This formation occurs in the Rosario and La (Hatten, 1957; Pszcz6tkowski and de Albear, 1982;
Esperanza belts only. Pszcz6lkowski, 1990; Iturralde-Vinent, 1994).

Fig. 4. Generalized lithostratigraphic scheme of the Guaniguanico terrane in western Cuba (for location of belts see Fig. 2C); lithology:
1 -- sandstones and shales with intercalations of limestone, 2 -- mafic rocks, 3 = fossiliferous limestones and shales, 4 ---=thick-bedded
to massive carbonates (Jurassic in age), 5 = thin- and medium-bedded limestones, 6 = limestones, shales and sandstones (Polier
Formation), 7 = radiolarian cherts and shales, 8 = massive, shallow-water limestones (Cretaceous in age), 9 = shales and sandstones
(Late Cretaceous and Paleogene in age), 10 - detrital limestones and breccia (Cacarajfcara Formation), 11 = limestone and chert
breccias (Anc6n Formation), 12 - Paleogene olistostrome; lithostratigraphic units (circled letters): SC = San Cayetano Formation, A C --
Arroyo Cangre Formation (equivalent to the San Cayetano Formation), ES = E1 Sfibalo Formation, J Jagua Formation, F = Francisco
=
Formation, S V = San Vicente Member of the Guasasa Formation, G = Guasasa Formation, AR -- Artemisa Formation, PL -- Polier
Formation, L = Lucas Formation, ST = Santa Teresa Formation, GB -- Guajaib6n Formation, P N = Pons Formation, CT = Carmita
Formation, PA = Pinalilla Formation, MR = Moreno Formation, PS = Pefias Formation, CA = Cacarajfcara Formation, A N -- Anc6n
Formation, M N -- Manacas Formation.
The presence of the Upper Jurassic shallow-water higher tectonic units of the Sierra de los Organos
limestones in the Pinar 1 well (Fig. 5) is a strong belt ( 3 0 0 - 6 0 0 m), and eventually wedge out in the
argument in favor of the above-mentioned tectonic overlying tectonic units of the Southern Rosario belt
restoration. This well was located 4 km south of (Pszcz6tkowski, 1978).
Pons in the Sierra de los Organos (L6pez Rivera
et al., 1987; Pszcz6lkowski, 1994), in a tectonic
window occurring in the central, most uplifted zone OUTLINE OF STRATIGRAPHY
of the Guaniguanico terrane. The Valle de Pons unit
(VP in Fig. 2B) is the lowermost tectonic element R e m a r k s on lithostratigraphic s c h e m e
exposed in the Sierra de los Organos (Piotrowska,
1978) and probably in the whole Guaniguanico The lithostratigraphic scheme for the Guanigua-
terrane. At least three tectonic units were drilled in nico terrane, as used in this paper (Figs. 4 and 6), was
the Pinar 1 deep well (Fig. 5). The bottom unit, developed during the last 18 years (Pszcz6tkowski
reaching below 3300 m, contains very thick Upper et al., 1975; Pszcz6tkowski, 1978, 1982, 1994a).
Jurassic shallow-water limestones (1500 m). L6pez Other authors (Iturralde-Vinent, 1994, 1996; Co-
Rivera et al. (1987) suggested that these limestones biella-Reguera, 1996) accepted this scheme, although
belong to the autochthonous unit. However, there with some minor changes. The results of a recent mi-
is no conclusive evidence that the Pinar 1 well cropaleontological study on the Paleogene deposits
really penetrated the whole Guaniguanico nappe (Bralower et al., 1993; Bralower and Iturralde-Vinent,
pile and entered into the autochthonous sedimentary 1997) also have been considered.
succession. The work of Hatten (1957, 1967) was fun-
In any case, the Upper Jurassic shallow-water damental for developing the m o d e m stratigraphic
carbonates attain their m a x i m u m thickness (1500 m) scheme for the Sierra de los Organos belt (see also
in the subsurface. These rocks are thinner in the Khudoley and Meyerhoff, 1971). Herrera (1961)

98 A. PSZCZOLKOWSKI
Sierra de los Southern Northem

PINAR 1 Organos belt Rosario belt Rosario belt
m 0 EARLY
EOCENE
PALEOCENE
MANACAS FM. ;'i... . . . . . . .
i
Vieja Member
......................
Pica Pica Member
1I
IVlAASTRICHTIAb CACARAJiCARA FM.

Lower Cretaceous (Aptian - Albian)
CAMPANIAN
Lower Cretaceous (Valanginian)
S~-E~-%5]q]-R~--
Lower Cretaceous (Aptian - Albian) CONIACIAN
1000 "-~,~'~.~'t TURONIAN
CARMITA FM.
CENOMANIAI~
Upper Jurassic
APTIAN-ALBIAI~ PONS SANTA TERESA FM.
BARREMIAN
FORMATION
VALANGINIAN
! Tumbitas Mb. I
~<,.,,< BERRIASIAN <Z~ ~u-r~i~aer~' ' u_m.i,:r~Mb' :1
Upper Jurassic 0B Member ~ ARTEMISA
2000 '< ''< ''<
,~-.~-) < ~- FE~i~,?rie-a-ca-n~
....... q
TITHONIAN u) ~ : , Member
< p ............... La Zarza I
,'-,>,3 ,,-< D ~ i i San Vicente Member ', FORMATION
KIMMERIDGIAh
o o, Member ~;:, ',
Lower Eocene i .... i
- 5
Lower Cretaceous (Aptian - Albian) FRANCISCO FM.
300C --
Lower Cretaceous (Berriasian - ?Barremian) -L_ FM.
Upper Jurassic " I SAN CAYETANO

Lower Cretaceous (Berriasian - Valanginian)
Lower Cretaceous (Aptian - Albian) FORMATION
Lower Cretaceous (Berriasian - ?Barremian)
N M'-N'-M
~ - ~ - ~
Fig. 6. Lithostratigraphic scheme of the Sierra de los Organos
and Rosario belts in the Guaniguanico terrane: L. = lower (Ox-
mm-- 5 fordian), U. = upper (Oxfordian), S.V. = San Vicente Member of
~ - ~ - ~
the Artemisa Formation.
Upper Jurassic
~ - ~
of the San Cayetano Formation. The rocks are

5000 --' ~ ~ ~
dark-gray to black; they are rhythmically bedded
(Fig. 7). The formation is deeply weathered so that
Fig. 5. Lithologic column of the Pinar 1 well, located 4 km south the exposures favorable for sedimentological obser-
of Pons in the Sierra de los Organos belt (after L6pez Rivera vations occur mainly in some streams and rivers. The
et al., 1987, simplified): 1 = massive, shallow-water limestones, sedimentary structures vary among distinct tectonic
2 - Berriasian-?Barremian pelagic limestones, 3 -- Aptian-
units (Meyerhoff and Hatten, 1974). Any reliable
Albian pelagic limestones, 4 -- Lower Eocene olistostrome, 5 --
tectonic contacts. lithostratigraphic subdivision of the monotonous and
often severely tectonized formation, 1000 to 3000(?)
m thick, has not been well established so far.
introduced many lithostratigraphic names for the Haczewski (1976) proposed a general descriptive
Sierra de los Organos area, but only few were model of sedimentation of the San Cayetano For-
valid and have been incorporated into the modern mation. This author distinguished nine facies (A-I)
scheme (Fig. 6). Imlay (1942), de la Torre (1960, within the studied sections. In his opinion, the fa-
1988), Furrazola-Bermddez (1965), Judoley and Fur- cies A - F occur in the Sierra de los Organos belt.
razola-Bermddez (1968), Wierzbowski (1976), My- However, the type sections of the facies A - C and
czyfiski (1976, 1977, 1989, 1994a,b), Myczyfiski E are now considered to belong to the Southern
and Pszcz6tkowski (1976, 1990, 1994) and Kutek et Rosario belt (Pszcz61kowski, 1994b). The facies A -
al. (1976) studied fossils and/or biostratigraphy of D consists of sandstones with some subordinate
the Jurassic and Lower Cretaceous formations in the pebbly sandstones or conglomerates, fine-grained
Sierra de los Organos belt and Rosario belts. sandstones with trough cross-lamination, of bed-
ded siltstones, shales, and rarely very fine-grained
San Cayetano Formation (?Lower Jurassic to sandstones, siltstones, fine-grained sandstones and
middle Oxfordian) thin-bedded shales. The rocks belonging to these
facies have been interpreted as deposited in a flu-
Lithology and facies model. This formation vial environment, and partly in a shallow marine
consists of shales, siltstones and sandstones with or beach environment. The facies E and F consist
some intercalations of conglomerates and lime- of black shales, sometimes with septarian nodules
stones. Limestones occur mostly in the upper part and pyrite spheres in facies E deposits. Both facies

Fig. 7. Shales, siltstones and fine-grained sandstones of the San Cayetano Formation at Cinco Pesos (Southern Rosario belt). These
deposits are similar to facies G and H of Haczewski (1976).
most likely have originated in extensive lagoons with spp., Gervillaria sp., and Neocrassina spp. In the
restricted circulation. Sierra de los Organos belt, bivalves belonging to the
The facies G-I (rhythmic sandstones and shales genus Gryphaea Lamarck probably also appear in
with graded and ripple bedding, alternating graded the coquinid limestones of the upper part of the San
sandstones and shales, and thick-bedded sandstones) Cayetano Formation.
occur only in the Southern Rosario belt. The facies Facies interpretation. According to Haczewski
G consists of fine-grained sandstones, siltstones and (1976), the San Cayetano rocks were deposited on a
shales. Deposits similar to facies G and H are ex- coastal alluvial plain (facies A-C) by a river trans-
posed in the Cinco Pesos area (Fig. 7). The thick-bed- porting the material a few hundred kilometers from
ded, coarse-grained sandstones, sometimes with peb- the south (or southwest). The sediments debouched
bles up to 7 cm long, are also known in the Cinco to the sea formed an arcuate delta and some of it
Pesos area; the sandstones belong to the facies I of was redistributed by a longshore drift and turbidity
Haczewski (1976). Fossiliferous pebbles, containing currents. Deposits distinguished as facies G probably
late Paleozoic foraminifers and bryozoans, have been accumulated on the slope of the continental margin.
found in these sandstones (Pszcz6tkowski, 1989b). Facies H and I were characterized as deposits of a
The foraminifers belong to Fusulinacea (Schwage- submarine fan accumulating at the base of the slope.
rina sp. and Parafusulina sp.) of Permian age. One The development of facies H is intermediate be-
specimen was identified as Tetrataxis sp. tween normal and proximal turbidites, while facies I
Fossils and age. The age of the San Cayetano is of a proximal character (Haczewski, 1976).
Formation (?Early Jurassic-middle Oxfordian) is The relationship between the diagrams of pale-
well established in the uppermost part of this unit ocurrent measurements and the geographical distri-
only. In the Sierra del Rosario belt, Myczyfiski and bution of the localities studied by Haczewski (1976)
Pszcz6tkowski (1976) have found some ammonites changes after the restoration of tectonic units. The
in the uppermost part of the San Cayetano Forma- resulting paleocurrent pattern indicates that the di-
tion, southeast of La Palma. These ammonites be- rections from south or south-southwest predominate
long to the following taxa: Perisphinctes (?Dichoto- in the northwestern part of the San Cayetano sedi-
mosphinctes) cayetaensis Myczyfiski, 1976, P. (?Di- mentary basin (APS in the Sierra de los Organos belt
chotomosphinctes) cf. anconensis Sfinchez Roig, and m Fig. 2B), and the directions from northeast prevail
P. (Discosphinctes) cf. pichardoi Chudoley et Fur- in its southeastern sector (Southern Rosario belt).
razola-Berm6dez. The ammonites indicate that the
uppermost part of the San Cayetano Formation is of El S~ibalo Formation (Oxfordian)
middle Oxfordian age. In the Sierra de los Organos
belt, the siliciclastic deposits of the San Cayetano Lithology and boundaries. The E1 Sfibalo
contain very scarce macrofossils, mainly bivalves. Formation occurs in the Northern Rosario belt
Pugaczewska (1978) identified the following taxa: (Pszcz6tkowski, 1994a). This unit, up to 400 m
Eocallista (Hemicorbula) spp., Vaugonia (Vaugonia) thick, consists of basalts and diabases with interbed-

100 A. PSZCZOLKOWSKI
ded limestones, marls and sometimes shales. These well-bedded coquinas and bioclastic limestones. The
horizons of pyroclastic rocks also have been ob- beds and lenses of silicified limestones are com-
served. The basalts are massive or pillowed flows mon. The coquinas beds are 0.2 to 1.5 m thick. The
(Pszcz6tkowski and de Albear, 1983). The lower bivalve-echinoderm and bivalve-oolitic microfacies
boundary of the E1 S~ibalo Formation is tectonic; are very common in the Pan de Azfcar Member. The
this unit contacts with various Cretaceous and Pa- coquina beds consist mainly of Gryphaea mexicana
leogene formations. Locally, the E1 S~ibalo Formation Felix (Pugaczewska, 1978). The bivalve shells are
is overlain by thin San Cayetano Formation silici- commonly encrusted by agglutinated foraminifers.
clastics, but often contacts with the ?late Oxfordian- The limestones of this member were deposited at
Kimmeridgian limestones of the Artemisa Formation. a depth ranging from a few up to some tens of
The E1S~ibalo/Artemisa boundary is tectonically dis- meters, in proximity of some oolitic shoals. The
turbed in some sections, but in places a thin bed of Zacarfas Member, not shown in Fig. 6, consists of
sedimentary breccia separates the two formations. ammonite-beating shales with thin intercalations of
Age. The Jurassic age of the E1 S~ibalo Forma- siltstones and bivalve coquinas with Liostrea, Os-
tion (Oxfordian-?early Kimmeridgian) was defined trea, Exogyra and Plicatula (Wierzbowski, 1976).
on the basis of infrequent microfossils occurring This member attains 40 m in thickness and is of
in the limestones intercalating with the basalts and middle Oxfordian age (op. cit.). The Jagua Vieja
diabases (Pszcz6tkowski, 1989a, 1994a). According Member comprises black shales and marly lime-
to Cobiella-Reguera (1996), these rocks overlie the stones, up to 60 m thick. The calcareous concretions
San Cayetano Formation clastics and may be as old contain numerous ammonites, fish remains, marine
as Callovian in the lowermost part of the E1 S~ibalo reptile bones (Iturralde-Vinent and Norell, 1996) and
Formation. The alleged position of the E1 S~ibalo bivalves. The Pimienta Member occurs in the upper
Formation rocks above the San Cayetano Formation part of the Jagua Formation (Fig. 6). This unit, up
is not supported by the geological relations visi- to 60 m thick, consists of well-bedded, dark-gray
ble in outcrops described so far from the Sierra to black micritic limestones with some minor shale
del Rosario. In fact, the San Cayetano-type clastic intercalations in the lower part of the member.
deposits (up to 15 m thick) were observed in few Age. The ammonites indicate that the Zac-
places above the rocks of the E1 S~ibalo Formation in arras Member represents the middle Oxfordian
the Northern Rosario belt (Pszcz6tkowski, 1994b). (Wierzbowski, 1976). As judged from the relation
Therefore, it seems that the E1 S~ibalo Formation is to the Zacarfas and Jagua Vieja members, the Pan
(locally?) older than the Francisco Formation (late- de Azt~car Member is probably middle Oxfordian
middle to late Oxfordian). Consequently, the age of in age. The Jagua Vieja Member has been assigned
the E1 S~ibalo Formation should be pre-late Oxfor- to the middle Oxfordian on the basis of well pre-
dian, in some sections at least (Fig. 6). The supposed served ammonites (Wierzbowski, 1976). Myczyfiski
Callovian age of the lower part of the E1 S~ibalo (1976) described from the Pimienta Member several
Formation is, however, still not confirmed by any ammonite species assigned to Mirosphinctes and
paleontological or radiometric data. Cubaspidoceras. One specimen of Taramelliceras
Facies interpretation, The limestones of the E1 (Metahaploceras) sp. also has been found. This
S~ibalo Formation have been deposited in anaerobic ammonite assemblage was assigned to the upper
conditions, below the wave-base, in the outer neritic Oxfordian (Myczyfiski, 1994a).
zone. This origin of this formation was related to the Facies interpretation, The bioclastic limestones
more advanced rifting stage during ?Middle to early and coquinas of the Pan de Azfcar Member were
Late Jurassic times. According to Iturralde-Vinent deposited at a paleodepth ranging from a few to
(1995a), the magmatic activity of continental passive some tens of meters, in the proximity of some
margin represented in Cuba was time coincident oolitic shoals. The shales of the Zacarfas Member
with continental rifting until the Oxfordian, and with were accumulated in a deeper part of the shelf
oceanic spreading in the Caribbean area since the environment adjacent to a delta. The shales and
Late Jurassic. limestones of the Jagua Vieja Member and the
limestones of the Pimienta Member were deposited
Jagua Formation (middle and upper Oxfordian) in an outer shelf environment.
Subdivision and lithology. The Jagua Forma- Francisco Formation (middle to upper
tion occurs in the Sierra de los Organos belt. This Oxfordian)
formation consists of limestones and shales subdi-
vided into four subunits, namely the Pan de Azfcar, Lithology. In the Sierra del Rosario, the Fran-
Zacarfas, Jagua Vieja and Pimienta members. The cisco Formation is an equivalent of the Jagua Forma-
Pan de Azficar Member (Fig. 6) consists of black, tion. The Francisco Formation consists of shales, mi-

limestone breccia (sharpstone) separating the mas-

sive limestones of the Guasasa Formation from the
Jagua Formation (Hatten, 1957). A description of
the microfacies of San Vicente Member was given
by Pszcz6lkowski (1978).
The E1 Americano Member comprises well-bed-
ded, dark-gray to black limestones, up to 45 m thick.
The Saccocoma-Didemnidae microfacies is charac-
teristic for the limestones of the lower part of this
member, while biomicrites with calpionellids and
radiolarians occur in its upper part (Myczyfiski and
Pszcz6lkowski, 1990). The E1 Americano Member
terminates the Jurassic part of the Guasasa For-
mation (Fig. 6). The Tumbadero Member consists
Fig. 8. Spicules of Didemnidae (Didemnum carpaticum Migik of well-bedded, often laminated, radiolarian biomi-
et Borza and Didemnum sp.) in a microsparitic limestone of crites and calcilutites with black chert intercalations.
the Francisco Formation (middle and upper Oxfordian), Southern
The thickness of this unit ranges from 20 to 50
Rosario belt; • 220.
m. The Tumbitas Member consists of thick-bedded,
light-gray calpionellid, calpionellid-radiolarian and
critic limestones, and thin sandstone intercalations. nannoconid biomicrites with infrequent thin inter-
Calcareous concretions occur in shales, sometimes calations of dark-gray or reddish limestones. The
with bivalves and/or ammonites. At Cinco Pesos, thickness of these deposits is about 40 m, but in
a volcanic rock (basalt) half a meter thick occurs some sections attains 80 m.
within limestones and sandstones. The formation Age. The ?late Oxfordian-Kimmeridgian age of
attains 25 m in thickness. the San Vicente Member is defined mainly on the
Age. The Francisco Formation contains some am- basis of the lithostratigraphic position of this unit
monites, rare bivalves, fish and plant remains. Some (Fig. 6).
limestone beds contain Globochaete alpina Lombard The late Oxfordian age of the basal San Vicente
and spicules of Didemnidae (Fig. 8), while radio- Member limestones is suggested by the ammonite-
larians were not observed in thin sections. The am- beating Jagua Formation occurring below the mas-
monites indicate the upper part of the middle Oxfor- sive limestones of this unit (Wierzbowski, 1976;
dian and also the lower part of the upper Oxfordian Myczyfiski, 1976). However, the limestone breccia
(Kutek et al., 1976; Myczyfiski, 1976, 1994a). locally separating the San Vicente Member from the
Facies interpretation. The deposits of the Fran- Pimienta Member of the Jagua Formation indicates
cisco Formation were accumulated in an outer that upper Oxfordian limestones had been partly
shelf environment. This environment was proba- eroded before the deposition of the shallow-water
bly slightly deeper than in the case of the Jagua San Vicente carbonates. A stratigraphic hiatus of
Formation deposits. unknown duration could be related with this (latest
Oxfordian?) erosive event. The Kimmeridgian age of
Guasasa Formation (?upper Oxfordian- the San Vicente Member carbonates was confirmed
Valanginian) by some microfossils (Fermindez Carmona, 1989).
The E1 Americano Member of the Guasasa For-
Subdivision and lithology. The Guasasa Forma- mation is Tithonian in age. Ammonites collected at
tion has been distinguished in the Sierra de los the base of the E1 Americano Member are early
Organos belt. This formation is subdivided into four Tithonian in age (Houga, 1974; Myczyfiski, 1989).
subunits, namely the San Vicente, E1 Americano, Recently, few specimens belonging to the genus
Tumbadero and Tumbitas members (Fig. 6). Hybonoticeras have been identified in the Sierra
The San Vicente Member, 300-650 m thick, oc- de los Organos belt (Myczyfiski, 1996a). These
curs in the lower part of the Guasasa Formation. This ammonites indicate the Hybonoticeras-Mazapilites
member consists of massive or thick-bedded, gray Zone in the Sierra de los Organos belt (Fig. 9).
to black limestones, often dolomitized, sometimes Their presence suggests that the age of the upper
with chert nodules and lenses. Micritic limestones boundary of the San Vicente Member is close to
with Favreina predominate in the lower part of the the Kimmeridgian-Tithonian boundary. The upper
member, while oncolitic and algal calcarenites oc- Tithonian ammonite assemblage (Myczyfiski, 1989,
cur in the upper part. At the top, there are also 1994a, 1996a,b; Myczyfiski and Pszcz6~kowski,
well-bedded limestones up to a dozen meters thick. 1990) contains some cosmopolitan taxa (Duran-
The San Vicente Member includes a sedimentary gites, Corongoceras, Kossmatia), those known from

102 A. PSzCZOLKOWSKI
LU were deposited on the shallow-water carbonate bank

Substage Ammonite zones (Pszcz6ikowski, 1978, 1981) or platform. The Lower
!--
O9 Tithonian S a c c o c o m a - D i d e m n i d a e biocalcisiltites of
the E1 Americano Member accumulated in a deeper
(outer neritic) environment. The upper Tithonian
Durangites - Himalayites -
z Upper biomicrites of this member were deposited in an
< Hildoglochiceras (Salinites) outer neritic to bathyal environment. The radiolarian
z and/or calpionellid biomicrites of the Tumbadero
0 Paralytohoplites carribeanus and Tumbitas members are bathyal deposits and
-1-
!- Pseudolissoceras spp. were laid down below the aragonite compensation
~_ Lower
depth.
Hybonoticeras - Mazapilites Artemisa Formation (upper Oxfordian-

Valanginian)
Fig. 9. Ammonite zones established in the Tithonian limestones
of the E1 Americano Member (the Guasasa Formation), eastern Subdivision and lithology. The Artemisa Forma-
part of the Sierra de los Organos belt (Myczyfiski, 1996a,b and tion is subdivided into three members: San Vicente,
pers. commun., 1996). La Zarza and Sumidero (Fig. 6). The San Vicente
Member was recognized but in a few sections of the
Southern Rosario belt. In the Northern Rosario belt,
Mexico (Hildoglochiceras (Salinites), Proniceras), the thick-bedded dolomitized limestones occurring
as well as endemic taxa (Vinalesites rosariensis in the lower part of the Artemisa Formation are
(Imlay), Protancyloceras hondense Imlay, Butticeras lithologic equivalent to the San Vicente Member.
butti (Imlay) and Butticeras antilleanum (Imlay)). The La Zarza Member consists of bedded (0.1 to
Also bivalves Buchia aft. B. okensis (Pavlov) and 0.8 m), gray to black micritic limestones (Figs. 10
Buchia aft. B. piochii (Gabb) have been reported and 11) with some intercalations of shale, siltstone
from the upper Tithonian limestones of the Sierra de and fine-grained sandstone in the lower part of this
los Organos belt (Myczyfiski, 1989). unit. The limestone beds contain rare aptychi and
The ammonites are rare and poorly preserved in fish remains. In the upper part of the member,
the deposits of the Tumbadero and Tumbitas mem- there are calcilutites interbedded with dark-gray to
bers of the Guasasa Formation. Therefore, stratig- black bioclastic limestones and coquinas composed
raphy of these members is based on calpionellids. of ammonite shells and aptychi. The thickness of
The Tumbadero Member is Berriasian in age, while the La Zarza Member attains 200 m. The limestones
the Tumbitas Member is of late Berriasian-early of this member are overlaid by light-brown, rose
Valanginian age (Pszcz61kowski, 1978). and gray to black biomicrites with intercalations of
Facies interpretation. The ?late Oxfordian- radiolarian chert, assigned to the Sumidero Mem-
Kimmeridgian rocks of the San Vicente Member ber (Pszcz6tkowski, 1978). The limestones contain
Fig. 10. Kimmeridgian limestones of the La Zarza Member (Artemisa Formation) exposed in a quarry west of La Palma, Southern
Rosario belt.

The calpionellids document a late Berriasian-

early Valanginian age of the lower and middle parts
of the Sumidero Member. Ammonites attributed to
Thurmanniceras cf. novihispanicus (Imlay) indicate
a Valanginian age (Myczyfiski, 1977) and nannofos-
sils (Nannoconus spp.) suggest a late Valanginian
age for the upper part of the Sumidero Member.
The mentioning of the presence of Southern
Boreal/Northern Tethyan faunas (30~ in the
Valanginian deposits of the Sierra de los Organos
and Sierra del Rosario (Pessagno et al. in Chapter
5) needs a comment. Two specimens of Buchia sp.
figured by Myczyfiski (1977, pl. 8: 6-7) were col-
lected from the Artemisa Formation at La Catalina.
The Tithonian limestones are exposed at this locality
(Myczyfiski, 1989; Myczyfiski and Pszcz6tkowski,
1994), but no Cretaceous rocks with macrofauna are
known there. Therefore, these specimens of Buchia
sp. are late Tithonian (or earliest Berriasian?) in
age. The specimens of Vinalesites rosariensis (Im-
lay) figured by Myczyfiski (1977, pl. 3: 1), were
also found in the Artemisa Formation at La Catalina
Fig. 11. Black Tithonian limestones of the La Zarza Member,
Artemisa Formation, exposed in a road-cut west of Cinco Pesos, and are late Tithonian in age. In fact, specimens
Southern Rosario belt. of Buchia from the (stratigraphically well docu-
mented) Lower Cretaceous deposits of the Sierra del
Rosario and Sierra de los Organos, were not fig-
abundant radiolarians. Calpionellids are common in ured in any paper. In this situation, the presence of
the lower part of the member, while numerous apty- a Southern Boreal/Northern Tethyan macrofauna in
chi occur in its upper part. The Sumidero Member the Valanginian deposits of the Guaniguanico terrane
attains 200 m in thickness in some sections. needs confirmation or should be rejected.
Fossils and age. The Cubaspidoceras-Miro- Facies interpretation. The deposits of the lower
sphinctes assemblage has been identified in the part of the La Zarza Member accumulated in rel-
basal limestone beds of the Artemisa Formation atively quiet conditions of a partly restricted (la-
(Kutek et al., 1976). These ammonites indicate gunal?) environment. The limestones of the upper
the late Oxfordian age of these deposits. Conse- part of the La Zarza Member were deposited in the
quently, the lower part of the La Zarza Mem- outer neritic environment, as indicated by its faunal
ber is late Oxfordian-Kimmeridgian in age. The content. Pelagic deposits of the Sumidero Member
Tithonian ammonite assemblage occurs in the up- accumulated in bathyal zones below the aragonite
per part of the La Zarza Member. These am- compensation depth, as evidenced by the relative
monites were studied by Imlay (1942), Judoley abundance of aptychi and scarcity of ammonites in
and Furrazola-Bermfidez (1968), Houga (1974), and the radiolarian limestones.
Myczyfiski (1989, 1994b). The taxa Paralytoho-
plites, Butticeras, Vinalesites, Protancyloceras and Polier Formation (upper Berriasian-?Albian)
Hildoglochiceras (Salinites) are characteristic for
the Tithonian of the Rosario belts (Myczyfiski and Lithology. This formation consists of gray
Pszczdtkowski, 1994; Myczyfiski, 1996a,b). The bi- pelagic limestones with intercalations of turbiditic
valves occurring in the Tithonian limestones of sandstones and shales (Fig. 12). Radiolarian or
the Rosario belts belong mainly to the genus In- radiolarian-spicule microfacies are typical of the
oceramus (Parkinson, 1819), although representa- limestones. Sometimes the sandstones and shales
tives of Buchiidae are also present (Myczyfiski and may be distinguished as a distinct subunit in the
Pszczdtkowski, 1994; Myczyfiski, 1994b). topmost part of the formation (the Roble Member).
The Saccocoma-Didemnidae microfacies pre- The Polier Formation attains its maximum thickness
dominates in the lower Tithonian limestones of the (about 300 m) in the Cangre tectonic unit (CE in
Southern Rosario belt, being replaced gradually by Fig. 13). The Polier Formation occurs in the North-
the radiolarian microfacies in the Chitinoidella spp. ern Rosario belt. Similar rocks of Early Cretaceous
Zone and lower part of the Crassicollaria Zone age are also known in the Esperanza belt (Fig. 2).
(Myczyfiski and Pszcz6tkowski, 1994). Age. Calpionellids indicate the late Berriasian age

104 A. P S Z C Z O L K O W S K I
Fig. 12. Limestones, sandstones and shales of the Polier Formation (Lower Cretaceous), exposed west of Cayajabos, Northern Rosario
belt; hammer 28 cm long.
Fig. 13. Palinspastic restoration of relative position and width of the major tectonic units occurring in the Guaniguanico terrane (Fig.
2B). The highest units of the Northern Rosario belt were originally located far to the southeast (Iturralde-Vinent, 1994; Pszcz6tkowski,
1994b), while the structurally lowermost units exposed in the Sierra de los Organos belt (VP, /) are shown at the opposite, northwest
extreme of this pre-tectonic reconstruction. Tectonic units: VP = Valle de Pons, I = Infierno, V = Vifiales, A = Anc6n, PG & SG =
Pico Grande and Sierra de la Gtiira, A P S = Alturas de Pizarras del Sur, CB -- Cangre belt (Cerro de Cabras, Mestanza and Pino Solo
metamorphosed units), C + T = Caimito and Taco Taco, Z = La Zarza, A P N = Alturas de Pizarras del Norte (Loma del Muerto and
La Paloma units), CP -- Cinco Pesos, MA = Mameyal, B V = Bel6n Vigoa, NO -- Naranjo, CE = Cangre, CH = Sierra Chiquita, QS =
Quifiones. Lithology: 1 = sandstones and shales, 2 -- mafic rocks, 3 = fossiliferous limestones and shales, 4 -- massive shallow-water
limestones, 5 = thin- to medium-bedded limestones, 6 = limestones, shales and sandstones (Polier Formation), 7 = radiolarian cherts and
shales, 8 = sandstones and shales (Late Cretaceous and Paleogene in age), 9 -- detrital limestones and breccia (Cacarajfcara Formation),
10 = limestone and chert breccias (Anc6n Formation), 11 = Paleogene olistostrome; lithostratigraphic units (encircled letters) as in
Fig. 4.

of the basal beds of the Polier Formation. Ammonite brown and sometimes black radiolarian cherts and
imprints are common in the middle part of this radiolarites with thin interbeds of silicified shales.
formation. Valanginian to ?Albian taxa have been Nannoconus-radiolarian biomicrites and thin tur-
identified by Myczyfiski (1977). However, the bulk bidites occasionally occur in some sections. The
of the deposits is Hauterivian-Barremian in age. thickness of the formation attains 40 m.
Rhyncholites (Hou~a, 1969) and aptychi are com- This formation is exposed in the La Esperanza
mon in the Polier Formation. The lower boundary of belt, the Northern and Southern Rosario belts, in
the Polier Formation is diachronous. the Matanzas Province, in central Cuba and the
Facies interpretation. The turbidites of siliciclas- Camagtiey Province. The author has not seen the
tic and mixed (siliciclastic and calcarenitic) compo- Santa Teresa Formation deposits in the Sierra de los
sition are the most characteristic constituent of the Organos belt. Nevertheless, Iturralde-Vinent (1996)
Polier Formation. The sandstone beds are 0.02-1.0 suggests the presence of a radiolarian chert unit
m thick and most of them are graded, with horizontal within the pelagic limestones and cherts of the Pons
and cross-lamination in their upper part (the Bouma Formation.
sequence). Flute, groove and prod casts are very fre- Age. Infrequent planktonic foraminifers of
quent on the soles of sandstones. Except some thick Albian-Cenomanian age occur in the Santa Teresa
sandstone beds of the topmost part of the formation, Formation. The lithostratigraphic position suggests
these deposits display features of distal turbidites. that deposition of this formation began during the
Paleocurrent measurements indicate paleoflow to- Hauterivian in the Southern Rosario belt and during
ward the south and southeast (Pszcz6tkowski, 1982). the Aptian in the Northern Rosario belt. Radiolari-
The pelagic limestones and turbidites accumulated ans identified by Aiello and Chiari (1995) indicate
in the deep-water part of the basin, above the calcite a Valanginian to middle Aptian age of one sam-
compensation depth. ple taken in the transitional beds from the Polier
Formation to Santa Teresa Formation.
Lucas Formation (upper Hauterivian- Facies interpretation. The siliceous rocks of this
Barremian) formation originated in the deep-water environment,
near (and sometimes below) the calcite compensa-
Lithology. The Lucas Formation consists of tion depth. Minor terrigenous input was still active
thin-bedded black limestones with occasional inter- during deposition of the lower part of the Santa
calations of fine-grained sandstones and shales. The Teresa Formation in the Northern Rosario belt.
limestones are radiolarian biomicrites. The thickness
of this formation attains 300 m. Pons Formation (?upper Valanginian-Turonian)
Age. The Lucas Formation is the stratigraphic
equivalent of the Polier Formation (Fig. 6) in the Lithology. The Pons Formation consists of gray
Quifiones tectonic unit (Fig. 13). The aptychi and to black micritic limestones interbedded with cherts.
ammonite imprints are the only macrofossils col- The amount of chert intercalations is variable verti-
lected from the Lucas Formation. The ammonites cally and laterally. The nannofossil-radiolarian lime-
indicate the late Hauterivian to early Barremian age stones and radiolarian cherts are common in the
of this formation (Myczyfiski, 1977). The top of Pons Formation, although other microfacies types
the Polier and Lucas formations marks the upper are also present in this pelagic sequence. The Pons
boundary of the Upper Jurassic-Lower Cretaceous Formation is developed in the lower tectonic units
limestone sequence in the Northern Rosario belt. of the Sierra de los Organos belt only (Fig. 13).
Both formations occur directly below the radiolarian The total thickness of Pons Formation is 120 to
cherts and shales of the Santa Teresa Formation. 150 m.
In the NR this important facies change occurred Age. The ?late Valanginian age of the lowermost
significantly later than in the SR (Figs. 4 and 6). part of Pons Formation may be assumed on the basis
Facies interpretation. The ammonite-bearing ra- of (1) the lithostratigraphic position of this unit,
diolarian limestones of the Lucas Formation are in- and (2) the presence of calpionellids (Tintinnopsella
terpreted as pelagic deposit accumulated in bathyal carpathica Murgeanu et Filipescu) in the limestones
environment, less influenced by influx of terrigenous exposed at the base of the type section south of Pons
sediment. (Pszcz6tkowski, 1978; de la Torre, 1988). Hatten
(1957) considered the upper boundary of the Pons
Santa Teresa Formation (?Hauterivian- Formation to be of Turonian age.
Cenomanian) Facies interpretation. The pelagic limestones
and radiolarian cherts were deposited in a deep
Lithology and distribution. The Santa Teresa bathyal environment, between the aragonite com-
Formation is composed of green, red or reddish- pensation depth and calcite compensation depth.

Terrigenous influx was negligible during deposition in the type section situated south of Pons, in the
of the Pons Formation limestones. Sierra de los Organos belt. This section occurs in
the Valle de Pons tectonic unit (Fig. 13), while the
Carmita Formation (Cenomanian-Turonian) rocks of the Pefias Formation have been eroded in
the higher tectonic units of the Sierra de los Organos
Lithology. The Carmita Formation is composed belt.
of green, red and gray limestones (biomicrites and Age. The planktonic foraminifers, studied in thin
calcarenites), radiolarian cherts and shales. The cal- sections, indicate a Campanian-Maastrichtian age
carenites (calciturbidites) are up to 10 m thick. for this formation (de la Torre, 1988). According
These graded limestones contain minor amounts of to Hatten (1957) and Meyerhoff (in Khudoley and
sharp-edged quartz, wacky sandstone fragments, and Meyerhoff, 1971) the Pefias Formation is Turonian-
plagioclase detrital grains. In thin sections, ooids, Campanian in age. Recently, Iturralde-Vinent (1994,
shallow-water and pelagic limestone fragments are 1996) reported the presence in the Sierra de los
also seen. Organos belt of a hiatus that spans the Coniacian,
Maximum thickness of this formation is 70 m. Santonian and Campanian. Unfortunately, the results
This unit is better developed in the Northern Rosario of the paleontological study mentioned by this au-
belt, mainly because of Maastrichtian submarine thor are still not published. In the present paper, a
erosion in the Southern Rosario belt (Figs. 3 and 6). hiatus that comprises the late Turonian-Santonian is
Age. The planktonic foraminifers identified in accepted (Fig. 6).
thin sections by de la Torre (1988) indicate a Facies interpretation, Pelagic deposits of the
Cenomanian-Turonian age for the Carmita Forma- Pefias Formation were accumulated in a bathyal en-
tion. The younger age of the marls and calcareous vironment, below the aragonite compensation depth.
shales of the topmost part of this formation, although This part of the basin was effectively protected from
possible, has not been confirmed so far. the terrigenous influx.
Facies interpretation. The deep-water deposits
of the Carmita Formation were accumulated in Moreno Formation (Campanian)
bathyal environment, below the aragonite compen-
sation depth. The calciturbidites contain a shallow- Lithology. The Moreno Formation is composed
water debris transported from a carbonate platform. of marly limestones, detrital limestones, shales, silt-
stones and sandstones. The marly limestones prevail
Pinalilla Formation (Cenomanian-Turonian) in the lower part of this formation. The detrital
limestones are calciturbidites, up to 5 m thick.
Lithology. This formation is built of thick-bed- These deposits contain common shallow-water bio-
ded to massive, gray-green limestones, up to 170 m clasts. Volcanic lithoclasts and quartz are minor
thick. These carbonate rocks are mainly biomicrites constituents of calciturbidites. Shales and sandstones
containing planktonic foraminifers. The Pinalilla (siliciturbidites) constitute the upper part of the for-
Formation is developed in the Quifiones tectonic mation. The sandstones contain abundant angular
unit of the Northern Rosario belt only (Fig. 13). volcanic lithoclasts and plagioclase. This formation
Age. The planktonic foraminifers identified in in known mainly in the Northern Rosario belt. The
thin sections demonstrate a Cenomanian-Turonian maximum thickness of the Moreno Formation rocks
age. Limestones of the upper part of the formation is 240 m, but in places this unit was eroded partly
are middle to late Turonian in age. or completely at the base of the late Maastrichtian
Facies interpretation. The thick-bedded pelagic Cacarajfcara Formation (Figs. 4 and 6).
limestones of the Pinalilla Formation do not contain Age. The Campanian age of the Moreno For-
any chert intercalations. These limestones were de- mation is based on identification of planktonic
posited in a bathyal environment, probably less deep foraminifers in thin sections (Pszcz6tkowski, 1994a).
than in the case of the Santa Teresa and Carmita There is a hiatus between the Pinalilla Formation and
formations. the overlying Moreno Formation, which spans the
Coniacian and Santonian (Fig. 6). A similar hiatus
Pefias Formation (Campanian-Maastrichtian) exists between the Carmita and Moreno formations,
although the age of the uppermost deposits of the
Lithology. This formation consists of dark-gray former unit is still not too precise.
to black, thin-bedded limestones with abundant Facies interpretation. The deep-water, hemi-
black chert intercalations (Fig. 14). The limestones pelagic deposits of the lower part of the Moreno
are biomicrites containing profuse calcified radiolar- Formation originated in a bathyal environment, in
ians and less numerous planktonic foraminifers. The conditions of increasing influx of shallow-water and
thickness of the limestones and cherts attains 80 m volcaniclastic debris from a volcanic arc terrane.

Fig. 14. Pelagic limestones and cherts of the Pefias Formation (Campanian-Maastrichtian); the Las Piedras river, south of Pons, Sierra de
los Organosbelt. Hammerlength 28 cm.
The shales and sandstones of the upper part of the formations, in western and central Cuba, respec-
formation are interpreted as an evidence for conver- tively, are stratigraphic and facies equivalents of the
gence between the Upper Cretaceous passive margin Cacarajfcara Formation. The origin of these massive
and a volcanic arc terrane approaching from the megabeds was probably related to an extraordinary
southwest. These terrigenous deposits were accumu- earthquake and tsunami wave at the end of the Maas-
lated in front of this arc terrane. They are preserved trichtian. That unusual event could be associated
mainly in the southernmost tectonic units of the with the Chicxulub structure of YucaUin (Hildebrand
Guaniguanico terrane (Fig. 13). et al., 1991), although an accurate stratigraphic cor-
relation of the Cuban megabeds with the suggested
Cacarajicara Formation (upper Maastrichtian) impact site is still not known.
Lithology. This formation is developed as detrital Anc6n Formation (Paleocene)

limestones composed of breccia, calcarenite and cal-
cisiltite or calcilutite (Hatten, 1957; Pszcz6tkowski, Subdivision and lithology. In the Sierra de los
1978, 1986c, 1994a). The breccia consists of lime- Organos belt, the Anc6n Formation is subdivided
stone and chert clasts 0.1 to 5 m across, subordi- into the La Gtiira and La Legua Members (Fig. 6),
nately of shale fragments; upward, it passes grad- and the informal member of micritic and marly lime-
ually into fine calcirudite and coarse calcarenite. stones. In the Southern Rosario belt, the formation
Locally, this breccia is up 180 m thick (the Los is undivided, although the terrigenous rocks occur-
Cayos Member). The fine calcarenite and calcisiltite ring locally in the Cinco Pesos tectonic unit may be
constitute the upper part of the formation. Abundant considered as an informal unit (member).
redeposited shallow-water bioclasts of Cretaceous The Anc6n Formation consists of biomicrites,
age and relatively infrequent grains of volcanic rocks marly limestones, breccias and locally also of ter-
occur throughout the Cacarajfcara Formation. The rigenous deposits. Gray, green, and reddish biomi-
planktonic foraminifers are frequent in the upper, crites contain planktonic foraminifers, calcareous
fine-grained part of the formation. The thickness of nannoplankton and, sometimes, radiolarians. In the
the Cacarajfcara Formation rises from 5-30 m in the Sierra de los Organos belt, there are some chert
SR up to 450 m in the NR. nodules and/or lenses in the green biomicrites of the
Age. Planktonic foraminifers indicate the upper lower part of the formation. The breccias composed
Maastrichtian Abathomphalus mayaroensis Zone. of limestone and chert clasts are a characteristic
The list of the identified taxa was given by lithology of the formation in the Sierra de los
Pszcz6tkowski (1994a). Organos belt. The terrigenous rocks occurring in
Facies interpretation. The Cacarajfcara Forma- the Southern Rosario belt contain polymictic sand-
tion was interpreted as a clastic unit that origi- stones with abundant volcaniclastic fragments. The
nated as a result of a single, unusual depositional total thickness of the Anc6n Formation attains 50
event close to the Cretaceous/Tertiary boundary m in the Sierra de los Organos and 120(?) m in
(Pszcz6tkowski, 1986b). The Pefialver and Amaro the Southern Rosario belt. Only few outcrops of this

formation are known in the Northern Rosario belt belt (Pszcz6~kowski, 1994b). The Vieja Member is
and the yellowish shales of the Manacas Formation Early Eocene in age (Pszcz6~kowski, 1994a,b), but
are the main component of the Paleocene deposits in some sections its lower part may be as old as
there. Late Paleocene (Bralower et al., 1993; Bralower and
Age. Thin section analyses of planktonic Iturralde-Vinent, 1997).
foraminifers indicate that in the Sierra de los Facies interpretation. Large amounts of volcanic
Organos belt the Anc6n Formation is Paleocene in rock fragments, serpentinite and other types of exotic
age (Pszcz6~kowski, 1978; de la Torre, 1988). In the material in the Manacas Formation define this unit
Southern Rosario belt, this formation is of late Early as a foreland basin sedimentary fill (Iturralde-Vinent,
Paleocene to Late Paleocene age (Pszcz6~kowski, 1995b). The deep-water stage with four phases char-
1994a). The earliest Paleocene deposits are missing acterize evolution of this basin: (1) deposition of
in the Southern Rosario belt. fine-grained sediments (mostly shales), (2) turbiditic
Facies interpretation. The breccia units origi- sedimentation, (3) development of multi-component
nated in proximity of synsedimentary faults affecting olistostrome and (4) serpentinite slide. Phase (4)
the Jurassic-Cretaceous limestones (Pszcz61kowski, of the Guaniguanico foreland basin evolution was
1978). Some uplifted limestone blocks supplied proposed by Iturralde-Vinent (1996). The deposits
a carbonate debris during the Paleocene. Pelagic of a shallow-water stage, typical of many foreland
biomicrites and marly limestones accumulated in basins (Covey, 1986), have not been reported from
deeper areas of the sea bottom (submarine depres- the Manacas Formation.
sions). The volcaniclastic debris locally appeared in
the Southern Rosario belt beginning an early stage
of the foreland basin development during the (late) JURASSIC TO EARLY PALEOCENE EVOLUTIONOF
Early Paleocene. THE PASSIVE MARGIN IN WESTERN CUBA
Manacas Formation (Paleocene-Lower Eocene) Plate tectonic reconstructions of the Caribbean

region
Subdivision and lithology. This formation is sub-
divided into the Pica Pica Member (lower) and Vieja The Jurassic to Early Paleocene evolution of the
Member (upper) (Fig. 6). The Pica Pica Member North American margins exposed in Cuba are con-
is composed of layered terrigenous deposits: shales, sidered within the tectonic history of the Yucat~in
sandstones and conglomerates. Usually, yellow- or and Bahamas platform margins, as well as the proto-
red-weathering shales are developed in the lower part Caribbean basin development. The plate tectonic
of this member. In these shales, there are some brec- reconstructions of the Caribbean region (or the Gulf
cia interbeds in the southwestern part of the Sierra of Mexico-Caribbean system) have been outlined in
de los Organos belt. The lithoclastic sandstones and several papers during the last 15 years (e.g. Pindell
conglomerates, sometimes with infrequent intercala- and Dewey, 1982; Anderson and Schmidt, 1983; K1-
tions of marly and detrital limestones, predominate itgord et al., 1984; Ross and Scotese, 1988; Pindell
in the upper part of the Pica Pica Member. et al., 1988; Pindell and Barrett, 1990; Marton and
The Vieja Member consists of cobbles, boulders Buffler, 1994).
and larger olistoliths of various rocks enclosed in In the present paper, the plate tectonic reconstruc-
an argillaceous and/or silty matrix. These olistoliths tion by Marton and Buffler (1994) is accepted as a
were derived from the igneous, metamorphic and general paleotectonic basis for the Jurassic. These
sedimentary rocks. Fragments of rocks derived from authors presented the six-stage (Early Jurassic(?) to
the Guaniguanico stratigraphic successions occur Berriasian) rift and drift evolution of the Gulf of
throughout the whole Vieja Member. Olistoliths of Mexico and adjacent regions using the plate recon-
the serpentinite and other rocks of the ophiolitic struction software Plates 2.0. A rotation pole for the
complex are very frequent in this unit. Also volcanic Yucat~in block in the southeastern Gulf of Mexico
rocks, derived from the Cretaceous volcanic arc, are is proposed for the Callovian to Berriasian drifting
common, especially in the Rosario belts. The total stage (Marton and Buffler, 1994).
thickness of the Manacas Formation attains about The plate tectonic models proposed by Ross and
500 m in the most complete sections. Scotese (1988) and Pindell and Barrett (1990) are
Age. Planktonic foraminifers occurring in the adopted for the Cretaceous. These models assume
Manacas Formation are Paleocene-Early Eocene (1) the formation of a proto-Caribbean Sea due to
in age (Pszcz6ikowski, 1978, 1994a, 1988). The separation of North and South America, (2) the
lower boundary of the formation is diachronous; development of the Greater Antilles Arc along the
the oldest deposits (late Early Paleocene) of the (south)western margin of this sea, and (3) the in-
Manacas Formation occur in the Northern Rosario sertion of the Farallon plate between North and

~ooovv Callovian ,oovv

borderland represents the eastern extension of the
l /t~ /-~",, I v Yucatfin platform. However, the northern part of this
30ON borderland and western Cuba may be structurally
-- ,'~ . B
continuous (Fig. 1B). The Guaniguanico terrane
==== --~ ,,
rocks and the northern part of the Yucatfin border-
land together form a large wedge-shaped block now
isolated within the transform domain (Rosencrantz,
I / p," 1996, figs. 2 and 4). Deformed metaterrigenous
'7 I -- '
rocks dredged from the northern Yucatan borderland
(Pyle et al., 1973) may represent equivalents of the
Cangre belt located in the Guaniguanico terrane and
in the subsurface of the Guanahacabibes Peninsula.
b*+*+*+1 % % % " According to Pessagno et al. (Chapter 5) the
~ 1 1.+_,.+,+.,_.112 [. =.=,...eJ 3 I"' J4 ~ 5 Jurassic and Early Cretaceous successions in west-
Fig. 15. Reconstruction of the continental blocks and rift basins ern Cuba (Sierra del Rosario and Sierra de los Or-
between North America an South America during the Callovian ganos) show lithostratigraphic, paleobathymetric, and
(after Marton and Buffler, 1994, simplified and partly mod- paleolatudinal signatures which are nearly identical
ified by the author): BFZ = Bahamas fracture zone" BP =
Blake-Bahamas plateau; C = Camajuanf belt of north-central to those of San Pedro terrane remnants in central
Cuba (Upper Jurassic-Paleogene); E = Escambray terrane; G = Mexico. Pessagno et al. conclude: "These Cuban
Guaniguanico terrane; MSM = Mojave-Sonora megashear; NA remnants are allochthonous when compared to sur-
= North America; NWB = northwestern Bahamas; P = Pinos rounding Central Tethyan successions in the Blake
terrane; PS -- Placetas belt of north-central Cuba (Proterozoic Bahama basin and elsewhere in Cuba. They contain
marbles, Early to Middle Jurassic granitoids and Late Juras-
high latitude bivalves such as species of B u c h i a that
sic to Paleogene deposits)" R = Remedios belt of north-central
and northeastern Cuba (Jurassic?-Late Cretaceous)" SA = South can only be derived (exclusive of Greenland) from
America; T M V - - Trans-Mexican Volcanic belt; WMT = Western a Pacific source. The presence of Southern Bore-
Main Transform; 1 -- continental blocks, 2 -- Proterozoic mar- al/Northern Tethyan faunas (30~ in the Sierra de los
bles and Early to Middle Jurassic granitoids of the Placetas belt, Organos and Sierra del Rosario remnants as late as the
3 = San Cayetano Formation and its metamorphic equivalents
now exposed in western and south-central Cuba, 4 - rift basins Early Cretaceous (Valanginian) suggests much later
in the Gulf of Mexico area, 5 = Central Atlantic. tectonic transport by northwest to southeast move-
ment along the Walper Megashear and by subsequent
southwest to northeast movement as the Caribbean
South America resulting in the subduction of proto- plate plowed its way through the gap between the
Caribbean crust. North American and South American plates."
The position of the Guaniguanico terrane in re- Also Pszcz6tkowski and Myczyfiski 2 suggested
spect to the Yucatfin block during the Jurassic is that the occurrence of the bivalves B u c h i a and the
presented herein (Figs. 15-17) mainly after Itur- radiolarians P a r v i c i n g u l a sp. and Pantanellidae in
ralde-Vinent (1994, 1996). This author located the the Tithonian of western Cuba may indicate (ac-
Guaniguanico, Pinos and Escambray terranes on the cording to the criteria of Pessagno et al., 1993) the
eastern margin of the Yucatfin platform. In fact, the Southern Boreal Province or the Northern Tethyan
exact original position of these Cuban terranes dur- Province, and that the Jurassic sequences of western
ing the Jurassic and Cretaceous is still uncertain. Cuba could have been located in the Pacific during
The location of the Guaniguanico terrane very close Tithonian time. They considered, however, that this
to the present northeastern coast of the Yucatan interpretation requires further investigations.
Peninsula, as suggested by Iturralde-Vinent (1994, A paleomagnetic study of hand-samples collected
1996), may be difficult to sustain, mainly because from the Jurassic rocks of the Sierra de los Organos
the remnants of the Mesozoic successions occurring and Sierra del Rosario revealed postfolding magneti-
in western Cuba were not reported from the eastern sation in mafic rocks of the E1 Sfibalo Formation, but
margin of the Yucat~.n block (Viniegra, 1971; Lopez no meaningful results were derived from samples
Ramos, 1975). of the San Cayetano and Artemisa formations due
Probably, the Guaniguanico Mesozoic succes- to the weak magnetization and/or large scatter of
sions were deposited about 100 km to the east and paleomagnetic data (Bazhenov et al., 1996). P6rez
northeast of the present northeast Yucatan coast. Lazo et al. (1995) studied their samples from the San
Such a possibility is not contradicted by the exist- Cayetano Formation of the Sierra del Rosario, and
ing data on the Yucatan borderland topography and from the Collantes Formation marbles of the Escam-
structure (Rosencrantz, 1990, 1996). This province
extends from western Cuba to Honduras and has an 2 'Information and 1994 Annual Report', Institute of Geological
average width of 100 km. In general, the Yucatfin Sciences, Polish Academy of Sciences, Warsaw, 1995, p. 15.

Fig. 16. (A) Reconstruction of the continental blocks around the proto-Caribbean Sea during the middle Oxfordian (partly adapted from
Marton and Buffler, 1994 and Iturralde-Vinent, 1994): NR = Northern Rosario belt, SR = Southern Rosario belt, SO = Sierra de los
Organos belt, E = Escambray terrane, P -- Pinos terrane, BFZ = Bahamas fracture zone. (B) Enlarged rectangle shown in Fig. 1A; 1
= oceanic crust, 2 -- E1 Sfibalo Formation (mainly mafic rocks), 3 = clastic deposits of the San Cayetano Formation and equivalent
metamorphosed units (Arroyo Cangre Formation, and others), 4 -- ammonite-bearing clastic deposits of the uppermost San Cayetano
Formation in the Southern Rosario and La Esperanza belts, 5 = bivalve-bearing deposits of the uppermost part of the San Cayetano
Formation in the Sierra de los Organos and southwestern part of the Southern Rosario belt (mainly the Loma del Muerto tectonic unit), 6
= land areas.
Fig. 17. (A) Reconstruction of the proto-Caribbean basin for the Tithonian, partly adapted after Marton and Buffler, 1994. (B) Enlarged
area indicated by rectangle in (A). Yucatfin passive margin: SO = Sierra de los Organos belt, SR = Northern and Southern Rosario
belts; Florida-Bahamas passive margin: P = Placetas belt, C -- Camajuanf belt, R = Remedios belt; PC = proto-Caribbean Sea, BFZ
--- Bahamas fracture zone; 1 = oceanic crust, 2 -- pelagic ammonite-bearing biomicrites and calcarenites, 3 = terrigenous deposits, 4 =
shallow-water carbonates, 5 - land areas.
b r a y terrane. T h e s e s a m p l e s y i e l d e d p a l e o m a g n e t i c (1995) inferred f r o m their p a l e o m a g n e t i c data con-

i n c l i n a t i o n indicative of a p a l e o l a t i t u d e of about 12 ~ c e r n i n g the L o w e r C r e t a c e o u s and A p t i a n - T u r o n i a n
Pdrez L a z o et al. (1995) c o n c l u d e d that their pale- rocks, that the d e v e l o p m e n t of the volcanic arc in
o m a g n e t i c result c o r r o b o r a t e s the earlier g e o l o g i c a l C u b a t o o k p l a c e at p a l e o l a t i t u d e values o f a b o u t
i n t e r p r e t a t i o n s c o n c e r n i n g the original l o c a t i o n of 16-17~ C o n s e q u e n t l y , the e n t r a n c e to the proto-
the C u b a n Jurassic rocks " n o t too far f r o m H o n d u r a s C a r i b b e a n b a s i n c o u l d be b l o c k e d by the C u b a n or
and G u a t e m a l a . . . " (op. cit.), that is, s o u t h w e s t of its the G r e a t e r A n t i l l e s volcanic arc (see also R e n n e et
p r e s e n t - d a y position. M o r e o v e r , Pdrez L a z o et al. al., 1991 for the A p t i a n - C e n o m a n i a n p a l e o l a t i t u d e

of their Zaza terrane). This result constrains (or even The Oxfordian faunal assemblages of the San
contradicts?) a possibility of tectonic transport of Cayetano and Jagua formations do not contain
the 'Sierra de los Organos and Sierra del Rosario any taxa characteristic for high latitudes (see also
remnants' by southwest (from Pacific) to northeast Wierzbowski, 1976 and Myczyfiski, 1976, 1994a).
movement after the Barremian?-Aptian. These taxa were also not reported from the Kim-
The location of the San Cayetano basin along meridgian carbonate rocks of the Sierra de los
the Yucat~in margin during Middle and Late Jurassic Organos and Sierra del Rosario belts. In the Titho-
(Oxfordian) times cannot be defined exactly by the nian limestones of the Sierra de los Organos belt
available geological data. The Mesozoic successions there are frequent specimens of bivalves belong-
of western and south-central Cuba suffered severe ing to the genera Anopaea and Buchia (Myczyfiski,
tectonic deformations, including large-scale thrust- 1999). Some specimens of Anopaea sp. resemble
ing and metamorphism. During the Paleogene, these taxa described earlier from Tithonian deposits of
successions were thrust to the north, and probably Antarctica, Himalayas, New Zealand, and Sula Is-
sheared off from the Yucatan block, as the inactive lands (Myczyfiski, 1999). In the Sierra del Rosario,
volcanic arc passed along the eastern margin of this the Tithonian bivalves belong mainly to the genus
platform (Ross and Scotese, 1988; Pindell and Bar- Inoceramus (Parkinson, 1819), although represen-
rett, 1990; Hutson et al., 1998). The presence of tatives of Buchiidae are also present (Myczyfiski,
a transform boundary parallel to the Yucatan bor- 1994b). The occurrence of Tithonian bivalves char-
derland shows that the arc moved northward from acteristic for northern and southern high latitudes
a location south of the Yucat~in Peninsula (Rosen- may be explained by upwelling of deeper waters in
crantz, 1996). Nevertheless, various indirect geolog- the northwestern part of the proto-Caribbean basin
ical evidences constrain the inferred position of the (Coleman et al., 1995).
Guaniguanico terrane during the Jurassic. Only a The Tithonian ammonites of western and central
part of these arguments may be expressed herein, Cuba (Imlay, 1942, 1980; Millfin and Myczyfiski,
because of limited space. 1978; Myczyfiski, 1989, 1994a; Myczyfiski and
The San Cayetano Formation facies configura- Pszcz6tkowski, 1990, 1994) are so similar, that it
tion, reconstructed from the present-day internal would be unrealistic to place them in two different,
tectonostratigraphic pattern of the Guaniguanico ter- widely separated, oceanic basins. Therefore, any at-
rane, results as roughly parallel to the Yucat~in tempt to locate the Guaniguanico terrane at a higher
eastern margin. The directions of sediment trans- paleolatitudinal position (30~ during the Titho-
port from south or south-southwest measured in the nian, far from the proto-Caribbean basin, should be
San Cayetano Formation sandstones of the Sierra de accompanied by a similar shift of the Escambray
los Organos belt (Haczewski, 1976), are compati- terrane, as well as of the Placetas and Camajuanf
ble with the Caribbean paleotectonic reconstructions belts of central Cuba. In this case, the Guaniguanico,
proposed by Ross and Scotese (1988) and Pin- Escambray (and Pinos) terranes and the Placetas and
dell and Barrett (1990) for the Middle Jurassic- Camajuanf belts must be considered as parts (frag-
Oxfordian and Late Triassic-Early Jurassic, re- ments) of a composite terrane, or superterrane, 800-
spectively. In general, the composition of the San 1000 km long and 150-200 km wide, transported
Cayetano sandstones (Pszcz6tkowski, 1986a; Hut- (1) from NW to SE, and later (2) from SW to NE
son et al., 1998) can be explained on the basis of the (or NNE). However, this tectonic shift would leave
geological structure and Jurassic history of Yucat~in the Bahamas platform and the Yucatan block with-
(Lopez Ramos, 1975). out deep-water successions of their proto-Caribbean
The shallow-water to neritic San Cayetano, Jagua continental slope and adjacent basin floor. Consider-
and Francisco formations gradually disappear from ing the aforementioned problems, it seems that the
northwest to southeast. These units are laterally hypothesis of tectonic transport of the Sierra de los
replaced by the deep neritic E1 S~ibalo Forma- Organos and Sierra del Rosario successions along
tion. The carbonate bank of the Sierra de los the Walper Megashear still needs a lot of evidences
Organos belt (Pszcz6lkowski, 1978, 1981) is a and probably additional studies.
clear evidence of shallowing phase close to the
Oxfordian/Kimmeridgian boundary; this phase is Syn-rift stage (Lower Jurassic-?Callovian/early
characteristic for the Late Jurassic successions of Oxfordian)
western Cuba. A deepening phase occurred at the
Kimmeridgian/Tithonian boundary or during the South America was close to the Yucatan block
earliest Tithonian. Both events seems to be unknown during the late Middle Jurassic (Fig. 15). Some left-
in Mexico, especially in the sections described by lateral transform motion probably occurred between
Pessagno et al. (Chapter 5) as belonging to the San these two continental blocks (Marton and Buffler,
Pedro del Gallo terrane. 1994). Such a situation could exist also during the

early Oxfordian with a rift zone (Iturralde-Vinent, source for the San Cayetano Formation. The extent
1994, 1996) or rift/spreading center (Marton and of the San Cayetano depocenter to the southwest,
Buffler, 1994) between the two continental blocks. along the Yucat~in margin, is still uncertain. The
The basement of the Mesozoic sedimentary suc- model of propagating westward rift/spreading cen-
cessions, that originally belonged to the Yucat~in ter between Yucat~in and South America, with si-
eastern margin, is unknown in the Guaniguanico multaneous counterclockwise rotation of the former
terrane. The thrust units of this terrane do not block (Marton and Buffler, 1994), requires facies
contain any basement rocks at their base, because shift from the northeast to the southwest in the
the initial tectonic detachment occurred above the San Cayetano Formation. Indeed, some existing data
basement/sedimentary cover boundary. agree with this paleotectonic model. The most obvi-
During the sedimentation of the San Cayetano ous observation pertains to the facies changes within
deltaic deposits, their source areas probably con- the San Cayetano deposits in the Southern Rosario
sisted of metasedimentary and terrigenous rocks, and belt between La Palma and Mantua (Pszcz6ikowski,
also granitoids (Pszcz6tkowski, 1978). A prolonged 1994b). From NE to SW, the facies G-I disappear
transport, recycling of the pre-Middle Jurassic ter- and facies A-C and E (Haczewski (1976) are domi-
rigenous rocks, and weathering resulted in the high nant southwest of Minas de Matahambre, mainly in
content of quartz in the San Cayetano sandstones the Loma del Muerto tectonic unit (Fig. 2). These
(Pszcz6tkowski, 1986a). In the Southern Rosario facies changes are parallel to the regional strike of
belt, rare pebbles of the heavily silicified limestones the present-day tectonic structures (NE-SW). The
occur in the thick-bedded pebbly sandstones. Some San Cayetano Formation deposits, about 1400 m
of those pebbles contain the late Paleozoic fossils thick, also occur in the subsurface of the south-
(Pszcz6tkowski, 1989b). Evidently, these pebbles western part of the La Esperanza belt (Los Arroyos
were derived from the limestone succession situated 1 w e l l - Fern~indez et al., 1987; for location see
some hundreds kilometers from the San Cayetano Fig. 1). Deposition of the San Cayetano clastics was
basin during the early to middle Oxfordian. Ac- accompanied by syn-sedimentary magmatic activ-
cording to Donnelly et al. (1990) a correlation of ity, mainly of a mafic character (Piotrowski, 1977;
fossiliferous pebbles found in the San Cayetano Pszcz6tkowski, 1978; Iturralde-Vinent, 1995a).
Formation with Permian beds of eastern Guatemala
is not apparent. However, any conclusive explana- Drift stage (?Callovian/middle Oxfordian-
tion of the provenance of these pebbles is difficult, Santonian)
mainly because of small amount and size of fossilif-
erous clasts. During the middle Oxfordian, the gap between
At present it is not possible to demonstrate any Yucat~in and northwestern South America widened,
direct connection between the San Cayetano basin and the narrow proto-Caribbean seaway was formed.
and the northwestern edge of South America as A paleogeographic location of the Guaniguanico,
a source of the clastic material for the Jurassic Escambray and Pinos terranes during middle Oxfor-
deposits in western Cuba. The hypothesis for a South dian time is shown in Fig. 16A. The presence of
American provenance for the San Cayetano clastics the mafic rocks in the Northern Rosario belt (the
expressed by some authors (Anderson and Schmidt, E1 S~ibalo Formation) suggests that the oceanic crust
1983; Klitgord et al., 1984; Ryabukhin et al., 1984) formation in this gap commenced during the middle
is still to be proved. Apparently, the fossils and Oxfordian or earlier, south of the Northern Rosario
microfacies of the silicified limestone pebbles from and La Esperanza belts.
the San Cayetano Formation are not similar to the Recently, the Callovian sea-floor spreading was
Permian limestones occurring in the Palmarito and interpreted to have started simultaneously in the Gulf
Tucutunemo formations known in northern South of Mexico and in the Caribbean (Marton and Buffler,
America (Benjamini et al., 1987). 1994).
Instead, some data are consistent with a possible Marine conditions developed before the middle
connection of the San Cayetano depocenter with Oxfordian in all belts of the Guaniguanico terrane.
Central America (Meyerhof, in Khudoley and Mey- In the Sierra de los Organos succession, the first lime-
erhoff, 1971; Meyerhoff and Hatten, 1974; Iturralde- stone intercalations with marine fauna occur about
Vinent, 1975; Haczewski, 1976; Pszcz6ikowski, 400 m below the top of the San Cayetano Formation.
1987). Prior to the Oxfordian, the San Cayetano Nevertheless, exact timing of the onset of marine de-
clastics were located in the originally narrow, but position in all belts of the Guaniguanico terrane is
steadily widening, rift zone formed between Yu- still to be established. Fig. 16B shows a reconstruc-
cat~in and South America (Fig. 15). According to tion of facies during the middle Oxfordian. Infre-
Hutson et al. (1998), the presence of grains with quent ammonites have been found only in the upper-
Taconic and Grenvillian ages supports a Yucat~in most San Cayetano clastics in the Southern Rosario

belt. These ammonite-beating clastic deposits accu- ual, with appearance of some pelagic microfossils
mulated between the E1 S~balo limestones and basalts (Saccocoma sp., Colomisphaera spp. ) and deposi-
to the southeast and the shallow-water bivalve-bear- tion of a few thin-bedded limestone units within the
ing sandstones and shales to the northwest. thick-bedded to massive calcarenites of the upper
The advance of the middle Oxfordian trans- part of the San Vicente Member. In the North-
gression resulted in a major facies change, when ern Rosario belt, the lower Tithonian limestones
the San Cayetano deltaic sediments were replaced overlay the Kimmeridgian shallow-water dolomitic
by limestones with bivalves and shales with am- limestones.
monites (Sierra de los Organos and Cangre belts) Drowning of the shallow-water carbonates re-
and ammonite-bearing limestones and shales (South- sulted in a considerable uniformity of facies in all
ern Rosario belt). Facies differentiation in the W - E belts of the Guaniguanico terrane (Fig. 17). At Titho-
(or NW-SE) direction existed during deposition of nian time, the sedimentation rate (8-10 m/m.y.) of
the Jagua and Francisco formations. Surprisingly, the black pelagic limestones (Fig. 11) was some
marine macrofossils were not found in the sedi- ten times lower than that of the Kimmeridgian
mentary rocks of the E1 S~balo Formation. The shallow-water carbonates (80-100 m/m.y.). The
limestones occurring between the mafic rocks of Saccocoma-Didemnidae microfacies that predom-
this formation contain an impoverished microfos- inated in the lower Tithonian ammonite-bearing
sil assemblage dominated by Globochaete alpina limestones, was replaced gradually by a radio-
Lombard indicating a deeper (and partly restricted?) larian microfacies (Myczyfiski and Pszcz6~kowski,
depositional environment. The contact between the 1994). The latter microfacies is typical for the upper
E1 S~balo and Artemisa formations is erosional and, Tithonian and Lower Cretaceous limestones in all
in some sections, tectonic. Locally, thin breccia with Guaniguanico belts. In the Rosario belts, favorable
volcanic clasts occurs at the E1 S~balo/Artemisa conditions for radiolarians existed since the early
formations boundary. This boundary may be inter- Tithonian. Moderate fertility of the northern proto-
preted as the unconformity below the basal Artemisa Caribbean surface waters is also suggested by an
Formation of ?late Oxfordian-early Kimmeridgian elevated content of phosphatic grains, mainly abun-
age roughly correlatable with the limestone breccia dant fish debris (bones, scales and teeth) in the upper
at the base of the Guasasa Formation in the Sierra de Tithonian limestones of the Sierra de los Organos.
los Organos belt. Radiolarian limestones are also common in the Ca-
The onset of the carbonate shallow-water sedi- majuanf succession of central Cuba (Fig. 17) and in
mentation in the Sierra de los Organos and Cangre the Camagfiey Province.
belts occurred in the late Oxfordian or earliest Kim- In some sections, there are frequent specimens of
meridgian. Shallow-water carbonates are known also inoceramids belonging to the genera Anopaea and
from the Southern Rosario and Northern Rosario Buchia (identified by Dr. R. Myczyfiski). The occur-
belts. Development of the shallow-water bank above rence of Tithonian bivalves characteristic for high
the E1 S~balo Formation in the Northern Rosario belt latitudes is consistent with upwelling of cold, oxy-
(in the Bel6n Vigoa and Naranjo tectonic units genated, and nutrient-rich deeper waters in the north-
Fig. 13), with manifestations of erosion of basalts ern (or northwestern) part of the proto-Caribbean
and diabases, indicates that this volcano-sedimentary basin. According to Baumgartner (1987), off-shore
sequence has been locally uplifted during late Oxfor- winds created upwelling and high fertility of the sur-
dian or earliest Kimmeridgian time. These local up- face waters in the Late Jurassic 'Caribbean Tethys'.
lifts (rotated fault blocks?) could be a barrier inhibit- A Jurassic/Cretaceous boundary event, marked
ing free communication of the Sierra de los Organos by positive shift in the both carbon and oxygen
and Southern Rosario belts with the open, but still curves seems to be specific for the proto-Caribbean
narrow, proto-Caribbean Sea. The fine-grained lime- basin, or even for the Caribbean-Gulf of Mexico
stones (Fig. 10) and clastics (mainly shales and region (Coleman et al., 1995). The possible expla-
siltstones) accumulated in the inner, semirestricted, nation for this proto-Caribbean event is the invasion
part of the Southern Rosario belt. During the Kim- of geochemically different water masses across the
meridgian, subsidence kept pace with the relatively Jurassic-Cretaceous boundary (op cit.).
high rate of sedimentation. About 400 to 650 m of During the Berriasian and Valanginian, pelagic
shallow-water limestones and dolomitic limestones limestones with chert interbeds accumulated in the
formed in the Sierra de los Organos belt (San Vi- northwestern part of the proto-Caribbean basin. By
cente Member in Figs. 6 and 13). late Berriasian-early Valanginian, gray nannoconid
The transition from shallow-water deposition to limestones with abundant calpionellids were de-
pelagic conditions of sedimentation occurred close posited in the Sierra de los Organos belt. Basin-
to the Kimmeridgian/Tithonian boundary. In the ward, these thick-bedded, pure pelagic limestones
Sierra de los Organos, this change was rather grad- passed gradually into black, thin-bedded radiolar-

ian limestones occurring in the Northern Rosario terial in the above-mentioned belts. Probably, the
belt. The ammonites are uncommon in the pelagic Northern Rosario and Placetas belts belonged to the
limestones of Berriasian and Valanginian age (My- deep-water sector of the basin which extended be-
czyfiski, 1977). At the Tithonian/Berriasian bound- tween the Yucatfin and Bahamas passive margins and
ary, the basin floor probably descended below the the speculative spreading zone (ridge?) responsible
aragonite compensation depth. for the generation of the proto-Caribbean oceanic
In the Northern Rosario belt (farthest south sec- crust. This deep-water part of the basin was much
tions: BV, NO, CE, CH and QS, in palinspas- narrower at the Yucatfin-Florida Straits, creating a
tic reconstruction Fig. 13), the first turbidite fan-like mode of the sediment transport and disper-
sandstones appeared in the late Berriasian deposits. sal. Nevertheless, turbidity currents could not reach
Nevertheless, the main influx of siliciclastic mate- more marginally (and upslope?) located sedimen-
rial in this belt occurred during the Valanginian- tary successions now occurring in the Sierra de los
Barremian. A similar petrographic composition of Organos and Southern Rosario belts.
Hauterivian-Barremian siliciturbidites occurs in the In some sections, thin turbidites occur also among
Northern Rosario belt, La Esperanza belt and Plac- the radiolarian cherts in the lower part of the Santa
etas belt in the Matanzas Province and north-central Teresa Formation. The influx of the terrigenous ma-
Cuba indicating a common source for the clastic terial ceased during the Aptian-Albian(?). At the
material (Fig. 18). The Northern Rosario and La end of the Early Cretaceous, the siliceous deposition
Esperanza belts were situated nearer to the source extended across the entire deeper part of the north-
area. This conclusion results from established differ- western proto-Caribbean basin. The Santa Teresa
ences in abundance and thickness of siliciturbidites Formation appears in all belts related to the Yucat~in
between the Northern Rosario and La Esperanza passive margin, except the Sierra de los Organos
belts and Placetas belt (Pszcz6{kowski, 1982, 1987). and the Guajaib6n-Sierra Azul belts. This formation
The source area probably was located at the north- occurs also in the Placetas belt, originally located
eastern end of the Yucatfin block (Fig. 18). However, south of the Bahamas platform, represented in cen-
terrigenous deposits do not appear in the Sierra de tral Cuba mainly by the Remedios belt (Fig. 18).
los Organos and Southern Rosario belts, although These radiolarian cherts of Early Cretaceous to early
they were situated relatively close to this hypo- Cenomanian age are not known in the southeast-
thetical source area. Accepting the idea that the ern Gulf of Mexico. Probably, their accumulation in
Paleogene thrusting completely reversed the relative the northwestern part of the proto-Caribbean basin
positions of the belts and tectonic units (Fig. 13), was a net result of several different reasons (basin
one should account for the lack of terrigenous ma- margins subsidence since the Tithonian, eustatic sea-
Hauterivian - Barremian
~Gulf of Mexico):~ ~ "~o#/^'~,/'"-,. I

FB
, "~-~
' / ' """. " I- V-.-2..:.:.'. :...." ~ ,
~ 500 km
!
Fig. 18. (A) Lower Cretaceous (Hauterivian-Barremian) paleogeography of the proto-Caribbean and adjacent areas. (B) Enlarged area
indicated by rectangle in (A), with suggested provenance and distribution of terrigenous sediments (siliciturbidites). NA - North America,
SA = South America, FB -- Florida-Bahamas block, SO = Sierra de los Organos belt, SR = Southern Rosario belt, NR = Northern
Rosario belt, LE = La Esperanza belt, P -- Placetas belt, C = Camajuanf belt, R - Remedios belt; 1 - oceanic crust, 2 = radiolarian
cherts and shales, 3 -- pelagic limestones, 4 - shallow-water carbonates, 5 = suggested distribution of siliciturbidites, 6 = land area.
Black arrow indicates probable provenance of the terrigenous sediments.

level rise, paleogeographic and paleoceanographic

conditions).
During the Cenomanian, the pelagic carbonate
sedimentation was restored in the Rosario and Plac-
etas belts. The Turonian deposits occur in the
Pinalilla Formation and in the upper part of the
Carmita Formation, and were also reported from
the Pons Formation of the Sierra de los Organos
belt (Hatten, 1957). The Coniacian-Santonian, or
even late Turonian-Santonian, deposits are very
scarce, or even entirely missing (?) in western Cuba,
due to non-deposition and, sometimes, Late Creta-
ceous and/or Paleocene erosion (Figs. 4 and 6). The
Carmita Formation (Cenomanian-Turonian) pelagic
limestones are more clayey in their uppermost part
and are overlain by the marls and limestones of the
Campanian Moreno Formation. The thick-bedded Fig. 19. Generalized paleogeographic reconstruction of the proto-
pelagic limestones of the Pinalilla Formation occur Caribbean basin for the late Campanian (tectonic framework
immediately below the thin-bedded limestones and adapted from Ross and Scotese, 1988 and Pindell et al., 1988).
SP = Sepur foredeep, MR = Moreno depocenter, VB = Vfa
shales of the Moreno Formation, and no traces of
Blanca deep-water basin (western Cuba), SJM = San Juan y
erosion are discernible along the Pinalilla/Moreno Martfnez shallow-water basin (western Cuba); 1 = Caribbean
formations boundary. plate oceanic crust, 2 = proto-Caribbean and Atlantic oceanic
The origin of a regionally extensive late crust, 3 = pelagic limestones and cherts (Pefias Formation in the
Turonian-Santonian hiatus in the deep-water, Sierra de los Organos belt), 4 = shales, volcaniclastic sandstones
pelagic sequence is probably related to paleoceano- (turbidites) and marly limestones of the Moreno Formation, 5 =
lack of the Campanian deposits, 6 = shallow-waterlimestones of
graphic conditions existing during Late Cretaceous the Remedios belt in north-central Cuba, 7 -- deep-water clastics
times in the northwestern part of the proto-Caribbean of the Vfa Blanca Formation, 8 = shallow-water limestones and
basin. For example, this hiatus is coincident with the conglomerates of the San Juan y Martfnez Formation. Heavy line
maximum flooding of the South American con- with triangles denotes subduction of the proto-Caribbean crust
tinent in northwestern Venezuela during the Late beneath the Greater Antilles Arc.
Cretaceous highstand of sea level (Lugo and Mann,
1995). In addition, during the Turonian-Santonian,
the Nicaraguan Rise-Greater Antilles Arc partially material, mainly in its upper part (Pszcz6~kowski,
closed the connection of the proto-Caribbean basin 1994a,b). The calciturbidites with volcanic litho-
with the Pacific (Pindell and Dewey, 1982; Pin- clasts occur in the lower part of this formation. The
dell, 1991). However, the influence of this pale- petrographic composition of the detrital limestones
ogeographic change on the sedimentation in the and sandstones clearly indicates the volcanic arc
proto-Caribbean basin is still to be evaluated. source for the Moreno Formation turbidites.
During the Campanian, the volcanic arc was lo-
The beginning of the active margin stage cated east of the Yucatan block margin and south
(Campanian-Paleocene) of the Moreno depocenter. This arc could be the
westernmost part of the Greater Antilles Arc (GAA),
According to Pindell and Dewey (1982), Ross as proposed by Pindell and Dewey (1982) and Pin-
and Scotese (1988) and Pindell (1991), the proto- dell et al. (1988). The position of the GAA shown
Caribbean ocean basin has been progressively sub- in Fig. 19 is, in general, in accordance with the
ducted beneath the Caribbean plate during the Late tectonic reconstructions by Ross and Scotese (1988)
Cretaceous-Eocene. An arc terrane collided with and Pindell and Barrett (1990). The results of the pa-
the southern Yucatan margin during the Campa- leomagnetic investigations published so far (Renne
nian to early Maastrichtian (Donnelly, 1989; Pindell et al., 1991; Chauvin et al., 1994; P6rez Lazo et al.,
and Barrett, 1990). The Campanian Sepur clastics 1995; Bazhenov et al., 1996) indicate between 550
filled the foredeep in front of the arc terrane (Sp in and 1600-+-600 km northward displacement of the
Fig. 19). Zaza volcanic arc (or Zaza terrane) during the Late
The northeastern passive (proto-Caribbean) mar- Cretaceous, Paleocene and Early Eocene.
gin of Yucatan was also affected by the approach- Although the Moreno depocenter could be con-
ing arc terrane in the Campanian. In the Northern nected with the Sepur foredeep, their Campanian
Rosario belt of western Cuba, the Moreno Formation deposits were accumulated in different tectonic and
(MR in Fig. 19) contains abundant volcaniclastic paleobathymetric settings (Fig. 19). The Sepur For-

116 A. PSzCZOLKOWSKI
form, the limestones and cherts of the Lutgarda

Formation are known in the Camajuanf belt, while
the Amaro Formation is a characteristic unit oc-
curring in the deep-water Placetas belt (Am in
Fig. 20). The Amaro Formation is an equivalent
of the Cacarajfcara and Pefialver formations in west-
ern Cuba. The peculiar character and origin of the
late Maastrichtian Cacarajfcara, Amaro and Pefialver
megabeds was studied by Pszczdlkowski (1986b)
and was also discussed by Iturralde-Vinent (1992).
The latest Maastrichtian paleogeography, shown in
Fig. 20, is partly based on conclusions presented
in these papers. However, the position of GAA in
respect to the Bahamas platform is more southerly in
Fig. 20 than that assumed earlier by Pszcz6tkowski
(1986b) and Iturralde-Vinent (1992). Such a posi-
tion of the GAA at the end of the Cretaceous is
Fig. 20. Paleogeographic reconstruction of the proto-Caribbean
basin for the latest Maastrichtian; modified tectonic framework inferred from the paleotectonic reconstruction and
adapted from Ross and Scotese (1988, fig. 9): Ca -- Cacarajfcara lithology of the late Maastrichtian deposits, as the
Formation (western Cuba), Am -- Amaro Formation (north-cen- Cacarajfcara Formation does not contain a significant
tral Cuba), Pr -- Pefialver Formation (western Cuba), SJM =
amount of a coarse-grained volcaniclastic material
San Juan y Martfnez shallow-water basin (western Cuba), Cf (> 1 cm) derived from the extinct volcanic arc, even
= Cienfuegos basin (south-central Cuba); 1 = Caribbean plate
oceanic crust, 2 -- proto-Caribbean and Atlantic oceanic crust, in the Cangre and Sierra Chiquita tectonic units.
3 = pelagic limestones and cherts of the Pefias Formation, 4 The thickest sections of the Cacarajfcara megabed
-- detrital limestones of the Cacarajfcara, Amaro and Pefialver clastic deposits (450 m) were measured in these
megabeds (maximum thickness of 200-450 m), 5 -- pelagic two tectonic units (Fig. 13), and this fact shows
limestones and cherts of the Camajuanf belt in north-central that their position was well to the south (Fig. 20),
Cuba, 6 = shallow-water limestones of the Remedios belt in
north-central Cuba, 7 = shallow-water limestones of the San not at the entrance of the Gulf of Mexico between
Juan y Martfnez Formation in western Cuba and marls of the Florida and Yucatfin, as proposed by Iturralde-Vinent
lowermost part of the Vaquerfa Formation in the Cienfuegos (1992).
basin. Heavy line with black triangles denotes subduction of the During the Paleocene, the convergence of the
proto-Caribbean crust, while heavy line with open triangle marks extinct GAA segment with the Bahamas platform
underthrusting.
margin continued (Fig. 21). The position of the ex-
tinct, westernmost GAA segment (occurring now in
mation was deposited on a carbonate shelf in south- the Bahia Honda terrane) at the Yucatfin margin as
ern Yucatan. This shelf was depressed and buried indicated in Fig. 21, results from a tectonic and
by the Sepur serpentinite-bearing flysch (Pindell lithostratigraphic analysis of the Lower Paleocene
and Barrett, 1990). The Moreno marly limestones deposits in the Guaniguanico terrane. In the North-
and siliciturbidites were laid down in a deep-water ern Rosario belt, the Manacas Formation shales
basin on the thinned continental crust and partly overlying the Cacarajfcara Formation are evidence
on oceanic crust. During the Campanian, the Sierra of a major change from pelagic conditions during
de los Organos belt was not affected by the influx the Cretaceous to the Paleogene foreland basin en-
of volcaniclastic debris, as this material was not vironment. The narrow northwestern sector of the
reported from the pelagic limestones and cherts of proto-Caribbean basin was either a peripheral or
the Pefias Formation. Campanian sediments were not a retroarc foreland basin, located in front of the
preserved (or deposited?) across the vast area situ- thrust belt along the southern side of the remnant
ated between the Moreno depocenter, Sierra de los proto-Caribbean Sea. The discrimination of ancient
Organos belt and the Bahamas platform and slope peripheral and retroarc foreland basins is difficult
(Fig. 19). (Ingersoll, 1988). Within the plate tectonic model
The Maastrichtian deposits of the passive mar- accepted herein (Figs. 19-21), the Paleocene basin
gin of Yucatan are represented by the Cacarajfcara of western and central Cuba was a peripheral fore-
Formation in the Rosario belts, and by the pelagic land basin. However, according to an alternative
limestones and cherts of the Pefias Formation in plate-tectonic model (Iturralde-Vinent, 1994, 1996)
the Sierra de los Organos belt (Figs. 4, 6 and an retroarc foreland basin formed in western and
14). The location of these deposits is shown in central Cuba during the Paleocene.
paleogeographic reconstruction for the latest Maas- The material derived from the volcanic suites and
trichtian (Fig. 20). South of the Bahamas plat- ophiolite contributed to the foreland basin deposits

CONCLUSIONS
The Mesozoic successions of western Cuba, now

exposed in the Guaniguanico terrane, were deposited
more than 100 km to the east of the present north-
east Yucat~in coast. The evolution of these Yucat~in
passive margin successions encompasses the syn-
rift stage (Lower Jurassic-?Callovian/early Oxfor-
dian), drift stage (?Callovian/middle Oxfordian-
Santonian), and the beginning of the active margin
stage (Campanian-Paleocene). Prior to the middle
Oxfordian, the San Cayetano basin was located in
a narrow, but steadily widening, rift zone formed
between Yucatfin and South America. The advance
of the middle Oxfordian transgression resulted in a
major facies change, when the San Cayetano deltaic
Fig. 21. Simplified paleogeographic reconstruction of the north- sediments were replaced by shallow-water lime-
ern Caribbean region for the Early Paleocene: Vb -- Vfbora basin stones with bivalves and/or by deeper ammonite-
(western Cuba), Cf = Cienfuegos basin (south-central Cuba), bearing deposits. The restoration of the carbonate
T.f = La Trocha fault in central Cuba (see Hatten, 1967); 1
shallow-water sedimentation in the Sierra de los
-- Caribbean plate oceanic crust, 2 = proto-Caribbean and At-
lantic oceanic crust, 3 = pelagic biomicrites and breccias of the Organos and Cangre belts and its onset in the
Anc6n Formation (western Cuba), 4 = shales and claystones Rosario belts occurred in the late Oxfordian or ear-
(Manacas and Vega Alta formations), 5 - breccias and calciru- liest Kimmeridgian. Subsidence kept pace with the
dites of the Vega Formation (Camajuanf belt of the Bahamas relatively high rate of sedimentation in the Sierra de
platform margin), 6 = Remedios belt in north-central Cuba, 7 =
los Organos belt; about 400 to 650 m of shallow-
syn-sedimentary normal faults. Other symbols as in Fig. 20.
water limestones and dolomitic limestones formed
during the Kimmeridgian.
Drowning of the shallow-water carbonates in the
early Tithonian resulted in a considerable uniformity
in western Cuba. Pelagic limestones prevailed in the of pelagic facies in all belts of the Guaniguanico
Southern Rosario belt (Anc6n Formation), although terrane. The Hauterivian-Barremian siliciturbidites
with a clear influence of the arc-originated detritus that occur in the Northern Rosario, La Esperanza
in the Cinco Pesos tectonic unit. Pelagic limestones and Placetas belts of western and central Cuba are
and breccias are widespread in the Sierra de los interpreted to have a common source for the clas-
Organos belt. The limestone and chert breccias were tic material. These belts probably belonged to the
formed as a result of a considerable erosion of the deep-water sector of the basin, which extended be-
underlying Cretaceous limestones (in places also tween the Yucat~in and Bahamas passive margins.
Tithonian), along syn-sedimentary fault escarpments During the Aptian-Albian, the siliceous deposition
(Pszczdtkowski, 1978). These faults, schematically extended across the entire northwestern, deeper part
shown in Fig. 21, could originate in a zone of of the northwestern proto-Caribbean basin. In the
extension induced by bending of the underthrusting Cenomanian, pelagic carbonate sedimentation was
plate during arc-passive margin collision (Bradley restored in the Rosario and Placetas belts. The late
and Kidd, 1991). The Paleocene-Middle Eocene Turonian (or Coniacian)-Santonian deposits are very
limestone breccias with a considerable thickness are scarce, or even entirely missing in western Cuba, due
also known in the Camajuanf belt of west-central and to non-deposition and the Late Cretaceous and/or
central Cuba (Pszcz6tkowski, 1983). The Jurassic- Paleocene erosion. During the Turonian-Santonian,
Cretaceous sedimentary successions deposited on the Nicaraguan Rise-Greater Antilles Arc partially
(and along) the passive margin of Yucatfin, now closed the connection of the proto-Caribbean basin
exposed in western Cuba, formed the foreland basin with the Pacific. Among other factors, this paleo-
substrate during the Paleocene-Early Eocene. In geographic change could create specific paleoceano-
the Sierra de los Organos belt, a change from the graphic conditions in the northwestern part of the
passive margin to a foreland basin occurred during proto-Caribbean basin.
the Early to Late Paleocene. The terminal collision The eastern passive margin of Yucatfin was af-
of the extinct volcanic arc with the passive margin fected by approaching arc terrane in the Campanian.
occurred in the Late Paleocene to Early Eocene in In the Northern Rosario belt of western Cuba, the
western Cuba (Bralower et al., 1993; Bralower and Moreno Formation contains abundant volcaniclas-
Iturralde-Vinent, 1997; Gordon et al., 1997). tic material, mainly in its upper part. During the

C a m p a n i a n , the volcanic arc was located east of R.B. Hargraves, and R. Shagam (Editors), The Caribbean-
the Yucatfin b l o c k m a r g i n and south of the M o r e n o South American Plate Boundary and Regional Tectonics. Geol.
depocenter. This arc could be the w e s t e r n m o s t part Soc. Am. Mem., 162: 1-30.
Case, J.E., MacDonald, W.D., Fox, EJ., 1990. Caribbean crustal
of the G r e a t e r Antilles Arc ( G A A ) , as p r o p o s e d by
provinces; Seismic and gravity evidence. In: J.E. Case and
Pindell and D e w e y (1982) and Pindell et al. (1988). G. Dengo (Editors), Decade of North American Geology, Vol.
T h e late M a a s t r i c h t i a n deposits of the passive mar- H. The Caribbean Region. Geological Society of America,
gin of Yucatfin are r e p r e s e n t e d by the Cacarajfcara Boulder, CO, p. 15-36.
F o r m a t i o n in the Rosario belts. A m o r e southerly Chauvin, A., Bazhenov, M.L. and Beaudouin, T., 1994. A re-
connaissance paleomagnetic study of Cretaceous rocks from
position of the extinct volcanic arc at the end of
central Cuba. Geophys. Res. Lett., 21: 1691-1694.
the C r e t a c e o u s is inferred from the p a l e o t e c t o n i c re- Cobiella-Reguera, J.L., 1996. E1 magmatismo junisico
c o n s t r u c t i o n and lithology of the late M a a s t r i c h t i a n (caloviano?-oxfordiano) de Cuba occidental: ambiente de for-
deposits. maci6n e implicaciones regionales. Rev. Asoc. Geol. Argent.,
D u r i n g the P a l e o c e n e , the s e d i m e n t a r y succes- 51 (1): 15-28.
Coleman, M.L., Gruszczyfiski, M., Pszcz61kowski, A. and
sions of the G u a n i g u a n i c o terrane, originally de-
Myczyliski, R., 1995. Geochemical isotopic and elemental
posited on (and along) the passive m a r g i n of Yu- record from Cuban Late Jurassic-earliest Cretaceous: a proto-
catfin, f o r m e d the foreland basin substrate. This fore- Caribbean sea-water evolution. 1st SEPM Congr. Sedimentary
land basin was located in front of a thrust belt along Geology, August 13-16, 1995, St. Pete Beach, FL, Progr.
the southern side of the r e m n a n t p r o t o - C a r i b b e a n Abstr. Vol. 1, p. 41.
Covey, M., 1986. The evolution of foreland basins to steady
Sea.
state: evidence from the western Taiwan foreland basin. In:
EA. Allen and E Homewood (Editors), Foreland Basins. Int.
Assoc. Sedimentol., Spec. Publ., 8: 77-90.
ACKNOWLEDGEMENTS de la Torre, A., 1960. Fauna de la formaci6n Cayetano del
Junisico Medio de Pinar del Rio. Mem. Soc. Cubana Hist.
Nat., 25 (1): 65-72.
T h e author is grateful to Richard T. Buffler,
de la Torre, A., 1988. Resultados micropaleontol6gicos m~s
T h o m a s W. Donnelly, John E L e w i s and G y 6 r g y importantes del levantamiento geol6gico de la provincia de
M a r t o n for their review of the manuscript, and to Pinar del Rfo. In: Contribuci6n a la geologfa de la provincia
Paul M a n n for his useful c o m m e n t s on the text and de Pinar del Rfo. Edit. Cient. T6cn., La Habana, pp. 170-184.
figures. T h e discussions with R y s z a r d Myczyfiski Donnelly, T.W., 1989. Geologic history of the Caribbean and
Central America. In: J.E. Case and G. Dengo (Editors),
and M a n u e l Iturralde-Vinent on s o m e p r o b l e m s con-
Decade of North American Geology, Vol. A. The Geology
cerning the g e o l o g y of Cuba are appreciated. of North America - - An Overview. Geological Society of
America, Boulder, CO, pp. 299-321.
Donnelly, T.W., Horne, G.S., Finch, R.C. and L6pez-Ramos,
E., 1990. Northern Central America; the Maya and Chortis
REFERENCES
blocks. In: J.E. Case and G. Dengo (Editors), Decade of North
American Geology, Vol. H. The Caribbean Region. Geological
Aiello, I.W. and Chiari, M., 1995. Cretaceous radiolarian cherts Society of America, Boulder, CO, pp. 37-76.
of western Cuba. Ofioliti, 20:123-128. Fernandez Carmona, J., 1989. Primer reporte de microf6siles
Anderson, T.H. and Schmidt, V.A., 1983. The evolution of Mid- fndices del Kimmeridgense en Cuba. In: Geologfa/Geology
dle America and the Gulf of Mexico-Caribbean Sea region '89 (Primer Congreso Cubano de Geologfa, La Habana), Re-
during Mesozoic time. Geol. Soc. Am. Bull., 94 (8): 941-966. sumenes y programa, p. 64.
Baumgartner, EO., 1987. Age and genesis of Tethyan Jurassic Fermindez, J., Areces, A. and Dfaz, M.L., 1987. Estratigraffa
radiolarites. Eclogae Geol. Helv., 80 (3): 831-879. del ~irea Los Arroyos, provincia de Pinar del Rfo basada en
Bazhenov, M.L., Pszczolkowski, A. and Shipunov, S.V., 1996. datos de perforaci6n profunda. In: Memorias del III encuentro
Reconnaissance paleomagnetic results from western Cuba. cientffico-t6cnico de geologfa en Pinar del Rfo. Soc. Cubana
Tectonophysics, 253: 65-81. de Geologfa, pp. 51-56.
Benjamini, C., Shagam, R. and Menendez V., A., 1987. (Late?) Furrazola-Bermfidez, G., 1965. Tres nuevas especies de Tin-
Paleozoic age for the 'Cretaceous' Tucutunemo Formation, tfnidos del Jur~isico Superior de Cuba. Inst. Cubano Rec. Min.,
northern Venezuela: stratigraphic and tectonic implications. Publ. Esp., 2: 1-29.
Geology, 15: 922-926. Gradstein, EM., Agterberg, EE, Ogg, J.G., Hardenbol, J., van
Bradley, D.C. and Kidd, W.S.E, 1991. Flexural extension of the Veen, E, Thierry, J. and Huang, Z., 1994. A Mesozoic time
upper continental crust in collisional foredeeps. Geol. Soc. scale. J. Geophys. Res., 99:24051-24074.
Am. Bull., 103: 1416-1438. Gordon, M.B., Mann, E, C~iceres, D. and Flores, R., 1997. Ceno-
Bralower, T.J., Hutson, E, Mann, E, Iturralde-Vinent, M., Renne, zoic tectonic history of the North America-Caribbean plate
E and Sliter, W.V., 1993. Tectonic of oblique arc-continent boundary in western Cuba. J. Geophys. Res., 102: 10055-
collision in western Cuba, 1. Stratigraphic constraints. Eos, 10082.
74: 546. Haczewski, G., 1976. Sedimentological reconnaissance of the
Bralower, T.J. and Iturralde-Vinent, M.A., 1997. Micropaleon- San Cayetano Formation: an accumulative continental margin
tological dating of the collision between the North American in the Jurassic of Western Cuba. Acta Geol. Pol., 26:331-353.
plate and Greater Antilles arc in western Cuba. Palaios, 12: Hatten, C.W., 1957. Geologic report on Sierra de los Organos
133-150. (unpubl.). Fondo Geol. (MINBAS), La Habana.
Case, J.E., Holcombe, T.L. and Martin, R.G., 1984. Map of Hatten, C.W., 1967. Principal features of Cuban geology: discus-
geologic provinces in the Caribbean region. In: W.E. Bonini, sion. Am. Assoc. Pet. Geol. Bull., 51: 780-789.

THE E X P O S E D PASSIVE M A R G I N OF N O R T H A M E R I C A IN W E S T E R N C U B A 119
Herrera, N.H., 1961. Contribuci6n a la Estratigraffa de la Provin- Caribbean Region. Geological Society of America, Boulder,
cia de Pinar del Rfo. Rev. Soc. Cub. Ing., 61: 22-44. CO, pp. 77-140.
Hildebrand, A.R., Penfield, G.T., Kring, D.A., Pilkington, M., Lopez Ramos, E., 1975. Geological summary of the Yucatan
Camargo, Z.A., Jacobsen, S.B. and Boynton, W.V., 1991. Peninsula. In: A.E.M. Nairn and EG. Stehli (Editors), The
Chicxulub crater: a possible Cretaceous/Tertiary boundary im- Ocean Basins and Margins, 3. The Gulf of Mexico and the
pact crater on the Yucatan Peninsula, Mexico. Geology, 19: Caribbean, Plenum Press, New York-London, pp. 257-282..
867-871. L6pez Rivera, J.G., L6pez Quintero, J.O., Fernandez Carmona
Hou~a, V., 1969. Neocomian rhyncholites from Cuba. J. Paleon- J. and Fernandez Rodriguez, G., 1987. Analisis geol6gico
tol., 43:119-124. del corte del pozo param6trico Pinar 1. In: Memorias del III
Houga, V., 1974. Informe final sobre los trabajos de campo Encuentro Cientffico-t6cnico de Geologfa en Pinar del Rfo.
realizados para el estudio bioestratigrafico y la recolecci6n de Soc. Cubana de Geologfa, pp. 40-45.
los ammonites del Tithoniano y Cretacico Inferior en algunas Lugo, J. and Mann, E, 1995. Jurassic-Eocene tectonic evolution
localidades en la Provincia de Pinar del Rfo, Cuba. Unpubl. of Maracaibo basin, Venezuela. In: A. Tankard, G. Suarez and
Rep., Inst. Geol. Paleontol. MINBAS, La Habana. H. Welsink (Editors), Petroleum Basins of South America.
Howell, D., Jones, D. and Schermer, E., 1985. Tectonostrati- Am. Assoc. Pet. Geol., Mem., 62: 699-725.
graphic terranes of the circum-Pacific region. In: D. Howell Mann, E, Draper, G. and Lewis, J.E, 1991. An overview of
(Editor), Tectonostratigraphic Terranes of the Circum-Pacific. the geologic and tectonic development of Hispaniola. In: E
Circum-Pacific Council for Energy and Mineral Resources, Mann, G. Draper and J.F. Lewis (Editors), Geologic and
Houston, Texas, Earth Science Series 1, pp. 3-30. Tectonic Development of the North America-Caribbean Plate
Hutson, F., Mann, E and Renne, E, 1998. 4~ dating Boundary in Hispaniola. Geol. Soc. Am. Spec. Pap., 262: 1-
of single muscovite grains in Jurassic siliciclastic rocks (San 28.
Cayetano Formation): constrains on the paleoposition of west- Martfnez, D. and Vazquez, M., 1987. Alturas Pizarrosas del
ern Cuba. Geology, 26: 83-86. Norte, extremo occidental de la subzona 'Sierra del Rosario'.
Imlay, R.W., 1942. Late Jurassic fossils from Cuba and their In: Memorias del III Encuentro Cientffico-t6cnico de Geologfa
economic significance. Geol. Soc. Am. Bull., 53: 1417-1478. en Pinar del Rfo. Soc. Cubana de Geologfa, pp. 57-71.
Imlay, R.W., 1980. Jurassic paleobiogeography of the contermi- Marton, G. and Buffler, R.T., 1994. Jurassic reconstruction of the
nous United States in its continental setting. Geol. Surv. Prof. Gulf of Mexico Basin. Int. Geol. Rev., 36: 545-586.
Pap., 1062: 1-134. Meyerhoff, A.A. and Hatten, C.W., 1974. Bahamas salient
Ingersoll, R.V., 1988. Tectonics of sedimentary basins. Geol. of North America: Tectonic framework, stratigraphy, and
Soc. Am. Bull., 100: 1704-1719. petroleum potential. Am. Assoc. Pet. Geol. Bull., 58: 1201-
Iturralde-Vinent, M.A., 1975. Problems in application of modern 1239.
tectonic hypotheses to Cuba and Caribbean region. Am. Assoc. Millan, G. and Myczyfiski, R., 1978. Fauna jurasica y consid-
Pet. Geol. Bull., 59: 838-855. eraciones sobre la edad de las secuencias metam6rficas del
Iturralde-Vinent, M., 1992. A short note on the Cuban late Maas- Escambray. Inf. Cient. T6cn. Acad. Cien. Cuba, 80: 1-14.
trichtian megaturbidite (an impact-derived deposit?). Earth Mossakovskiy, A. and de Albear, de, J.E, 1979. Estructura de
Planet. Sci. Lett., 109: 225-228. cabalgamiento de Cuba occidental y norte e historia de su
Iturralde-Vinent, M.A., 1994. Cuban geology: a new plate-tec- formaci6n a la luz del estudio de los olistostromas y las
tonic synthesis. J. Pet. Geol., 17: 39-70. molasas. Cien. Tierra Espacio, 1:3-31.
Iturralde-Vinent, M.A., 1995a. Implicaciones tect6nicas de mag- Myczyfiski, R., 1976. A new ammonite fauna from the Oxfordian
matismo de margen continental pasivo en Cuba. In: A.C. of the Pinar del Rfo Province, western Cuba. Acta Geol. Pol.,
Riccardi and M.E Iglesia Llanos (Editors), Jurasico de Cuba 26:261-299.
y Am6rica del Sur. Rel. Conf. de Campo, PICG 322 en Cuba, Myczyfiski, R., 1977. Lower Cretaceous ammonites from Sierra
Noviembre 22-29, 1994, Buenos Aires, pp. 14-30. del Rosario (western Cuba). Acta Paleontol. Pol., 22: 139-
Iturralde-Vinent, M.A., 1995b. Cuencas sedimentarias del 173.
Paleoceno-Eoceno de Cuba. Bol. Soc. Venez. Geol., 20: 75- Myczyfiski, R., 1989. Ammonite biostratigraphy of the Tithonian
80. of western Cuba. Ann. Soc. Geol. Pol., 59: 43-125.
Iturralde-Vinent, M.A., 1996. Introduction to Cuban geology and Myczyfiski, R., 1994a. Caribbean ammonite assemblages from
tectonics. In: M.A. Iturralde-Vinent (Editor), Ofiolitas y arcos Upper Jurassic-Lower Cretaceous sequences of Cuba. Stud.
volcanicos de Cuba (Cuban ophiolites and volcanics arcs). Geol. Pol., 195:91-108.
IUGS/UNESCO Project 364, Contrib. 1, pp. 3-35. Myczyfiski, R., 1994b. Lower Tithonian ammonites from the
Iturralde-Vinent, M. and Norell, M.A., 1996. Synopsis of Late Sierra del Rosario, Western Cuba. Palaeopelagos Spec. Publ.
Jurassic marine reptiles from Cuba. Am. Mus. Novitates, 3164, 1, Proc. 3rd Pergola Int. Symp., pp. 287-298.
17 pp. Myczyfiski, R., 1996a. Discusi6n sobre la presencia de los
Judoley, K.M. (= Khudoley, K.M.) and Furrazola-Berm6dez, G., g6neros Parodontoceras Spath, 1923 y Lytohoplites Spath,
1968. Estratigraffa y fauna del Jurasico de Cuba. La Habana, 1925 en el Tithoniano de Cuba. In: E O16riz and EJ. Ro-
126 pp. drfguez-Tovar (Editors), Cephalopods - - Present and Past. IV
Khudoley, K.M. and Meyerhoff, A.A., 1971. Paleogeography and Int. Symp. Granada, July 15-17, 1996, Abstracts Volume, pp.
geological history of Greater Antilles. Geol. Soc. Am. Mem., 125-126.
129: 1-199. Myczyfiski, R., 1996b. Interpretaci6n estratigrafica y revisi6n
Klitgord, K.D., Popenoe, E and Schouten, H., 1984. Florida: sistematica de algunos ammonites del Tithoniano de Cuba.
a Jurassic transform plate boundary. J. Geophys. Res., 89: Terra Nostra, 8/96, p. 103.
7753-7772. Myczyfiski, R., 1999. Inoceramids and buchids in Tithonian de-
Kutek, J., Pszcz6[kowski, A. and Wierzbowski, A., 1976. The posits of western Cuba and their paleogeographic significance
Francisco Formation and an Oxfordian ammonite faunule from for the Caribbean region (in prep.).
the Artemisa Formation. Acta Geol. Polon., 26:299-319. Myczyfiski, R. and Pszcz6~kowski, A., 1976. The ammonites
Lewis, J.F. and Draper, G., 1990. Geology and tectonic evolution and age of the San Cayetano Formation from the Sierra del
of the northern Caribbean margin. In: J.E. Case and G. Dengo Rosario, western Cuba. Acta Geol. Pol., 26:321-330.
(Editors), Decade of North American Geology, Vol. H. The Myczyfiski, R. and Pszcz6~kowski, A., 1990. Tithonian stratigra-

120 A. P S z C Z O L K O W S K I
phy in the Sierra de los Organos, western Cuba: correlation of raphy in the western part of the Cuban miogeosyncline. Acta.
the ammonite and microfossil zones. Atti II Conv. Int. EE.A. Geol. Pol., 32: 135-161.
Pergola, 1987, pp. 405-415. Pszcz6tkowski, A., 1983. Tect6nica del miogeosinclinal cubano
Myczyriski, R. and Pszcz6lkowski, A., 1994. Tithonian stratigra- en el ~ea limftrofe de las provincias de Matanzas y Villa
phy and microfacies in the Sierra del Rosario, Western Cuba. Clara. Cien. Tierra Espacio, 6: 53-61.
Stud. Geol. Pol., 105: 7-38. Pszcz6~kowski, A., 1985. Sobre la edad del metamorfismo y la
Pardo, G., 1975. Geology of Cuba. In: A.E.M. Nairn and EG. estructura tect6nica de la faja Cangre, provincia de Pinar del
Stehli (Editors), The Ocean Basins and Margins, 3. The Gulf Rio, Cuba. Cien. Tierra Espacio, 10:31-36.
of Mexico and the Caribbean. Plenum Press, New York, pp. Pszcz6tkowski, A., 1986a. Composici6n del material chistico de
553-615. las arenitas de la Formaci6n San Cayetano, en la Sierra de
P6rez Lazo, J., Fundora Granda, M., Garcfa, A., Krop~i~ek, los Organos (Provincia de Pinar del Rio, Cuba). Cienc. Tierra
V. and Honi6ek, J., 1995. Paleomagnetic investigations in Espacio, 11: 67-79.
Cuba from late Jurassic to middle Eocene times and tectonic Pszcz6tkowski, A., 1986b. Secuencia estratignifica de Placetas en
implications. Acta Univ. Carolinae Geol. 38 (1994): 3-19. el ~ea limitrofe de las provincias de Matanzas y Villa Clara
Pessagno, E.A., Jr., Hull, D.M., Longoria, J.E and Kelldorf, (Cuba). Bull. Pol. Acad. Sci., Earth Sci., 34: 67-79.
M.E., 1993. Tectonostratigraphic significance of the San Pedro Pszcz61kowski, A., 1986c. Megacapas del Maestrichtiano de
del Gallo area, Durango, western Mexico. In: G. Dunn and Cuba occidental y central. Bull. Pol. Acad. Sci., Earth Sci., 34:
K. McDougall (Editors), Mesozoic Paleogeography of the 81-87.
Western United States, II. Pac. Sect. SEPM, Book 71, pp. Pszcz6tkowski, A., 1987. Paleogeography and paleotectonic evo-
141-156. lution of Cuba and adjoining areas during the Jurassic-Early
Pessagno, E.A., Jr., Cantd-Chapa, A., Hull, D.M., Kelldorf, M., Cretaceous. Ann. Soc. Geol. Pol., 57: 127-142.
Longoria, J.E, Martin, C., Meng, X., Montgomery, H., Fucu- Pszcz6{kowski, A., 1989a. La edad y posici6n de la secuencia
gauchi, J.U. and Ogg, J.G., 1999. Stratigraphic evidence for vulcan6geno-sedimentaria (Formaci6n E1 S~ibalo), en la es-
northwest to southeast tectonic transport of Jurassic terranes tructura geol6gica de la Sierra del Rosario (Cuba occidental).
in central Mexico and the Caribbean (western Cuba). In: E Geologfa/Geology '89 (Primer Congreso Cubano de Geologfa,
Mann (Editor), Caribbean Basins - - Sedimentary Basins of La Habana), Resumenes y Programa, p. 66.
the World, vol. 4. Elsevier, Amsterdam, pp. 123-149. Pszcz6tkowski, A., 1989b. Late Paleozoic fossils from pebbles in
Pindell, J.L., 1991. Geologic rationale for hydrocarbon explo- the San Cayetano Formation, Sierra del Rosario, Cuba. Ann.
ration in the Caribbean and adjacent regions. J. Pet. Geol., 14 Soc. Geol. Pol., 59: 27-40.
(3): 237-257. Pszcz6ikowski, A., 1990. Late Cretaceous volcanic arc in the
Pindell, J.L. and Barrett, S.E, 1990. Geological evolution of Bahia Honda terrane, western Cuba. In: Tectonostratigraphic
the Caribbean region; A plate-tectonic perspective. In: J.E. Correlation of Late Cretaceous-Tertiary Island Arc Rocks in
Case and G. Dengo (Editors), Decade of North American the Caribbean Region. Annual Meeting of the Geological
Geology, Vol. H. The Caribbean Region. Geological Society Society of America, Dallas, Abstracts with Programs, T 23, p.
of America, Boulder, CO, pp. 405-432. A338.
Pindell, J.L. and Dewey, J.E, 1982. Permo-Triassic reconstruc- Pszcz6{kowski, A., 1994a. Lithostratigraphy of Mesozoic and
tion of western Pangea and the evolution of the Gulf of Palaeogene rocks of Sierra del Rosario, western Cuba. Stud.
Mexico/Caribbean region. Tectonics, 1:179-211. Geol. Pol., 105: 39-66.
Pindell, J.L., Cande, S.C., Pitman III, W.C., Rowley, D.B., Pszcz6{kowski, A., 1994b. Geological cross-sections through the
Dewey, J.E, LaBreque, J.L. and Haxby, W., 1988. A plate- Sierra del Rosario thrust belt, western Cuba. Stud. Geol. Pol.,
kinematic framework for models of Caribbean evolution. 105: 67-99.
Tectonophysics, 155: 121-138. Pszczd~kowski, A. and de Albear, J.E, 1982. Subzona estruc-
Piotrowska, K., 1978. Nappe structures of Sierra de los Organos, turo-facial de Bahfa Honda, Pinar del Rio; su tect6nica y
western Cuba. Acta Geol. Pol., 28: 97-170. datos sobre la sedimentaci6n y paleogeograffa del Cret~icico
Piotrowska, K., 1993. Interrelationship of the terranes in western Superior y del Pale6geno. Cien. Tierra Espacio, 5: 3-24.
and central Cuba. Tectonophysics, 220: 273-282. Pszcz6{kowski, A. and de Albear, J.E, 1983. La secuencia vul-
Piotrowski, J., 1977. First manifestations of volcanism in the canogeno-sedimentaria de la Sierra del Rosario, provincia de
Cuban geosyncline. Bull. Acad. Pol. Sci., S6r. Sci. Terre, 24: Pinar del Rio, Cuba. Cien. Tierra Espacio, 6: 41-52.
227-234. Pszcz6lkowski, A., Piotrowska, K., Myczyfiski, R., Piotrowski,
Pszczolkowski, A. (-- Pszcz6lkowski, A.), 1994. Interrelation- J., Grodzicki, J., Skupinski, A., Haczewski, G., Danilewski,
ship of the terranes in western and central Cuba - - Comment. D., 1975. Texto explicativo para el mapa geol6gico a escala
Tectonophysics, 234: 339-344. 1:250000 de la provincia de Pinar del Rio. Unpubl. Rep.,
Pszcz61kowski, A., 1971. Jurassic, Cretaceous and Paleogene Inst. Geol. Paleontol. Fondo Geol. (MINBAS), La Habana.
deposits of Sierra del Rosario (Cuba). Bull. Acad. Pol. Sci., Pugaczewska, H., 1978. Jurassic pelecypods from Cuba. Acta
S6r. Sci. Terre, 19 (3/4): 249-259. Palaeontol. Pol., 23:165-186.
Pszcz6lkowski, A., 1977. Nappe structure of Sierra del Rosario Puscharovsky, Yu. (Editor), 1988. Mapa geol6gico de Cuba de la
(Cuba). Bull. Acad. Polon. Sci., S6r. Sci. Terre, 24 (3-4): Reptiblica de Cuba escala 1" 250000 (42 sheets). Acad. Sci.
205-215. Cuba.
Pszcz6lkowski, A., 1978. Geosynclinal sequences od the Pyle, T.E., Meyerhoff, A.A., Fahlquist, D.A., Antoine, J.W.,
Cordillera de Guaniguanico in western Cuba; their lithostratig- McCrevey, J.A. and Jones, EC., 1973. Metamorphic rocks
raphy, facies development, and paleogeography. Acta Geol. from north-western Caribbean Sea. Earth Planet. Sci. Lett., 18:
Pol., 28 (1): 1-96. 339-344.
Pszcz6tkowski, A., 1981. E1 banco carbonatado junisico de la Renne, ER., Scott, G.R., Doppelhammer, S.H., Linares Cala,
Sierra de los Organos, Provincia de Pinar del Rio: su desar- E. and Hargraves, R.B., 1991. Discordant Mid-Cretaceous
rollo y situaci6n paleotect6nica. Cien. Tierra Espacio, 3: 37- paleomagnetic pole from the Zaza terrane of central Cuba.
50. Geophys. Res. Lett., 18: 455-458.
Pszcz6{kowski, A., 1982. Cretaceous sediments and paleogeog- Rigassi-Studer, D., 1963. Sur la g6ologie de la Sierra de los

THE E X P O S E D PASSIVE M A R G I N OF N O R T H A M E R I C A IN W E S T E R N C U B A 121
Organos, Cuba. Arch. Sci. Soc. Phys. Hist. Nat. G6n~ve, 16: model of the Gulf of Mexico and Caribbean region. Tectono-
339-350. physics, 155: 139-148.
Rodriguez, E, 1987. Divisi6n estratigrfifica de la Formaci6n Ryabukhin, A.G., Tchekhovich, V.D., Zonenshein, L.P. and
Esperanza y comparaci6n de los cortes de las subzonas Es- Khain, V.E., 1984. Development of the Caribbean basin and
peranza y Sierra del Rosario. In: Memorias del III Encuentro the western part of the Tethys. 27 Int. Geol. Congr., Coll. 03,
Cientffico-t6cnico de Geologfa en Pinar del Rfo. Soc. Cubana Palaeoceanography, Rept. 3, Moscow, pp. 104-113.
de Geologfa, pp. 46-50. Segura Soto, R., Millfin, E. and Fern~indez, J., 1985. Complejos
Rosencrantz, E., 1990. Structure and tectonics of the Yucatan litol6gicos del extremo noroccidental de Cuba y sus implica-
basin, Caribbean Sea, as determined from seismic reflection ciones estratigrfificas de acuerdo con los datos de las perfora-
studies. Tectonics, 9: 1037-1059. ciones profundas. Rev. Tecnol., XV, Ser. Geol., 1: 32-36.
Rosencrantz, E., 1996. Basement structure and tectonics in the Viniegra, O.F., 1971. Age and evolution of salt basins of south-
Yucatan basin. In: M.A. Iturralde-Vinent (Editor), Ofiolitas eastern Mexico. Am. Assoc. Pet. Geol. Bull., 55: 478-494.
y arcos volcanicos de Cuba (Cuban ophiolites and volcanics Wierzbowski, A., 1976. Oxfordian ammonites of the Pinar del
arcs). IUGS/UNESCO Project 364, Contrib. 1, pp. 36-47. Rfo Province (western Cuba); their revision and stratigraphical
Ross, M.I. and Scotese, C.R., 1988. A hierarchical tectonic significance. Acta Geol. Pol., 26:137-260.

Chapter 5
Stratigraphic Evidence for Northwest to Southeast Tectonic

Transport of Jurassic Terranes in Central Mexico and the
Caribbean (Western Cuba)
EMILE A. PESSAGNO, JR., ABELARDO CANTIJ-CHAPA, DONNA M. HULL, MICHAEL KELLDORF,

JOSE E LONGORIA, CHRISTOPHER MARTIN, XIANGYING MENG, HOMER MONTGOMERY,
JAIME URRUTIA FUCUGAUCHI and JAMES G. OGG
Jurassic and Early Cretaceous stratigraphic data from terranes in Central Mexico situated southwest of the Walper Megashear
demonstrate similar records of paleobathymetry and tectonic transport. In general, each of these terranes shows the same
paleobathymetric fingerprint: (1) marine deposition at inner neritic depths during the Callovian to early Oxfordian (Middle to Late
Jurassic); (2) marine deposition at outer neritic depths during the late Oxfordian (Late Jurassic); (3) sudden deepening to bathyal
or upper abyssal depths (ACD = aragonite compensation level) from the early Kimmeridgian (Late Jurassic) until the end of the
Cretaceous. This paleobathymetric fingerprint differs markedly from that occurring to the east-northeast of the Walper Megashear
in the Coahuiltecano terrane (emended herein: ~Sierra Madre Oriental terrane). In the Coahuiltecano terrane (e.g., Peregrina
Canyon near C. Victoria, Tamps.), no Mesozoic marine deposits older than late Oxfordian occur. The paleobathymetric fingerprint
of this terrane was (1) inner neritic during the late Oxfordian (Late Jurassic) to ~Barremian (Early Cretaceous) and (2) bathyal to
abyssal during the remainder of the Cretaceous (Aptian to Maastrichtian). Though varying in detail, each succession that has been
examined in the mosaic of suspect terranes to the southwest of the Walper Megashear shows evidence of tectonic transport from
higher latitudes to lower latitudes during the late Middle Jurassic, the Late Jurassic, and the Early Cretaceous. For example, the
paleolatitudinal signature of the San Pedro del Gallo terrane (Durango) supplied by faunal data (radiolarians and megafossils) and
preliminary paleomagnetic data indicates that this terrane was transported tectonically from higher paleolatitudes (Southern Boreal
Province: ~40~ during the Late Jurassic (Oxfordian) to lower paleolatitudes (Tethyan Realm: Northern Tethyan Province) by
the Early Cretaceous (Berriasian). The Jurassic and Lower Cretaceous successions at Mazapil (Zacatecas), Sierra de la Caja
(Zacatecas), Sierra de Zuloaga (Zacatecas), Symon (Durango), and Sierra de Catorce (San Luis Potosi) are all genetically related
to that at San Pedro del Gallo. They are regarded as representing dismembered remnants of the San Pedro del Gallo terrane.
Faunal data (radiolarians and megafossils) from the Mazapil succession (Sierra Santa Rosa) indicate that this remnant of the San
Pedro del Gallo terrane was situated at Southern Boreal paleolatitudes (>30~ during the Oxfordian and Kimmeridgian and
at Northern Tethyan paleolatitudes (22 to 29~ during the Tithonian and Berriasian. Preliminary paleomagnetic data from the
upper Tithonian to Berriasian part of the Mazapil succession indicates 25~ Farther to the southeast (San Luis Potosi, Hidalgo,
Veracruz, Puebla) in the Huayacocotla segment of the Sierra Madre Oriental, previous investigations indicate tectonic transport
from Southern Boreal paleolatitudes (>30~ during the Callovian to Northern Tethyan paleolatitudes (22 ~ to 29~ during the
Kimmeridgian and Tithonian and to Central Tethyan paleolatitudes (<22~ during the latest Tithonian (Late Jurassic) and the
Berriasian (Early Cretaceous).
Jurassic and Early Cretaceous successions in western Cuba (Sierra del Rosario and Sierra de los Organos, Pifiar del Rfo
Province) show lithostratigraphic, paleobathymetric, and paleolatitudinal signatures which are nearly identical to those of San
Pedro del Gallo terrane remnants in central Mexico. They clearly represent portions of the North American Plate and are treated
as remnants of the San Pedro del Gallo terrane herein. The Cuban remnants of the San Pedro del Gallo terrane were carried
to eastern Yucatan by the Walper Megashear. By the Middle Cretaceous terrane amalgamation had occurred between the San
Pedro del Gallo and Coahuiltecana terranes and all movement along the Walper Megashear had ceased. Subsequent southwest to
northeast movement of the Caribbean Plate during the Late Cretaceous and Early Tertiary bulldozed the Cuban remnants of the
San Pedro del Gallo terrane into their present position. Once the Cuban San Pedro del Gallo remnants were carried northward by
the advancing Caribbean Plate, it is likely that they became part of an Atlantic-type margin.
Caribbean Basins. Sedimentary Basins of the World, 4 edited by P. Mann (Series Editor: K.J. Hs~i), pp. 123-150.

124 E.A. PESSAGNO et al.
INTRODUCTION San Pedro del Gallo terrane along the southwest

side of this megashear from higher Boreal paleo-
Mexico is the key component in plate tectonic re- latitudes (>30~ to lower Tethyan paleolatitudes
constructions involving the break-up of Pangea and during the Middle Jurassic, Late Jurassic, and Early
the subsequent formation of the Gulf of Mexico and Cretaceous.
the Caribbean (Figs. 1-3 and 7). The well-known
overlap position of Mexico and South America in
Atlantic reconstructions necessitates a mechanism IMPORTANCE OF FAUNALAND FLORAL DATAIN
for moving much of the Mesozoic succession of PALEOGEOGRAPHIC RECONSTRUCTIONS
Mexico away from its present-day position (see
Fig. 7, Inset A). Finding such mechanism has been Much of Mexico west of the Walper Megashear
a major task in geology since the first fit of the of Longoria (1985a,b, 1986, 1987, 1994) consists of
continents was suggested more than three decades suspect terranes or displaced terranes (Fig. 1). Pre-
ago by Carey (1958) and Bullard et al. (1965). vious studies by Taylor et al. (1984), Pessagno and
Walper and Rowett (1972) suggested that the off- Blome (1986), Pessagno et al. (1986, 1993a,b), and
set of the Mexico-Marathon-Ouachita-Appalachian Montgomery et al. (1992, 1994a,b) have established
structural belt resulted from transform movement the importance of faunal and floral data in pale-
along the Texas and Wichita megashears. Almost all ogeographic reconstructions in North America as
plate tectonic reconstructions of the Atlantic, Gulf of well as in the Caribbean. Recognition of displaced
Mexico, and Caribbean regions invoke megashears tectonstratigraphic terranes depends primarily on pa-
or transcurrent faults as mechanisms to transport leolatitudinal data derived from paleontology and
crustal blocks in Mexico and to explain the overlap paleomagnetism. Faunal and floral data can be used
position of Mexico in the reconstruction of Pangea to constrain existing paleomagnetic data and in some
(Fig. 7, Inset A). More than twenty tectonic models cases can also help determine whether tectonos-
have been proposed during the last two decades to tratigraphic terranes originated in the Northern or
explain the paleogeographic reconstruction of Mex- Southern Hemisphere or in the Eastern or Western
ico, the Gulf of Mexico, and the Caribbean (e.g., Pacific. For example, paleomagnetic data presented
Bullard et al., 1965; Dietz and Holden, 1970; Walper by Jones et al. (1977) indicate that the Wrangellia
and Rowett, 1972; Van der Voo et al., 1976; Buffter terrane originated 15~ north or south of the Triassic
et al., 1981; Dickinson and Coney, 1981; Walper, paleoequator. During the Late Triassic and Early
1981; Anderson and Schmidt, 1983; Pindell, 1985; Jurassic both the molluscan and radiolarian assem-
Longoria, 1985a,b, 1987, 1994; and so forth). blages tend to be predominantly Tethyan in origin
The purpose of this report is to present strati- (Tipper, 1981; Taylor et al., 1984; Pessagno and
graphic data that not only demonstrate the presence Blome, 1986). However, the discovery by Taylor et
of the Walper Megashear, but also demonstrate the al. (1984, pp. 128, 135) of very rare Boreal am-
northwest to southeast translation of remnants of the monites (amaltheids) in the upper Pliensbachian por-
Fig. 1. Map showingapproximateposition of area of displaced and suspect terranes west of WalperMegashear of Longoria.

N W TO SE T E C T O N I C T R A N S P O R T OF JURASSIC TERRANES IN MEXICO AND THE CARIBBEAN 125
Fig. 2. Index map showing important Jurassic localities in Mexico and Cuba. The most important localities for this report are 3-18,
24. Key to localities: 1 = Tlaxiaco: Sierra Madre del Sur, Oaxaca. 2 = Pletalcingo: Sierra Madre del Sur, Puebla. 3 = Huayacocotla
Anticlinorium: Taman-Tamazunchale, San Luis Potosi; Huayacocotla, Veracruz; Huachinango, Puebla. 4 -- Sierra Catorce, San Luis
Potosi. 5 = Sierra Santa Rosa, Zacatecas. 6 = Sierra de la Caja, Zacatecas. 7 = Sierra Cadnelaria, Zacatecas. 8 = Sierra Sombretillo
and Sierra Zuloaga, Zacatecas. 9 = Sierra de Ramirez, Zacatecas-Durango. 10 - Sierra de Chivo, Durango. 11 -- Sierra de Palotes,
Durango. 12 --- San Pedro del Gallo, Durango. 13 = Santa Maria del Oro, Sierra de la Zarca, Durango. 14 = Sierra Vieja-Arroyo Doctor,
Tamaulipas. 15 = Huizachal Anticlinorium, Tamaulipas. 16 =- Sierra Galeana-Iturbide, Nuevo Leon. 17 = Sierra de Parras, Coahuila. 18
- Sierra de Jimulco, Coahuila. 19 = Sierra Menchaca, Cohuila. 20 = Sierra Plomosas-Place de Guadalupe, Chihuahua. 21 -- Sierra E1
Cuchillo Parado, Chihuahua. 22 = Sierra de Samalayuca, Chihuahua. 23 = Sierra de Cucurpe, Sonora. 24 = Cordillera de Guaniguanico,
Cuba. Base map partly derived from that in Salvador et al. (1992).
-30 ~ \ ~. ~._ \ ~ ~.'~ \,------., .....\ . 30 ~

J \ r " - 7 "'~43 '~'~. "31
$,),.,,,,
9 if,, g '/~\ \
,.~.._ .,',,,,,> ~,, ~. \!
e "-h
Fig. 3. Megashear map of Longoria (1994).

tion of the Maude Formation of the Queen Charlotte RADIOLARIAN PALEOLATITUDINAL MODEL
Islands (B.C.) indicates that Wrangellia was situated
in the Northern Hemisphere. Moreover, the presence It is now apparent from our analyses of radi-
of the pectenacid Weyla in Lower Jurassic strata olarian faunal data from North America and else-
in the Queen Charlotte and other remnants of the where in the world that radiolarians can be utilized
Wrangellia terrane indicates that this terrane origi- in paleobiogeographic investigations and to moni-
nated in the Eastern Pacific (Smith, 1980; Taylor et tor the tectonic transport of terranes both in the
al., 1984; Pessagno and Blome, 1986; Pessagno et Northern and Southern Hemispheres. Much of the
al., 1986). Circum-Pacific margin is comprised of a collage of
Although the focus of the present report is on tectonostratigraphic terranes. Many of these terranes
geologic terranes immediately southwest of and ad- have been displaced paleolatitudinally for hundreds,
jacent to the Walper Megashear, it is worth noting or in some cases, possibly thousands of kilometers.
that faunal data from a variety of sources suggest Circum-Pacific Jurassic paleogeography, as a result,
either northwest to southeast movement or south- is difficult to discuss in simplistic terms and must be
east to northwest movement along other possible viewed through this complex mosaic.
megashears (Fig. 3). In the states of Oaxaca and In the Northern Hemisphere Pessagno and Blome
Guerrero Burckhardt (1927, 1930) was the first to (1986) and Pessagno et al. (1986, 1987, 1993a,b)
record the presence of a Middle Jurassic (Bajocian divided the Tethyan Realm into a Central Tethyan
to Callovian) ammonite assemblage which strongly Province characterized by a radiolarian assemblage
resembles that in the Andes, specifically Argentina with high pantanelliid abundance and diversity and
(Figs. 1 and 2). Subsequently, the strong Andean the absence of Parvicingula/Praeparvicingula and
affiliation of the Middle Jurassic ammonite assem- into a Northern Tethyan Province with high pan-
blage of Oaxaca and Guerrero has been noted by tanelliid abundance and diversity and common
Arkell (1956), Imlay (1980), Sandoval and West- Parvicingula/Praeparvicingula (Fig. 4). The Bo-
ermann (1988), Sandoval et al. (1990), and von real Realm was subdivided into a Southern Bo-
Hillebrandt et al. (1992). The fact that no Middle real Province and a Northern Boreal Province.
Jurassic ammonite faunas with strong Andean af- The Southern Boreal Province is characterized by
filiation are known from elsewhere in Mexico and a sharp decline in pantanelliid abundance and diver-
North America suggests that the Middle Jurassic sity and by the abundance and diversity of species of
(Bajocian to Callovian) succession in these states Parvicingula/Praeparvicingula; the Northern Boreal
has undergone southeast to northwest tectonic trans- radiolarian assemblage is distinguished by abundant
port along a more outboard megashear paralleling Parvicingula/Praeparvicingula and by its total lack
the Walper Megashear (i.e., the Cserna Megashear of pantanelliids. In Fig. 4 the boundary between
of Longoria, 1994; Fig. 3). the Tethyan Realm and Boreal Realm is placed at
In the Vizcaino Peninsula (Baja California Sur; ~30~ the boundary between the Central Tethyan
Fig. 1), radiolarians, though abundant and well-pre- Province and the Northern Tethyan Province is es-
served in strata of Early Jurassic (Pliensbachian) tablished by associated paleomagnetic data at ,~22~
to Late Cretaceous (Cenomanian) age, are repre- (Pessagno et al., 1987; Yeh and Cheng, 1996). Pes-
sentative of a poorly diversified Boreal assemblage sagno and Blome (1986) originally proposed that the
(Whalen and Pessagno, 1984; Whalen, 1985; Davila- model for the Southern Hemisphere is the mirror
Alcocer, 1986; Pessagno et al., 1986). Foraminifers image of that in the Northern Hemisphere. Subse-
occurring in the Late Jurassic to Late Cretaceous quently, new data from the Southern Hemisphere has
strata of the Vizcaino Peninsula are likewise Boreal substantiated this model from Argentina (Pujana,
in nature and show strong affinity to those of the 1989, 1991, 1993, 1996), New Zealand (Aita and
California Coast Ranges. In fact, studies made by Grant-Mackie, 1992), Antarctica (Kiessling, 1995;
Longoria (unpublished PEMEX report) indicate that Kiessling and Scasso, 1996), and the Sula Islands
the planktonic foraminiferal assemblage occurring (Pessagno and Hull, in prep.).
in the Upper Cretaceous Valle Formation is like
that described by Douglas (1969) from the Great
Valley Supergroup (California Coast Ranges). This PALEoLATITUDINAL RECONSTRUCTIONS USING
fauna is not well-developed south of the latitude of MULTIPLE CRITERIA
Bakersfield in California and occurs northward to
Alaska and to Japan. In addition to the foraminifers Although radiolarian paleobiogeographic recon-
and radiolarians the Late Jurassic (Tithonian) to structions are useful and can stand alone, they are
Early Cretaceous (Valanginian) strata of the Euge- far more effective when combined with information
nia Formation contain a Boreal bivalve assemblage derived from paleomagnetism, analysis of the to-
characterized by species of Buchia (cf. Fig. 5). tal faunal and floral assemblage, and other criteria

NW TO SE TECTONIC TRANSPORT OF JURASSIC TERRANES IN MEXICO AND THE CARIBBEAN 127
Fig. 4. Paleolatitudinal model based on distribution of selected radiolarians from the Jurassic and Lower Cretaceous.
having paleolatitudinal or paleolongitudinal signif- tudinal position of a given terrane from the study
icance (see Fig. 5). The tenet stressed herein is of the radiolarian assemblage alone. Conversely,
that paleogeographic reconstructions should use all in successions where megafossils are present and
criteria available, where possible, and should not the radiolarians remain unstudied, it is possible to
focus on any one facet (e.g., analysis of only the predict the character of the radiolarian assemblage
ammonite assemblage). Nevertheless, even in eugeo- from that of the paleogeographic character of the
clinal terranes, where other fossils are absent, it is megafossil assemblage. This was, in fact, the case in
often possible to determine the relative paleolati- our initial studies of the San Pedro del Gallo area.
Because there was a mixture of Boreal ammonites
such as Amoeboceras sp., cf. alterans (von Busch)
(Burckhardt, 1930; Imlay, 1980) and Buchia associ-
TETHYAN REALM BOREAL REALM ated with Tethyan ammonites, the senior author was
I
CENTRAL TETHYAN
[
! NORTHERN SOUTHERN NORTHERN BOREAL
able to predict that the radiolarian assemblage, if
PROVINCE
i
I,
TETHYAN
PROVINCE
BOREAL
PROVINCE I
PROVINCE
present in these strata, would be assignable to the
- ~-- TETHYAN AMMONITES ............... ~ > !
Southern Boreal Province.
< I
.d < ............................ ...............................
BOREAL AMMONITES ........
o~ ~ Weyla ? ~ ...... ~,~,~~Buchia spp. ,........

I ,
C R I T I C I S M O F T H E M O D E L O F P E S S A G N O AND
9 ~ ............ PANTANELLIIDS~ ..............................~I BLOME (1986) BY H A G S T R U M AND M U R C H E Y (1996)
~ ~ Acanthocircus dciranocanthos=~,l .... i I
z I ~ ...... Parvicingula/Praeparvicingula .................... Hagstrum and Murchey (1996) in a report entitled

'Paleomagnetism of Jurassic radiolarian chert above
" o I
the Coast Range Ophiolite at Stanley Mountain,
California and implications for its paleogeographic
origins' challenged the validity of the methodology
~ "' . . . . . . Xiphostylus .: . . . . ~s,
u
uJ
oO
of Pessagno and Blome (1986) and Pessagno et al.
.............. Tripocyclia ...... , ~ (1987, 1993a). In spite of such criticism, we are
i appreciative of the paleomagnetic data that these
i U3
CALPIoNELLIDS,
.... ~>, z~- authors obtained from the upper part of the vol-
i . .~. . GINKGoALES canopelagic succession at Stanley Mountain (i.e.,
t g Dakota)
'e.-., '~-'---' ..... ~ .....
i
I
Z 32 + 8~ These data support the previous conclu-
sions of Pessagno et al. (1984) and Hopson et al.
- ~ . . . . . ~ ..... ~ . . . . . (1996) that this Coast Range Ophiolite remnant and
:!i:~ ......... ........N .... " i ....... ,:iT:'~;1 its overlying sedimentary cover was at mid to high
............
Fig. 5. Multiple criteria for use in paleobiogeographic recon- latitudes (Southern Boreal Province) by Tithonian
structions in the Northern Hemisphere. times. As noted in the rebuttal by Hull et al. (1997),

Hagstrum and Murchey's interpretation of the paleo- tary environments ranging from open ocean beyond
geographic model presented by Pessagno and Blome the reach of terrigenous or volcanic input (red rib-
(1986) and Pessagno et al. (1986) is inaccurate. The bon cherts and also some pelagic limestone such
following points merit discussion herein. as that occurring in the pillow lava at La D6sir-
(1) The distribution of Praeparvicingula and Par- ade) to open ocean down-wind from an island arc
vicingula (= 'homed parvicingulids' of Hagstrum (e.g., tuffaceous chert above Coast Range Ophio-
and Murchey) cannot be linked to the environment lite at Point Sal, Santa Barbara County, California)
of deposition based on rock type as suggested by to backarc, interarc, and forearc environments (e.g.,
these authors. 'chert member' of La Caja Formation at San Pedro
(2) The distribution of pantanelliids as a criterion del Gallo (= backarc), Rogue Formation, Klamath
is still useful although it must be used with caution. Mountains, Southwestern Oregon (= interarc), and
(3) Pessagno and Blome (1986) and Pessagno Snowshoe Formation, Izee terrane, east-central Ore-
et al. (1986, 1987, 1989, 1993a,b 1993b)stressed gon (= backarc: see Pessagno and Blome, 1986)).
the use of multiple criteria (see P A L E O L A T I T U D I - (2) The distribution of pantanelliids as a crite-
NAL RECONSTRUCTION USING MULTIPLE CRITE- rion is still useful although it must be used with
RIA) rather than just the presence of pantanelliids caution. As noted by Pessagno et al. (1986, p. 8),
and Praeparvicingula and Parvicingula as indicated the abundance and diversity of pantanelliids in ra-
by Hagstrum and Murchey. diolarian chert, siliceous mudstone is controlled by
(4) The Pessagno and Blome model is not depen- diagenesis. However, it is also influenced by the
dent on paleomagnetic data from Stanley Mountain method of extracting the microfossils from rock
as indicated by these authors. samples using the hydrofluoric acid (Blome and
These points will be discussed below. For a more Reed, 1993). The fragile nature of many pantanelliid
in depth discussion the reader should refer to Hull et taxa prevents them from being preserved in sedimen-
al. (1997). tary strata which were metamorphosed or underwent
(1) The distribution of Praeparvicingula and Par- lithification subsequent to deposition. Crushing and
vicingula (= 'horned parvicingulids' of Hagstrum stretching of specimens extracted from radiolarian
and Murchey) cannot be linked to the environment of chert and shale is common and usually results in the
deposition based on rock type as suggested by these total destruction of all fragile radiolarians. Only radi-
authors. Hagstrum and Murchey (1996, p. 649) fol- olarians with thick-walled, sturdy tests are preserved
low Baumgartner (1987) in stating that "In general, (e.g., Parvicingula, Praeparvicingula, Mirifusus, Ar-
homed parvicingulids occur in hemipelagic rocks chaeocenosphaera) although these forms may often
such as tuffaceous mudstone or gray, mudstone, and be quite abundant. The best recovery of well-pre-
siltstone". This statement is erroneous. It infers that served pantanelliids, and indeed all radiolarians oc-
'homed parvicingulids' only occur in coastal en- curring in Mesozoic strata, comes from limestone.
vironments. Parvicingula and Praeparvicingula not Radiolarian diversity is at least three times greater
only occur in hemipelagic strata, but they also oc- in limestone strata than it is in adjacent chert or
cur in pelagic strata such as red manganiferous mudstone layers. This is in part due to two factors:
ribbon cherts at a number of localities throughout (1) the use of the hydrofluoric acid technique (Pes-
the world, some of which occur in Hagstrum and sagno and Newport, 1972) to extract the radiolarians
Murchey's own backyard in the California Coast from siliceous strata and the use of HC1 to extract
Ranges. Examples of such occurrences are as fol- radiolarians from limestone, and (2) the early lithi-
lows: (a) thousands of meters of Upper Jurassic fication of limestone strata at the time of deposition
oceanic plateau-type basalt interbedded with red as opposed to the post-depositional lithification of
manganiferous ribbon chert in the Franciscan Com- the chert. Blome and Reed (1993) demonstrated that
plex at Wilbur Springs and Stoneyford (locs. 17, the use of the HF technique invariably results in the
18 of Hopson et al., 1981); (b) red ribbon cherts destruction of radiolarians with more fragile tests
throughout the Greater Antilles in the Caribbean (e.g., most pantanelliids). The most dramatic exam-
(e.g., volcanic member of ophiolite complex in La ple exemplifying the early lithification of limestone
D6sirade, ophiolite remnants in Bermeja Complex close to the time of deposition comes from a study
of southwestern Puerto Rico, and ophiolite remnants of Pessagno et al. (1993a) of the volcanopelagic
in the Duarte Complex of the Dominican Republic succession overlying the Josephine Ophiolite (Smith
(Montgomery et al., 1992, 1994a,b); (c) subduction River Subterrane, Klamath Mountains, Northwestern
complex in the Philippines (Yeh and Cheng, 1996). California). At this locality (Middle Fork of Smith
Red ribbon cherts from all of these localities con- River) volcanopelagic strata consisting of dark-gray
tain Parvicingula and Praeparvicingula and clearly to greenish-gray tuffaceous chert and light-gray
lack any sort of terrestrial input. It is apparent that pelagic limestone overlie the Josephine Ophiolite
these taxa flourished in a wide range of sedimen- and underlie the flysch of the Galice Formation.

Identical radiolarian chert occurs within the volcanic all radiolarians, is greatly influenced by the care
member of the Josephine Ophiolite. In the inter- taken in sample processing. The recovery of abun-
val including the volcanic member of the Josephine dant pantanelliids in the Oxfordian and Kimmerid-
Ophiolite and the overlying volcanopelagic strata, gian parts of other CRO remnants could also be
the limestone strata produced about three times more influenced by other factors, such as preferred high-
radiolarian taxa than did the chert strata. Moreover, fertility upwelling areas. Although Hagstrum and
about four times more pantanelliid taxa occur in Murchey selectively point to pantanelliids as favor-
limestone strata than in chert strata. Other examples, ing high-fertility upwelling areas, it could be equally
of this sort are cited in studies by Blome and Reed expected that parvicingulids flourished in such ar-
(1993) and Hull (1995). eas, subject, however, to different water temperature
Although there are undoubtedly cases where pan- (paleolatitude) controls. The published literature no-
tanelliids can not be used in paleogeographic recon- tably lacks any reference to the comparative pref-
structions because of the factors noted above, it is erence of pantanelliids over parvicingulids for high
important to point out that this group of radiolarians fertility upwelling areas.
has been successfully utilized to establish paleolat- Regarding this link between high fertility and
itudes by Pessagno and Blome (1986) for the Izee radiolarian assemblages, we also question Hagstrum
terrane (east-central Oregon) during the Late Trias- and Murchey's use (their fig. 8) of Lisitizin's 1972
sic, Early Jurassic and Middle Jurassic, by Pessagno map, showing spatial distribution of the annual pro-
et al. (1984, 1987) for the Huayacocotla remnant duction of silica by marine organisms in the world
of the San Pedro del Gallo terrane during the Late ocean, for the purpose of advancing their thesis
Jurassic and Early Cretaceous, and by Pessagno et that 'the tuffaceous Stanley Mountain cherts (CRO)
al. (1993a) for the Smith River Subterrane during the were likely deposited at ~30~ within a high pro-
Middle and Late Jurassic. In all of these examples, ductivity zone near the western margin of North
radiolarians were extracted from either bedded lime- America'. This map for the modem ocean illus-
stone or from limestone nodules using hydrochloric trates silica production predominantly for diatoms
acid, were exposed to no higher than prehnite- near continental coast lines. Diatoms are chief sil-
pumpellyite to greenschist metamorphism, and were ica producers in modem oceans and are responsible
well-preserved. In the Izee terrane, for example, for more than seventy percent of the total marine
well-preserved, abundant and diversified pantanelli- silica (Kennett 1982); silica production from radio-
ids associated with Tethyan megafossils characterize larians ranks a distant second among siliceous plank-
the Late Triassic (Karnian-Norian) Rail Cabin For- ton, and presumably, silicoflagellates and siliceous
mation, the Early Jurassic Nicely Formation (late sponge spicules contribute silica as well. Moreover,
Pliensbachian), the Early Jurassic Hyde Formation it is well known that radiolarians generally are not
(early to middle Toarcian), and the Early Jurassic as abundant in nearshore waters of modern oceans
part of the Warm Springs Member of the Snowshoe (Kennett 1982) as diatoms in such settings, partic-
Formation (middle to late Toarcian, part). Moreover, ularly in eastern boundary regions. We believe that
such an assemblage also characterizes the Aalenian, the high production of biogenic silica in the coastal
early Bajocian, and late Bajocian (Middle Jurassic) regions bordering North America during the Recent
parts of the Snowshoe Formation (Warm Springs (Hagsrum and Murchey, fig. 8) reflects diatom pro-
Member to lower part of the South Fork Member. duction and therefore, once again, provides no clue
However, as noted by Pessagno and Blome (1986) to the geographic distribution of Stanley Mountain
pantanelliid diversity and abundance and diversity radiolarians in the Late Jurassic ocean."
drops dramatically in the late Bathonian part of the (3) Pessagno and Blome (1986) and Pessagno
South Fork Member of the Snowshoe Formation et al. (1986, 1987, 1989, 1993a)stressed the use
and in early Callovian Lonesome Formation. The of multiple criteria rather than just the presence
great drop in diversity and abundance of pantanel- of pantanelliids and Praeparvicingula and Parvic-
liids can be directly related to the first occurrence ingula as indicated by Hagstrum and Murchey. As
of Boreal ammonites such as Kepplerites and Pseu- noted by Hull et al. (1997), in their discussion of
docardoceras in the upper part of the South Fork the paleogeographic foundation for the Pessagno
Member (Pessagno and Blome, 1986; Pessagno et and Blome model Hagstrum and Murchey neglect
al., 1986). to mention that the model's realms and provinces,
Hull et al. (1997) stated "Thus facing both suc- are by definition (Pessagno and Blome, 1986), al-
cesses and questions about pantanelliids, we agree ways constructed on multiple criteria. This thesis
that much remains to be solved concerning the pa- was repeatedly stressed subsequently in a series of
leoceanographic and/or paleolatitudinal preferences reports by Pessagno et al. (1986, 1987, 1993a,b)
of this group of radiolarians. It must be remem- and Blome (1987) (see PALEoLATITUDINAL RE-
bered that the recovery of pantanelliids and, indeed, CONSTRUCTION USING MULTIPLE CRITERIA).

(4) The Pessagno and Blome model is not de- is flanked to the north by the flysch of the Parral
pendent on paleomagnetic data from Stanley Moun- terrane (Coney and Campa, 1984) and to the east
tain as indicated by these authors. Hagstrum and by the Sierra Madre Oriental terrane (= Coahuilte-
Murchey (p. 650) indicate that model presented cano terrane of Sedlock et al., 1993; emend, herein;
by Pessagno and Blome (1986) incorporates lit- see Fig. 7). According to Coney and Campa (1984,
tle quantitative paleolatitudinal control. Moreover, p. D-3) the 'Parral Terrane' includes "highly de-
they suggest that the placement of the Central formed, partly calcareous Upper Jurassic and Lower
Tethyan-Northem Tethyan boundary at 22~ was Cretaceous turbiditic sandstone". "The Sierra Madre
based on the Stanley Mountain paleomagnetic data Oriental Terrane" includes "deformed upper Meso-
(i.e., 14 -4- 7~ presented by McWilliams and zoic sedimentary rocks of the Gulf of Mexico trans-
Howell (1982). This statement is totally erroneous. gressive sequence and their diverse basement rocks
The province boundary was tentatively placed at which include, at different places, Pre-Cambrian
22~ because of the presence of Central Tethyan crystalline rocks and structurally juxtaposed Paleo-
Berriasian faunas occurring at 20~ at DSDP Site zoic sedimentary rocks, Lower Jurassic sedimentary
534 (Blake Bahama Basin: Ogg, 1983; Baumgart- rocks, structurally associated with Pre-Cambrian and
ner, 1984; Pessagno et al., 1987, p. 7). In addition, Paleozoic rocks, and pre-Late Jurassic red beds and
it is now supported by new data from the Philip- volcanic rocks." Our data indicate that the San Pedro
pines (Yeh and Cheng, 1996). Quantitative data was del Gallo area includes strata that were deposited in
also cited by Hopson et al. (1996) for the Llanada the Boreal Realm during the Late Jurassic. The San
and Point Sal remnants of the CRO and by Pes- Pedro del Gallo succession is juxtaposed (along the
sagno (1995) for west-central Mexico. There is little Walper Megashear) against strata to the east ('Sierra
question, however, that more paleomagnetic data are Madre Oriental Terrane') which are both miogeo-
needed by workers in future studies. It should be clinal and Tethyan in character. West-southwest of
pointed out, however, that 'quantitative' paleomag- the Walper Megashear the geology is considerably
netic data for the Phanerozoic is completely de- more complex and is indicative of a back-arc setting.
pendent on biostratigraphically derived chronostrati- The presence of common basaltic andesite clasts
graphic data. The Phanerozoic chronostratigraphic in coeval Upper Jurassic to Lower Cretaceous olis-
scale is based on fossil biozones. Moreover, the tostromal deposits coupled with beds of green and
geochronologic scale ('geologic time scale') is de- red silty tuff and associated graywacke at the village
rived from the integration of biostratigraphic and of Cinco de Mayo to the north of San Pedro del
chronostratigraphic data with geochronometric data Gallo (Fig. 8) reflects a more proximal back-arc
(e.g., U/Pb dates). origin.
Pessagno et al. (1993b) suggested that the Juras-
sic successions at Mazapil and in the Huayacocotla
REMNANTS OF THE SAN PEDRO DEL GALLO segment of the Sierra Madre Oriental (Longoria,
TERRANE 1984) are genetically related to that at San Pedro
del Gallo and were possibly remnants of the SPG
The San Pedro del Gallo terrane (SPG) was first terrane. Subsequent investigations have proven this
defined by Pessagno et al. (1993b) for the area first to be the case. In this report we regard the Upper
studied in 1910 by the famous Swiss geologist, Car- Jurassic and Lower Cretaceous successions at San
los Burckhardt in the state of Durango (Fig. 2: Loc. Pedro del Gallo, Symon and Sierra Ramirez, Maza-
12; Fig. 8). Burckhardt's contributions to the geology pil (Sierra Santa Rosa), Sierra de la Caja, Sierra
of San Pedro del Gallo were five-fold: (1) he estab- Zuloaga and Sierra Sombretillo, Sierra Cadnelaria,
lished the succession/superposition of Jurassic and Sierra de Catorce, and in the Huayacocotla Anti-
Cretaceous strata by utilizing ammonite biostratig- clinorium to represent remnants of the SPG terrane
raphy and chronostratigraphy; (2) he monographed (see Fig. 2: Locs. 12, 9, 5, 8, 7, 6, and 3; Bur-
the rich megafossil assemblage (largely ammonites; ckhardt, 1930, 1931; Imlay, 1980). Moreover, it is
Burckhardt, 1912); (3) he made the first geologic clear that the Middle Jurassic to Lower Cretaceous
map as well as the first topographic map of the succession in western Cuba (Fig. 2: Loc. 24) also
area; (4) he interpreted the structure and produced includes remnants of the SPG terrane. In addition
numerous structural profiles; and (5) he assessed the to displaying the similar lithostratigraphic and pa-
mineral deposits of the area. Burckhardt's study was leobathymetric signatures, all of the remnants of
so thorough that few workers have been able to the SPG terrane show evidence of tectonic transport
improve on it. from higher latitudes to lower latitudes. The mecha-
Fig. 6 shows a comparison of the geology of the nism for this displacement in Central Mexico is the
San Pedro del Gallo terrane (SPG) to that of adjacent Walper Megashear (Figs. 2, 3 and 7). Tectonostrati-
Parral and Coahuiltecana terranes. The SPG terrane graphic data to support this thesis is presented below.

N W T O SE T E C T O N I C T R A N S P O R T O F J U R A S S I C T E R R A N E S IN M E X I C O A N D T H E C A R I B B E A N 131
San Pedro del Gallo Coahuilatecana

Characteristics Parral terrane
terrane terrane
Flysch: Rhythmically bedded Massively bedded quartzite and Thin-bedded to massively
graywacke and shale. interbedded nerineid bedded gray sandstone,
"TurbiditJc sandstone" of limestone, organic-rich, black pink silty siliceous shale and
Lithologic shale, and mudstone with
Coney and Campa (1984). mudstone with interbedded
Characteristics numerous limestone nodules,
Thickness unknown. Rocks pink to buff, silty micritic
partially coeval with those of black radiolarian chert, limestone, and chalky
San Pedro del Gallo terrane. graywacke, tuff, and thin mudstone, etc.
bedded micritic limestone.
Domain Eugeoclinal: Back Arc Eugeoclinal: Back Arc Miogeoclinal
Bathyal to upper abyssal.

Bathyal Neritic. Mostly shallow
Water Depth Above depth of
to Abyssal ned'dc
compensation of aragonite
Paleolatitudinal Signature
during Unknown Southern Boreal to Tethyan Entirely Tethyan
Jurassic and Cretaceous
Hummocky, irregularly Hummocky, irregularly Parallel, linearly arranged

Geomorphic Expression
trending ranges trending ranges ranges
Folds relatively small scale. Folds relatively small scale.
Folds open, parallel, more
Folding tight, intricate. Folding tight, intricate.
Structural Style regularly trending, very large
Rocks more faulted than those" Rocks more faulted than those
scale..
of SMO terrane of SMO terrane
Fig. 6. Comparison of San Pedro del Gallo terrane to Coahuiltecano terrane.
[----~ .... r ...... ~...... . ....... ~ ...... v,..
"~... /.. :,
j,,'<//\ Inset A
[ .... \,,.,..',.. ~ ' , , ~ % '.,)~;-, Inset B
Li 7-.! ,7i~
' i '!:;i,.,,:.i-;,,-7:~,.
~{~.,' ~---~-...../.~-..
.... " A .Z : : "~. ' "-~-
"hi '<!t I';,,-~><.,, .~--.~.',.'-; ] ~.1 ' ~, ~'- ~" ~ ~i ,,~ k "-~, "'; ~-,
i> -.,':...It
.:," "; " . ~;<,i:../ ",
i ". '.....!-:::-<"/-. ! /""
'~.. .t,t" ,'---.-.~ .... j

i~...........7;,.... I / "
E / ..........i.;'A../ !-"/ i "-~~~?-<;.">--~---~-.' <:
g "~I > "_.,,...... /.."
_ 30 ~
.,.;" ( GULF
OF
Cuba
MEXICO
........................................ Tropic of Cancer ............................
Cuban r e m n a n t s SPG T e r r a n e ~---J'r-~i~~'-"~[
N PACIFIC
T OCEAN CARIBBEAN
+""~// I t:-,,:'~
-,o ~......~Y--" t -
,~ # 9 " ""-':'.-
re, 9
+2
S C A L E
'":'~ SEA
o leo
t j I ..............'%\ ( Guatemala
Fig. 7. New map showing terranes, first-order megashears, Sedlock et al.'s terrane map, and Pangea at end of Pe.
ANALYSES OF SAN PEDRO DEL G A L L O T E R R A N E the Walper Megashear. Moreover, Fig. 7 shows the
REMNANTS terranes utilized by Sedlock et al. (1993), some of
which are emended or abandoned herein (see In-
Figs. 2 and 7 shows the position of San Pedro set B of Fig. 7). Fig. 9 is a correlation chart show-
del Gallo (SPG) terrane remnants to the west of ing the correlation of lithostratigraphic units in each

104 ~
26 ~
/z
'~I :~ceanls~cGoeaOl~~176
l _ Gom~ ,~i
"~% ~ U S a n Luis del Cordero
I
104 ~ l
0 Kilometers 100
Fig. 8. Index map for San Pedro del Gallo.
Mazapil Sierra Santa San Pedro del Gallo, Sierra Catorce, Taman--Tamazunchale, Western Cuba Megafossil Radiolarian
San Luis Potosi Si ,~rra Sierra de
Rosa, Zacatecas Durango San Luis Potosi del Fosario los Organos Biohorizons Biozones
,,,,i Chapulhuacan Chapulhuacan Chapulhuacan ~E~ ~
._.1 Be ins Limestone Limestone unnamed Ims. Limestone ~ ~ ~ Subzone
PimientaFormation 4or
.s
La Caja .~_
o Formation '~
z ~-, (= "Capas de San E El Verde Taman Formation
(./.) ,-, o "upper thin-bedded C
'< = La Caja Pedro", part, " Member
c/3
oz Formation Burckhardt, 1910) ~' member" o Subzone
rv" -1- Units E-B o sensu
.~ N ~ ~
N
-- Pessagno et al.
C 4p
223 ~ E
1987a
__ ,.. .._ , ... ,-...
!::!::i::iiiiii!iiiii!ii::!!!ii!
:::::::::::::::::::::::::::: ,
Pastor Mum.
... ... ,.,
o~ /i:!:!:i:!~!;;:!~i:i:!:i:i:i: :!:i:!:!:i:i:i:i:!:!:!:!:i:i:l TamanFm.

cr-
w ,._ iiiii;ii~!~i~ii!i!ii:iiiiiiiiiSi~tij~,iiiiiil ..~,~....~. ~ .
Zone 3
13... :::::::i:::!:! ~ Sz. 2cz 1
13_ ~ ~ LaCajaFmUnitE+
, upperUnilF
. - - La Ca'a Fm PastorMum . . a. .m. . a. .n. . . .Fm
T
!!~!i!i!~ii!ii~ Sz. 2ot2
_o ~ : . . . . i T "lowermassiv_.___~e
mum."
i!~i~iI~Siiiiliiiiiiiiiii
Sz. 213
z ~ ~:~i~~i::!::::ii::iHiatLi.S::ii::iii:: C
"::1: - -~ unnamedredshale,siltstone, unnamed
- - red shale
unnamed_redshale, [::ii!i!i!i::iii!::!iiiiiiiii!ii ~rancisco Ja0ua
~ &limestone(UnitsF&GI siltstone,&limestone ::::::::::::::::::::::::::::::::::::::: Fn. Fm t~ - Sz. 2y
or'v" ~F: ......." " _. . 1' ""~"'''"" ""
u_ ~ ZuloagaLimestone ZuloagaLms. tongue I Santiano Sz. 26
oX ~,.-- tonguebetweenupper in imddle of unnamed I Formation %--
o & lowerquartziteunits quartzite unit J Zuloaga " - -- Zone 11
o
z
z
g. ........ -.... " " -.... -........ ;:::~ -........ -.-I
, , ~ ; ,r-.,
= " //
~. ii5 "5) i15 ~. 5.1......15 ii/iI
.//- //" .//- ./// //-
Limestone
?
ii!i!iiiiiiiii!i!iiiii!iiii
ii!ii!ilHi~itL~Sil;iiiil
=o
Zone 1 H
C ) , ~
- - ~ i ~
- . . o
E~ --, = ........ .......... ......... / . ...... az ........ / ! i !iiii!i!iii!!iii!iiiii!i
c,-- O ~ "/ z..-" ..... ~s~n,~!e.~,.,..se Zone 1G
~L__
c_)!
< S" ...... ........ . :!:?i?i?i?i?i?i?:?:?ii:?:!:
9 i~
9 Zz ~ ,
c)l~
-7- i N-
t--. r--c-~ Zone 1 F
03
Fig. 9. Correlation chart showing radiolarian biozones and ammonite and Buchia markers, and correlation of litho units in all SPG terrane
remnants.
SPG remnant with chronostratigraphic and biostrati- metric interpretations. Fig. 11 shows the correlation
graphic units. Fig. l0 shows the depth distribution of of lithostratigraphic units with paleobathymetry and
important Mesozoic fossil groups used in paleobathy- paleogeographic position; moreover, it attempts to

Fig. 10. Depth distribution of various Mesozoic fossil groups.
demonstrate the significance of lithostratigraphic fea- as middle Oxfordian or older via the superposition
tures (e.g., unconformities) that record regional North of overlying strata containing middle Oxfordian am-
American tectonostratigraphic events. Figs. 12-17 monites (Figs. 9, 11 and 12). Martin's (1996, p.
are detailed stratigraphic summaries of each terrane 105) microfacies analysis of the Zuloaga Limestone
remnant that supplement Figs. 9 and 11. indicates that it consists of bioturbated micrite ('lime
mudstone') and peloidal lime grainstone. The faunal
San Pedro del Gallo remnant assemblage of the Zuloaga Limestone is characteris-
tic of inner neritic depths. It contains nerineids, other
The analysis of the San Pedro del Gallo remnant gastropods, rare corals, echinoid fragments, and rare
is based on ongoing investigations since 1990 (Pes- benthonic Foraminiferida (Miliolina and Textulari-
sagno et al., 1993b; Martin, 1996; Meng, 1997) as ina) (Martin, 1996). The presence of miliolids in the
well as previous observations by Burckhardt (1910, Zuloaga Limestone indicates that it was deposited at
1912, 1930) and Imlay (1939). Fig. 2 shows the po- depths no greater than 100 m (Fig. 10). Martin noted
sition of the San Pedro del Gallo terrane to the west that most of the allochems are well-sorted peloids.
of the Walper Megashear (see Fig. 2: Loc. 12). The He noted that many of the peloids have poorly de-
understanding of the stratigraphy of the San Pedro fined rims that may indicate that they were formed
del Gallo succession was hampered for many years by the rolling action of near-shore waves. Moreover,
by miscorrelation of several lithostratigraphic units. he suggested that the bioturbated micrite was de-
Web Fig. 5.11 is a space shuttle image showing the posited upon a gently sloping carbonate bank in an
western front of the Sierra Madre Oriental (right). area that was isolated from the winnowing action of
The boundary between the Coahuiltecana terrane wave energy. The 'upper quartzite unit' consists of
(emended herein) and displaced terranes to the west thick-bedded, well-sorted quartz arenite with sym-
occurs immediately west of the mountain front along metrical ripple marks, trough cross-beds (Martin,
the Walper Megashear. 1996, p. 105), and occasional large gastropods. Mar-
Examination of the type Zuloaga Limestone in the tin indicated that the uppermost beds of this unit
Sierra Sombreretillo (Zacatecas) has established that are peloid lime grainstone similar to those occurring
Burckhardt's 'nerineid limestone' at San Pedro del within the Zuloaga Limestone tongue. Moreover, ac-
Gallo is its lithostratigraphic equivalent. At San Pe- cording to Martin the well-sorted nature of the quartz
dro del Gallo the Zuloaga Limestone (Web Figs. 5.2 sand and micrite peloids suggests that the 'upper
and 5.3) occurs as a tongue in Burckhardt's (1910) quartzite unit' may have been deposited at even shal-
quartzite unit which Imlay incorrectly correlated lower inner neritic depths near a shore face where
with the La Gloria Formation (type locality -- Sierra sediments would be subjected to higher wave energy.
la Gloria, ~50 km east-southeast of Parras (Fig. 2: The 'upper quartzite unit' (Figs. 9, 11 and 12) is
Loc. 17), Imlay, 1937). overlain by a unit consisting of pink siltstone, mud-
The base of the succession at San Pedro del Gallo stone, and silty limestone with middle and upper
is not exposed. In that neither the Zuloaga Limestone Oxfordian ammonites, brachiopod shell fragments,
nor the lower and upper quartzite units possess age Buchia concentrica, common nodosarids (benthonic
diagnostic fossils, their age can only be established foraminifers), and miliolids (benthonic foraminifers)
(senior author's observation and those of Martin,
1 Available at: http://www.elsevier.nl/locate/caribas/ 1996). Martin noted that the calcareous siltstone is

134 E.A. P E S S A G N O et al.
Fig. 11. Lithostratigraphy, chronostratigraphy, paleobathymetry, and paleobiogeography. (1) "Lower quartzite unit" of Burckhardt (1910).
Unfossiliferous massively bedded, white to pink sandstone. (2) "Upper quartzite unit" of Burckhardt (1910). Unfossiliferous massively
bedded sandstone. Overlies Burckhardt's "nerineid limestone" -- Zuloaga Limestone of Imlay (1938). (3) Unnamed pink silty limestone,
mudstone, and siltstone. Contains Buchia, common ammonites, and Radiolaria (upper part only). Chronostratigraphically significant
megafossils include the ammonites Dichotomosphinctes and Discosphinctes and the Buchia concentrica (middle to upper Oxfordian).
(4) Lower Kimmeridgian Ataxicoceras Zone and probably part of Idoceras Zone (ammonites) missing. Upper part of upper Oxfordian
probably missing. Regional unconformity in much of western North America. Corresponds approximately to onset of deposition of flysch
during middle Oxfordian times in Klamath Mountains of northwestern California and southwestern Oregon (Galice Formation) and in
Sierra Nevada (Mariposa Formation, Monte del Oro Formation). See Pessagno et al. (1993a). Possibly reflecting pre-Nevadian orogenic
pulse in backarc domain. (5) "Lower shale member" of La Caja Formation = lower part of "Capas de San Pedro" of Burckhardt (1910).
Dark gray calcareous to siliceous mudstone with micrite nodules containing abundant Radiolaria, common ammonites and Buchia. Basal
strata assignable to Idoceras Zone (upper half of lower Kimmeridgian) and radiolarian Subzone 2 alpha-1 (Meng, 1997; Meng and
Pessagno, in prep.). (6) Regional unconformity in western North America recognizable in Nevadian back arc terranes (e.g., all San
Pedro del Gallo remnants in Mexico and in Cuba) and in Nevadian forearc terranes (e.g., volcanopelagic (VP) strata overlying Stanley
Mountain remnant of Coast Range Ophiolite, San Luis Obispo Co., California and Point Sal remnant of Coast Range ophiolite, Santa
Barbara Co., California). See Hull (1991, 1995), Hull et al. (1993); Hopson et al., (1996). (7) Includes upper part of "Capas de San
Pedro" of Burckhardt (1910) and "chert", "upper shale", and "Cerro Panteon quarry unit 2" members of La Caja Formation herein (see
Figs. 8 and 10 for more detailed litho description, biostratigraphic data and chronostratigraphic data). Note that the La Caja Formation
at San Pedro del Gallo was miscorrelated lithostratigraphically by Imlay (1939) with his La Casita Formation. All members of La Caja
Formation at San Pedro del Gallo with abundant Radiolaria. Sudden influx of silty wacke at Cerro Panteon and at La Pefia (10 km north
of San Pedro del Gallo) reflects onset of Nevadian orogeny. Contact of Great Valley Supergroup (flysch) and underlying VP sequence at
Stanley Mountain above Stanley Mountain remnant of Coast Range Ophiolite occurs in lower part of Subzone 4 alpha (uppermost upper
Tithonian). At San Pedro del Gallo equivalent strata contain the ammonite Durangites and Buchia piochii. (8) Unnamed thin-bedded, tan
to pink mudstone and micrite with common ammonites and calcified Radiolaria. These strata overly the massive to medium bedded tan
micrites of the Chapulhuacan Formation (type area in Taman-Tamazunchale area to southeast. Imlay (1937) miscorrelated these strata
with shallow neritic Taraises Formation (type area = Sierra de Parras). (9) Paleomagnetic data ("upper quartzite unit") from Ogg indicates
~40~ of Jurassic paleoequator (Pessagno, 1995). Presence of the Boreal megafossils Buchia concentrica and Amoeboceras sp. in
overlying Oxfordian strata associated with Tethyan ammonites such as Dichotomospinctes indicate Southern Boreal Realm. Overlying
La Caja Formation with Buchia concentrica, B. rugosa, B. mosquensis, and B. piochii associated with Southern Boreal radiolarian
assemblage characterized by high diversity and abundance of Parvicingula and Praeparvicingula. Upper part of La Caja Formation with
Buchia associated with Parvicingula/Praeparvicingula and abundant calpionellids (Tethyan: see Pessagno et al., 1996). Late Tithonian
portion of La Caja Formation formed "~ at boundary between Boreal Realm and Tethyan Realm. (10) Abundant calpionellids together
with lack of Buchia and presence of only Tethyan ammonites suggests Northern Tethyan Province (Meng, 1997). (11) Units G and F
of Fig. 13. (12) Regional unconformity noted in (4) above. (13) Unit E (pt. cf. Fig. 11). The discovery of the lower Kimmeridgian
ammonite Idoceras in a limestone nodule 1.5 m below contact between units E and F, indicate that the uppermost part of unit F
is lower Kimmeridgian (identification by Dr. Cant4-Chapa, Instituto Politecnico Nacional, Mexico); the middle Oxfordian ammonite,
Dichotomosphinctes was recovered 7 m below this horizon (identification by Dr. Cantf-Chapa, Instituto Politecnico Nacional, Mexico).
(14) Units E (pt)-B of Fig. 13. Note that silty wacke occurs in upper part of Unit B in lower Subzone 4 alpha. This horizon occurs below
final occurrence of Durangites and Substeueroceras (Pessagno et al., in prep.; identification by Dr. Canti-Chapa, Instituto Politecnico

slightly laminated and contains 10 to 35% angular to nodules in its lower part. Although the 'Capas de
subangular, well-sorted quartz grains. The upper part San Pedro' was informally n a m e d by Burckhardt,
of this u n n a m e d Oxfordian unit contains c o m m o n it is clearly genetically related to the La Caja For-
radiolarians. The sudden occurrence of the radiolari- mation. We have observed black radiolarian chert in
ans in this part of the succession reflects a rapid shift the La Caja Formation thus far at Cation San Matias
in paleobathymetry from inner neritic to outer neritic (Sierra Santa Rosa) near Mazapil, at its type locality
depths during the late Oxfordian (Figs. 10 and 11). in the Sierra de la Caja, in the Sierra de Catorce, and
Imlay (1939) correlated the informal unit which at other localities where the La Caja Formation has
Burckhardt called the 'Capas de San Pedro' with been reported (see Imlay, 1980).
the La Casita Formation (see Figs. 9, 11 and 12). The La Caja Formation ( = 'Capas de San Pedro'
Although these lithic units are approximately equiv- part of Burckhardt, 1910) is divided into this report
alent chronostratigraphically, it is clear from our into four informal m e m b e r s (in ascending order): (1)
examination of the La Casita in its type area (Sierra the 'lower shale m e m b e r ' , (2) the 'chert m e m b e r ' ,
de Parras, Fig. 2: Loc. 17) as well as in the Sierra (3) the 'upper shale m e m b e r ' , and (4) the 'Cerro
Jimulco (Fig. 2: Loc. 18) that the 'Capas de San Panteon quarry unit 2' member.
Pedro' are not correlative lithostratigraphically with (1) 'Lower shale member'. The 'lower shale
the La Casita Formation (see Martin, 1996). The m e m b e r ' consists of 52 m (min.) of dark-gray
La Casita Formation consists of gypsiferous gray to siliceous to calcareous mudstone with c o m m o n
pinkish-gray silty, calcareous to siliceous mudstone, dark-gray micrite nodules. Lenticular masses of
silty micritic limestone, and siltstone deposited at thin-bedded, dark-gray micrite are present locally.
inner neritic depths containing ammonites, bra- The bedded micrite and micrite nodules contain
chiopods, bivalves, and a sparse, poorly diversified abundant radiolarians, rare to c o m m o n ammonites,
foraminiferal assemblage (five species, largely Tex- and Buchia (see Figs. 11 and 12). This unit is
tulariina: senior author's observations). In contrast, assignable to late early K i m m e r i d g i a n to the early
Burckhardt's 'Capas de San Pedro' consists of upper late K i m m e r i d g i a n (Figs. 9, 11 and 12). It rests un-
abyssal dark-gray calcareous to siliceous mudstone conformably on the middle to early late Oxfordian
with c o m m o n black, thin-bedded radiolarian chert in strata of the u n n a m e d red, siltstone, limestone and
its upper part and c o m m o n radiolarian-rich micrite shale unit (Figs. 9, 11 and 12).
Nacional, Mexico). (15) Unnamed limonitic mudstone and limestone of Burckhardt (1930 = Burckhardt's unit B). (16) Change from
inner neritic to outer neritic occurs in upper part of unit F (Fig. 13). Martin (1996) noted the first occurrence of common Radiolaria in
upper unit F. This horizon may also correspond to regional unconformity noted in 12 above. (17) Called La Joya Formation by Imlay
(1980). Lower Jurassic strata below this unit contain ammonites and probably are equivalent to the Huayacocotla Group (see Imlay,
1980). (18) E1 Pastor Member of La Caja Formation (Verma and Westermann, 1973). Massively, bedded medium gray micrite with
thin-beds of black radiolarian chert and wacke. Wacke often with displaced shallow neritic megafossils. Overlying E1 Verde member
consisting of thin-bedded dark gray micrite and black radiolarian chert together with wacke. Graded-bedding and displaced shallow
water fossils noted in wacke (19). Incorrectly correlated with shallow neritic La Taires Formation by Verma and Westermann (ibid.).
(20) Overlies Huayacocotla Group in Huayacocotla remnant. (21) Canti-Chapa, (1969) recovered the Boreal ammonite Kepplerites in
the subsurface of the Huayacocotla remnant from the shallow neritic Palo Blanco Formation. This ammonite is common in terranes in
the Sierra Nevada, in the Izee Terrane of east-central Oregon, and in western terranes north to Alaska. It is indicative of the uppermost
Bathonian or lower Callovian (Imlay, 1980) In the surface the Cahuasas Formation is overlain by the Tepexic calcarenite (see Fig. 13 for
more detailed description). (22) All but uppermost part of Santiago Formation contains an inner neritic molluscan assemblage. Common
Radiolaria (including Praeparvicingula) first occur near top of unit (Pessagno et al., 1987a). (23) Taman Formation characterized by
abundant Radiolaria, common pectenacids, and rare ammonites. The rarity of ammonites suggests deposition in the upper abyssal depth
zone just below the depth of composition of aragonite. See Pessagno et al. (1987a). Co-occurring throughout the Taman Formation in
the area south of Taman (e.g., near Huauchinango Puebla) are discontinuous masses of inner neritic calcarenite (San Andres Member,
Canti-Chapa, 1969, 1971; Imlay, 1980) that either represent shallow neritic carbonate sedimentation on sea mounts or turbidites.
Definition of Taman Formation following that of Pessagno et al. (1987a). (24) The Pimienta Formation differs from the Taman Formation
by consisting of light to medium gray, thin-bedded micrite interbedded with black radiolarian chert and occasional layers of green vitric
tuff. (25) See annotation 21 above. Taman Formation with rich Northern Tethyan radiolarian assemblage including Parvicingula and
Praeparvicingula and abundant diversified pantanelliids associated with Tethyan ammonites and calpionellids (Tithonian). (See Pessagno
et al. 1987a). (26) Pimienta Formation and overlying Chapulhuacan Limestone with abundant calpionellids, tethyan ammonites, and
lacking Parvicingula/Praeparvicingula. Central Tethyan Province. (27) Data from Imlay (1980), Haczewski (1976), Lewis and Draper
(1990), and Myczyfiski and Pszcz61kowski (1994). Imlay indicates that marine bivalves of probable Middle Jurassic age occur in upper
part of this unit. (28) San Vincente Member (Myczyfiski, 1994) is anomalous and appears to be an analog of San Andres Member
of Taman Formation in Huayacocotla remnant. See Annotation 23 above. (29) Radiolarian-rich strata comprising all of the Artemisa
Formation and most of Guasasa Formation (except for San Vincente, see above) were deposited at upper abyssal depths either above or
slightly above the CCD of aragonite. (30) The presence of Parvicingula/Praeparvicingula in the radiolarian assemblage associated with
Tethyan ammonites and Buchia (Myczyfiski, 1994) indicates that the Cuban SPG remnants were at Southern Boreal paleolatitudes. (31)
Northern Tethyan paleolatitudes indicated by same association as in (30), but with common to abundant calpionellids (Myczyfiski and
Pszcz61kowski, 1994).

DiagnosticFaunal Faunal
Lith0 Unit Description Age Elements
Paleobathymetry
Realm/Province
Medium to massive bedded light gray to tan, very aphanitic
Chapulhuacan micrite with abundant calpionellids, common Radiolaria, and rare I Berriasian
Abundant calpionellids N. Tethyan
Limestone (1)
ammonites. Some horizons with phosphate nodules. Province
r
oo~ Durangites, Substeueroceras
.,..,
c'- ..,,.~_
c" =~_ Red, pink, and pinkish gray limestone, calcareous J belemnites, Buchia pioch# +
siltstone (wacke), and calceous mudstone with common ammonites, abundant calpionellids,
late Tithonian
0 ~.- common belemnites, Buchia, common Radiolaria, and abundant calpionellids. abundant Radiolaria
Thickness = 6 to 77 m. ct3
= (Subzone 40 0 .
Dark gray siliceous mudstone minor thin-bedded, dark gray micrite Ourangites
= "uppershale and dark gray micrite nodules. Rare ammonites. + Buchiarugosa, B.
late Tithonian
o member" Buchia rugosa & B. mosquensis. Common calpionellids. Thickness = 36 m. mosquenfsis Q
.,,_, Radiolaria
+ Subzone 4or
C
E Buchia rugosa, B.
k..
o mosquensis
LL. Dark gray siliceous mudstone interbedded with black, thin-bedded O
late early Tithonian 4- L_
radiolarian chert and minor dark gray micrite. Common ammonites Kossmatia, Durangites
to
& Buchia + abundant Radiolaria. Common calpionellids +
late Tithonian
Thickness = - 316 m. c-.} cO
Subzone 413 Radiolaria.
J See Meng (1997).
O
r'n
Dark gray siliceous to calcareous mudstone with common dark gray micrite Buchia concentrica, B.
~r = , _ nodules throughout. Lenticular masses of thin-bedded, dark gray micrite early Kimmeridgian rogusa, & B. mosquensis e--
present locally. Micrite nodules and bedded limestone with abundant to in same bed with Glochicera,, L._
Radiolaria, rare to common ammonites and Buchia. Rests unconformably early late gp. fialar sensu Burckhardt e---
o
=
E on Oxfordian strata below. Kimmeridigan Idoceras spp.
Thickness = 52 m (minimum). Subzone 2otl Radiolaria
O
unnamed red Discospinctes

Outer Neritic
Interbedded red sility limestone, silty mudstone, and siltstone containing middle Oxfordian
siltstone, limestone, Buchia and common ammonites. Radiolaria first occurring in Buchia concentrica
to late Oxfordian
and shale upper part. Thickness = 21.2 m. Dichotomosphinctes
Massivelybedded white to red sandstone. Cross-bedsand symmetricalripple
"upper quartzite unit" marks. See Martin (1996). Paleoma.clneticdata = 40~ =3.3 m.
No fossils Inner
Zuloaga Massively bedded micritic limestone with nodules of black chert.
middle Oxfordian Nerinea, bivalves, corals, Neritic
or and sponge spicules
Limestone older See Burckhardt (1910, 1 9 3 0
"lower quartzite unit" Massively bedded red and white fine grained sandstone. Base not exposed. No fossils
Fig. 12. Stratigraphic summary for SPG remnant.
(2) 'Chert member'. This unit includes ~ 3 1 6 m emnites, calpionellids, and radiolarians (reddish color
of early to late Tithonian dark-gray siliceous mud- probably result of hydrothermal alteration by Tertiary
stone interbedded with thin-bedded, black radiolar- intrusives) (Web Fig. 5.5). At La Pefia, 10 km to
ian chert, and minor dark-gray micrite which rest the north of San Pedro del Gallo, the senior author
unconformably on the underlying 'shale member' observed 92 m of interbedded black siliceous shale,
(Web Fig. 5.4). Occasional thin-layers of quartz- thin-bedded siltstone (wacke), and thin-bedded dark
rich silty wacke often display graded bedding and gray micrite containing belemnites, abundant radio-
may possibly represent turbidites. This unit contains larians, Buchia, and common ammonites. Abundant
common ammonites, abundant radiolarians, abun- calpionellids were reported from this unit by Adatte et
dant siliceous sponge spicules, common calpionel- al. (1995). Contreras-Montero et al. (1988)recorded
lids, and common Buchia (Fig. 1). abundant ammonites, Buchia piochii as well as abun-
(3) 'Upper shale member' (= top of Burckhardt's dant radiolarians and belemnites from this locality.
'Capas de San Pedro'). The 'upper shale member' Imlay (1939) correlated strata assignable to
consists of 36 m of dark-gray siliceous mudstone 'Cerro Panteon quarry unit 2' and the Chapulhuacdn
and minor amounts of thin-bedded dark-gray micrite. Limestone (Figs. 9, 11 and 12) with the inner neritic
The mudstone contains common micrite limestone Taraises Formation (type area -- Sierra de Parras,
nodules. Abundant radiolarians and rare ammonites Fig. 2: Loc. 17). Where the latter unit has been
occur in the siliceous mudstone and in the micrite. observed during the course of this study, it con-
Rare ammonites, Buchia, and abundant calpionellids sists of rhythmically bedded chalky mudstone and
occur in the micrite beds and nodules. The late interbedded medium-gray, medium-bedded micritic
Tithonian strata of the 'upper shale member' rest limestone. Whereas the Taraises Formation contains
conformably above the 'chert member' and below a poorly diversified foraminiferal assemblage, bra-
'Cerro Panteon quarry unit 2' (Figs. 9, 11 and 12). chiopods, echinoids, and ammonites, the San Pedro
(4) 'Cerro Panteon quarry unit 2' member. This units contain common ammonites and foraminifers
member of the La Caja Formation at San Pedro del as well as abundant radiolarians and calpionellids
Gallo consists of 6 to 77 m of red, pink, and pink- (see Fig. 10).
ish-gray micritic limestone, calcareous siltstone, and The Chapulhuacdn Limestone (type area = Cha-
calcareous mudstone with common ammonites, bel- pulhuacdn, Hildago near Taman-Tamazunchale, San

Luis Potosi; Fig. 2: Loc. 3) consists of ~20 m of The Zuloaga consists of massive to medium-bed-
medium- to massively-bedded light-gray to tan very ded, medium-gray micritic limestone strata with
aphanitic micrite with abundant calpionellids, com- nodules of black chert (Web Fig. 5.4). Microfacies
mon radiolarians, and rare ammonites assignable to analysis of the Zuloaga at this locality indicates
the Berriasian. Some horizons contain large phos- that the micrite contains encrusting coralline al-
phate nodules (12 cm). gae, nerineid gastropods, bivalves, foraminifera, and
The La Caja Formation as well as the Cha- siliceous sponge spicules (Martin, 1996, p. 67). The
pulhuac~in Limestone contain abundant radiolari- faunal and floral data suggests that the Zuloaga
ans and siliceous sponge spicules, rare benthonic Limestone at Canyon San Matias was deposited at
foraminifers, and common ammonites (see Figs. 9, inner neritic depths (<50 m). on a carbonate bank
11 and 12). Deposition took place at upper abyssal free of wave energy. The age of the Zuloaga, like
depths somewhat above the ACD (compensation the oldest beds at San Pedro del Gallo, can only
level of aragonite) during early Kimmeridgian to be established as middle Oxfordian or older via the
Berriasian times and continued at these depths superposition of overlying strata (Units G and F)
through the Late Cretaceous (see Burckhardt, 1930). containing middle Oxfordian ammonites (Fig. 13).
The radiolarian cherts usually contain nearly 50% Unnamed units G and F consists of pink, silty
radiolarian tests. As a result, it is likely that they mudstone, micritic limestone, and siltstone rich in
formed as radiolarian ooze. bivalves and with common ammonites. Common
The Oxfordian to upper Tithonian part of the suc- radiolarians are present in the upper part of Unit E
cession is characterized by containing a mixture of All of the Zuloaga Limestone, Unit G, and all but
Tethyan and Boreal ammonites (e.g., Amoeboceras, the upper part of Unit F were deposited at inner
Idoceras, Durangites), common Buchia (e.g., Buchia neritic depths. The sudden appearance of common
concentrica, B. mosquensis, B. rugosa) as well as radiolarians in the upper part of Unit F reflects a
an abundance of Parvicingula and Praeparvicingula rapid change in paleobathymetry from inner neritic
and rare pantanelliids among the radiolarians. The depths to outer neritic depths (~200 m) in the late
megafossil and radiolarian assemblage coupled with Oxfordian (Web Figs. 5.6 and 5.7).
preliminary paleomagnetic data indicate that this ter- Units E, D, C, and B (lower Kimmeridgian to
rane remnant originated at Southern Boreal paleolat- Berriasian) are included in the La Caja Formation of
itudes (~40~ Ogg, in Pessagno et al., 1995) during Imlay (1938, 1939); see Figs. 9, 11 and 13. All La
the Oxfordian (see Figs. 4 and 5). The appearance Caja units at Canyon San Matias are characterized
of abundant calpionellids coupled with the presence by the presence of common to abundant beds of
of Buchia and the presence of common Parvicingula thin-bedded black, radiolarian chert identical to that
and Praeparvicingula in the 'Cerro Panteon quarry in the 'chert member' of Burckhardt's (1910) 'Capas
unit 2' member of the La Caja Formation demon- de San Pedro' (Web Figs. 5.8 and 5.9). The chert is
strate that the San Pedro del Gallo remnant had interbedded with thin- to medium-bedded, dark-gray
been transported to close to the boundary (~30~ micritic limestone and dark-gray siliceous to cal-
between the Northern Tethyan Province and the careous mudstone commonly containing dark-gray
Southern Boreal Province by the latest Tithonian micritic limestone nodules. Unit D, as noted by Bur-
(Late Jurassic) (Figs. 4 and 5). The lack of Boreal ckhardt (1930), is unique in that it is characterized
elements such as Buchia in overlying Early Creta- by the presence of beds of phosphate and phosphatic
ceous strata may suggest transport of the San Pedro limestone. All La Caja strata are characterized by
del Gallo remnant to the Northern Tethyan Province containing a microfauna with abundant radiolarians,
(>22 ~ to <30~ by the Berriasian. Based on Ogg's abundant siliceous sponge spicules, and rare ben-
paleomagnetic data and the faunal data cited above, thonic foraminifera and a megafossil assemblage
Meng (1997) estimated the rate of movement of with common to abundant ammonites. Deposition
the San Pedro del Gallo remnant along the Walper of La Caja strata at this locality during the Late
Megashear to be 4.9 cm/yr. Jurassic and Early Cretaceous (Berriasian) occurred
at upper abyssal depths, or perhaps lower bathyal
The Mazapil remnant depths, above the ACD (compensation level of arag-
onite) and continued at these depths until the end
The Mazapil remnant of the SPG terrane was of the Cretaceous (see Burckhardt, 1930). Radiolar-
examined at Canyon San Matias in the Sierra Santa ian chert formed consists of about 50% by volume
Rosa (Fig. 2: Loc. 5). As in the case of the San of radiolarian tests and test fragments; hence, it
Pedro del Gallo remnant, the base of the succes- is likely that it formed as a radiolarian ooze. The
sion at Canyon San Matias is not exposed (Figs. 9, phosphate horizon occurring in Unit D is puzzling.
10 and 14). The oldest unit exposed at this local- Frequently, phosphate-rich sediments occur today
ity is the Zuloaga Limestone (Unit H in Fig. 13). along coast lines with narrow continental shelves

Diagnostic Faunal Faunal

Litho Unit Description Age Elements Paleobathymetry
,
Realm/Province
Thurmannites spp.,
Unnamed limestone Medium bedded buff calcareous mudstone and micritic limestone. Micrite Valanginian Asteria aft. psilostoma
Unit (1) with common limonite nodules. Thickness (fide Burckhardt, 1930+ 50-70 m. .- fide Burckhardt (1930). ,
Chapulhuacan Massively bedded, very fine grained micritic limestone weathering to cream Calpionellids
Limestone or buff color. Calpionellids, Radiolaria. Sparse ammonites. Berriasian ? calcified Radiolaria o
c--
"UNIT A" Thickness (fide Burckhardt 1930) = 15 m. ,
--~ ,
Upper 5.64 m consisting of medium-bedded light gray micrite and
thin-beddded black chert, Remainder of unit consisting of thin-bedded, late Tithonian Substeueosceras spp. ~ O
' i
"UNIT B" buff-weathering calcareous siltstone and black chert (2). Abundant Radiolaria upper part may be Paradontoceras aft. i :>~
and ammonites. Berriasian c-
callistoides, Durangites sp. <s c~
Thickness = 23.5 m
c- ' i c--
Kossmatia spp.
O
-~
E
"UNIT C" Medium-bedded to thick-bedded (0.9--1.2 m) micrite, thin to medium-bedded
black chert, and minor siltstone Abundant Radiolaria and ammonites.
late Tithonian Boundary between Subzone
413 and 4~ in upper part of 5
c---
Thickness = 21.1 m. Unit B. i
o
ii
!
Interbedded phosphatic limestone, black chert and red calcareous mudstone. Hybonoticeras spp . . . . e-
...._, "UNIT D" Abundant Radiolaria and ammonites. lateTithonian Kossmatia spp.
L..
Thickness = 13.3 m +Subzone 4~ Radiolaria. i ~c'~ o
Z
o:I
early late Idoceras spp, Glochiceras
Thin-bedded black chert, dark gray, silicerous mudstone often with Kimmeridgian grp. fialar, Buchia
interbedded limestone nodules (up to - 4 m in maximum dimension). Red + concentrica, Hyboniticeras,
"UNIT E" calcareous mudstone in upper part. Some mudstone beds up to 0.6 m
hiatus and Zone 2o~1, Zone 3,
in lower part. Abundant Radiolaria in all lithofacies. Abundant ammonites. o
early to late and Zone 4, Subzone4J3 c-
Thickness = 27.9 m.
Tithonian Radiolaria. !
, i o
Outer Neritic
Red silty calcareous mudstone with 1.5 m dark gray micrite nodules in CL.
late early Idoceras spp. at top of
"UNIT F" upper part. Common ammonites, bivalves. Common Radiolaria at top.
Kimmeridgian at top 'UNIT F"
Thickness = 7.8 m c~
cD
i
Red medium-bedded silty limestone and mudstone.
"UNIT G" Thickness = 6.6 m
middle Oxfordian Dichotomosphinctes Inner O
__, Neritic e-
L..
cD
"UNIT H" middle Oxfordian Nerinea,bivalves, corals, t--
Massively bedded micritic limestone with nodules of black chert.
Zuloaga or and sponge spicules
Base not exposed. --5
Limestone older See Burckhardt (1910, 1930)
o
Fig. 13. Stratigraphic summary for Mazapil remnant.
and steep continental slopes at sites of upwelling (<30~ to >22~ during the early Kimmeridgian
of nutrient-rich waters. Whether this scenario could to late Tithonian interval. It should be noted that
exist in the distal backarc setting characterizing Ogg's preliminary paleomagnetic data for the late
all San Pedro del Gallo remnants is questionable. Tithonian indicate 25~
Phosphatic limestones and shales were recorded by
Burckhardt (1930) in the Sierra Santa Rosa, Sierra Sierra de Catorce remnant (Fig. 2" Loc. 4; Figs.
de la Caja, and Sierra de Zuloaga (Fig. 2.: Locs. 5, 9, 11, 14)
6, 8). They are not known from San Pedro del Gallo,
Sierra Catorce, the Huayacocotla Anticlinorium, or Erben (1956, p. 46), Carrillo-Bravo (1961, p. 42),
from western Cuba (Fig. 2: Locs. 3, 4, 12, 24). An and Imlay (1980) noted the presence of red phylitic
alternative to the upwelling model may be a large shale with Sinemurian ammonites in the Sierra de
kill of fish and other organisms by a red tide, produc- Catorce. These strata are the chronostratigraphic
ing an abundance of phosphatized bones and other equivalent (part) of the Huayacocotla Formation of
material at lower bathyal or upper abyssal depths. east-central Mexico.
The upper Oxfordian to lower upper Tithonian The Upper Jurassic succession at the Sierra de
part of the La Caja Unit E (Fig. 7) contains Catorce is much like that of other remnants in terms
Buchia, Tethyan ammonites, abundant Parvicin- of its lithostratigraphic and paleobathymetric signa-
gula/Praeparvicingula, and poorly diversified pan- tures. The Sierra de Catorce succession begins with
tanelliids indicative of Southern Boreal paleolati- the massively bedded, micritic, inner neritic Zuloaga
tudes. The remainder of Unit E and all of Units Limestone which is identical to that at San Pedro del
D, C, B, and A are assignable to the Northern Gallo and at Mazapil (Web Fig. 5.10). Verma and
Tethyan Province based on the presence of abun- Westermann (1973) divided the La Caja Formation
dant, diversified pantanelliids, abundant to common in the Sierra de Catorce into two members: a lower
Parvicingula/Praeparvicingula, and the presence of E1 Pastor Member and upper E1 Verde Member. The
calpionellids (Units A and B, Fig. 5). These data E1 Pastor Member consists of massively bedded dark-
indicate that the Mazapil remnant of the SPG terrane to medium-gray micritic limestone with interbedded
has been transported from Southern Boreal paleolat- black chert, siltstone, sandstone, and shale. The E1
itudes (>30~ to Northern Tethyan paleolatitudes Verde Member includes thin-bedded micritic lime-


Litho Unit Description Age Paleobathymetry Realm/Province
Elements
Light gray, massively bedded micrite with yellow, oval chert concretions.
unnamed limestone Correlated incorrectly by Verma and Westermann (1973) and Imlay (1980) Valanginian
with the La Taraises Formation which has its type area in Sierra de La Parra. Olcostephanus sp p. O3
unit
Probably the Chapulhuacan Limestone. Thickness is unknown. Berriasian? c-
Thin-bedded medium gray micrite, black chert, light gray siltstone and Kossmatiaspp, c--
oo
sandstone. Chert and micrite with abundant Radiolaria. Siltstone and Durangitesspp. oo
c~
El Verde sandstone are carbonate wacke (turbidite) frequently showing graded Corongocerasspp, ..K:=
bedding withgrading of molluscan fragments. Chert and micrite with late Tithonian
Member Substeuerocerasspp., cD
abundant Radiolaria. Abundant ammonites occur in micrite, Berriasellaspp.
Thickness = - 25 m, c-
E .+_,
o
u_ o
late early Mazapilites mexicanus, z
Massively bedded medium to dark gray micrite interbedded with thin-bedded Kimmeridgian Pseudolissoceras zitteli
-~,
black radiolarian chert and siltstone or sandstone. Siltstone and sandstone to
representing carbonate wacke occurs in layers up to 0.4 m thick and shows early Tithonian;
late Kimmeridgian
, El Pastor grading of molluscan fragments. Massive micrite beds averaging about Abundant Radiolaria
is missing; early
Member 0.8. Closely resemble micrites of "lower massively bedded limestone
Tithonian strata
member"of the Taman Formation in Huayacocotla remnant. Micrite and assumedly overlie
black chert with abundant Radiolaria Micrite with abundant ammonites. early Kimmeridgian Idocerasspp, in lower part
Buchia noted by Burckhardt (1930). Thickness = - 28 m. strata together with Buchia
unconformably concentrica
Massively bedded micritic limestone with nodules of black chert.. middle Oxfordian
Zuloaga Nefinea, bivalves, corals,
According to Imlay (1980, p. 45) Zuloaga rests unconformably on the or
Limestone continental red beds of the Cahuasas Formation.
and sponge spicules Inner Neritic
older See Burckhardt (1910, 1930)
Thickness = 200 m fideVerman and Westermann (1973).
Continental red beds. Red shale, sandstone, and siltstone.

Probably
Congomerate at base. Overlies Lower Jurassic (Sinemurian) strata of
Cahuasas Bathonian to
Huayacocotla Group. The latter unit contains Sinemurian ammonites. This unit
Bajocian based on Unfossiliferous NON MARINE
Formation was assigned to the La Joya Formation by Imlay. However, because it overlies
age in Huayacocotla
the Huayacocotla Group, it is assigned here to the Cahuasas. See Figure 13.
remnant.
Thickness = 120 to 160 m.
Fig. 14. Stratigraphic summary for Sierra Catorce remnant.
stone, black thin-bedded radiolarian chert, thin-bed- del Gallo remnant, the Mazapil remnant, and the
ded siltstone and sandstone, and shale. The siltstone Huayacocotla remnant.
and sandstone in both members often display graded
bedding with graded molluscan fragments and are Huayacocotla remnant (Figs. 9, 11, 15)
interpreted as turbidites (Web Figs. 5.10 and 5.11).
These strata were deposited at upper abyssal depths The Mesozoic succession begins in this area with
rather than inner neritic depths as incorrectly sug- the deposition of the Lower Jurassic (Sinemurian to
gested by Verma and Westermann (1973) and Sal- lower Pliensbachian) Huayacocotla Formation which
vador et al. (1992). Most stratigraphers in the past consists of black shale, mudstone, and graywacke. A
have totally ignored the far more abundant micro- rich ammonite assemblage occurs in the lower and
fauna. Because they failed to examine the rocks in middle parts of the unit (Burckhardt, 1930; Erben,
thin-section, they reached erroneous conclusions con- 1956; Imlay, 1980). Bivalves and land plants have
cerning water depth. been recorded from the upper part.
Burckhardt (1930) recorded the presence of The Huayacocotla Formation overlies the Missis-
several species of Buchia from the Kimmerid- sippian to Lower Permian strata of the Guacamaya
gian portion of La Caja Formation at Sierra de Formation with marked hiatus associated with an an-
Catorce. These Boreal bivalves are associated with gular unconformity. According to Nestell (1979) the
Tethyan ammonites and Parvicingula/Praeparvi- Guacamaya Formation contains fusulinids with South
cingula (thin-section analysis). Hence, we conclude American affinities. This likewise seems tobe true of
that the Sierra de Catorce remnant was situated fusulinids occurring in the Guacamaya Formation at
in the Southern Boreal Province during the Ox- Peregrina Canyon near Ciudad Victoria (Tamps.). The
fordian to Kimmeridgian interval. Tithonian La thickness of the unit varies from 560 to 1200 m.
Caja strata contain Tethyan ammonites, Parvicin- As far as can be determined, references to the
gula/Praeparvicingula and lack Buchia. Hence, dur- presence of the Upper Triassic Huizachal Formation
ing the Tithonian the Sierra de Catorce remnant was (continental red beds) in this area are erroneous. The
probably situated in the Northern Tethyan Province. Huizachal has largely been confused with the Middle
Our investigations of the strata in this area are not, Jurassic Cahuasas Formation which also consists of
however, as complete as they are in the San Pedro continental red beds (cf. Imlay, 1980).


Litho Unit Description Age Elements Paleobathymetry Realrn~rovince
Medium to massively bedded very fine-grained cream to light gray micrite Berriasian Subthurmannia sp. r"-
Chapulhuacan c~
with abundant Radiolaria, calpionellids, and rare ammonites. Micrite with black to i Neolissoceras sp. Spiticeras sp.
Limestone chert nodules and lenses. Thickness = - 30 m. Valanginian. Thurmanniceras sp, t"
:Paradontocerasaff, callistoides
Thin-bedded cream colored to light gray micrite interbedded with dark Durangites, I
gray shale, common black radiolarian chert, and light green vitric tuff. Substeueroceras j
Pimienta Formation late Tithonian
Micrite with abundant Radiolariaand calpionellids together with rare + c,o c
ammonites and common sponge spicules. Thickness = - 200 m. SubzoneZletRadiolaria , >,
0
at some localitiesto south '-'
!
d Durangites, Kossmatia,
o Thin-bedded dark gray to medium gray micrite with thick interbedds of dark Salinites grossicostatum r
t-"
oi "upper thin-bedded to medium gray shale. Shale layers with abundant micrite nodules. Micrite +
late Tithonian
~ limestonemember" nodules and beds with abundant Subzone 4~ Radiolariasiliceous very abundant Subzone O
II
sponge spicules and commonammonites. 4~ Radiolaria. a..
b-- t'-
c---,
Massively bedded dark gray to medium gray micrite interbedded with thin- Ataxicoceras, Idoceras, c-, c-"
bedded to medium bedded shale. Upper part of unit with numerous dark gray
~ "lowermassively micrite nodules. Massive micrites and micrite nodules with abundant Radiolaria early Kimmeridgian
"Glochiceras fialar"
Mazapilites,
t-.,
?
"c-- beddedlimestone rare to common ammonites and common pectenacids (Aulacomyella). Hyaline to Virgatosphinctes + Radiolaria t---
=~ member" early Tithonian assignable to Subzone 2ctl,
calpionellids occur in basal Zone 4, Subzone 4[3 strata at same horizon as el3
t"-
lower Tithonian ammonite Mazapilites. Zone 3, & Subzone 4l].
i i
O
Outer Neritic Z
Reineckeia,
Santiago Silty black shale, mudstone, micrte. Containing ammonites. Radiolaria early Callovian Dichotomosphinctes,
occurring in upper-most part. to Discosphinctes, I
Formation
Thickness = N169 m. late Oxfordian Ochetoceras. !
I
Tepexic Calcarenite containing ammonites and bivalves. Reineckeia

Thicknes -- - 39 m. late early Callovian
Limestone Neuqueniceras !
t
Continental red beds. Dominantly red shale, siltstone, sandstone, and

Cahuasas conglomerate. Commonly cross-bedded. Overlies Lower Jurassic Bathonian to
Bajocian. Fossil plants. NON MARINE
Formation (Sinemurian) strata of HuayacocotlaGroup.The latter unit contains
Sinemurian ammonites. Thickness = 40-1200 m. (1)
Fig. 15. Stratigraphic summary for H u a y a c o c c o t l a remnant.
The Cahuasas Formation consists of 40 to 1200 The Tepexic Limestone is overlain conformably
m of red arkosic sandstone, conglomerate, and shale by silty black shale, siltstone, and silty micritic lime-
that rest with angular unconformity on the Huayaco- stone constituting the Santiago Formation (middle
cotla Formation (Imlay, 1980, p. 49). Imlay indicates Callovian to upper Oxfordian). The lower and mid-
that the Cahuasas must be older than Callovian dle parts of the Santiago contain bivalves (e.g., small
because where it crops out on the surface it lies Ostrea, senior author's observations) and ammonites;
disconformably below marine beds of early to mid- microfacies analysis by Longoria (1984) indicates
dle Callovian age. In the subsurface, however, it that most of this unit was deposited at inner ner-
underlies latest Bathonian to early Callovian marine itic depths. The uppermost (upper Oxfordian) part
shale of the Palo Blanco Formation (see Palo Blanco of the Santiago Formation (e.g., at Taman, S.L.E)
below). Imlay also indicates that the Cahuasas must contains common radiolarians as well as ammonites
be younger than Toarcian (Early Jurassic) in that it (Pessagno et al., 1987). These Santiago strata reflect
passes downward into plant-bearing beds which are the same sudden change in water depths from inner
early Middle Jurassic. neritic to outermost neritic during the late Oxfor-
The Cahuasas Formation is overlain discon- dian that was noted in the San Pedro del Gallo and
formably in surface outcrops by the inner neritic Mazapil remnants. The Santiago Formation is over-
early Callovian Tepexic Limestone. The Tepexic is lain conformably by the Taman Formation (sensu
a calcarenite containing common to abundant Li- Pessagno et al., 1984, 1987).
ogryphaea nebrascaensis and ammonites such as The Taman Formation (thickness about 30-60 m)
Neuquenisceras neogaeum and Reineckeia (Can- consists of two informal units (Web Figs. 5.12 and
td-Chapa, 1969, p. 19; Imlay, 1980, p. 50). In 5.13): (1) a massively bedded to medium-bedded
the subsurface and at some surface localities a inner micritic limestone member (lower Kimmeridgian to
neritic black shale unit, the Palo Blanco Formation upper Tithonian), and (2) a thin-bedded micritic
(Cantd-Chapa, 1969, p. 5; Imlay, 1980, p. 49), under- limestone member (upper Tithonian) (Pessagno et
lies the Tepexic Limestone and rests disconformably al., 1984, 1987). Both members of the Taman
on the Cahuasas. The Palo B lanco Formation con- Formation contain profusely abundant radiolarians,
tains the late Bathonian to early Callovian ammonite rare foraminifera (chiefly Textulariina), common
Kepplerites (Cantd-Chapa, 1969, p. 5; Imlay, 1980). siliceous sponge spicules, ammonite aptychi, and

occasional ammonites. The abundance of radiolari- ing abundant calpionellids and nannoconids lacking
ans together with the sparse benthonic foraminiferal Parvicingula/Praeparvicingula. This association of
assemblage and the rarity of ammonites suggests that faunal elements is indicative of the Central Tethyan
Taman strata were deposited at upper abyssal depths Provinces (Figs. 4 and 5).
at or somewhat below the ACD (aragonite compen- These data indicate that the Huayacocotla rem-
sation level) (see microfacies analysis in Longoria, nant of the SPG terrane underwent tectonic transport
1984). from Southern Boreal paleolatitudes (> 30~ during
The Taman is overlain conformably by the lat- the late Bathonian (Middle Jurassic) to Northern
est Tithonian (Late Jurassic) to Berriasian (Early Tethyan paleolatitudes by the early Kimmeridgian
Cretaceous) Pimienta Formation (sensu Pessagno et (Late Jurassic) to Central Tethyan paleolatitudes
al., 1984, 1987) and overlain conformably by the (<22~ during the Berriasian (Early Cretaceous).
Chapulhuac~in Limestone Berriasian to Valanginian).
The Pimienta Formation includes 200-400 m of Remnants of the San Pedro del Gallo terrane in
light-gray thin-bedded micritic limestone with thick western Cuba (Figs. 9, 11, 16)
shale intervals, thin-bedded black radiolarian chert,
and light green vitric tuff; it contains abundant ra- Remnants of the SPG terrane in western Cuba
diolarians, calpionellids, siliceous sponge spicules, crop out in the Sierra del Rosario and the Sierra
and common ammonites (Fig. 10). Pimienta depo- de los Organos (Figs. 9, 11, 16 and 17). Our data
sition likewise took place at upper abyssal depths are derived from field observations by Longoria and
somewhat above the ACD (compensation level of by examination of Jurassic rocks from western Cuba
aragonite). by Pessagno in the collections of the U.S. Geolog-
The Chapulhuac~in Limestones consists of about ical Survey. Moreover, they are derived from data
30% of medium to massively bedded, very fine- presented by Br6nnimann (1954), Arkell (1956),
grained, cream to light-gray micrite with abundant Meyerhoff (1964), Khudoley and Meyerhoff (1971),
radiolarians, calpionellids, nannoconids, and plank- Kutek et al. (1976), Imlay (1980), Myczyfiski (1989,
tonic foraminifera, and rare ammonites at most local- 1994), Lewis and Draper (1990), Myczyfiski and
ities (senior author's observations and those of Lon- Pszcz6~kowski (1976, 1994).
goria, 1984, p. 69); Chapulhuac~in strata were also
deposited at upper abyssal depths somewhat above The Sierra del Rosario remnant
the ACD. Deposition continued at these depths dur-
ing the remainder of the Cretaceous. The succession in the Sierra del Rosario begins
The upper Bathonian (Middle Jurassic) to upper with the Middle Jurassic (Bajocian?) to Upper Juras-
Oxfordian (Upper Jurassic) part of the succession sic (middle Oxfordian) San Cayetano Formation.
contains Boreal megafossils such as the ammonite The San Cayetano Formation includes 1500 to 3000
Kepplerites in the Palo B lanco Formation (Can- m of reddish weathering carbonaceous shale, white
tf-Chapa, 1969, p. 5; Imlay, 1980, p. 50). Elsewhere to grayish quartzose siltstones and sandstones, mica-
in western North America this ammonite is known rich gray shales, and friable arkoses (see Lewis and
from Middle Jurassic strata in the Sierra Nevada, Draper, 1990 and Haczewski, 1976). Imlay (1980,
from the upper Bathonian part of the Snowshoe For- p. 39) indicates that its upper part contains marine
mation, Izee terrane (east-central Oregon), and from bivalves of probable Middle Jurassic age. The San
Boreal Middle Jurassic strata as far north as Alaska Cayetano is the lithic equivalent of the Cahuasas
(Imlay, 1980; Pessagno and Blome, 1986; Pessagno Formation in the Huayacocotla remnant of the SPG
et al., 1986, 1987). terrane in east-central Mexico. Most of the San
The lower Kimmeridgian to upper Tithonian (Up- Cayetano except for its upper 609 m appears to be
per Jurassic) part of the succession (Taman For- non-marine. The upper 609 m of the San Cayetano
mation sensu Pessagno et al., 1984, 1987) con- contains bivalves like Ostrea and Vaugonia which
tains a rich Northern Tethyan radiolarian assemblage are interpreted herein as being inner neritic.
characterized by the abundance and diversity of The overlying Francisco Formation includes 13
pantanelliids and by the presence of common to to 25 m of shale, limestone, and some sandstone
abundant Parvicingula and Praeparvicingula. Calpi- with middle to early late Oxfordian ammonites
onellids (Tethyan) occur in the upper Tithonian part such as Discosphinctes and Dichotomosphinctes.
of the Taman Formation. Moreover, the megafossil These strata are interpreted herein as being ner-
assemblage is Tethyan in aspect (Figs. 4 and 5) itic. Whether these late Oxfordian strata contain
(see Imlay, 1980; Cantf-Chapa, 1989). The Pimienta radiolarians as do coeval strata in the Mexican rem-
Formation as well as the overlying Chapulhuac~in nants of the San Pedro del Gallo terrane cannot
Limestone is characterized by a Tethyan ammonite be established at present. The Francisco Formation
assemblage and by a microfossil assemblage includ- is overlain conformably by the late Oxfordian to

D i a g n o s t i c Faunal Faunal
Litho Unit Description Age Elements
Paleobathymetry
Realm/Province
~o Leptoceras
Berriasian
Protancyloceras
Thin-bedded black micrite and black radiolarian chert. to
I:D
Olcostephanus r
co Valanginian t-'-
Buchia
o
1_
r
..-.-.. r-
c~
.c:
v (D
b'-
03
c (33 r'-
o ..Q 03
E
Peeddolissoceraszitteli
r'-
Virgatosphinctes .....,
E L..
Corongocerassp. o
o
u_ Salinites
Thin-bndded dark gray to black micritic limestone with thin interbeds of early Tithonian Durangites
shales, siltstones, and sandstones. Abdundant Radiolaria and ammonites. to Microacanthoceras
Thickness -- 40 m. Berriasian Paradontoceras
c~
N
Buchia
co Calpionellids (1)
Radiolaria
Dichotomosphinctes Neritic.
Francisco Shale, limestone and sandstone with ammonites. early to late
4- Proably mostly inner
Formation Thickness = 13 -- 25 m. 0xfordian
Discosphinctes neritic.
miildle All but upper 609 m

Reddish weathering carbonaceous shale, white to grayish quartzone siltstone
San Cayetano 0xfordian consisting of
and sandstone, mica-rich shale, and friable quartzite. Upper 609 m with continental red beds.
Formation to None
marine bivalves (e.g., Ostrea). See Lewis and Draper (1990).
(part) Bajocian Upper 609 m inner
Thickness =1500 -- 3000 m. neritic.
Fig. 16. Stratigraphic summary for Sierra de Rosario, Cuba.
D i a g n o s t i c Faunal Faunal
Litho Unit Description Age Elements
Paleobathymetry
Realrn~rovince
d3
Tumbitas Light gray, thin-bedded limestone with intercalations of dark gray chert. Valanginian Ammonites poorly preserved
Member Abudant calpionellids i
+ 0
Tumbadero Dark gray, laminated, medium-bedded with thin intercalations of shale, chert, Radiolaria
Berriasian 13-
..--..
Member and numerous dark gray chert nodules. Thickness = - 15 m. C
4-,
O3
03 F-
.+.-,
v
t-- I--
o
4...,
c13
E Mazapilites
Hard, compact dark gary to black micritic, thin to thick-bedded micrititc Hyboniticeras
o 0
LI_ limestone. Abundant Radiolaria throughout. Abundant calpionellids in upper Salinites grossicostatum Z
lower Tithonian and upper Tithonian. Abundant ammonites. This unit includes early to late
Durangites
the Vifiales Limestone of older literature. Tithonian
Kossmatia
Thickness = 300-400 m. Buchia
C3..
c.)
o
~._
m
Poorly dated. Inner Neritic
S Vicente Dark gray to black mostly massively bedded micrite with some lenses and None
Possibly (2)
Member concretions of chert. Poorly fossiliferous. Kimmeridigian.
o
1:13
Divided into 3 members (ascending order):1] Azucar Membv (48-76 m): Gray c--
i to black, thinibedded micritic limestone Ioally sandy & oolitic (Hatten, 1967,
p. 782); 2] Jagua Vieja Member (50-60 m): dark gray silty to sandy shale, early to late Dichotomosphinctes Mostly inner c-
Jagua 4- .4--,
mudstone, and limestone with many limestone nodules; 3] Pimienta Member Oxfordian neritic.
Formation Discosphinctes (1) 0
(40-60 m): Gray dense, platey limestone. Overlies San Cayetanof Formation CO
unconformably (See Figure 9).
miildle All but upper 609 m
Reddish weathering carbonaceous shale, white to grayish quartzone siltstone i consisting of
San Cayetano Oxfordian
and sandstone, mica-rich shale, and friable quartzite. Upper 609 m with i continental red beds.
Formation to None
marine bivalves (e.g., Ostrea). See Lewis and Draper (1990). Upper 609 m inner
(part) Thickness =1500 -- 3000 m. Bajocian
neritic.
Fig. 17. Stratigraphic summary for S. de los Organos, Cuba.

Valanginian Artemisa Formation. The Artemisa For- As in the case of the Sierra del Rosario remnant
mation as a whole consists of dark-gray, mostly the succession begins in the Sierra de los Organos
thin to medium-bedded, dense cherty limestone, and with the deposition of the Middle Jurassic (Bajo-
tuffaceous shale (Imlay, 1980, p. 39). Pszcz6~kowski cian?) to Upper Jurassic San Cayetano Formation
(1978) divided the formation into three members; (see description above). The San Cayetano Forma-
they are in ascending order: (1) the San Vicente tion in the Sierra de los Organos is overlain by
Member (upper Oxfordian-Kimmeridgian; (2) the the Jagua Formation (Hatten, 1967; Wierzbowski,
La Zarza Member (Tithonian); and (3) the Sum- 1976). The Jagua Formation is divided locally into
idero Member (Berriasian to Valanginian). The San three members: (1) a lower Azucar Member; (2) a
Vincente Member consists of massive inner ner- middle Jagua Vieja Member; and an upper Pimienta
itic limestone probably formed as a bank deposit Member. The Azucar Member (48 to 76 m)consists
(Myczyfiski, 1994). The massive limestone strata of gray to black, thin-bedded micritic limestone that
are apparently localized in their distribution and is in some places oolitic and sandy (Hatten, 1967, p.
may be analogous to those of the San Andres 782). The Jagua Vieja Member (50 to 60 m) consists
Limestone of the Huayacocotla remnant (SPG) in of dark-gray silty to sandy shale, marl, and lime-
east-central Mexico (see Canttl-Chapa, 1969). The stone and contains many limestone nodules in the
Zarza Member in the southern part of the Sierra shale beds. The upper Pimienta Member (40-60 m)
del Rosario includes about 40 m of thin-bedded consists of gray, dense platy thin-bedded limestone.
black to dark-gray micritic limestone with thin in- Imlay (1980), p. 39) indicates that ammonites from
terbeds of shale, siltstone, and sandstone (see My- the Azucar Member are of middle or late Oxfordian
czyfiski and Pszcz6~kowski, 1994). These strata in- age. The Jagua Formation appears to be lithic equiv-
clude black ammonite-beating limestones, aptychi, alent of the Santiago Formation in the Huayacocotla
and ammonite shell coquina in the upper part of the remnant (SPG).
member. The Zarza Member grades up into the over- The remainder of the Jurassic succession (Kim-
lying Sumidero Member. The Sumidero Member meridgian to upper Tithonian) above the Jacaguas
includes 'ammonite-free' black, thin-bedded micrite Formation in the Sierra de los Organos is included
with interbedded radiolarian chert. Examination of in the Guasasa Formation (Herrera, 1961; Kutek et
limestones from the Tithonian and Berriasian parts al., 1976; Wierzbowski, 1976; Imlay, 1980). The
of the Artemisa Formation in Mesozoic USGS col- Guasasa Formation in the more recent literature
lections from Cuba indicate that many of these rocks includes the Vifiales Limestone of the older liter-
contain abundant radiolarians. This observation was ature. The Jurassic part of the Guasasa includes
also confirmed by Myczyfiski and Pszcz6~kowski two members: (1) a lower San Vicente Member
(1994, p. 11, fig. 3). The presence of radiolarians (1000 m) consisting of dark-gray to black, mostly
as well as calpionellids in the La Zarza Member massively bedded micrite with some lenses and con-
indicates that deposition occurred at bathyal to up- cretions of chert, and (2) an upper E1 Americano
per abyssal depths above the ACD (compensation Member (300-400 m) consisting of hard, compact
level of aragonite). The absence of the ammonites in gray to black micritic, highly fossiliferous thin-
the Sumidero Member coupled with the presence of to thick-bedded limestone. The presence of com-
abundant radiolarian chert suggests that deposition mon to abundant radiolarians in the E1 Americano
during the early Berriasian (Early Cretaceous) was at Member in association with ammonites and calpi-
upper abyssal depths below the ACD (compensation onellids suggests that these strata were deposited
level of aragonite). at upper abyssal to bathyal depths above the ACD
For the most part the Upper Jurassic fau- (compensation level of aragonite). The Lower Creta-
nal assemblage of the Artemisa Formation in- ceous part of the Guasasa Formation includes three
cludes Tethyan ammonites and calpionellids. How- members: the Tumbadero Member (Berriasian), the
ever, species of the Boreal bivalve Buchia are Valanginian Tumbitas Member, and the Albian-
present throughout the Tithonian (Upper Juras- Aptian Infierno Member (cf. Myczyfiski, 1989). The
sic) to Valanginian (Lower Cretaceous) interval. Tumbadero Member consists of 15 m of dark and
Praeparvicingula/Parvicingula was observed by dark-gray, laminated medium-bedded limestone with
Pessagno in micritic limestones from the Meso- thin intercalations of shale, chert, and numerous
zoic collections of the USGS formerly housed at the dark-gray chert nodules. The Tumbitas Member con-
US National Museum. The composite faunal data tains about 50 m of light gray, thin-bedded limestone
suggest that the Sierra del Rosario remnant of the with intercalations of dark-gray chert. The upper-
SPG terrane remained near the boundary between most member, the Infierno Member, consists of
the Northern Tethyan Province and Southern Boreal about 50 m of dark-gray micritic limestone inter-
Province during the Late Jurassic and Sierra de los layered with lighter-colored micritic limestone and
Organos remnant (Figs. 9, 11 and 17) dark-gray chert. The occurrence of radiolarians as

well as calpionellids in the Tumbadero and Tumbitas tratigraphic, paleobathymetric, and paleolatitudinal
members coupled with the absence of ammonites signatures which are nearly identical to those of San
suggests that these strata were deposited below the Pedro del Gallo terrane remnants in central Mex-
ACD (compensation level of aragonite) at upper ico (Figs. 9, 11, 16 and 17). Even the presence of
abyssal depths. the inner neritic 'San Vicente Member' (Guasasa
The presence of Tethyan ammonites together Formation) has an analogue in the San Andres Lime-
with calpionellids, Parvicingula/Praeparvicingula, stone of the Huayacocotla remnant (Fig. 2: Loc. 3).
and Buchia (Myczyfiski, 1994) in the Upper Jurassic This record, at least during the Jurassic and Early
Tithonian part of the Guasasa Formation indicate a Cretaceous, is distinctly North American. Unconfor-
paleolatitudinal position close to the boundary be- mities and hiatuses are regional in distribution and
tween the Northern Tethyan Province and Southern can be traced as far north as the California Coast
Boreal Province (~30~ It should be noted that Ranges and Klamath Mountains (see Fig. 11). More-
Buchia is unknown from the North Atlantic Province over, these unconformities reflect tectonic events that
except for occurrences in Greenland. Elsewhere it is affect the Nevadian forearc, interarc, and backarc in
known from Jurassic and Lower Cretaceous strata western North America. Although a backarc origin
from Baja California Sur to Alaska. Hence, its oc- is advocated for the pre-latest Tithonian (radiolarian
currence in the Jurassic of western Cuba is entirely Subzone 4 alpha: Fig. 9), such an origin is related
anomalous. to the Nevadian island arc and not the Antillean
island arc. Once the Cuban San Pedro del Gallo
Origin of San Pedro del Gallo terrane remnants were carried northward by the advancing
(= Guaniguanico terrane) in western Cuba Caribbean Plate, it is likely that they became part of
an Atlantic-type margin as suggested by Gordon et
In Chapter 4, Pszcz6tkowski includes what we al. (1998).
refer to as remnants of the 'San Pedro del Gallo
terrane' in the 'Guaniguanico terrane' of Iturralde-
Vinent (1994, 1996). We treat the Guaniguanico ANALYSIS OF PREVIOUSLY DESCRIBED TERRANES
terrane as a junior synonym of the San Pedro del IN MEXICO (SEE FIG. 7: INSET B): VALIDITY OF MAYA,
Gallo terrane (Pessagno et al., 1993b) herein. GUACHICHIL, TEPEHUAlqO, AND COAHUILTECANO
Pszcz6tkowski followed Iturralde-Vinent (1994, TERRANES OF SEDLOCK ET AL. (1993)
1996) in suggesting that the Guaniguanico terrane
(= San Pedro del Gallo terrane) was situated along According to Howell et al. (1985, p. 4; see also
the eastern margin of the Yucat~in platform. He Howell, 1995): "A tectonostratigraphic terrane is a
admits, however, that the original position of the fault-bounded package of rocks of regional extent
Guaniguanico terrane is difficult to establish with characterized by a geologic history which differs
any degree of certainty during the Jurassic and Cre- from that of neighboring terranes. Terranes may be
taceous. In this report we advocate a similar origin characterized internally by a distinctive stratigraphy,
for the Guaniguanico terrane as that advocated by but in some cases a metamorphic or tectonic over-
Iturralde-Vinent (1994, 1996). As can be seen from print is the most distinctive characteristic. In cases
the examination of Figs. 2, 3 and 7, the Walper where juxtaposed terranes possess coeval strata, one
Megashear cuts the Yucat~in Peninsula. Hence, there must demonstrate different and unrelated geologic
is a strong case for an eastern Yucat~in origin for histories as well as the absence of intermediate litho-
the Guaniguanico terrane as advocated by Iturralde- facies that might link the two terranes. In general,
Vinent (1994, 1996). As noted, previously both the the basic characteristic of terranes is that the present
San Pedro del Gallo terrane and the Coahuilte- spatial relations are not compatible with the inferred
cana terrane show similar paleobathymetric records geologic histories."
and stratigraphic records by the Middle Cretaceous. The Mexican SPG terrane remnants occur within
We suggest that terrane amalgamation had occurred the Tepehuafio, Guachichil, and Maya terranes of
by the Middle Cretaceous and that all movement Sedlock et al. (1993). It is clear that these ter-
along the Walper Megashear had ceased. Subsequent ranes, as presently defined, are incompatible with
southwest to northeast movement of the Caribbean the SPG terrane. The focus of Sedlock et al. (1993)
Plate during the Late Cretaceous and Early Tertiary in their definition of these terranes seems to have
bulldozed the Cuban remnants of the San Pedro been mostly on the Paleozoic and Pre-Cambrian
del Gallo terrane into their present position (see basement rocks rather than on important differences
Montgomery et al., 1992, 1994a,b). in the Mesozoic stratigraphic record. This Mesozoic
Jurassic and Early Cretaceous successions in stratigraphic record is critical in plate-tectonic re-
western Cuba (Sierra del Rosario and Sierra de constructions beating on the break-up of Pangea, the
los Organos, Pifiar del Rio Province) show lithos- opening of the Gulf of Mexico, and the origin of the

Caribbean Plate. The break-up of Pangea heralds a Huayacocotla terrane of Longoria (1994). The north-
new cycle of terrane formation that crosscuts terrane ern subterrane corresponds to the Victoria segment of
and plate boundaries that characterized the Pre-Cam- the Sierra Madre Oriental of Longoria (1984). We see
brian and Paleozoic. The Tepehuafio and Guachichil no significant differences in the Mesozoic (Pangea
terranes are too generalized and all encompassing and post-Pangea) paleobathymetric, paleolatitudinal,
to be useful in plate reconstructions. Sedlock et al. and lithostratigraphic signatures of the northern sub-
(1993) failed to relate the geologic history of coeval terrane of Sedlock et al.'s Guachichil terrane and their
rock packages (see Longoria, 1994). Coahuiltecano terrane. In this report the northern sub-
The Maya terrane and the Coahuiltecano terrane in terrane is included within the Coahuiltecano terrane
an emended form may still be usable. The Coahuilte- and the term Guachichil is abandoned. The Coahuil-
cano terrane is emended herein (see below). tecano terrane is emended herein to include the north-
ern subterrane and exclude the southern subterrane
Objections to the Maya terrane of the Guachichil terrane; moreover, it is emended
to include all parts of the Tepehuafio terrane east of
Sedlock et al. (1993, p. 28) divide their Maya the Walper Megashear. As so defined, the southern
terrane into three 'geographic' provinces: boundary between the Coahuiltecano terrane and the
(1) A northern province. This province includes Huayacocotla remnant of the SPG terrane would cor-
southern Tamaulipas, Veracruz as far southeast as the respond to the Walper Megashear (Fig. 7).
Isthmus of Tehuantepec, and thin transitional crust
along the western margin of the Gulf of Mexico. Objections to Tepehuafio terrane
(2) The Yucat(m platform. The Mexican states
of Tabasco, Campeche, Quintana Roo, and Yucatan The Tepehuafio terrane as defined by Sedlock et
as well as Belize, northern Guatemala, and thinned al. (1993) encompasses the San Pedro del Gallo rem-
transitional crust in the adjacent Gulf of Mexico. nant, the Symon remnant, and the Sierra de Catorce
(3) A southern province. This province is said to remnant of the SPG terrane. Moreover, it includes
include central Guatemala, Chiapas, and northeast- the Parral and Sierra Madre Oriental Terranes of
ern Oaxaca. Campa (1983), Campa and Coney (1983) and Coney
Although all three provinces of the Maya terrane and Campa (1984) (= in part Coahuiltecano terrane
may show remnants of Pre-Cambrian and Paleo- emended herein). As presently defined it embraces to-
zoic Gondwana crust that remained behind after tally unrelated scraps of real estate on both sides of the
the break-up of Pangea and the opening of the Walper Megashear. It is obvious that the chief prob-
Gulf of Mexico, the northern province displays two lem with the definition of the Tepehuafio terrane is the
completely different stratigraphic records during the placement of the Mojave Sonora Megashear in that
Jurassic and Early Cretaceous. Most of the Mayan the alleged position of this structure defines the Tepe-
terrane in the state of Veracruz should be assigned huafio terrane's eastern boundary. As noted under the
to the SPG terrane (Huayacocotla remnant) as de- discussion of the Coahuiltecano terrane above, all of
scribed herein. That portion of Mayan terrane said the Tepehuafio terrane east of the Walper Megashear
to be in the state of Tamaulipas should be reas- is included in the Coahuiltecano terrane.
signed to the Coahuiltecano terrane (emend. herein); At present the geology of the area west of the
it is clearly a portion of the 'Victoria segment' of Walper Megashear is still too poorly understood and
the Sierra Madre Oriental as described by Longoria too complex to warrant establishing terranes as large
(1985a,b, 1986, 1987, 1994). In east-central Mexico as the Tepehuafio terrane. Stratigraphic packages are
the Huayacocotla remnant of the SPG terrane has present in different structural blocks each represent-
been juxtaposed against the Coahuiltecano terrane ing different tectonostratigraphic settings: Parral ter-
along the Walper Megashear at approximately the rane (Upper Jurassic and Lower Cretaceous flysch:
latitude of the Tampico-Ciudad Valles line. backarc? or forearc?), SPG terrane (Upper Jurassic
to Upper Cretaceous distal backarc), Zacatecas area
Objections to the Guachichil terrane and (Triassic to Jurassic Interarc). The northern bound-
Coahuiltecano terrane ary of the San Pedro del Gallo terrane occurs at
the village of Cinco de Mayo (Fig. 8) and is tenta-
Sedlock et al. (1993) divided the Guachichil ter- tively taken to be Longoria's (1994) San Pedro del
rane into provisional northern and southern subter- Gallo Fault (see Fig. 3). At Cinco de Mayo one
ranes based on the outcrop of Paleozoic sedimentary can observe a complex jumble of structural blocks
and metamorphic rocks. The southern terrane corre- that include Upper Jurassic-Lower Cretaceous sand-
sponds to the Huayacocotla remnant of the SPG ter- stones (graywacke), shales, limestone, green and
rane herein and to the Huayacocotla segment of the red tuffaceous siltstone, and olistostromal units with
Sierra Madre Oriental of Longoria (1985a,b) and the basaltic andesite clasts.

COMPARISON OF PALEOBATHYMETRY IN THE SAN like Mazaplites zitteli Burckhardt. These strata litho-
PEDRO DEL GALLO TERRANE (SPG) AND THE logically appear to be similar to those of the Taman
ADJACENT COAHUILTECANO (COAH) TERRANE Formation and the La Caja Formation. Ostrea, bry-
(EMEND. HEREIN) ozoans, and the remains of conifers occur at two
horizons. These may represent inner neritic forms
Fig. 18 shows a comparison of the composite that have been displaced by turbidity currents to
paleobathymetry of the SPG terrane and the COAH bathyal or abyssal depths.
terrane along opposing sides of the Walper Megas- By Late Cretaceous times the paleobathymetric
hear. The paleobathymetry of the SPG remnants has record of the Mexican SPG terrane remnants and
been discussed above. As can be seen in Fig. 18, the COAH terrane remnants become similar. Both ter-
paleobathymetric signature of the COAH is totally ranes show a similar lithostratigraphic record during
different. The COAH paleobathymetric signature can the remainder of the Late Cretaceous. At this point
be substantiated by examining the succession ex- in time (~Albian/Cenomanian) it would appear that
posed at Peregrina Canyon and elsewhere along the terrane amalgamation had occurred and movement
eastern front of the Sierra Madre Oriental. More- along the Walper Megashear had ceased.
over, it can be documented by examining published Anomalies to the scenario described above are
well records from the Tampico Embayment area pre- inner neritic platform deposits (e.g., E1 Abra Lime-
sented by Burckhardt (1930), Muir (1936), Imlay stone: rudistid reef complex in Sierra del Abra west
(1980), L6pez-Ramos (1985), and numerous other of Tampico) that formed during the Albian to Turo-
workers. Burckhardt (1930, p. 95) reported inner nian interval (Murray, 1961). These strata (rudistid
neritic megafossils (including Ostrea and hydroco- and miliolid limestones) were probably deposited on
rals) in upper Tithonian oolitic limestone from a seamounts at inner neritic depths and relate to a
depth of 986-1029 m in well Chocoy No. 2 about remnant horst and graben topography resulting from
50 km northwest of Tampico. Inner neritic strata previous rifting.
continue upward into the Berriasian and Valanginian
(lower part of Tamaulipas Formation).
To the southwest of Tampico near Panuco Well CONCLUSIONS
Panuco No. 82 includes Upper Jurassic (lower Titho-
nian) black carbonaceous limestones and shales with (1) Stratigraphic data from displaced terranes sit-
Aptychus, pectenacid Aulacomyella, and ammonites uated to the west of the Walper Megashear (Mexico)
demonstrate similar records of lithostratigraphy, pa-
leobathymetry, and tectonic transport from higher
latitudes to lower latitudes.
(2) In general, the stratigraphic successions in
each of these areas show the same paleobathymet-
ric fingerprint: (a) marine deposition at inner ner-
itic depths during the Callovian to early Oxfordian
(Middle to Late Jurassic); (b) marine deposition at
outer neritic depths during the late Oxfordian (Late
Jurassic); (c) sudden deepening to bathyal or upper
abyssal depths (above the ACD of aragonite) from
the early Kimmeridgian (Late Jurassic) until the end
of the Cretaceous.
(3) This paleobathymetric fingerprint differs
markedly from that occurring to the east of the
Walper Megashear in the Coahuiltecano terrane
(emend.) (e.g., Sierra Jimulco, Coahuila; Peregrina
Canyon, Tamaulipas).
(4) In this report we regard the Upper Juras-
sic and Lower Cretaceous successions at San Pe-
dro del Gallo, Symon and Sierra Ramirez, Mazapil
(Sierra Santa Rosa), Sierra de la Caja, Sierra Zuloaga
and Sierra Sombretillo, Sierra Cadnelaria, Sierra de
Catorce, and in the Huayacocotla Anticlinorium to
represent remnants of a single terrane, the San Pedro
Fig. 18. Comparison of paleobathymetry between Cohuiltecana del Gallo terrane, that has undergone dismember-
terrane and SPG terrane. ment and tectonic transport to varying degrees (NW

to SE) along the west (southwest) side of the Walper western North America. Although a backarc origin
Megashear (see Fig. 2: Locs. 3, 5, 6, 7, 8, 9, and is advocated for the pre-latest Tithonian (radiolarian
12; Burckhardt, 1930; Imlay, 1980). The San Pedro Subzone 4 alpha: Fig. 9), such an origin is related
del Gallo remnant of the San Pedro del Gallo terrane to the Nevadian island arc and not the Antillean
originated at Southern Boreal paleolatitudes (~40~ island arc. In addition, these Cuban remnants are al-
according to preliminary paleomagnetic data) during lochthonous when compared to surrounding Central
the Oxfordian and was tectonically transported to Tethyan successions in the nearby Blake Bahama
Northern Tethyan paleolatitudes (22 ~ to 29~ by Basin and elsewhere in Cuba. They contain high lati-
latest Tithonian or earliest Berriasian times. Fau- tude bivalves such as species of Buchia that can only
nal data (radiolarians and megafossils) from the be derived (exclusive of Greenland) from a Pacific
Mazapil succession (Sierra Santa Rosa) indicates source. The presence of Southern Boreal/Northern
that this remnant of the San Pedro del Gallo ter- Tethyan faunas (~30~ in the Sierra de los Organos
rane was situated at Southern Boreal paleolatitudes and Sierra del Rosario remnants as late as the Early
(>30~ during the Oxfordian and Kimmeridgian Cretaceous (Valanginian) suggests much later tec-
and at Northern Tethyan paleolatitudes (22 ~ to 29~ tonic transport by northwest to southeast movement
during the Tithonian and Berriasian. Preliminary along the Walper Megashear and by subsequent
paleomagnetic data from the upper Tithonian to southwest to northeast movement as the Caribbean
Berriasian part of the Mazapil succession indicates Plate plowed its way through the gap between the
~25~ Farther south in the state of San Luis Po- North American and South American plates. As
tosi, the Sierra de Catorce remnant was situated in suggested by Pszcz6lkowski (see Chapter 4) and
the Southern Boreal Province during the Oxfordian by Iturralde-Vinent (1994, 1996), the Guaniguanico
to Kimmeridgian interval. During the Tithonian the terrane (= San Pedro del Gallo terrane) was situ-
Sierra de Catorce remnant was probably situated in ated along the eastern margin of the Yucatfin plat-
the Northern Tethyan Province. Farther to the east form. This hypothesis is supported by the fact that
(San Luis Potosi, Hidalgo, Veracruz, Puebla) in the the Walper Megashear cuts the Yucatfin Peninsula.
Huayacocotla segment of the Sierra Madre Orien- We suggest that terrane amalgamation had occurred
tal previous investigations indicate tectonic transport by the Middle Cretaceous and that all movement
from Southern Boreal paleolatitudes (> 30~ during along the Walper Megashear had ceased. Subsequent
the Callovian (Middle Jurassic) to Northern Tethyan southwest to northeast movement of the Caribbean
paleolatitudes (22~ 29~ during the Kimmerid- Plate during the Late Cretaceous and Early Tertiary
gian and Tithonian (Late Jurassic) to central Tethyan bulldozed the Cuban remnants of the San Pedro del
paleolatitudes (<22~ during the Berriasian (Early Gallo terrane into their present position (see Mont-
Cretaceous). The kinetics of terrane remnants along gomery et al., 1992, 1994a,b). Once the Cuban San
the Walper Megashear can be likened to blocks of Pedro del Gallo remnants were carried northward by
ice in an ice flow with most blocks being episodi- the advancing Caribbean Plate, it is likely that they
cally rotated during transport, some blocks moving became part of an Atlantic-type margin as suggested
along at steady rate (e.g., Huayacocotla remnant and by Gordon et al. (1998).
San Pedro del Gallo remnant), and still others mov-
ing very little while rotating in place (e.g., Mazapil
remnant). ACKNOWLEDGEMENTS
(5) In western Cuba the Sierra del Rosario and
Sierra de los Organos successions are likewise re- This investigation was in part supported by
garded to be remnants of the SPG terrane and show grants from the National Science Foundation:
stratigraphic, paleobathymetric, and paleolatitudinal EAR-9418194 to Pessagno and Montgomery and
signatures which are nearly identical to those of San EAR-9304459 to Pessagno, Hull, and Ogg. It has
Pedro del Gallo terrane remnants in Mexico (Figs. 9, also been supported by a grant from CONOCYT
11, 16 and 17). Even the presence of the inner neritic to Urrutia Fucugauchi. We also wish to thank
'San Vicente Member' (Guasasa Formation) has an Dr. Jurgen Remane (Institut de G6ologie, Uni-
analogue in the San Andres Limestone of the Huay- versit6 de Neuch~tel, Switzerland) for identifying
acocotla remnant (Fig. 2: Loc. 3). This record at Late Jurassic and Early Cretaceous calpionellids.
least during the Jurassic and Early Cretaceous is Finally, we would like to thank Dr. Norman Sil-
distinctly North American. Unconformities and hia- berling (U.S.G.S., Denver, Colorado), Dr. Charles
tuses are regional in distribution and can be traced D. Blome (U.S.G.S., Denver, Colorado), and Dr.
as far north as the California Coast Ranges and Kla- Gyorgy L. Marton, Amoco Exploration and Produc-
math Mountains (see explanation for Fig. 11). More- tion Company, Houston, Texas for their review of
over, these unconformities reflect tectonic events that the manuscript. Contribution No. 826, Programs in
affect the Nevadian forearc, interarc, and backarc in Geosciences, The University of Texas at Dallas.

148 E.A. P E S S A G N O et al.
REFERENCES Continental Drift Hypothesis. Symposia of the Geology De-

partment, University of Tasmania, 2, 375 pp.
Adatte, T., Stinnesbeck, W. and Remane, J., 1995. The Jurassic- Carrillo-Bravo, J., 1961. Geologia del Anticlinorio Huizachal-
Cretaceous boundary in Northeastern Mexico. Confrontation Peregrina al N.W. de Ciudad Victoria, Tamaualipas. Asoc.
and correlations by microfacies, clay minerals, mineralogy, Mex. G6ol. Pet. Bol., 13" 1-98.
calpionellids, and ammonites. Geobios, 17: 37-56. Coney, P.J. and Campa, M.E, 1984. Lithotectonic terrane map of
Aita, Y. and Grant-Mackie, J.A., 1992. Late Jurassic Radiolaria Mexico. In: N.J. Silberling and D.L. Jones (Editors), Litho-
from Kowhai Point Siltstone, Murihku terrane, North Island, tectonic Terrane Maps of the North American Cordillera.
New Zealand. In: K. Ishizaki and T. Saito, (Editors), Centenary U.S.Geol. Surv., OF84-0506, pp. D 1-D 14.
of Japanese Micropaleontology. Terra Scientific Publishing Contreras-Montero, B., Martinez-Cortez, A. and Gomez-Luna,
Co., Tokyo, pp. 375-382. M., 1988. Bioestratigrafia y sedimentologia del Jurasico Supe-
Anderson, T.H. and Schmidt, V.A., 1983. The evolution of Mid- rior en San Pedro del Gallo, Durango, Mexico. Rev. Inst. Mex.
dle America and the Gulf of Mexico-Caribbean sea region Pet., 20 (3): 5-49.
during Mesozoic time. Geol. Soc. Am. Bull., 94: 941-966. Davila-Alcocer, V., 1986. Biostratigraphic Studies of Jurassic,
Arkell, W.J., 1956. Jurassic Geology of the World. Oliver and Cretaceous Radiolaria of the Eugenia and Asuncion Forma-
Boyd, London, 806 pp. tions, Vizcaino Peninsula, Baja California Sur. M.S. Thesis,
Baumgartner, P.O., 1984. A Middle Jurassic-Early Cretaceous University of Texas at Dallas, 157 pp.
radiolarian zonation based on Unitary Associations and age of Dickinson, W.R. and Coney, P.J., 1981. Plate tectonic constraints
Tethyan radiolarites. Eclogae Geol. Helv. 77: 729-837. on the origin of the Gulf of Mexico. In: R.H. Pilger, Jr.
Baumgartner, P.O., 1987. Age and genesis of tethyan Jurassic (Editor), The Origin of the Gulf of Mexico and Early Opening
radiolarites. Eclogae Geol. Helv., 80:831-879. of the Central North Atlantic Ocean. Proc. Symp. Louisiana
Blome, C.D., 1987. Paleogeographic significance of lower Meso- State University, Baton Rouge, March 1980, pp. 27-36.
zoic radiolarians from the Brooks Range, Alaska. In: L. Dietz, R.S. and Holden, R.C., 1970. Reconstruction of Pangea:
Taileur, and P. Weimer (Editors). Alaska North Slope Geol- Breakup and dispersion of continents, Permian to present. J.
ogy. Society of Economic Paleontologists and Mineralogists, Geophys. Res., 75: 4937-4956.
Pacific Section. 1: pp. 371-380. Douglas, R.G., 1969. Upper Cretaceous planktonic foraminifera
Blome, C.D. and Reed, K.M., 1993. Acid processing in pre-Ter- in Northern California. Micropaleontology, 15 (2): 151-209.
tiary radiolarian cherts and its impact on faunal content and Erben, H.K., 1956. E1 Junisico Inferior de M6xico y sus amoni-
biozonal correlation. Geology, 21: 177-180. tas, M6xico D.F. Int. XX Congr. Geol. Internacional, 393 pp.
BrOnnimann, P., 1954. On the occurrence of calpionellids in Gordon, M.B., Mann, P., C~iceres, D., Flores, R., 1998. Ceno-
Cuba. Ecologae Geol. Helv., 46 (2): 263-268. zoic tectonic history of the North American-Caribbean plate
Buffler, R.T., Watkins, J.S., Shaub, EJ. and Worzel, J.L., 1981. boundary zone in western Cuba. J. Geophys. Res. (in press).
Structure and early geologic history of the deep central Gulf Haczewski, G., 1976. Sedimentological reconnaissance of the
of Mexico Basin. In: R.H. Pilger, Jr. (Editor), The Origin of San Cayetano Formation; an accumulative continental margin
the Gulf of Mexico and Early Opening of the Central North in the Jurassic of Western Cuba. Acta Geol. Pol., 26:331-353.
Atlantic Ocean. Proc. Symp. Louisiana State University, Baton Hagstrum, J.T. and Murchey, B.L., 1996. Paleomagnetism of
Rouge, March 1980, pp. 41-51. Jurassic radiolarian chert above the Coast Range ophiolite at
Bullard, E.C., Everett, J.E. and Smith, A.G., 1965. The fit of the Stanley Mountan, California, and implications for its paleo-
continents around the Atlantic: a symposium on continental geographic origins. Geol. Soc. Am. Bull. 108: 643-652.
drift. Philos. Trans. R. Soc. London, A, 258: 3-16. Hatten, C.W., 1967. Principal features of Cuban geology. Am.
Burckhardt, C., 1910. Estudio geologico de la regi6n de San Assoc. Pet. Geol. Bull., 51: 780-789.
Pedro del Gallo (Durango). Instituto Geol. de M6xico, Par- Herrera, N.M., 1961. Contribuci6n a la estratigraffa de la provin-
ergones, III, 6, pp. 307-357. cia de Pinar del Rfo. Soci. Cubana Ing., 61 (1-2): 1-24.
Burckhardt, C., 1912. Faunes jurassiques et cr6taciques de San Hopson, C.A., Mattinson, J.M. and Pessagno, E.A., Jr., 1981.
Pedro del Gallo. Inst. Geol. M6x., 29:264 pp. Coast Range ophiolite, western California. In: W.G. Ernst
Burckhardt, C., 1927. Cefal6podos del Jur~isico medio de Oaxaca (Editor), The Geotectonic Development of California. Pren-
y Guerrero. Inst. Geol. M6x. Bol., 47:108 pp. tice-Hall, Englewood Cliffs, NJ. Rubey 1: pp. 419-510.
Burckhardt, C., 1930. l~tude synth6tique sur le M6sozoique M6x- Hopson, C.A., Pessagno, E.A., Jr., Mattinson, J.M., Luyendyk,
icain. Soc. Pal6ontol. Suisse M6m., 49: 1-123. B.P., Beebe, W., Hull, D.M., Mufioz, I.M. and Blome,
Burckhardt, C., 1931. l~tude synth6tique sur le M6sozoique M6x- C.D., 1996. Coast Range ophiolite as paleoequatorial mid-
icain. Soc. Pal6ontol. Suisse M6m., 50: 123-280. ocean lithosphere. In: W.R. Dickinson, C.A. Hopson and J.B.
Campa, M.E, 1983. The Mexican thrust belt. In: D.G. Howell Saleeby (Editors), Alternate Origins of the Coast Range Ophi-
(Editor), Tectonostratigraphic Terranes of the Circum Pacific olite (California): Introduction and Implications. GSA Today,
Region. Circum-Pacific Council for Energy and Mineral Re- 6 (2): 1-10.
sources, Earth Science Series, 1, pp. 299-313. Howell, D.G., 1995. Principles of Terrane Analysis: New Appli-
Campa, M.F. and Coney, P.J., 1983. Tectonostratigraphic terranes cations for Global Tectonics. Topics in the Earth Science 8,
and mineral resource distributions in Mexico. Can. J. Earth Chapman and Hall, London, 2nd ed., 245 pp.
Sci., 20:1040-1051. Howell, D.G., Jones, D.L. and Schermer, E.R., 1985. Tectonos-
Cantti-Chapa, A., 1969. Estratigraffa del Junisico medio-superior tratigraphic terranes of the Circum-Pacific region. In: D.G.
del subsuelo de Poza Rica, Ver. Inst. Mex. Pet. Rev., 1 (1): Howell (Editor), Tectonostratigraphic Terranes of the Cir-
3-9. cum-Pacific Region. Circum-Pacific Council for Energy and
Cantti-Chapa, A., 1971. La serie Huasteca (Jur~isico medio-supe- Mineral Resources, Earth Science Series, 1, pp. 1-581.
rior) del centro este de Mexico. Inst. Mex. Pet. Rev., 3 (2): Hull, D.M., 1991. Upper Jurassic Radiolarian Biostratigraphy
17-40. of the Lower Member of the Taman Formation, East-Central
Cantti-Chapa, A., 1989. Precisiones sobre el limite Jurasico- Mexico and of Volcanopelagic Strata Overlying the Coast
Cretacico en el subsuelo del este de M6xico. Rev. Mex. Range Ophiolite, Stanley Mountain, Southern California Coast
Paleontol., 2 (1): 26-69. Ranges. Ph.D. dissertation, The University of Texas at Dallas,
Carey, S.W., 1958. Symposium on the Present Status of the 696 pp.

N W TO SE T E C T O N I C T R A N S P O R T OF JURASSIC T E R R A N E S IN M E X I C O AND THE C A R I B B E A N 149
Hull, D.M., 1995. Morphologic diversity and paleogeographic Longoria, J.E, 1994. Recognition and characteristics of a strike-
significance of the Family Parvicingulidae (Radiolaria). Mi- slip fault system in Mexico and its Mesozoic transpressional
cropaleontology, 41 (1): 1-48. regime: implications in plate tectonics and paleogeographic
Hull, D.M., Pessagno, E.A., Jr., Blome, C.D., Hopson, C.A. reconstruction. Bol. Dept. Geol. Uni-Son., 11 (1): 77-104.
and Mufioz, I.M., 1993. Chronostratigraphic assignment of L6pez-Ramos, E., 1985. Geologfa de M6xico, Tomo II. Impre-
volcanopelagic strata above the Coast Range ophiolite. In: G. siones Resendiz, Mexico, D.E, 453 pp.
Dunn and K. McDougall (Editors), Mesozoic Paleogeography Martin, C.B., 1996. Tectonostratigraphic analysis of the Up-
of the Western United States, II. Pacific Section SEPM, Book per Jurassic through Lower Cretaceous stratigraphy in the
71, pp. 157-170. Mazapil, Sierra de Jimulco, and San Pedro del Gallo regions
Hull, D.M., Blome, C.D. and Pessagno, E.A., Jr., 1997, Paleo- of North-central Mexico. Thesis (M.S.), Geosciences Depart-
magnetism of Jurassic radiolarian chert above the Coast Range ment, The University of Texas at Dallas: pp. 1-148.
ophiolite at Stanley Mountain, California, and implications for McWilliams, M.O. and Howell, D.G., 1982. Exotic terranes of
its paleogeographic origins: Discussion and reply. Discussion. western California. Nature, 297:215-217.
Geol. Soc. Am. Bull., 109 (12): 1633-1639. Meng, X., 1997, Radiolarian biostratigraphy of the Upper Juras-
Imlay, R.W., 1937. Geology of the middle part of the Sierra de sic of San Pedro del Gallo terrane, north-central Mexico and
Parras, Coahuila, Mexico. Geol. Soc. Am. Bull., 48: 587-630. the Lower Cretaceous of Nooksack Group, Nooksack Terrane,
Imlay, R.W., 1938. Studies in the Mexican Geosyncline. Geol. northwestern Washington. Ph.D. Dissertation, The University
Soc. Am. Bull., 49:1651-1694. of Texas at Dallas, 351 pp.
Imlay, R.W., 1939. Upper Jurassic ammonites from Mexico. Meyerhoff, A.A., 1964. Review of the geological formations of
Geol. Soc. Am. Bull., 50: 1-78. Cuba (Bermudez, 1961). Int. Geol. Rev., 6 (1): 149-156.
Imlay, R.W., 1980. Jurassic paleobiogeography of the contermi- Montgomery, H.A., Pessagno, E.A., Jr. and Mufioz, I., 1992.
nous United States in its continental setting. U.S. Geol. Surv. Jurassic (Tithonian) Radiolaria from La D6sirade (Lesser An-
Prof. Pap., 1062: 1-134. tilles): preliminary paleontological and tectonic implications,
Iturralde-Vinent, M.A., 1994. Cuban Geology, A new plate-tec- Tectonics, 11 (6): 1426-1432.
tonic synthesis. J. Pet. Geol., 17: 39-70. Montgomery, H.A., Pessagno, E.A., Jr. and Pindell, J.L., 1994a.
Iturralde-Vinent, M.A., 1996. Ofiolitas y arcos volcanicos de A 195 Ma terrane in a 165 Ma sea: Pacific origin of Caribbean
Cuba: Project 364, Caribbean ophiolites and volcanic arcs. Plate. GSA Today, 4 (1): 3-6.
Special Contribution No. 1, Miami, FL: pp. 1-254. Montgomery, H., Pessagno, E.A., Jr., Lewis, J.A. and
Jones, D.L., Silberling, N.J. and Hillhouse, J., 1977. Wrangellia Schellekens, J.H., 1994b. Paleogeography of the Jurassic frag-
- - a displaced terrane in northwestern North America. Can. J. ments in the Caribbean, Tectonics, 13: 725-732.
Earth Sci., 14 (11): 2565-2577. Muir, J.M., 1936. Geology of the Tampico Embayment area. Am.
Kiessling, W., 1995. Pal0kologische Verwertbarkeit Ober-Juras- Assoc. Pet. Geol., 280 pp.
sisch Unterkretazischer Radiolarienfaunen mit Beisp. Ph.D. Murray, G.E., 1961. Geology of the Atlantic and Gulf Coastal
dissertation, Erlangen University, Erlangen. Province of North America. Harper, New York, pp. 1-692.
Kiessling, W. and Scasso, R., 1996. Ecological perspectives of Myczyfiski, R., 1989. Ammonite biostratigraphy of the Tithonian
Late Jurassic radiolarian faunas from the Antarctic Peninsula. of western Cuba. Ann. Soc. Geol. Pol., 59: 43-125.
In: A.C. Riccardi (Editor), Advances in Jurassic Research. Myczyfiski, R., 1994. Caribbean ammonite assemblages from
GeoRes. Forum, 1-2:317-326. Upper Jurassic-Lower Cretaceous sequences of Cuba. Stud.
Khudoley, K.M. and Meyerhoff, A.A., 1971. Paleogeography Geol. Pol., 105:91-108.
and geological history of the Greater Antilles. Geol. Soc. Am. Myczyfiski, R. and Pszcz6~kowski, A., 1976. The ammonites
Mem., 129:200 pp. and age of the San Cayetano Formation from the Sierra del
Kutek, J., Pszcz6tkowski, A. and Wierzbowski, A., 1976. The Rosario, western Cuba. Acta Geol. Pol., 26 (2): 321-329.
Francisco Formation and an Oxfordian ammonite faunule from Myczyfiski, R. and Pszcz6lkowski, A., 1994. Tithonian stratigra-
the Artemisa Formation, Sierra del Rosario, Western Cuba. phy and microfacies in the Sierra del Rosario, western Cuba.
Acta Geol. Pol., 26 (2): 299-319. Stud. Geol. Pol., 105: 7-38.
Lewis, J.E and Draper, G., 1990. Geology and tectonic evolution Nestell, M.K., 1979. Lower Permian fusulinids from the vicin-
of the northern Caribbean margin. In: G. Dengo and J.E. Case ity of Tianguistengo, Hidalgo, Mexico. Texas Academy of
(Editor), The Geology of North America, H. The Caribbean Sciences Meeting, Arlington, Texas, Prog. Abstr., p. 40.
Region. Geological Society of America, Boulder, Colo., pp. Ogg, J.G., 1983. Magnetostratigraphy of the Upper Jurassic and
77-140. lowest Cretaceous sediments, Deep Sea Drilling Site 534,
Longoria, J.F., 1984. Mesozoic tectostratigraphic domains Western North Atlantic. In: S. Orlofsky, Initial reports of the
in east-central Mexico. In: G.E.G. Westermann (Editor), Deep Sea Drilling Project, Washington, D.C., U.S. Govern-
Jurassic-Cretaceous Biochronology and Paleogeography of ment Printing Office. 76: pp. 685-697.
North America. Geol. Assoc. Can. Spec. Pap., 27: 65-76. Pessagno, E.A., Jr., 1995. Stratigraphic evidence for Northwest
Longoria, J.E, 1985a. Tectonic transpression in the Sierra Madre to Southeast movement along the west side of the Walper
Oriental, northeastern Mexico: an alternative model. Geology, Lineament. Geol. Soc. Am., Abstr. Prog., 27 (6): A75.
13 (7): 453-456. Pessagno, E.A., Jr. and Blome, C.D., 1986. Faunal affinities
Longoria, J.E, 1985b. Tectonic transpression in northeast Mex- and tectonogenesis of Mesozoic rocks in the Blue Mountains
ico: its relation to sea floor spreading in the Gulf of Mexico. Province of eastern Oregon and western Idaho. In: T.L. Vallier
Gulf Coast Assoc. Geol. Soc., Trans., 35: 199. and H.C. Brooks (Editors), Geology of the Blue Mountains
Longoria, J.E, 1986, Tectonic transpression in the Sierra Madre Region of Oregon, Idaho, and Washington: Geologic Implica-
Oriental, northeast Mexico: an alternative model. Reply. Geol- tions of Paleozoic and Mesozoic Paleontology and Biostratig-
ogy 14 (9): 809-810. raphy. U.S. Geol. Surv. Prof. Pap., 1435: 65-78.
Longoria, J.E, 1987. Oblique subduction and kinematics of the Pessagno, E.A., Jr. and Hull, D.M., 1999. Upper Jurassic (mid-
American Plate: evidence from the stratigraphic record of dle Oxfordian Radiolaria) from the Sula Islands (East Indies):
Mexico. In: T.W.C. Hilde and R.L. Carlson (Conveners), Geo- their taxonomic, biostratigraphic, and paleogeographic signifi-
dynamics Symposium. Texas A and M University, College cance (in prep.).
Station, 2 pp. Pessagno, E.A., Jr. and Newport, R.L., 1972. A technique for ex-

tracting Radiolaria from radiolarian cherts, Micropaleontology, Sandoval, J. and Westermann, G.E.G., 1988. The Bajocian
18 (2): 231-234. (Jurassic) ammonite fauna of Oaxaca, Mexico. J. Paleontol.,
Pessagno, E.A., Jr., Blome, C.D. and Longoria, J.E, 1984. A 60: 1220-1271.
revised radiolarian zonation for the Upper Jurassic of western Sandoval, J., Westermann, G.E.G. and Marshall, M.C., 1990.
North America. Bull. Am. Paleontol., 87 (320): 1-51. Ammonite fauna, stratigraphy, and ecology of the Bathonian-
Pessagno, E.A., Jr., Whalen, EA. and Yeh, K.Y., 1986. Juras- Callovian (Jurassic) Tecocoynca Group, south Mexico. Pale-
sic Nassellariina (Radiolaria) from North American Geologic ontographica, A210: 93-149.
Terranes. Bull. Am. Paleontol., 91 (326): 1-75. Sedlock, R.L., Ortega-Guti6rrez, E and Speed, R.C., 1993.
Pessagno, E.A., Jr., Longoria, J.E, MacLeod, N. and Six, W.M., Tectonostratigraphic terranes and tectonic evolution of Mex-
1987. Upper Jurassic (Kimmeridgian-upper Tithonian) Pan- ico. Geol. Soc. Am. Spec. Pap., 278: 1-153.
tanelliidae from the Taman Formation, east-central Mexico: Smith, EL., 1980. Correlation of the members of the Jurassic
tectonostratigraphic, chronostratigraphic, and phylogenetic im- Snowshoe Formation in the Izee basin of east-central Oregon.
plications. In: S.J. Culver (Editor), Studies of North American Can. J. Sci., 17 (12): 1603-1608.
Jurassic Radiolaria, Part I. Cushman Found. Foraminiferal Taylor, D.G., Callomon, J.H., Hall, R., Smith, EL., Tipper, H.W.
Res. Spec. Publ., 23:1-51. and Westermann, G.E.G., 1984. Jurassic ammonite biogeogra-
Pessagno, E.A., Jr., Blome, C.D., Hull, D. and Six, W.M., Jr., phy of western North America: the tectonic implications. In:
1993a. Middle and Upper Jurassic Radiolaria from the Western G.E.G. Westermann (Editor), Jurassic-Cretaceous Biochronol-
Klamath terrane, Smith River subterrane, northwestern Cali- ogy and Paleogeography of North America. Geol. Assoc. Can.
fornia: their biostratigraphic, chronostratigraphic, geochrono- Spec. Pap., 27: 121-141.
logic, and paleolatitudinal significance. Micropaleontology, 39 Tipper, H.W., 1981. Offset of an upper Pliensbachian geographic
(2): 93-166. zonation in the North American Cordillera by transcurrent
Pessagno, E.A., Jr., Hull, D.M., Longoria, J.E and Kelldorf, fault movement. Can. J. Earth Sci., 18: 1788-1792.
M.E., 1993b. Tectonostratigraphic significance of the San Pe- Van der Voo, R., Mauk, EJ. and French, R.B., 1976. Permian-
dro del Gallo area, Durango, Western Mexico. In: G. Dunn Triassic continental configurations and the origin of the Gulf
and K. McDougall (Editors), Mesozoic Paleogeography of the of Mexico. Geology, 4: 177-188.
Western United States, II. Pacific Section SEPM, Book 71, pp. Verma, H.M. and Westermann, G.E.G., 1973. The Tithonian
141-156. (Jurassic) ammonite fauna and stratigraphy of Sierra de
Pessagno, E.A., Jr., Six, W.M. and Yang, Q., 1989. The Catorce, San Luis Potosi, Mexico. Bull. Am. Paleontol., 63:
Xiphostylidae Haeckel and Parvivaccidae, n. fam., (Radio- 107-278.
laria) from the North American Jurassic. Micropaleontology, von Hillebrandt, A., Smith, E, Westermann, G.E.G. and Cal-
35: 193-255. lomon, J.H., 1992. Ammonite zones of the circum-Pacific
Pindell, J.L., 1985. Alleghenian reconstruction and subsequent region. In: G.E.G. Westermann (Editor), The Jurassic of the
evolution of the Gulf of Mexico, Bahamas, and Proto- Circum-Pacific. Cambridge University Press, Cambridge, 676
Caribbean. Tectonics, 4: 1-39. PP.
Pszcz6tkowski, A., 1978, Geosynclinal sequences of the Walper, J., 1981. Tectonic evolution of the Gulf of Mexico. In:
Cordillera de Guaniguanico in western Cuba; their lithostratig- R.H. Pilger, Jr. (Editor), The Origin of the Gulf of Mexico
raphy, facies development and paleogeography. Acta Geol. and Early Opening of the Central North Atlantic Ocean. Proc.
Pol., 28: 1-96. Symp. Louisiana State University, Baton Rouge, March 1980,
Pujana, I., 1989. Stratigraphical distribution of the multicyrtid pp. 27-98.
Nassellariina (Radiolaria) at the Jurassic-Cretaceous boundary Walper, J. and Rowett, C.L., 1972. Plate tectonics and the origin
in the Neuqu6n Basin, Argentina. Zbl. Geol. Pal~iontol. I (5- of the Caribbean Sea and the Gulf of M6xico. Gulf Coast
6): 1043-1052. Assoc. Geol. Soc. Trans., 22: 105-116.
Pujana, I., 1991, Pantanelliidae (Radiolaria) from the Tithonian Whalen, EA., 1985. Lower Jurassic Radiolarian Biostratigraphy
of the Vaca Muerta Formation, Neuqu6n, Argentina. Neues of the Kunga Formation, Queen Charlotte Islands, British
Jahrb. Geol. Paleontol. Abh., 180 (3): 391-408. Columbia and the San Hipolito Formation, Baja California
Pujana, I., 1993. Middle Jurassic (Bathonian-Callovian) Radio- Sur. Ph.D. dissertation, University of Texas at Dallas, 440 pp.
laria from Chacay Melehue, Cordillera del Viento, Province of Whalen, EA. and Pessagno, E.A., Jr., 1984. Lower Jurassic
Neuqu6n, Argentina. Master of Science thesis, Univ. of Texas Radiolaria, San Hipolito Formation, Vizcaino Peninsula, Baja
at Dallas, pp. 1-87. California Sur. In: V.A. Frizzell, Jr. (Editor), Geology of the
Pujana, I., 1996. Occurrence of Vallupinae (Radiolaria) in the Baja California Peninsula. Pacific Section SEPM, pp. 53-65.
Neuqu6n Basin: Biostratigraphic Implications. In: A.C. Ric- Wierzbowski, A., 1976. Oxfordian ammonites of the Pifiar del
cardi (Editor), Advances in Jurassic Research. GeoRes. Forum, Rfo Province (western Cuba); their revision and Stratigraphical
1-2: 459-456. significance. Acta Geol. Pol., 26 (2): 138-260.
Salvador, A., Westermann, G.E.G., O16ritz, E, Gordon, M.B. Yeh, K. and Cheng, Y., 1996. Jurassic Radiolarians from the
and Gursky, H.J., 1992. Meso-America. In: G.E.G. Wester- northwest coast of Busuanga Island, North Palawan Block,
mann (Editor), The Jurassic of the Circum-Pacific. Cambridge Philippines. Micropaleontology, 42 (2): 93-124.
University Press, Cambridge, pp. 1-676.

Chapter 6
Cretaceous Carbonate B iostratigraphy and Environments in

Honduras
ROBERT W. SCOTT and R.C. FINCH
A carbonate platform developed on the Chortfs block beginning in the Berriasian-Aptian and ending in the Albian. This
platform is represented in Honduras by carbonate strata of the Yojoa Group, mainly the Atima Formation. We report new
stratigraphic and paleontological data from exposures of this carbonate package and superjacent Upper Cretaceous strata in
easternmost Honduras. In addition, new paleontological data from west-central Honduras are presented confirming previous
interpretations there.
In west-central Honduras, carbonate deposition of the Atima Formation was interrupted in the Aptian by an influx of
non-marine, terrigenous sediments of the 'Mochito shale', and terminated at the beginning of the Cenomanian by coarse redbed
strata of the lower Valle de Angeles Group. The exact nature of the Atima/Valle de Angeles contact in west-central Honduras is
not known, but may be gradational. In the Montafias de Col6n area in eastern Honduras, no equivalent of the 'Mochito shale' has
yet been identified, and late Albian deposition of the Atima was ended by an influx of fine-grained, shallow marine, terrigenous
sediments of the Cenomanian 'Krausirpi beds'. Although the Krausirpi beds are only known locally, we include them in the Yojoa
Group as an informal member. After deposition of the Krausirpi beds, local subaerial exposure and erosion occurred prior to
deposition of the redbeds of the lower Valle de Angeles Group.
In west-central Honduras, carbonate platform deposition was resumed during the Cenomanian represented by the Jaitique
Formation and by the slightly younger and lithologically distinct Esqufas Formation just east of the outcrop area of the Jaitique.
These limestone units conveniently divide the Valle de Angeles Group into lower and upper redbed sections. No equivalent
limestone unit has been found in the Montafias de Col6n area. Instead, deposition of continental redbeds began here in the
Cenomanian and continued into latest Cretaceous, as indicated by radiometrically dated volcanic flow units within the redbeds.
The Jaitique Formation consists of a lower thick-bedded limestone member overlain by a thin shaly limestone member in the
area south and west of Lake Yojoa. In central Honduras the Esqufas Formation consists of interbedded limestone, marl and shale.
The remainder of Cretaceous deposition was fluvial clastic sediments and volcanic rocks which belong to the upper Valle de
Angeles Group.
New paleontologic data from limestone of the Atima Formation, shale of the Krausirpi beds, limestone conglomerate in the
Valle de Angeles Group, and from the Jaitique Formation confirm the ages and depositional environments. The Atima ranges
from Barremian-Aptian to late Albian based on ten foraminiferal taxa, one caprinid rudist and three calcareous algae. The
Krausirpi beds are Albian-Cenomanian based on nine palynomorph taxa and two planktic foraminifers. The Jaitique is middle
to late Cenomanian with the benthic foraminifer Biconcava, which is reported for the first time in the Caribbean Province. A
Cenomanian dasyclad, Dissocladella undulata, was recovered in a limestone clast in the Valle de Angles Group suggesting the
possibility that Cenomanian as well as older limestone units were eroded during the Late Cretaceous. The Atima records at least
two shoaling upward depositional cycles, and the Jaitique was deposited in an open shelf environment that became restricted near
the end of deposition.
INTRODUCTION the Chortfs block, and published detailed strati-

graphic and paleontologic data were, until recently,
Cretaceous marine and non-marine strata cover restricted mainly to the western and central portions
much of the Chortfs block of northern Central Amer- of the republic. The geology of easternmost Hon-
ica, and represent Early to mid-Cretaceous carbonate duras, however, was known from reconnaissance
platform sedimentation buried by Late Cretaceous work by Mills et al. (1967), Mills and Hugh (1974),
terrigenous redbeds. Honduras comprises most of and Weiland et al. (1992). The first detailed strati-

152 R.W. SCOTT and R.C. FINCH
graphic data from this part of the Chortfs block has its present position. Consensus is growing, based on
recently been provided by Rogers (1994, 1995) and paleomagnetic data, lithologic similarities between
Mills and Barton (1996). Mills and Barton (1996) Chortfs and the Guerrero block of Mexico, and plate
used these Cretaceous carbonate strata in southeast- reconstructions, that Chortfs formerly lay along the
ern Honduras to define the Col6n platform. present southwest coast of Mexico (e.g, Anderson
The geology and tectonic history of the Chortfs and Schmidt, 1983; Dengo, 1985; Gose, 1985a;
block has been summarized by Azema et al. (1985), Fourcade and Michaud, 1987; Pindell and Barrett,
Donnelly et al. (1990), Finch and Dengo (1990), 1990; Burkart, 1994). As a consequence of its loca-
and Gordon and Muehlberger (1994). The Chortfs tion west of Mexico during the Early Cretaceous, the
block, which today forms the northwest comer of new paleontological data reported here have biogeo-
the Caribbean plate, is bounded on the southwest by graphic significance. The biota of the Chortfs block
the Middle America Trench subduction zone and on is very similar to that of the Caribbean Province.
the north by the Cayman transform and its continen- The primary purpose of this paper is to re-
tal extension in the Motagua-Chixoy-Polochfc fault port new paleontological results from samples col-
zones. To the south the Chortfs block is thought to be lected in easternmost Honduras by R.C. Finch and
separated from the Chorotega block of the Caribbean T.J. Weiland in a traverse across the Montafias de
plate by the Santa Elena fault-Hess escarpment. The Col6n fold belt, and along the Rio Patuca and Rio
eastern limit is undefined, but the Nicaraguan rise Wampfi, and to place these results in the context of
is generally included as a submarine extension of stratigraphic interpretations made by Rogers (1994,
the Chortfs block (e.g., Pindell and Barrett, 1990). 1995). This report provides new data from a region
Prior to its Late Cretaceous suturing against the lying some 200 km east of most of the previously
Maya block (Donnelly et al., 1990), the Chortfs was studied areas of the Honduran Cretaceous. A sec-
independent of the Maya, as indicated by differences ondary purpose is to describe other collections yield-
in basement rocks (Gordon, 1989a,b; Gordon and ing new paleontological data from the region south
Gose, 1989; Donnelly et al., 1990) and paleomag- and west of Lake Yojoa in west-central Honduras
netic data (Gose, 1985a,b; Gordon and Gose, 1989). (Fig. 1).
Gose's paleomagnetic data also strongly imply that In 1984 Amoco Production Co. sponsored a field
prior to its emplacement adjacent to the Maya block, party to eastern Honduras to investigate and sample
Chortfs was not part of the Caribbean plate of today outcrops in the Montafias de Col6n-Rio Wampfi area
(Gose and Finch, 1992). Since mid-Cretaceous time near the border with Nicaragua (Fig. 1). This area
the Chortfs block has rotated counterclockwise into consists of three distinct physiographic elements: (1)
~, I CARIBBEAN SEA
MEXICO
\
"-,
i BELl
84" W
+ ,"-- _t + '' / -P 1 6 ~ N
/ L-- +
, r #" San Pedro Sula
" /~' ~,-oc.,,o i ~ ,,,,,,,,,, ,,. o.o HONDURAS ~ 2 13.,t

/"
r
. . = , o .... ,,
,t
k. Guateemlla G .r_ ~v,~. ~ *Ta~'n." /. " ~ t ",,,~ MQSQUITI&
+',,,.GUATEMALA /"
.9.Metaphn "*~
,4,
,
. . . . .o ./.-
~.'~~"
Co,t
.-.,~.
+~'J L ~1,.,,.._jp-~.- 1~. Tegulligalpa ::. .." ""
"---4 " ~ EL SALVADOR r Jr- +

?
PACIFIC OCEAN ~ 1,. ~ . . . ) NICARAGUA
0 50 100 200
3 ~- _ _ I
% km
- 4 - 12 ~ N
Fig. 1. Location map of selected quadrangles, geographic reference points; stippled pattern shows the approximate distribution of
Mesozoic strata in Olancho east of the Guayape fault (from Donnelly et al., 1990, figs. 5 and 6). The Montafias de Col6n area is northeast
of its name. Numbered quadrangles: ! = Confluencia Rfos Patuca y Wampfi quadrangle; 2 = Krausirpi quadrangle; 3 = Wampusirpi
quadrangle; 4 = Agalteca quadrangle.

CRETACEOUS CARBONATE BIOSTRATIGRAPHY AND ENVIRONMENTS IN HONDURAS 153
[ I I I I I I $ i ic.- I I I I 5~
: ~0 ~/ h Cerro
Rio Wampu ~/" /Asangbusna .
,~ oo,OO
X, ,4" ~ .
"~ Boca
~A~_._...- // Ie Wankibila Are~ ""~
"/" J /
_ _ . , .. VVdl i I pU / .. 9 9 ~,r~ ,,~
- ~ ,~-~, ............... /.; .: 9 9 9 ~@" ~" 55
. . . . . . . . . .. ... . . . . . . . 58
! /,:L-I9 " ' - .9. , "9 .o, 9e ~ 9 ~ 9
- ~f~g,r.. - . - ~ "~" M...., "..~A-3 ~ .o % "~ X2,~.'o ,-,Y ~"/ ~, --
9 9 ml,uJ.,,L'l.,,uummuun1,...unuukv '~*' ~ ,,. -,..=
9,,- - Tirisne r" A 10
_ .,r . . . . Montanas de Colon "~'~ HA-1 "'.l

l,- u 1 z Km ~ I
./..
6 ~"~., " , 20 J_ 25 219
"1 I I I I I I -I1- I I I I 50 ..-J
Fig. 2. Location map of key samples in the Rfo Sutawala area; base map from the Confluencia Rfos Patuca y Wamp6 quadrangle scale
1:50.000. Southwest map corner begins at UTM grid coordinates 16PGM150407.
a northeast-trending belt of folded and faulted lime- STRATIGRAPHY

stone mountains (the Montafias de Col6n), flanked
on the northwest by (2) a low terrain of clastic strata Lithostratigraphy
flooring the broad northeast-trending valley of the
Rio Patuca, flanked, in turn, to the northwest by (3) The Cretaceous stratigraphic terminology of Hon-
a mountainous mixed metamorphic/sedimentary ter- duras (Fig. 3) was revised by Finch (1981) and
rain drained by the Rio Wamp6, a southeast-flowing summarized by Donnelly et al. (1990). The mid-Cre-
tributary to the Patuca (Fig. 2). taceous platform carbonates are widely exposed
Paleontological results were obtained from a mea- across Honduras and have been designated the Yojoa
sured section representing a thin interval of the up- Group (Mills et al., 1967), consisting of the locally
permost part of the Atima Formation exposed at developed Cantarranas Formation overlain by the
Tirisne Cliffs and from other locations along the widespread Atima Formation (see Gose and Finch,
Sutawala Valley cutting across the Montafias de 1992, fig. 4, and Donnelly et al., 1990, fig. 7, for
Col6n (Fig. 2). Additional collections were made stratigraphic columns in west-central Honduras).
along the Rios Patuca and Wamp6. The Cantarranas Formation (Carpenter, 1954) is
Paleontological results from the present study a shaly limestone of variable lithology ranging in
demonstrate that Chortfs mid-Cretaceous carbonate thickness from 30 to 190 m. The unit is dated
strata were part of the same biogeographic province as Valanginian-Hauterivian by ammonites (Gordon
as coeval carbonate rocks of the Caribbean Province and Young, 1993) and represents a shelf lagoon.
in Jamaica, northern Mexico and the Gulf of Mex- Outside of its type area near Talanga northeast of
ico. Previous paleontological data from west-central Tegucigalpa, the Cantarranas has been found to be
Honduras (Mills et al., 1967; Lozej, 1976) suggested of limited value as a map unit at the 1:50,000
such a relationship. Likewise, outcrops of Atima For- scale used for geologic quadrangle mapping in Hon-
mation limestone reported by Emmet et al. (1992) duras, but has been mapped by Gordon (1990).
in central Honduras (some 50 km north-northeast We did not recognize the Cantarranas in the Rio
of Tegucigalpa in the Agalteca quadrangle, 4 in Wampfi-Montafias de Col6n area, nor did Rogers
Fig. 1) contain late Aptian to middle Albian planktic (1994, 1995), although Mills et al. (1967) reported
foraminifers, colomiellids, and calcispheres. These Cantarranas exposed on a tributary to the Wampfi.
forereef basinal fossils also demonstrate biogeo- The Atima Formation, named by Mills (1959), is
graphic similarities and exchange with coeval car- a widely recognized map unit consisting primarily of
bonate environments in Guatemala, Jamaica, Mexico thick-bedded micritic limestone ranging in thickness
and southern United States. from 90 to 1400 m (Finch, 1981). In the area of the

t-
...=
r-
E pper redbeds, conglomerate

o
t.- volcanics
8
Cenomanian =tique Formation

limestone FTC-3, P-l, 2, 4, A-11
UpperAlbian- wer redbeds

Lower Cenomanian
Krausirpi beds
shale A-9, HA-1
Albian ~.tima Formation T-1 to-29, A-3,-7.-4,-5
FSB-48, -47, -3, -4;
limestone FZ-2S.-32
FM-21, -33; FSB~100
Pre-Albian "Mochito Shale"
Honduras Group
sandstone, shale &
metasedimentary rocks
above basement rocks
,m
O
N
O Agua Fria Formation
(9
=E
Basement Rocks
igneous-metamorphic rocks
Fig. 3. Generalized, composited stratigraphic section of Mesozoic strata in eastern and western Honduras showing distribution of key
samples listed in Tables 1 and 2. Krausirpi beds found only in the eastern region and the Jaitique Formationonly in the western region.
Mochito Mine west of Lake Yojoa (Fig. 1) the Atima Mochito Mine near E1 Mochito (Fig. 1) and in the
is divided into informal lower and upper intervals area north and west of the mine (Finch, 1973, 1985).
separated by about 100 m of green and gray mud- This shale was described by Lozej (1976) as "of
stone with limestone interbeds and some red shale. 'red bed' affinity (somewhat similar lithologically
In the Mochito area Lozej (1976) dated the lower to the overlying and underlying formations)...". He
limestone of the Atima as upper Barremian through reported conformable boundaries with the lower and
Aptian by "possible 'early' Orbitolina" sp. and by upper Atima limestones "marked by green and/or
Tintinnopsella cf. carpathica (Murgeanu and Fil- red mottled mudrock interbeds". The transitional
ipescu), which in the Gulf Coast ranges from Berri- beds at the top of the lower Atima contain ostracodes
asian through Barremian. A rudistid facies is present and the base of the upper limestone contains charo-
about 210 m below the top of the lower limestone. phytes, gastropods and ostracodes (Lozej, 1976). He
The upper Atima was dated as latest Aptian through concluded that the Mochito was late Aptian by its
upper Albian (Lozej, 1976) by the presence of or- position between the Barremian-lower Aptian lower
bitolinids, Cuneolina, Coskinolina/Dictyoconus, and Atima limestone and the Albian upper Atima lime-
Dicyclina. An interval between the orbitolinid facies stone. At three widely separated exposures in central
contains the offshore shelf protozoan Colomiella Honduras, Gallo and Van Wagoner (1978) described
recta Bonet, which ranges from Barremian through sections 130 to 150 m thick of predominantly clas-
lower Albian in the Gulf Coast (Bonet, 1956). Rud- tic strata within the Atima, which they correlated
istid facies are developed about 150 m below the top with the Mochito shale. In the Agalteca quadrangle
of the upper limestone. (4 in Fig. 1), Emmet (1983a,b) also subdivided the
The shale and mudstone interval within the Atima Atima into upper and lower intervals separated by
Formation has been informally designated the "Mo- a shaly unit. It is not known if Emmet's un-named
chito shale", and comprises a mappable unit at the shale correlates with the Mochito shale, but it is

possible to infer that at least one major shale break et al., 1992; Rogers, 1994, 1995; Mills and Barton,
in carbonate deposition is relatively widespread in 1996). Ammonites from marine portions of the Agua
the Atima Formation of central Honduras. This shale Frfa indicate a Middle Jurassic age (Ritchie and
interval represents a regressive depositional phase, Finch, 1985; Gordon and Young, 1993).
which contrasts with a widespread transgressive ma- The Honduras Group underlies Lower to middle
rine shale break between mid-Cretaceous carbonates Cretaceous carbonates of the Yojoa Group with no
in northern Mexico and the U.S. Gulf Coast (Scott detected unconformity or in fault contact. If there is
et al., 1988). In the Montafias de Col6n area no Mo- no hiatus in deposition between the two groups, then
chito shale-equivalent has been identified (Rogers, deposition of the upper Honduras strata continued
1995). into the Early Cretaceous. However, in several areas,
For the present study, the upper Atima Forma- including the Rfo Wamp6 area, strata of the Agua
tion was sampled in the Montafias de Col6n in Frfa have undergone a weak dynamic metamorphism
a traverse along the Rfo Sutawala that drains a (Kozuch, 1989; Gordon, 1990; Rogers, 1994, 1995)
prominent valley transecting the folded and faulted that did not affect the Yojoa Group limestone, leav-
limestone mountains approximately perpendicular to ing open the possibility that a significant hiatus
stratigraphic and structural strike (Fig. 2). Along occurred between deposition of the Honduras Group
this traverse the strata dip southeastward 20 to 65 and the Yojoa Group. At the present time we do not
degrees and strike generally northeast-southwest. know if Honduras Group deposition continued into
Brownish-gray limestone is found near the conflu- the Early Cretaceous.
ence of the Rfo Sutawala with the Rfo Patuca, in In west-central Honduras carbonate strata of the
the Sutawala channel. Several kilometers upstream, Atima Formation are overlain, apparently gradation-
and about 100 m higher topographically, a section of ally, by red terrigenous clastic strata of the Valle
54.8 m of the upper part of the Atima was measured de Angeles Group, named by Carpenter (1954) for
and sampled bed by bed at the Tirisne Cliffs expo- exposures between Tegucigalpa and San Juancito
sure, which is 4.5 km east of the junction of Sutwala (Fig. 1). In the area west and south of Lake Yojoa
with Rfo Patuca directly south of the Sutawala Val- (Fig. 1) the Valle de Angeles redbeds are divided
ley trail in the Confluencia Rfos Patuca y Wamp6 into lower and upper intervals by the presence of
quadrangle (Fig. 1, quadr. 1; 16PGM253530; Fig. 2, a prominent limestone unit, the Jaitique Formation
Table 1). Farther up the Sutawala valley from this (Finch, 1981). Somewhat to the east of Lake Yojoa
measured section, shaly strata of the 'Krausirpi and north of Tegucigalpa, the Esqufas Formation
beds' are exposed, apparently in sedimentary contact (Home et al., 1974) divides the Valle de Angeles.
overlying the upper Atima (Rogers, 1994, 1995). The Jaitique Formation includes two units, an un-
Additional collections were made northeast of Rfo named cliff-forming limestone member, overlain by
Wamp6 in the Krausirpi and Wampusirpi quadran- a thin section of distinctive petroliferous limestone
gles (Fig. 1, quadr. 2, 3). designated the Guare Member (Finch, 1981). The
The base of the Yojoa Group is not exposed in the lower member is about 100-150 m thick, comprised
Montafias de Col6n. However, wherever the basal of thick-bedded shelfal limestone, well-dated pale-
contact can be seen, the limestone overlies siliciclas- ontologically as Cenomanian. No unconformity was
tic strata now generally assigned to the Honduras seen in the field at the base of the Jaitique, and
Group (Ritchie and Finch, 1985; Finch and Ritchie, paleomagnetic data suggest that the contact between
1990; Donnelly et al., 1990). The Honduras Group the basal Jaitique and the underlying redbeds of
includes: (1) strata mapped as E1 Plan Formation the lower Valle de Angeles Group is conformable
near Tegucigalpa (Carpenter, 1954), (2) unnamed (W.A. Gose, in Finch, 1981). The Guare is about
siliciclastic beds (Simonson, 1977, 1981), 3) strata 10 m thick, consisting of thin- to medium-bedded,
formerly mapped on the Chortfs block as Todos flaggy weathering, oil-stained, laminated, dark-gray
Santos Formation (Finch, 1985; Ritchie and Finch, to black limestone with black shale interbeds. The
1985; Emmet et al., 1992; Rogers, 1995), and (4) a Guare is in turn overlain by laminated gypsum that
thick succession of dark marine shale, marine and grades upwards into fine-grained red clastic strata of
non-marine sandstone and conglomerate designated the upper Valle de Angeles Group (Finch, 1981).
the Agua Frfa Formation in eastern and southeastern Farther east, in central Honduras near the village
Honduras. Esqufas (Fig. 1) marly and argillaceous limestone
The Agua Frfa Formation was named by Ritchie of the Esqufas Formation (Weaver, 1942; Atwood,
and Finch (1985) after a long section exposed near 1972; Home et al., 1974; Finch, 1981) divides
Danlf, which was first described by Roberts and the redbeds. Like the Jaitique Formation, the Es-
Irving (1957). Agua Frfa strata have been widely qufas apparently is in conformable contact with the
mapped in eastern Honduras (Kozuch, 1989, 1991; enclosing redbeds (Home et al., 1974). The age
Gordon, 1992; Finch and Ritchie, 1990; Weiland of the unit is not well defined, paleontologically,

156 R.W. S C O T T and R.C. F I N C H
Table 1
Age-diagnostic taxa in samples from the Tirisne Cliffs measured section, Honduras, along cliffs directly south of the Sutawala valley trail
and about 5 km east-southeast of the Rfo Patuca
Field No.: T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 Tll T12 T13 T14

Cum. thickn. (m): 0-1.5 3.6 3.9 4.5 5.4 6.3 7.9 8.9 9.5 10.0 10.6 13.4 1 6 . 1 17.8
Barkerina barkerensis R?
Cuneolina walteri R?
Globigerina delrioensis R? R R R R? R R? R?
Globigerina planispira? R
Globochaeta alpina R R R
Lenticulina sp. R R R R
Heterohelix globulosa R F R F F R F R R R
Nezzazata simplex
Nodosaria sp. R R R
Praeg lob igerina delrioensis R R R R? R
Pseudocyclammina hebergi R? R? R? F R F R R
Radiolaria R
Sponge spicules A Oc F F F R Oc Oc
Micritosphaera ovalis
Pithonella ovalis R
Pithonella sphaerica
Saccomma sp. R R R R R R
Indet. calc. algae R
Indet. caprinids
Field No." T15 T16 T17 T18 T19 T20 T21 T22 T23 T24 T25 T26 T27 T28 T29
Cum. thickn. (m)" 20.6 22.0 22.4 24.3 27.8 29.1 31.8 33.4 35.7 39.9 41.8 44.8 50.0 52.1 54.8
Barkerina barkerensis
Cuneolina walteri R? R? R? R? R? R R F
Globigerina delrioensis R R? R?
Globigerina planispira?
Globochaeta alpina R R R R
Lenticulina sp. R R R
Heterohelix globulosa R R R R R R
Nezzazata simplex R? R F
Nodosaria sp. R R
Praeglobigerina delrioensis R
Pseudocyclammina hebergi F R? R? R
Radiolaria
Sponge spicules F F Oc R R R
Micritosphaera ovalis R R R R R
Pithonella ovalis
Pithonella sphaerica R R
Polystrata alba R?
Saccomma sp. R R R R R
Indet. calc. algae R R R
Indet. caprinids Oc
Indet. radiolitids? R? 9
Long. 84~ lat. 14o 56 ! 35 I! N. Collected by T. Weiland, measured from lowest bed exposed at west edge of cliff face. Relative
abundance scheme: R = rare, F = few, Oc = occasional, A -- abundant.
but a Cenomanian-Turonian age is suggested, and Formation of mainly Albian age, and below the Jai-
its paleopole position indicates that the Esqufas is tique of Cenomanian age or the Esqufas of upper
somewhat younger than the well-dated Cenomanian Cenomanian or Turonian age. The age of the lower
Jaitique Formation (Gose and Finch, 1992) Valle de Angeles redbeds are, therefore constrained
Although the Jaitique and Esqufas formations dif- to upper Albian to Lower Cenomanian. Thus, in
fer in lithology and somewhat in age, where present, western and central Honduras, Yojoa carbonate de-
each conveniently divides the thick Valle de Angeles position appears to have ceased toward the close of
Group into upper and lower redbed intervals, and the Albian and to have been superseded by redbed
provide some constraint on the age of the lower clastic deposition by early Cenomanian time.
redbeds. The lower Valle de Angeles redbeds lie In west-central Honduras the upper contact of the
above the paleontologically well-dated upper Atima Valle de Angeles Group is an angular unconformity,

with the redbeds overlain by middle Tertiary vol- glomerate, with the clasts embedded within a redbed
canic rocks of the Matagalpa Formation or by the matrix. These limestone conglomerates, commonly
Miocene-Pliocene Padre Miguel Group (Williams designated the Ilama Formation (Mills et al., 1967;
and McBirney, 1969; Everett, 1970; Finch, 1972, Southernwood, 1986; Mills and Barton, 1996) are
1981; Emmet, 1983b). It is not known if redbed found throughout Honduras. Available paleontologi-
sedimentation continued into the early Tertiary. cal evidence indicates that the limestone clasts were
Neither the Jaitique nor the Esqufas Formation derived from the Atima Formation and the con-
has been found within the Valle de Angeles Group glomerates certainly indicate widespread erosion of
in the Rfo Wampfi-R/o Patuca area (Weiland et al., Yojoa Group carbonates during Valle de Angeles
1992; Rogers, 1994, 1995). Instead, in the Mon- time. However, there is no good evidence that lime-
tafias de Col6n region of easternmost Honduras, stone conglomerate production was coeval across
the upper Albian Atima Formation is overlain ap- Honduras, or that the various conglomerate expo-
parently conformably by non-red, olive tan shale sures should all be assigned to one formation. In-
and silty, shallow marine strata. These strata were deed, Southernwood's stratigraphic correlation chart
designated the 'Krausirpi beds' by Rogers (1994, (Southernwood, 1986, plate 4) indicates a wide va-
1995), who reported the unit to include gray and riety of stratigraphic positions for the conglomerates
tan shale, arkosic sandstone, graywacke, and minor of the 'Ilama Formation', as would be expected for
conglomerate. Rare wood fragments were observed such coarse, fluvial deposits, which possibly were
in clastic beds. However, the shale and sandstone deposited as alluvial fans shed from fault blocks.
beds are calcareous, and thin limestone beds oc- 'Ilama-type' limestone conglomerate deposits oc-
cur within the unit. In thin section one sample is cur in the valley of the Rio Patuca, along the north-
a silty planktic foram lime mudstone. A sample west flank of the Montafias de Col6n, within redbeds
collected from the Krausirpi beds by Finch and Wei- assigned to the Valle de Angeles Group (Rogers,
land contained a latest Albian to Early Cenomanian 1994, 1995; Mills and Barton, 1996). Limestone
biota. At the village of Krausirpi (Fig. 2), Rogers clasts in these conglomerates sampled by Finch
found an erosional surface developed on top of the and Weiland yielded Cretaceous microfossils and
Krausirpi beds with the Valle de Angeles redbeds megafossils that suggest that most clasts were de-
overlying them in a slightly angular unconformity. rived from the Atima Formation. A single sample
Where the Krausirpi beds are missing, Valle de An- contains a dasyclad alga known only in Cenomanian
geles redbeds directly and unconformably overlie rocks elsewhere in the Tethys suggesting that Ceno-
the Atima (Rogers, 1995). South of Cerro Wampfi manian limestone was a local source in the Montafias
(Fig. 2) karstic features are developed at the top of de Col6n, although no in-situ outcrops are mapped.
the Atima (Rogers, 1995). Rogers suspects that the The Krausirpi beds contain local limestone beds that
Atima was subaerially exposed prior to deposition may be of Cenomanian age and could have been a
of the Valle de Angeles in this region (R.D. Rogers, source. Rogers (1994, 1995) has demonstrated that
pers. commun., 1996). Thus, in the Montafias de the volcanic units dated by Weiland et al. (1992; K -
Col6n region, marine deposition continued into the Ar dates from unaltered plagioclase separations) at
Cenomanian, and deposition of the Valle de Angeles 80-70 Ma are interbedded with the Valle de Angeles
did not commence until latest Cretaceous. redbeds. Thus, the Valle de Angeles redbed sedimen-
Although Rogers (1995) noted that it is "possible tation continued at least into the latest Cretaceous
that the Krausirpi beds are a local unit of the Yojoa in easternmost Honduras, but at present there is, as
Group", he concluded that they "should not be in west-central Honduras, no evidence to show that
correlated with other mapped units in Honduras". redbed deposition continued as late as Tertiary time.
However, we feel that, in spite of their local nature,
it is appropriate to include the Krausirpi beds in Biostratigraphy
the Yojoa Group as an informal member because
of their marine character, the absence of redbeds, The Atima Formation in the Montafias de Col6n
their conformable contact with the Atima, and their region is as young as upper Albian. The age of the
unconformable contact with the Valle de Angeles. 55-m-thick section of the uppermost Atima Forma-
Within the Valle de Angeles Group throughout tion at Tirisne Cliffs section is upper Albian based
Honduras, conglomerates of various compositions on the co-occurrence of Cuneolina walteri Cushman
form a significant portion of the unit. In west-central and Applin (Fig. 4A-C), Pseudocyclammina hed-
Honduras the lower Valle de Angeles section is more bergi Maync (Fig. 4D,E), and Praeglobotruncana
conglomeratic than the upper, with quartz pebble delrioensis (Plummer). The first two species occur
conglomerate beds being prominent ledge-formers in in middle and upper Albian carbonates in northwest-
outcrop. However, the most distinctive conglomerate ern Mexico (Scott and Gonzales-Leon, 1991) and
units consist of limestone pebble- to boulder-con- elsewhere in the Gulf Coast (Conkin and Conkin,


1958). The third species, a planktic foraminifer, 1990) contain key taxa of rudists and foraminifers
ranges from late Albian to late Cenomanian (Bolli that confirm the age of the Atima to range from
et al., 1985). Samples from the limestone section in Barremian-lower Aptian to upper Albian. Poorly
the Rfo Sutawala stratigraphically below the Tirisne preserved rudist specimens from a caprinid pack-
Cliffs section yield Pseudonummoloculina [Num- stone in the upper 50 m of the Atima Formation
moloculina] heimi (Bonet) (Fig. 4F), which ranges are close to either Kimbleia sp. (late Albian), or
through Albian and lower Cenomanian rocks in the Caprinula sp. (Albian-Turonian) (Table 2). Other
Gulf Coast region (Scott and Gonzales-Leon, 1991). samples of the Atima contain the Barremian-early
The genetic reassignment of this species was made Aptian foraminifer Choffatella decipiens Schlum-
by Hottinger et al. (1989) because the apertural berger with Pseudocyclammina hedbergi, the bivalve
structure differs from the Neogene genus Nummolo- Chondrodonta sp., and the alga Lithocodium aggre-
culina. gatum Elliott. All of these taxa are common in
From west-central Honduras we can demon- theGulf of Mexico part of the Caribbean Province.
strate the presence of the key benthic foraminifer The post-Atima Krausirpi beds exposed in the
Orbitolina (Mesorbitolina) subconcava Leymerie bottom of Rfo Sutawala are represented by two
(Fig. 5D,E) in the Atima Formation in several samples. One yielded late Albian-early Cenomanian
places, which probably was the more advanced type palynomorphs identified by R.W. Aurisano (written
reported by Lozej (1976). This species ranges from commun., 1985): Reticulisporitesjardinus, Xenascus
the late Aptian to the late Albian and is reported in ceratioides, Dinopterygium sp., Diconodinium sp.,
lower Albian strata of Texas and Mexico (Schroeder Oligosphaeridium sp., Classopollis sp., Coronifera
and Neumann, 1985). This species occurs in the up- sp., Scriniodinium sp., Trichodinium sp., and tri-
permost Atima a few meters below its contact with colpate pollen. A thin section of a second sample
the Valle de Angeles Formation southwest of Lake contains Globigerinoides cushmani and Heterohelix
Yojoa (in the Montafia de Santa B~irbara, Table 2). globulosa, which together indicate a latest Albian to
This suggests that the upper Atima is not younger early Cenomanian age for the Krausirpi beds, show-
than upper Albian. The species also occurs in a ing that shallow marine sedimentation continued into
limestone clast within a conglomerate in the Valle de the Cenomanian in the eastern part of the Chortfs
Angeles along the Rfo Patuca (Fig. 2). block.
Other benthic foraminifers support the lower to Limestone conglomerate in the Valle de Ange-
upper Albian age of the upper Atima limestone. les Group exposed along the Rfo Patuca in the
Coskinolinoides sp. cf. C. texanus Keijzer is found Krausirpi and Wampusirpi quadrangles also contains
southwest and west of E1 Mochito near Lake Yojoa late Albian fossils (Table 2). Important taxa are Or-
(in the San Pedro Zacapa and the Santa Bfirbara bitolina subconcava, Cuneolina walteri, Pseudonum-
quadrangles, Table 2). This species occurs in the moloculina [Nummoloculina] heimi, Pseudocyclam-
middle to lower-upper Albian Fredericksburg Group mina hedbergi, Favusella [Globigerina] washitensis
in central Texas and unnamed species of this genus (Carsey), and Globigerinoides cushmani, the caprinid
range up into the upper Albian part of the Washita Mexicaprina sp., and calcareous algae, Polystrata
Group (Coogan, 1977). Praechrysalidina infracre- alba (Pfender) and Parachaetetes texana Johnson
tacea Luperto Sinni, which we found in the Atima (Fig. 5A). Evidently these cobbles were eroded from
with Orbitolina texana (Roemer) and Cuneolina the Atima Formation
walteri southwest of Lake Yojoa (in the Santa Bfir- At Boca Wampfi a conglomerate with a reddish
bara quadrangle), is a Hauterivian-Albian species brown sandy matrix and cobbles of red siltstone
in the Middle East (Banner et al., 1991). Another and well rounded, light-gray limestone was identi-
Albian taxon in the upper Atima in this area is Coski- fied in the Valle de Angeles Group, although the
nolinella sp., which is well illustrated by Schroeder limestone cobbles are lighter colored than most of
and Neumann (1985). the Atima Formation and they contain miliolids, tex-
Outcrop grab samples from the Olancho district tulariids, and bivalves. A limestone clast from this
(Fig. 1, in the area indicated by the name Olan- conglomerate yields the calcareous dasyclad alga,
cho) collected by M.B. Gordon (pers. commun., Dissocladella undulata (Raineri) (Fig. 4G), which is
Fig. 4. (A-G) Photomicrographs of key fossils found in the Atima Formation in eastern Honduras and the Jaitique Limestone in central
Honduras. (A-C) Cuneolina walteri Cushman and Applin (1947); (A) proloculus, sample A-4, 75 x; (B) transverse section, sample A-4,
75• (C) longitudinal section, sample T-29, 47x. (D-E) Pseudocyclammina hedbergi Maync (1953); (D) axial view, sample T-7, 30x;
(E) longitudinal view, sample T-7, 30x. (F) Pseudonummoloculina [Nummoloculina] heimi (Bonet, 1956), emend. Conkin and Conkin
(1958), generic reassignment by Hottinger et al. (1989), sample A-4, 75x. This specimen has flatter chambers than is typical of P.
heimi. (G) Dissocladella undulata (Raineri, 1922; Pia, 1936), sample A-11, Valle de Angeles Group, 55x. (H) Biconcava n. sp., Jaitique
Limestone, sample FTC-3, 60•


Table 2
Location and description of key samples collected in Honduras by R.C. Finch during several field seasons (locations given using the
Universal Transverse Mercator Grid system)
Krausirpi quadrangle
P-1 Valle de Angeles Gp. at 16PGM257617, Krautara on Rfo Patuca; cobble conglomerate of light-gray limestone and
lesser amounts of volcaniclastics, quartz, and quartzite of silicified siltstone in a matrix of reddish sandstone. Fossils in
limestone clasts: Orbitolina subconcava, Nummoloculina heimi, Cuneolina walteri, Pseudocyclammina hedbergi,
Parachatetes texana, Lithocodium aggregatum.
P-2 Valle de Angeles Gp. at 16PGM346681, Pimienta on Rfo Patuca; 'Ilama' limestone conglomerate as above. Fossils in
limestone clasts: Mexicaprina sp., Lithocodium aggregatum.
Wampusirpi quadrangle
P-4 Valle de Angeles Gp. at 16PGM523723, Walpatanta on Rfo Patuca; 'Ilama' limestone conglomerate as above. Fossils
in limestone clasts: Globigerina delrioensis, Favusella washitensis, Globigerellinoides cushmani, Parachaetetes texana,
Polystrata alba, Lithocodium aggregatum.
Confluencia Rfos Patuca y Wampft quadrangle
A series Atima Fm. collected by core; all are limestone except A-9 which is shale of the Krausirpi beds and A-11 which is
conglomerate in Valle de Angeles Gp.; A-l: 16PGM221544; A-2: 16PGM221543; A-3: 16PGM223540; A-4:
16PGM234539; A-6: 16PGM227540; A-7: 16PGM231542; A-8: 16PGM247533; A-9: 16PGM262529; A-10:
16PGM279520; A-11: 16PGMI74547.
T series Atima Fm. collected in measured section at Tirisne Cliffs by T. Weiland; T-1 (base) through T-29 (top) at
16PGM253530, 84054, 19"W, 14~ HA-locality in Rfo Sutawala estimated at 16PGM289512.
Santa B6rbara quadr. (west side of Lake Yojoa in Fig. 1)
FSB-3 Uppermost Atima Fm. no more than a few tens of meters below contact with Valle de Angeles Fm. at 16PCM711409.
Orbitolina subconcava.
FSB-4 Uppermost Atima Fm., no more than a few tens of meters below contact with Valle de Angeles Fm. in a quarry near
16PCM844433. Coskinolinella sp.
FSB-46 Upper Atima Fm., very probably the upper part on top of Montafia Santa B~irbara near 16PCM788454. Coskinolinella
texanus, Praechrysalidina infracretacea.
FSB-47 Upper Atima Fm., probably the upper part, from near 16PCM828419 in the Mochito graben. Coskinolinella texanus.
FSB-100 Atima Fm., low on the southwest flank of Montafia Santa B~irbara at about 16PCM717535 apparently above the
'Mochito shale'. Orbitoliina texana, Praechrysalidina infracretacea.
FM-21 Upper Atima Fm., possibly lower part from inside the Mochito Mine at level 1350, cross-cut 9444E. Praechrysalidina
infracretacea (16PCM9394230).
FM-33 Upper Atima Fm., lower part near contact with 'Mochito shale' inside Mochito Mine at level 850, cross-cut 992.
Colomiella coahuilensis (16PCM939423).
Taulab# quadr. (southeast of Lake Yojoa in Fig. 1)
FTC-3 Uppermost part of lower, unnamed member of Jaitique Fm. directly below basal contact of Guare Mbr. at
16PCM998218 near Carrizal-La Mision. See fig. 4 in Finch (1981) for stratigraphic position. Specimens of Biconcava
n. sp. in thin section.
San Pedro Zacapa quadr. (southwest of Lake Yojoa in Fig. 1)
FZ-28, 32 Uppermost Atima Fm. near 16PCM870343; FZ-32 is approximately 50 m above FZ-28. Coskinolinella texanus.
characteristic of the C e n o m a n i a n in the Tethys (Bas- T h e rudist M e x i c a p r i n a sp. is r e p o r t e d f r o m H o n -

soullet et al., 1978). This is the first report of this duras for the first time. This g e n u s is charact eris tic
species in the C a r i b b e a n P r o v i n c e . This l i m e s t o n e of u p p e r A l b i a n rocks in M e x i c o and the U.S. T h e
clast m a y have b e e n d e r i v e d f r o m one of the C e n o - s p e c i m e n was f o u n d in a c o n g l o m e r a t e within the
m a n i a n l i m e s t o n e units within the Valle de A n g e l e s , Valle de A n g e l e s G r o u p and is c o m p l e t e l y recrys-
or f r o m a l i m e s t o n e b e d within the K r a u s i r pi beds. tallized so that internal structures are not visible
If this is the case, it is the first t e n u o u s e v i d e n c e (Fig. 5F). This s p e c i m e n has the distinctive ridges
that C e n o m a n i a n l i m e s t o n e d e p o s i t i o n m a y h a v e at the corners similar to M e x i c a p r i n a sp. illustrated
e x t e n d e d this far east. To date, no l i m e s t o n e unit has by C o o g a n (1977, pl. 17, figs. 8), w h i c h is f r o m
b e e n m a p p e d within the Valle de A n g e l e s r e d b e d s the E1 A b r a F o r m a t i o n at Taninul quarry in M e x i c o .
in the W a m p d - P a t u c a r e g i o n ( W e i l a n d et al., 1992; T h e H o n d u r a n s p e c i m e n , h o w e v e r , is a b o u t 10 m m
R o g e r s , 1995). across and the M e x i c a n s p e c i m e n is a b o u t 23 m m
Fig. 5. Photomicrographs of fossils from limestone clasts in conglomerates of the Valle des Angeles Group (A, D, E, F) in eastern
Honduras and the Jaitique Limestone (B, C) in west-central Honduras. (A) Parachaetetes texana Johnson, sample P-4, 55• (B, C)
Biconcava n. sp, sample FTC-3, sagittal and axial sections, 151 x. (D, E) Orbitolina (Mesorbitolina) subconcava Leymerie, sample P-1,
47 x. (F) Mexicaprina sp., sample P-2, transverse section, 7.4 x.

in longest diameter. A similar species, Mexicaprina noid wackestone with pycnodont oysters; red shale
minuta Coogan, is 8 to 11 mm across, but it does with plant debris (probably not Atima); silty peloid-
not have the large ridges. The Honduran specimen ostracode packstone.
may be a new species of this very specialized genus. Farther upstream in the Sutawala Valley, at
Coogan (1973) described the genus from the E1Abra Tirisne Cliffs, 55 m of upper Atima Formation were
Formation and from the Stuart City Formation in measured and closely sampled. This section records
the subsurface of Texas (Coogan, 1977, pl. 17, fig. a shoaling upwards facies succession (Fig. 6). Two
5). The age of the Stuart City is middle to lower- major microfacies were collected: a lower peloid-
upper Albian by its foraminifers and rudists (Scott, foram-spicule wackestone/packstone and an upper
1990). The age of the E1 Abra is less clear; Coogan peloid-foram-bivalve wackestone/grainstone. These
(1973, 1977) placed it in the Cenomanian based on facies indicate an environmental change from the
the radiolitid rudists, Pecten roemeri and Parkeria middle shelf to the inner shelf. Overall, taxo-
sphaerica, and the absence of characteristic Albian nomic diversity decreases slightly near the top of
taxa. The age of the oldest radiolitids is indefinite; the section where carbonate sands were more com-
Pecten roemeri is found in lower Cenomanian rocks mon. The upwards shoaling is indicated by the
of north Texas, and Parkeria sphaerica is originally decrease in number of planktic taxa, the decrease in
known from the uppermost Albian beds in England sponge spicules, and the replacement of Lenticulina
and its Cenomanian attribution comes from its oc- sp. and Nodosaria sp. by shallow-water benthic
currence in the E1 Abra (Dieni and Turn~ek, 1979). foraminifers. Bottom waters were well oxygenated
Recently, Mexicaprina minuta and Mexicaprina cor- and water energy increased up-section where grain-
nuta have been found with the late Albian rudists stone is more common.
Kimbleia and Caprinuloidea in northern Mexico The post-Atima Krausirpi beds are poorly ex-
(Alba and L6pez-Casillas, 1993). Therefore, the age posed. This unit represents an influx of terrige-
significance of Mexicaprina spp. is late Albian. nous sediments into a shallow marine environment
The Jaitique Formation has yielded specimens terminating the Atima carbonate platform. Bottom
of the foraminifera Biconcava n. sp. (Fig. 4H, waters may have been poorly oxygenated. The de-
Fig. 5B,C), which is reported from the Caribbean positional environment may have been prodeltaic
Province for the first time. The sample was collected (Rogers, 1994, 1995).
southeast of Lake Yojoa in the Taulab6 area from The limestone conglomerates within the Valle
the uppermost part of the lower, unnamed member de Angeles Group along the Rfo Patuca in the
directly below basal contact of Guare Member (fig. Krausirpi and Wampusirpi quadrangles were de-
4 in Finch, 1981). These specimens possess the posited in non-marine to nearshore environments.
major features of the genus as defined by Hamoui The limestone clasts were derived mainly from the
and Saint-Marc (1970): the planispiral test is slightly Atima Formation, which was deposited in a shallow-
involute and expands gradually; the proloculus is water, moderate- to high-energy environment with
simple, globular; chamber cross-section is lunate but rudists. The common facies are rudist packstone,
the periphery of these specimens is more convex intraclast-peloid grainstone and rudist-peloid grain-
than in the genotype, Biconcava bentori Hamoui stone. Some clasts represent quiet water lagoonal or
and Saint-Marc (1970). The wall of both species shelf environments with foram wackestone. One
is microgranular calcite. The aperture appears to be clast contains radiolarian-planktic foram wacke-
a simple pore in the septal face (Fig. 5B, third- stone in contact with peloid grainstone turbidite.
last septa in final whorl). B. bentori ranges through So the cobbles represent a suite of environments
the middle and upper Cenomanian in its type area ranging from the shelf margin to the slope.
of Lebanon and Israel, but similar specimens have A major shift in depositional regime from car-
been reported in upper Albian and Senonian strata bonate shelf to terrigenous shelf, possibly prodeltaic,
(Schroeder and Neumann, 1985), so the range of the in eastern Honduras is recorded by the termination
genus is yet to be defined precisely. of Atima deposition by an influx of terrigenous
sediments of the lower Cenomanian Krausirpi beds.
Water depths were up to approximately 50 m to ac-
PALEOENVIRoNMENTS commodate the planktic foraminifers and dinoflagel-
lates. Deposition of the Krausirpi beds was followed
The upper Atima Formation exposed near the by a period of exposure, during which an erosional
mouth of the Rfo Sutawala records a transgressive unconformity developed that cut out part of the
succession. The beds dip southeast and the sam- Krausirpi beds and, locally, exposed the top of the
ples were collected from outcrops in the valley Atima Formation. This exposure event on the eastern
floor. Top: rudist-peloid packstone; rudist pack- part of the Chortfs block may correspond with one
stone-wackestone; quartz-peloid packstone; echi- of several middle Cenomanian sequence boundaries

M
55 W
"=''I I I I I I I I I I I I I I I
UJZ
",2 I=6.
< ,,-I,
50-
~,,,, U.I
O Z
u.O
g
OW
- S
40" ~3
Pseudoc~
_ .,.~edber!
- ~..
7
N "n oer"
ILl
20'
w
!11 z e~
! g ,,
,,,
UJ
10 ="1 IJ.l
-Praeglob
- delrio ,PBarkerina ~ ~ "
-j ,
q~
~ w
~ v
,barkerensis ~
0
5 lO
, 15
i o' ' ' ' '5'0' o ' ~ ' ' A' 0 .... ' A
'
BASEOF NO. TAXA % PLANKTIC SPICULES BENTHIC DEPOSITIONAL
EXPOSURE TAXA FORAMS ENVIRONMENT
Fig. 6. Paleoenvironmental interpretation of the Atima Formation at the Tirisne Cliffs section, Sutwala valley, Rio Wampd, Honduras.
Relative abundance ranges from absent (0), to rare, few, occasional, common, and abundant (A).
(Haq et al., 1988); the unconformity at the lower- ported in Honduras: Mexicaprina sp., Kimbleia sp.
middle Cenomanian boundary is widespread in the (late Albian), or Caprinula sp. (Albian-Turonian),
U.S. Gulf Coast and Arabian shelf and platform and Praeglobotruncana delrioensis. The presence of
sections (Scott et al., 1988). Deposition of redbeds Choffatella decipiens with Pseudocyclammina hed-
of the Valle de Angeles Group buried this expo- bergi on the western margin of the Montafias de
sure surface, which, locally, overlies karstic Atima Col6n indicates that the lowermost part of the Atima
(Rogers, 1995). In west-central Honduras a similar, crops out there.
but gradual environmental shift deposited conglom- In west-central Honduras the upper Aptian to
eratic redbeds with Atima limestone clasts prior to upper Albian Orbitolina subconcava is documented
deposition of the Jaitique and Esqufas formations. for the first time, as is Coskinolinoides sp. cf.
However, in the area south and west of Lake Yojoa C. texanus, Orbitolina texana, Cuneolina walteri,
some occurrences of limestone conglomerate con- Coskinolinella, and Coskinolinella sp.
taining abundant light-gray limestone clasts that do The Atima Formation in eastern Honduras is
not look like typical Atima limestone were mapped overlain conformably by a marine shale, the
by Finch (1972, 1985) as within the upper Valle de Krausirpi beds (Rogers, 1994, 1995), that contain
Angeles. So a second, younger local erosional event late Albian to early Cenomanian dinoflagellates and
may be recorded in this area. planktic foraminifers. The limestone conglomerates
in the Valle de Angeles Group were derived mainly
from the Atima; however, some may have been
SUMMARY derived from a Cenomanian limestone unit, as sug-
gested by the presence of the dasyclad alga Disso-
New collections from Cretaceous carbonate strata cladella undulata, which is reported here from the
in eastern Honduras yield taxa new to the area Caribbean Province for the first time.
that confirm the upper Albian age of the up- Accepting the placement of Chortfs near the
per part of the Atima Formation. Some of these Guerrero block on the Pacific side of Mexico dur-
taxa have been reported to the generic level, but ing the mid-Jurassic (Anderson and Schmidt, 1983;
not illustrated nor identified to species: Cuneolina Dengo, 1985; Pindell and Barrett, 1990), the biota
walteri (Fig. 4A-C), Pseudocyclammina hedbergi reported here confirm that by Aptian and Albian
(Fig. 4D,E), and Pseudonummoloculina [Nummolo- time, the Chortfs block had moved into biogeo-
culina] heimi. Some have never before been re- graphic connection with the fauna of the Caribbean

P r o v i n c e . A n e a s t w a r d c o m p o n e n t o f t r a n s l a t i o n is Region. The Geology of North America, Vol. H, Geological

i m p l i e d , likely s o u t h e a s t w a r d a l o n g a structure like Society of America, pp. 37-76.
the A c a p u l c o - G u a t e m a l a megashear of Anderson Emmet, P.A., 1983a. Mapa geol6gico de Honduras, cuadningulo
de Agalteca, escala 1:50.000. Tegucigalpa, D.C., Instituto
and S c h m i d t (1983).
Geognifico Nacional, 1 sheet.
Emmet, P.A., 1983b. Geology of the Agalteca Quadrangle, Hon-
duras, Central America. Unpubl. MS thesis, University of
REFERENCES Texas at Austin, 210 pp.
Emmet, P.A., White, R.J. and Curry, R.P., 1992. Measured strati-
Alba, J.A. and L6pez-Casillas, 1993. Mid-Cretaceous rudists graphic sections, biostratigraphy and sequence stratigraphy of
(Bivalvia-Hippuritacea) from Durango, north-central Mexico. Mesozoic strata in the Agalteca Quadrangle, Honduras. Geo-
Proc. 3rd Int. Conf. Rudists, UNAM, Inst. de Geologfa, pp. logical Society of America, South-Central Section, Abstracts
2-3. with Programs, p. 10.
Anderson, T.H. and Schmidt, V.A., 1983. The evolution of Mid- Everett, J.R., 1970, Geology of the Comayagua quadrangle, Hon-
dle America and the Gulf of Mexico-Caribbean Sea region duras, Central America. Unpubl. PhD. dissertation, University
during Mesozoic time. Geol. Soc. Am. Bull. 94: 941-966. of Texas at Austin, 152 pp.
Atwood, M.G., 1972. Geology of the Minas de Oro Quadrangle, Finch, R.C., 1972. Geology of the San Pedro Zacapa Quadran-
Honduras, Central America. Unpubl. M.A. thesis, Wesleyan gle, Honduras, Central America. Unpubl. PhD. dissertation,
University, Middletown, CT, 88 p. University of Texas at Austin, 238 pp.
Azema, J., Biju-Duval, B., Bizon, J.J, Carfantan, J.C., Mas- Finch, R.C., 1973. Mochito number 1 exploration 1973. Unpubl.
clr, A.J. and Tardy, M., 1985. Le Honduras (Am6rique report, Rosario Resources Corp., E1 Mochito, Honduras, 28
centrale nucl6aire) et le bloc d'Oaxaxa (Sud du Mexique): PP.
deux ensembles comparables du continent Nord-Am6ricain s6- Finch, R.C., 1981. Mesozoic stratigraphy of central Honduras.
par6s par le jeu d6crochant s6estre des failles du syst~me Am. Assoc. Pet. Geol., Bull., 65: 1320-1333.
Polochic-Motagua. Symp. Geodynamique des Caribes, Paris, Finch, R.C., 1985. Mapa Geol6gico de Honduras, Cuadningulo
5-8 Fevrier 1985. Editions Technipp, Paris, pp. 427-438. de Santa B~irbara, escala 1" 50.000. Instituto Geognifico Na-
Banner, ET., Simmons, M.D. and J.E. Whittaker, 1991. The cional, Tegucigalpa, D.C., Honduras, 1 sheet.
Mesozoic Chrysalidinidae (Foraminifera, Textulariacea) of the Finch, R.C. and Dengo, G., 1990, NOAM-CARIB plate bound-
Middle East: the Redmond (Aramco) taxa and their relatives. ary in Guatemala: a Cretaceous suture zone reactivated as
Bull. Br. Mus. Nat. Hist. (Geol.), 47: 101-152. a Neogene transform fault. Geological Society of America
Bassoullet, J.P., Bernier, P., Conrad, M.A., Deloffre, R. and Annual Meeting, Field Trip No. 17 Guidebook, 46 pp.
Jaffrezo, M., 1978. Les algues dasycladales du Jurassique et Finch, R.C. and Ritchie, A.W., 1990. Mapa Geol6gico de Hon-
du Cr6tac6. Geobios, M6m. Sp6c., 2, 330 pp., 40 pl. duras, Cuadr~ingulo de Danlf, escala 1:50.000. Instituto Ge-
Bolli, H.M., Saunders, J.B. and Perch-Nielsen, K., 1985. Plank- ognifico Nacional, Tegucigalpa, D.C., Honduras, 1 sheet.
ton Stratigraphy. Cambridge Univ. Press, London, 1032 pp. Fourcade, E. and Michaud, E, 1987. L'ouverture de l'Atlantique
Bonet, E, 1956. Zonifacaci6n microfaunistica de las calizas et son influence sur les peuplements des grands foraminif~res
cret~icicas del este de M6xico. Asoc. Mex. Ge61. Petrol. Bol., de plates-formes p6ri-oc6aniques au M6sozoique. Geodin.
8: 389-488. Acta (Paris), 1 (4/5): 247-262.
Burkart, Burke, 1994. Northern Central America. In: S.K. Don- Gallo, J. and Van Wagoner, J.C., 1978. Stratigraphy and fa-
van and T.A. Jackson (Editors), Caribbean Geology: An In- cies analysis of Honduras. Exxon Production Research Co.,
troduction. U.W.I. Publisher's Association, Kingston, Jamaica, unpubl, special report.
pp. 265-284. Gordon, M.B., 1989a. The Chortfs block is a continental,
Carpenter, R.H., 1954. Geology and ore deposits of the Rosario pre-Mesozoic terrane. In: D.K. Lame and G. Draper (Editors),
mining district and the San Juancito Mountains, Honduras, Transactions of the 12th Caribbean Geological Conference, St.
Central America. Geol. Soc. Am. Bull., 65: 23-38. Croix, U.S.V.I. Miami Geological Society, pp. 505-512.
Conkin, J.E. and Conkin, B.M., 1958. Revision of the genus Gordon, M.B., 1989b. Mesozoic igneous rocks on the Chortfs
Nummoloculina and emendation of Nummoloculina heimi block: implications for Caribbean reconstructions (abstr.). Eos,
Bonet. Micropaleontology, 4: 149-150. 70: 1342.
Coogan, A.H., 1973. Nuevos rudistas del Albian y Cenomaniano Gordon, M.B., 1990. Strike-slip faulting and basin formation at
de Mexic6 y del sur de Texas. Rev. Inst. Mex. Pet. 5:51-82. the Guayape fault-Valle de Catacamas intersection, Honduras,
Coogan, A.H., 1977. Early and middle Cretaceous Hippuritacea Central America. Ph.D. dissertation, University of Texas at
(rudists) of the Gulf Coast. In: D.G. Bebout and R.G. Loucks Austin, 260 pp.
(Editors), Cretaceous Carbonates of Texas and Mexico. Bur. Gordon, M.B., 1992. Northern Central America: The Chortfs
Econ. Geol., Univ. Texas Austin, Rept. Invest., 89: 32-70. block. In: G.E.G. Westermann (Editor), Jurassic of the Cir-
Cushman, J.A. and Applin, E.R., 1947. Two new species of cum-Pacific Region. World and Regional Geology 3, Cam-
Lower Cretaceous foraminifera from Florida. Cushman Lab. bridge University Press, New York, pp. 93-121.
Foraminiferal Res. Contrib., 23: 29-30. Gordon, M.B. and Gose, W.A., 1989. The Chortfs block: a raft
Dengo, G., 1985. Mid America: tectonic setting for the Pacific of Mesozoic sediments and Cenozoic volcanics on a solid
margin from southern Mexico to northwestern Colombia. In: foundation. Geol. Soc. Am. Abstr. Prog., 21: 12.
A.E.M. Nairn and EG. Stehli (Editors), The Ocean Basins and Gordon, M.B. and Muehlberger, W.R., 1994. Rotation of the
Margins. Plenum Press, New York, 7, pp. 123-180. Chortfs block causes dextral slip on the Guayape fault. Tecton-
Dieni, I. and Turngek, D., 1979. Parkeria sphaerica Carter, 1877 ics, 13: 858-872.
(Hydrozoan) in the Vraconian (Lower Cretaceous) of Orosei Gordon, M.B. and Young, K., 1993. Bathonian and Valanginian
(Sardinia). Boll. Soc. Paleontol. Ital., 18: 200-206. fossils from Honduras. In: S. Elmi, C. Mangold and
Donnelly, T.W., Horne, G.S., Finch, R.C. and L6pez-Ramos, Y. Alm6ras (Editors), 3~me Symposum International:
E., 1990. Northern Central America: The Maya and Chortfs C6phalopodes Actuels et Fossiles, Symposium F. Roman.
blocks. In: G. Dengo and J.E. Case (Editors), The Caribbean G6obios, M6m. Sp6c., 15: 175-179.

CRETACEOUS C A R B O N A T E B I O S T R A T I G R A P H Y AND E N V I R O N M E N T S IN H O N D U R A S 165
Gose, W.A., 1985a. Paleomagnetic results from Honduras and of Tripoli (North Africa). J. Paleontol., 10: 3-13.
their bearing on Caribbean tectonics: Tectonics, 4: 565-585. Pindell, J.J. and Barrett, S.E, 1990. Geological evolution of the
Gose, W.A., 1985b. Caribbean tectonics from a paleomagnetic Caribbean region; a plate tectonic reconstruction. In: G. Dengo
perspective. In: EG. Stehli and D. Webb (Editors), The Great and J.E. Case (Editors), The Caribbean Region. The Geology
American Biotic Interchange. Plenum Press, New York, pp. of North America, Vol. H, Geological Society of America, pp.
285-301. 405-432.
Gose, W.A. and Finch, R.C., 1987. Magnetostratigraphic studies Raineri, R., 1922. Alghi sifonee fossili della Libia. Soc. Ital. Sci.
of Cretaceous rocks in Central America. Actas Fac. Ciencias Nat., Milano, Atti, 61: 72.
Tierra U.A.N.L. Linares, E1 Cret~icico de M6xico y Am6rica Ritchie, A.W. and Finch, R.C., 1985. Widespread Jurassic strata
Central, Resumenes, 2, pp. 233-241. on the Chortfs block of the Caribbean plate. Geol. Soc. Am.,
Gose, W.A. and Finch, R.C., 1992. Stratigraphic implications of Abstr. Prog., 17:700-701.
palaeomagnetic data from Honoduras. Geophys. J. Int., 108: Roberts, R.J. and Irving, E.M., 1957. Mineral deposits of Central
855-864. America. U.S. Geol. Surv. Bull., 1034, 205 pp.
Hamoui, M. and Saint-Marc, E, 1970. Microfaunes et micro- Rogers, R.D., 1994. Preliminary stratigraphy and structure along
facies du C6nomanien du Proche-Orient. Bull. Cen. Rech. the Rfo Patuca and Rfo Wampd, La Mosquitia, Honduras.
Pau-SNPA, Pau, 4 (2): 257-352. Geol. Soc. Am., Abstr. Prog., 26 (7): A-247.
Haq, B.U., Hardenbol, J. and Vail, ER., 1988. Mesozoic and Rogers, R.D., 1995. Geology along the Rfo Patuca and Rfo
Cenozoic chronostratigrapy and cycles of sea-level change. Wamp6, La Mosquitia, Honduras. Open-file report, Instituto
Soc. Econ. Paleontol. Mineral. Spec. Publ., 42: 71-108. Geogr~ifico Nacional, Tegucigalpa, D.C., 21 pp.
Horne, G.S., Atwood, M.G. and King, A.E, 1974. Stratigraphy, Schroeder, R. and Neumann, M., 1985. Les grands foraminif~res
sedimentology and paleoenvironment of Esqufas Formation in du Cr6tac6 moyen de la r6gion M6diterran6nne. Geobios,
Honduras. Am. Assoc. Pet. Geol., Bull., 58: 176-188. M6m. Sp6c., 7, 161 pp., 68 pl.
Hottinger, L., Drobne, K. and Caus, E., 1989. Late Cretaceous, Scott, R.W., 1990. Models and stratigraphy of Mid-Cretaceous
larger, complex miliolids (Foraminifera) endemic in the Pyre- reef communities, Gulf of Mexico. Soc. Econ. Paleontol. Min-
nean faunal province. Facies, 21 : 99-134. eral., Concepts Sedimentol. Paleontol., Vol. 2, 102 pp.
Kozuch, M.J., 1989. Mapa Geol6gico de Honduras, Cuadr~in- Scott, R.W. and Gonzales-Leon, C., 1991. Paleontology and
gulo de San Francisco de Becerra, escala 1:50.000. Instituto biostratigraphy of Cretaceous rocks, Lampazos area, Sonora,
Geogr~fico Nacional, Tegucigalpa, D.C., 1 sheet. Mexico. Geol. Soc. Am., Spec. Pap., 254:51-67.
Kozuch, M.J., 1991. Mapa Geol6gico de Honduras, Segunda Edi- Scott, R.W., Frost, S.H. and Shaffer, B.L., 1988. Early Creta-
ci6n, escala 1:50.000. Instituto Geogr~ifico Nacional, Teguci- ceous sea level curves, Gulf Coast and southeastern Arabia.
galpa, D.C., 3 sheets. Soc. Econ. Paleontol. Mineral., Special Publ., 42: 275-284.
Lozej, G.E, 1976. Stratigraphy and petrography of the Mochito Simonson, B.M., 1977. Geology of the E1 Porvenir quadran-
limestone, central Honduras (preliminary report). Unpubl. re- gle, Honduras, Central America. Open-file report, Instituto
port, Rosario Resources Corp., E1 Mochito, 22 pp. Geogr~fico Nacional, Tegucigalpa, D.C., 84 pp.
Maync, W., 1953. Pseudocyclammina hedbergi n. sp. from Simonson, B.M., 1981. Mapa geol6gico de Honduras, cuad-
the Urgo-Aptian and Albian of Venezuela. Cushman Found. rangle de E1 Porvenir, escala 1:50.000. Instituto Geogr~fico
Foraminiferal Res. Contrib., 4:101. Nacional, Tegucigalpa, D.C., 1 sheet.
Mills, R.A., 1959. Habr~ petr61eo en Honduras?. Pet. Interam., Southernwood, R., 1986. Late Cretaceous limestone clast con-
17 (5): 39-44. glomerates of Honduras. Unpubl. MS thesis, University of
Mills, R.A. and Barton, R., 1996. Geology of the Ahuas area Texas at Dallas, 299 pp.
in the Mosquitia Basin of Honduras: preliminary report. Am. Weaver, C.E., 1942. A general summary of the Mesozoic of
Assoc. Pet. Geol., Bull., 80:1627-1640. South America and Central America. Proc. 8th Am. Sci.
Mills, R.A. and Hugh, K.E., 1974. Reconnaissance geologic Congr., Geol. Sci., 4:179-180.
map of Mosquitia region, Honduras and Nicaragua Caribbean Weiland, T.J., Suayah, I.B. and Finch, R.C., 1992. Petrologic,
coast. Am. Assoc. Pet. Geol., Bull, 58: 189-207. stratigraphic and tectonic significance of Mesozoic volcanic
Mills, R.A., Hugh, K.E., Feray, D.E. and Swolfs, H.C., 1967. rocks in the Rfo Wampd area, Eastern Honduras. J. S. Am.
Mesozoic stratigraphy of Honduras. Am. Assoc. Pet. Geol., Earth Sci., 6: 309-325.
Bull., 51:1711-1786. Williams, H. and McBirney, A.R., 1969. Volcanic history of
Pia, J., 1936. Calcareous green algae from the Upper Cretaceous Honduras. Univ. Calif. Publ. Geol. Sci., 85, 101 pp.

Chapter 7
Jurassic-Cenozoic Structural Evolution of the

Southern Great B ahama Bank
JOSE L. MASAFERRO and GREGOR E EBERLI
Multichannel seismic reflection lines from the southern Great Bahama Bank (GBB), with a complete record down to 5 s
(two way travel time), reveal the nature of basement and the evolution of the bank, which was strongly influenced by tectonic
activity. The reflection seismic profiles display a fragmented internal anatomy of the bank that is tectonically controlled by
deep basement faults of different ages. Three distinct episodes of deformation have shaped the architecture of the southern
GBB. During the first episode high-amplitude reflections overlaying the acoustic basement are displaced by faults creating a
fault-bounded topographic relief. The seismic facies of the basement with faint continuous, horizontal internal reflections overlain
by a continuous high-amplitude reflection horizon is reminiscent of continental or transitional crust with a sedimentary cover. The
episode of extensional tectonism that affected these two seismic facies probably corresponds to the rifting phase in the Jurassic.
The GBB established on this faulted crust but subsequent growth of the bank immediately leveled the topographic relief. During
the growth of the bank in the Cretaceous about 5 km of shallow-water platform carbonates were accommodated by passive-margin
subsidence. The second deformational episode occurred probably in mid-Cretaceous and was characterized by the reactivation of
some pre-existing structures which first segmented GBB. In southern GBB it created a WNW-ESE-trending margin towards the
Tongue of the Ocean. We speculate that this tectonic phase is a consequence of the reorientation of the stress regime that caused
the plate reorganization in the Caribbean realm. The third deformational episode occurred during the Late Cretaceous-Middle
Eocene when Cuba collided with the southern edge of the Bahamas platform. As a result of the collision, the bank was dissected
by WNW-striking oblique-slip faults forming long, narrow, symmetric depressions within the bank. The deepest depressions were
produced by a long WNW-ESE straight master fault zone consisting of oblique-slip faults with both normal and left-lateral
strike-slip components of displacement. The faults offset the entire bank and diverged upwards in convergent fault slices, forming
negative flower structures. Activity of most faults ceased with the end of the collision in the Late Eocene. The subsequent infilling
by the highly productive carbonate environment healed the depressions masking the scars produced by tectonism. Since the late
Cenozoic the southern GBB is mostly a flat-topped aggrading carbonate bank.
INTRODUCTION trough configuration of the Bahama banks. Some sci-

entists have argued that the modern bank-channel
The northwestern portion of the Bahamas configuration is the result of an inherited Early Juras-
archipelago has generally been considered as a car- sic rift pattern in which the intraplatform seaways are
bonate platform on a passive continental margin that remnants of the grabens, while the platforms grow on
was not significantly affected by tectonism. Some au- ancient horsts (Mullins and Lynts, 1977). Others have
thors have, however, reported the existence of both speculated that the Jurassic structures were healed by
compressional structures close to the Cuban orogen Early Cretaceous platform growth and are buried be-
(Austin, 1983; Ball et al., 1985) and faulting along the low a Cretaceous megabank. They view the modern
Great Abaco fracture zone, in the Providence Chan- pattern of platforms as the result of a partial drowning
nel and in the Tongue of the Ocean (Mullins and of the megabank in the mid-Cretaceous caused by a
Lynts, 1977; Mullins and Van Buren, 1981; Sheri- worldwide crisis in carbonate and reef growth where
dan et al., 1981; Ladd and Sheridan, 1987). Although only isolated platforms survived (Bryant et al., 1969;
most scientists have recognized that a rift topography Paulus, 1972; Meyerhoff and Hatten, 1974; Sheridan,
underlies the Bahamas archipelago, a controversy has 1974; Schlager and Ginsburg, 1981).
emerged around the question of how the Early Juras- With the first multichannel seismic reflection data
sic rift structures influence the present-day bank- from the top of the Great Bahama Bank (GBB) it be-

168 J.L. MASAFERRO and G.E EBERLI
came obvious that the modern platform configuration during the break-up of Pangea. During the break-up,
is not the result of one or the other event alone, but is the Bahama-Cuban area was a transform zone link-
controlled by episodic tectonic segmentation and coa- ing the young Gulf of Mexico and the Atlantic
lescence of the banks during their growth (Eberli and Ocean (Klitgord et al., 1984; Dillon et al., 1987;
Ginsburg, 1987, 1989). Although Eberli and Ginsburg Ross and Scotese, 1988; Sheridan et al., 1988;
(1987, 1989) recognized the role of tectonic deforma- Buffler and Thomas, 1994). Sedimentation in this
tion in the segmentation of the GBB, the limited depth passive-margin setting was characterized by the de-
control of their seismic data did not provide informa- position of volcaniclastic and terrigenous material
tion about possible deeper structures. (Meyerhoff and Hatten, 1974; Mullins and Lynts,
For this study on the southern GBB seismic 1977). In the Late Jurassic, the southern edge of
reflection data that extend to a depth of 5 s two- the Bahamas archipelago and northern Cuba were
way travel time (twtt) are available, allowing us to part of a large evaporite basin (Lewis, 1990; Walles,
evaluate structural deformation of the bank since 1993). Carbonate sedimentation began with shal-
its inception in the Jurassic (Fig. 1A). We develop low-water carbonates that alternated with evaporites,
a structural interpretation based on recognition of indicating a restricted marine environment, and on
structural styles from seismic reflection profiles in localized basement highs. Shallow-water conditions
the southern GBB (Fig. 1B). In particular, we inves- persisted throughout the Early Cretaceous in some
tigate the structural response of the southern GBB areas of the Bahamas archipelago, although deep-
to major tectonic changes. The purpose of this pa- water sediments have been described in the Cayo
per is to document how the carbonate platform has Coco area in northern Cuba (Roque-Marrero and
preserved a record of the major tectonic events that Iturralde-Vinent, 1987; Hempton and Barros, 1993),
have affected the southern portion of the Bahamas and in the northeast Providence Channel (Austin et
archipelago since the breakup of Pangea. al., 1988), and south of San Salvador (Schlager et al.,
1984), suggesting the existence of deep re-entrants
Database in the Bahamas carbonate province at that time.
Seismic reflection data in the Straits of Florida also
The data for this study consist of two seismic re- suggest the existence of such a re-entrant between
flection datasets and a set of gravity data from the GBB and Cay Sal Bank (Masaferro and Eberli,
same area (Fig. 1). A grid of 1300 km of unmigrated 1995).
multichannel seismic profiles was acquired with air The Mid-Cretaceous was a time of change in plate
guns with variable volumes (1600-2600 cub. in.) at motions and global plate re-arrangements (Ross and
intervals of 25 m. 400 km of migrated seismic pro- Scotese, 1988). In the Pacific, intraplate volcanism
files were acquired with an average charge size of caused an increase in volume of oceanic plateaus
520 cubic inch at intervals of 100 m. These data resulting in an increase of global spreading rates and
were provided by Texaco, Inc. In addition, 180 km a relative rise of sea level (Larson, 1991; Mullins
of migrated multichannel seismic profiles were pro- et al., 1992; Vaughan, 1995). In the Caribbean re-
vided by Pecten International. The spacing in the grid gion, a change in the polarity of the subduction zone
varies between 9 km in the northern portion and 5 km caused the insertion of the Farallon plate into the
in the south. The seismic data are not connected to proto-Caribbean realm carrying the Greater Antilles
a well. Thus, age assignments to the main reflections towards the northwest (Burke et al., 1984; Ross and
in the upper part of the seismic reflection profiles Scotese, 1988; Pindell and Barrett, 1990). Move-
were extrapolated from the Doubloon Saxon 1 well ment of the Greater Antilles arc and consumption
which is located 80 km to the east of our seismic grid of the proto-Caribbean crust continued up to the
(Walles, 1993). Correlation of the deeper part of the Late Cretaceous-Paleocene when the arc collided
profiles was based on similarities of reflection char- with the southern edge of the Bahamas platform
acter with seismic reflection data from the available (Burke, 1988; Ross and Scotese, 1988), forming the
literature (Ladd and Sheridan, 1987; Schlager et al., Cuban orogenic belt. The Cuban arc, which consists
1988; Sheridan et al., 1988). Time to depth conversion of obducted ophiolites, imbricated platform carbon-
of seismic profiles was performed using ProMAX TM ates, melanges and arc-derived plutonic and volcanic
processing software. Gravity data from the study area, rocks, was thrust onto the passive margin of the Ba-
also released by Texaco Inc., provided additional ev- hamas archipelago in a northward-verging fold and
idence for some of our interpretations. thrust system (Gealey, 1980). The timing and the po-
larity of subduction is still controversial and several
Regional setting and paleogeographic evolution interpretative models have been proposed (Goodell
and Garman, 1969; Malfait and Dinkelman, 1972;
The Bahamas carbonate platform was probably Mattson, 1973, 1974, 1979; Gealey, 1980; Pindell
established in the Late Jurassic on a thinned crust and Dewey, 1982; Pindell, 1985; Dillon et al., 1987;

CENOZOIC STRUCTURAL EVOLUTION OF THE SOUTHERN GREAT BAHAMA BANK 169
Fig. 1. (A) Location map of the seismic profiles, industry boreholes and core borings across the Great Bahama Bank. A - A t is a trace of
cross-section in Fig. 20. (B) Multichannel seismic data used in this study. Bold line segments indicate parts of the seismic profiles shown
in this paper.
Burke, 1988; Draper and Barros, 1994; Iturralde- unconformably on the highly deformed collisional
Vinent, 1994). With the collision and loading of deposits (Iturralde-Vinent, 1972, 1975, 1988; Pardo,
Cuba onto the margin the Bahamas area became 1975; Pindell and Barrett, 1990; Lewis, 1990). From
a foreland basin (Ball et al., 1985; Walles, 1993; this time onward the NW Bahamas-Cuban area has
Denny et al., 1994). been tectonically stable and the foreland basin be-
The collision ceased in the middle Tertiary, as came again part of the passive continental margin of
indicated by the stratigraphic relationships in Cuba, the North American plate (Fig. 2). The platforms of
where Middle Eocene and younger sediments lie the Bahamas archipelago were influenced by these

170 J.L. MASAFERRO and G.R EBERLI
Fig. 2. Present-day plate boundary faults of the northern Caribbean plate modified from Mann et al. (1995). Shaded area indicates study
area.
tectonic changes but succeeded for the most part to The S.2 seismic facies is bounded at the top by
exist. Subsequent growth of the platforms and the a high-amplitude reflection (R.1, Fig. 3). Occasion-
maintenance of their flat-topped geometry hide the ally, this seismic package contains internally strong,
structural features, which are revealed by the multi- more continuous reflections that can be used for
channel seismic reflection data we present below. correlation within the otherwise chaotic package
(correlation horizons in Fig. 3).
Overlying the high-amplitude reflection R.1 at
SEISMIC REFLECTION DATA approximately 0.8 s (twtt), is the fourth seismic
facies that contains fairly continuous, moderate- to
Seismic facies description high-amplitude reflections, which are locally inter-
rupted by chaotic and incoherent reflections (S.3,
The seismic data contain five distinct seismic fa- Fig. 3). A lateral change of the acoustic charac-
cies defined by the acoustic character of the seismic ter within this unit indicates lateral facies changes.
reflections. The first seismic facies (S.0) is the basal The most common change is from thin low-ampli-
seismic facies occurring below 3.3 s (twtt); it con- tude and transparent reflections to more continuous
sists of horizontal discontinuous, medium-amplitude reflections of a higher amplitude. S.3 is overlain
to chaotic reflections. It is overlain by seismic facies seismic facies S.4 at the top of the seismic reflection
(S.1), which is a group of strong regional high- profiles. S.4 is transparent at the base and contains
amplitude seismic reflections that extend regionally horizontally continuous high-amplitude reflections at
and appear in the lower half of the seismic profiles the top.
from 2.8 to 3.3 s (twtt) (Fig. 3). Displacement of
the S.1 horizons displays the geometry of the deep Seismic facies interpretation
structure.
The third seismic facies (S.2) consists of an in- The nature of the crust underneath the modem
terval of approximately 2 s (twtt) in thickness with banks is unknown as no drill hole ever reached the
incoherent to chaotic seismic reflections (Fig. 3). basement. Refraction seismic data suggest a 12-
Using interval velocities, we estimate the average 14-km-thick transitional crust (Sheridan, 1974). The
thickness of S.2 to be about 5000 m (Table 1). seismic facies S.0 on the reflection seismic data gives

CENOZOIC STRUCTURAL EVOLUTION OF THE SOUTHERN GREAT B A H A M A BANK 171
Table 1
Seismic facies units identified in the southern GBB and their internal descriptions
Seismic facies Seismic facies Interval velocities Estimated ages (top of Interpretation of seismic facies
units (km/s) seismic facies)
S.4 Transparent low-amplitude 1.9-2.5 Recent Platform carbonates

S.3 High-amplitude continuous 3.2-3.6 Middle Paleocene Periplatform basin fill
S.2 Chaotic 4.0-5.1 Late Cretaceous Platform carbonates
S.1 High-amplitude continuous 5.4 Middle Jurassic? Evaporites, clastics and carbonates
S.0 Discontinuous chaotic 5.5 Early Jurassic? Continental to transitional crust
evidence that the southern GBB might be located on progradation from the south and southwest is at least
such a transitional or extended continental crust but as common (Fig. 4). This progradation pattern differs
not oceanic crust. Seismic images of oceanic crust markedly from the prevailing east-to-west prograda-
are characterized by chaotic internal reflections and tion in the northeastern Great Bahama Bank (Hine
a hummocky top (Sheridan et al., 1981; Shipley et and Neumann, 1977; Eberli and Ginsburg, 1987).
al., 1989; Mountain and Tucholke, 1989). Such a The top portions of all the seismic profiles are
chaotic facies is not present in the southern GBB but characterized by horizontal reflections (S.4), indicat-
horizontally continuous reflections are observed being homogeneous platform aggradation in the interior
low the continuous, high-amplitude reflection hori- of southern GBB. Shallow core borings from other
zons S.1. These seismic facies suggest a continental parts of the GBB penetrated this seismic facies and
crust overlain by a sedimentary unit, consisting of consisted of shallow-water carbonates punctuated by
lithologies with large impedance variations. These exposure horizons (Beach and Ginsburg, 1980).
lithologies could be alternations of volcaniclastics,
evaporites and limestones. Age assignments to seismic horizons
The chaotic seismic reflection signature (S.2) has
been interpreted by many authors as shallow-water In order to give relative ages to the deformational
platform carbonates (Ball et al., 1985; Eberli and events and to better describe the structural features,
Ginsburg, 1987, 1989; Denny et al., 1994). This we have divided the section into four stratigraphic
interpretation is corroborated by seismic modeling units based on both the character of the seismic
which indicates that the lack of contrast of acoustic facies and the geometry of the reflections that char-
impedance within the shallow-water platform car- acterize each unit: (1) a high-amplitude seismic unit
bonates is the main factor that causes the incoherent (S.1), (2) a thick, chaotic, intermediate seismic unit
character of the reflections in the seismic sections we interpret as platform carbonate facies (S.2), (3)
(Anselmetti et al., 1997). Thus, the seismic facies at an upper, more continuous seismic unit that corre-
the top 2.7 s (twtt) of the seismic profiles probably sponds to the filling of the depressions (S.3), and
consists exclusively of carbonates (Walles, 1993). (4) an overlying upper relatively undeformed seis-
Across reflection horizon R.1, however, occurs a mic unit (S.4). Determination of geologic ages and
drastic change in seismic facies from low-ampli- lithologies of the seismic facies are difficult because
tude and chaotic reflections to more continuous, of limited well control in the area. No well is situ-
high-amplitude reflections (S.3). This facies is best ated in the area of our seismic reflection data. As a
developed in areas were depressions formed within result, our age assignments are based on the correla-
the bank. The depressions were filled by onlap- tion of the depth of the seismic reflections to depths
ping and prograding packages (Fig. 4), while more of the ages in the nearby Doubloon Saxon 1 well
shallow-water carbonates were deposited on the ad- and similarities in seismic character with reflections
jacent topographic highs. Within the depressions, the described in the literature (Sheridan et al., 1981,
change in seismic facies probably reflects the transi- 1988; Ladd and Sheridan, 1987; Walles, 1993).
tion from platform deposits to more basinal deposits The S.1 seismic facies in the southern GBB is
and might be a mixture of periplatform ooze and similar in seismic character and position with respect
various amounts of neritic debris and redeposited to the overlying seismic facies to a group of reflec-
material (Eberli and Ginsburg, 1989). The continu- tions observed under the Tongue of the Ocean, Ex-
ous reflections on the adjacent highs are probably a uma Sound and northeast Providence Channel (Ladd
result of alternating shallow-water and deeper-water, and Sheridan, 1987; Sheridan et al., 1988). The
chalky lithologies. The prograding packages of S.3 tops of these high-amplitude reflections were cor-
indicate margin progradation into the depressions. related with Middle Jurassic limestones, dolomites
No predominant direction of progradation is ob- and evaporites (Sheridan et al., 1981, 1988). We
served, and although westward progradation is seen, have used line 337 of Ladd and Sheridan (1987),

172 J.L. M A S A F E R R O and G.R EBERLI
II II
~r.,5
~g ~A
O O
. ,..,,~
~99 O
O U=
.,,a
.,..~
. ,...~
,,.o
cA
= 9
9.= .~
.,..~
. ,...,
II O
=
,.~ ..,
(D
....~
9 ,..,
U:: =
O O
" ~ ~
.,.~ ~
,~ ~ 0
~',~
~ ~.~ ~
~
9,'-~
e'~
~.~ ~
=
N ~
i
9 "~
9 ~ ~
9, - 9 9

Fig. 4. Uninterpreted and interpreted unmigrated seismic profile showing the first segmentation of the southern GBB. Direction of
progradation is towards the NE. See Figs. 1B and 4 for location.

174 J.L. MASAFERRO and G.E EB ERLI
located in the southern part of the Tongue of the (Fig. 6) of reflection R. 1, which marks the base of in-
Ocean, to extrapolate these reflections into our seis- tra-platform depressions, shows five platforms sepa-
mic grid, and assign a Middle Jurassic age to the top rated by depressions of different size and orientation.
of seismic facies S. 1. A large, asymmetric depression, termed here
As the Tongue of the Ocean is a basin and our Cochinos Sound (Fig. 6), was a wide N W - S E
grid is within the shallow-water platform of the depression south of the modem Tongue of the
GBB, it is not possible to trace younger reflections Ocean from which it was separated by a narrow
from southern Tongue of the Ocean into our seismic platform ridge. This narrow WNW-ESE-trending
reflection grid. Therefore, we compare the depth margin formed a barrier between the Tongue of the
of three reflections (R.1, R.2 and R.3) of a depth- Ocean and the Cochinos Sound (Figs. 4 and 5). To
converted seismic section to ages in the Doubloon the south, the Cochinos Sound borders the northern
Saxon 1 well (Fig. 5). The Doubloon Saxon 1 is a platform, and turns to the south into a more N-S
deep (6631 m TD) exploration well located 80 km trend to form a narrow N-S depression on the east-
west of our seismic reflection grid (Fig. 1A) (Walles, ern side of the eastern platform (Fig. 6). We spec-
1993). From the depth section, we correlated the ulate that this depression was probably connected
age of reflection R.1 with the top of the Upper Cre- with the Old Bahama Channel in the south. Using
taceous section in the Doubloon Saxon 1 (Fig. 5). average interval velocities the depth of the Cochinos
Reflection R.2 can be correlated with the top of the Sound is estimated to be about 300-500 m and the
middle Paleocene section and reflection R.3 is just depth of the eastern depression approximately 700 m
above the top of the Middle Eocene and could be as In the area southwest of the Cochinos Sound,
young as Early Oligocene. where the coverage by seismic lines is extensive, the
platform is dissected by narrow, more symmetric de-
Internal anatomy of the southern GBB pressions (Figs. 3 and 6). The northwestern depres-
sion extends N W - S E for about 15 km. Toward the
The seismic reflection profiles reveal the compli- east, this long depression turns into a more E - W di-
cated internal structure of the Great Bahama Bank rection (Fig. 6). The depth of the northwestern de-
south of Tongue of the Ocean. An isochron map pression is estimated at about 500 m in its axial val-
SSW NNE
Doubloon Saxon 1 Depth-converted seismic profile from Figure 3 (see location map)
Fig. 5. Age correlation of the Doubloon Saxon 1 well (Walles, 1993) with a depth-converted seismic profile. See text for discussion.

Fig. 6. Isochron map on top of reflection R.1 (Late Cretaceous) showing the orientation of both depressions and adjacent platforms.
Contour interval is 50 ms.
ley. Farther to the south, at the southern limit of our the observed gravity and to estimate the regional gra-
seismic reflection grid, a third depression (southern dient. A regional gradient was estimated from each
depression, Fig. 6) dissects the southern portion of the of the profiles and residual gravity profiles were
platform. It is possible that this depression was an an- obtained by subtracting the estimated regional from
cient northern boundary of the Old Bahama Channel. the observed gravity at all points along the profiles.
The residual gravity map (Fig. 7) displays the
enhanced local anomalies caused by the density con-
GRAVITY DATA trast between high and less-dense sediments. The
areas of negative values coincide approximately with
The gravity data (Web Fig. 7.1, available the depressions while positive values follow the ad-
at http://www.elsevier.nl/locate/caribas/), provided jacent shallow-water areas. Thus, the residual gravity
by Texaco Inc., consist of an uncorrected Bouguer anomalies support the seismic facies interpretation
anomaly map. To remove the regional effect that where the chaotic, incoherent seismic reflections are
obscures the local anomalies, we corrected the data. thought to represent shallow-water platform carbon-
The regional gradient is a component of the gravity ates (higher density), and the more continuous, high-
anomaly that has longer wavelength and is thought to amplitude seismic reflections form the less-dense
be caused by density contrasts of deep-seated masses sediments in the depressions.
located at any depth. Removal of this regional effect
from the Bouguer gravity map yields the residual
anomalies which represent the anomalies created by STRUCTURAL ANALYSIS
a density contrast at or near the surface. Different
methods can be used to remove the regional gra- The structural style of the southern GBB is domi-
dient from the Bouguer gravity map. We applied a nated mainly by faults of different ages accompanied
two-dimensional surface fitting technique (Nettleton, by localized compressional structures. Each of the
1971) to fifteen N-S cross-sections of the gravity four stratigraphic units (S.1-S.4) contains a charac-
field, using a polynomial of second order to smooth teristic set of structural features that formed during

Fig. 7. Residual gravity map obtained subtracting the estimated regional from the observed gravity (Web Fig. 7.1). The minimumvalues
coincide with depressions and the maximumvalues with adjacent platforms (shaded areas, compare with Fig. 6). Contour interval is 0.7
mGal.
different tectonic stages in the evolution of the car- faults have a predominantly NNE strike and most of
bonate platform. them dip steeply to the west. The lack of continuous
reflections within the overlying chaotic unit S.2 does
Fault geometries not show whether these faults propagated upward
into this unit. However, some migrated seismic lines
Seismic reflection profiles through the southern with more continuous reflections in the chaotic fa-
GBB image three generations of faults that disrupt cies show propagation of some faults into this unit
the carbonate platform at different levels. The oldest where they seem to die out.
set of faults has a planar to more curved profile The second generation of faults observed in the
and displace the top of the high-amplitude reflection area dissects S.2 seismic facies. An example is
horizon of S.1. Fig. 8 shows a migrated seismic pro- shown in Fig. 10 from the northern end of our
file that crosses two of these faults, which displace seismic reflection grid, where two steep normal
the S.1 reflections from 2.8 s to 3 s (twtt) and from 3 faults form a WNW-ESE-striking narrow asymmet-
s to 3.2 s (twtt) which translates into approximately ric trough. The southern fault disrupts the S.2 unit
600 m of vertical displacement. These faults separate at approximately 1.7 s (twtt) based on the southward
extensional fault blocks. Fig. 9 is a structure map inclined reflection terminations of unit S.3 (shaded
that shows the distribution and orientation of the area, Fig. 10). The dipping reflections thicken to the
faults that affected the S.1 seismic facies. The old southwest into the fault plane. The northern fault

C E N O Z O I C S T R U C T U R A L E V O L U T I O N OF THE S O U T H E R N GREAT B A H A M A B A N K 177
Fig. 8. Uninterpreted and interpreted migrated seismic profile showing old faults affecting S.1 seismic facies (Middle Jurassic?). See text
for discussion. See Fig. 1B for location.
Fig. 9. Isochron map contoured on top of the S.1 seismic facies. Inferred pre-Cretaceous faults have NNE-SSW orientation. Contour
interval is 50 ms.

17 8 J.L. MASAFERRO and G.E EBERLI
Fig. 10. Part of migrated seismic profile and interpreted line drawing showing a narrow, fault-bounded trough. The SW-dippingreflections
within the shaded area of S.3 thicken towards the fault plane indicating that the deposition of the strata was coeval with fault activity.
Subsequent filling of the trough, however, is by NE-prograding sequences. See Fig. 1B for location.
displays contraction in the upper part whereas in the In map view, this main fault system consists of three
deeper part the same fault shows normal separation. fault zones (fault zones A, B and C, Fig. 14) char-
A third generation of faults is widespread and acterized by their linearity, by the presence of local-
displaces strata of Paleocene and younger age. ized contractional structures such as folds and reverse
High-angle faults of this generation have mostly nor- faults coexisting with the normal dip-slip faulting, and
mal separation and propagate towards the surface, dis- by an uplifted zone that blocks the main fault corridor.
rupting almost the entire section of S.3 (Figs. 11-13). The major W N W - E S E fault zone A is about 10 km in

Fig. 11. Uninterpreted and interpreted part of unmigrated seismic profile. Faults dip steeply and have both normal and reverse separation.
Steep faults seem to offset S. 1, while reverse faults continue to younger strata. Box is enlarged in Fig. 12. See Fig. 1B for location.
width and 65 km in length and displays a sharp bend In cross-section, the entire fault system is char-
in its northwest portion where contractional struc- acterized by steep faults that show both normal and
tures such as reverse faults and folds are concentrated reverse separation where the S.1 unit is involved in
(Fig. 14). To the east, fault zone A merges with the E - the deformation. The narrow area where the S.1 unit
W-trending fault zone B to form a braided geometry. is affected by splay-upward structures is generally
This combined fault zone widens southeastward and devoid of significant reflections, producing a sharp
terminates at an uplifted and folded structure. A third change in the continuity of the reflections. Four seis-
fault zone, C, is located in the southernmost extension mic reflection profiles (Figs. 11-13 and 15) cross
of our seismic grid and is not connected to the other fault zone A, and cut the fault system at a fight angle
two fault zones (Fig. 14). illustrating the marked lateral changes in fault geom-

Fig. 12. Uninterpreted and interpreted migrated upper part of seismic profile shown in Fig. 11, showing normal and reverse fault
separation, some contractional features and the creation of a depression within the fault system. These are characteristics of a negative
flower structure. Note how the infilling sequences heal the structural deformation and how the seismic reflections are horizontal above
R.2.
etry within a distance of 25 km (Fig. 14). Figs. 12 converge at depth. North of this branching-upward
and 13 show faults that have an almost symmetri- structure a fault with normal separation (called here
cal upward-branching form showing reverse faults D), displaces both seismic unit S.1 (with throws
and folding on the northern side. These fault splays up to 1200 m) and probably the lower part of the

Fig. 13. Uninterpreted and interpreted part of seismic profile shown in Fig. 11. Some faults displace S.1 seismic facies. Fault D is an old
normal dip-slip fault with approximately 1000 m of displacement. See Fig. 1B for location.
chaotic unit S.2. Fault D has a slightly listric profile mic grid (Figs. 8 and 16). The geometry of this fault
at depth with a gentle rollover in the hanging wall zone is characterized by a series of normal faults
(Fig. 13). Passing from the eastern side to the central that converge at depth. In contrast to fault zone A,
area of the fault zone A, the geometry changes from no contraction or reverse faults are observed. Fig. 16
an upward-branching form to a single strand fault illustrates the structural style of faults in zone B
where the strata thicken toward its plane (Fig. 15). and the style of the infilling sequences. The de-
The second fault zone B strikes east-west and velopment of aggrading and prograding sequences
probably continues to the southwest out of our seis- within this fault-controlled depression seems to be

Fig. 14. Tectonic map of structures mapped on the R.1 reflection and position of seismic profiles. Three fault zones can be identified.
Fault zones A and B merge towards the east and terminate at an area of tectonic uplift. Fault zone C parallels the two other zones further
in the south. Shaded areas indicate extension.
related to the temporal evolution of the fault system. ticline is down faulted (fault B) almost coeval with the
Initial and in some cases syntectonic deposition is local convergent phase. A local angular unconformity
aggradational (Fig. 16). As soon as the fault system separates the folded sequences from the flat-lying se-
becomes inactive prograding sequences start to fill quences of unit S.4 above (Fig. 17).
the depression from south to north (Fig. 16). Progra-
dation fills the deepest portion of the depression Folds
and shifts the basin axis to the north. Before the
infilling is completed another tectonic pulse deepens Two types of folds are recognized based on their
again the basin. Onlapping sequences fill this young cross-sectional geometry: (1) folds related to reverse
embayment. Finally, shallow-water conditions re- faults that lie within the main fault system, and are
established over the entire bank, masking the former mostly parallel and subparallel to the displacement
depression and infilling sequences. zone (Figs. 12 and 13), and (2) localized anticlines
The third fault zone, C (Figs. 14 and 17), has the and monoclinal flexures associated with vertical to
following geometric characteristics: a narrow fault subvertical normal faults (Fig. 18). These flexures
zone, convergent steeply dipping faults showing nor- are located outside and at the southeastern end of the
mal separation at depth, and coeval normal and remain fault zone (Fig. 14). Folds that are associated
verse faults with folding in the upper unit. On the mi- with the upward-branching fault system lie within
grated seismic profile (Fig. 17) that crosses this fault or immediately adjacent to the fault zone (Fig. 14).
zone, the base of unit S.3 is down-faulted and tilted The crestal traces of the folds trend subparallel and
towards the southernmost fault creating a steep south- oblique to the main fault zone direction.
westward-dipping ramp. The sediments start to fill in
the space created by extensional faults progressively
onlapping the ramp towards the northeast. After de- STRUCTURAL INTERPRETATION
position of the onlapping sequences the strata in the
center of the depression are locally folded producing The seismic reflection data show two main struc-
a symmetrical anticline. The southern flank of the an- tural styles in the southern GBB: a predominantly

Fig. 15. Uninterpreted and interpreted part of seismic profile illustrating a change in fault geometry across fault zone A. See Figs. 1B and
14 for location.
extensional style with NNE-SSW-striking normal Most of the extensional faults are confined to the
faults, and a late sinistral divergent strike-slip S. 1 unit that we interpret to be pre-Early Cretaceous
style within a main WNW-ESE-striking deforma- in age. Therefore their origin is probably related to
tion zone. The deformation is dominated by faults the late stages of rifting caused by the separation be-
that show mostly normal separation on the seismic tween North America and Africa (Mullins and Lynts,
profiles. 1977; Ross and Scotese, 1988; Buffer and Thomas,
1994). The timing of faulting is not constrained by
Extensional phase well control, but rifting is reported to begin during
the Early-Middle Jurassic (Sheridan and Osburn,
Due to the limited amount of dip seismic reflec- 1975; Sheridan et al., 1981, 1988; Klitgord et al.,
tion profiles that cross the extensional fault system it 1984; Ladd and Sheridan, 1987). The extensional
is difficult to infer the regional extension direction. faults are buried by a thick pile of platform car-

t-

C~
t~
t-
Fig. 16. Uninterpreted and interpreted upper part of migrated seismic profile showing the geometry of fault zone B. Faults with normal separation form a depression that is subsequently filled by prograding
sequences. See text for discussion.
bonates deposited during the Late Jurassic-Early (5) The abrupt change in the geometry along-
Cretaceous. Some of the old faults propagate into strike of the main deformation zone, from an up-
the chaotic unit (S.2), e.g. creating a WNW-striking ward-branching structure to a single-strand fault
margin at the southern end of Tongue of the Ocean (Figs. 13 and 15).
(Fig. 6), and narrow depressions within the Upper (6) External folds oriented oblique and outside
Jurassic-Lower Cretaceous shallow-water platform the main deformation zone (Fig. 18) that look like
(Fig. 10). We interpret these propagations as related flexures or monoclinal knees in seismic sections
to a younger tectonic event in the mid-Cretaceous (Harding et al., 1985) indicate local convergence
that reactivated some of the Jurassic faults. associated with the displacement of the fault zone.
(7) The change in both seismic reflection facies
Transtensional phase and thickness across the fault zone from continuous,
high-amplitude reflections in the southwestern side
The extensive coverage of the main deformation to a chaotic, thicker, discontinuous reflections in the
zone by the seismic reflection grid defines both northeastern side (Fig. 12).
the orientation and the geometry of the structural (8) The presence of local unconformities with
features of a left-lateral strike-slip system with an a limited lateral extent within the subsided areas
important component of extension. The difficulty indicates pulses of syntectonic sedimentation associ-
in the interpretation of areas where strike-slip is a ated with the temporal development of the strike-slip
main component of deformation from vertical sec- system (Fig. 17).
tions has led different authors to establish numerous This structural assemblage of the southern GBB
subsurface criteria to identify such areas (Harding, cannot be interpreted in terms of either a pure ex-
1983, 1985, 1990; Gibbs, 1986, 1987). The follow- tensional nor a pure strike-slip model. Most of the
ing characteristics of structural features suggest a deformation is dominated by faults that look exten-
left-lateral strike-slip component in the deformation sional on the seismic data, but the coexistence of
of the southern GBB. folds, reverse and normal faults is reminiscent of a
(1) The main deformation zone is restricted to strike-slip system with a component of extension.
an elongated, straight, narrow area about 8 km This mixture of structural styles of both extensional
wide. Fig. 3 shows the width of this zone in a and associated contractional elements is characteris-
seismic reflection line that crosses the structure at tic for divergent wrench zones (Harding et al., 1985).
almost a right angle. The horizontal reflections north
and south of the deformation zone are tectonically Structural model for the southern GBB
undisturbed.
(2) High-angle faults that converge at depth dis- Experimental models on fault reactivation by
rupt the entire section down and beyond the S.1 Richard and Krantz (1991) and oblique slip by
seismic facies. On several seismic reflection profiles Mandl (1988) support our interpretation. Richard
(e.g., Figs. 12 and 16) the continuation of these and Krantz's models suggest that a previously
faults is seen in the upper S.3 unit. faulted zone can control the deformation of a strike-
(3) The bending geometry of the main displace- slip zone, and the pre-existing basement structure
ment fault zone defines an area of local convergence controls also the location of the deformation im-
that can be interpreted as a restraining bend (Crow- posed during the subsequent strike-slip episode. The
ell, 1974b) (Fig. 14). Folds and faults with reverse set of structures that Richard and Krantz (1991)
separation are concentrated in this area. The ex- obtained from the experimental models are analo-
istence of these localized contractional structures gous to the structures we described in the main fault
allows us to infer a general sinistral sense of dis- system (Fig. 19A). Mandl (1988) interpreted the for-
placement (Fig. 14). The fault trace also bends into a mation of 'tulip'-type cross-sectional fault structures
more E - W orientation to the southeast, and creates or negative flower structures (Harding, 1985) as a re-
a zone of extension or an incipient releasing bend sult of oblique slip induced by a basement fault. His
(Crowell, 1974b). models suggest that the fault pattern in the overbur-
(4) The upward-branching geometry observed in den depends on the predominance of the strike-slip
some seismic reflection profiles (Figs. 12, 13, 16 or dip-slip component in the basement. If the strike-
and 17) across the deformation zone that includes slip component dominates, the result is a 'tulip'-type
both normal and contractional elements in the upper fault structure, whereas if the strike-slip component
S.3 unit defines negative flower structures (Harding, is small the structure resembles a single graben.
1983, 1985; Harding et al., 1985). The almost coeval In either model the pattern of the overburden faults
development of folds and extensional faults within caused by a strike-slip component in the deformation
the negative flower structure is characteristic of of the basement is comparable with the fault pattern
strike-slip systems. that characterizes our seismic sections. By analogy,

t-"

9
Fig. 17. Uninterpreted and interpreted migrated part of a seismic profile showing the geometry of fault zone C. Seismic profile illustrates the coeval coexistence of both extensional and contractional ~z
structures within the fault system. See text for discussion.
t~
t-
Fig. 18. Uninterpreted and interpreted migrated part of a seismic profile showing folding outside the main deformation zone.
the strike-slip structures seen in the southern GBB can platform into a foreland tectonic setting. Multichan-
be interpreted to be induced by a basement fault sys- nel seismic reflection data across the southern GBB
tem located at depth. In particular, fault D (Fig. 19B) suggest that the bank underwent three major tec-
may be such a pre-existing dip-slip fault subsequently tonic events during its geologic history. These three
reactivated in a strike-slip mode creating extensional tectonic episodes can also be seen in other parts of
and contractional structures in the overburden similar the GBB. In the following we discuss the timing
to the structures observed in the models. of these events and a possible regional correlation.
The first deformation event was extensional associ-
ated with a transform system that operated probably
DISCUSSION through the Early-Middle Jurassic in a broad zone
across the Florida-Bahamas area (Klitgord et al.,
The internal configuration of the southern GBB 1984) (Fig. 20A). Klitgord et al. (1984) proposed
documents the change of a passive-margin carbonate a transform plate boundary located between the

realm (Ross and Scotese, 1988; Scotese et al., 1989).

In the southern GBB we see evidence that this
reorganization of the stress regime caused the reacti-
vation of older Jurassic faults and their propagation
into the Upper Jurassic-Lower Cretaceous carbonate
platform (Fig. 20D). Some of these faults die out
within the Upper Jurassic-Lower Cretaceous section
without producing significant changes in the archi-
tecture of the platform. Others propagated through
the entire section creating deep, narrow, WNW-strik-
ing troughs within the Upper Jurassic-Lower Creta-
ceous shallow-water carbonate platform (Fig. 20D).
Further north, both Andros Bank and Bimini Bank
were segmented and backstepped probably during
this time as a result of extensional faulting (Fig. 21B)
(Eberli and Ginsburg, 1987, 1989). In the Santaren
Channel, seismic reflection profiles show that the
Lower Cretaceous megabank was also segmented by
deep, NW-striking extensional faults (Fig. 21B) with
throws up to 1200 m (Masaferro and Eberli, 1995)
creating a northeast-facing platform margin.
The third deformational episode occurred dur-
Fig. 19. Tectonic model for the southern GBB. Base- ing the Late Cretaceous/middle Tertiary, when
ment-induced strike-slip structures. (A) Reactivation of a nor- the Bahamas archipelago became involved in the
mal dip-slip basement fault in strike-slip mode (modified from Caribbean-American collision (Fig. 20E). The
Richard and Krantz, 1991). (B) Part of seismic profile (Fig. 13) southern GBB started to be affected by subduction
showing similarities with model. See text for discussion. beneath the Caribbean plate, which some authors
postulated to be oblique under the Caribbean plate
producing a left-lateral wrench system (Mullins and
Bahamas and Cuba fracture zones that connected Sheridan, 1983; Mullins, 1984; Pindell and Barrett,
the Central Atlantic and Gulf of Mexico spread- 1990; Mann and Burke, 1990; Mann et al., 1995).
ing systems. The southern GBB established on a In the southern GBB, this collision resulted in re-
stretched transitional/continental crust controlled by newed movements on older Jurassic faults and the
a set of NW-striking strike-slip faults and NNE ex- formation of new WNW-striking strike-slip zones.
tensional faults created by the Jurassic transform The pre-existing Jurassic transform faults (Klitgord
zone (Fig. 20B). Rift-related sediments consisting of et al., 1984; Ladd and Sheridan, 1987; Sheridan et
clastics, evaporites and hemipelagic carbonates were al., 1988) probably played an important role in defin-
deposited soon after breakup (Sheridan et al., 1981; ing the newly formed strike-slip system reactivated
Ladd and Sheridan, 1987). In the northwestern GBB, by the Cuban convergent orogen. The structural ex-
this extensional event is indicated in seismic reflec- pression of this deep wrenching in the carbonate
tion profiles by two NNE-striking normal faults that platform was a composite of extension and strike-
probably first segmented the bank and formed the slip elements. This phase of left-lateral strike-slip
Straits of Andros (Fig. 21A). Seismic reflection pro- faulting, which was primarily accommodated on
files from the Straits of Florida and Santaren Chan- three narrow fault zones, caused the fragmentation
nel suggest that part of this area was a fault-con- of the Cretaceous carbonate platform in the area of
trolled deep-water re-entrant (Masaferro and Eberli, the southern GBB. Individual faults branched out
1995). Schlager et al. (1984) also reported the exis- from the deep basement showing extensional and
tence of Early Cretaceous deep-water sediments in compressional elements that defined negative flower
the northeast Providence Channel (Fig. 21A) giving structures. Transtension associated with these faults
evidence for an old re-entrant at this place. controlled localized subsidence and sedimentation.
The second deformational episode probably oc- Sedimentary sequences developed within the sub-
curred during the middle Cretaceous after the depo- sided depressions. They show in some cases internal
sition of Lower Cretaceous shallow-water carbonate angular unconformities as a result of the episodic
platform sediments in a post-rift subsidence phase tectonic pulses associated with the activity of the
(Fig. 20C). The middle Cretaceous was a time of strike-slip faults. During this time, further north the
major plate rearrangement and a change in the ori- Straits of Andros started to be infilled by aggrading
entation of tectonic stresses affecting the Caribbean sequences followed by successive pulses of progra-

Fig. 20. Interpretation of the tectonic evolution of the southern GBB along cross-section A-A' (Fig. 1A). A rift topography is buried by
Late Jurassic-Early Cretaceous sedimentation. Mid-Late Cretaceous tectonism segments the platform during the collision of Cuba with
the North American plate. In the tectonic quiet post-Eocene to Recent time the platform heals these tectonic scars again. See text for
discussion.
dation (Fig. 21C) (Eberli and Ginsburg, 1989). In the form became part of a foreland basin (Fig. 20F).
Santaren Channel area, a major extensional border Soon after the collision the productive carbonate
fault east of Cay Sal that had controlled the location system responded hiding the depressions created
of the platform margin, was inverted due to com- by tectonism. During the Middle/Late Eocene, the
pressional forces of the Cuban-Bahamas collision convergence ceased when Cuba was detached from
(Masaferro and Eberli, 1995). Reactivation led to the the Caribbean plate and was added to the North
development of faults with reverse separation and American plate (Ross and Scotese, 1988; Mann et
broad anticlines along this margin (Masaferro and al., 1995). In the seismic sections, the cessation of
Eberli, 1995). tectonic activity is well represented by a sharp, hori-
During the collision and subsequent orogeny in zontal reflection (R.2) at approximately 0.4 s (twtt).
Cuba, the southern extension of the Bahamas plat- Horizontal seismic reflections above this reflection

190 J.L. M A S A F E R R O and G.E E B E R L I
Fig. 21. Compilation of structural features and paleogeographic evolution of the Great Bahama Bank area. Arrows indicate main direction
of progradation. Data compiled from Schlager et al. (1984), Eberli and Ginsburg (1987, 1989), Sheridan et al. (1988) and Masaferro and
Eberli (1995).
record vertical growth of the carbonate platform up CONCLUSIONS

to the present. Some faults, however, seem to affect
very young seismic reflections that might indicate re- Seismic reflection profiles across the southern
cent tectonic activity. Such Cenozoic tectonic events GBB show that the internal architecture of the bank
are also documented in various places in the north- is primarily a result of the interaction between tec-
west Bahamas, e.g. in Walkers Cay (Mullins and tonic destructive processes and platform recovery
Van Buren, 1981), Bimini Bank (Eberli and Gins- through sedimentation. Seismic data suggest that
burg, 1987) and at the western side of the Straits the southern GBB experienced three segmentation
of Florida close to the eastern edge of the former events. The first segmentation event was probably
larger Cay Sal Bank (Eberli et al., 1995). The origin related to the Jurassic rifting and created a series of
of these post-collisional tectonic events is uncertain, fault blocks. Subsequent sedimentation buried large
but it might imply that deformation was still oc- portions of these rift structures to form the south-
curring between the Caribbean and North American ern GBB. Probably during mid-Cretaceous time, a
plates during the Neogene. second tectonic event that was also extensional re-

activated WNW-trending old structures as normal Bahaman-Cuban collision zone. Am. Assoc. Pet. Geol. Bull.,
faults. It created in the southern GBB a NE-fac- 69: 1275-1294.
Beach, D.K. and Ginsburg, R.N., 1980. Facies succession of
ing margin towards the Tongue of the Ocean and
Pliocene-Pleistocene carbonates, northwestern Great Bahama
several deep water re-entrants further north. The Bank. Am. Assoc. Pet. Geol. Bull., 64: 1634-1642.
third segmentation event, during the Late Creta- Bryant, W.R., Meyerhoff, A.A., Brown, N.K., Furrer, M.A., Pyle,
ceous to middle Tertiary, was transtensional and T.E. and Antoine, J.W., 1969. Escarpments, reef trends, and
synsedimentary. Oblique-slip faults with reverse and diapiric structures, eastern Gulf of Mexico. Am. Assoc. Pet.
Geol. Bull., 53: 2506-2542.
sinistral strike-slip components disrupted the bank,
Buffler, R.T. and Thomas, W.A., 1994. Crustal structure and
forming symmetric intraplatform depressions. Sub- evolution of the southeastern margin of North America and
sequently, sedimentation filled the depressions with the Gulf of Mexico basin. In: R.C. Speed (Editor), Phanero-
aggrading/prograding packages burying the inactive zoic Evolution of North American Continent-Ocean Transi-
structures. tions. Geol. Soc. Am., Decade of North American Geology
Project, Continent-Ocean Transect Volume. Geological Soci-
The internal structure of the GBB reveals a
ety of America, pp. 219-264.
competition between relief-forming tectonism and Burke, K., Cooper, C., Dewey, J.E and Pindell, J.L., 1984.
masking sedimentation. Beneath the flat top of the Caribbean tectonics and plate motions. In: W.E Bonini, R.B.
southern GBB lies a complex history of changing Hargraves and R. Shagam (Editors), The Caribbean-South
geological regimes from a passive continental mar- American Plate Boundary and Regional Tectonics. Geol. Soc.
Am. Mem., 162:31-63.
gin into a collisional plate boundary, the effects of
which became masked by sediments of the highly Earth Planet. Sci. Lett., 16:201-230.
productive carbonate platform. Crowell, J.C., 1974b. Sedimentation along the San Andreas fault.
In: R.H. Dott Jr. (Editor), Modern and Ancient Geosynclinal
Sedimentation. Soc. Econ. Paleontol. Mineral. Spec. Publ., 19:
ACKNOWLEDGEMENTS 292-303.
Denny, W., Austin, J.A. and Buffler, R.T., 1994. Seismic stratig-
raphy and geologic history of Mid-Cretaceous through Ceno-
We thank Texaco Inc. for providing us with the zoic rocks, Southern Straits of Florida. Am. Assoc. Pet. Geol.
seismic data and the gravity data, and Pecten In- Bull., 78: 461-487.
ternational for additional migrated seismic profiles. Dillon, RS., Austin, J.A., Scanlon, N. and Klitgord, K.D., 1987.
Numerous discussions with Chris Avenius, Steve Geology of the Caribbean. Oceanus, 30: 42-52.
Draper, G. and Barros, J.A., 1994. Cuba. In: S. Donovan and
Hook, Tim Dixon and John Hurst were of great T.A. Jackson (Editors), Caribbean Geology An Introduc-
benefit to the ideas presented in the paper. Critical tion. University of West Indies Publishers' Association, pp.
and thorough reviews by Henry Mullins, Robert 65-83.
Sheridan, Jim Dolan, and Paul Mann improved Eberli, G. and Ginsburg, R.N., 1987. The segmentation and
the manuscript. Kevin Cunningham read the final coalescence of Cenozoic carbonate platforms, northwestern
Great Bahama Bank. Geology, 15: 75-79.
version and improved the English. Financial sup- Eberli, G. and Ginsburg, R.N., 1989. Cenozoic progradation of
port for this project was provided by NSF Grant northwestern Great Bahama Bank, a record of lateral platform
OCE-9314586 (to G.P. Eberli), and several smaller growth and sea-level fluctuations. In: RD. Crevello, J.A. Wil-
grants from GSA, 1APG, Sigma Xi and GCAGS son, J.F. Read and F.J. Sarg (Editors), Controls on Carbonate
(to J.L. Masaferro). The Fulbright Commission is Platform and Basin Systems. Soc. Econ. Paleontol. Mineral.
Spec. Publ., 44: 339-351.
acknowledged for supporting J.L. Masaferro in the
Eberli, G., Anselmetti, E and Swart, R, 1995. The Bahamas
early stages of the project. transect: Neogene/Quaternary sea-level fluctuations and fluid
flow in a carbonate platform. Safety Package for Leg 166,
ODP data file (unpublished).
Gealey, W.K., 1980. Ophiolite obduction mechanism. Proc. Int.
REFERENCES
Ophiolite Symp., Nicosia, pp. 228-243.
Gibbs, A.D., 1986. Strike-slip basins and inversion: a possible
Anselmetti, ES., Eberli, G.R and Bernoulli, D., 1997. Seismic model for the southern North Sea gas areas. In: J. Brooks, J.C.
modeling of a carbonate platform margin (Montagna della Goff and B. Van Hoorn (Editors), Habitat of Paleozoic Gas in
Maiella, Italy): variations in seismic facies and implications N.W. Europe. Geol. Soc. London Spec. Publ., 23: 23-35.
for sequence stratigraphy. In: I. Palaz and K.J. Marfurt (Edi- Gibbs, A.D., 1987. Development of extensional and mixed mode
tors), Carbonate Seismology. S.E.G. Geophys. Dev., 6: 373- sedimentary basins. In: M.R Coward, J.E Dewey and EL.
406. Hancock (Editors), Continental Extensional Tectonics. Geol.
Austin, J.A., Jr., 1983. OBC 5-A. Overthrusting in a deep-water Soc. London Spec. Publ., 28: 19-33.
carbonate terrane. In: A.W. Bally (Editor), Seismic Expression Goodell, H.G. and Garman, K., 1969. Carbonate geochemistry of
of Structural Styles. Am. Assoc. Pet. Geol., Stud. Geol., 3: Superior deep test well, Andros Island, Bahamas. Am. Assoc.
3.4.2-167-172. Pet. Geol. Bull., 53: 513-536.
Austin, J.A., Jr. et al., 1988. Proceedings of the Ocean Drilling Harding, T.R, 1983. Divergent wrench fault and negative flower
Program, Scientific Results, Vol. 101, 501 pp. structure, Andaman Sea. In: A.W. Bally (Editor), Seismic
Ball, M.M., Martin, R.G., Bock, W.D., Sylwester, R.E., Bowles, Expression of Structural Styles. Am. Assoc. Pet. Geol. Bull.,
R.M., Taylor, D.E. Coward, L., Dodd, J.E. and Gilbert, L., Studies in Geology, Series 15, 4.2-1-4.2-8.
1985. Seismic structure and stratigraphy of northern edge of Harding, T.E, 1985. Seismic characteristics and identification

192 J.L. M A S A F E R R O and G.R E B E R L I
of negative flower structures, positive flower structures and Mountain, G.S. and Tucholke, B.E., 1989. Abyssal sediment
positive structural inversion. Am. Assoc. Pet. Geol. Bull., 69: waves. In: A.W. Bally (Editor), Atlas of Seismic Stratigraphy,
582-600. Vol. 3. American Association of Petroleum Geologists, Tulsa,
Harding, T.R, 1990. Identification of wrench faults using subsur- OK, pp. 233-236.
face structural data; criteria and pitfalls. Am. Assoc. Pet. Geol. Mullins, H.T., 1984. Structural controls of contemporary carbon-
Bull., 74: 1590-1609. ate continental margins. In: Platform Margin and Deep Water
Harding, T.R, Vierbuchen, R.C. and Christie-Blick, N., 1985. Carbonates. Soc. Econ. Paleontol. Mineral., Short Course 12,
Structural styles, plate-tectonic setting, and hydrocarbon traps pp. 57.
of divergent (transtensional) wrench faults. In: K.T. Biddle and Mullins, H.T. and Lynts, G.W., 1977. Origin of the northwestern
N. Christie-Blick (Editors), Strike-Slip Deformation, Basin Bahama platform: review and reinterpretation. Geol. Soc. Am.
Formation and Sedimentation. Soc. Econ. Paleontol. Mineral. Bull., 88: 1447-1461.
Spec. Publ., 37: 51-77. Mullins, H.T. and Sheridan, R.E., 1983. Wrench tectonic origin
Hempton, M.R. and Barros, J.A., 1993. Mesozoic stratigraphy for the northern Bahama Platform. Geol. Soc. Am. Abstr.
of Cuba: depositional architecture of a southeast facing conti- Prog., 15: 648-649.
nental margin. GCSSEPM Foundation 13th Annu. Res. Conf. Mullins, H.T. and Van Buren, H.M., 1981. Walkers Cay Fault,
Proc., pp. 193-209. Bahamas: evidence for Cenozoic faulting. Geo-Mar. Lett., 1:
Hine, A.C. and Neumann, A.C., 1977. Shallow carbonate bank 225-231.
margin growth and structure, Little Bahama Bank. Am. Assoc. Mullins, H.T., Breen, N., Dolan, J., Wellner, R., Petruccione,
Pet. Geol. Bull., 61: 374-406. L., Gaylord, M., Andersen, B., Melillo, A., Jurgens, A. and
Iturralde-Vinent, M., 1972. Principal characteristics of Oligocene Orange, D., 1992. Carbonate platforms along the southeast
and lower Miocene stratigraphy of Cuba. Am. Assoc. Pet. Bahamas-Hispaniola collision zone. Mar. Geol., 105: 169-
Geol. Bull., 56: 2369-2379. 209.
Iturralde-Vinent, M., 1975. Problems of the application of two Nettleton, L.L., 1971. Elementary gravity and magnetics for
modern tectonic hypotheses to Cuba and the Caribbean region. geologists and seismologists. Monogr. Ser., Soc. Explor. Geo-
Am. Assoc. Pet. Geol. Bull., 59: 838-855. phys., 1,121 pp.
Iturralde-Vinent, M., 1988. Naturaleza geologica de Cuba. La Pardo, G., 1975. Geology of Cuba. In: A.I.M. Nairn and E
Editorial Cientifico-Tecnica, Habana, 246 pp. Stehli (Editors), The Ocean Basins and Margins, The Gulf
Iturralde-Vinent, M., 1994. Cuban geology: A new plate-tectonic and Mexico and the Caribbean. Plenum Press, New York, pp.
synthesis. J. Pet. Geol., 17: 39-70. 553-615.
Klitgord, K.D., Popenoe, R and Shouten, H., 1984. Florida: Paulus, EJ., 1972. The geology of Site 98 and the Bahamas
a Jurassic transform plate boundary. J. Geophys. Res., 89: platform. Init. Rep. DSDP, 11: 877-897.
7753-7772. Pindell, J.L., 1985. Alleghenian reconstruction and subsequent
Ladd, J.W. and Sheridan, R.E., 1987. Seismic stratigraphy of the evolution of the Gulf of Mexico, Bahamas, and proto-
Bahamas. Am. Assoc. Pet. Geol. Bull., 7: 719-736. Caribbean. Tectonics, 4: 1-35.
Larson, R.L., 1991. Latest pulse of Earth: evidence for a Pindell, J.L. and Barrett, S.E, 1990. Caribbean plate tectonic
mid-Cretaceous superplume. Geology, 19: 547-550. history. In: G. Dengo and J.E. Case (Editors), The Caribbean
Lewis, J.E, 1990. Cuba. In: J.E Lewis and G. Draper (Editors), region. The Geology of North America, Vol. H, Geological
Geological and Tectonic Evolution of the Northern Caribbean Society of America, Plate 12.
Margin, the Caribbean. Geol. Soc. Am., Vol. H, pp. 77-140. Pindell, J.L. and Dewey, J.E, 1982. Permo-Triassic reconstruc-
Malfait, B.T. and Dinkelman, M.G., 1972. Circum-Caribbean tion of western Pangea and the evolution of the Gulf of
tectonic and igneous activity and the evolution of the Mexico/Caribbean. Tectonics, 1:179-211.
Caribbean plate. Geol. Soc. Am. Bull., 83: 251-271. Richard, P. and Krantz, R.W., 1991. Experiments on fault reacti-
Mandl, G., 1988. Mechanics of tectonic faulting, models and ba- vation in strike-slip mode. Tectonophysics, 188:117-131.
sic concepts. In: H.J. Zwart (Editor), Developments in Struc- Roque-Marrero, E and Iturralde-Vinent, M., 1987. Redefinicion
tural Geology, 1. Elsevier, Amsterdam, pp. 1-407. de la zona de Cayo Coco en la provincia de Camaguey. Rev.
Mann, R and Burke, K., 1990. Transverse intra-arc rifting: Pale- Tecnol., XVII (1): 18-22.
ogene Wagwater belt, Jamaica. Mar. Pet. Geol., 7: 410-427. Ross, M. and Scotese, C., 1988. A hierarchical tectonic model
Mann, R, Taylor, E W., Edwards, R.L. and Ku, T., 1995. Ac- of the Gulf of Mexico and Caribbean region. Tectonophysics,
tively evolving microplate formation by oblique collision and 155: 139-168.
sideways motion along strike-slip faults: an example from the Schlager, W. and Ginsburg, R.N., 1981. Bahamas carbonate
northeastern Caribbean plate margin. Tectonophysics, 246: 1- platforms m the deep and the past. Mar. Geol., 44: 1-24.
69. Schlager, W., Austin, J.A., Jr., Corso, W., McNulty, C.L.,
Masaferro, J. and Eberli, G., 1995. Tectonic destruction and FRiegel, C.V., Renz, O. and Steinmetz, J.C., 1984. Early Cre-
segmentation of the Bahamas carbonate platform. Geol. Soc. taceous platform re-entrant and escarpment erosion in the
Am., Abstr. Prog., Session 81, p. A228. Bahamas. Geology, 12: 147-150.
Mattson, RH., 1973. Middle Cretaceous nappe structures in Schlager, W., Bourgeois, E, Mackenzie, G. and Smith, J., 1988.
Puerto Rico ophiolites and their relation to the tectonic history Boreholes Great-Isaac-l, ODP 626 and the history of the
of the Greater Antilles. Geol. Soc. Am. Bull., 84:21-37. Florida Straits. Proc. ODP, Sci. Results, 101: 425-438.
Mattson, RH., 1974. Cuba. In: A.M. Spencer (Editor), Scotese, C.R., Gahagan, L.M. and Larson, R.L, 1989. Plate
Mesozoic-Cenozoic Orogenic Belts. Geol. Soc. London Spec. tectonic reconstructions of the Cretaceous and Cenozoic ocean
Publ., 4: 625-638. basins. Tectonophysics, 155: 27-48.
Mattson, RH., 1979. Subduction, buoyant breaking, flipping and Sheridan, R.E., 1974. Atlantic continental margins of North
strike-slip faulting in the northern Caribbean. J. Geol., 87: America. In: C.A Burk and C.L. Drake (Editors), The Geology
21-37. of Continental Margins. Springer-Verlag, New York, pp. 391-
Meyerhoff, A.A. and Hatten, C.W., 1974. Bahamas salient of 407.
North America. In: C.A. Burk and C.L. Drake (Editors), The Sheridan, R.E. and Osburn, W.L., 1975. Marine geological and
Geology of Continental Margins. Springer-Verlag, New York, geophysical studies of the Florida-Blake Plateau-Bahamas
pp. 429-446. area. In: C. Yorath, E.R. Parker and D. Glass (Editors),

Canada's Continental Margins and Offshore Petroleum Ex- Shipley, T.H., Mayer, L.A. and Winterer, E.L., 1989. Seismic
ploration. Can. Soc. Pet. Geol., Mem., 4: 9-33. reflections and sedimentation in the pelagic equatorial Pacific.
Sheridan, R.E., Crosby, J.T., Bryan, G.M. and Stoffa, EL., 1981. In: A.W. Bally (Editor), Atlas of Seismic Stratigraphy, Vol.
Stratigraphy and structure of southern Blake plateau, Northern 3. American Association of Petroleum Geologists, Tulsa, OK,
Florida Straits, and Northern Bahama Platform from multi- pp. 224-232.
channel seismic reflection data. Am. Assoc. Pet. Geol. Bull., Vaughan, A.E, 1995. Circum-Pacific mid-Cretaceous deforma-
65: 2571-2593. tion and uplift: a superplume-related event? Geology, 23: 491-
Sheridan, R., Mullins, H., Austin, J.A., Jr., Ball, M. and Ladd, 494.
J., 1988. Geology and geophysics of the Bahamas, the Atlantic Walles, EE., 1993. Tectonic and diagenetically induced seal
Continental Margin, Vols. 1-2. U.S. Geological Society of failure within the south-western Great Bahamas Bank. Mar.
America, pp. 329-364. Pet. Geol., 10: 14-28.

Chapter 8
Deformation History of Roat in Island: Implications for the Origin

of the Tela Basin (Honduras)
HANS G. AVI~ L A L L E M A N T and M A R K B. G O R D O N
The rocks of Roat~n Island, Honduras, have experienced several phases of ductile (D1 to D3) and brittle (F1 to Fs)
deformations related to the NE to E passage of the Caribbean plate along the Motagua and Swan Islands faults which lie in
the North American-Caribbean plate boundary zone. The older, ductile structures formed at metamorphic conditions during Late
Cretaceous-early Tertiary left-oblique collision of the Chortfs and Maya blocks. Most brittle structures are younger and formed
after uplift and exhumation of the metamorphic rocks. New isotopic data indicate that the exhumation occurred in Late Eocene
or Early Oligocene time. The most prominent faults on Roat~in Island are NNE- to NNW- and WNW- to ENE-trending normal
faults. The younger NNE- to NNW-trending normal faults may have formed when the island was lying in a releasing bend of
the Swan Islands fault zone, whereas the older normal faults formed by 'transtension' consistent with displacement partitioning
of the relative divergence rate vector along the Swan Islands fault zone. The deformation history of Roat~in Island suggests that
the Tela Basin lying between the island and the mainland of Honduras subsided since Eocene time first by N-S extension and
subsequently by E-W extension.
INTRODUCTION secondary goal of the study was to investigate if

and how the deformation history of Roat~n Island
The Bay Islands of Honduras, consisting from constrains the origin and evolution of the Tela Basin.
west to east of Utila, Roat~in, Barbareta, Guanaja,
and many smaller islands, together with the Swan
Islands to the east are part of the ENE-trending REGIONAL GEOLOGY AND TECTONICS
Bonacca Ridge which lies south of the similarly
ENE-trending Swan Islands fault zone (Figs. 1 and Prominent features of the North A m e r i c a n -
2). This fault zone is a left-lateral transform and Caribbean plate boundary (Figs. 1 and 2) in the
is the southern boundary of the Cayman Trough, region are the western Cayman Trough, the Swan
a deep-marine basin formed by sea-floor spread- Islands fault zone, and the Motagua fault zone. The
ing (Holcombe et al., 1973). South of the Bonacca Motagua fault is currently the most active of the
Ridge lies the Tela Basin on the northern continental left-lateral strike-slip faults cutting across northern
margin of Honduras and the northern margin of the Central America. South of the Motagua fault zone
Nicaraguan Rise. The Cayman Trough, Swan Is- lies the Chortfs block. North of it lies the Maya
lands fault zone, Bonacca Ridge, Tela Basin, and the block (e.g., Dengo, 1969).
northern portion of mainland Honduras are all part of
the North A m e r i c a n - C a r i b b e a n plate boundary zone Chortis b l o c k
(Figs. 1-3). The present study of the rocks of Roat~in
Island was undertaken to investigate their deforma- The Chortfs block (Fig. 1) occupies all of
tion history and relate the structures to the relative Honduras and E1 Salvador, the southern part of
motions between the North American and Caribbean Guatemala, and the northern part of Nicaragua and
plates as the present motions (e.g., DeMets et al., it extends eastward to the Nicaraguan Rise (e.g.,
1994) and Cretaceous-Tertiary ones (e.g., Pindell et Dengo, 1969; Arden, 1975; Case et al., 1984;
al., 1988; Pindell, 1993) are quite well known. A Donnelly et al., 1990). It is thought to be an al-
Caribbean Basins. Sedimentary Basins of the World, 4 edited by P. Mann (Series Editor: K.J. Hsti), pp. 197-218.

198 H.G. AVlff L A L L E M A N T and M.B. G O R D O N
Fig. 1. Simplified tectonic map of the Caribbean from Case and Holcombe (1980). Plate motion vectors are given with respect to a fixed
Caribbean plate (DeMets et al., 1994). Based on the spreading rate of the Cayman Trough (15 mm/yr; Rosencrantz et al., 1988) plus the
slip on strike-slip faults south of the plate boundary, Rosencrantz and Mann (1991) determined that the North America-Caribbean plate
motion is 20 mm/year which is notably higher than the Nuvel-lA rate shown here. Box outlines map of Fig. 2. Abbreviations: OF =
Oriente fault; P R T = Puerto Rico Trench; SIF -- Swan Islands fault.
Fig. 2. Tectonic map of the Gulf of Honduras and adjacent areas (location on Fig. 1) after Pinet (1976), Case and Holcombe (1980),
Manton (1987), and Mills and Barton (1996). Bathymetric contours in km. Thick lines with arrows are strike-slip faults (from Case and
Holcombe, 1980; Manton, 1987; Gordon and Muehlberger, 1994; Mills and Barton, 1996); thick lines with tick marks are normal faults
(from Pinet, 1976; Manton, 1987; Kozuch, 1991). Abbreviations: A F = Agufin fault; B = Barbareta Island; CF -- Chamelec6n fault; EF
= La Esperanza fault; G -- Guanaja Island; GF -- Guayape fault; GoH = Gulf of Honduras; L C F -- La Ceiba fault; M F = Motagua
fault; OF -- Oriente fault zone; R L G -- Rio Lenin graben; R V F -- Rfo Viejo fault; R -- Roat~in Island; SG -- Sula graben; SI = Swan
Islands; SIF -- Swan Islands fault zone; U -- Utila Island. Focal-mechanism diagrams are from Harvard CMT Database (see Table 1).
Box outlines area of Fig. 3.
lochthonous terrane of continental character (Case Mesozoic a n d C e n o z o i c f o r m a t i o n s (e.g., G o r d o n ,

et al., 1984; G o r d o n , 1991) that h a d a c o m p l i c a t e d 1991). I s o t o p i c a g e s o f the m e t a m o r p h i c r o c k s (of-
m i g r a t i o n h i s t o r y ( G o s e , 1985). T h e C h o r t f s b l o c k ten o f q u e s t i o n a b l e q u a l i t y ) v a r y f r o m P r e c a m b r i a n
consists of a generally poorly dated metamorphic ( H o m e et al., 1976a) to C e n o z o i c ( S o u t h e r n w o o d ,
basement overlain unconformably by well-dated 1986). T h e o v e r l y i n g s e d i m e n t a r y units c o n s i s t o f

DEFORMATION HISTORY OF ROATAN ISLAND 199
Fig. 3. Simplified geologic map of Gulf of Honduras region (location on Fig. 2) after McBirney and Bass (1969a), Manton (1987), and
Kozuch (1991). Faults in Tela Basin from Pinet (1976), Tyburski (1992), and Paul Mann (pers. commun., 1996). Bathymetric contours (in
m) from Pinet (1976) and von der Hoya (1986) (200 m contour along Honduran coast and 1200 and 1400 m contours in Tela Basin are
dashed; Tela Basin deeper than 1000 m is shaded. Symbols: i = Late Cretaceous or early Tertiary intrusives; K -- Cretaceous sedimentary
rocks; m --- Precambrian, late Paleozoic, or Late Cretaceous-Tertiary metamorphic rocks; Q --- Quaternary sedimentary rocks; v --
Tertiary or Quaternary volcanic rocks.
Middle Jurassic and Lower Cretaceous clastic rocks, rocks occurred in the Paleozoic, but Precambrian in-
Lower Cretaceous (Aptian and Albian) carbonates, heritance has been shown by Gomberg et al. (1968).
and middle to Upper Cretaceous red beds with The basement is intruded by early Paleozoic granitic
interbedded Cenomanian limestones (Mills et al., plutons (Steiner and Walker, 1996). The sedimentary
1967; Home et al., 1974; Finch, 1981; Gordon, sequence consists of upper Paleozoic siliciclastics
1993). These rocks were intruded by Cretaceous and carbonates, Upper Jurassic to Lower Cretaceous
and Tertiary granitic plutons (Home et al., 1976b; evaporites and red beds, Lower Cretaceous platform
Southernwood, 1986). Cenozoic deposits consist carbonates, Upper Cretaceous flysch, and Tertiary
mostly of volcanic rocks with some intercalated clastic rocks (Burkart, 1994). Late Cretaceous to
(volcani)clastic rocks and red beds (Olson and Mc- Paleogene granitic intrusions have been reported
Grew, 1941; Williams and McBirney, 1969; Heiken (Donnelly et al., 1990; Burkart, 1994).
et al., 1991; Bargar, 1991).
The Nicaraguan Rise is often thought to be the Cayman Trough and the Swan Islands and
easterly extension of the Chortfs block. From west Motagua fault zones
to east the crust thins and becomes more island arc
in character (Arden, 1975). The Cayman Trough (Figs. 1 and 2) is an ENE-
trending, approximately 100-km-wide basin locally
Maya block more than 6000 m deep (e.g., Case and Holcombe,
1980). In the north, the Cayman Trough is bounded
The Maya block (Fig. 1) lies north of the Mo- by the Oriente transform fault zone which contin-
tagua fault zone and occupies the northern part of ues eastward all the way to the Puerto Rico Trench
Guatemala, all of Belize and the Yucatfin Peninsula (Fig. 1) (e.g., Burke et al., 1984; Calais and Mercier
of eastern Mexico (Dengo, 1969). It consists of de Ldpinay, 1991; Leroy et al., 1996). The trough
an igneous-metamorphic basement, overlain uncon- is bounded to the south by the Swan Islands trans-
formably by upper Paleozoic, Cretaceous, and Ceno- form fault zone (Figs. 1 and 2) which continues
zoic sedimentary rocks (e.g., Donnelly et al., 1990; westward through the Gulf of Honduras into Hon-
Burkart, 1994). The metamorphism of the basement duras and Guatemala. The basin formed by sea-floor

200 H.G. AVE LALLEMANT and M.B. GORDON
spreading (e.g., Holcombe et al., 1973; Perfit, 1977). 1969; Kanamori and Stewart, 1978; White, 1991;
N-S-trending magnetic anomalies (e.g., MacDonald Deng and Sykes, 1995) and surface faulting (PlaNer,
and Holcombe, 1978) indicate E - W spreading. Rift- 1976) studies have both indicated that the Swan Is-
ing may have started at 55 Ma (Land, 1979). Results lands and Motagua fault zones are left-lateral trans-
of earthquake focal-mechanism studies are consis- form faults (Fig. 2).
tent with the location and orientation of the spread-
ing center (e.g., Molnar and Sykes, 1969; Deng and Amalgamation of the Chortis and Maya blocks
Sykes, 1995). Spreading rates are difficult to con-
strain, but may have been 15 m m / y e a r since 30 Ma According to Donnelly et al. (1990), the Chortfs
and 30 m m / y e a r from 44 to 30 Ma (Rosencrantz et block, part of the 'Great Arc of the Caribbean' of
al., 1988). Total strike-slip displacements may be of Burke (1988), collided in Late Cretaceous time with
the order of 1100 km (e.g., Rosencrantz and Sclater, the Maya block approximately along the Motagua
1986; Rosencrantz et al., 1988) whereas normal fault fault zone causing partial to total overprint of the
displacements may amount to several kilometers. Paleozoic metamorphic structures in a divergent oro-
The generally ENE-trending Swan Islands fault genic belt with N-vergent thrust faults in the north
zone consists of two en-echelon, right-stepping fault and S-vergent ones in the south. Because of the
segments between which a restraining bend de- post-Eocene displacements along the Motagua fault
veloped with NW-trending folds (Rosencrantz and zone, it is likely that the terrane that collided with
Mann, 1991). On the basis of side-scan sonar and the Maya block is the Nicaraguan Rise rather than
seismic reflection studies, Mann et al. (1991) have the Chortfs block proper. Slabs of Late Cretaceous
shown that the fault segment through the Swan ophiolites (El Tambor Group; McBirney and Bass,
Islands is transpressive. The islands are not only un- 1969b; Donnelly et al., 1990), possibly representing
derlain by a major ENE-trending anticline, but also oceanic crust and mantle that once occurred between
by several ENE-trending, N-vergent thrust faults the Maya and Chortfs blocks, were emplaced both
consistent with the thrust mechanism deduced from north and southward onto the Maya and Chortfs
the earthquake of December 25, 1995 (No. 13 on blocks (e.g., Burkart, 1994). Many plate tectonic
Fig. 2 and Table 1). models (e.g., Pindell, 1993) suggest that the Chortfs
The Swan Islands fault zone trends west as it block and Nicaraguan Rise represent the plate that
continues into mainland Honduras and Guatemala, overrode the Maya block and the intervening ocean
where it is comprised of several major faults, with basin. Based on the age of sedimentary rocks in
the Motagua fault currently the most active (e.g., strike-slip basins, Burkart (1994) suggested that the
Schwartz et al., 1979; Burkart, 1994). Earthquake Motagua fault zone and the other faults parallel to
focal-mechanism studies (e.g., Molnar and Sykes, the Motagua became active in Eocene time.
Table 1
Mechanisms of selected earthquakes which occurred since 1976 in the region from the Harvard CMT Database (Dziewonski et al., 1981)
and a single event from 1969 (Dean and Drake, 1978)
No. Yr Mon Dy Hr Mn Sec Lat Lon D Mo Mw strl dipl rkl str2 dip2 rk2
1 69 2 25 7 39 2.0 15.30 -87.40 24 5.4a 339 42 86 75
2 77 8 20 2 46 11.8 16.61 -86.85 14 1.58 • 1025 6.1 157 68 -168 62 79 -23
3 77 8 20 3 51 54.7 16.70 -86.61 36 3.81 x 1025 6.4 156 61 -168 60 80 -29
4 80 3 20 16 54 18.3 16.84 -85.71 27 1.48 • 10 24 5.4 74 80 9 342 81 170
5 80 8 9 5 45 9.5 15.89 --88.52 22 6.66 x 1025 6.5 352 67 --162 255 74 --25
6 80 9 2 10 28 8.8 15.91 --88.29 25 9.62 x 1023 5.3 176 90 180 266 90 0
7 81 6 11 18 34 20.6 16.72 --86.11 20 8.89 x 1024 5.9 72 72 1 341 89 162
8 82 4 10 16 25 34.1 17.53 --83.37 10 3.71 x 1025 6.3 79 70 --16 175 75 --159
9 90 7 27 0 54 57 16.06 --86.25 15 1.62 • 10 24 5.4 357 17 --136 224 78 --78
10 91 1 29 5 29 3.6 16.88 --85.53 33 1.16 x 1024 5.3 254 81 3 164 87 171
11 92 7 21 15 31 58.7 17.52 --83.66 10 6.29 x 1023 5.2 176 90 --180 266 90 0
12 94 6 8 20 53 25.9 16.76 --85.85 10 2.01 x 10 24 5.5 67 70 --17 163 74 --159
13 95 12 25 18 18 17 17.51 --82.87 10 1.43 • 10 24 5.4 257 37 41 133 67 120
Most mechanisms have been published in Physics of Earth and Planetary Interiors from 1983 to 1997, except for those in press and the
1969 one (No. 1) from Dean and Drake (1978). Abbreviations: No., Yr, Mon, Dy, Hr, Mn, Sec, Lat, Lon, D, Mo, Mw, strl, dipl, rkl, str2,
dip2, and rk2 stand for number, year, month, day, hour, minute, second, latitude, longitude, depth (in km), seismic moment (in dyne-cm),
moment magnitude (calculated from the seismic moment), strike of first fault plane, dip of first plane, slip direction (given as rake) of
first plane, strike of second plane, dip of second plane, and slip direction (rake) of second plane, respectively. Time is given as Universal
Time.
a mb.

DEFORMATION HISTORY OF ROAT/kN ISLAND 201
The northern portion of the Chortfs block is dis- Quaternary deposits. Most of Utila Island is covered
rupted by many approximately ENE-trending strike- by Quaternary deposits, but the most easterly tip
slip faults and N- to NNE-trending normal faults. of the island is underlain by Holocene alkali-basalt
The strike-slip faults are all thought to be left-lateral. (Wadge and Wooden, 1982).
The Rio Viejo fault has left-lateral kinematic indica-
tors (M.B. Gordon, unpubl, data) and an earthquake Lithology of Roatfin Island
focal-mechanism plot shows that the Aguan fault is
left-lateral as well (Gordon and Muehlberger, 1994; Roatan Island is underlain by metamorphic and
see also Fig. 2). Some of the normal faults are seis- igneous rocks. McBirney and Bass (1969a) divided
mically still active, e.g., the Sula graben (Osiecki, the metamorphic rocks into two packages: a high-
1981; Mann and Burke, 1984; White, 1991; Gordon grade package overlain by a low-grade one, the latter
and Muehlberger, 1994) and the Rio Lean graben containing large masses of serpentinite (Fig. 5a).
(Manton, 1987). The fact that the Rio Lean graben They assumed that the contact between the two was
abuts against the La Ceiba fault suggests that the a south-dipping low-angle thrust fault.
latter is a left-lateral strike-slip fault as well. The high-grade, amphibolite facies assemblage
consists of biotite gneiss (often with large plagio-
clase augen), biotite and muscovite schists with
STRUCTURAL GEOLOGY OF ROATAN ISLAND bands and lenses of metagabbro, pyroxene horn-
blendite, amphibolite, and marble. They are invaded
The Bonacca Ridge stretching from Utila Island by many quartz and rare pegmatite veins.
in the west to the Swan Islands in the east (Fig. 2) is The low-grade, greenschist facies assemblage
an ENE-trending ridge separated from the Cayman consists of phyllite, (stretched) pebble conglom-
Trough by the sinistral Swan Islands transform fault erate, chlorite (with or without muscovite) schist,
as well as by a similarly ENE-trending normal fault serpentinite, (meta)sandstone, chert, and marble and
(Figs. 3 and 4). The ridge, consisting of several limestone. The smaller island Barbareta, just east of
en-echelon, right-stepping segments (Pinet, 1976) is Roatan, is underlain by chlorite schist, serpentinite,
mostly submarine, except for the Bay Islands and and granite porphyry.
the Swan Islands. The deepest sites of the ridge
are only about 650 m below sea level (Case and Age constraints
Holcombe, 1980). The ridge is separated from the
Tela Basin by a normal fault. Thus, the Bonacca Marble
Ridge is a horst, but one that has been interpreted We collected a sample of low-grade marble from
to be in part antiformal in structure (Pinet, 1976). the eastern part of Roatan (sample H-94-14; locality
The major Bay Islands, consisting from west to east No. 33; for location see Fig. 5b, Table 2). J. Tom
of Utila, Roatan, Barbareta, and Guanaja islands Dutro, Jr. (pers. commun., 1995) identified gastro-
have been mapped by McBirney and Bass (1969a). pod fragments that constrain the age of the marble
Roatan, Barbareta, and Guanaja islands are underlain to between Triassic and Recent. Because Cretaceous
by metamorphic rocks, a few plutonic rocks, and limestones are common in the Chortfs block, we as-
LI2
~2
L~ Ud
g o
lz
I3
I ~J ~ o 0
SIA 3 ~
,~ Q:
- - 3 Tela Basin ~~176Coymon ~ Trough A/N
~174 ................i :
-I- W/v + q- ~
.......
. . . . . . .
'0 + ' ~~
L' , + + ', ~ r + ' , N + + + ~__z
,o + -.+ +',,+ +
k
+ 2),+ +"" + . I0 km.
.... 7 .
Fig. 4. Speculative N-S cross-section through Cayman Trough, Roat~n Island, Tela Basin, and northern portion of Honduras along line
A-A' on Fig. 3. Thickness of continental crust (crosses) and of oceanic crust in Cayman Trough (v-pattern) from Case et al. (1984);
thickness of sediments in Tela Basin from yon der Hoya (1986). Faults from Manton (1987), Kozuch (1991), and Tyburski (1992);
fault dips are speculative. Faults merging into basal detachment along the Moho-discontinuity (M) is speculative as well9 Earthquake
mechanisms (lower-hemisphere, equal-area) are projected onto a vertical N-trending plane (numbers 3 and 9 refer to same earthquake
plots in Fig. 3).

202 H.G. AVI!~LALLEMANT and M.B. GORDON
.Q
o
L ~ ~
"i 9
0 (D
,,_ .Iti Ol~
0 1,1 " " 9
~| ~ --~-~
9,-- Y:,,| ~ ? 9
| ,..
L 0, "0~ ,.,
~ . . _
oR E br
114111111 \ ',..(,
9
.,..o
9
.,,=~
~o ' glo 90 ' 'gloge~

~
~..,.~
I'
...-:.
,,0s '~o9 "~163

,,z~>,,ge..
=I-"
<~
$
I" ~g
eat
0
7~
.,..a
.,.,~
i-i
.Io
~=-~-.><-....,b ==i
) I=
. i ~~v,.~ "-
i=
v, ~'..~'~-.-.,d~
V', ~ *.W~"..,#
~ ~
#
b
i
i

Table 2
Geographic location of localities mentioned in the text and figures
Site No. Site name UTM coordinates Latitude Longitude Rock type Result
oN oW
1 Walk Point 16QEP741163 16~ ' 86018.4' metaconglomerate fault mversmn

4 Wild Mines 16QEP514046 16~ 86~ muscovite schist fault inversion
4B Wild Mines 16QEP511043 16o19.2' 86031.3 ' muscovite schist fault inversion
5B Jonesville Ranch 16QEP678141 16o24.5' 86021.9' metaconglomerate fault inversion
7 Hong Kong Hill 16QEP651127 16023.7' 86023.5 ' amphibolite fault mverslon
7B Further Wetl Hill 16QEP646126 16o23.7' 86023.7 ' amphibolite fault inversion
8 West End Point 16QEN424984 16~ ' 86036.2 ' amphibolite and marble 4~ sample, conodont
analysis, Fig. 10c
11 Bodden Bight 16QEP674116 16024.0' 86~ metalimestone conodont analysis
12 Blue Rock 16QEP892172 16~ ' 86009.9 ' serpentinite fault inversion
12A Smooth Bay 16QEP906182 16o26.6' 86~ ' granite porphyry zircon fission-track sample
13 Coxen Hole 16QEP503039 16o19.0' 86031.8 ' chlorite schist fault inversion
14 Coxen Hole 16QEP494043 16o19.2' 86032.3 ' serpentinite fault inversion
15 Coxen Hole 16QEP495048 16o19.4' 86032.2' serpentinite fault inversion
20B Big Bight 16QEP602103 16o22.4' 86026.2 ' biotite gneiss Fig. 10a,b
22 Big Plan 16QEP610104 16o22.5' 86025.7 ' quartz-feldspar schist fault inversion, Fig. 14
24 Mid Isle 16QEP617109 16022.7' 86025.3 ' amphibolite and pegmatite fault inversion
24C Mid Isle 16QEP622108 16022.7' 86~ ' amphibolite and pegmatite fault inversion
25 Key Point 16QEN435998 16~ 86035.6 ' metalimestone conodont analysis
25C West End 16QEN428988 16o16.2' 86036.0 ' metalimestone conodont analysis
27 Mid Isle 16QEP630110 16022.8' 86024.6' biotite gneiss fault inversion, Figs. 15, 16
32 Dixon Hill 16QEP503052 16o19.7' 86031.7 ' chlorite schist fault inversion
33 Diamond Rock 16QEP709148 16024.9' 86020.2 ' metalimestone fossil analysis
(Jurassic-Recent)
34 Alligator Nose 16QEP732164 16025.7' 86o18.9 ' metalimestone and chlorite fault inversion and conodont
schist analysis
sume that the Roat~in limestone is Cretaceous as well Ductile deformation structures
in agreement with this fossil identification. We col-
lected several other samples of low-grade limestone Most rocks o f Roatfin I s l a n d h a v e u n d e r g o n e
for conodont analyses, but all samples were barren. several ductile f o l d i n g (D) and brittle faulting (F)
p h a s e s o f d e f o r m a t i o n (Table 5). T h e first ductile
Amphibolite d e f o r m a t i o n (D1) is s y n m e t a m o r p h i c w h i l e the sec-
A sample of amphibolite was collected from the o n d (D2) and third (D3) p o s t - d a t e the p e a k m e t a -
w e s t e n d o f Roat~in I s l a n d ( s a m p l e n u m b e r H - 9 3 - 1 2 ; morphism and refold the first g e n e r a t i o n struc-
locality No. 8; for location, see Fig. 5b, Table 2). tures. Five g e n e r a t i o n s o f faults (F1 to Fs) w e r e
H o r n b l e n d e o f this r o c k w a s d a t e d b y the 4 ~
'ot
m e t h o d and y i e l d e d a L a t e E o c e n e p l a t e a u age o f
36.0 :i: 1.2 M a (Fig. 6, Table 3). I I I I I I I I
7O
H-93-12 hornblende
Granite porphyry
A sample of granite porphyry was collected from 60
m
B a r b a r e t a I s l a n d ( s a m p l e H - 9 3 - 1 3 ; l o c a l i t y No. 13; v
50--
for location, see Fig. 5b, Table 2), but it c o n t a i n e d ,,r
ol m
insufficient z i r c o n for a U - P b age d e t e r m i n a t i o n . 40--

r
T h e z i r c o n c o n t e n t w a s sufficient t h o u g h for fission- I--

30--
Q.
track a n a l y s i s (Table 4), w h i c h p r o v i d e d an age o f Q.
<: 20-- 36.0 + 1.2 M a
39.4 :k 2.8 M a ( L a t e E o c e n e ) .
10--
Conglomerates
O-- I I I I I I I I I
McBirney and Bass (1969a) assigned a Tertiary 0 20 40 60 80 100
age to deposits of apparently unmetamorphosed con- Cumulative % 39 Ar Released
glomerates and sedimentary breccia from a small
Fig. 6. 4~ release spectrum of hornblende from amphi-
island off Guanaja Island. The cobbles and frag- bolite of Roat~in Island (locality No. 8; location on Fig. 5b and
ments contain fossils of Cretaceous age. Table 2; more data in Table 3).

204 H.G. AVE LALLEMANT and M.B. GORDON
-d
m.
qq
9
0
0
<
"0
,<
o'b
"0
O~
o
=o
0
9
70 7 07 7~7 7 7 707 07 7 7~7 7~ 70 0 ~ 0 0 0 0

7
,<
0
o'b X X X X X X X X X X X X X X X
0
0
9
o
0
0
o
O~
,< 0
0
r r 0 u'b o'b ,--~ c q O~ 0 ~'b t'~ ~ O~
,<
<
0 <
0
"0
0
o
<
0
,< r 0 0 0 0 ~ 0 ~ 0 ~ 0 0 0 ~
00
"0
.< 0
9 <
r~ "0
o
0
o [-
0 0 0
0
X X
9
OO 0 ~rb ,~" O~ OO t'~. ~ r r r r r r r
0 ~ ~ 0 "0
, - d d o o o o d o d o o d o o
0
~ It3 0
0 0
"0
O O U ' b O 0 0 0 0 0 0 0 0 0 0 0
~ o. o. o o t'.. t"~ OO O~ O O O ,"~ ~ ~ C',l C",l O'b c~'b ~'~
9 0
0
~dd~
"0 "~ "~' ~" ~" 6
m
,<
~~~
<<<
~ ~'
0 0 0
r
0
r
0 0 0
r t'xl t'~l C',I r

0
sssssssssssss
0 0
r
0 0
r r
O 0
r r
~
cq
2:
<<< 0
C',I ~',1 ~',1 ~',1 ~,1 t"q t",l ~q ~-1 ~1 ~',1 ~,1 ~,1 t',l ~',1 9

Table 4
Zircon fission-track data from granite porphyry on Barbareta Island
Sample Mineral No. of grains Standard track density Fossil track density Inducedtrack density Chi square prob. Fission-track age
(xl06 cm-2) (xl06 cm-2) (xl06 cm-2) (%) (Ma)
H-93-13 zircon 11 0.21 556.30 502.42 89 38.9 4- 2.8
(1359) (413) (373)
Locality No. 12b, for location see Fig. 5b and Table 2. Brackets show number of tracks counted. Standard and induced track densities
measured on mica external detectors (g - 0.5), and fossil track densities on internal mineral surfaces. Ages determined by Ann E. Blythe
using r = 335 + 20 (zircon) for dosimeter glass SRM 962a (e.g. Hurford and Green, 1983). Ages are pooled.
Table 5
Summary of deformation structures of Roatfin Island and their kinematic interpretation
Phase Structure Orientation Kinematic interpretation

D1 S1 cleavage NE strike, moderate SE dip Shear on Sl with top to SW sense
B1 fold axis shallow SW plunge
L1 lineation shallow SW plunge
D2 $2 axial planes (1) E-W strike, moderate S dip NNW-SSE contraction
(2) N-S, strike, moderate W dip
B2 fold axis shallow SW plunge
L2 lineation shallow SW plunge
D3 $3 kink bands NNW strike, steep ENE-WSW contraction
B3 fold axis shallow NNW and SSE plunge
Fl Thrusts faults (1) N-S strike, shallow W dip E-W contraction
(2) N-S strike, shallow E dip
F2 Strike-slip faults (1) (LL) NNW strike, steep NNE-SSW extension
(2) (RL) ENE strike, steep
F3 Normal faults (1) E-W strike, moderate N dip N-S extension
(2) E-W strike, moderate S dip
F4 Strike-slip faults (1) (RL) NW strike, steep E-W extension
(2) (LL) NE to ENE strike, steep
F5 Normal faults (1) N-S strike, moderate W dip E-W extension
(2) N-S strike, moderate E dip
LL and RL stand for left-lateral and right-lateral, respectively.
observed to cut across the ductile deformation struc- dipping to the south and west (Fig. 7b, Fig. 9). In
tures. fact, the entire island is a major D2 antiform which
The first recognized phase of folding on Roatfin plunges to the west-southwest (Figs. 8 and 9).
(D1) is synmetamorphic and is related to myloniti- Rare kink bands (D3) refold all the previous
zation. It is expressed by small isoclinal folds (B1) structures and their axial planes ($3) are steep and
with a strong axial-planar cleavage (S1) and min- strike NNW. Kink axes (B3) are sub-horizontal and
eral, stretching, and intersection lineations (L1) all plunge to the north-northwest or south-southeast
parallel to the fold axes (Fig. 7a, Figs. 8 and 9). (Fig. 7c, Figs. 9 and 10b).
Generally the cleavage trends ENE, parallel to the Inspection of the geological map of Guanaja Is-
long axis of Roatfin Island and it dips moderately land of McBirney and Bass (1969a) suggests that the
to the south-southeast. The lineations and fold axes rocks of Guanaja have undergone at least the D1 and
tend to plunge gently to the west-southwest. The D2 deformations. The orientations of these structures
deformation is clearly non-coaxial as plagioclase are very similar to the D1 and D2 structures on
augen in biotite gneiss and quartz lenses in mar- Roat~n, implying that no appreciable vertical-axis
ble have asymmetric tails indicating shear with a rotation has occurred between the two islands.
top-to-the-SW sense (Fig. 10). Metamorphic rocks south of the Motagua fault
The S1 cleavage is folded by the D2 deformation, zone in Honduras have all three (D1, 02, and D3)
expressed by a conjugate set of chevron-style folds folds approximately of similar orientation as those
with fold axes (B2) generally coaxial with the first on Roatfin Island, although the $1 cleavage is steeper
fold axes (B1) and axial planes ($2) moderately on the mainland (unpubl. data, M.B. Gordon and

206 H.G. AV]~LALLEMANT and M.B. GORDON
N b N ?
II 9 | | I ~
| 9
9 9 II
illl I I 9 || e 9
e . a# e ~e la" el
9 d | i 9 . |
" 0 | e |
D, D2 D~
Fig. 7. Equal-area, lower-hemisphere projections of mesoscopic ductile fabric elements from Roat~in Island. Solid symbols are poles to
planes (dot = cleavage; square - axial plane of fold); open symbols are lineations (dotted circle = fold axis; open circle --- stretching
lineation [arrows show sense of displacement; arrows pointing away from center of diagram show 'normal' motion and arrows pointing
toward center show 'thrust' motion]; triangle -- mineral lineation; square -- intersection lineation). Diagrams (a), (b), and (c) show the
data of the first (D1), second (D2), and third ductile deformation (D3), respectively.
N b N
Fig. 8. Equal-area, lower-hemisphere projections of poles to D1 cleavages (S]) in southeastern (a) and northwestern domain (b); domain
boundary follows the divide of Roat~in. Contours in (a) at 3, 6, 9, 12, 15, and 18% per 1% area and in (b) at 5.5, 11.1, 16.5, 22.2, and
27.8% per 1% area. Dashed great circle connects the two maxima of (a) and (b); dotted circle is pole to dashed great circle and probably
megascopic B2 fold axis.
or tilting of the Bay Islands probably has occurred

though (see below).
Brittle deformation structures
Roat~in Island is covered by tropical vegetation

and plantations. Therefore mesoscopic analysis of
brittle structures relies mostly on road and coastal
outcrop. Megascopic faults that might be present are
not exposed because of erosion and vegetation. We
Fig. 9. Interpretative block diagram showing schematically the
did not observe the low-angle thrust fault inferred
D1 to D3 structures as shown in Fig. 7. S and L stand for axial by McBirney and Bass (1969a) between the low-
planes or cleavage and lineation, respectively, with subscripts re- and high-grade metamorphic rocks. However, rare
ferring to the phases D1, D2, and D3. L1 (lozenges) is stretching mesoscopic thrust faults do occur near the contact.
lineation; L2 and L3 are intersection lineations ($2 with $1 and
$3 with $1, respectively).
Aerial photographs
H.G. Av6 Lallemant, 1994). Thus, vertical-axis rota- To assess the presence of megascopic faults, aerial
tions between the mainland and the Bonacca Ridge photographs were analyzed to find preferred orien-
have not occurred either. Horizontal-axis rotation tations of lineaments. Lineaments on aerial pho-

DEFORMATION HISTORY OF ROAT,/~N ISLAND 207
Fig. 10. Photographs of biotite gneiss [(a) and (b); locality No. 20B, see Fig. 5b, Table 2] and marble [(c); locality No. 8; see Fig. 5b,
Table 2]. Asymmetric plagioclase augen in gneiss and asymmetric quartz lenses in marble show top-to-the-WSW shear (top to the left).
Kink fold in (b) is D3 structure.
tographs may be the result of lithological contacts, ered in interpreting aerial photographs taken at low
but generally they result from differential weathering angles of incidence of sun rays: (1) shadows may
along fault planes. Two factors have to be consid- cause apparent lineations; and (2) linear valleys may

208 H.G. AVI~ LALLEMANT and M.B. GORDON
N nally consistent sets. In this study, the results of the

inversions are not given as principal stresses, but as
principal strains, because stresses are calculated and
thus derivatives of the displacements and strains. The
fault data are presented in Fig. 12 and Table 6, while
simplified block diagrams are shown in Fig. 13.
Our fault data from Roat~in and Barbareta Islands
can be separated into five sets, but these sets do
not occur in all outcrops. We found some relative
age relationships between the fault sets, but not
sufficient to uniquely constrain the relative timing of
the deformation history.
No relative age relationship between the F1 fault
I0 5 0 5 I0
% / I0" set and the other sets could be established. The only
Fig. 11. Rose diagram of trends of lineaments determined from relationships that are certain between the fault sets
air photos (Fig. 5b) expressed as percentage of lineament length F2 to F5 are: (1) F2 > (is older than) Fs; (2) F3
per 10~ Total length of lineaments is 241 km. Note the large > F4; and (3) F4 > Fs. There are thus only three
maximum of 12.2% at N25W and two lesser maxima of 9.6%
possibilities for sets F2 to Fs: (1) F2 > F3 > F4 >
and 6.4% at N85W and N65E, respectively.
Fs; (2) F3 > F2 > F4 > Fs; and (3) F3 > F4 > F2
> F5 (Angelier, 1991b). How then was the relative
be invisible if they are parallel to the direction of timing scheme F1 > F2 > F3 > F4 > F5 chosen?
incidence, but valleys perpendicular to that direction Timing relationships of set F1 are completely absent,
are strongly accentuated. Fortunately, both factors but the kinematics of this set of faults are compatible
are not important in the present study, because the with those of the third phase of folding (D3). None
photos were taken on April 1, 1989 between 11:00 of the other fault sets is compatible with the earlier
AM and 1 : 0 0 PM.; thus, the incidence angle of the folding phases (D] and D2). Thus we speculate that
sun rays was generally greater than 75 ~. the F] faults are the oldest brittle structures. In
The lineaments observed in the aerial pho- all areas F5 is evidently the youngest. Because the
tographs were plotted on the geological map of strains of F4 and F5 are similar (both are related to
Roat~in Island on Fig. 5b. The orientation and length E - W extension) and because both sets F2 and F3 are
of the lineaments were measured and plotted in a characterized by N - S extension, we suggest that the
rose diagram (Fig. 11) showing that the most com- first scheme (F1 > F2 > F3 > F4 > F s ) a p p l i e s to
mon lineaments trend N25W, almost perpendicular Roat~in Island.
to the Swan Islands fault zone. Two lesser-preferred Only four F1 thrust faults (Fig. 12a, Fig. 13a)
orientations are N85W and N65E, the latter direction were found and only in a small area (localities Nos.
parallel to the Swan Islands fault zone. 13, 14, and 15; see Fig. 5b, Table 2). They indicate
E - W contraction. They occur near the contact of
Mesoscopic faults the high- and low-grade metamorphic rocks that
As many fault planes and slickenside striations as McBirney and Bass (1969a) had interpreted as a
possible were measured in each outcrop or locality, thrust fault. Although our F1 data are consistent with
with the sense of slip of each fault determined by that interpretation, the superposition of low-grade
techniques described by Petit (1987) and Twiss and on high-grade metamorphic rocks suggests that the
Gefell (1990). The relative age of a fault with respect contact is a low-angle normal fault.
to other faults was determined by studying crosscut- The mesoscopic F2 faults (Fig. 12b, Fig. 13b)
ting relationships of the fault planes or slickenside occur in two orientations. The steep NNW-striking
striations. Using the stress inversion technique of An- ones are left-lateral strike-slip faults, whereas steep
gelier (1990, 1991 a), the fault data for each locality ENE-striking ones are right-lateral strike-slip faults.
or a set of localities were separated into several inter- These structures indicate N N E - S S W extension.
Fig. 12. Equal-area, lower-hemisphereprojections of mesoscopic faults (great circles) and slip directions; small arrows pointing outwards
indicate normal faults, pointing inwards indicate thrust or reverse faults; half arrows indicate strike-slip faults (sense of displacement
follows usual practice). Fault analysis was performed using the method of Angelier (1990, 1991a); resulting principal strain axes are
shown as 3-, 4-, and 5-pointed stars indicating, respectively, the maximum principal extension (X), the intermediate principal strain axis
(Y), and the maximum principal shortening axis (Z). Large solid arrows pointing outward show the trend of the X-axes and those pointing
inward are the trend of the Z-axes. Localities where fault analysis was performed are shown here and in Table 2. Diagrams (a) to (e)
show F1 to F5 faults, respectively. Symbols: triangles = poles to joints; open squares = poles to tension gashes; solid squares = fiber
lineations.

DEFORMATION HISTORY OF ROAT/~N ISLAND 209

210 H.G. AVt~ LALLEMANT and M.B. GORDON
t",l
tt~
~h
9
.,-~
b - .... ~ ~ /__ ,,1 0
Zd
.,..~
~5
.,..~
,.o
~
,.o
Fig. 13. Interpretative block diagrams showing graphically the 9

five sets (F1 to Fs) of mesoscopic faults depicted in Fig. 12.
(a) F1 thrust faults resulting in E-W shortening. (b) Set of
conjugate strike-slip faults (F2) resulting in N-S extension. (c)
Set of normal faults (F3) resulting in NNE extension. (d) Set of
conjugate strike-slip faults (F4) causing E-W extension. (e) Set
of conjugate normal faults (Fs) causing E-W extension.
The F3 faults are WNW- to ENE-striking normal 9
faults dipping moderately to the north and south

(Fig. 12c, Fig. 13c). They are consistent with NNE
to NNW extension suggesting that F3 is related to F2. ,.o
These faults may be correlated with the N85W and

N65E lineaments observed on the aerial photographs
(Fig. 5b, Fig. 11). Well developed vertical joints 2= c~
striking parallel to the faults may be related to the F3

event.
The F4 faults consist of steep NE- to E-strik-
ing left-lateral strike-slip faults and steep NW- to t",l
N-trending left-lateral ones (Fig. 12d, Fig. 13d).

These faults indicate E - W extension. The NE- to
ENE-striking left-lateral faults are parallel to the
Swan Islands fault and they may be correlated 9
with the N65E lineaments on the aerial photographs

(Figs. 5 and 11).
~.~
The F5 structures are N- to NNW-striking normal
faults (Figs. 14-16) and may be related to F4 as
displacements along both caused E - W extension. ~.~
This generation of faults is by far the most common ~'~
structure on Roatfin Island and can be seen in many
roadside outcrops (e.g., Fig. 14). These faults are
not only the most numerous ones as the aerial
photographs show (Fig. 5b, Fig. 11), but they appear N~
to have caused very large strains. In one outcrop
(Fig. 15, Fig. 16c) E - W extensile strains are at least r~
80%. North-trending joints and tension gashes with

W-trending mineral fibers (generally quartz) may be 0
related to this deformation. 9

DEFORMATION HISTORY OF ROAT.A,NISLAND 211
Table 6
Orientation data and statistics of fault analysis of Roatdn and Barbareta Islands using inversion technique of Angelier (1990, 1991 a)
Phase Program o1 02 03 fi0 Ang Rup No. faults Site name Site number
5 INVD 277.0 86.3 170.0 1.1 79.9 3.5 0.558 2 12 5 Alligator Nose 34
5 INVD 11.9 62.1 180.9 27.4 273.3 4.5 0.244 5 24 6 Blue Rock 12
3 INVD 351.1 79.0 97.2 3.1 187.8 10.6 0.795 8 18 5 Blue Rock 12
4 INVD 3.9 32.7 154.5 53.6 264.5 14.2 0.511 3 30 4 Blue Rock 12
5 INVD 113.6 73.1 342.6 11.3 250.1 12.4 0.395 6 19 7 CoxenHole 14
5 INVD 230.1 71.3 14.5 15.4 107.4 10.4 0.390 9 29 10 Coxen Hole 15
4 INVD 22.6 18.0 157.9 65.4 287.2 16.2 0.771 12 44 5 CoxenHole 15
2 INVD 286.4 4.3 29.5 71.7 195.1 17.8 0.372 12 27 7 CoxenHole 13, 14,15
3 INVD 17.5 80.4 283.0 0.8 192.9 9.6 0.402 9 31 12 Coxen Hole 13, 14, 15
1 INVD 270.1 0.4 180.1 8.7 2.5 81.3 0.409 25 54 4 Coxen Hole 14, 15
4 INVD 187.7 27.1 41.2 58.5 285.5 14.9 0.428 9 37 7 Coxen Hole 13, 14, 15
5 R4DT 272.2 73.2 157.4 7.2 65.4 15.1 0.537 2 37 5 Hong Kong Hill 5B, 7
5 R4DT 156.0 87.2 357.2 2.6 267.2 1.0 0.159 13 29 4 Hong Kong Hill 7
2 INVD 294.4 32.3 88.7 55.0 196.6 12.2 0.697 5 22 13 Mid Island 24C
3 INVD 296.1 83.2 111.2 6.8 201.3 0.6 0.540 7 22 8 Mid Island 24C
5 INVD 357.8 76.6 157.9 12.6 248.9 4.4 0.205 15 30 4 Mid Island 24C
4 INVD 189.8 21.1 302.3 44.8 82.4 37.7 0.387 8 31 4 Mid Island 24C
5 INVD 39.3 71.2 184.4 15.6 277.3 10.2 0.456 4 15 6 Mid Island 27
3 INVD 295.2 84.1 111.5 5.9 201.5 0.4 0.533 7 21 9 Mid Island 24C, 27
5 INVD 52.1 70.7 185.5 13.5 278.8 13.5 0.421 5 27 12 Mid Island 24, 24C, 27
3 INVD 120.3 87.4 332.6 2.2 242.5 1.4 0.638 10 32 6 Big Plan 22
4 R4DT 221.4 1.2 352.5 88.1 131.3 1.4 0.425 9 50 8 Walk Point 1
3 INVD 302.3 63.7 63.9 14.5 159.7 21.4 0.219 4 30 5 Wild Mines 4
5 INVD 44.4 70.3 206.0 18.8 298.0 5.8 0.625 13 41 7 Wild Mines 4
3 INVD 305.5 63.7 70.7 15.9 166.8 20.4 0.160 5 30 6 Wild Mines and Dixon Hill 4, 32
5 INVD 44.7 74.2 206.2 15.0 297.5 4.8 0.563 13 37 9 Wild Mines and Dixon Hill 4, 32
Note that in the text, principal strain axes X, Y, and Z are being used instead of 03, 0"2, and 0.1, respectively.
,- i i
I EN E ~ ~ - ---- -- ~ - = - - ~ ~ ~ ~~-,~~'-%=-- ~- - - : W SW
. ----...5.
Fig. 15. ENE-trending road section showing stretching (F5 deformation) of competent quartz-feldspar schist (boudins A to D) in incom-
petent phyllite. Extensional strain parallel to outcrop measured from center of boudin B to center of boudin D is about 80%. Note quartz
in pressure shadows of boudins and the late F5 faults post-dating the boudinage; in locality No. 27 (location on Fig. 5b and Table 2).
Tectonic evolution of Roat~n Island and Av6 Lallemant, 1995). Thus, Roat~in Island may
have been displaced eastward for much less than
It has been proposed that left-lateral displacement 1100 km, but for how much cannot be ascertained
along the Swan Islands and Motagua fault zones because the metamorphic rocks on Roatdn Island are
since the 55 Ma (Early Eocene) initial opening of lithologically different from those in the Chortfs and
the Cayman Trough (e.g., Rosencrantz et al., 1988) Maya blocks. The rocks most similar to the rocks
is about 1100 kin. Thus, Roat~in Island could have exposed on Roatdn are the ophiolitic rocks of the
originated far to the northwest off the west coast E1 Tambor Group (first defined as the E1 Tambor
of Mexico. However, the island may not always Formation diabase dikes, basalts and chert have not
have been part of the Caribbean plate, but could been encountered in the Bay Islands. However, they
at times have been attached to the North American are relatively rare in the Motagua Valley and may
plate, if major left-lateral strike-slip displacements be metamorphosed beyond recognition in the Bay
occurred along the Agu~in, La Ceiba, La Esperanza, Islands. The chlorite schist found in the Bay Islands
or Rio Viejo faults (Figs. 2 and 3; e.g., Gordon is not generally considered part of the E1 Tambor

212 H.G. AVI~ L A L L E M A N T and M.B. G O R D O N
Fig. 16. Photographs of F5 structures. (a) Boudins of quartz-feldspar gneiss in mica schist showing ENE extension; locality No. 8, Fig.
5b). (b) Northward view of conjugate set of N- to NW-trending normal faults in quartz-feldspar gneiss with quartz veins at the fault
intersection (locality No. 22, Fig. 5b). (c) Southward view of boudins of quartz-feldspar gneiss in mica schist with quartz in the pressure
shadows (locality No. 27, Fig. 5b and 15).

Group. However, Donnelly et al. (1990) state that the west-southwest. They are related to NNW-SSE
the San Diego phyllites of the Motagua Valley are contraction and thus may have formed during the
commonly difficult to distinguish from rocks of Late Cretaceous collisional event as well. The S1
the E1 Tambor Group; thus this group is similar cleavages on Roatfin dip moderately to the south-
to the rock sequence found in the Bay Islands. southeast and the sense of shear is ENE over WSW.
Wilson (1974) showed that the 'Sanarate limestone' To bring these structures into coincidence with the
is in normal contact with the graywacke-chlorite D1 structures in the Motagua fault zone, the island
schist of the E1 Tambor Formation, and Lawrence must have been tilted to the south which is most
(1975) included it in the E1 Tambor Group calling easily done by back thrusting (Fig. 17b). Loading by
it the Cerro de la Virgen Limestone. Wilson (1974) this back thrust may have caused subsidence to the
found rudist fragments of certain Cretaceous age and south and the birth of the Tela Basin.
probable Coniacian-Campanian age. Thus a major
metalimestone unit of the E1 Tambor Group may D3 folds and F1 faults
also be similar to the metalimestone on Roatfin. During the third folding phase kink folds devel-
oped that plunge moderately to the north-northwest
Displacement partitioning and south-southeast. The F1 faults are N-trending
It is at present well known that deformation struc- thrust faults. Both folds and faults caused E - W con-
tures in boundary zones of two obliquely converging traction that may have taken place in a restraining
plates are not directly related to the relative con- bend along the Motagua fault zone (Fig. 17c). The
vergence rate vector, but to the two components F~ faults are the only mesoscopic thrust faults that
parallel and sub-perpendicular to the boundary (e.g., we encountered, and although they may be related to
Fitch, 1972; Walcott, 1978; Beck, 1983; Av6 Lalle- the megascopic thrust fault that McBirney and Bass
mant and Guth, 1990; McCaffrey, 1991; Cashman et (1969a) proposed for the contact between the high-
al., 1992). Many deformation structures on Roat~in and overlying low-grade metamorphic rocks, such
Island can be interpreted to be the result of parti- contact is better explained as a low-angle normal
tioning of the convergence rate vector between the fault (see below).
Caribbean and North American plates. Other struc-
tures seem to indicate that during oblique divergence Unroofing
the relative plate motion vector was partitioned as The D2 and D3 deformations could have been
well. responsible for uplift of the metamorphic rocks of
Roatfin. Erosion, tectonic denudation, or both caused
D~ folds the exhumation. The uplift, exhumation, and cool-
The original metamorphic ages of the rocks along ing through the closure temperatures of hornblende
the Motagua fault zone are Precambrian to Paleozoic (500 ~ to 550~ McDougall and Harrison, 1988)
(Gomberg et al., 1968), but they were overprinted and zircon (fission track; ~280~ M. Brix, pers.
during Late Cretaceous time as a result of the commun., 1996) occurred in Late Eocene to Early
collision of the Chortfs block with the Maya block Oligocene times (36 Ma 4~ hornblende and
(Donnelly et al., 1990; Burkart, 1994). Although 39 Ma zircon fission track). The contact between the
N- and S-vergent thrust faults, north and south of high-grade and the overlying low-grade metamor-
the Motagua fault zone, respectively, are the main phic rocks has been interpreted by McBirney and
results of the collision, mylonites resulting from Bass (1969a) as a thrust fault. We observed a few
left-lateral strike-slip displacement are ubiquitous in mesoscopic thrust faults near the contact consistent
the fault zone (Meschede et al., 1993). We propose with their model. Generally, however, such contacts
that the metamorphic structures along the Motagua are interpreted as low-angle normal faults. We found
fault zone can be correlated with the D~ structures many mesoscopic F3 normal faults with the correct
on Roatfin Island. We believe that the convergence orientation for a megascopic normal fault near the
between the two blocks was left-oblique and that contact (Fig. 12c), and although most F3 faults are of
the thrust faults in the Chortfs and Maya blocks are Miocene and younger age (see below), tectonic de-
related to the plate boundary-normal component of nudation and exhumation of the metamorphic rocks
convergence, while the mylonites in the Motagua cannot be ruled out.
fault zone and the Bay Islands are related to the plate
boundary-parallel vector component (Fig. 17a). F2 and F3 faults
The F2 faults are a mostly conjugate set of strike-
D2 folds slip faults and the F3 faults a mostly conjugate
The second generation of folds on Roatfin re- set of normal faults (Fig. 12b,c, Fig. 13b,c). They
folded the D~ folds and the mylonites. They are are discussed together, because they both caused
chevron-type folds with axes plunging moderately to NNS-SSW to NNW-SSE extension (transtension).

214 H.G. AVI~ L A L L E M A N T and M.B. G O R D O N
The formation of F3 faults may be related to dis- imately conjugate normal faults that strike NNW
placement partitioning. Heubeck and Mann (1991) to NNE. It is the most penetrative fault system on
have shown that most Caribbean plate motion mod- Roatfin Island: e.g., Fig. 14 shows a representative
els (e.g., Stein et al., 1988) applied to the Swan E - W road section with numerous, approximately
Islands fault zone imply extension along the fault NNW-trending normal faults and tension gashes (see
zone (at least since 5 Ma). It is possible that the also Fig. 12e, locality No. 22) and Fig. 15 shows
relative plate motion vector (possibly oriented E - W ) a road cut with E-W-trending extensile strain of at
was partitioned into a component of left-lateral slip least 80% (see also Fig. 16c). Normal faults of this
along the Swan Islands fault zone and a component orientation can easily form near major strike-slip
perpendicular to the fault causing the formation of faults (e.g., Harding, 1974), but they tend to dis-
normal faults parallel to the Swan Islands fault zone appear at short distances from the strike-slip fault.
on and just south of Roat~in Island. The WNW trend However, similar normal faults occur across the Tela
of many normal faults may be related to the WNW Basin in northern Honduras (Figs. 2 and 3). Further-
trend of the Honduran coast line (Fig. 17d). The F2 more, ENE-trending strike-slip faults extend toward
faults could be interpreted as tear faults. As the F3 the northern Honduran coastline, but are not shown
faults are related to the major phase of subsidence to continue into the Tela Basin (Kozuch, 1991). We
of the Tela Basin, the age of the faulting is Miocene speculate that they do continue into the basin and
(vonder Hoya, 1986). that the E - W extension is related to left-stepping
of the Swan Islands fault zone, and thus, that the
F4 and F5 faults Tela Basin during F4 and F5 faulting is a pull-apart
The F4 and F5 faults both caused approximately basin (Fig. 17e). As the Swan Islands fault as well
E - W extension. The F4 phase consists of a mostly as the La Ceiba and Agu~in faults are seismically
conjugate set of strike-slip faults, one set of which active, it seems that E - W as well as N-S extension
(the left-lateral one) parallels the Swan Islands fault is occurring at present.
zone. The F5 phase consists of a set of approx-
Fig. 17. Tectonic model of deformation history of Roat~in Island

and Tela Basin. (a) Late Cretaceous to early Tertiary NE con-
vergence of Chortfs and Maya blocks was left-oblique resulting
in displacement partitioning: Chortfs block is thrust northward
along fault No. 1 upon Maya block and it moved eastward
along E-W-trending proto-Motagua left-lateral strike-slip fault
(fault No. 2) (D1 deformation). (b). Major southward tilting and
folding (D2) of Roat~in Island segment of proto-Motagua fault
zone caused by back thrusting along fault No. 3 which may
have caused loading and subsidence and the origin of the Tela
Basin; strike-slip displacements occur now along fault No. 4. (c).
E-W contraction, possibly in restraining bend, resulted in N -
S-trending (D3) folds and (F1) thrusts. During the Late Eocene
deformation D2 and/or D3 may have caused uplift followed
by erosional exhumation of metamorphic and igneous rocks of
Roat~in Island (MF = Motagua fault; PF = Polochic fault). (d).
In Oligocene time, the relative displacement direction between
the North American and Caribbean plates changed to approxi-
mately E-W causing left-oblique divergence along the Motagua
(MF) and Swan Islands fault (SIF) zones. Displacement parti-
tioning resulted in strike-slip displacements along these faults
and normal faulting along parallel ENE-trending faults; WNW-
trending normal faults formed along the Honduran coast line.
Displacement along these normal faults caused the second phase
subsidence of the Tela Basin. (e). The youngest and most per-
vasive brittle deformation on Roat~in Island is related to approx-
imately E-W extension by a conjugate set of strike-slip faults
(F4) and N-S-trending normal faults (Fs). The Rio Lenin graben
(RLG) on the Honduran mainland may be related to F5 and as
it terminates against the La Ceiba fault (LCF), the latter must
be a left-lateral strike-slip fault. It is speculated that during the
Miocene-Recent, the Tela Basin was lying in a releasing bend
of the Swan Islands fault zone. Seismic activity along many of
these faults indicate, however, that both E-W and N-S extension
is still occurring today.

Tela Basin and the tilted rocks were unconformably overlain by

fine-grained clastic material (von der Hoya, 1986).
Previous work At about 10 Ma deepening of the basin accelerated
The Tela Basin lies between the Bonacca Ridge and turbidites were deposited in the deepest part of
and mainland Honduras (Figs. 2-4). It is wedge- the basin (Pinet, 1976; von der Hoya, 1986).
shaped narrowing toward the west. Few data are
available in the public domain. In the 1960s to Structure and origin of the Tela Basin
1980s the petroleum industry carried out several
geophysical surveys (gravity, magnetics, seismic re- The only structures in the Tela Basin, discussed
flection) and drilled test wells in the Tela Basin in the literature, are ENE- to WNW-trending normal
and on the mainland of Honduras (e.g., Caceres faults (Fig. 3; Pinet, 1975; Paul Mann, pers. com-
Avila et al., 1984). Although economic hydrocar- mun., 1996). The earthquake of July 27, 1990 (No. 9
bon deposits were not found, few data have been in Figs. 3 and 4 and Table 1) in the Tela Basin might
released. Pinet (1975) published several interpreta- be the result of slip along such normal fault.
tions of N-S-trending seismic reflection lines. Von The early history of the Tela Basin may have
der Hoya (1986) showed several seismic reflection been 'transpressional' because it formed during the
lines, however, without location maps. Paul Mann collision and overthrusting of the Chortfs block onto
provided us with an unpublished E-W-trending seis- the Maya block. The back thrusting event that we
mic reflection line through the central part of the proposed above (D2) may have caused loading and
basin (with location map) as well as SeaMARC II down-warping. The uplift of axial regions of the
side-scan sonar data. Magnetic studies were done by basin (von der Hoya, 1986) may have been related
Pinet (1971) and vonder Hoya (1986). Bathymetric to late D2 folding as well. This period may have
studies were carried out by Pinet (1976) and vonder started in the Late Eocene (our new isotopic ages)
Hoya (1986). and continued until the Middle Miocene ( v o n d e r
The deepest part of the Tela Basin is a narrow Hoya, 1986).
E-W-trending trough which at present is the main On the basis of the existence of ENE- to WNW-
depocenter. It is segmented by N-S-trending ridges trending normal faults on Roatfin Island (F3) and
(Figs. 3 and 4; Pinet, 1976). SeaMARC II side-scan in the Tela Basin as deduced from seismic reflec-
sonar data of the central deep, south of Roat~n Is- tion lines (Pinet, 1976; v o n d e r Hoya, 1986; Paul
land, indicate that the sea floor is quite smooth with Mann, pers. commun., 1996), we suggest that in the
only a few ENE-trending lineaments (Paul Mann, Miocene the Tela Basin started to subside again due
pers. commun., 1996). Some of these lineaments to N-S stretching related to transtension across the
may be seismically active faults (see focal-mecha- Swan Islands fault zone. The ENE trend of the faults
nism plot of earthquake in the Tela Basin on Fig. 2 is related to the ENE trend of the Swan Islands fault
and Table 1). zone and the WNW trend to the WNW trend of the
Pinet (1971) presented a magnetic anomaly map Honduran coast line.
of the Tela Basin and suggested that three positive The most penetrative brittle deformation on
anomalies may be related to serpentinite bodies Roatfin Island (Fs) caused major E - W extension
thrust onto non-magnetic sedimentary rocks. Based along N-S-trending faults. These faults have to ex-
on the seismic reflection characteristics of an area tend to the south until they abut against an E -
where such magnetic anomaly occurs, von der Hoya W-trending strike-slip fault. All approximately E -
(1986) suggested that they are better explained as W-trending faults in the Tela Basin have been iden-
being caused by basaltic intrusives, perhaps related tified as normal faults. The Rio Lefin graben on
to the basalts exposed on Utila Island. the mainland of Honduras (Fig. 3) trends NNE to-
Seismic reflection lines were interpreted by Pinet ward the Tela Basin in which, however, it has not
(1975) and von der Hoya (1986) as indicating that been identified. Toward the south the Rio Le~n
the Tela Basin is underlain by a basement of meta- graben abuts against the ENE-trending La Ceiba
morphic and plutonic rocks overlain unconformably fault which consequently must be a strike-slip fault.
by up to 5000 m of Tertiary sedimentary rocks. Von The La Ceiba fault has been mapped eastward until
der Hoya (1986) proposed that in the early Tertiary it reaches the coastline. We speculate that the La
the Tela Basin was formed by down-warping and Ceiba fault continues eastward along the Honduran
that the basin was not controlled by faulting. He shelf into the Tela Basin (Fig. 17e). Furthermore, we
proposed that during the Oligocene (35 to 30 Ma) speculate that the Rio Lefin graben continues north-
basement uplifts were formed in the axial portions ward (Fig. 17e) until it reaches the Swan Islands
of the Tela Basin along ENE trends. The older sed- fault zone west of Utila Island. We also suggest
imentary rocks were faulted and tilted and eroded. that the ridge separating the two subbasins is a N -
The basin started subsiding in Middle Miocene time S-trending horst. Thus, we propose that during the

216 H.G. AVt~ LALLEMANT and M.B. GORDON
F5 phase the Tela Basin was lying in a releasing E-W-trending relative plate motion vector was par-
bend of the Swan Islands fault zone. titioned into a N-S component (normal faulting)
The areas where we propose the existence of and an ENE-trending component (strike-slip on the
normal and strike-slip faults in the Tela Basin have Swan Islands fault zone). These normal faults caused
not been mapped by side-scan sonar and we do renewed subsidence of the Tela Basin in Miocene
not have access to seismic reflection lines through time. An earlier generation of F3 faults may have
these areas to test our speculative model. However, a been partly responsible for tectonic denudation and
similar deformation scheme that we propose for the exhumation of the metamorphic rocks of Roat~in.
Tela Basin has previously been proposed for another The fourth generation of faults (F4: a conjugate
part of the plate boundary: Aldrich et al. (1991) set of strike-slip faults) and the fifth (F5: conjugate
described E-W-trending grabens cut across by N - set of normal faults) resulted in E - W extension. The
S-trending ones in northern Honduras, just south of normal faults can be correlated with the NNE-trend-
the Motagua fault zone and the time of deformation ing Rfo Lenin graben on the mainland of Honduras.
is about the same as in the Tela Basin. Toward the south this graben abuts against the La
Ceiba fault which is a left-lateral strike-slip fault
suggesting that Roat~in Island and the Tela Basin
SUMMARY AND CONCLUSIONS were lying at the time in a releasing bend of the
Swan Islands fault zone. As the Swan Islands fault
Three generations of folds (D1 to D3) and five between Roatfin and the Motagua fault is seismically
generations of faults (F1 to Fs) were identified in the active, it seems that both E - W and N-S extension
metamorphic rocks of Roat~in Island. The folding are still occurring at present.
occurred in Late Cretaceous to early Tertiary and
the faulting from early Tertiary to the Present. Dis-
placement partitioning is an important factor when ACKNOWLEDGEMENTS
interpreting the structural evolution of Roat~in Island,
because the deformation structures are generally not This study was made possible by a grant from
directly related to the relative convergence or diver- the National Science Foundation (EAR-9219384).
gence rate vectors between the North American and We thank A.R. McBirney for air photos, maps, and
Caribbean plates, but to their components parallel comments; E. Phelgar, Defense Mapping Agency,
and perpendicular to the plate boundary: the Swan Instituto Geogrfifico Nacional, and Direcci6n Gen-
Islands and Motagua fault zones. eral de Minas e Hidrocarburos (Honduras) for lo-
The first folding (D1) is synmetamorphic and re- gistical support; R.J. Phillips for supplying us with
lated to mylonitization along the left-lateral Motagua the thesis of v o n d e r Hoya; R Mann, for sharing
and Swan Islands fault zones. The D1 structures are with us unpublished material (side-scan sonar map
similar in style and orientation as those on Guanaja and seismic reflection line) of the Tela Basin; J.T.
Island and along the Motagua fault zone in Hon- Dutro Jr. for fossil identification; G.K. Merrill for
duras and Guatemala. N- and S-vergent thrusting in helping us in our (unsuccessful) search for con-
Honduras and Guatemala may be coeval with the odonts; Peter Copeland for the 4~ dating;
formation of the mylonites. Ann Blythe for the fission-track dating; Jinny Sis-
The second generation of folds (D2) is related to son for the amphibole analysis; and D. Chu for
S-vergent thrusting and refolding. Loading by these help with the focal-mechanisms and plate motions.
thrust sheets may have caused subsidence and the We are especially grateful to Burke Burkart, Gren
origin of the Tela Basin. Draper, Pete Emmet, and Paul Mann for their ex-
The D3 structures indicate E - W contraction pos- tensive reviews of the original manuscript which
sibly in a restraining bend along the Motagua fault improved it considerably. However, errors, mistakes,
zone. These two contractional phases (D2 and D3) and misinterpretations are clearly our responsibility.
resulted in uplift of the metamorphic rocks which
were subsequently exhumed by erosion and possibly
by tectonic denudation. New isotopic ages indicate REFERENCES
that this event occurred in the Late Eocene.
The first generation faults (F1) are N-S-trending Aldrich, M.J., Jr., Adams, A.I. and Escobar, C., 1991. Structural
thrust faults that resulted in E - W contraction. This geology and stress history of the Platanares geothermal site,
phase may be coeval with the D3 folding. Honduras: implications on the tectonics of the northwestern
Caribbean plate boundary. J. Volcanol. Geotherm. Res., 45:
The second generation faults (F2: a mostly conju-
59-69.
gate set of strike-slip faults) and the third one (F3: Angelier, J., 1990. Inversion of field data in fault tectonics to
a mostly conjugate set of normal faults) are related obtain the regional stress, III. A new rapid direct inversion by
to N-S extension. These structures indicate that the analytical means. Geophys.J. Int., 103: 363-376.

D E F O R M A T I O N HISTORY OF ROAT,/~N I S L A N D 217
Angelier, J., 1991a. Inversion directe et recherche 4-D: compar- Society of America, Golden, CO, pp. 37-76.
ison physique et math6matique de deux modes de d6termina- Dziewonski, A.M., Chou, T.-A. and Woodhouse, J.H., 1981. De-
tion des tenseurs des pal6oconstraintes en tectonique de failles. termination of earthquake source parameters from waveform
C. R. Acad. Sci. Paris, Ser. II, 312: 1213-1218. data for studies of global and regional seismicity. J. Geophys.
Angelier, J., 199 lb. Analyse chronologique matricielle et succes- Res., 86: 2825-2852.
sion r6gionale des 6v6nements tectoniques. C. R. Acad. Sci. Finch, R.C., 1981. Mesozoic stratigraphy of central Honduras.
Paris, Ser. II, 312 1633-1638. Am. Assoc. Pet. Geol. Bull., 65: 1320-1333.
Arden, D.D., 1975. Geology of Jamaica and the Nicaragua Rise. Fitch, T.J., 1972. Plate convergence, transcurrent faults, and in-
In: A.E.M. Nairn and EG. Stehli (Editors), The Ocean Basins ternal deformation adjacent to southeast Asia and the western
3: The Gulf of Mexico and the Caribbean. Plenum, New York, Pacific. J. Geophys. Res., 77: 4432-4469.
pp. 617-662. Gomberg, D.M., Banks, EO. and McBirney, A.R., 1968.
Av6 Lallemant, H.G. and Guth, L.R., 1990. Role of extensional Guatemala: preliminary zircon ages from Central Cordillera.
tectonics in exhumation of eclogites and blueschists in an Science, 162: 121-122.
oblique subduction setting. Geology, 18: 950-953. Gordon, M.B., 1991. The Chortfs block is a continental,
Bargar, K.E., 1991. Fluid inclusions and preliminary studies pre-Mesozoic terrane. In: D.K. Larue and G. Draper (Editors),
of hydrothermal alteration in core hole PLTG-1, Platanares Transactions of the 12th Caribbean Geological Conference, St.
geothermal area, Honduras. J. Volcanol. Geotherm. Res., 45: Croix, U.S. Virgin Islands. Miami Geological Society, Miami,
147-160. FL, pp. 505-512.
Beck, M.E., Jr., 1983. On the mechanism of tectonic transport in Gordon, M.B., 1993. Revised Jurassic and Early Cretaceous
zones of oblique subduction. Tectonophysics, 93: 1-11. (Pre-Yojoa Group) stratigraphy of the Chortfs block: Paleo-
Burkart, B., 1994. Northern Central America. In: S.K. Donovan geographic and tectonic implications. In: J.L. Pindell and B.F.
and T.A. Jackson (Editors), Caribbean Geology: An Introduc- Perkins (Editors), Mesozoic and Early Cenozoic Development
tion. Univ. West Indies Publ. Assoc., Kingston, Jamaica, pp. of the Gulf of Mexico and Caribbean Region. Trans. 13th
265-284. Annu. Res. Conf. Gulf Coast Sect., Society of Economic Pa-
Burke, K., 1988. Tectonic evolution of the Caribbean. Annu. Rev. leontologists and Mineralogists Foundation, Austin, TX, pp.
Earth Planet. Sci., 16:201-230. 143-154.
Burke, K., Cooper, C., Dewey, J.F., Mann, E and Pindell, Gordon, M.B. and Av6 Lallemant, H.G., 1995. Cryptic strike-slip
J.L., 1984. Caribbean tectonics and relative plate motions. faults of the Chortfs block. Geol. Soc. Am. Abstr. Progr., 27:
In: W.E. Bonini, R.B. Hargraves and R. Shagam (Editors), 226-227.
The Caribbean-South American Plate Boundary and Regional Gordon, M.B. and Muehlberger, W.R., 1994. Rotation of the
Tectonics. Geol. Soc. Am. Mem., 162, 31-64 pp. Chortfs block causes dextral slip on the Guayape fault. Tecton-
Caceres Avila, E, Tappmeyer, D.M., Aves, H.S., Gillett, M. and ics, 13: 858-872.
Klenk, C.D., 1984. Recent studies of basins are encouraging Gose, W.A., 1985. Paleomagnetic results from Honduras and
for future exploration in Honduras. Oil Gas J., 82: 139-140. their bearing on Caribbean tectonics. Tectonics, 4: 565-585.
Calais, E. and Mercier de L6pinay, B., 1991. From transtension Harding, T.E, 1974. Petroleum traps associated with wrench
to transpression along the northern Caribbean plate boundary faults. Am. Assoc. Pet. Geol. Bull., 58: 1290-1304.
off Cuba: implications for the Recent motion of the Caribbean Heiken, G., Ramos, N., Duffield, W., Musgrave, J., Wohletz, K.,
plate. Tectonophysics, 186: 329-350. Priest, S., Aldrich, J., Flores, W., Ritchie, A., Goff, E, Eppler,
Case, J.E. and Holcombe, T.L., 1980. Geologic-tectonic map D. and Escobar, C., 1991. Geology of the Platanares geother-
of the Caribbean region. U.S. Geol. Surv. Misc. Invest. Map mal area, Departamento de Cop~n, Honduras. J. Volcanol.
I- 1100, Scale 1 : 2,500,000. Geotherm. Res., 45: 41-58.
Case, J.E., MacDonald, W.D. and Fox, P.J., 1984. Caribbean Heubeck, C. and Mann, E, 1991. Geologic evaluation of plate
crustal provinces; seismic and gravity evidence. In: G. Dengo kinematic models for the North American-Caribbean plate
and J.E. Case (Editors), The Caribbean Region. The Geology boundary zone. Tectonophysics, 191: 1-26.
of North America, Vol. H, Geological Society of America, Holcombe, T.L., Vogt, ER., Matthews, J.E. and Murchison, R.R.,
Golden, CO, pp. 15-36. 1973. Evidence for sea-floor spreading in the Cayman Trough.
Cashman, S.M., Kelsey, H.M., Erdman, C.E, Cutten, H.N.C. and Earth Planet. Sci. Lett., 20: 357-371.
Berryman, K.R., 1992. Strain partitioning between structural Horne, G.S., Atwood, M.G. and King, A.E, 1974. Stratigraphy,
domains in the forearc of the Hikurangi subduction zone, New sedimentology, and paleoenvironment of Esquias Formation of
Zealand. Tectonics, 11: 242-257. Honduras. Am. Assoc. Pet. Geol. Bull., 58: 176-188.
Dean, B.W. and Drake, C.L., 1978. Focal mechanism solutions Horne, G.S., Clark, G.S. and Pushkar, E, 1976a. Pre-Cretaceous
and tectonics of the Middle America arc. J. Geol., 8 6 : 1 1 1 - rocks of northwestern Honduras: basement terrane in Sierra de
128. Omoa. Am. Assoc. Pet. Geol. Bull., 60: 566-583.
DeMets, C., Gordon, R.G., Argus, D.E and Stein, S., 1994. Home, G.S., Pushcar, E and Shafiqullah, M., 1976b. Laramide
Effect of recent revisions to the geomagnetic reversal time plutons on the landward continuation of the Bonacca ridge,
scale on estimates of current plate motions. Geophys. Res. northern Honduras. Publ. Geol. ICAITI, 5: 84-93.
Lett., 21: 2191-2194. Kanamori, H. and Stewart, G.S., 1978. Seismological aspects of
Deng, J. and Sykes, L.R., 1995. Determination of Euler pole the Guatemala earthquake of February 4, 1976. J. Geophys.
for contemporary relative motion of Caribbean and North Res., 83: 3427-3434.
American plates using slip vectors of interplate earthquakes. Kozuch, M.J., 1991. Mapa Geol6gico de Honduras. Tegucigalpa,
Tectonics, 14: 39-53. Honduras. Instituto Geogr~fico Nacional, Tegucigalpa scale
Dengo, G., 1969. Problems of tectonic relations between Central 1 : 500,000.
America and the Caribbean. Trans. Gulf Coast Assoc. Geol. Land, L., 1979. The fate of reef-derived sediment on the north
Soc., 19:311-320. Jamaican Island slope. Mar. Geol., 29: 55-71.
Donnelly, T.W., Horne, G.S., Finch, R.C. and L6pez-Ramos, Lawrence, D.E, 1975. Petrology and Structural Geology of the
E., 1990. Northern Central America; the Maya and Chortfs Sanarate-E1 Progreso Area, Guatemala. Ph.D. Thesis, Univ.
blocks. In: G. Dengo and J.E. Case (Editors), The Caribbean New York, Binghamton, NY, 255 pp.
Region. The Geology of North America, Vol. H, Geological Leroy, S., Mercier de L6pinay, B., Mauffret, A. and Pubellier,

218 H.G. AVt~ L A L L E M A N T and M.B. G O R D O N
M., 1996. Structural and tectonic evolution of the eastern 155: 121-138.
Cayman Trough (Caribbean Sea) from seismic reflection data. Pinet, ER., 1971. Structural configuration of the northwestern
Am. Assoc. Pet. Geol. Bull., 80: 222-247. Caribbean. Geol. Soc. Am. Bull., 82: 2027-2032.
MacDonald, K.C. and Holcombe, T.L., 1978. Inversion of Pinet, ER., 1975. Structural evolution of the Honduras conti-
magnetic anomalies and sea-floor spreading in the Cayman nental margin and the sea floor south of the western Cayman
Trough. Earth Planet. Sci. Lett., 40:407-414. trough. Geol. Soc. Am. Bull., 86: 830-836.
Mann, E and Burke, K., 1984. Cenozoic rift formation in the Pinet, ER., 1976. Morphology of northern Honduras, northwest-
northern Caribbean. Geology, 12: 732-736. ern Caribbean Sea. Deep-Sea Res., 23: 839-847.
Mann, E, Tyburski, S.A. and Rosencrantz, E., 1991. Neogene Plafker, G., 1976. Tectonic aspects of the Guatemala earthquake
development of the Swan Islands restraining-bend complex, of 4 February 1976. Science, 193: 1201-1208.
Caribbean Sea. Geology, 19: 823-826. Rosencrantz, E. and Mann, E, 1991. SeaMARC II mapping
Manton, W.I., 1987. Tectonic interpretation of the morphology of of transform faults in the Cayman Trough, Caribbean Sea.
Honduras. Tectonics, 6: 633-651. Geology, 19: 690-693.
McBirney, A.R. and Bass, M.N., 1969a. Geology of Bay Islands, Rosencrantz, E. and Sclater, J.C., 1986. Depth and age in the
Gulf of Honduras. Am. Assoc. Pet. Geol. Mem., 11: 229-243. Cayman Trough. Earth Planet. Sci. Lett., 79: 133-144.
McBirney, A.R. and Bass, M.N., 1969b. Structural relations of Rosencrantz, E., Ross, M.I. and Sclater, J.G., 1988. Age and
Pre-Mesozoic rocks of northern Central America. Am. Assoc. spreading history of the Cayman Trough as determined from
Pet. Geol. Mem., 11: 269-280. depth, heat flow, and magnetic anomalies. J. Geophys. Res.,
McCaffrey, R., 1991. Slip vectors and stretching of the Sumatran 93: 2141-2157.
forearc. Geology, 19: 881-884. Schwartz, D.E, Cluff, L.S. and Donnelly, T.W., 1979. Quaternary
McDougall, I. and Harrison, T.M., 1988, Geochronology and faulting along the Caribbean-North American plate boundary
Thermochronology by the 4~ Method. Oxford Univ. in Central America. Tectonophysics, 52:431-445.
Press, New York, 212 pp. Southernwood, S., 1986. Late Cretaceous Limestone Clast Con-
Meschede, M., Ratschbacher, L. and Frisch, W., 1993. Kine- glomerates of Honduras. MS Thesis, Univ. of Texas, Dallas,
matic information from fault-slip data in southern Mexico and 299 pp.
along the Motagua-Polochic fault system in Guatemala. In: Stein, S., DeMets, C., Gordon, R.G., Brodholt, J., Argus, D.,
E Ortega-Guti6rrez, EJ. Coney, E. Centeno-Garcfa and A. Engelen, J.E, Lundgren, E, Stein, C., Wiens, D. and Woods,
G6mez-Caballero (Editors), Transactions of the 1st Circum- D.E, 1988. A test of alternative Caribbean plate relative mo-
Pacific and Circum-Atlantic Terrane Conference, Univ. Nac. tion models. J. Geophys. Res., 93: 3041-3050.
Aut6n. M6xico, Inst. Geol., pp. 81-85. Steiner, M.B. and Walker, J.D.,1996. Late Silurian plutons in
Mills, R.A. and Barton, R., 1996. Geology of the Ahuas area in Yucatan. J. Geophys. Res., 101: 17,727-17,735.
the Mosquitia Basin of Honduras. Am. Assoc. Pet. Geol. Bull., Twiss, R.J. and Gefell, M.J., 1990. Curved slickenfibers: a new
80:1627-1640. brittle shear sense indicator with application to a sheared
Mills, R.A., Hugh, K.E., Feray, D.E. and Swolfs, H.C., 1967. serpentinite. J. Struct. Geol., 12: 471-482.
Mesozoic stratigraphy of Honduras. Am. Assoc. Pet. Geol. Tyburski, S.A., 1992. Deformational Mechanisms along Active
Bull., 51: 1711-1786. Strike-Slip Faults: SeaMARC II and Seismic Data from North
Molnar, E and Sykes, L.R., 1969. Tectonics of the Caribbean and America-Caribbean Plate Boundary. MA Thesis, Univ. of
Middle America regions from focal mechanisms and seismic- Texas, Austin, 195 pp.
ity. Geol. Soc. Am. Bull., 80: 1639-1684. vonder Hoya, H.A., II, 1986. A Reflection Seismic and Magnetic
Olson, E.C. and McGrew, EO., 1941. Mammalian fauna from the Investigation of the Tela Basin: Northern Offshore Honduras.
Pliocene of Honduras. Geol. Soc. Am. Bull., 52: 1219-1244. MS Thesis, Southern Methodist Univ., Dallas, Texas, 112 pp.
Osiecki, ES., 1981. Estimated intensities and probable tectonic Wadge, G. and Wooden, J.L., 1982. Late Cenozoic alkaline
sources of historic (pre-1898) Honduran earthquakes. Bull. volcanism in the northwestern Caribbean: tectonic setting and
Seism. Soc. Am., 71: 865-881. Sr isotopic characteristics. Earth Planet. Sci. Lett., 57: 35-46.
Perfit, M.R., 1977. Petrology and geochemistry of mafic rocks Walcott, R.I., 1978. Geodetic strains and large earthquakes in the
from the Cayman Trench: evidence for spreading. Geology, 5: axial tectonic belt of North Island, New Zealand. J. Geophys.
105-110. Res., 83: 4419-4429.
Petit, J.E, 1987. Criteria for the sense of movement on fault White, R.A., 1991. Tectonic implications of upper-crustal seis-
surfaces in brittle rocks. J. Struct. Geol., 9: 597-608. micity in Central America. In: D.B. Slemmons, E.R. Engdahl,
Pindell, J.L., 1993. Regional synopsis of Gulf of Mexico and M.D. Zoback and D.D. Blackwell (Editors), Neotectonics of
Caribbean evolution. In: J.L. Pindell and B.E Perkins (Edi- North America. The Geology of North America, Decade Map
tors), Mesozoic and Early Cenozoic Development of the Gulf Vol., Geological Society of America, Golden, CO, pp. 323-
of Mexico and Caribbean Region. Trans. 13th Annu. Res. 338.
Conf. Gulf Coast Sect., Society of Economic Paleontologists Williams, H. and McBirney, A.R., 1969. Volcanic history of
and Mineralogists Foundation, Austin, TX, pp. 251-274. Honduras. Univ. California Publ. Geol. Sci., 85: 1-101.
Pindell, J.L., Cande, S.C., Pitman, W.C., Rowley, D.B., Dewey, Wilson, H.H., 1974. Cretaceous sedimentation and orogeny in
J.E, LaBrecque, J. and Haxby, W., 1988. A plate-kinematic Nuclear Central America. Am. Assoc. Pet. Geol. Bull., 58:
framework for models of Caribbean evolution. Tectonophysics, 1348-1396.

Chapter 9
The Southern Flank of the Tela Basin,

Republic of Honduras
W.I. MANTON and R.S. MANTON
A wide strike-slip zone is formed at the northern margin of the Caribbean plate. Its northern portion, the Tela Basin, is
submerged, but the north coast of Honduras, from Tela to Trujillo, is characterized by high, fault-bounded mountains. Shear zones
exposed in drainages attest to a long history of faulting. A feature of some interest is the Agu~n fault, which lies south of the
mountains, and today forms the southern boundary of the zone. It curves through 35-45 ~ and is in the west transpressional. In
the east, however, it is transtensional and gives rise to the wide valley of the Agu~n River, which is probably a 'lazy' S-type of
pull-apart basin. Poorly bedded conglomerates of unknown age outcropping at the eastern end of the valley may represent part of
its infilling with fluvial sediments. Also at its eastern end are near-vertical mudstones and shales with thin interbeds of limestone,
that were deposited in a hemipelagic to pelagic environment. Poorly preserved globeriginids suggest that the rocks are Middle
Miocene or younger. About 50 km south of this occurrence, on the road between Trujillo and San Esteban, are exposures of rocks
with a similar lithology that have been thrust from the northwest. They contain poorly preserved fossils, some of which may be
Tertiary, and from both their environment of deposition and their direction of transport are correlated with the rocks of the Agu~n
Valley. Both sequences are cut by mafic dikes, probably intruded before deformation. The presence of such rocks suggests that in
the Miocene the north coast of Honduras was submerged with the formation of deep, turbidite-filled basins. When the region was
uplifted some of these rocks were preserved by transtensional faults; others were deformed into flower structures by transpressive
faults and were thrust to the south.
INTRODUCTION colonial roads that for the most part no longer exist,
by Powers (1918) and Foye (1918), and by Roberts
High mountains whose northern slopes are cov- and Irving (1957), Williams and McBirney (1969)
ered with dense tropical vegetation, wide valleys and Manton (1987) along the present-day roads. In
filled with alluvium, few roads, and a geology contrast, the offshore region is much better known.
consisting of a monotonous succession of poorly The Bay Islands have been mapped (McBimey and
exposed phyllites cut by plutons or overlain by Bass, 1969), as have the Swan Islands (Ivey et al.,
deeply weathered lavas have discouraged geologi- 1980), and several offshore wells have been drilled
cal mapping of the north coast of Honduras, even to the Mesozoic basement (Caceres Avila et al.,
though the region lies on the margin of a major 1984). Marine geophysics has detailed the topog-
strike-slip (Fig. 1) boundary that has accommodated raphy, the sediment-filled basins, and the principal
more than 1000 km of movement during the Tertiary faults of the submerged part of the plate boundary
(Pindell and Barrett, 1990). Satellite images, how- (Pinet, 1975, 1976; Rosencrantz and Mann, 1991;
ever, show with clarity the great arcuate faults that Mann et al., 1991).
shape the topography (Fig. 2), and several papers The area between the Bay Islands and the main-
interpreting them have been written (Muehlberger, land is known as the Tela Basin, and the geology
1976; Letouzey, 1985; Manton, 1987; Gordon and of its northern margin is discussed elsewhere in this
Muehlberger, 1994). Apart from the mapping of volume (Av6 Lallemant and Gordon, Chapter 8).
the westernmost part of the coast by the United Here we discuss the geology of its southern margin
Nations (part of their map is reproduced in Home and elaborate on the occurrences of both continen-
et al., 1976a), published geologic observations are tal and marine sediments on mainland Honduras that
restricted to those made by Sapper (1905) along were mentioned but not described by Manton (1987).

220 W.I. M A N T O N and R.S. M A N T O N
. . . . . . ' ~dw ' 86'w ' e4'w '
Figs:3&4~
17N -
/
15N ' I HONDURAS ....~ _//',_..j,,~-'"'- '%

15N -
-13N ~ l ~ ~ ~t 13N-
, .....,
8B.w , \ e6w
.... _,,_, , o 4,. - -
Fig. 1. Map of portion of Central America showing Honduras and the locations of the Landsat image (Fig. 2) and of the area discussed in
this paper (Figs. 3 and 4).
THE NORTH COAST: PHYSIOGRAPHY AND A C C E S S Esteban. From E1 Progreso (Fig. 3) the tarred road
continues south to Santa Rita where it turns east to
The coast of Honduras between Omoa and Tru- Yoro. Yoro and the western end of the Agmin Valley
jillo consists of a narrow plain backed by mountain are linked by an unpaved road that passes through
ranges, the Sierra de Omoa behind Puerto Cortez and Olanchito to meet the Jutiapa-Trujillo road at Saba.
the Sierra Nombre de Dios between Tela and Trujillo Distances are: San Pedro Sula-Tela, 99 km; Tela-La
(Fig. 3). The ranges are breached between Omoa and Ceiba, 103 km; La Ceiba-Trujillo, 171 km.
Tela by the Sula Valley and south of La Ceiba by the
gorge of the Cangrejal River. To the east of La Ceiba
the Cordillera Nombre de Dios narrow, and a low PRINCIPAL FEATURES OF THE STRIKE-SLIP MARGIN
saddle is eroded in them. South of the ranges lies the
valley of the Agu~in, a broad, flat-floored depression, The kinematics of the North American-
that curves gently northward to meet the Caribbean Caribbean (NOAM-CARIB) plate boundary have
coast east of Trujillo. The town of San Pedro Sula, proved particularly difficult to determine, both be-
which lies in the Sula Valley, is the second city cause the data are few and because the calculated
in Honduras and has an international airport with rotation parameters are strongly biased by the initial
car rental agencies. The north coast is reached by choice of data. While the models of MacDonald
driving to E1 Progreso on the east side of the valley (1976), Minster and Jordan (1978) and Stein et al.
and taking the tarred road that runs northeastwards to (1988) predict fairly well the azimuths of bounding
Tela (Fig. 3). This road follows the coast for 90 km Swan Islands and Oriente faults, those of Sykes et
before turning inland and crossing over the saddle al. (1982) and of the NUVEL-1 model do not, al-
into the Agu~in Valley, the southern margin of which though the latter can be brought to agreement if the
it follows for 50 km before terminating at Trujillo. Lesser Antilles data are omitted from the inversion
At the east end of the valley an unpaved road leads (DeMets et al., 1990). Heubeck and Mann (1991),
southwards through the town of Bonito Oriental to however, have shown that even the models yielding
the Sico River and thence to E1 Carbon and San good fits fail to predict the pattern of transpression
Fig. 2. (A) Landsat image of the north coast of Honduras in the region of La Ceiba and Trujillo. Honduras P18 R49, 19 December, 1973.
Color composite 1, 2, 3. (B) Some features of the Landsat image. C.V. = Cangrejal Valley; A.B. = Agu~in beds, Agu~in Valley locality.
Linear features in the image parallel to the strike of the beds suggest that they extend for several kilometers southwest of where they
are exposed by the road. C.T. = Cerro de Tarros (see Fig. 6). Triangle = Pico Bonito, 2643 m. D = possible dolerite dikes occupying
deep linear valleys (within the Cangrejal Valley the relationship between dikes and valleys has been observed in the field); V = linear
valleys of unknown origin; L = linear features of unknown origin; R.S.L. = Agu~in beds, Rio Sico locality. For reference, the approximate
location of the Swan Islands fault is shown as is the NUVEL-1 NOAM-CARIB plate boundary vector (DeMets et al., 1990).

THE SOUTHERN FLANK OF THE TELA BASIN, HONDURAS 221
~u o~
~ Roatan
I (~Utila
LA CEIBA
/ ///"~2 ~
._........_ L

222 W.I. MANTON and R.S. M A N T O N
1 ; I I ~ ' i
88W 871W 86]W
/
o 20 40 60 8,0 loo
: . . . . . . . . I
KILOMETRES ISLANDS
~GUANAJA
~UTI LA
- 16 N ,~ o ~ 16N -
~ ~ C O R T E Z ~ ~'J . ~~ ~) -i~~~o~uon
' ~ " ~0 /
~.~,o,~~./ ~.~ _.sAN , " ~ ~ ~

s j ,.,... ,o
J.OSta. 4 \ ! / .....
/ ,,,.,.,,,...,,,'.. r( H 0 N D U R,4 S DE LA PAZ
L.Yojoo .J T Tegucigalpa To Jutiopo
88W 87W 86W

, i 1 ... I. I I I I
Fig. 3. Location map showing principal towns, roads, mountains and rivers.
and transtension observed along the plate bound- with a gentle restraining bend that previously existed
ary, and this observation has led them to divide between the two faults produced the broad anticlinal
the Caribbean plate into three segments each having uplift upon which the islands lie (Mann et al., 1991).
slightly different poles of rotation. Flanking the southern margin of the western portion
An interpretation of the geology of the north coast of the trough is a discontinuous series of en-echelon
has been given by Manton (1987) whose map is ridges (the Bonacca Ridge of Banks and Richards,
reproduced with some revisions in Fig. 4. Principal 1969) which break sea level to form the Bay Islands
changes are: (1) the addition of the Precambrian of Roatan and Guanaja and elsewhere rise to within
(Manton, 1996); (2) the change of the age of the 1000 m of the surface. Between these ridges and
Agu~n beds from Late Cretaceous to Tertiary (this the Honduran shelf break lies a triangular area com-
paper, below); (3) the change of the age of the posed of ridges and deep, fault-bounded, sediment-
volcanics between La Ceiba and Olanchito from filled basins elongated in the direction of the plate
Cretaceous to Tertiary to make them consistent with boundary (Pinet, 1976). On seismic reflection pro-
the geological map of Honduras (Kozuch, 1991); files Pinet (1975) identified an eroded pre-Eocene or
and (4) the addition of faults to each side of the Cretaceous surface unconformably overlain by two
Agu~n Valley to make the feature a lazy S pull-apart units, which from their widespread distribution and
basin (Mann et al., 1983). uniform thickness he ascribed to deposition during
a tectonically quiet interval during the middle Ter-
Offshore tiary. He assigned a Pliocene age to the faulting that
produced the present bottom topography of ridges
The present-day boundary between the North and basins and argued that the sediments filling
American and Caribbean plates is an active fault that the basins are turbidites, at the same time noting
appears on side-scan sonar images as a well defined that some are faulted. Pinet (1971) recorded three
linear feature extending from the Motagua Valley of magnetic anomalies, one between Roatan and the
Guatemala to the Swan Islands and closely following mainland, and the others off the mouth of the Agu~n
the base of the southern wall of the Cayman Trough and off the coast at longitude 84~ 30". These he
(Rosencrantz and Mann, 1991). At the Swan Islands attributed to en-echelon offsets of an elongate mass
the fault fight steps to a parallel fault that continues of ultramafic rock. In addition, Pinet (1972) drew
to the Cayman spreading centre. Stresses associated attention to piercement structures lying off the coast,

/ \871W /I 861W
/ 50o
~~176 ~.~OU % y
Basalt flows
and cones Alluvium 4000 ~
Aguan beds, ~-~ Fluvial c o n g l o m e r a t e s
marine shale ~ limestone unnamed beds
Andesitic
volcanics
-- ~ Cretaceous sediments
~o~ undifferentiated
Schists, phyllites
8~ marbles
l.OGa metamorphosed
quartzo- felspathic rocks
-r \
Granitic plutons CASTILLA +
of various ages / No I CASTANA
Nol
--16N
Geology Unknown
Geology On now,,o ~'~b. ~ ,;~; v ~

: <
. . ~ - / _ \ Z _ g.,"4 Tv ..17
~.'-~
t,t. >~-.r n
-15N
88 W 87W 86W
I I I I I I
Fig. 4. Geological sketch map of the region shown in Fig. 2, modified from Manton (1987). The earthquake solution from the western por-
tion of the Agufin fault is from Gordon and Muehlberger (1994), and the arrows are the directions of the NUVEL-1 NOAM-CARIB plate
boundary vector (DeMets et al., 1990). The vector of the 'alternative model' would be within 2~of the azimuth of the Swan Islands fault.
which coincided with the western and eastern mag- of Middle Miocene shales and white quartzose sands
netic anomalies but were not observed off the Agu4n interpreted as being deposited in a neritic environ-
mouth. He postulated that they were salt domes. ment. These may be the widespread Middle Miocene
Von der Hoya (1986) interpreted propriety seismic units of Pinet (1975). They are unconformably over-
and magnetic data obtained between Roatan and the lain by fine-grained clastic sediments deposited in
coast and concluded that the magnetic anomalies ever increasing depths.
and the piercement structures were manifestations
of one or more laccolith-shaped igneous bodies that Onshore
reached to within 1.5 km of the top of the sediments.
He argued that they were intruded in the Quaternary. The strike-slip nature of the north coast is spec-
The island of Utila is partly formed by a Quater- tacularly displayed by the Cordillera Nombre de
nary volcano, and as such may be a manifestation Dios which abruptly rise from the narrow coastal
or the same activity that produced the intrusions plain and attain their highest point behind La Ceiba,
discovered by Von der Hoya (1986), but Roatan and where the bare, pointed peak of Pico Bonito rises
Guanaja are composed of metamorphic rocks similar to 2435 m (Fig. 5). On the Landsat images (Fig. 2)
to those found in Guatemala (McBirney and Bass, the northern flank of the cordillera is demarcated
1969). The Swan Islands (Fig. 1), on the other hand, by a sharp line that extends from the Sula Valley
are composed of Oligocene to Lower Miocene tur- to east of La Ceiba and is clearly a recent fault,
bidites containing a large amount of volcanic glass called by Muehlberger (1976) the La Ceiba fault.
(Ivey et al., 1980). Southeast of La Ceiba is an equally strong linear
Two wells, Castilla No. 1 and Castana No. 1, feature that marks the Rfo Viejo fault, and between
have been drilled on the continental shelf north- these two faults the cordillera have been recently
east of Trujillo (see Caceres Avila et al., 1984, and uplifted (Manton, 1987). (For the radar image of this
Fig. 4). Both penetrated approximately 3000 m of region, see Gordon and Muehlberger, 1994.) East of
Tertiary sediments before bottoming in red to brown La Ceiba, the La Ceiba fault makes a 45 ~ clockwise
shales, siltstones, and sandstones correlated with the turn and almost intersects the Rio Viejo fault, but
Upper Cretaceous Valle de Angeles group (Aves, thereafter the trace of both faults becomes unclear,
1983). The lowermost units consisted of 670-820 m with the La Ceiba fault probably turning offshore

224 W.I. MANTON and R.S. MANTON
Fig. 5. The Cordillera Nombre de Dios photographed from La Ceiba airport. The pointed peak in the backgroundis Pico Bonito, 2643 m.
and the Rfo Viejo extending into a deep linear valley of the tonalite, its sense of slip is sinistral (W.I.
in the eastern part of the cordillera. Manton, unpubl, information). The southern portion
The present-day faults closely follow the traces of the mylonite is developed in a granite of unknown
of older faults. For example, in Quebrada Juana age, which contains pseudo-tachylite veins up to 5
Leandra, 11 km east of the bridge over the Cangrejal cm wide. Whether these were produced during the
at La Ceiba, a waterfall exposes a ductilely deformed uplift of the cordillera along the Rfo Viejo fault or
shear zone containing large boudins of the granitic whether they belong to some earlier event is un-
country rock. It is about 100 m wide and strikes known. At some time the region experienced north-
at 82 ~ with a 60N dip. Deformed leucocratic shear south extension accompanied by the intrusion of a
veins indicate that it is a high-angle thrust fault with swarm of dolerite dikes along the N60E strike of
upthrow to the north. It is cut by a highly fractured the gneissosity of the Piedras Negras tonalite. These
and altered dolerite dike of unknown age. dikes weather more easily than the surrounding rock
The Cangrejal River, which cuts through the and give rise to the deep, narrow valleys that are
cordillera south of La Ceiba, reveals more of the conspicuous on the Landsat image (Fig. 2). Similar
long history of tectonism experienced by the re- closely spaced, deep valleys occur in the cordillera
gion. About 15 km south of the town, a gneissose further west in the vicinity of Pico Bonito. Although
tonalite possessing a well developed C-S fabric in- they have a more easterly orientation it seems likely
dicative of dextral slip (Manton and Manton, 1989) that they were produced by weathering of a related
is intruded by an undeformed granodiorite. Both swarm of dikes. All that is known of their age is
rocks have been dated by Home et al. (1976b). that they are younger than the mylonite. They appear
Conventional K-Ar ages on hornblende gave 72.2 to be absent from the eastern end of the cordillera
Ma for the deformed tonalite (their Piedras Negras because the Landsat image of the region (Fig. 2)
tonalite) and 57.3 Ma for muscovite from the un- shows none of the deep linear valleys that would
deformed granodiorite (their Las Mangas tonalite). reveal their presence. It may be significant that they
The age obtained from biotite from the Piedras Ne- are restricted to the part of the coast opposite the
gras tonalite was 57.3 Ma. To the south the Piedras volcanic island of Utila and the intrusions discovered
Negras tonalite passes into a 900-m-wide zone of by Von der Hoya (1986).
mylonite. (For a general view of this rock, see fron- The Agu~in Valley has by virtue of its length,
tispiece, Tectonics, 9 (2), 1990.) In contrast to that breadth, and curvature been long regarded as the

T H E S O U T H E R N F L A N K OF T H E T E L A BASIN, H O N D U R A S 225
~.~
| '1
2'o 21 2'2 23 2; 2'6 28
LAGUNA
6 3 ~ ~2 63-
FAULT BR E C C IA c,,,o Lo Piedro
61 -,',~,_~.__~. . . . . 61 -
~ AGUAN BEDS ~ c e ~
60-
GRANITE
QUARR'Y
1 J~;~~
~ _~_/~
59-
58-
Monte Cristo
57-
CONGLOMERATES
AND
SANDSTONES 56-
55
CARIBBEAN SEA :/ ,/
54 Wta
54-
TRUJ I
5:5 53-
-'-----~ Fig6~L__L,L. _
52 52-
20 21 22 2~ 24 25 26 27 28
I I I I I I I I I
Fig. 6. Location of outcrops at the eastern end of the Agu~n Valley.
site of a major strike-slip fault (Muehlberger, 1976), of a transtensional regime, and the valley itself is
but only on its southeastern flank is it bounded probably a 'lazy' S-type of pull-apart basin (Mann
by a youthful looking scarp. Elsewhere its walls et al., 1983). See also Heubeck and Mann (1991).
are dissected and eroded back as if the faults that There are no published geophysical data to confirm
caused them have not been active for some time this interpretation, but at its eastern end are low
(Fig. 2), although several hot springs do occur along hills (Fig. 6) made up of southward-dipping coarse
its length (Manton, 1987). The direction of its con- sandstones and quartz-pebble conglomerates (Fig. 7)
tinuation to the west is unknown as there is no which may represent part of a sedimentary filling.
means of access to the region. Manton (1987) drew Some of these rocks are exposed in a road cut
it following the course of the Guiamas River. If this through Cerro La Penita and, because they are the
extrapolation is correct the fault changes direction only known examples of their kind in the Agu~in
by 35-45 ~ along its length, so that its western por- Valley, they are described below.
tion, by virtue of its orientation with respect to the The cut exposes 250 m of conglomerate with mi-
NOAM-CARIB rotation vector (Fig. 4), is in trans- nor amounts of sandstone and shale that strike 40-
pression while its eastern portion is in transtension. 60 ~ and dip south at 36 ~. The first rock encountered
This interpretation is supported by an earthquake at the northern end of the cut is a coarse-grained
with a strong thrust component plotted by Gordon quartz sandstone which grades upwards into an in-
and Muehlberger (1994) on its presumed western durated, weakly bedded or massive quartz-pebble
extension (Fig. 4). On the other hand, the wide conglomerate. The rock is clast-supported and con-
floor of the eastern part of the valley and the gener- sists of white vein quartz pebbles up to 4.5 cm in
ally low relief of the surrounding region are typical greatest dimension set in a coarse-grained sand or

Fig. 7. Northward-dipping, poorly bedded conglomeratic section exposed at Cerro La Penita, eastern end of the Agu~in Valley. The
apparent decrease in dip results from distortion brought about by a wide-angle lens.
pebble matrix. With increasing height in the sec- 1986). The undeformed rock is coarse-grained and
tion, the clast composition begins to include gray composed of large phenocrysts of pink orthoclase
limestone and a small quantity of brown sandstone; set against a matrix of white plagioclase and quartz.
then the conglomerates become matrix-supported Biotite, however, is pervasively altered to green
and sandstone clasts up to 15 cm appear. The con- chlorite. As the shear zone is traversed, the gran-
glomerates at the top of the section are massive and ite becomes progressively deformed, first cataclas-
vary from matrix- to clast-supported. Clasts include tically, then ductilely until the quartz has flowed
several varieties of sandstone and weakly indurated into stringers with aspect ratios between 5:1 and
claystone which could not have been transported a 10: 1. The orthoclase is also extended by about 5:1
great distance. but has done so brittlely by slipping along closely
Shales in the lower part of the section are for spaced microfaults. The zone, which is about 3 km
the most part gray, becoming maroon in the up- wide, strikes at 55 ~ and the minerals are stretched
per portion, and may be associated with sandstones. in the vertical direction, implying the motion was
Their most characteristic feature, however, is the dominantly dip-slip. Zircon separated from the un-
way they have been diapirically deformed by the deformed granite is strongly discordant with inter-
heavier, rapidly loaded conglomerates. Soft-sedi- +15 and 30 +6
cepts at 127_13 _ Ma (Appendix). The lower
ment structures produced in the deformed shales intercept, which falls in the Oligocene, may reason-
include large-scale protrusions into the base of the ably be interpreted as resulting from Pb loss during
overlying conglomerates, and a spectacular balloon- hydrothermal activity associated with the sheafing,
shaped diapir (Fig. 8) forcibly intruded into the raising the possibility that in the Miocene the strike-
overlying beds. The coarseness of the sediments, the slip margin extended as far south as E1 Carbon.
lack of bedding, and the rapidity of loading are all
characteristic of alluvial deposition at the margin of
a rapidly deepening fault-generated valley. THE AGUAN BEDS: POSSIBLE TERTIARY MARINE
Today the Agu~in Valley is the southern limit SEDIMENTS
of the strike-slip margin, but in the vicinity of
E1 Carbon (Fig. 4) the road crosses a major fault Manton (1987) has already reported the presence
zone developed in a granitic pluton (Southemwood, of marine sediments in northern Honduras, naming

Fig. 8. Large balloon-shaped diapir of shale and thinly bedded sandstones disrupting conglomerates at Cerro La Penita. For scale, see
man standing in the middle of the diapir.
them the Agufin beds. They are well exposed in two tion between the hills. Primary sedimentary geopetal
localities, the one in road cuts through low hills structures were searched for in the field without
(Fig. 6) in the Agufin Valley, southeast of Trujillo, success, but piercement structures associated with a
and the other in a deep cut on the road between Tru- dike suggest that the section youngs to the south.
jillo and E1 Carbon, near the Sico River (Fig. 4). At The rocks lie structurally beneath the shales and
each locality sections were measured, thin sections sandstones of Cerro La Penita, but the contact is
were made for petrographic study, the mineralogy obscured.
of fine-grained samples was determined by X-ray Indurated, green-gray to dark tan, silty siliceous
diffraction, and interbedded limestones were ana- mudstones and thin bands of fissile gray shales
lyzed for Sr isotope ratios with a view to determin- dominate the lower part of the section (Fig. 9).
ing their age by comparison with published seawater The mudstones may be massive or may have planar
curves. Samples containing microfossils were sub- lamination. Beds generally have thicknesses between
mitted to micropaleontologists for identification. 2.5 and 5 cm, with some units ranging up to 20
cm. Many mudstone beds have a pinch and swell
Agu~n Valley locality appearance, but it is unclear whether this is a primary
feature.
Cuts made in a succession of low hills that strad- Thin- to medium-bedded, black (but gray weath-
dle the road 4.5 km north of Cerro La Penita (Fig. 6) ering), argillaceous, micritic limestones and in-
expose steeply dipping, thinly bedded mudstones terbedded mudstones dominate the upper half of
and shales and thicker limestone units, intruded by the section. A few beds of coarser-grained calcaren-
deeply weathered basic dikes (Fig. 9). The total ite are also present. Planar lamination is the most
length exposed in outcrop is about 520 m. The ex- common sedimentary structure in the limestones. In
posures end on the north against a thickly bedded, places, however, laminae appear to be disturbed,
green sandstone containing abundant spheroidal con- giving the rocks a mottled appearance. En-echelon,
cretions. The contact is not exposed, nor does the calcite-filled microfractures are found in some beds
rock resemble any in the measured section. There is and thicker veins filled by sparry calcite infill frac-
no evidence to suggest duplication of the sequence tures in the more thickly bedded limestones such as
by folding. Faults may exist, given gaps in the sec- S- 138 (see Fig. 10).

Fig. 9. View of the mudstones and shales of the Agu~in beds exposed at the Agu~in Valley locality. South is to the right of the photograph.
The mudstones are too fine-grained for their com- The strike of the rocks is about 60 ~. They ex-
position to be resolved microscopically. In thin tend northeastwards as far as Cerro Las Lomas, 2
section they vary from silty packstones (S-114) km from the road. They have not been traced to
to recrystallized pelagic foraminiferal wackestones the west, but linear features on the Landsat image
(S-144). Bioclastic and organic content varies as (Fig. 2) suggest that they continue in that direc-
does quartz content. Sedimentary structures were tion for several kilometers. Their total thickness is
not generally recognized in thin section, though unknown, but must exceed the 520 m of outcrop
some samples contained darker layers of organic exposed on the road.
material. Compaction and broken bioclastic grains The contact between the Agu~in beds and the
were frequently seen, and miniature flame struc- overlying strata exposed at Cerro La Penita is not
tures were found in S-126. X-ray diffraction showed exposed, but the departure from a deep-water marine
the mudstones to be composed predominantly of environment to a continental piedmont environment
quartz with minor amounts of muscovite and cal- is so profound that a major unconformity must exist
cite. between them.
The limestones vary from micrites to packstones
depending on fossil content and several samples Rio Sico locality
contain abundant globigerinids (Figs. 10 and 11)
some of which may be Orbulina (Tony Eva, written This locality, which is 50 km south-southeast of
commun.). In samples such as S-139, the fossils have the Agu~in Valley locality described above, lies on
been compacted and flattened. The composition of the road joining the towns of Bonito Oriental and E1
the limestones as determined by X-ray diffraction is Carbon. A cutting 4 km south of the Sico River ex-
calcite and quartz. poses about 25 m of limestone and mudstones. The
Pink and yellow weathering, very fine-grained rocks, which strike east-west and dip gently to the
material weathering to clay, or powdery, loosely south, are cut by minor faults and deeply weathered
consolidated material was encountered at places in dikes. At the base of the section a fold is developed,
the section and may be layers of volcanic ash. Unfor- and the contact with the underlying schists is a thrust
tunately, none of the friable material was collected (Fig. 12). The measured section is shown in Fig. 13,
for X-ray analysis. and it is seen that, as in the Agu~in locality, two rock

l ~t
~ ,o , i ~
~ .J
<
r
i|-~-'I
il, ]t'!;
I!I "
,~ 1
,
II iji
t!~!i:ll; :ll! [i;I~. !!" ~ - "~
0
o
-I l l- ~
E
o o
iil~.
z
Z "'
.J
I,LI
:=i r ---
9,...~ 0
< =" =o 'iT= ' ! o

"E o 7
~ > o
z
0< <
= nD o
<
......................................... ~ '-~
o ~i.:~ i~ .~i
;-ilt ,..1 . "" .... -
i~ I~ L~. , - . ~ t..a o "~i ,.~ .Q .,.. l.i. i %=
~ ~ - ~ . . . . . .~ ~
,-e=i
~>t "I
i
~I ~i :l.0--l.
~ =is.,<~ I! ~ .- ~ J..~
-" ~
l =
1
.,,~
= :,~~~
!ijt :--:.= i]ii,=~ ltl'~ '~.~-=+. = -.'~?it lll= t,..-~,~ o >: o
.o-.. i!|
~,~111~ : ."; ! . l i =
~,,__~,j,,~i,:,<~=-,~ " ,~,=,- ,..~-,. :.,~ .ll 8 1..11.. = ~ = o,..~ _

.
. '>
~:'~ : o
-" 0 0
~ m
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ rT~"
9i-~ li .-- ,~ ~
!i '~ ~ ~'~ :<
9 ~.. ,.- ,.
-- l,,_ .-, t t [ ~
l.,.~_l~"x."fllliil!llilllll;:illlp: 9: ;4tlliiiiiiiii!li!!!!iiiiiilbi~i!l{li!iIii!l':))':..)':i
" ~ N ik~!!',!!!l!i!!i!i ~ t:l!iJl ~l!!!!!i!)l z ~" i~!~;iil
'""" .~
~ ~ :
N!NIIii!i!IltiiliI[ 9 ;]ttf!ii i !il i i !i i i!iiti-:;;::;i;T4

!iki i iTi i ji,i ~ iN~li i iJi!7i lfll!tlff!!ili!ilri{l!~ ETI ,~=~,"~
EN!!!i!tllll!{ :',::,,: 9... ;t.!Ii!i!.!i!I . i!!i![!j!j(ill!{!P!!!i!li!i!l:!t::~;~7;(;~;::~i
...... :~ ~ ik~i!i',!!Jf!!iii!~ 1!f!:!i !!!l!!!itt =-~ iEiItl':: o_o=.
':':,,,,,"" < -~~~-~~:
~ fl It R t t ~l fi1 t -~ 0 ~I It ~
~- t t ~i fil IiI -~ ~ It t ~-
~ t t ~ fil lil ~~ ~ ~I t R t t ~ I t ~ ..~ ~~" :~

Fig. 11. Photomicrograph of sample S-144 showing abundant globigerinids some of which may be Orbulina.
Fig. 12. Mudstones and shales at the Rio Sico locality folded and thrust over schists of the basement. The thrust plane is at the base of
the Jacob staff. The fold does not persist upwards, so that in the area above the photograph the rocks are completely undisturbed. View
from the north, i.e. west is on the right of the photograph.

78' AGUAN BEDS

RIO SICO L O C A L I T Y
75'-- ......
_~____.-~- . . . .
._-_-_-_- -
- ':"~ ~ , S 62-3
u . _.
70'_ :2 Section ~ more rn~gNe;
LEGEND
~J
:5 :2 green oolour
m --__2
Thickest bed I 8/10', lam mudstone"
f~~--] Mudstone
65'_ _4
:- ~-:--~-- > v
~ ~ - - ~ 3 Sh ale
-_ . . . . . . ---- '~ ,~ ___~ - - -
60'_
[ ! !1 Limestone
ll':.'. : . f : .'11
55' Tough gray-green siliceous indurated i:.1: :'l(:J Impure Limestone
;,.-. mud=done, non-calcareous; calcite
rifled shear surfaces interbedded with
-~,S 59 1-2" shady beds; small faults in outcrop; ~= Stylolite
gray colour but often weathered red -
50'_ Fe stained: shears 132/56E & 60/50SW;
~ , S 58 v fine grained ss, tough, could be 1i
silicified tutf(~ I Calcareous
-~.S 57
~.$60
45'_ Break between "benches" but section "~'-' Fossiliferous
may not be missing
I i I i I l
~ ~" >. " Igneous Dike
40'-- ! ! I I 'i I I I I
Massive gray micritic fossilif Is, calcite
! 1 ! ] '.--F/ veined, fractured, and stylolitized.
1 1 1 1 I I ] , ,,r Shell fragments, foraminifera?, contact F~ Faulted Contact
I ~ I .~,S 70of limestone with =li calcareous green
1 I i 1 I I I
',, I 1 1 1 1 1 lain mudstone or rnicritic shale (?)
1 ! 1 1 ! I I
i I I i I i Mudstone used her~
I I I
" ,
:.][ ~
i
~
i
~ --~..'~-
-,=.S 71
as a textural term
'L -q,S 72-3
30'
~.$64 Amorphous non-calcaleous mudstone
-.*S 65 (containing some fossils in thin section)
25'
-_ ..... "5~C::: ..................... Mudstone (not $ilicified); sandstone
chertified (?), ss is medium grained
20'm
___:_:_ _ _: _ _ N - - -:
(S 67) non calcareous, v fine gr fiat
lamin mudstone; laminae disappearing upwards
15'-- Fining up sequence - beginning of bench 2
I [ I I I I i l I l II
[ 1 i t i i 1 I ! i Section down-faulted about 10',
i t ! I=l~. I J t I ~ l l]
t i I "i;, l l I lq'-I I i Tough, gray, calcite veined micriSc
i l i I t i I l i ! It
10'-- ! 1
[
!
I s
t I
t
I
a t
!
i
i
J
i
i
! !/
I I
S
limestone
i i 94 I i i I i J i id~'i ! i i I rE- ./ 75
I I" i I ! I 1":'.! I I ) $74 Interbedded medium bedded micritic
I l ! i i i 1 I I tv limestone and marly shale (field
i t i t i i ! i l'~ d~-~ion)
i I i , I i i , i J) .'q.S 76
Limestone & mudstone in westward

dipping thrust contact with basement
F
?-9 Highly contorted schists
Fig. 13. Agufin beds as measured at the Rio Sico locality. (The map reference is approximately 40350E,13600N, E1 Carbon sheet, HOJA
3062II.) The beds are near-horizontal; so true thickness is shown.

types are present. The one is a tough, gray-green, both mudstones and limestones contain calcite and
indurated, calcite-veined mudstone with individual quartz; thus there are no pure calcareous or siliceous
beds up to 50 cm thick. Sedimentary structures rocks present in the section.
within the mudstones are limited to planar parallel The state of preservation in the rocks is
lamination with rare ripples and cross-lamination. poor and fossil identification difficult, but sample
Millimeter-scale bands of coarser-grained silt-sized S-59, a siliceous wackestone, contains (Fig. 14)
material were sometimes present. The other rock spumellarian radiolaria and foraminifera, some of
type is a massive, gray, stylolitized, micritic lime- which resemble Orbulina (Tony Eva, written com-
stone which occurs in two units, each about 5 m mun.).
thick. Within each unit individual beds may be up About 7.5 km north-northeast of the Sico River
to 30 cm thick and are separated by thin beds locality a road cut exposes about 5 m of folded,
of argillaceous shale. The rocks, however, are much thinly bedded sediments thrust over andesitic lavas
more indurated and silicified than those of the Agu~in of the Oligocene Matagalpa formation (Fig. 15).
Valley locality. They have not been examined in any detail, but
The mudstones are microcrystalline and range in superficially resemble the rocks studied at the Rio
texture from packstones to wackestone; none con- Sico locality. The apparent strike of the thrust plane
tain more than about 15% bioclastic material. Some is east-west, whereas at the Rio Sico locality it is
of the more silty varieties contain plant fragments. approximately north-south, so that it seems that the
In thin section the mudstones, e.g. samples S-59, rocks have been transported from a source lying
S-63, and S-67, are siliceous or cherty, and contain within the northwestern quadrant.
recrystallized spherical microfossils filled with mi-
critic or siliceous material. In a few samples, such as Paleoenvironment
S-65, millimeter-scale planar or wavy laminae and
pseudo-breccias are preserved. Thin sections of the Composition, sedimentary structures, and fauna
limestones, such as S-70, show them to be micrites are consistent with the interpretation that the rocks
containing abundant plant fragments and recrystal- of the two localities are hemipelagic to pelagic
lized fossil fragments. X-ray diffraction showed that slope deposits formed near a carbonate platform. No
Fig. 14. Photomicrograph of sample S-59 showing spumellarian radiolaria and poorly preserved foraminifera. The globular specimens
may be Orbulina.

complete turbidite units were seen, but parallel lam- Table 1

ination indicates modification by bottom currents. 87Sr/86Sr ratios of limestones from the Agu~n beds
The minor beds of calcarenite were probably em-
placed by some type of grain flow mechanism or are Sample 87Sr/86Sr
the result of reworking by currents. R~ Sico locali~
$66A 0.70636
Lithologic age $79 0.70606
Agudn Riverlocali~
The microfossils are poorly preserved. If Or- S-134 0.70657
bulina has been correctly identified the rocks are S-136 0.70603
S-141 0.70598
no older than Middle Miocene (zone N9 of Blow, S-145 0.70663
1969) but could be as young as his zone N23, i.e. S-148 0.70693
Holocene. No finer control is possible due to the S-159 0.70686
absence of age-diagnostic globorotalids.
Analyzed on Finnigan MAT 261 multicollector mass spectrome-
In an attempt to overcome the shortcomings of ter. Analysis of SRM987 yielded 87sr/g6sr + 0.71023 • 1.
the micropaleontology, strontium isotope ratios were
measured in limestones from the Agu~in Valley lo-
cality. The 87sr/g6sr ratios range from 0.70598 to different histories and may have been derived from
0.70693 (Table 1). These values are far lower than different basins. We correlate them on the basis
those encountered in seawater over the Tertiary and of being deposited in similar environments, from
are found only in the Late Jurassic (McArthur, rather tenuous fossil evidence, and in the knowledge
1994). Most of the samples, however, were collected that the Rfo Sico rocks were transported from the
from a part of the section that had been extensively northwest.
intruded by mafic dikes (Fig. 10), and it is likely If the assumption is made that these rocks were
that Sr of low isotope ratio derived from these equi- laid down in marine basins associated with the
librated with the pool of Sr of high isotopic ratio strike-slip margin, it may be argued from a tectonic
originally incorporated in the sediments. standpoint that they can be no older than Miocene,
From the difference in induration between the because before this time the part of Honduras where
rocks of the two localities it is clear that they had they are found was bounded on the north by the
Fig. 15. Agu~in-typebeds folded and thrust over Oligocene Matagalpa lavas. South is to the left of the photograph.

continental mass of Guatemala and Belize. This is problematic, but it could be due to their being
fact may be deduced from the magnetic anomalies restricted to the region between the valley walls, so
in the Cayman Trough (Rosencrantz et al., 1988), that they would be hidden beneath alluvium. Their
which indicate that the present location of Trujillo presence is suggested by the magnetic anomaly off
did not lie east of the Guatemala-Belize coastline the mouth of the Agufin (Pinet, 1971). To explain the
before the Early Miocene, i.e. before 20 Ma. See thrust sequences near the Sico River it is proposed
Rosencrantz et al. (1988, fig. 11) and Mann et al. that the transtensional phase ended and new strike-
(1991, fig. 2C). A Miocene age is compatible with slip faults were initiated. Where the basins were
the presence of ash in the rocks of the Agufin cut by segments of these faults that were in trans-
locality, which could have been derived from the pression, the sediments were deformed into flower
voluminous ignimbritic volcanism associated with structures, and uplift and thrusting of the deepest
the Middle America Trench that produced the Padre parts of the basin occurred, cf. Mann and Gordon
Miguel volcanics coveting much of central Honduras (1996).
(Donnelly et al., 1990).
Timing of dike emplacement ACKNOWLEDGEMENTS
A critical question is whether the deformation We thank Tony Eva and Donna Meyerhoff Hull
of the Agufin beds occurred before or after the for assistance with the paleontology and J.B. Barton
emplacement of the dikes. Unfortunately, no direct of Rand Afrikaans University for use of an X-ray
evidence exists because in neither locality were the diffractometer. Rob Rogers kindly checked on some
dikes seen to cut across folds. In the Agufin locality, of our observations.
however, the dike with which piercement structures
were associated presently dips north at 30 ~ but the
upturning of the beds is symmetric on either side of APPENDIX
it as if the sediments had been horizontal when it
was intruded. Zircon was separated from a sample of E1 Carbon granite and
was sized into fractions. Ten to twenty milligrams of each were
No petrography, geochemistry or dating was done
spiked with 235U and were decomposed in Teflon capsules at
on the dikes because they are extensively altered. 200~ An aliquot was taken and spiked with 2~ to measure
It may be stated that the apparent homogenization Pb concentration. Lead and uranium were separated by standard
of strontium isotopes is consistent with the dikes ion-exchange techniques. Isotope ratios were measured on a
having been intruded before faulting, because the single collector NBS-type mass spectrometer. The Pb blank was
1 ng Pb, large by today's standards but nonetheless an acceptable
environment of a fault or thrust zone, where fluids
fraction (1 to 2%) of the Pb processed. The results are shown
readily circulate through fractured rock, presents in Table A1, and are plotted in a Tera-Wasserberg diagram in
ideal conditions for homogenization. Fig. A1. The points are highly discordant, which is attributed to
Pb loss from zircon having exceptionally high contents of U.
Tectonic implications
The occurrence of deep-water marine strata on

the north coast of Honduras is not easily explained,
and it is difficult to make regional tectonic syntheses
from sedimentologic observations at two small road
cuts. The most conservative and least provocative
hypothesis is one which envisages that the strike-slip
margin extended as far south as E1 Carbon in the
Tertiary and was totally submerged, so that the Tela
Basin was twice the width it is today. In its southern
0.04780
portion a number of deep strike-slip basins formed Q_
8 ~ ~~i ~ ~ .5*-z3my
that were filled with turbidites. The region was up-
o
e~
Q_
0.04775
lifted, but faults curving at a steep angle to the e,l
plate boundary cut across some of these basins, pre- 004770 5O
serving the sediments by downfaulting. Such faults

necessarily produced highly transtensional regions
which were locally intruded by mafic dikes. In this ' ',oo . . . . ,;o . . . . 2oo'
scenario the Agufin locality would represent a block 27,8U / 206 Pb
of sediments preserved by downfaulting. The lack of Fig. A1. Tera-Wasserberg plot of zircon data from E1 Carbon
evidence for dikes at the eastern end of the valley granite sample, X-3.

THE S O U T H E R N F L A N K OF THE T E L A BASIN, H O N D U R A S 235
Table A 1
Uranium and lead concentrations, lead isotope ratios and corrected daughter/parent ratios for zircons from E1 Carbon granite
Mesh range U Pb 2~176 2~176 2~176 2~ 2~

(~tm) (ppm) (ppm)
100-200 1477 20.0 0.0031 0.09449 0.2214 0.07460 0.01122
170-200 1747 20.2 0.0027 0.08778 0.2101 0.06472 0.00980
200-230 1261 15.8 0.0014 0.06864 0.1589 0.07630 0.01152
230-270 1447 17.7 0.0014 0.06889 0.1654 0.07427 0.01120
270-325 2489 18.7 0.0023 0.08179 0.2010 0.04222 0.00647
Common Pb correction from K-feldspar: 2~176 -- 18.695; 2~176 = 15.646; 2~176 -- 38.566.
REFERENCES (Editors), Subduction: Top to Bottom. Am. Geophys. Union,

Geophys. Monogr., 96:143-154.
Aves, H.S., 1983. Tela Basin. Unpublished Report. Helchris Mann, E, Hempton, M.R., Bradley, D.C. and Burke, K., 1983.
Associates, Dallas, TX, 38 pp. Development of pull-apart basins. J. Geol., 91: 529-554.
Banks, N.G. and Richards, M.L., 1969. Structure and bathymetry Mann, E, Tyburski, S.A. and Rosencrantz, E., 1991. Neogene
of western end of Bartlett Trough, Caribbean Sea. In: A.R. development of the Swan Islands restraining-bend complex,
McBirney (Editor), Tectonic Relations of Northern Central Caribbean Sea. Geology, 19: 823-826.
America and the Western Caribbean - - the Bonacca Expedi- Manton, R.S. and Manton, W.I., 1989. A major Late Cretaceous
tion. Am. Assoc. Pet. Geol. Mem., 11: 229-243. dextral strike-slip zone on the north coast of Honduras. Geol.
Blow, W.H., 1969. Late Middle Eocene to Recent planktonic Soc. Am., Abstr. Progr., 21: 203-204.
foraminiferal biostratigraphy. Proceedings First International Manton, W.I., 1987. Tectonic interpretation of the morphology of
Conference on Planktonic Microfossils Geneva, 1967, Vol. 1, Honduras. Tectonics, 6: 633-651.
pp. 199-422. Manton, W.I., 1996. The Grenville of Honduras. Geol. Soc. Am.,
Caceres Avila, E, Tappmeyer, D.M., Aves, H.S., Gillett, M. and Abstr. Progr., 28: 187.
Klenk, C.D., 1984. Recent studies of basins are encouraging McArthur, J.M., 1994. Recent trends in strontium isotope stratig-
for future exploration in Honduras. Oil Gas J., 42:139-149. raphy. Terra Nova, 6: 331-358.
DeMets, C., Gordon, R.G., Argus, D.E and Stein, S., 1990. McBirney, A.R. and Bass, M.N., 1969. Geology of the Bay
Current plate motions. Geophys. J. Int., 101: 425-478. Islands, Gulf of Honduras. In: A.R. McBirney (Editor), Tec-
Donnelly, T.W., Horne, G.S., Finch, R.C. and Lopez-Ramos, tonic Relations of Northern Central America and the Western
E., 1990. Northern Central America; the Maya and Chortis C a r i b b e a n - the Bonacca Expedition. Am. Assoc. Pet. Geol.
blocks. In: G. Dengo and J.E. Case (Editors), The Caribbean Mem., 11: 229-243.
Region. The Geology of North America, Vol. H, The Geologi- Minster, J.B. and Jordan, T.H., 1978. Present-day plate motions.
cal Society of America, Boulder, CO, pp. 37-76. J. Geophys. Res., 83: 5331-5354.
Foye, W.G., 1918. Notes on a collection of rocks from Honduras, Muehlberger, W.R., 1976. The Honduras depression. Publ. Geol.
Central America. J. Geol., 26:524-531.
ICAITI, 5: 43-54.
Gordon, M.B. and Muehlberger, W.R., 1994. Rotation of the
Pindell, J.L. and Barrett, S.E, 1990. Geologic evolution of the
Chortis block causes dextral slip on the Guayape fault. Tecton-
Caribbean region. In: G. Dengo and J.E. Case (Editors), The
ics, 13: 858-872.
Caribbean Region. The Geology of North America, Vol. H,
Heubeck, C. and Mann, E, 1991. Geologic evaluation of plate
The Geological Society of America, Boulder, CO, pp. 405-
kinematic models for the North American-Caribbean plate
432.
bounding zone. Tectonophysics, 191: 1-26.
Pinet, ER., 1971. Structural configuration of the northwestern
Horne, G.S., Clark, G.S. and Pushkar, E, 1976a. Pre-Cretaceous
Caribbean plate boundary. Geol. Soc. Am. Bull., 82: 2027-
rocks of Northwestern Honduras: basement terrane in Sierra
2032.
de Omoa. Am. Assoc. Pet. Geol. Bull., 60: 566-583.
Horne, G.S., Pushkar, E and Shafiquallah, M., 1976b. Laramide Pinet, ER., 1972. Diapirlike features offshore Honduras: impli-
plutons on the landward continuation of the Bonacca Ridge. cations regarding tectonic evolution of Cayman Trough and
Publ. Geol. ICAITI, 5: 84-90. central America. Geol. Soc. Am. Bull., 83: 1911-1922.
Ivey, M.L., Breyer, J.A. and Britton, J.C., 1980. Sedimentary Pinet, ER., 1975. Structural evolution of the of the Honduras
facies and depositional history of the Swan Islands, Honduras. continental margin and the sea floor south of the western
Sediment. Geol., 27:195-212. Cayman Trough. Geol. Soc. Am. Bull., 86: 830-838.
Kozuch, M.J., 1991. Mapa geologico de Honduras. Segunda Edi- Pinet, ER., 1976. Morphology off northern Honduras, northwest-
tion. Instituto Geografico Nacional, Tegucigalpa, Honduras. ern Caribbean Sea. Deep-Sea Res., 23: 839-847.
Letouzey, E, 1995. Interpr6tation gdologique de la r6gion du Powers, S., 1918. Notes on the geology of eastern Guatemala
Rio de Aguan (Honduras) et du bassin de la basse Mag- and northwestern Spanish Honduras. J. Geol., 26: 507-523.
dal6na, Sierra de San Jacinto (Colombie) ~ l'aide des images Roberts, R.J. and Irving, E.M., 1957. Mineral deposits of central
Landsat multispectrales tra~tdes numdriquement h l'6chelle America. U.S. Geol. Surv. Bull., 1034: 1-205.
du 1/500,000. In: A. Mascle (Editor), Gdodynamique des Rosencrantz, E. and Mann, E, 1991. SeaMarc II mapping of
Cara'/bes. Editions Technip, Paris, pp. 439-450. transform faults in the Cayman Trough, Caribbean Sea. Geol-
MacDonald, W.D., 1976. Cretaceous-Tertiary evolution of the ogy, 19: 690-693.
Caribbean. Transactions of the 7th Caribbean Conference, Rosencrantz, E., Ross, M.I. and Sclater, J.G., 1988. Age and
Guadeloupe, pp. 69-81. spreading history of the Cayman Trough as determined from
Mann, E and Gordon, M.B., 1996. Tectonic uplift and exhuma- depth, heat flow, and magnetic anomalies. J. Geophys. Res.,
tion of blueschist belts along transpressional strike-slip fault 93: 2141-2157.
zones. In: G.E. Bebout, D.W. Scholl, S.H. Kirby and J.E Platt Sapper, K., 1905. Uber Gebirgsbau und Boden des s~idlichen

Mittelamerikas. Petermanns Mitteilungen, Erganzungsheft Caribbean Plate during the last 7 million years and implica-
151, Justus Perthes, Gotha, 82 pp. tions for earlier Cenozoic movements. J. Geophys. Res., 87:
Southernwood, R., 1986. Late Cretaceous Limestone Clast Con- 10,656-10,676.
glomerates of Honduras. M.S. Thesis, Univ. of Texas at Dallas, Von der Hoya, H.A., 1986. A Reflection and Seismic and
300 pp. Magnetic Investigation of the Tela Basin: Northern Offshore
Stein, S., DeMets, C., Gordon, R.G., Brodholt, J., Argus, D., Honduras. M.S. Thesis, Southern Methodist Univ., Dallas, TX,
Engeln, J.E, Lundgren, P., Stein, C., Wiens, D.A. and Woods, 112 pp.
D.E, 1988. A test of alternative Caribbean Plate relative mo- Williams, H. and McBirney, A.R., 1969. Volcanic history of
tion models. J. Geophys. Res., 93:3041-3050. Honduras. Univ. Calif. Publ. Geol. Sci., 85: 1-101.
Sykes, L.R., McCann, W. and Kafka, A.L., 1982. Motion of

Chapter 10
Cretaceous Microfaunas of the Blue Mountains, Jamaica, and of

the Northern and Central Basement Complexes of Hispaniola
HOMER MONTGOMERY and EMILE A. PESSAGNO, JR.
Radiolaria and foraminifera isolated from matrix and observed in thin section refine or extend previously known or suspected
chronostratigraphic assignments for Mesozoic rocks from the Blue Mountains of Jamaica and from the Puerto Plata basement
complex, the Dajabon area, the Rfo San Juan complex, and the Siete Cabezas basaltic rocks of the Dominican Republic.
Microfossils are found only rarely and in isolated sites some of which have questionable field relations with the surrounding
complex. New chronostratigraphic assignments indicate that at least some of the basalt of the Blue Mountains, Jamaica, is older
than suggested by previous paleontological studies. In the Dominican Republic, Upper Cretaceous radiolaria in the San Marcos
Formation and within volcanics of the Puerto Plata basement complex refine previous chronostratigraphic assignments. Upper
Cretaceous radiolaria from chert associated with the Siete Cabezas basaltic rocks of Arroyo Bermejo apparently constrain the age
of basalt in this area. Upper Cretaceous foraminifera from Dajabon and from the Rfo San Juan complex are the first reported
fossils from these areas.
INTRODUCTION Isotopic ages, where determined, may date un-

certain events which could be wide-scale magma-
The Late Cretaceous to Eocene Great Arc (Fig. 1) tism and/or plate collision but are not necessarily
of the Caribbean existed as an east-facing vol- ages of deposition. Paleontological control associ-
canic island chain spanning the gap between North ated with isotopic ages remains almost nonexis-
and South America. During the Eocene-Oligocene tent. Rare sedimentary rock bodies, many of which
the arc collided with the Bahama platform in a di- go unrecognized, have never been sampled or at
achronous event that produced an east-west-trending least have never been rigorously investigated for
strike-slip plate boundary on the north- and the fossils.
east-facing subduction zone present today along the A note relevant to GPS locations of outcrops
eastern boundary. in this report is in order. All locations in Jamaica
Recognition, dating, and correlation of compo- and locations in the Dominican Republic of DR92
nents of the Great Arc remain an on-going puzzle. and DR95 samples were located using map datum
Several relatively well-studied terrane fragments of WGS84. Use of this datum produces a map error
prominence in discussions of plate tectonic interac- of approximately 300 m in latitude. DR96 samples
tions remain so poorly age-constrained that labels were located using NAD 1927-Caribbean. These
such as Mesozoic may be applied only with caution. localities should be accurately located.
Examples of such units include the metamorphic
complexes of Rio San Juan and Samami in the Do-
minican Republic. Other areas are broadly mapped MICROFAUNAS AND AGES
with little lithological inventory and with no age
control of some of the units. An example is the Da- The Cretaceous faunas presented in this report
jabon area of the Dominican Republic. In this report are derived from rocks collected in Jamaica and in
we add to the body of paleontological information the Dominican Republic (Fig. 2). Most of the rocks
from problematic Cretaceous units in Jamaica and were collected in situ. As noted below, two samples
the Dominican Republic. collected in Jamaica are from fiver gravels. The

238 H. MONTGOMERY and E.A. PESSAGNO
Fig. 1. Late Cretaceous snapshot of possible paleogeographic setting of the Great Arc of the Caribbean.
gravels were sampled because no fossiliferous rocks in one of those samples, a cobble from the Devil's
were discovered cropping out in this area of the Blue River, Jamaica. Thin sections of limestone revealed
Mountains. identifiable larger and planktic foraminifera.
Radiolaria and foraminifera were isolated from The paleontological details presented modify or
siliceous rocks via extraction in dilute hydrofluoric amplify what was previously known about sev-
acid. Fossils vary from poorly to well-preserved. eral units (Blue Mountains, Jamaica; San Marcos
Foraminifera extracted from chert via the same pro- Formation, and Arroyo Bermejo, Dominican Re-
cess are fluoritized and are only rarely identifiable or public) or are the first assignments made (Puerto
are identifiable only to genus. Thin sections were cut Plata basement complex of Maimon Bay, Dajabon
of most of the siliceous samples that had produced cherts, and Rfo San Juan u all of the Dominican
no microfossils in residue. Radiolaria were present Republic).
Fig. 2. Location map of outcrops sampled for this report.

CRETACEOUS MICROFAUNAS OF JAMAICA AND HISPANIOLA 239
MICROFAUNAS OF THE DOMINICAN REPUBLIC Kg

DR95.8a
Puerto Plata basement complex (DR92.8a)
Limestone nodules in volcanic matrix are found Maimon Bay

along the margin of Maimon Bay (Fig. 3). Lime- Puerto
stone is present in four distinct forms. Light gray Nm Plata
nodules approximately 10 to 15 cm in diameter are
too highly metamorphosed to yield fossils. A few
dark gray nodules produced radiolaria and planktic
foraminifera. Buff-colored limestone adjacent basalt
pillows yielded no fossils. Select pieces of brec-
ciated, greenish-gray limestone (sample DR92.8a)
and crudely bedded but discontinuous dark gray,
silicified limestone proved variably fossiliferous.
--x Kv f N
Two pieces of greenish-gray limestone within highly
weathered volcanic matrix and adjacent crumbly red
umber yielded well-preserved radiolaria and planktic
foraminifera. Pieces of all limestone samples were
dissolved in HC1 and thin sectioned.
The planktonic foraminifera of sample DR92.8a Pi
Nm
0
l2 km
include Globotruncana arca, Globotruncana bul- i I I
loides, Globotruncana hilli, Globotruncana lappar- "N to Santiago
enti s.s., Globotruncana linneiana, Globotruncanita
calcarata, Globotruncanita elevata, Globotrun- Fig. 3. Location map for sampled outcrops of the Puerto Plata
area, Dominican Republic. Kv -- volcanics; Kg -- gabbro; Ks --
canita stuartiformis, and Heterohelix sp. The bios- serpentinite; Pi = Imbert Fm." Nm = San Marcos unit.
tratigraphic determination is upper Campanian.
Radiolaria of sample DR92.8a are some of the
best we have observed in Mesozoic Caribbean
rocks. HC1 residues include Alievium gallowayi from Puerto Plata to Santiago (Fig. 3). The exposed
(base of Coniacian-top of Campanian), Amphipyn- portion of the boulder is the size of a small car and is
dax pseudoconulus (base of Campanian-top of associated with a mass of fractured serpentinite and
Maastrichtian), Archaeodictyomitra squinaboli (up- limestone in muddy matrix. Radiolaria from sample
per Albian-top of Campanian), Dictyomitraformosa DR95.1a are lower Santonian to lower Campanian
(base of Coniacian-top of Campanian), Patellula (Table 1) and are the best-documented radiolarian
verteroensis (Campanian), Pseudoaulophacus par- assemblage of the San Marcos unit.
gueraensis (base of Santonian-top of Campanian).
Although ranges vary somewhat among authors, Dajabon subcomplex(?) of the Upper Duarte
DR92.8a radiolaria are clearly Campanian. complex(?) (DR95.12), UTM: 2167651; 219445
San Marcos Formation, Puerto Plata basement Abundant, multicolored (red, orange, blue, green,
complex (DR95.1a; Pindell and Draper (1991) black, brown, gray) chert rubble is scattered across
locality No. 31), UTM: 2188790; 307775 several low hills northeast of the town of Dajabon
(Fig. 4). No bedded rocks crop out in this area of
Abundant and well-preserved radiolaria are low relief. Dissolution of a few of the chert samples
present in a dark red jasper boulder located near yielded Albian to upper Cenomanian foraminifera
Imbert along the east side of the highway that runs (Table 2).
Table 1
Radiolaria from San Marcos Fm., Puerto Plata basement complex, Dominican Republic sample DR95.1a
Radiolaria present Chronostratigraphic assignment

Alievium gallowayi base Santonian-middle-upperCampanian
Amphipyndax stocki
Dictyomitraformosa Coniacian-lower Campanian
Praeconocaryoma universa base Coniacian-middle Campanian
Pseudoaulophacus lenticulatus lower Coniacian-upper Campanian

Table 4
Foraminifera from the Rio San Juan complex, Dominican Re-
public sample DR96.1
Foraminifera present Chronostratigraphicassignment

Orbitoides sp. upper Campanian-upper Maastrichtian
Heterohelix sp. upper Albian-upper Maastrichtian
Rio San Juan complex (DR96.1), UTM: 2165512;

DR95.12 380118
Limestone at this site is located in the fiver at the

end of the road south of Magante (Fig. 5). Much of
the limestone is metamorphosed, especially where
N interbedded with schist. Farther away from the schist
the limestone is sporadically but richly fossiliferous.
Among other microfossils and broken megafossil
fragments, thin sections revealed upper Campanian
Dajabon
t'4 km to Maastrichtian foraminifera (Table 4).
I
Arroyo Bermejo (DR96.9), UTM: 2056799;

385880
Fig. 4. Location map for chert sampled in the Dajabon area,
Microfossiliferous, black, gray, and greenish bed-
Dominican Republic.
ded chert of the Siete Cabezas Formation is found
in the cut bank of the stream approximately 100
m south of the bridge that leads east of the Duarte
highway (Fig. 6). Radiolaria are probably middle
Rio San Juan complex (DR95.7a), UTM: 216820;
Campanian to Maastrichtian (Table 5). The critical
37675
specimen for this chronostratigraphic assignment is
Dictyomitra multicostata, and while slightly bro-
Sandy red mudstone dug from the muddy shale
ken, is clearly this taxon. Other specimens suggest
in the roadcut at the top of the first hill south of the
a Coniacian assignment. Problematically, we have
town of Magante (Fig. 5) yielded poorly preserved
not observed any specimens of Dictyomitra formosa
foraminifera (Table 3). The identifications, if correct,
which is usually found associated with Dictyomitra
assign the sample to the uppermost Cenomanian on
multicostata. We suggest the Coniacian specimens
the basis of the concurrence of Dicarinella and
are reworked.
Thalmanninella.
Table 2 SIGNIFICANCE OF THE DOMINICAN REPUBLIC

Foraminifera from the Dajabon subcomplex(?) of the Upper MICROFAUNAS
Duarte complex(?), Dominican Republic sample DR95.12
Puerto Plata basement complex
Forams p r e s e n t Chronostratigraphic
assignment
Heterohelix moremani upper Albian-lower Turonian The Puerto Plata basement complex is composed
Hedbergella sp. Barremian-Maastrichtian
of fault-bounded sections of serpentinized peridotite,
Spongodiscacea radiolaria
harzburgite, layered cumulate ultramafics and gab-
Table 3
Foraminifera from the Rfo San Juan complex, Dominican Republic sample DR95.7a
Foraminifera present Chronostratigraphic assignment

Hedbergella sp. cf. delrioensis
Thalmaninella (poorlypreserved) uppermost Cenomanian
Double keeled forms such as: Dicarinella sp., Marginotruncana sp. uppermost Cenomanian

ATLANTIC OCEAN
to
ar Hernandez to
Rio San Juan
DR95.7a
DR96.1
Fig. 5. Location map for sampled outcrops of the Rfo San Juan area, Dominican Republic. S = Gaspar Hernandez serpentinites; H =
Hicoteca schists.
bros, massive gabbros, and basic to intermediate plate remained conjectural without a chronostrati-
volcanics (Pindell and Draper, 1991). The age of graphic framework. The discovery of Campanian
the Puerto Plata basement complex is an enduring faunas allows comparison with several other circum-
problem. Caribbean terranes.
The Puerto Plata basement complex was inter- Campanian radiolaria are well-known in the
preted as a pre-Paleocene ophiolite complex (Pin- Caribbean and surrounding regions from Puerto
dell and Draper, 1991) emplaced as part of a Rico, Cuba, Texas, Panama, and DSDP Sites 146)
forearc/accretionary prism of the Great Caribbean (Venezuelan basin), 4 (adjacent to the Bahama plat-
Arc (Mann et al., 1991). But the paleogeographic reform), 95 (200 km northwest of Cuba), and 24
lationship of the Puerto Plata basement complex both (western Brazil basin). Recently, Campanian radio-
to other Hispaniola terranes and to the Caribbean laria identical to the Puerto Plata basement complex
fauna were isolated from red chert (samples pro-
vided by M. Iturralde-Vinent of the Museo Nacional
Ud\ de Historia Natural, Habana) in a chert and basalt
sequence of the Santa Teresa or Quifiones Formation
Bonao \ < ) \ ]
,-o-
\ , o \, of the Bahfa Honda quadrangle (H. Montgomery, un-
publ. data). Brecciated greenish gray foraminiferal
N K \ U s i~ -.
,i~ .. Villa'~k--~ ~ us and radiolarian limestone of Campanian age drilled
I Altagracia-~ \X / ~' at Site 146 in the Caribbean Sea between Hispaniola
and Venezuela (Edgar et al., 1973) and examined by
0"Q" Morin (1982) are the age and lithological equivalents
0' 4 km I~" Ws~X of the Puerto Plata basement complex limestones. A
fauna examined by Foreman (1973) containing many
9 to I, Id ",~X._~ \ of the same species as the Puerto Plata basement
Santo Domingo \ ~ "N~a complex was recovered in the Gulf of Mexico at
Fig. 6. Location map for chert sampled along Rfo Isabela at DSDP Site 95 from olive-gray, dolomitic limestone,
Arroyo Bermejo in the Siete Cabezas Formation. Us = Siete some of which was interbedded with black chert.
Cabezas Fm.; Ud = Duarte Fm. Based on the planktic microfaunal assemblage, the

Table 5
Radiolaria from Arroyo Bermejo, Dominican Republic sample DR96.9

Acanthocircus spp. Triassic-Upper Cretaceous
Amphipyndax sp. cf. A. stocki
Alievium praegallowayi lower Coniacian-lower Santonian
Archaeosponoprunum triplum Coniacian
Archaeodictyomitra sp. Middle Jurassic-Upper Cretaceous
Artostrobium sp.
Dictyomitra multicostata middle Campanian-Maastrichtian
Microsciadiocapsa cortinaenis Coniacian
Patellula sp. Cenomanian-Maastrichtian
Pseudoaulophacus sp. middle Turonian-upper Campanian
Pseudodictyomitra sp. Upper Jurassic-middle Turonian
Theocampe sp. cf. tina
Puerto Plata basement complex fauna was deposited phosed version of the Duarte may be confused with
at abyssal depths probably slightly above the CCD. basaltic rocks of the Magua Formation (considered
Eocene). There is even a chance that these rocks are
San Marcos Formation part of the Maimon-Amina schists complex. Further
mapping will be required.
The San Marcos Formation was described as hav-
ing been emplaced by mud diapirism and lateral Rio San Juan complex
flow sampling all of the pre-Quaternary stratigraphic
units of the Puerto Plata area (Pindell and Draper, The Rio San Juan complex of the northeastern
1991). Within this emplacement scenario the jasper Dominican Republic is composed predominantly of
block sampled for this report would have been de- igneous and metamorphic rocks. Conglomerate in
rived from the Puerto Plata basement complex. Such sandstone/tuff sequences is similar to strata of the
a scenario is consistent with chronostratigraphic as- Imbert Formation of the Puerto Plata area (Draper
signments for Campanian limestone of the Puerto and Nagle, 1991). No fossil-bearing rocks have been
Plata basement complex at Maimon Bay. reported for the Rio San Juan complex. Superposi-
Ages reported for the San Marcos are Early tion is not particularly age-defining as sedimentary
Cretaceous (Bourgois et al., 1982) and Miocene rocks surrounding the Rio San Juan complex are of
(Nagle, 1966). Lower Santonian to lower Campanian the Upper Eocene La Toca Formation.
jasper blocks such as our sample DR95.1a have not Microfossils of DR95.7a and DR96.1 are the first
been previously recognized in any other part of the reported from the Rio San Juan complex. In spite
San Marcos. of this seemingly important discovery, it remains
unclear if the fossils date metamorphic rocks of
Dajabon subcomplex(?) of the Upper Duarte the complex, or are themselves two separate over-
complex(?) lying or incorporated units. The site is mapped as
La Toca Formation of the Mamey group which
The age of the Dajabon rocks is important to is Upper Eocene, presenting the possibility that
decide whether they are a westerly component of the foraminifera are somehow reworked. G. Draper
the Duarte complex or are they of some other (written communic., 1997) suggested the sample
association. A Jurassic age for the cherts would lend may be nonconformable on the coherent part of the
credence to a Duarte association. Dajabon sample Rio San Juan high-pressure rocks comparing such an
DR95.12 presented an older age for the Dajabon occurrence to a similar relationship from the Purial
area than any previously reported, but the chert is area in Cuba.
clearly not as old as the Jurassic red ribbon chert The significance of the upper Cenomanian
of the Duarte complex in the Central Cordillera chronostratigraphic assignment for DR95.7a remains
(Montgomery et al., 1994a). tentative. The foraminifera are poorly preserved.
The Dajabon region remains poorly mapped and Having been collected from a mud matrix not unlike
there is a problem with identification of the Duarte the San Marcos Formation in appearance, we assume
lithologies in this region. Other units in the Dajabon that this rock is not a genetic component of the Rio
area are mapped as Tertiary, and DR95.12 is clearly San Juan metamorphic complex. The Imbert Forma-
not associated with these rocks. A less metamor- tion is found in the northern part of the Rio San Juan

Kbd Spring Bank

Bath Fountain i
g.,-:
Kcp
/ ,
/ Kcp Kbd (/)\ DR95.5
Kcp j
.
~-...~.,.
\
!
! DR9S.10
Kbd
~--_~_~.~,-,,___: \. K b d ~
N
Bath
I
to Port Morant
Fig. 7. Location m a p for s a m p l e d outcrops near Bath, Jamaica. Kcp -- Cross Pass Fm.; Kr : Rfo G r a n d e Fm.; Kbd : B a t h - D u n r o b i n Fm.
complex (Draper and Nagle, 1991), but this material and Coniacian assignments. Chris Sinton (J. Lewis,
is unlike the sandstone, serpentinite conglomerate, written commun., 1996) obtained Ar-Ar ages of
and tuff components of the Imbert. The Imbert is 69 + 0.7 Ma whole rock and 68.5 + 0.5 Ma pla-
also younger, being Paleocene(?)-Eocene (Pindell gioclase for the Siete Cabezas basalt located several
and Draper, 1991). kilometers north of Arroyo Bermejo. Assuming re-
The limestone of sample DR96.1 is rich in working of the Coniacian specimens, our assignment
calcium carbonate fossils and fossil fragments agrees relatively well with the Ar-Ar ages.
including lagoonal miliolid foraminifera, larger
foraminifera, various algae, echinoderm fragments,
and other calcareous debris normally encountered M I C R O F A U N A S OF J A M A I C A
in thin sections of sediments deposited in shal-
low water. The seemingly inconsistent element is All of the samples were collected in the Blue
the presence of deeper-water planktic foraminifera. Mountains of southeastern Jamaica (Fig. 7).
Such faunal mixing is not unknown in the Caribbean
in various sections with shallow-water fossils having Devil's River (JAM95.5a), UTM: 1976631;
cascaded into adjacent deeper waters (Montgomery, 344535
1997).
The Campanian to Maastrichtian chronostrati- A dark red jasper cobble collected from the
graphic assignment for the DR96.1 limestone is Devil's River stream bed where it crosses the road
consistent with the 4~ cooling age of 85.1 east of Bath yielded Santonian to lower Campanian
Ma for a hornblende-phengite-bearing block from radiolaria (Table 6) with possibly reworked Pseu-
the Jagua Clara M61ange (Draper and Nagle, 1991). doaulophacus sp. cf. praefloresensis. The origin of
The dated block is from the m61ange part of the this sample is uncertain but the Devil's River drains
Rfo San Juan, but no dates were forthcoming from the Blue Mountains block.
the area that contains sample DR96.1. Draper (G.
Draper, written commun., 1996) is uncertain whether Devil's River (JAM95.5b), UTM: 1976631;
the fossiliferous limestone is part of the metamor- 344535
phic part of the Rfo San Juan complex. Over the hill
from the fossiliferous limestone are outcrops of Im- A thin section of a red jasper cobble from the
bert Formation. The question remains whether this Devil's River stream bed at the above locality
fossiliferous limestone is a part of an unmetamor- (JAM95.5a) yielded Coniacian to lowermost San-
phosed Upper Cretaceous to Paleocene sequence. tonian radiolaria (Table 7).
More mapping will be required to resolve this prob-
lem. Devil's River (No. 802), locality: third cascade of
the Devil's River northwest of the village of
Arroyo Bermejo Spring Bank
This chronostratigraphic assignment for the Ar- This sample of reddish calcareous mudstone was
royo Bermejo chert is not completely consistent with collected by John Lewis and Edward Robinson (Uni-
that of Boisseau (1987) who presented Cenomanian versity of the West Indies museum sample No. 802).

Table 6
Radiolaria from Devil's River, Jamaica sample JAM95.5a

Alievium gallowayi base Santonian-middle-upperCampanian
Dictyomitra formosa Coniacian-lower Campanian
Dictyomitra torquata Coniacian-lower Campanian
Pseudoaulophacus sp. cf. praefloresensis middle-upper Turonian
Patellula sp.
Table 7
Radiolaria from Devil's River, Jamaica sample JAM95.5b

Alievium praegallowayi/ superbum base Coniacian-lower Santonian
Dictyomitra torquata Coniacian-lower Campanian
Table 8
Radiolaria from Devil's River, Jamaica sample No. 802

Alievium superbum/praegallowayi middle Turonian-upper Campanian
Archaeodictyomitra sp. Jurassic-Cretaceous
Crucella sp. Jurassic-Cretaceous
Pseudodictymitra n. sp. Upper Jurassic-middle Turonian
Dissolution produced upper Turonian to lower Co- SIGNIFICANCE OF THE JAMAICAN MICROFAUNAS
niacian (Table 8) radiolaria. This rock probably
contains the oldest Jamaican microfossils collected Blue Mountains
for this report.
The Bath-Dunrobin Formation and Wild Cane
complex of Jamaica is composed mostly of basalt,
Bath (JAM95.10), locality: adjacent to dolerite, gabbro, and tonalite with rare limestone
Churchill's Vineyard along the road from the and chert. Foraminifera from the Back Rfo Grande
town of Bath to Bath Spring limestone of the Blue Mountains inlier were reported
as upper Campanian (Krijnen and Chin, 1978). The
The Wild Cane complex outcrop sampled is red, three radiolarian assemblages presented in this re-
thinly bedded chert. Relatively well-preserved mid- port are all older than Campanian. Assuming that
dle-upper Turonian to upper Coniacian radiolaria the chert, jasper, and red mudstone of this report
(Table 9) suggest that JAM 95.10 is one of the are/were interbedded with the basalts, then some of
older samples collected for this report from the Blue the Bath-Dunrobin basalts are older than previously
Mountains block. thought. The JAM 95.10 chert sampled is clearly
Table 9
Radiolaria from Bath, Jamaica sample JAM95.10

Alievium superbum/gallowayi middle Turonian-upper Campanian
Artostrobium urna
Dictyomitra napaensis middle-upper Turonian-upper Coniacian
Vitorfus sp. Albian-?Maastrichtian

incorporated within a volcanic sequence. Thus the (3) The various Cretaceous rocks of Jamaica, the
middle-upper Turonian to upper Coniacian chronos- Dominican Republic, and other Cretaceous Great
tratigraphic assignment for JAM 95.10 indicates the Arc subduction complexes and associated sedimen-
same assignment for the volcanic sequence. The age tary sequences are a collection of Atlantic bank and
for JAM 95.10 suggests that the ophiolitic sequence basin material with no incorporated rocks of Pacific
of the southeast mountains is older than Campanian, affinity. Multicolored Cretaceous chert and jasper are
a proposition long suspected by G. Draper (written common. Red mudstone is present, a rock not ob-
commun., 1997). The Campanian limestone is at the served in the Jurassic sequences. Most obvious is the
top of, or unconformable on, the ophiolitic sequence. abundance of limestone in Great Arc material. No
carbonate rocks are known from the oldest Jurassic
terrane fragments which are found in the Dominican
CONCLUSIONS Republic and Puerto Rico. While not abundant nor
apparent, the oldest limestone is interbedded with
Tectonic scenarios describing the complex origin red ribbon chert in the youngest Jurassic terrane
and development of the Caribbean plate are difficult fragment in La D6sirade.
to construct partly because of poor age-control. One (4) Based on the chronostratigraphy presented in
persistent problem involves the earliest paleogeo- this report and assuming an approximate 90 Ma
graphic history of the Great Arc of the Caribbean date for polarity reversal, various components of the
(for an overview see Mann et al., 1991). Mont- Great Arc were deposited before reversal within the
gomery, Pessagno, and co-workers (Montgomery et arc or on the Atlantic side. The JAM samples are
al., 1992, 1994a,b) described pre-Great Arc paleo- pre-Great Arc. The Rfo San Juan mudstone and the
geography with emphasis on the Pacific origin of Dajabon cherts are pre-Great Arc. The other samples
sedimentary rocks that collected in the west-facing are too close to call or they formed after polarity
trench during the Jurassic. The timing of arc polar- reversal.
ity reversal from west-facing to east-facing marking (5) Based on the chronostratigraphic assignment
the beginning of the Great Arc remains specula- of limestone in the Rio San Juan complex, and as-
tive. Based on the age of the youngest chert in suming that this limestone is incorporated within the
the Bermeja accretionary complex of Puerto Rico, subduction complex, the Great Arc of the Caribbean
Montgomery et al. (1994a) suggested that the flip apparently encountered and subducted part of a
occurred shortly after 90 Ma. carbonate platform (too early to be the Bahama
Incorporating the results of this report with other platform?) during the latest Cretaceous. The striking
current information, observations are forthcoming admixture of shallow- and deep-water constituents
better characterizing the origin of the Jamaican and may suggest tectonic disruption.
Dominican units as regards the earliest history of the
Great Arc.
(1) Most if not all of the sedimentary rocks ACKNOWLEDGEMENTS
described in this report formed in deep basins
some probably below the CCD. This work was supported by National Science
(2) Lithologies are significant. The importance Foundation grants NSF-EAR 9418194 and NSF-
of multicolored chert, jasper, red mudstone, and EAR-9117397 to E.A. Pessagno and H. Mont-
limestone of the Upper Cretaceous metamorphic gomery. Some initial collection and site evaluation
and sedimentary complexes of the leading (and by H. Montgomery was supported by funding from
colliding) edge of the Caribbean plate should be the University of Puerto Rico through the NSF EP-
clear. Nowhere have we or has anyone else reported SCoR program. J. Lewis and G. Draper are due
finding post-Jurassic red ribbon chert in the Antilles much credit for their guidance during many days
nor has red ribbon chert been dredged or drilled of field work and during many nights of highly
in the Atlantic basin. This is notable because red constructive contemplation. Falconbridge Domini-
ribbon chert at E1 Aguacate, Dominican Republic cana provided vehicles, lodging, and advice. Special
(Montgomery et al., 1994a), Sierra Bermeja, Puerto thanks go to Salvador Brouwer for his help. Paul
Rico (Montgomery et al., 1994b), and La Ddsirade, Mann offered a thorough and constructive review
Guadeloupe (Montgomery et al., 1992) is remnant, which greatly improved the manuscript.
sub-CCD, oceanic floor originating in the Pacific
basin. As far as concerns the Caribbean, red ribbon
chert is exotic and is a defining characteristic of REFERENCES
Pacific rather than Atlantic origin. Red ribbon chert
accumulated only during the west-facing phase of Boisseau, M., 1987. Le flanc nord-est de la Cordillbre Cen-
subduction. trale dominicaine (Hispaniola, Grandes Antilles); un 6difice de

246 H. M O N T G O M E R Y and E.A. P E S S A G N O
nappes Cr6tac6 polyphase. Doctoral Thesis 3rd Cycle, Univer- in Hispaniola. Geol. Soc. Am. Spec. Pap., 262: 1-28.
sit6 Marie et Pierre Curie, Paris, 215 pp. Montgomery, H., 1997. Paleogene stratigraphy and sedimentol-
Bourgois, J., Vila, J.-M., Llinas, R. and Tavares, I., 1982. Datos ogy of the North Coast, Puerto Rico. In: E. Lidiak and D.K.
geologicos nuevos acera de la region de Puerto Plata (Repub- Larue (Editors), Tectonics of the Northeastern Caribbean.
lica Dominicana). In: Transactions, 9th Caribbean Geological Spec. Publ. Geol. Soc. Am. 322: 177-192.
Conference, Santo Domingo, pp. 39-50. Montgomery, H., Pessagno, E.A., Jr. and Mufioz, I.M., 1992.
Draper, G. and Nagle, E, 1991. Geology, structure, and tec- Jurassic (Tithonian) Radiolaria from La D6sirade (Lesser An-
tonic development of the Rfo San Juan Complex, northern tilles): preliminary paleontological and tectonic implications.
Dominican Republic. In: P. Mann, G. Draper and J.E Lewis Tectonics, 11: 1426-1432.
(Editors), Geologic and Tectonic Development of the North Montgomery, H., Pessagno, E.A., Jr. and Lewis, J.A.,
America-Caribbean Plate Boundary in Hispaniola. Geol. Soc. Schellekens, J.H., 1994a. Paleogeography of the Jurassic
Am. Spec. Pap., 262: 77-95. fragments in the Caribbean. Tectonics, 13: 725-732.
Edgar, N.T., Kaneps, A.G. and Herring, J.R., 1973. Initial Re- Montgomery, H., Pessagno, E.A., Jr. and Pindell, J.L., 1994b.
ports of the Deep Sea Drilling Project, Vol. 15. U.S. Govern- A 195 Ma terrane in a 165 Ma ocean: Pacific origin of the
ment Printing Office, Washington, DC, 1137 pp. Caribbean Plate. GSA Today, 4: 1-6.
Foreman, H.P., 1973. Radiolaria of Leg 10 with systematics Morin, K.M., 1982. Analysis and Definition of Campanian (Late
and ranges for the families Amphypyndacidae, Artostrbiidae, Cretaceous) Radiolarian Populations Characteristic of Tropical
and Theoperidae. In: J.L. Worzel, W. Bryant et al. (Editors), to Subtropical Latitudes. Ph.D. Dissertation, Univ. of Texas at
Initial Reports of the Deep Sea Drilling Project, Vol. 10. U.S. Dallas, 375 pp.
Government Printing Office, Washington, DC, pp. 407-474. Nagle, E, 1966. Geology of the Puerto Plata Area, Dominican
Krijnen, J. and Chin, A., 1978. The geology of the northern Republic. Ph.D. Thesis, Princeton University, Princeton, NJ,
central and southeastern Blue Mountains, Jamaica with a pro- 171 pp.
visional compilation of the entire inlier. Geol. Mijnbouw, 57: Pindell, J.L. and Draper, G., 1991. Stratigraphy and geological
243-250. history of the Puerto Plata area, northern Dominican Republic.
Mann, P., Draper, G. and Lewis, J.E, 1991. An overview of the In: P. Mann, G. Draper and J.E Lewis (Editors), Geologic and
geologic and tectonic development of Hispaniola. In: P. Mann, Tectonic Development of the North America-Caribbean Plate
G. Draper and J.E Lewis (Editors), Geologic and Tectonic Boundary in Hispaniola. Geol. Soc. Am. Spec. Pap., 262:
Development of the North America-Caribbean Plate Boundary 97-114.

Chapter 11
Cenozoic E1 Mamey Group of Northern Hispaniola:

A Sedimentary Record of Subduction, Collisional and
Strike-Slip Events within the
North America-Caribbean Plate Boundary Zone
RUURDJAN DE ZOETEN and PAUL MANN
Cretaceous and Cenozoic paleogeographic and plate tectonic reconstructions of the Greater Antilles (Hispaniola, Cuba,
Jamaica, and Puerto Rico) are complicated by large-offset, Eocene? to Recent strike-slip movements between the North America
and Caribbean plates. Moreover, Eocene?-Recent oblique subduction of the Bahama carbonate platform presently affects the
Hispaniola region and further complicates the reconstruction of this wide and complex plate boundary zone. This paper describes
a detailed sedimentological study of deformed and uplifted Eocene to Lower Pliocene sedimentary rocks (El Mamey Group)
between the North America and Caribbean plates in northern Hispaniola (Dominican Republic). Paleocene to Lower Pliocene
siliciclastic and carbonate rocks of the E1 Mamey Group, which crop out within a 500 km 2 area of the central Cordillera
Septentrional and are well exposed in road and stream cuts, formed the object of this regional tectonic study.
On the basis of its compositional, age and facies character, we divide the sedimentary succession of the E1 Mamey Group
of the central Cordillera Septentrional into three lithologically distinct, stratigraphic sequences which we relate to three tectonic
phases that affected this segment of the North America-Caribbean plate boundary from the Eocene to Recent.
Phase 1 (Paleocene to Middle Eocene). Sedimentary and facies characteristics of an approximately 250-m-thick section of
Upper Paleocene to Lower Eocene siliciclastic and carbonate rocks (Los Hidalgos ~ormation), suggest that these rocks were
deposited in a deep-marine, hemipelagic environment adjacent to an active volcanic arc. Calc-alkaline volcanic flows and sills are
interbedded with these deep-marine sedimentary rocks. Termination of deposition and volcanism in Early to Middle Eocene time
coincides with a major folding and uplift event, which we believe was caused by the early attempted subduction of the Bahama
carbonate platform beneath the arc-related basin. This event terminated arc activity in the Hispaniola volcanic arc and forearc.
Phase 2 (Late Eocene to Early Miocene). Sedimentary and facies characteristics of a 4000-m-thick, Upper Eocene to Lower
Miocene siliciclastic succession (Altamira and Las Lavas formations of the E1 Mamey Group) suggest that these rocks were
deposited as submarine turbidites and other types of mass-flow deposits within a west-northwest-trending, elongate basin.
Petrographic analysis of framework grains of sandstones within the section shows two distinct sandstone populations separated by
a linear, 100-400-m-wide left-lateral strike-slip fault zone. Petrographic differences across the fault zone are especially prominent
in coeval Oligocene sedimentary rocks and suggest that the two basins were juxtaposed by lateral fault movement sometime after
Oligocene time. The end of deep-marine siliciclastic deposition in both basins coincides with a gentle Middle Miocene folding
event believed to be related to transpressional strike-slip faulting.
Phase 3 (Late Miocene to Recent). Sedimentary and facies characteristics of an approximately 250-m-thick section of Upper
Miocene to Lower Pliocene carbonate rocks (Villa Trina Formation) suggest that these rocks were deposited as a shallow
carbonate bank above slightly folded, Early Miocene siliciclastic rocks. Carbonate deposition was terminated in Early Pliocene
time by a folding and uplift event believed to be related to transpression along a restraining bend in the Septentrional fault zone.
INTRODUCTION metamorphic, and sedimentary basement blocks of

Late Cretaceous to Middle Eocene age that formed
Objectives in an intra-oceanic island-arc setting (cf. Bowin,
1975; Sykes et al., 1982; Mann et al., 1991, for
The island of Hispaniola consists of a 250-km- regional reviews; Fig. 1). The basement blocks are
wide tectonic collage of fault-bounded igneous, overlain by a cover of Upper Eocene to Pliocene
Caribbean Basins. Sedimentary Basins of the World, 4 edited by E Mann (Series Editor: K.J. Hsfi), pp. 247-286.

248 R. DE ZOETEN and E MANN
Fig. 1. Present-day plate structure of the Caribbean region. Direction and rates of plate motion relative to the Caribbean plate are from
DeMets et al. (1990) and Dixon et al. (1998). The island of Hispaniola straddles the active left-lateral strike-slip zone separating the North
America and Caribbean plates. The large amount of plate convergence and topographic uplift of Hispaniola is related to transpression
between two thick crustal blocks: the Bahama carbonate platform to the north and the Cretaceous Caribbean oceanic plateau (hatched
pattern) on the Caribbean plate to the south. Box shows map area shown in Fig. 2.
siliciclastic and carbonate sedimentary rocks that Based on the same outcrops discussed in this
post-date island-arc activity and mainly record paper, de Zoeten and Mann (1991) published a
the initiation of the present period of left-lateral Cenozoic structural history of the area using data
strike-slip motion between the North America and on major and minor faults and folds observed in
Caribbean plates (Fig. 2). Many previous tectoni- the same set of outcrops as described in this paper.
cally oriented studies in northern Hispaniola have Dolan et al. (1991) presented a summary of some of
focused on the composition and structure of arc- the preliminary results of this study in their discus-
related basement rocks in order to better understand sion of the pattern of regional basin formation during
the origin and tectonic history of the Greater An- the Cenozoic in Hispaniola and Puerto Rico. Neither
tilles island-arc system (for example, Nagle, 1979; of the previous studies presented the primary sedi-
Palmer, 1979; Pindell and Draper, 1991; Joyce, mentological data collected by de Zoeten (1988) that
1991; Fig. 2). are fundamental to many tectonic interpretations.
Following the earlier sedimentation and tecton-
ics study of Dolan et al. (1991), the objective of Significance of this study
this paper is to reconstruct the depositional his-
tory of Paleocene to Lower Pliocene sedimentary The Cordillera Septentrional is a critical area for
rocks of the central Cordillera Septentrional that studies of post-Middle Eocene island-arc tectonics in
either post-date or accompany the final stages of is- Hispaniola because the area straddles the Septentri-
land-arc activity in Hispaniola (Fig. 1). The method onal fault zone, which is presently the main strike-
of this paper is to reconstruct the depositional his- slip fault zone separating the North America and
tory and water depth of sedimentary basins overlying Caribbean plates (Mann et al., 1991, 1998) (Fig. 2).
arc rocks through the use of: (1) regionally corre- Marine studies have shown that the Septentrional fault
lated measured sections of continuous stratigraphic zone continues westward as fault zones bounding the
sequences; (2) integration of biostratigraphic data Middle Eocene to present Cayman Trough pull-apart
as a way to reconstruct sedimentary environments, basin (Rosencrantz et al., 1988; Calais and Mercier
determine water depths through time, and date sedi- de L6pinay, 1995) and eastward as active fault zones
mentary, tectonic, and eustatic events; (3) integration along the southern edge of the Puerto Rico Trench
of paleocurrent and sandstone petrographic data to (Grindlay et al., 1997; Dolan et al., 1998; Fig. 1).
determine siliciclastic provenance. The Cordillera Septentrional provides the largest sub-

CENOZOIC EL MAMEY GROUP OF NORTHERN HISPANIOLA 249
aerial exposure of this 3600-km-long, interplate fault GEOLOGIC SETTING OF THE CORDILLERA
system outside of Central America (Fig. 1). SEPTENTRIONAL
Ages and distribution of rock units

TECTONIC SETTING OF THE CORDILLERA
SEPTENTRIONAL Basement units
Igneous, metamorphic, and sedimentary rocks
Plate-scale tectonic setting ranging in age from Cretaceous to Early Pliocene
are exposed in the Cordillera Septentrional whereas
The Cordillera Septentrional ('Northern Range') siliciclastic sedimentary rocks of Mio-Pliocene age
forms an elongate, east-northeast-trending mountain are exposed in the adjacent, asymmetric Cibao basin
range that rises to a maximum elevation of 1249 m (de Zoeten and Mann, 1991) (Figs. 2B and 3). The
(Fig. 2). The range is partially bounded by seismi- oldest, arc-related rocks of the Cordillera Septen-
cally active, strike-slip and reverse faults related to trional occur in three basement complexes of the
left-lateral displacement between the North Amer- Cordillera Septentrional that include the Samana
ica and Caribbean plates across Hispaniola (Mann Peninsula (Joyce, 1991) and Rio San Juan complex
et al., 1991, 1998; Fig. 2B). Transpression across to the east of the study area, and the Puerto Plata
northern Hispaniola is probably a response to highly complex (Pindell and Draper, 1991) to the north of
oblique subduction of the Bahama carbonate plat- the study area (Fig. 3). Mann et al. (1991) and Mann
form (Mullins et al., 1992; Mann et al., 1995; Dolan and Gordon (1996) proposed that all three inliers
et al., 1998) (Fig. 2A). The relation between the may have been uplifted by late Neogene restraining
general shape of the unsubducted Bahama Platform bend tectonics along the Septentrional fault zone
and the thrust front north of Hispaniola suggests that (Fig. 3).
as much as 100 km of Bahamanian crust may been Basement rocks of the Cordillera Septentrional
subducted beneath Hispaniola (Dolan et al., 1998). and the Samana Peninsula can be divided into two
Local convergence in Hispaniola may also be related lithologic provinces: a blueschist-serpentinite-sedi-
to the location of the island between the Bahama mentary complex interpreted as an outer forearc-
carbonate platform to the north and thicker-than- trench assemblage by previous workers (Nagle,
average oceanic plateau seafloor of the Caribbean 1979; Bowin and Nagle, 1982; Pindell and Draper,
Sea to the south (Mann et al., 1995; Diebold and 1991; Draper and Nagle, 1991; Joyce, 1991); and a
Driscoll, Chapter 19; Figs. 1 and 2). volcanic-plutonic-sedimentary complex interpreted
to be a forearc assemblage (Bowin and Nagle,
Island-scale tectonic setting 1982; de Zoeten and Mann, 1991; Calais et al.,
1992). Based on a compilation of data, Mann et al.
A regional, unbalanced cross-section modified (1991) have classified these two areas as tectonos-
from Mann et al. (1991) illustrates several important tratigraphic terranes originating as fragments of the
features about the Cenozoic structural history of forearc or accretionary prism of the arc (Fig. 2B).
Hispaniola (Fig. 2B). We follow the block terminology of the cen-
(1) A prominent folding and thrusting event in tral Cordillera Septentrional that was defined by de
central Hispaniola is Late Miocene and younger in Zoeten and Mann (1991). This scheme subdivides
age and verges southward to southwestward. the basement of our study area into the Altamira
(2) Late Miocene and younger reverse and block to the west and the La Toca block to the
oblique-slip faulting is responsible for the present east separated by the left-lateral Rio Grande fault
pattern of morphotectonic units in central Hispan- zone (Fig. 4C). Basement of the La Toca block in
iola, including the distribution of the three major the eastern part of the study area consists of Upper
ramp, or thrust-bound, basins the Cibao, San Cretaceous to Eocene andesitic tuff and tonalite of
Juan-Azua, and Enriquillo (Mann et al., 1991; Mann the Pedro Garcfa Formation (Eberle et al., 1982;
et al., Chapter 12). Peralta-Villar, 1985) (Fig. 4C). Basement of the
(3) Cretaceous-Eocene island-arc terranes of the Altamira block in the western part of study area con-
northern and central part of the island are topograph- sists of Upper Paleocene to Lower Eocene pelagic
ically high-standing and deeply eroded; the Creta- carbonate rocks of the Los Hidalgos Formation that
ceous oceanic plateau terrane of the southern part of are crosscut by dikes and sills of the Palma Picada
the island is relatively low-standing and less deeply intrusives (Muff and Hernandez, 1986) (Fig. 4C).
eroded. The lower elevation of the oceanic plateau in The Los Hidalgos Formation correlates well with
the south may also reflect its footwall position relative the Paleocene-Eocene E1 Cacheal tufts described
to the higher-standing hanging wall block represented by Calais et al. (1992) in the western Cordillera
by the island-arc terranes in the north (Fig. 2B). Septentrional.


Fig. 3. Map of northern Hispaniola showing ages of exposed rocks and the four main physiographic provinces which include the
Cordillera Septentrional, Samana Peninsula, Cibao Valley, and Cordillera Septentrional. The area mapped in detail for this study is
boxed and lies in the central part of the Cordillera Septentrional (cf. Fig. 4 for detailed maps). Major high-angle faults include: HFZ
-- Hispaniola fault zone; SFZ = Septentrional fault zone; CFZ = Camti fault zone. Labelled basement complexes of the Cordillera
Septentrional are discussed in the text.
Relation of major structures to outcrop d o m e along the Septentrional fault zone accounts
pattern of sedimentary units for the northeast dips o b s e r v e d over m u c h of the
Younger, s e d i m e n t a r y units above these three A l t a m i r a b l o c k (Fig. 5B). F o l d i n g related to the Pe-
b a s e m e n t c o m p l e x e s g e n e r a l l y dip away from the in- dro Garcfa anticline accounts for the northeast dips
liers or are b o u n d e d by high-angle faults (Figs. 3 and observed over m u c h of the La Toca block. The A1-
4C are based on a 1 : 1 5 0 , 0 0 0 geologic c o m p i l a t i o n tamira fault zone abruptly truncates fold axes devel-
map of the Cordillera Septentrional by de Z o e t e n oped in the central part of the study area (Fig. 5A).
et al., 1991). The ages of E o c e n e through Pliocene Well-dated, U p p e r E o c e n e to L o w e r M i o c e n e
sedimentary units in the western Cibao basin and d e e p - m a r i n e siliciclastic s e d i m e n t a r y rocks of the
central and eastern Cordillera Septentrional defines a A l t a m i r a and Las Lavas formations u n c o n f o r m a b l y
large synclinal structure with the axis of the syncline overlie arc-related b a s e m e n t of the A l t a m i r a block
a p p r o x i m a t e l y parallel to the long axis of the Cibao (Fig. 4B). T h e s e two formations together consist of
basin (Fig. 3). The orientation and age of this post- about 4000 m of thin- to m e d i u m - b e d d e d sandstone
Pliocene fold is consistent with the regional pattern i n t e r b e d d e d with c o n g l o m e r a t e . About 1200 m of
of n o r t h e a s t - s o u t h w e s t shortening across Hispaniola O l i g o c e n e to L o w e r M i o c e n e siliciclastic sedimen-
seen on the cross-section in Fig. 2B. tary rocks of the La Toca F o r m a t i o n u n c o n f o r m a b l y
In our study area, ages of rock units indicate two overlie, or are locally faulted against, igneous rocks
large h a l f - d o m e or anticlinal structures adjacent to of the Pedro Garcfa F o r m a t i o n (Fig. 5).
the Septentrional and Rio Bajabonico fault zones M i d d l e M i o c e n e to L o w e r Pliocene shallow-
(Figs. 4C and 5). Tilting related to the Paradero half- marine l i m e s t o n e of the Villa Trina F o r m a t i o n forms
Fig. 2. (A) Map of the northeastern Caribbean plate margin modified from Dolan et al. (1998). Crystalline basement rocks represent the
exhumed core of the Caribbean island-arc that became inactive in Eocene to Oligocene time in this area. Rocks of the Bahama carbonate
platform on the North America plate north of Hispaniola are part of a passive margin sequence formed following the Mesozoic rifting
of North and South America. Rocks of the Caribbean plate and oceanic plateau were mainly formed as part of a large oceanic plateau
in Late Cretaceous time. Convergence in Hispaniola is related to the local impedance of the eastward migration of the Caribbean plate
by the salient formed by the Bahama Platform. Box shows map of northern Dominican Republic in Fig. 3. Key to abbreviations: OFZ
= Oriente fault zone; SDB = Santiago deformed belt; EPGFZ = Enriquillo-Plantain Garden fault zone; SFZ = Septentrional fault zone;
NHDB = North Hispaniola deformed belt; PRT = Puerto Rico trench; LMOB = Los Muertos deformed belt. (B) Cross-section across
Hispaniola modified from Mann et al. (1991). Convergence between the Bahama Platform and the Caribbean oceanic plateau across the
island of Hispaniola has led to extreme topographic uplift and erosion of the extinct arc core and the formation of three, thrust-bound
ramp basins of Neogene age: the Enriquillo, San Juan and Cibao. Abbreviations of fault zones from north to south: CFZ = Cam6 fault
zone; R G F Z = Rfo Grande fault zone; SFZ = Septentrional fault zone; HFZ - Hispaniola fault zone; GFZ = Guacara fault zone; SJRFZ
= San Josd-Restauraci6n fault zone; SJLPFZ San Juan-Los Pozos fault zone; EPGFZ Enriquillo-Plantain Garden fault zone; and
- - =
BAGFZ = Bahoruco fault zone. See Mann et al. (1991) for detailed descriptions of faults and terranes.


the youngest sedimentary unit in the Cordillera Eberle et al. (1982) compiled a lithologic and
Septentrional and is at least 250 m thick (de Zoeten structural map of the central and eastern Cordillera
and Mann, 1991; Calais et al., 1992) (Fig. 6). The Septentrional, based on reconnaissance mapping and
limestone exhibits both conformable and uncon- detailed biostratigraphic sampling of calcareous nan-
formable contacts with the underlying siliciclastic nofossils. One formation, the E1 Mamey, was estab-
rocks of the La Toca and Altamira formations, respec- lished to include all siliciclastic rocks of Eocene to
tively. Dips in the Villa Trina Formation define a large, Early Miocene age. The E1 Mamey Formation was
post-Early Pliocene anticline that coincides with the divided into two, laterally equivalent 'facies': the
topographically highest part of the Cordillera Septen- Luperon facies north of the Cam6 fault zone, and the
trional (Fig. 6A). The angular contact between the Altamira facies to the south (Fig. 7). Eberle et al.
Villa Trina Formation and the underlying siliciclastic (1982) were also the first to recognize the intrusive
formations is generally found in the western and cen- rocks in the Palma Picada area south of E1 Mamey
tral Cordillera Septentrional whereas conformable or (Figs. 4C and 7).
disconformable contacts are found in the eastern part The study by Redmond (1982) focused on the sedi-
of the range (Calais et al., 1992). mentology of Cenozoic sedimentary rocks of the cen-
tral Cordillera Septentrional and their relation to the
occurrence of amber deposits in the area. Redmond
PREVIOUS WORK AND STRATIGRAPHIC refers to these turbiditic rocks as the Altamira Forma-
FRAMEWORK OF THE CENTRAL CORDILLERA tion after the town of Altamira along the Santiago-
SEPTENTRIONAL Altamira highway (Fig. 4A). A Late Eocene age
was determined for the Altamira Formation based
Previous studies on the planktonic and benthonic foraminifera from
ten samples (E. Robinson, in Redmond, 1982). Calais
The stratigraphic nomenclature developed in pre- et al. (1992) carried out stratigraphic and structural
vious studies of the central Cordillera Septentrional mapping in the western Cordillera Septentrional that
is summarized in Fig. 7. Previous work by Eberle et dated a Lower Paleocene-Eocene fine-grained sec-
al. (1982) and Redmond (1982) on the sedimentary tion equivalent to the Los Hidalgos Formation (El
rocks of the Cordillera Septentrional has been on a Cacheal tufts) unconformably overlain by a Lower
regional scale (1 : 100,000) with little emphasis on Miocene to Upper Pliocene siliciclastic section (Gran
detailed mapping of smaller areas. Mangle and Villa Vasquez series).
Vaughan et al. (1921) published the first recon-
naissance map of the Cordillera Septentrional. Their Stratigraphic n o m e n c l a t u r e of this study
proposed stratigraphy was based mainly on work
done along the southern margin of the Cibao Valley We propose a stratigraphic scheme shown in the far
and correlated units exposed in the Cibao Valley. right column of Fig. 7 based on data presented in this
Oil exploration by Dohm (1943) and Beall (1943) paper. We measured over 9000 m of stratigraphic sec-
assembled a more detailed geologic map at a scale tions and integrated these with 199 biostratigraphic
of 1 : 100,000 and used the nomenclature of Vaughan analyses. Of these 199 biostratigraphic analyses, 136
et al. (1921). were done on samples collected specifically for this
Bermudez (1949) analyzed 22 samples from the study (all biostratigraphic data shown on Fig. 4B are
Cordillera Septentrional for planktonic foraminifera compiled in Appendix 1 of de Zoeten, 1988).
to revise formation ages established previously by Combining biostratigraphic data with detailed
Dohm (1943); (Fig. 7). Bermudez (1949) found that lithostratigraphy, we recognized that the Altamira
the oldest sedimentary rocks of the 'Abuillot For- and La Toca blocks were characterized by dis-
mation' (the Los Hidalgos Formation in this paper) tinct stratigraphy and sandstone composition (de
consisted of hard, thin-bedded limestone containing Zoeten and Mann, 1991) (Fig. 4C). These two
Lower Eocene radiolaria and planktonic foraminifera. blocks are separated by the Rfo Grande fault zone, a
Fig. 4. (A) Map of central Cordillera Septentrional showing major towns, road system and streams with outcrops that are referred to in
the text of this paper. Paved roads are shown by heavy lines; unpaved, secondary roads in the mid-1980's are shown as dashed lines.
(B) Map of the central Cordillera Septentrional showing the age of exposed rock units based on microfossils from 140 sample localities
shown by circles. Patterns indicate ages which and are constrained by both biostratigraphy and stratigraphic relationships. Microfossils
used in this compilation include calcareous nannofossils, planktonic foraminifera, and benthonic foraminifera. Appendix 1 in de Zoeten
(1988) provides a list of microfossils identified from each sample site. (C) Map showing lithostratigraphy and formations of the central
Cordillera Septentrional. The study area is limited to the area between the Septentrional fault zone (SFZ) and the Camfi fault zone (CFZ),
and is separated into the Altamira block and the La Toca block by the Rfo Grande fault zone (RGFZ). Key to abbreviations: A F Z =
Altamira fault zone; RBFZ = Rio Bajabonico fault zone; CBA = Canada Bonita anticline; LS = Llanos syncline.

254 R. D E Z O E T E N and R M A N N
Fig. 5. (A) Major structural features of the central Cordillera Septentrional modified from de Zoeten and Mann (1991). Note the en
echelon arrangement of major anticline and half-domes, which are shaded in gray. Key to numbered folds: 1 = Paradero half-dome;
2 = Canada Bonita syncline; 3 = Llanos syncline; 4 = Ocampo anticline; 5 = Pedro Garcfa anticline; 6 = Sonador anticline. Lines
A - A I, B - B I, and C - U indicate cross-sections shown in (B). (B) One-to-one cross-sections of the central Cordillera Septentrional. Key
to rock units: KTpg = Pedro Garcfa Formation (shaded); Tlh = Los Hidalgos Formation (shaded); Tp = Palma Picada intrusive rocks
(shaded); Ta = Altamira Formation; Tt = La Toca Formation; T1 = Las Lavas Formation; Tvt = Villa Trina Formation. Heavy dot pattern
indicates Mio-Pliocene sedimentary rocks of the Cibao basin; wavy lines are unconformities. SFZ -- Septentrional fault zone; R G F Z =
Rfo Grande fault zone; CFZ = Camti fault zone. Dip symbols represent the dip of beds measured in outcrop.

Fig. 6. Map showing the distribution and structure of the Villa Trina Formation in the Cordillera Septentrional. The elevation of the base
of the Villa Trina Formation above sea level was determined from 1:50,000 topographic maps.
Fig. 7. Comparison of stratigraphic columns for the central Cordillera Septentrional including the stratigraphy proposed on the basis of
data presented in this paper. Most previous work was reconnaissance in nature. This study used systematic biostratigraphy and detailed
measured sections to better establish the character and age of lithologic units and correlate them across the central part of the mountain
range.
100-400-m-wide, oblique-slip shear zone (de Zoeten comprises the Altamira and Las Lavas formations
and Mann, 1991). and the La Toca block comprises the La Toca and
The name E1 M a m e y Group was proposed by Luperon formations (Figs. 4C and 7). The Luperon
Dolan et al. (1991) and in this paper to include all the Formation has been described previously by Nagle
Upper Eocene to Lower Miocene sedimentary rocks (1979), Eberle et al. (1982), and Pindell and Draper
in the Cordillera Septentrional. The Altamira block ( 1991) (Fig. 7).

256 R. DE ZOETEN and R MANN
r~
~ c~
r~
c~
m 0
.~ o
~ ~ .
~:~ .,~ . ~
~ ~ ~ .,.,
.,., ,.~
~ -~ 9 ~
_ ~ ..~
~ _=~.~:~ ~, ~
,..., ;>
m.o
9 ~ :~
o
o r,.) ~
o
~ =~ '~ , . - ,.~
o ~ ~ .,.-,
z~ =~ ~ ~=~ E
~
;>,
r~ ,..C:~ ~ 0
,o 9 ~;~ =o-~
c~ ~ ~o
:~~
~ m
o
.~.-~..'~.~.= ~ "~ :~ ~ "~ ~ = .- o

m o
i i i i
b~ b~ o
o ~
~a
,"~ .
"~ .~ o o .~
.~=.~ =o~ ~o ~o'~ . ~ ~
.~'~ ~
~ ~s
~o ~.~ =~.~
~ ~ ~ 9,-. .,,_. .,.. ,.r ~ 0"~ 0 o;~ r~
o o= ; ~ ;~.~ ~
~ . ~ .~
o~,~ o~
o,..
~ M
*" "c:~ , ' ~ .~
~'~
o ~, o~ .~
~.~ ..,
.~ ,~ ,~ ~ ~,~E ~ .
.~
~=
r~l m
~" .~- ..~ .~ .~
'" o "~
r~ 0 ~ 0
.-, o c:= Ei ;~,. ~ .o.~ _ o,.~ 9
m ~
.,..~., ,..-~ r~ o
o =,-~ ~ o
=o ~ ~" . - . ~ 9,-. i= o
r~ o 0 o .,--~
~, .-
o ~"=
r./3 ..c:~
i i i
;>,
9" - ~E ="= = =~'~ =.-= ,~== -

=~
, .-.= o ~- o -.- ==~i==~
~ "" o 9
.~ ~'~ o
= ~.~ ~ .~-.-.-~ ~ ~ ~ o ~ =~ ~ =~ .. 0 ~
o . ~ " ~ ~" O~ 0 "~ ~ ~ o 0
" 0 r~ !
.. 0 ~
_o~.~ ~~ 0
~ ~..= ~ .~ ~ . ~
=. == . . . . ~ . ~ = ~o o ~ ~ . ,.._ ~ ~., ~o ,-~ 0 0
'c~ .~ c~ ~.~ ~ r~ ~ c~ r~ ~ r~ ~ .~...,

'-'- ~r~ 0 ~1 ~ r ""
== ~ . ~
iaa== == ;=" " - -
r.~ ,. 0
, , , . . . . .
r cn
~,T..

Facies classification used in this study rocks south of E1 Mamey (UTM 847816). The Las
Lavas Formation, which conformably overlies the
In our stratigraphic and sedimentologic study of Altamira Formation, consists of calcareous and ter-
the Altamira, Las Lavas and La Toca formations, we rigenous, deep-marine turbidite and other mass-flow
used the facies classification for deep-marine silici- deposits. The topographically higher regions of the
clastic rocks that was developed by Pickering et al. Altamira block are capped by Upper Miocene to
(1986). This classification, a modification of Mutti Lower Pliocene, shallow-marine carbonate rocks of
and Ricci Lucchi's (1978) lithofacies classification, the Villa Trina Formation (Fig. 6).
facilitates facies descriptions and interpretations in
the field. The Pickering et al. (1986) classification is Basement of the Altamira block
a comprehensive and purely descriptive scheme used
to subdivide lithologies into mappable facies. Four Los Hidalgos Formation
of the seven facies classes proposed by Pickering et The Los Hidalgos Formation is composed of
al. (1986) were recognized in the central Cordillera thinly laminated to medium-bedded, dark gray, red
Septentrional. These commonly seen facies are sum- and green recrystallized biomicrite interbedded with
marized on Fig. 8. The letter-number code shown in minor amounts of volcaniclastic calciturbidites and
Fig. 8 modified from Pickering et al. (1986) is used tuffaceous shale (argillite). The base of the Los Hi-
for all measured sections and outcrop photographs dalgos Formation is not exposed, but the formation
presented in this paper. is at least 250 m thick in single outcrops. The forma-
tion is exposed in three localities in the study area:
Biostratigraphic age determinations used in this (1) in an irregular-shaped area 3-12 km south of
study E1 Mamey (UTM 847816); (2) on a ridge 0.4 km
wide, and 5 km long just north of Altamira (UTM
Ages of stratigraphic units are based on their 096752); and (3) as a fault-bounded sliver within the
fossil content (Fig. 4B). Reworked older fauna, Rio Grande fault zone about 3 km north of Santiago
however, are an inherent problem associated with (UTM 299594; Fig. 4C). Based on its outcrop dis-
resedimented turbidite deposits characteristic of the tribution at these three localities, the Los Hidalgos
E1 Mamey Group and so the youngest ages were Formation appears to underlie most of the Altamira
picked to represent the time of deposition. block.
Thin parallel laminae in the recrystallized biomi-
Location grid used in this study crites are laterally continuous, but locally biotur-
bated by Skolithos and Planolites. Laminated lime-
The Universal Transverse Mercator (UTM) grid stones contain matrix-supported, silt-sized, angular
location system is used throughout this paper to plagioclase grains and ellipsoidal silt-sized grains,
precisely locate outcrops discussed in the text. This interpreted to be calcified radiolarians or recrystal-
grid is printed on U.S. Defense Mapping Agency lized planktonic globigerinid foraminifera. Acarin-
1:50,000 topographic maps which were used as ina and Morozovella spp. have been identified in
basemaps for our field investigations (see Preface of thin-sections (E. Robinson, pers. commun., 1988)
Mann et al., 1991, for a key to these maps in the suggesting that the rocks range in age from Late
Dominican Republic; the grid zone is 19Q). Paleocene through Early Eocene and that they were
deposited in bathyal to abyssal water depths (150-
6000 m; cf. Appendix 1 of de Zoeten, 1988). These
STRATIGRAPHY OF THE ALTAMIRABLOCK rocks contain stylolites which are crosscut by com-
plex micro- to mesoscopic calcite veins. The struc-
Definition of the Altamira block ture of the Los Hidalgos Formation is described in
more detail by de Zoeten and Mann (1991) and
The Altamira block is bounded on the north by Calais et al. (1992).
the Cam~ fault zone (Pindell and Draper, 1991), Thin- to medium-bedded calciturbidites (1-30
on the east by the Rio Grande fault zone (de cm) with partial Bouma sequences (Tad, Tae) are
Zoeten and Mann, 1991), and on the south by interbedded with the laminated limestone unit. The
the Septentrional fault zone (Mann et al., 1998; coarser fraction consists of poorly graded, fine- to
Fig. 5A). The basement rocks of the Altamira block medium-grained plagioclase crystals, volcanic rock
comprise Upper Paleocene to Lower Eocene, sed- fragments and bioclasts. An abrupt grain-size change
imentary and igneous rocks of the Los Hidalgos occurs between the basal sand unit and the overlying
Formation. The deep-marine siliciclastic rocks of the recrystallized limestone, which composes the pelagic
Altamira Formation sit unconformably over the Los division (Te) of the sand bed. The limestone unit is
Hidalgos Formation and its intrusive Palma Picada commonly bioturbated.

Thin parallel laminae and abundant planktonic Sandstone and siltstone of the Canada Bonita
fossils in the Los Hidalgos Formation indicate slow Member are blue-gray, calcite-cemented, feldspathic
deposition from suspension in a deep-marine envi- litharenite, which weathers to an orange-tan color.
ronment. Pelagic to hemipelagic accumulation in the The Bouma facies Tde of the sandstone and siltstone
basin was periodically interrupted by low-density couplet of the Altamira Formation typically lack
turbidity flows or other mass-flow processes intro- mud-sized particles. These upper divisions consist
ducing the coarser clastic material of the Ta turbidite predominantly of coarse- to fine-grained silt with
intervals. only minor clay.
Clasts in conglomerates of the Canada Bonita
Palma Picada intrusive rocks Member range in size from granules to boulders,
Porphyritic rocks of the Palma Picada Formation but most commonly range in size from pebbles to
intrude the Los Hidalgos Formation sedimentary cobbles. The clasts are equidimensional or oblate in
rocks near Palma Picada (UTM 922778) (Fig. 4C). shape, and subrounded to well rounded. In general,
These compositionally diverse, porphyritic rocks clasts are composed of: (1) recrystallized limestone
consist of a series of vertical dikes and horizontal (~60%), including biomicrite, dark-gray, green, and
sills, which are approximately 250 m thick (Eberle banded argillite, and carbonaceous silt derived from
et al., 1982; Muff and Hernandez, 1986). the underlying Los Hidalgos Formation; (2) plu-
tonic porphyries (~20%); (3) bioclastic limestone
Stratigraphy of the Altamira Formation (~10%); and (4) sandstone and volcanic fragments
(,-10%). The conglomerate matrix is a gray to light
Outcrop distribution and general stratigraphy brown, fossiliferous volcaniclastic sand.
The Altamira Formation extends over a 200 km 2
area from the Rfo Grande fault zone to a poorly Age and paleobathymetry
defined western limit near E1 Mamey (Fig. 4C). The Biostratigraphic analysis on 37 samples indicates
Altamira Formation consists of thin- to medium-bed- that the Altamira Formation ranges in age from Mid-
ded sandstone and siltstone couplets, with minor dle or Late Eocene to Late Oligocene (Fig. 4B). Ten
interbedded conglomerate and thick-bedded sand- samples from the lower part of the Altamira Forma-
stone. The Altamira Formation is divided into two tion exposed near E1 Mamey have been well dated
members: (1) a 50-m-thick basal conglomerate, the as Middle to Late Eocene using calcareous nanno-
Ranchete Member, which makes up a minor part fossils and foraminifera (Appendix 1 in de Zoeten,
of the total thickness of the Altamira Formation, 1988). These dates constrain the upper age limit of
and lies unconformably above rocks of the Los Hi- the underlying Ranchete Member as Late Eocene.
dalgos Formation, and the Palma Picada intrusions The lower age limit of the Ranchete Member is con-
(UTM 906815; Fig. 4C); and (2) the Canada Bonita strained by the Early Eocene age of the underlying
Member, an approximately 2500-m-thick section of Los Hidalgos Formation.
alternating sandstone and siltstone with interbedded Several samples from the Canada Bonita Member
conglomerate; this member composes most of the collected along the Santiago-Puerto Plata highway
thickness of the Altamira Formation (Fig. 4C). suggest Late Eocene ages (Appendix 1 in de Zoeten,
The Ranchete Member of the Altamira Formation 1988) (Fig. 4C). However, Bourgois et al. (1982,
is named here after the village of Ranchete, which 1983) and S. Monechi (pers. commun., 1988) deter-
is located 3.5 km southwest of E1 Mamey (UTM mined that three samples from beds stratigraphically
838824; Fig. 4A). In this area, the Ranchete Member below the Upper Eocene sample localities contain
is exposed as a thin (<130-m-wide), 3.5-km-long distinct Oligocene faunas (Appendix 1 in de Zoeten,
belt rimming the northern margin of the Los Hidal- 1988). No northwest-striking faults separating the
gos Formation (Fig. 4C). sample localities along the Santiago-Puerto Plata
Sedimentary rocks of the Altamira Formation highway were recognized (de Zoeten and Mann,
overlie the basal conglomerate of the Ranchete 1991) (Fig. 4C). This suggests that Upper Eocene
Member in an arcuate belt stretching 40 km across calcareous nannofossils are reworked and that depo-
the center of the study area (Fig. 4C). This unit is sition of the Altamira Formation continued into Late
here named the Canada Bonita Member of the A1- Oligocene time as we show on Fig. 7. Paleoenvi-
tamira Formation, after the village of Canada Bonita, ronmental studies on the benthonic foraminiferal as-
7 km north of Navarette (UTM 064703; Fig. 4A). semblage indicate upper bathyal water depths (150-
The Canada Bonita Member is composed of 80% 500 m).
very thin- to medium-bedded, sandstone and silt-
stone couplets (facies types: C2.2, C2.3, D2.2), 15% Measured sections of the Altamira Formation
conglomerate (AI.1, A2.1), and 5% thick-bedded Seven sections were measured from west to east
sandstone (B2.1, C2.1). to better characterize rapid lateral facies changes

CENOZOIC EL M A M E Y GROUP OF NORTHERN HISPANIOLA 259
within the Altamira Formation. The measured sec- Plata highway (Fig. 11, inset map). The rocks
tions are divided into three sets based on geographic are gently folded into an anticline-syncline pair
location and similar facies patterns. From west to whose structure is described in detail by de Zoeten
east, the three sets are: (1) E1 Mamey and Guanan- and Mann (1991). These measured sections all de-
ico, (2) Southern Canada Bonita, Northern Canada scribe variations within a 1000-1500-m-thick sec-
Bonita, and Calabaza, and (3) Rio Perez and Llanos tion of lateral equivalent rocks of the upper Canada
syncline (Fig. 4A). Bonita Member of the Altamira Formation (Fig. 7).
The basal conglomerate of the Ranchete Member The Calabaza measured section is separated from
was only measured in the westernmost E1Mamey sec- the Northern Canada Bonita and Southern Canada
tion (Fig. 9, inset map; Fig. 10). The Guananico sec- Bonita sections by the north-south striking Altamira
tion lies 10 km to the west of Altamira and lies along fault zone (Eberle et al., 1982; Figs. 4C and 5).
strike of the E1Mamey section (Fig. 9, inset map). The Despite the structural complexity related to fold-
Guananico and E1 Mamey sections are dominated by ing and the presence of the Altamira fault zone, there
a repetitive series of thin- to medium-bedded sand- are many similarities between all five sections near
stones and siltstones (C2.2, C2.3, D2.2 facies). the Santiago-Puerto Plata highway that allow the
Five sections of the Altamira Formation were various sections to be correlated. For example, all
measured along or adjacent to the Santiago-Puerto three section are overlain by the basal conglomerate
A. El M a m e y section" Altamira Fm. B. Guananico section: Altamira Fm.

90C
C2.1 TOP
m
r,, !
400 2 - 2.3
\ 400
)2.1
\
r
3 \ 800
\ 80C
\ \
\ ~t
l
\
300 \ \ 300 \
' ~2 2-'23/D22 \ \
\ \ \
\ 700
\ 700 \
\ \ \ )G
200 \ \ \ ' - 2.3 / D2.2
\ 200 \
=E, ,2..2-2.3 \ \ \
(~ F - G \
r~ , A1.1 600 - 2.3 / 1~2.2 600
t TOP B2.1 / C2.1 \
~ OffsettoB \\ u. Eoc
\m \
100 ~A~ 1u Eoc \\\ ~ooo 100
\
\ \ )G
C2..2 2387 C 2 . 2 - 2 . 3 ! D2.2
\
~ A22 500t
soo \ C2.3
..... ', ~ Breccia (Fig. 10) 1 \ ~-"- F
LOS tt=calg(s IJ~ U P a l l Eoc BASE
ForlTl(tion 0 ~ l i " ~ F " "" I
(Eocene) III ....
Grain Size I I Gravel =~o,- I Gravel
IMed. Sand Med. Sand
Silt Silt
Offset in El Mamey
G groove cast F flute cast Very thin to med. bedded I "measured section 0 5 km I
IE'MameY]--//~ '' I
R ripplemark C a-axis clast orientation Med. to very thick bedded I 41 J l l " / / " f = ~ Guananico]
13;El Covered interval
Mio Miocene u upper " - ~ Stratigraphyinferred I ~,Mamo,sec,~on iJ "~ I
Olig Oligocene m middle Fault II Guananico[
Eoc Eocene I lower A2.1 Facies type
~ se='~ I
2387 Sandstone sample I~Y:;," ~ %~ I
(~ Paleocurrent direction - Map trace of measured section
Fig. 9. Measured sections from the E1 Mamey (A) and Guananico (B) localities (cf. inset map for location of sections). The sections
summarize facies types, paleocurrent indicators, biostratigraphic samples with age determinations, and sandstone sample localities used
for point-count analyses. See Fig. 8 for explanation of letter-number codes of facies types. Note along-strike offset in the E1 Mamey
section at 120 m above the base.

Fig. 10. Normally graded, sedimentary breccias from the Middle(?) to Late Eocene Ranchete Member of the Altamira Formation
(location of photograph shown on section in Fig. 9). Clasts are composed of grey, recrystallized biomicrites of the underlying Upper
Paleocene to Lower Eocene Los Hidalgos Formation. Outcrop is exposed in roadcut just south of Los Hidalgos Pass (UTM 847816).
Jacob's staff is divided into 0.5 m increments. Dip to northeast. Color version at http://www.elsevier.nl/locate/caribas/
of the Las Lavas Formation and all three sections imentary breccia (AI.1), which is approximately
consist predominantly of thin- to medium-bedded 15 m thick. The breccia grades into thick-bedded,
sandstone and siltstone (C2.2, C2.3, D2.2) and as- coarse-tail graded conglomerates (A2.3; Fig. 10) and
sociated conglomerate (AI.1, A2.1). The Calabaza is capped by medium-bedded, parallel-stratified con-
section consists of a higher percentage of conglom- glomerates (A2.1). Clast rounding increases upward
erate (~50%) than the Northern Canada Bonita and in the section from angular in the A2.3 facies to
Southern Canada Bonita sections (12%). subrounded clasts in the A2.1 facies.
The Rfo Perez and Llanos syncline sections are Gravel to cobble size, angular clasts near the base
south and southwest of Altamira (Fig. 11, inset map). of the Ranchete Member are composed entirely of
These sections consist of thick-bedded sandstone- gray and dark gray, recrystallized limestone, which
siltstone couplets (B2.1, C2.1 facies), conglomerates reflect the lithology of the underlying basement rocks
(AI.1, A2.1 facies) and thin- to medium-bedded of the Los Hidalgos Formation (Fig. 10). The per-
sandstones and siltstones (C2.2, C2.3, D2.2 facies). centage of recrystallized limestone clasts decreases
upward in the section at the expense of marly pack-
El Mamey and Guananico measured sections of stone clasts composed mainly of red algae and larger
the Altamira Formation reefal foraminifera. Lepidocyclina foraminifera in the
packstone clasts suggest that they formed in a back-
Ranchete Member of the Altamira Formation reef environment (E. Robinson, Appendix 1 in de
The E1Mamey section is a composite section based Zoeten, 1988). Laterally equivalent conglomerate to
on two traverses (Fig. 9, inset map). The basal 54 m the northwest (UTM 906815) contain up to 50% dark
of the section forms the type section of the Ranchete purple porphyry clasts of the Palma Picada intrusions.
Member, which is exposed along the road near Los This conglomerate also exhibits an upsection increase
Hidalgos Pass (UTM 846817) (Fig. 4A). The over- in shallow-water carbonate clasts.
lying section was measured in the Arroyo Berraco
Blanco stratigraphic section (UTM 927844; Fig. 4A). Canada Bonita Member of the Altamira
The contact between the Ranchete Member and Formation
the underlying Los Hidalgos Formation is uncon- The overlying Canada Bonita Member in the E1
formable, although we found that this contact is Mamey and Guananico sections consists mostly of
locally modified by faulting at two localities (UTM thin- to medium-bedded, alternating sandstones and
846817; 871825). The lower part of the Ranchete siltstones, which are mostly facies D2.2, D 1.2, C2.3,
Member consists of a massive, tightly packed sed- with minor C2.2 (Fig. 12). These facies are interbed-

CENOZOIC EL M A M E Y GROUP OF NORTHERN HISPANIOLA 261
O ,.~
O ""'~
O O
o.~
O .,,~
,,,.o
or,.)
E~
rs~
.,.-~
O,, ~
~Z
O
~ 9
~ o
-~ r,..) ~
~<
0
~Zo
=~
~ ~,~ .~,.~
NN~
o ~
~c~ o
~ -~ ,.~
r~
" ~ ~
~ g.~

262 R. DE ZOETEN and R M A N N
r,.)
9
.,,,.
9
~,,,,~
9
9
.=.
9
.<
r,.) .~
.,,a
r,,) ~:
O
(D
Z
O
<~
r,.) ~
,x::

ded with minor, thick-bedded sandstone and siltstone section to the south of the anticlinal axis (Canada
couplets (C2.1). Conglomeratic facies (AI.1, A2.3) Bonita anticline) (Fig. 11, inset), a conglomeratic
are rare, and mostly found near the base of the sequence 40 m thick displays two depositional
Canada Member. units. The base of the upper conglomerate unit is
Thin-bedded sandstone and siltstone couplets of scoured at least 1 m into the underlying conglomer-
the Canada Bonita Member have a sand/silt ratio ate (Fig. 13A). Both conglomerate units grade up-
greater than 1 : 1, are tabular, laterally continuous for wards from predominantly pebble to boulder, clast-
at least 10 m, exhibit sharp basal contacts, and are supported disorganized conglomerates (AI.1 facies)
overlain by medium- to coarse-grained, structureless (Fig. 14A), to pebble to cobble, poorly stratified con-
or poorly graded sand (Tbd, Tabd). Thin parallel lam- glomerate (A2.1, A2.5 facies) (Fig. 14B). Although
inae are very common throughout the lower division poorly stratified conglomerate appears massive, the
of beds. These laminae are defined by 0.5-4 mm bio- preferred orientation of the a-axis of conglomerate
genic carbonate grains composed mostly of red algae clasts parallel to bedding produces a stratified ap-
and benthonic foraminifera. The silty upper division pearance in the conglomerate (Fig. 14B). Smaller
of C2.2 and C2.3 sandstone and siltstone couplets conglomerate bodies (A 1.1, A2.1) form 2-5-m-thick
contain isolated stringers of medium to coarse sand, concave-down channel structures within alternating
are locally parallel-laminated, and are commonly bio- thin-bedded sandstones and siltstones (630 m from
turbated by Skolithos, Planolites, and Glockeria. the base of the Northern Canada Bonita section, and
The thick-bedded sandstones of the Canada 170 m from the base af the Southern Canada Bonita
Bonita Member have sand/silt ratios greater than section) (Fig. 12A).
1:1 and exhibit erosional basal contacts with local The lower part of the Northern Canada Bonita sec-
load and groove casts. The lower division of beds tion (Fig. 12B) is dominated by thin- to medium-bed-
are massive or coarse-tail graded and capped by ded sandstone and siltstone couplets (C2.2, C2.3,
parallel laminae which are defined by concentrations D2.2). These couplets have a 1:3 sand/silt ratio.
of coarser-grained carbonate material (Tbd, Tabd). Basal sands are medium- to very coarse-grained and
Siltstones are commonly thin (3-10 cm), mud-poor, massive to thinly laminated. Basal sands occasionally
parallel-laminated, and occasionally show bioturba- exhibit normal, inverse, and coarse-tail grading (Tdb).
tion by Skolithos and Glockeria. Beds are tabular and A sharp contact divides the upper and lower divi-
exhibit good lateral continuity at the outcrop scale. sions of couplets. The upper division of most couplets
No vertical cycles were recognized in the Canada are silt-rich, but increasing clay-sized fraction causes
Bonita Member, except for rare, thin (1-3 m) thick- the beds to be transitional into the D2.2 and D1.2
ening-upward cycles within rhythmically bedded, facies. The siltstone contains parallel laminations,
thin, sand-silt couplets (Fig. 12). Two 2-7-m-thick, commonly bioturbated by Skolithos and Planulites,
thinning-upward intervals of C2.1-C2.3 facies are and contains lignite fragments with a-axes ranging
found 450 and 470 m above the base of the Guanan- from 0.5 to 10 cm. The a-axis of lignite fragments
ico section (Fig. 9). Both intervals are composed of are oriented parallel to laminations. In the Td inter-
lithic-rich calcarenites (Tabd). These inferrals include vals, coarse-grained sand stringers and lenses are ori-
the only calciturbidites found in the Altamira For- ented parallel to bedding. Locally, outsized carbonate
mation. Their lateral extent or geometry could not be blocks with longest axes ranging from 25 to 160 cm
determined. are found isolated in the thin- to medium-bedded cou-
Twenty-eight groove casts measured from the bot- plets (410 and 585 m from the base of the Northern
tom of medium- to thick-bedded sandstones indicate Canada Bonita section, 225 m from the base of the
a northwest to southeast (110 ~ mean paleocurrent Southern Canada Bonita section).
trend. In the Guananico section, eight unimodal flute In general, conglomerate displays thinning- and
casts indicated paleoflow to the southeast (Fig. 9). fining-up cycles with increasing internal organiza-
tion of clasts (Fig. 13A). Thinning- and fining-up cy-
Northern and Southern Canada Bonita measured cles are poorly expressed in the thin- to medium-bed-
sections of the Altamira Formation ded sandstone and siltstone couplets (C2.2, C2.3,
D2.2), which range in thickness from 5 to 300 m.
Canada Bonita Member of the Altamira Thirty-three paleocurrent indicators were mea-
Formation sured using sole marks, tipple marks, and orientation
The Northern and Southern Canada Bonita sec- of a-axes of conglomerate clasts. At least fifteen
tions consist mostly of very thin- to medium-bedded measurements of the a-axes of conglomerate clasts
sandstone and siltstone couplets (C2.2, C2.3, D2.2), were averaged to a mean direction, which is shown
interbedded with minor conglomeratic facies (AI.1, on the section in Fig. 12. The orientations of bimodal
A2.1), and thick-bedded sandstones (C2.1) (Fig. paleocurrent measurements were scattered, but gen-
12A,B). At the base of the Southern Canada Bonita erally show either N-S or NNW-SSE bidirectional

Fig. 13. (A) Drawing from photographs of conglomerate exposed in a roadcut along the Santiago-Puerto Plata highway (UTM 062706).
Outcrop consists of two vertically stacked, fining-upwards conglomerate bodies within the Upper Oligocene Canada Bonita Member. This
outcrop occurs 60 m above the base of the Southern Canada Bonita measured section shown in Fig. 12A. Letter-number codes indicate
facies types of Pickering et al. (1986) (see Fig. 8 for explanation). Location of photograph shown in Fig. 14A is indicated by box. (B)
Drawing from photographs of a poorly stratified conglomerate (A2.1) which forms the basal unit of the E1 Limon Member of the Las
Lavas Formation and crops out in a roadcut along the Santiago-Puerto Plata highway (UTM 072690). Note the sharp contact with thinly
bedded sandstones and siltstones (C2.3).
paleoflow. Three unimodal measurements indicate 30-90-cm-thick beds (C2.1) are interbedded with
paleocurrent flow both to the south and north. the thin- to medium-bedded sand and silt couplets.
These thicker beds are graded, and commonly con-
Calabaza measured section of the Altamira tain groove marks (Tabd).
Formation The Calabaza section shows a distinct 1400 m
thinning- and fining-up cycle, in which conglomerate
Canada Bonita Member of the Altamira decreases in thickness and abundance upward (Fig.
Formation 12C). Only one 10-m-thick thinning- and fining-up
The base of the section starts at a 350-m- cycle was recognized.
thick, faulted conglomerate ridge (AI.1, A2.1) Eleven bidirectional paleocurrent indicators sug-
and is capped by a 400-m-thick section of thin- gest northeast-southwest (70 ~ paleoflow (Fig. 12C).
to medium-bedded sandstone and siltstone cou- Well-preserved flute casts measured within 10 m
plets (Fig. 12C). This conglomerate is predomi- of the base of the Las Lavas Formation indicate a
nantly massive, disorganized, clast-supported, and paleoflow towards the northwest (285~
decreases in thickness upward in the section. The
few basal contacts observed are erosive. Rio Perez measured section of the Altamira
Thin- to medium-bedded sandstone and siltstone Formation
(C2.1, C2.2, D2.2) is tabular and laterally contin-
uous for 5 m or more. These beds consist of silt Canada Bonita Member of the Altamira
to coarse-grained sand, and exhibit a graded or Formation
structureless lower division and an upper division The R/o Perez section (Fig. l lA) is laterally
characterized by parallel laminae (Tabd, Tbd). Rare equivalent to the Northern Canada Bonita, Southern

Fig. 14. (A) Disorganized, clast-supported conglomerate (type A I.1) observed in the Upper Oligocene Canada Bonita Member. Location
of photo is shown in drawing of Fig. 13A. Bar scale on card is 15 cm long. (B) Poorly stratified, clast-supported conglomerate of the
Upper Oligocene Canada Bonita Member. Note absence of stratal boundaries and well-developed, parallel orientation of oblate clasts. Bar
scale on card is 15 cm long. Color version at http://www.elsevier.nl/locate/caribas/
Canada Bonita, and Calabaza sections (Fig. 12). It is and have a sandstone to siltstone ratio much greater
composed of 60% alternating, thin- to medium-bed- than 1:1 (average 4 : 1 ) (Fig. l l A ) . Bouma inter-
ded sandstone and siltstone (C2.2, C2.2, D2.2), 20% vals include Tbd and Tabd. The lower Ta intervals
thick-bedded sandstone and siltstone couplets (B2.1, are medium- to coarse-grained, massive, coarse-
C2.1), and 20% conglomeratic facies (AI.1, A2.1). tail graded, or amalgamated. Tb intervals are very
Characteristics of the thin- to medium-bedded fa- common. Higher concentrations of red algae, larger
cies are identical to those described for the Northern reefal foraminifera, and plant fragments define par-
Canada Bonita and Southern Canada Bonita sections. allel laminae and rare cross-bedding. The overlying
Thick-bedded sandstones (B2.1, C2.1) are tabu- siltstone of the Td interval also contains abundant
lar and laterally continuous over distances of 10 m plant and lignite debris. The siltstone is parallel-lam-

inated, bioturbated, and rarely exhibit ripple marks. Facies analysis of the Altamira Formation
Conglomerate of the Rio Perez section con-
sist of massive, clast-supported, disorganized facies Facies of the Ranchete Member
(AI.1), and thick-bedded, clast-supported, stratified The basal Ranchete Member breccia suggests lo-
and poorly graded facies (A2.1, A2.3) (Fig. l lA). cal erosion and limited transport of the underlying
Conglomerate horizons range in thickness from 1 to Los Hidalgos Formation (Fig. 9A). The lithologi-
50 m. Basal contacts are scoured. Lateral geometry cally homogeneous, angular clasts suggest that the
is difficult to discern on outcrop scale, but some breccia was deposited as a rock fall (Fig. 10). It
smaller-scale beds are discontinuous over distances is unclear if the rockfall was subaerial or sub-
of 5-20 m and exhibit convex-downward channel marine. The graded nature of the overlying A2.3
cross-sections (e.g., 80 m from the base of Rio Perez conglomerates indicates a transition from a dis-
section; Fig. 11). organized deposition to a more organized deposi-
A total of 32 groove marks and flute clasts were tion from submarine, high-concentration turbidity
measured from the bottom of C2.1 sandstone beds or other mass-flow currents (Fig. 9A). The upward
in the Rio Perez section (Fig. l lA). Twenty-five increase in clast rounding may reflect increased
bidirectional measurements indicate a northwest- grain-to-grain contacts during transport or increasing
southeast (~120 ~ paleocurrent trend, and seven distance from source area, or a combination of both
unimodal current indicators suggest flow to the factors.
southeast. There are five coarsening- and thicken-
ing-upward sequences (5-20 m thick) capped by Facies of the Canada Bonita Member
conglomerates (A1.1, A2.1). This coarsening- and Several sedimentary features displayed by rocks
thickening-upwards trend appears to reverse to a fin- of the Canada Bonita Member rocks support Red-
ing- and thinning-upwards trend toward the top of mond's (1982) interpretation that these rocks are
the section (400 m from the base), where beds define deep-marine turbidity deposits. These features in-
three, 10-50-m-thick, thinning- and fining-up cycles. clude graded beds, poor to well developed Bouma
sequences (Taba, Tba), bathyal depths based on ben-
Llanos syncline measured section of the Altamira thic foraminifera (Appendix 1 in de Zoeten, 1988),
Formation and lateral and vertical facies relationships. The ab-
sence of organization and the very coarse nature
Canada Bonita Member of the Altamira of A1.1 and A2.1 conglomerate horizons indicates
Formation rapid deposition from high-concentration turbidity
The Llanos syncline section is composed of about currents or debris flows; AI.1 facies suggests rapid
70% thick-bedded sandstone and siltstone couplets sedimentation by frictional freezing processes, and
(C2.1, B2.1), 25% thin- and medium-bedded facies A2.1 facies implies deposition from traction bedload
(C2.2, C2.3, D2.2) and 5% conglomeratic facies processes (Picketing et al., 1986). Smaller, con-
(AI.1, A2.1) (Fig. llB). The basal contact of the vex-down conglomerate beds (1-7 m thick; A I.1,
Canada Bonita Member with the Los Hidalgos For- A2.1) suggest that the coarsest-grained material
mation is obscured by vegetation. Thick-bedded moved as channelized flows. Internal stratal fea-
sandstone is predominantly C2.1 facies type, tab- tures of thick-bedded sandstone (B2.1, C2.1 facies)
ular-shaped, and laterally continuous. However, a suggest deposition from high-concentration turbidity
few sandstone beds form discontinuous, broad con- currents. The B2.1 sandstone facies indicate more
vex-down channels (450 m above the base of the rapid deposition than the C2.1 sandstones (Pickering
section). Although facies assemblages are quite dif- et al., 1986). Deposition of the repetitive, thin- to
ferent between the Llanos syncline and the Rio Perez medium-bedded sandstones and siltstones resulted
sections, individual facies characteristics are similar. from low-concentration turbidity currents. The pre-
In the Llanos syncline section alternating sandstone dominance of parallel-laminated and the structure-
and siltstone beds (B2.1, C2.1, C2.2, C2.3) form less lower division of these beds suggests rapid
packages of 3-20-m-thick, thickening- and coarsen- deposition from high-velocity flows.
ing-upwards cycles. These cycles are rarely capped Three distinct facies assemblages are recognized
by conglomerate-filled (AI.1), concave-down scours in the Canada Bonita Member of the Altamira For-
which range from 1 to 5 m in thickness (Fig. 11B). mation. The first is characterized by conglomerates
B imodal paleocurrent indicators, based on 27 (AI.1, A2.1) interbedded with alternating thin- to
groove casts, suggest an east-west to northwest- medium-bedded sandstone and siltstone (C2.2, C2.3,
southeast (90-120 ~ flow direction (Fig. l lB). Uni- D2.2) found in the Northern Canada Bonita, South-
modal measurements from flute casts and tipple ern Canada Bonita, and Calabaza sections (Fig. 12).
marks indicate that the paleocurrents flowed towards Facies D2.2, C2.3, and C2.2 appear to be orga-
the southeast (135~ nized into a few poorly developed, thinning- and

fining-upward sequences, which are about 100 m members: the E1 Limon and the overlying La Pocil-
thick. Individual conglomeratic units average about guita Member (Fig. 7). The E1 Limon Member forms
5 m in thickness and exhibit greater internal orga- a resistant hogback which strikes 110~ for approxi-
nization and fining-upwards trends. We infer that mately 6 km. The ridge-forming E1 Limon Member
the thin-bedded sandstone and siltstone turbidite is offset by north-south striking faults (Fig. 4C). The
couplets represent channel/overbank or interchannel type section is found in the Arroyo Las Lavas near
deposits (Mutti and Normark, 1987). In this inter- the village of E1 Limon (UTM 094689), 7 km to
pretation, the conglomerate-filled channels serve as the northeast of Navarette (Fig. 4A). The E1 Limon
conduits for sediment transport from the slope. The Member disconformably overlies the Altamira For-
coarse nature of the overbank sediment and channel mation near the village of E1 Limon (UTM 097688)
deposits requires either very competent flows or a (Fig. 4A). The E1 Limon Member is approximately
nearby source. 260 m thick and consists of a basal conglomerate
The Rio Perez and the Llanos syncline sections (A1.1, A2.1 facies comprising 40% of the mem-
(Fig. 11) illustrate the second major facies assem- bers), thin- to medium-bedded couplets (C2.2, C2.3,
blage in the Canada Bonita Member, which consists D facies comprising 30% of the members), capped
of coarsening- and thickening-up packages of tab- by interbedded lithic-rich calcarenites (AI.1, A2.3,
ular-shaped, thick-bedded sand (B2.1, C2.1). This C2.1, C2.2 facies comprising 30% of the members).
facies assemblage is interpreted as representing pro- The La Pocilguita Member conformably overlies
grading depositional lobes in a submarine fan system the E1 Limon Member and is named after the village
(Mutti and Normark, 1987). Channel-fill conglom- La Pocilguita del Limon, 1 km south of E1 Limon
erates capping the coarsening- and thickening-up (UTM 088678) (Fig. 4A). The La Pocilguita Mem-
packages suggest that lobes were prograding (Shan- ber is 1300 m thick with more than half the section
mugan and Moiola, 1988). Fan lobe progradation is consisting of alternating thin- to medium-bedded
further supported by the reversal in vertical cycles siliciclastic sandstone and siltstone (C2.2, C2.3, D
at the top of the Rio Perez section, which suggests facies) and the remainder consisting of lithic-rich
a transition from lobe facies assemblage to more calcarenite (AI.1, A2.3, A2.7, C2.1, C2.2 facies)
proximal channel/overbank deposits (Fig. 11). and minor lithic conglomerate (A1.1).
The final facies assemblage consists chiefly of West-northwest striking shear zones increase in
thin- to medium-bedded, sandstone and siltstone number southward in the Las Lavas Formation to-
couplets (C2.2, C2.3, D2.2) which characterize the ward the Septentrional fault zone (de Zoeten and
E1 Mamey and Guananico sections. Lithologically Mann, 1991). The regional effect on these shear
and sedimentologically these deposits resemble the zones on the stratigraphy of the Las Lavas Forma-
overbank deposits described in the first facies as- tion appears minor, because biostratigraphic dating
semblage, although the absence of conglomerates is consistent with southward younging in the south-
or thick-bedded sandstone beds argues against a di- dipping section (Fig. 4B).
rect correlation. These rocks are more likely basinal
deposits at the distal regions of the fan or distal Age and paleobathymetry of the Las Lavas
interchannel deposits. Formation
Biostratigraphic analysis of four samples indicate
Sedimentary rocks of the Las Lavas Formation that deposition of the E1 Limon Member occurred
during Late Oligocene time (Appendix 1, in de
Outcrop distribution and general stratigraphy Zoeten, 1988). Fourteen samples from the La Pocil-
of the Las Lavas Formation guita Member indicate a Late Oligocene to Early
The Las Lavas Formation crops out over 800 Miocene age (Appendix 1, in de Zoeten, 1988). Very
km 2 in the Cordillera Septentrional in a belt extend- few depth-distinctive benthic foraminifera were rec-
ing from Monte Cristi in the west to northeast of ognized. One sample suggests deposition at bathyal
Santiago (Figs. 3, 4C). This study focused on the depths.
best-exposed sections along the south-central flank
of the Cordillera Septentrional. Approximately 450 Measured sections of the Las Lavas Formation
m of section is exposed in isolated outcrops along Three sections were measured from the Las Lavas
the Santiago-Puerto Plata highway, and more than Formation. From west to east, these sections in-
1600 m is exposed in the type section along the clude the Navarette, Las Lavas, and Rio Jacagua
Arroyo Las Lavas (UTM 087645) (Fig. 4A). (Fig. 15). The Navarette section was measured along
The Las Lavas Formation consists of lithic and the Santiago-Puerto Plata highway south of the
carbonate conglomerate, lithic-rich calcarenite and Southern Canada Bonita section (Fig. 11). The Las
thin- to medium-bedded sandstone and siltstone cou- Lavas measured section is the type section for the
plets. The Las Lavas Formation is divided into two Las Lavas Formation (Fig. 15A). The Las Lavas sec-

A:Z o
~9 . O,...,
o~
r.gZ
g~
Z o~ ,..z.
z~0~
r
,.q
o ..~
~ o
,.tzZ ,x:Z
9 ,.,,~

Fig. 16. Measured section of the Lower Oligocene to Lower Miocene La Toca Formation. Inset map shows location of the measured
section near Pedro Garcfa and location of the three along-strike offsets in this composite section (A, B, C). Also shown is the location of
the Rfo Jacagua measured section that appears in Fig. 15C.
tion overlies the Calabaza section measured of the Las Lavas measured section: type locality of the
Altamira Formation in the Arroyo Las Lavas (Fig. Las Lavas Formation
11, inset map). The Rfo Jacagua section is bounded
on the north by the Rio Grande fault zone and on the El Limon Member of the Las Lavas Formation
south by the Septentrional fault zone (Fig. 16, inset The basal conglomerate of the E1 Limon M e m b e r
map). exposed in Arroyo Las Lavas thickens westwardly

from 50 to at least 150 m before being truncated reefal foraminifera, coral and shell fragments with
by the north-south striking Altamira fault zone (Fig. lesser lithic fragments. A few foraminiferal rud-
4C). The contact between the basal conglomerate stones, found as outsized boulders (1-5 m), are
and the underlying Altamira Formation is inferred to composed entirely of large shallow-marine Eulepad-
be an erosional disconformity based on lateral along- ina foraminifera which are diagnostic of Oligocene
strike variations in thickness and distinct lithologic reef, forereef, or shelf environments (Appendix 1 in
differences across the conglomerate. The conglomer- de Zoeten, 1988).
ate is massive to thick-bedded (Fig. 15A). Conglom-
erate beds of the E1 Limon Member are disorganized, La Pocilguita Member of the Las Lavas
clast-supported (AI.1), with some A2.3 beds show- Formation
ing marked scouting and poor grading, and some Very thin- to medium-bedded siliciclastic fa-
A2.1 beds showing tangential cross-stratification. cies (C2.2, C2.3, D2.2) conformably overlie the
Conglomerate clasts in the E1 Limon Mem- E1 Limon Member and make up more than half of
ber range in size from pebbles to cobbles. Clasts the La Pocilguita Member (Fig. 15). Bed geometry
are equidimensional or oblate, subrounded to well and internal organization of these facies are iden-
rounded, and have a poorly defined a-axis orientation. tical to those previously described in the Altamira
Near E1 Limon (UTM 094689), clasts are composed Formation. Thin-bedded facies are interbedded with
predominantly of recrystallized limestone (>60%) individual and stacked lithic-rich clastic limestone
(which include gray, black, green, and banded argillite beds, which increase toward the south. Calcaren-
derived from the Los Hidalgos Formation) and minor ites range in grain size from silt to small cobbles,
amounts of volcanic (15%), plutonic (5%), sandstone and are organized into conglomerate (A2.3, A2.7)
(5%), and packstone (5%) clasts. Some conglomerate and sandstone and siltstone couplets (C2.1, C2.2,
beds contain higher concentrations of igneous clasts B2.1). Thinner-bedded (0.5-2 m) conglomeratic fa-
(e.g., andesite 32%, tonalite 12%). cies (A2.3, A2.5, A2.7) define 1-4-m-thick fining-
The basal conglomerate of the E1 Limon Member upward cycles, which begin at 420 m above the base
is overlain by about 80 m of monotonous thin- of the section. The basal contact of conglomerate
to medium-bedded sandstones and siltstones (C2.2, beds are erosional, but on the scale of the out-
C2.3, D2.2) (Fig. 15A). The beds are tabular and crop appear to be tabular and laterally continuous.
laterally continuous. The basal division of beds are Carbonate-rich, lithic pebble conglomerate grades
massive or poorly graded, and the upper parts are up into structureless coralline gravel beds, paral-
mud-rich siltstones exhibiting parallel laminae and lel-laminated calcarenite, silty calcilutite, and back
more rare bioturbation by Planolites and Skolithos. into thin-bedded siliciclastic rocks. The thicker and
The thin-bedded facies of the E1 Limon Member coarser conglomerate is mostly organized into A I.1
is overlain by about 60 m of medium-bedded to and A2.1 facies and consist of very few carbonate
massive lithic-rich carbonates. Lithic-rich carbonate clasts.
conglomerate is clast-supported and exhibits both Gravel- to silt-sized calcarenite in the La Pocil-
disorganized (AI.1) and organized (A2.3) internal guita Member commonly shows well developed
character (Fig. 15A). Clasts are predominantly lithic Bouma sequences (Tabd, Tabcd). Beds are tabular and
to carbonate clasts. The texture and lithology of laterally continuous. Calcarenite beds have sharp
the lithic component resembles that of the basal stratal boundaries and display few sole marks. Lithic
conglomerate. The lithic fragments decrease upward fragments are typically concentrated in the lower
in the graded beds. Larger carbonate clasts (7- portion of the bed (Ta) and decrease in abundance
60 cm) are irregularly shaped and are composed upward. Basal sands of the Ta interval are coarse-tail
of coral heads, branching corals, and fossiliferous graded or structureless, and grade into parallel lam-
wackestone and mdstone, while the smaller grains inae, tipple marks, and laminated calcilutite. Com-
are mainly red algae, larger reefal foraminifera, and positionally, the calcarenite is identical to that found
coral fragments. in the E1 Limon Member. Calcarenite is composed
Calciturbidites and calcarenite interbedded with mainly of red algae, larger reefal foraminifera, coral
the carbonate conglomerates are medium- to very fragments, and fossiliferous packstone.
thick-bedded, and contain abundant (5-40%) sand- It is difficult to discern any large-scale verti-
to pebble-sized lithic fragments. Medium- to cal organization in the Las Lavas Formation (Fig.
thick-bedded, clastic limestone is tabular, later- 15). However, on a smaller scale, calcarenite beds
ally continuous, and has a sharp basal contact. increase in frequency upsection and are organized
Clastic limestone is commonly massive, exhibits into small (1-5 m thick) thinning- and fining-up
coarse-tail grading and lacks sole marks. Calcitur- cycles with conglomeratic bases. The only matrix-
bidites and calcarenite are predominantly composed supported, disorganized conglomerate (A1.4) recog-
of biogenic material, including red algae, larger nized in the Altamira or Las Lavas formations is

present near the top of the Las Lavas measured sec- lying calcarenite of the E1 Limon Member: (1) thin-
tion (Fig. 15A). Conglomerate clasts are composed to medium-bedded sandstone and siltstone (C2.2,
of cobble-sized coral heads with a minor amount of C2.3, D2.2 comprising 75% of the La Pocilguita
lithic pebbles supported in a calcareous shale matrix. Member); (2) interbedded medium to very thick cal-
Paleocurrent indicators were measured on a few careous sandstone beds (C2.1 comprising 10% of
medium- and thick-bedded calcarenite and silici- the member); (3) 5-15-m-thick calcarenite packages
clastic strata. Measurements on 24 bimodal current comprising 10% of the member; and (4) pebble-
features, including groove casts and a-axis clast cobble lithic conglomerate comprising 5% of the
alignment of conglomerate indicate bidirectional, member. The thin-bedded rocks (C2.2, C2.3, D2.2)
northwest-southeast (~ 130 ~ trending flow. Twenty- have sharp basal contacts and are overlain by paral-
three unimodal paleocurrent structures indicate that lel-laminated to massive, medium- to coarse-grained
currents flowed predominantly to the northwest. sand. The thick sand beds (C2.1) are laterally contin-
uous, although some are lenticular-shaped. The thick
Navarette measured section of the Las Lavas sandstone is composed of poorly graded to massive,
Formation coarse-grained carbonate-rich sand (Tabd, Tdb).
Thick (5-15 m) packages of lithic-rich calcaren-
El Limon Member of the Las Lavas Formation ite and pebble conglomerate are found 240 m and
Along the Santiago-Puerto Plata highway, the 430 m above the base of the section (Fig. 16). At
basal conglomerate of the E1 Limon Member is ap- 240 m above the base, a 15-m-thick clastic limestone
proximately 50 m thick (Fig. 15). The conglomerate unit is in fault contact with underlying thin-bedded
is massive to poorly bedded, clast-supported, and siliciclastic rocks. Stratal boundaries are poorly de-
disorganized to organized (A2.1, AI.1) (Fig. 13B). veloped in the calcarenite, which is predominantly
Poor stratification is defined by oblate clasts pref- massive and contains few lithic fragments. Carbon-
erentially oriented parallel to bedding (Fig. 14B). ate material consists predominantly of red algae,
The basal contact with the underlying Altamira For- larger reefal foraminifera, coral fragments and car-
mation is faulted south of Canada Bonita (UTM bonate rock fragments.
071692). The lateral geometry of the conglomerate The 20-m-thick conglomerate exposed 310 m
horizon is difficult to discern in the field (e.g. Fig. above the base of the section is poorly stratified
14B). The conglomerate body appears to pinch out (A2.1 facies). Clasts are composed of similar litholo-
over a lateral distance of 10 m, and is replaced gies as those of the basal conglomerate of this sec-
by pebble stringers, which are found interbedded tion. Secondary gypsum fills some fractures in the
with very thin- to thin-bedded sandstone-siltstone clasts and in the surrounding thin-bedded siliciclastic
couplets. Conglomerate clasts have similar charac- rocks.
teristics to those described in the Las Lavas section. Vertical organization is not apparent in this sec-
Approximately 40 m of facies D siltstone and tion, although exposures are limited. Poorly defined
shale and minor amounts of C2.3 sandstone and a-axis clast orientations in the basal E1 Limon con-
siltstone overlie the basal conglomerate of the E1 glomerate suggests varied north-south to northwest-
Limon Member (Fig. 15). These siltstones and shales southeast paleocurrent trends.
are pervasively sheared, which results in a distinctive
blocky weathering pattern. There does not appear Rio Jacagua measured section of the Las Lavas
to be any vertical organization in this thin-bedded Formation
sequence.
Along the Santiago-Puerto Plata highway, the Beds of the Las Lavas Formation (Fig. 15) could
relationship between the calcarenite and siliciclas- not be directly correlated between the Arroyo Las
tic units is less clear. Carbonate conglomerate, Lavas and the Rio Jacagua sections (location maps in
foraminiferal rudstone and lithic-rich calcarenite (20 insets of Figs. 11 and 16). The thick lithic conglom-
m thick) are found in the Arroyo Guanabano (UTM erate that characterized the base of the Las Lavas
073687), east of the highway (Fig. 4A). West of and Navarette sections was not observed in the Rfo
the highway (UTM 068687), limestones of the Late Jacagua section. For this reason, the medium-bedded
Miocene-Pliocene Villa Trina Formation caps the to massive, lithic-rich calcarenite and carbonate con-
surrounding hilltops and occur as talus blocks at the glomerate was inferred to form the base of the Las
level of the road. Lavas Formation in the Rio Jacagua section. These
clastic limestones are about 15 m thick and ex-
La Pocilguita Member of the Las Lavas hibit similar characteristics to those in the E1 Limon
Formation Member in the Lavas and Navarette sections. The
In the Navarette section (Fig. 15B) the following Rio Jacagua section is dominated by small-scale (1-
four units comprise the La Pocilguita Member over- 10 m) thinning- and fining-up cycles composed of

carbonate-rich conglomerates (AI.1, A2.3), calcitur- because tectonic uplift has been less on the La Toca
bidites (C2.1, C2.2, C2.3) and siliciclastic turbidites block (Fig. 6A).
(C2.1, C2.2, C2.3, D2.2) (Fig. 15C).
Over 70 bidirectional paleocurrent structures in-
dicate a uniform northwest-southeast (130 ~ mean STRATIGRAPHY OF THE LA TOCA BLOCK
trend to paleoflow direction. The five unidirectional
indicators suggest a paleoflow towards the south- Definition of the La Toca block
east.
The basement rocks and sedimentary cover of the
Facies analysis of the Las Lavas Formation La Toca block is abruptly separated from the base-
ment rocks and sedimentary cover of the Altamira
Vertical facies assemblages and lateral facies rela- block by the Rio Grande fault zone (Fig. 4C). The
tionships in the Las Lavas Formation are not clearly La Toca block is bounded on the north by the Camt~
developed and appear to be more heterogeneous than fault zone, on the southwest by the Rio Grande fault
those described above from the Altamira Forma- zone, and on the south by the Septentrional fault
tion. The stratigraphy and lithology in the E1 Limon zone. The basement rocks and sedimentary cover of
Member indicate rapid deposition of coarse terrige- the La Toca block extend to the Rio San Juan area
nous and intrabasinal clastic sediment from high- of the eastern Cordillera Septentrional (Draper and
concentration turbidity currents and debris flows. Nagle, 1991).
The facies of the La Pocilguita Member are simi- The basement of the La Toca block consists of
lar to those found in the Altamira Formation. The two types of Upper Cretaceous to lower Tertiary
two formations differ mainly in their lithologic com- rocks: (1) volcanic rocks of the Pedro Garcfa Forma-
position, with the Las Lavas Formation containing tion in the central Cordillera Septentrional (Eberle
abundant calciturbidites and the Altamira Formation et al., 1982; Peralta-Villar, 1985); and (2) a het-
containing siliciclastic turbidites. The influx of intra- erogeneous assemblage of igneous and metamorphic
basinal, carbonate biogenic material in the Las Lavas rocks of the Rio San Juan complex in the eastern
Formation suggests an actively prograding carbonate Cordillera Septentrional (Draper and Nagle, 1991).
platform at the basin edge. The basal section of the La Toca Formation consists
Deposition by deep-marine turbidity currents in of interbedded sandstone, siltstone and conglomerate
the Las Lavas Formation is supported by the occur- and is faulted against both basement assemblages.
rence of graded bedding, good lateral continuity of The contact between the basement rocks and sed-
beds in larger outcrops, development of partial to imentary cover of the La Toca Formation is in-
complete Bouma sequences, vertical facies cycles, ferred to be a fault-modified unconformity, because
and diverse facies associations. Calciturbidites of the compositions of the Upper Eocene to middle
the Las Lavas Formation were deposited by grav- Oligocene basal conglomerates directly reflect the
ity-driven processes similar to those which deposited lithologies of the underlying Upper Cretaceous to
the siliciclastic turbidites because the two rock types lower Tertiary basement rocks.
share the same sedimentary structures.
Basement complex of the La Toca block
Carbonate rocks of the Villa Trina Formation
Pedro Garcia Formation
Siliciclastic rocks of the Altamira block are In the central Cordillera Septentrional, basement
capped by a widespread, little studied, --~250-m-thick rocks of the La Toca block consist of the Pedro Gar-
Upper Miocene to Lower Pliocene shallow-water cfa Formation, which is exposed in a 45 km 2 inlier
carbonate unit, the Villa Trina Formation (Fig. 6). west of the village of Pedro Garcfa (UTM 266670)
The Villa Trina Formation is composed of a lower (Fig. 4A). The fault-bounded igneous rocks are com-
unit of medium-bedded to massive, marly limestones posed mostly of volcanic rocks (aphanitic andesite,
with few isolated deposits of reefal material. The amygdaloidal andesite, tuff, and pyroclastic) with a
Villa Trina Formation is capped by well-indurated, minor amount of intrusive rocks (tonalite, basaltic
medium- to thick-bedded wackestones and pack- dikes, and porphyritic volcanic rocks) (Eberle et al.,
stones, which are interbedded with massive reefal 1982; Peralta-Villar, 1985). A single K-Ar radio-
deposits, and reefal talus deposits. This upper unit metric date indicates that this igneous complex is at
exhibits a karst topography. Similar lithologies of least 72 4- 6 Ma (Bowin and Nagle, 1982).
the Villa Trina Formation cap the La Toca block
to the east of the Rio Grande fault zone (Fig. 7). Rio San Juan complex
The Villa Trina Formation is more extensive on the The modal distribution of sandstone grains stud-
La Toca block than the Altamira block probably ied in thin-section and the composition of clasts

within conglomerates of the La Toca block exposed Age and paleobathymetry of the La Toca
in the central Cordillera Septentrional suggests a Formation
metamorphic and igneous source area. Likely silici- Thirty-six biostratigraphic analyses were per-
clastic source areas include the Rfo San Juan com- formed on twenty samples collected from the La
plex to the east (Draper and Nagle, 1991) and/or Toca Formation (Appendix 1 in de Zoeten, 1988)
the Puerto Plata basement complex to the north (Fig. 16). No microfossils from the basal conglomer-
(Pindell and Draper, 1991). Both complexes consist ate were identified. The overlying sediments range in
of serpentinite, gabbro, and mafic and felsic schist. age from Early Oligocene to early-Middle Miocene.
Based on similar lithologic assemblages and some Sedimentary rocks of the La Toca Formation could
similar rock types, Draper and Nagle (1991) believe be as old as Late Eocene, if they correlate with
that the Rfo San Juan and the Puerto Plata basement compositionally similar rocks described by Draper
complexes formed in the Upper Cretaceous to lower and Nagle (1991) that overlie the Rfo San Juan com-
Tertiary forearc-trench environment and were offset plex. Studies of benthic foraminifera in the La Toca
by the Camfi fault zone in Neogene time. Because Formation indicate deposition in middle to upper
of their lithologic similarities and the possibility of bathyal water depths (150-1500 m; Appendix 1 in
lateral fault offset, Mann et al. (1991) included both de Zoeten, 1988).
basement complexes as part of the Rfo San Juan-
Puerto Plata-Pedro Garcfa disrupted terrane. Measured sections of the La Toca Formation
One composite section was measured for the La
Stratigraphy of the La Toca Formation Toca Formation. Because of structural complexities,
this section consists of three parts (Fig. 16, inset
Outcrop distribution and general stratigraphy map). The base of the section (part A) lies northeast
of the La Toca Formation of the village of Altamira, where the Rfo Grande
The La Toca Formation crops out over an area of cuts through a steeply dipping basal conglomerate
100 km 2 in the central Cordillera Septentrional. The ridge (Fig. 4A). The majority of the section (parts B
La Toca Formation is bounded on the south and west and C) was measured to the east in outcrops along
by the Rfo Grande fault zone, and on the north by the the Rfo Yaroa (Fig. 4A).
Cam6 fault zone (Fig. 4C). The base of the La Toca The base of the La Toca section consists mostly
Formation is faulted against the Upper Cretaceous of a 300-m-thick package of amalgamated conglom-
igneous rocks of the Pedro Garcfa Formation (UTM eratic facies (A 1.4, A1. l, A2.3, A2.7) with minor in-
233663) (Fig. 4C). The La Toca Formation consists terbedded sandstones (B2.1; Fig. 16). Disorganized,
of an approximately 300-m-thick basal conglomerate matrix-supported facies (A1.4) and a minor clast-
(AI.1, A1.4, A2.3, A2.5 facies), which crops out as supported facies (A1.1) comprise the lower 200 m of
hogback ridges, up to 4 km in length (Fig. 16). the conglomerate. Most conglomeratic beds are tab-
Sedimentary rocks overlying the basal conglom- ular, medium-bedded to massive, with planar basal
erate (and possibly laterally equivalent to the con- contacts which rarely drape over clasts protruding
glomerates) are about 500-m-thick and composed from the underlying bed. Internal organization of
mostly of alternating, very thin- to medium-bed- the conglomerate appears to increase upward in the
ded, sandstone and shale couplets (C2.2, C2.3, D 1.2, section. Locally, upsection, inversely graded facies
D2.2) (Fig. 16). These couplets are capped by a (A2.3, A2.7) are interbedded with clast-supported,
300-m-thick package of thick- to very thick-bed- parallel-stratified conglomerates. Conglomerate hog-
ded, sandstones (B2.1, C2.1). These thick-bedded backs, extending for 5 km along strike, are sep-
sandstones are exposed in a faulted synclinal ridge arated by laterally equivalent, alternating, thin- to
(La Cumbre Ridge), striking northwest, which sepa- medium-bedded sandstones and shales (C2.2, C2.3,
rates north- and south-flowing streams in the central D2.2, D 1.2).
Cordillera Septentrional (Fig. 4A). Clasts in the base of the La Toca Formation range
High-angle faulting north of the Rio Grande fault from granule to boulder in size. They are equidi-
zone has complicated the stratigraphy of the La mensional to oblate in shape, and are subangular to
Toca Formation (Fig. 5B). To the north of the Rio rounded. The composition of clasts directly reflects
Grande fault zone, strata generally strike north- to the heterogeneous lithologies in the underlying Pe-
northwesterly and dip to the east. In this area, dro Garcfa Formation (Eberle et al., 1982). Clasts in
sedimentary rocks of the La Toca Formation are outcrops near the village of Pedro Garcfa (Fig. 16,
deformed by high-angle faults. Of the three silici- inset) are composed of approximately 70% volcanic
clastic formations of the E1 Mamey Group described rocks, including tuff, andesite and amygdaloidal
in this chapter, the La Toca Formation received the lava, 20% tonalite, and 10% sandstone, argillite (re-
least amount of study, and the following descriptions crystallized limestone), vein quartz, serpentinite, and
should be considered as a preliminary report. coralline rudstone. This indicates that the Pedro Gar-

cfa Formation was the major sediment source of the Few paleocurrent indicators were found in
La Toca Formation. the La Toca Formation. Nine measurements of
Above the basal conglomerate, the measured sec- groove marks indicate a northwest-southeast mean
tion continues in the east up the R/o Yaroa (Fig. 16). paleocurrent trend. Another twelve paleocurrent
350 m above the base of the section is a sedimentary structures were measured between anastomosing
package, 500 m thick, dominated by very thin- to high-angle faults in the Rfo Grande fault zone and
medium-bedded sandstones and shales (D1.2, D2.2, therefore are possibly subject to tectonic rotation.
C2.3 facies). Beds are tabular, laterally continuous, They suggest both northwest-southeast and north-
and exhibit sharp basal contacts. Basal sands are rel- east-southwest mean paleocurrent trend (Fig. 16).
atively coarse-grained, graded or structureless, and
have mud-rich, massive, and bioturbated upper di- Facies analysis of the La Toca F o r m a t i o n
visions (Tabd, Tbd). A few thin- to medium-bedded, The internal organization of the basal conglom-
lens-shaped calcareous sandstone beds are interbed- erate of the La Toca Formation is distinct from
ded with the sandstone and shale couplets. conglomerate described in both the Altamira and
Higher in the section lithic-rich calciturbidites Las Lavas formations. Matrix-supported, basal con-
and lithic conglomerate are interspersed between glomerate of the La Toca Formation indicates rapid
thin-bedded siliciclastic couplets (Fig. 16). Calci- deposition from cohesive debris flows (Middleton
turbidites are organized into C2.1 and C2.2 facies and Hampton, 1976). Whether deposition occurred
(Tabd, Tbcd), but are less abundant than in the Las from broad, unconfined sheets or from channelized
Lavas Formation (e.g. Fig. 15). Together with the flows is not clear.
conglomerate (1-2-m-thick AI.1 facies), the calci- Very thin- to medium-bedded sandstones and silt-
turbidites form small (1-5 m thick) coarsening- and stones (C2.2, C2.3, D2.2, D2.1) in the lower part of
thickening-up cycles. the La Toca Formation suggest deposition from low-
Thick-bedded sandstones (B2.1, C2.1, C2.2) and concentration turbidity currents and from high-con-
lesser amounts of conglomerate beds (AI.1, A2.7) centration silt-dominated turbidity currents (Fig. 16).
form the upper 300 m of the La Toca Formation Coarsening and thickening of sandstone upsection
(Fig. 16). Sandstone beds have high sand to silt suggests progradation of a marine depositional sys-
ratios ( 2 : 1 - 1 0 : 1), are commonly tabular, laterally tem. Thick-bedded sandstone capping the La Toca
continuous, and thick- to very thick-bedded (rang- Formation section was deposited by high-concentra-
ing from 0.5 to 3.5 m thick). Basal sands range tion turbidity currents. Because vertical organization
from gravel to medium sand in grain size. Lower is lacking in these thick, tabular sandstone beds, it is
division of beds are coarse-tail graded, or massive, unclear if they were deposited as lobes, channel-lobe
and grade up into parallel-laminated sands, rarely transitional facies, or as delta frontal sands.
containing tipple marks (Tabd, Tbd). Load casts are
the dominant type of sole marks. The upper division
in many beds contain concentrations of lignite and PALEOCURRENTS FROM THE EL MAMEY GROUP
amber fragments that define parallel laminae in the
siltstones. The finer fraction of beds is mud-poor and More than 280 bidirectional and unidirectional
parallel-laminated or massive. paleocurrent indicators were recorded from the Up-
Locally interbedded with the sandstone beds are per Eocene to Lower Miocene siliciclastic rocks in
1-10-m-thick pebble to small cobble conglomerate the E1 Mamey Group (Fig. 17). For completeness
beds (AI.1, A2.7 facies) (Fig. 16). Clast-supported, another 90 measurements from work by Dolan et
disorganized conglomerates are dominant. Clasts are al. (1991) were added to the measurements col-
composed of a wide assortment of lithologies, which lected during this study. Measurements taken from
include volcanic, plutonic, metamorphic, and sedi- structurally tilted beds have been rotated around hor-
mentary rock fragments. Vein quartz and serpentinite izontal axes. Sole marks are generally scarce in the
fragments are present. study area, but occur mostly in facies C2.1 and C2.2.
In the 800-m-thick section above the basal con- (Fig. 8). Unimodal measurements are based on flute
glomerate, increasingly coarse sand and conglom- casts and ripple marks. Groove casts, channel axes
erate beds define a thickening- and coarsening-up and a-axis clast orientation provided bimodal current
cycle (>400 m) that is capped by almost 300 m indicators.
of thick-bedded sandstones. Thin (1-5 m) isolated Rose plots of bidirectional measurements shown
sandstone packets also reflect thickening-upward in Fig. 17 indicate a uniform northwest-southeast
trends. The thick-bedded sandstone at the top of the trend in paleocurrents continued from the Late
section shows no large-scale thickening- or thinning- Eocene through the Early Miocene. Unimodal pale-
upward trend, but locally there are both thinning- ocurrent indicators, however, suggest that the pale-
and thickening-up cycles (Fig. 16). oslope reversed during latest Oligocene time. Uni-

C E N O Z O I C EL M A M E Y G R O U P OF N O R T H E R N H I S P A N I O L A 275
Fig. 17. Paleocurrent data from siliciclastic marine rocks from the Upper Eocene to Lower Miocene Altamira, Las Lavas, and La Toca
formations in the central and western Cordillera Septentrional (modified from Dolan et al., 1991). Rose diagrams differentiate between
unidirectional (shown in white) and bidirectional (shown in black) paleocurrent indicators. Unidirectional paleocurrent indicators from
the Lower Miocene Las Lavas Formation near Monte Cristi and the Lower Miocene La Toca Formation from near Moca show a mean
northwesterly paleoflow.
modal indicators found in the Altamira Formation SANDSTONE PETROGRAPHY OF THE EL MAMEY
(Upper Eocene to Upper Oligocene) show pale- GROUP
oflow towards the southeast (Fig. 17). Paleocurrents
measured in the Las Lavas and La Toca forma- Methodology
tions (middle Oligocene to Lower Miocene) suggest
paleoflow to the northwest (Fig. 17). A signifi- Thirty-six medium- to coarse-grained sandstone
cant number (8%) of paleocurrent indicators are samples were selected for point-counting to deter-
oriented at high angle (N-S) to the predominant mine the petrographic mode (Dickinson, 1970; Gra-
current orientation, and suggest the possibility of ham et al., 1976). Samples are representative of the
lateral (north or south) sediment source areas. Alter- 5-km-thick siliciclastic section of the Altamira, Las
natively, anomalous north-south oriented paleoflow Lavas, La Toca formations, and range from Upper
structures may simply have formed from interchan- Eocene to Lower Miocene (location of petrographic
nel deposits. samples shown on all measured sections). Samples
Dolan et al. (1991) compiled paleocurrent data were collected over a 500 k m 2 geographic area.
from four Late Cretaceous to Miocene basins in The Gazzi-Dickinson point-count method de-
both Hispaniola and Puerto Rico and included some scribed by Ingersoll et al. (1984) was followed.
of these data from the E1 Mamey Group. Dolan et Using this method, a point is counted as a rock
al. (1991) showed that all four basins are charac- fragment if the cross-hairs fall on the aphanitic part
terized by elongate shapes with most paleocurrents of a rock fragment. A point is counted as a mineral
oriented in a basin-parallel orientation. The elongate grain if the cross-hair falls on a phenocryst greater
basins in Puerto Rico appear to have formed as in- than 0.0625 mm. For each thin-section, 300-350
tra-arc basins which accompanied volcanic activity points were identified. Thin-sections were stained
in the arc. The elongate basins of Hispaniola have with sodium cobaltinitrite to facilitate recognition of
largely post-dated magmatic activity of the Hispan- orthoclase. Because of the inherent difficulty of rec-
iola arc. ognizing chert from felsic volcanic rock fragments,

276 R. D E Z O E T E N and E M A N N
Table 1
Point-count data based on 300-350 point counts per thin section following the conventions of the Gazzi-Dickinson method (Ingersoll,
1984)
No. Sample no. Age Quartz Feldspar Lithicfragments Qm F Lt Qp Lvm Lsm Lv Lm Ls
Altamira Formation
1 15387 Early Eocene 1 19 80 1 19 80 0 100 0 100 0 0
2 2387 Early Eocene 1 3 96 0 3 97 1 99 0 100 0 0
3 1687 Early Oligocene 2 22 76 1 22 77 1 85 14 85 0 15
4 4908 Early Oligocene 18 50 32 13 50 37 15 57 28 70 0 30
5 -746 Early Oligocene 1 14 85 1 14 85 0 93 7 93 0 7
6 -745 Early Oligocene 0 14 86 0 14 86 0 82 18 82 0 18
7 -744 Early Oligocene 1 9 90 0 9 91 1 70 29 70 0 30
8 4687 Early Oligocene 0 13 87 0 13 87 0 92 8 92 0 8
9 15687 Early Oligocene 0 46 54 0 46 54 0 100 0 100 0 0
Las Lavas Formation

10 10187 Eocene? 3 33 64 2 33 65 2 97 1 99 0 1
11 6687 Oligocene 1 58 41 0 58 42 1 98 1 99 0 1
12 7 Early Oligocene 1 24 75 1 24 75 0 98 2 98 0 2
13 6187 Early Oligocene 2 20 78 2 20 78 1 98 1 98 0 1
14 6087 Early Oligocene 8 10 82 2 10 88 7 89 4 96 1 3
15 12387A Early Oligocene-Early Miocene 2 17 81 1 17 82 1 98 1 99 0 1
16 5787 Early Oligocene-Early Miocene 1 10 89 1 9 90 1 99 0 100 0 0
17 6887 Early Miocene 15 37 48 10 37 53 9 89 2 98 0 2
18 1B Early Miocene 18 42 40 16 42 42 6 92 2 99 0 1
19 10987 Early Miocene 27 25 48 13 25 62 24 75 1 93 5 2
20 10887 Early Miocene 0 30 70 0 30 70 0 100 0 100 0 0
21 8387 Early Miocene 0 39 61 0 39 61 0 100 0 100 0 0
22 7387 Early Miocene 34 56 10 14 56 30 64 4 32 89 0 11
23 1787 Oligocene? 18 20 62 7 20 73 16 84 0 100 0 0
24 1887 Oligocene? 27 38 35 13 38 49 29 71 0 100 0 0
25 -521 Oligocene? 3 23 74 1 23 76 3 86 8 90 0 10
26 14287 Oligocene? 1 39 60 1 39 60 0 99 1 99 0 1
La Toca Formation
27 9487 Oligocene 5 16 79 4 16 80 2 97 1 99 0 1
28 7887 middle Oligocene 40 21 39 9 21 70 53 46 1 70 28 1
29 8687 middle Oligocene 33 46 21 26 46 28 25 70 5 90 1 9
30 13987 Late Oligocene 51 41 8 9 41 50 92 8 0 100 0 0
31 8887 Late Oligocene 30 60 10 21 60 19 49 51 0 100 0 0
32 13187 Late Oligocene 41 39 20 20 38 42 50 50 0 100 0 0
33 10587 Early Miocene 74 12 14 15 11 74 93 1 6 28 41 31
34 8287 Early Miocene 29 60 11 7 60 33 69 29 2 96 0 4
35 9587 Early Miocene 28 54 18 18 54 28 35 56 9 91 1 8
Percentages are recalculated to 100%. Data indicate that sands from the La Toca Formation consist of much more quartz than coeval
sandstones from the Altamira Formation. The quartz grains from the La Toca Formation are predominantly polycrystalline. Key to
abbreviations: Q -- quartz, F = feldspars, L = lithic fragments, Qm = monocrystalline quartz, Lt = total lithic fragments and
polycrystalline quartz, Qp = polycrystalline quartz, Lvm -- total volcanic and metavolcanic lithic fragments, Lsm -- total sedimentary
and metasedimentary lithic fragments, Lv = total volcanic lithic fragments, Lm = total metamorphic lithic fragments, Ls = total
sedimentary lithic fragments.
the former was grouped with the felsic fragments. lithic arkose (Fig. 18). Sandstone of the Late Eocene
Recalculating framework modes by assigning the to Late Oligocene age Altamira Formation are char-
questionable polycrystalline grains to chert showed acterized by: (1) abundant volcanic lithic fragments;
no significant (<5%) change in their ternary distri- (2) feldspars; (3) biogenic fragments; and (4) the
bution. Recalculated point-count data are shown in absence of quartz (Fig. 19A).
Table 1.
Las Lavas Formation
Results The composition of Las Lavas Formation sand-
stone changed from Late Oligocene to Early
Altamira Formation Miocene time (Fig. 18A). Quartz and plagioclase
Framework modal data indicate that the siliciclas- grains gradually increase in abundance upsection,
tic rocks in the central Cordillera Septentrional range whereas the concentration of volcanic lithic frag-
in composition from volcanic arenite to quartz-rich, ments remains constant. Similar to the Altamira For-

Fig. 18. (A) Ternary plots of quartz, feldspar, and lithic fragments (QFL) which illustrate change in composition with time for the (1)
Altamira Formation, (2) La Toca Formation, and (3) Las Lavas Formation. From Late Eocene to Early Miocene the underlying arc and
forearc crust of the area is subjected to progressively deeper levels of erosion related mainly to tectonic events. (B) QFL ternary diagram
of sandstone framework grains from all three formations. Tectonic provenance fields from Dickinson et al. (1983) are shown.
mation, sandstone of the Las Lavas Formation (Up- Detrital mineral grains such as hornblende, serpen-
per Oligocene) is volcanic arenite. Lower Miocene tinite, epidote, zoisite, and several types of mica
sandstone from the Las Lavas Formation, on the make up a significant proportion (5-30%) of the
other hand, is feldspathic litharenite (Fig. 19B). framework grains.
Detrital serpentine was found in three samples
near the top of the Lavas section (two samples of Comparison of all formations of the El Mamey
Late Oligocene age; one sample of Early Miocene Group
age). In these three samples, serpentinite fragments Sandstone from the Altamira, Las Lavas, La Toca
make up almost 30-+- 10% of the framework grains formations exhibit several similar characteristics.
in each sample. Several Miocene sandstones from They contain abundant intrabasinal biogenic mate-
the Las Lavas Formation contain albite-rich plagio- rial, such as red algae, larger reefal foraminifera,
clase, which contain parallel-oriented needle-shaped and lesser amounts of coral and shell fragments.
microlites, along with zoisite and epidote inclusions. The lithic fragments are predominantly volcanic, al-
These types of plagioclase grains are also com- though, significant quantities of sedimentary rock
monly found in plagioclase grains from the La Toca fragments are recognized in sandstones from the
Formation sandstone. Altamira Formation. Terrestrial organic debris (lig-
nite and amber) is a minor constituent in all the
La Toca Formation sandstones. The majority of the lignite and am-
Sandstone from the La Toca Formation has a ber deposits are concentrated in the thick-bedded
much lower volcanic lithic component (24-1-3%), sandstone at the top of the La Toca Formation (Red-
and a significantly higher quartz concentration mond, 1982; Iturralde-Vinent and MacPhee, 1996)
(37 :i: 3%), than sandstone from the Altamira and (Fig. 16). Overall, the sandstone of the E1 Mamey
Las Lavas Formations (Figs. 18A and 19C). Poly- Group are well compacted, contain very little clay-
crystalline quartz is the most abundant and is proba- or mud-sized fraction (<5%), and are moderately
bly metamorphic in origin. Plutonic quartz makes up well cemented with sparry calcite and much lesser
most of the inclusion-rich monocrystalline quartz. amounts of laumonite and analcite. All sands have


very low porosities (<3%). Chloritization, epidotiza- DISCUSSION

tion, and clay alteration of volcanic rock fragments
are likely to be pre-depositional, because alteration Three-phase tectonic and sedimentary history of
either appears to be localized within grain bound- northern Hispaniola
aries, or slightly affecting the matrix (Fig. 19A,B).
Some plagioclase grains show partial calcite replace- Based on sedimentary data presented in this pa-
ment and/or seritization. per, the tectonic and basin evolution of northern
Hispaniola can be divided into three phases, each
of which is marked by distinct depositional facies
Provenance of the Altamira Formation
(Fig. 20): (1) Paleocene to Middle Eocene phase
records the termination of arc activity and uplift
The Altamira Formation sandstones are charac-
of arc basement rocks probably as the result of
terized by very low quartz contents, high plagio-
early interaction between the Hispaniola arc and
clase/orthoclase values, and a volcanic-rich lithic
the Bahama carbonate platform; (2) Late Eocene to
component. These grain parameter values all point
Early Miocene phase marks the first major deposi-
to a volcanic provenance (Fig. 18B). The volcanic
tion of deep-marine siliciclastic rocks; and (3) Late
rock fragments are composed predominantly of lath-
Miocene to Early Pliocene phase records tectonic
work and felsitic (65-97%) grains with a lesser
uplift of northern Hispaniola to near sea level and
amount of microlitic grains. These grains suggest
subsequent deposition of shallow-marine limestones
a diverse volcanic source with a composition that
of the Villa Trina Formation (Fig. 6).
ranges from silicic to basaltic lavas (Dickinson,
Data from this study of the central Cordillera
1970). Framework grains of sandstones plotted on
Septentrional has refined the timing and extent of
a QFL diagram fall within the provenance field of
these phases and, therefore, the tectonic history of
a magmatic arc setting as described by Dickinson
plate interactions in this part of the North America-
and Suczek (1979) and Dickinson et al. (1983) (Fig.
Caribbean plate boundary zone
18B). Plotting the modal distribution of sandstone
from the Las Lavas Formation on a QFL ternary dia-
Phase 1: Paleocene to Middle Eocene
gram (Fig. 18B) indicates that Miocene sandstone is
derived from a transitional magmatic arc setting. The
Tectonics
composition of Altamira Formation and Las Lavas
Coeval igneous and metamorphic rocks in north-
Formation sandstone of Late Oligocene age suggests
ern Hispaniola are interpreted to have formed in an
an undissected arc source (Dickinson et al., 1983).
intra-oceanic island-arc environment at the leading
edge of the Caribbean plate (Bowin, 1975; Nagle,
Provenance of the La Toca Formation 1979; Draper and Nagle, 1991; Pindell and Draper,
1991; Dolan et al., 1991; Calais and Mercier de
La Toca Formation sandstone is more heteroge- Ldpinay, 1995) (Fig. 20A). Isotopic ages for north-
neous and their framework modes plot in the feldspar ern Hispaniola suggest that island-arc development
and quartz realms of the QFL diagram, which corre- was continuous from a mid-Cretaceous orogenic
lates to the dissected magmatic arc field (Dickinson event (Draper et al., 1996) to the Late Eocene or
et al., 1983) (Fig. 18B). The presence of volcanic Early Oligocene (Kesler et al., 1991). Near the end
lithic fragments and the high plagioclase/orthoclase of the arc phase, initial opening of the Cayman
ratio suggests that the sedimentary source for the La Trough marks a change from northeast to eastward
Toca Formation continued to be a volcanic arc. The Caribbean plate motion and the transition from a
increase in compositional complexity in the La Toca convergent to the present east-west strike-slip mar-
Formation implies breaching of the shallow levels gin (e.g., Sykes et al., 1982; Mann et al., 1995).
of the arc with exposure of plutonic-metamorphic Several authors believe that the collision of the
rocks (Fig. 18B). Caribbean plate with the buoyant Bahama Platform
Fig. 19. (A) Thin-section photomicrograph with crossed-polars of an Upper Eocene to Upper Oligocene Altamira sandstone consisting
of felsitic and microlitic volcanic rock fragments, plagioclase, and carbonate rock fragments. The width of the photo is approximately
1.65 mm (sample no. 7-46). (B) Thin-section photomicrograph with crossed-polars of a typical Lower Miocene sandstone of the Las
Lavas Formation which consists of polycrystalline quartz, plagioclase, felsite and lathwork type volcanic rock fragments, and bioclasts.
The width of the photo is approximately 3.3 mm (sample no. 10887). (C) Thin-section photomicrograph with crossed-polars of a Lower
Oligocene to Lower Miocene sandstone from the La Toca Formation. This photomicrograph shows polycrystalline and monocrystalline
quartz, plagioclase, volcanic rock fragments, and zoisite. The large, twined plagioclase grain near the center of the photomicrograph
contains unidentified, needle-shaped microlites which exhibit a preferred parallel orientation. The width of the photo is about 1.65 mm
(sample no. 10587). Color version at http://www.elsevier.nl/locate/caribas/

280 R. DE Z O E T E N and R M A N N
Fig. 20. Block diagrams summarizing three main tectonic phases in the evolution of the North America-Caribbean plate boundary zone
in northern Hispaniola. (A) Phase 1 is marked by Paleogene to Early Eocene deposition of hemipelagic, fine-grained turbidites (Los
Hidalgos Formation = number 5) which are interbedded with arc-related dikes and sills of intermediate composition. Similar deep-marine
sedimentary rocks are found to the north (Imbert Formation = number 4) and to the south (Magua Formation = number 6) of the study
area and suggest a regionally extensive basin at least 40 km wide. The substrate of the Imbert Formation is a heterogeneous basement
consisting of serpentinite, gabbros, volcanic rocks and blueschists (Puerto Plata basement and Rio San Juan complexes = number 3),
whereas the substrate of the Magua Formation is greenschist metamorphic rocks intruded by granodiorite plutons (Duarte complex =
number 7). Tuffaceous horizons are common in the Imbert, Los Hidalgos, and Magua formations and suggest an active arc environment
probably to the south along the Hispaniola segment of the volcanic arc. We interpret these geologic relationships in terms of a forearc
basin developed above a south- to southwest-dipping slab of subducted Atlantic ocean floor (number 2). Large-scale, Middle Eocene
folding and uplift will terminate Phase 1 deposition. This compressive event is related to attempted subduction of the Bahama Platform
(number 1) beneath the forearc area. (B) Phase 2 is marked by Upper Eocene to Lower Miocene deposition of several kilometers of
siliciclastic turbidites (El Mamey Group = number 8; Tabera Group -- number 9) in west-northwesterly striking, elongate basins. Arrows
indicate paleoflow directions based on paleocurrent studies in turbiditic rocks (cf. Fig. 17). Source areas for the Tabera Group include
Lower Cretaceous metasedimentary rocks (Amina schists = number 10) and volcanic arc rocks exposed to the east. Source area for the E1
Mamey Group include folded, hemipelagic rocks of the Los Hidalgos Formation to the south (number 5), the Puerto Plata basement and

terminated subduction and arc-related volcanism and ship was recognized by us in the central Cordillera
initiated east-west movement (Pindell and Draper, Septentrional between the tightly folded Upper Pa-
1991; Dolan et al., 1991; Mann et al., 1995). Calais leocene to Lower Eocene Los Hidalgos Formation
and Mercier de L6pinay (1995) have correlated the and less folded overlying Upper Eocene Altamira
Late Eocene deformation seen in northern Hispan- Formation. In the western Cordillera Septentrional,
iola with a convergent event of the same age in Calais et al. (1992) recognized this same unconfor-
southern Cuba and northern Haiti. mity separating the Paleocene-Eocene E1 Cacheal
tufts from the overlying Lower Miocene series.
Paleocene-Eocene rock record
The oldest sedimentary rocks in northern Hispan- Interpretation of phase 1 tectonic and
iola are mainly turbidites and hemipelagic sedimen- sedimentary events
tary rocks of Paleocene to Middle Eocene age. These Unconformable contacts between the basement
rocks were deposited in a deep-marine environment and overlying sedimentary rocks indicate that base-
above igneous and metamorphic arc basement rocks ment blocks were locally uplifted and exposed by
(Fig. 20A). This deep-marine depositional phase Late Eocene time, and possibly as early as the Late
seems to have affected all components of the arc Paleocene. Lithologies of the Imbert and Magua
system, including: (1) deposition of the Imbert For- formations indicate that arc-derived sediments were
mation in the 'outer forearc-trench setting' (Nagle, deposited on the edges of the deep-marine forearc
1979; Pindell and Draper, 1991); (2) deposition of basin (Fig. 20A). The composition of the Los Hidal-
the Los Hidalgos Formation in the 'inner arc set- gos Formation indicates that very little terrigenous
ting'; and (3) deposition of the Magua Formation sediment reached its position in the deep basin.
on the volcanic arc (Palmer, 1979). Unlike the Los Three reasons may be responsible for the absence
Hidalgos Formation, the Imbert and Magua for- of terrigenous sediment in the Los Hidalgos For-
mations contain a significant amount (10-60%) of mation: (1) the basin was effectively isolated from
siliciclastic rocks. a siliciclastic source, possibly because of an irreg-
Igneous activity during deposition of the Imbert ular bottom topography; (2) the basin was a great
and the Los Hidalgos formations is shown by in- distance from the source of siliciclastic material; or
terbedded calcareous tuffaceous rocks in the Imbert (3) because only a small area of arc rocks were
Formation and Palma Picada porphyritic rocks in- subaerially exposed during this time.
truded into the Los Hidalgos Formation (Fig. 20A). The deep-marine origin and present proximity of
Although no radiometric ages have been determined the Imbert and Los Hidalgos formations suggest that
for these igneous rocks, pelagic foraminifera from these formations may be laterally equivalent. Less-
the tuffaceous rocks of the Imbert Formation indi- folded shallow-marine limestone deposits of the La
cate a Paleocene-Early Eocene age (Nagle, 1979). Isla Formation overlie folded rocks of the Imbert
The Palma Picada igneous rocks intruded the cal- Formation and suggest that the Imbert Formation
careous sediments of the Los Hidalgos Formation was uplifted with an outer 'arc-trench' assemblage
either during or after their deposition and must be to near sea level during Late Paleocene to Middle
Late Paleocene to Middle Eocene in age. Eocene time (Pindell and Draper, 1991). A simi-
Pindell and Draper (1991) report that the Ira- lar uplift history for the Los Hidalgos Formation
bert Formation is overlain unconformably by Upper is suggested based on the Middle Eocene angular
Eocene sedimentary rocks of the Luperon Forma- unconformity separating the folded Los Hidalgos
tion. The stratigraphic relationship between shallow- Formation and the less-folded Altamira Formation.
marine, Lower to Middle Eocene limestone of the The mixture of shallow-marine carbonate clasts and
La Isla Formation and the Imbert Formation is un- clasts derived from the Los Hidalgos Formation at
clear, but the La Isla Formation limestone is believed the top of the basal conglomerate (Ranchete Mem-
to post-date Imbert deposition and pre-date Upper ber) of the Altamira Formation further suggests that
Eocene rocks above the unconformity (Pindell and the Los Hidalgos Formation had been uplifted to
Draper, 1991). A similar, unconformable relation- near sea level by Late Eocene time. Thus, the fore-
Rfo San Juan complexes and Pedro Garcfa Formation to the north (number 3). Regional uplift in Middle Eocene time is attributed to
the initial attempted subduction of the Bahama Platform (number 1) beneath the Hispaniola arc and Oligocene to Miocene left-lateral,
strike-slip faulting along the Rio Grande fault zone (RGFZ) and the Septentrional fault zone (SFZ). (C) Phase 3 is marked by Upper
Miocene to Lower Pliocene deposition of shallow-watercarbonate rocks (Villa Trina Formation -- number 11). This limestone appears to
have covered most of northern Hispaniola as shown by the wide distribution of its remnants (cf. map in Fig. 6). Late Pliocene to Present
uplift of the Cordillera Septentrional along the transpressional Septentrional fault zone has folded the Villa Trina Formation and uplifted
it to an elevation of 1250 m. Uplift of the Cordillera Septentrional has accompanied subsidence of coeval rocks in the Cibao basin to
depths greater than 3500 m below sea level (Yaque Group = number 12).

arc may have uplifted as a single block during this would require paleocurrents flowing from the south
time period. This folding and uplift event coincides and southwest. This prediction, however, conflicts
with the cessation of most subduction-related pro- with the mean paleocurrent trend (125 ~ measured
cesses and probably resulted from the early oblique from 170 structures, 40 of which indicate that the
collision of the Caribbean plate with the Bahama current flowed to the southeast (Fig. 17).
Platform (Pindell and Draper, 1991). High concentrations of calcareous biogenic mate-
rial in the turbidite rocks of the E1 Mamey Group
Phase 2: Late Eocene to Early Miocene indicate shallow-water carbonate production near the
shelf margin. Limestone turbidites and debris sheets
Tectonics recognized in slope and basinal settings have com-
Phase 2 covers the depositional period from the monly been reported in both ancient rocks (Cook and
Late Eocene to the Early Miocene (that is, the pe- Mullins, 1983) and modem environments (Schlager
riod of deposition of the Altamira, Las Lavas, and and Chermak, 1979). Deep-marine fossiliferous
La Toca formations; Fig. 7). The tectonic regime limestones, like those from the Las Lavas Formation,
during this period has been interpreted by Sykes et commonly emanate from a line source dissected by
al. (1982) and Mann et al. (1995) as transitional: several small channels and are rarely associated with
collision with the Bahama Platform was ending, and submarine fans (Cook and Mullins, 1983).
the Caribbean plate was moving along strike-slip Vertical facies relationships show that the
faults in a more easterly direction. In the west- Oligocene section exposed along the Santiago-
ern Cordillera Septentrional, Calais et al. (1992) Puerto Plata highway is more conglomeratic than the
provided important structural confirmation of this Upper Eocene section near E1 Mamey. A gradual,
transition period. They mapped an older set of col- upsection transition from lobe to channel/overbank
lision-related folds affecting Paleocene and Eocene deposition suggests that southeasterly prograding
rocks equivalent to the Los Hidalgos Formation with submarine fan systems developed during Late
folds having north- to east-northeast-trending axial Oligocene time. Vertical relationships in the La
traces and a younger, strike-slip set of folds affecting Toca Formation point to a prograding submarine fan
Miocene rocks that have northwest-trending axial or delta during Early Miocene time.
traces and sub-vertical fold axes. A major, Late Oligocene depositional event is
marked by sudden appearance of limestone con-
Late Eocene to Early Miocene rock record glomerate and calciturbidite in the E1 Limon Mem-
Systematic lateral facies and facies assemblage ber at the base of the Las Lavas Formation (Fig. 15).
changes which commonly characterize submarine Intrabasinal, clastic limestone beds increase upward
fans (e.g., Walker, 1984) are not well expressed in the section. The section is overlain by platform
in the siliciclastic deposits exposed in the central carbonate of the Villa Trina Formation (Fig. 6).
Cordillera Septentrional. The best documentation of Sandstone petrography from the Altamira, Las
lateral facies relationships is recorded in the mea- Lavas and La Toca formations on the QFL diagram
sured sections from the Altamira Formation (Figs. 9, indicates that these sediments came from an arc
11 and 12). The channel (AI.1, A2.1) and overbank environment (cf. Dickinson and Suczek, 1979) (Fig.
deposits (C2.2, C2.3, D2.2) in the Northern Canada 18B). Modal distribution of framework grains sug-
Bonita, Southern Canada Bonita, and Calabaza sec- gest two compositional trends: (1) sandstone from
tions resemble middle submarine fan deposits (Mutti the Altamira and Las Lavas formations have a dif-
and Ricci Lucchi, 1978; Nilsen and Abbot, 1981). ferent sand grain composition from coeval sandstone
The stacked tabular sandstones (B2.1, C2.1) seen from the La Toca Formation; and (2) quartz con-
in the Rio Perez and Llanos syncline sections are tent increases upsection from the Altamira through
similar to deposits interpreted as outer fan lobes the Las Lavas Formation. Compositional differences
or as large crevasse-splay lobes in the middle fan in the sandstones from the La Toca and Altamira
environment (Mutti and Ricci Lucchi, 1978; Shan- formations imply two distinct source areas, with
mugan and Moiola, 1988). Thin- to medium-bedded apparently no mixing between the two. This in-
sandstone and siltstone facies (C2.2, C2.3, D2.2) terpretation is further supported by the absence of
seen in the E1 Mamey and Guananico sections may serpentinite grains in the Altamira Formation sand-
represent basin plain or distal overbank deposits. stones. In contrast, serpentinite clasts are found in
The Rio Perez, Llanos Syncline, Southern Canada the La Toca Formation as well as in the Lower to
Bonita, Northern Canada Bonita, and Calabaza sec- Middle Eocene sedimentary rocks to the north of the
tions are all laterally equivalent. Their areal dis- Camfi fault zone. Only the very youngest sandstone
tribution in the E1 Mamey Group suggests north to (Lower Miocene) in the Las Lavas Formation have
northeast fan progradation from the Northern Canada similar modal distributions as the sandstone from
Bonita to the Rio Perez section. Such progradation the La Toca Formation. Superimposed on this trend

is a time-dependent unroofing sequence from Upper Miocene sandstones from both the La Toca and Las
Eocene undissected magmatic arc provenance to a Laves formations (Fig. 18A); and (2) the widespread
Lower Miocene dissected arc provenance (Fig. 18B). cover of Upper Miocene to Lower Pliocene Villa
Calais and Mercier de L6pinay (1995) have corre- Trine Formation limestones over the siliciclastic for-
lated Upper Eocene to Lower Miocene sedimentary mations (Fig. 6).
formations in northern Hispaniola (Altamira Forma- The consistent northwest-southeast trend of pa-
tion) with similar rocks in northern Haiti and south- leocurrents measured in Upper Eocene to the Lower
ern Cuba. They interpret this period as a time of tec- Miocene indicates a long-lived, elongate basin as pro-
tonic quiescence when topographic relief generated posed by Dolan et el. (1991) for the E1 Mamey and
by the Late Eocene uplift event was eroding to sea three other coeval deformed basin complexes in His-
level. The Late Eocene uplift and folding event may paniola and Puerto Rico. Deposition of the E1Mamey
explain why there is little or no Oligocene preserved Group rocks within an elongate depositional basin is
in the western part of the Cordillera Septentrional further supported by: (1) complex lateral facies pat-
(Calais et el., 1992). terns, which do not form classical deep-sea fan mor-
phologies (Link, 1982); and (2) regional stratigraphy
Interpretation of phase 2 tectonic and which suggests that the 'outer forearc-trench' assem-
sedimentary events blage was bathymetrically shallow to the north of the
Sandstone petrography and facies relationships basin from Early Eocene time to the Present (Pindell
suggest that the La Toca Formation was probably and Draper, 1991), and that the area of the Los Hidal-
deposited in a separate basin from the basin in gos Formation to the south of the basin also formed
which the Altamira and Las Lavas formations were a bathymetric high during at least Late Eocene time
deposited (Fig. 20B). However, it is also possible and possibly into the Oligocene (Fig. 20B).
that these formations may have been deposited in The mechanism that produced the complex basins
different localities within a larger basin with a wide of the E1 Mamey Group is unclear. Although the
range of source areas. The La Toca Formation is basin is situated in the forearc region of Hispan-
separated from the coeval Altamira and Las Lavas iola, siliciclastic deposition began during the waning
formations by the 100-400-m-wide, left-lateral Rfo stages of arc activity in Late Eocene time (Fig.
Grande fault zone (de Zoeten and Mann, 1991). 20B). The basin may have formed by down-warping
The composition of sandstone and conglomerate between the uplifted 'outer forearc-trench' assem-
changes abruptly across the Rfo Grande fault zone. blage (Imbert Formation) and the 'inner forearc'
Quartz-rich, La Toca Formation sandstone lie north (Los Hidalgos Formation)' during the final stages
of the Rfo Grande fault zone, whereas lithic-rich of the collisional event with the Bahama Platform.
Altamira Formation sandstone occur to the south of Folding and uplift of these arc environments would
the fault. Distinct Upper Cretaceous to lower Tertiary provide igneous and metamorphic source areas that
basement complexes underlying sedimentary rocks seem to be absent prior to the Middle Eocene folding
are likewise separated by the Rfo Grande fault zone. and uplift event. Calais et el. (1992) show that an
Differences in basement rock lithologies, com- older set of folds affecting the Eocene Los Hidal-
position of overlying sandstone, and sedimentary gos equivalent rocks in the area (El Cacheal tufts)
facies profiles suggest distinct origins for the sedi- of the western Cordillera Septentrional has north to
mentary rocks now juxtaposed along the Rfo Grande east-northeast axial trends while a younger set of
fault zone. Left-lateral oblique-slip motion on the folds with northwest-trending axial traces affecting
Rfo Grande fault zone appears to have juxtaposed post-Eocene rocks. They interpreted the older set
the Altamira and La Toca blocks which Mann et of folds as related to the Bahama collision and the
al. (1991) have interpreted as part of much larger younger set as related to post-collisional strike-slip
terranes (Fig. 20B). Similar relationships, in which movement.
blocks with distinctly different geologic histories are Alternatively, the highs flanking the basin may
in fault contact, are documented for several of the have been formed by strike-slip faults roughly paral-
ten other terranes proposed by Mann et al. (1991) lel to the axis of the basin (Fig. 20B). Whether the
in Hispaniola. Movement of blocks along multi- basin formed in a convergent or strike-slip setting is
ple high-angle faults is consistent with large-offsets not clear from the sedimentary results of this study,
documented further west in the Cayman Trough but it is likely that both processes may have affected
(Rosencrantz et al., 1988) (Fig. 1). early basin formation and sedimentation given the
The time of 'docking' or present juxtaposition transition from arc to strike-slip tectonics at the time
of the La Toca and Altamira blocks is not well of deposition of most of the sediments. Based on
constrained. The La Toca and Altamira blocks were regional plate reconstructions and the presence of
possibly juxtaposed by Miocene time for two rea- the Eocene to Recent Cayman Trough west of the
sons: (1) similar detrital compositions of Lower area, strike-slip faulting may have played an impor-

tant role in the development of Upper Eocene source stones are found throughout the Cordillera Septentri-
areas and basins (Fig. 1). onal and Samana Peninsula (Fig. 6).
Eustatic controls on deposition in the El Interpretation of phase 3 tectonic and

Mamey Group sedimentary events
Coarse-grained, channel/overbank deposits in the De Zoeten and Mann (1991) divided the Neogene
Altamira Formation correlate with a global sea level uplift history of Hispaniola into two major events:
lowstand in the Late Oligocene (Haq et al., 1987) (1) Middle Miocene uplift event, and (2) a post-Early
and may represent a 'lowstand fan'. The introduction Pliocene uplift event. The Middle Miocene uplift is
of calciturbidites of the Las Lavas Formation into the recorded by a ubiquitous change from deep-marine
basin may indicate that the drop in sea level forced deposition (Las Lavas and La Toca formations) in
carbonate progradation on to the slope environment. Early Miocene time to shallow-marine deposition
An unstable slope environment would provide a (Villa Trina Formation) in Late Miocene and Early
source area for carbonate-rich gravity flows. An- Pliocene time. Post-Early Pliocene restraining bend
other possibility is that rising sea level, following tectonics continues to uplift the Cordillera Septentri-
the lowstand, produced a quiescent period inhibiting onal to its present position at 1250 m above sea level
siliciclastic transport into the basin and increasing (Fig. 6).
carbonate production on the shelf. Carbonate pro-
duced on the shelf could be mobilized by localized
tectonic activity and moved into the deeper basins. CONCLUSIONS
Although Oligocene sedimentation patterns appear
to coincide with a Late Oligocene fall in sea level, Detailed studies of several Paleocene through
tectonic influences on sedimentation may have also Miocene sedimentary formations exposed in the cen-
been important. Upper Oligocene olistostromes con- tral Cordillera Septentrional, Dominican Republic,
taining outsized platform carbonate blocks have been indicate at least three, distinct tectonic phases in the
documented in the Tavera Group (Palmer, 1979; Cenozoic evolution of the North America-Caribbean
Groetsch, 1983) and in the Ocoa Group (Heubeck et plate boundary zone. Each tectonic phase is charac-
al., 1990) south of the study area (Fig. 2). terized by deposition of characteristic sedimentary
facies and is punctuated by a short-lived folding
Phase 3: Late Miocene to Recent event. The three phases include:
(1) Paleocene to Early Eocene deposition of at
Tectonics least a 250-m-thick section of hemipelagic, fine-
Miocene strike-slip faulting and related deforma- grained turbidites (Los Hidalgos Formation) in-
tion is documented in several areas of northern His- terbedded with arc-related dikes and sills of inter-
paniola (de Zoeten and Mann, 1991). Gentle folding mediate composition (Palma Picada intrusive rocks).
of the siliciclastic units of the E1 Mamey Group, Sedimentation was terminated by a folding and up-
which is not recognized in the overlying shallow- lift event, which is thought to be related to early
water carbonate rocks, indicates Middle Miocene attempted subduction of the Bahama Platform be-
folding, probably associated with transpressional neath the Hispaniola arc (Fig. 20A).
strike-slip faulting. Calais et al. (1992) report the (2) Late Eocene to Early Miocene deposition of
same folding event affecting rocks of the western approximately 4000 m of deep-marine, siliciclastic
Cordillera Septentrional. turbidites (Altamira, Las Lavas, and La Toca forma-
During post-Early Pliocene time, northern His- tions) into at least two elongate basins subsequently
paniola underwent transpression associated with the juxtaposed by strike-slip faulting. We interpret sand-
strike-slip 'restraining bend' in the plate boundary stone compositions suggesting two distinct source
zone near Hispaniola (Mann et al., 1991). Oblique- areas during Late Eocene and Oligocene time as
slip movement on the Septentrional fault zone up- evidence that the basins were isolated from one an-
lifted Upper Miocene to Lower Pliocene limestones other and later juxtaposed by a 400-m-wide, linear,
to 1200 m above sea level (de Zoeten and Mann, strike-slip fault (Rfo Grande fault zone). Similari-
1991). ties in sandstone composition indicate that the two
basins were juxtaposed in Miocene time. Siliciclas-
Late Miocene to Pliocene rock record tic sedimentation was terminated by a folding and
Few Middle Miocene rocks are found in northern uplift event, which is thought to be associated with
Hispaniola (Figs. 4B, 7). This time period marks a transpressional strike-slip faulting related to North
change from Lower Miocene deep-marine sedimen- America-Caribbean plate motion (Fig. 20B).
tation to Upper Miocene shallow-marine carbonate (3) Late Miocene to Early Pliocene deposition
deposition. Upper Miocene to Lower Pliocene lime- of more than 250 m of shallow-marine limestones

C E N O Z O I C EL M A M E Y G R O U P OF N O R T H E R N H I S P A N I O L A 285
(Villa Trina Formation). Carbonate sedimentation Cook, H.E. and Mullins, H.T., 1983. Basin margin. In: EA.
was terminated by a folding and uplift event related Scholle, D.G. Bebout and C.H. Moore (Editors), Carbonate
to the current pattern of restraining bend tectonics Depositional Environments. Am. Assoc. Pet. Geol. Mem., 33:
539-618.
and active collisional underthrusting of the Bahama DeMets, C., Gordon, R.G., Argus, D.F. and Stein, S., 1990.
Platform. Maximum uplift of the limestone is asso- Current plate motions. Geophys. J. Int., 101: 425-478.
ciated with a large fold in the convex, uplifted side De Zoeten, R., 1988. Structure and Stratigraphy of the Central
of the restraining bend (Fig. 20C). Cordillera Septentrional, Dominican Republic. MA thesis,
University of Texas at Austin, 168 pp. (unpubl.).
De Zoeten, R. and Mann, E, 1991. Structural geology and Ceno-
zoic tectonic history of the central Cordillera Septentrional,
ACKNOWLEDGEMENTS Dominican Republic. In: E Mann, G. Draper and J. Lewis
(Editors), Geologic and Tectonic Development of the North
This work formed part of a master's thesis by America-Caribbean Plate Boundary Zone in Hispaniola. Geol.
R. de Zoeten that was supervised by E Mann, E. Soc. Am. Spec. Pap., 262: 265-279.
De Zoeten, R., Draper, G. and Mann, E, 1991. Geologic map
McBride and M. Cloos at the University of Texas
of the northern Dominican Republic (scale 1:150,000). In:
at Austin (de Zoeten, 1988). M. Perez, L. Pena, E Mann, G. Draper and J. Lewis (Editors), Geologic and
J. Guzman and G. Draper provided assistance in Tectonic Development of the North America-Caribbean Plate
the field and W. Eberle, E Cepek, S. Monechi, Boundary Zone in Hispaniola. Geol. Soc. Am. Spec. Pap.,
B. Redmond, and E. Robinson generously provided 262: Plate 1.
Dickinson, W.R., 1970. Interpreting detrital modes of greywacke
us with unpublished map and biostratigraphic data.
and arkose. J. Sediment. Petrol., 40: 695-707.
We thank J. Dolan, G. Draper, E. McBride, C. Dickinson, W.R. and Suczek, C.A., 1979. Plate tectonics
Heubeck, J. Lewis, M. Cloos and J. Pindell for and sandstone composition. Am. Assoc. Pet. Geol., 63:
useful discussions and J. Dolan, G. Draper, and 2164-2182.
E. Calais for their careful reviews of this paper. Dickinson, W.R., Beard, L.S., Brakenridge, G.R., Erjavec, J.L.,
Ferguson, R.C., Inman, K.F., Knepp, R.A., Lindberg, EA.
This work was supported by Grant 17068-AC2 from
and Ryberg, ET., 1983. Provenance of North American
the Donors of the Petroleum Research Fund of the Phanerozoic sandstones in relation to tectonic setting. Geol.
American Chemical Society to E Mann. University Soc. Am. Bull., 94: 222-235.
of Texas Institute for Geophysics contribution 1423. Dixon, T., Farina, E, DeMets, C., Jansma, E, Mann, E and
Calais, E., 1998. Relative motion between the Caribbean
and North American plates and related plate boundary zone
deformation based on a decade of GPS measurements. J.
REFERENCES Geophys. Res., 103: 15,157-15,182.
Dohm, C.F., 1943. Geologic map of and report on the western
Beall, R., 1943. Geologic map of the eastern portion of the portion of the Cibao Basin, Dominican Republic. Scale
Cibao Basin, Dominican Republic. Scale 1 : 100,000. Domini- 1 : 100,000. Dominican Seaboard Oil Co., New York Office,
can Seaboard Oil Co., New York Office, Santo Domingo Santo Domingo (unpubl.).
(unpubl.). Dolan, J.E, Mann, E, de Zoeten, R., Heubeck, C., Shiroma,
Bermudez, EJ., 1949. Tertiary smaller foraminifera of the Do- J. and Monechi, S., 1991. Sedimentologic, stratigraphic, and
minican Republic. Cushman Lab. Foraminiferal Res. Spec. tectonic synthesis of Eocene-Miocene sedimentary basins,
Publ. 25, 322 pp. Hispaniola and Puerto Rico. In: E Mann, G. Draper and
Bourgois, J., Vila, J.-M. and Tavares, I., 1982. Datos geol6gicos J. Lewis (Editors), Geologic and Tectonic Development
nuevos acerca de la regi6n de Puerto Plata (Republica Domini- of the North America-Caribbean Plate Boundary Zone in
cana). 9th Caribbean Geological Conference, Santo Domingo, Hispaniola. Geol. Soc. Am. Spec. Pap., 262:217-263.
Dominican Republic, Vol. 2, pp. 633-635. Dolan, J., Mullins, H. and Wald, D., 1998. Active tectonics of
Bourgois, J., Blondeau, A., Feinberg, H., Glaqon, G. and Vila, the north-central Caribbean region: oblique collision, strain
J., 1983. The northern Caribbean plate boundary in Hispan- partitioning and opposing slabs. In: J. Dolan and E Mann
iola: tectonics and stratigraphy of the Dominican Cordillera (Editors), Active Strike-Slip and Collisional Tectonics of the
Septentrional (Greater Antilles). Soc. G6ol. Fr. Bull., 25 (1): Northern Caribbean Plate Boundary Zone. Geol. Soc. Am.
83-89. Spec. Pap., 326: 1-61.
Bowin, C.O., 1975. The geology of Hispaniola. In: A.E.M. Draper, G. and Nagle, E, 1991. Geology, structure and
Nairn and EG. Stehli (Editors), The Gulf of Mexico and the tectonic development of the Rfo San Juan complex, northern
Caribbean. The Ocean Basins and Margins, Vol. 3, Plenum, Dominican Republic. In: E Mann, G. Draper and J. Lewis
New York, pp. 501-552. (Editors), Geologic and Tectonic Development of the North
Bowin, C.O. and Nagle, F., 1982. Igneous and metamorphic America-Caribbean Plate Boundary Zone in Hispaniola. Geol.
rocks of the northern Dominican Republic: an uplifted sub- Soc. Am. Spec. Pap., 262: 77-95.
duction zone complex. 9th Caribbean Geological Conference, Draper, G., Guti6rrez, G. and Lewis, J.F., 1996. Thrust emplace-
Santo Domingo, Dominican Republic, Vol. 1, pp. 39-50. ment of the Hispaniola peridotite belt: orogenic expression of
Calais, E. and Mercier de L6pinay, B., 1995. Strike-slip tectonic the mid-Cretaceous Caribbean arc polarity reversal? Geology,
processes in the northern Caribbean between Cuba and His- 24:1143-1146.
paniola (Windward Passage). Mar. Geophys. Res., 17: 63-95. Eberle, W., Hirdes, W., Muff, R. and Pelaez, M., 1982. The
Calais, E., Mercier de L6pinay, B., Saint-Marc, E, Butterlin, geology of the Cordillera Septentrional (Dominican Republic).
J. and Schaaf, A., 1992. La limite de plaques d6crochante 9th Caribbean Geological Conference, Santo Domingo,
nord cara'fbe en Hispaniola: 6volution pal6og6ographique et Dominican Republic, Vol. 2, pp. 619-632.
structural c6nozoYque. Bull. Geol. Soc. Fr., 163: 309-324. Graham, S.A., Ingersoll, R.V. and Dickinson, W.R., 1976.

286 R. DE Z O E T E N and E M A N N
Common provenance for lithic grains in Carboniferous intrusion at Palma Picada in the Cordillera Septentrional,
sandstones from Ouachita Mountains and Black Warrior Dominican Republic. Natural Resources and Development
Basin. J. Sediment. Petrol., 46: 620-632. (Ttibingen, West Germany), Vol. 26, pp. 83-94.
Grindlay, N., Mann, E and Dolan, J., 1997. Researchers investi- Mullins, H. and nine others, 1992. Carbonate platforms along
gate submarine faults north of Puerto Rico. EOS, 78: 404. the southeast Bahamas-Hispaniola collision zone. Mar. Geol.,
Groetsch, G.J., 1983. Resedimented Conglomerates and Tur- 105: 169-209.
bidites of the Represa and Janico Formations, North-Central Mutti, E. and Normark, W.R., 1987. Comparing examples of
Dominican Republic. M.S. Thesis, George Washington modern and ancient turbidite systems: problems and concepts.
University, Washington DC, 108 pp. In: J.K. Leggett and G.G. Zuffa (Editors), Marine Clastic
Haq, B.U., Hardenbol, J. and Vail, ER., 1987. Chronology Sedimentology: Concepts and Case Studies. Graham and
of fluctuating sea levels since the Triassic. Science, 235: Trotman, London, pp. 1-37.
1156-1167. Mutti, E. and Ricci Lucchi, E, 1978. Turbidites of the northern
Heubeck, C.E., Mann, E, Dolan, J. and Monechi, S., 1990. Apennines: introduction to facies analysis (translation by T.H.
Diachronous uplift and recycling of sedimentary basins during Nilsen). Int. Geol. Rev., 20: 125-166.
Cenozoic tectonic transpression, northeastern Caribbean plate Nagle, E, 1979. Geology of the Puerto Plata area, Dominican
margin. Sediment. Geol., 70: 1-32. Republic. In: B. Lidz and E Nagle (Editors), Hispaniola:
Ingersoll, R.V., Bullard, T.E, Ford, R.L., Grimm, J.E and Sares, Tectonic Focal Point of the Northern Caribbean, Three Tec-
S.W., 1984. The effect of grain size on detrital modes: a test tonic Studies in the Dominican Republic. Miami Geological
of the Gazzi-Dickinson point-counting method. J. Sediment. Society, Miami, FL, pp. 1-28.
Petrol., 54:103-116. Nilsen, T.H. and EL. Abbot, 1981. Paleogeography and
Iturralde-Vinent, M.A. and MacPhee, R.D.E., 1996. Age and sedimentology of Upper Cretaceous turbidites, San Diego,
paleogeographical origin of Dominican amber. Science, 273: California. Am. Assoc. Pet. Geol., 65 (7): 1256-1284.
1850-1852. Palmer, H.C., 1979. Geology of the Moncion-Jarabacoa area,
Joyce, J., 1991. Blueschist metamorphism and deformation on Dominican Republic. In: B. Lidz and E Nagle (Editors),
the Samana Peninsula: a record of subduction and collision Hispaniola: Tectonic Focal Point of the Northern Caribbean,
in the Greater Antilles. In: E Mann, G. Draper and J. Lewis Three Tectonic Studies in the Dominican Republic. Miami
(Editors), Geologic and Tectonic Development of the North Geological Society, Miami, FL, pp. 29-68.
America-Caribbean Plate Boundary Zone in Hispaniola. Geol. Peralta-Villar, J., 1985. Geologie und Erzffihrung in der Umge-
Soc. Am. Spec. Pap., 262: 47-76. bung des Intrusionsbrekzienkorpers von Los Jobos/Pedro
Kesler, S., Sutter, J., Barton, J. and Speck, R., 1991. Age Garcia. Dominikanische Republik. M.S. Thesis, Institut der
of intrusive rocks in Hispaniola. In: E Mann, G. Draper Universit~it Heidelberg, 111 pp.
and J. Lewis (Editors), Geologic and Tectonic Development Pickering, K., Stow, D., Watson, M. and Hiscott, R., 1986.
of the North America-Caribbean Plate Boundary Zone in Deep-water facies, processes and models: a review and
Hispaniola. Geol. Soc. Am. Spec. Pap., 262: 165-172. classification scheme for modern and ancient sediments.
Link, M.H., 1982. Slope and turbidite facies of the Miocene Earth-Sci. Rev., 23: 75-174.
Castaic Formation and the lower part of the Marple Canyon Pindell, J.L. and Draper, G., 1991. Stratigraphy and geological
Sandstone Member, Ridge Route Formation, Ridge Basin, history of the Puerto Plata area, northern Dominican Republic.
southern California. In: J.L. Crowell and M.H. Link (Editors), In: P. Mann, G. Draper and J. Lewis (Editors), Geologic and
Geologic History of Ridge Basin, Southern California. Pacific Tectonic Development of the North America-Caribbean Plate
Section of the Society of Economic Paleontologists and Boundary Zone in Hispaniola. Geol. Soc. Am. Spec. Pap.,
Mineralogists, Los Angeles, CA, pp. 79-88. 262:97-114.
Mann, E and Gordon, M., 1996. Tectonic uplift and exhumation Redmond, B.T., 1982. Sedimentary Processes and Products: an
of blueschist belts along transpressional strike-slip fault zones. Amber-Beating Turbidite Complex in the Northern Dominican
In: G. Bebout, D. Scholl, S. Kirby and J. Platt (Editors), Republic. Ph.D. Dissertation, Rensselaer Polytechnic Institute,
Dynamics of Subduction Zones. Am. Geophys. Union New York, 495 pp.
Monogr., 96: 143-154. Rosencrantz, E., Ross, M.I. and Sclater, J.G., 1988. Age and
Mann, E, Draper, G. and Lewis, J., 1991. An overview of the spreading history of the Cayman Trough as determined from
geologic and tectonic development of Hispaniola. In: E Mann, depth, heat flow, and magnetic anomalies. J. Geophys. Res.,
G. Draper and J. Lewis (Editors), Geologic and Tectonic De- 93: 2141-2157.
velopment of the North America-Caribbean Plate Boundary Schlager, W. and Chermak, A., 1979. Sediment facies of
Zone in Hispaniola. Geol. Soc. Am. Spec. Pap., 262: 1-28. platform-basin transition, Tongue of the Ocean, Bahamas. In:
Mann, E, Taylor, E, Edwards, L. and Ku, T., 1995. Actively L.L. Doyle and O.H. Pilkey (Editors), Geology of Continental
evolving microplate formation by oblique collision and Slopes. Soc. Econ. Paleontol. Mineral. Spec. Publ., 27:
sideways motion along strike-slip faults: an example from the 193-208.
northern Caribbean plate margin. Tectonophysics, 246: 1-69. Shanmugan, G. and Moiola, R.J., 1988. Submarine fans:
Mann, E, Prentice, C., Burr, G., Pena, L. and Taylor, E, characteristics, models, classification, and reservoir potential.
1998. Tectonic geomorphology and paleoseismology of the Earth-Sci. Rev., 24: 383-428.
Septentrional fault system, Dominican Republic. In: J. Dolan Sykes, L.R., McCann, W.R. and Kafka, A.L., 1982. Motion of
and E Mann (Editors), Active Strike-Slip and Collisional Caribbean plate during last 7 million years and implications
Tectonics of the Northern Caribbean Plate Boundary Zone. for earlier Cenozoic movements. J. Geophys. Res., 87:
Geol. Soc. Am. Spec. Pap., 326: 63-123. 10,656-10,676.
Middleton, G.V. and Hampton, M.A., 1976. Subaqueous Vaughan, T.W., Cooke, W., Condit, D.D., Ross, C.E, Woodring,
sediment transport and deposition by sediment gravity flows. W.E and Calkins, E C., 1921. A Geological Reconnaissance of
In: D.J. Stanley and D.J.E Swift (Editors), Marine Sediment the Dominican Republic. Gibson Brothers, Inc., Washington,
Transport and Environmental Management. Wiley, New York, 268 pp.
pp. 197-218. Walker, R.G., 1984. Turbidites and associated coarse clastic
Muff, R. and Hernandez, M., 1986. The hydrothermal alteration deposits. In: R.G. Walker (Editor), Facies Models (2nd Ed.).
and pyrite-galena-sphalerite mineralization of a porphyrite Geological Association of Canada, Toronto, ON, pp. 171-188.

Chapter 12
Tectonic and Eustatic Controls on Neogene Evaporitic and

Siliciclastic Deposition in the Enriquillo Basin,
Dominican Republic
PAUL MANN, PETER E MCLAUGHLIN, JR., W.A. VAN DEN BOLD, S.R. LAWRENCE,
and MICHAEL E. LAMAR
There are two distinct facies of Early Pliocene evaporites which formed in the center and edges of the tectonically active
Enriquillo basin of the southwestern Dominican Republic (island of Hispaniola). The basin-central deposit may be a late Neogene
analog to poorly understood ancient saline giants known in many other parts of the world. Both the basin-center and basin-edge
evaporite deposits are important for understanding the complex interplay between tectonic and eustatic effects on the stratigraphy
of this part of the North America-Caribbean oblique-slip plate boundary zone.
The basin-central evaporite deposit (Angostura Formation) is composed mainly of halite and is approximately 1500 m thick
in the Charco Largo-1 well in the center of the basin. A basin-margin evaporite composed mainly of gypsum interbedded in a
1-km-thick section of shallow-marine siliciclastic and carbonate rocks (Arroyo Blanco Formation) is exposed on the margins of
the basin and in the Charco Largo-1 well (Las Salinas Formation).
Analysis of the seismic reflection data, tied to the Charco Largo-1 and other exploration wells, shows that the deposition of
these Lower Pliocene evaporites occurred in a major sub-circular deep in the center of the valley. The location of this deep was
controlled by west-northwest to north-striking reverse faults and was separated from the Caribbean Sea to the east by a shallow
sill. Source areas for the interbedded siliciclastic rocks lay to the north and northeast in the topographically elevated central range
of Hispaniola (Cordillera Central). The depositional and climatic setting of the Enriquillo basin during the Early Pliocene was
probably similar to the modern Enriquillo Valley, which is 80 m below sea level (BSL) at its lowest point, separated by a shallow
sill from the Caribbean Sea, and receives siliciclastic sediment derived from the Cordillera Central by the Rfo Yaque del Sur.
Outcrop exposures of these evaporites were examined along the southern edge of the basin at Loma Sal y Yeso. The exposures
occur along a faulted diapir in which halite of the Angostura Formation is extruded as a narrow, 100-m-wide strip along a
reverse fault separating the Angostura and Las Salinas Formations. Primary textures in halite and gypsum exposed there are
overprinted by alteration, faulting and folding. Interbedded sedimentary rocks are fine-grained mudstone lacking sedimentary
structures. The monoclinal Las Salinas Formation overlies the more deformed rocks of the Loma Sal y Yeso and consists of a
lower shallow-marine siliciclastic interval, 180 m in thickness, which is overlain by a brackish interval, 1670 m in thickness,
containing both siliciclastic and carbonate rocks. The lower shallow-marine interval terminated deposition of the basin-central
halite of the Angostura Formation. Correlations between distinctive lithologies cropping out in the Angostura and Las Salinas
Formations can be made to the lithologies present in the Charco Largo-1 well. A notable difference is the lack of gypsum in the
well but its presence in the Loma Sal y Yeso.
The Early Pliocene northeast margin of the Enriquillo basin consists of an originally southwest-facing shelf-slope which
was tilted to the north and northwest 20-40 ~ in post-Early Pliocene times by tectonic folding. The tilted beds now form a
natural cross-section of the gently sloped, Early Pliocene margin. The sedimentary rocks of the shelf-slope can be divided into
a gypsum and oolite-bearing shallow-marine facies and a deeper-marine facies. Nine distinct lithologic groups are recognized,
including overlying non-marine rocks of the Late Pliocene-Pleistocene Arroyo Seco Formation. The primary textures of gypsum
in the Arroyo Blanco Formation suggests that these were deposited in several meters or less of water in a tidal-flat setting.
The nine lithologic groups can be interpreted as three shoaling-upwards cycles consisting of: (1) a basal deeper-marine facies
association characterized by basinward-prograding clinoforms composed of grainstones and coral debris; (2) a shallow marine
oolitic limestone facies association; and (3) an upper gypsum facies association. The shallow-marine gypsum and oolitic limestone
facies associations are wedge-shaped in cross section and occupy a position at the base of the slope. Interbedded marine units
onlap and truncate the underlying gypsum units.

288 E MANN et al.
The three shallow-marine facies associations of the basin-edge evaporite deposit are interpreted as having been deposited during
three eustatic falls in sea level during the Early Pliocene. These eustatic sea-level drops correlate reasonably well to the
tectonically active Cibao basin of the northern Dominican Republic. Correlation between the basin-margin and basin-central
evaporites in the Enriquillo basin is difficult because of uncertainty in the age of the units. Despite this correlation problem, we
propose that the Enriquillo evaporites formed in a shallow water-deep basin setting for three reasons: (1) the 'ramp basin' setting
of the basin would promote a 'deep' sub-sea level depression in a coastal setting; (2) shallow water within this 'deep' basin would
promote efficient evaporation of seawater spilled over the coastal sill; and (3) multiple spills during times of higher Pliocene sea
level would replenish brines to sustain the formation of thick basin central evaporites and explain the three cycles of basin margin
evaporites observed.
INTRODUCTION itic facies a desirable target for p e t r o l e u m explo-

ration.
Significance of evaporites for studies of Studies of the relationship b e t w e e n evaporite oc-
sedimentary basins currence and paleogeography, tectonic setting, eu-
static sea level, and hydrocarbons are best carried
The study of evaporites and their associated de- out in geologically young basins which have not
posits can provide important information on the been extensively modified by repeated tectonic de-
paleogeography, eustatic sea-level position and tec- formation or diagenesis related to either deformation
tonic setting of ancient sedimentary basins. More- or subaerial exposure. The Enriquillo basin of the
over, several authors have pointed out the close southern D o m i n i c a n Republic on the island of His-
relationship b e t w e e n hydrocarbon occurrence and paniola (Fig. 1) is an ideal place to study the controls
the presence of evaporites in ancient sedimentary on evaporite formation for the following reasons.
basins (for example, Kirkland and Evans, 1981; (1) The basin contains a thick (up to 4.5 km) suc-
Friedman, 1982; Warren, 1986). Evaporites can cession of mainly shallow-marine siliciclastic and
serve as seals and interbedded anoxic sedimen- evaporitic rocks deposited since the Late M i o c e n e
tary rocks can serve as hydrocarbon source rocks. (McLaughlin et al., 1991). The rocks are slightly
The potential for seals and sources within com- d e f o r m e d by gentle folds and high-angle faults along
m o n l y thick stratigraphic sections makes evapor- the edges of the basin that were produced in an in-
Fig. 1. Present-day plate tectonics of the Caribbean region. Direction and rates of plate motion relative to the Caribbean plate are from
DeMets et al. (1990) and Dixon et al. (1980). The island of Hispaniola straddles the active left-lateral strike-slip zone separating the North
America and Caribbean plates. The large amount of plate convergence and topographic uplift of Hispaniola is related to transpression
between two thick crustal blocks: the Bahamas carbonate platform to the north and the Cretaceous Caribbean oceanic plateau to the
south. Box shows map area shown in Fig. 3.

NEOGENE EVAPORITIC AND SILICICLASTIC DEPOSITION IN THE ENRIQUILLO BASIN 289
terplate, transpressional setting (Mann et al., 1991b,

1995) (Fig. 1).
(2) The basin contains both a massive basin-
central halite and gypsum deposit up to 1500 m
thick and three intervals of much thinner-bedded,
coeval basin-margin gypsum deposits which occur
in a siliciclastic section about 1 km in thickness.
The basin-central deposit is known from wells in the
basin center and from outcrops in a diapir (Loma Sal
y Yeso) along the southern margin of the basin (Lli-
nas, 1972a,b). One deep exploration well, Charco
Largo-1, penetrates the entire undeformed thickness
of the basin-central evaporite deposit and provides
an opportunity to compare the geologic history of the
basin center with the history of the more deformed
basin edges known from outcrop mapping.
(3) Sedimentary textures in the basin-central
evaporite deposit are recrystallized, but textures in
the basin-margin evaporites are primary and closely
resemble textures described in other, better-studied
areas like the Neogene of the western Mediterranean
(Schreiber, 1988).
(4) The present-day tectonic, physiographic, and
climatic setting of the Enriquillo basin (Fig. 1) is
similar to classic conceptions of silled, coastal evap-
orite basins described in other modern settings like
Lake MacLeod, Australia (Logan, 1987) and postu-
lated in ancient settings including the Middle De-
vonian Elk Point basin of western Canada (Kendall,
1989), the Early Cretaceous South Atlantic (Burke
and Seng6r, 1988), the Late Jurassic Louann salt of
the Gulf of Mexico (Winker and Buffler, 1988), the Fig. 2. Three models for the deposition of massive evaporite de-
Permian Delaware basin of west Texas (Lowenstein, posits like the one present in the Enriquillo basin (modified from
1988), and the Late Miocene Mediterranean Sea Kendall, 1984). Because it is presently about 50 m below sea
level, the Enriquillo basin is a likely shallow-water, deep-basin
(Hsii et al., 1973; Schreiber, 1988). Although mas- setting for evaporites.
sive basin-central evaporites have formed in all of
these settings, a modem or late Neogene analog for
these ancient 'saline giants' has never been identified OBJECTIVES AND METHODS OF THIS STUDY
with certainty and remains a major paradox in the
study of evaporite deposits. Kendall (1984) proposed This study follows on several previously pub-
three, idealized models for silled basins which he lished studies by the authors and their colleagues on
proposed as possible depositional settings of sub- the surface and subsurface geology of the Enriquillo
aqueous evaporites found in 'saline giants' (Fig. 2). basin (Norconsult, 1983; Mann et al., 1984; Taylor
(5) Rates of plate movement between the et al., 1985; Mann et al., 1991a,c; McLaughlin et al.,
Caribbean and North America plates in Hispaniola 1991; Fig. 3A). This study builds on these previous
are relatively slow (--~2 cm/year, Fig. 1) and it is likely surface studies by attempting the following.
that the physiography of the Enriquillo basin has not (1) Integrate subsurface seismic reflection data
changed significantly since the formation of the basin collected by industry in the center of the Enriquillo
about 5 million years ago (Mann et al., 1995; Dixon et basin (Norconsult, 1983) with surface mapping and
al., 1998). A recent cycle of marine flooding and des- section measuring of the age-equivalent sections ex-
iccation, similar to that postulated for the origin of the posed by Neogene deformation at the basin margins.
Pliocene evaporite deposits, has occurred as recently (2) Integrate the stratigraphy of the Charco
as 4000 years BP (Taylor et al., 1985). Largo-1 well in the center of the basin with sur-
(6) There are reports of oil and gas seeps from rounding seismic reflection data and surface out-
the margins of the basin (Guerra Pena, 1956; Llinas, crops. Use the biostratigraphy of the well to resolve
1972a,b). The source, reservoir and relation of these correlation and age estimates of evaporitic and silici-
evaporties is unknown but deserves further study. clastic sedimentation in the center of the basin.

290 E M A N N et al.
9 ,.
~~o~
~~~
o~
,s::
o
9 O
~J

(3) Establish the thickness, paleobathymetry, and major, offshore thrust front subparallel to the north-
facies of siliciclastic sedimentary rocks that enclose ern coast of Hispaniola (Mann et al., 1995; Dolan
the evaporite deposits in the center of the basin and et al., 1998) (Fig. 3A). The relation between the
at its edges. Can these data be used to distinguish general shape of the unsubducted Bahamas Platform
between the three types of silled basins proposed and the thrust front north of Hispaniola suggests
by Kendall (1984) (Fig. 2) which may have been that as much as half of the island of Hispaniola
the depositional setting for the 'saline giant' found is underlain by the subducted Bahamas Platform.
today in the center of the Enriquillo basin? Local convergence in Hispaniola with present topo-
(4) Establish the textural variations in evaporite graphic elevations up to 3 km may also be related
deposits in various parts of the valley and relate to the location of the island between the Bahamas
these regional variations to depositional environment carbonate platform to the north and thicker-than-
and water depth within the Enriquillo basin (for average oceanic plateau seafloor of the Caribbean
example, Schreiber and Friedman, 1976; Schreiber, Sea to the south (Fig. 3). Mann et al. (1995) propose
1988). These data, in conjunction with paleobathy- that the Enriquillo-Plantain Garden strike-slip fault
metric estimates derived from benthic foraminifera zone formed in late Neogene time as a response to
preserved in siliciclastic rocks enclosing the evapor- oblique subduction of the Bahamas Platform because
ites, can also be used to evaluated the silled basin the subduction of the thicker-than-average crust of
models shown in Fig. 2. the Bahamas Platform resisted the eastward motion
(5) Identify stratigraphic sequences and sequence of the central and northern parts of Hispaniola.
boundaries and to relate these to changes in Miocene A regional, unbalanced cross-section modified
and Pliocene sea level. Traditionally, evaporites are from Mann et al. (1991b) and shown in Fig. 3B
interpreted as forming during times of major re- illustrates several important features of the Cenozoic
gression. However, in a sub-sea-level depression set- structural history of Hispaniola. (a) The most promi-
ting like the modern Enriquillo Valley and as shown nent folding and thrusting event in Hispaniola is Late
schematically in Fig. 2 (bottom panel), marine trans- Miocene and younger in age and verges southward
gression would produce a large area of seawater sub- to southwestward. (b) South to southwest vergence
ject to evaporation in a dry, basinal setting. Do evap- of the central mountain range of the island, the
orites in these settings mark sea-level highs and spills Cordillera Central, is reflected in its slightly asym-
into such depressions or do they mark regressions? metric topographic profile with straighter and steeper
(6) Compare sedimentation in the Enriquillo basin slopes along its southwestern flank. (c) Late Miocene
to three classic models of silled evaporite basins and younger reverse and oblique-slip faulting is re-
proposed by Kendall (1984) (Fig. 2). Does this sponsible for the present pattern of morphotectonic
comparison suggest that the Enriquillo basin could units in central Hispaniola, including the distribution
be considered a late Neogene analog to massive of the three major ramp, or thrust-bound, basins
basin-central evaporite deposits known from the Gulf the Cibao, San Juan-Azua, and Enriquillo (Mann
of Mexico, Permian basin, and other well-studied et al., 1991a). (d) Cretaceous-Eocene island-arc
localities? terranes of the northern and central part of the is-
land are topographically high-standing and deeply
eroded; the Cretaceous oceanic plateau terrane of the
TECTONIC AND GEOLOGIC SETTING OF THE southern part of the island is relatively low-stand-
ENRIQUILLO BASIN ing and less deeply eroded. The lower altitude of
the oceanic plateau in the south may also reflect
Tectonic setting its foot-wall position relative to the higher-stand-
ing hanging-wall block represented by the island-arc
The Enriquillo basin of the southern Dominican terranes in the north (Fig. 3B).
Republic is a thick (~5 kin), Plio-Pleistocene basin
containing both massive halite and gypsum deposits Geologic setting
and thin-bedded unrecrystallized gypsum deposits
(McLaughlin et al., 1991) (Fig. 3). The basin forms Present-day basin physiography
an elongate valley bounded by active strike-slip Most of the 2000 km 2 floor of the Enriquillo
and reverse faults related to left-lateral displace- basin and its westward extension into Haiti (Cul-de-
ment between the North America and Caribbean Sac basin) falls within rain shadows of bordering,
plates across Hispaniola (Fig. 3B) and trends ap- fault-bounded mountain ranges up to 2 km high in
proximately east-west across the southern part of contrast to the semiarid valley that receives only
the island (Mann et al., 199 l a). Transpression in the ~60 cm of rainfall per year, with most falling in
Enriquillo basin appears to be a response to oblique the month of June (Garcia, 1976) (Fig. 4). Most
collision of the Bahamas carbonate platform at a of the moisture transported off the Atlantic Ocean

292 E MANN et al.
Fig. 4. Amount of annual rainfall based on rainfall gauges around the Dominican Republic (modified from Garcfa, 1976). Areas of high
rainfall correspond to mountain ranges while areas of low rainfall correspond to rain shadows in semiarid valleys separating mountain
ranges. Note the correlation between Pliocene evaporite occurrence and the belt of low rainfall in the Enriquillo Valley.
by northeasterly trade winds falls as rain on top of deposits appear as a large light-colored delta-shaped
the Cordillera Central north of the Enriquillo basin area at the eastern end of the valley.
(Fig. 4). This close association of high topography, As sea level rose about 10,000 years ago, the
rain shadows, and semiarid intermontane valleys is Enriquillo Valley was flooded as the Caribbean Sea
the reason why the eastern Enriquillo Valley is the at Bahia de Neiba breached the sill at the eastern
site of the only Neogene occurrence of evaporites in end of the valley (Mann et al., 1984; Taylor et
the northern Caribbean. al., 1985; Mann et al., 1995). A pristine coral
The 262 k m 2 Lago Enriquillo in the Dominican reef, now subaerially exposed in the area around
Republic and the 120 km 2 Etang Saumatre in Haiti Lago Enriquillo (Fig. 5), records marine conditions
occupy closed drainage basins in the central part of from about 10,000 to 5000 years BE From the
the valley and are separated by a sill a few hundred period from about 5000 to 2800 years BE conditions
meters wide at the international border (Fig. 5). The became increasingly brackish as the eastern end of
sill between the two lakes lies between 20 and 40 m the valley became blocked by deltaic deposits of the
above sea level (ASL). All of the Haitian Cul-de-Sac Rio Yaque del Sur (Taylor et al., 1985) (Fig. 5A).
basin is above sea level with the surface of Etang Evaporation of the seawater in the semiarid valley
Saumatre at an elevation of about 14 m ASL. About produced Lago Enriquillo which presently has about
half of the Enriquillo Valley is below sea level twice the salinity as seawater. Ostracodes from short
(BSL) with the surface of Lago Enriquillo at about cores (= 1.15 m) from the lake bed reflect the
- 4 2 m BSL. The northern, deeper part of the Lago present period of hypersalinity of the lake waters
Enriquillo has a maximum depth of - 4 0 m and (Bold, 1990). There are no modem evaporites in or
forms the deepest known part of the valley at about around Lago Enriquillo.
- 8 0 m BSL (Taylor et al., 1985). Lago Enriquillo is Mann et al. (1984) speculated that Pliocene sea-
separated from the Caribbean Sea at Bahia de Neiba level incursions, similar to the Holocene transgres-
at the eastern end of the valley by a sill about 4 m sion into the Enriquillo Valley, may have resulted in
ASL (Fig. 5). This relief results mainly from flood the deposition of evaporites reported from field stud-
plain deposition near the mouth of the Rio Yaque del ies in the eastern Enriquillo Valley (Llinas, 1972a,b;
Sur, a major river draining into the Bahia de Neiba. Bold, 1975) and from deep wells in the center of the
In the LANDSAT image of Fig. 5A, these floodplain basin (Bowin, 1975; de Leon, 1983).

N E O G E N E E V A P O R I T I C A N D S I L I C I C L A S T I C D E P O S I T I O N IN T H E E N R I Q U I L L O B A S I N 293
Fig. 5. (A) LANDSAT image of southern Hispaniola (Haiti and Dominican Republic). (B) Interpretation of LANDSAT image modified
from Mann et al. (1995). Key to numbered localities and tectonic features" 1 = Cibao Valley (Dom. Rep.); 2 -- Bonao fault zone; 3
= Cordillera Central (Dom. Rep.) vertical hatching indicates areas over 2 km in topographic elevation; 4 -- Los Pozos-San Juan
fault zone; 5 --- Massif du Nord (Haiti); 6 = Plateau Central (Haiti)" 7 = San Juan Valley (Dom. Rep.); 8 = Azua Valley (Dom. Rep.)"
9 = Muertos trench; 10 -- Beata fault zone; 11 -- Sierra Martfn Garcfa (Dom. Rep.); 12 = late Holocene delta of the Rfo Yaque
(Dom. Rep.); 13 -- Enriquillo Valley (Dora. Rep.); 14 = southern Sierra de Neiba (Dom. Rep.); 15 = northern Sierra de Neiba; 16
-- Montaignes Noires-Chaine de Marmelade (Haiti); 17 = late Holocene delta of Rfo Artibonite (Haiti)" 18 = northern Chaine des
Matheux; 19 = southern Chaine des Matheux; 20 = Cul-de-Sac Valley (Haiti); 21 -- Gulf of Gonave and offshore Rfo Artibonite delta;
22 -- Gonave Island; 23 = Enriquillo-Plantain Garden fault zone; 24 = Sierra de Bahoruco (Dom. Rep.) and Massif de la Selle (Haiti);
25 = Mirogoane Lakes (late Quaternary pull-apart basin); 26 = eastern Massif de la Hotte. (C) Interpretation of proposed boundaries for
the Gonave microplate based on major fault features seen in (A) and (B).

294 R M A N N et al.
~D
9
E
,s=
,s=
~
~0
9
9
~D
,s=
~0
.=.
,..Z,
,-4
,SD
,.~ - ~
~-~ c~
0 ,.~
~'~
~ ~
.,..~
~.~
~
~E
~.~

Main lithologic units of the Enriquillo basin CENTRAL NE MARGIN

Previous workers including McLaughlin et al. ENRIQUILLO ENRIQUILLO
BASIN BASIN
(1991) have mapped several major lithologic units
in south-central Hispaniola that range in age from
Late Cretaceous to Quaternary (Fig. 6). These units JIMANI FM.
PLEISTOCENE (210 M) ARROYO
are shown on the geologic compilation map in SECO FM.
Fig. 6 that has been simplified from a much more (~ 500 M)
detailed l:150,000-scale geologic map compiled
by Mann et al. (1991c). The present physiographic
Enriquillo basin or valley is continuous with the
Azua basin to the east and the San Juan basin to
the north (Fig. 5). Many of the same lithologic units LAS SALINAS FM.
are present in all three basins although thicknesses (2182 M) ARROYO
and ages vary between the central Enriquillo basin BLANCO FM.
PLIOCENE (~ 830 M)
and the northeastern margin of the Enriquillo basin
(McLaughlin et al., 1991) (Fig. 6).
Igneous basement
ANGOSTURA FM. QUITA CORAZA FM
The lowest unit consists of Cretaceous to Eocene (1562 M) (2oo- 700 M)
igneous and volcanisiliciclastic rocks that crop out
to the south of the Enriquillo basin in the Sierra de
Bahoruco and to the north in the Sierra de Neiba
TRINCHERA FM.
(Fig. 6). The rocks in the Sierra de Bahoruco have TRINCHERA FM. (~ 2500 M)
been correlated with the Caribbean oceanic plateau (271 M)
province that underlies much of the areas of the
Caribbean Sea (Diebold and Driscoll, Chapter 19) MIOCENE
(Fig. 1) while the rocks to the north have more GAJO LARGO
arc-like affinities and appear related to the Caribbean MB. (~ 200 M)
SOMBRERITO FM.
arc of Hispaniola (Mann et al., 1991b). Deep wells in (322 M) SOMBRERITO FM.
the Enriquillo basin have not penetrated crystalline (~ 2500 M)
basement so it is not known whether the basin is
Fig. 7. Correlation, ages and stratigraphic thicknesses proposed
underlain by the oceanic plateau crust, arc crust, or by McLaughlin et al. (1991) and McLaughlin and Sen Gupta
both. In the regional cross-section of Fig. 3B, the (1994) and used in this paper for Neogene formations of the
lower-standing oceanic plateau basement is inferred Enriquillo basin.
to extend northward as part of an underthrusting
foot-wall block beneath northeast-dipping reverse
faults along the southern flanks of the Sierra de the youngest, the Sombrerito Formation of Early
Neiba and Cordillera Central. Miocene to latest-Middle Miocene age (Fig. 7). This
A fold-thrust belt of Paleocene-Eocene deep- unit, consisting of about 500 m of mainly micrite,
marine rocks called the Peralta belt is found north- was deposited in 500 to 1500 m of water in a basinal
east and north of the Enriquillo basin (Fig. 6). Dolan setting (McLaughlin et al., 1991). Deep wells in
et al. (1991) and Heubeck et al. (1991) interpret the Enriquillo basin have also penetrated rocks of
these rocks as back-arc basin deposits that were de- the Sombrerito Formation, so it appears to underlie
formed during the late Neogene closure of the basin most, if not all, of the Enriquillo basin.
(Fig. 3B).
Neogene rocks
Paleogene rocks Neogene siliciclastic rocks of the Enriquillo basin
The most extensive of the lithologic units of the are divisible into three main units (Fig. 7). A
Enriquillo basin shown on the map in Fig. 6 con- thick (~2 km), latest-Middle Miocene to earliest
sists of Paleocene-Middle Miocene carbonate rocks. Pliocene section of deep-marine siliciclastic tur-
These rocks crop out in the cores of the Sierra de bidites (Trinchera Formation) is exposed on the
Bahoruco, the Sierra de Neiba, and the interven- northern flank of the Sierra Martfn Garcfa at the
ing Sierra Martfn Garcfa (Fig. 6). The Paleogene eastern end of the Enriquillo basin (Fig. 6). This unit
carbonate units have not been studied in detail but also crops out on the southern and northern flanks
consist of limestone and local dolomite indicative of the Sierra de Neiba and has been encountered
of shallower-water facies (van den Berghe, 1983). in most of the wells of the eastern Enriquillo Val-
The best-studied carbonate unit in the area is also ley. Previous studies by McLaughlin (1989, 1991)

296 R MANN et al.
have shown that this unit formed a large, time- between the northern and southern outcrop areas dis-
transgressive siliciclastic wedge produced during the cussed in this paper (Fig. 6). Llinas (1972a,b), Bowin
Late Miocene-Early Pliocene uplift and erosion of (1975) and de Leon (1983) present a summary of the
the Cordillera Central north of the Enriquillo basin logs of the Mella 1 and 2 wells, which were drilled
(Fig. 3). The present outcrop area of the unit on the in the early 1970s. These wells bottomed at depths
northern flank of the Sierra Martfn Garcfa appears of 2673 m and 3328 m, respectively, but did not
to have been the main depocenter of the siliciclastic penetrate the entire thickness of the massive salt and
wedge as thicknesses of the unit drop off dramati- gypsum. De Leon (1983) presents a summary log of
cally to the west in the Enriquillo Valley and to the the Charco Largo-1 well, which was drilled by Su-
east in the Azua basin. perior Oil Company (now Mobil) in 1980 and pen-
Early Pliocene to Quaternary shallow-marine and etrated a 1966-m-thick section of massive salt and
fluvial siliciclastic and evaporitic rocks conformably gypsum, and bottomed at a total depth of 4830 m.
overlie the deep-marine Trinchera Formation. These Prior to drilling the well, Superior Oil Company ac-
rocks crop out most extensively along the north flank quired a dense grid of multichannel seismic profiles
of the Sierra Martin Garcfa (Arroyo Blanco Forma- over the central part of the valley (Fig. 6, inset map).
tion), the north flank of the Sierra de Bahoruco
(Las Salinas Formation), and in the Azua and San
Juan basins (Arroyo Blanco Formation) (Fig. 6; for METHODS
a detailed review of all stratigraphic nomenclature,
see Mann et al., 1991a). All wells drilled in the En- Field-based mapping
riquillo and Azua area have penetrated the equivalent
Arroyo B lanco and Las Salinas Formations which Evaporites of Loma Sal y Yeso were mapped at
range in thickness from about 1 to 2 km. a scale of 1:25,000 using 2x enlargements of U.S.
The Early Pliocene to Quaternary unit includes Defense Mapping Agency 1:50,000 topographic
both outcrop areas of Pliocene evaporites in the En- maps available through the Dominican Cartographic
riquillo Valley: a southern area in the Loma Sal y Institute in Santo Domingo. Aerial photographs at
Yeso ('Hill of Salt and Gypsum') along the southern a scale of 1:40,000 are used in this paper and by
edge of the basin, where it is known as the Las Sali- Mann et al. (1991a) to interpret the along-strike
nas Formation, and a northern area near the eastern continuity of strike ridges for sequence stratigraphic
edge of the basin where it is known as the Arroyo interpretation and to identify major faults. Detailed
Blanco Formation (Fig. 7). The Loma Sal y Yeso is field mapping focused on the area of an open-pit salt
an elongate anticline bounded by thrust faults separat- mine on the north-central flank of the Loma Sal y
ing the Sierra de Bahoruco from the Enriquillo basin Yeso and the active open-pit gypsum mines oper-
(Fig. 3B). Llinas (1972a,b) interpreted the massive ated by the Cooperaci6n Dominicana de Empresas
and bedded salt and gypsum along the axis of the fold Estatales (CORDE) of the Dominican government
as a diapir mobilized during folding and classified on the eastern and southeastern flanks of the hill
the structure as a 'diapiric anticline'. In more modem near the town of Las Salinas. We measured one
terminology, the anticline probably represents a hang- long (1850 m) section of siliciclastic rocks north of
ing-wall anticline developed along the frontal thrust the Loma Sal y Yeso (Arroyo del Pozo), which dip
of the Sierra de Bahoruco (Vann et al., 1986). northwards into the Enriquillo basin and presumably
The northern area of basin-edge evaporites in- intersect the wells in the basin center at depth.
terbedded with siliciclastic rocks (Arroyo Blanco We carried out a similar style of mapping of the
Formation) lies about 30 km to the northeast where evaporites to the west of the Rio Yaque del Sur. Be-
they crop out along the eastern bank of the Rio cause of the exceptional natural exposures of later-
Yaque del Sur. Evaporites consist of beds of gypsum ally continuous strike ridges of gypsum, we made a
0.1 to 10 m thick and are interbedded in a siliciclas- special effort to identify and measure the thickness
tic sequence up to 1 km in thickness. The section of sedimentary sequences using a Brunton compass
strikes east to northeast, has an average dip of 20 ~, mounted on a Jacob's staff. We measured five sec-
and is well exposed in about a dozen stream sections tions which ranged from 140 m to 930 m in thickness.
draining eastward into the Rio Yaque del Sur. The
exceptionally continuous exposure of the resistant Subsurface mapping
beds of gypsum, conglomerate, and sandstone visi-
ble on aerial photographs is attributed to planation In order to attempt correlations between the basin-
by the Rio Yaque del Sur prior to the incision of the edge exposures of evaporites at Loma Sal y Yeso and
present-day modem fiver channel. along the western side of Rio Yaque del Sur, we in-
Evaporites are reported from three deep wells in tegrated the results of subsurface mapping done by
the center of the Enriquillo basin about equidistant S. Lawrence (Norconsult, 1983) using seismic lines

originally collected by Superior Oil Company in the reflectors that are correlated to certain horizons at the
late 1970s and now owned by Mobil. Mobil released base of or within sedimentary formations described
the use of these lines to Mann for use in this study. from wells and surface mapping (Fig. 8). T i m e -
structure maps were then made in two-way time
Analysis of Charco Largo-1 well cuttings to these horizons. We illustrate these key reflectors
using some of the longer lines in the seismic data
Mobil also made available the well report on the set (locations of these lines shown on Fig. 8). Key
Charco Largo-1 well and released cuttings from the seismic reflectors in the Enriquillo basin seen on
well to E McLaughlin and W. Bold who made a com- selected lines shown on the inset map on Fig. 6
plete analysis of the age, facies and paleobathymetry include the following units.
of the well using planktonic foraminifera and ostra-
codes. These results are discussed with previously
published biostratigraphic results from outcrop sam- Top of the Neiba-Plaisance Formation
ples (Bold, 1975; McLaughlin et al., 1991) and shal- This unit corresponds to the P a l e o c e n e - M i d d l e
low core samples from Lago Enriquillo (Bold, 1990). Miocene carbonate section shown on the regional
map in Fig. 6. This contact has not been mapped
in detail in the field so the origin of the seismic
GEOLOGY OF THE BASIN CENTER INFERRED FROM reflector remains unknown. This reflector is not
INTERPRETATION OF MULTI-CHANNEL SEISMIC present on all lines in the study area and for that
LINES TIED TO WELLS reason no time map was made to the base or top of
this unit. The reflector is present on regional line 111
Key reflectors used in mapping and their through the area of the Charco Largo-1 and Mella
correlation to outcrops wells (Fig. 9) as well as on lines 1410 and 410 in the
Lago Enriquillo area (Fig. 10). On these lines, the
Subsurface analysis of seismic data is based on top Neiba-Plaisance reflector is conformable with
the basin-wide recognition of several major seismic the overlying top Sombrerito reflector.
Fig. 8. Time structural map (contour interval 50 ms) of the reflector interpreted as the top of the earliest-Late Miocene Sombrerito
Formation. Complete track map of lines used to map this reflector is shown in Fig. 6. Major faults subdivide the subsurface of the basin
into several major structural blocks. The two central blocks have undergone the greatest amount of subsidence in post-Miocene time
because they are being dynamically depressed by thrust or strike-slip faults at their margins. Much of the present-day surface of the Lago
Enriquillo block is about 50 m below sea level and is isolated from the Caribbean Sea by a low sill about 10 m above sea level at the
Bahfa de Neiba and Mella blocks. The lined pattern at the top of the map indicates the zone of recent left-lateral shearing along the
Enriquillo-Plantain Garden fault zone of Mann et al. (1995). The inset box shows the GPS-derived rate and direction of Caribbean plate
motion relative to a fixed point on the North America plate (Dixon et al., 1998).

298 P. MANN et al.
Fig. 9. Regional migrated multi-channel seismic line 111 from the central part of the Enriquillo basin (location on Fig. 8). The line
crosses the deepest three blocks of the central part of the basin (Angostura, Mella, and Vicente Noble). Prominent reflectors are indicated
in the left margin. Mapping of the upper four of these reflectors was used to make the four time structural maps presented in this paper.
Fig. 10. Regional migrated multi-channel seismic line 1410 (to left) and 410 (to right) from the central part of the Enriquillo basin
(location on Fig. 9). The line crosses the Lago Enriquillo block and the edge of the Mella block. Prominent reflectors are indicated in the
center margin. Mapping of the upper four of these reflectors was used to make the four time structural maps presented in this paper. The
lenticular seismic anomaly is interpreted as a reefal mound at the top of the Sombrerito Formation.
Top of the Sombrerito Formation rating the m a i n l y carbonate rocks of the Sombrerito
This horizon forms a p r o m i n e n t reflection sur- F o r m a t i o n from the overlying siliciclastic rocks of
face over m o s t of the basin and offshore in Bahfa the Trinchera F o r m a t i o n (Fig. 7). The horizon is
de N e i b a and has b e e n penetrated by all the wells present on regional line 111 through the area of the
shown on Fig. 8. The p r o m i n e n c e of the horizon Charco Largo-1 and M e l l a wells (Fig. 9), on lines
m a y be related to the m a j o r velocity change sepa- 1410 and 410 in the L a g o Enriquillo area (Fig. 10),

Fig. l l. Regional migrated multi-channel seismic line 133 across the westernmost part of the Enriquillo basin (location on Fig. 8).
The line crosses the Lago Enriquillo block and the left-lateral Enriquillo-Plantain Garden fault zone from north to south (see Mann
et al., 1995, for three other seismic lines crossing the EPGFZ to the east of Isla Cabritos). The fault is inferred to be the prominent
down-to-the-south fault adjacent to the Cabritos-1 well. Two prominent reflectors are indicated.
and on line 133 across the Cabritos anticline in Lago Sombrerito Formation from the overlying evaporitic
Enriquillo (Fig. 11). On these lines, the top Som- rocks of the Angostura Formation (Fig. 7).
brerito reflector is conformable with the overlying
base evaporites reflector but is disconformable with Mid-Las Salinas Formation
the overlying top evaporites reflector, probably as a This horizon forms a prominent reflection surface
result of deformation of the evaporite unit. In the over most of the basin and offshore in Bahia de
Lago Enriquillo area, the top of the Sombrerito For- Neiba and has been penetrated by all the wells shown
mation is marked by a lenticular seismic anomaly on Fig. 8. Although this horizon has been penetrated
interpreted as a reefal mound (Fig. 10). in all of the wells, its lithology is not clear since
there are several abrupt lithologic changes between
Base of evaporites (Angostura Formation) isolated beds of limestone and evaporites in the
This reflector is present only in the area of the middle part of the Las Salinas Formation both in the
basin-central evaporite deposit between the eastern Charco Largo-1 well and in the Loma Sal y Yeso
edge of Lago Enriquillo and Laguna Rinc6n. The section (Table 1). We speculate that the Mid-Las
reflector is present on regional line 111 through Salinas reflector may correspond to a 3-m-thick
the area of the Charco Largo-1 and Mella wells limestone bed penetrated at a depth of 7075 ft (2156
(Fig. 9) and on line 410 in the area east of Lago En- m) in the Charco Largo-1 well (Table 1). This is
riquillo (Fig. 10). The horizon was penetrated in the the highest occurrence of limestone in the well and
Charco Largo-1 and Mella wells. The prominence suggests that this might have been the last marine
of the horizon may be related to the major velocity influence until the Holocene marine incursion about
change separating the mainly carbonate rocks of the 10 ka (Taylor et al., 1985). The horizon is present

300 R MANN et al.
Table 1
Correlation between Angostura and Las Salinas Formations exposed in Loma Sal y Yeso and in the Charco Largo-1 well (see Fig. 8 for
location)
Loma Sal y Yeso Charco Largo-1 (Fig. 15)
Angostura Formation (Fig. 19)

Unit 1 - halite (200 m-minimum thickness) interval 8700-13,750 ft or 2652-4191 m (1539 m of halite and
shale; shallow-marine fossils present at 8390 ft or 2557 m)
Unit 2 - gypsum (350 m-minimum thickness) not present in thick sections
Unit 3 - gypsum and massive gray mudstone (50 m) interval 8625-8700 ft or 2628-2651 m (23 m of shale)
Las Salinas Formation in Arroyo el Pozo (Fig. 21)

Unit 4 - shallow-marine limestone of 'Razorback Ridge' (3 m) interval 8610-8625 ft or 2625-2629 m (4.6 m of dense white
limestone)
Unit 5 - ripple-marked sandstone and interbedded siltstone interval 8510-8610 ft or 2594-2624 m (30 m of shale capped
(150 m) by a 6-m-thick bed of anhydrite)
Unit 6 - trough-cross bedded sandstone and interbedded interval 7075-8510 ft or 2156-2484 m (437 m of 25-100 ft or
siltstone (390 m) 8-30-m-thick beds of sandstone interbedded in shale)
Unit 7 - coquina beds interbedded with siltstone (95 m) 10 ft or 3-m-thick limestone bed at 7075 ft or 2156 m (?)
Unit 8 - growth position coral reefs, reworked coral debris 7075 ft or 2156 m (?) to (?); no coral or limestone
interbedded in siltstone (225 m) distinguished above this horizon
Unit 9 - recrystallized micrite and calcarenite interbedded in
siltstone (955 m)
on regional line 111 through the area of the Charco in the area between the high and low in the approxi-
Largo-1 and Mella wells (Fig. 9) and on lines 1410 mate center of the block (Fig. 8). A zone of complex
and 410 in the Lago Enriquillo area (Fig. 10). northeast to northwest-striking faults in the eastern
part of the block is perhaps related to movements
Main subsurface structural blocks of the along the northeast-striking Beata fault zone.
Enriquillo basin
Vicente Noble block
Mapping of all of these reflectors indicates that This block on the northeastern edge of the basin
the subsurface of the Enriquillo basin is divisible is isolated between the South Martfn Garcfa fault
into four fault-bounded structural blocks. We discuss zone to the southeast, fault A, and the left-lateral
each of these blocks and their bounding faults as Enriquillo-Plantain Garden fault zone to the north
defined by the depth in two-way travel time to the (Fig. 8). The Sierra Martfn Garcfa forms the proba-
top of the Sombrerito surface (Fig. 8). ble extension of this block to the southeast as the in-
tervening area is occupied by unfaulted strata of the
Bahia de Neiba block Trinchera and Arroyo Blanco Formations dipping to
This block is isolated between the South Martfn the northwest (Fig. 6). The steep plunge of the Sierra
Garcfa fault zone, a probable oblique-slip thrust zone Martfn Garcfa anticline may be terminated in the
along the southern edge of the 1200-m-high Sierra subsurface by fault A which exhibits a reverse throw
Martin Garcfa, fault A, and the Bahoruco fault zone, to the southwest. Reverse throw on fault A may be
the frontal thrust of the Sierra de Bahoruco (Fig. 8). the mechanism for monoclinally tilting the exposed
To the east, this block is probably truncated by late Neogene stratigraphic section of the western Vi-
the Beata fault zone, a northeast zone of probable cente Noble block to the northeast (Fig. 9). Fault A
oblique-slip normal faulting that defines the straight, is a subsurface fault with a transverse northwesterly
northeast-trending coast and shelf edge of the south- strike in the area between outcrops of the Arroyo
central Dominican Republic (Mann et al., 1991a; Blanco and Trinchera Formations on the northwest
Mauffret and Leroy, Chapter 21). Fault A is a subsur- flank of the Sierra Martfn Garcfa and the area of
face fault with a transverse northerly to northwest- Quaternary deltaic sedimentation of the Rfo Yaque
erly strike that is present in the sill area separating del Sur (Fig. 6). Fault A is shown on seismic line
Bahia de Neiba from the interior of the valley. The 111 in Fig. 9 as a broad zone of transparent to
Bahfa de Neiba block exhibits a large, closed struc- chaotic reflections that bounds a closed structural
tural high along the Bahoruco fault zone and a low high on its northeastern edge. Most of the reflections
along the South Martin Garcfa fault zone (Fig. 8). from the Vicente Noble block appear to originate
The Palo Alto-1 well is a dry exploration well drilled from the Arroyo Blanco Formation (Fig. 9). Deeper

reflectors are not apparent, possibly because of the source areas in the Cordillera Central and the thinner
steeper dip of the strata within the block. evaporitic section probably results from its greater
distance from the Caribbean Sea at Bahfa de Neiba
Mella block (Fig. 8). Fault C exhibits minor normal throw up
This basin-central block is isolated between the to the level of the base of the evaporites as seen
subsurface and transverse faults A to the east and on line 410 in Fig. 10. The structural contours are
fault B to the west (Fig. 8). The block is bounded to relatively flat across the Lago Enriquillo block as
the north and south by the Bahoruco and Enriquillo- seen on line 1410 in Fig. 10 although a central sag
Plantain Garden range-front faults, respectively. The defines a large structural low between Isla Cabritos
block is strongly downthrown along fault A with and the Bahoruco fault zone. This structural low is
about 4 seconds of structural relief between the manifested in a large gravity minimum seen plotted
closed structural high on the upthrown Vicente No- along line 133 crossing the Cabritos faulted anticline
ble block and the closed structural low on the down- from north to south (Fig. 11). North-striking faults of
thrown Mella block (Fig. 9). The block rises to the fault zone D, en-echelon normal faults in the area of
north and south and therefore exhibits a broad saddle Isla Cabritos, and the anticline defining Isla Cabritos
shape with its largest low in the approximate center itself may be secondary structures developed by left-
of the basin. The western edge of the Mella block is lateral shear along the Enriquillo-Plantain Garden
formed by reverse throw on fault B to the approxi- fault zone (Fig. 11). Mann et al. (1995) document a
mate level of the Mid-Las Salinas reflector (Fig. 9). 500-m-wide swath of tectonically uplifted Holocene
The Charco Largo-1 and Mella wells are drilled into coral reef deposits above these subsurface features.
a hanging-wall anticline on the upthrown side of
fault B. Two closed structural highs are present north
of the well area and one closed structural high is
Evolution of subsurface blocks of the Enriquillo
present south of the well area (Fig. 8).
basin through time
Angostura block
Comparison of the three subsurface structural
This basin-central block is isolated between the
maps for the times of latest-Middle Miocene
subsurface and transverse faults B to the east and C to
(top Sombrerito reflector), Early Pliocene (top
the west (Fig. 8). The block is bounded to the north
evaporites/Angostura reflector) and mid-Pliocene
and south by the Bahoruco and Enriquillo-Plantain
(Mid-Las Salinas reflector) reveals information on
Garden range-front faults, respectively. Fault B ex-
the evolution of the subsurface fault systems active
hibits much less reverse throw (~250 ms) than fault
in the basin during these times. Because deformation
A to the east (~4 s) (Fig. 9). Large structural lows
is cumulative, each map including those of the oldest
are present in the center of the block along with
top Sombrerito reflector reflects the youngest defor-
some short, discontinuous faults with apparent nor-
mation event in addition to the effects of all older
mal throws (Fig. 10). The structural contours define
events. For this reason, these maps do not provide
an upward slope towards the Enriquillo-Plantain Gar-
a true picture of the structure at the time each sur-
den and Bahoruco blocks from the structural lows
face formed. However, the maps do show how fault
in the approximate center of the basin. Strong west-
movement ceased in some areas during the times of
ward deflection of the structural contours near the
the younger, Mid-Las Salinas reflector.
Enriquillo-Plantain Garden fault zone may reflect
shear effects related to left-lateral displacement on
this fault (Fig. 8). These deflected contours underlie a
zone of en-echelon anticlines and tectonically uplifted Top Sombrerito reflector
Holocene reef deposits in the area of Lago Enriquillo It is unclear from the structural maps how ac-
(Mann et al., 1995). tive the subsurface faults were during late-Middle
Miocene times of deposition, since reflectors below
Lago Enriquillo block the top Sombrerito reflector are not well imaged
This block is isolated between the subsurface and (Figs. 9 and 12). Surface studies of the Sombrerito
transverse faults C to the east and fault zone D to Formation indicate a relatively uniform pelagic car-
the west (Fig. 8). The block is bounded to the south bonate environment ranging in water depth from
and north by the Bahoruco and Enriquillo-Plantain 1000 to 1500 m over a large area of south-cen-
Garden range-front faults, respectively. The Lago tral Hispaniola (McLaughlin et al., 1991). The large
Enriquillo block is distinguished from the basin slopes observed on the top Sombrerito structural
blocks to the east by a much thinner late Neogene map therefore were likely produced by deforma-
siliciclastic and evaporitic section. The thinner silici- tion after the deposition of the upper Sombrerito
clastic section results from its greater distance from Formation in late-Middle Miocene time (Fig. 12).

302 P. MANN et al.
Fig. 12. Time structural map (contour interval 50 ms) of the reflector interpreted as the top of the Middle-earliest-Late Miocene
Sombrerito Formation. Complete track map of lines used to make this map is shown in Fig. 7. Boxes indicate areas of detailed mapping
along basin margins.
Base evaporites (Angostura Formation) part of the Las Salinas Formation (Fig. 14). Fault A,
It is unclear from the structural maps how active a much wider zone of deformation, extends much
the subsurface faults were during the Early Pliocene, higher into an area of unresolved reflectors above
since stratigraphic growth adjacent to these faults one second of two-way travel time (Fig. 9). The
is not imaged (Figs. 9 and 13). The Angostura arcuate pattern of closed structural lows on the An-
Formation exhibits a large variation in thickness gostura and Lago Enriquillo blocks may reflect in-
some of which may be related to deformation and/or creased overthrusting of the similarly shaped frontal
compaction by the overlying siliciclastic formations thrust of the Sierra de Bahoruco. Flowage of salt and
(Fig. 9). The localization of the evaporites in the gypsum southwestward towards the subaerial expo-
central area of the basin adjacent to subsurface sures at Loma Sal y Yeso from the Angostura block
faults A and B suggest that reverse movement on may also contribute to this large, closed structural
these faults may have been active at this time and low in the center of the basin (Fig. 14).
created the sub-circular topographic low in which Large folds in the Mid-Las Salinas reflector be-
the evaporites formed (Fig. 13). neath Lago Enriquillo may reflect increasing activ-
It is interesting to note that these north-northwest ity and movement along the left-lateral Enriquillo-
to northwest-striking faults would be optimally ori- Plantain Garden fault zone (Fig. 14). Three folds are
ented to accommodate reverse fault movement if one present with east-west fold axial traces sub-parallel
assumes that the same approximately east-west di- to the Enriquillo-Plantain Garden fault zone. The
rection of interplate slip operated in the mid-Pliocene most northerly syncline north of the interpreted trace
as observed today with GPS geodesy (Dixon et al., of the Enriquillo-Plantain Garden fault zone is a
1998) (Fig. 8). The more northwesterly striking faults large syncline that overlies the deepest part of the
bounding the evaporites in the central part of the Lago Enriquillo and the lowest part of the valley
valley would therefore be prone to reactivation as at - 8 0 m BSL. The central fold is a large, dou-
reverse faults whereas the more east-west-trending bly plunging anticline with geomorphic expression
faults bounding the western part of the valley would on Isla Cabritos in the center of Lago Enriquillo.
be prone to reactivation as oblique-slip faults. This fold tectonically elevates Holocene reef and
algal tufa deposits that fringe the island (Mann et
Mid-Las Salinas reflector al., 1995). A dry exploration well, Cabritos-1, was
During the mid-Pliocene, fault B is covered by drilled on the crest of this structure at the eastern
undeformed reflectors correlated with the middle end of the island (Mann et al., 199 l a) (Fig. 11). The

N E O G E N E E V A P O R I T I C A N D S I L I C I C L A S T I C D E P O S I T I O N IN THE E N R I Q U I L L O BASIN 303
Fig. 13. Time structural map (contour interval 50 ms) of the reflector interpreted as the base of the Early Pliocene massive evaporites
of the Angostura Formation. Complete track map of lines used to make this map is shown in Fig. 7. Boxes indicate areas of detailed
mapping along basin margins.
Fig. 14. Time structural map (contour interval 50 ms) of the reflector interpreted as the middle part of the middle to Late Pliocene
siliciclastic rocks of the Las Salinas Formation. Complete track map of lines used to make this map is shown in Fig. 7. Boxes indicate
areas of detailed mapping along basin margins.

304 R MANN et al.
southernmost fold is a large syncline that underlies MOBIL CHARCO LARGO NO. 1
the part of the lake between the island and the frontal
thrust zone of the Bahoruco fault zone (Fig. 14). This
syncline corresponds to the large gravity minimum
plotted on the seismic line in Fig. 11. JtMANI
FORMATION
AND QUAT.
A
-, ._-, ._-, ._-,
ENVIRONMENT
MARINE ?
',MESTONC
Regional pattern of strain partitioning in the
61o- 2 ii
Enriquillo basin
The large folds seen in the structural map of the

Mid-Las Salinas reflector may form by a process 1219- 4 BRACK, SH
of 'strain partitioning', whereby part of the regional
transpression across this part of the island is accom-
modated by left-lateral slip along the Enriquillo-
Plantain Garden fault zone and part of the transpres- 1829i 6
sion is accommodated by shortening normal to the
trend of the Enriquillo basin (cf. regional cross-sec-
tion in Fig. 3B). Based on GPS observations and
modeling, Dixon et al. (1998) predict a minimum
2438 "~, 8 L I M E S T O N E,~
~ VERY SHALLOW
MARINE
rate of 8 -t- 4 mm/year of left-lateral slip along the ,,.....,..,.,

.'1".,;, ,T. -"2g. 9 ,'l', -"
Enriquillo-Plantain Garden fault zone. This fault ,.,-o, ,- 9 ,- 9 ,-.,

. . . . . . . .
3048- 10 -.y.=.y.,.y.=y.
.................
therefore accommodates about 38% of the total rate

of 21 -t- 1 mm/year between the Caribbean and North HYPER-
SALINE
America plates at the longitude of Hispaniola. ,.'.,~'A'A'.,"
"r
Well-developed folding at the Mid-Las Salinas 3658- 12
,.'A"A'A'.,~
........
horizon in the western part of the valley and waning ..-.......-..........-
j% =_%=_%=..'~=~
thrust movement on fault B in the central part of the EARLIEST
PLIOCENE "V'.'~.'V'.V,
''''";';';'""
valley may indicate a relative increase in the amount
4267- 14' ' '=~'==""'=-
of strike-slip faulting at the expense of regional !
LATE
!MIOCENE TRINCHERA
........
"-.--'.'.'-, OPEN
shortening along faults like fault B. The direction 1
i =
"
........
" MARINE
/ MIDDLE il
of the plate vector determined from the GPS study / M I O C E N E ![ SOMBRERITO ~=
j4877. 16 J 1 =
by Dixon et al. (1998) indicates that the majority
of the interplate motion is strike-slip in a roughly Fig. 15. General stratigraphic section of the Charco Largo-1
east-west direction (Fig. 8). well of the central Enriquillo basin. Lithologies are based on
a detailed well log and an accompanying report provided to us
by Mobil; ages and environments of deposition are based on
micropaleontologic work by McLaughlin and Bold using sample
STRATIGRAPHY AND MICROPALEONTOLOGYOF material taken from well cuttings. Key to symbols: brick pattern
THE CHARCO LARGO-1 WELL = deep-water limestone; dashed pattern = shale; v pattern --
massive halite; stipple = sandstone and shale; heavy black dots
-- conglomerate.
Charco Largo-1 well
The Charco Largo-1 well, drilled on the Mella where possible. The biostratigraphic findings in this
structural block (Fig. 8), provides a complete strati- study, which are summarized on Fig. 17, support the
graphic record of sedimentation and tectonics in ages and water depths proposed in previous studies
the central part of the Enriquillo basin. The well by Bold (1975) and McLaughlin et al. (1991) using
was drilled to a depth of 4830 m by Superior Oil surface samples mainly from the eastern part of the
Company (now Mobil) in 1980 and penetrated five Enriquillo basin along the flanks of the Sierra Martfn
formations: Sombrerito, Trinchera, Angostura, Las Garcfa. A summary of the formation names and ages
Salinas, and Jimanf (Fig. 15). from previous surface studies is given in Fig. 7.
Lithologic descriptions are taken from unpub-
lished company reports used with the permission of Lithology and age of the Sombrerito Formation
Mobil. Lithologic records allowed us to compile a in the Charco Largo-1 well
histogram of the thickness of halite beds in the well
(Fig. 16). Micropaleontologic analysis for ostracodes The Sombrerito Formation in the Charco Largo-1
(by Bold) and foraminifera (by McLaughlin) was un- well extends from 4508 m to the bottom of the hole
dertaken for 143 cuttings samples representing 9.15 at 4830 m (Fig. 15). As in outcrop, the well samples
m (30 ft) composites at intervals of 30.5 m (100 ft), consist of a light tan to light buff, cryptocrystalline

limestone rich in planktonic foraminifera (Fig. 17). and Orbulina universa. The only benthic foraminifer
The lowest part of the Sombrerito Formation, recovered was Siphonina sp. This poor fauna con-
from 4783.5 to 4830 m, is placed in Middle Miocene trasts with the generally more abundant foraminifera
planktonic foraminiferal zone M11 of Berggren et reported from the Trinchera Formation to the north-
al. (1995) based on the highest occurrence of the east in the Azua basin and on the northern flanks
nominate taxa (Fig. 17). Higher in the well from of the Sierra Martfn Garcfa (McLaughlin and Sen
4570.1 to 4656.1 m, the Sombrerito Formation is Gupta, 1994). This relation suggests the possibility
characterized by the presence of Sphaeroidinellopsis of a barrier between the Azua area, where up to 6 km
multiloba in the absence of Neogloboquadrina may- of open-water deposits occur, and the basin-central
eri, which places this interval at or just above the area of the Charco Largo-1 well, where a the thin,
Middle Miocene-Upper Miocene boundary, in the impoverished section is found. Based on regional
M12 or M13a zone (Fig. 17). The uppermost sample stratigraphic correlation to these outcrops, this in-
in this formation (4509.1-4512.2 m) is interpreted terval in the Charco Largo-1 well is interpreted as
to lie in the Upper Miocene M13b zone based on the Upper Miocene to basal Pliocene and therefore cor-
presence of Sphaeroidinellopsis seminulina and the relates well to previous dating of surface exposures
absence of S. multiloba. (McLaughlin et al., 1991; Fig. 7).
Benthic microfauna recovered from the Som-
brerito Formation are indicative of a deep-water Lithology and age of the Angostura Formation in
setting. The presence of Cibicides wuellerstorfi at the Charco Largo-1 well
4692-4695 m indicates middle bathyal or deeper
depths. Higher in the formation, the presence of The Angostura Formation in the Charco Largo-1
Cibicidoides mundulus, Cibicidoides bradyi, Sig- well consists of 1562 m of clear to milky, moderately
moilopsis schlumbergeri, Anomalina flintii, and Ori- hard, anhedral halite interbedded in greenish-gray,
dorsalis umbonatus indicates an upper bathyal or indurated, silty, slightly calcareous shale (Fig. 15)
deeper setting. Similar ages are described from out- (Table 1). The base of the formation is taken as
crops of the Sombrerito Formation in the Sierra the lowermost bed of halite of approximately 7.5
de Neiba along the northern edge of the basin m thickness at a depth of 4192 m; the top of the
(McLaughlin et al., 1991). formation is taken as a hard, buff-white limestone
of approximately 4.6 m thickness at a depth of 2630
Lithology and age of the Trinchera Formation in m. The Angostura Formation is interpreted to un-
the Charco Largo-1 well conformably overlie the Trinchera Formation in this
well.
The Trinchera Formation in the Charco Largo-1 Because of its marginal marine environment and
well was encountered from 4237 to 4508 m and impoverished microfauna, there is little fossil con-
consists of 271 m of greenish shale containing Late trol on the age of the Angostura Formation in
Miocene foraminifera (Fig. 15) (Table 1). We pro- the Charco Largo-1 well. However, surface samples
pose that this interval of Trinchera Formation is from the Angostura Formation in the Loma Sal y
equivalent to the Trinchera Formation and possibly Yeso contain an impoverished, brackish water or
the overlying shales of the Quita Coraza Formation hypersaline ostracode fauna, considered indicative
of the Azua basin and northern flank of the Sierra of Late Miocene to Early Pliocene in age by Bold
Martfn Garcfa (Figs. 7 and 8) (McLaughlin et al., (1975). The ostracode fauna is interpreted by Bold
1991). The 271 m thickness in the well is signif- (1975) to represent a brackish or hypersaline envi-
icantly less than the 1300 to 2200 m of interbed- ronment consistent with the evaporitic lithology. We
ded deep-marine mudstones and turbidite sandstones consider the lower part of the Angostura Formation
present in outcrops 30 km to the east in the Azua to be equivalent to gypsum-bearing deposits of the
basin (McLaughlin et al., 1991). In outcrops in both Arroyo Blanco Formation exposed along the north-
the Azua and San Juan basins, a progression may eastern flank of the basin (Fig. 13). The base of
be traced from outer turbidite-fan facies in the lower the Arroyo Blanco Formation in this area is well
part of the Trinchera Formation to inner-fan facies constrained biostratigraphically to lie near the top of
in the upper part. This relation records the north-to- the Lower Pliocene (McLaughlin et al., 1991).
south or northeast-to-southwest progradation of the The halite section of the Angostura Formation
Trinchera Formation over the deep-water carbonate appears massive on the generalized well lithologic
environment recorded by the underlying Sombrerito column of Fig. 15 but the Charco Largo-1 well
Formation. log shows individual halite beds of halite ranging
The Trinchera Formation samples in the well con- in thickness from 1.5 to 76 m and averaging 3 m
tain an impoverished microfauna (Fig. 17). Plank- (Fig. 16). The thickest beds of halite are generally
tonic species present are limited to Globigerina spp. confined to the lower one-third and upper one-third

306 P. M A N N et al.
Fig. 16. Histogram showing thickness of halite beds in the Charco Largo-1 well as reported on the detailed well log. See text for
discussion.
~ ~ ~ ~
9 [] .[] [] II l 9 m [][][] 9 [] 9 salebrosa

. . . . -~ . . . . . . . . ~ -- -- -- [] [] -- i - - C~::o~,on, =.
l l II Pensscc/lhendeasuDrugosa
. . . . . . . . . . . . . . . . . i . . . . . l [] [] [] _ ~ - - m _~ _ _~ m _ P . , ~ = c ~ . , ~ . a ~=..~a
-- [] [] 9 [] [] [] [] [] Cyonce,ss,mOis ,,
CypnCe~s mexlcana
m cyom~oos~s vidua
m Hem~c'/ons reticulata
mmm m m Penssocythendea so. A
m--mm- m r.Jmnoc./lhere stap/inl
mmmi m m ~arwmula so.
mmmm m ~leterocyons sO.
I-,~ll m :~ensscc'/thendea ex~lis
m immm m Cative/la haws
m i iN CypdcYe~sso.
iiimmmmmmmmmmmmm [] i m [] ~edssocythendea sO.
m Doleroc/pna so.
mm ~udana gatunensis
mmmmmmmmmmmmmm mmm [_mmmmmmmmmmmm- Ammonia sO.

mm mm m-mnnmm-nmmmmm- E!ohidium sO.
Archaias so.
Siphon9 sO.
Ondorsalis unbonatus
I,
Sigmoiioosis schlumergen
m- Anomalina flintii
-m Cibicicloides brady9
:m CibiG~doidesmundulus
-m
Cibic/de$ wuellerstorfl
GloOigenna sO.
OrOulina umversa
31oOorotalia rnenradii
HastigerYna siphonifera
3pharoidine//opsis seminu/ina
Glotjigennoides trY/oOus
31obigennoides obliquus obliquus
31oDoquadtfna altispira
31obigerma venezuelana
12 31obigenna nepenthes
-i SplTaeroidinallopsi$ multiloDa
Globorotalia mayed
o,;o o, = o, ~ -~ N ==== c, u -,4 -,4 cn o o 0~ ~ ~ o = 0= ~ ' 9 :=~ .F,
(3) ; .~ I'~ (JO J~ 0 clO U1 .4~ "-,4 CO P,J C. O1

9 1~ (~ l,~ .~ r9 oo . pI 9-, p i ' - ; p ~ .~ m
~_ ,
QUATERNARY
SOMBRERITO TRINcHER ANGOSTURA L.AS SALINAS JIMANI FORMATION
LACUSTRINE DEPOSITS
PLIO-
I ~ ~ I ~= UPPER PLiOCENE UPPER PLJOcENE PLEIST QUA TERNARY STAGE
~9 ~ ~ ~ MIOCENE
Fig. 17. List of foraminiferal and ostracode fauna recovered from well cuttings of the Charco Largo-1 well by McLaughlin and Bold.

of the formation. In the middle one-third of the formation is a 4.5-m-thick anhydrite at approximately
mation, the halite bed are generally thinner and are 2595 m depth; this is approximately 535 m above
interbedded with siltstones that make up 5 to 10% the highest Angostura Formation halites that occur
of the formation's thickness. The interbedded shale near 2630 m. The base of the Las Salinas Formation
comprises approximately 25% of the formation, with in the well is marked by a 4.6-m-thick, hard buff-
shaley zones 3 to 30 m in thickness. The thinner white limestone (Table 1). In outcrops at Loma de
halite beds of the middle one-third of the forma- Sal y Yeso on the southwest side of the basin, this
tion are interbedded with siltstone. The Angostura basal limestone is correlative to a 2-m-thick mol-
Formation is virtually devoid of sandstone. lusc-shell-rich sandstone bed with symmetrical wave
There is little gypsum and anhydrite reported ripples (McLaughlin et al., 1991) (unit 4 on map in
from the halite interval in the Charco Largo-1 well. Fig. 18A and on section in Fig. 19A). Because of its
There are, however, two sills of hornblende-diorite resistance to weathering, vertical to overturned bed-
intruded into the central part of the Angostura For- ding attitude, and topographic prominence among
mation. The sills range from 3 to 9 m in thickness more erodible rocks of the Angostura Formation and
and probably are related to the same Late Pliocene- basal Las Salinas Formation, we informally named
Pleistocene episode of alkalic basaltic volcanism that this unit 'Razorback Ridge'.
created a number of volcanic bodies, whose rocks Overlying this unit in the well, the lower 100 m of
extend across south-central Hispaniola (Wadge and Las Salinas Formation consists of 10-20-cm-thick
Wooden, 1982; Mann et al., 1991a). medium-grained sandstone beds alternating with 5 -
The Angostura Formation in this well contains 10-cm-thick siltstone beds. The section above this
a microfauna that reflects deposition in a marginal is characterized by 15-25-m-thick cycles of in-
marine setting during the Pliocene. The association terbedded, trough-cross-bedded pebbly conglomer-
of the ostracodes Cyprideis salebrosa and Cyprideis ate, sandstone, and wave-rippled maroon and green
mexicana indicates a chronostratigraphic position no siltstone. In outcrop, sandstone and conglomerate
lower than the upper part of the Lower Pliocene. beds are generally more resistant to erosion and pro-
This is slightly younger than previous reports from duce the prominent hogback topography seen in the
outcrops of the formation; Bold (1975) recorded aerial photograph of the Loma Sal y Yeso area in
Cyprideis subquadraregularis and Cyprideis pascag- Fig. 18B.
oulaensis, which together were considered to suggest The Las Salinas Formation includes a more di-
a position near the Miocene-Pliocene boundary. verse association of ostracodes than the underlying
Cyprideis salebrosa is indicative of low-salin- Angostura Formation, but occurrences are similarly
ity environments. The foraminifera Elphidium and scattered. The faunas recovered are mostly composed
Ammonia are present in some of the samples and of marginal marine species of Cyprideis and Perisso-
could indicate brackish-water or hypersaline con- cytheridea. The foraminifera Elphidium and Ammo-
ditions. In addition, Archaias was found in one nia were also recovered. Both these forms are tol-
sample, suggesting the presence of shallow-marine erant of variable salinity conditions. Cyprideis sale-
Thalassia sea-grass environments (Sen Gupta and brosa occurs throughout the formation and Cyprideis
Schaefer, 1973). Although the formation includes a similis occurs near the top, indicating a mid-Pliocene
great thickness of evaporite, reflecting deposition in or higher position. The above ostracode species are
an evaporative lake system, the varying representa- consistent with placement of this formation by Bold
tion of brackish-water and shallow-marine faunas in (1975) in the Pliocene to Holocene Cyprideis sale-
a succession suggests that significant variations in brosa zone. The Angostura Formation in the Charco
salinity occurred in the overall evaporative system. Largo-1 well and in outcrops of the Loma Sal y Yeso
is considered Early Pliocene in age.
Lithology and age of the Las Salinas Formation
in the Charco Largo-1 well Lithology and age of the Jimani Formation in the
Charco Largo-1 well
The Las Salinas Formation consists of a 2182-
m-thick, coarsening-upward section of shale, clay, The Jimanf Formation comprises 210 m of in-
sandstone, conglomerate, and minor anhydrite terbedded shales, sandstones, conglomerates and
(Fig. 15). The formation is present at a depth of limestones in the well, similar to the facies described
448-2630 m in the Charco Largo-1 well (Table 1). for the type section near Jimanf in the westernmost
The conglomerate contains clasts of quartz and vari- Enriquillo basin (McLaughlin et al., 1991) (Fig. 15).
ous types of igneous and metamorphic rocks derived The Jimanf Formation is not present in the exposed
from the erosion of Cretaceous to Eocene arc rocks section to the northeast of the Loma Sal y Yeso
in the Cordillera Central (McLaughlin et al., 1991). (Fig. 18). Samples of Jimanf Formation from the
The single evaporite bed in the Las Salinas For- well have scattered occurrences of ostracodes simi-


lar to those of the underlying Las Salinas Formation. salt diapirism between the Sierra de Bahoruco and
Diversity is slightly higher in the Jimanf Formation, the topographic depression of the Enriquillo basin
with as many as seven species in a single sample. (Fig. 18). Thrust faults of the Bahoruco fault zone
Cyprideis salebrosa and Perissocytheridea spp. are extend in a narrow belt along the mountain front to
present in all ostracode-bearing samples, suggesting the Haitian border (Fig. 6). Loma de Sal y Yeso is
a reduced-salinity environment where samples were the only exposed salt-cored fold in the belt along
recovered. this mountain front and supports subsurface data
None of the samples produced species indicated showing that evaporites are restricted to a small area
by Bold (1975) as typical of the Jimanf Forma- of the eastern Enriquillo basin (Fig. 4). Mapping in
tion (e.g., Cyprideis portuprospectuensis, 'Campylo- the quarry areas of Loma Sal y Yeso was carried out
cythere' perieri). However, they do include species to determine the fault structure of this section of the
considered typical of the Jimanf or upper Las mountain front and to improve correlations between
Salinas Formations, including Cyprideis similis, C. rocks exposed in the diapir and those penetrated by
salebrosa, Perissocytheridea subrugosa, and Lim- the Charco Largo- 1 well.
nocythere staplini. The Jimanf Formation is there- Mapping shows three major lithologically distinc-
fore considered Late Pliocene to Pleistocene, in the tive fault-bounded lithologic units in the central part
Cyprideis salebrosa zone, based on these occur- of the Loma de Sal y Yeso that form a composite
rences and on the presence of the Quaternary form section at least 350 m thick (Fig. 19A). Deformation
'Campylocythere' perieri in outcrop samples. associated with folding and diapirism on Loma de
Sal y Yeso appears to have eradicated the primary
Lithology and age of the undifferentiated fabrics, sedimentary structures, and most of the bed-
Quaternary rocks in the Charco Largo-1 well ding characteristics of both the halite and gypsum.
This makes interpretation of evaporative depositional
The Jimanf Formation is overlain by Quaternary environments difficult.
cover (Fig. 15). The ostracode fauna of the Qua-
ternary section is similar to that of the Jimanf but
more samples are fossiliferous and exhibit slightly
higher diversities, with up to eight species in a Unit 1
sample. Cyprideis spp. and Perissocytheridea spp. This unit is composed of a minimum of 200 m of
are the most notable faunal elements and suggest the massive halite of the Angostura Formation that
continued marginal marine conditions. This unit in- occurs as an elongate strip along the northeastern
cludes many of the same Plio-Pleistocene species edge of the hill (Fig. 19A). The halite exhibits a
listed by Bold (1975) as typical of the Jimanf and recrystallized fabric composed of interlocking, irreg-
Upper Las Salinas, including Cyprideis similis, C. ular halite crystals 1-2 cm across. Fresh exposures
salebrosa, Perissocytheridea subrugosa, and Limno- of halite reveal gray to black flow bands in the
cythere staplini. It also includes the only occurrences halite composed of thin layers of concentrated or-
in the well of Cyclocypris sp., Cypridopsis vidua, ganic debris. This banding probably formed by salt
Hemicypris reticulata, and Darwinula sp. flowage from depth as suggested by seismic line 111
that is adjacent to the Loma Sal y Yeso (Fig. 9).
Flow bands vary from 1 to 3 mm in thickness,
SURFACE GEOLOGY OF THE LOMA DE SAL Y YESO and can be traced over distances of several me-
AREA AND STRATIGRAPHIC CORRELATION WITH ters across single outcrops. These bands commonly
THE CHARCO LARGO-1 WELL exhibit small-scale open folds, and dip either ver-
tically or steeply to the south (Fig. 19C). Locally,
Stratigraphy of L o m a de Sal y Yeso the massive flow-banded halite contains intervals of
interbedded gray mudstone. We interpret these rocks
Loma de Sal y Yeso forms a low range of hills as shallow-water halite deposits in an enclosed basin
elevated by a combination of thrust faulting and subject to periodic influxes of terrigenous rocks.
Fig. 18. (A) Geologic map of Loma Sal y Yeso in the area of the open-pit salt mine and measured section in Arroyo del Pozo.
Key to numbered horizons: 1 - recrystallized, flow-banded halite and overlying shale of the Angostura Formation; 2 = recrystallized
massive to thinly bedded gypsum and interbedded shale of the Angostura Formation; 3 -- finely laminated gypsum and unfossiliferous,
massive, gray silty mudstone overthrust by halite of the Angostura Formation; 4 = 'Razorback Ridge', a resistant bed of shallow-marine
grainstone taken as the base of the Las Salinas Formation; 5 -- ripple-marked sandstone and interbedded siltstone; 6 = trough-cross
bedded sandstone and interbedded siltstone; 7 = coquina beds interbedded in siltstone; 8 -- growth position coral reefs and reworked
coral debris interbedded in siltstone; 9 = recrystallized micrite and calcarenite interbedded in siltstone. (B) Aerial photograph is Marena
series (1983), roll 43, flight 21, no. 212 (original scale: 1:40,000).

Z
o

Unit 2 Structure of Loma Sal y Yeso

This unit makes up most of the hill and is com-
posed of a minimum of 350 m of massive and Mapping in the gypsum unit indicates a synform
bedded gypsum and interbedded siltstone of the An- along the top of Loma de Sal y Yeso (Fig. 19A).
gostura Formation (Fig. 19A). The bedding of the The contact between the halite of unit 1 and the
gypsum of unit 2 is locally visible both on 1 : 40,000 locally horizontally dipping gypsum of unit 2 is a
scale aerial photographs and in the field (Fig. 18A). fault that is either vertical or dipping steeply to the
Packages of gypsum up to 20 m in thickness are south where exposed (Fig. 19A). The syncline may
more resistant and preferentially weather out of be a local drag effect caused by reverse motion along
thinner, poorly exposed siltstone beds. Individual this fault contact. In other areas to the east, gypsum
gypsum beds appear to range from 1 cm to 2 0 - 3 0 of unit 2 appears to have an anticlinal structure at the
cm in thickness. The beds are always weathered to crest of the hill (Llinas, 1972a,b).
nodular alabastrine gypsum as defined by Schreiber A large-scale, transverse right-lateral fault visible
et al. (1976) which are composed of centimeter- on aerial photographs offsets the hill to the southeast
scale masses of white gypsum surrounded by gray of the open-pit salt mine (Fig. 18). Mann et al.
shrinkage cracks, and fine-grained gray gypsum and (1991a) named this fault the Las Salinas fault zone
anhydrite (?), and organic inclusions. We interpret after the town in the fault valley.
these rocks as shallow-water gypsum deposits in At the outcrop scale, a well exposed 1-2-m-wide
an enclosed basin subject to periodic influxes of shear zone separates the unit 1 halite from the in-
terrigenous rocks. terbedded siltstone and gypsum of unit 3 (Fig. 19A).
This shear zone strikes parallel to the trend of Loma
Unit 3 Sal y Yeso and dips vertically or steeply to the south-
This unit is a 30-m-thick section of interbedded, west. Small-scale folds within the shear zone are
gray, finely laminated mudstone ('paper shale'), and asymmetric and indicate up-to-the-southwest reverse
interbedded siltstone, and thin (1-cm-thick) gypsum motion across the zone (Fig. 19C). Flow banding
beds (Fig. 19A). The paper shale section is faulted indicative of ductile deformation is present within
against the halite, which itself contains thin interbeds the sheared halite and in brown mudstone interbeds
of a similar paper shale. The paper shales exhibit adjacent to the shear zone. Bedding in unit 3 and the
very fine-grained, light and dark laminations which overlying strata of the basal Las Salinas Formation
form parting surfaces. Although extensively sampled is vertical or slightly overturned to the northeast.
in previous studies (e.g., McLaughlin et al., 1991), Bedding dips in the Las Salinas Formation gradually
unit 3 is barren of microfossils. We interpret the decrease in a basinward direction (Fig. 19A). No
rocks of unit 3 as anoxic lacustrine deposits. other major faults were mapped in the monoclinally
Gypsum is the most resistant unit and forms the dipping section of the Las Salinas Formation north
highest elevations of the hill (Fig. 19A). Where of Loma Sal y Yeso (Fig. 19A). For this reason,
unmined, the halite forms a low depression along the these rocks would project downdip into the Charco
crest of the hill. Where mined along the northeastern Largo-1 well (Fig. 13 and Table 1).
edge of the hill, large sinkholes caused by salt
dissolution have formed at the lowest points of the Economic significance of salt and gypsum mined
excavated areas (Fig. 19B). The basinward edge of from Loma Sal y Yeso
the salt area is delineated by a 3-m-thick vertical
bed of highly resistant limestone that marks the base The salt and gypsum mined in the Loma Sal y
of the Las Salinas Formation ('Razorback Ridge', Yeso has been exploited for the past 30 years and
or unit 4 on Fig. 19A). The flat area northeast of is of local economic importance. Halite mined at
Loma Sal y Yeso is underlain by outcrops of the Las the quarry has a 96 to 98% purity. The mechanized
Salinas Formation (Figs. 18 and 19A). government salt mine, which formerly exported salt
Fig. 19. (A) Cross-section of the Loma Sal y Yeso in the area of the open-pit salt mine. Tadpoles indicate measured dips of bedding.
Numbered horizons are the same as Fig. 18. The downdip extension of the halite and thrust faults is speculative but consistent with the
geometry of halite seen on seismic line 111 north of Loma Sal y Yeso (compare to line in Fig. 10). (B) View to the northwest along
the open-pit salt mine (see approximate location in A). The deep valley is formed by extraction of halite by mining and by solution of
halite by rainwater (note natural sinkholes formed in the bottom of the mined valley). The width of the halite deposit corresponds to the
approximately 50-m-wide valley. The outcrop to the left is almost flat-lying recrystallized gypsum; the steep ridge to the right corresponds
to a steep reverse fault which thrusts halite on massive silty mudstone cropping out on the vegetated hill to the right (compare this view
to the cross-section shown in A). (C) View to northwest of recrystallized halite along the edge of the open-pit salt quarry. Note hammer
for scale in the center of the photo; small-scale open folding indicates thrusting with southwest-side up. Note flow banding within sheared
halite, and brown mudstone interbeds within halite to left of photo. Color photographs at http://www.elsevier.nl/locate/caribas/

312 E MANN et al.
to the northeastern U.S. for use on winter streets, was

closed in 1963. Subsequent salt mining reoccupied
the former quarry but uses only manual labor. The
salt is broken into gravel-sized pieces, filled in burlap
bags, and loaded by hand onto trucks. The salt is
used for domestic consumption.
The gypsum quarry on the southeastern side
of Loma Sal y Yeso is being actively mined by
Cooperaci6n Dominicana de Empresas Estatales
(CORDE) using fully mechanized methods. Most
of the gypsum is exported by ship through the port
of Barahona to Santo Domingo, other Caribbean
countries, and northern South America for use as a
hardening agent in cement. Four different commer-
cial grades of gypsum are recognized by CORDE ge-
ologists on the basis of colors ranging from white to
dark brown or black. The color variations reflect the
amount of siliciclastic impurities within the gypsum.
Stratigraphy and sedimentology of the Las

Salinas Formation in the Arroyo del Pozo
The three evaporitic units of the Loma Sal y

Yeso are overlain by a 1820-m-thick section (Las
Salinas Formation) of monoclinally basinward-dip-
ping sandstone, shale, and conglomerate of the Las
Salinas Formation. These strata form a relatively flat
area along the southern edge of the Enriquillo basin
northeast of the Lome Sal y Yeso (Fig. 19A). Five
subdivisions of the formation (units 4-9) are based
on distinct assemblages of lithologies and sedimen-
tary structures. The distribution of lithologic units Fig. 20. Key for lithology and sedimentary structures for Fig. 21
and all subsequent measured sections in this paper.
1-9 are shown on the map of Fig. 18 and cross-
section of Fig. 19A. Fig. 20 provides a key for all
lithologies and sedimentary structures of the five The limestone bed appears to conformably overlie
units (units 4-9) we identified from the Las Salinas approximately 50 m of gray massive siltstone and
Formation (Fig. 21). Fig. 22 provides an expanded, 'paper shale' with interbeds of gypsum (unit 3 of
more detailed section of the lower, well exposed 27 Loma Sal y Yeso). The limestone bed is made
m of unit 4 and the base of unit 5 at the head of the up of multiple beds of medium- to coarse-grained
Arroyo del Pozo. Because of long covered intervals calcarenite 5-10 cm thick, with pervasive larger
between most units, the boundaries and thicknesses shell material. This includes numerous intact and
of each unit are approximate. broken bivalves and gastropods (Fig. 22E). Bedding
Approximately 50% of sections 5-9 of the Las units are defined by alternations in grain size of
Salinas Formation consist of easily erodible gray coarse- and fine-grained shell hash.
to bluish siltstone that is covered or poorly ex- The limestone of unit 4 is interpreted as a basal
posed in most places (Fig. 21). Limestone, sand- transgressive deposit marking a marine or brackish
stone, and conglomerate form the most resistant water incursion over the Angostura evaporite basin
beds which are exposed in outcrops along the Ar- of the central Enriquillo basin. We correlate this bed
royo del Pozo and along narrow, heavily vegetated with the hard, buff-white, 4.6-m-thick limestone
strike ridges extending to the east and west of the penetrated at 2625 m in the Charco Largo-1 well at
stream (Fig. 18A). the transition between the Angostura and Las Salinas
Formations (Fig. 15 and Table 1).
Unit 4
The base of this unit and of the Las Salinas Unit 5
Formation is marked by the 3-m-thick, resistant bed This unit is a ~ 180-m-thick interval of symmetri-
of shallow-marine grainstone and coquina which cal, wave-ripple-marked sandstone and interbedded
forms 'Razorback Ridge' (unit 4 in Fig. 22A). siltstone recording low-energy, nearshore conditions

in a shallow-marine and/or lacustrine environment open-marine influence (Fig. 22A). Where exposed,
(Fig. 21). The section consists mainly of 0.5- the upper 100 m of unit 5 consists of bluish silty
1.0-m-thick packages of gray to greenish, friable mudstone. The top of unit 5 is defined by the base of
quartz-rich litharenite in beds ranging from 1-2 a sand-rich section of trough cross-bedded sandstone
cm up to 25-30 cm (Fig. 21). The sandstones are at 180 m (Fig. 21).
fine- and medium-grained, with quartz and lithic
fragments showing a moderately well-rounded char- Unit 6
acter. These sandstone packages alternate with 1.0 This unit consists of 390 m of trough cross-bed-
to 0.25-1.0-m-thick horizons of bluish to greenish ded sandstone, ripple-marked, channelized sandstone
siltstone and silty mudstone. The sandstone-rich in- and interbedded siltstone that record higher-energy,
tervals in the lower part of the section shown in deeper-water shallow-marine conditions than unit 5.
Fig. 22A show packages arranged into two thinning- Unit 6 is distinguishable from overlying and under-
and fining-upward cycles over intervals of 4 to 9 m. lying units by its thick sandstone beds and its lack
Bedding contacts are planar. Sand size is fine- and of marine shell horizons. Sandstones form packets
medium-grained. ranging in thickness from 2 to 10 m. These sandstone
Symmetrical wave ripple-marks are found on packages are slightly more resistant than interven-
most bedding planes in the sandstone-rich packages ing siltstone beds and form low ridges which control
(Fig. 22D). The vertical or slightly overturned atti- the direction of Arroyo del Pozo (Fig. 18A). Each
tude of bedding at the base of the section allows ex- sandstone package consists of a thinning- and fining-
cellent views of extensive ripple marks on the upper upward cycle marked by a thicker, trough cross-bed-
surfaces of the bedding planes. Sand is concentrated ded sandstone at the base overlain by thinner-bed-
in the troughs of the ripples marks. Paleocurrent ded, ripple-marked sandstone beds, passing upwards
measurements on 24 ripple marks restored to hori- into siltstone and silty mudstone at the top of the
zontal indicate a paleowave direction of ENE-WSW cycle. The basal sandstone beds generally exhibit
(Fig. 22A). This is the approximate present-day di- pronounced basal scour or channelization into the
rection of prevailing winds which blow westward siltstones of the upper part of the underlying cycle.
from the Caribbean Sea into the Enriquillo Valley. Individual sandstone beds in the packets range in
Vertical worm burrows (Skolithos?) characteristic of thickness from 10 to 50 cm. Sandstone is brown to
intertidal or beach environments are found in several greenish in color, and is composed of friable, fine- to
of the ripple-marked sandstone beds (Fig. 22C). medium-grained, moderately well-rounded litharen-
Interbedded within the sandstone beds near the ite. Cement is a dusty-appearing matrix of clay min-
base of unit 5 is a resistant, ripple-marked bed erals derived from the breakdown of unstable rock
of dense, fractured, buff-white chert, 1 m thick, fragments. The thicker sandstone beds are planar to
with oil seeps and small-scale soft-sediment 'di- broadly lens shaped, especially those with promi-
apirs' protruding through the bed into overlying nent trough cross-bedding. The sandstone beds ap-
beds (Fig. 22B). The 'diapirs' are 10 to 15 cm in pear continuous for tens of meters in the field and on
height and are rooted in an underlying bed of the aerial photographs (Fig. 18A) and are assumed to be
same composition. This rock is similar to descrip- equally extensive in the downdip direction (Fig. 18B).
tions of 'Magadi-type chert' of Jurassic, Eocene, Paleocurrents were measured from oblique
and Pleistocene age in Wyoming (Surdam et al., streambed exposures of 24 trough cross-beds in
1972), of Paleozoic age in Scotland (Parnell, 1988), unit 6 using the method of DeCelles et al. (1983)
and Cambrian age in Australia (White and Youngs, (Fig. 21). This method allows more precise estimates
1980). Magadi-type chert is precipitated from alka- of the paleocurrent direction for two dimensional
line lakes which store large amounts of silica in outcrop exposures of trough cross-beds. The pale-
solution. Conversion to chert involves loss of sodium ocurrent direction is E N E - W S W and is consistent
and water, and a volume loss of at least 25% that with the paleocurrent of tipple marks in unit 5.
causes soft-sediment deformation. These sediments
form in stratified bodies of standing water, under Unit 7
conditions conducive to the preservation of organic This unit consists of a largely covered 100-m-
matter. As a result, they commonly exhibit oil seeps thick section of siltstone with two 20-cm-thick beds
and are associated with petroleum source rocks. of marine shell fragments identified (Fig. 21). The
The section becomes more sand-rich with thicker interval is distinguishable from unit 6 by its lack of
(1 m) cross-bedded sandstone beds 25 m above the thick sandstone beds.
base of the Arroyo del Pozo section (Fig. 22A). A
single coquina bed composed of broken and whole Unit 8
marine shells is present at 21.5 m above the base Unit 8 consists of a 225-m-thick, coral-rich lime-
of the section and may indicate an increasingly stone and siliciclastic section. The base is marked


by section beginning with a l 1-m-thick coral reef unit 9 suggests that marine influence and water depth
in growth position. The coral reef contains delicate may have varied through time as it has done in the
branches of coral, clearly in growth position. The Holocene (Taylor et al., 1985).
reef unit is overlain by 212 m of bluish siltstone with
a few 10-20-cm-thick beds of litharenite. Correlation between Angostura and Las Salinas
Formations exposed in Loma de Sal y Yeso and
Unit 9 in the Charco Largo-1 well
This is the highest unit measured in the Arroyo
del Pozo section and is largely covered. Limited Broad similarities in the nine distinct lithologic
exposures suggest that it seems to consist mainly of units exposed in outcrops near Loma Sal y Yeso
bluish siltstone with infrequent, 5-20-cm-thick beds and in lithologies described in the well log and
of litharenite. The unit is distinguishable from unit 8 reports from the Charco Largo-1 well allow correla-
by its lack of coral beds. The unit also includes 0.2- tions to be made between the two sections (Table 1)
1.3-m-thick beds of recrystallized micrite and cal- (Fig. 21). A significantly greater thickness of An-
carenite containing marine mollusc-shell fragments. gostura Formation evaporites occurs in the Charco
These resistant beds form strike ridges which can Largo-1 well (1562 m, Fig. 15) than in the Loma de
be followed for several kilometers on aerial pho- Sal y Yeso outcrops (350 m, Fig. 19A). Evaporitic
tographs (Fig. 18A). Above this unit, Quaternary units 1, 2, and 3 are probably thinner than evaporitic
alluvium of the Enriquillo Valley onlaps the rocks at units in the Charco Largo-1 well because of local
the top of the section. structural complexities and salt flowage effects as
seen on seismic line 111 (Fig. 9) and as seen in flow
Facies interpretation of Las Salinas Formation textures of evaporites cropping out on Loma Sal y
in the Arroyo del Pozo Yeso (Fig. 19C).
The progression of facies through this section For example, the 200 m of massive halite of unit
reflects an overall increase in marine influence. An 1 contrasts with 1539 m of interbedded halite and
initial marine transgression recorded by the high- shale in the Charco Largo-1 well. Unit 1 may be
energy limestone of unit 4 ends the hypersaline significantly thinner than the basin-central deposits
restricted environment recorded by the evaporitic because it represents a thin diapiric 'offshoot' of the
facies of Loma Sal y Yeso; this transgression termi- parental evaporite body as suggested on seismic line
nated evaporitic deposition throughout the Enriquillo 111 in Fig. 9.
basin. High-energy wave-dominated shoreline de- Massive gypsum of the type and thickness (350
position marked by well sorted, rippled sandstones m) seen in Loma Sal y Yeso (unit 2) is not present
at the base of unit 5 pass upwards into a more re- in the Charco Largo-1 well. The evaporite section in
stricted, possibly alkaline and lacustrine environment the well consists entirely of halite. It is possible that
marked by the inferred Magadi-type chert. This re- the basin-center facies were more saline than more
striction of the basin may have been short-lived and basin-edge areas like Loma Sal y Yeso.
was superseded by more open-marine conditions as The 50-m-thick massive gray mudstone and gyp-
supported by the reappearance of marine macrofos- sum horizon (unit 3) that is faulted against the
sils 10 m above the chert (Fig. 22). A deeper-water halite along the northern edge of Loma Sal y Yeso
marine environment is marked by the alternating (Fig. 19A) may correlate to a 23-m-thick shale and
sandstone ridges and intervening shaley horizons gypsum interval in the well that separates the top of
dominates sections 5 and 6. We interpret this fa- the halite at 2652 m from the base of a 4.6-m-thick,
cies as a shallow-marine, fluvial-deltaic setting. The dense, white limestone bed at 2630 m in the well.
lower part of the section with numerous sandstone The 3-m-thick shallow-marine limestone of 'Ra-
ridges is interpreted as a coastal deltaic complex zorback Ridge' (unit 4), which marks the base of the
characterized by thick, sandstone bodies showing Las Salinas Formation in the Arroyo del Pozo, may
thinning-upward profiles (Fig. 21). Upward in the correlate to a 4.6-m-thick bed of limestone in the
section in units 7, 8, and 9, coral becomes more well (Table 1). Both units mark the end of significant
prominent and suggests the presence of a siliciclas- halite formation in the Enriquillo basin.
tic-dominated shelf setting with scattered coral reefs The 150-m-thick interval of ripple-marked and
(Fig. 21). The absence of growth position corals in burrowed sandstone and interbedded siltstone at the
Fig. 21. Measured section from the Las Salinas Formation in the Arroyo del Pozo north of Loma Sal y Yeso (see Figs. 18A and 19A for
locations). Base of section (unit 4) and top of section (unit 9) are shown on the map in Fig. 18A. A detailed measured section of the
lower part of the section is shown in Fig. 15. Paleocurrents are measured from trough cross-beds. See Fig. 20 for explanation of lithology
and sedimentary structures.

316 E MANN et al.
Fig. 22. (A) Detailed measured section of the lower 28 m of the Las Salinas Formation in the Arroyo del Pozo section (see Figs. 18A and
19A for locations). Paleocurrents are from ripple marks. See Fig. 20 for explanation of lithology and sedimentary structures. Locations
of photographs that follow are indicated by their letter to left of column. (B) Magadi-type chert bed interbedded with dark siltstone near
the base of the Las Salinas Formation (bed is vertical with top to the right or northeast). Note two dark stains (at arrows) which are oil
seeps derived from the chert bed. Oil seeps and soft-sediment deformation features are typical of this type of chert. (C) Vertical burrow
of Skolithos (?) with structureless fill. This bed is both overlain and underlain by beds with symmetrical ripple marks. (D) Symmetrical
wave ripple marks with continuous crestlines on a vertical bed at the base of shallow-marine beds of the Las Salinas Formation 2 m
above the limestone of 'Razorback Ridge' in Arroyo del Pozo. Note that the troughs of the ripples are filled with sand and the round
exit hole of Skolithos (?) in right-center of photo. (E) Resistant grainstone forming 'Razorback Ridge', a highly resistant bed forming the
approximate boundary between the evaporitic beds of the Angostura Formation and shallow-marine beds of the Las Salinas Formation
(see Figs. 18A and 19A for locations). Color photographs at http://www.elsevier.nl/locate/caribas/
base of the lower part of the Las Salinas Formation in lower half of the Las Salinas Formation correlates
outcrop (unit 5) may correlate to a 30-m-thick inter- to a 4387-m-thick section of 8-60-m-thick beds of
val of shale capped by a 6-m-thick bed of anhydrite at sandstone interbedded in shale between 2157 and
2595 m in the Charco Largo-1 well. In their study of 2595 m in the well. Some of the thicker sandstone
the Charco Largo-1 cuttings, McLaughlin and Bold beds may form the prominent reflectors visible in the
identified shallow-marine microfauna from unit 4. Las Salinas Formation on seismic line 111 across the
The anhydrite may mark a drying cycle in the basin at area near the well (Fig. 9).
the close of the marine interval of marked by unit 4. Correlations between the upper part of the Las
The 390-m-thick section of trough-cross bedded Salinas Formation in outcrop and in the section
sandstone and interbedded siltstone of unit 6 in the above the sandstone interval in the Charco Largo-1

well become more tenuous because of the basinward are consistent with the study by McLaughlin et al.
thickening of units, as is evident from the seismic (1991): (1) Upper Miocene-Lower Pliocene sed-
data, and loss of exposures at the northern end of imentary rocks record a marine-continental transi-
the Arroyo del Pozo outcrop. In general the upper tion that is slightly older than the marine-continental
part of the Las Salinas Formation (units 7, 8, 9) fines transition seen further south in the valley of the Rio
upward in the Arroyo del Pozo while the Las Salinas Yaque del Sur; and (2) sand and gravel composi-
Formation above 1220 m depth in the well coarsens tions record erosion of Upper Cretaceous-Eocene
upward. It is possible that a 3-m-thick limestone island-arc rocks of the Cordillera Central.
bed reported in the well at a depth of 2157 m may
correlate to one of the two 1-m-thick coquina beds Structure
in unit 7. Because coral clasts are not distinguished
on the well log, it is difficult to attempt correlations Resistant beds of Upper Miocene sandstone and
between unit 8 and the well. conglomerate of the upper Trinchera Formation form
the eastern edge of the meander belt of the Rio
Yaque del Sur (Figs. 6, 23). Rocks of the Pliocene
SURFACE GEOLOGY OF THE NORTHEASTERN Quita Coraza Formation occupy the meander belt of
MARGIN OF THE ENRIQUILLO BASIN the river because they are easily erodible and lack
resistant sandstone or conglomerate interbeds. Re-
Stratigraphy sistant beds of calcarenite and sandstone beds of the
lower Arroyo Blanco Formation form the northwest-
The northeastern margin of the Enriquillo basin in ern edge of the meander belt (Fig. 23). Dips on beds
the valley of the Rio Yaque del Sur consists of north- in all formations vary between 20 ~ and 45 ~ (Mann
to northwest-dipping beds of four sedimentary rock et al., 1991a,c). Several minor high-angle faults cut
formations: the Trinchera, Quita Coraza, Arroyo all three formations but none have significant offset
Blanco, and Arroyo Seco (Fig. 23). The formations (Mann et al., 1991 a).
crop out on the southern limb of the Los Guiros Planar erosion of the Arroyo Blanco Formation
syncline, a major northeast to east-west-trending in Quaternary time by the meandering Rio Yaque
syncline separating the anticlinal Sierra Martin Gar- del Sur has produced excellent exposures of strike
cfa to the south from the anticlinal Sierra de Neiba ridges formed by resistant gypsum, sandstone, and
to the north (Mann et al., 1991a) (Fig. 6). conglomerates (Fig. 23). In a few areas, gravel-
The four formations have been previously corre- and cobble-covered river terraces less than 5 m in
lated by McLaughlin et al. (1991) with outcrops in thickness cover the strike ridges. Because the rocks
the Loma Sal y Yeso and in the Azua and San Juan of the Quita Coraza and Arroyo Blanco Formations
basins (Fig. 7). The facies and micropaleontology of are tilted north and northeast, they present a natural
outcrops of the Trinchera and Quita Coraza Forma- outcrop section across the Early Pliocene north-
tions in the valley of the Rio Yaque del Sur have eastern shallow-marine margin of the Enriquillo
been studied in detail by McLaughlin et al. (1991). basin (Fig. 23). Detailed interpretations of large-
This study concluded that: (1) the two formations scale trends in strike ridges can be integrated with
represent an Upper Miocene-Lower Pliocene ma- surficial mapping and measured sections to recon-
rine sequence which shallows upward from a middle struct the geometry of the margin. Less resistant
bathyal to inner neritic environment; (2) coarser- lithologies which form strike valleys separating the
grained facies in the Trinchera Formation represent resistant ridges are poorly exposed or are covered.
a southwest-prograding submarine fan derived from
the erosion of Upper Cretaceous-Eocene island-arc Facies associations of the Arroyo Blanco and
rocks of the Cordillera Central; and (3) finer-grained Arroyo Seco Formations
facies in the Quita Coraza Formation represent shelf
and upper slope deposits. For the purpose of this study, sedimentary rocks
Sedimentary facies in the overlying shallow- of the Arroyo Blanco Formation and the overlying
marine and non-marine rocks of the Arroyo Blanco Arroyo Seco Formation are grouped into four fa-
and Arroyo Seco Formations have only been mapped cies associations shown on the map in Fig. 23A:
in a reconnaissance manner by Cooper (1983) and (1) a deep-marine facies association; (2) a shal-
form the basis of the detailed description that follow-marine facies association characterized by the
lows. Studies of the sedimentary structures, sand- presence of gypsum beds; (3) a shallow-marine fa-
stone petrography and conglomerate clast counts of cies association characterized by the presence of
these two formations have been carried out 50 to oolitic limestone; and (4) a non-marine facies asso-
100 km to the north in the San Juan basin by Harms ciation. The first three facies associations alternate
(1989). The main results of the Harms (1989) study in the 700-930-m-thick Arroyo Blanco Formation

OO

Fig. 23A. Schematic map of major groups of sedimentary facies along the eastern margin of the Enriquillo basin based on aerial photograph interpretation and field mapping. Key to numbered facies groups"
unit 1 -- massive marine siltstone of Lower Pliocene Quita Coraza Formation; units 2 through 8 = shallow-marine oolite and gypsum facies associations of the Lower Pliocene Arroyo Blanco Formation;
unit 9 = non-marine facies association of the Upper Pliocene-Pleistocene (?) Arroyo Seco Formation. Letters identify resistant ridges of gypsum, oolitic limestone, or calcarenite-coral debris flows that are
used to correlate the measured sections which are identified by name.
Z
O
C~
>
Z
>
-]
:Z
ZZ
~Z
0
u~

>
Fig. 23B. Example of aerial photograph used as a base map for the map shown in (A) (photograph number 721, ICM-I(58)-13-V14 12-29 December 1958, original scale: 1 "60,000).
',D
320 E MANN et al.
while the fourth association composes the entire The prominent change in strike of bedding from
1000-m-thick Arroyo Seco Formation (Fig. 23). northeast to northwest near the Enriquillo basin is
related to bending about a vertical axis associated
Water depths with the formation of the Los Guiros syncline (Mann
et al., 1991a) (Fig. 6). The Quita Coraza Formation
The use of the terms 'shallow-marine' and 'deep- (unit 1) and oolite and gypsum shallow-marine fa-
marine' are relative. There are few constraints on cies associations (units 2, 4, 5, 7, and 8) thin towards
the absolute water depth in which each of the four the northeast. The lower deep-marine facies associ-
facies associations shown on the map in Fig. 23A ation (unit 3) thickens to the northeast. The wedge
was deposited. McLaughlin et al. (1991) determined geometry of the upper deep-marine facies associa-
that benthic foraminifera identified in samples taken tion (unit 6) is unclear because of the thinness of the
from the base to the top of the Quita Coraza For- unit and poor exposure.
mation in the map area shown on Fig. 23 indicate The lower deep-marine facies association (unit 3)
a transition from an outer to inner neritic setting. contains resistant calcarenite and coral debris flows
Benthic foraminifera in the lower part of the for- which form prominent southwest-dipping clinoforms
mation suggest depths greater than 100 m; those (Fig. 23A). The clinoforms prograde upward in the
in the upper part suggest depths less than 100 m. section from northeast to southwest. To the west, the
Microfauna are rare in the Arroyo Blanco Formation coral debris flows and calcarenite horizons lose their
and estimates of water depth must rely on analogies clinoform shape and become planar.
to modem environments of deposition for oolites The contacts between facies associations appear
and gypsum. The main field criteria for deep-marine parallel-sided in all but one case. On detailed aerial
facies association are the presence of marine fos- photographs, the eastern part of unit 6 clearly trun-
sils, gravity-driven sedimentary structures, such as cates the eastern part of unit 5.
graded bedding, and the absence of gypsum and
oolitic limestone. Measured sections in the Arroyo Blanco
Formation
Subdivision of the Arroyo Blanco Formation
In order to document the nature of the seven units
In order to facilitate description and interpre- of the Arroyo Blanco Formation, five measured sec-
tation, the Arroyo Blanco Formation in the map tions were measured over an along-strike distance of
area shown in Fig. 23 is subdivided into six units 20 km (locations of sections shown on Fig. 23A).
which are numbered 2-7 on Fig. 23A (unit 1 is the Sections were measured using a Brunton compass
marine Quita Coraza Formation and unit 8 is the mounted on a Jacob's staff. Strike and dip of bed-
non-marine Arroyo Seco Formation). Units 3 and ding were frequently measured to insure that the true
6 are deep-marine facies associations characterized thickness of the section was measured. Where bed-
by the presence of marine fossils, units 2, 5, and 7 ding was not exposed, regional strike was estimated
are shallow-marine facies associations characterized using the strike of resistant ridges on aerial pho-
by the presence of gypsum, and units 4 and 7 are tographs. Most of the sections generally follow a ma-
shallow-marine facies associations characterized by jor dry stream bed. The areas outside the stream beds
the presence of oolitic limestone. The units alternate are covered by a semiarid, dry forest-type of vegeta-
in three shoaling-upwards cycles: tion characterized by mesquite, low trees, and cactus.
(1) deep-marine facies association of the Quita In the E1 Granado, Barrero, and Honduras-Facolina
Coraza Formation (unit 1) - - shallow-marine gyp- sections, vegetation or cover in the stream bed or
sum facies association (unit 2); parallelism of the stream bed to regional strike re-
(2) deep-marine facies association (unit 3) quired that the section be measured out of the stream
shallow-marine oolite facies association (unit 4) bed. All of the sections follow a single direction
shallow-marine gypsum facies association (unit 5); across strike, except for the Honduras section which
(3) deep-marine facies association (unit 6) is offset 1 km along-strike to the northeast to the
shallow-marine oolite facies association (unit 7) Facolina section (Fig. 23A). The section was offset
shallow-marine gypsum facies association (unit 8) because Arroyo Honduras forms a heavily vegetated
non-marine facies association (unit 9, Arroyo 'box canyon' with few outcrops. Well-used trails
Seco Formation). follow all of the dry stream beds and penetrate many
of the densely vegetated areas between the streams.
Stratal geometry Correlation between stream sections is based on
the identification of lithologically distinct and resis-
Mapping and interpretation of aerial photographs tant beds of gypsum, calcarenite, and oolitic lime-
indicates that units 1-8 are wedge-shaped (Fig. 23). stone which can be traced on aerial photographs

and walked out in the field (beds A - J labelled on Marker beds A and B
Fig. 23A). Distinctive gypsum types in the sections Roadcuts at the base of the section expose 2 -
are correlated in the chart on Fig. 24 over along- 5-cm-thick beds of gray-brown mudstone with in-
strike distances of about 13 km. terbeds of 2-3-cm-thick beds of siltstone or very
The tops of the gypsum ridges are not covered fine-grained sandstone and 2-cm-thick white clay
with soil or vegetation and form elevated walkways layers. Coarsely crystalline gypsum layers up to 3
through otherwise heavily vegetated areas. In the cm thick occur as fibrous, secondary overgrowths
lower deep-marine facies association (unit 3), three along bedding planes. The 35-m-thick section above
distinct and resistant beds of calcarenite named X1, the roadcut forms a grass-covered slope. We infer
X2, and X3 could be correlated over a distance of that this section forms the top of unit 3 of the
5 km between the Barranca and Honduras-Facolina deep-marine facies association (Fig. 23A).
sections (Fig. 23A). In the shallow-marine gypsum The first resistant units in the E1 Granado sec-
and oolite facies associations, distinct and resistant tion are two marker beds (A and B in Fig. 23A) of
beds of gypsum and oolite could be traced for 10 oolitic limestone ranging from 2 to 3 m in thickness
km from the E1 Granado section to the Honduras- (Fig. 25B). The lower 60 cm of bed A consists of mas-
Facolina section where the units pinch out on the sive oolitic limestone which is buff-white on broken
underlying marine section (unit 3) (Fig. 24). surfaces and weathers to a gray color. Parallel lami-
nations are seen at the base of this unit. The overly-
ing, 60-cm-thick unit consists of flaggy, 4-6-cm-thick
beds of calcarenite with multiple sets of symmet-
rical, straight-crested ripples with heights of 5 mm
El Granado section and trough-to-trough lengths of 3 - 6 cm. The base of
this unit is marked by a 2-cm-thick layer of marine
Location shell hash. The outcrop is pockmarked with cavities,
The E1Granado section is 143 m thick (Fig. 25A) or oomolds, formed by the solution of oolites which
and lies within unit 3 of the deep-marine facies as- are preferentially deposited in the troughs of the rip-
sociation, unit 4 of the shallow-marine oolite facies ple marks (Fig. 25B). Close inspection reveals the
association, and unit 5 of the shallow-marine gyp- presence of thin remnant coatings of undissolved and
sum facies association (Fig. 23A). The E1 Granado uncemented oolites within some of the cavities. The
section can be reached by driving north from San- orientations of 30 ripple marks in both beds A and
tana on the road to E1 Granado. Because there is no B suggest a N E - S W wave direction (Fig. 25A). The
named stream nearby, the section was named after uppermost unit of bed A consists of a 140-cm-thick
E1 Granado, the nearest village. bed of massive, weathered fine-grained calcarenite
Fig. 24. Along-strike thickness and textural variations in single horizons of gypsum in the Rfo Yaque area of the eastern Enriquillo basin.
Over a distance of 13 km, gypsum beds thin and pinch out to the northeast and thicken toward the center of the Enriquillo basin to the
southwest (location of this area relative to basin center is shown on Fig. 13).

Fig. 25. (A) E1 Granado measured section. See Fig. 20 for key to lithology and sedimentary structures. Vertical scale is in meters.
Locations of photographs that follow are indicated to left of column. (B) Oolitic limestone 40 m above the base of the E1 Granado section
at the E1 Granado-Santana road (marker bed A on map in Fig. 23A). Lenticular dark cavities in the center of the outcrop are formed by
solution of oolites which preferentially fill troughs of cross-beds. Scale is 10 cm long. (C) Photomicrograph of oolitic limestone from
horizon A in the E1 Granado section (outcrop is shown in A). The rock is an oomoldic sparite with few of the original ooids remaining.
Note the increase in size of the spar cement from the ooid molds into the primary pore space. Dimensions of the field of view is
approximately 4 mm by 6 mm. Color photographs at http://www.elsevier.nl/locate/caribas/

with horizontal burrows on the tops of some bedding channel is overlain by beds of fine-grained litharenite
planes. The rock is massive and has a mottled appear- exhibiting hummocky cross-bedding (Fig. 26A).
ance which appears to be related to pervasive burrow- Marker bed C, unit 1. This consists of a 2-m-
ing. Thin sections reveal that the massive parts of the thick section of gypsum and sandstone. The lower
rock are now supported by secondary dolomite ce- gypsum unit consists of three 4-10-cm-thick beds
ment which formed between the oolites prior to their of primary, coarsely crystalline gypsum layers of the
dissolution (Fig. 25C). grass-type interbedded in gray mudstone containing
Limestone marker bed B is 10.2 m above the well-preserved leaf fragments (Fig. 26E).
top of bed A and consists of 1 m of ripple-marked Marker bed C, unit 2. This consists of a 1-m-
oolitic limestone (Fig. 25A). The interval between thick section of massive, spheroidally weathering
marker beds A and B is covered and assumed to be gray mudstone with molds of gypsum crystals which
gray-brown litharenite of bed B 1, which is seen in resemble three-pronged stars.
outcrop and in float in other areas along strike. Marker bed C, unit 3. This consists of a 1-m-
thick ledge-forming bed of gypsum (Fig. 26A,B)
Contact between marker beds B1 and C composed of 2-3-cm-thick beds of gypsum growing
The first gypsum bed (bed C) occurs 28 m above on laminae of a slightly muddier gypsum crystal hash
the upper oolitic limestone of bed B (Fig. 25A). (average crystal length is ~ 1 cm) (Fig. 26C,D). The
Sandstone of bed B1, which separates limestone of upper 50 cm of the bed contains at least 20 horizons
beds A and B from gypsum of bed C, is covered in of well-organized and continuous layers of grass-type
the line of the measured section but is well exposed gypsum (term of Schreiber et al., 1976) (Fig. 26C).
in a stream cut of the Arroyo de la Salvia near Bed C1. Bed C1 consists of fine- to medium-
the crossing of the Santana-E1 Granado road 200 m grained brown litharenite which is 11.9 m above the
to the west of the line of the measured section. A top of the gypsum of bed C. The sandstone forms
detailed section was measured at the contact on the a low cliff on the side of a ridge but does not
road (Fig. 26A). form a prominent strike ridge similar to the gypsum
and limestone marker beds B and C. The total ex-
Sandstone-gypsum contact in Arroyo de la posed thickness is 5.1 m. The geometry of the entire
Salvia outcrop suggest a 20-m-long and 5-m-high channel
The Arroyo de la Salvia outcrop can be divided incised into covered rocks assumed to be mudstone
into seven distinct lithologic units which are well or siltstone. The lower 3 m of bed C 1 consists of two
exposed in the streamcut (Fig. 26A). The first two fining-upward cycles composed of basal units of grit
units are part of bed B while the upper five units are overlain by large trough cross-beds (20 cm high and
part of bed C. 80 cm wide) in medium- to fine-grained litharen-
Marker bed B2, unit 1. This consists of par- ite which are, in turn, overlain by better-sorted,
allel-sided, 5-20-cm-thick, poorly sorted, brown, medium-grained litharenite with plane-parallel lami-
calcite-cemented litharenite separated by interbeds nations. Measurements of the long axis of the trough
of brownish-gray siltstone (Fig. 26A,B). The sand- cross-beds suggests N E - S W paleoflow. The top 2 m
stone layers are intensely bioturbated and mixed of the outcrop consists of medium- to fine-grained
with the over- and underlying siltstone. The bur- litharenite with large trough cross-beds.
rows are stained red with hematite cement. Sandy Marker bed D. This bed is 5 m above bed C 1 and
beds thicken to 10-20 cm near the top of the unit consists of a 10-cm-thick bed of coarsely crystalline
and consist of 1-2-cm-thick beds of ledge-forming gypsum overlying fine-grained muddy sandstone. In
sandstone alternating with mudstone. Sandy beds are views parallel to bedding, the gypsum consists of a
clearly truncated by a channel of the overlying unit mosaic of sutured stylolitic contacts between square
2. Sandy layers of unit 1 contain minor fine-grained crystals 3 • 3 cm in size. The gypsum is similar
fossil hash and some well-preserved, very small (--~2 to facies 3 (massive selenite) described by Vai and
mm) gastropod and pelecypod shells along with one Ricchi Lucchi (1977; their fig. 13) in the Messinian
internal mold of a gastropod. Vena del Gesso basin of the northern Apennines.
Marker bed B2, unit 2. This consists of a A similar, 5-cm-thick bed is present 2 m above
2-m-thick channel which is 2.4 m deep and 5 m the lower gypsum horizon. This gypsum horizon
wide (Fig. 26B). The base of the channel is filled is overlain by a 1-m-thick section of platy, black
by 5-10 cm of a poorly sorted muddy grit with mudstone, or 'paper shales'.
pebbles averaging 5 mm in diameter and up to 4 Marker bed E. This bed is a 4.91-m-thick gyp-
cm in maximum diameter. The lag is thickest at the sum bed 23 m above the top of the upper gypsum
base of the channel and thins to the walls of the horizon of horizon D. The gypsum forms a promi-
channel. The upper part of the channel is filled with nent ridge and is composed of crystals ranging in
a brown fine- to medium-grained litharenite. The length from 1 to 2 cm.


Interpretation of El Granado section muddy siltstone with finely broken, 1-3-mm-size

The section is interpreted as a shoaling-upward shell hash. Paleocurrent measurements were made in
cycle which begins with shallow-marine oolitic two 1-m-thick debris flows where the long axis of
limestone (beds A, B, B1), passes upwards into branching coral clasts show a consistent orientation.
a more restricted evaporitic interval (beds C, D, The paleoflow was from northwest to southeast.
E) (Fig. 23A). The sandstone bed B2 marks the The top of the coral debris horizon is gradationally
boundary between the shallow-marine oolite and overlain by 6 m of massive siltstone, containing
gypsum association and probably reflects the regres- scattered bivalves, some of which appear in place.
sion responsible for this transition. The C 1 sandstone Marker bed A. Bed A of unit 4 is 132 m above
bed within the gypsum section probably records a the top of the coral debris horizon and consists of
shorter-lived transition from marine to evaporitic 60-cm-thick beds of buff-white limestone. Bed A
environments. can be traced on air photos to the oolitic lime-
stone outcrops of bed A in the E1 Granado section
Arroyo Barrero section (Fig. 24). The limestone is flaggy, composed of 5 -
15-cm-thick beds, and has a gray, micritic matrix
Location containing 1-5-mm-long gastropods and bivalves.
The Arroyo Barrero section is 750 m thick and Marker bed B. Bed B of unit 4 is 4.9 m above
crosses unit 4 of the shallow-marine oolite facies the top of bed A. The basal unit of consists of 40 cm
association, unit 5 of the shallow-marine gypsum fa- of recrystallized, flaggy oolitic limestone. Ghosts of
cies association, unit 6 of the deep-marine facies as- oomolds similar to those observed in the E1 Granado
sociation, unit 7 of the shallow-marine oolite facies section are present (Fig. 25B). Individual limestone
association, unit 8 of the shallow-marine gypsum fa- beds are 2 cm thick.
cies association, and unit 9 of the non-marine facies Marker bed C. The limestone bed is directly
association (units 4 - 8 comprise the Arroyo Blanco overlain by 3.2 m of a poorly exposed and altered
Formation while unit 9 belongs to the Arroyo Seco gypsum bed with remnant crystals up to 2 cm in
Formation) (Fig. 23A). The Arroyo Barrero section length that is correlated to bed C. Most of the
can be reached by driving north on dirt access roads gypsum is altered to a fine, white powder. The
through sugar cane fields north of Santana. The sec- gypsum is overlain by 2 m of mudstone which is
tion follows Arroyo Barrero for most of its length. in turn overlain by a 1-cm-thick limestone bed and
The main features of the complete measured section a 10-cm-thick bed composed of gypsum crystals
shown in Lamar (1990) are summarized below and disseminated in mudstone. Solution of the crystals
shown on the gypsum correlation chart in Fig. 24. produces a 'worm-eaten' texture in the mudstone.
Marker bed C1. Bed C1 of unit 5 consists of a
Description of the Arroyo Blanco Formation poorly exposed layer of greenish-brown litharenite
Base of the section. The base of the section is 25 m above bed C. The exact thickness of this
marked by a low ridge composed of a 10-m-thick layer is unknown because it crops out as loose float
section of reworked coral reef and shell rubble. The blocks of sandstone on a slope maintained by the
unit is well exposed in a cut made for an irrigation overlying resistant gypsum bed D. The lithology of
canal near the mouth of Arroyo Barrero. Bedding the float blocks is very similar to the bed C1 in the E1
in the unit thickens upward from 6-8-cm-thick beds Granado section (Fig. 25). None of the float blocks
near the base to 2-m-thick beds in the upper part of sandstone is greater than 30 cm in thickness.
of the outcrop. The beds are made up of 9 0 - 9 5 % Marker bed D. Bed D of unit 5 consists of
branching coral fragments which average 3-5 cm in a 3-m-thick bed of massive, altered selenite. Al-
length and 1 cm in diameter. No corals in growth teration to a white powder is most intense on the
position were observed, although several articulated upper surface of the bed and along joints and frac-
bivalves were found. Calcite is present in the pore tures in the lower part of the bed. The selenite
spaces of shells. The matrix is a poorly sorted, is composed of 2 x 2 cm to 3 x 3 cm square
Fig. 26. (A) Detailed measured section of the contact between marker horizon B 1 and horizon C of the outcrop shown in (B). See text
for description of the seven lithologic units. Locations of photographs that follow are indicated to the left of column. (B) Photograph
of outcrop showing contact between horizon B1 and horizon C at outcrop near the crossing of Arroyo de la Salvia and the E1
Granado-Santana road at approximately the same scale as the measured section of this outcrop in (A). (C) View looking perpendicular to
bedding in grass-type gypsum from horizon C in the E1 Granado section (unit 5 on measured section in A). Top of bed is towards top of
photograph. Note the even banks of crystals which can be traced across the outcrop. (D) View looking down on a bedding plane surface of
grass-type gypsum from same horizon as the bed shown in (C). Note fine needles of gypsum suspended in mud matrix. Diameter of coin
is 17 mm. (E) Well-preserved leaf fragment found in siltstone bed (unit 3). Leaf and other plant fragments are probably well-preserved
because of high salinity of the brine pools in which they were deposited. Color photographs at http://www.elsevier.nl/locate/caribas/

326 E MANN et al.
crystals when viewed perpendicular to the bedding Marker bed J. Bed J of unit 8 is 81 m above
plane. Immediately overlying the top of bed D are bed I and consists of a 2-m-thick of very coarsely
at least two beds of poorly exposed limestone which crystalline massive selenite with crystals up to 3 x
form a small knob of the resistant gypsum ridge. 3 cm in size. The crystals are disorganized and may
The two limestone beds which are recrystallized and be reworked as a 'gypsrudite'. Bed J is overlain
coarse-grained may have originally been shell hash. by brown sandstone and mudstone. Bed J is only
Marker bed E. Bed E of unit 5 is 72 m above 100-200 m below non-marine rocks of the Arroyo
the top of bed D and consists of a 110-cm-thick Seco Formation (Fig. 23A).
bed of massive, altered selenite composed of 1 x 1
cm crystals. The slope beneath bed E is covered but Interpretation of Arroyo Barrero section
appears to be weathered, massive gray mudstone. The section is interpreted as two shoaling-upward
Marker bed F. Bed F of unit 6 is 85 m above cycles (Fig. 23A). The base of the section is marked
bed E and consists of a 1-m-thick interval of poorly by a coral debris horizon of unit 3 which can be
exposed, flaggy oolitic limestone composed of 2 - correlated on aerial photographs to resistant clino-
5-cm-thick beds. The base of the bed appears to be forms of calcarenite and coral debris beds to the
bioturbated. Most of the limestone is recrystallized northeast. The flatness of the horizon in this area
but casts of oolites about 1 mm across are visible (Fig. 23A) suggests that it may have occupied a
on unweathered surfaces of beds about 1.5 cm in basinward position relative to the slope clinoforms.
thickness. Also present is a 2-cm-thick bed of micrite This deep-marine facies association of unit 3 shoals
containing fossil hash about 2 mm in diameter. rapidly upward into a shallow-marine facies associ-
An unnamed bed above bed F consists of a ation defined by oolitic limestone in marker beds A
15-cm-thick bed of medium-grained brown litharen- and B of unit 4. As in the E1 Granado section, these
ite overlain by a 30-cm-thick bed of silty micrite limestones shoal upward into a shallow-marine gyp-
containing fossil hash. These two beds are not trace- sum facies association defined by the gypsum beds
able on aerial photographs. C, D, and E of unit 5. Marine limestone of bed D and
Marker bed G1. Bed G of unit 7 consists of beds marine siliciclastic rocks of bed C1 suggest that the
G1 and G2, and, together with bed H, forms a 3-m- gypsum association is actually composed of smaller
thick triplet of beds occurring in rapid succession cycles defined by: (1) marine regression (sandstone
(Fig. 23A). Bed G1 is 88 m above the top of bed F channels); (2) gypsum and associated mudstone;
and consists of a 42-cm-thick marly coquina made and (3) marine transgression (oolitic limestone with
up of four bedding units. The lower unit of bed G1 is marine shells). However, the overall trend is shoal-
15-cm-thick and consists of a sandy mudstone con- ing-upwards.
taining marine shells. The unit is intensely biotur- The shallow-marine gypsum association of unit
bated by 75-mm-wide round burrows. The overlying 5 is overlain by a 'deep-marine' facies association
unit is 17 cm thick and contains marine shells and defined by the dominance of shallow-marine oolitic
load casts up to 2 cm deep. This unit is overlain by a limestones present in marker bed F of unit 6. The
2-cm-thick mudstone layer and an 8-cm-thick poorly 'deep-marine' interpretation here is used in a relative
sorted calcareous siltstone. The shell units appear to sense to distinguish this predominantly marine lime-
be rapidly deposited 'tempestites' or storm deposits stone section from the underlying predominantly
which were later subjected to bioturbation. gypsum section of unit 5. Moreover, gypsum bed E
Marker bed G2. Bed G2 of unit 7 is 1 m above of unit 5 is clearly truncated on aerial photographs
bed G1 and consists of a 70-cm-thick bed of very by limestone bed F of unit 6 to the east of Arroyo
well sorted, brown litharenite. The bed is composed Barrero (Fig. 23A). This is interpreted as a marine
of two 20-30-cm-thick beds which are laminated transgression of marine limestone bed F of unit 6
in 2-cm-thick partings and contain possible trough over the underlying gypsum of unit 5. The regression
cross-beds. is marked by sandstone G2 and the overlying gyp-
Marker bed H. Bed H of unit 7 is 1 m above bed sum bed H which marks the base of the overlying
G2 and consists of a 60-cm-thick bed of weathered shallow-marine gypsum facies association of unit 8
gypsum. The upper part of the bed contains 1-cm- (Fig. 23A).
thick mud laminations which suggest that the bed
was originally a grass-type gypsum. Arroyo Barranca section
Marker bed I. Bed I of unit 8 is 85 m above bed
H and consists of a weathered gypsum horizon about Location
1-2 m thick which is altered to a white powder. The Barranca section is 930 m thick and in-
Some loose crystals up to 2-3 cm in diameter are cludes three deep-marine facies associations, three
present as float and suggest that bed I was originally shallow-marine gypsum facies associations, and two
a massive selenite. shallow-marine oolite facies associations of the up-

Fig. 27. Arroyo Barranca measured section. See Fig. 20 for key to lithology and sedimentary structures. Vertical scale is in meters.
Locations of some of the photographs in Figs. 28 and 29 are indicated in the left margin of the section.
permost Quita Coraza Formation and the Arroyo Arroyo Barranca. The upper part of the section can
Blanco Formation (Fig. 27). The base of the section be reached by walking along the Arroyo Barranca to
begins at the top of the Quita Coraza Formation at the northwest.
the bluff along the west bank of the Rfo Yaque del
Sur and can be reached by driving to the bridge Upper Quita Coraza Formation
across the Arroyo Barranca on the Tamayo-Vuelta The base of the section consists of 15 m of
Grande dirt road and walking southeast along the turbiditic sandstone and minor conglomerate over-

328 E MANN et al.
lying the massive bluish siltstone which makes up cm in thickness which are similar to those seen at
most of the Quita Coraza Formation (Fig. 27). The Barrero. Average coral size is 5-10 cm long and
section consists of massive to trough cross-bedded 2-3 mm thick. Cementation of the uppermost 1 m
beds, 20 cm to 3 m thick, of fine- to medium- of the bed gives a coarsening-upward appearance to
grained litharenite, grit, and conglomerate. Clasts the outcrop. Coral fragments in this bed showed a
in conglomerate consist of volcanic, plutonic, and preferred alignment, presumed to be the result of
sedimentary rocks derived from Cretaceous-Eocene transport. Paleocurrent studies on this bed suggests
rocks of the Cordillera Central. Sandstone beds ex- north-to-south paleoflow.
hibit mm-scale laminations defined by variations in Marker beds X1, X2, and X3. Beds X1, X2,
grain size and normal grading. The sandstone ap- and X3 of unit 3 are flaggy, graded, well ce-
pears darker in color than the sandstones seen higher mented grainstones which occur 50-90 m above
in the Arroyo B lanco and in the Las Salinas Forma- the coral conglomerates described above. The three
tion to the south because of a much higher content of beds range in thickness from 40 to 60 cm but form
lithic grains (up to approximately 50% as estimated resistant strike ridges which can be traced on aerial
visually in the field). photographs from Arroyo Barranca for several kilo-
meters to the east and west (Figs. 23A and 29A).
Description of the Arroyo Blanco Formation Handsample examination of the grainstones shows
Base of section. Marine rocks of the Quita Coraza bioclastic and limestone grains. Lower contacts may
Formation are overlain by the lowest of the shallow- have been slightly scoured but show no extreme
marine units of the Arroyo Blanco, a 30-cm-thick incision (Fig. 29A).
alabastrine gypsum bed cropping out in a roadcut Marker bed C. The overlying shallow-marine
on the Tamayo-Vuelta Grande road. This outcrop interval of unit 5 includes gypsum marker beds C
is 160 m above the uppermost exposed beds of the and D (Figs. 23A and 24). Gypsum marker bed E
Quita Coraza Formation (Fig. 27). A second gypsum of unit 5 from Arroyo Barrero is not present and
bed crops out 90 m above the lower bed and consists is seen to be truncated by the limestone bed F of
of a 9-m-thick bed of grass- and spear-type selen- marine unit 6 on aerial photographs (Fig. 23A). The
ite. These two gypsum beds and their associated gypsum at marker bed C consists of 4.32 m of
deposits are significant because they mark the first grass-type (Fig. 28B) and bladed selenite (Fig. 28C).
shallow-marine conditions of unit 2 in the Neogene Beds of the grass-type gypsum are typically 0.5-1
history of the Enriquillo basin (Fig. 23A). Both gyp- m thick and are found near the base of the out-
sum beds are seen to pinch out on aerial photographs crop. The vertically oriented crystals are 1-3 cm in
and are not present in streams to the northeast of length (Fig. 28B). Beds of bladed selenite overlie the
Arroyo Barranca (Fig. 23A). grass-type gypsum, are 1-2 m thick, and are com-
The deep-marine facies association of unit 3 over- posed of stacks of smaller beds with the thickness of
lying the gypsum association of unit 2 consists the vertically oriented crystals, 20-40 cm in length
of poorly exposed massive siltstone, thick trough (Fig. 28C). These are the longest crystals of gyp-
cross-bedded sandstone (Fig. 28A), calcarenite and sum observed anywhere in the study area (Fig. 24).
conglomerate composed mainly of coral and shell Above gypsum C is a 12-m-thick section of massive
fragments (Fig. 27). The sandstone consists of brown mudstone and siltstone (Fig. 27).
litharenite composed of fine- to medium-grained, Marker bed D. This unit consists of a 1-m-thick
well-rounded quartz, volcanic, and carbonate grains. bed of jumbled, 1-3-cm-long gypsum crystals in
Sandstone beds range from 20 to 50 cm and are a buff, muddy matrix which is 40 m above bed
interbedded with gritty siltstone. Trough cross-beds C (Fig. 27). As in Arroyo Barrero, gypsum D is
were measured in the more massive sandstone beds abruptly overlain by a flaggy limestone (bed F).
and indicate N E - S W paleoflow. There is one thick Marker bed E As in Arroyo Barrero to the east,
(8 m) conglomerate in the sandy section at 350 m. the base of the overlying 'deeper-marine' interval of
The conglomerate has a highly scoured, erosional unit 6 is taken as the oolitic limestone of bed F as
lower contact. The matrix is a fine- to medium- in the Arroyo Barrero (Fig. 23A). Gypsum bed E is
grained sandstone which supports clasts ranging in truncated by limestone bed F between Arroyo Bar-
size from granules to boulders 20 cm to 30 cm in rero and Arroyo Barranca. Beds F and G, in turn, are
diameter. not found in streams to the northeast of Arroyo Bar-
The coral debris at 475 m in unit 3 consists of a ranca because the 'deeper-marine' interval onlaps
3-m-thick bed composed of 90% coral-stick rubble the underlying shallow-marine interval and pinches
(Fig. 29B). The coral does not appear as abraded out (Fig. 23A). This onlap is consistent with a thin-
as coral rubble in the Barrero section 4.9 km to ning of the 'deeper-marine' interval between Barrero
the west (Fig. 23A). The deposit is clast-supported and Barranca. Bed F is a flaggy limestone about 3 m
in a sandy mud matrix and occurs in beds 5-10 thick which shows oomolds in hand sample.

N E O G E N E EVAPORITIC AND S I L I C I C L A S T I C D E P O S I T I O N IN THE E N R I Q U I L L O BASIN 329
Fig. 28. Photographs of different siliciclastic and evaporitic facies at 380 m in the Arroyo Barranca measured section (see Fig. 27 for
section and locations). (A) View of light-colored, resistant coral limestone conglomerate bed overlying less resistant, massive trough
cross-bedded sandstone in marine unit 3 in the Arroyo Barranca section (compare with regional map of units in Fig. 23A). (B) Grass-type
gypsum of marker bed C in evaporitic unit 4 in the Arroyo Barranca section. Top of bed is towards top of photograph and pencil is 14
cm long. (C) Spear-type bladed and twinned selenite of evaporitic unit 4 in the Arroyo Barranca section. Top of bed is towards top of
photograph. Length of crystals represents water depth in modern settings of bladed selenite such as Lake MacLeod, Australia. (D) Bed of
unsorted gypsrudite of evaporitic unit 8 in the Arroyo Barranca section suggesting nearby erosion of a coarse-grained selenite. Pencil for
scale is 14 cm long. (E) Poorly sorted conglomerate composed of well rounded pebbles and cobbles and devoid of marine shells. This
unit is taken as the base of the non-marine section (unit 9, Arroyo Seco Formation) in the Arroyo Barranca section. Pencil is 14 cm long.
Color photographs at http://www.elsevier.nl/locate/caribas/

330 R MANN et al.
Marker bed G. Bed G is a 1.5-m-thick grain-

stone containing small intraclasts of broken shell
material.
Marker bed I. The highest gypsum bed (I) in the
overlying shallow-marine interval of unit 8 contains
a 15-cm-thick bed of gypsrudite (Fig. 28D) similar
to that described by Schreiber et al. (1976) and Vai
and Ricchi Lucchi (1977) in the Messinian of Sicily.
This is the only gypsrudite seen in the area. Gypsum
crystals are 2 cm to 5 cm long and show no preferred
orientation within a fine matrix of gypsum powder.
Description of the Arroyo Seco Formation

The top of the Arroyo Barranca section is marked
by non-marine conglomerates which are completely
lacking in gypsum or marine fossils (Fig. 27). Clasts
within the conglomerates are less than 5 cm in
diameter with the majority only 1 cm to 2 cm in
diameter (Fig. 28E). Clasts are supported by a fine-
to medium-grained sandy matrix and are primarily
volcanic and metamorphic fragments.
Interpretation of Arroyo Barranca section

The section is interpreted as the result of three
shoaling-upwards cycles numbered 1-3 on Fig. 23A.
The lowest cycle consists of marine sandstone, silt-
stone, and conglomerate of the upper Quita Coraza
Formation (unit 1) which was deposited at inner
neritic water depths (McLaughlin et al., 1991). This
marine section shoals rapidly upward into the shal-
low-marine gypsum facies association marked by
the two gypsum horizons near the Tamayo-Vuelta
Grande road (unit 2 on Fig. 23A). The gypsum sec-
tion is overlain by a deep-marine facies association
of unit 3 marked by horizons of coral conglomerate,
trough cross-bedded sandstone, and grainstone (beds
X1, X2, and X3) (Fig. 23A). This section shallows
upward into an unexposed but inferred shallow-
marine oolite facies of unit 4 and an association of
gypsum beds C and D of unit 5 is in turn overlain
by the 'deep-marine' facies association of beds F
and G of units 6 and 7 (Fig. 23A). These marine
beds are overlain by the highest gypsum beds (bed I)
of unit 8 which show some evidence for reworking
(Fig. 28D).
Honduras-Facolina section
Fig. 29. Photographs of different marine siliciclastic facies from Location

several measured sections (see Fig. 27 for section and locations). The Honduras-Facolina composite section is 850
(A) Turbiditic calcarenite of bed X1 in marine unit 3 in the m thick and includes a deep-marine unit and a
Arroyo Barranca. (B) Coral reef material in debris flow of shallow-marine gypsum unit near the area where
marine unit 3 in Arroyo Barranca section. In-situ coral reefs are
present 12 km to the northeast (shelfward and along strike) of the gypsum pinches out above the deep-marine unit
this outcrop (cf. Fig. 23A). Length of pen shown is about 7 cm. (Fig. 23A). A complete measured section is pre-
Color photographs at http://www.elsevier.nl/locate/caribas/ sented in Lamar (1990).

Description of the Arroyo Blanco Formation Marker bed D. Bed D of unit 5 is a poorly
Base of the section. The base of the section exposed 30-cm-thick bed of recrystallized gypsum.
begins near the bridge over the Arroyo Honduras The remainder of this section is covered until the
on the dirt road from Tamayo to Vuelta Grande. 820 m level, where a thin recrystallized gypsum bed
The measured section is offset because of lack of marks the end of subaqueous deposition and the
exposure in the upper part of the Arroyo Honduras. beginning of deposition of non-marine sandstones
The base of the section is marked by a resistant and conglomerates of the Arroyo Seco Formation
debris flow deposit of reworked coral and shell de- (Fig. 23A).
bris which is 3 m thick. The debris flow deposit
exhibits local grading and is intensely bioturbated Interpretation of Arroyo Honduras-Facolina
by Ophiomorpha. Matrix consists of silty to fine- section
grained carbonate sand. Fossils include both coral This section mainly (unit 3) consists of a thick
and bivalve fragments. deep-marine facies association composed of sand-
On aerial photographs, this bed and several oth- stone, coral debris beds, and grainstone. The upper
ers to the east form prominent southeast-dipping part of the section shoals rapidly into a shallow-
clinoforms which climb the section in an eastward marine gypsum association of unit 5 marked by two
direction and suggest westward progradation of car- gypsum beds (beds C and D) which can be seen
bonate-mantled slope deposits (Fig. 23A). onlapping the marine section on aerial photographs.
The overlying marine rocks of unit 3 consist of The shallow-marine oolitic limestone beds of unit
turbiditic calcarenite beds similar to those in the 4 appear to pinch out before reaching the Arroyo
Arroyo Barranca (Fig. 29A), coquinas, and trough Facolina (Fig. 23A).
cross-bedded sandstone beds. The section is three
times thicker than the equivalent section in the Ar- Rio Yaque section
royo Barranca (Fig. 23A). A thick siliciclastic sec-
tion begins 12 m above the grainstone beds near the Location
base of the section and is composed of: (1) 15 m of The Rio Yaque section (Fig. 30A) is a shal-
bedded (beds a few cm thick) bioturbated medium- low-marine section which is laterally equivalent
grained sandstone; (2) fine-grained calcarenite 1 to deeper-marine rocks of unit 3 described in the
m thick; and (3) 80 m of massive, poorly ex- Arroyo Honduras-Facolina and Barranca sections
posed, fine- to medium-grained, dark quartz lith- (Fig. 23A). Resistant beds can be followed directly
arenite. between the two areas on aerial photographs. The
Marker beds X1, X2, and X3. These three beds section crops out in the Arroyo Boca de los Guiros
occur over an interval of 235 m of largely covered and can be easily reached by walking northward
sandstone and siltstone of unit 3. The interval is from the village of Sierracita just off the Azua-
much thicker than the correlative interval in the Barahona highway.
Arroyo Barranca to the west (Fig. 23A). The three
beds are similar and consist of 1-1.25 m of turbiditic Description of the Arroyo Blanco Formation
limestone composed of fine-grained carbonate grains Base of the section. The base of the section
and 1-2 mm fragments of shell hash and occasional is a 10-m-thick interval of marine of siltstone and
2-5 cm fragments of oyster shells. Bedding thickens sandstone that represents the top of the Quita Coraza
from 5-10-cm-thick beds at the base of the bed to Formation. The 10-m-thick interval consists of 4-m-
50-cm-thick beds at the top of the bed. The three thick debris flows of coral and gray siltstone frag-
beds are interbedded with poorly exposed, massive, ments in a gray silty matrix; gritty, coarse-grained
medium-grained litharenite in beds from 2 to 3 m volcanic sandstone; and 10-30-cm-thick beds of
in thickness which contain trough cross-beds. Coral fine-grained sandstone interbedded with blue silt-
debris beds and shell beds containing Turritella are stone (Fig. 30A).
interbedded in the sandstone. Growth position coral reef. The base of the Ar-
Marker bed C. Gypsum beds C and D of unit royo B lanco Formation is marked by a 10-m-thick
5 occupy a condensed shallow-marine section which coral reef composed of branching and head corals
onlaps the underlying marine unit containing beds (Fig. 30A). The upper 1-2 m of the reef is tightly ce-
X1, X2, and X3 (Fig. 23A). Bed C consists of mented and the reef forms a prominent, resistant ridge
a 15-cm-thick bed of greenish, finely crystalline, which can be traced eastwards on aerial photographs
grass-type gypsum which contrasts significantly to (Fig. 23A). The base of the reef is transitional with
the thickly bedded, bladed selenite in Arroyo Bar- the underlying siltstone section of the Quita Coraza
ranca (Fig. 24). The abrupt change in crystal size section (unit 1). One meter above the base of the reef
may reflect the proximity of an influx of non-marine is the first appearance of head corals in growth posi-
water (B.C. Schreiber, pers. commun., 1989). tion and delicate, upright-branching corals (Fig. 30B).

332 P. MANN et al.
Fig. 30. (A) Rfo Yaque measured section. See Fig. 20 for key to lithology and sedimentary structures. Vertical scale is in me-
ters. Locations of photograph (B) is indicated in the left margin of the section. (B) Growth-position reef forming base of ma-
rine unit 3 (compare to map in Fig. 28). Note upright, growth-position coral heads amongst reef rubble. Color photograph at
http://www.elsevier.nl/locate/caribas/
Above the reef is a 20-m-thick section of 10- stone layers exhibit very low-angle to hummocky
50-cm-thick coquina and coral debris flows interbed- to trough cross-bedding; the sand-conglomerate unit
ded with brown silty mudstone. This reef and asso- tends to fine upwards and is typically scoured into
ciated coral debris beds probably supplied the coral the underlying thinly bedded siltstone and sandstone
clasts and finer-grained carbonate material to grain- or exhibits load casts at its base; (2) a thin 5 -
stones including beds X1, X2, and X3 observed in 10-cm-thick coquina bed commonly occurs at the
measured sections to the southwest (Fig. 23A). top of the resistant conglomerate-sandstone unit; (3)
Sandy, marine section. Above the coral-rich the upper part of the cycle consists of 5-10 m of
basal part of the section, there is a 290-m-thick in- thinly bedded fine-grained sandstone and siltstone
terval of mainly litharenite and interbedded siltstone and is usually covered.
which are associated with intraclasts and layers of Sandy to conglomeratic, non-marine section.
marine fossils (Fig. 30A). This interval is thought The upper 390 m of the section consists of alternat-
to represent the shelfward equivalent of more bas- ing ridges of conglomerate-sandstone ranging from
inward deep-marine facies of unit 3 which were 3 to 10 m in thickness and intervening thinly bedded
measured in sections to the southeast of this section. sandstone and siltstone which exhibits distinct red
The sandstone occurs in characteristic 3-10-m- and green alterations in color (Fig. 30A). Grain size
thick cycles composed of: (1) a resistant 3-5-m- in the lower, coarser-grained beds increases upsec-
thick basal massive bed of medium- to coarse- tion from medium-grained sandstone to conglomer-
grained litharenite which contains pebble conglom- ate. There are no marine shells associated with this
erate in a coarse-grained sandstone matrix; the sand- section and this observation, along with the distinct

red and green color of the intervening finer-grained section (Fig. 24). The length of these crystals and
layers suggests that the section is non-marine. the thickness of this horizon rapidly decreased to the
Interpretation of the Rio Yaque section. The northeast and probably reflect the onlap of the gyp-
lower half of the section is interpreted as a shallow- sum horizon onto the marine rocks of the underlying
marine siliciclastic section formed in a shelf setting. section (Fig. 23A). Evaporite deposition was proba-
The presence of a growth position coral reef shown bly suddenly terminated by influxes or 'freshenings'
in Fig. 30A establishes the water depth at the base of marine water as seen today at Lake MacLeod, Aus-
of the section at several meters below sea level. The tralia (Logan, 1987).
presence of marine shells in 290 m of section above
the reef indicates a prolonged marine influence. The
lack of marine fossils and the red-green color in DISCUSSION
the upper part of the section suggests a non-marine
environment of deposition. Application of sequence stratigraphic
terminology to the Enriquillo basin
Environment of deposition of the Arroyo Blanco
Formation Sarg (1988) has proposed two settings for evap-
orites in the framework of sequence stratigraphy: (1)
The environment of deposition of the Arroyo onlapping lowstand and shelf-margin wedges; and (2)
Blanco Formation consisted of a shelf edge on the lagoonal-sabkha facies in the interior or back-shelf
northern margin of the basin which faced southwest positions of carbonate banks. The evaporites of the
towards the center of the Enriquillo basin (Fig. 31). northeastern margin of the Enriquillo basin appear to
The edge of the shelf is marked by the reef at fall into the first category because: (1) they are de-
the base of the Rfo Yaque section. The slope is posited basinward of the shelf break suggested by
marked by clinoforms in marine interval 3. The prograding clinoforms in the underlying deep-marine
clinoforms flatten in a basinward direction as the facies association (unit 3 in Fig. 31); and (2) the evap-
grade of the slope decreases in deeper water of the orite units exhibit wedge-shaped geometries which
Enriquillo basin. The clinoforms are composed of onlap underlying marine rocks in both map view
carbonate material which was derived from the reef (Fig. 23A) and in cross-section (Fig. 31). The on-
and mantled the slope. Paleocurrents were mainly lapping geometry of the evaporites is similar to the
directed from northeast to southwest down the slope. Triassic Dolomita Principale peritidal complex of the
The shallow-marine facies associations are con- Dolomites of Italy (Bosselini, 1984) and the Permian
fined to the basin and slope whereas the earlier Salado Formation of west Texas (Sarg, 1988).
deeper-marine facies extends across both the slope In the terminology of sequence stratigraphy, the
and shelf. As in the Messinian section of Sicily, basal contact of the evaporite horizon would form
the evaporite rocks are sandwiched between marine a sequence boundary (for example, boundaries be-
rocks with the higher evaporite beds overstepping tween units 1 and 2, units 4 and 5, and units 7
the underlying evaporite units (Fig. 31). In contrast and 8 in Fig. 31). The deep-marine units would
to the lower marine unit, which progrades basinward, represent highstand deposits which prograded out
the evaporite units prograde landward. onto a downlap surface. This surface is the boundary
between units 1 and 3 (Fig. 31).
Textural variations in gypsum deposits of the The geometry and spacing of the evaporite inter-
Arroyo Blanco Formation vals suggest three discrete drops in sea level affected
the northeastern margin of the basin and presum-
Grass-type and bladed selenite in the gypsum lay- ably the basin-central evaporite deposit less than 30
ers and oolitic limestone suggest water depths of km to the southwest (Fig. 13). The first two falls
probably less than a few meters (Schreiber, 1988). In are followed by rises in sea level and transgression
Fig. 24, we have compiled the measured thicknesses of the evaporites by a marine horizon (Fig. 31).
and primary textures of all gypsum horizons observed This transgression produced an angular truncated
in the four measured sections. All of the gypsum lay- contact between evaporite unit 5 and marine unit 6
ers probably formed in shallow brine pools on broad (Fig. 23A). The toe-of-slope position of the evapor-
mud flats. It is possible that the length of the larger ite units suggests that sea level dropped significantly
crystals in the selenite beds may reflect the water to the base of the slope (~ 100-200 m) (Fig. 31).
depth at the time of deposition as is observed in mod-
em environments of deposition like Lake MacLeod in Correlation to the Vail sea-level curve
Australia (Logan, 1987; Schreiber, 1988). If this is the
case, the deepest part of the tidal flat (<50 cm) during It is interesting to compare the timing of the
the deposition of bed C was in the Arroyo Barranca three transgressive deep-marine units of the Quita

334 R MANN et al.
Fig. 31. Summary of measured sections and schematic cross-sections from the northeastern margin of the Enriquillo basin. Pattern of
facies association is the same as shown in map view on Fig. 23. Letters identify key marker horizons discussed in text. Vertical scale is
in kilometers. The locations of the five long sections described in the text are indicated by boxes. Location of shelf, slope, and basin are
based on the geometry of clinoforms in the lower marine facies of the Arroyo Blanco Formation.
Coraza and Arroyo Blanco Formations to a eustatic eustatic sea-level fluctuations during the deposition
sea-level curve (Haq et al., 1987) (Fig. 32). Gen- of the Arroyo Blanco evaporites or whether the basin
erally, low sea levels in the Late Miocene coincide was isolated from the sea and was unresponsive to
with coarsening-upward submarine fan rocks of the sea-level changes. Kendall (1989) has emphasized
Trinchera Formation (McLaughlin et al., 1991). A the point that all saline giants must be completely
highstand of sea level in the Early Pliocene coincides isolated from the sea (Fig. 2). In the Enriquillo basin,
with the Quita Coraza Formation, which would be the correlation of evaporites to Early Pliocene high-
interpreted as a highstand condensed section. This stands (Fig. 32) suggests that preiodic connections to
highstand is also the same age as the halite of the the open ocean were achieved. Three distinct evap-
central Enriquillo Valley, which may have been fed orite cycles at the basin edge (Fig. 31) suggest that
by marine waters spilling over a sill at the mouth of these connections may have been shortlived, perhaps
the valley (Fig. 13). A similar shaley unit is present as a result of wide eustatic sea level fluctuations
.at the same time in the Cibao Valley of the northern during the Pliocene.
Dominican Republic (Evans, 1986) (Fig. 3). In the
Pliocene, there are two significant drops of sea level Models for shallow-water evaporites
which may coincide with the regressive episodes
in the Arroyo Blanco Formation. Imprecise dating There are two popular facies models for deposi-
in the Arroyo B lanco Formation, however, makes tion of shallow-water evaporites. Hardie and Eugster
correlations difficult. (1971) proposed that deposition of coarsely crys-
An important question to answer is whether the talline selenite in the Messinian of Sicily occurred in
basin was connected to the sea and was responsive to the quiet waters of a shallow lagoon or gulf adjacent

Fig. 32. Comparison of sea-level curve of Haq et al. (1987) to stratigraphy of the Enriquillo and Cibao basins of the Dominican Republic.
Ages from the Enriquillo basin are from McLaughlin et al. (1991) and this study. Ages from the Cibao basin are from Evans (1986).
to a littoral belt of laminated gypsum. Gypsum in bed. This suggests that each gypsum bed represents
the laminites and in associated gypsum sand bodies a short regressive episode during a much longer
was derived from the area of selenite deposition and shallow-marine episode. The stratigraphy is best ex-
transported shorewards onto marginal evaporitic flats plained by three shoaling-upward cycles composed
during storms. of a lower, 'deep-marine' interval, an intermedi-
Vai and Ricchi Lucchi (1977) worked on a se- ate shallow-marine interval characterized by oolitic
quence in the Messinian of Italy that lacked gypsum limestone, and an upper shallow-marine interval
laminites and suggested transport of gypsum toward characterized by gypsum (cycles number 1-3 on
the basin center. Gypsum was reworked from older, Fig. 23A).
emergent beds of selenitic gypsum by ephemeral
slope-controlled agents such as torrential streams Relation of basin-margin evaporites to
and debris flows which built up shallow alluvial basin-central evaporites
cones that encroached the basin.
As Kendall (1984) points out, the models may For reasons discussed at length by Kendall
be complementary and apply to a transgressive- (1988), it is often difficult to relate basin-margin
regressive cycle: the first model applies to times evaporites to basin-central evaporites. Based on lim-
of transgression and highstand or when the regres- ited dating of the basin-central evaporites, it appears
sion occurs entirely as a consequence of sediment that the basin-central evaporites (Angostura Forma-
outbuilding (stream gradients are low); the second tion) are coeval with the Quita Coraza or lowermost
model may apply to times of regression and lowstand Arroyo B lanco Formation in the eastern Enriquillo
when older evaporites are exposed and reworked in basin (Fig. 7). The basin-central evaporites may
marginal areas (stream gradients are high). correlate with the lowermost of the three shallow-
In the marginal evaporites of the Enriquillo basin, marine intervals or, alternatively, they may correlate
there is little evidence for reworking of older, emer- with all three of the intervals.
gent beds of selenitic gypsum. Only one isolated
example was identified in the upper part of the Bar- Subsidence analysis of the Charco Largo-1 well
ranca section (Fig. 28D). On vertical scales of 10
to 20 m, a characteristic pattern shows a bed of Backstripping, or removal of sediment load, and
channeled marine sandstone overlain by a gypsum compaction effects is one approach that can be used

336 E MANN et al.
Fig. 33. Subsidence history of the Charco Largo-1 well. See text for discussion.
to isolate the tectonic control on subsidence in the ites range from 0.2 to 2 km in thickness. They
central part of the Enriquillo basin (Mella structural consist of largely shallow-water and 'desiccated'
block in Fig. 8). A subsidence plot of the Charco deposits interbedded with open-marine sedimentary
Largo-1 well, based on integration of water depth rocks. Intermediate- and some deep-water evaporites
information gained from this study along with back- are also recognized (Schreiber, 1988). Most of the
stripping, is shown in Fig. 33. Pulses of maximum gypsum in subaerially exposed basins of Messinian
tectonic subsidence occur in the latest Miocene, co- age in the Mediterranean has not been buried deep
eval with the deposition of the Trinchera and lower enough to convert the primary depositional fabrics
Angostura Formations, and in the mid-Pliocene to to nodular anhydrite. The alternation of kilometer-
Recent time, coeval with the deposition of the upper thick open-marine turbidites with Messinian gypsum
Angostura, Arroyo B lanco, Las Salinas, Arroyo Seco without halite in the Apennines of northern Italy (Vai
and Jimanf Formations (Fig. 7). The Early Pliocene and Ricchi Lucchi, 1977) is similar to the northeast
is characterized by a pulse of tectonic uplift of the margin of the Enriquillo basin (Fig. 31). Accord-
basin. ing to Schreiber (1988), seismic profiles in the
A simple interpretation involves an early pe- Mediterranean show a similar 'stratigraphic sand-
riod of latest Miocene thrusting and crustal loading wich' consisting of deep-marine marls passing into
related to the oblique collision of the Bahamas Plat- evaporites (with a thick halite section) and back into
form with Hispaniola (Fig. 3). The uplift phase in a deep-marine section.
the Early Pliocene may be related to fold-related Studies of the Messinian in Sicily by Schreiber
topographic uplift in the area of the Mella salt-cored and Friedman (1976) and Schreiber et al. (1976)
anticlines seen on seismic line 111 in Fig. 9. The have shown that the lowermost part of the basi-
subsequent mid-Pliocene through Recent subsidence nal evaporite section, which directly overlies deep-
is consistent with rapid siliciclastic sedimentation marine marls, begins with massive layers of crys-
corresponding to the Las Salinas, Arroyo B lanco and talline gypsum. This succession is similar to gypsum
Arroyo Seco Formations (Fig. 7). Burial of the faults occurrence along the margin of the Enriquillo basin
bounding the Mella folds in the center of the basin both in its lateral continuity, primary textures, and
is consistent with the end of tectonic uplift and the in its intercalated shallow-marine carbonate rocks.
renewal of tectonic subsidence (Fig. 33). According to Schreiber (1988) and Schreiber, pers.
commun. (1989), continuous beds of bladed gypsum
Comparison of Dominican evaporites to other (or 'ferro-di-lancia' or 'spears of iron') in this lower
areas unit formed in channels or sub-basins where there
was active water movement that supplied a continu-
Messinian of the Mediterranean ous ionic feed. These 'ferro-di-lancia' of Sicily look
Some of the best studied and most controver- exactly like the long, bladed selenite (marker bed C
sial evaporite deposits occur in Messinian (Late of unit 5) of the Arroyo Barranca section (Fig. 28C).
Miocene) sections of the Mediterranean. Evapor- Schreiber further observes that the length of these

crystals (20 cm to 50 cm) probably reflects minimum cycles (Lowenstein, 1988). The Castile Formation
water depths at the time of deposition (Fig. 24). overlies deep-marine turbiditic rocks and appears to
In areas where greater amounts of freshwater en- have formed as a sharp drop in sea level affected the
tered the basin as shown by higher clay and sand silled basin. As in the Quita Coraza Formation of the
content, the beds of gypsum crystals become ir- Enriquillo basin, marine fossil assemblages of the
regular and pass into gypsum-cemented mud and Permian marine turbidite units reflect increasing re-
sand layers with some thin stringers or lenses of striction of the basin. Varve layers in the evaporites,
larger, nearly vertically oriented crystals (2 cm to which may record annual changes in circulation,
5 cm high). The deposits from these water bodies were deposited at a rate of 1.4 mm/year, can be
have great lateral continuity and can be followed traced great distances of 70-110 km, and have been
for kilometers. As in the Dominican Republic, the used as an argument for deep-water deposition (cf.
onset of evaporite deposition in many areas of the Kendall, 1984, for a review). The geometry and
Mediterranean was preceded by the appearance of thickness of the Castile Formation suggests that it
coral reefs, which were subsequently buried beneath filled topographic depression of the Delaware basin
the evaporites and associated siliciclastic sedimen- of 600 m depth in a period of about 260,000 years.
tary rocks (Warren, 1988; Schreiber, 1988). The In this respect the Castile Formation may be similar
lower level of crystalline gypsum of the Messinian to the basin-central evaporites of the Enriquillo basin
of Sicily is overlain by: (1) halite and potassic which also appear to have filled a deep topographic
salts; these contain halite crusts and desiccation depression in the center of a silled basin (Fig. 2).
polygons and probably formed in a shallow-water
to dried-basin setting; and (2) an upper evaporite Lake MacLeod, western Australia
layer consisting of seven or eight cycles beginning Lake MacLeod is a halite-filled salina that is sim-
with a brackish-water marl and passing upward into ilar in dimensions and sub-sea-level setting to the
shallow-water gypsum which oversteps and onlaps present-day Enriquillo Valley. The bed of the lake
the underlying evaporites. Although the upper layer lies 2.8 m to 4.0 m below sea level and is fed by sub-
resembles the basin-margin evaporites of the Do- surface flow generated from the nearby Indian Ocean
minican Republic, there is no three-part division of by evaporative drawdown from (Logan, 1987). The
either the basin-margin or basin-central evaporite lake is isolated by tectonic folds which form a low
deposit. In the Charco Largo-1 well, the halite in- sill along the coast. Much of the lake is filled by
terval directly overlies marine siltstone and a single a massive gypsum bed 5-7 m thick, or halite lay-
limestone bed and is terminated by a single anhy- ers. These evaporite deposits mark areas of brine
drite bed (Fig. 15). The marginal evaporites overlie ponding, whereas areas of inflow of marine wa-
marine siltstone but are not directly associated with ter are marked by carbonate rocks. The dominance
halite (Fig. 31). The cycles in the basin-margin evap- and preservation of thick halite in Lake MacLeod
orite of the Dominican Republic exhibit the same is a result of the extreme aridity as compared to
pattern of onlapping the underlying evaporite cycles more temperate, gypsum-dominated lakes in south
of the Messinian in Sicily. Australia (Warren, 1988). The lack of substantial
meteoric input during the history of the lake allowed
Lac Assal, Djoubouti, Persian Gulf the lake waters to remain in the halite precipitation
According to B.C. Schreiber (pers. commun., field for an extended period of time. Likewise, for-
1989), continuous beds of bladed gypsum as those mation and preservation of halite in the central part
observed in bed C of unit 5 (Arroyo Barranca of the Enriquillo basin may result from the extreme
section) in the basin-margin evaporites presently aridity of this area and a lack of meteoric input.
grow in evaporative water bodies which make up
Lac Assal. The gypsum is fed by marine influx, Dead Sea of Israel
the crystals are uniform within the bed, and pass The Dead Sea of Israel is similar to the pre-
laterally into muddy, finer gypsum beds where the sent-day slip-strike setting of the Enriquillo basin
basin is fed by freshwater. This lateral change in and contains modern basinal sediments which are
character is also seen as one passes along bed C similar to anoxic paper shales associated with halite
from Arroyo Barranca to Arroyo Facolina (Fig. 24). of Loma Sal y Yeso (Fig. 19A). The water surface
Crystal size decreases dramatically while the muddy of the Dead Sea is 403 m BSL (as compared to
fraction increases. 42 m BSL for Lago Enriquillo) and the Miocene-
Holocene sedimentary fill is 8-10 km thick (as
Delaware basin, Texas compared to approximately 5 km for the Enriquillo
The Permian Castile Formation of the Delaware basin). The modern lake sediments of the Dead Sea
basin consists of 550 m of gypsum, dark lime- are laminated muds with dark detrital-rich and light
stone, and halite deposited in four depositional aragonite-rich laminae similar to the paper shales of

338 E MANN et al.
Loma Sal y Yeso. The laminated bottom sediment Deep water-deep basin model
contains deep-water accumulations of halite which In the 'deep water-deep basin' model, a deep
make up less than 10% of the bottom sediment restricted basin is filled with seawater passing over
and occur as tiny clear cubes (Warren, 1986, 1988). a shallow sill (Fig. 2, upper panel). The water in
The Dead Sea is often cited as a likely environ- the basin attains a level of hypersaline equilibrium,
ment for the accumulation of laminar or bedded perhaps through stratification or brine reflux, and
deep-water halite and may be analogous to halite evaporites are deposited at either a brine-air or
accumulation in the central Enriquillo basin. In the brine-basin floor contact. Many authors have evoked
Charco Largo-1 well (Fig. 15), much of the halite this model to explain 'saline giants' such as the Per-
occurs in beds ranging in thickness up to several mian Castile Formation of west Texas, the Jurassic
meters which are interbedded with unfossiliferous Louann Salt of the Gulf of Mexico, and the Permian
dark shale (Fig. 16). It is possible that the halite Zechstein of northwest Europe (cf. Kendall, 1984,
accumulated in a basinal environment similar to the for review). Intercalated black shales, lateral bedding
Dead Sea. continuity, and the suggestion of great topographic
relief within basins have been taken as evidence
Paleogeographic evolution of the Enriquillo basin supporting this model.
A major weakness of this model is the difficulty
Fig. 34 presents an interpretation of depositional of maintaining a deep-water body at or near halite
events in the Enriquillo basin and its northeast saturation. Water saturated with salts is hygroscopic
margin which is slightly modified from McLaughlin and will dilute itself by attracting moisture from
et al. (1991) using the results of this paper. an atmosphere with a relative humidity of 65% or
The Middle Miocene was a time of deep-water, greater (Schreiber, 1988). In deep water with low
normal-marine conditions recorded by the deposi- ratios of surface area to volume, it is extremely
tion of the pelagic limestone of the Sombrerito difficult to reach and maintain halite saturation.
Formation (Fig. 34A,B). Generally lower stands of
sea level in the early-Late Miocene through earliest Shallow water-shallow basin model
Pliocene and tectonic uplift of the Cordillera Central Models for 'shallow water-shallow basin' set-
coincided with submarine siliciclastic deposition of tings involve influx of marine water across a sill
the Trinchera Formation (Fig. 34C-E). Rising sea into a rapidly subsiding basin filled with sediment
level in the Early Pliocene (Fig. 32) flooded the area (Fig. 2, middle panel). This concept is obviously not
and ended siliciclastic deposition of the Trinchera applicable to evaporites such as those of the west-
Formation. This transgression is marked by siltstone ern Mediterranean which clearly formed in a deep
of the Quita Coraza Formation, and seems to have basinal setting (Ryan and Cita, 1978).
been the flooding event with which the following
cycle of flooding and desiccation of the central Shallow water-deep basin model
Enriquillo basin began. The late part of the Early The 'shallow water-deep basin' model was in-
Pliocene saw the deposition a massive halite section voked by Hsti et al. (1973) to explain the Messinian
in the basin center during these series of marine evaporites of the Mediterranean (Fig. 2, bottom
influx and evaporation events (Fig. 34F). Later re- panel). A shallow-water environment facilitates rapid
gressions are recorded along the northern margin of evaporation per unit volume of water, more efficient
the Enriquillo basin. During the late-Early Pliocene, solar heating of the water, and a quicker return to
while gypsum and interbedded siliciclastics of the evaporite saturation levels after freshwater influx or
Angostura Formation were being deposited in the climate change. The deep setting of the basin pro-
basin center, similar sediments of the Arroyo B lanco motes periodic spills of seawater during highstands
Formation were deposited on the northeast basin of sea level which replenish brines within the basin.
margin. In these formations, times of near-normal Deep basins below sea level such as the modem En-
marine conditions often saw influx of sediments riquillo Valley and the Dead Sea are fairly common
from the northeast. in tectonically active areas and are likely settings for
seawater spills.
Relevance of silled basin models to the Enriquillo
basin Preferred model for Enriquillo evaporites
The 'shallow water-deep basin' model (Fig. 2,
Kendall (1984) presents three models for silled bottom panel) is the preferred model for explaining
basins which may be the depositional settings of sub- both the basin central and basin edge evaporites of
aqueous evaporites found in 'saline giants' (Fig. 2). the Enriquillo basin for the following reasons: (1)
Could the Enriquillo basin be considered a late the present-day 'ramp basin' structure of the basin
Neogene analog for one of these conceptual models? characterized by a subsea-level depression about 80

Fig. 34. Schematic diagram of paleoenvironments in Miocene and Pliocene time of the Enriquillo basin based on McLaughlin et al.
(1991) and this study. (A) middle Middle Miocene; (B) late Middle Miocene; (C) early Late Miocene; (D) earliest Pliocene; (F) late Early
Pliocene; (G) middle Pliocene; (H) Present.
m BSL and isolated from the sea by a shallow also existed during the Pliocene evaporite deposi-
coastal sill (Fig. 6) approximates the basinal con- tion; (2) a shallow water setting would promote
ditions suggested by the shallow water-deep basin more efficient evaporation and rapid formation of
model; subsidence history of the Charco Largo-1 1500 m thick basin evaporites documented by the
well (Fig. 33) and regional stratigraphic studies Charco Largo-1 well (Fig. 15); and (3) a 'deep' basin
(Fig. 34) indicate that these 'ramp basin' conditions tectonically depressed by thrust faults and isolated

340 E MANN et al.
from the sea by a coastal sill would provide a likely (7) The northeast Early Pliocene margin of the
site for multiple seawater spills that would replenish Enriquillo basin consists of a southeast-facing shelf-
brines and sustain the formation of the types of thick slope which was tilted to the north and northeast 20-
basin central halite (Fig. 16) and the three cycles of 40 ~ in post-Early Pliocene times by tectonic folding.
basin margin evaporites observed (Fig. 31). The tilted beds now form a natural cross-section of
the Early Pliocene margin (Fig. 23).
(8) The sedimentary rocks of the shelf-slope can
CONCLUSIONS be divided into gypsum and oolite-bearing shallow-
marine facies and a deeper-marine facies (Figs. 23,
(1) There are two distinct facies of Early Pliocene 31). Nine distinct lithologic groups are recognized
evaporites which formed in the Enriquillo basin: including overlying non-marine rocks of the Arroyo
(a) a basin-central evaporite deposit composed Seco Formation. The primary textures of gypsum
mainly of halite (Angostura Formation), approxi- suggests that these were deposited in several meters
mately 1500 m thick, which is known from the or less of water in a tidal-flat setting (Fig. 24).
Superior Charco Largo-1 well in the center of the (9) The nine lithologic groups can be interpreted
basin and from the diapiric anticline of Loma Sal as three shoaling-upwards cycles consisting of: (a)
y Yeso along the southern edge of the basin; (b) a a basal deep-marine section characterized by bas-
basin-margin evaporite composed mainly of gypsum inward-prograding clinoforms composed of grain-
up to 5 m thick interbedded with shallow-marine stones and coral debris; (b) an intermediate oolitic
siliciclastic and carbonate rocks (Arroyo Blanco limestone section; (c) an upper gypsum section
Formation) (Fig. 13). (Fig. 31).
(2) Seismic reflection and well data show that (10) The shallow-marine gypsum and oolitic
the Early Pliocene depositional basin was a major, limestone sections are wedge-shaped and occupy
fault-controlled deep in the center of the valley a position at the base of the slope (Fig. 23). Interbed-
separated from the Caribbean Sea to the east by ded marine units onlap and truncate the underlying
a shallow sill (Figs. 13, 14). Siliciclastic source gypsum units. The three shallow-marine sections are
areas lay to the north in the Cordillera Central. interpreted as being deposited during three drops in
The depositional and climatic setting was probably sea level during the Early Pliocene. These sea-level
similar to the modem Enriquillo Valley, which is 80 drops correlate reasonably well to the Cibao basin
m BSL at its lowest point, separated by a shallow sill of the northern Dominican Republic (Evans, 1986)
from the Caribbean Sea, and receiving siliciclastic (Fig. 32).
sediment derived from the Cordillera Central by the (11) Correlation between the basin-margin and
Rio Yaque del Sur (Fig. 5). basin-central evaporites in the Enriquillo basin is
(3) The structure of the Loma Sal y Yeso consists difficult because of lack of accuracy in the age of
of a faulted diapir. Halite of the Angostura Forma- the units. Despite this limitation, we favor a 'shallow
tion is extruded as a narrow, 100-m-wide strip along water-deep basin' model (Fig. 2) for both areas of
a reverse fault separating the Angostura and Las evaporites for three reasons: (1) the 'ramp basin'
Salinas Formations (Fig. 19). setting of the basin would promote a 'deep' sub-sea
(4) Primary textures in halite and gypsum exposed level depression in a coastal setting; (2) shallow
in Loma Sal y Yeso are overprinted by alteration and water within this 'deep' basin would promote effi-
tectonic flowage related to diapirism and folding cient evaporation of seawater spilled over the coastal
(Fig. 19). Interbedded sedimentary rocks are fine- sill; and (3) multiple spills during times of higher
grained mudstone lacking sedimentary structures. Pliocene sea level would replenish brines to sustain
(5) The Las Salinas Formation consists of a the formation of thick basin central evaporites and
lower shallow-marine siliciclastic interval 180 m in explain the three cycles of basin margin evaporites
thickness which is overlain by a brackish interval observed
1670 m in thickness containing both siliciclastic
and minor carbonate rocks (Fig. 21). The lower
shallow-marine interval terminated deposition of the ACKNOWLEDGEMENTS
basin-central halite of the Angostura Formation.
(6) Correlations between distinctive lithologies in This work represents a distillation of a 15-year-
the Angostura and Las Salinas Formations can be long effort by the authors. Initial field studies were
made to the lithologies described by geologists of conducted as part of Mann and McLaughlin's Ph.D.
Superior Oil Company in the Charco Largo-1 well dissertations in the 1980s and were followed up by
(Table 1). A notable difference is the lack of gypsum a 1990 masters study by Mike Lamar that was car-
in the well and the presence of gypsum in the Loma fled out at the University of Texas at Austin. Our
Sal y Yeso. 1980s work followed up on ostracode and strati-

N E O G E N E EVAPORITIC AND S I L I C I C L A S T I C D E P O S I T I O N IN THE E N R I Q U I L L O BASIN 341
graphic studies of Bold in the 1960s and 1970s Dolan, J.E, Mann, E, de Zoeten, R., Heubeck, C., Shiroma, J.
which attempted to integrate outcrop data with early and Monechi, S., 1991. Sedimentologic, stratigraphic, and tec-
tonic synthesis of Eocene-Miocene sedimentary basins, His-
exploration wells. Lawrence worked on the subsur-
paniola and Puerto Rico. In: E Mann, G. Draper and J. Lewis
face data as part of the Norconsult group in the (Editors), Geologic and Tectonic Development of the North
early 1980s (Norconsult, 1983). Mann, McLaughlin, America-Caribbean Plate Boundary Zone in Hispaniola. Geol.
and Bold were individually supported by NSF, the Soc. Am. Spec. Pap., 262: 217-263.
Petroleum Research Fund and Mobil Exploration Dolan, J.E, Mullins, H. and Wald, D., 1998. Active tectonics
and Production Company. We thank Mobil for pro- of the north-central Caribbean region: oblique collision, strain
partitioning, and opposing slabs. In: J. Dolan and E Mann
viding Mann and Lamar with funds for field work (Editors), Active Strike-Slip and Collisional Tectonics of the
in the Dominican Republic and for providing the Northern Caribbean Plate Boundary Zone. Geol. Soc. Am.
subsurface data. We also thank Mobil for allow- Spec. Pap., 326: 1-61.
ing McLaughlin and Bold access to the core of Evans, C.C., 1986. Facies Evolution in a Neogene Transpres-
the Charco Largo-1 well in their core storage fa- sional Basin: Cibao Valley, Dominican Republic. Ph.D. Thesis,
University of Miami, Miami, FL, 103 pp.
cility in Dallas, and for granting us permission to
Friedman, G.M., 1982. Evaporites as source rock for petroleum.
publish this combined data set. We also thank the In: Depositional and Diagenetic Spectra of Evaporites; a Core
Direcci6n General de Minerfa in Santo Domingo Work-Shop. SEPM Core Workshop 3, Calgary, pp. 385-395.
for their continued cooperation and permission to Garcfa, S., 1976. Geograffa Dominicana. Amigo del Hogar,
publish these data. The careful reviews of Christoph Santo Domingo, 272 pp.
Guerra Pena, F., 1956. Las principales cuencas sedimentarias
Heubeck, Steve Pierce, and Charlotte Schreiber sub-
de la Republica Dominicana y sus posibilidades petroliferas.
stantially improved this paper. UTIG contribution Symposium sobre Yacimientos de Petroleo y Gas, Tomo IV,
number 1424. XX Congreso Geologico International, Mexico City, pp. 141-
159.
Haq, B.U., Hardenbol, J. and Vail, ER., 1987. Chronology of
fluctuating sea levels since the Triassic. Science, 235:1156-
REFERENCES
1167.
Hardie, L.A. and Eugster, H.E, 1971. The depositional envi-
Berggren, W.A., Kent, D.V. and Aubry, M.-E, 1995. A re- ronment of marine evaporites: a case for shallow, siliciclastic
vised Cenozoic geochronology and chronostratigraphy. In: accumulation. Sedimentology, 16:187-220.
W.A. Berggren, D.V. Kent, et al. (Editors), Geochronology, Harms, F.J., 1989. Konglomeratisches Jungterti~ir im Valle de
Time Scales, and Global Stratigraphic Correlation. Soc. Econ. San Juan (stidwestliche Dominikanische Republik, Grosse An-
Paleontol. Mineral. Spec. Publ., 54:129-212. tillen). Zusammensetzung, Herkunft, und Alter der Ger611e,
Bold, W. van den, 1975. Neogene biostratigraphy (Ostracoda) of Fazies. Ph.D. Thesis, University of Hannover, Hannover, 141
southern Hispaniola. Bull. Am. Paleontol., 66, No. 286, 39 pp.
PP.
Bold, W.A. van den, 1990. Late Holocene ostracoda in and Heubeck, C., Mann, E, Dolan, J. and Monechi, S., 1991. Di-
around Lake Enriquillo, Dominican Republic. In: D.K. Larue achronous uplift and recycling of sedimentary basins during
and G. Draper (Editors), Transactions 12th Caribbean Ge- Cenozoic tectonic transpression, northeastern Caribbean plate
ological Conference, St. Croix, U.S. Virgin Islands. Miami margin. Sediment. Geol., 70: 1-32.
Geological Society, Miami, FL, pp. 163-173. HsiJ, K.J., Cita, M.B. and Ryan, W.B.F., 1973. The origin of the
Bosselini, A., 1984. Progradation geometries of carbonate plat- Mediterranean Evaporites. Init. Rep. DSDP, 13:1203-1231.
forms: examples from the Triassic of the Dolomites, northern Kendall, A.C., 1984. Evaporites. In: R.G. Walker (Editor), Facies
Italy. Sedimentology, 31: 1-24. Models (2nd ed.). Geosci. Canada Repr. Ser., 1: 259-296.
Bowin, C.O., 1975. The geology of Hispaniola. In: A.E.M. Nairn Kendall, A.C., 1988. Aspects of evaporite basin stratigraphy.
and G.G. Stehli (Editors), The Ocean Basins and Margins: The In: B.C. Schreiber (Editor), Evaporites and Hydrocarbons.
Gulf of Mexico and the Caribbean. The Ocean Basins and Columbia University Press, New York, pp. 11-65.
Margins, Vol. 3, Plenum, New York, pp. 501-552. Kendall, A.C., 1989. Brine mixing in the Middle Devonian
Burke, K. and Seng6r, A.M.C., 1988. Ten meter global sea-level of western Canada and its possible significance to regional
change associated with South Atlantic Aptian salt deposition. dolomitization. Sediment. Geol., 64: 271-285.
Mar. Geol., 83:309-312. Kirkland, D.W. and Evans, R., 1981. Source-rock potential of
Cooper, J.C., 1983. Geology of the Fondo Negro Region, Do- evaporite environment. Am. Assoc. Pet. Geol. Bull., 65: 81-
minican Republic. M.S. Thesis, State University of New York, 190.
Albany, 128 pp. Lamar, M.E., 1990. Geology of Pliocene Evaporitic Rocks, En-
DeCelles, EG., Langford, R.E and Schwartz, R.K., 1983. Two riquillo Valley, Dominican Republic. Unpublished MA Thesis,
new methods of paleocurrent determination from trough cross- University of Texas at Austin, 162 pp.
stratification. J. Sediment. Petrol., 53: 629-642. Llinas, R.A., 1972a. Geologia del area Polo-Duverg6, Cuenca de
De Leon, R., 1983. Aspectos geologicos e hidrogeologicos de la Enriquillo, Codia, Part 1. Publication of Colegio Dominicano
region suroeste. Publicationes Especiales, Museo de Historia de Ingenieros, Arquitectosa, y Agrimensores, Santo Domingo,
Natural, Santo Domingo, 25 pp. No. 31, pp. 55-65.
DeMets, C., Gordon, R.G., Argus, D.E and Stein, S., 1990. Llinas, R.A., 1972b. Geologia del area Polo-Duverg6, Cuenca de
Current plate motions. Geophys. J. Int., 101: 425-478. Enriquillo, Codia, Part 2. Publication of Colegio Dominicano
Dixon, T., Farina, F., DeMets, C., Jansma, E, Mann, E and de Ingenieros, Arquitectosa, y Agrimensores, Santo Domingo,
Calais, E., 1998. Relative motion between the Caribbean and No. 32, pp. 40-53.
North American plates and related plate boundary zone defor- Logan, B.W., 1987. The MacLeod evaporite basin, western Aus-
mation based on a decade of GPS measurements. J. Geophys. tralia. Am. Assoc. Pet. Geol. Mem., 44, 140 pp.
Res., 103:15,157-15,182. Lowenstein, T.K., 1988. Origin of depositional cycles in a Per-

342 E M A N N et al.
mian 'saline giant': the Salado (McNutt zone) evaporites of kilometer-deep Mediterranean in the Miocene. Mar. Geol., 27:
New Mexico and Texas. Geol. Soc. Am. Bull., 100: 592-608. 193-230.
Mann, P., Taylor, E W., Burke, K. and Kulstad, R., 1984. Subaeri- Sarg, J.E, 1988. Carbonate sequence stratigraphy. In: C.K. Wil-
ally exposed Holocene coral reef, Enriquillo Valley, Domini- gus, B.S. Hastings, C.G.St.C. Kendall, H.W. Posamentier, C.A.
can Republic. Geol. Soc. Am. Bull., 95: 1084-1092. Ross and J.C. Van Wagoner (Editors), Sea-Level Changes: An
Mann, P., McLaughlin, P.P., Jr. and Cooper, J.C., 1991a. Geol- Integrated Approach. Soc. Econ. Paleontol. Mineral. Spec.
ogy of the Enriquillo-Azua basins, Dominican Republic, 2. Publ., 42: 155-181.
Structure and tectonics. In: P. Mann, G. Draper and J.E Lewis Schreiber, B.C., 1988. Subaqueous evaporite deposition. In: B.C.
(Editors), Geologic and Tectonic Development of the North Schreiber (Editor), Evaporites and Hydrocarbons. Columbia
America-Caribbean Plate Boundary in Hispaniola. Geol. Soc. University Press, New York, pp. 182-255.
Am. Spec. Pap., 262: 367-389. Schreiber, B.C. and Friedman, G.M., 1976. Depositional envi-
Mann, P., Draper, G. and Lewis, J.E, 1991b. An overview of ronments of upper Miocene (Messinian) evaporites of Sicily as
the geologic and tectonic development of Hispaniola. In: P. determined from analysis of intercalated carbonates. Sedimen-
Mann, G. Draper and J.E Lewis (Editors), Geologic and tology, 23: 255-270.
Tectonic Development of the North America-Caribbean Plate Schreiber, B.C., Friedman, G.M., Decima, A. and Schreiber,
Boundary in Hispaniola. Geol. Soc. Am. Spec. Pap., 262: 1- E., 1976. Depositional environments of Upper Miocene
28. (Messinian) evaporite deposits of the Sicilian Basin. Sedi-
Mann, P., Lebr6n, M., Rodriguez, J. and Heubeck, C., 1991c. mentology, 23: 729-760.
Geologic maps of the southern Dominican Republic. In: P. Sen Gupta, B.K. and Schaefer, C.T., 1973. Holocene benthonic
Mann, G. Draper and J.E Lewis (Editors), Geologic and foraminifera in leeward bays of St. Lucia, West Indies. Mi-
Tectonic Development of the North America-Caribbean Plate cropaleontology, 19: 341-365.
Boundary in Hispaniola. Geol. Soc. Am. Spec. Pap., 262, Surdam, R.C., Eugster, H.P. and Mariner, R.H., 1972. Mag-
Plates 4a, 4b, and 4c, scale: 1 : 150,000. adi-type chert in Jurassic and Eocene to Pleistocene rocks,
Mann, P., Taylor, E W., Edwards, R.L. and Ku, T.L., 1995. Wyoming. Geol. Soc. Am. Bull., 83: 2261-2266.
Actively evolving microplate formation by oblique collision Taylor, E W., Mann, P., Valastro, S., Jr. and Burke, K., 1985.
and sideways motion along strike-slip faults: an example from Stratigraphy and radiocarbon chronology of a subaerially ex-
the northern Caribbean plate margin. Tectonophysics, 246: 1- posed Holocene coral reef, Dominican Republic. J. Geol., 93:
69. 311-332.
McLaughlin, P.P., Jr., 1989. Neogene planktonic foraminiferal Vai, G.B. and Ricchi Lucchi, E, 1977. Algal crusts, au-
biostratigraphy of the southwestern Dominican Republic. J. tochthonous and siliciclastic gypsum in a cannibalistic evap-
Foraminiferal Res., 19:294-310. orite basin: A case history from the Messinian of northern
McLaughlin, P.P., Jr., 1991. Migration of Neogene marine en- Apennines. Sedimentology, 24:211-244.
vironments, southwestern Dominican Republic. Geology, 19: van den Berghe, B., 1983. Evolution s6dimentaire et structurale
222-225. depuis le Pal6oc~ne du secteur 'Massif de la Selle' (Haiti),
McLaughlin, P.P., Jr. and Sen Gupta, B.K., 1994. Benthic fo- Baoruco (R6publique Dominicaine), 'Nord de la Ride Beata'
raminiferal record in the Miocene-Pliocene sequence of the dans l'Orog~ne Nord Caribe (Hispaniola, Grandes Antilles).
Azua basin, Dominican Republic. J. Foraminiferal Res., 24: Ph.D. Thesis, l'Universit6 Pierre et Marie Curie, Paris, 205 pp.
75-109. Vann, I.R., Graham, R.H. and Hayward, A.B., 1986. The struc-
McLaughlin, P.P., Jr., Bold, W.A. and Mann, P., 1991. Geol- ture of mountain fronts. J. Struct. Geol., 8:215-227.
ogy of the Enriquillo-Azua basins, Dominican Republic, 2. Wadge, G. and Wooden, J.L., 1982. Late Cenozoic alkaline
Foraminiferal/ostracode biostratigraphy and depositional his- volcanism in the northwestern Caribbean: tectonic setting and
tory. In: P. Mann, G. Draper and J.E Lewis (Editors), Geologic Sr isotopic characteristics. Earth Planet. Sci. Lett., 57: 35-46.
and Tectonic Development of the North America-Caribbean Warren, J.K., 1986. Perspectives: shallow-water evaporitic envi-
Plate Boundary in Hispaniola. Geol. Soc. Am. Spec. Pap., 262: ronments and their source rock potential. J. Sediment. Petrol.,
337-366. 56: 442-454.
Norconsult, 1983. Dominican Republic Petroleum Exploration Warren, J.K., 1988. Evaporite sedimentology: Importance in hy-
Appraisal (Vol. 1), Norconsult Internal Report, Sandvika, Nor- drocarbon accumulations. Prentice-Hall/IHRDC, 304 p.
way, 79 pp. White, A.H. and Youngs, B.C., 1980. Cambrian alkali playa-
Parnell, J., 1988. Significance of lacustrine cherts for the en- lacustrine sequence in the northeastern Officer basin, South
vironment of source-rock deposition in the Orcadian basin, Australia. J. Sediment. Petrol., 50:1279-1286.
Scotland. In: A.J. Fleet, K. Kelts and M.R. Talbot (Editors), Winker, C.D. and Buffler, R.T., 1988. Paleographic evolution
Lacustrine Petroleum Source Rocks. Geol. Soc. London Spec. of early deep-water Gulf of Mexico and margins, Jurassic
Publ., 40:205-217. to Middle Cretaceous (Comanchean). Am. Assoc. Pet. Geol.
Ryan, W.B.E and Cita, M.B., 1978. The nature and distribution Bull., 72:318-346.
of Messinian erosional surfaces - - indicators of a several-

Chapter 13
Evolution of the Neogene Kingshill Basin of St. Croix,

U.S. Virgin Islands
I V A N G I L L , P E T E R P. M C L A U G H L I N , JR. a n d D E N N I S K. H U B B A R D
The sedimentary rocks of the Neogene Kingshill basin of St. Croix record part of the evolution of the tectonically complex
region at the eastern edge of the North American-Caribbean plate boundary zone. The Kingshill basin is a northeasterly oriented
graben or half-graben that contains a thick section of Neogene carbonates bounded by fault blocks of Cretaceous siliciclastic and
intrusive rocks. Significant details of basin development have been added by the inclusion of data from a drilling program that
included fourteen test holes with cumulative footage exceeding 533 m and a maximum depth of 91 m. Additional information
came from outcrop sampling over the ca. 80 km 2 basin, subsurface records, and samples from engineering and water wells
donated to the project.
Previous models of basin development suggest that the carbonate rocks of the Kingshill basin were deposited (1) in shallow
water or (2) entirely within the confines of an insular graben system. These models assume an isolated insular basin with
self-contained sediment source. Instead, subsurface evidence suggests that early Kingshill basin sedimentation started in deep
marine conditions prior to faulting on the basin margins and includes incursions of coarse, reef-derived sediment from a nearby
source. The period of pre-rift sedimentation is documented to extend into the early Middle Miocene, but probably extends into
the Oligocene or earlier. The faulting that formed the basin margins was initiated no earlier than the late Middle Miocene. After
rifting, the Kingshill basin underwent significant shallowing and uplift in Late Miocene to Early Pliocene time. Basin development
culminated in the establishment of a Pliocene reef tract and several episodes of subaerial exposure.
The Jealousy Formation, the lowest formation described, is an entirely subsurface Middle Miocene unit of dark marls deposited
at middle bathyal depths. The Kingshill Limestone conformably and diachronously overlies the Jealousy Formation and is divided
into two members. The La Reine Member is characterized by buff pelagic limestones and marls with an upward increasing
proportion of intercalated shelf-derived sediment flows. It ranges from basal Middle Miocene to uppermost Miocene and exhibits
a transition from middle bathyal to upper bathyal environments. The Mannings Bay Member is composed of skeletal debris-rich
carbonate slope deposits and lies near the Miocene-Pliocene boundary. The Blessing Formation overlies the Kingshill Limestone
and represents a reef system that existed on the south coast of St. Croix during the Early Pliocene.
Stratal relations on the basin margins indicate that the Jealousy Formation and at least the lower part of the Kingshill
Limestone were deposited prior to graben formation near the end of the Middle Miocene. Subsidence analysis of the Neogene
section indicates that 400 m of vertical uplift occurred on St. Croix between 10.5 and 3.5 Ma.
A right-lateral model of movement between St. Croix and the Puerto Rico platform has been suggested by several recent
workers. This model is consistent with the geomorphology of the Virgin Islands Trough and the Anegada Passage with right-lateral
strike-slip motion in the Anegada Passage opening the Virgin Islands Trough as a pull-apart basin. However, an older left-lateral
model of island movement is consistent with the northeasterly orientation of the normal fault system of St. Croix and the St. Croix
Ridge. In addition, left-lateral motion would locate pre-rift St. Croix south of the known extra-basinal sources of Cretaceous and
Tertiary shelf sediment required by the timing of Kingshill basin sedimentation. In this model, the Puerto Rico platform area could
act to disperse slip between the North American and Caribbean plates. A variety of models are possible, but each should take into
account geologic details of the Kingshill basin development.
INTRODUCTION AND GEOLOGIC SETTING d i s t i n c t f r o m the m a j o r i t y o f the p r i m a r i l y i g n e o u s

i s l a n d s o f the L e s s e r A n t i l l e s . A t h i c k s e c t i o n o f
St. C r o i x is the s o u t h e r n m o s t o f the U.S. V i r g i n Neogene c a r b o n a t e s o c c u p i e s a c e n t r a l g r a b e n or
I s l a n d s , l o c a t e d at the e a s t e n d o f the G r e a t e r A n t i l l e s h a l f - g r a b e n , h e r e r e f e r r e d to as the K i n g s h i l l b a s i n
a n d the n o r t h w e s t e d g e o f the L e s s e r A n t i l l e s arc (Fig. 2). T h i s b a s i n lies b e t w e e n fault b l o c k s o f
(Fig. 1). T h e i s l a n d is t e c t o n i c a l l y a n d g e o l o g i c a l l y C r e t a c e o u s s i l i c i c l a s t i c a n d i n t r u s i v e r o c k s o f the

344 I. GILL et al.
Fig. 1. Location map of St. Croix, the Virgin Islands platform, and the Virgin Islands basin. Bahymetry and structure are after Houlgatte
(1983). Inset: NOAM = North American plate; SOAM -- South American plate; CARIB = Caribbean plate (after Burke et al., 1984).
Fig. 2. Generalized geologic map of St. Croix from Whetten (1974). Exposed strata mapped as Jealousy Formation by Whetten (1966)
are re-mapped as Kingshill Limestone in this paper.
Mt. Eagle Group that comprise the mountainous tions (Fig. 3): the blue-gray marls of the Jealousy
East End and Northside Ranges. Formation; marls and limestones of the Kingshill
The Neogene carbonate section, which is the Limestone; and reef limestones of the Blessing For-
focus of this paper, is divided into three forma- mation (Gill et al., in press). The stratigraphy of this

EVOLUTION OF THE NEOGENE KINGSHILL BASIN OF ST. CROIX, U.S. VIRGIN ISLANDS 345
basin provides clues to the tectonic evolution of the STRATIGRAPHY AND DEPOSITIONAL SETTING OF
eastern end of the North American-Caribbean plate THE NEOGENE FORMATIONS OF THE KINGSHILL
boundary zone (PBZ). BASIN
The purpose of this paper is to trace the evolution
of the Kingshill basin from the Miocene to Recent Jealousy Formation
based on observations from test holes and outcrops
in the central plain of the island. This investigation Lithology and distribution
also evaluates the implications of these findings, The Jealousy Formation is a unit of blue-gray
tied to marine geology studies in the area, for marls that underlies much of the central plain of
plate tectonic models of the northeastern part of the St. Croix. The top of the formation is marked and
Caribbean region. Up to this point, there have been abrupt in the subsurface; water-well drillers treat the
few integrated biostratigraphic-stratigraphic studies top of this 'blue clay' as hydrologic basement and
that have incorporated subsurface information. generally stop drilling when it is reached.
Previous studies of the geology of St. Croix have, The blue-gray marls are rich in planktonic
for the most part, been based solely on outcrop foraminifera and other deep-water microfauna. The
data. These studies considered the carbonates of the formation includes a number of conglomeratic lime-
Kingshill basin to record deposition in an isolated stone and thin limestone layers in the deep subsur-
Oligocene-Miocene graben system (Multer et al., face, below the reach of the drilling conducted for
1977; Gerhard et al., 1978; Lidz, 1984a). Whetten this study (Cederstrom, 1950). These coarse-grained
(1966) produced a detailed geologic map of St. Croix beds are bracketed above and below by pelagic blue
and a particularly detailed description of its Creta- marls, so are considered allochthonous deposits of
ceous section; some of this work is summarized in down-slope transported debris.
Fig. 2. Multer et al. (1977) and Gerhard et al. (1978) Test well data indicate the Jealousy Formation is
provided modern models for the carbonate sedimen- present in the subsurface throughout the central plain
tation, including the structural and sedimentological region, both inside and outside the fault boundaries
model of the basin. Gerhard et al. (1978) designated of the Kingshill basin graben. The type section
the type section and provided detailed petrologic was defined by Cederstrom (1950) in the deepest
descriptions of the Kingshill Limestone that are still of several test wells (Test Well 41) drilled by the
pertinent today. Lidz (1982, 1984a, 1988) defined Civilian Conservation Corps (CCC) in 1939, where
the biostratigraphic relationships within the basin, a thickness of more than 426 m was encountered
tied them to the basin model of Gerhard et al. (1978) (Fig. 4). It is present in wells M1, M2, and M10 of
and suggested ties to global eustasy. In the only this study, all west of the eastern fault boundary of
work to include subsurface information, Cederstrom the graben, and was reported by Cederstrom (1950)
(1950) provided an early geologic map and a de- at 18 m below sea level in CCC Test Well C26,
tailed description of early deep drilling work on the approximately 1 km east of this fault (Fig. 4).
island. This work includes the type-section descrip- A maximum thickness of 450 m has thus far
tion of the Jealousy Formation. Most of these studies been recognized for this unit (Cederstrom, 1950).
either did not address a wider tectonic framework, or Although its base has never been reached in the
have related the late Cenozoic tectonic evolution of center of the basin, gravity surveys indicate that
St. Croix solely to vertical uplift. more than 1800 m of Jealousy and older sedimentary
However, cores from a drilling program con- rocks may underlie the central plain (Shurbet et
ducted in the 1980's furnish new subsurface data al., 1956; R.C. Speed, written commun., 1994).
that, in conjunction with outcrop data, help to more Although the top of the Jealousy Formation exhibits
clearly delineate the sedimentary and structural evo- considerable relief in the subsurface (Fig. 5), no
lution of the Kingshill basin during the Neogene. apparent change in bulk mineralogy, microfauna, or
The drilling program included fourteen test wells grain size is observed across the Jealousy/Kingshill
drilled to depths of up to 91 m as well as data boundary. No hiatus or missing section is evident
from a number of private wells donated to the within the resolution of available biostratigraphic
project (Fig. 4). These wells were logged during control.
drilling, and samples were collected for sedimento- Extensive areas of Jealousy Formation outcrop
logic, micropaleontologic, and geochemical analysis exposures have been mapped in some previous
at intervals of 1.5 or 3 m in the wells (Gill and Hub- studies (Cederstrom, 1950; Whetten, 1966). How-
bard, 1986, 1987; McLaughlin et al., 1995). Core ever, we recognize the Jealousy Formation as an
material and logs from pre-existing wells provided exclusively subsurface unit and suggest that these
additional data on the carbonate units underlying the outcrops are more correctly mapped as Kingshill
southeastern portion of the central plain. Limestone, following Gerhard et al. (1978).

4~

Fig. 3. Stratigraphic column, chronostratigraphic framework, and paleoenvironments of the Neogene section of the Kingshill basin (after McLaughlin et al., 1995). Planktonic foraminiferal zonation based
t"
on Bolli and Saunders (1985). Chronostratigraphy, coastal onlap curve, and eustatic cycles after Haq et al. (1988). t"*
Fig. 4. Locations of outcrops, test wells and water wells used in the stratigraphic cross-sections. A/P = Airport/Penitentiary; AQ --
Airport Quarry; FC = Five Corners; HC = Hess Cut; MS = Morningstar; R/B -- Rattan/Belvedere; SR --- Salt River valley; VR = Villa
La Reine and Fredensburg Quarry; WR = Work and Rest. Core sample locations designated with M are test holes drilled by Gill, the
others are from previous studies. Cutting sample locations are noted for water well cuttings studied by the authors. Test holes drilled in
1939 by the Civilian Conservation Corps are designated by C.
Age and paleoenvironment The Jealousy Formation is a dominantly hemi-

Our recent micropaleontologic studies of bore- pelagic, deep-water unit. The benthic foraminiferal
hole samples from the Kingshill graben place the fauna indicates deposition at 600 to 800 m water
Jealousy Formation in the lower part of the Middle depth (McLaughlin et al., 1995). Most species recov-
Miocene, ranging from the P r a e o r b u l i n a g l o m e r o s a ered are generally associated with middle and upper
zone to the G l o b o r o t a l i a f o h s i f o h s i zone (McLaugh- bathyal environments; several species are present that
lin et al., 1995). The Jealousy Formation has previ- indicate an environment no shallower than the middle
ously been referred to as Oligocene (Cushman, 1946; bathyal zone. This differs from previous interpreta-
Cederstrom, 1950) and even as low as Middle Eocene tions of the Jealousy Formation as an estuarine de-
(Lidz, 1984b). However, the Oligocene citations are posit (Van den Bold, 1970; Multer et al., 1977) based
based on older notions of the age significance of cer- on outcrop samples previously mapped as Jealousy.
tain benthic macrofauna and larger foraminifera. The
Middle Eocene citation is based on the planktonic Source area and paleocurrents
foraminiferal fauna found by Lidz (1984b) in an al- The coarse, shelf-derived carbonates in the Jeal-
lochthonous shale clast in the Kingshill Limestone ousy Formation are sandwiched between large in-
that is presumed to have been derived from the Jeal- tervals of foram-rich basinal sediments. The coarse
ousy Formation. Although no in-place paleontologic carbonates are therefore allochthonous, and require
evidence exists for an age any older than Miocene for a nearby shelf source. The thickness of Jealousy
the Jealousy Formation, the estimated 1800 m thick- strata makes the uplifted horst blocks of St. Croix
ness of sedimentary fill in the Kingshill basin (Shurbet an unlikely source area (Gerhard et al., 1978) even
et al., 1956) leaves open the possibility that the Jeal- if they had existed during Jealousy deposition. No
ousy Formation and any underlying units could ex- data exist on presumed paleocurrent directions in the
tend as far as the Oligocene or lower. Jealousy Formation.

348 I. GILL et al.
146 m
M2
M8 North
South
A A'
30 m - M3 /~ ,--30 m
i---_----------_~
SL-
-N
;,.o.O'.o.. o'.,~~;;
N SL
-30 m-- o . ; O * Oo-, O* oo-o

-30 m
;-o-O"o. o'.o:. ;;
o:o..Z" ~z.',
c~; o * o d
;'o%~ i ,,i i i j,_ ,/ / | DOLOMITE
o . p ~ / I I ! I I 7-'--I"
~-~;;/'-~-I ~ ' '-"' ~ v-'-r:'r TEST
HOLE ~ ALLUVIUM
1000 rn
I~.| ~eL-l-,.J~/-- I -- I-- I i i FE~ KINGSHILL FM (MANNINGS BAY MBR) & BLESSING FM
~ ]~ ! . l .I I ./ I--I I--
I I
~ iI J
i
i
t Y'-t-:-'i-
I/-I'i--I
-I
-- VERT. EXAG. KINGSHILL FM (LA REINE MEMBER)
' I I I I~ l --I-- I--I-- I 50 X
I I I I/l--/ I:1 I-
JEALOUSY FORMATION
r~i CRETACEOUS SlLIClCLASTICS
West East
B B'
30 m
Fairplain Fault
M5 1-30
WD272 PA7 PAl0
M4
B7 M9
SL I I "~,~W~ &~' &~.':." ~ "o-

_ : i i Ii~-." o...% I~-.'-~ g * . / l - - - - - I ~ ,~
,,,, ~ - ; : ~ h ~ ~ ~ ~
, ~ 1,, ~.o....:~ i~:..q.~ z
-30 9- ~.~;~" ~.~" 3
I J I~ J I~/ ~'~ 9
,,. I I..I I I I ~ ' ~ "]
__~ .I,.[ i i i I I
./ / . DOLOMITE
TEST
HOLE ALLUVIUM
1000 m
FE~ KINGSHILL FM (MANNINGS BAY MBR) & BLESSING FM
VERT. EXAG. KINGSHILL FM (LA REINE MEMBER)
50 X
' 1 JEALOUSY FORMATION
CRETACEOUS SILICICLASTICS
. . . . . . . . . . . . ,R .
Fig. 5. Geological crossections through the Kingshill basin. (A) North-south cross-section A-A'. Note that the Jealousy Formation
surface roughly follows the topography of the Kingshill Limestone. (B) East-west cross-section B-B'. A normal fault (the Fairplain
fault) forms the western boundary of the small graben on the south coast occurs between test wells M 1 and M4. The Jealousy Formation
was not reached to the east of this fault.
Kingshill Limestone plain and can also be mapped in the subsurface

based on well data. It is composed of limestones
The blue-gray marls of the Jealousy Formation and buff pelagic marls with an upward-increasing
are succeeded upward by the more carbonate-rich proportion of shelf-derived sediment gravity-flows.
succession of the Kingshill Limestone. The Kingshill Lithologic variations permit it to be divided into
Limestone crops out over large areas of the central two members (Gill et al., in press): the interbedded

Rattan/Belvedere, and Morningstar sections). In the

subsurface, this interval is dominated by plank-
tonic foraminiferal packstone, with less common
lithic-pebble or foraminifera-rich wackestones. The
boundary between it and the underlying Jealousy
Formation is marked by a distinct color change from
tan above to blue-gray below. However, the signif-
icance of this change is unclear; sedimentological
and micropaleontological evidence reveal no notable
change in lithology, mineralogy, or depositional en-
vironment, nor is any hiatus resolvable.
The stratigraphically highest part of the La Reine
Member is exposed in the Airport/Penitentiary sec-
tion along the Melvin Evans Highway, where it
is disconformably overlain by the Mannings Bay
Member (Fig. 7). This interval is characterized by
regularly bedded intercalations of softer, planktonic
foraminifera-rich beds and more indurated, graded
beds of shelf-derived debris. The quantity of shelf-
derived sand is greater than lower in the member,
and burrowing appears to be more pervasive.
The La Reine Member in the St. John's/Judith
Fancy area includes beds of calcareous conglomerate
composed of rounded terrigenous gravel and a fauna
of shallow-water echinoids and benthic foraminifera.
Previously, these and nearby outcrops in the North-
side Range have been considered to be shelf and
Fig. 6. Interbedded planktonic foraminiferal packstones and sed- lagoon deposits or part of the Jealousy Formation,
iment gravity flows of shallow-water debris at Villa La Reine based on their similarity to conglomerates encoun-
outcrop, type section of the Kingshill Limestone (La Reine
Member). Biogeographic control places this outcrop near the tered in the CCC Test Well 39 (Gerhard et al., 1978;
boundary of the Middle Miocene and Upper Miocene. This is Lidz, 1982; Andreieff et al., 1986). However, be-
denoted as location VR in Fig. 4. cause the conglomerate beds are overlain in outcrop,
and underlain in Well M10 (Fig. 4), by planktonic
foraminiferal packstones, we interpret them as al-
marls and limestones of the La Reine Member and lochthonous beds occurring within a succession of
the benthic-foram-rich, burrowed limestone of the typical La Reine Member deep-water strata. In ad-
Mannings Bay Member. dition, structural relations support inclusion of these
strata in the Kingshill Limestone rather than the
Lower Kingshill Limestone- La Reine Member Jealousy Formation. Because these exposures occur
at elevations similar to outcrops of the La Reine
Lithology and distribution Member only a few kilometers away (Fig. 4), fault-
The La Reine Member is composed of interbed- ing within the graben would be required to raise
ded planktonic foraminifera-rich marls and shallow- the stratigraphically lower Jealousy beds to the same
marine limestone debris beds with increasing propor- elevation as the nearby Kingshill. However, there is
tions of downslope-transported material upsection. no evidence of such faulting.
Typical lithologies of the La Reine Member are ex- The maximum thickness of the La Reine Member
posed at the type section of the Kingshill Limestone at encountered in the test wells drilled for this study is
Villa La Reine (Fig. 6). The outcrop is a rhythmically approximately 140 m (Fig. 5A). Cederstrom (1950)
bedded alternation of polymictic packstones (Gerhard reported a thickness range from 0 to 180 m for the
et al., 1978), some with boulder-sized coral heads, Kingshill Limestone, the larger figure referring to
and deep-water planktonic foraminiferal chalks and extrapolated thickness in the carbonate highlands of
marls. Similar lithologies occur at Fredensburg the Rattan Hill area. Isopach patterns reveal thinning
Quarry and Estate Work and Rest, but breccia beds of the La Reine Member toward the north and
at the latter include terrigenous material presumably northwest margins of the basin. The formation shows
derived from the Cretaceous Mt. Eagle Series. a pronounced thickening in the carbonate highlands
The lower part of the La Reine Member is similar close to the northern coast of St. Croix and less
but includes less transported debris (Five Corners, pronounced thickening toward the south, interrupted

350 I. GILL et al.
Fig. 7. Disconformable contact between the La Reine and Mannings Bay Members of the Kingshill Limestone at the Airport/Penetentiary
outcrop, along Melvin Evans Highway, southern St. Croix. Disconformity occurs midway up the outcrop, approximately 7 m above road
level. This is denoted as location A/P in Fig. 4.
by post-depositional faulting along the south coast ically within the range of the La Reine Member,
(Fig. 5). In general, these isopach patterns follow supporting the lithologic and structural arguments
the trends of the top of the Jealousy Formation. against their inclusion in the Jealousy. Van den Bold
If deformation is ignored, the Kingshill Limestone (1970, in Gill, 1989, and in McLaughlin et al., 1995)
isopach patterns imply a basin opening to the south considers the ostracode fauna indicative of a position
but with a deep area under the modem carbonate near the Lower Miocene-Middle Miocene boundary.
highlands in the north. The fauna is completely different from that of the
subsurface Jealousy Formation, but contains several
Age and paleoenvironment species in common with the lower part of the La
The La Reine Member extends from basal Middle Reine Member in the subsurface.
Miocene to approximately the Miocene-Pliocene Middle Eocene to Early Miocene foraminifera
boundary (McLaughlin et al., 1995). The subsur- were described by Lidz (1984b) from a mud clast in
face section sampled in our drilling program is the the La Reine Member. Its presence in these deep-
stratigraphically lowest, ranging from basal Mid- water strata indicates that older Tertiary sediments
dle Miocene (Praeorbulina glomerosa zone) to the were being eroded and transported by sediment
medial Middle Miocene (Globorotalia fohsi robusta gravity flows during the deposition of the La Reine
zone). The stratigraphically lowest outcrops are in Member. This is consistent with the occurrence of
the northern part of the island (Salt River Valley and pebbles in both the Jealousy Formation and King-
Five Comers, Fig. 4), where the La Reine Member shill Limestone that are assumed to be derived from
is placed in the lower part of the Middle Miocene the Cretaceous strata presently exposed in the high-
(Globorotaliafohsifohsi zone, possibly to Praeorbu- lands of the basin-bounding fault blocks.
lina glomerosa zone). The type section of the King- The benthic foraminifera of the La Reine Mem-
shill Formation at Villa La Reine represents the mid- ber in Wells M1, M2, and M10 comprise a middle
dle part of the formation, with faunas indicative of bathyal fauna (600-800 m water-depth) that differs
the upper part of the Middle Miocene (Globorotalia little from that of the underlying Jealousy Formation
mayeri zone, possibly to the Globorotalia menardii (McLaughlin et al., 1995). No significant paleoenvi-
zone). The top of the member crops out on the ronmental shift is evident at the boundary. Although
south side of the island (Airport/Penitentiary sec- there are significant numbers of shallower-water
tion), where it is placed near the top of the Miocene species in some of the stratigraphically higher out-
(upper part of Globorotalia humerosa zone). crop samples from the La Reine Member, the faunas
The outcrops in the St. John area previously in the finer-grained beds that over- and underlie these
mapped as Jealousy Formation fall biostratigraph- samples are middle bathyal types, indicating that the

shallow-water forms are present due to downslope parts of Evans Highway, notably the upper part of
transport. the Airport/Penitentiary roadcut (Figs. 4 and 7). It
was also examined in several of the test wells drilled
Source area and paleocurrents for this project.
Paleocurrent indicators in the lower Kingshill The Mannings Bay Member rests disconformably
Limestone at the Villa La Reine type section show on the La Reine Member, from which it was distin-
west-southwest flow. Four measurements included a guished based on higher abundance of shallow-water
pebble halo around a boulder and orientations from carbonate material (Gill, 1989; Gill et al., in press).
cross-lamination and oscillation ripples (Gerhard et At the Airport/Penitentiary section, the disconfor-
al., 1978). Clasts larger than 4 mm are concentrated mity is evident as a scour surface with more than
in the northeast portion of the basin (Gerhard et al., 1 m of relief. This surface appears to have been
1978). None of the reef-derived material, including scoured by partly channelized submarine flows of
boulder-sized coral heads, is in-place, and this ma- shelf-derived sediment (Lidz, 1984a; Gill, 1989; Gill
terial must therefore be derived from a nearby shelf et al., 1990). The Mannings Bay Member includes
area. the strata referred to as a "benthic foraminiferal
wackestone and grainstone facies" in the Kingshill
Upper Kingshill Limestone: Mannings Bay Limestone by Gerhard et al. (1978) and those strata
Member separated from the Kingshill Limestone as "post-
Kingshill" limestones by Lidz (1982) and Andreieff
Lithology and distribution et al. (1986).
The Mannings Bay Member of the Kingshill The Mannings Bay Member and the overlying
Limestone is characterized by channelized beds of Blessing Formation are best developed in a small
grainstones rich in shelf debris, interbedded with graben on the south coast of St. Croix (Fig. 8). They
softer wackestones and packstones. The grainstones can be difficult to differentiate from one another in
contain abundant Operculinoides cojimarensis and core but together total more than 50 m thickness
Paraspiroclypeus chawneri (Behrens, 1976; Gerhard in some of the wells (Gill and Hubbard, 1986,
et al., 1978; S. Frost, oral commun., 1986). Many 1987). The westernmost documented extent of the
specimens show signs of transport or reworking, Mannings Bay Member is near the western fault
such as fracturing, abrasion, and imbrication. The boundary of this graben near Fairplain. The eastern
foraminiferal wackestones and the softer packstones edge of this graben is perhaps indicated by where
also include significant quantities of planktonic the stream flow makes an abrupt southerly turn to
foraminifera. These lithologies are well exposed the coast against the exposed Pliocene reef complex
in the type section at the quarry on the southeast (Fig. 9).
side of Mannings Hill (Gill et al., in press) and along
Fig. 8. Disconformable contact between the Mannings Bay Member of the Kingshill Limestone (below) and the Blessing Formation
(above) at the Airport Quarry outcrop. This is denoted as location AQ in Fig. 4.

352 I. GILL et al.
Fig. 9. Facies map for south coast industrial area. Dolomite in the vadose zone or exposed in outcrop is patchily distributed in an arcuate
region following the Pliocene reef trend. Dolomite presently in the phreatic zone is found in off-shore facies. The western boundary of
the small, south coast graben is well-defined by a normal fault. The northern and eastern boundaries are poorly known.
Age and paleoenvironment itic species and forms associated with shallow-water
Planktonic foraminifera are generally uncommon carbonate environments (McLaughlin et al., 1995).
and poor in the Mannings Bay Member. The basal In outcrop and core samples, the larger benthic
part of the member, in the Airport/Penitentiary foraminifera Operculinoides and Paraspiroclypeus
section, contains faunas suggestive of the lower are abundant; these forms were likely photic-zone
Pliocene Globorotalia margaritae zone (McLaugh- inhabitants (S. Frost, oral commun., 1986). Other
lin et al., 1995), consistent with Lower Pliocene bioclasts that contribute to the facies are coralline al-
findings for this same section by Andreieff et al. gal crusts, rhodoliths, echinoid spines and plate frag-
(1986, 1987) and Lidz (1982). A subsurface sample ments, coral fragments, and molluscan debris. These
from Well M4 yields a fauna indicative of an inter- forms, the poorly developed planktonic fauna, and
val near the Miocene-Pliocene boundary, between the evidence for transport in the larger foraminifera-
the Globorotalia humerosa zone and the Globigeri- rich beds together indicate that the shallow-water
noides trilobus fistulosus zone (McLaughlin et al., material was carried into a deeper shelfal setting of
1995). approximately 100 m depth.
The foraminiferal control from above and be-
low the disconformity at the base of the Mannings
Bay Member suggests a general chronostratigraphic Source area and paleocurrents
position at or near the terminal-Miocene (5.5 Ma) The beds of larger benthic forams show sort-
eustatic fall, but does not provide sufficient resolu- ing and imbrication, the result of extensive cur-
tion to tie it exactly to this event, as has been pro- rent working. Sediment transport direction in the
posed by Lidz (1982). The biostratigraphic control Kingshill basin was dominantly to the west-north-
is also insufficient to determine whether a signif- west, and ranged from southwest to north-north-
icant chronostratigraphic interval is missing at the east. This is based on 42 measurements of imbri-
disconformity. cated large benthic foraminifers on Mannings Hill,
Smaller benthic foraminifera recovered from the probably the same outcrop referred to here as the
Airport/Penitentiary section are a mix of outer ner- Airport/Penitentiary outcrop (Gerhard et al., 1978).

Blessing Formation (Lidz, 1982), its position above the Lower Pliocene
Mannings Bay Member (Andreieff et al., 1986;
Lithology and distribution McLaughlin et al., 1995), and the occurrence of pre-
The highest stratigraphic unit, the Blessing For- Pleistocene species of the scleractinian corals Telio-
mation, represents a Pliocene reef tract that extended phyllia, Stylophora, and Thysanus (Behrens, 1976).
across the south and west coastlines of St. Croix. No planktonic foraminifera were recovered from our
The reef tract consisted of interspersed reefs and samples. The macrofaunal assemblages within the
shelf systems similar to the arrangement of reefs Blessing Formation represent co-existing reef, for-
around the modem south coastline of St. Croix. The ereef, and lagoon environments that extended along
classic reef model with flanking fore- and backreef the south and west coastlines of St. Croix.
facies does not appear to apply here. Reef systems
on St. Croix apparently formed planar deposits with Source area and paleocurrents
little topographic relief. This planar geometry is ap- Much of the exposed Blessing Formation appears
parently common in Caribbean Tertiary reef deposits to be in-place. The morphology of the reef tract
(S. Frost, oral commun., 1986). mimics the present shoreline, and the reefs would
The Blessing Formation was sampled in outcrops have been affected by open-ocean conditions to the
near the south coast and in cores. Based on subsur- south and to the west.
face data, its greatest thickness is in a small graben,
just east of the Fairplain fault, where it may be up
to 30 m thick. Scattered exposures of reef facies EVOLUTION OF THE KINGSHILLBASIN
also occur west of the Fairplain fault along Evans
Highway, a location near Fredericksted, and at an In previous studies, the Kingshill basin was
exposure described by Gerhard et al. (1978); its thought to have formed in the Oligocene as a result
maximum thickness in that area is estimated to be of vertical tectonic movement. Whetten (1966) char-
between 10 and 20 m. The best exposures of this acterized the carbonate section of the Kingshill basin
unit are in its type section at a road cut next to as reefal and estuarine deposits that accumulated in
the Hess Oil refinery (Gill et al., 1990, in press) a graben in the central part of the island follow-
(Figs. 4 and 10). Reefal facies are predominantly ing a period of low-rank metamorphism, faulting,
composed of coralline boundstones characterized folding, igneous intrusion, and uplift. He concluded
by external molds of scleractinians, gastropods and that there was no significant evidence for strike-slip
pelecypods, as well as skeletal debris. The scler- motion north of St. Croix, and that the Neogene
actinians include species of extant genera such as section was affected only by vertical tectonics. More
Agaricia, Diploria, Montastrea, and Siderastrea, as recent studies (e.g. Multer et al., 1977; Gerhard
well as the extinct forms Stylophora, Teliophyl- et al., 1978; Lidz, 1982) have accepted this struc-
lia, and Thysanus (Behrens, 1976). Lagoonal facies tural framework; for example, Multer et al. (1977)
include skeletal wackestones composed of shallow- envisioned the Northside Range and the East End
water foraminifera, coralline algae and a wide vari- Range as subaerially exposed horst blocks that pro-
ety of shallow-water invertebrates. vided both terrigenous and shelf-derived carbonate
The Hess outcrop is marked by several well- debris to the basin. However, these studies recog-
cemented undulatory layers (Fig. 10) distinguished nized that deposition took place in a deep-marine
by abrupt light stable-isotopic excursions, an on- setting, which they envisioned as a seaway opening
lap surface (Gill, 1989) and karstification (Lidz, to the northeast and southwest in a basinal setting
1984a; Gill, 1989). Nearby exposures were appar- similar to modem deep basins north of St. Croix.
ently marked by terra rosa beds within the Blessing
Formation (Behrens, 1976) and underneath it (S. Basin-margin faulting and basin formation
Frost, pers. commun., 1986; L. Gerhard, written
commun., 1997). These surfaces indicate that the Subsurface stratigraphic evidence indicates that
south coast reef trend of St. Croix was exposed the Kingshill basin graben began to form no earlier
several times during the Pliocene (Gill, 1989; Gill et than the latest part of the Middle Miocene, during
al., 1990). deposition of the upper part of the La Reine Mem-
ber. The Jealousy Formation and the lower part of
Age and paleoenvironment the La Reine Member show no evidence of tectonic
The Blessing Formation is loosely placed by bios- activity. These strata were deposited at water depths
tratigraphic data in the interval between the upper of approximately 600 m during the early part of the
part of Lower Pliocene and the top of the Pliocene. Middle Miocene (McLaughlin et al., 1995). Previous
This assignment is based on the occurrence of Glo- studies have suggested the existence of an active
bigerina nepenthes in the Hess refinery outcrop graben at this time with subaerially exposed mar-

354 I. GILL et al.
Fig. 10. Reefal facies of the Blessing Formation at the Hess Cut outcrop, southern St. Croix. HG = coral reef hardground, with onlap
onto exposure surfaces (ES). CAL -- undulatory caliche layer under karst cavities. Backpack and hammer for scale at arrow. This is
denoted as location HC in Fig. 4.
gins. However, the existence of tentatively identified either requires significant lateral translation of the
deep-water marls of the Jealousy Formation outside St. Croix platform.
the graben in CCC Test Well C-26 (Cederstrom, We suggest that initiation of the St. Croix fault
1950) indicates that faulting on the graben margins system occurred no earlier than in the latest part of
must post-date these early Middle Miocene deposits. the Middle Miocene (Fig. 11). The contacts between
In addition, exposed graben margins would re- the lower Kingshill Limestone and the Cretaceous
quire a marginal slope exceeding 45 ~ based on the rocks on the east and west sides of the graben have
measured distance from the test well samples in been interpreted as faults by Multer et al. (1977)
the basin to the hypothetical exposed margin on the although they were mapped simply as stratigraphic
northwest side of the graben. This angle is compa- contacts on the geologic map by Whetten (1966).
rable to that of the modem slope north of St. Croix, The western contact, along the edge of the Northside
which is dominated sedimentologically by input of Range, is mostly obscured by alluvial cover (Fig. 4).
shelf-derived material (Hubbard et al., 1981; Gill, Gerhard et al. (1978) have suggested that displace-
1983), including reef foraminifera such as Amphis- ment along the eastern fault boundary of the graben
tegina (B. Sen Gupta, oral commun., 1984). Our was greater than that along the western boundary.
samples from the Jealousy Formation and the lower The eastern fault contact is sharp and characterized
part of the La Reine Member near this boundary by offset of both Cretaceous and Kingshill strata.
do not show the major input of shelfal material that This age of graben formation is constrained by
would be expected with deposition at the foot of a fault relations between the La Reine Member and
similarly steep slope. the Cretaceous strata along the eastern boundary of
If a steep-sided Kingshill basin did not form the graben, which indicates that at least the lower
before the middle Miocene, then the scattered con- part of the La Reine Member was deposited prior to
glomerates and shelf-derived sediments that do oc- basin faulting (Gill and Hubbard, 1986, 1987). Beds
cur in the Jealousy Formation and lower part of the of coral debris and rounded pebble conglomerate
Kingshill Limestone must be derived from some- exist low in the exposed section near Judith's Fancy
where other than St. Croix. These lithologies suggest and St. John and were penetrated by test hole
that St. Croix was close to a land mass capable of M10. These materials were interpreted as in-place
supporting reef growth and supplying clastic materi- deposits by Gerhard et al. (1978) and Lidz (1982),
als to the deep-water environment during the Middle but are interpreted as allochthonous deposits here
Miocene. Puerto Rico and the Virgin Islands plat- (Gill, 1989). Similarly, the type-section at Villa La
form, to the northwest of St. Croix, and Anguilla Reine (Gerhard et al., 1978) records the influx of
and Saba to the northeast, are possible source areas; shallow-marine and terrigenous debris in the basin.

Fig. 11. Block models of the evolution of the Kingshill basin of St. Croix from the Early Miocene to Recent. The model reflects
overall tectonic quiescence until near the end of the Middle Miocene when significant tectonism, uplift, and erosion of uplifted Jealousy
Formation sediments apparently began. The Late Miocene to Plio-Pleistocene diagrams trace shoaling in the basin from bathyal to
shallow-marine and reefal deposition.
This section is placed near the Middle-Late Miocene tion (Fig. 4), angular clasts of typical Cretaceous Mt.
boundary (Andreieff et al., 1986; McLaughlin et al., Eagle Group lithologies form breccia layers between
1995), and includes a number of beds that contain beds of the La Reine Member. The Estate Work and
large coral heads and siliciclastic material probably Rest section is placed biostratigraphically near the
derived from exposed and eroded rock outside of Middle Miocene-Late Miocene boundary, between
the graben. Together these data suggest a shift from the Globorotalia mayeri and Globorotalia acostaen-
relative tectonic quiescence to uplift activity near the sis zones (McLaughlin et al., 1995). This breccia may
end of the Middle Miocene. provide the earliest evidence for fault activity on the
Breccia beds occur in strata of approximately the eastern boundary of the graben.
same age at Estate Work and Rest along the eastern
edge of the basin. These beds have been interpreted as Eustatic events and basinal shallowing
syntectonic breccia by Gerhard et al. (1978) in an out-
crop that could not be located for this study. In nearby The effects of eustasy superimposed on tectonic
exposures, including the Estate Work and Rest sec- uplift produced the disconformity separating the La

356 I. GILL et al.
Reine and Mannings Bay Members at the Airport/ below sea level in CCC Test Well 45a, located 60 m
Penitentiary outcrop. This disconformity was inter- to the east of Well M1 (Fig. 4); and deeper than the
preted by Lidz (1984a) as evidence of the Messinian 80 m below sea level bottom-hole depth of Well M4
eustatic fall of Haq et al. (1988). Our sedimentologic less than 180 m to the east of M1. Minimum vertical
and paleontologic data indicate shoaling across this fault displacement is 68 m, based on the occurrence
surface, from depths of approximately 200-300 m of the La Reine Member-Mannings Bay Member
in the La Reine Member to 100 m in the Mannings contact at 24 m above sea level on Mannings Hill
Bay Member. There is no evidence of soil forma- west of the fault, and approximately 44 m below sea
tion, dissolution, or karsting to indicate subaerial level in Well M4, east of the fault.
exposure. Based on this evidence, we believe this The presence of a fault at this location is also sup-
disconformity represents submarine erosion during ported by surface features. Strata in nearby outcrops
the uplift of the island in the Late Miocene. The along Evan's Highway (Fig. 4) dip toward the fault
global sea-level fall associated with the Messinian line and ephemeral streams locally approximate the
event is of approximately the same age and could trend of the fault near the coastline (Fig. 9). The east-
have triggered erosion via sediment gravity flows. ern edge of the presumed graben is not well marked.
However, biostratigraphic control (McLaughlin et Stream flow north of Well M5 is from west to east,
al., 1995) cannot precisely tie the timing of this oddly parallel with the coastline. The eastern edge of
unconformity to the Messinian event, nor confirm a this graben may be marked by the ephemeral stream
single event as its cause. drainage turning abruptly south where it meets the
The strata above the unconformity record the de- exposed Plio-Pleistocene reef trend (Fig. 9). Surface
velopment of extensive foraminiferal-coralline algal features also suggest that a northern hinge-line exists
bank environments on St. Croix during the early for this small graben, just north of the industrial
part of the Pliocene (Fig. 11). The environments areas along the coast. Together, these fault orienta-
served as shallow-water sources of the larger benthic tions suggest that normal faulting on the margins of
foraminifera present in the Mannings Bay Mem- this graben was produced by the same extensional
ber, in particular Operculinoides cojimarensis and tectonic regime that initiated faulting on the margins
Paraspiroclypeus chawneri (Gerhard et al., 1978). of the Kingshill basin during the latest part of the
This environment is not present in modem St. Croix, Middle Miocene.
where coral ecosystems predominate. It is likely that The stratal relations across the Fairplain fault sug-
foram-algal ecosystems were supplanted by corals gest that the south-coast subsidiary graben existed
with the extinction of many larger foram groups in as an entity during and after deposition of the Man-
the Neogene (Frost, 1977). It is also possible that nings Bay Member foraminiferal-algal facies. The
the relative lack of coral debris could be the result fact that the greatest thickness of these deposits is
of (1) the upslope storage of coral reef sediments, preserved in the graben has two possible explana-
with minor deposition only at sporadic intervals (e.g. tions: the graben formed a marine embayment along
Moore et al., 1976), (2) changes in circulation, nu- the south coast where these facies accumulated; or
trient or temperature conditions yielding competitive the strata were preferentially preserved within the
advantage to the foraminifera-algal community, or subsiding graben during island uplift. The former
(3) rapid uplift and eustatic variation suppressing the alternative suggests that the faulting produced topo-
establishment of coral reef systems. graphic relief prior to and during deposition of the
foraminiferal-algal facies, whereas the latter alterna-
Fairplain fault tive requires only post-depositional faulting. We be-
lieve that both processes are likely to have occurred.
The tectonism that produced the shallowing
across the boundary between the La Reine and Pliocene reef tract and subaerial exposure
Mannings May members is also reflected in faulting
within the Kingshill basin. The Fairplain fault, which Continued tectonic uplift and shoaling of the
marks the western margin of the small graben on the Kingshill basin resulted in deposition of the Blessing
south coast, cuts through the Mannings Bay Mem- Formation reef tract (Fig. 11). The greatest thickness
ber and Blessing Formation, indicating that motion of reef growth is found in the Krause Lagoon area
occurred on this fault during the Late Pliocene or on the south-central coastline where the arcuate
later. The orientation of the Fairplain fault is roughly distribution of reef and lagoonal facies suggest the
parallel to the orientation of the northeast-trending existence of an embayment (Fig. 9). The size and
main basin boundary faults. It dips at least 20~ to the shape of this embayment was probably controlled by
east, as indicated by the depth to the Jealousy For- faulting along the margins of the south coast graben.
mation at the fault contact in several well sections: The Blessing Formation contains indications of
29 m below sea level at Well M1 (Fig. 13); 53 m more than one period of Pliocene subaerial exposure

along its southern coastline (described in a previ- and coastline of the island are less steep and less
ous section), as well as an onlap surface near the suggestive of recent faulting.
Hess Oil Refinery. We suggest that eustatic vari- On submersible dives in DSRV Alvin along the
ations superimposed on the overall tectonic-uplift south wall of the Virgin Islands basin off the north
trend account for the evidence of exposure noted coast of the island, Dill (1977) encountered struc-
in our field studies in St. Croix, the same as Lidz tures he interpreted as fault gouge in St. Croix
(1984a). The accumulation of the Pliocene sedi- basement rocks. In similar dives off the northwest
ments only on the south coast, the apparent extent of coast, a vertical escarpment greater than 10 m in
erosion/non-deposition in the northern central plain height was observed at a depth of greater than 2600
at the same time, and the general southerly dip of m (Hubbard et al., 1981; Gill, 1983). The escarpment
Neogene strata in the Kingshill basin suggest that was composed of dark, terrigenous rock similar in
Late Pliocene uplift preferentially raised the northern appearance to the Cretaceous Mt. Eagle Group rocks
part of the island. that form the east and west hills of St. Croix. If
in-place, the face of the escarpment suggests the
Subsidence analysis role of fault-offset in creating the north slope of St.
Croix.
Subsidence analysis (Fig. 12) indicates that the Like the southern margin of the Virgin Islands
majority of the Neogene shallowing in the Kingshill basin off St. Croix, the northern margin off Vieques
basin is due to tectonic uplift. The Kingshill basin and St. Thomas is also characterized by very steep
shallowed from as much as 800 m of water depth in slopes (Fig. 13). If the Virgin Islands basin formed as
the Middle Miocene (ca. 10.5 Ma) to approximately a result of rifting, these slopes may represent scarps
100 m in the Early Pliocene (ca. 3.5 Ma). Given formed during the rifting event. If these scarps were
a modem stratigraphic thickness of 250 m for the initially juxtaposed, the present position of St. Croix
study interval, and correcting for the effects of could have been achieved by movement of the island
sediment loading, water loading, and compaction, south and east relative to its initial position in the
we calculate 400 m of tectonic uplift in this interval, Virgin Islands platform. This motion would require a
which translates into a rate of 57 m/Ma. combination of left-lateral movement and tensional
separation, which is consistent with the oblique
left-lateral strike-slip motion for the formation of the
TECTONIC EVOLUTION OF ST. CROIX AND THE Virgin Islands basin.
NORTHEASTERN CARIBBEAN
Fault orientation
Left-lateral oblique motion models All documented Neogene faults on St. Croix are
normal faults, including the graben-bounding faults,
Stratigraphic and structural evidence indicates and strike in a northeasterly direction, oblique to the
that the Kingshill basin graben began to form no south margin of the Virgin Islands basin. On the St.
earlier than the late Middle Miocene. St. Croix may Croix Ridge, seismic profiles and GLORIA imagery
have began to rift away from Puerto Rico during this (Masson and Scanlon, 1991) indicate that the ridge
period by oblique left-lateral faulting, movement that is broken into a series of block-fault 'piano key'
could have opened the Virgin Islands basin (Fig. 13). structures with the same northeasterly orientation as
Similar left-lateral faulting may have also occurred the St. Croix faults (Fig. 13). These structures are
between St. Croix and the Saba Bank area to the interpreted to be the products of normal faulting
northeast, but structural and bathymetric relations similar to those in the Neogene section of St. Croix
in the intervening St. Croix basin (Fig. 1) are less (Holcombe, 1977).
clear. A left-lateral tectonic model is consistent with The northeasterly orientation of the apparently
several lines of evidence. continuous set of tensional fractures is consistent
with the type of deformation expected in a left-lat-
Bathymetry eral wrench-fault zone aligned along the Anegada
Bathymetric profiles along the north coast of St. Passage. Such fractures tend to form parallel to the
Croix are rugged and steep, sloping between 23 and short axis of the strain ellipse in clay models (Wilcox
45 ~ to the center of the Virgin Islands basin and et al., 1973). The northeasterly fault orientation is
dropping off at nearly vertical angles near the shelf inconsistent with right-lateral strike-slip movement,
edge (Fig. 13). The northwestern shoreline is carved which would produce normal faulting of the oppo-
from cliffs of the Northside Range, prompting Mey- site orientation. Alternately, a model similar to that
erhoff (1927) to suggest relatively recent faulting of Geist et al. (1988) may apply here.
and uplift for the northern coast, probably no older The consistent orientation of the St. Croix and
than the Pliocene. Gradients of the southern shelf St. Croix Ridge fracture system implies a tensional

358 I. GILL et al.
Fig. 12. Subsidence analysis for the Neogene of the Kingshill basin. Upward trend of tectonic subsidence curve indicates uplift beginning
at approximately 10.5 Ma, coincident with initiation of formation of the Kingshill basin boundary faults. Tectonic subsidence is calculated
based on stratigraphic thicknesses, age, and paleobathymetry of each unit, decompacted thickness of the strata based on lithology, and
removing the effects of sediment and water loading. The sea floor line represents the position of the sea bottom based on paleobathymetry
with eustatic contributions removed.
origin under a consistent tectonic regime (Fig. 13). Meyerhoff, 1971). This area is presently more than
The sharply defined walls of the Virgin Islands 60 km to the northwest of St. Croix. Given these
basin, including the north slope of St. Croix, suggest distances, St. Croix was probably much closer to
relatively recent tectonic activity along this area. these potential sources of coarse clastics during the
Assuming that the origin of this faulting is connected Middle Miocene than it is today. Such a reconstruc-
to movement in the Virgin Islands basin/Anegada tion would require tens of kilometers of left-lateral
Passage, the orientation of the fault system on St. movement to place St. Croix in its present position.
Croix suggests left-lateral wrench faulting north of A second potential sediment source is the
St. Croix beginning in the late part of the Middle Anguilla/Saba Bank area to the east. Shelf carbon-
Miocene. ates of the same age as the Jealousy Formation and
To the north of the Virgin islands basin, the the Kingshill Limestone exist in Anguilla (Van den
islands of St. Thomas and St. John both show ex- Bold, 1970) and contain very similar ostracode fau-
tensive faulting. Donnelly (1966) mapped a graben nas. Saba Bank is underlain by rocks interpreted to
structure on both islands that also strikes north- be early Tertiary carbonates and fluvio-deltaics (Ne-
easterly and displaces Cretaceous rocks. Left-lat- mec, 1980; Warner, 1989) with a possible sediment
eral strike-slip displacement is apparent within the source on Puerto Rico. If St. Croix was originally
graben. Although these faults may pre-date St. Croix juxtaposed with either of these areas, fault mo-
strata, they show a similar orientation. tion would also be left lateral, assuming movement
crudely parallel to the Anegada Passage. Lateral
Sediment sources for the Kingshill basin movement from Anguilla to the present location of
Because no uplifted horst blocks were available St. Croix would require a greater travel distance than
as sediment sources before the Middle Miocene would movement from Puerto Rico. Unfortunately,
formation of the Kingshill basin graben, an exter- information on the basins between St. Croix and
nal source of coarse clastics is required to explain Saba Bank is too sparse to allow more thorough
the significant quantities of conglomeratic deposits evaluation.
present in the type section of the Jealousy Formation Interestingly, the problem of a sediment source
(Cederstrom, 1950). The southeastern part of Puerto exists for Cretaceous St. Croix as well. Conglom-
Rico contains exposures of Tertiary carbonates that eratic deposits of rudistid bivalves are found within
extend eastward of Puerto Rico only as far as the deep marine sedimentary rocks on St. Croix (Whet-
southern coastline of Vieques Island (Khudoley and ten, 1966). The closest documented source for these

E V O L U T I O N OF THE N E O G E N E K I N G S H I L L BASIN OF ST. CROIX, U.S. V I R G I N I S L A N D S 359
Fig. 13. Migration model for Tertiary St. Croix and opening of the Virgin Islands basin showing hypothetical positions for St. Croix
between the Early Miocene and Recent, assuming an initial position south of Vieques. The bottom diagram contrasts this model with an
alternative model that assumes more of a north-south relative motion across the Virgin Islands basin. Position numbers for the same time
periods are indicated with italics.

360 I. GILL et al.
rudists of this type is Puerto Rico (H. Santos, pers. age of 11 Ma for initiation of motion and a distance
commun., 1998). of lateral movement of 91 km. This rate is somewhat
slower than the estimated 20 m m / y r or greater
Seismicity rate of movement between the North American and
Seismic activity today is detectable only in the Caribbean plates cited by Rosencrantz and Mann
shallow zones in the north wall from 0 to 50 km (1991). This slower rate is consistent with the idea
deep. These seismic events occur in swarms and are of Puerto Rico and the Virgin Islands platform being
generally less than magnitude 3.2 (Frankel et al., decoupled from the Caribbean plate (McCann et al.,
1980). Historic records indicate that the potential 1987) and moving eastward with a slower relative
exists for much larger earthquakes in the Virgin motion. The resulting differential motion between
Islands basin/Anegada Passage area. Two major the Virgin Islands platform and the main body of
earthquakes caused damage in both St. Croix and St. the Caribbean plate, including St. Croix, may have
Thomas in 1867 and were calculated on the basis of caused the opening of the Virgin Islands basin.
tsunami arrival times to have originated in the north
wall of the Virgin Islands basin (Reid and Tabor,
Rotating platelet models
1920, cited in Frankel et al., 1980).
In general, the Virgin Island basin and Anegada
An alternative group of models call for counter-
Passage are no more active than the Puerto Rico
clockwise rotation of a Puerto Rico microplate or
Platform, and do not show seismic patterns corre-
terrane. The idea of a separate Puerto Rico platelet
lated with their bathymetry. To our knowledge, no
was proposed by McCann et al. (1987), who sug-
fault-plane solutions have been calculated for the
gested that this platelet had a westward (left-lateral)
seismic events currently taking place in the Virgin
motion relative to the main Caribbean plate. Lithgow
Islands basin. For these reasons, present seismicity
et al. (1987) suggested that the Virgin Islands basin
patterns do not support any one tectonic model or
formed as a pull-apart in response to these relative
significant movement along the Anegada Passage
motion differences.
today (J. Joyce, pers. commun., 1998).
Tectonic context Seismic profiling and sidescan sonar data

The Caribbean plate is interpreted to have an east- Based on seismic profiles and GLORIA sidescan
ward present-day motion relative to the North Amer- sonar surveys (EEZ Scan Scientific Staff, 1987),
ican plate. This plate motion is generally manifested Scanlon and Masson (1988) proposed that the Puerto
by sinistral slip along the northern Caribbean plate Rico microplate has undergone counterclockwise ro-
boundary, and by dextral slip zones along the south- tation, with a pole of rotation south of Puerto Rico
ern boundary (Stephan et al., 1986). Active docu- at the juncture between the St. Croix Ridge and the
mented subduction in the northeastern Caribbean is Muertos Trough (Fig. 14C). This model does not
presently taking place only along the Lesser Antilles address relative motion between the microplate and
arc. In the area near the Puerto Rico trench, the the Caribbean and North American plates. Based on
northern boundary of the Caribbean plate is char- paleomagnetic studies in Puerto Rico, Reid et al.
acterized overall by slip but with some evidence (1991) have documented approximately 25 ~ of coun-
for compression (Frankel et al., 1980; Burke et al., terclockwise rotation of Puerto Rico relative to North
1984). Given the modem left-lateral motion of the America between 11 and 4.5 Ma. They concluded
northern Caribbean plate boundary, it is reasonable that the Puerto Rico microplate behaved as a 'roller
to predict that the opening of the Virgin Islands basin beating' between the North American and Caribbean
between the Virgin Islands platform and St. Croix plates during this period as it either became uncou-
also reflects left-lateral movement. The consistency pled from Hispaniola or responded to changes in
and simplicity of this model is perhaps the reason relative plate motions. They suggested that this rota-
that left-lateral motion in the Anegada Passage was tion ceased as the Puerto Rico microplate detached
suggested by Hess (1933, cited in Whetten, 1966), from the Caribbean plate and transferred extensional
Hess (1966), Burke et al. (1984, table 7), and in stress to the Anegada Passage and Mona Canyon.
Case et al. (1984). Speed and Larue (1991) proposed that the north-
ern Caribbean PBZ has been in left-lateral transten-
Estimation of rifling rate sion for the last 15 to 20 m.y., with much of the
For a model in which St. Croix moves from motion dispersed by counterclockwise rotation of
a position south of Vieques, we estimate that the terranes within the PBZ such as the Puerto Rico-
Virgin Islands basin has opened with rate of lateral Virgin Islands terrane. In their model, as in the oth-
motion of approximately 8 mm/yr since the late ers, this rotation caused extension in the Anegada
Middle Miocene. This rate is based on an assumed Passage/Virgin Islands basin area.

E V O L U T I O N OF THE N E O G E N E K I N G S H I L L BASIN OF ST. CROIX, U.S. V I R G I N ISLANDS 361
Fig. 14. Three models for the tectonic evolution of the northern Caribbean. (A) Plate tectonic model for the northern Caribbean with
dextral slip in the Virgin Islands basin/Anegada Passage via a 'Puerto Rico Festoon' (after Stephan et al., 1986). Note that documented
sinistral faults through Puerto Rico are not indicated in this diagram. The inset shows a mechanical analog for the formation of a 'Puerto
Rico Festoon' with dextral slip to the east, i.e. Virgin Islands basin/Anegada Passage, and sinistral slip to the west (from Stephan et al.,
1986). (B) Plate tectonic model for the northern Caribbean with dextral slip in the Virgin Islands basin/Anegada Passage. Note proposed
triple junction to the southeast of Puerto Rico (after Jany et al., 1990). (C) Rotating microplate model of Scanlon and Masson (1988).

362 I. GILL et al.
Discussion Seismic profiling

With an axis of rotation situated to the southwest Based on north-northeasterly oriented seismic
of St. Croix (Scanlon and Masson, 1988), counter- profiles, Mauffret et al. (1986) have interpreted a
clockwise rotation of a separate Puerto Rico platelet structure on the north side of the Virgin Islands basin
would produce a zone of extension in the Anegada as a northward-verging reverse-fault zone. They cite
fault zone that widens to the northeast. In fact, the this faulting as evidence for fight-lateral slip during
Anegada Passage has narrow, subparallel walls until the formation of the basin. However, if this structure
it empties into the Sombrero basin to the northeast trends normal (east-southeasterly) to this seismic
and into the Virgin Islands basin to the southwest. profile, fight-lateral slip should produce extension
For this reason, the bathymetry of the region does rather than compression. Therefore, this structure
not support simple rotation alone, and if rotation did could actually represent sinistral rather than dextral
occur, it may have been coupled to other motion. slip.
In any case, rotational models do not preclude ei- Jany et al. (1990) have interpreted a similar
ther right or left-lateral movement between St. Croix structure - - perhaps the same one on an inter-
and the Virgin Island platform. The direction of slip secting north-northwesterly oriented seismic line as
would be dependent on relative motion between the a 'flower structure' indicative of strike-slip faulting.
rotating Puerto Rico microplate and the Caribbean This structure overlies what is interpreted as the
plate. If Puerto Rico rotated counterclockwise rela- Late Miocene sedimentary surface. Although strike-
tive to a fixed or slowly eastward-moving Caribbean slip motion may be a reasonable interpretation for
plate, this motion would most likely produce right- the structures on these two lines, there is no way
lateral slip in the Anegada Passage and Virgin Is- to assess accurately the sense of strike-slip motion
lands basin. However, if the rotation of Puerto Rico for structures normal to such generally northerly
is accommodating part of the left-lateral movement oriented transects.
between the North American and Caribbean plates, Several studies have noted the predominance of
as proposed by Speed and Lame (1991), then the generally easterly to northeasterly oriented normal
Caribbean plate would have a faster eastward rela- faulting in the marine basins off St. Croix. Jany et
tive motion. The result would be a left-lateral sense al. (1987) identified northeasterly oriented normal
of motion in the Anegada Passage and Virgin Islands faults on a northwesterly oriented seismic cross-sec-
basin. The extensional nature of the northeasterly tion across the St. Croix basin east of St. Croix. In
trending structures in that area supports the latter addition, several other seismic traverses published by
model. Houlgatte (1983) similarly document normal fault-
ing within the Virgin Islands basin. The sense of
Right-lateral models movement on these faults appears to be dominantly
dip-slip; the degree of strike-slip motion is diffi-
An alternative model for the origin of the Virgin cult to document. Although these data were related
Islands basin proposes that motion in the Virgin to a model for right-lateral movement in the plate
Islands basin and Anegada Passage is fight-lateral boundary zone, the occurrence of northeasterly ori-
(Fig. 14A) (Houlgatte, 1983; Stephan et al., 1986). ented extensional features in an easterly oriented
Mauffret et al. (1986) and Jany et al. (1987) propose strike slip zone is more consistent with left-lateral
that motion along the Anegada Passage was origi- movement.
nally sinistral, but reversed during the Pliocene or
later. GPS data
Global Positioning data exist for stations on
Basin morphology Puerto Rico and St. Croix. Present data, although
Jany et al. (1990) suggested that the Virgin Is- still preliminary, indicates that Puerto Rico may be
lands basin shows a rhomboidal 'lazy Z' shape moving to the east-northeast slightly faster than St.
(Fig. 14B), a shape indicated by Mann et al. (1983) Croix (Dixon et al., 1998). These data support fight-
to be diagnostic of fight-lateral shear zones. If strike- lateral displacement between St. Croix and Puerto
slip motion is parallel to the Anegada Passage, the Rico. However, the measurement error is presently
oblique orientation of the Virgin Island basin is con- too large to make reliable conclusions (P. Jansma,
sistent with a fight-lateral pull-apart basin. However, pers. commun., 1998).
although the geometry of the Virgin Islands basin
and Anegada Passage is clear, the extension of the Discussion
southwestern part the basin is not. Depending on the The dextral slip model suffers from several prob-
placement of the strike-slip zone to the southwest of lems when incorporated into a regional model of
the basin, the shape of the Virgin Islands basin could tectonics. Primary among these is the difficulty in
suggest opening of the basin by left-lateral motion. reconciling a dextral strike-slip fault in the Vir-

gin Islands basin with the compression established strike-slip movement directly along the Anegada
along the length of the Muertos Trough (McCann Passage. In addition, recent GPS data may be more
et al., 1987). This would require an extension of supportive of present fight-lateral displacement than
the Anegada Passage/Virgin Islands basin fault zone with left-lateral models. In contrast, rotating platelet
westward of its present termination, and there is no models are supported by paleomagnetic data, and are
seismic or bathymetric evidence to support this. Jany consistent with either right- or left-lateral movement
et al. (1990) suggest a triple-junction south of Puerto in the Virgin Islands basin. We feel left-lateral mo-
Rico to accommodate right-lateral plate movements tion is most consistent with the long-term evolution
(Fig. 14B). However, the deformation front along the of the Virgin Islands basin area, but all the models
Muertos Trough can be seen in the GLORIA data to have serious weaknesses. The only documented type
extend nearly to the longitude of St. Croix, well east of faulting between St. Croix and the Virgin Islands
of where Jany's triple junction is shown (K. Scanlon, platform is extensional.
written commun., 1997). Subsidence analysis in the Kingshill basin indi-
Dextral motion along the Anegada fault zone cates that a period of rapid tectonic uplift took place
would also require Puerto Rico to move eastward between 10.5 and 3.5 Ma. The timing of this uplift
faster than the Caribbean plate, and a driving mech- is consistent with the relative plate motion changes
anism for this movement requires a complicated between the Puerto Rico platelet and the Caribbean
model. Similarly, it is difficult to reconcile right-lat- plate cited by McCann et al. (1987), the age of ex-
eral motion in the Anegada Passage with established tension in the Virgin Islands basin noted by Lithgow
left-lateral motion for the northern Caribbean plate et al. (1987), and tectonic changes north of Hispan-
boundary zone (Stephan et al., 1986; Burke et al., iola (Dillon et al., 1992). The lithologies of the lower
1984). The structural geology of the islands in the Kingshill Limestone suggest that uplift of the mar-
area is more consistent with left-lateral than right- gins of the Kingshill basin graben may have begun
lateral motion; easterly oriented terrestrial faults of within this time interval, and St. Croix may have
Puerto Rico and the northern Virgin Islands are rifted away from shelf areas of the Virgin Islands
mapped with left-lateral displacement and the nor- platform and Saba Bank. Further discussion of tec-
mal faults on St. Croix and on the St. Croix Ridge tonic models involving the Virgin Islands platform is
are oriented northeasterly. If dextral faulting is oc- contained in van Gestel et al. (1998).
curring in the Anegada fault zone, it must post-date
the faulting on the Puerto Rico/Virgin Islands plat-
form and on St. Croix; such a change in the latest CONCLUSIONS
Neogene or Quaternary would require a major rever-
sal of plate motion in the northeastern Caribbean. If (1) St. Croix and the Virgin Islands basin
such a reversal did occur, it apparently left no trace were produced by transtension and associated up-
in the rocks exposed on St. Croix, which record lift and subsidence along the northern Caribbean
deposition and faulting through at least the Early plate boundary zone during the Neogene. The strati-
Pliocene. graphic and structural evolution of the Neogene
Kingshill basin record a major tectonic event in this
Discussion of tectonic models plate boundary zone near the Middle Miocene-Late
Miocene transition.
An oblique sinistral model for the opening of (2) The Jealousy Formation is a strictly subsur-
the Virgin Islands basin and the Anegada Passage face unit of blue-gray marls deposited in a deep-
satisfies structural evidence on St. Croix such as the water (600-800 m) environment before the onset of
northeasterly orientation of the normal fault system. tectonism. All documented occurrences from well
It also permits the Kingshill basin to be paleo- samples place this unit in the lower part of the
geographically situated near a known extra-basinal Middle Miocene, although the formation may extend
sediment source. Left-lateral motion, if oblique, pro- lower, perhaps into the Oligocene, below current
vides a mechanism for the opening of the Virgin depths of well penetration.
Islands basin. Left-lateral transtension across the (3) The lithologic transition between the Jealousy
Virgin Islands basin may disperse the slip between Formation and the Kingshill Limestone is abrupt
the North American and Caribbean plates. Our es- and distinct, but is time-transgressive and does not
timate of 8 mm/yr of movement across the Virgin appear to indicate any major mineralogic, faunal, or
Islands basin is slower than the 20 mm/yr rates of environmental change.
motion between the North American and Caribbean (4) The Kingshill Limestone records the shal-
plates estimated by Rosencrantz and Mann (1991). lowing of the Kingshill basin and the initiation of
Dextral motion is consistent with most inter- uplift of basin-bounding horst blocks near the end
pretations of present basin morphology assuming of the Middle Miocene. There is an increase in

364 I. GILL et al.
the abundance of coarse-grained debris in the mid- REFERENCES

dle of the formation and a progressive change to
shallower-water microfaunas. Calculated uplift rates Andreieff, R, Mascle, A., Mathieu, Y. and Muller, C., 1986.
Les carbonates n6og~nes de Sainte Croix (Iles Vierges): 6tude
(backstripped, decompacted) for the Late Miocene
stratigraphique et p6trophysique. Rev. Inst. Fr. Pet., 41" 336-
to Early Pliocene are 57 m / M a . 350.
(5) St. Croix acquired its present shoreline con- Andreieff, E, Bouysse, E and Westercamp, D., 1987. G6ologie
figuration by the Pliocene, and an extensive reef and de l'arc insulaire des Petites Antilles, et 6volution g6ody-
lagoon tract had established itself along the west and namique de l'est-Caraibe. Th~se de Doctorat d'Etat es Sci-
ences, Universit6 de Bordeaux I, 465 pp.
south shorelines of the island.
Behrens, G.K., 1976. Stratigraphy, Sedimentology and Paleoecol-
(6) Structural control of the coastline in the form ogy of a Pliocene Reef Tract: St. Croix, U.S. Virgin Islands.
of a small graben allowed the accumulation and Unpubl. Masters thesis, Northern Illinois University, 93 pp.
preservation of reef and platform Pliocene sediments Bold, W.A. van den, 1970. Ostracoda of the lower and middle
along the south coast. Normal faulting has continued Miocene of St. Croix, St. Martin, and Anguilla. Caribbean J.
Sci., 10:35-61.
at least into the Pliocene.
Bolli, H.M. and Saunders, J.B., 1985. Oligocene to Holocene
(7) The Virgin Islands basin is a structure formed low latitude planktic foraminifera. In: H.M. Bolli, J.B. Saun-
by extension and sinistral faulting that rifted St. ders and K. Perch-Nielsen, (Editors), Plankton Stratigraphy.
Croix away from Puerto Rico, the Virgin Islands Cambridge University Press, Cambridge, pp. 155-262.
platform and perhaps Saba Bank. Separation rates Burke, K., Cooper, C., Dewey, J.E, Mann, E and Pindell,
J.L., 1984. Caribbean tectonics and relative plate motions.
are estimated to have been 8 m m / y r .
In: W.E. Bonini, R.B. Hargraves and R. Shagam (Editors),
(8) Extension related to left-lateral plate motion is The Caribbean-South American Plate Boundary and Regional
most consistent with the long-term evolution of the Tectonics. Geol. Soc. Am. Mem., 162:31-63.
Virgin Islands basin area and St. Croix, but none of Case, J.E., Holcombe, T.L. and Martin, R.G., 1984. Map of
the regional tectonic models discussed is consistent geologic provinces in the Caribbean region. In: W.E. Bonini,
R.B. Hargraves and R. Shagam (Editors), The Caribbean-
with all the available data. The only solid tectonic
South American Plate Boundary and Regional Tectonics. Geol.
evidence for formation of the Virgin Islands basin Soc. Am. Mem., 162: 1-30.
is extensive normal faulting. No definitive evidence Cederstrom, D.J., 1950. Geology and groundwater resources
for strike-slip motion in the Virgin Islands basin or of St. Croix, U.S. Virgin Islands. U.S. Geolog. Surv. Water
Anegada Passage exists. Supply Pap., 1067, 117 pp.
Cushman, J.A., 1946. Tertiary foraminifera from St. Croix, Vir-
gin Islands. U.S. Geol. Surv. Prof. Pap. 210-A, 17 pp.
Dill, R.E, 1977. Deep water erosional feature, bedrock of St.
ACKNOWLEDGEMENTS Croix, U.S. Virgin Islands, as seen from the research submeri-
ble Alvin. 7th Caribbean Geol. Conf., Abstr., Addendum.
The authors thank Lee Gerhard, Nancy Grind- Dillon, W.E, Austin, J.A., Jr., Scanlon, K.M., Edgar, N.T. and
Parson, L.M., 1992. Accretionary margin of north-western
lay, Barbara Lidz, Kathryn Scanlon, and Robert
Hispaniola: morphology, structure, and development of the
Speed for helpful, critical reviews of this paper, and northern Caribbean plate boundary. Mar. Pet. Geol., 9: 70-88.
Paul Mann for his constructive and patient editing. Dixon, T.H., Farina, E, Demets, C., Jansma, E, Mann, E and
Pamela Jansma and James Joyce generously dis- Calais, E., 1998. Relative motion between the Caribbean and
cussed various aspects of Caribbean tectonics and North American plates and related boundary zone deformation
from a decade of GPS observations. J. Geophys. Res. B, 103:
provided unpublished data. Clyde Moore served as
15,157-15,182.
major professor to the senior author throughout field Donnelly, T.W., 1966. Geology of St. Thomas and St. John, U.S.
and laboratory work. The aid and cooperation of the Virgin Islands. In: H.H. Hess (Editor), Caribbean Geological
West Indies Laboratory, the U.S. Geological Survey Investigations. Geol. Soc. Am. Mem., 98: 85-176.
Water Resources Division in San Juan, ER., and K. EEZ Scan Scientific Staff, 1987. Atlas of the U.S. Exclusive
Econonic Zone, Gulf of Mexico and Eastern Caribbean Areas.
Eastman and the staff of Caribbean Geological Ser-
U.S. Geol. Surv. Misc. Invest. Ser., I- 1864-A, B, 171 pp.
vices in undertaking this study is gratefully acknowl- Frankel, A., McCann, W.R. and Murphy, A.J., 1980. Observa-
edged. The assistance of Marc Lowman with illus- tions from a seismic network in the Virgin Islands region:
trations is also appreciated. Funding for St. Croix tectonic structures and earthquake swarms. J. Geophys. Res.,
drilling and field work was provided to Gill by: 85 (B5): 2669-2678.
Frost, S.H., 1977. Cenozoic reef systems of the Caribbean m
the Virgin Islands Water Resource Center; SOHIO,
prospects for paleoecologic synthesis. In: S.H. Frost, M.E
Chevron, and Shell field research grants; Geological Weiss and J.B. Saunders (Editors), Reefs and Related Carbon-
Society of America and American Association of ates - - Ecology and Sedimentology. Am. Assoc. Pet. Geol.,
Petroleum Geologists student grants; D. Eby and Stud. Geol., 4: 93-110.
Champlin Petroleum; and the Applied Carbonate Geist, E.L., Childs, J.R. and Scholl, D.W., 1988. The origin of
summit basins of the Aleutian Ridge: implications for block
Research Program, the Department of Geology, and
rotation of an arc massif. Tectonics, 7:327-341.
the Basin Research Institute at Louisiana State Uni- Gerhard, L.C., Frost, S.H. and Curth, EJ., 1978. Stratigraphy and
versity. This work is part of a dissertation project of depositional setting, Kingshill Limestone, Miocene, St. Croix,
the senior author at Louisiana State University. U.S. Virgin Islands. Am. Assoc. Pet. Geol. Bull., 62:403-418.

E V O L U T I O N OF THE N E O G E N E K I N G S H I L L BASIN OF ST. CROIX, U.S. V I R G I N ISLANDS 365
Gill, I.E, 1983. The Sedimentological Controls on Organic Car- stratigraphic sequence, northeast Caribbean (St. Croix, U.S.
bon in the Virgin Islands Trough, U.S. Virgin Islands. Unpubl. Virgin Islands). Geol. Soc. Am. Bull. 100: 282-298.
MS thesis, Univ. of Rochester, 70 pp. Lithgow, C., McCann, W.R. and Joyce, J., 1987. Extensional tec-
Gill, I.E, 1989. The Evolution of Tertiary St. Croix. Unpubl. tonics at the eastern edge of the Puerto Rico Platelet (abstr.).
Ph.D. dissertation, Louisiana State University, 287 pp. Eos, 68 (44): 1483.
Gill, I.E and Hubbard, D.K., 1986. Subsurface geology of the Mann, E, Hempton, M.R., Bradley, D.C. and Burke, K., 1983.
St. Croix carbonate rock system. Caribbean Research Institute Development of pull-apart basins. J. Geol. 91: 529-554.
Technical Report 26, College of the Virgin Islands, 86 pp. Masson, D.G. and Scanlon, K.M., 1991. The neotectonic setting
Gill, I.E and Hubbard, D.K., 1987. Subsurface geology of the St. of Puerto Rico. Geol. Soc. Am. Bull., 103:144-154.
Croix carbonate rock system, phase II. Technical Report No. Mauffret, A., Jany, I., Mercier de Lepinay, B., Bouysse, E,
28, Water Resources Research Center, College of the Virgin Mascle, A., Renard, V. and Stephan, J.-F., 1986. Releve au
Islands, St. Thomas, U.S. Virgin Islands, 79 pp. sondeur multifaisceaux du bassin des Iles Vierges (extremit6
Gill, I.E, Hubbard, D., McLaughlin, EE and Moore, C., 1990. orientale des Grandes Antilles): role de l'extension et des
Sedimentological and tectonic evolution of Tertiary St. Croix. ddcrochements. C.R. Acad. Sci. Paris, 303, Ser. II, 10: 923-
In: D.K. Hubbard (Editor), Terrestrial and Marine Geology of 928.
St. Croix, U.S. Virgin Islands. Special Publication No. 8, West McCann, W.R., Joyce, J. and Lithgow, C., 1987. The Puerto Rico
Indies Laboratory, Teague Bay, St. Croix, U.S. Virgin Islands, Platelet at the northeastern edge of the Caribbean Plate. Eos,
pp. 49-72. 68 (44): 1483.
Gill, I.E, Hubbard, D., McLaughlin, EE and Moore, C.H., in McLaughlin, EE, Gill, I.E and Bold, W.A. van den, 1995.
press. Geology and hydrogeology of the St. Croix carbon- Biostratigraphy and paleoenvironments of the Neogene of St.
ate aquifer system. In: R. Renken (Editor), Aquifers of the Croix, U.S. Virgin Islands: implications for stratigraphic evo-
U.S. Virgin Islands and Puerto. U.S. Geol. Surv., Prof. Pap., lution. Micropaleontology, 41 (4): 293-320.
PP1419-A (30 ms. pages, 10 figs). Meyerhoff, H.A., 1927. The physiography of the Virgin Islands,
Haq, B.U., Hardenbol, J. and Vail, ER., 1988. Mesozoic and Culebra and Vieques. New York Academy of Science, Scien-
chronostratigraphy and cycles of sea level change. In: C.K. tific Survey of Porto Rico and the Virgin Islands, Vol. 4, no. 2,
Wilgus, B.S. Hastings, C.G.St.C. Kendall, H.W. Posamen- pp. 152-156.
tier, C.A. Ross and J.C. Van Wagoner (Editors), Sea-Level Moore, C.H., Jr., Graham, E.A. and Land, L.S., 1976. Sediment
Changes: An Integrated Approach. Soc. Econ. Paleontol. Min- transport and dispersal across the deep fore-reef and island
eral., Spec. Publ., 42: 71-108. slope ( - 5 5 to -305 m), Discovery Bay, Jamaica. J. Sediment.
Hess, H.H., 1933. Interpretation of geological and geophysical Petrol., 46:174-187.
observations. U.S. Hydrological Office Navy-Princeton grav- Multer, H.G., Frost, S.H. and Gerhard, L.C., 1977. Miocene
ity expedition to the West Indies in 1932, pp. 27-54. 'Kingshill Seaway' - - a dynamic carbonate basin and shelf
Hess, H.H., 1966. Caribbean research project, 1965, and bathy- model, St. Croix, U.S. Virgin Islands. In: S.H. Frost, M.E
metric chart. In: H.H. Hess (Editor), Caribbean Geological Weiss and J.B. Saunders (Editors), Reefs and Related Car-
Investigations. Geol. Soc. Am. Mem., 98: 1-10. bonates--Ecology and Sedimentology. Am. Assoc. Pet. Geol.,
Holcombe, T.L., 1977. Geomorphology and subsurface geology Stud. Geol., 4: 329-352.
west of St. Croix, U.S. Virgin Islands. Am. Assoc. Pet. Geol. Nemec, M.C., 1980. A two-phase model for the tectonic evolu-
Mere., 29: 353-362. tion of the Caribbean. 9th Caribbean Geol. Congr. Trans., 2:
Houlgatte, E., 1983. Etude d'une partie de la fronti6re nord-est 23-34.
de la plaque Caraibe. Unpubl. Masters thesis, L'Universite de Reid, H. and Tabor, S., 1920. The Virgin Islands earthquakes of
Bretagne Occidentale, 69 pp. 1867-1868. Bull. Seismol. Soc. Am., 10: 9-30.
Hubbard, D.K., Suchanek, T.H., Gill, I.E, Cowper, S., Ogden, Reid, J.A., Plumley, EW. and Schellekens, J.H., 1991. Paleomag-
J.C., Westerfield, J.R. and Bayes, J., 1981. Preliminary studies netic evidence for late Miocene counterclockwise rotation of
of the fate of shallow-water detritus in the basin north of St. north coast carbonate sequence, Puerto Rico. Geophys. Res.
Croix, USVI. Proc. 4th Int. Coral Reef Symp., Manila, 1, pp. Lett., 18 (3): 565-568.
383-387. Rosencrantz, E. and Mann, E, 1991. SeaMARC II mapping
Jany, I., Mauffret, A., Bouysse, E, Mascle, A., Mercier de of transform faults in the Cayman Trough, Caribbean Sea.
Lepiany, B., Renard, V. and Stephan, E, 1987. Releve Geology, 19 (7): 690-693.
bathymetrique Seabeam et tectonique en decrochements au Scanlon, K.M. and Masson, D., 1988. Seafloor deformation at
sud des Iles Vierges (Nord-Est Caraibes). C.R. Acad. Sc. the northern Caribbean Plate boundary and rotation of a Puerto
Paris, 304, Ser. II, 10: 527-532. Rico microplate (abstr.). Eos, 69 (16): 462.
Jany, I., Scanlon, K.M. and Mauffret, A., 1990. Geological Shurbet, G.L., Worzel, J.L. and Ewing, M., 1956. Gravity mea-
interpretation of combined Seabeam, GLORIA and seismic surements in the Virgin Islands. Geol. Soc. Am. Bull., 67:
data from Anegada Passage (Virgin Islands, North Caribbean). 1529-1536.
Mar. Geophys. Res., 12: 173-196. Speed, R.C. and Larue, D.K., 1991. Extension and transtension
Khudoley, K.M. and Meyerhoff, A.A., 1971. Paleogeography and in the plate boundary zone of the NE Caribbean. Geophys.
geological history of Greater Antilles. Geol. Soc. Am. Mem., Res. Lett., 18: 573-576.
129, 199 pp. Stephan, J.F., Blanchet, R. and Mercier de Lepinay, B.,
Lidz, B.H., 1982. Biostratigraphy and paleoenvironment of 1986. Northern and southern Caribbean festoons (Panama,
Miocene-Pliocene hemipelagic limestone, Kingshill Seaway, Colombia-Venezuela and Hispaniola-Puerto Rico), inter-
St. Croix, U.S. Virgin Islands. J. Foram. Res., 12: 205-233. preted as pseudo subductions induced by the east-west short-
Lidz, B.H., 1984a. Neogene sea-level change and emergence, ening of the peri-Caribbean continental frame. In: E-C. Wezel
St. Croix, Virgin Islands: evidence from basinal carbonate (Editor), The Origin of Arcs. Elsevier, New York, pp. 401-
accumulations. Geol. Soc. Am. Bull., 95:1268-1279. 422.
Lidz, B.H., 1984b. Oldest (early Tertiary) subsurface carbonate van Gestel, J.E, Mann, E, Dolan, J. and Grindlay, N.R., 1998.
rocks of St. Croix, USVI, revealed in a turbidite mudball. J. Structure and tectonics of the upper Cenozoic Puerto Rico-
Foram. Res., 14:213-227. Virgin Islands carbonate platform as determined from seismic
Lidz, B.H., 1988. Upper Cretaceous (Campanian) and Cenozoic reflection studies. J. Geophys. Res., 103: 30,505-30,530.

366 I. G I L L et al.
Warner, A.J., 1989. The Cretaceous age sediments of the Saba In: H.G. Multer and L.C. Gerhard (Editors), Guidebook to
Bank and their petroleum potential. Trans. 12th Caribbean the Geology and Ecology of Some Marine and Terrestrial
Geol. Conf., pp. 341-354. Environments, St. Croix, U.S. Virgin Islands. West Indies Lab.
Whetten, J.T., 1966. The geology of St. Croix, U.S. Virgin Spec. Publ., 5: 129-143.
Islands. Geol. Soc. Am. Mem., 98: 177-239. Wilcox, R.E., Harding, T.E and Seely, D.R., 1973. Basic wrench
Whetten, J.T., 1974. Field guide to the geology of St. Croix. tectonics. Am. Assoc. Pet. Geol. Bull., 57 (1): 96.

C h a p t e r 14
Review of the Tectonic Controls and Sedimentary Patterns in

Late Neogene Piggyback Basins on the Barbados Ridge Complex
PASCALE HUYGHE, JEAN-LOUIS MUGNIER, ROGER GRIBOULARD, YANN DENIAUD,

E L I A N E G O N T H I E R and J E A N - C L A U D E F A U G E R E S
A review of the tectonic control and sedimentary patterns of the late Neogene piggyback basins of the Barbados Ridge
Complex is proposed, mainly based on the analysis of seismic reflection profiles and sidescan sonar images. The geometrical
relationships between the thrust faults and the syntectonic deposits show that the frontal piggyback basins are controlled by active
thrusting. Deformations, with distinct scales, control their evolution: (1) a rapid tilting, strictly localized at the back limb of
anticlines, and attributed to migration of active axial surfaces during fault-bend fold propagation; (2) a complex activation of the
major thrust system, at a scale of a few kilometers: blind thrusts corresponding to frontal propagation develop seaward whereas
motion along thrusts occurs backward. At about 50 km back of the deformation front, abundant muddy material raises up through
pre-existing faults and disturbs arcward piggyback basins. The superimposition of the diapiric structures upon deformations linked
to tectonic accretion (development of backthrusts and reactivating of forward verging thrusts) and inherited oceanic basement
ridges leads to the individualized development of sub-basins bounded by steep topographic features.
Most of the sediments of the South Barbados piggyback basins originate from the South American continent and are massively
transported to the abyssal plain through canyons. Their course is driven by the main regional structures and their morphology
reflects the tectonic activity of the features where they run. Oceanic ridges, by damming and collecting turbidite material control
the thickness of sediments added to the complex and then the depth of the decollement and size and filling of the piggyback
basins. Tectonics, by generating routes along the faults and excess fluid pressures control the main location and importance of
mud diapirs and authigenic deposits. These stiff carbonate crusts preferentially develop on diapiric domes or on the back limb
of anticlines up to the edge of supra-prism basins. Clayey diapiric material may be found within the basins where they form
important mud flows as well as sliding masses. Tectonic and diapiric structures control gravity reworking processes, whereas
structural relieves locally disturb bottom currents and hence control some erosion processes.
The development of the piggyback basins of the BRC is closely linked to the evolution of a thrust wedge. The formation of
frontal basins is mainly controlled by a forward-verging thrust system that forms a brittle wedge, whereas the development of the
arcward basins is mainly controlled by subcretion of deep muds that induces mud diapirism, ductile deformation in the lowermost
part of the wedge, a regional gentle slope, and the occurrence of both backward and forward verging thrusts. The evolution of the
piggyback basins of the BRC also reflects the north to south changes in width and thickness of the wedge which are mainly related
to variations of the increase of the sediment supply. Piggyback basins then evolve from minor depressions filled with very poor
sediments in the north to about 10 km long overfilled basins in the south of the complex where abundant sedimentation occurs.
INTRODUCTION and p r o p a g a t i o n of n e w thrusts ( R o u r e et al., 1990;

Z o e t e m e i j e r and Sassi, 1992).
P i g g y b a c k basins d e v e l o p in the external part of T h e s o u t h e r n L e s s e r A n t i l l e s (Fig. 2) is a g o o d
o r o g e n i c belts, at the e d g e of the f o r e d e e p basin. Ori place to study the e v o l u t i o n of p i g g y b a c k basins as
and F r i e n d (1984) defined t h e m as basins that f o r m h i g h s e d i m e n t influx and active faulting o c c u r in
and fill w h i l e b e i n g carried on the b a c k of m o v i n g the B a r b a d o s R i d g e C o m p l e x (BRC). H i g h s e d i m e n t
thrust sheets (Fig. 1). P i g g y b a c k basins are charac- influx p r o v i d e s an a c c u r a t e stratigraphic r e c o r d of
terized by a s y n t e c t o n i c filling and their g e o m e t r y the history of basins d u r i n g late N e o g e n e t i m e and
evolves with g r o w t h of the a c c r e t i o n a r y w e d g e . Mi- g e o p h y s i c a l t e c h n i q u e s p e r m i t the o b s e r v a t i o n of
gration of the d e p o c e n t e r s of the s y n t e c t o n i c filling the p i g g y b a c k basin f o r m a t i o n . This study m a i n l y
furnishes a r e c o r d of the tilting of the thrust sheets c o n c e r n s the basins that develop, f r o m the toe of

370 R HUYGHE et al.
PIGGYBACK BASIN POSSIBLE BY PASSING

MAIN BASIN
I - " .....
ONLAP THRUST
SEQUENCECONTEMPORANEOUS
SHEETS ~ < ~ ~ ~ - - - WITHTHRUSTING
NOT TO SCALE
Fig. 1. A sketch of piggyback basin (modified from Ori and Friend, 1984).
the accretionary complex to the forearc basin (Fig. 3 data also provide constraints upon the growth of the
adapted from Westbrook et al., 1988) and which accretionary wedge.
were recognized as piggyback basins (Mascle et al.,
1990; Huyghe et al., 1996).
The aim of this paper is to present the sedimen- GEOLOGIC SETTING
tary pattern of the piggyback basins in the Southern
Barbados Ridge Complex (SBRC). We present here The BRC (Fig. 2) develops as a result of
single-channel seismic (SCS), 3.5 kHz and sides- the subduction of the North American (HA) and
can sonar (SAR) data which were acquired in 1987 South American (SA) plates beneath the Caribbean
and 1990 during French surveys with the NO Nadir plate (CA) since the Middle Eocene (Westbrook,
and Suroit (Faugbres et al., 1991; Griboulard et al., 1982; Stephan et al., 1990). The movement of the
1991; Faug~res et al., 1993; Huyghe et al., 1996). Caribbean plate relative to the South American plate
As the evolution of piggyback basins is representa- is complex: obliquity of convergence (Jordan, 1975;
tive of the history of thrust-sheet movements, these Stein et al., 1982; Speed, 1985; DeMets et al., 1990;
Fig. 2. Geodynamic setting of the eastern Caribbean. Solid arrows represent the Orinoco delta and major continental influxes. Open
arrows show relative motion of the Caribbean plate with respect to the South American plate (from DeMets et al., 1990). Hatched area:
collision and/or strike-slip motion between the Caribbean plate and South American margin (from Biju-Duval et al., 1982); dashed line:
limit between continental crust and crust of uncertain nature (from Bouysse and Westercamp, 1988).

LATE NEOGENE PIGGYBACK BASINS ON THE BARBADOS RIDGE COMPLEX 371
Russo et al., 1993; DeMets et al., 1994), transten-

sion (Sykes et al., 1982) and intraplate deformations
(Weber et al., 1995) are still under discussion. In
the southern part of the prism, these movements
may lead to the development of additional structures
such as transcurrent faults (Robertson and Burke,
1989, 1991; Griboulard et al., 1991). Moreover, the
boundary between continental and oceanic crusts is
not firmly established in the southern part of the
complex (Bouysse and Westercamp, 1988).
Large volumes of sediments coming mainly from
the Orinoco and Amazon systems (Damuth, 1975;
Leonard, 1983) are involved in the frontal accretion
at the toe of the Barbados wedge. As a result of the
southern location of these sources, the B RC narrows
from 450 km at its southern extremity to 100 km
north of 18~ This variation in width is associated
with a change of sediment thickness on the Atlantic
c~
ocean floor from 7000 m south of l l~ to 200
m in the north at 19~ (Ewing et al., 1973; Pe-
,.o
ter and Westbrook, 1976; Mascle and Moore, 1988;
(D
Moore and Mascle, 1990). This decrease is also lo-
9
cally controlled by the occurrence of N 140~
ridges which act as barrage to the sediment transport
proceeding from south to north (Westbrook et al.,
E
9
1984; Wright, 1984). These structures have been
z evidenced by gravity, bathymetry and seismic reflec-
tion data (Birsh, 1970; Peter and Westbrook, 1976)
,.o
and probably represent remnant transform faults in
the oceanic basement (Fig. 3) that generate troughs
(D
and asymmetrical flanking ridges (Westbrook et al.,
E
9
1984). Such features induce lateral ramps and lead
to the global eastward shift of the deformation front
9
....,
,.o
(Calassou et al., 1993; Huyghe et al., 1996). The
(D
best known of these are the Barracuda Ridge and the
(D
Tiburon Rise (Fig. 2) which still affect the abyssal
~ plain.
9
Mud diapirism is a widespread phenomenon on
the SBRC (Fig. 4) (Fontas et al., 1985; Brown and
Westbrook, 1988; Langseth et al., 1988; Le Pichon
(D
et al., 1990). It is linked to high pore fluid pressures
9
favored by the rapid tectonic burial of undercom-
pacted sediment at the deformation front (Bangs and
~
9 Westbrook, 1990), the high horizontal compression
(D occurring in a prism and by subcretion that releases
~ fluids into the accretionary complex (Brown and
Westbrook, 1988). In the SBRC, most of the diapiric
features seem to be controlled by tectonic structures
such as faults that provide routes to the surface for
~ mud (Biju-Duval et al., 1982; Brown and Westbrook,
1988). Depending on the viscosity of mud, diapiric
forms vary from mud volcanoes fed by low-viscosity
9
mud and generating flows, to higher-viscosity mud
~
ridges or domes forming prominent relieves (Shih,
1967; Brown and Westbrook, 1988). This mud is
~D
~ probably provided by Miocene beds as onshore in
.1
e4 Trinidad (Higgins and Saunders, 1974). Propagation
~

372 E HUYGHE et al.
Fig. 4. Regional map (adapted from Brown and Westbrook, 1987 and Fontas et al., 1985). Numbers refer to piggyback basins shown in
Figs. 8, 10 and 16. Color version at http://www.elsevier.nl/locate/caribas/
of excess pore pressure from the decollement below overflows the channels and deposits mainly in sedi-
the accretionary complex seaward into the unde- mentary basins contemporaneous with the growth of
formed ocean basin sediments also generates mud the SBRC (Biju-Duval et al., 1982; Mascle et al.,
diapir fields and venting seaward of the deforma- 1990). As a consequence the late Neogene filling of
tion front (Westbrook and Smith, 1983; Silver et basins consists of turbiditic layers intercalated with
al., 1986; Langseth et al., 1988; Henry et al., 1990; fine hemipelagic muds (Faug~res et al., 1993). The
Lallemant et al., 1990). stratigraphic pattern of these syntectonic deposits
In the SBRC, the thick sedimentary series lying and their geometrical relationships with faults per-
above the decollement are intersected by a network mit to distinguish two main types of basins that are
of submarine canyons feeding the Orinoco deep-sea- described below.
fan (Belderson et al., 1984; Faug~res et al., 1993).
The canyons generally run northward (Embley and
Langseth, 1977; Mascle et al., 1990) but their course FRONTAL BASINS
is also driven by the main regional tectonic features
(Fig. 4) and changes from NE to ENE and ESE. Frontal basins develop in the zone of initial
At a smaller scale, the course of the canyons is accretion (Fig. 3) defined by Westbrook et al. (1984).
strongly influenced by structures having prominent They form between major anticlinal ridges from
relief on the seafloor (Mascle et al., 1990), such the wedge toe to 50 km west of the deformation
as anticlines and mud diapirs so that the course of front. The folds delimiting the frontal basins develop
the canyons appears sinuous (Fig. 5). The coarsest above major thrust structures (Biju-Duval et al.,
turbiditic material (medium sands and gravels) com- 1982) and may be considered as thrust fault related
ing from the Guyana Margin is funnelled down to folds above a major decollement (Fig. 6) (Mascle
the Atlantic Abyssal Plain through major canyons et al., 1990). On the SBRC, spacing of the folds
(Fig. 4). The finest material (fine sands and clay) evolve from 5-6 km at 12~ (Fig. 7) to 8-10 km

LATE N E O G E N E P I G G Y B A C K BASINS ON THE BARBADOS RIDGE C O M P L E X 373
~-~ BASIN~ANYON ~-~BASlN ~_1 ~r BACK LiMB OF THE ANTICLINE--)~
"~'~ --- ~THE CANYON |~-".,.~'-.~"~'~

.~ '
~ " " I:',.-:,~'.:-:~.~
""
,~"1,O~ ~~~ d ~ ~ ' ~~ - ~ "
',,~'-~r,,,,'
~ ~2~'\ ':': " , " ;,~:~_'~~
II~~.~,: --- o,~ ~STRUCTURAL
)~" ,~'fl' ~ ~I~--.~-~SlEDIMENTARY LEVEES ~ \ .~':-!:-;:]/~!.-"t~ !~ LINEATIONS
{.., c... """ ~"
:.-..~:, ',
'
. 4'z
S_s___
,...., "~ ~ i -, .g,:.::.~"~ ,,..:,.'.~,, -
. . . . - - ";"-" 9 u ~' r :,'"" :/'.'~."
I 9 ",v" ,4",: ,':.'~ ~ 2 . ~ " .._~.

. 6oo M
' " .,u~,,;"~/ ""Y
.r~".'~-'~~ .-~tt.-~',:
~ - 9 ~.* " ~* ' ~ ~ S Z l F F CRUSTS
Fig. 5. Line drawing of SAR images representing the back limb of a fault related fold and the edge of the associated piggyback basin.
Near the anticline, the canyon changes direction. Its meanderings are disturbed by tectonic structures which also affect the back limb of
the fold covered with stiff crusts (dotted areas).
KM W E
-4
FOLDSPACING
I-.1KM--.I 15~ N
5
0 I I I I I
0 2 4 6 8 10 KM
KM NW SE
.2 I
-4
-6
-8
0 5 :10 15 20 2'5 ;30 3,5 z~0 KM
KMI NW FOLDSPACING SE
- 2 -t I 8 KM I 10~ N
"4- J
-6
-8
0 5 10 15 20 25 30 35 40 KM
Fig. 6. Balanced cross-sections of the Barbados Ridge Complex (from Endignoux and Mugnier, 1990): (a) at the latitude 15~ (CRV
128, Leg 110 transect, rate of shortening 25%); (b) at latitude 10~ (CRV 108 A, rate of shortening 15%); and (c) at latitude 10~
(CRV105, rate of shortening 5%).

374 E HUYGHE et al.
Fig. 7. (a) Diapicar UV water-gun seismic profile (non migrated) at 12~ (b) Detail of (a) showing that thrust fault 1 cuts through
sediments onlapping the frontal sheet.
at 10~ (Fig. 8) and controls the width of basins tectonic sheet. Relationships between this thrust
(Endignoux and Mugnier, 1990; Mascle et al., 1990) fault and syntectonic filling is difficult to observe
respectively varying from a few thousand meters to because of 'artefacts' or slumps linked to the slope.
6-8 km. The frontal piggyback basins show a mainly Fig. 7 shows the relationships between a major
N10~176 direction parallel to the frontal thrust fault (thrust 1) and the sediments deposited
thrusts. Their filling reaches up to 1 s TWT at about at its footwall. Recent sediments onlap the frontal
10~ (Fig. 8) and shows a wedge-shaped geometry thrust sheet and postdate its tilting. As thickness of
characteristic of syntectonic deposits of piggyback these sediments is continuous and regular over 1 kin,
basins undergoing active thrusting. their origin has not to be linked to gravity sliding
from the above westside fold related fault. Thrust
Stratigraphic pattern at the edge of the basins 1 cuts through the sediments onlapping the frontal
sheet. Its recent propagation is consequently subse-
Inner edge quent to that of the frontal thrust and shows that
The inner edge of the piggyback basins is formed reactivation occurs backward from the deformation
by the thrust fault bounding the nearby backward front.

Fig. 8. Diapicar KL water-gun seismic profile (non migrated) at 10~ Horizontal scales are the same on the seismic profile and line
drawing. Note the presence and gentle size of a piggyback basin in the zone of initial accretion, whereas a smaller piggyback basin is
observed at 12~ (Fig. 7). Dots represent depocenters.
Outer edge minor thrusts localized at the edge of the basin cut
The outer edge of the piggyback basin is formed through recent sediments.
by folds that have frequently kink-shaped geometries The pattern described above is characteristic of
with rounded hinges. Interlimb angles of about 160~ the growth of fault related folds (Suppe et al., 1992)
have locally been evaluated in the stratigraphically and the syntectonic strata record the evolution of
higher strata (Fig. 9). Numerous minor tectonic the fold relative to the history of sedimentation
features affect the limb of the anticlines but there is (Burbank and Verg6s, 1994). Growth folding im-
no major cliff of the sea floor delimiting the edge of plies axial surfaces migration (Suppe, 1983) and
the basins as shown by S AR images. The basin edge involves progressive rotation of the back limb of the
is only marked by a change of sediment reflectivity anticlines. From the angular relationships (Fig. 9)
(Fig. 5) linked to stiff crusts covering the back limb and the average of Quaternary sedimentation rate, a
of the fold and soft sediments within the basin. mean tilting rate of 1~176 yrs has been esti-
3.5 kHz seismic profiles have a resolution better mated (Huyghe et al., 1996). Such tilting gives birth
than 5 m and evidence the characteristics of recent to angular unconformities which develop in growth
strata. The stratigraphic pattern at the edge of the strata localized on the back limb of the fold and
basin shows the following (Fig. 9): (1) a succession reflects the distinct phases of fold growth. Sedimen-
of five sedimentary bodies with a wedge-shaped tary bodies 1, 2, 3 and 5 onlap progressively onto
geometry deposited on the 15~ back limb of the back limb whereas stratigraphic offlap occurs for
the anticline (circled numbers in Fig. 9b); (2) the unit 4, a stratigraphic pattern that reveals that sed-
individual body itself thickens towards the center of iment accumulation exceeds uplift of the anticline
the piggyback basin; (3) an increase in dip from as long as onlap occurs but that uplift suddenly and
stratigraphically higher to lower strata is seen in the shortly exceeds accumulation during deposition of
syntectonic sedimentary bodies depositionally above unit 4. This abrupt stage of fold growing may be re-
the back limb; (4) angular disconformities occur sponsible for minor displacement along thrusts in the
between sedimentary bodies in the thinned domain syncline and slumping of unconsolidated material on
at the back limb of the anticlinal ridge; (5) surfacial the back limb of the rising anticline (Fig. 9).

376 E HUYGHE et al.
Fig. 9. (a) Diapisar 9036 seismic profile (3.5 kHz) showing a thrust fault related fold and the recent filling of the associated piggyback
basin. Vertical exaggeration 5 x. Location in Fig. 5. (b) Line drawing and detail of (a). Note the presence of a thrust in the syncline. Grey
dotted areas indicate slumps. Disconformities between the different sedimentary units 1 to 5 are related to the successive tilting of the
ridge. Dips drawn under seismic interpretation take into account 5 x vertical exaggeration of seismic line. (c) Kink-shaped construction of
the thrust fault related fold from the 3.5-kHz seismic data. Horizontal scale is the same as in (a). Interlimb angle is about 160~ Location
of the western axial surface has not been determined. Occurrence of thrusts in the syncline in the piggyback basin is probably evidence
for positioning it more westward. No vertical exaggeration.
Global stratigraphic pattern filling of the frontal basins 1, 2 and 3 (see Fig. 4 for
location) f r o m east to w e s t and s h o w n in Figs. 8 and
T h e global stratigraphic pattern of the syntectonic 10. In every basin, the global b a c k w a r d m i g r a t i o n
deposits of the p i g g y b a c k basins is a n a l y s e d f r o m the of d e p o c e n t e r s p r o v i d e s e v i d e n c e for a w e s t w a r d

Fig. 10. Diapicar OP water-gun seismic profile (non migrated) and its interpretation showing the backward migration of depocenters in
piggyback basins 2 and 3. Dots represent depocenters. Horizontal scales are the same on the seismic profile and line drawing.
tilting of the 7-10-km-long underlying thrust sheets. the movement of nearby thrust faults inducing fold
The geometrical relationships between syntectonic growing and the tilting of individual thrust sheets.
deposits and thrusts display a complex activation The successive positions of the depocenters and
of the major thrust system: (a) development of the onlap of the distinct sedimentary bodies indicate
piggyback basin 2 (seismic units 2.1 and 2.2) linked a complex activation of the major thrust system
to the growth of the eastern fault-bend fold (T2); relative to sedimentation history. Stages of higher
(b) development of another sub-basin as a result of displacement rate take turn with increase of sedi-
the propagation of a second ramp anticline (T2') mentation rate and indicate out-off-sequence reacti-
west of the latter. In the main basin, depocenter vation whereas blind thrusts corresponding to frontal
2.3 is temporally shifted eastward as a result of propagation develop seaward under the abyssal plain
the formation of the fault related fold (T2'); (c) (Mascle et al., 1990). Close-to-the-surface-motion
growing of the western anticline (T2') ceased. The results either in a whole ramp reactivation from
depocenter of unit 2.4 is located east of the blind decollement to the surface or in a passive re-adjust-
ramp anticline (T2'), as a result of its asymmetry. ment linked to the imbrication of new frontal sheets
A large-scale backward tilting of the whole basin (Huyghe et al., 1996).
leads to the westward thickening of units 2.3 and Finally, the stratigraphic pattern of frontal piggy-
2.4. back basins reveals a strong tectonic control: (1) a
The successive onlap positions of the sedimen- rapid tilting strictly localized at the back limb of
tary bodies reflect changes in the evolution of the anticlinal ridges; (2) a global backward migration of
folds relative to sedimentation rate. This is clearly depocenters linked to the tilting of the whole sheet
recorded in the thick filling of piggyback basin that they are carried on; (3) a complex activation
3: seismic units 3.0 and 3.3 are associated with of the major thrust system relative to the history of
stages of an uplift rate higher than sedimentation sedimentation.
rate.
Tectonic control of frontal piggyback basins ARCWARD BASINS
Both the detailed and global patterns show that Piggyback basins develop in a more arcward
basin depocenters are dynamically controlled by position, at about 50 km in back of the deformation

378 E HUYGHE et al.
front, where the bathymetry rises gently to the west, intrusions. Globally, the sedimentary filling is poorly
from water depths of 2000 m to 1400 m (Fig. 4). deformed except at the edges of the basins where
Numerous mud intrusions occur and the number of sediments are locally overturned because of mud
diapirs increases both southward and eastward in intrusions (Fig. 14). Mud flows with an extend of
relation respectively with the extent and thickness about 3 km are interbedded within tabular sedi-
of the Orinoco submarine fan on the abyssal plain ments (Fig. 13). High pore fluid pressure and steep
and the onset of subcretion (Brown and Westbrook, slopes associated with diapiric structures promote
1988). mass movements from the flanks of the diapiric
edifices to the basins. Downslope accumulation is
Stratigraphic pattern of arcward piggyback very chaotic (Fig. 15) showing the heterogeneous
basins nature of these sliding masses. The mean thickness
of the displaced sediments is about 20 m which
The sedimentary filling of these basins is thicker yields a total volume of more than 120 x 106 m 3.
than that of frontal basins (1.5 to 2 s TWT). At about These sliding masses are strictly localized at the
12~ late Neogene sedimentary bodies show west- edge of the basins as shown by Fig. 15 where the
ward or eastward wedge-shaped geometries above sliding chaotic sheet changes to undeformed sedi-
steep faults (Fig. 11) recognized as thrusts and ments when the slope smoothes and shows again the
back-thrusts (Biju-Duval et al., 1982; Brown and N40~ direction characteristic of that of the
Westbrook, 1987). The youngest sediments observed frontal piggyback basins. Diapiric edifices undergo
in Fig. 11 and also with 3.5-kHz seismic profiles short-lived catastrophic events and recur every few
locally overlap and seal the faults indicating that tens of years without any steady cyclicity and the
displacement occurs episodically. volume of material averages 3 x 106 m 3 (Higgins and
Southward, piggyback basins of about 150 km 2 Saunders, 1974). These observations could explain
show a complex arcuate shape (Fig. 4) and mud the occurrence of superimposed slides at the toes of
diapirism becomes more intensive. As fault planes diapirs rather than at the foot of tectonic structures
form suitable routes to drain clay to the surface, whose reactivation in the case of a thrust fault re-
arcward piggyback basins are bounded with diapiric lated fold is far less frequent and less catastrophic
structures (Figs. 12 and 13). Sedimentary bodies (Fig. 9).
sometimes show wedge-shaped geometries thinning
towards the borders (Fig. 12), but it is here difficult Evolution of the arcward piggyback basins
to assess if it is linked to the displacement along
a thrust or back-thrust fault or to the progressive Mud rises up through NS- or N40~ faults
up-rising of mud. The last 0.5 s TWT of the filling (Figs. 14 and 15) parallel to the direction of the
is rather tabular and its local and moderate deforma- deformation front. These faults show a recent nor-
tion seems to be related to the growing of the mud mal downthrow probably due to late mud up-rising
features. Recent mud intrusions through pre-existing (Huyghe et al., 1996).
or new forming faults lead to the individualisation The intrusion of mud into N40~ faults
of sub-basins between the diapiric edifices (Fig. 13). bounding the arcward basins raises the question
As a result, their sedimentary filling does not seem whether they are inherited frontal piggyback basins
to result from tectonic subsidence but rather from which suffered a long displacement above the major
borders uprising linked to fault tilting and to mud thrust system and which were deformed by mud
Fig. 11. CEPM seismic profile from Biju-Duval et al. (1982). Basins equally form back of thrusts or back-thrusts.

LATE N E O G E N E P I G G Y B A C K B A S I N S O N T H E B A R B A D O S R I D G E C O M P L E X 379
Fig. 12. Diapicar MN water-gun seismic profile (non migrated). Horizontal scales are the same on the seismic profile and line drawing.
intrusions. On the other hand, the influence of N W - PIGGYBACK BASINS AND LARGE SCALE EVOLUTION
SE- and NE-SW-trending structures oblique to the OF THE WEDGE
deformation front has also been observed or invoked
(Biju-Duval et al., 1982; Griboulard et al., 1989; Piggyback basins show a major tectonic control
Huyghe et al., 1996) in the arcward zone of the in the zone of initial accretion. Deformation of dis-
wedge. These structures have been related either tinct length scales (described above) controls their
to inherited oceanic ridges (Peter and Westbrook, sedimentary pattern. At a regional scale, numeri-
1976), basement faults extending from the Venezue- cal (Chalaron et al., 1995) and analogic modelling
lan margin to the wedge (Valery et al., 1985), or (Storti and McClay, 1995; Mugnier et al., 1997)
transcurrent faults associated with the E1 Pilar fault suggest that their repartition reflects the dynamic of
system in the southern part of the prism (Griboulard the accretionary prism.
et al., 1991). These structures may be reactivated
as transtensive or transpressive faults so that they Evolution of the piggyback basins from north to
generate local uplifting or sinking and may be then south
responsible for bulges alignment and for the com-
plex arcuate shape of the arcward basins. As mud The Barbados wedge shows north to south varia-
intrudes pre-exiting and new forming faults, it is tions. The most obvious one is its width, that varies
hard to assess the origin of the piggyback basins from less than 100 km in the northern part to more
occurring in the arcward areas. than 200 km in the southern part. A similar increase

380 E HUYGHE et al.
Fig. 13. Diapicar HI water-gun seismic profile (non migrated). Horizontal scales are the same on the seismic profile and line drawing.
Grey-shaded area: argilokinesis clay. Mud structures intrude poorly deformed sediments and determine sub-basins. East of the section,
sediments are interbedded with mud flows.
of the width and extent of the piggyback basins from few thousand meters at 12~ to 6-8 km at
and of the sedimentation rate occurs (see Table 1). 10ON.
The north-south distribution of the basins and the Shortening rate velocities vary from north to
evolution of the accretionary complex are controlled south due to the plate kinematics: earthquake mech-
by the volume balance between sediments entering anisms (Russo and Speed, 1992) and NUVEL-1
the wedge (sediments accreted at the toe, deposited plate model (DeMets et al., 1990, 1994) provide
upon the prism and incorporated to the wedge by strong support for the fight oblique collision model
subcretion) and those lost by subduction and surfi- of Caribbean-South American plate interaction but
cial erosion (Le Pichon et al., 1990; Speed, 1990). the prolongation towards the east of the Caribbean-
Sediment infilling of the piggyback basins mainly South American plate boundary zone (CA-SA PBZ)
originates from the South American shelf. Therefore is still debated (Speed et al., 1991). A reasonable
sediment supply above the wedge decreases from hypothesis is to infer that the shortening rate at
south to north, and basin geometry evolves from the back of the wedge decreases from the lati-
nearly overfilled basins to the south where recent tude of the CA-SA PBZ towards the south, leading
sediments frequently overlap all the structures, to to the arcuate westward shape of the deformation
northern areas with very poor sediments trapped front and occurrence of tear faults in the displaced
between thrusts sheets. wedge. The arcward basins located between 11~
Thickness of the sediment column incorporated and 10~ are frequently characterized by a cres-
within the wedge varies with the distance to the cent-shaped pattern (Fig. 4) and presumably develop
southern location of continental sources. The in- above dextral transcurrent shear zones (Gfiboulard
creasing thickness, from few hundred meters to et al., 1991).
nearly 7000 m, favors the forward propagation of
the deformation front and generates a general east- Evolution of the piggyback basins from east to
ward shift of the deformation front from 16~ south west
to 1 l~ Furthermore, the increase of the strati-
graphic thickness of the thrust sheets from north The above study shows that the sedimentary
to south induces an increase of the fold spacing pattern of frontal basins is controlled by out-off-
(Fig. 6) and an increase of the width of the basins sequence reactivation along forward verging thrusts.

Z
O
C~
>
Z
0
Z
,H
>
>.
C~

Fig. 14. (a) Sidescan sonar (SAR) images corresponding to the Diapisar 9020 seismic profile shown below. The size of mud volcano is about 8 km 2. Horizontal scales are the same for (a), (b) and (c). (b)
Diapisar 9020 seismic profile (3.5 kHz) across the northeastern end of a piggyback basin. The N40~ fault that crosses the section, controls the depression and the mud uprising. South of the mud
volcano, sediments of the basin have been turned up by the mud intrusion. Vertical exaggeration of seismic document x5. (c) Line drawing of (b).
382 E HUYGHE et al.
In the arcward areas of the wedge, the influence of influx and to maintain a constant mean slope (Davis
mud diapirism increases and tectonic disturbances et al., 1983). Assuming a constant subduction rate
are induced by fault reactivation and secondary and a constant basin sedimentary column, Le Pichon
faulting. The superimposition of clay diapiric struc- et al. (1990) have shown that the horizontal accre-
tures on deformations linked to tectonic accretion tion velocity dramatically decreases through time
and oceanic basement ridges leads to complex basin whereas 'internal thickening' increases. This 'inter-
morphologies bounded by steep topographic features nal thickening' is mainly related to motion along
(Fig. 16). The progressive tectonic evolution from thrusts (Mulugetta, 1988) and out-off-sequence re-
frontal to arcward basins is associated to a change in activation (Chalaron et al., 1995). The development
the regional topography with a slope angle varying of prominent relief on the sea floor at the hanging-
from about 1.5 ~ in the zone of initial accretion to wall of the faults and the sedimentary pattern of the
about 0.25-0.4 ~ in the arcward zone. This flattening SBRC piggyback basins attest for such reactivations.
of the topography is evidenced in more than 100-km- The source of the extensive clay diapirism of the
long profiles (Fig. 16) at a larger scale than the SBRC is provided by subcretion of muddy Miocene
convex shape induced by compaction and cohesion sediments occurring at the footwall of a major ramp
increase in the very frontal part (Zhao et al., 1986). of the decollement (Brown and Westbrook, 1988;
Dia et al., 1997). These muds are overpressured and
Mechanics of the accretionary wedge induce ductile deformation in the lowermost part of
the wedge beneath the arcward basins. Adaptations
In active margins, wedges develop and deform of the critical wedge model (Dahlen, 1990; Williams
until a critical taper is attained. They then slide sta- et al., 1994) suggest that such a ductile deformation
bly, continuing to grow self-similarly as additional in the lowermost part of a wedge would induce the
material is accreted at the toe. A critically tapered flattening of its topography.
wedge that is accreting fresh material deforms in- We suggest that the brittle-ductile wedge model
ternally while sliding in order to accommodate the applies to the SBRC. The frontal areas are char-
Fig. 15. (a) SAR image of the gravity structures on the flank of mud domes. 1 - scarps; 2 -- compressional ripples; (b) -- track of
seismic profile shown in (b). (b) 3.5-kHz seismic profile 9031 showing the slumps and the associated erosion that affected the departure
zone of the features. Horizontal scales are the same for (a) and (b). (c) Line drawing of SAR image shown in (a). The SAR interpretation
is drawn on the bathymetric map (1 = slumps; 2 = compressional ripples). N-S grey line outlines the axis of the mud volcanoes
alignment. (d) Line drawing of (b). Horizontal scales are the same for (c) and (d).

Fig. 15 (continued).

Or
a) b) c)
Iw MUDVOLCANO~ MUD VOLCANO ~ EI NW MUD VOLCANO S E]/
' ' o~oo~o % ~, , ~
MUD voLcANo /
I
. oo,
o oo
oo ,,oo x
"14~ '~ ( O9 I~. I Z

,,,,, (~ k o =,,~L
CARIBBEAN PLATE • . / ~ ,..a 0-~ l" o_
~0 ~ l ~ 0 I -J
12~ ~ >..
m
I
d) ARCWARD BASINS /
O
| g)
-4000~,"~'---- I
' I ' I , ' I
0.2 ~ /
0 400000 200 000 "300000 t 400000 6 00000
~O~::~:ER,CXN:.ii!~.iliiilii~'~~'%"..4"~?~
"~,~
............P . ~ T E ........... o~ :..-l~e, ~-o
-4000 ~ 1 ~:i IL~ ,~J"' ~ RISINGFLUIDS (1~ )

' I '' I ' I '
0 10 0000 200000 3 00000 400000 =500000 I ~,~ NO (RE)ACTIVATION
/ ~ (RE)ACTIVATION
~ DEEP MUD
-4000 I FRONTAL BASINS

I ' I ' I ' , I ' I ' )
o ,ooooo ~ooooo ~ooooo .| ~ooooo Ie f)
~1 NW PIGGYBACK BASIN3 SE I ~ PIGGYBACK BASIN1 ~ SI::l
! j -.

-4000
' I ' I ' I ' I , I
0 1000 O0 200000 300 000 400000 500 000
Fig. 16. Distinction between frontal and arcward piggyback basins. (a) Line drawing of the Diapicar HI seismic profile (Fig. 13). Grey shaded area: argilokinesis clay. Mud structures intrude poorly
deformed sediments and determine sub-basins. (b) Line drawing of the Diapicar FG seismic profile (from Griboulard et al., 1991). Mud (indicated by shading) rises up through pre-existing faults and
intrudes the cores of the anticlinal structures. It may generate thick mud flows that feed piggyback basins. (c) Location of bathymetric profiles 1 to 5. (d) Bathymetric profiles across the SBRC realized
from GEBCO bathymetric map (Fontas et al., 1985). The flattening of the topography in the arcward zone of the wedge fits with the occurrence of numerous mud diapirs. (e) Line drawing of the Diapicar
OP seismic profile (Fig. 10) showing the backward migration of depocenters in piggyback-back basins. (f) Line drawing of the Diapicar KL seismic profile (Fig. 8) showing the backward migration of
depocenters in piggyback basins. (g) Schematic evolution of the wedge and piggyback basins. Frontal basins are mainly controlled by forward-verging thrust system that form a brittle wedge, whereas ~Z
arcward basin development is mainly controlled by subcretion of deep muds that induces mud diapirism, a gentle regional slope and the occurrence of both backward and forward verging thrusts.
LATE NEOGENE PIGGYBACK BASINS ON THE B A R B A D O S RIDGE COMPLEX 385
acterized by a brittle behavior of the wedge and ~D
mean slope angle of about 1.4 ~ (Fig. 16g) and the 9~ = 9
major principal stress direction (cq in Fig. 16g)

is not horizontal, instead making an angle relative ~D
r
to the basal decollement (Davis et al., 1983). This
angle is nonetheless small as fluids pressure along
the decollement zone (Moore, 1989; Labaume et al.,
1995; Bangs et al., 1996) reduce the shear stress
along the decollement. The high traction along the
decollement near the toe favors tectonic activity on o
forward verging thrusts (Davis and Lillie, 1994) A
rather than the initiation of steeper backthrusts.
In the arcward zone, the flattening of the re-
gional topography reflects the ductile deformation
of the lowermost part of the wedge. The major
principal stress direction becomes nearly horizontal
above the ductile zone of deformation. As a result,
backthrusts and pre-existing thrusts may equally be
reactivated (Davis and Lillie, 1994), as evidenced by . ,...~
.A
the opposite wedge-shaped geometries of the sedi-

mentary bodies developed back of both thrusts and
back-thrusts (Fig. 11).
Therefore distinction between arcward and frontal
basins is based on the occurrence of abundant mud
diapirism and on the regional flattening of the to-
pography that are both related to changes in the
rheology of the deep material of the wedge. ~D
~D
c'-I C',l -~- ~
PIGGYBACK BASINS OF THE BARBADOS RIDGE r162
COMPLEX COMPARED TO OTHER PIGGYBACK

BASINS
Sedimentary environment
E
Piggyback basins have been initially defined in k9
shallow marine conditions (Fig. 1, adapted from .,.,~
9
Ori and Friend, 1984) inducing temporary erosions
>
under subaerial conditions. Piggyback basins also 9
developed under continental conditions as in the
Subandean Belt of Bolivia (Baby et al., 1995),
where the greatest piggyback basins of the world
occur. The late Neogene piggyback basins of the
SBRC formed under deep marine conditions. As a C ~D
result, the sediment supply is small compared to k9

o
other piggyback basins (see Table 1). The size and
sedimentation rate of the SBRC piggyback basins C'-I
9
are similar to those of the Sub-Himalayan piggyback <

~D
basins. The reasons for such a small supply are quite = "~
different: in the Sub-Himalayan Belt (Chalaron et al., tt')
1995; Bilham et al., 1997) the sedimentation rate is

==
controlled by the base-level of the rivers, whereas in
the SBRC most of the terrigenous sediment supply 9
transits from the Guyana Margin to the abyssal 0
plain through canyons (Damuth, 1975) and a small tt')
sedimentation occurs in the piggyback basins (0.12

mm/yr for the late Quaternary at 11~ latitude, from ~ Z ~ < < <

386 E HUYGHE et al.
I antetectonic sediments and Royden, 1994; Artoni and Casero, 1997) are
characterized by a ratio Se/Sh close to 20-40%. The
piggyback basins of the SBRC (1, 2 and 3 in Fig. 4)
are characterized by a small Se/Sh value (less than
0 5em
t I 5%). The Se/Sh ratio varies dramatically from north
to south in the BRC, and a comparison of geome-
try of the analogic models (Fig. 17) and schematic
cross-sections of the zone of initial accretion of the
BRC (Fig. 6) outlines the influence of syntectonic
sedimentation in the Barbados prism. The model of
Fig. 17a, performed without sedimentation, shows
analogy with the 15~ cross-section of Fig. 6a.
The model of Fig. 17b performed with a 5% Se/Sh
ratio looks like the 1 I~ cross-section also charac-
terized by a 5% Se/Sh ratio, whereas the section at
Fig. 17. Analogue models showing the influence of syntectonic 10~ and the model of Fig. 17c show very steep
sedimentation on thrust wedge evolution (adapted from Storti
thrusts linked to high syntectonic sedimentation.
and McClay, 1995). The three analogue experiments have been
performed with the same experimental conditions in a sandbox
apparatus. The three final geometries result from a horizontal
shortening applied to the same initial thickness of the antetec- SUMMARY
tonic sediments (drawn in white), but the syntectonic sediments
supply (grey shaded) varies: (a) experiment F8 was performed
The development of the SBRC frontal piggyback
without sedimentation and with 57% contraction; (b) experiment
F4 was performed with low syntectonic sedimentation and 43% basins is mainly controlled by a forward-verging
contraction; syntectonic sediments form piggyback basins above thrust system formed above a decollement, whereas
thrust sheets; (c) experiment F7 was performed with high syntec- the development of arcward basins is mainly con-
tonic sedimentation and 17% contraction. Thrusts dip very steep trolled by the subcretion of deep muds that induces
through the syntectonic sediments.
mud diapirism, a very gentle regional slope and the
occurrence of both backward and forward verging
Faug~res et al., 1991) consisting of fine material thrusts. In both areas, features with prominent re-
transported as suspension within the water masses. lief produce lumpiness in the mean regional slope
and consist of either hanging-wall of faults or mud
Rate of d e v e l o p m e n t of the p i g g y b a c k basins domes. They are locally eroded during reworking
of slope sediments and form barriers to terrigenous
Piggyback basins development and thrust wedge supply from the South American continent. There-
geometry are linked by a strong feed-back: the width fore these reliefs favor the trapping of sediments in
and depth of the depression is controlled by the the basins of the southern part of BRC, whereas its
thrust wedge thickness whereas the abundant sedi- northern part is poorly nourished.
ment supply dramatically changes the shape of the
Coulomb thrust wedge, decreasing the number of
thrust slices and the critical taper of the wedge ACKNOWLEDGEMENTS
(Fig. 17, from Storti and McClay, 1995). Simultane-
ously, syntectonic sedimentation ahead of the thrust We would like to thank N. Bangs, E. Calais, D.
front increases the thickness of the incipient thrust. Davis and E Mann for helpful reviews of the initial
A key parameter in the understanding of these feed- draft of the paper.
back phenomena is the ratio between the sediment
supply efficiency and the back-to-the-wedge velocity
(column 3 of Table 1) as inferred from numerical REFERENCES
modelling (Leturmy et al., 1995). The parameters
listed in Table 1 are crude estimations and are Artoni, A. and Casero, E, 1997. Sequentialbalancing of growth
structures, the late Tertiary example from the central Apen-
affected by great uncertainties. Nonetheless the sedi-
nine. Bull. Soc. Geol. Fr., 168: 35-50.
mentation rate at the toe of the accretionary complex Baby, E, Coletta, B. and Zubieta, D., 1995, Etude g6om6trique
(Se) partly reflects the sediment supply efficiency, et exp6rimentaled'un bassin transport6: exemple du synclinar-
whereas the adimensional parameter Se/Sh is useful ium de l'Alto Beni (Andes Centrales). Bull. Soc. Geol. Fr.,
for a comparison of natural piggyback basins or 166: 797-813.
Bangs, N. and Westbrook, G.K., 1990. Seismic modeling of
analog experiments. The large piggyback basins of the decollement zone at the base of the Barbados Ridge
the Subandean Belt (Baby et al., 1995) or Apennines Accretionary Complex. J. Geophys. Res., 96: 3853-3866.
(Pieri, 1989; Mugnier and Endignoux, 1991; Kruse Bangs, N., Shipley T.H. and Moore, G.F, 1996. Elevated fluid

LATE N E O G E N E P I G G Y B A C K BASINS ON THE BARBADOS RIDGE C O M P L E X 387
pressure and fault zone dilatation inferred from seismic models accretionary prism. Mar. Geol., 96: 247-267.
of the northern Barbados Ridge decollement. J. Geophys. Res., Faugbres, J.C., Gonthier, E., Griboulard, R. and Mass6, L., 1993.
101: 627-642. Quaternary sandy deposits and canyons on the Venezuelan
Belderson, R.H., Kenyon, N.H., Stride, A.H. and Pelton, C.D., margin and south Barbados accretionary prism. Mar. Geol.,
1984. A 'braided' distributary system on the Orinoco deep-sea 110:115-142.
fan. Mar. Geol., 56: 195-206. Fontas, E, Griboulard, R. and Prud'Homme, R., 1985. Mor-
Biju-Duval, B., Le Quellec, E, Mascle, A., Renard, V. and phogdn6se d'un prisme d'accrdtion et variabilit6 du style des
Valery, R, 1982. Multi-beam bathymetric survey and high res- ddformations sur une marge active. Apports et spdcificit6 des
olution seismic investigations on the Barbados ridge complex rdsultats d'une analyse cartographique interprdtative sur la ride
(Eastern Caribbean): a key to the knowledge and interpretation de la Barbade. In: A. Mascle (Editor), G6odynamique des
of an accretionary wedge. Tectonophysics, 80: 275-304. Carafbes, Symposium. Technip, Paris, pp. 221-233.
Bilham, R., Larson K., Freymueller J. and Project Idylhim mem- Griboulard, R., Faug6res, J.C., Blanc G., Gonthier, E. and Ver-
bers, 1997. GPS measurement of present-day convergence nette, G., 1989. Nouvelles 6vidences sddimentologiques et
across the Nepal Himalaya. Nature, 386: 61-64. gdochimiques de l'activit6 actuelle du prisme Sud Barbade.
Birsh, ES., 1970. The Barracuda Fault Zone in the Western C.R. Acad. Sci., Paris, 308:75-81.
North Atlantic: geological and geophysical studies. Deep Sea Griboulard, R., Bobier, C., Faug6res, J.C. and Vernette, G., 1991.
Res., 17: 847-859. Clay diapiric structures within the strike-slip margin of the
Bouysse, E and Westercamp, D., 1988. Effets de la subduction de southern leg of the Barbados prism. Tectonophysics, 192:
rides ocdaniques sur l'6volution d'un arc insulaire: l'exemple 383-400.
des Petites Antilles. G6ol. Fr., BRGM Eds., 2-3" 3-38. Henry, E, Le Pichon, X., Lallemant, S., Foucher, J.E, Westbrook,
Brown, K.M. and Westbrook, G.K., 1987. The tectonic fabric of G.K. and Hobart, M.A., 1990. Mud volcano field seaward of
the Barbados ridge accretionary complex. Mar. Pet. Geol., 4: the Barbados accretionary complex: a deep-towed side scan
71-81. sonar survey. J. Geophys. Res., 95: 8917-8929.
Brown, K.M. and Westbrook, G.K., 1988. Mud diapirism and Higgins, G.E. and Saunders, J.B., 1974. Mud volcanoes - - their
subcretion in the Barbados ridge accretionary complex: the nature and origin. In: Contributions to the Geology and Pale-
role of fluids in accretionary processes. Tectonics, 7: 613-640. obiology of the Caribbean and Adjacent Areas. Naturforsch.
Burbank, D.W. and Verg6s, J., 1994. Reconstruction of topogra- Ges. Verh., Basel, 84: 101-152.
phy and related depositional systems during active thrusting. J. Huyghe, E, Griboulard, R., Faug~res, J.C., Gonthier, E. and Bo-
Geophys. Res., 99: 20281-20297. bier, C., 1996. Gdom6trie des bassins du prisme Sud-Barbade.
Calassou, S., Larroque, C. and Malavielle, J., 1993. Transfer Bull. Soc. G6ol. Fr., 167: 345-359.
zones of deformation in thrust wedges: an experimental study. Jordan, T.H., 1975. The present-day motions of the Caribbean
Tectonophysics, 221: 325-344. Plate. J. Geophys. Res., 80: 4433-4439.
Chalaron, E., Mugnier, J.L. and Mascle, G., 1995. Control on Kruse, S.E. and Royden, L.H., 1994. Bending and unbending of
thrust tectonics in the Himalayan foothills: a view from a an elastic lithosphere: the Cenozoic history of the Apennine
numerical model. Tectonophysics, 248: 139-163. and Dinaride foredeep basins. Tectonics, 13: 278-302.
Dahlen, F.A., 1990. Critical taper model of fold-and thrust belts Labaume, E, Henry, E and scientist party of Leg ODP 156,
and accretionary wedges. Annu. Rev. Earth Planet. Sci., 18: 1995. Circulation et surpression de l'eau interstitielle dans le
55-99. prisme d'accrdtion nord-Barbade: r6sultats du Leg ODP 156.
Damuth, J.E., 1975. Echo-character of the western equatorial at- C.R. Acad. Sci. Paris, 320: 977-984.
lantic floor and its relationship to the dispersal and distribution Lallemant, S.J.C., Henry, E, Le Pichon, X. and Foucher, J.E.
of terrigenous sediments. Mar. Geol., 18: 17-45. 1990, Detailed structure and possible fluid paths at the toe
Davis, D. and Lillie, J.R., 1994. Changes in mechanical response of the Barbados accretionary wedge (ODP Leg 110 area).
during continental collision: active example from the foreland Geology, 18: 854-857.
thrust belts of Pakistan. J. Struct. Geol., 16: 21-34. Langseth, M., Westbrook, G.K. and Hobart, M.A., 1988. Geo-
Davis, D., Suppe, J. and Dahlen F.A., 1983. Mechanics of physical survey of a mud-volcano seaward of the Barbados
fold-and-thrust belts and accretionary wedges. J. Geophys. ridge accretionary complex. J. Geophys. Res., 93:1049-1061.
Res., 88:1153-1172. Leonard, R., 1983. Geology and hydrocarbon accumulation,
DeMets, C., Gordon, R.G., Argus, D.E and Stein, S., 1990. Colombus Basin, offshore Trinidad. Am. Assoc. Pet. Geol.
Current plate motions. Geophys. J. Int., 101: 425-478. Bull., 67: 1081-1093.
DeMets, C., Gordon, R.G. and Vogt, E, 1994. Location of the Le Pichon, X., Foucher, J.P., Boul6gue, J., Henry, E, Lallemant,
Africa-Australia-India triple junction and motion between the S., Benedetti, M., Avedik, F. and Mariotti, A., 1990. Mud
Australian and Indian plates: results from an aeromagnetic volcano field seaward of the Barbados Accretionary Complex:
investigation of the Central Indian and Carlsberg ridges. Geo- a submersible survey. J. Geophys. Res., 95:8931-8944.
phys. J. Int., 119: 893-930. Leturmy, P., Mugnier, J.L. and Chalaron, E., 1995. Erosion et
Dia, N.A., Castrec, M. and Boul6gue, J., 1997. Trinidad mud sddimentation au voisinage d'un anticlinal de rampe: apport
volcanoes: where do the expelled fluids come from? Terra d'un module numdrique. Bull. Soc. G6ol. Fr., 166: 345-359.
Nova, 9, Abstr. Suppl., 1:319. Mascle, A. and Moore, J.C., 1988. Proceedings of the Ocean
Embley, R.W. and Langseth, M.G., 1977. Sedimentation pro- Drilling Project, Initial Reports (Pt. A). College Station, TX,
cesses of the continental rise of north-eastern South-America. 110, 603 pp.
Mar. Geol., 25: 279-297. Mascle, A., Endignoux, L. and Chennouf, T., 1990. Frontal
Endignoux, L. and Mugnier, J.L., 1990. The use of a forward accretion and piggyback basin development at the southern
kinematic model in the construction of balanced cross-sec- edge of the Barbados ridge accretionary complex. Proc. ODE
tions. Tectonics, 9: 1249-1262. Sci. Results, 110:17-28.
Ewing, M., Carpenter, G., Windisch, C. and Ewing, J., 1973. Moore, J.C., 1989. Tectonics and hydrogeology of accretionary
Sediment distribution in the oceans: the Atlantic. Geol. Soc. prisms: role of the ddcollement zone. J. Struct. Geol., 11, 1/2:
Am. Bull., 84:71-88. 95-106.
Faug6res, J.C., Gonthier, E., Mass6, L., Parra, M., Ports, J.C. and Moore, J.C. and Mascle, A., 1990. Proceedings of the Ocean
Pujol, C., 1991. Quaternary deposits on the South Barbados Drilling Program, Scientific Results. College Station, T X , ,

388 E H U Y G H E et al.
110, 448 pp. R. and Dercourt, J., 1990. Paleogeodynamic maps of the
Mugnier, J.L. and Endignoux, L., 1991. Cin6matique et Carribean: 14 steps from Lias to Present. Bull. Soc. G6ol. Fr.,
vitesse d'6volution des nappes superficielles: une simulation VI: 915-919.
num6rique. Rev. Fr. G6otech., 56: 23-32. Storti, E and McClay, K., 1995. Influence of syntectonic sedi-
Mugnier, J.L., Baby, E, Colletta B., Vinour, E, Bale, E and mentation on thrust wedges in analogue models. Geology, 23:
Leturmy, E, 1997. Thrust geometry controlled by erosion and 999-1002.
sedimentation: a view from analogue models. Geology, 25 (5) Suppe, J., 1983. Geometry and kinematics of fault bend folding.
(in press). Am. J. Sci., 283: 684-721.
Mulugetta, G., 1988. Modelling the geometry of Coulomb thrust Suppe, J., Chou, G.T. and Hook, S.C., 1992. Rates of folding
wedges. J. Struct. Geol., 10 (8): 847-860. and faulting determined from growth strata. In: K.R. McClay
Ori, G.G. and Friend, EE, 1984. Sedimentary basins formed and (Editor), Thrust Tectonics. Chapman and Hall, London, pp.
carried on active thrust sheets. Geology, 12: 475-478. 105-121.
Peter, G. and Westbrook, G.K., 1976. Tectonics of southwestern Sykes, L.R., MCcann, W.R. and Kafka, A.L., 1982. Motion of
North Atlantic and Barbados Ridge Complex. Am. Assoc. Pet. Caribbean Plate during last 7 million years and implications
Geol. Bull., 60:1078-1106. for earlier Cenozoic movements. J. Geophys. Res., 87: 10656-
Pieri, M., 1989. Three seismic profiles through the Po Plain. 10676.
In: A.W. Bally (Editor), Atlas of Seismic Stratigraphy, Vol. 3. Valery, E, Nely, G., Mascle, A., Biju-Duval, B., Le Quellec, E
Am. Assoc. Pet. Geol. Stud. Geol., 27:90-110. and Berthon, J.L., 1985. Structure et croissance d'un prisme
Robertson, R.E and Burke, K., 1989. Evolution of southern d'accr6tion tectonique proche d'un continent: la ride de la
Carribean plate boundary, vicinity of Trinidad and Tobago. Barbade au sud de l'arc Antillais. In: A. Mascle (Editor),
Am. Assoc. Pet. Geol. Bull., 73 (1): 490-509. G6odynamique des Cara'fbes. Technip, Paris, pp. 173-186.
Robertson, E and Burke, K., 1991. Evolution of the southern Weber, J.C., Ambeh, W.A., Dixon, T.D., Speed, R.C., Lynch,
Caribbean plate boundary, vicinity of Trinidad and Tobago: L.L., Saleh, J. and Webb, E, 1995. Historic geodetic con-
reply. Am. Assoc. Pet. Geol. Bull., 75 (11): 1795-1796. straints on Caribbean-South America relative plate motion,
Roure, E, Howell D.G., Guellec, S. and Casero, E, 1990. plate boundary zone kinematics and seismic risk. 3rd Geolog-
Shallow structures induced by deep-seated thrusting. In: J. ical Conference of the Tobago and 14th Caribbean Geological
Letouzey (Editor), Petroleum Tectonics in Mobile Belts. Tech- Conference, Port of Spain, Trinidad, pp. 83-84.
nip, Paris, pp. 15-30. Westbrook, G.K., 1982. The Barbados ridge complex: tectonics
Russo, R.M. and Speed, R.C., 1992. Oblique collision and tec- of a mature forearc system. In: J. Leggett (Editor), Trench and
tonic wedging of the South American continent and Caribbean Forearc Geology. Geol. Soc. London Spec. Publ., 10: 275-
terranes. Geology, 20: 447-450. 290.
Russo, R.M., Speed, R.C. and Okal, E.A., 1993. Seismicity and Westbrook, G.K. and Smith, M.J., 1983. D6collements and mud
tectonics of the Southeastern Caribbean. J. Geophys. Res., 98: volcanoes: evidence from the Barbados Ridge complex for
14299-14319. high pore fluid pressure in the development of an accretionary
Shih, T.T., 1967. A survey of the active mud volcanoes in Taiwan complex, Geology, 11: 279-283.
and a study of their type and the character of the mud. Pet. Westbrook, G.K., Mascle, A. and Biju-Duval, B., 1984. Geo-
Geol. Taiwan, 5: 259-311. physics and the structure of the Lesser Antilles Forearc. Init.
Silver, E.A., Breen, N.A. and Prasetyo, H., 1986. Multibeam Rep. DSDE 78A: 23-38.
study of the Flores back arc thrust belt, Indonesia. J. Geophys. Westbrook, G.K., Ladd, J.W., Buhl, E, Bangs, N. and Tiley, G.J.,
Res., 91: 3489-3500. 1988. Cross section of an accretionary wedge: Barbados Ridge
Speed, R.C., 1985. Cenozoic collision of the Lesser Antilles arc complex. Geology, 16: 631-635.
and continental South America and the origin of the E1 Pilar Williams, C.A., Conners, C., Dahlen, E, Price, E. and Suppe
Fault. Tectonics, 4 (1): 41-69. J., 1994. Effect of the brittle-ductile transition on the topog-
Speed, R.C., 1990. Volume loss and defluidization history of raphy of compressive mountain belts on Earth and Venus. J.
Barbados. J. Geophys. Res., 95: 8983-8996. Geophys. Res., 99: 19947-19974.
Speed, R.C., Russo, R., Weber, J. and Rowley, K.C., 1991. Wright, A., 1984. Sediment distribution and depositional pro-
Evolution of Southern Caribbean Plate Boundary, Vicinity of cesses operating in the Lesser Antilles intraoceanic island arc,
Trinidad and Tobago: discussion. Am. Assoc. Pet. Geol. Bull., Eastern Caribbean. Init. Rep. DSDE 78A: 301-324.
75: 1789-1794. Zhao, W.L., Davis, D.M., Dahlen, EA. and Suppe, J., 1986. Ori-
Stein, S., Engeln, J.F. and Wiens, D.A., 1982. Subduction, seis- gin of convex accretionary wedges: evidence from Barbados.
micity and tectonics in the Lesser Antilles Arc. J. Geophys. J. Geophys. Res., 91: 10246-10258.
Res., 87 (B10): 8642-8664. Zoetemeijer, R. and Sassi, W., 1992. 2-D reconstruction of thrust
Stephan, J.E, Mercier, J., De Lepinay, B., Calais, E., Tardy, M., evolution using the fault-bend fold method. In: K.R. McClay
Beck, C., Carfantan, J.Ch., Olivet, J.L., Vila, J.M., Bouysse, E, (Editor), Thrust Tectonics. Chapman and Hall, London, pp.
Mauffret, A., Bourgeois, J., Thery, J.M., Tournon, J., Blanchet, 133-140.

Chapter 15
Tectonic Evolution of the Grenada Basin
DALE E. BIRD, STUART A. HALL, JOHN E CASEY and PATRICK S. M I L L E G A N
Detailed analyses of gravity, seismic reflection and refraction data are integrated with an earlier interpretation of magnetic data
to produce a coherent model for the tectonic evolution of the Grenada basin that suggests that the basin formed by near east-west
extension.
Although the seafloor of the Grenada basin changes from smooth and undisturbed in the south to rugged with relatively high
relief in the north, Bouguer anomalies and two-dimensional and three-dimensional gravity models, based upon seismic refraction
and reflection data, reveal that the crust gradually thins in an east-west sense towards the center of the basin. Typical back-arc
crust is observed in the southern part of the basin, but refraction data are not sufficiently reliable in the northern part to adequately
determine the nature of the crust.
Several curvilinear discontinuities in magnetic, gravity and bathymetric trends are observed. These discontinuities, when
integrated with two-dimensional and three-dimensional modeling and analyses of Bouguer gravity anomalies, are interpreted to
be due to late Tertiary compressional forces in the northern part of the region. These compressional forces have resulted in the
bifurcation of the Lesser Antilles island arc north of 15~ the westward displacement of part of the Aves Ridge (a remnant
island arc), and the crustal deformation observed in the northern Grenada basin. The compressional forces also appear to have
sufficiently disrupted the crust in the northern Grenada basin such that earlier magnetic anomaly patterns have been modified to
yield the observed magnetic signature.
INTRODUCTION If the Grenada basin formed by e a s t - w e s t exten-

sion, then magnetic anomalies produced by e a s t -
Various orientations ranging from north-south to west seafloor spreading in the back-arc region
east-west have been proposed for the direction of would be characterized by small amplitudes. Fur-
extension and subsequent formation of the Grenada thermore, if these north-south spreading centers
basin (Tomblin, 1975; Bouysse, 1988; Pindell and were offset by east-west-trending features, then
Barrett, 1990). Large-amplitude, long-wavelength anomalies produced by these features would ex-
magnetic anomalies, oriented generally e a s t - w e s t hibit large amplitudes that would overwhelm the
over the basin, have been used by authors to ar- north-south-trending anomalies.
rive at these conflicting interpretations of the basin's We have previously analyzed shiptrack and grid-
evolution. ded magnetic data over the Grenada basin, includ-
The Grenada back-arc basin (Fig. 1) formed in ing anomaly trend analyses and three-dimensional
early Tertiary time at approximately 12~ (Duncan forward modeling (Bird et al., 1993). Subtle, sub-
and Hargraves, 1984; Ghosh et al., 1984; Pindell et parallel to the Lesser Antilles island arc anomalies
al., 1988; Ross and Scotese, 1988). Anomalies pro- have been identified over the southern part of the
duced by east-west-trending features at low mag- basin and are interpreted to be due to near e a s t -
netic latitudes have significantly larger amplitudes west extension. Amplitudes of these anomalies are
than anomalies produced by north-south-trending approximately 40 nT while amplitudes of e a s t -
features. The orientation of the Earth's magnetic west-oriented anomalies are about 300 nT. This
field at low latitudes is nearly horizontal and seven-fold increase, the difference between north-
its direction is essentially to the north, therefore south- and east-west-oriented profiles, was further
east-west-oriented features produce large perturba- supported by three-dimensional forward modeling.
tions in the Earth's magnetic field while north- Although results from our analyses of magnetic
south-oriented features result in minor anomalies. data support near e a s t - w e s t extension, and probable

390 D . E . B I R D et al.
20"N
0 100 200 300 km

I I t ~ I
~t r
9.o
~_.,,_ ~ ~ ._
Northh American Plate
18"N ,,~=:~, ~ ~ 0~,~
f4~.--~J~ ,3 Saba Bank ' q ~ ~~ ~5~L
[\ (~Caribbean Plat, k ~r
, ,.
16"N "-" I
= ', 2. J :,
"~ I
c o " -~ c
14"N ) A
/ i / I ~ ,",..
12"N~ 2"J ~2 o/,5

~
I 9 ~ U
.
,m,, I . . . . ~- : -. - ~ - i i 4",
'".:, '.";..~
~ ...
.~ ~--!.:.!~~South i
,o.,- ' .....
...'~:" " " " ' ~ American t
i ,
"
i
i:~
::.-r.,,~ i
Plate \ I
~'w 64"w 62ow 6o'w 58"w
Fig. 1. Physiographyof the eastern Caribbean with 2, 4 and 5 km isobaths contoured (after Bouysse, 1984). The outline of the study area,
trace of the subduction zone, and strike-slip fault zones which define the North American/Caribbean and South American/Caribbean
plate boundaries are displayed. Heavy dashed lines indicate probable locations for plate boundaries. The inner and outer arcs are
represented by dashed and dotted lines, respectively.
back-arc spreading for the formation of the Grenada MODELS FOR THE FORMATION OF THE GRENADA
basin, the results should be integrated with analyses BASIN
of gravity and seismic data to provide more evidence
with which to examine the evolution of the region. Various kinematic models for the formation of
In this study, interpretations of seismic and gravity the Grenada basin are described by Tomblin (1975),
data are combined with our earlier interpretation Bouysse (1988), Pindell and Barrett (1990) and Bird
of magnetic data. Then two-dimensional and three- et al. (1993). These models outline the formation
dimensional forward models, incorporating gravity of the basin by near east-west, northeast-southwest
and seismic data (both reflection and refraction), extension, north-south extension and east-west ex-
are constructed for the region. Comparison of these tension, respectively.
results for the Grenada basin with the morphol-
ogy of the younger Andaman Sea back-arc basin East-west extension
(11 Ma) provide a compelling interpretation for
the evolution of the Grenada basin (Mukhopadhyay, Tomblin (1975) describes two possible scenarios
1984). for east-west extension. The first involves an early

TECTONIC EVOLUTION OF THE GRENADA BASIN 391
.....
North American Plate Ridge to the west) requires the formation and subse-
quent spreading from a north-south-oriented median
ridge. He reports that no such ridge is observed.
However, since extension is no longer evident in the
Caribb o.x, 5 Grenada basin, the loss of heat at spreading centers

would cause the ridge system to cool and subside.
Therefore a ridge system may not be obvious in
the data, but the center of the basin would still
produce a Bouguer gravity high when compared to
(A) thicker crust on either side. Intuitively, east-west
extension as proposed by Tomblin seems most rea-
South American Plate
sonable, and magnetic anomaly lineations over many
back-arc basins support this conclusion (i.e., the
North American Plate
South Sandwich, Lau, Havre, North Fiji, Banda Sea,
Parece-Vela, Shikoku, and Sea of Japan basins).
North-south extension
Caribbean Plate ~ , Pindell and Barrett (1990) feel that the Lee-
ward Antilles were coupled to the northern edge
of the South American plate and that north-south
spreading in the Grenada basin is a result of con-
(B) tinued eastward progression of the Caribbean plate.
South American Plate The basin therefore formed by right lateral shear
(Fig. 2B). In this model the Leeward Antilles may
North American Plate have been part of the Aves Ridge prior to the for-
mation of the basin and represents fragmentation
of the arc as the Caribbean plate progressed east-
ward. If the long-wavelength, high-amplitude, east-
west-trending magnetic anomalies over the basin
are produced by seafloor spreading, then this model
C a r i b b ~ } Pla
appears to fit the magnetic data.
For this model differences in the nature of the
crust between the northern and southern parts of
the basin are important. Pindell and Barrett (1990)
(C) suggest that the northern part of the basin is block
South American Plate faulted with no development of oceanic crust; how-
0 300 km ever, the southern part of the basin may be under-
lain by oceanic crust. Although Pindell and Bar-
Fig. 2. (A) Possible east-west extension due to a westward shift
of the Aves Ridge for the opening of the Grenada basin (Tomblin,
rett's model may appear overly complex, magnetic
1975). (B) Possible north-south extension for the opening of anomaly patterns over the Andaman Sea basin ap-
the basin (Pindell and Barrett, 1990). (C) Possible northeast- pear to be oriented at a high angle to the trench line
southwest extension for the opening of the basin (Bouysse, of its subduction zone.
1988). Large arrows indicate the relative motions of the North
American, Caribbean, and South American plates and small
arrows indicate the directions of extension for the formation of
Northeast-southwest extension
the basin.
Similar to Pindell and Barrett's model, Bouysse
(1988) describes a possible mechanism for extension
Tertiary eastward shift of the subduction zone, and in which coupling of the southern part of the Lesser
the second involves a westward shift of the Aves Antilles with the South American plate precedes the
Swell (Fig. 2A). In the first model the older rocks opening of the basin (Fig. 2C). He suggests that the
of la D6sirade would either have been part of an Netherlands-Antilles, Lesser Antilles and Greater
older orogeny and moved eastward with the sub- Antilles formed a continuous Mesozoic arc prior
duction zone, or they may represent obduction of to the injection of the Caribbean plate between the
part of the Atlantic floor onto the eastward mov- American plates. Bouysse further theorizes that sub-
ing Caribbean plate. Tomblin's alternative scenario sequent seafloor spreading was oriented northeast-
for the formation of the basin (a shift of the Aves southwest at the onset of the Cenozoic in a seg-

392 D.E. BIRD et al.
mented manner such as described by Tamaki (1985) North American Plate

for the Sea of Japan basin. Initial spreading was in
the southernmost portion of the basin and gradually
progressed northward over time.
Bouysse's model provides for contemporaneous
formation of the Yucatan and Grenada basins. This
development occurred when the northeast-travelling
Caribbean plate, with respect to the North American
plate, was wedged between the North American and
Caribbean Plat / i/~JJ~ J
South American plates in Late Cretaceous/early
Tertiary time. Subsequent to this collision, the
Caribbean plate rotated clockwise and began travel-
ling in an east-west direction.
South AmericanPlate
_J ~
fJ (A)
Near east-west extension
0 300 km
" " I l
Bird et al. (1993) interpret the basin to have
formed by near east-west extension similar to that
described by Tomblin (1975) with late Tertiary tec-
tonic movements disrupting the crust, and magnetic North AmericanPlate
signature, over the northern part of the basin (Fig. 3).
These conclusions are supported by forward three-
dimensional magnetics modeling and identification
of subtle anomaly trends over the southern part of the
basin. These low-amplitude (about 40 nT) anoma-
lies are interpreted as those produced by a roughly
north-south-oriented spreading center(s) near the Caribbean
geomagnetic equator. The chaotic, patchy anomalies
over the northern part of the basin are thought to
have formed by seafloor spreading also, but later
were disrupted by the late Tertiary event responsible
for the bifurcation of the Lesser Antilles.
South AmericanPlate
(B)
Fig. 3. Two-step model for the formation of the Grenada basin by

DATA BASE
east-west extension. Large arrows indicate directions of relative
plate motion and small arrows indicate directions of extension
The data base for this study includes gridded and basin formation. (A) The basin is formed fairly uniformly by
gravity and bathymetry data (Figs. 4 and 5, and Web- early Tertiary seafloor spreading. (B) Late Tertiary compressional
Figs. 15.1-41), shipboard gravity and bathymetry tectonism disrupts the northern part of the basin (indicated by
long northeast-oriented arrows).
profile data (Fig. 6), multiple channel seismic reflec-
tion sections (Fig. 7A), and reversed and unreversed
seismic refraction profiles (Fig. 7B). Grids of total Multi-channel seismic reflection lines were ob-
intensity magnetic anomalies (2 km) and free-air tained from the Institute Franqais du Petrole (IFP),
gravity (6 km) were compiled in 1987 by the Geo- Lamont-Doherty Geological Observatory (LDGO) of
logical Society of America Decade of North Ameri- Columbia University, and the University of Texas at
can Geology (DNAG) Committees on the Magnetics Austin (UT) Institute for Geophysics (Fig. 8). The
and Gravity Maps of North America, respectively. reflection lines obtained from UT were originally
These grids are available from the NOAA/National acquired by Gulf Oil Company (GULFREX lines).
Geophysical Data Center (NGDC). The bathymetry Identification, fold, orientation, length, and year of
grid is a portion of the ETOPO5 data. The ETOPO5 acquisition of seismic reflection sections used are dis-
data set is a 5-minute grid of topography and played in Table 1. Seismic reflection line RC1904
bathymetry for the entire world and is available from (LDGO) was available in digital form as well as its
the NGDC as well. Since the grid spacing is not con- corresponding velocity analyses. Nine horizons of
stant as latitude varies, the data were regridded to 9 this line were analyzed by LDGO and those within
km utilizing a moving average least squares method. the basin were used to interpret a velocity function
for modeling.
1 Available at http://www.elsevier.nl/locate/caribas Refraction information used for the study comes

T E C T O N I C E V O L U T I O N OF T H E G R E N A D A BASIN 393
Fig. 4. Free-air gravity anomalies over the study area. Heavy solid and dashed lines indicate interpreted curvilinear zones of disruption
(dashed lines indicate reduced confidence). Contour interval is 10 mGal.
Table 1
Parameters of multiple channel reflection seismic data
Organization Line No. Fold Direction Year Length (km)
LDGO C 1904 24 E 1975 870.50

IFP A4 48 SE 1973 116.66
IFP 118A 48 E 1974 169.77
IFP 131A 48 SE 1974 222.73
IFP 217A 48 N 1976 111.36
UT (GULFREX) LS-11 48 NE 1975 169.85
UT (GULFREX) LS- 12 48 NE 1975 143.96
UT (GULFREX) LS- 14 24 E 1975 304.84
UT (GULFREX) LS-15A 24 NE 1975 165.40
UT (GULFREX) VB-11 24 NE 1975 138.73
UT (GULFREX) VB-12 24 SE 1975 132.89
Total 2546.69

Fig. 5. Bathymetryof the study area. Heavy solid and dashed lines indicate interpreted curvilinear zones of disruption (dashed lines also
indicate reduced confidence). Contour interval is 100 m.
from published studies (Ewing et al., 1957; Officer Taylor and Karner (1983), and Tamaki (1985). A
et al., 1959; Edgar, 1968; Boynton et al., 1979; back-arc basin is defined here as an extensional
Speed and Westbrook, 1984). In general, coverage of basin located at the edge of an overriding plate of
refraction data is sparse, particularly in the northern a subduction zone such that the basin formed after
part of the basin where the velocity structure is initiation of subduction.
represented by a single, unreversed profile (Fig. 7B). At least nineteen basins have formed around the
world by extension, or seafloor spreading, behind is-
land arcs (Table 2). Back-arc basins formed by or-
TECTONICS OF BACK-ARC BASINS ganized seafloor spreading should produce magnetic
anomalies parallel to spreading centers and those with
Models for the formation of back-arc basins are well defined magnetic anomaly lineations include
discussed by, among others, Karig (1971), Sleep and the South Sandwich, Shikoku, Bismark, South Fiji,
Toskoz (1971), Poehls (1978), Uyeda and Kanamori Lau, Havre, Banda, and Andaman basins (Taylor,
(1979), Dewey (1980), Cross and Pilger (1982), 1979; Weissel, 1980, 1981; Barker and Hill, 1981;

z ........ 64:W . . . . ., 62~ 60"W 64:W 62,W 60~ m
0
:Z
"-%
It
m
%. ... '
<
...... . "'***.. o
% ." o.
9,, ..
0
Z
0
z "-'~" "" >,~, I

o ,"
*" i . .....-. . . . . . . . . . c:'--~:."" 'r""-t . . . . J . . ~ ' . ' : ~ ,-,.':..J,
*" ~ " " -'\ ;~'~I Z

\~ ~,~ ,...,
>.
) 9 ":':':.,,.. ,
:>
_ _ ,
,:.:: ..... \ .~ -... i >
..... ''" I j
Z
' ~ l~:'. . . . . . .
,.{..t, . . . . . . "
zo ~
.,::.-:":.............................................. ~ ~ i "" ..... .. Zo
~" : ',,, "".~ .....
\
9 . ............................................. ) ............... :....................
/ /
/ /
/ i
,o. _,t--
9~ , ! i
./ .... ;
i --. . !! II "- . ]
/ . , "... ; ,:.-C ..... ;~..
/,,, ,>. / .."
/ ..."" ". "~.. I'Q ~... .,~ ......... : . ~ , . / .,~ ,
.: . . . . . ~.... ,.. ~ . ~ " : ........... ,
-. %" \
L . ...~%..~."
9 -....

-..~,,.~.....:. ..... _~+"
".. o . "'
................... :..? .... I""
I
i "--'"...'-,,- :' ~-L..
/ i , ......
9 % %
/
........ " . . . . . . . . ~ ..... k 1~~
I ""':J .... I \ ,,/ --
(A) (B)
0 100 200 km 100 200 km
I I I ! I L/I
Fig. 6. S h i p t r a c k data c o v e r a g e in the study area: (A) gravity, (B) b a t h y m e t r y .

ta3
I "-I ql I
z 66"W 64"W 62"W 60"W 58"W

q,- I "; I I t
l "
LS-15A/ ~O B
i,
O
F_
QD
-- _
~~ ,,. LS-14
;-12
b c d \
-=iil -=-II --,,0 '=-0
LS-11 C1904
F_ LASP B
118A
i
29
b
2kin ) Z;
63
0 100 200 km

I i I 0I 50
,
100
I
150
1
km
P i= dl . I --
64"W 62"W 60"W
(A) (B)
Fig. 7. Multi-channel seismic reflection data and refraction profiles. (A) Reflection data. LS = GULFREX, VB = GULFREX, A = IFP, C = LDGO. (B) Refraction data. For the Lesser Antilles Seismic
Project (LASP), circles and triangles indicate shot and receiver locations, respectively, otherwise circles indicate receiver locations for the reversed and unreversed profiles. Velocities for profiles LASP B,
21, 22, 29, a, b, c, d, e, and f are shown in Fig. 15.
Table 2 Table 3
Back-arc basins of the world Means and standard deviations for the velocity structure of se-
lected back-arc basins (regarding transition layers, means and
Basin Subduction zone standard deviations are only calculated for the total of all transi-
tion layers)
Aegean Hellenic
Andaman Andaman
Back-arc basin Layer2 Layer3 Transition Mantle
Banda Java
Bismark New Britian M STD M STD M STD
Fiji Plateau New Hebrides
Sea of Japan 5.5 0.2 6.4 0.4 7.4, 7.5 8.0 0.2
Grenada Lesser Antilles
Okinawa 5.1 0.3 6.0 0.2 7.1, 7.4 - -
Havre Kermadec
Shikoku 4.9 0.6 6.7 0.3 - 8.2 0.3
Kurile Kurile
Parece-Vela 5.3 0.5 6.9 0.1 7.7 8.4 0.2
Lau Tonga
Andaman 5.4 0.3 6.2 0.3 - - -
Mariana Mariana
Sulu 5.4 0.3 6.4 0.3 7.2 8.3 -
Okinawa Ryuku
Celebes 5.2 0.2 6.7 0.3 - 8.1 0.3
Parece-Vela Mariana
Banda 5.1 - 6.6 - - 8.0 -
Sea of Japan Japan
Grenada 4.9 0.5 6.2 0.2 7.4, 7.4 8.2 -
Shikoku Bonin
Mariana 5.1 0.6 6.2 0.4 7.2, 7.4 8.2 0.4
South Aleutian Aleutian
Havre 4.4 - 6.6 - - 8.5 -
South Fiji Kermadec, Tonga
Means 5.1 0.3 6.4 0.3 7.3, 0.2 8.2 0.2
South Sandwich South Sandwich
Tyrrhenian Hellenic
Yucatan Cuban
Japan, and Sulu basins (Ewing et al., 1957; Karig,
1971; Hayes et al., 1978; Bibee et al., 1980; Hus-
Bandy and Hilde, 1983; Brooks et al., 1984; Mc- song and Uyeda, 1981; Curray et al., 1979), appear
Cabe et al., 1985, 1986). Some back-arc basins have to contain an additional layer exhibiting intermedi-
anomaly patterns characterized by weak or subtle ate velocities between layer 3 and mantle velocities
trends including the Grenada, Sea of Japan, Parece- (defined as the transitional layer for this study).
Vela, and West Fiji basins (Weissel, 1981; Brooks Table 3 displays results from the eleven back-arc
et al., 1984; Speed and Westbrook, 1984; Tamaki, basins studied by analyses of refraction velocities.
1985; Bird et al., 1993). Finally, some back-arc Means and standard deviations of the data are dis-
basins have no predominant magnetic pattern over played for each basin as well as for the entire data
them such as the Kurile, Okinawa, and Mariana set. The standard deviations illustrate the variability
basins (Lee et al., 1980; Weissel, 1981; Brooks et al., of the data, while the means support the concept of
1984; Okuma et al., 1990). Magnetic anomaly pat- layered crusts, similar to normal oceanic.
terns produced by back-arc spreading may be related
to the orientations of back-arc extension, which in
turn may be related to the interactions between over- REGIONAL SETTING
tiding plates and their respective subducting slabs.
Four scenarios that lead to a lack of coherent Geology
magnetic lineations over back-arc basins are: (1)
complex rifting and segmentation of spreading cen- The Grenada basin is bounded to the north by
ters producing incoherent anomaly trends (Tamaki, the Saba Bank at the junction of the Greater and
1985); (2) young basins, such as the Mariana and Lesser Antilles, and to the south by the continental
Okinawa, not sufficiently developed to produce well rise of northern Venezuela (Fig. 1). The Aves Swell
defined trends; (3) basins formed by e a s t - w e s t ex- (or Ridge) and the Lesser Antilles arc form the
tension near the magnetic equator (the Grenada basin western and eastern limits of the basin, respectively.
is a probable candidate for this scenario); and (4) The shape of the basin is arcuate with approximate
a basin formed during a time when there were no dimensions of 640 km (north-south) by 140 km
geomagnetic reversals, such as in Mid-Cretaceous (east-west), its water depth ranges from about 2
time (about 118-84 Ma). to 3 km. Sediment thickness ranges from 2 km in
Refraction data from the Andaman, Banda, the north to 9 km in the south (Bouysse, 1988).
Celebes, Grenada, Havre, Mariana, Okinawa, Morphologically, the ocean floor of the Grenada
Parece-Vela, Sea of Japan, Shikoku, and Sulu basins basin can be divided into northern and southern parts
reveal that the crustal structure of back-arc basins at about 15.5~ The bathymetry of the northern part
is similar to 'typical' oceanic crust described by is rugged while the southern part of the basin is
Ludwig et al. (1971). The nature of back-arc basin characterized by a near-horizontal, smooth seafloor.
crust, however, is more variable. Six back-arc basins, The nature of the sediments in the Grenada basin
Grenada, Mariana, Okinawa, Parece-Vela, Sea of is not known; however, refraction data indicate that

sediments of the Aves Ridge extend and thicken into Boynton et al. (1979) reports that the Grenada basin
the basin (Westbrook, 1975). crust is approximately 14 km thick near 14~ with
The Aves Swell, an extinct island arc (Bouysse, an average velocity of 4.8 km/s. Descriptions of
1984, 1988), is oriented north-south and dips steeply the velocity structure vary; however, general struc-
into the Venezuela basin to the west. Its eastern edge ture and probable rock types follow: (1) 2.2 and
is arcuate, similar to the Grenada basin, and de- 3.7 km/s m unconsolidated sediments and partially
scends in steps into the Grenada basin. Fox and lithified sediments (Boynton et al., 1979); (2) 5.3
Heezen (1975) report volcanic rocks consisting of km/s m (upper oceanic crust), basaltic flows and
andesites, basalts, dacites and volcanic breccias re- dikes (observed only in the southern part of the
covered from pedestals and scarps of the Aves Ridge. basin (Speed and Westbrook, 1984); (3) 6.2 km/s
They also report Middle Eocene fossiliferous lime- (lower oceanic crust), gabbros; and (4) 7.4 to 7.5
stones, marls and chert recovered from dredge hauls. km/s lower crust to mantle velocity transition
Late Cretaceous to Paleocene granodiorites, diabases zone.
and basalts, dredged from the southern part of the Upper and lower crusts of the Lesser Antilles
Aves Ridge, may be part of the Aves Ridge or the arc are defined as 6.3 km/s and 6.9 km/s layers
northern edge of the South American platform (Fox totaling about 35 km in thickness (Westbrook, 1975;
and Heezen, 1975). Boynton et al., 1979). Rocks of the lower crust may
Like the eastern edge of the Aves Swell and the be basic in composition while the rocks overlying the
Grenada basin, the general shape of the Lesser An- upper crust (3.4-4.5 km/s layer) may be composed
tilles island chain is arcuate. It bifurcates at about of limestones, pyroclasts and sediments (Boynton et
15~ with a maximum separation of about 50 km at al., 1979). Officer et al. (1957) describe the seismic
its northern limit. The outer arc is older (generally structure of the Aves Ridge as similar to that of the
middle and late Paleogene to early Neogene) and Lesser Antilles.
inactive, while the inner arc is younger (generally
Neogene to Quaternary) and presently active (Fox Seismic reflection
and Heezen, 1975). The western limit of the arc
is marked by steep bathymetric gradients into the Neither of the prominent Caribbean reflectors,
Grenada basin. Bouysse (1988) reports two episodes A' or B', can be traced across the Aves Ridge
of volcanism: the first in Late Cretaceous time prior into the Grenada basin; however "... a lower or
to back-arc spreading (84-66.4 Ma), and the sec- middle Miocene horizon can be followed throughout
ond from Eocene to present. Warner (1991) reports the Grenada basin and west to the crest of the
that Early Paleocene to Early Oligocene volcanic Aves Ridge..." (Speed and Westbrook, 1984). Two
rocks have been recovered from the Saba Bank. The outer ridges of the Aves Swell enclose horizontal
episodic nature of volcanism is clear because vol- layers of sediment which are thickest near 13.4~
canic rocks younger than Early Paleocene to Early (Kearey, 1974). Reflection data also suggest that the
Oligocene (Warner, 1991) have not been collected Aves Ridge continues northward with its topographic
from the Aves Swell and volcanic rocks older than expression buried by layers of sediment (Kearey,
Middle Eocene (with one exception) have not been 1974).
collected from the Lesser Antilles (Bouysse, 1988). For the purpose of this work, the most important
This age difference indicates that the Grenada basin aspect of the seismic reflection data is the character
probably formed in early Tertiary time. The duration of sediments overlying deep structures in the basin.
of back-arc spreading was restricted, as found in The rugged bathymetry of the northern part of the
other back-arc basins. Jurassic age basalts on the is- basin is produced by deep structures in several loca-
land of la D6sirade are thought to represent obducted tions. Conversely, reflection horizons of the southern
oceanic crust (Fink, 1968, 1970). part of the basin are smooth and relatively undis-
turbed (Fig. 7). Sediment thickness increases to the
Seismic refraction south from approximately 2 to 9 km.
The thickness of the crust increases beneath the Magnetics

Aves Ridge and Lesser Antilles and decreases be-
neath the Grenada basin (Kearey, 1974; Kearey et Magnetic anomalies over the Grenada basin
al., 1975; Westbrook, 1975). Modeling also suggests (Fig. 9 and Web-Figs. 15.5,6) have been carefully
that the base of the crust generally mirrors the topog- examined by Bird et al. (1993). Anomaly amplitudes
raphy and bathymetry (Boynton et al., 1979; Bird, of hundreds of nanoteslas and wavelengths ranging
1991). Oceanic crustal layers 2 (4.9-5.3 km/s) and from 10 to over 50 km are observed. Anomalies
3 (6.2-6.4 km/s) are present in the Grenada basin over the southern part of the basin display longer
south of about 15~ (Speed and Westbrook, 1984). wavelengths and smaller amplitudes than those over

the northern part. Similarly, shapes and trends of (McCann and Sykes, 1984), or differential motion
anomalies change from north to south. The shape between the North American and South American
of the anomalies over the northern part are typically plates (Bougault et al., 1988), in the late Tertiary
oblong with an east-west trend degrading to patchy has caused the subducting slab(s) to shoal under
and more disorganized farther south. The magnetic the northern part of the overriding Caribbean plate.
anomalies over the Aves Swell are similar to those McCann and Sykes (1984) have further suggested
over the northern part of the basin except that they that the Barracuda and Main Ridges are continuous
are oriented north-south with larger amplitudes. The and have been recently overridden by the relative
magnetic anomalies over the Lesser Antilles range eastward motion of the Caribbean plate. In either
in amplitude from 150 to 600 nT with wavelengths case this shoaling has resulted in a westward shift
from 5 to 40 km. Short-wavelength anomalies (20 to of the center of volcanism beneath the northern part
50 km) clearly delineate the island chain. of the Lesser Antilles island arc. In addition to tec-
tonic compression, this shoaling of the subducting
Gravity slab(s) may have caused sections of the overrid-
ing plate (i.e., northern Aves Ridge, Grenada basin,
Several high-amplitude (approximately 80 to 150 and Lesser Antilles Arc) to be displaced and/or
mGal) free-air gravity anomalies, parallel to and just disrupted. Therefore curvilinear discontinuities are
east of the island arc, are shown in Fig. 4 and Web- interpreted to be related to disruptions within the
Figs. 15.1,2. The wavelengths of these anomalies crust of the basin, which are in turn interpreted to
increase from about 20 km in the south to 50 km be related to the tectonic event responsible for the
in the north. Similarly, several north-south-trending bifurcation of the arc.
free-air gravity anomalies are observed over the
Aves Ridge reflecting bathymetric variations. These Gravity
anomalies display wavelengths and amplitudes of
about 20 km and 50 mGal, respectively. North of High-amplitude free-air gravity anomalies near
the Aves Ridge, the free-air gravity field is subdued, the Lesser Antilles arc are interpreted to be produced
displaying a broad positive northeast gradient (0.6 by a combination of shallow bathymetry as well as
mGal/km) over the area. The average free-air grav- dense volcanic rocks. Hence the contrast between
ity value ranges from about 0 to - 2 0 mGal over the the magmatic arc with the surrounding water and
northern half of the basin, then gradually decreases sediments results in positive gravity anomalies as
to about - 8 0 mGal over the southern half. seen in the results of 2-D modeling. The long-wave-
Airy isostatic reduction was performed by Kearey length low over the southernmost part of the basin
(1974). Isostatic anomalies over the Aves Ridges are is also interpreted to be produced by a combination
generally negative ( - 1 5 mGal) with some positive of effects. Sediment thickness increases to about 10
values to the south. Isostatic anomalies over the kin, causing the crust to warp down into the mantle.
Grenada basin decrease from +10 to - 3 0 mGal, Kearey (1974) reports that negative isostatic anoma-
north to south. Kearey (1974) calculated 50 reGal lies (less than - 3 0 mGal) over the southern part
positive anomalies over the Lesser Antilles. of the Grenada basin may be related to downward
flexing of the crust into the mantle.
Bouguer gravity anomalies were calculated in or-
INTERPRETATION der to compensate for the water bottom interface
and allow for the interpretation of crustal variations.
The northern part of the Aves Ridge (north of Bouguer gravity anomalies were calculated for two
15~ appears to be displaced to the west and may cases: (1) substituting rock with a density of 2.0
be related to the late Tertiary tectonic event which g/cm 3 for the water layer, and (2) substituting rock
resulted in the westward shift of the Lesser Antilles with a density of 2.67 g/cm 3 for the water layer
(Figs. 4 and 5 and Web-Figs. 15.1-4). Therefore, (Fig. 10 and Web-Figs. 15.7,8). In both cases a
the data are inspected for features which support broad high (i.e., positive anomaly) is located over
this hypothesis. Subtle, curvilinear 'discontinuities' the center of the basin, with a north-south elonga-
are interpreted for total intensity magnetic anomaly, tion, from about 13~ to 15~ This high is slightly
free-air gravity anomaly, and bathymetry data sets. displaced to the south as the density used to cal-
Since these data sets are physically different and culate the Bouguer anomalies is increased. A large,
related to physically different rock properties, trends triangular-shaped residual low is produced by the
are not coincident between the data sets. Discontinu- Bouguer calculations over the southernmost part of
ities consist of connected gradients, highs and lows, the Grenada basin and adds confidence to the notion
and/or connected truncations of gradients, highs and that the crust is downwarped into the mantle. In gen-
lows. Subduction of the aseismic Barracuda Ridge eral, the only differences between utilizing a rock

Fig. 9. Total intensity magnetic anomalies over the study area. Heavy solid and dashed lines indicate interpreted curvilinear zones of
disruption (dashed lines indicate reduced confidence). Contour interval is 50 nT.
density of 2.0 or 2.67 g/cm 3 in the Bouguer calcu- is compensating for the effect of the water bottom,
lation are the amplitudes of the resultant anomalies. then cross-cutting northeast-trending discontinuities
Anomaly shapes and wavelengths are essentially also seen in these maps are interpreted to be related
unaffected by the calculations. to deep-seated, crustal features
The large Bouguer gravity high over the center
of the basin is interpreted to be caused by crustal Two-dimensional models
thinning and the formation of the Grenada basin.
This gravity high migrates southward as the rock Two-dimensional models were constructed us-
density which replaces the water layer is increased, ing profile gravity, bathymetry and seismic (both
because as the water column increases, the increased reflection and refraction) data. Basement horizons,
density assigned to it produces increased anomaly interpreted from multiple channel seismic reflection
amplitudes. Small wavelength anomalies over the lines, were converted to depth prior to being incor-
Lesser Antilles and Aves Ridge are relatively un- porated in models. Oceanic crustal layers 2 and 3,
changed by the calculation. If the Bouguer gravity and a probable transition from lower crust to man-

Table 4 of oceanic crustal layers 2 and 3 decreases from

Densities used in 2-D and 3-D forward models about 15 km in the south, to 8 km in the center of
B-C (near 14.5~ and then increases to 15 km to
Layer Density (g/cm 3) the north. In general, the thickness of the transitional
2-D and 3-D layer is modeled to thicken away from the central
Water 1.03 portion of the basin. East-west-oriented models (F-
Sediments 2.30 F' and G-G') reveal that the minimum depth to the
Layer 2 2.57
(basement or upper crust)
base of oceanic crustal layer 3 ranges from 13 to
15 km (Fig. 12). These minima are located at about
2-D only
62.5~ and 61.5~ for models F-F' and G-G',
Layer 3 2.74
(lower crust) respectively. The combined thickness of layers 2 and
Transition 3.05 3 ranges from about 17 km (Aves Ridge) to 7 km
Mantle 3.30 (Grenada basin) to 28 km (Lesser Antilles arc).
For the two-dimensional model F-F', the depth
to the base of layer 3 decreases to about 15 km
tle velocity, were incorporated utilizing refraction at 15.75~ 62.5~ and increases to about 17 km
information. Densities were interpolated from the to the east (Fig. 12A). The minimum combined
Nafe-Drake curve (Ludwig et al., 1971). In order to thickness of combined layers 2 and 3 is 11 km.
maintain consistency throughout the study area, den- Dramatic variations in the combined thickness of
sity variations were not introduced within individual layers 2 and 3 modeled in the eastern part of F-F'
rock layers of the models. Table 4 displays horizons are interpreted to be related to the discontinuities.
and densities used in modeling. Other dramatic variations in the combined thickness
Depth conversion of two-way seismic reflection of layers 2 and 3 are observed in the western part
time is achieved by utilizing a velocity function of F-F'. These variations are interpreted to represent
derived from velocity analyses of seismic line C1904 density variations within the Aves Ridge. In model
(LDGO). This velocity function is developed in a G-G', the shape of the base of layer 3 is asymmetric
two-step process. First, the velocity analyses in the and shallowest (about 14 km) in the eastern part of
basin are converted to depth using the Dix formula the model (14.25~ 61.5~ From here the depth
(Dix, 1955). Time versus depth curves are then increases east and west to about 18 km (Fig. 12B).
overlain and a 'best fit' curve is determined. Second, Overall, the two-dimensional models display
the interpreted velocity curve is modified to satisfy good correlations between calculated gravity and
seismic refraction data. Reversed seismic refraction free-air gravity profiles. Long-wavelength gravity
line 29 (Officer et al., 1957) indicates that the top of anomalies over this area are considered to be
oceanic seismic layer 2 is at about 8 km below sea two-dimensional because of the overall north-south
level near LDGO line 15, cruise RC1904. The lower trough-like shape of the gravity field over the study
limit of the velocity function was set at this point. area. The excellent correlation of gravity profiles
The use of a single velocity function for bas- adds confidence to the overall concept of the basin's
inwide time-to-depth conversions is easily imple- crustal structure. That is, the crust of the Grenada
mented, but there are inherent pitfalls. The sediment basin thins toward the center and thickens under
and water layer thicknesses vary from 0 to 9 km the Aves Ridge and Lesser Antilles island arc. The
and 0 to 3 km, respectively. The effect of an average observed gravity data are consistent with such a
velocity function tends to make shallow horizons model.
too deep, and deep horizons too shallow. A high
degree of accuracy with respect to shallow depth Three-dimensional models
conversions is not critical for the purpose of this
study. Fortunately the velocity analyses from cruise A three-dimensional forward model was con-
RC1904 are tied to seismic refraction velocities, structed to help interpret gridded gravity data. Grid-
hence the accuracy of depth conversions for deep ded bathymetry data and an interpreted depth to
horizons is good (:k: 0.5 km) when considering the acoustic basement surface (Speed and Westbrook,
overall dimensions of the basin (640 by 140 km). 1984) were incorporated into a two-and-one-half
Connecting models (A-B, B-C, and C - D - E ) layer model representing the water column, sed-
over the Grenada basin, oriented north-south and iments, and basement respectively. The depth to
concentric with the Lesser Antilles Arc, reveals that acoustic basement surface was interpreted from two-
the depth to the base of oceanic crustal layer 3 way travel time using many of the multi-channel
(Ludwig et al., 1971) decreases from about 20 to 18 seismic lines utilized in this study (Speed and West-
km in C - D - E and remains nearly constant at about brook, 1984). Fig. 13, and Web-Figs. 15.9,10, show
15 km thereafter (Fig. 11). The combined thickness the depth-to-acoustic basement surface used in 3-D

4~
0

Fig. 10. Bouguer gravity anomalies over the study area calculated by substituting rock density of (A) 2.0 g/cm 3 and (B) 2.67 g / c m 3 for the water layer. Contour interval is 10 mGal.
F 64"W 62"W 60"W

=,.. I "--I ~, I I I
~
q~
% ,,
4)
A o
.z
qp,
(O
-~(-,)')B
14 (!1) J
~
.~ 1 3 (8) \eqp 9
G (15.5) 14 (10)
-----.~) G
.,- ~,) 13(8) 17( l l ...--
H""5) ..5 ,,,, I
C "",, J
r
./= j i ,
Q /
E
__- ,,A I 1 I
P
o loo 200 km
Fig. l l. Locations for 2-D models of the basin. Posted values coincide with depth (km) to the base of layer 3 (Ludwig et al., 1971) and
combined thickness (in parentheses) of both layers 2 and 3 from results of 2-D forward modeling.
modeling. Speed and Westbrook (1984) interpreted short-wavelength anomalies. With respect to long-
this basement surface in time (Web-Fig. 15.11) uti- wavelength anomalies, the calculated gravity field
lizing extensive multiple- and single-channel seismic (Fig. 14A and Web-Fig. 15.12) correlates well with
reflection data sets. This surface was converted to the observed free-air gravity field; however, there are
depth utilizing the velocity function described above. several important differences between the calculated
The calculated gravity field is produced by the and observed fields. The location of the gravity min-
water layer, sediment layer, and the upper surface ima over the Grenada basin is displaced to the north
of layer 2. The observed free-air gravity field is for the calculated field. A north-south-trending, sub-
produced by these same layers as well as layer 3, the tle, broad high is superimposed along the center of
transitional layer, and deeper sources. The observed the basin for the observed free-air gravity. The area
free-air gravity field also reflects density variations of the continental shelf of Venezuela exhibits an
within the sediments and basement which are un- anomaly high in the calculated gravity. The observed
known, as well as variations in crustal thickness. free-air gravity field has high frequency anomalies,
In general, density variations within the sediments particularly over the Aves Ridge and Lesser Antilles,
and shallow parts of the basement should produce which are not observed in the calculated field.

i i i i i ! i i I I l
/ s~ .... .~ 1 1=
GRAVITY (reGal)
- 2 " 1 ~/~'t'/' ~ tr L \ "~qf ~" , ' , ~~
I I I I I i I
F'
, , I lp = 1.O3 Ik =0 I I l .I
= 2.s _ A= 0 ~~___-~
p = 2.74 k = 8000
p = 3.05 k = 0
p=3.3 k=O
O
~t
I , I I I I I I I I ' (A)
100 200
Distance (km)
0
1 I i i I I I l
~'2 ~ GRAVITY (mGal)

, , ~~o
~
-- 1
(.3
t i I I I I I I I I I
G G'
O i I
--. i I I i I I I p=1.03 I k = 0 I I i 9
p =,2.3 k= O r
p = 2.57 k = 8000
p = 2.74 k = 8000
p = 3.05 k = 0
p=3.3 k=O
O _
0 1 O0 200 240
Distance (km)
Fig. 12. (A) 2-D model F - F ' oriented east-west in the northern part of the basin. (B) 2-D model G - G f oriented east-west in the central
part of the basin. Horizontal and vertical scales of the models are 1" 1.5 and 1 "0.75 million (2" 1 vertical exaggeration); 1 = observed
gravity (free-air) profile; 2 -- calculated gravity profile.
To increase confidence in the 3-D model, the gravity (Web-Figs. 15.13,14). This operation isolates
free-air gravity field is filtered (high cut: 60 km) to anomalies which are produced by deep crustal and
remove short-wavelength anomalies, then the calcu- upper mantle variations. Residual gravity anomalies
lated gravity is subtracted from the filtered free-air over the basin are similar to Bouguer gravity anoma-

F 64"W 62"W 60~

yi
I I "l~ I I I
Z
9 B
F_
F
Q
---" ~ I I l
0 100 200 km
Fig. 13. Depth to acoustic basement with heavy lines indicating faults (after Speed and Westbrook, 1984). Contour interval is 500 m.
lies and are defined here as residual Bouguer anoma- structure of the basin is broadly similar from north
lies. A broad high exists over the basin and lows over to south. Although seismic refraction coverage is
the Aves Ridge and Lesser Antilles arc. Also, the tri- sparse in the north, when utilized as control for two-
angular-shaped anomaly over the southernmost part dimensional and three-dimensional modeling, these
of the basin correlates well with Bouguer anomalies. data also indicate similarities from north to south.
The only mantle velocity (8.2 km/s) recorded for
the basin is Lesser Antilles Seismic Project (LASP)
DISCUSSION Profile B (Boynton et al., 1979) and is oriented ap-
proximately east-west near 13~ (Fig. 15). Two-D
A key to understanding the Grenada basin is an model G-G' is crossed by the north-northeast-ori-
understanding of the differences and similarities be- ented reversed Profile 29 (Ewing et al., 1957). There
tween the northern and southern parts of the basin. is no doubt that the crust in this part of the basin
The differences are dramatic for bathymetry, seis- is similar to typical oceanic layering. Average ve-
mic reflection, and magnetic data. However, free-air locities of back-arc basins are 5.1, 6.4, 7.3, and 8.2
and Bouguer gravity data suggest that the crustal km/s for layer 2, layer 3, transition, and mantle,

O

~7
Fig. 14. (A) Calculated gravity anomalies (free-air) over the study area. (B) Filtered free-air minus calculated gravity anomalies = residual Bouguer anomalies. Contour interval is 10 mGal.
~~ .54
.66
2.61
5.44 I~ ~ 1 ~ ! 2"15
~\\\-,= 1.69 1.55 to 3 . 5 0 km/s
3.00
~] 5.95 3.86
l 4.42 5 . 5 0 to 6.45 km/s
..~ 4.95
I 1~
5.86
21
I / 1.54
91.67
~x,,\'N ~ . .
6.24
1.63
1.83 2"37
I _
1.53
2.41
9
3.33
9 4.40 ,N•• 3.00 ~ 2.97
22
I//'//
5.42
5.85
6.41
J4.00
1155 !
t . , ~I 2.421.50
I
3.76 ~\\\\\',,i 1.69 , - - - - - ! 1.60
....... 1.93 1.79
I ] 5.28 = J 2.25 2.47
I~\\\\~1 2.95 ,~\\~1300
I 14.-i ! 15.22
I I
7.41
29 e f
i" ' - i T
7.40
< ii
8.20
LASP B
Fig. 15. Refraction velocities in the Grenada basin. Locations of velocity profiles are shown in Fig. 7B.
respectively (Table 3). Similarly, lower velocities for with typical oceanic crustal layers 2 (5 km/s) and 3
profiles LASP B and 29 are 5.30, 6.20, 7.40, 8.2 (6.4 to 7.1 km/s) described by Ludwig et al. (1971).
km/s and 5.28, 6.22, 7.41 km/s, respectively. Veloc- In their models for the formation of the Grenada
ities 5.28 or 5.30 and 6.22 or 6.30 km/s correlate basin, Bouysse (1988) and Pindell and Barrett (1990)

suggest that the crust of the northern part of the (discussed above) could correlate, but the overlying
basin is composed of stretched and thinned arc layers do not correlate. Similarly, the 4.98 km/s
material rather than accreted oceanic crust. Reversed layer of reversed Profile 22 (Officer et al., 1959) to
refraction Profile 21 (Officer et al., 1959) displays the south-southwest may be a match for the 4.95
lower velocities 5.44, 5.95, and 6.52 km/s. Velocities km/s layer, but velocities above and below do not
5.44 and 6.52 km/s are similar to typical oceanic correlate either. Furthermore, to the south-southeast,
layers 2 and 3. Although the intermediate 5.95 km/s an unreversed profile displays lower velocities 4.4,
layer of this profile could be interpreted to represent 5.42, 5.85, and 6.41 km/s. This area has been sug-
an intermediate velocity from 5.44 to 6.52 km/s, the gested to be the northernmost extent of oceanic crust
location of the profile (over the Aves Ridge) dictates (Bouysse, 1988; Pindell and Barrett, 1990). The 5.42
that the crust is composed of island arc material. and 6.41 km/s layers are probably interpreted to
The only other velocity information for the north- coincide with typical oceanic layers 2 and 3, respec-
ern part of the basin is an unreversed profile near tively, with the thin 5.85 km/s layer representing a
16~ 63~ and oriented approximately east-west possible intermediate velocity.
(Speed and Westbrook, 1984). The lower veloci- Since the morphology of the northern part of the
ties from this profile (Fig. 15a), 4.42, 4.95, and basin is rugged, and the refraction data are sparse
5.86 km/s do not correlate well with other profiles with no clear correlations to surrounding refraction
for the region. The 5.95 km/s layer of Profile 21 data, the lone profile near 16~ 63~ is interpreted
83"W
9 1
l Il I
z
m
~,i'km
(A) (C) ~D
F[_
62"W
(B)
150nTI, 20km ~
Fig. 16. Magnetic anomaly profiles: (A, C) correlated; (B, D) displayed in their acquisition position. Heavy lines in (A) and (B) indicate
correlated anomalies. Dots in (B) and (D) correspond to locations of correlated anomalies in (A) and (B). Dashed lines correspond to
interpreted curvilinear zones of disruption (Fig. 3).

to be less reliable than the rest of the refraction in- mation of the Grenada basin was by near east-west
formation. Therefore, the nature of the crust for the extension. And though the nature of the crust of the
northern part of the basin remains relatively uncer- northern part of the basin cannot be determined con-
tain. Two-dimensional and three-dimensional gravity clusively, a mechanism for which oceanic crust was
models suggest that crustal layers gradually thin to later disrupted by tectonism related to the bifurcation
the north in this region. The Bouguer gravity over of the Lesser Antilles arc is suggested.
the Grenada basin has been calculated four times Strike-slip motion (Speed and Westbrook, 1984)
using two methods in this work as well as twice by along prominent east-northeast faults in the northern
Kearey (1974; Kearey et al., 1975). Anomaly pat- part of the basin may have caused the magnetic
terns produced by all these calculations are similar. signature over this part of the basin to be dis-
The resultant long-wavelength high over the central rupted. McCann and Sykes (1984) identified mag-
part of the basin diminishes gradually north to the netic anomalies having amplitudes of over 400 nT
Saba Bank, and south to the continental slope of produced by fracture zones in oceanic crust, during
Venezuela. This indicates that the thickness of the the Cretaceous quiet period, just north of Hispaniola
crust of the basin also changes gradually over the and Puerto Rico. In order to test this hypothesis,
length of the basin. magnetics profiles were re-interpreted with the spe-
In his discussion regarding the magnetic anoma- cific purpose to identify, and correlate displaced
lies over the Grenada basin, Bouysse (1988) points anomalies, or packages of anomalies. Fig. 16A,C
out that, "... the present great depth to the oceanic displays correlated anomaly profiles over the basin
basement due to sedimentary overloading combined and Fig. 16B,D shows the same profiles plotted in
with a possible location of the Eastern Caribbean in their acquisition position. For Fig. 16A,B the north-
the vicinity of the geomagnetic equator during the Pa- ern six profiles may be displaced about 45 km to the
leocene may contribute to significantly lowering the southwest with respect to the southern five profiles
anomalies' intensity and to blur the original pattern." over this area. The central high (dots) is shown in
Although the geophysical data in the Grenada basin both figures. Fig. 16C,D is a somewhat less convinc-
are complex, our interpretation suggests that the for- ing correlation. For this set of profiles, dots follow
% %
'U
4 ,Q
O
%
q
IAI J cBi j
Fig. 17. East-west reconstruction of the Grenada basin using 2 km isobath: (A) present-day configuration; (B) east-west closure of the
basin.

the correlated prominent low. These anomalies may nisms. Although Brooks et al. (1984) have classified
be displaced about 25 kin. the Sulu and Celebes Sea basins as possibly being
Fig. 17 shows a simple east-west closing of the back-arc basins with the westward dipping Philip-
Grenada basin along the 2 km isobath. Although pine plate related to extension, these basins are
the fit is generally good, the basin does not close probably trapped, small ocean basins (McCabe et
completely in the south and the isobaths overlap in al., 1985, 1986; Lee and McCabe, 1986; McCabe
the north. Knowledge of seafloor spreading geome- and Cole, 1987). The only other back-arc basin in
tries and the amount of extension of the arc prior the world which exhibits extension oblique to the
to spreading would allow a better constrained recon- trench line of the subduction zone is the Andaman
struction. Furthermore, the poor fit in the north is Sea basin. However, there is no counterpart for the
consistent with our interpretation that the island arc Aves Ridge here. That is, a remnant island arc is
has been tectonically modified since the formation not bounded by basins on both sides. If oblique
of the basin. subduction resulted in oblique back-arc extension
There are two important aspects of the north- and formation of the Grenada basin, then it is rea-
south (Pindell and Barrett, 1990) and northeast- sonable that the Aves Ridge would be composed of
southwest (Bouysse, 1988) spreading models sug- oblique-oriented segmented arc sections.
gested for the formation of the Grenada basin. There Fig. 18A shows a trace of the subduction zone
is no real world analogy for these suggested mecha- and ridge/transform sections in the Andaman Sea
'5\ \ \
L
k-,, 1
(A) (B)
Fig. 18. Modified after Mukhopadhyay (1984). (A) Trace of subduction zone and transform/ridge segments in the Andaman Sea back-arc
basin. The large vector (/) represents the relative motion of the Indian plate with respect to the basin. Small arrows and line segments
represent projections of ridge segments toward the subduction zone. (B) Upside-down trace of subduction zone and transform/ridge
segments in the Andaman Sea back-arc basin. The large vector (A) represents the relative motion between the North/South American
plates and the Lesser Antilles subduction zone included for comparison of Andaman Sea and Grenada Basins.

(after Mukhopadhyay, 1984). The large vector (I) in these results with analyses of gravity and seismic
Fig. 18A represents the direction of motion of the data to develop a more coherent model for the evo-
Indian plate with respect to the island arc. Oblique lution of the Grenada basin. This evolution involves
to the arc back spreading is apparent; however, if two major tectonic episodes. The basin is interpreted
one projects individual ridge segments back to the to have formed by near e a s t - w e s t back arc seafloor
subduction zone perpendicular to ridge segments spreading in early Tertiary time. In late Tertiary
(small arrows and connecting segments), then these time, the northern part of the basin was affected
segments are subparallel to the arc for the most by compressional forces causing the Lesser Antilles
part. In Fig. 18B the previous trace of the Andaman island arc to split, displacing a section of the Aves
Sea is turned upside down. The second vector (A) Ridge, and disrupting the crust of Grenada basin,
represents suggested oblique subduction for the At- including its magnetic signature.
lantic plate and the Lesser Antilles (Pindell and
Barrett, 1990). The similarity between the Andaman
Sea and Grenada basins is obvious. Furthermore, REFERENCES
the ridge/transform pattern is similar to the tectonic
model for the evolution of the Grenada basin (Fig. 6) Bandy, W.L. and Hilde, T.W.C., 1983. Structural features of
and the pattern predicted by Dewey (1980) regarding the Bonin Arc: implications for its tectonic history. Tectono-
physics, 99: 331-353.
the mechanism for back-arc spreading.
Barker, E and Hill, I.A., 1981. Back-arc extension in the Scotia
There are at least nineteen basins which have Sea. Philos. Trans. R. Soc. London, A300: 249-262.
formed via back-arc extension in the world today. Bibee, L.D., Shor, G.G. and Lu, R.S., 1980. Inter-arc spreading
All but possibly two of these basins exhibit exten- in the Mariana Trough. Mar. Geol., 35: 183-197.
sion perpendicular to the subduction zone. There- Bird, D.E., 1991. An Integrated Geophysical Interpretation of
Grenada Basin. M.S. Thesis, Univ. of Houston, 316 pp.
fore, with the physiography of the study area in
Bird, D.E., Hall, S.A., Casey, J.E and Millegan, ES., 1993.
mind, near e a s t - w e s t extension for the formation Interpretation of magnetic anomalies over the Grenada Basin.
of the Grenada basin is favored. Second, it is felt Tectonics 12 (5): 1267-1279.
that conditions which cause back-arc spreading are Bougault, H., Dmitriev, L., Schilling, J.G., Sobolev, A., Joran,
primarily a result of the interaction between the sub- J.L. and Needham, H.D., 1988. Mantle heterogeneity from
trace elements: MAR triple junction near 14~ Earth Planet.
ducting slab and the overriding plate. Both north-
Sci. Lett., 88: 27-36.
south (Pindell and Barrett, 1990) and northeast- Bouysse, E, 1984. The Lesser Antilles arc: structure and geody-
southwest (Bouysse, 1988) spreading models for the namic evolution. Init. Rep. DSDP, 78A: 83-103.
formation of the basin rely on coupling along the Bouysse, E, 1988. Opening of the Grenada back-arc basin and
Caribbean/South American plate transform bound- evolution of the Caribbean plate during the Mesozoic and
ary for back-arc extension. In the Southwest Pacific, Early Paleocene. Tectonophysics, 149: 121-143.
Boynton, C.H., Westbrook, G.K., Bott, M.H.E and Long, R.E.,
the Fiji Plateau, South Fiji basin, and Lau basin are 1979. A seismic refraction investigation of crustal structure
back-arc regions which have formed juxtaposed to beneath the Lesser Antilles island arc. Geophys. J. R. Astron.
transform faults. Magnetic lineations over the Fiji Soc., 58: 371-393.
Plateau are oriented subparallel to the subduction Brooks, D.A., Carlson, R.L., Harry, D.L., Melia, EJ., Moore,
zone (Brooks et al., 1984). Extension in the South R.E, Rayhorn, J.E. and Tubb, S.G., 1984. Characteristics of
back-arc regions. Tectonophysics, 102: 1-16.
Fiji basin is from a r i d g e - r i d g e - r i d g e triple junc-
Clark, T.F., Korgen, B.J. and Best, D.M., 1978. Heat flow in the
tion (Weissel, 1981). The spreading ridge of the eastern Caribbean. J. Geophys. Res., 83:5883-5891.
Lau basin is parallel to the Tonga Trench (Taylor Cross, T.A. and Pilger, R.H., Jr., 1982. Controls of subduction
and Karner, 1983) and forms a small transform- geometry, location of magmatic arcs, and tectonics of arc and
ridge-transform triple junction near the bounding back-arc regions. Geol. Soc. Am. Bull., 93: 545-562.
Curray, J.R., Moore, D.G., Lawver, L.A., Emmel, EJ., Raitt,
transform fault to the north. These back-arc regions
R.W., Henry, M. and Kieckhefer, R., 1979. Tectonics of the
have not formed as predicted by the trench transform Andaman Sea and Burma: In: J.S. Watkins, L. Montadert
model. Hence, the coupling along a transform fault and EW. Dickerson (Editors), Geological and Geophysical
may only have a minor influence on the development Investigation of Continental Margins. Am. Assoc. Pet. Geol.,
of back-arc basins. pp. 189-198.
Dewey, J.E, 1980. Episodicity, sequence, and style at convergent
plate boundaries. In: D.W. Strangway (Editor), The Continen-
tal Crust and its Mineral Deposits. Geol. Assoc. Can., Spec.
CONCLUSION Pap., 20: 553-573.
Dix, C.H., 1955. Seismic velocities from surface measurements.
Thorough analyses of magnetic, gravity and seis- Geophysics, 20: 68-86.
mic data support near e a s t - w e s t extension for the Duncan, R.A. and Hargraves, R.B., 1984. Plate tectonic evolu-
tion of the Caribbean region in the mantle reference frame.
formation of the Grenada basin. Interpretation of In: W.E. Bonini, R.B. Hargraves and R. Shagan (Editors),
magnetic data included trend analyses and forward The Caribbean-South American Plate Boundary and Region
modeling (Bird et al., 1993). We have integrated Tectonics. Geol. Soc. Am. Mem., 62: 81-93.

Edgar, N.T., 1968. Seismic Refraction and Reflection in the Geophys. Res., 89:4493-4519.
Caribbean Sea. Ph.D. Thesis, University of Columbia. Mukhopadhyay, M., 1984. Seismotectonics of subduction and
Ewing, J.I., Officer, C.B., Johnson, H.R. and Edwards, R.S., back-arc rifting under the Andaman Sea. Tectonophysics, 108:
1957. Geophysical investigations in the eastern Caribbean: 229-239.
Trinidad Shelf, Tobago Trough, Barbados Ridge, Atlantic Officer, C.B., Ewing, J.I., Edwards, R.S. and Johnson, H.R.,
Ocean. Bull. Geol. Soc. Am., 68: 897-912. 1957. Geophysical investigations in the eastern Caribbean:
Fink, L.K., 1968. Marine Geology of the Guadeloupe Region, Venezuelan Basin, Antilles Island Arc, and Puerto Rico
Lesser Antilles Arc. Ph.D. Dissertation, University of Miami, Trench. Bull. Geol. Soc. Am., 68: 359-378.
FL. Officer, C.B., Ewing, J.I., Hennion, J.E, Harkrider, D.G. and
Fink, L.K., 1970. Field guide to the island of La Desirade with Miller, D.E., 1959. Geophysical investigations in the eastern
notes on the regional history and development of the Lesser Caribbean: summary of 1955 and 1956 cruises. Phys. Chem.
Antilles island arc. Int. Field Inst. Guideb. to the Caribbean Earth, 3: 17-109.
Island-Arc System, Am. Geol. Inst. N.S.P., pp. 287-302. Okuma, S., Nakatsuka, T., Makino, M. and Morijini, R., 1990.
Fox, P.J. and Heezen, B.C., 1975. Geology of the Caribbean Aeromagnetic constraints on the basement structure of the
crust. In: A.E.M. Nairn and EG. Stehli (Editors), The Ocean Okinawa Trough and East China Sea Basin (abstr.). Soc.
Basins and Margins. Plenum Press, London, 3, pp. 421-466. Explor. Geophys. Expanded Abstr. Biogr., 1: 594-597.
Ghosh, N., Hall, S.A. and Casey, J.E, 1984. Seafloor spreading Pindell, J.L. and Barrett, S.E, 1990. Geological evolution of the
magnetic anomalies in the Venezuelan Basin. In: W.E. Bonini, Caribbean region; a plate tectonic perspective. In: G. Dengo
R.B. Hargraves and R. Shagan (Editors), The Caribbean- and J.E. Case (Editors), The Caribbean Region. Geol. Soc. N.
South American Plate Boundary and Regional Tectonics. Geol. Am., H, pp. 405-432
Soc. Am. Mem., 162: 65-80. Pindell, J.L., Cande, S.C., Pitman III, W.C., Rowley, D.B.,
Hayes, D.E., Houtz, R.E., Jarrard, R.D., Mrozowski, C.L. and Dewey, J.E, LaBrecque, J. and Haxby, W., 1988. A plate-kine-
Watanabe, T., 1978. Crustal structure. In: D.E. Hayes (Editor), matic framework for models of Caribbean evolution. Tectono-
A Geophysical Atlas of East and Southeast Asian Seas. Geol. physics, 155: 121-138.
Soc. Am., Map Chart Ser., MC-25. Poehls, K.A., 1978. Intra-arc basins: a kinematic model. Geo-
Hussong, D.M. and Uyeda, S., 1981. Tectonic processes and the phys. Res. Lett., 5: 325-328.
history of the Mariana Arc: a synthesis of the results of deep Ross, M.I. and Scotese, C.R., 1988. A hierarchical tectonic
sea drilling Leg 60. Init. Rep. DSDP, 60: 909-929. model of the Gulf of Mexico and Caribbean region. Tectono-
Karig, D.E., 1971. Origin and development of marginal basins in physics, 155: 139-168.
the western Pacific. J. Geophys. Res., 76: 2542-2561. Sleep, N.H. and Toskoz, M.N., 1971. Evolution of marginal
Kearey, E, 1974. Gravity and seismic reflection investigations basins. Nature, 233: 548-550.
into the crustal structure of the Aves Ridge, eastern Caribbean. Speed, R.C., Westbrook, G.K., et al., 1984. Lesser Antilles
Geophys. J.R. Astron. Soc., 38: 435-448. arc and adjacent terranes. Atlas 10, Ocean Margin Drilling
Kearey, E, Peter, G. and Westbrook, G.K., 1975. Geophysical Program, Regional Atlas Series, Marine Science International,
maps of the eastern Caribbean. J. Geol. Soc. London, 131: Woods Hole, Mass., 27 sheets.
311-321. Tamaki, K., 1985. Two modes of back-arc spreading. Geology,
Lee, C.S. and McCabe, R.J., 1986. The Banda-Celebes-Sulu 13: 475-478.
Basin: a trapped piece of Cretaceous-Eocene oceanic crust? Taylor, B., 1979. Bismark Sea: evolution of a back-arc basin.
Nature, 322:51-54. Geology, 7: 171-174.
Lee, C.S., Shor, G., Bibee, L.D., Lu, R.S. and Hilde, T.W.C., Taylor, B. and Karner, G.D., 1983. On the evolution of marginal
1980. Okinawa Trough, origin of a back-arc basin. Mar. Geol., basins. Rev. Geophys., 21: 1727-1741.
35: 219-241. Tomblin, J.E, 1975. The Lesser Antilles and Aves ridge. In:
Ludwig, W.J., Nafe, J.E. and Drake, C.L., 1971. Seismic refrac- A.E.M. Nairn and EG. Stehli (Editors), The Ocean Basins and
tion. In: A.E. Maxwell (Editor), The Sea, 1. John Wiley, New Margins, Plenum Press, London, 3, pp. 467-500.
York, 4, pp. 53-84. Uyeda, S. and Kanamori, H., 1979. Back-arc opening and the
McCabe, R.J. and Cole, J.T., 1987. Speculations on the Late mode of subduction. J. Geophys. Res., 84: 1049-1061.
Mesozoic and Cenozoic evolution of the southeast Asian mar- Warner, A.J., Jr., 1991. The Cretaceous age sediments of
gin. Trans. 4th Circum-Pacific Energy and Mineral Resources the Saba Bank and their petroleum potential. Trans. 12th
Conf., 4: 375-394. Caribbean Geol. Conf., St. Croix, U.S.V.I., Miami Geological
McCabe, R.J., Lee, C.S. and Hilde, T.W.C., 1985. The Sulu- Society.
Celebes-Banda b a s i n - a trapped piece of oceanic crust Weissel, J.K., 1980. Evolution of the Lau Basin by the growth of
(abstr.). Eos, 66:1078. small plates. In: M. Talwani and W.C. Pittman III (Editors), Is-
McCabe, R.J., Hilde, T.W.C., Cole, J.T., Sager, W. and Lee, land Arcs, Deep Sea Trenches, and Back-arc Basins. Maurice
C.S., 1986. Sulu-Celebes-Banda Basins: a trapped piece of Ewing Ser., Am. Geophys. Union, 1: 429-436.
Cretaceous to Eocene oceanic crust (abstr.). Bull. Am. Assoc. Weissel, J.K., 1981. Magnetic lineations in marginal basins of the
Pet. Geol., 70: 930. west Pacific. Philos. Trans. R. Soc. London, A300: 223-247.
McCann, W.R. and Sykes, L.R., 1984. Subduction of aseismic Westbrook, G.K., 1975. The structure of the crust and upper
ridges beneath the Caribbean Plate: implications for the tec- mantle in the region of Barbados and the Lesser Antilles.
tonic and seismic potential of the northeastern Caribbean. J. Geophys. J.R. Astron. Soc., 43: 201-242.

Chapter 16
Sequence Stratigraphy of the Eastern Venezuelan Basin
J. DI CROCE, A.W. BALLY and E VAIL
The Eastern Venezuelan Basin and its offshore continuation is a Neogene foredeep superimposed on a Mesozoic passive
margin. The basin contains petroleum reserves amounting to some 35,000 million bbls and is the focus of very active exploration.
Four major tectonically controlled Phanerozoic unconformities separate distinct structural regimes, i.e. (1) an ill-defined
Paleozoic/pre-Jurassic pre-rift phase, (2) a Jurassic syn-rift phase, (3) Cretaceous to Oligocene passive margin phase, and (4) and
a Neogene foredeep phase. The pre-Cretaceous unconformity corresponds to the classical breakup unconformity often found on
passive margins. The pre-Neogene unconformity is the basal foredeep unconformity encountered in most foreland basins of the
world.
The Cretaceous to Paleocene passive margin sequence of eastern Venezuela is subdivided into five second-order transgressive-
regressive cycles bounded by a 131 Ma (basal Cretaceous) sequence boundary, four maximum flooding surfaces with the inferred
age of early Aptian (111 Ma), late Albian (98 Ma), middle Cenomanian (95 Ma), middle Turonian (91.5 Ma) and a Late Paleocene
sequence boundary (58.5 Ma). A Late Paleocene to Eocene second-order cycle (58.5 Ma-36 Ma) is followed by the Oligocene
which is subdivided into two third-order cycles bounded by 36 Ma, 30 Ma and 25.5 Ma sequence boundaries. The second-order
sequence boundaries reflect sea-level fluctuations that are superimposed on a thermally subsiding passive margin.
The purely structurally controlled post-Oligocene to pre-Lower Miocene (25.5 Ma) basal foredeep unconformity is associated
with the sudden deepening of the passive margin in response to the incipient emplacement of the ancestral Serranfa del Interior
far to the northwest of its present-day position. The Miocene of the eastern Venezuela foredeep is characterized by aggrading and
prograding deltaic sequences located in the western part of the basin. These sequences consist of three second-order sequences
defined by 25.5 Ma, 16.5 Ma and 10.5 Ma boundaries. Two regional flooding events occurred at 16 Ma and 13.4 Ma. However,
the Serravallian (13.4 Ma) event can only be recognized in the onshore. Following this flooding event subsidence of the foredeep
continued and second- and higher-order sequences formed reflecting sea-level changes and increased sediment supply from the
rising mountain ranges of Venezuela. In the offshore a Late Miocene (5.5 Ma) unconformity is associated with deeply incised
submarine canyons. This deep-water unconformity is not obviously related to a structural event; instead it may reflect a worldwide
Messinian sea-level lowering. The Plio-Pleistocene is characterized by an overall prograding deltaic front in the Orinoco platform
area. About sixteen Neogene third-order sequences are recognized on seismic profiles and are tentatively correlated with Haq et
al.'s (1987) sequence boundaries.
INTRODUCTION 45,000 million bbls of which some 11,000 million

bbls have already been produced.
The onshore Eastern Venezuelan Basin (Fig. 1) is The total area (i.e. onshore and offshore) of the
the second richest hydrocarbon province of South Eastern Venezuelan Basin is about 200,000 km 2.
America after the Maracaibo Basin. More than Elongate and asymmetric, the basin contains up
12,000 wells have been drilled and 35 giant fields to 8-km-thick Tertiary sediments. The Precambrian
and 260 minor fields have been developed during underlying the south flank of the Eastern Venezuelan
90 years of exploration and production (Gonzalez de Basin dips gently northward. To the north of the
Juana et al., 1980). Since 1980 many other fields basin the fold thrust-belts of the Serranfa del Interior
have been discovered. The generalized distribution (Interior Mountain Range), the Gulf of Paria and
of the major fields is shown in Fig. 2. Excluding Trinidad involve sediments that are of the same age
the reserves of the Orinoco tar belt which occupies as the basin fill.
parts of the updip margin of this basin the ultimate The onshore Eastern Venezuelan Basin is sub-
reserves of the Eastern Venezuelan Basin amount to divided into two sub-basins: the Gu~irico sub-basin
Caribbean Basins. Sedimentary Basins of the World, 4 edited by R Mann (Series Editor: K.J. Hsti), pp. 419-476.

420 J. D I C R O C E et al.
Fig. 1. Northern Venezuela, physiographic provinces and sedimentary basins. The dashed outline is the study area bounded to the north
by the front of the Monagas foothills. Abbreviations: A F -- Anaco Fault; B A P = Barbados Accretionary Prism; B F -- Bocon6 Fault; E P F
= E1 Pilar Fault; I F = Icotea Fault; L B F : Los Bajos Fault; S F F = San Francisco Fault; SER. I N T = Serranfa del Interior; O F = Oca
Fault; O D F -- outer deformation front of Monagas foothills" UF -- Urica Fault.
to the west and the Maturfn sub-basin to the east with the sequences proposed by Haq et al. (1987).
(Fig. 1). The boundary that separates the Gu~irico Combining regional tectonics with sequence stratig-
sub-basin from the folded belt of the Serranfa del raphy we will attempt to illustrate the interaction
Interior and Maturfn sub-basin is the Utica fault of structural regimes with eustatic and/or erosional
system, i.e. the buried lateral ramp of the Serranfa events.
del Interior (Roure et al., 1994). Farther south the The conventional stratigraphy of the area has been
Gmirico and Maturfn sub-basins are separated by reviewed in great detail in Gonzalez de Juana et al.
the complex structures associated with the Anaco (1980) and the International Stratigraphical Lexicon
fault system and its associated inversion structures (1956). A more detailed glossary is also contained
(Villaroel, 1993; Bejarano et al., 1996). The offshore in the thesis that forms the basis of this paper (Di
continuation of the Eastern Venezuelan Basin is the Croce, 1995).
offshore Orinoco platform or Plataforma Deltana
which to the south merges with the South Atlantic
passive margin of Guyana. DATA BASE
All data for this study were kindly provided by

P U R P O S E O F T H E STUDY Lagoven S.A. and Corpoven S.A., both affiliates
of Petroleos de Venezuela S.A. (ED.V.S.A.). Some
This study was primarily undertaken to provide an 14,000 km of seismic profiles and more than 150 on-
integrated view of the onshore and offshore subsur- shore and two offshore wells were used. The uniform
face stratigraphy of the Maturfn sub-basin (in short geographical distribution and the overall quality of
the Maturfn Basin!) and its offshore continuation. these data provided an adequate grid to support a re-
The area was selected to illustrate the transition from gional study. Fig. 3a shows the location of eighteen
a pristine Atlantic-type passive margin to a trans- key wells that were selected based on the quality of
pressional foredeep. This transition may be used as a paleontological data, depth of penetration, availabil-
model for the development of earlier transpressional ity of synthetic seismograms, time-converted logs and
foredeeps that developed farther west in Venezuela. regional position. The composite logs of these wells
Another key objective of this study was the applica- include spontaneous potential (SP), gamma-ray (GR),
tion of modem sequence-stratigraphic principles to resistivity (Res) and conductivity curves. Selected pa-
an area where the biostratigraphic control is limited leontological data from these wells will also be sum-
but where the seismic data permit reasonable resolu- marized in this study. The seismic reflection profiles
tion into sequences that at least may be compatible used for this paper were acquired between 1969 and

SEQUENCE STRATIGRAPHY OF THE EASTERN VENEZUELAN BASIN 421
Fig. 2. Major oil fields of the Eastern Venezuelan Basin. Excluding the Orinoco tar belt to the south, ultimate reserves for this basin are
estimated to be in the order of 45,000 million bbls.
1993. Sixty percent of these seismic lines are located Geosyncline' for the Eastern Venezuelan Basin and
offshore and forty percent are onshore. Records range until the late sixties most regional stratigraphic stud-
between 6 and 9 s two-way travel time (twt) and most ies were done in the context of geosynclinal theories.
of the profiles are migrated. During the 30's and 50's, knowledge of the Eastern
Most of the original seismic data in this paper are Venezuelan Basin greatly increased thanks to the
shown in the form of line drawings. The location work of many geologists (e.g. Hedberg, 1937, 1942;
of these line drawings is shown in Fig. 3b. The Hedberg and Pyre, 1944; Gonzalez de Juana, 1947;
drawings were directly derived from the seismic pro- Funkhouser et al., 1948). These authors described
files and represent our subjective interpretation. The the general stratigraphy of the basin and defined
correlations shown have been traced with reasonable most of the sedimentary units and their type lo-
confidence throughout the study area. Line drawings calities on the outcrop or in subsurface. Hedberg
permit legible reproduction and avoid the inclusion (1950) provided a remarkable synthesis, outlining
of bulky seismic profiles in this paper. However, the geologic history of the basin. He distinguished
to further substantiate our interpretations a few se- the present eastern Venezuelan fold and thrust belt
lected segments of the original seismic profiles are from a series of depositional basins which existed
included. Fig. 3c shows the location of figures that throughout the Cretaceous and early Tertiary. Hed-
illustrate these seismic details. berg (1950) further recognized the profound differ-
ence in the depositional history and character of the
sedimentary units which was controlled largely by
A REVIEW OF PAST WORK the migrated position of the axis of the 'Eastern
Venezuela Geosyncline'.
Early exploration and geosynclinal concepts Renz et al. (1958) in an other outstanding contribu-
tion describes the evolution of the Eastern Venezuelan
During the 90-year long history of hydrocarbon Geosyncline beginning with a Late Jurassic or Early
exploration many regional and local studies have Cretaceous orthogeosynclinal phase. The Late Creta-
been carried out in the Eastern Venezuelan Basin and ceous orogeny transformed the basin into a geanticli-
the folded belts of eastern Venezuela and Trinidad. nal welt which forced the migration of the geosyncli-
Most of the published key references are included in nal axis to the south and formed an exogeosyncline or
the bibliography of Gonzalez de Juana et al. (1980) foredeep which appeared during the Early Oligocene.
and the course notebook of Audemard and Lugo The authors also include a series of restored cross-sec-
(1997); however, much of the fine work done by the tions, paleogeographic and isopach maps for different
staff of many oil companies remains unpublished. time intervals.
Liddle (1928), in his classic 'Geology of Ven- Salvador and Stainforth (1968) presented an in-
ezuela and Trinidad' introduced the term 'Orinoco tegrated synthesis of the stratigraphy of the Eastern

422 J. DI C R O C E et al.
t~,O 62 ~ 61 ~ "7 59 ~
MARGARITAI. CARIBBEAN SEA
GULF 3P
EPF OF TRINIDAD
o .,4..._.. 0
MAT~RIN
liT
,,,,
eS
ORINOCO
eB
,\o o
DELTA
eA
~-9~ Ge eF 7-
eE
0 80 km
eD I ~ |
a) 640 60 ~ 59 ~
r~,3o ;2~ 61 ~ 590

MAI:K~RITA I. CARIBBEAN SEA
Fig. gb
GULF
EPF OF
Fig. 11a
TRINIDAD
Fig. 7a
MATURIN
Fig. 25a
DELTA ~TLANTIC
OCEAN __
9 9a Fig. 6 a
80 krn
9 I I I
b) 60 ~ ~ 59 ~
1~ 630 2~ 61 ~ 60 ~ 590
MAI:II~RITA I. CARIBBEAN SEA
11b
GULF
EPF
TRINIDAD Fig.24
MATURIN
9 0
ATLANTIC
OCEAN
ORINOCO
DELTA Fig. 16a & 16b

t ~ R g . 22a & 22b Fig. 18a & 18b
Fig. 20b Fig. 21a Fig. 21b & 21c
.23a o |
I
c) 640 620 59 ~
Fig. 3. Index maps for the Eastern Venezuelan Basin. (a) Base map for well data used in this study. Offshore bathymetric contours in
meters. Point J is a composite surface section for the Serranfa del Interior (Fig. 17). Abbreviations: AF = Anaco Fault; EPF = E1 Pilar
Fault; LBF -- Los Bajos Fault; ODF -- outer deformation front of Monagas foothills; SFF = San Francisco Fault; UF -- Urica Fault.
(b) Map showing the location of all the line drawings presented in this paper. (c) Map displaying the location of the segments of seismic
profiles included in this paper.

Venezuelan Basin and Trinidad based on the plank- trast to western Venezuela from where since the Late
tonic foraminiferal zonation and the interpretation Paleocene (Fig. 4a) transpressional deformation ad-
of surface exposures and wells in Trinidad. The vanced diachronously toward the east along the north-
most important stratigraphic synthesis was done by ern border of South America, in direct response to the
Gonzalez de Juana et al. (1980) who summarized eastward migration of the Caribbean Plate (e.g. Pin-
the stratigraphy of all of Venezuala in great detail. dell and Barrett, 1990; Lugo and Mann, 1995). The
The reader is referred to this publication for the main onset of transpression in western Venezuela was
definition of formation names and their type locali- during early-Middle Eocene (Fig. 4b) but progres-
ties. Arnstein et al. (1985) updated the stratigraphic sively younger transpression affected eastern Ven-
and structural framework of the Eastern Venezuelan ezuela and Trinidad during Late Oligocene-Middle
Basin in a petroleum geologic context. Miocene (Fig. 4b,c) and continued until today.
Other major events recorded during Eocene time
The advent of plate tectonics also include the initiation of the volcanism in the
Lesser Antilles Arc (Pindell and Barrett, 1990). The
Beginning in the midsixties plate tectonics Oligocene is a tectonically quiet period and per-
brought about a basic change in the understand- haps reflects a slowdown of the eastward motion
ing of eastern Venezuela culminating with the fine of the plate and eastern Venezuela survived as a
reconstructions of Pindell and Dewey (1982), Pin- passive margin (Stephan et al., 1990). During early-
dell (1985), Ross and Scotese (1988), Pindell and Middle Miocene to present the Maturfn Basin and
Barrett (1990), Stephan et al. (1990) and Dercourt its offshore continuation finally became defined as
et al. (1993). In a plate tectonic context the Eastern a foredeep in response to southeastward thrusting
Venezuelan Basin formed as the result of the com- of the Serranfa del Interior and foreland loading
plex interaction between the South America, North (Fig. 4c,d). Today the Eastern Venezuelan Basin is
America and Caribbean lithospheric plates. located near the junction of the South American
Three major tectonic stages control the evolution Plate, the Caribbean Plate and the Atlantic Plate.
of the basin. These stages are linked to the post-
Jurassic relative motions of South America with re- Exploration and stratigraphy in the eighties and
spect to North America and to the Tertiary eastward nineties
motion of the Caribbean Plate. The first stage from
Middle Jurassic to Late Cretaceous (165 Ma-80 Ma) The discovery of the giant E1 Furrial fields in
shows a northwest-southeast divergence. The sec- 1986 gave a renewed impetus to exploration in east-
ond stage from Campanian to Eocene (80 Ma-49 ern Venezuela which was followed by the publica-
Ma) has negligible relative motions between South tion of a new generation of papers that focussed on
and North America. However, this stage also marks the hydrocarbon habitat and the stratigraphy of the
the inception of the very important relative eastward foothills of the Serranfa del Interior (i.e. the Mona-
motion of the Caribbean Plate which dominates from gas foothills). Carnevali (1988, 1989) described the
the Eocene to the present (49 Ma-0 Ma). hydrocarbon habitat of the new fields in the context
Based on this plate tectonic scenario the fol- of two tectonic phases, i.e. a passive margin phase
lowing tectonic settings are now differentiated: (1) and a foreland fold-thrust phase (see also Chevalier
the Triassic to Late Jurassic rift phase (breakup of et al., 1995). Erlich and Barrett (1992) outlined the
Pangea) (Bartok, 1993); (2) the passive margin phase petroleum geology of the Eastern Venezuela Foreland
which is Late Jurassic to Late Cretaceous in western Basin from a regional perspective. Their work when
Venezuela and Late Jurassic to Oligocene in eastern combined with detailed stratigraphic data compiled
Venezuela) (e.g. Erikson and Pindell, 1993); (3) the during the past four decades, has helped to constrain
active margin phase (Late Cretaceous to present). and refine models of the geohistory of northeastern
This active margin phase is characterized by an Venezuela and substantial undiscovered hydrocarbon
eastward-shifting flexural foredeep depression (Au- resources may still be found along the present trend of
demard, 1991; Lugo, 1991; Lugo and Mann, 1995; giant fields. Parnaud et al. (1995) and Gallango and
Audemard and Lugo, 1996). Fig. 4 shows the Ter- Parnaud (1995) provide an excellent and up-to-date
tiary evolution of the Venezuelan margin based on overview over the petroleum geology of the East-
the reconstructions of Stephan et al. (1990) and the ern Venezuelan Basin. For a structural study of the
position of these shifting foredeep depocenters. Serranfa del Interior and its foothills (the Monagas
During the Early Paleocene, sea floor spreading foothills) refer to Passalacqua et al. (1995).
between North and South America ceased and the Erikson and Pindell (1993) based on the sub-
Caribbean Plate began to migrate toward the north- sidence analysis of a composite stratigraphic sec-
east. Initially this transpressional collision has no tion of the northern Serranfa del Interior conclude
effect on the eastern margin of Venezuela, in con- that the Cretaceous-Eocene interval is characterized

424 J. DI CROCE et al.
a) LATE PALEOCENE (59 Ma)
j3
c) EARLY MIOCENE(20 Ma) . _ ~ ~ . _.;~ d) PRESENT"

o
YB W
...._
... //~BRK,..x VB ~.d ~ .f. JJBR~ VB [A] ~
" ~" ~. ,, ~,,COC__';MB

.'" SA ll,,
StephanC
~~WICO~3
jc'ACe,
al{19,0) CCC SA rvo "~
Stephan et a ~ ~ . : . . ' J J - ~ .,
Fig. 4. Four stages of the development of the northern and southern Caribbean boundary from Eocene to Present associated with the
eastward migration of the Caribbean Plate modified after Stephan et al., 1990. The diagonal pattern shows the center of the foreland
foredeep depression. (a) During the Paleocene the Caribbean island arc system was located at the northwestern corner of the northern
Venezuela passive margin. The remainder of northern Venezuela was part of a passive/(?)transform margin. (b) During the Middle
Eocene compressional deformation due to the Caribbean-South American convergence affected much of northwestern Venezuela where
a foredeep formed in the Lake Maracaibo area. (c) Continuation of the oblique convergence of Caribbean Plate and South America. The
result is the emplacement of a transpressional folded belt and the associated development of an eastward migrating foredeep. (d) From
Late Pliocene to Present the kinematic change in the Eastern Venezuelan Basin is associated with a decrease of the N W - S E contraction
and increase in the strike-slip deformation. The center of the foredeep moves into the Orinoco platform area. Abbreviations: A C = Andes
Cordillera; AR = A v e s Ridge; BR -- Barbados Ridge; BR = Beata Ridge; CA = Central America; CB =- Colombian Basin; CCC =
Central Colombian Cordillera; CP = Cocos Plate; CT -- Cayman Trough; CU = Cuba; GB = Grenada Basin; GYB = Guyana Basin; HI
-- Hispaniola; LA = Lesser Antilles; MB = Maracaibo Basin; NP -- Nazca Plate; OCC = Occidental Colombian Cordillera; PR = Puerto
Rico; SA = South America; VB -- Venezuelan Basin; YU - Yucatan; YB = Yucatan Basin.
by Atlantic-type passive margin thermal subsidence Upper Miocene (lower-middle La Pica Formation)
of the lithosphere. According to these authors the shelf-slope deposits and the fluvial-deltaic Plio-
Oligocene-Miocene interval reveals a markedly in- Pleistocene (upper La Pica, Mesa and Las Piedras
creased subsidence due to tectonic loading. An un- formations) deposits.
conformity separating the passive margin from the Bejarano et al. (1996) and Crux et al. (1996)
foredeep interval is interpreted as the passage of a emphasize a depocenter switch from the western
peripheral bulge associated with the tectonic load of Late Eocene to Early Miocene Gufirico sub-basin to
the Venezuelan Coastal Ranges. In this paper we will the eastern Maturfn sub-basin around 12.5 Ma, i.e.
present seismic-stratigraphic evidence to add detail considerably later than the obvious basal foredeep
to Erikson and Pindell's (1993) concept. unconformity which will be discussed in detail in
Regarding the stratigraphy of our area Azavache this paper.
et al. (1996) recognize a basal foredeep uncon- Gonzales et al. (1996a,b) also did a very detailed
formity that separates the Lower Miocene (Oficina sequence-stratigraphic study in the Maturfn sub-
Formation) from the underlying Cretaceous. A deep- basin supplemented with chronostratigraphic charts
water stage is represented by the Lower and Middle for parts of the onshore of eastern Venezuala. We
Miocene upper Oficina and Freites formations in the will suggest a somewhat different correlation for
foreland, and the Carapita Formation in the Mona- the late Neogene but perhaps more importantly our
gas foothills to the north. These are overlain by the paper tries to tie the onshore stratigraphy of east-

ern Venezuela with the stratigraphy of the Orinoco separates an Upper Jurassic to Lower Cretaceous
offshore. passive margin megasequence from the overlying
Much less is written about the eastern part of Neogene megasequence.
the area, i.e. the offshore Orinoco Delta. Only two
papers provide most of the information about this
area. Leonard (1983) described the geology and REGIONAL TRANSECTS ACROSS THE EASTERN
hydrocarbon accumulation of the Columbus Basin, VENEZUELAN BASIN
which corresponds to the southeastern Trinidad off-
shore. He described the structure and stratigraphy of I n t r o d u c t i o n and structural setting
the basin during the Neogene and the major factors
controlling hydrocarbon accumulation. In his thesis Although this paper is mainly a sequence-strati-
Prieto (1987) described the subsurface geology of graphic analysis of the Eastern Venezuelan Basin the
the Orinoco platform and, based on some of the regional setting will first be illustrated with line draw-
same data used for this paper, established in some ings of selected regional transects. The structural set-
detail the timing and distribution of growth-fault ting for the regional transects across the Eastern Ven-
provinces in the area. ezuelan Basin is shown in Fig. 5 which shows the
Numerous authors have discussed the stratigraphy end product of an oblique convergence of the rel-
of Trinidad (e.g. Kugler, 1953, 1959; Barr and Saun- atively eastward-moving Caribbean Plate with the
ders, 1965; Carr-Brown and Frampton, 1979; Persad, northwestward-moving South American Plate. The
1979, 1985). Our own study is best compared with Eastern Venezuelan Basin is the foredeep (or fore-
Aden and Bierley (1996) who recognize in Trinidad land basin) directly associated with the transpres-
a Cretaceous to Upper Oligocene deep-water passive sional Coast Ranges of northern Venezuela which are
margin sequence, which correlates with our pas- a product of this collision. The Precambrian to (?) Pa-
sive margin sequence. This sequence is overlain by leozoic basement underlying the foredeep dips gently
an Upper Oligocene-Middle Miocene synorogenic underneath the frontal portions on the foothills of the
deep-water sequence and an Upper Miocene through Serranfa del Interior and their eastern continuation
Pliocene shelfal sequence. In contrast, Babb et al. in southern Trinidad. The Eastern Venezuelan Basin
(1996) single out a Paleogene megasequence that merges to the southeast with the Atlantic passive mar-
Fig. 5. Block diagram of the southern Caribbean showing the major tectonic elements as related to the Eastern Venezuelan Basin.
The Atlantic lithosphere dips to the west and the Caribbean lithosphere dips to the south. The relationship between the two subducted
lithospheric slabs is poorly understood.

NW FRONTAL SE
THRUST LINE 9b B A LINE 7a
a) o FAULT
m ............ [...~. . . . . . . . . . . ;
, z._. ............. . ...... . .................
2 LOWER MIOCENE-PLEISTOCENE 2
A
(,)
(/'}
I.-
4 4
- ::,:,.
51 ~ U N I T V
10 km
i;,,.;i~.~;
_
....
L E G E N D
S TRA TIGRAPH Y LOCATION MAP

41o' ~ r~o\ ~o 6,~o 6~o c~o 5~o - -
A KEY WELL var~A~~ ~ C A R I B B E A N S E A ..__,,,~ / "~ 11~
=--= TOP OF PLIOCENE
A WELL PROJECTED .......... BASE OF PLIOCENE
(~ CROSS POINT WITH = - , T O P OF MIDDLE MIOCENE
' ~ , ~ EPF ~ ~k ~F ~TRINIDAD
THE SECTION ( ~
.... ~-TOP OF LOWER MIOCENE
~" ~'~-~ ~''"-.~__-c'~Z
....i
~-,SB,d-~ SEQUENCE BOUNDARY .10 A 10~
~--TOP OF OLIGOCENE
REFLECTION TERMINATION
~ EVERSFEAULT ....
......
TOP OF PALEOCENE-EOCENE
TOP OF CRETACEOUS
NORMAL FAULT
INTRA-CRETACEOUS
o 9o'
.............. K-3 K-2 K-1
~
j~." j.,/~-." p' ", o ~o ~
SHALE DIAPIR/MUD VOLCANO Jr JURASSIC
l I J
BASEMENT Pz PALEOZOIC
OCEANIC CRUST ..... TOP OF BASEMENT
b) STRATIG RAPHY =
SEQUENCE STRATIGRAPHY
i
AGE PHASE ONSHORE SERRA~N~

OFFSHORE i UNITS Sequence Maximun
FORELAND Floodingi
Boundary Surface ~
NW SE. =
QUAT PLEISTOCENE ~3.16 3.--~- ~'

" I I ]~~MESNLAS PIEDRAS Fms 3.14 3 . - ~ -
PLIOCENE 13_
VIII 3.12 3.-~-5_
i
,3.10
UM UJ 3.9 --
FOREDEEP a VII 3.8 --
3.7 - -
'.~ FREITES LU
LU 3.6 --
MIOCENE MM Fm r'r'~ VI 3.5 -- -- M2 -
O! 3.4 --
OFICINA Fm. ~. LL 3.3 - - - M1--
rr LM \~~MERECURE Fm. n'< I 3.2 - -
V 3.1 - -
SB-3
Oligocene IV 2.3m
' L I L,')~Os j~w 2.2 - - - O1 - -
Eocene ! i ~4~ARATAS Fm.
Paleocene
' ! ~ i / VIDONO Fm. i"r III -- 2.1 _ _ - - P 1 - -
i L/ SAN JUAN Fm. '~
i SAN ANTONIO Fm.
U POST-RIFT ! 7 / QUERECUAL Fm. UJ -- K4 --
~ j l I CHIMANA F m . . - - ~ ' -- K3 --
CRETACEOUS- ' < ~ TEMBLADOR Gr. EL CANTIL Fm. --K2--
i
~ N Q U I N Fm. [~ K1--
SB-2
; I ~ '"
JURASSIC RIFTING ' "lm"l~ RIFT
'-- 1.1 --
PALEOZOIC PRE-RIFTING w
"/Vk/k/Vk/Vk/Vk/k/k/V~ 'v ~Vk~
I i-?- J-: 7 L 1, ~ J w v w v w v w ~ SB-1
PRECAMBRIAN CRYSTALLINE BASEMENT
Fig. 6. The superposition of the eastern Venezuela foredeep sequence on the passive margin of South America. (a) NW-SE regional
transect across the offshore Orinoco platform. The onlap on the basement is the SB-1 surface (shown on the diagram below),
which here is merged with the SB-2 surface. The basal foredeep unconformity, SB-3 in (b), separates the underlying passive margin
sequence from the overlying foredeep sequence. (b) Stratigraphic chart showing general chronostratigraphy, tectonic phases and major
sequence-stratigraphic boundaries of the Eastern Venezuelan Basin. Common onshore formation names (e.g. Gonzalez de Juana et al.,
1980) are shown in their correct position. There probably is a significant hiatus separating the pre-rift Paleozoic from the syn-rift Jurassic.

"OlUoJd oql jo pJ!ql uJoqlJOtl oql uo u~oqs
SlOlgO]jOl atI@9Ot.lA[olpp!lAI pug Jo~oq gu!ddeluo oql ql!~ u!seq oql jo aOlOlglgt]o doopoJoj oql soz!seqdtuo 19OStlgll S - N oq~L (q) "ut.~jetu JlOqs OllgtloqJ133 9.~ozosolA[ poumoJp pug popoJo ug oq o1 poloJdJOltlt, st ltlOttldl139so s~.q~L "SlUOtu!pos
gJe!IJO~L £q poddeluo s! qa!q~ 'dJeaSOOled poaunouoJd e s[ odols igluot~!luoa Joa¢oI Oql gu!~lJopUFl "sa!lst.JolaeJeqa u[gJetu oa[ssgd Jeoi a s~oqs laosugJ1 A%S-~N oqj~ (e) "tuJojleld oaou!J O oJoqsjjo oql ssoJae slaosueJ1 IeUO!gON "L "~!~I
(qz
/ J Cml cj }.s
IN31a3SV8 dO d O l ISUUO OINV3OO ~ 17.
-t,
OlOZO3lVd Zd IN3~3SV8 ~ ~ " IN31AI3SV8 NVIhlSIAIVO31do, -I
DISSVhlNr Jr ONVOIOA UNIAI/UIdVla 3-1VHS I £.
L-)I ~-)I £-)I ................ 1 3 A 3 l INB~IBqqOO3Q -- (]-- ~1~' .
SIqOBOVIBIdO-VEIINI
SNO3OVI3UO dO d O l . . . . . .
•~o~.~o~ # ~-~
3NBOOB-BNBOOBIVd 40 d O l . . . . IliA/IN
NOllVNIV~U31 NOIIOBqJBu ~ I."
3N3009130 30 d O I - - - -
AIdVUNNO8 3 0 N B n 0 3 s "=~SS "-'-~
3N3001V~ U3MO7 dO d O l . . . . . .
(~ NO,IO=tS =IHI A O"
3N3001~ 3qQuI~ 9 0 a 0 1 . . . . H I I M I N l O d SSOEIO kZ) ~t
3NBOOIqd 90 3SV8 .......... o a ± o a r o u a qqBM q6 9 N I 7 eZ 9 N I 7
aNBOOI7d 30 d O l . . . . ]-]am AaN
V I 7 9 (2 0 0 0 N I /:/ 0
AHdVlt~911VI:tl S S70BMIIS
0N9997
(ez

tu~t 01. I
g INqlAIqSV~ NVlhlglAIVOqldd
~.,,, -~.~ -~,,z o s~~~~so~
--~
o.9..
£.
-o6 N 'e" 9 0 0 I ~}.~~//~/% o~
OIZNV71M 'A IINN
oo~ ~, ~:~ ,.-. ~ ' , 4 ~ # ~ -'~ IIA
(3VOINIUI "' IliA l l N f l
"oLL
-..-
~ 063 009~
d VIAl
/
op
N 0 I1
V 9 S
0,~9
VO
N V :7 @ 8 / ht V 0
0 7
o~
......J
vs,~v~
~o~C~7 o~L BN
Ms[. qz 9NI7 e9 9NI7
V I 7 9 (2 0 O 0 N I &l 0
"0! pue 6 "s~!ct u! un~oqs sllnej Ftuaou apls!i aql jo S,~Z!A Z~I!JlS aae (IA l!UCl) auaaO!lA[ zlPP!lAI aq] u!qI!A~ sllne~ z~t!l-dooas zq~L "lJedJmunoa uJa:lsa,~ sl! o] J,el[.tu[.s s[ ]3asueJl EIS-A~N
~aoqsuo uamsea aqj. (q) "sllneJ Ftuaou se pmaJdamu! aq osie pInoa uo!la~s aql jo pua uJ~qlaou aql uo u~oqs sllneJ aSJZAaJ aq&, "aau~nbas m~Jetu zMssed aql jo l!tu!i p.~eA~pUeI lSOtU papoJa aq] s! q3!qA~ a~pzA~ ,Qmumu!p~s snoz3elaa D
u.[q] X.[aA e UO pm!sodap sa!lse[a d~zp~aoj ~Ia!ql sn~oqs laasueal EIS-A~N uJalsa~ aqd, (e) "snoaael&~3 aql pmeJlauad suopaas qloq uo UA~OqSsip,v, aql jo lSOlN "u!se~t u e p n z z u % uamse~I zaoqsuo aql ssoa3e sl3~suea] ieuo!~z~i "8 "~!cI
3ldI/-Inlhl ~ ~" w~! 9 <°"t:
LN31/~35V8 -I0 dO£ ........... zsnklo OINV3OO J~>~:i:[':':
9,
......... -...... : ' ' ~ - .LN=IIAi3SV8 N V I E I S ~ V O 3 E l d
Olsswnr ,r' ONVO7oAanw~lavla a7VHS I £. .s
~ " " ~ ~ ! ! ! r ~ : : .............. oq31HS VNVAf7D
O"
~'* "~ ~'~ ................ ,~^~,~.~,,oo~o--O--
sno3ovI3EIO-VEIZNI I#,
zlnV:l 7VI/~EION
£no3ov13EIO 40 <:101. . . . . .
19r1v--I 39EIBAEIEI ,8
3N3OO3-3N3OO37Vc140 dO1 . . . .
NOIIVNIIAIE131 NOIlOBq=IBEI
3N=1009170 =10 c101- - - - - - -
AWVONn08 30N317039 ~ 88"" g, ~ I I I A J.INA ~__
3N3001~ EI3M07 90 dO1 . . . . . . ( ~ N011038 3H1
~N3OOiV~ ~700i~ ~O aO± . . . . H±IN~±NiOa SSOUO k~
L"
~N~OOna ~o 3sv~ .......... o~±o~r'oua 77~a ~
i
EIN3OOl7d 40 dO1 . . . . 773M ,,k3H iii 0
~-6 3NI"I
MN 3S
AHd'dldt~l.I. ~1:11S $70~1/111S
S "7 - 1 1 H . L 0 0 _ - I - ,N I S V Et Et n s N I IE n J. v 141
ON.~Dg'I
S V t g V N O ~
Z
~ ~ (e8
,9
9- \ ~ ~ ,
11
o. ".....o~ o~ ~ ~ o~ ~ o~
.
~ ' \ m8 "o!d
~' ¢" ~.,,,,~:z--.~.2"~_.~'~..~ ~ . . ~ " ~ ~ ~ ..........~.,,. 1N:I~EISVB NVIEISIAIVO3EId I '¢ ~
/ ~ tY~9 NV~8~IIdVO ~ ~ " "~ ~ ~ ".Z, ........................ : ' " : ' " ~ ' : " ' : ' "
>l~ o~ oO/ op op o,~ ~,o,~'~ o~. 0 W
d Vi/V NOIJ. VO 0-1 e6 3NI"I
MN N I S' V 8 8 n S' N / E/ n .L V /4/ as
gin of South America. On the eastern margin it is also posed of delta and prodelta sediments and associated
shown in Fig. 5 that the oceanic Atlantic lithosphere is lowstand wedges; (6) a dominantly alluvial-deltaic
subducted towards the west underneath the Antillean ('molasse') phase. Three important and structurally
island arc. Using the nomenclature of Bally and Snel- controlled unconformity types separate three differ-
son (1980) the limited northward continental litho- ent structural regimes, i.e. the pre-rift unconformity,
spheric A-subduction of South America merges with the breakup unconformity (both related to the earlier
the westward B-subduction of the oceanic lithosphere passive margin development) and the all - impor-
of the central Atlantic. The southward B-subduction tant basal foredeep unconformity which marks the
of the Caribbean Plate underneath the South Ameri- inception of the foredeep.
can Plate is conjugate to the A-subduction of northern The basal foredeep unconformity was defined
Venezuela. Thus, the transpressional Neogene folded by Bally (1989) as "a complex unconformity un-
belts of eastern Venezuela, Trinidad and the domi- derlying the foredeep clastic wedge (phases 4, 5,
nantly transtensional northern Caribbean offshore are and 6) and overlying the preceding platform se-
formed at a triple junction of three subduction zones. quence (phase 3)". The basal foredeep unconformity
Structural studies of the collisional zone associ- is often characterized by regional updip stratigraphic
ated with this triple subduction junction have been truncation below and by the progressive onlap of
made by numerous authors (e.g. Audemard et al., deep-water sequences, and/or condensed downlap
1985; Rossi, 1985; Lilliu, 1990; Chevalier, 1993; surfaces. This downdip portion of the foredeep se-
Roure et al., 1994; Passalacqua et al., 1995; Cheva- quence represents onlap on a flexed and suddenly
lier and Spano, 1996; Av6 Lallemant, 1997) and deepened former platform top. However, in the up-
several tectonic maps have been published (Bellizia dip portion erosion of the basal unconformity is
et al., 1978; Case et al., 1984; Speed et al., 1984; due to eustasy-modulated exposure of the peripheral
Mascle and Letouzey, 1990). bulge that is associated with flexural basins (e.g.
Beaumont, 1981; Tankard, 1985). In its dynamic
Comments on foredeeps evolution, the downdip onlap surface and the updip
erosional surface are often linked. The many shelfal
The Eastern Venezuelan Basin is a foredeep (or unconformities that separate the alluvial, deltaic and
foreland) basin. The term 'foredeep' or 'foreland prodelta sequences (phases 5 and 6) are third-order
basin' as re-defined by Bally and Snelson (1980) to fourth-order unconformities that respond mainly
corresponds to the class of perisutural basins that are to sea-level changes. These unconformities may fre-
associated with and adjacent to A-subduction zones quently be enhanced when they intercept growing
(i.e. subduction zones where limited amounts of compressional anticlines.
continental lithosphere are subducted). These basins In its simplest form the Neogene foredeep as-
develop on the continental or transitional lithosphere sociated with the East Venezuelan A-subduction
that underlies fold and thrust belts and their adja- boundary is best illustrated by Fig. 6a across the
cent foreland. Subsidence of these flexural basins Orinoco platform where a Lower Cretaceous to
occurs in response to loading and is controlled by Oligocene passive margin section is separated by
the rheology of the underlying lithosphere and rates a regional basal foredeep unconformity from the
of deformation in the adjacent folded belt (e.g. overlying Neogene foredeep sequence. An over
Beaumont, 1981; Flemings and Jordan, 1990; Jor- 7000-m-thick wedge-shaped Neogene section fills
dan, 1995; Miall, 1995). Foreland basins remain the foredeep which unconformably overlies the Cre-
underfilled if the rate of subsidence exceeds the taceous to Paleogene passive margin platform both
rate of sediment supply, a situation that is often onshore and offshore. The Neogene section thickens
realized in the early stages of the foredeep devel- toward the orogenic belt of Trinidad. In our area the
opment. The foredeep stratigraphy is modulated by passive margin stratigraphy is a mixed siliciclastic-
eustasy which particularly affects the upper shallow- carbonate margin, and the Neogene foredeep is en-
water portions of these basins. Bally (1989) showed tirely siliciclastic and characterized by an overall
an idealized cross-section that sums up some of longitudinal progradration from west to the Orinoco
the main characteristics of foreland basins. From Delta. Fig. 6b summarizes the tectonic context of the
bottom to top the following unconformity-bounded deposition of the ' sequence-stratigraphic units that
tectono-stratigraphic units can be differentiated: (1) will be discussed in the main part of this paper.
the top basement unconformity; (2) a syn-rift se-
quence topped by a breakup unconformity; (3) a Regional transects
tilted passive margin sequence topped by a basal
foredeep unconformity; (4) a deep-water ('flysch') In the following we will first discuss two offshore
phase that corresponds to the inception of the fore- transects across the Orinoco platform (Fig. 7) to
deep; (5) a prograding ('transitional') phase corn- contrast them with two onshore north-south tran-

t.aO
,41 o 64 o 62 ~ 61 ~ 60 ~ 59~ _ ~ 11~ W E I
~ ~ GULF - " ,
LOWER MIOCENE __ ,,__ = ~ ,~ --- ---- "_~.".., " ' ~ , ~ ,._~
-- ~ EPF _ . . " . 9 . . -SB-3
t
,.,,, , ,SB- .
9 _
,,,
"~~ ORI~CO / A T L A N T I C E
/ \ ~, o~,~. // ...
. OCEAN 9o.
/ /Fig. 8a
j Fig. On , 8o, km
~ o ~
I
/ / \
I I \ ~'-~
I i |
I \ H
~~ i
..,... ~ w . . . . . . . i=
9".-;'.;"Z~:):i. !:~~:;~,.:: . . . . . . . . . . . .!.." ......... :.:;:~-:i:::";-:', .........
/
/
, ,:,0
i9 , # " (I i~~':~i~~:~~::~~!~':~~:;~:~;~:'~~~-~
I ~"
\ .w

\ ~ ~~.,oo~.~ .......
". ,~ -
9, i.i.'~;:,.,.~..... .......Gii.~!....:~:.:.:,. " 9ii.i.,i..i...:.i~;;-;~.

. . . . . . .
Mmou~ MO:=a~e / "" \ ~ =~

SHELF BREAK / '., ~'"---.-2.4
r
lOkm ~ ' - (a) (b) G)

| ii _ i i ,
Fig. 10. The Mapirito listric normal fault system. (a) Time-structure map showing the location and the general trend of the listric normal fault system in the Mapirito area (redrawn from Daza and Prieto,
1990). (b) Schematic evolution of a west-east cross-section across the Mapirito system showing sequence of faulting and the pronounced truncation of these faults during a major Late Miocene erosional
event. The listric faults are younger in the west and older in the east, which contrasts with the classical development of sedimentary load-induced growth fault systems.
sects (Fig. 8). Finally, a very long east-west transect shown in Fig. 7a. The overlying Miocene section
(Fig. 9) will illustrate the transition from the onshore progrades from west to east. Also, from west to east
Maturfn Basin into the offshore Orinoco platform. several structural features can be seen as follows
The offshore Orinoco transects (Figs. 6a and 7) (1) A major late Neogene reverse fault is related
show a pristine passive margin. The Cretaceous on to the Anaco inversion structure trend which con-
the passive margin is characterized by a substantial sists of a series of elongated northeast-trending and
thickness (ranging from about 2 km updip to over 4 southeast-verging domes. Funkhouser et al. (1948)
km downdip). A thin Paleogene section overlies the and Murany (1972) suggested that the dip of the
Cretaceous. To the east a steep submarine erosional principal fault plane diminishes with depth. The
scarp marks the margin of the Orinoco platform structural history of the Anaco trend was interpreted
(Fig. 7a). The deep-water onlap of Neogene sed- by Banks and Driver (1957), who considered its
iments onto this scarp is reminiscent of seismic origin as a major northwest-dipping syn-sedimentary
profiles across the Blake plateau (e.g. Dillon et al., normal fault that began to form at least in Early
1985, 1988) and the Florida scarp of the Gulf of Oligocene time. According to Villaroel (1993) this
Mexico (Wu et al., 1990). In Fig. 7a Neogene sub- growth fault ceased to be extensional by the end of
sidence appears to be related to the development of the Middle Miocene and inverted into a compres-
the Atlantic passive margin as it is loaded by the sional reverse fault during the Plio-Pleistocene.
thick Neogene sediments of the Orinoco Delta sys- (2) Minor half-grabens of possible Jurassic age.
tem. However, in Figs. 6a and 7b the same Neogene (3) Basement-involved reverse faults that offset
infill is wedge-shaped, expanding to the north and the Lower Cretaceous to Oligocene section. Note
onlapping towards the south onto the basal foredeep that these faults could also be interpreted as normal
unconformity in the offshore southern Trinidad area. faults as already mentioned for Fig. 8.
Two onshore transects in Fig. 8 illustrate the tec- (4) The listric normal faults shown on the onshore
tonic setting of the onshore Maturfn foredeep. The transect (Fig. 9a) are referred to as the Maparito
northern end of both transects corresponds to the faults (Lilliu, 1990; Daza and Prieto, 1990). On
leading edge of the foothills of the Serranfa del In- E - W seismic profiles, the basal detachment of this
terior. The wedge-shaped Neogene directly overlies listric normal fault system steps down from 3.5 s
and onlaps a truncated Cretaceous section that is in the west to 5.0 s in the east and involves mostly
more deeply eroded and thinner than the Cretaceous Lower Miocene sediments (Fig. 9a). The listric
of the Orinoco platform. In Fig. 8 the structure of the faults trend northwest-southeast and dip toward the
southern flank of the basin is characterized mostly northeast.
by southward dipping normal faults often associated The stratigraphic relationship of sequence bound-
with some important oil-bearing structures of the aries associated with this extensional system suggests
area. Typically, these normal faults trend east-west a backstepping listric normal fault system with no
and southwest-northeast (N60 ~ to 70~ and they perceptible growth. Note the pronounced truncation
involve the basement, the overlying Cretaceous and of the beds. Fig. 10 shows the location, structure and
the Neogene sediments. On seismic dip profiles this evolution of this system. The key point here is that the
fault system displays subvertical throws frequently faults are younger in the west and older in the east.
less than 50 m (Audemard et al., 1985). Farther north Thus, in contrast to typical growth fault systems the
in Fig. 8 we show north-dipping reverse faults. The oldest faults are found basinward. This type of struc-
dip of these faults is poorly defined and a normal ture is generated by gravitational collapse or slides
fault interpretation may be equally acceptable. The on a gentle submarine slope that is gravitationally
margin of the folded belt is shown on the north end unstable. The listric faults were formed during the
in Fig. 8a and b. Note that the structural growth of latest Middle Miocene to Late Miocene. Note also
the southern edge of the folded belt is shown by the that the extensional system overlies a complex sys-
updip convergence of Plio-Pleistocene strata on the tem of apparently compressional reverse faults and
southern flank of the first major anticline. Thus, very folds. The faults are difficult to understand. It may be
young deformation marks the outer margin of the reasonably inferred that the deeper reverse faults orig-
folded belt. inally were normal basement-involved normal faults
Fig. 9 is a line drawing of a 800 km longitudinal that were subsequently compressed either during the
set of seismic profiles corresponds to the onshore Middle Miocene or perhaps later.
Maturfn Basin. A gap of about 70 km separates this The offshore section (Fig. 9b) shows from left to
profile from its offshore continuation. This profile right:
ties the offshore Orinoco platform profiles to the (1) The west to east progradation of the Pliocene
Maturfn Basin profiles.The relatively thin Cretaceous depocenters.
section onshore and the poorly defined pinchout of (2) A set of pronounced growth-faults and pos-
the Oligocene contrast with the thick Cretaceous sibly related toe thrusts of the Columbus channel

area (Fig. 11). In the eastern Venezuela offshore the STRATIGRAPHIC OVERVIEW
Neogene depocenter overlies the Cretaceous passive
margin platform and its thin Paleogene cover. Note Our interpretation of the sequence stratigraphy of
the syn-sedimentary rotation of the Neogene beds eastern Venezuela is based mostly on well and seis-
into the fault planes (Fig. 1 l a). We suggest that the mic correlations and will be organized on the base
growth faults may be synchronous with at least one of sequences as defined by well and seismic data.
compressional fold at the toe of the deltaic system Fig. 6b shows the three major tectono-stratigraphic
developed on the slope (Fig. 1 lb). However, lacking regimes that control the stratigraphy of the Eastern
a tighter grid of seismic data to provide a measure of Venezuelan Basin, i.e. a poorly defined rift regime,
three-dimensional control, it is difficult to differenti- a passive margin and a foredeep sequence regime.
ate toe thrusts with a strike that parallels the growth A systematic study of regional seismic profiles cor-
faults from the more complex folds of the Barbados related with well logs permits to identify and date
accretionary wedge. numerous sequence boundaries for a chronostrati-
(3) At the eastern end is an oblique cross section graphic framework. These are also summarized in
(Fig. 9b) across the north-south-trending Barbados Fig. 6b.
accretionary wedge (e.g. Speed et al., 1984; Bouysse A top basement unconformity (SB-1) commonly
and Westercamp, 1989). These folds can be easily underlies the mixed siliciclastic-carbonate passive
mapped as part of that accretionary wedge, but seen margin megasequence. Only in the presence of
on a single section they falsely mimic common toe rather vaguely defined Paleozoic and/or Jurassic
thrusts associated with growth fault systems. The reflectors an intermediate unconformity (SB-2) sep-
updip convergent reflectors of these folds indicate arates the older earlier sediments from the overlying
continuous growth which occasionally is intercepted Cretaceous. The basal foredeep unconformity (SB-3)
by deep-water erosional events, that form unconfor- separates a foredeep siliciclastic from the underlying
mities. mixed siliciclastic-carbonate passive margin pack-
age (Fig. 6b). A line drawing (Fig. 6a) of a profile to
Fig. 11. Schematic illustration of growth-fault related toe-thrusts and folds and thrust related to the Barbados accretionary wedge. (a)
SW-NE close-up of growth faults. For location see Fig. 3b. (b) Schematic model of the offshore Orinoco platform-Barbados accretionary
complex transect. See Fig. 3c for location. Note that the toe-thrust shown in the center of the diagram is not related and probably strikes
at a right angle to the strike of the Barbados accretionary wedge, folds here shown on a very oblique profile. The question mark indicates
in a strike view the area that separates the pristine passive margin from the area underlain by the subducting Atlantic slab.

the south of Trinidad that displays the significance of (1977) and summarized by Gonzalez de Juana et al.
these two major basin-forming sequence boundaries (1980). Feo-Codecido et al. (1984) emphasize that
has been discussed earlier. the extent of the Precambrian basement to the north
of the Maturfn Basin is not known and that one can-
not determine whether undeformed Paleozoic and
THE BASEMENT Paleozoic metamorphic belts shown in Fig. 12 as
underlying much of the Gufirico sub-basin to west
On most seismic profiles the 'top of the basement' would project underneath the Serranfa del Interior.
is the deepest 'strong' reflector which is character- Offshore, to the east, the Precambrian basement
ized by its high amplitude. The top of basement extends eastward underneath the Orinoco platform
is the sequence boundary SB-1. Over much of the and the Atlantic passive margin, where it eventually
area, Cretaceous sediments onlap on this surface and is involved in the transition to Atlantic oceanic crust
record the beginning of the Mesozoic encroachment (Fig. 7a). Onshore the top of the basement is observed
cycle of this area. More than 50 wells in the area dipping to the north. Underneath the frontal folds of
confirm that the basement is the crystalline Pre- the Serranfa del Interior (Fig. 13) the downdip contin-
cambrian of the Guyana Shield, composed mainly uation of the basement is poorly defined on seismic
of meta-sedimentary and meta-igneous rocks in an profiles and mostly deeper than the 6 s (twt) penetra-
amphibolite to granulite facies intruded by gran- tion provided by most seismic profiles (Lilliu, 1990;
ite intrusions (Feo-Codecido et al., 1984; Fig. 12). Chevalier, 1993; Passalacqua et al., 1995).
The reported age for these crystalline rocks ranges Fig. 13a is a time-structure map of the top base-
from 3600 Ma to 800 Ma. Complete descriptions ment and Fig. 13b is a depth-converted basement map
and discussions of these Precambrian rocks have that encompasses a much larger area, including the
been provided by Martfn-Bellizia (1974), Mendoza Guyana offshore. The general trend is characterized
C A R I B B E A N
660 65~
S E A
64~Margarita [ . 63~ 6~~
TobagoI. ll~ ]
EPF
II' II~
%,,,~
9 9 9
"
. 9 j
"lb%
,9 j
9 % 9
9 #" .
"*. 9
ORINOCO
j 1 p , f,=- j .~ j .., 9
% .=,~..~-.. % % 9 9
DELTA
% % % 9 %%% % %% % %
9*'
.9 J J J .# .*" 9 .~ J ,*" J .." ." J ..* ,." .# #" s
9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 -
J S J J #t J S ~ .~ ~ J J i," J J ~" .*" 9 9
9 % 9 9 9 9 9 9 9 9 9 1 4 9 91 49 99 %9 %"
9 9 ,,. 9 9 .r .# ,s 9 9 9 .~ 9 #, .s -
9
99 9 1 9 99 99 9/ 99 9 ,
9 % ~ I I V / /V KI&
-~II/-~I l l 19 "
9 9 9 9 9 9 9 9 9 9 % % 9 9 9 9 ",
~ '-'-'0"... . . . ..., 9 9 9 9 9 9149 9 9 9
/
' 9 ar ar ~ J S aP ar .,1' ~"
9 % % 9 %% 9 %%',,,,
9* ' % 9 1 4 9 1 4 9 1 4 9 9 9 9
9 -" 9 "
9 9 9 9
9 "
9
-" "
.,
9 " SHIELD 8o
Stratioraohic C o l l ) m n LEGEND
CTertiary I ~ (2800 m)
retaceous~ 9 WELL
Eguivale_nt ]~asalticFlow ~ PzMETAMORPHICFy---[ THRUSTFAULT
La QuintaFm. (162 Ma) FOLDEDBELT
(1645 m)
Carboniferous - " (740 m) 0
Sandstones ~ UNDEFoRMED
PALEOZO
CI ~ NORMALT
A
U
L
F
CarrizalFm. ~iiii!iiiii!i'ii!~i!i!!ii!! (1,830 m) [~~ ESPINO ~ OUTERDEFOR
100 km
I
GRABEN MATIONFRONT-
Precambdan % % %
~PRECAMBRIAN ~ PALEOZOIC
Basement % % %
BASEMENT THRUST SYSTEM
Fig. 12. Generalized pre-Mesozoic subcrop map including a stratigraphic column illustrating the presence of Paleozoic and Jurassic rocks
in the Anaco trend. Redrawn from Feo-Codecido et al., 1984. Jurassic half-grabens are not differentiated in any detail within the study
area because their distribution is limited and poorly defined on seismic reflection profiles.

440 J. D I C R O C E et al.
63 ~ 62~
CARIBBEAN SEA T O P OF B A S E M E N T
9
3.0 ='
2.5
2.0
~.0
o..~ '-"'-.', "," ' - " " ' '~ "-" " ' - " ' - ' ",'" ".-: ' - " " ' '" "-" "'-" ".', '," .s \
/ ~ ~~ . : !~ ~ ~ : : ....- ,...
'..'-. -,'~ -.- ,_~,;; ....- '..",'-i-:'
', -.- - . '~ -. - - 6= ~ ', ,. - - '. . . . . - . . . . . 1 o , , -.- - _-
~ .. -- ,. , -."."" 6oo
__ ,~, ", sso \ \ 58o
]~1 Outer margin of I~1 Stdke-slin fault 51 Contour crystalline Contour units sec ( T W T ) }
~v I folded belt I >'~ I " ~ basement I
~ I b~,,I . . . . . . I',',?,'J Exposure Precambrian 8 Contour oceanic Contour interval 0.5 sec ( T W T ) I
r /nrus]: raull: , ,,,- 9 9. . . . . . . . .
64 U 62 o 60 ~ 58 ~ 56 ~
M.,o~ CARIBBEAN /'SEA BARBADOS
ACCRETIONARY T O P OF B A S E M E N T
COMPLEX
TRINIDAD
4 r ,q4,be IIDSDP-1441
OOE4,, v
.;.;.;.;.;.:.:.:.i.;.;.;.;.;.;.;..
.,,,,,
. ,., .,...,,..,v
,.-".,,",, ",'.:~.:,.,.,,-
%.,.,"-#%/%g% ,% ,%/%g%.,.~%g"*/%g%/%r ..% .%
,:,',,.,.,,,..,..
....,, . ,.GUYANA.;,;
VENEZUELA
:.-.-.
0 50 100
I I I
b) Km
"-,',. ] .~-.',.',,',,',',.;.,'.. ~ . . . . SURINAM',',,',','/.,;,' ""~'n-. ". ' ~ 4
Fig. 13. Structure maps on top of basement. (a) Time-structure map of the top of basement derived from regional seismic profiles. For
abbreviations see caption of Fig. 3. (b) Depth to the top basement of Eastern Venezuelan Basin, extended to include the basement of the
Guyana Basin (based on Petroconsultants, 1989). The Soledad-1 well (Point 'S') is the deepest well to reach basement in the study area.
by a broadly arcuate pattern gently dipping north- lantic Ocean suggesting that the eastern Venezuela
northeast with a hinge-zone marked at 5 s (twt) that passive margin may have originated as a transform
turns into a southeasterly direction along the offshore margin as also suggested by George and Sams (1993).
Guyana passive margin. The deepest and northern-
most well to reach basement in the Matuffn Basin
is Soledad-1 or p o i n t ' S ' in Fig. 12 and Fig. 13b. P R E - C R E T A C E O U S S T R A T I G R A P H Y ( P R E - R I F T AND
The well encountered an undated granite at a depth R I F T PHASE)
of 4267 m (Feo-Codecido et al., 1984). Note that
the roughly east-west-trending basement strike to the Onshore a few seismic profiles suggest the pres-
north parallels the fracture zones in the adjacent At- ence of pre-Cretaceous sediments. Occasionally in

the northern Maturfn Basin a strong regional reflec- consolidated, sorted, medium- to coarse-grained cal-
tor can be observed about 6 s below the base of carenaceous orthoquartzite which is composed of
the Cretaceous. Together with other profiles it can rounded quartz, shell debris, nonskeletal granules
be shown that this reflector has an overall north- with minor amounts of glauconite) were dredged
westerly dip. The top of the crystalline Precambrian from a 4400 m deep scarp at the northern edge of
on these sections has been designated as SB-1 and the Demerara plateau (Fox et al., 1970; Gouyet et
the base of the Cretaceous as SB-2 (Di Croce, al., 1992). The possible presence of Jurassic rifts is
1995). Farther east, a few seismic sections both further supported by observations from the Tacutu
onshore and offshore show locally poorly defined rift system of northeastern Brazil which parallels the
diverging reflectors (e.g. western third of regional Espino Graben. According to Eiras and Kinoshita
profile Fig. 9a). It is useful to differentiate the deep (1989) the Tacutu Basin is bottomed by extensive
regional pre-Cretaceous reflector from other more Middle to Upper Jurassic basaltic volcanics (150 Ma
local reflectors that obviously diverge. The more lo- to 180 Ma), overlain by about 1250 m of Juras-
cal diverging reflectors may correspond to possible sic clastics and evaporites and 4900 m of possibly
Jurassic half-grabens and the deeper reflector may Lower Cretaceous clastics. These observations are
represent the base of the lower Paleozoic cover of relevant because they support an opening of the
the Guyana Craton. For convenience, the lower re- Atlantic facing the Orinoco Delta during the Late
flector was designated as SB-1 and the base of the Jurassic (Heezen and Freeman-Lynde, 1976).
more local diverging reflectors was designated as To sum up, we suggest that the basement re-
SB- 1.1. The relationship between SB- 1 and SB- 1.1 flector SB-1 at the western end of the area is
cannot be observed on any single profile and the probably overlain by a thick wedge of lower Pa-
proposed interpretation by Di Croce (1995) is rather leozoic, mostly Cambrian sediments. These could
vague. possibly correlate with the Paleozoic sequences of
Pre-Cretaceous sediments have not been drilled the Bove Basin of southern Senegal (Villeneuve et
in the study area but are known farther to the west. al., 1989; Villeneuve and Komara, 1991) and its
Much of the relevant information was summarized western extension in Florida which on many Pangea
by Feo-Codecido et al. (1984) and was incorporated reconstructions were adjacent to the north of our
in Fig. 12. Accordingly a zone underlain by an un- study area (e.g. Dercourt et al., 1993). Farther east
deformed wedge of lower Paleozoic sediments has in the Eastern Venezuelan Basin, sporadic diverging
been penetrated by a number of wells. According to reflectors suggest the occurrence of limited rifting
Feo-Codecido et al. (1984) the Carrizal-2X well pen- which may be coeval with the rifting events of the
etrated 1827 m of Carrizal clastics. We believe that Espino Graben to the west and the Tacutu Graben of
the above-mentioned northwesterly dipping reflector northeastern Brazil. The sequence boundary SB-I.1
may correspond to the base of the Cambrian Carrizal separates the underlying Precambrian craton from
and Hato Viejo formations. Note that to the north of the overlying 'Jurassic' graben fill. We conclude that
the Espino Graben, the Paleozoic is deformed and a rifting phase preceded the Upper Jurassic opening
slightly metamorphosed and the Apure thrust fault of the Venezuelan Atlantic margin. However, over
separates the deformed Paleozoic deformed belt to most of the area the pre-Cretaceous basement is the
the north from the Guyana Craton and its subsurface Precambrian craton and therefore we refer to the
Paleozoic cover to the south. amalgamated SB-1/SB-2 simply as the pre-Creta-
The Jurassic Espino Graben is filled with an un- ceous basement top.
fossiliferous red-bed section with intercalated basalt
flows (162 Ma, see Feo-Codecido et al., 1984).
These flows are underlain by some Carboniferous THE CRETACEOUS-PALEOGENE (PASSIVE MARGIN
clastics and the Cambrian Carrizal Formation. By PHASE)
analogy to sections farther west, the Jurassic is as-
signed to the 'La Quinta' Formation of the Lake During the Cretaceous and Paleogene siliciclastic
Maracaibo area (Gonzalez de Juana et al., 1980). sequences were deposited along the passive mar-
Although the evidence for extension shown on our gin of Venezuela responding to tectonic subsidence
seismic profiles is far from satisfactory, we speculate and worldwide eustatic sea-level changes. A sea-
that a subdued Jurassic rifting event may affect much ward thickening wedge of sediments represents this
of the Eastern Venezuelan Basin. Large Jurassic rifts tectono-stratigraphic megasequence. The chronos-
have been reported from offshore French Guyana tratigraphic calibration of the offshore seismic pro-
(Gouyet et al., 1992). In offshore Guyana (southeast files is mostly based on offshore wells A and B
of the study area) none of the wells drilled pene- (Fig. 14a). The proposed stratigraphic correlation
trated this sequence. However, Upper Jurassic sam- with the outcrops of the Serranfa del Interior (In-
ples (angular pebble-sized fragments of light-green, terior Range) is shown in Fig. 14b, which also

b~
(b)
SCALE
(a)_ ___+~+~WELL"B" .~I:.,i"A"~
WELL_ Datum: Base Miocene
Carapita Fro. + i ++!i++++++ i +?i+!:++ i+
i ++++
i +++' +::+,i
(m)o (feet)
Naricual Fm. .;.-_-_-.T-_ ,,
,?,
Areo Fm. ~i::!!:i.:!::!:;..:.:~ Oligocene UNIT
300. .1000 Los Jabillos Fm. i:..-:..:.-:..:.;
I L ' , , . ' , , _ - , ,_+r " - " I
--+,~: - ...... _ - _ ~ ~ +_+ . . . . . . . . . -! + - - ' ~ ~ . i-.. Tinaj#as Mb.

~ ~ . . . . . i = : PIlO-Pleistoce~ 600, 92(X)O . ~ ~ ~Fm.~ ~ w e r
V~aleeeeeeee~c~e .
San Juan Fm. Maestrichtian
900, .3000 ~ampani~n
s
~UNIT VIii ~ : -~ Santonian
J
Albian
1200. 94000 San Antonio Fm.
1500. 5000 Querecual Fm.

__Inial . ,
1800- 6000 , , ~ Albian

Chimana Fm.
..~ ~. ~ ~ 2100- 7000 UNIT II Aptian
El Cant# Fm.
2400- 84 Garcia Mb
Aptian
2700- ,9OOO
3000- ,100(X)
Barranquin Fm.
3300- .11000
"++"= _ Z ~ -~= " z ' ~ U N I T I I ~j_ _ _ ' _ :+'-,"+,, ,

3600.
912000 ~ Barremian
3900. .13000 LEGEND
,--, ,,~",,--,'-- ".,-, '~ ~-. ,;y~,,, Sandstone

~ ~ ~ ~ ~ - ~ , ~ ,-,,,,-;'~,,-,y-
! . , . . . . . ~....... :~ .... 4200.
.14000
Shale
LOCATION MAP Limestone
Marg=~ I. 6'3~ 6'2~ 6'1~ 6()0 '"' mfs= 91.5 Ma

911"~,(~ CA~ISaEAN SEA I mfs= 95.75 Ma
mfs= 98.25 Ma

mfs= 111 Ma
SB-3= 25.5 Ma
SB-2.2= 36 Ma
SB-2.1= 58.5 Ma
l,."
I vEN~zuE~ ~
UNIT IV O L I G O C E N E
/ L~
J A. Well Offshore Orinoco Delta 0 100 Km UNIT III ( P A L E O C E N E - E O C E N E )
JJ. Composite Stratigraphic
J Column Interior Range UNIT II (CRETACEOUS)
Fig. 14. Correlations of the Orinoco platform stratigraphy with the stratigraphy of the Serranfa del Interior. (a) Schematic stratigraphic column of offshore wells A and B (tied to a line drawing of Fig. 6a) 0
showing the calibration used in this study. See Fig. 3c for location. (b) Proposed correlation of passive margin stratigraphy of offshore well 'A' with a simplified composite stratigraphic column of the
Serranfa del Interior (Chevalier, 1993).
shows the relation of the formations known from middle Cenomanian (95.75 Ma) and middle Turo-
the Serranfa del Interior to the offshore Orinoco nian (91.5 Ma). From bottom to top the characteristic
platform. Calibration of the seismic profiles permits of each subunit is as follows.
subdivision of the passive margin Cretaceous into (1) The basal subunit II-A (inferred to be 'pre-
three stratigraphic units, i.e. from oldest to youngest: lower Aptian') is a wedge-shaped section that
Unit II (Cretaceous), Unit III (Paleocene-Eocene) pinches out in an updip direction, due to the
and Unit IV (Oligocene) (Fig. 6b). The age attri- progressive onlap on the crystalline basement (the
bution in Ma to the proposed sequence boundaries SB-1/SB-2 or pre-Cretaceous unconformity) of di-
and maximum flooding surfaces is merely an infer- vergent, discontinuous and variable-amplitude seis-
ence because, in fact, the available paleontological mic reflectors (Fig. 16a).
control is not adequate to fully support the assign- (2) The overlying subunit II-B (inferred to be
ment of specific ages. In other words, the proposed 'Aptian-upper Albian') exhibits an aggradational
subdivision is a working hypothesis that needs to stacking pattern with a nearly uniform thickness. The
be amplified by future paleontological studies from high-amplitude seismic reflectors are subparallel,
both old and new wells. continuous and can be traced throughout the area
The description of Unit II (Cretaceous) and its (Fig. 16a).
subdivision will be discussed separately for the off- (3) The following subunit II-C (inferred to be
shore and the onshore areas, respectively. Unit III 'upper Albian to lower Cenomanian') displays an
and Unit IV are mostly present in the offshore area aggradational stacking pattern and pinches out in a
because in the onshore area (i.e. south of the de- downdip direction (Fig. 16a). The seismic reflectors
formation front) a hiatus corresponds to much of are subparallel but discontinuous and have low to
the Paleogene. This hiatus may be due to a combi- medium amplitudes.
nation of nondeposition and erosion preceding the (4) The subunit II-D (inferred to be 'lower Ceno-
deposition of the onlapping Early Miocene foredeep. manian to middle Turonian') is a thin package
corresponding to two or three reflectors (Fig. 16b)
Unit II (Cretaceous) offshore with suggestions of a downlap surface at its base.
(5) The subunit II-E (inferred to be 'middle
On reflection profiles it is observed that the off- Turonian-Senonian') is characterized by a thin pro-
shore stratigraphic Unit II (Cretaceous) onlaps on grading wedge of sediments with a pronounced
the basement and locally on some Jurassic half- sigmoidal stratal pattern that downlaps on the K4
grabens corresponding to sequence boundary SB-2 surface and thins basinward. The low- to medium-
(i.e. the breakup unconformity of some authors) amplitude reflectors are fairly continuous but locally
which merges with sequence boundary SB-1 to form chaotic (Fig. 16b).
a major regional stratigraphic break, i.e. the pre-Cre- Fig. 17a shows the relation of well 'A' to our
taceous unconformity. At its top Unit II is bounded proposed seismic stratigraphic subdivision of the
by sequence boundary SB-2.1 (58.5 Ma or the basal offshore Cretaceous Unit II, which includes the
Late Paleocene). Unit II is an overall seaward- Lower and Upper Cretaceous. Two of the above-
thickening wedge (Fig. 7b) trending north-northeast mentioned seismic-stratigraphic packages, i.e. 'pre-
where its thickness in the best preserved portion lower Aptian' and the 'Aptian to upper Albian', were
reaches almost 3.0 s (about 12,500 m). A time-struc- penetrated by well A where the 'Lower Cretaceous'
ture map of the top of the Cretaceous (Fig. 15a) is consists of a thick unit of multicolored and mottled
characterized by a monotonous seaward-dipping atti- shales with sandstones and silty shale. The biostrati-
tude with widely separated contours to the south (0.0 graphic characteristics of this unit are mainly repre-
to 4.5 s), becoming closer-spaced to the north-north- sented by palynomorphs. The occurrence of species
east (5.0 to 7.0 s), reflecting the gradient change such as Cellassopollis, Ephedrapites, Liliacidites,
from platform to slope. In fact, the 5-s contour ap- Cicatricosisporites, among others, suggests that the
proximates the seismic shelf edge of the Cretaceous Aptian to Albian was penetrated (Furrer, 1979). Pa-
platform (Fig. 7a). lynological and sedimentological analyses indicate
Based on internal stratal configurations a tenta- that these sediments were deposited in a continental
tive subdivision into five distinct packages is made. environment.
These packages amount to transgressive-regressive The Upper Cretaceous subunits II-D and II-E in-
(T/R) cycles bounded by four major continuous ferred to be 'Cenomanian' and 'Turonian-Senonian'
high-amplitude reflectors, labeled K1, K2, K3, K4, consist of thinly bedded light-colored algal lime-
and SB-2.1 (Fig. 16). The K1-K4 reflectors are stone, rich in forams and pelecypods, interbedded
interpreted as downlap surfaces that represent max- with micritic limestone, glauconitic shale and sand-
imum flooding events corresponding to the lower stone. The study of its fossils (Furrer, 1979) has
Aptian (111 Ma), the uppermost Albian (98.5 Ma), given the following ages: (1) Cenomanian-Turonian,

64 ~ 63 ~ 9~
62 e-~o _~\~ eo~ s9o
Margarita.l.,M~ C A R I B B E A N S E A Tobagol.
Cumana
J
,8.0 1
~.
'~ J . / . - ~ ~ ; , , ~ . . . - . ' : ~ . .... : ' I 0 40 80 120 K m r~. -. ..... _-: , ~. % 8~
-., : ' : -; -,-~: :.:. : "64: -~: :, ;:.:. .' ,: - - .'_ ; ,: : : : - ~ o -':' - " ' ""'" '. -'" " 62 . . '~
. .... -. .. '. . ' - , ,-_-
'~ .. .- . .61
. ~ ': :.::,--, ': '" 6o o ': 59 ~ 58o
9 " " II ' " - I ' " - "I I I " II I I I
,9" 9
! .3 .10 9
l~r162 I. C A R IB B E A N S E A Tob~o J
C.ma,,a ~ TRINIDAD ~ i
9 ) La Crm
~ 2.. 9
_ ~ o o .o . so . l,oo~,.
F ~ Outer margin of I ~ Strike-slip fault ~ 21 Contour I

folded belt
i
i j~si Thrust fault ~ Exposure Precambdan
crystalline basement
Contour units sec (TWT)
Contour interval 0.5 sec (TWT)
I
Fig. 15. Structure map and isopachs of the Eastern Venezuelan Basin. (a) Time-structure map of the top of Cretaceous. (b) Time-isopach
map of the Lower Miocene (Unit V). Note the isopach maximum to the west (1.25 s). (c) Time-isopach map of the Pliocene (Unit VIII).
Compared with B the depocenter migrated eastward with the highest isopach values (5.0 s) located to the southwest of Trinidad.

Fig. 16. Seismic expression of Cretaceous sequences (see Fig. 3c for location). (a) Uninterpreted and interpreted dip-oriented segment of
a seismic profile showing the subdivision of Unit II (Cretaceous). (b) Uninterpreted and interpreted dip-oriented segment of a seismic
profile (see Fig. 3c for location) showing details of the forestepping prograding pattern of the Upper Cretaceous Unit II-E. The downlap
surface (K4) marks the inception of the overall regression from the Upper Cretaceous to the Tertiary.
characterized by the presence of lnoceramus, Ner- of Globotruncana, Pseudotextularia, Heteroelix and

inea sp., miliolids and Acicularia; (2) Campanian, Rugoglobigerina. Based on sedimentological and
characterized by the presence of Orthokarstenia paleontological analyses, the thin carbonate-shale
bramletti cretacea, Orthokarstenia cretacea and facies of the Cenomanian, the Turonian, and the
Orthokarstenia parva, Gaudryina palmula, and Senonian is interpreted to be deposited on a sta-
Globotruncana fornicata; (3) Maastrichtian, char- ble shallow-water platform with sequence bound-
acterized by the presence of a planktonic assemblage aries which based on the limited paleodata are

Fig. 17. Regional Cretaceous and Paleogene correlation from Guyana to the Eastern Venezuelan Basin. (a) Passive margin stratigraphy
based on correlation of selected key wells across the Guyana Basin, the Eastern Venezuelan Basin and outcrop sections of the Serranfa del
Interior. Stratigraphic columns simplified after Petroconsultants (1989), unpublished Lagoven reports and Chevalier (1993). (b) Onshore
Eastern Venezuelan Basin: stratigraphic correlation of Unit II (Cretaceous) penetrated by selected key wells.
compatible with the Haq et al. (1987) sea-level seismic shelf margin observed on the northeastern
curve. segment of Fig. 7a consists of carbonates as sug-
Wells A and B penetrated a dominantly siliciclas- gested by the overall similarity of the seismic image
tic Cretaceous that was derived from the Guyana with other eroded carbonate margins from the Blake
Shield. Only a few carbonate layers were encoun- Plateau (Dillon et al., 1985, 1988) or west Florida
tered. It is not demonstrated that the eroded steep (Wu et al., 1990).

Unit II (Cretaceous) onshore silty shale of Barremian to Aptian age was recovered
by piston core and the oldest sedimentary unit so far
Onshore Unit II consists of a northward-thicken- (Upper Jurassic) (Fox et al., 1970) was dredged near
ing wedge which rests unconformably and onlaps DSDP Site 144.
the Precambrian crystalline basement (Fig. 8a). At The Upper Cretaceous of the Guyana Basin con-
its base, Unit II is underlain by SB-1/SB-2 pre-Cre- sists of Cenomanian deep-water shales and marls
taceous unconformity. At its top Unit II is bounded and of Turonian-Maastrichtian platform-type clas-
by SB-3 which corresponds to a major regional un- tics with carbonates at the periphery of the basin and
conformity, which has already been mentioned as the deep water clastics with minor carbonates beyond
'basal foredeep unconformity'. Unit II also includes that margin (Petroconsultants, 1989).
a couple of strong, continuous and parallel reflec- By far the most complete Cretaceous section of
tors. These reflectors are probably due to an acoustic the former passive margin is outcropping in the
impedance contrast of clastics and carbonates. In Serranfa del Interior (e.g. Rossi et al., 1985; Vi-
general, Unit II is uniformly thick along strike but vas, 1986; Chevalier, 1993; Erikson and Pindell,
onlaps the basement on dip profiles. The sequence is 1993), where the lowermost subdivision of the Bar-
truncated at the top. Based on well information Unit ranquin Formation and its four cycles cannot be
II is equivalent to the Aptian-Santonian Temblador easily correlated with the subsurface data of the
Group (well 'H' in Fig. 17b). Unit II reaches 600 m Eastern Venezuelan Basin. However, a comparison
to the north. The same well also permits subdivision of the outcrop section with wells A and B from
of this unit into two subunits (Fig. 17b). The lower the offshore (Fig. 14b) supports the plausibility of
subunit correlates with the middle Aptian to Albian the inferred assignment of the offshore downlap sur-
Canoa Formation which consists of mottled coarse- faces, i.e. K1 (111 Ma) is equivalent to the Aptian
grained sandstones interbedded with siltstones that Garcfa Member, K2 (98.5 Ma) to the upper Albian
were deposited in a continental environment. The Chimana Formation, K3 and K4 (95.75 Ma and
upper subunit correlates with the Cenomanian to 91.5 Ma, respectively) to the Cenomanian-Turonian
Campanian Tigre Formation. This upper subunit is Querecual Formation.
composed of two lithofacies. A basal lithofacies con- In conclusion, the Cretaceous is onlapping on
sists of sandstones interbedded with shales. The up- a crystalline basement over most of the area and
per lithofacies consists of dolomitic limestones and only locally overlies the inferred Jurassic rift Unit I.
glauconitic shales with Exogira, Lingula, Plicatula The Cretaceous Unit II is best subdivided into the
sp., Astarte sp. (Gonzalez de Juana et al., 1980). Sed- following five transgressive-regressive (T-S) cycles
imentological and biostratigraphic analyses indicate (Figs. 14a and 17a).
that this subunit was deposited in an environment (1) A wedge-like clastic pre-lower Aptian II-A
ranging from lagoonal/near shore/marginal marine cycle (SB-2 to K1, i.e. from ?132 to 111 Ma) corre-
at its base to the outer-shelf towards the top. sponds roughly to the siliciclastic upper Barranquin
section of the Serranfa del Interior.
Unit II: onshore-offshore correlation (2) A clastic inferred 'Aptian-upper Albian' II-B
cycle (K1 to K2, i.e. 111-98.25 Ma) with a basal
A correlation between the onshore and the off- condensed downlap sequence may be equivalent to
shore Cretaceous of the Eastern Venezuelan Basin the Garcfa Member of the Serranfa del Interior.
with the stratigraphy of the Serranfa del Interior and In the Serranfa these clastics are replaced by the
the wells of the Guyana offshore is illustrated in carbonates of the E1 Cantil Formation.
Fig. 17. Based on seismic correlation and age, on- (3) A clastic inferred 'upper Albian to lower
shore Unit II appears to be equivalent to most of the Cenomanian' II-C cycle (K2 to K3, i.e. 98.5-95.75
offshore Unit II. Onshore, the top of Unit II with its Ma) with a basal condensed downlap sequence is
omission of documented Maastrichtian may be more equivalent to the Chimana Formation of the Ser-
eroded. ranfa del Interior. For a similar correlation see also
Toward the southeast, the continuation of the Galea-Alvarez et al. (1996).
study area is the Guyana Basin, a portion of the pas- (4) A thin carbonate shale 'lower Cenomanian
sive margin basins of northeast and east South Amer- to middle Turonian' II-D cycle (K3-K4, i.e. 95.75-
ica (Jankowsky and Schlapak, 1983; Veeken, 1983). 91.5 Ma) with an ill-defined downlap surface at its
The Guyana offshore basin shows stratigraphic char- base is equivalent to the lower Querecual Formation
acteristics similar to the Orinoco platform. Thus, the of the Serranfa del Interior. The Turonian marine
Lower Cretaceous unit consists of mostly Aptian- flooding surface is also documented paleontologi-
Albian shallow-water clastics and carbonates. How- cally in Trinidad by Huang (1996) who correlates
ever, in the DSDP Site 144 on the northern edge of the surface with a worldwide Turonian marine flood-
the Demerara Plateau (east of the Guyana Basin) a ing event.

(5) A thin wedge of prograding clastics and to reworked glauconitic limestone. The faunas again
carbonates of the 'middle Turonian-Senonian' II-E (Furrer, 1979) permit subdivision of this subunit into
cycle (K4-SB-2.1, i.e. 91.5 to ?66.5 Ma) with a a Lower Eocene unit, based on the occurrence of
pronounced downlap surface at its base is probably Globorotalia formosa, Globorotalia broennimanni
equivalent to the combined upper Querecual, San and Globorotalia palmerae, a Middle Eocene unit,
Antonio and San Juan formations of the Serranfa based on the occurrence of Globorotalia spinuloin-
del Interior. Significant parts of the II-E cycle may flata, Globorotalia centralis and Truncatulinoides
be eroded in much of the onshore of the Eastern rohri, and Upper Eocene unit (only in the well B)
Venezuelan Basin. Note that much of the Cretaceous based on the occurrence of Globorotalia cerroazu-
in the Maturfn sub-basin and of the Orinoco shelf is lensis cerroazulensis (Furrer, 1979). The same au-
dominated by shallow-water deposits, while the cor- thor interpreted an Oligocene unit in well A resting
responding deeper-water lowstand sediments appear on top of the Middle Eocene. However, the very poor
on outcrops and in wells in the Southern Basin of state of preservation of the planktonic foraminifera
Trinidad (Sprague et al., 1996). did not permit a clear definition of the age of this in-
terval. Sedimentological analysis and the occurrence
Unit III (Late Paleocene-Eocene) offshore and of benthic foraminifera, such as Cyclammina, sug-
onshore gest a bathyal (500 m) depositional environment for
these sediments. The presence of reworked limestone
Unit III is a thin and condensed section defined that includes shallow-water forms such as Lepido-
by two strong reflectors with onlap on the lower cyclina, Nummulites and Discocyclina, suggests that
sequence boundary SB-2.1 (58.5 Ma) (Fig. 18a). these sediments were transported and reworked in
The top is the SB-2.2 downlap surface. Offshore, the a deep marine setting, possibly by gravity flows.
thickness of this sequence is nearly constant (i.e. less Note that onshore w e l l ' S ' reports Middle to Upper
than 100 ms) and can be followed throughout the Eocene pelagic faunas from a sandstone-shale se-
offshore area. On the other hand, onshore Unit III quence. This suggests that subunit III-B extends well
is mostly eroded or never was deposited. However, into the onshore area.
only in onshore w e l l ' S ' over 450 m of Middle and Fig. 17a illustrates the regional correlation of
Upper Eocene pelagic shales and sandstones have the Paleocene and the Eocene. To sum up, in the
been reported. Unit III can be subdivided into two offshore the Late Paleocene overlies the SB-2.1.
depositional subunits. Much of the Paleocene, the Danian and in places
The data from offshore wells A and B (Fig. 14a) the Maastrichtian appears to be missing. A similar
show that a lower subunit III-A (90 m thick) consists unconformity in the Guyana offshore is less well
of two distinctive lithofacies. A basal lithofacies constrained (Petroconsultants, 1989). The lower Unit
is characterized by shallow-water limestone with al- III-A corresponds to a middle Thanetian sequence,
gae, ostracods and abundant planktonic foraminifera. presumably sandwiched between the 58.5 Ma and
Based on the occurrence of Globorotalia pseudo- the 56.5 Ma or 54.2 Ma sequence boundary of Haq
menardii, this basal lithofacies appears to be Late et al. (1987). The upper subunit III-B includes most
Paleocene (early Thanetian) in age. The upper litho- of the Eocene, with pelagic faunas of the Upper,
facies consists mostly of silty shale and dark shale Middle and Lower Eocene. Specific correlations
with abundant benthic and planktonic foraminifera. with the Guyana offshore are difficult to make, but
The assemblage includes Bathysiphon, Glomospira, it appears reasonable that all units reported from
Cyclammina, Globorotalia aegua, Globorotalia an- the Venezuelan offshore area are also present in the
gulata, Globorotalia acuta and Globorotalia velas- Guyana offshore.
coensis. These paleontological data (Furrer, 1979) The offshore 'Upper Paleocene' subunit III-A
suggest deep (bathyal) water conditions for this should be correlated with the top of the Vidofio
lithofacies which is dated as Upper Paleocene Formation of the Serranfa del Interior, because this
(Thanetian). There is a pronounced hiatus sepa- formation shows a very similar paleontological as-
rating Unit III from the underlying Unit II. This semblage and lithological features. Subunit III-B
hiatus omits the Lower Paleocene and locally the appears to be equivalent to the Caratas Formation
Maastrichtian. Thus, the base of Unit III could be of the Serranfa, but the lithofacies of these units
considered to be equivalent to the 58.5 Ma sequence appears to be somewhat different. Note also that in
boundary of Haq et al. (1987) and the base of the Serranfa (stratigraphic column J, Fig. 14b) a sig-
the upper facies could correspond to the 56.5 Ma nificant unconformity causes the omission of much
maximum flooding event of these authors. of the Upper Eocene which is reported in well A
The upper subunit III-B (115 m thick) consists of (Fig. 14a). Subunit III-B is also represented in the
glauconitic shale interbedded with thin-beds of silty- Eastern Venezuelan Basin and has been penetrated
shale and fine-grained sandstone, that grades upward by well 'S'.

Fig. 18. Seismic expression of Paleogene sequences (see Fig. 3c for location of this section which is very close to well A shown
in Fig. 14). (a) Uninterpreted and interpreted dip-oriented segment of seismic profile showing Unit III (Paleocene-Eocene). (b)
Uninterpreted and interpreted dip-oriented segment of seismic profile showing Unit IV (Oligocene).
Unit IV (Oligocene) offshore and onshore sequence IV is correlated with a regional unconfor-
mity described as the 'basal foredeep unconformity'
Offshore Unit IV is a sedimentary wedge down- that separates the overlying foredeep tectono-strati-
lapping onto sequence boundary SB-2.2 (36 Ma of graphic unit from the underlying passive margin unit.
Haq et al., 1987). The top of Unit IV is the sequence In the offshore area Unit IV occurs mostly to the
boundary SB-3 (probably 25.5 Ma corresponding to south and no Oligocene has been reported in other
the lowermost Miocene) which is characterized by wells of the offshore. Therefore, it is suggested that
truncated reflectors and overlain by onlapping re- in the downdip offshore the section is eroded or never
flectors (Figs. 6a and 18b). This upper boundary of was deposited. In the onshore area except for well

S no Oligocene fossils have been reported from the the Serranfa del Interior with the Venezuela offshore.
Maturfn Basin, but significant siliciclastic sections are Fig. 17a ties the Venezuelan onshore-offshore to
reported from the oil fields of the folded belt. the Guyana Basin. It is important to note that in
Offshore Unit IV can be subdivided into two Fig. 14a the faunally documented Oligocene in well
subunits separated by a high-amplitude reflector, A is absent in well B, suggesting a significant hiatus,
represented by sequence boundary SB-2.3 (inferred as already suggested by Prieto (1987). An alternative
to be 30 Ma by Haq et al., 1987). The lower subunit is that the Oligocene downdip of the prograda-
IV-A consists of subtle sigmoid seismic reflection tional wedge shown on the seismic (Fig. 18b) is
patterns that pinch out in a downdip direction. The so condensed that its detection was not possible in
upper subunit IV-B displays a progradational config- well B. Onshore well 'S', which did encounter the
uration and shows a pronounced downlap geometry Oligocene, places the zero edge of the formation to
(Fig. 18b). the south of the well. At this time it is not known
Based on well 'A' (Fig. 14a) Unit IV (max. 200 whether the Oligocene has been encountered in other
m) consists of two lithofacies. A basal lithofacies wells that were recently drilled farther north in the
is composed of glauconitic shale with abundant basin.
fossils (gastropods and foraminifers) that grade up The twofold division observed in offshore well
into skeletal and glauconitic limestone (stromato- A and on adjacent seismic lines can be reasonably
poroids and foraminifers) interbedded with fine- to correlated with wells in the E1 Furrial-Carito trend
medium-grained calcareous white sandstone. This and from there with a section in the Serranfa del
lower lithofacies is considered to be Early Oligocene Interior (Fig. 17a). Many authors have accepted
(Rupelian) in age based on the occurrence of Bu- a threefold subdivision in the Serranfa, i.e. from
limina sculptiles and poorly preserved specimens of bottom to top Los Jabillos clastics, the Areo shales
Globigerina euapertura (Furrer, 1979). The upper and the Naricual clastics (e.g. Gonzalez de Juana et
lithofacies consists of a coarsening-upward sequence al., 1980; Sams, 1995). Conventionally, these three
of poorly consolidated medium- to coarse-grained formations were all included in the Merecure Group
white sandstone and appears to be Late Oligocene (e.g. Gonzalez de Juana et al., 1980), which has its
(Chattian) in age, based on the occurrence of Sipho- type locality in the Santa Ana field of the Anaco
generina senni, Cassigerinella chipolensis and pos- trend. Because no fossils are reported from the
sible specimens of Globigerina ciperoensis (Furrer, Merecure type section, the overall correlation with
1979). In addition, the sandiest portion contained the sections of the Serranfa may be debatable.
the macrofossils of Heterostegina antillea, Lepi- In conclusion, it appears that the Oligocene fau-
docyclina and Nummulites. These sedimentological nas reported from wells A and S, the faunas reported
and paleontological analyses indicate that both units from E1 Furrial area, and the faunas reported from
were deposited in shallow-water conditions closely the Areo Formation are all roughly correlatable.
associated with a carbonate platform. Additional paleontological details from the Orocual
Based on correlation of the seismic facies and field were reported by Giffuni and Castro-Mora
sedimentological characteristics, Unit IV is subdi- (1996). Based on offshore data, the Oligocene sec-
vided into: (1) a lower subunit IV-A (equivalent to ond-order cycle may be split into two third-order
Lower Oligocene) that consists of very condensed cycles bounded respectively by SB-2.2 (36 Ma),
transgressive and highstand deposits which contain SB-2.3 (30 Ma) and SB-3 (25.5 Ma). Throughout
the maximum flooding surface 35 Ma and over- the area in the updip direction and toward the craton,
lie sequence boundary SB-2.2 (36 Ma); these two the Oligocene is absent due to Lower Miocene to
surfaces merge updip in a single seismic reflector; Upper Miocene erosion. Downdip of the prograda-
(2) an upper subunit IV-B (equivalent to the Up- tional wedge shown near well A, the Oligocene may
per Oligocene) bounded at its base by sequence be either absent because of sediment starvation or so
boundary SB-2.3 (30 Ma). Seismic data show that a thin and condensed that the interval escaped detec-
lowstand deposit (mostly sandstone) associated with tion in well B. The paleogeography of the Oligocene
this sequence boundary pinches out updip near the will be discussed below.
relict offlap break of the previous highstand deposit
and becomes thinner basinward (Fig. 18b). The only
onshore record of the Oligocene of the Eastern Ven- THE NEOGENE (FOREDEEP PHASE)
ezuelan Basin is reported from well 'S', where some
325 m of sandstone and pelagic shales are reported From uppermost Oligocene to Early Miocene a
to contain Oligocene faunas including Globigerina major change of tectonic subsidence regimes oc-
ciperoensis. curred in the Eastern Venezuelan Basin. In the
Fig. 14b shows the correlation of Oligocene Unit context of the oblique convergence between the
IV between the composite stratigraphic column of Caribbean and South American plates (see Fig. 4)

O0
0
.,.-, .. ©
0
,=.- __
0
t~
~ 0
I I r.~ I I I C I eq ;~
m I.~:.~ .~~
~1 .~
~-~
.~
~~-~ ~.~
.~ ~!
~_~ 0
.~-~
A H V I 1 ~1 3 I
I ~-~ E'~ ~ ~-~ ~1 ~ ~ I I ua
0 C~ ~_ ~ ~
~~ .~.-_ ~: ~ .~ ~ o~ 1 o.~ O
-
c~ ~ o I I~~ I N ~
r2 .~ ft. rO rO ~'~-~ "~, auaao~,s!ald-O!ld ~_ ~o = ~~ 0 Z
I I ~ I I I Zl ~ - g
~:¢ .~o, ~y #
© ~-~
:12 o :~7aa,~,
q~ ~ :. ! ::~.::i!i:i~!!.17:!!!i
i !.!:::/:i:~:/:::.!:.!<i.i::i::i:.::i~:i.~i~:i{!~:~.~:~.~%..~ih~!:~i!~:.~~:!ii!i:!~!i
iii:~•i::i!i:!!!:.!:!k:~.
f:i~:~: ~I:.<Ii!."I:.IIEY~:!!I
!!!:i!::!i:iiiiiiliiii!;!i
l ':ii<iS'C!::!::i~i2:i~!!::i[ :ii)!i!:!i;i;<':i::!:::.:ili.::lF~!!;ii!~ 0 "~
kFIddN
IT"
>Z o ~ u31no i~ ~ i:??<:::i:!::;!ii:~i
i:: ::i~i:S:i!!::i:!!:!:!ii!:ii:i:!:iiii::ii:!~i~
:i :i~:i?!ii2:ii!'{:"!~i~~{:ii":?ii2!:i:!}ff!!i2:~i!:~ :il;::.F!i:i~;%.i?ii?~:E::iS.:::i;i!~:~;!L
::::i:;::!:i~i:i:!!??!.i!!:i:/:i:5:i?!s:i!i. i:i:~i?i/:?~i
i::.:.il!~i~2::i71}i?
,.C
0
IN~NI.LNOO
J ,x::
'696L'MO,O °
NOIIVNOZ z z z ~" Z ~::~!: :~!!:ii!ii!::!ii!i~.i!i~!!!iii!i~,i:i:i{}.-~iiii~:i.~:~:.~
i:::ii~:ii:~. :~ :i co ~ ~~~ i!i :! : {::!!ii~
(~pn~s s!ql) o~ ~~ co,
XUVaNnO8
3ON3n03s ~
, ~ m
O9 ZZ" a2~
I .... :!:.: :!~!~i!~:!:i~;:~:!<:~!~!~!;:i}~::i!~! ! :! i!~! :!~i:!:,::!;i
~i<@~: i i !!!}!i~y:~! !i:~i~!~:! !i!!ili!!:~:,~!!!ii i!:~!}:!!2!2
eueoo~s!eld I eueoO!ld auaoo!lAI
Jaddlq auaoo!lAI elPP!IAI-JeMo7 eueooS!lO
~addn gb~I:!:~f!!::~i~:Eiitl~ii i i i!iljii i [~I1 ~ 0
-~.~_
~ 0
~'~
SVWO31d
NOIIVIAIWO:I VS31AI VOid V9 VIIdV~VO 3wno3w3~N
SV9
0
,..o ~~
~g
: ~:!iI i::!!,!!~, !i!!ii!i ,~!i~ii!i:!i!i~i:i!~!ii: :,ili:=ii!,i~i i:!i~~!ii~ i!i!i!! ~:ii!!ii ~i'~i!il~ii::iil:~:~!i~ '~ i'~:~,~¢~:i!!i~:i~~:
:i! ::~:::!~ii~i~:~i!i!~ii~
i !i =~
..~ >
m
©
iiii~I!iii!iliii!iiiiiii!iii!!ii!ili!il
i}i!~!iiiiiiiiii!i!!~i!iii~:iii!i
!ii ili !ii!!!ii i i!i!i~~:~i~ii!iii
ii!il,,~i i~
°:tq,, ! i iii
:":'~' iill
!ii~i;~i!ii!ii~i'!i!iii4iiiii~i~i i '~!!i~::~!t!
i!!ii!!iiii~?iii i i ~i i !i i!~,~!)i
~!i;i!iiiil
,~;ili ii~/:
! i~i u ~
~ t,..
"13
=~ ~
-..,..,
0 ,-~
u~
0
?~- -- __ ,,
~ > .,.~
3N30019d 3N3OOIIAI W3ddn
d e a p e J o :1 ~- m ~
,.o to3
>m !
.~_ ~
,© ©
©
Z~ Z .
~'~
~ euaoO!ld euaoo!lAI Jaddn aueoo!lAI elPP!IAI
'~ A ~1 V I 1 El 3 1

the east-west-trending Cretaceous-Paleogene pas- unconformity) and at its top by sequence boundary
sive margin of South America collided with the SB-3.3 (16.5 Ma). The basal Lower Miocene (Aqui-
transpressional front of the Caribbean Plate. Thus, tanian) sequence boundary SB-3 is a composite
the northern Venezuelan transpressional folded belts surface characterized on the regional scale by three
became associated with a foredeep and its depocen- main features: (1) onlap that recorded the deepening
ters that migrated from west to east. Lugo (1991) of the basin at the beginning of the foredeep phase
illustrated the Paleogene foredeep of the Lake Mara- (Fig. 14a and Fig. 20a); (2) on the distal stable
caibo area and Audemard (1991), Lugo and Mann platform a set of prograding sequences; and (3) on
(1995) and Audemard and Lugo (1996) sketched the onshore proximal stable platform an erosional
how this foredeep migrated farther east into the area surface defined by truncation of Cretaceous strata
of this study. Three directions of sediment transport associated with the southward-migrating peripheral
characterize the foredeep basin fill as follows: (1) an bulge (Fig. 20b).
important longitudinal east-west-directed transport; Fig. 15b shows an isopach in seismic twt-time of
(2) a southerly sediment source from the adjacent the Lower Miocene-Middle Miocene, which is char-
Guyana Shield; and (3) a north-northwesterly source acterized by its overall wedge-shaped geometry with
of sediments from the emergent fold belt of the Ser- values ranging from less than 0.5 s to over 2.5 s. On
ranfa del Interior which reworked the Cretaceous- the longitudinal profile (Fig. 9a), which obliquely
lower Tertiary units that outcrop to the north. Note crosses the axis of the foredeep basin, Unit V (i.e.
that a significant amount of Plio-Pleistocene sed- the SB-3-SB-3.3 interval) is relatively thick (3000
iments derived from the Serranfa is also trapped m) to the west (near the Anaco trend) and grad-
in compressional satellite ('piggyback') basins that ually thins towards the SSE. On NW-SE-oriented
overlie the Miocene accretionary wedge of the Mon- profiles (Fig. 8) Unit V pinches out towards the
agas foothills and do not reach the foreland basin south as it onlaps the truncated Cretaceous (Unit
itself (e.g. Parnaud et al., 1995). II). The seismic and sedimentological characteris-
tics differ between the onshore and offshore areas
Well calibration and key chronostratigraphic (see Fig. 20a). Therefore in the following sections
seismic horizons of foredeep stratigraphy the onshore and offshore areas will be described
separately.
As will be shown later the estimated age attribu-
tions for the Neogene stages differ and there is no Unit V onshore
agreement among various authors on the chronos- On west-east-oriented seismic profiles (see
tratigraphic timing of stratigraphic stages and series. Fig. 21a) the stacking pattern of key wells per-
For the purpose of this paper the assigned ages in mits the subdivision of Unit V into at least three
Ma of the chronostratigraphic horizons on figures backstepping depositional sequences (i.e. third-order
and seismic profiles have been made in accordance sequences) which are bounded by the sequence
with the global cycle chart of Haq et al. (1987). boundaries SB-3 (lowermost Miocene, approx. 25.5
Because changes in the time scale are anticipated, Ma), SB-3.1 (21 Ma), SB-3.2 (17.5 Ma) and SB-3.3
sequence boundaries are numbered and the inferred (16.5 Ma). These boundaries are characterized by
age is given in brackets. Fig. 19a is a simplified local truncation of the underlying reflectors and by
summary showing key onshore chronostratigraphic onlapping reflectors. Internally, these depositional
horizons interpreted on well logs and seismic lines. sequences are characterized by highstand and trans-
It also shows major facies cycle subdivisions of gressive system tracts. However, the transgressive
the stratigraphic column. The foredeep stratigra- system tracts and maximum flooding surfaces are
phy is characterized by two overall transgressive- poorly developed. The following stratigraphic sum-
regressive cycle wedges, subdivided into four major mary of Unit V is derived from a sedimentological
units: the lower Unit V, corresponding to the Lower report of well L (see Fig. 3a) shown in a gener-
Miocene; Unit VI, Middle Miocene; Unit VII, Up- alized manner in Fig. 19a and located in the axis
per Miocene; and Unit VIII, Plio-Pleistocene to the of the western portion of the basin. From bottom
present. In the following sections a more detailed to top Unit V consists of two major lithofacies:
description of these units is given, including com- (1) lithofacies A, overlying, but not reaching the
ments about their overall geometry, seismic facies, SB-3 (i.e. basal foredeep unconformity), a section
stratigraphy and depositional environment. of blocky (15-25 m thick) and massive, medium- to
coarse-grained sandstone interbedded with thin (2-4
Unit V (Lower Miocene) m thick) layers of shale and occasional lignite which
is bounded at its top by sequence boundary SB-3.1;
Seismostratigraphic Unit V is bounded at its base this lithofacies was deposited in a fluvial environ-
by sequence boundary SB-3 (i.e. the basal foredeep ment; and (2) lithofacies B, bounded by SB-3.1,

Fig. 20. Details of profiles showing the basal foredeep unconformity. (a) Uninterpreted and interpreted dip-oriented segment of seismic
profile showing the configuration of the Neogene foredeep in the offshore area. For location see Fig. 3c. Note in the downdip direction the
onlap of deep water facies on relatively thick Cretaceous passive margin sequence. Abbreviations: SB-3 = basal foredeep unconformity;
SB-3.10 -- Upper Miocene (5.5 Ma) sequence boundary. See Fig. 3c for location. (b) Uninterpreted and interpreted dip-oriented segment
of seismic profile showing the configuration of the Neogene foredeep in the onshore area. See Fig. 3c for location. Here a Neogene
shallow-water facies onlaps SB-3 which overlies a thin truncated proximal passive margin Cretaceous sequence.
SB-3.2 and SB-3.3, and consisting of two prograd- and ending with fine- to m e d i u m - c o a r s e sandstone.
ing packages of coarsening-upward sequences which These packages were deposited in a littoral to shal-
are characterized by basal shales grading upward low marine environment as coastal bars prograding
into alternating facies of siltstone and sandstone progressively landward.

Unit V offshore during the latest Early Miocene. This corresponds

The stratigraphic configuration of Unit V in the with SB-3.3 or the 16.5 Ma sequence boundary of
offshore area differs considerably from the onshore Haq et al. (1987). Laterally and upward the sec-
but is still characterized by a northward-thickening ond lithofacies of Unit V occurs basinward to the
wedge (Fig. 6a). In this part of the basin Unit V cor- northeast and consists of mainly brown-olive silty
responds to the deep-water phase which illustrates shale. Sedimentological and paleontological analy-
the inception of the foredeep (Fig. 20a). Like in the ses indicate that these sediments were deposited in
onshore area, offshore Unit V is bounded at its base an upper bathyal to outer shelf water depth and are
by SB-3 (i.e. the basal foredeep unconformity) with considered to be Early Miocene (Burdigalian) based
its deep-water Lower Miocene sediments onlapping on the occurrence of Globigerinatella insueta. In
onto thin Oligocene or probably older sediments. light of the above description and patterns observed
This interpretation is particularly important because on seismic profiles it is suggested that the upper
a similar situation can be observed in wells of the E1 lithofacies of Unit V is represented by lowstand de-
Furrial-Carito oil fields, i.e. within the folded belt posits characterized by muddy slope fans (Fig. 2 l b).
to the north of the study area. There, the well data The sequence boundaries that bound each of these
(Fig. 19b) show that a substantial deepening of the lowstand deposits are from base to top as follows:
basin occurred during the Early Miocene as recorded SB-3 dated 25.5 Ma, SB-3.1 dated 21.0 Ma and
by an important unconformity separating underlying SB-3.3 dated 16.5 Ma.
Oligocene neritic sediments (lower Merecure For-
mation) from the overlying middle to upper bathyal Summary and comments
shales and thin turbidites of the Lower Miocene Unit V has an overall transgressive character and
(lower Carapita Formation). In the offshore to the is bounded at its base by SB-3 dated approximately
north and northeast, Unit V and upper units have 25.5 Ma (i.e. basal foredeep unconformity) and its
been deeply eroded by submarine currents during top by SB-3.3 dated 16.5 Ma. Regional sequence
the Upper Miocene (Fig. 20a). Consequently, part stratigraphy allowed the subdivision of Unit V
of Unit V is directly overlain by sequence bound- (Lower Miocene) as follows: in the onshore, Unit V
ary SB-3.10 (5.5 Ma). Farther southeast still in the consists of three depositional sequences, character-
offshore but in the passive margin domain, Unit V ized mostly by highstand and transgressive deposits.
exhibits a uniform thickness (approx. 110 ms) along This unit includes part of the Merecure Formation
strike profiles and thins updip and downdip (i.e. and the basal portion of the Oficina Formation. In the
toward the south-southwest and toward the north- offshore Unit V consists of a thin carbonate platform
northeast). On seismic dip profiles Unit V can be located to the south and two depositional sequences
subdivided into two depositional sequences which characterized by lowstand deposits. Proceeding from
consist of a backstepping mound-shaped configura- west to east along the axis (i.e. from onshore to
tion, each of them bounded by onlap surfaces and offshore) of the basin, the seismic reflection patterns
underlain by truncated reflectors (Fig. 21b). Unit of these depositional units change. In the western
V is characterized by low- to medium-amplitude area, the seismic reflection patterns are represented
and moderately continuous reflectors which dis- by medium- to high-amplitude continuous reflectors
play bi-directional terminations (i.e. onlap updip and and the sequences boundaries are defined by on-
downlap basinward). Based on seismic and well cor- lap and local truncation. The configuration changes
relations (Fig. 14a) in this portion of the basin, Unit laterally and basinward, to low-medium-amplitude
V is composed of two lithofacies. A first lithofacies discontinuous reflectors. The sequence boundaries
present to the south-southwest consists of gently are difficult to follow and some may merge into a
dipping shallow-water carbonates that overlie Unit single surface or seismic reflector. The seismic facies
IV (Oligocene). On seismic this facies is not visi- of the west and southwest are interpreted as coastal
ble, and only a single high-amplitude reflector with to platformal facies, gradually deepening to bathyal
moderate continuity is present. The carbonate sec- water depth to the north and northeast. This facies
tion consists of white to beige skeletal coral micritic marks the inception of the foredeep phase in the
limestones that contain macroforaminifera such as basin.
Miogypsina, Lepidocyclina and, Nummulites, gas-
tropods, echinoderms, bryozoa and algae (Halimeda Unit VI (uppermost Lower Miocene-Middle
and Lithothamniun). Sedimentological analysis and Miocene)
the faunas suggest the development of a coral reef
complex during the Early Miocene (Furrer, 1979). Unit VI is bounded at its base by sequence
The carbonate section shows cave and dissolution boundary SB-3.3 (16.5 Ma) and at its top by se-
porosity suggesting subaerial exposure that perhaps quence boundary SB-3.7 (10.5 Ma). On dip-oriented
has been related to a sea-level fall that occurred seismic profiles (Fig. 9a) Unit VI shows an overall

Fig. 21. Interpreted segments of seismic profiles illustrating Miocene sequences. (a) Interpreted segment of an onshore seismic profile showing depositional sequences and sequence boundaries of Unit V

(Lower Miocene) and Unit VI (Upper Miocene) For location see Fig. 3c. (b) Interpreted segment of offshore seismic profile in the offshore area showing the configuration of Unit V (Middle Miocene). See
Fig. 3c for location. (c) Interpreted segment of seismic profile in the offshore area showing the configuration of Unit VI (Middle Miocene) and Unit VII (Upper Miocene). See Fig. 3c for location.
0
wedge shape, trending west-east, very similar in geof siltstone and sandstone and ending with fine-
ometry to the underlying Unit V. In the western and to medium-coarse sandstone. These sediments were
best preserved portion of the basin the unit is over 2 deposited in a litoral to shallow marine environment
s thick and thins toward the east. On N-S-oriented and consist of coastal bars which prograded progres-
profiles (Fig. 8), Unit VI thickens toward the north sively landward. This section is bounded by SB-3.4
and pinches out towards the south as it onlaps the (15.5 Ma) and SB-3.5 (13.5 Ma).
truncated Cretaceous (Unit II) and the crystalline (3) Abruptly, the previously described lithofacies
basement. Onshore, two different seismic facies changes to a massive thick section (approx. 600
characterize Unit VI. The western portion exhibits m) of silty shales with abundant planktonic and
a pattern with high- to medium-amplitude reflectors benthic faunas. The planktonic assemblage includes
with excellent continuity. Toward the east Unit VI is Globorotalia praemenardii, Globorotalia periphe-
formed by reflectors with low- to medium-amplitude roacuta, Globorotalia prefohsi, Globorotalia fohsi
and disrupted/irregular reflection configurations. On lobata, and Globorotalia obesa, which corresponds
the other hand, in the offshore the seismic facies to Zones N10-N12 (Blow, 1969). The benthic fau-
exhibits a pattern with low amplitude and moderate nas consist of Cyclammina cancellata, Bathysiphon,
continuity. Lenticulina subpapillosa, Valvulina flexilis, Bolivina
imporcata, Globulina ovata, among many others,
Unit VI onshore indicating that these sediments were deposited in
Based on internal configuration and well data, middle to upper bathyal water depths and mark the
Unit VI consists of four depositional sequences maximum deepening of the basin during the Middle
(third order) defined from bottom to top by sequence Miocene (Serravallian). This section is bounded by
boundaries SB-3.3 (16.5 Ma), SB-3.4 (15.5 Ma), SB-3.5 (13.5 Ma) and SB-3.6 (12.8 Ma) and con-
SB-3.5 (13.8 Ma), SB-3.6 (12.5 Ma) and SB-3.7 tains the maximum flooding surface of 13.4 Ma of
(10.5 Ma) (Fig. 21a). These sequence boundaries Haq et al. (1987) which is shown in Fig. 19a.
show onlap reflection terminations and SB-3.7 is (4) A thick (between 450 and 750 m) sec-
a well-defined seismic horizon (particularly in the tion of gray-olive shale, locally interbedded with
western portion of the study area) characterized by a few thin layers of fine-grained sandstone. Ben-
an erosive surface showing underlying truncated re- thic foraminifera such as Lenticulina, Bathysiphon,
flectors and overlying onlapping reflectors (Figs. 2 l a Haplophragmoides, Textularia, Miliammina suggest
and 22a). Unit VI is characterized by two different that this section was deposited in water depths rang-
trends. The lower two depositional sequences exhibit ing from upper bathyal/outer shelf to inner neritic.
a backstepping behavior, building up from prograd- The planktonic foraminifera within this section are
ing higher-order sequences (Fig. 21a). The upper abundant and diversified. However, only one distinct
two depositional sequences are formed by subtle marker was found, corresponding to the Globoro-
progradational forestepping but an overall aggra- talia aft. siakensis, Zones N13-N14 (Blow, 1969),
dational configuration (Fig. 22b). Note also that i.e. the upper part of the Middle Miocene. This zone
some of these depositional sequences are composed is close to SB 3.7 which correlates with the sequence
of lowstand deposits. The ages of the sequence boundary of 10.5 Ma of Haq et al. (1987).
boundaries and of the maximum flooding are well
constrained by paleontological data contained in the Unit VI offshore
following stratigraphic summary of Unit VI which is The seismic stratigraphic configuration of Unit
derived from sedimentological reports of key wells VI of the offshore passive margin domain differs
and particularly of well L (Figs. 3a and 19a). From from the onshore. Offshore Unit VI is a single de-
bottom to top the lithofacies that characterize Unit positional unit bounded at its base by SB-3.3 (16.5
VI consist of the following. Ma) and at its top by an onlap surface, correspond-
(1) Gray-brownish shale with occasional thin lay- ing to sequence boundary SB-3.7 (10.5 Ma). To the
ers of sandstone and profuse pellets of glauconite north in the offshore foredeep domain Unit VI is
throughout the interval yielded abundant and diverse thin as a result of widespread deep erosion during
faunas of an latest Early Miocene age. This age is the Early Pliocene. In the updip area Unit VI ex-
mainly based on the occurrence of Globigerinoides hibits a very subtle sigmoidal pattern with moderate
sicanus which is restricted to Zones N7-N8 (Blow, progradational offlap, characterized by low-ampli-
1969). The section was deposited in a middle to outer tude and moderate continuous reflectors with north-
shelf environment. It includes the maximum flooding east-dipping downlap terminations (Fig. 21c). Unit
surface of 16 Ma of Haq et al. (1987) and is bounded V (Lower Miocene) and Unit VI are separated by a
by SB-3.3 (16.5 Ma) and SB-3.4 (15.5 Ma). thin continuous drape sheet (approx. 50 ms) which
(2) Coarsening-upward stacking patterns with represents the downlap surface. Correlation with
basal shales grading upward into variable facies nearby wells suggests that this surface corresponds

Fig. 22. Uninterpreted and interpreted segment of onshore seismic profiles showing sequence boundaries and relations of Middle Miocene
to Pliocene. For location see Fig. 3c. (a) Onshore, the Unit VII (Upper Miocene) is characterized by three deeply incised unconformities
(i.e. SB-3.7, SB-3.9 and SB-3.10). (b) This is another view of the relationship of Unit VI and Unit VII showing a more subdued onlap on
SB-3.7.

to the 16 Ma maximum flooding surface (M1) of 1990; Daza and Prieto, 1990). In the offshore, in
Haq et al. (1987). some areas mostly to the south, Unit VII shows a
In offshore wells (Fig. 14a), Unit VI consists of prograding sigmoidal configuration with a northeast-
thick (~ 180 m) mostly fossiliferous gray-olive shale dipping attitude (Fig. 21c). Note also that most of
with a few layers of sandstone and an increased glau- the sigmoid strata and the successive offlap breaks
conite content towards the lower part. Sedimentolog- were partially eroded by canyons, thus displaying
ical and paleontological analyses indicate that these only remnant slope clinoforms truncated at the top
sediments were deposited in an outer-shelf setting. (Fig. 23a).
Based on the occurrence of planktonic foraminifers,
such as (from bottom to top) Globorotalia fohsi Unit VII onshore
fohsi, Globorotalia fohsi lobata and Globorotalia Overlying the SB-3.7 (10.5 Ma), a fining-upward
siakensis, which correspond to Zones N10 through stacking pattern on logs and reported paleowater-
N14 (Blow, 1969), the whole Middle Miocene is depths suggest that the lower portion of Unit VII
represented. represents an overall transgressive phase (Fig. 19c).
Two major lithofacies are differentiated in Unit VII,
Summary as shown in wells located in the axis of the basin,
To sum up, Unit VI together with the underly- which penetrated a 1200-m-thick section. The lower
ing Unit V appear to form a complete second-order portion consists of siltstone and shale interbedded
transgressive-regressive cycle wedge in the onshore with white fine- to medium-grained poorly sorted
area (Fig. 19a) as well as in the offshore. The sandstone. Plant remains and occasional lignite lay-
transgressive phase comprises the Lower Miocene to ers are present. Sedimentological analysis suggests
the middle portion of the Middle Miocene. During that these sediments were deposited in environments
the transgressive phase two flooding events are in- ranging from continental to coastal plains. This por-
terpreted. The older event is the 16-Ma maximum tion is bounded at the base by SB-3.7 (10.5 Ma)
flooding surface of Haq et al. (1987) which can and at its top by SB-3.8 (8.2 Ma). Its depositional
be correlated along the basin. The second flooding sequence consists of transgressive and highstand de-
event, which culminates with a major paleobathy- posits. The upper portions contain predominantly
metric deepening, is interpreted to mark the peak gray-olive shale with abundant benthic foraminifera,
transgression and is recorded as the 13.4-Ma (Ser- siltstone and poorly sorted sandstone. Sedimento-
ravallian) maximum flooding surface of Haq et al. logical and paleontological analyses suggest that
(1987). Following that maximum transgression the this upper portion was deposited in environments
overall regression began. The regressive phase com- ranging from shallow marine to outer-shelf/upper
prises the upper portion of the Middle Miocene and bathyal. Two depositional sequences have been in-
SB-3.7 (10.5 Ma) is interpreted as the peak regres- terpreted within this upper portion, bounded succes-
sion for this cycle. The Middle Miocene (Unit VI) sively by SB-3.8 (8.2 Ma), SB-3.9 (6.3 Ma) and
in the onshore corresponds to the Freites Formation at its top by SB-3.10 (5.5 Ma). These depositional
(e.g. Gonzalez de Juana et al., 1980). sequences are characterized mostly by lowstand de-
posits (Fig. 19c). A second deepening event is ob-
Unit VII (Upper Miocene) served within this upper portion. However, the peak
transgression is paleontologically poorly constrained
Unit VII on seismic profiles is best defined in the due to the lack of reliable biostratigraphic succes-
western onshore where three depositional sequences sions that would permit a more precise dating of
are observed (Fig. 22a). Overall aggradational to sequence boundaries and maximum flooding sur-
subtle onlapping patterns predominate and involve faces. Based on seismic interpretation and well data
from bottom to top the sequence boundaries SB-3.7 (Figs. 14a and 19c) the peak transgression occurred
(10.5 Ma), SB-3.8 (8.2 Ma), SB-3.9 (6.3 Ma) and between sequence boundaries SB-3.8 (8.2 Ma) and
SB-3.10 (5.5 Ma). Seismic data in the west show SB-3.9 (6.3 Ma) and, consequently, the maximum
that these sequence boundaries are characterized by transgression M2 is tentatively correlated with the
erosional surfaces and medium- to high-amplitude 7-Ma maximum flooding event of Haq et al. (1987).
reflectors with moderate continuity. Moving to the
east, along the axis of the basin, Unit VII shows seis- Unit VII offshore
mic facies characterized by low-amplitude, discon- The stratigraphic configuration of Unit VII in the
tinuous reflectors. While the sequence boundaries offshore differs from its onshore equivalents. In the
are reasonably well defined in the western part of offshore passive margin domain, Unit VII is char-
the area they are difficult to follow eastward because acterized by a single depositional sequence with a
they are intercepted and offset by a complex system lowstand system tract of slope fan and lowstand
of listric faults (Figs. 9a and 10; see also Lilliu, prograding wedge deposits (Figs. 2 l c and 23a). Unit

Fig. 23. Canyon systems associated with SB-3.10 (5.5. Ma). (a) Uninterpreted and interpreted segment of an offshore seismic profile.
The SB-3.10 (5.5 Ma) sequence boundary marks a major erosional event. The line drawing also shows Unit VII (Upper Miocene)
characterized by a prograding sigmoidal configuration (see Fig. 3c for location). (b) The detailed time-structure map of the top Upper
Miocene shows the geometry and orientation of the canyons in the offshore area.
VII is bounded at its base by SB-3.7 (10.5 Ma) gives an Late Miocene to earliest Pliocene age for
and at its top by SB-3.10 (5.5 Ma) with its deeply this unit.
incised valleys, which farther basinward (i.e. north-
east) develop into large-scale submarine canyons. Summary
In some areas only isolated remnants of Unit VII Unit VII is overall transgressive and bounded by
can be observed on seismic profiles (Figs. 6a, 7b two major regional unconformities, i.e. at its base by
and 19d). To the north, where Unit VII has been SB-3.7 (10.5 Ma) and at its top by SB-3.10 dated
penetrated by wells (Fig. 14a), mostly gray-green- 5.5 Ma. Regional sequence stratigraphy allowed the
ish shale interbedded with siltstone and fine-grained subdivision of Unit VII (Upper Miocene). In the
sandstone has been reported. The presence of ben- onshore, Unit VII consists of three depositional
thic foraminifera such as Cyclammina, Haplophrag- sequences: the two lower depositional sequences,
moides, Alveo-Valvulinella suteri, Uvigerina, Lenti- represented mostly by transgressive and highstand
culina, Bolivina, Recurvoides, Buliminella, among deposits, and the upper depositional sequence, char-
many others, indicates that these sediments were acterized by lowstand deposits. The second peak
deposited in upper to middle bathyal water depths transgression is interpreted to have occurred during
characteristic for lowstand deposits. The limited the Late Miocene and is correlated with the 7 Ma of
occurrence of planktonic species of Globigerina maximum flooding surface of Haq et al. (1987). Unit
nephenthes and Sphaeroidinellopsis paenedehiscens VII includes the La Pica Formation in the onshore

portion of the basin (e.g. Gonzalez de Juana et al., Unit VIII (Plio-Pleistocene)
1980). In the offshore area Unit VII consists of a sin-
gle depositional sequence characterized by lowstand The overall regional pattern of Unit VIII (i.e. both
deposits. west to east and north to south) is characterized by
its well-defined wedge-shaped configuration (Figs.
SB-3.10 (5.5 Ma), a Messinian erosional event 8 and 9a). Fig. 15c is a time-isopach (two-way-
time) of the unit. A monoclinal trough characterizes
Unit VIII, the youngest part of the stratigraphic the isopachs. Considerable thickness variations dis-
column, is separated from Unit VII by the se- tinguish this unit which pinches out to the south
quence boundary SB-3.10, which no doubt is the increasing to thicknesses in excess of 7000 m (~5
most conspicuous feature of the basin. On seis- s) in the offshore area up to the shelf-break, where
mic profiles, this unconformity is easily identified, the sequence is involved in a major growth fault
showing large-scale erosion with deep-water chan- zone (Figs. 9a and l la). The maximum depth of
nels and submarine canyons that create an irregular more than 5 s in its northwestern part reflects the
scoured surface with truncations of the parallel-pla- subsidence and infill of the foredeep by the eastward
nar reflectors of the subcropping depositional units. advance of the Neogene depocenter.On reflection
Fig. 23b shows a time-structure sketch map of the seismic profiles of the western part of the basin
main canyon systems and the physiographic features Unit VIII consists of parallel to divergent medium-
of these northeast-trending canyons characterized to high-amplitude reflectors with moderate conti-
by closer contours and landward V-shaped valleys. nuity. Moving to the east (i.e. offshore) forestep-
The canyons are better visible on shelf-parallel-ori- ping progradational sigmoidal configurations domi-
ented profiles which generally show the U-shaped nate (Figs. 24 and 25).
canyon flanks. On perpendicular, i.e. dip-oriented
profiles, the truncation appears minimal because the Unit VIII onshore
canyon surfaces often mimic the underlying slope Fig. 19c is a composite log summary of well
clinoforms (Fig. 21c). The seismic facies of the K that defines the chronostratigraphic horizons in-
canyon fill consists of low-amplitude, discontinu- terpreted on seismic profiles and is tied to other
ous to chaotic and irregular reflectors which often wells. Unit VIII represents an overall regressive
pinch out on the slope. Upward and on dip pro- regime showing a distinctive change in log-facies
files a series of prograding clinoforms plunge from from a fining-upward trend in the underlying Unit
the wall canyon head and pinch out within the VII to a coarsening-upward trend (Fig. 19c). Unit
upper slope (Fig. 21c). The lower portion of the VIII shows great lateral variation of seismic facies
canyon fill consists of slumps and slope deposits, which is related to the progressive eastward move-
and the upper portion exhibits isolated lowstand pro- ment of the Caribbean Plate relative to the South
grading wedges. Well data suggest that the canyon American Plate during the Neogene (Fig. 4c and d).
fill is overwhelmingly dominated by shales which Unit VIII is bounded at its base by SB-3.10 (5.5
are characterized by monotonous high values on Ma) and at its top by the actual topography. At least
gamma-ray logs and low resistivity values occa- three depositional sequences have been interpreted
sionally interrupted by sandstone spikes (Fig. 19d). within Unit VIII. These are bounded by SB-3.10
When the shaley infill is superposed on slope and (5.5 Ma), SB-3.11 (4.2 Ma), SB-3.12 (3.8 Ma) and
basin shale, the well-log response of canyon fill and SB-3.13 (3.0 Ma) which are defined seismically by
outside-canyon shales is virtually the same, i.e. with- regional onlap and local truncation and tied to well
out showing any abrupt changes in the log patterns. logs. These depositional sequences are dominated by
Based on well reports Prieto (1987) suggests that transgressive and highstand system tracts. The ages
the canyons developed between 7.0 Ma and 4.5 Ma. of the sequence boundaries are poorly constrained.
The approximate age of the canyon formation can Based on well reports, the lithofacies of Unit VIII
be reasonably related to a major sea-level fall that consists of stacked patterns of coarsening-upward se-
occurred during the Messinian, which corresponds quences of lignitic shale, red-brown micaceous silt-
to a 5.5 Ma sequence boundary of Haq et al. (1987). stone and shale, poorly sorted fine- to coarse-grained
Note particularly in Figs. 6a and 20a that in the sandstone, and conglomeratic sandstone. Faunas are
downdip direction the SB-3.10 surface is overlain by very poor and marine fossils are lacking within this
a pronounced regional deep-water onlap that mimics unit although fossil turtles, Corbicula sp. and fish
the underlying basal foredeep unconformity. This teeth (Funkhouser et al., 1948), sparse species of Mil-
suggests that the deep-water canyon cutting event is iammina fusca (Hedberg et al., 1947) and the fluvial
strongly enhanced by the differential subsidence of mollusk of Hyria trinitaria (Palmer, 1945) all sug-
the foredeep and therefore leads to the formation of gest a Pliocene age. Unit VIII was deposited in a
a 'secondary basal foredeep unconformity'. littoral/marginal marine to mostly continental setting.

Fig. 24. Seismic expression of Plio-Pleistocene (Unit VIII) prograding sequences on the offshore Orinoco platform. For location see
Fig. 3c. SB-3.15 is the inferred top of the Pliocene. SB-3.10 is the Messinian canyon-forming event.
Unit VIII offshore shows the geographic position of the shelf edge
The stratigraphy of Unit VIII in well A (Fig. 19d) of each sequence. The depositional framework was
shows an overall coarsening- and thickening-upward controlled by the advance of the Orinoco Delta since
trend. On seismic profiles, this unit is repre- the Late Miocene. The shelf edges display a con-
sented by a thick (~7000 m) progradational sig- tinued shift from the southwest to the northeast,
moidal pattern consisting of multiple offlapping de- which is due to the high rate of sediment supply
positional sequences that represent high-frequency from the west. Furthermore, the progressive devel-
Plio-Pleistocene glacio-eustatic sea-level fluctua- opment of a growth fault system was established
tions (Fig. 24). Reliable paleontological data are with a base that is controlled by the top of the un-
not available but the seismic resolution is good and derlying passive margin (Fig. 9b). Unit VIII consists
several sequence boundaries were interpreted. The of a typical deltaic progradation composed of the
dating was inferred by comparing the interpretation following.
with the idealized Neogene model of Vail et al. (1) A lower portion characterized by gray-olive
(1991). Seven Plio-Pleistocene sequences are rec- shale with abundant foraminifers such as Orbulina
ognized. These sequences are mainly third-order, universa, Globigerinoides obliquus, Lenticulina sp.,
but some may be fourth-order sequences (Vail and Bathysiphon sp., and Cyclammina interbedded with
Wornardt, 1990; Mitchum and Van Wagoner, 1991; a few thin layers of turbidites and fine-grained sand-
Vail et al., 1991). The inferred dates can be obtained stone. Sedimentological and paleontological analy-
by combining Fig. 25b and Fig. 26. In general, ses indicate that these sediments were deposited in
the depositional sequences bounded by these se- an upper- to middle-bathyal water depth.
quence boundaries show a well-developed lowstand (2) A middle portion consisting of fine- to
system tract with slope fan and prograding wedge medium-grained sandstone interbedded with gray
deposits, a very thin transgressive system tract and siltstone and shale. The presence of Sphaeroidinella
variable thickness highstand system tracts. Fig. 25b seminulina marks the top of the Pliocene. In addi-

w 0 R I N 0 C 0 D E T A
a~! LINE 6a LINE 7b
0
UNIT VII
-!
-i
1_.3"
~K4~ t~
4-
l lOKm I ~
Fig. 25. Offshore Orinoco platform, prograding systems of Unit VIII (Pliocene). (a) Line drawing of an offshore W-E profile adjacent to
the Orinoco Delta showing the sequence stratigraphy of Unit VIII (Plio-Pleistocene). The heavy dashed line marks the interpreted top of
the Pliocene, i.e. SB-3.15. (b) Map showing the position of the prograding shelf-margin from Late Miocene to Present. The growth fault
system to the east is in essence a Quaternary growth system.
tion these shales contain pelecipods, gastropods and TERTIARY PALEOGEOGRAPHY OF EASTERN
foraminifera such as Ammonia, Elphidium, Florilus, VENEZUELA
Haplophragmoides, Eponides, Globigerina and Glo-
bigerinoides. Sedimentological and paleontological The map of Fig. 27a shows the Oligocene fa-
analyses indicate that these sediments were de- cies distribution, its inferred updip zero edge, and
posited in a middle to outer shelf environment that the starved downdip continuation of that formation.
probably represents a prodeltaic facies. Because of the limited data the Oligocene paleo-
(3) An upper portion that consists of coarsen- geography is difficult to reconstruct. The Oligocene
ing- and thickening-upward strata of fine- to coarse- zero edge is due to erosional truncation. Downdip
grained sandstone with frequent microconglomerates from the lowstand deltas of well A, the Oligocene is
and abundant fossil and wood fragments, chert, pel- absent in well B probably due to sediment starvation.
lets and pyrite nodules. Interbedded with this facies, To the northwest of the starved area more complete
thin layers of gray siltstone and shale are present, pelagic sediments are reported from the Serranfa del
with planktonic foraminifera such as Globorotalia Interior and from Trinidad. Combined with the plate
truncatulinoides, suggesting a Pleistocene to Recent tectonic reconstruction shown in Fig. 4b this paleo-
age. Sedimentological and paleontological analyses geographic map suggests that the area now occupied
indicate that these sediments were deposited in a by the Serranfa del Interior and the Venezuelan off-
shallow-water depth. They represent the delta front shore to the north was still an intact passive margin
and delta plain facies of the Orinoco Delta. Based on during the Oligocene.
stacking patterns in well logs, the paleo-water depth The paleogeographic map of Fig. 27b depicts the
and seismic configuration, Unit VIII represents the main depositional regimes of the Early Miocene (25
overall regressive Pliocene to Present phase of the Ma), at the onset of the foredeep phase. To the
Eastern Venezuelan Basin. west and south of the basin, continental to coastal

ii ii ii iiiiii
CHRONOSTR~TIGRAPHY SEQUENCE
SEQUENCE
I
S(QU~NC~ S T R A T I G R A P H Y BOUNDARYI
SYSTEM SERIES STAGES(1) (2) ! (3) I (4) I (S) l (s) BOUNDARY o.s o I AG[ ,"
' ~ "O.Ol- , 0 . O l , ~O.Ol, =,,,+ '
QUATERNARY PLBSTOCEm 1 ~'
' 9 9 ,,, , .6 -"-1.64-' 1 9 "Pu~,sTo"1.65 1 T/-t.85"
SB-3.SB-3"'
514 . . . . . . . . . .: . - ' , : ~ ' ~ ~ ~ : ~ ~ ~
(~ 1-- 3.4 - - 3.4-- .z- -3.4 -. 3.8--3.5-
El ~ SB-3.13 "-.~-~ ~: ~ ~ .-'" ~ ~,'-- ""~s"
o. 5.3----5.2-- 5.1 - - 5.3-- 5.3--5.2- S B 3 1 2 9 9, ~ - - - : =
- , * +,~_ ~ ! ~
_L . . . . . . . :;_
' "~.'!~-_ ..~_ - ~.~- -~.~-

SB-3.1 0 "s." - ' " ~- . .%. . . . i:~.-
LLI
Z
U TORTOMAN
-u- , ~ . , ,., ~
~u-:J u." .-"-X ,
: I-2-- L
~., IMW_
SB-3.8-""" ~ . . . . r--" "T.-.

LU LLJ -10.4- i .I0.2.
lZl ---11.2-- -113" ;-11-0--11.2-
_14.2_ ,14.4.::
-14.5 - -14.8 -
---15.1-- -152-
-i, I.ANOHIAN
>- - - - 1 6 . 6 - - -16.3-
-16.0 - -16.1 - -16a?,-
SB-3.2'-"~._ .......... ,,,,_. ,,.,_z~.~

BU~.JAN i
<:C
m mm20.0- i2050i
,..y TR J
p- L - 20.5-
--21.8-- -21.5- rm,laa,
k ........ (2.2.5P-.- - 22 . . . . H .
AQUITANLAN
LLI .... 23.7__.'23.3- -23.5- "23.8-
-24.6-
I._ + -25.2-
III
Z c''"
!11 Iii -28.5-
:,,.-:,,,
- 29.3- -29.0- i
0 rao.o-
,
o o ,
i i
i i
-33.0 -
o L.U .
. . . . . . . . . _ _ + : - -_ _
i,,I. - 35.4 -
---~8.6---
~.0~
Fig. 26. Variations on the Neogene and Oligocene chronostratigraphy according to different authors. On the far right is the inferred
position of our sequence boundaries on the Haq et al. (1987) cycle chart. Column (6) shows the ages assigned to various stages by Haq et
al. (1987). Column (1) shows ages proposed by the Geological Society of America (DNAG 1989). Column (2) the same after Harland et
al. (1990). Column (3) the same after Bolli et al. (1991). Column (4) the same after Vail and Wornardt (1990). Column (5) the same after
Hardenbol et al. (in press).
plain sandstone and shale (upper Merecure For- The Serravallian paleogeographic map of Fig. 27c
mation) merge with shallow-water platformal shale shows that the sedimentation was strongly controlled
and fine- to coarse-grained sandstone littoral bar by subsidence of the eastward shifting foredeep.
facies (lower Oficina Formation). To the north, i.e. As a result the shelf margin, with its coastal facies,
closer to the present deformation front and within moved landward, i.e. toward the Guyana Shield. This
the folded belt, deep-water shales with thin layers period includes maximum flooding in the foredeep
of turbidites (lower Carapita Formation) onlap on domain. The basin shows a narrow trough geometry
the basal foredeep unconformity. To the east and to the east. To the west sediments are deposited in a
southeast limited terrigenous input to the shelf fa- broad coastal plain to littoral zone, while in the cen-
vored the development of a thin carbonate rim at tral part and towards the east deep-water sediments
the edge of the platform. Beyond the shelf-break, are deposited in an open-sea environment. The sym-
deposition occurs in upper bathyal water depths of bol coveting the Serranfa del Interior suggests that
muddy slope fans. The age of the basal foredeep an accretionary wedge composed mostly of Lower
unconformity which forms the base of the Lower Miocene imbricates was emplaced in that area by
Miocene suggests the presence of a structural load Serravallian time.
in the area of the present-day northern Venezuela Fig. 27d is a paleogeographic map of the Lower
offshore. The location and dimensions of that load Pliocene just after the SB-3.10 (5.5 Ma) erosional
remain largely hypothetical and are not shown in event. The Serranfa and the Monagas foothills are
Fig. 26c. close to their present-day position near the edge of

Fig. 27. Maps showing the paleogeographic evolution of the Eastern Venezuelan basin. (a) Paleogeographic map of the Upper Oligocene just after 30 Ma. This map marks the cessation of the passive margin regime. The present-day
area of the Serranfa del Interior is shown as an undeformed extension of the Atlantic passive margin. (b) Paleogeographic map of the Lower Miocene, after inception of the foredeep (25 Ma). The area of the present-day Serranfa del
Interior remains undeformed. Presumably the load that caused the initiation of the foredeep was positioned much farther northwest somewhere on the margin of the present-day Caribbean shelf. Compare this figure with the isopach
in time shown in Fig. 15b. (c) Paleogeographic map of the Middle Miocene (15 Ma) which shows that much of the Serranfa del Interior is now emplaced while southern Trinidad is not yet deformed. (d) Paleogeographic map of the
Early Pliocene just after the SB-3.10 (5.5 Ma). The Serranfa del Interior and Trinidad are now mostly in place. The delta front of eastern Venezuela has migrated from west to east. Compare this figure with the time isopach shown in
Fig. 15c.
Fig. 28. Eastern Venezuelan Basin chronostratigraphic charts. (a) A W-E-oriented chart that sums up the onshore to offshore chronostratigraphy of the longitudinal section shown in Fig. 9. The passive margin Cretaceous section is
thin to the west due to onlap below and truncation at the top. Also onshore a very large hiatus is associated with the basal foredeep unconformity. Farther east this hiatus is replaced by condensed sequences of the Paleogene. In the
west the Miocene is represented by relatively thick Miocene aggrading and prograding sequences which shale out to become a more condensed section towards the east. The dashed and dotted lines in the second-order cycle column
are the sequence boundaries and the maximum flooding surfaces used to separate individual second-order cycles. (b) A SSE-NNW-oriented chronostratigraphic chart that displays the main sequence-stratigraphic units in a section
across the onshore tbredeep. (c) A SSE-NNW-oriented chronostratigraphic chart that displays the main sequence-stratigraphic units in a section across the offshore foredeep.
the folded belt. The east Venezuela foredeep and regressive phase that is still active today. Such a
it offshore extension are now almost completely long-term global transgression-regression is often
developed. referred to as a major 'Continental Encroachment
Cycle'. Onshore the transgressive encroachment of
the Lower Cretaceous is well displayed as a thin
DISCUSSION OF THE CHRONOSTRATIGRAPHIC wedge of Cretaceous clastics (Figs. 9 and 28a). This
MODEL sedimentary package was deposited during landward
displacement of the shoreline with a retrograda-
Introduction tional geometry corresponding to a 'Backstepping
Transgressive Phase' (Cramez and Vail, 1990). On
Our conclusions are summarized in the form of all chronostratigraphic charts the Turonian K4 sur-
several chronostratigraphic charts that correspond to face corresponds to the maximum flooding event.
key transects across the area. Chronostratigraphic Thus, the K4 surface separates the first-order over-
charts link the spatial distribution of depositional all transgressive phase from the overlying first-order
sequences to geological time. Thus, chronostrati- regressive phase and is marked by a major down-
graphic charts display genetically related strata that lap surface corresponding to the 91.5-Ma (middle
were deposited during a given time period. They Turonian) event of Haq et al. (1987). The overlying
summarize a wide range of information such as: (a) Upper Cretaceous and Cenozoic package is char-
relations of sequences to bounding unconformities, acterized by a progressive seaward displacement of
showing onlap, downlap, toplap and truncation; (b) the shoreline, with progradational geometries that
correlation of sequences to standard geochronologic correspond to the 'Forestepping Regressive Phase'
subdivisions; (c) hiatal gaps along unconformities; (Cramez and Vail, 1990). On the southern passive
(d) distribution of facies and environment; and (e) margin of Venezuela sediments included in this
the relation and correlation of named lithostrati- overall regressive unit reach a maximum thickness
graphic units such as groups and formations within exceeding 5-6 km on the outer shelf and thinning
the sequences even though such units may be time- again towards the Atlantic Ocean.
transgressive and do not show as a distinct unit on The first-order passive margin encroachment cy-
the chronostratigraphic chart. cle is strongly modified within the Eastern Venezuela
Except for the symbols used for condensed se- foredeep and its offshore extension as shown on all
quences, the legend for the chronostratigraphic charts chronostratigraphic charts (Fig. 28). On these charts
is mostly self-explanatory. Ideally, a condensed se- and the corresponding seismic sections, it can be
quence no matter how thin ought to be repre- seen that the passive margin regime terminates with
sented by a symbol that fills the whole time repre- the deposition of thin Paleogene cycles characterized
sented by the condensed section. Such a procedure by narrow and thin siliciclastic northeastward-pro-
is not altogether desirable because large time seg- grading wedges and extensive condensed sequences
ments of a chronostratigraphic chart are often dom- and/or starved sedimentation. These cycles are trun-
inated by rather thin condensed sequences. In the cated updip by the basal foredeep unconformity. The
eastern Venezuela offshore the problem is further main characteristic of the basal foredeep unconfor-
compounded by the observation that much of the mity is the widespread deep-water onlap of Lower
downdip Oligocene appears to be missing, suggesting Miocene shales on the condensed or starved pas-
widespread starved sedimentation that is not easy to sive margin of the Paleogene as best illustrated by
explain. On the chronostratigraphic charts of this pa- Fig. 28b and c.
per, condensed sequences are represented by widely It is concluded that the Continental Encroach-
spaced dashed lines with the understanding that par- ment Cycle of the Eastern Venezuelan Basin is
ticularly for much of the Paleogene it remains ques- strongly modified by a superimposed Neogene fore-
tionable whether we deal with very condensed sedi- deep phase that began with the Early Miocene and
mentation or else with a form of sediment starvation. radically changed the geometry and stratigraphic ex-
pression of the subsequent higher-order cycles. This
First-order cycle change is directly related to the eastward displace-
ment of the transpressive orogen due to the oblique
Overall, the Cretaceous-Cenozoic of the Eastern collision of the Caribbean Plate with the South
Venezuelan Basin was deposited during a long-term American Plate.
transgressive-regressive cycle related to the opening
of the northern Atlantic Ocean that began in the Mid- Second-order and third-order cycles
dle to Late Jurassic. The cycle reached its peak in the
Turonian (Vail et al., 1977; Hallam et al., 1985; Haq A complex series of second-order cycles of dif-
et al., 1987) and ended with a post-Turonian overall ferent duration and magnitude is superposed on the

previously described first-order cycle. These major ond-order cycle bounded by SB-3 (25.5 Ma) at its
transgressive-regressive (T-R) facies cycles (sec- base and SB-3.3 (16.5 Ma) at the top. Based mostly
ond-order cycles) are the most important cycles on offshore data this second-order cycle is subdi-
controlling distribution of facies and are the strati- vided into three units, i.e. the SB-3-SB-3.1 (25.5-
graphic signature for combined eustatic and tectonic 21 Ma), the SB-3.1-SB-3.2 (21-17.5 Ma) and the
second-order cycles. Second-order tectono-eustatic SB-3.2-SB-3.3 (17.5-16.5 Ma) third-order cycles.
cycles typically cover a wide time span ranging from The Middle Miocene (Unit VI) as a whole may
3 m.y. to 50 m.y. Third-order cycles range from 0.5 be viewed as a short second-order cycle defined by
m.y. to 3 m.y. and are superposed on these second SB-3.3 (16.5 Ma) at its base and SB-3.7 (10.5 Ma)
order cycles. At least ten second-order cycles con- at its top. It may be subdivided into four third-order
trol the facies distribution of the study area. Seven cycles, i.e. SB-3.3-SB-3.4 (16.5-15.5 Ma), SB-3.4-
of these cycles occurred during the passive margin SB-3.5 (15.5-13.8 Ma), SB-3.5-SB-3.6 (13.8-12.5
phase and three cycles correspond to the foredeep Ma) and SB-3.6-SB-3.7 (12.5-10.5 Ma). Note that
phase of eastern Venezuela (Fig. 28a and b). the base of this second-order cycle is directly over-
The Cretaceous to Lower Paleocene (Unit II) is lain by the 16-Ma maximum flooding event. Another
bounded below by SB- 1/SB-2, the pre-Cretaceous major flooding event is represented onshore by the
unconformity with its lower Cretaceous onlap. The Serravallian 13.4-Ma flooding event which, however,
Cretaceous to Paleocene is subdivided into five sec- is not recognized in offshore eastern Venezuela.
ond-order T/R packages that are best defined by The Upper Miocene to Recent (Units VII and
maximum flooding (i.e. downlap surfaces) in the VIII) represents the youngest second-order cycle. Its
lower Albian (111 Ma), the uppermost Albian (98.5 base is bounded by SB-3.7 (10.5 Ma), its top is the
Ma), the middle Cenomanian (98.75 Ma), and the land surface and the seaftoor. This unit is an overall
middle Turonian (91.5 Ma). Sequence boundary regressive prograding cycle. Using overall regression
SB-2.1, the Late Paleocene (58.5 Ma) unconformity as a theme it could perhaps be argued that the base
and its conformable continuation, forms the upper of this second-order cycle should be placed earlier
boundary of the uppermost second-order cycle. The to coincide with SB-3.5 (3.8 Ma) and its overly-
sequences underlying the Late Paleocene SB-2.1 ing 13.4-Ma flooding surface. Because paleontogical
(58.5 Ma) unconformity are defined in terms of control is scarce, for the sake of simplicity it was
T/R cycles because seismic permits only to identify decided to place the boundary of the last second-
downlap surfaces which were tentatively correlated order cycle at the base of the Upper Miocene. A
with the base of major transgressive shale packages major canyon-forming erosional event separates an
recognized in the stratigraphy of the Serranfa del Upper Miocene from a Pliocene-Pleistocene stack
Interior. Overlying SB-2.1, most second- and third- of sequences.
order cycles are defined by unconformities which The Upper Miocene is split into three third-
are best recognizable on the seismic profiles. Over order cycles, i.e. the SB-3.7-SB-3.8 (10.5-8.2 Ma),
much of the offshore the overlying Late Paleocene- the SB-3.8-Sb-3.9 (8.2-6.3 Ma) and the SB-3.9-
Eocene second-order cycle (Unit III) includes domi- SB-3.10 (6.3-5.5 Ma). The canyon-forming ero-
nantly pelagic units and is bounded by SB-2.1 (58.5 sional event correlated with a worldwide Messinian
Ma) and SB-2.2 (36 Ma). Note that SB-2.1 corre- unconformity which in eastern Venezuela may be
sponds in places to a significant hiatus with omis- enhanced by erosional processes associated with up-
sion of the Lower Paleocene and the Maastrichtian. lifts of the Venezuelan orogen. The Plio-Pleistocene
A maximum flooding surface perhaps corresponding (Unit VIII) is bounded at its base by SB-3.10, i.e.
to the 56.5-Ma maximum flooding event of Haq et the 5.5-Ma erosional event. Lack of adequate faunal
al. (1987) permits to tentatively split this second- control permits only a very tentative age assignment
order cycle into two third-order T - R cycles. The to six third-order cycles that are separated respec-
Oligocene second-order cycle (Unit IV) has only tively by SB-3.11 (4.2 Ma), SB-3.12 (3.8 Ma), SB
limited distribution in the area and is subdivided 3.13 (3.0 Ma), SB-3.14 (2.4 Ma), SB-3.15 (1.6 Ma)
into two third-order cycles bounded, respectively, by and SB-3.16 (0.8 Ma).
SB-2.2 (36 Ma), SB-2.3 (30 Ma) and SB-3 (25.5
Ma).
The inception of the Neogene foredeep due to the CONCLUSIONS
initial emplacement of the transpressional folded belt
of Venezuela and Trinidad is heralded by the basal The Eastern Venezuelan Basin and its offshore
foredeep unconformity SB-3 (25.5 Ma) which mod- continuation is a Neogene foredeep superposed on a
ifies the Cretaceous-Present first-order cycle over Mesozoic passive margin. There is limited evidence
much of the Eastern Venezuelan Basin. The Lower suggesting that Jurassic rifting affected the South
Miocene (Unit V) as a whole represents another sec- American Precambrian craton and some of its Paleo-

zoic cover. This Jurassic rifting event has also been In many cases paleontological control in eastern
reported from the Espino Graben in Venezuela and Venezuela is not adequate to provide a substantial
the Tacutu Graben of northeastern Brazil. calibration of the postulated cycles. Future work
The stratigraphy of eastern Venezuela and its should aim at refining the paleontological control
offshore is encompassed by a major Cretaceous- and reconciling detailed log interpretation with seis-
Cenozoic first-order transgressive-regressive cycle mic profiles that provide a higher resolution.
that begins with the Early Cretaceous, culminates Structural deformation in the Eastern Venezuelan
with the middle Turonian 91.5-Ma maximum flood- Basin includes: (a) pre-Early Miocene and Neogene
ing event and ends with an overall Turonian to Recent normal faulting involving the basement; (b) late
regression. In the Venezuela foredeep this first-order Neogene reverse faulting involving the sediments
cycle is interrupted by the inception of the foredeep and the basement; (c) extensive d6collement fold-
phase with a 25.5-Ma basal foredeep unconformity. ing and thrust faulting associated with the Barbados
The Cretaceous to Paleogene passive margin accretionary complex; (d) N-S-trending down-to-
stratigraphy is displayed on seismic profiles of the the-basin pre-Pliocene listric normal faulting in the
Venezuelan Atlantic offshore and can be correlated onshore area; and (e) extensive Neogene growth
with outcrop sections in the Serranfa del Interior and faulting associated with the Orinoco depocenter in
in northern Trinidad. The Cretaceous to Paleocene is the offshore area that is superposed on deeper base-
best subdivided into five second-order transgressive- ment-involved compressional folds.
regressive cycles bounded by a 131-Ma (basal Creta-
ceous) sequence boundary, four maximum flooding
surfaces with inferred ages of early Aptian (111 Ma), ACKNOWLEDGEMENTS
late Albian (98 Ma), middle Cenomanian (95 Ma),
middle Turonian (91.5 Ma) and a Late Paleocene We thank Intevep S.A., Lagoven S.A. and Cor-
sequence boundary (58.5 Ma). An Late Paleocene poven S.A., all subsidiaries of Petroleos de Ven-
to Eocene second-order cycle (58.5 Ma-36 Ma) ob- ezuela (ED.V.S.A.) for making the data for this
served in the offshore area includes a dominantly study available for publication and for generously
pelagic unit. The Oligocene second-order cycle is supporting this project. At Lagoven we particularly
characterized by narrow prograding siliciclastic sys- thank Carlos Sanchez, Yves Chevalier, Max Furrer
tems which correlate with the Los Jabillos, Areo and and Antonello Lilliu. At Corpoven we are indebted
Naricual formations of the Serranfa del Interior and to Emira Cabrera. Very special thanks go to our
some of the reservoirs of the giant E1 Furial-Carito colleagues at Intevep S.A., Felipe Audermard, Pablo
oil fields. The Oligocene can be subdivided in two Klar, Mojtava Taheri and Irene MacQuahe for con-
third-order cycles bounded respectively by sequence tinuous encouragement during this project. Our re-
boundaries of 36 Ma, 30 Ma and 25.5 Ma. The viewers Amos Salvador, Keith James and Paul Mann
Oligocene pinches out toward the craton, so that made many constructive comments which led to a
onshore the Oligocene is partially absent. Offshore profound restructuring of the original manuscript.
the downdip equivalent of the prograding sequence We particularly appreciate Paul Mann's friendly per-
appears to be absent in wells suggesting either sedi- severance and Mrs Elaine Bally's patient help in
ment starvation and/or excessive condensation. editing this paper.
A 25.5-Ma basal foredeep unconformity is related
to the sudden deepening of the passive margin in
response to the incipient emplacement of the folds REFERENCES
and thrust sheets of the Serranfa del Interior and
Trinidad. These in turn are related to the easterly Aden, L.J. and Bierley, R.E., 1996. Structural Development of
displacement of the Venezuelan orogen in response the Southern Basin, Onshore Trinidad: Implications for Hy-
to the oblique convergence of the Caribbean and drocarbon Entrapment (abstr.). Am. Assoc. Pet. Geol. Bull.,
South American plates. 80: 1267.
The Neogene foredeep fill of eastern Venezuela is Arnstein, R., Cabrera, E., Russomano, E and Sanchez, H., 1985.
Revisi6n estratigr~ficade la Cuenca de Venezuela Oriental. VI
best characterized by three second-order sequences Conf. Geol. Venez., Mem., 1: 41-69.
bounded by sequence boundaries with inferred ages Audemard, EE., 1991. Tectonics of Western Venezuela. Ph.D.
of 25.5 Ma, 16.5 Ma and 10.5 Ma. In the offshore Thesis, Rice University,Houston, Texas (unpubl.).
a 5.5-Ma unconformity is associated with deeply Audemard, EE. and Lugo, J., 1996. The Venezuelan Foredeeps:
incised submarine canyons. Prominent maximum Structure (abstr.). Am. Assoc. Pet. Geol. Annu. Conv., A7.
flooding surfaces are at 16 Ma and 13.6 Ma. About Audemard, F.E. and Lugo, J., 1997. Petroleum geology of Ven-
ezuela. Am. Assoc. Pet. Geol. Annu. Conv., Short Course
sixteen Neogene third-order sequences can be recog- Notes.
nized on seismic and are tentatively correlated with Audemard, EE., Azpiritxaga, I., Bauman, E, Isea, A. and La-
sequence boundaries proposed by Haq et al. (1987). treille, M., 1985. Marco geol6gico del Terciario de la Faja

Petrolffera del Orinoco de Venezuela. VI Congr. Geof., Venez., Am. Assoc. Pet. Geol. Bull., 80:1281.
1: 70-109. Chevalier, Y., Gonzales, G., Mata, S., Santiago, N. and Spano,
Av6 Lallemant, H.G., 1997. Displacement partitioning and ex- F., 1995. Estratigraffa secuencial y tect6nica del transecto E1
humation in the eastern Caribbean/South American plate Pilar-Cerro Negro, Cuenca Oriental de Venezuela. VI Congr.
boundary northeastern Venezuela. Tectonics, 16: 272-289. Colombiano del Petr61eo, pp. 115-125.
Azavache, A., Flinch, J., Giffuni, R., Martinez, R., Matta, S., Cramez, C.D. and Vail, ER., 1990, Stratigraphy Cycles and
Mitacchione, V. and Pereira, J.G., 1996. Tectono-stratigraphic Major Marine Source Rocks, Short Course, September 5,
evolution of the eastern Maturin Foreland Basin (Eastern Ven- 1994, Houston, Texas, 82 pp.
ezuela) (abstr.). Am. Assoc. Pet. Geol. Bull., 80 (8): 1270. Crux, J., Sarzalejo, S., Audemard, E, Bejarano, C., Cabrera, S.
Babb, S., Mann, E and Buffler, R.T., 1996. Three-stage slope and Funes, D., 1996. Timing of sequence deposition related to
reversal in northern and central Trinidad and its effect on the development of the Eastern Venezuelan basin (abstr.). Am.
megasequences in Cretaceous-Cenozoic basins (abstr.). Am. Assoc. Pet. Geol. Bull., 80: 1284.
Assoc. Pet. Geol. Bull., 80 (8): 1271. Daza, J. and Prieto, R., 1990. Fallas de crecimiento en el ~ea
Bally, A.W., 1989. Phanerozoic basins of North America. In: de Mapirito-Monagas Central. V Congr. Venezolano de Ge-
A.W. Bally and A.R. Palmer (Editors), The Geology of North ofisica, Caracas, 21-25 Oct., pp. 142-149.
America--An Overview. Geol. Soc. Am., Boulder, CO, Vol. Dercourt, J., Ricou, L.E. and Vrielynck, B. (Editors), 1993. Atlas
A, pp. 397-446. Thetys Paleoenvironmental Maps. Gauthier-Villars, Paris, 14
Bally, A.W. and Snelson, S., 1980. Realms of subsidence. In: maps, 1 pl., 307 pp.
A.D. Miall (Editor), Facts and Principles of World Petroleum Di Croce, J., 1995. Eastern Venezuelan Basin: Sequence Stratig-
Occurrence. Can. Soc. Pet. Geol. Mem., 6: 1-94. raphy and Structural Evolution. Ph.D. Thesis, Rice University,
Banks, L.M. and Driver, E.S., 1957. Geologic history of Santa Houston, Texas (unpubl.).
Ana structure, Anaco structural trend, Anzo~itegui, Venezuela. Dillon, W.E, Paul, C.K. and Gilbert, L.E., 1985. History of
Am. Assoc. Pet. Geol. Bull., 41: 308-325. the Atlantic continental margin off Florida: the Blake Plateau
Barr, K.W. and Saunders, J.B., 1965. An outline of the geology Basin. In: Geologic Evolution of the United States Atlantic
of Trinidad. Trans., 4th Caribbean Geol. Conf. Trinidad and Margin. Van Nostrand Reinhold, New York, pp. 189-215.
Tobago, pp. 1-10. Dillon, W.E, Scanlon, K.M., Valentine, EG. and Huggett, Q.,
Bartok, E, 1993. Prebreakup geology of the Gulf of Mexico- 1988. Deep-sea erosion of a carbonate platform margin - - the
Caribbean: its relation to Triassic and Jurassic rift systems of Blake Escarpment (abstr.). Geol. Soc. Am. Bull., 20: 130.
the region. Tectonics, 12 (2): 441-459. Eiras, J.E and Kinoshita, E.M., 1989. Soluqao de um Problema
Beaumont, C., 1981. Foreland basins. Geophys. J.R. Astron. Estratigr~ifico atrav6s da Integra~o Regional, Um Exemplo da
Soc., 65: 291-329. Bacia do Tacutu, Norte do Brasil. I Seminario de Interpretacao
Bejarano, C., Funes, D., Sarzalejo, S., Audemard, E and Flores, Exploratoria, Petrobras, 1989, Rio de Janeiro, pp. 243-251.
G., 1996. Sedimentary sequence evolution in a foredeep basin: Erikson, J.E and Pindell, J.L., 1993. Analysis of subsidence in
Eastern Venezuela (abstr.). Am. Assoc. Pet. Geol. Bull., 80: northeastern Venezuela as adiscriminator of tectonic models
1273. for northern South America. Geology, 21:945 -948.
Bellizia, G.A., Pimentel, M.N. and Bajo, O.R., 1978. Mapa Erlich, R.N. and Barrett, S.E, 1992. Petroleum Geology of the
geologico-estructural de Venezuela. Ministerio de Minas e Eastern Venezuela Foreland Basin. In: R.W. Macqueen and
Hidrocarburos, scale 1:500,000. D.A. Leckie (Editors), Foreland Basins and Fold Belts. Am.
Blow, W.H., 1969. Late middle Eocene to Recent planktonic Assoc. Pet. Geol. Mem., 55: 341-362.
foraminiferal biostratigraphy. 1967 Proc., Int. Conf. Planktonic Feo-Codecido, G., Smith, ED., Aboud, N. and Di Giacomo, E.,
Microfossils, Geneva, pp. 199-422. 1984. Basement and Paleozoic rocks of the Venezuelan Llanos
Bolli, H.M., Beckmann, J.E and Saunders, J.B., 1991. Ben- Basin. In: W.E. Bonini et al. (Editors), The Caribbean-South
thic Foraminiferal Biostratigraphy of the Southern Caribbean. American Plate Boundary and Regional Tectonics. Geol. Soc.
Cambridge University Press, Cambridge, 1230 pp. Am. Mem., 162: 175-187.
Bouysse, E and Westercamp, D., 1989. Subduction of Atlantic Flemings, EB. and Jordan, T.E., 1990. Stratigraphic modeling of
aseismic ridges and Late Cenozoic evolution of the lesser foreland basin: interpreting thrust deformation and lithosphere
Antilles Island Arc. Tectonophysics, 175:349 -380. rheology. Geology, 18: 430-434.
Carnevali, J.O., 1988. Venezuela nor-oriental: exploraci6n en Fox, EJ., Heezen, B.C. and Johnson, G.L., 1970. Jurassic sand-
frente de montafia. III Simp. Bolivariano, Soc. Ven. Geol., stone from the tropical Atlantic. Science, 170: 1402-1404.
Caracas, pp. 69-89. Funkhouser, H.J., Sass, L.S. and Hedberg, H.D., 1948. Santa
Carnevali, J.O., 1989. Geology of new giant oil fields in moun- Ana, San Joaqufn, Guario and Santa Rosa oilfields (Anaco
tain front of northeastern Venezuela. Proc., 28th Int. Geol. fields), central Anzo~itegui, Venezuela. Am. Assoc. Pet. Geol.
Congr., Washington, 9-19 July, 1, pp. 1240-1241. Bull., 32:1851-1908.
Carr-Brown, B. and Frampton, J., 1979. An outline of the stratig- Furrer, M.A., 1979. Paleontology of well A and B. Lagoven,
raphy of Trinidad. 4th Latin Am. Geol. Con. Field Guide, pp. S.A., Internal report.
7-9. Galea-Alvarez, F., Coriano, M. and Aponte, A., 1996. Facies, age
Case, J.E., Holcombe, T.L. and Martin, R.G., 1984. Map of and sedimentary environment of the E1 Cantil, Chimana and
geologic provinces in the Caribbean region. In: W.E. Bonini, Querecual formations, Pertigalete, northeastern Anzoategui,
R.B. Hargraves and R. Shagam (Editors), The Caribbean- Venezuela (abstr.). Am. Assoc. Pet. Geol. Bull., 80: 1291.
South American Plate Boundary and Regional Tectonics. Geol. Gallango, O. and Parnaud, E, 1995. Two-dimensional computer
Soc. Am. Mem., 162: 1-30. modeling of oil generation and migration in a transect of
Chevalier, Y., 1993. A cross-section from the oil-rich Maturin the Eastern Venezuelan basin. In: Tankard et al. (Editors),
sub-basin to Margarita Island. The geodynamic relations be- Petroleum Basins of South America. Am. Assoc. Pet. Geol.
tween South American and Caribbean plates. Tectonics and Mem., 62: 727-740.
stratigraphy. Am. Assoc. Pet. Geol. Int. Conf., Caracas, March George, R.E, Jr. and Sams R.H., 1993. Eastern Venezuelan basin
11-14, Field Trip, 165 pp. post-Jurassic evolution as a passive transform margin basin
Chevalier, Y. and Spano, E, 1996. Tect6nica del transecto E1 (abstr.). Am. Assoc. Petr. Geol. Bull., 77: 320.
Pilar-Delta Amacuro, Cuenca Oriental de Venezuela (abstr.). Giffuni, G. and Castro-Mora, M., 1996. Biostratigraphic se-

S E Q U E N C E S T R A T I G R A P H Y OF THE E A S T E R N V E N E Z U E L A N BASIN 475
quence analysis of Oligocene-lower Miocene sections in the Kugler, H.G., 1959. Geologic map of Trinidad and geological
Orocual Field, Eastern Venezuelan basin (abstr.). Am. Assoc. sections through Trinidad. Orell Ftissli, Zurich, E. Stanford,
Pet. Geol. Bull., 80: 1294. London, scale 1 : 100,000.
Gonzales, G., Mata, S. and Santiago, N., 1996a. Sequence Leonard, R., 1983. Geology and hydrocarbon accumulation,
stratigraphy of the Maturin Subbasin (Eastern Venezuela) (ab- Columbus basin, offshore Trinidad. Am. Assoc. Pet. Geol.
str.). Am. Assoc. Pet. Geol. Bull., 80: 1295. Bull., 67: 1081-1093.
Gonzales, G., Mata, S., Odehnal, M. and Escandon, M., 1996b. Liddle, R.A., 1928. The Geology of Venezuela and Trinidad.
Chronostratigraphic Charts of the Eastern Venezuelan basin Mac Gowan, Fort Worth, Texas, 552 pp.
(abstr.). Am. Assoc. Pet. Geol. Bull., 80:: 1296. Lilliu, A.G., 1990. Geophysical Interpretation of Maturfn Fore-
Gonzalez de Juana, C., 1947. Elements of diastrophic history of land, Northeastern Venezuela. MSc. Thesis, University of
northeastern Venezuela. Geol. Soc. Am. Bull., 58: 689-702. Houston, Houston, Texas (unpubl.).
Gonzalez de Juana, C., Arozena, J. and Picard-Cadillat, X., Lugo, J., 1991. Cretaceous to Neogene Tectonic Control on
1980. Geologfa de Venezuela y de sus Cuencas Petrolfferas. Sedimentation: Maracaibo Basin, Venezuela. Ph.D. Thesis,
Foninves Ed., Caracas, 1 and 2, 1051 pp. University of Texas at Austin (unpubl.).
Gouyet, S., Unternehr, E and Mascle, A., 1992. The Demerara Lugo, J. and Mann, R, 1995. Jurassic-Eocene tectonic evolution
Rise and French Guyana, western equatorial Atlantic Ocean of Maracaibo Basin, Venezuela. In: Tankard et al. (Editors),
(abstr.). 4th Eur. Assoc. Pet. Geol. Conf., Paris, pp. 131-132. Petroleum Basins of South America. Am. Assoc. Pet. Geol.
Hallam, J.M., Hancock, J.M., LaBraque, J.L., Lowrie, W. and Mem., 62: 699-725.
Channel, J.E., 1985. Jurassic and Cretaceous geochronol- Marffn-Bellizia, C., 1974. Paleotect6nica del Escudo de Guyana.
ogy and Jurassic to Paleogene magnetostratigraphy. In: N.J. IX Conf. Geol. Inter-Guyanas, Ciudad Guyana, Venezuela,
Snelling (Editor), The Chronology of the Geological Record. Memoria, Bol. Geol., Publ. Esp., 6, pp. 251-305.
Geol. Soc. London Mem., 10:118-140. Mascle, A. and Letouzey, R, 1990. The Geological Map of The
Haq, B.U., Hardenbol, J. and Vail, ER., 1987. Chronology of Caribbean. Technip, Paris, 2 Plates.
fluctuating sea levels since the Triassic (250 million years ago Mendoza, V., 1977. Evoluci6n tect6nica del Escudo de Guyana.
to the present). Science, 235:1156-1167. II Congr. Latinoamericano Geol., Caracas, Noviembre, 1973,
Hardenbol, J., Thierry, J., Farley, M.B., Jaquin, T., de Grazian- Memoria, Bol. Geol., Caracas, Publ. Esp., 7, pp. 2237-2270.
sky, P.C. and Vail, ER., in press. The Mesozoic-Cenozoic Miall, A.D., 1995. Collision-related foreland basins. In: C.J.
chronostratigraphic framework. In: EC. De Graziansky, J. Busby and R.V. Ingersoll (Editors), Tectonics of Sedimentary
Hardenbol, T. Jacquin, ER. Vail and M.B. Farley (Editors), Basins. Blackwell, London, pp. 393- 424.
Sequence Stratigraphy of European Basins. Spec. Publ. Soc. Mitchum, R.M. and Van Wagoner, J.C., 1991. High frequency
Econ. Paleontol. Mineral. sequences and eustatic cycle in the Gulf of Mexico Basin.
Harland, W.B., Cox, EG., Craig, L.E., Smith, A.G. and Smith, In: J.M. Armentrout and B.E Perkins (Editors), Sequence
D.G., 1990. A geologic time scale 1989. Cambridge University Stratigraphy as an Exploration Tool - - Concepts and Practices
Press, New York, 265 pp. in the Gulf Coast. Gulf Coast Section, Society of Economic
Hedberg, H.D., 1937. Stratigraphy of the rfo Querecual section Paleontologists and Mineralogists Foundation, Program and
of northeastern Venezuela. Geol. Soc. Am. Bull., 48: 1971- Abstracts, Eleventh Annual Research Conference, Houston,
2024. Texas, pp. 257-267.
Hedberg, H.D., 1942. Mesozoic stratigraphy of northern South Murany, E.E., 1972. Tectonic basis for the Anaco fault. Am.
America. Am. Sci. Congr. VIII, U.S.A., 1940, Proc. IV, pp. Assoc. Pet. Geol. Bull., 56: 860-870.
195-227. Palmer, V.W., 1945. Fossil fresh-water mollusca from the State
Hedberg, H.D., 1950. Geology of the Eastern Venezuelan basin of Monagas. Bull. Am. Paleontol., 3: 1-34.
(Anzo~tegui-Monagas-Sucre-Eastern Gufirico portion). Geol. Parnaud, E, Gou, Y., Pascal, J.C., Truskowski, I., Gallango, O.,
Soc. Am. Bull., 61: 1173-1216. Passalacqua, H. and Roure, F., 1995. Petroleum Geology of
Hedberg, H.D. and Pyre, A., 1944. Stratigraphy of northeastern the central part of the Eastern Venezuelan Basin. In: Tankard
Anzo~tegui, Venezuela. Am. Assoc. Pet. Geol. Bull, 20: 1- et al. (Editors), Petroleum Basins of South America. Am.
128. Assoc. Pet. Geol. Mem., 62: 741-756.
Hedberg, H.D., Sass, L.C. and Funkhouser, H.J., 1947. Oilfields Passalacqua, H., Fernandez, F., Gou, Y. and Roure, E, 1995.
of the Greater Oficina area, central Anzofitegui. Am. Assoc. Crustal architecture and strain partitioning in the Eastern Ven-
Pet. Geol. Bull., 31: 2089-2169. ezuelan Ranges. In: Tankard et al. (Editors), Petroleum Basins
Heezen, B.C. and Freeman-Lynde, R.E, 1976. Sheet 22, Atlantic of South America. Am. Assoc. Pet. Geol. Mem., 62: 667-679.
Ocean 1 : 100,000. In: G. Choubert and A. Faure-Muret (Edi- Persad, K.M., 1979. Outline of the Geology of the Trinidad Area.
tors), Geological World Atlas. Commission for the Geological Trans., 4th Latin American Geol. Conf., Trinidad and Tobago,
Map of the World, UNESCO. pp. 738-758.
Huang, T., 1996. Nannofossil and sequence chronostratigraphy of Persad, K.M., 1985. Presentation of a new tectonic map (and
a marine flooding surface in the Turonian of Trinidad (abstr.). accompanying sections) of Trinidad and Tobago. 1st Geol.
Am. Assoc. Pet. Geol. Bull., 80:1302. Conf. Geol. Soc. Trinidad Tobago, Trans., pp. 1-6.
International Stratigraphical Lexicon, 1956. Boletfn de Geologfa. Petroconsultants, 1989. Guyana Basin--Geology and Petroleum
Special Publication No. 1, Ministerio de Minas e Hidrocar- Potential. Petroconsultants, Report, pp. 1-11.
buros, Editorial Sucre, Caracas, 664 pp Pindell, J.L., 1985. Alleghenian reconstruction and the subse-
Jankowsky, W.J. and Schlapak, G., 1983. Guyana offshore. In: quent evolution of the Gulf of Mexico, Bahamas and Proto-
A.W. Bally (Editor), Seismic Expression of Structural Styles. Caribbean Sea. Tectonics, 4: 1-39.
Am. Assoc. Pet. Geol., Stud. Geol., 15 (2): 2.2.3-47-2.2.3-50. Pindell, J.L. and Barrett, S.F., 1990. Geological evolution of the
Jordan, T.E., 1995. Retroarc foreland and related basins. In: C.J. Caribbean region; a plate-tectonic perspective. In: G. Dengo
Busby and R.V. Ingersoll (Editors), Tectonics of Sedimentary and J.E. Case (Editors), The Caribbean Region. The Geology
Basins. Blackwell, London, pp. 331-362. of North America, Vol. H, Geol. Soc. Am., pp. 405-432.
Kugler, H.G., 1953. Jurassic to Recent Sedimentary Environ- Pindell, J.L. and Dewey, J.E, 1982. Permo-Triassic reconstruc-
ments in Trinidad. Bull. Assoc. Suisse Geol. Ing. Petrol., 20: tion of western Pangea and the evolution of the Gulf of
27-60. Mexico/Caribbean region. Tectonics, 1: 179-212.

476 J. DI C R O C E et al.
Prieto, R., 1987. Seismic Stratigraphy and Depositional Systems Tankard, A.J., 1985. On the depositional response to thrusting
of the Orinoco Platform Area. Ph.D. Thesis, University of and lithospheric flexure: examples from the Appalachian and
Texas, Austin, Texas (unpubl.). Rocky Mountain basins. In: P.A. Allen and P. Homewood
Renz, H.H., Alberding, H., Dallmus, K.E, Patterson, J.M., Ro- (Editors), Foreland Basins. Int. Assoc. Sedimentol. Spec. Pub.,
bir, R.M., Weibord, N.E. and Masvall, J., 1958. The Eastern 8: 369-395.
Venezuela Basin. In: L.G. Weeks (Editor), Habitat of Oil. Am. Vail, P.R. and Wornardt, W.W., 1990. Well log-seismic sequence
Assoc. Pet. Geol., Symp., pp. 561-600. stratigraphy: an integrated tool for the 90's. In: J.M. Armen-
Ross, M.I. and Scotese, C.R., 1988. A hierarchical model of the trout and B.F. Perkins (Editors), Sequence Stratigraphy as an
Gulf of Mexico and Caribbean Region. Tectonophysics, 155: Exploration Tool n Concepts and Practices in the Gulf Coast.
139-168. Gulf Coast Section, Society of Economic Paleontologists and
Rossi, T., 1985. Contribution ~i l'6tude Geologique de la Fron- Mineralogists Foundation, Program and Abstracts, Eleventh
tibre Sud-Est de al Plaque Caraibe: la serrania del interior Annual Research Conference, Houston, Texas, pp. 307-320.
(Venezuela) sur le Transect Cariaco Mathrin. Ph.D. Thesis, Vail, P.R., Mitchum, R.M. and Thompson, S., III, 1977. Seismic
Univ. Bretange Occidentale. Brest, 338 pp. stratigraphy and global changes of sea level, part 3. Relative
Rossi, T., Stephan, J.E, Blanchet, R. and Hernandez, G., 1985. changes of sea level from coastal onlap. In: C.E. Payton
Etude g6ologique de la Serranfa del Interior oriental sur le (Editor), Seismic StratigraphymApplication to Hydrocarbon
transect Cariaco-Maturin. Rev. Inst. Fr. Pdt. Paris. Technip, 42 Exploration. Am. Assoc. Pet. Geol. Mem., 26:63-81.
(1): 3-30. Vail, P.R., Audemard, E, Bouwman, S.A., Eisner, P.N. and
Roure, E, Carnevali, J.O., Gou, Y. and Subieta, T., 1994. Geom- Perez-Cruz, G., 1991. The stratigraphic signatures of tec-
etry and kinematics of the North Monagas thrust belt (Venezu- tonics, eustacy and sedimentology n an overview. In: G.
ela). Mar. Pet. Geol., 11: 347-362. Einsele et al. (Editors), Cycles and Events in Stratigraphy.
Salvador, A. and Stainforth, R.M., 1968. Clues in Venezuela Springer-Verlag, Berlin, pp. 617-659.
to the geology of Trinidad and vice versa. 4th Geol. Conf. Veeken, P.C.H., 1983. Guyana Basin, offshore Northeast South
Caribe, Trinidad, pp. 31-40. America. Guyana offshore. In: A.W. Bally (Editor), Seismic
Sams, R.H., 1995. Interpreted sequence stratigraphy of the Los expression of structural styles. Am. Assoc. Pet. Geol., Stud.
Jabillos, Areo and (subsurface) Naricual formations, northern Geol., 15 (2): 2.3.3-1-2.3.3-2.
Monagas (area), Eastern Venezuelan basin. Bol. Soc. Ven. Villaroel, V., 1993. Santa Rosa Field, Venezuela, Eastern Ven-
Geol., 20 (1-2): 30-40. ezuela Basin. In: N.H. Foster and E.A. Beaumont (Editors),
Shepard, EE, 1981. Submarine canyons: multiple causes and Treatise of Petroleum Geology, Atlas of Oil and Gas Fields.
long-time persistence. Am. Assoc. Pet. Geol. Bull., 65: 1062- Am. Assoc. Pet. Geol., Structural Traps, VIII, pp. 193-210.
1077. Villeneuve, M. and Komara, S., 1991. Lower Paleozoic
Speed, R.C., Westbrook, G.K., Biju-Duval, B., Ladd, J.W., Mas- transgressions and regressions in the Bove Basin (Guinea
cle, A., Moore, J.C., Saunders, J.B., Schoomaker, J.E. and and Guinea Bissau, Africa) Stratigraphic, sedimentologic and
Stein, S., 1984. Lesser Antilles arc and adjacent terranes. Atlas paleogeographic data. J. Afr. Earth Sci., 12: 67-77.
10, Ocean Margin Drilling Program, Regional Atlas Series. Villeneuve, M., Diallo, M.C., Keleba, F., Kouroma, S., Paris, F.
Marine Science International, Woods Hole, 27 sheets. and Racheboeuf, P.R., 1989. New paleontological data from
Sprague, A.R., Lame, D., Faulkner, B., Sykes, M.A. and Aden, the Paleozoic series of the Bove Basin (Guinea, west Africa):
L.J., 1996. Sequence stratigraphy, facies architecture and reser- stratigraphical consequence. Compt. Rend. Ser., 309 (16):
voir distribution, Cretaceous lowstand fan reservoirs, Southern 1583-1590.
Basin, onshore Trinidad (abstr.). Am. Assoc. Pet. Geol. Bull., Vivas, V., 1986. Contribuci6n al estudio de la Serranfa Ori-
80 (8): 1338. ental (Venezuela), estratigraffa y tect6nica de la regi6n de
Stephan, J.E, De Lepinay, B.M., Calais, E., Tardy, M., Beck, C., Berganffn-Santa Ines (Edo Anzo~itegui). Thesis 3me cycle,
Carfantan, J.C., Olivet, J.L., Vila, J.M., Bouysse, E, Mauffret, Univ. Bretagne Occidentale (unpubl.).
A., Borgois, J., Thery, J.M., Tournon, J., Blanchet, R. and Wu, S., Bally, A.W. and Cramez, C., 1990. Allochthonous salt,
Dercourt, J., 1990. Paleogeodynamic maps of the Caribbean: structure and stratigraphy of the north-eastern Gulf of Mexico:
14 steps from Lias to Present. Bull. Soc. G6ol. Fr., 6: 915-919. structure. Mar. Pet. Geol., 7 (4): 334-370.

Chapter 1 7
Structure of the Gulf of Paria Pull-Apart Basin

(Eastern Venezuela-Trinidad)
J.F. FLINCH, V. RAMBARAN, W. ALI, V. DE LISA, G. HERNANDEZ,

K. RODRIGUES and R. SAMS
The Gulf of Paria lies entirely within the broad strike-slip plate boundary zone of the southeastern Caribbean. The structure
of the Gulf consists of a complex set of transtensional basins superimposed on a fold-and-thrust belt (Serranfa del Interior of
eastern Venezuela). This province extends to the east into the Caroni Basin of northern Trinidad and to the west into the San
Juan graben and other minor extensional basins of the Guanoco area. The main structural elements of the Gulf of Paria are: the
Casanay-Arima fault bounding the transtensional province to the north, the Warm Springs fault to the south and the Domoil and
Gopa Highs located in the central part of the Gulf of Paria. Rapid extensional collapse since the late Neogene resulted in deep
half-grabens, which are characterized by large offsets along the major extensional faults and by shallow-water sedimentary fill.
Transtensional tectonics in the Gulf of Paria was coeval with north-vergent thrusting in the Pedernales region to the south. The
north-vergent imbricates constitute a passive-roof duplex with respect to the underlying south-vergent thrusts of the Serranfa del
Interior of eastern Venezuela. Late Pleistocene-Holocene compression in the Gulf of Paria has caused minor positive inversions
involving the main basin-bounding faults. Seismic data used in this study have enabled us to significantly modify existing wrench
models.
REGIONAL SETTING lan Cordillera, forming the San Juan graben and
the Guanoco area (northern Guarapiche province),
The Gulf of Paria is located east and downplunge where normal faults overprint thrust sheets of the
of the eastern Venezuelan fold-and-thrust belt (Ser- eastern Venezuelan fold-and-thrust belt (Fig. 1). The
ranfa del Interior) (Fig. 1). Its northern limit is the structure of the Gulf of Paria extends eastward into
Paria Peninsula in Venezuela and the western ex- onshore Trinidad, west of the Barbados accretionary
tension of the Northern Ranges of Trinidad (Fig. 2). wedge (Fig. 1).
The Gulf of Paria is the link between the Serranfa del
Interior of Venezuela and the Central and Southern The South Caribbean plate boundary
Range of Trinidad.
The Paria Peninsula and the Northern Range con- The Gulf of Paria lies in the broad South
sist of allochthonous metamorphic terranes, assigned Caribbean Plate boundary (Burke et al., 1984).
to the Cordillera de la Costa Nappe (Bellizzia and The Caribbean Plate is moving in a fight-lateral
Dengo, 1990). This nappe consists of pre-Meso- sense, with respect to the South American Plate.
zoic basement rocks (Sebastopol Complex) uncon- While most of the eastward escape of the Caribbean
formably covered by thick Jurassic-Lower Creta- Plate is accommodated in the north by the Cayman
ceous metasediments (Caracas Group) and meta- pull-apart and the strike-slip systems of Motagua
morphosed igneous rocks, intruded by large granitic (Guatemala) and Jamaica, the southern boundary is
plutons (Bellizzia and Dengo, 1990). The Cordillera taken by several strike-slip systems, the F a l c o n -
de la Costa Nappe has been considered as an ac- Aruba (Macellari, 1995), Cariaco Trough (Schubert,
cretionary complex associated with oblique tectonic 1982; Mann and Burke, 1984) and the Gulf of Paria
convergence (Case et al., 1990) or a collisional arc being the most significant ones.
complex (Burke, 1988). The extensional faults of the In the past many geoscientists considered that
Gulf of Paria extend further west into the Venezue- the E1 Pilar fault represented the Caribbean-South

478 J.E FLINCH et al.
64 ~ 63 ~ 62 ~ 61 ~ 60 ~ 59 ~
11~
lo ~
8~
[, - .. ] Accretionary t- ---[ Barbados Accretionay wedge

metamorphic terranes ~ Thrust Fault
[~il Gulf of Paria ..1 Maturfn Foreland Basin

Extensional province Normal Fault
l Deformed South-American
[~_~] Undeformed Passive Margin ~,, Strike-slip fault
Passive Margin
Carapita-Nariva
[+++++ 1 GuayanaShield ...................................Shale Ridge
Accretionary Wedge
Fig. 1. Structural sketch map of the southeastern Caribbean region. The boundaries of the passive margin and the outline of the Barbados
accretionary wedge are traced after Di Croce (1995), the structure of the Maturfn Basin is based on LAGOVEN Internal Reports and the
structure of the Gulf of Paria is based on this study. Notice the location of the study area. A F = Arima fault; C F = Casanay fault; C R F --
Central Range fault; L B F = Los Bajos fault; S F F = San Francisco fault; S J G = San Juan graben; U F = Urica fault.
American Plate boundary (Molnar and Sykes, 1969; Many geoscientists as an example of a strike-slip
Schubert, 1982; Soulas, 1986). According to Vier- basin have cited the Gulf of Paria. In most of these
buchen (1984), fight-lateral displacement of at least models the E1 Pilar fault plays a major role as
150-300 km is required to explain the gravimetric the main northern bounding fault of the strike-slip
field distribution. Nevertheless, the E1 Pilar fault system (Fig. 2a,b); according to all models (Fig. 2 a -
has an ambiguous field expression, and shows no c), the Los Bajos fault as well as other N W -
evidence for significant right-lateral displacement. SE-trending faults, represent strike-slip faults related
According to Ball et al. (1971) and Gonzalez de to right-lateral transcurrent systems. Under these
Juana et al. (1980) the E1 Pilar fault is a steeply schemes, the Los Bajos fault joins the E1 Pilar
dipping normal fault that constitutes the southern fault with the Southern Range of Trinidad. The
boundary of the Cariaco Trough, while the Mor6n data presented in this study (Fig. 2d) significantly
fault would be its northern limit. Seismicity data do constrain the current models on the Gulf of Paria
not support continuity of the E1 Pilar fault through and the entire southeastern Caribbean region. Most
Trinidad (Speed, 1985). Burke (1988) considers the of the models shown on Fig. 2 are based on much
plate boundary to be at least 200 km or more wide. less data than our model which integrates data from
The strike-slip motion is distributed among several Venezuela and Trinidad. A common feature is that
faults, one of them being the North Coast fault zone, they all emphasize the role for the Los Bajos fault as
located north of Trinidad (Robertson and Burke, a link between the E1 Pilar fault and the South Coast
1989). fault (Salvador and Stainforth, 1968; Munro and
Smith, 1985; Algar, 1995). Most of these models
Previous work were based on a limited seismic data set. The
structural relationships presented by these models
Several models have been proposed to explain are not consistent with the new data set based on data
the complex structure of the Gulf of Paria (Fig. 2). from Trinidad and Venezuela presented in this study.

CARIBBEAN SEA CARIBBEAN SEA
El Pilar Fault LF El Pilar Fault ULF Spain
PARI
% % 9.....
Maturin
(a) 100 km (b) 100 km
CARIBBEAN SEA CARIBBEAN SEA

Northern Ran(
Arima
Casana,
El Pilar Fault GULF El Pilar Fault GULF
San Juan
Grabe~
. . . . .

100 km
Fig. 2. Four tectonic models of the Gulf of Paria in a time perspective: (a) Salvador and Stainforth (1968); (b) Munro and Smith (1985); (c) Algar (1995); (d) this study. Notice main differences regarding
the role of Los Bajos fault and the continuity of E] Pilar fault through Trinidad.
480 J.E F L I N C H et al.
STRATIGRAPHY OF THE GULF OF PARIA STRUCTURE
The stratigraphy of the Gulf of Paria is strongly A regional cross-section through the Gulf of Paria
controlled by its structural evolution. The stratigra- (see Fig. 1 for location) reveals the presence of two
phy of this region can be subdivided into a lower distinct structural provinces: a Southern Compres-
fold-and-thrust belt stratigraphy and an upper North- sional Zone and a Northern Transtensional Basin
em Transtensional Basin fill and the Southern Com- (Fig. 4). Both units are superimposed on an under-
pressional Zone basin successions (Fig. 3). With lying fold-and-thrust belt. These units are separated
respect to the folded belt, this is the area where the by a major fault system: the Warm Springs fault
classical Venezuelan and Trinidadian stratigraphic (Flinch and Rambaran, 1996) (labeled WSF in the
units meet. The superimposed transtensional fill has sections).
been stratigraphically subdivided in Trinidad, where Fig. 5 is a detailed structural map integrating data
most of the units are widely exposed on the is- from the Venezuelan and Trinidadian part of the Gulf
land. In this work we will use the classical Trinidad of Paria and the island of Trinidad. The map shows
stratigraphic names for the Northern Transtensional the most relevant present-day structural features of
Basin. The Southern Compressional Zone is oc- the region and the main tectonic provinces.
cupied by sedimentary basins that have equivalent
stratigraphy in the eastern Venezuelan Basin and Regional seismic sections
in southern Trinidad. Table 1 is based on twelve
wells in the area (see Fig. 5) and summarizes the The structure of the Gulf of Paria is illustrated by
stratigraphy of the region. a series of regional seismic profiles (Fig. 6). The
Southern Compressional Zone [ Northern Transtensional Basra

AGE S N
: .'_."2.'2.'2:- "2..2.'2 - f...2 : ~ ~ - " " . ~ ~-- "-.*: ( . " : : - : - - : ~ : - ' ~ ' L : 2 , i . f , ' 3 ; , ' ~ ;
LLI P L E I S T o . '';''~"~-'~'"~-''~;-':':::":':::"::"-'~
.- 9 " .-
9 .'.'.':..'..--. -- .-~ - . % . --
" .#
.':.-'.IL'..-
- 11D.il. - " ,dl
. k T . . ' . 1
'
Central
(.9 UPPER Campana Highs
0 0 High
Z -Q MIDDLE Paria
LOWER Peninsula
OLIGOCENE
uJ
(.9
O EOCENE L ...... ~,
iii
n PALEOCENE
O3
8Iii UPPER
O
<
I / / / ~ I / '
LOWER .. . / / X- - \ ,
/~ A A A A A A A A A A A A A A A A A A A, Ak\ "~ ,, \~ i
O , A A A A A A A A A A A A A A A A A A A A
~.~ ~ /x /k /k A. A. /k /k /k A.
,#k,, \
LEGEND ~'~"~'1 Marls Ii~o~ ol Fan-delta

conglomerates
Platform limestones
Deep-water Fluvial-alluvial
t i : ! : ! I Shallow-water shales sandstones
sandstones
1------ ~1 Shallow-water I ', 2 "1 Metamorphic Rocks
shales and claystones
Deep-water ABSENT
sandstones A n Evaporites I I /.,atus or erosion/
Fig. 3. Schematic litho-stratigraphic north to south section across the Gulf of Paria.

S T R U C T U R E OF THE G U L F OF PARIA P U L L - A P A R T B A S I N 481
Table 1
Description of the main stratigraphic units of the Gulf of P a r i a - correlation between eastern Venezuela and Trinidad
Thickness
Tectonic
Age Formation Lithology Environment Range
Events
(meters)
Paria/Mesa Mudstones,sandstones and occasional Fluvial-alluvial deposits 900-100 Inversion
PLEISTOCENE Erin/Morne L'Enfer conglomerate beds Delta Plain
o Transtension
PLIOCENE ~
._ Springvale ~. Clays and glauconitic sandstones Fluvio-deltaic ,,~ 800-200
13_ c
neritic N-vergent
Manzanilla ~ Calcareous silt, glauconitic sandstones
._1
1700-0
thrusting
UPPER La Pica/Forest Cruse Sandstones and siltstones with interbedded shales fluvial-littoral
Lengua Dark greenish grey marl rich in foraminifera

MIDDLE Deep-water deposits
Foredeep
9~. 9 grasso Shales, blue-grey mudstones and occasional 1500-0
LOWER ~
Cipero
sandstones lenses or channels Herrera, Retrench and
Karamat
S-vergent
O
Nariva
Calcareous clays and marls with abundant
foraminifera fauna Sandstones
Turbiditic sandstones thrusting
OLIGOCENE
San Fernando Glauconitic calcareous clays, silt sands and boulder beds
Caratas/Navet Light grey and greenish grey marls, white chalky
Pro-delta 600-0
EOCENE
? marls and occasional radiolaria rich marly-clays
Pyrite-bearing fossiliferous black shales and
mudstones, limely siltstones and glauconitic sandstones Outer shelf-slope 500-0
PALEOCENE > Lizard Springs Calcareous shale Deep-water c.-

San Juan Massive, yellow, fine to medium grained quartzy
sandstones Platform, coastal
o~ , San Antonio Siliceous shales and sandstones with chert levels
LIJ 700-0
O~
= Querecual/
c~ Nodular and micritic black to grey limestones and Deep-water
0
w Naparima Hill occasional black mudstone and siliceous claystone >
0
<
b- Chimana Micritic and bioclastic limestones, occasional sandstones u)
u.i El Cantil and calcareous shales Platform, r e e f a l 1600-600
iii Interbedded sandstones, limestones, foraminifera rich (Delta front-Prodelta 13_
garranquin/ black shales and reddish brown clay 700-500
Cuche Black silty micaceous shale with carbonaceous laminae Slope to deep-water
Couva Evaporites, mainly gypsum and anhydrite shallow-water coastal 500-0
Southern Compressional Zone I Northern Transtensional Basin

N
NW S
Gopa High Paria Peninsula
Pedernales Passive-Roof Duplex
0
1
2
3
4 4
5
6
7
8
9
Extensional Basin Fill Serrania del Interior

Maturin Foreland Basin Fold-and- Thrust Belt
Paleogene- AIIochthonous Metamorphic Terranes
Pleistocene
~ MiddleMiocene I:,-- ,t (Cordillera de la Costa Nappe)
Pliocene Cretaceous
Couva Evaporites WSF WarmSprings Fault
Upper Miocene I ! Pre'CretaceOus CAF Casanay-Arima Fault
Fig. 4. Generalized cross-section through the Gulf of Paria, showing the main structural provinces. The section is approximately three
times vertically exaggerated (V ~ 3H). See Fig. 1 for location. The section displays the relationship between the northern extensional
basin with the southern compressional zone and the underlying folded belt units of South America and the allochthonous metamorphic
terranes of Caribbean affinity.

sections below describe the main structural units and Pleistocene section is rotated due to normal fault-
tectonic elements of the region from west to east. ing, but the amount of growth is not significant.
Flat reflectors onlapping onto folded beds suggest
Section I that the northern half-graben has undergone positive
This N-S-trending section illustrates the contact Pleistocene inversion on its southern flank (the top
between the transtensional system of the central Gulf of the Pliocene is deformed). Pleistocene sediments
of Paria (Northern Transtensional Basin) and the directly onlap the Domoil and Gulf Highs dating the
Southern Compressional Zone. The contact between end of tectonic activity.
the two structural units is a set of northward-dipping
normal faults, referred to here as the Warm Springs Section IV
fault (WSF). To the north, mostly south-dipping listric This N-S- to NE-SW-trending section shows the
normal faults account for the rotation of the overly- contact of the allochthonous metamorphic terranes
ing Neogene succession. North-vergent thrusts and with the eastern Venezuelan fold-and-thrust belt that
related ramp anticlines are present in the southern part underlies the present-day basin. This northernmost
of the profile. The geometry of the piggyback basins area is occupied by a south-vergent thrust system.
suggests thrust propagation from south to north. The Thrust faults are offset by a narrow half-graben
relationships between thrusting and folding to normal previously described in section III, to the south
faulting suggest nearly coeval north-directed thrust- is the Domoil High. Seismic data suggest south-
ing and transtension. The contact between the basin ward thrusting within the Gulf and Domoil Highs
and the allochthonous metamorphic terranes is a (see sections III and IV of Fig. 6). Pleistocene
steeply dipping south-vergent normal fault referred sediments onlap onto the Lower Cretaceous strata,
to as the Casanay fault (see Fig. 5). which belonged to the underlying folded belt. This
is one of the few seismic evidences of south-verging
Section II thrusting underneath the present-day basin. South
This N-S section again shows the boundary be- of the Domoil High, down-to-the-basin listric to
tween the Southern Compressional Zone and the steeply dipping normal faults offset the Cretaceous
Northern Transtensional Basin, represented by the to Pliocene section. Miocene shale constitutes an
WSE From south to north this section shows south- important extensional decollement level (Brasso and
dipping compressional imbricates of post-Middle Nariva Formations). A shale ridge (well H) occupies
Miocene sediments. The seismic data suggest a the central part of the extensional basin. South of
basal decollement around 2.4 s (TWT). The imbri- these wells the southern branch of the extensional
cates were penetrated by well C which shows dupli- system is represented by normal faults that sepa-
cation of the Pliocene section. Thrust imbricates are rate a thick Upper Miocene-Pliocene section to the
offset by northward-dipping normal faults that con- north from a compression anticline constituted by
stitute the boundary of the Northern Transtensional Oligocene-Cretaceous strata.
Basin. The Gopa High (see Fig. 5 for location)
occurs in the central part of this large extensional Section V
basin. Well E penetrated the Gopa High where the This NE-SW to NW-SE zigzag section shows
Miocene section is most shallow. South- and north- the western part of the Avocado-Couva High. This
dipping normal faults delineate this high. Thus, the high consists of a WNW-ESE-trending anticline
Northern Transtensional Basin consists of a southern related to south-vergent thrusting. According to well
half-graben, the central Gopa High and a northern data the core of this high is occupied by the Couva
half-graben. To the north, southward-dipping normal evaporite. The prominent shale ridge, already seen
faults separate the Domoil High from the northern in the previous section, occupies the lower part of
half-graben of the basin. the half-graben fill. Notice the large offset along
the Warm Springs fault in the southern part of the
Section III profile. This fault can be pictured as the master fault
This N-S-trending section shows that the Do- of the transtensional system.
moil and Gulf Highs are separated by a narrow
half-graben. In map view (Fig. 5) this half-graben Section VI
has a NW-SE trend. South of the Domoil High the This roughly N-S-trending section shows the
section offers a dip view of the normal faulting style. structure of the Northern Transtensional Basin. The
Listric normal faults sole out into a shaly Lower Warm Springs fault system bounds the basin to
Miocene section that constitutes the main exten- the south, constituting the master fault of the ex-
sional detachment. Shale ridges bounded by steeply tensional system. Southward-dipping listric normal
dipping normal faults occupy the central part of the faults offset the Cunapo Conglomerate in the north
extensional system. The overlying Upper Miocene- and the Manzanilla and Springvale Formations in

/ s e q ! . 12.3/O]1200[/|U'J~!AOSIWA~A~A~//:dl]q 112 IJOlSJ~A JOlO~) "UO!112301 JOJ l "8!cl oaS "£pnls slql u! p~sn SllOA~ ~ql j o uo!112oo I ~q] pu12 slu~m~l~ 112Jn13nJls Ut121.I.t ~4:1 s~u!lmO
deuJ ~qj. "£pnls s!q] uo pas12q £[oJ!lUO st 12!J12d j o j[n O oql j o d e m 112Jnl:mJlS oqj, " ( I 9 6 I ' £ g 6 I ) Jol~nN uJoJj p ~ g ! p o u J pue p~U!lduJ!s s! pu121s! P12P!U!-LL ~ql j o ,~8OlO~8 ~qj, "p12p!u!Ji pu12 e!J12d j o j l n O ~ql j o deuJ |eJnlonJ] S "g "~!cI
line=! sGu!Jds uJJeAA _,ISAA euilouAs " . . . . . . .
llne=l so!eEl soq 4 8 q eGp!l=l eleqs

eU!lO!luv ~"
llne_,l/£eueseo _-!0 llnej d!!s-e>l!JlS ~ .
IIne-I eU.l!JV -IV Sl,lne~ ],snJql ],ueGJeA pJeMqlnos ~,
------- eueoo!iAI Jaddn eseEi seoueoloA pnlA! eullou/[s ~r euilOllUV +

se6p!EI eleqs ~ '
..~_- oueoO!ld eseEt
-. . . . . . eueools!elcl eseEt eu!l,')u/~s / eu!lo!luv UO!SJeAUl ~ / " >- sllnel lsnJqx lueBJe^ pJe~qlnos
+%
~,
]<lnet pelJeAUl ~ i ~ ou!IouXs [ eu!lo!luv v

s e ! 7! w J oj u o o u n J O[e lN sqG!H q
llnel IeLUJON f sMoq e sllneJ lsnJq/IUE)6JeA pJeMqlJON J
SOU18JJE)_L , .
o!qdJoLuelelAi snouoqiqoOllV I-':.-:-t u!seE! leUO!SUelSUeJl uJaqlJoN
- uenp' ,
~i-ii~:;i~i!!i!~ii!~iiiiiii!~;iii!!i~i!~iii!i!iiii~ii~iiiiii~iiii~;iii~!~!~i~i!~!!~ii~i!~!i!~!!~!i~ii!~!~i!
i:i~!iii!iiii!iiiiiiii!i:iii!i;i!ii
~iiTi!ii!!!iT!iT!i!!iiiii!iiiili~i~!ii!7~iiii!iT~i!iiiii!ii!iii!i!:i~iii!~
4SAA
S/V,
0
.._.I
! !/7: :: ! (
/ i I# .'" =========================== kxC_ - _ ......
l
I
t ,:
• lit
# u o a e o ,~ ~- / /- --. " I
-- "~
_
"-
j
/
,7-(- "" '" ~
.
¢
.
-- I
. .
/"
.
/
.
""
.
~,
.
~
.
"-~
.
.;
.
./
.
" ~ i
:,:~: ...... tl
S
ti
li
o
,
~P
•" . . . . . . tt U
U I l ~e ¢ CU
. . .
JO ,UOd
. . .
.
~,
. . . ':::=======================================:::::::
.. .., . ~ .... I ! -,, I" -
.
l
-
•
-. ~
'
- s
-
- . .
'1 ""
_
-+" "" \
~
\
_
....
; .....
~ "
_ .-. j
x -
.....
~> ""
s i :"
I ,,,s .~ i~. z ~ i i / -
/ \ 7 ~ .I i \ .. i / ~ / I ~ ;, i ~ / 1 \ s ". / ~ I / i % , t t I \ x '. / I l i ~ x .
I s ,,
....
.- I / "1 x / \ / ~ t I i
., x i " ]
....... .-" t
, 7 ...
. . . . . j.
I
. /
. . !
. I
. '
. . . . . . . / I
. !
. ..
. . /
.
I i . i I /
.. -- -- / " <" - /
/ . " ~~. ~ "_~" , , .. i . " " , s-.. / / ." " , . ' i . -£ a " " ~ ~"I"~- ." " " " -. I . ." " , , -.I / < -" , . . . . . . . . . . . . . . ,.," . .... . . . . . . . /. .
.,. . . - " . . . . . ..,, .
--.. -
I /' I i i ~ " - ..--- " ." , ~ " "
I - /
; I ~ / i I t .. / ~ t . / i / ... / I t ..- / i / i I / i / / I i .. / i l " i /

/seq!Jl~3/olp,3Ol/lU'JO!AOSlO'~,~//:duq m uo~sJo~ JOlO~) "sou!I o~]!JlS o.re
IIIA pu~ IIA suo!13os pu~ sou!I d!p oJe IA o] I suo!]3oS ,~pms s!ql u! u~oqs uo!13os 3~tus~os ll~UO!~Oj pul~ SliO~ jo uo!1~3o I oql SA~OqSd~uJ II~Jnl3nJls o q i "~!J~c[ jo JlnD oql q~noJql suo!13os 3[tustos [~uo!8oJ jo s~u!/~eJp-ou!"1 "9 "8!~
£
-t .,.
--!
8
m
0 ~ 0
I ~ ~I' ('[oJd) V ~ " A/IS
9SS
-,=$9 N O a q6lH
II I a q6!H euedweo opuewe4 ues I^
IliA
m
i. i. o
w>i O~
0 q6lH I I I '
9 ,owoa AI ~ ,14 9s~ II I ,M NA/I
III
IIA 17
SNOII O g S 9}II~IIS
"~~ ~' i~':~¸~¸~ii~'!!i~'i
~¸ ~:-~¸I~!~
fined , { e u e s e o 9 0 ~ m
ouooo!~ Joddn f;;;;;;;;;;;;:;;] o
lined sl~u!JdS LuJelH 9S/1/1
snoeoemJO
eueoo!~ elPp!~
-snoeoe:leJO
'
:~i~:,~ ~.,.,~-~i
o
~- eueoo~ld~ .
0
_,'3'N
q61H opeoo^v-e^noo
v
N MS A/IN I I A w
, v
]
-=,S
opueuJe=l ues
0
U6!H I L
~r~
-o,sse~nr [~:~-,?, -a eueools~elcl
O 17 ........ i~ ' ~ :. ' • ' , -:~ ~, . . . . ; , • :.i_ .:i~.i ;, ~ !'~' ,~",~ :~',~ : i : ~ ; ' :;-: ::.;~ ~, ~; ,:, ~: . ~ :~ ; , , • .:; :..~;,. :, ; 17
SHNVHM9..L OIHcl~Ol/llYJ..31,41 173S ISR~IHI-GNV-GTOJ 77/=/NISVEt
SNOIIOgS-SSOI:IO 0 N 9 9 9 7
oo
m
uol.sJaAul aAfl!SOcl IoAIuoo flaA/1 o
sq6!H / s~4o7
lined le~.~ON ~ sa6p!ld aleqS
u,se~#edV-llnct I:. . . . . . . I 0 qDH ,, V,,(bJd) l f ' r qDH ' 0
S~lOOJo!qd~ow~lal41 '" 9NN seue.ue_L • q6!H ['~o,o~r' I IA H opePlOS" A/IS
I
i
lsnJql ~ auoz ieuo!ssa~dtuoo uJaq;no S snouoq~qoollV I '" ii'1 o!qdJoLuelelAI JlnE) I!OWOO ,MSS I 9N AI
I
79Nb'HO Snm41n7oo
pla!:lo~SaleUJe~~-
o ~°
or)
m
UJ~l 13 I- 0 q6!a qDH 0

N ,linE) I,Ol.UOC]([°'d)r([o,d) I )4 " !11
qB!H opeplos o -"
q6!H eleid-eued~ueo
---I
oo
~ m
o
0 " 0 II
N
('[oJd) :t edo9 Ilia O S 9N aS II
IIA 1,
O3
.- .... . - ~ - ~ . . • "" " m
i. i..._Qo
0 I ~ , 0
q6!H eueduJeo e6p!EI eleqs
IIA IliA selewepecl
d~Ml N O U V O 0 7 SNOI.I.O_=,S dIG

STRUCTURE OF THE GULF OF PARIA PULL-APART BASIN 487
the south. Normal faults commonly sole out into Structural domains
a Miocene shaly section but locally cut into Cre-
taceous evaporite (Couva Formation). The Couva The structure of the Gulf of Paria can be subdi-
evaporites constitute the core of the anticline located vided into two main structural provinces or domains:
underneath the Avocado-Couva High. This high a Northern Transtensional Basin and a Southern
coincides with a change in vergence of the over- Compressional Zone. Both units are superimposed
lying extensional system. This lithologic unit has on an underlying fold-and-thrust belt that represents
been also recognized onshore Trinidad and the Paria the eastern prolongation of the Serranfa del Inte-
Peninsula. From a stratigraphic point of view, the rior of eastern Venezuela. Regional surface data in
most interesting feature provided by this section, is eastern Venezuela and Trinidad as well as seismic
the lateral facies change between the conglomerate and well log data presented in this study support
of the Cunapo Formation and sandstone and shale of southward thrusting of this underlying orogenic belt.
the Manzanilla and Springvale Formations. Normal The description of the Gulf of Paria will proceed
faulting and the lack of additional well data obscure from north to south.
the geometry of this contact.
Northern Transtensional Basin
This northern domain is bounded to the north
Section VII
by the Avocado-Couva High, the Gulf High and
This E-W-trending profile across the entire Gulf
the Paria Peninsula to the north and by the Warm
of Paria is located in the central part of the study
Springs fault to the south. The structure of this
area. The section illustrates the lateral ramps of
province is characterized by down-to-the-basin nor-
the extensional system and some accommodation
mal faults (Figs. 5 and 6). In the northernmost part of
zones associated with vergence changes. Eastward-
the Gulf of Paria, compressional structures, mainly
and westward-dipping listric normal faults are con-
south-directed thrust and folds, can be recognized
nected and combined, constituting extensional basin
underneath the extensional features. Thus, local ex-
fill with Upper Miocene-Pliocene strata. Notice sig-
tensional faults overprint previous compressional
nificant rotation but no growth of strata. In the
structures (Fig. 6).
central part of the section is the Domoil High,
The Northern Transtensional Basin is charac-
where Pleistocene sediments unconformably overlie
terized by a thick late Neogene section (Upper
Lower Cretaceous rocks. The different stratigraphy
Miocene-Pleistocene) and a variable, but often in-
of the wells M, I and J along with the seismic ex-
complete, pre-Neogene section. The pre-basin-fill
pression suggest thrusting within the Domoil High.
stratigraphy consists of Oligocene-Middle Miocene
West of the Domoil High, Upper Miocene sediments
deep-water shales (i.e. the Brasso, Cipero and Lower
unconformably overlie Upper Cretaceous marls, as
La Pica Formations) unconformably overlain by
revealed by well I.
the shallow-water Manzanilla, Springvale and Up-
per La Pica Formations. Several wells pierced the
Section VIII shale ridges (shale-cored footwall blocks) reveal-
This E - W to N E - S W section shows the E - ing the presence of deep-water shales and turbidites
W-trending Campana High. The easternmost part of the Cipero and Brasso Formations within the
of the transect shows a broad fold linked with a core. Structurally low areas, represented by exten-
deep decollement level. Data from well A suggest sional half-grabens, preserved more pre-extensional
a Lower Cretaceous decollement system. The struc- stratigraphic section (below the top of the Upper
ture of the high is characterized by a set of folds Miocene, 10.5 Ma) reaching the Oligocene and lo-
involving Miocene to Upper Cretaceous sediments. cally the Paleogene or Cretaceous. Structural highs
Thrusting is evidenced by the presence of two re- (i.e. Domoil, Gulf, and Posa Highs) did not preserve
peated sections in well B. Those thrusts are probably any Upper Cretaceous to Middle Miocene sediments
lateral ramps of a northward-vergent thrust system. (see sections II, III, IV, and VII of Fig. 6). In
Transtensional basins bound to the north and to the the Domoil area Pliocene sediments were also not
south by the Campana High. The relationship be- preserved (sections III, IV and VII of Fig. 6) and
tween wells A and B indicates that well A is located Pleistocene and Upper Miocene sediments uncon-
on the down-thrown block of a westward-dipping formably overlie Lower Cretaceous rocks. Wells M
extensional fault. The hangingwall of this exten- and I located on the edge of the Domoil High (see
sional detachment is characterized by domino-type Fig. 5) indicate the presence of Cunapo Conglomer-
normal faulting. East of well B, a synclinal feature ate associated with fault scarps.
located underneath Pleistocene strata may suggest The NNW-SSE-trending Avocado-Couva High
the presence of a relict compressional basin of the constitutes the most prominent structural feature in
underlying fold-and-thrust belt. the northern area. The hinge zone of this anticline

coincides with a change in the dip of the overlying The main decollement of the extensional system
extensional detachment and related normal faults. is located within Oligocene-Lower Miocene shales
According to well and seismic data, the extensional (i.e. Cipero and Brasso Formations). The southeast-
decollement coincides with the Cretaceous-Neogene ern part of the NTB (Trinidad side) is occupied by
contact. Obvious thrust structures are not seismically a complex pattern of NE-SW-trending shale ridges
clear south of the Avocado-Couva and Gulf Highs, that core footwall blocks. Well L (see Fig. 6) pen-
since they are crosscut and obscured by extensional etrated a shale ridge in this area (see sections IV
faults. The main decollement level of the thrust sys- and V of Fig. 6). Two main WNW-ESE-trending
tem is probably the Lower Cretaceous Couva evap- structural highs are located within the Northern
orite and Cuche shale (well exposed in Trinidad). Transtensional Basin: the Domoil-Gupe and Gopa-
The northernmost part of the Northern Transten- Posa Highs (Fig. 5). These structural highs are
sional Basin (i.e. the Avocado-Couva and Gulf bounded by down-to-the-basin normal faults and
Highs) has a thin and very incomplete stratigraphic are cored by Lower Cretaceous rocks. Anticlines
section. The Cunapo Conglomerate unconformably developed against the main normal faults and the
overlies Lower Cretaceous shales (Cuche Forma- thickness distribution of the Pleistocene section sug-
tion). A thick section of Lower Cretaceous evaporite gest Pleistocene positive inversion. Basin-bounding
(Couva Formation) is only present in this area and in normal faults were reactivated as reverse faults.
the Paria Peninsula. The Cunapo Conglomerate rep-
resents fan deltas and slope deposits associated with Southern Compressional Zone
fault scarps and were predominantly sourced from The Southern Compressional Zone (SCZ) is lo-
the north and are coeval with uplift of the Northern cated south of the Warm Springs fault system. This
Ranges and the Paria Peninsula. The wells N, P and domain is characterized by a thin Plio-Pleistocene
Q were drilled through these stratigraphic units. To stratigraphic section and predominantly north-ver-
the south, the conglomerate of the Cunapo Forma- gent Late Miocene to Pliocene thrusting (Figs. 5 and
tion interfingers with the Manzanilla and Springvale 6). The Southern Compressional Zone represents the
Formations, suggesting a Late Miocene to Pliocene continuation of the structures of the Central and
age for the conglomerate. Southern Ranges of Trinidad described by Kugler
The southern boundary of the Avocado-Couva (1961). Although the dominant vergence of thrusts
and Gulf Highs is a southward-dipping system of and folds is reported to be towards the south in the
anastomosing normal faults (Figs. 5 and 6). The island of Trinidad, north-vergent back-thrusts can
Gulf High plunges to the west and disappears. On be recognized in the Southern Range of Trinidad
the Venezuelan side of the Gulf of Paria, a ma- (Fig. 5).
jor basin-bounding fault (Casanay fault) separates The stratigraphic section of the Southern Com-
the rocks of the Paria Peninsula from the North- pressional Zone consists of basinal Upper Miocene
ern Transtensional Basin. The Casanay fault is a deep-water shale and turbiditic sandstone (Lower
major normal fault with minor Pleistocene positive La Pica Formation) overlain by a very thin Plio-
inversion and probably some strike-slip component, Pleistocene section. The Manzanilla Formation is
which is difficult to estimate (Fig. 6). absent in the east on the San Fernando High (section
The southern boundary of the Northern Transten- V of Fig. 6). The overlying Springvale and Talparo
sional Basin is the NE-SW-trending Warm Springs Formations are affected by folding and thrusting.
fault (WSF) which accounts for an important off- The pre-Neogene section is better known in the
set of the late Neogene and Cretaceous-Paleogene south because of the lack of thick superimposed
section (Fig. 6). The WSF does not consist of transtensional basins. Upper Cretaceous sediments,
a single fault but of a set of north-dipping nor- absent in the Northern Transtensional Basin, were
mal faults and related lateral ramps. We assume, encountered in wells located in the San Fernando
based on regional data, that the WSF has been Bay and the Soldado High (see Fig. 6).
active since Late Miocene time. The structure of The seismic expression of the Los Bajos fault
the transtensional basin is characterized by E-W- is that of a continuous NW-SE-trending shale
and NW-SE-trending normal faults (Fig. 5). North- ridge, only interrupted by the WSF (Fig. 5). In the
south-trending lateral and oblique ramps that join Morro area, where the Posa field is located, the re-
the Warm Springs fault are observed on the Trinidad gional decollement may be located above the Upper
side of the Gulf of Paria. The Northern Transten- Miocene unconformity (10.5 Ma sequence bound-
sional Basin (NTB) is tilted to the south against ary) (see section II of Fig. 6). This area consists
the Warm Springs fault, which constitutes the mas- of arcuate-shaped approximately E-W-trending and
ter fault of the transtensional system. Roll-overs south-vergent thrusts that constitute a thrust-imbri-
associated with basin-bounding faults are the most cate system (Fig. 5). The NE-SW-trending piggy-
conspicuous structures of this domain. back basins related to late Neogene thrusting are

characterized by the merging of unconformities and metamorphic terranes and the folded sediments of
the presence of progressive unconformities. Progres- the South American Plate is a northward-dipping
sive unconformities consist of folded unconformities extensional contact in the study area. Fission-track
in the lower part of the basin fill that are crosscut by data in the Northern Range (Algar, 1995) suggest
less steep unconformities in the upper part of the sec- Late Miocene uplift, which was coeval with exten-
tion. This type of unconformity records the timing sional tectonics in the Gulf of Paria and Caroni
of thrust emplacement. Shale-cored doubly verging Basin. The Northern Range of Trinidad, as well as
anticlines associated with thrusting are present in the the Paria Peninsula of eastern Venezuela can be pic-
Pedernales area and extend to Trinidad; mud volca- tured as core complexes bounded by the extensional
noes are common along this structural zone. One of basins of the Caribbean Sea to the north and by
the most prominent structural features in this area the transtensional basin of the Gulf of Paria in the
is the Campana High (Fig. 5). Wells A and B were south. The Central and Southern Ranges of Trinidad
drilled on structural highs constituted mostly by belong to the Southern Compressional Zone (Fig. 5).
Upper Cretaceous and Paleogene sediments (section This structural trend extends towards the west into
VIII of Fig. 6). This NE-SW-trending anticlinorium the Gulf of Paria, and can be traced further west
is interpreted to be related to a deep-seated thrust into the eastern Venezuelan Serranfa del Interior
rooted in the Cretaceous section. Nevertheless, seis- fold-and-thrust belt. This single folded belt involves
mic data are not conclusive regarding the vergence the deformed northern passive-margin sediments of
of the underlying thrust system. The seismic data the South American Plate. The so-called Nariva
display N-S-trending lateral ramps associated with fold-and-thrust belt of southern Trinidad represents,
north-vergent thrusts south of the Campana High like the Carapita overpressure shale belt of eastern
(see section VIII of Fig. 6). Venezuela, the westward continuation of the Bar-
In the Southern Compressional Zone, transten- bados accretionary wedge (Fig. 1). The continuity
sional structures are subordinated to thrusts, but between both is only interrupted by the Neogene
locally thrust sheets are overprinted by steeply dip- transtensional features of the Gulf of Paria. The
ping normal faults. Westward-dipping normal faults transtensional basins crosscut and are superimposed
offset the western edge of the Campana High (see onto the underlying and otherwise continuous folded
section VIII of Fig. 6). belt. This relationship not only applies to thrusts
and folds but also to the strike-slip faults that bound
Implications for the structure of Trinidad thrust units, i.e. tear faults like the Los Bajos fault.
This crosscutting relationship was observed by early
The knowledge of the structure of the Gulf of exploration works (Wilson, 1968). The Southern
Paria has important implications for the geology Basin and the Erin Basin represent piggyback basins
of the island of Trinidad. The Caroni Basin is the in the sense of Ori and Friend (1984) related to the
onshore continuation of the transtensional basins of emplacement of the Southern Range. These piggy-
the Gulf of Paria. The Warm Springs fault consti- back basins can be traced in the subsurface towards
tutes the southern boundary of this transtensional the west until the Campana High, where the Neo-
province. This apparent normal contact has been gene sediments onlap onto Cretaceous-Paleocene
subsequently modified by compression, like most of rocks. They are equivalent to the piggyback basins
the major normal faults of this northern domain. The of the Carapita Formation in eastern Venezuela. Mud
northern flank of the Central Range of Trinidad con- volcanoes are associated with thrust faults of this
stitutes the field expression of the positive inversion southern tectonic province (Fig. 5). Careful analy-
overcome by the Caroni Basin. Northward-dipping sis of the Southern Range geological maps reveals
faults and strata, on the hangingwall of the Warm the presence of doubly vergent thrusts and folds
Springs fault System, opposite to what would be (Fig. 5). In the southern part of the central Gulf of
expected for a normal fault, reveal compressional Paria the westward extension of this unit is repre-
reactivation after extension. The Guatapajaro An- sented by the Morro northward-vergent imbricates
ticline, located in the center of the Caroni Basin (backthrusts). From a geometric point of view the
(Fig. 5), is evidence of this late Neogene inversion contact between south-vergent thrusts and north-ver-
that we recognized offshore, in the Gulf of Paria. gent imbricates, can be interpreted as a 'passive-roof
The extensional character of the Casanay fault indi- duplex' in the sense of Banks and Warburton (1986).
cates to us that the nature of the E1 Pilar fault system This uppermost compressional system (backthrust
as the northern boundary of the Northern Transten- zone) involves slope and basinal deposits. The Ped-
sional Basin should be reviewed. The seismic data, ernales shale ridge located in the southern part of
as well as most of the surface data along the south- the SCZ (Fig. 1, Fig. 5, and section I of Fig. 6)
ern boundary of the Northern Range suggest that constitutes an ENE-WSW-trending doubly verging
the present-day contact between the allochthonous shale-cored anticline linked the basal decollement

490 J.F. FLINCH et al.
Fig. 7. Block diagrams showing the Neogene structural evolution of the Gulf of Paria. (a) Fold-and-thrust belt stage characterized by
south-vergent thrusting and related piggyback basins. (b) Transtension in the north coeval with north-vergent thrusting in the south. (c)
Positive inversion of transtensional basins and north-vergent thrusting. Color version at http://www.elsevier.nl/locate/caribas/

Fig. 7 continued.
of the passive-roof duplex. The shale ridge extends The structures underlying the Gulf of Paria repre-
from Maturfn in the eastern Venezuelan Basin to the sent the offshore continuation of the Central Range
Southern Range of Trinidad. of Trinidad and the eastern Venezuelan fold belt.
The Lower Cretaceous Couva evaporites (Avocado-
Tectonic evolution Couva High) may represent part of a much more
widespread evaporitic nappe that could occupy most
The data presented here permit establishment of of the basement of the Gulf of Paria. In fact,
an evolutionary model for the Gulf of Paria. The these evaporites could represent Triassic or Juras-
relative timing of the structures is reasonably well sic evaporites emplaced within Lower Cretaceous
known in the area, even though the absolute timing deep-water sediments like in the Gulf of Mexico.
of the main tectonic events is not fully determined. Broad synclinoria are located south of the thrust
The Neogene tectonic evolution of the Gulf of Paria that separates Lower Cretaceous from Tertiary sed-
can be subdivided into three main stages (Fig. 7): iments. Brasso shales and Nariva turbidites occupy
(1) fold-and-thrust belt (south-directed thrusting); the core of these synclinoria. The westward continu-
(2) transtension and north-directed thrusting; (3) ation of these synclinoria is occupied by the Carapita
positive inversion. Table 1 summarizes the timing Formation of eastern Venezuela. We interpret these
and nature of the main tectonic events within each synclinoria as relict piggyback basins of the eastern
structural domain. Venezuelan fold-and-thrust belt.
The presence of south-vergent thrusts underneath
Fold and thrust belt (Oligocene-Lower the basin fill can be inferred from seismic data
Miocene) (Fig. 6). The Campana High was developed before
Before Late Miocene time the Gulf of Paria was 10.5 Ma (initiation of half-graben tectonics) and
occupied by a peneplained (eroded) fold-and-thrust is probably associated with south-vergent thrusting
belt. The block diagram of Fig. 7a displays the hy- (sections I, II and VIII of Fig. 6). A well-developed
pothetical geometry of the fold-and-thrust belt that shelf-margin occurs in the southernmost part of the
occupied the region before transtension took place. Gulf of Paria (i.e. Erin Basin, Pedernales and Posa

areas) and represents the northern edge of the eastern are widespread within the Caroni Basin of Trinidad
Maturfn foreland basin. and can be recognized in surface exposures along the
northern border of the Central Range. North-vergent
Transtension and northward thrusting thrusting was still active in the south (Pedernales
(Middle Miocene-Pliocene) area) until Late Pleistocene-Holocene time.
From Late Miocene time on, half-grabens over-
printed the underlying thrust sheets in the northern
part of the Gulf of Paria. This tectonic stage was DISCUSSION
characterized by the development of transtensional
basins controlled by steeply dipping normal faults A valid model for the Gulf of Paria should ex-
(Fig. 7b). As transtensional collapse progressed, the plain: (1) the orientation of the Domoil, Gulf and
relief of lows and highs increased. Shale ridges form Posa Highs and the nature of their border faults;
shale-cored footwall blocks located in the central (2) the fact that the Los Bajos fault is crosscut and
part of the transtensional basin. The main struc- overprinted by the Warm Springs fault system, and
tural feature active during this time was the Warm therefore cannot be carried through the entire Gulf
Springs fault that separates the northern transten- of Paria; (3) that transtension in the northern Gulf
sional province from the southern compressional of Paria (half-graben development) was coeval with
province. The Casanay and Arima faults consti- north-vergent thrusting in the south, that is, exten-
tute the boundary between the NTB with the al- sion took place in an overall compressional scenario;
lochthonous metamorphic terranes exposed along (4) the Pleistocene positive inversion of major NW-
the Paria Peninsula and the Northern Range of SE- to E-W-trending bounding faults, suggesting
Trinidad. Extensional collapse induced by pull-apart widespread post-Pliocene N-S compression.
extended from the Caroni Basin of Trinidad to the
Guanoco area and the San Juan graben in eastern The pull-apart model
Venezuela (see Fig. 1).
Fission-track ages in northern Trinidad (Algar, From a geometric point of view fault patterns in
1995; D.A. Farrell, pers. commun., 1996) suggest the study area resemble pull-apart basins generated
that extensional collapse in the Gulf of Paria was by experimental models (McClay and Dooley, 1995)
coeval with the uplift of Trinidad's Northern Range. (Fig. 8). Following a comparison with an experi-
High tectonic subsidence rates were compensated mental pull-apart model, the E1 Pilar fault represents
by huge sediment supply, resulting in shallow-water the westward side of the Principal Displacement
Upper Miocene-Pliocene deposition (i.e. the Man- Zone (PDF), characterized by fight-lateral strike-
zanilla and Springvale Formations). Fan deltas, rep- slip displacement (Fig. 8). Steeply dipping normal
resented by the Cunapo Conglomerate, were shed faults such as the ones that bound the Domoil, Posa
from the uplifted Northern Range, Paria Peninsula and Gulf Highs would represent oblique-slip faults
and Domoil, Gulf and Avocado-Couva Highs. with normal and fight-lateral component (cross-
During Pliocene time, turbiditic wedges (devel- basin fault zones in the sense of McClay and Dooley,
oped in front of pre-existing south-vergent thrusts) 1995). The lack of reference lithologic or structural
were thrusted and folded by north-vergent imbricates markers makes it difficult to demonstrate the magni-
(Fig. 7b). Turbidites were encountered by drilling in tude of transcurrent displacement on most of these
the Pedernales and Posa oil fields (Fig. 5). Piggyback faults. The northern border of the pull-apart sys-
basins developed associated with this north-vergent tem is represented by the Casanay and Arima faults
thrusting. 0, while the southern border is the Warm Springs
fault (Southern Basin sidewall). The eastward side
Positive inversion (Pleistocene) of the strike-slip system would be represented by
During Late Pliocene-Early Pleistocene time, in- the Trinidad Central Range. The so-called Central
version of pre-existing normal faults took place in Ranges fault may represent the eastward PDZ of
the northern area (Fig. 7c). Normal faults were re- the presumed pull-apart system that generated the
activated with reverse displacement and anticlinal Gulf of Paria (Fig. 8). The pull-apart model also
features developed in the hangingwall of the ma- explains the eastward narrowing of the Caroni Basin
jor pre-existing normal faults (see sections III, IV and the orientation of the San Juan graben of east-
and VI of Fig. 6). The orientation of the inversion ern Venezuela, as well as the normal faulting along
structures is predominantly NW-SE to E-W coin- the eastern Venezuelan Serranfa del Interior. In our
ciding with the trend of major extensional faults model the Los Bajos fault does not have the same
(Fig. 5). N-S-trending normal faults were not in- significance as in the other models where it plays the
verted, suggesting that inversion is mainly related role of a major strike-slip fault across the entire Gulf
to a N-S-directed compression. Inversion structures of Paria (Munro and Smith, 1985; Algar, 1995).

Fig. 8. Comparison of the structure of the Gulf of Paria (lower panel) with the pull-apart experimental analogue model of McClay and
Dooley (1995) (upper panel). The westward PDZ is represented by the E1 Pilar fault while the eastward PDZ is the Central Range fault
of Trinidad. The Southern Basin sidewall is the Warm Springs fault and the northern the Arima-Casanay fault. Cross-basin fault zones
define the Domoil, Posa, Gulf and Avocado-Couva Highs.
The Los Bajos fault, like the Urica fault or San complication to simple pull-apart generated above
Francisco-Quiriquire faults (Fig. 1) is not related to non-deformed strata and differs from a conventional
the pull-apart system of the Gulf of Paria. They are basement-involved pull-apart basin. Pull-apart in the
tear faults related to the differential thrust transport Gulf of Paria overprinted an underlying fold-and-
displacement between thrust sheets of the eastern thrust belt.
Venezuelan fold-and-thrust belt (Fig. 1). The pull-
apart of the Gulf of Paria can be explained by step-
ping over between the E1 Pilar fault system and the CONCLUSIONS
Central Range fault of Trinidad (Fig. 8). The nature
of this releasing step-over could be an ancient tear The Gulf of Paria is characterized by NW-SE-
fault like the ones of the Serranfa del Interior. Epi- to E-W-trending transtensional basins. These basins
centers of earthquakes and related focal mechanisms developed during late Neogene time in the northern
indicate a change from a single E - W strike-slip fault part of the Gulf of Paria, and were coeval with
(El Pilar fault) to a distributed zone of deformation compression in the southern part. This permits the
along NW-SE-trending strike-slip faults (P6rez and definition of two tectonic provinces: the Northern
Aggarwal, 1981; Villasefior et al., 1992). High subsi- Transtensional Basin, dominated by a complex set
dence rates and parallel filling (no growth) with clas- of half-grabens, and the Southern Compressional
tic wedges (fan deltas) associated with fault scarps Zone, consisting of doubly vergent thrust imbricates.
(Cunapo Conglomerate) are typical characteristics of The Caroni Basin is the eastward continuation of
pull-apart basins (Link et al., 1985). the Northern Transtensional Basin, while the Central
The presence of south-vergent thrust sheets, un- Range and Southern Ranges fall into the Southern
derlying the present-day basin, implies additional Compressional Zone. The extensional structures of

494 J.E F L I N C H et al.
the Gulf of Paria extend towards the west and Flinch, J.E and Rambaran, V., 1996. Neogene extensional tecton-
are superimposed to Cretaceous thrust sheets of ics of the Gulf of Paria (eastern Venezuela-western Trinidad).
Am. Assoc. Pet. Geol. Bull., 80 (8): 1290.
the eastern Venezuelan Serranfa del Interior. The
Gonzalez de Juana, C., Iturralde, J. and Picard, X., 1980. Ge-
structural evolution of the Gulf of Paria, can be ologia de Venezuela y de sus cuencas petrolfferas. Ediciones
summarized into three main stages: Foninves, Caracas, 1031 pp.
(1) Fold-and-thrust belt (south-vergent thrusting). Kugler, H.G., 1953. Jurassic to recent sedimentary environments
(2) Transtension and northward thrusting. in Trinidad. Bull. Assoc. Suisse Geol. Ing. Pet., 20: 27-60.
Kugler, H.G., 1961. Geological Map of Trinidad, scale
(3) Positive inversion.
1:100,000. Edited by Orell Fussli Arts Graphiques S.A.
The structure of the Gulf of Paria can be in- Zurich, copyright of The Petroleum Association of Trinidad.
terpreted as a fight-lateral pull-apart caused by the Link, M.H., Roberts, M.T. and Newton, M.S., 1985. Walter Lake
step-over of the E1 Pilar and the Central Range Basin, Nevada: an example of Late Tertiary (?) to Recent
faults. sedimentation in a basin adjacent to an active strike-slip fault.
Soc. Econ. Paleontol. Mineral. Spec. Publ., 37: 105-125.
Macellari, C., 1995. Cenozoic sedimentation and tectonics of
the southwestern Caribbean pull-apart basin, Venezuela and
ACKNOWLEDGEMENTS Colombia. In: A.J. Tankard, R. Suarez-Soruco and H.J.
Welsink (Editors), Petroleum Basins of South America. Am.
We thank CVP (Corporaci6n Venezolana de Assoc. Pet. Geol. Mem., 62: 757-780.
Mann, E and Burke, K., 1984. Neotectonics of the Caribbean.
Petroleo), PETROTRIN and LAGOVEN (today
Rev. Geophys. Space Phys., 22 (4): 309-362.
PDVSA Oil and Gas) for permission to use the McClay, K. and Dooley, T., 1995. Analogue models of pull-apart
data presented here. Thanks also go to V. Hunter, basins. Geology, 23:711-714.
who provided biostratigraphic support. The article Molnar, E and Sykes, L.R., 1969. Tectonics of the Caribbean and
has benefitted from the suggestions of Albert W. Middle America regions from focal mechanisms and seismic-
ity. Geol. Soc. Am. Bull., 80: 1629-1684.
Bally and Max Furrer. We want also to thank Kevin
Munro, S.E. and Smith, ED., Jr., 1985. The Urica fault zone,
Burke, Paul Mann and Kris Meisling for helpful and northeastern Venezuela. In: W.E. Bonini, R.B. Hargraves and
constructive reviews. R. Shagam (Editors), The Caribbean-South American Plate
Boundary and Regional Tectonics. Geol. Soc. Am. Mem., 162:
213-215.
Ori, G.G. and Friend, EE, 1984. Sedimentary basins, formed
REFERENCES
and carried piggy-back on active thrust sheets. Geology, 12:
475-478.
Algar, S., 1995. Interaction of the Caribbean and South American
P6rez, O. and Aggarwal, Y.E, 1981. Present-day tectonics of
Plates as revealed in the Northern Range of Trinidad. Field
the southeastern Caribbean and northeastern Venezuela. J.
Trip Guide, 14th Caribbean Geological Conference. Port of
Geophys. Res., 86 (Bll): 10,791-10,804.
Spain, Trinidad, pp. 38-60.
Robertson, E and Burke, K., 1989. Evolution of southern
Ball, M.M., Harrison, C.G.A., Supko, ER., Bock, W.D. and
Caribbean plate boundary, vicinity of Trinidad and Tobago.
Maloney, N.J., 1971. Marine geophysical measurement on the
Am. Assoc. Pet. Geol. Bull., 73 (4): 490-509.
southern boundary of the Caribbean Sea. In: T.W. Donnelly,
Salvador, A. and Stainforth, R.M., 1968. Clues in Venezuela to
(Editor), Caribbean Geophysical, Tectonic and Petrological
the Geology of Trinidad, and vice versa. 4th Caribbean Geol.
Studies. Geol. Soc. Am. Mem., 130, 224 pp.
Conference Transactions, pp. 31-40.
Banks, C.J. and Warburton, J., 1986. 'Passive-roof' duplex ge-
Schubert, C., 1982. Origin of Cariaco Basin, Southern Caribbean
ometry in the frontal structures of the Kirthar and Sulaiman
Sea. Mar. Geol., 47: 345-360.
mountain belts, Pakistan. J. Struct. Geol., 8: 229-237.
Soulas, J.E, 1986. Neotect6nica y tect6nica activa en Venezuela
Bellizzia, A. and Dengo, G., 1990. The Caribbean mountain sys-
y regiones vecinas. VI Congreso Geol6gico Venezolano, Tomo
tem, northern South America. In: G. Dengo and J.E. Case (Ed-
X, pp. 6639-6656.
itors), The Caribbean Region. The Geology of North America,
Speed, R.C., 1985. Cenozoic collision of the Lesser Antilles and
H, Geological Society of America, Boulder, CO, pp. 167-175.
Continental South America and the origin of E1 Pilar fault.
Tectonics, 4:41-69.
Earth Planet. Sci., 16:201-230.
Vierbuchen, R.C., 1984. The geology of the E1 Pilar fault zone
Burke, K., Cooper, C., Dewey, J.E, Mann, E and Pindell,
and adjacent areas in northeastern Venezuela. In: W.E. Bonini,
J.L., 1984. Caribbean tectonics and relative plate motions.
R.B. Hargraves and R. Shagam (Editors), The Caribbean-
In: W.E. Bonini, R.B. Hargraves and R. Shagam (Editors),
South American Plate Boundary and Regional Tectonics. Geol.
The Caribbean-South American Plate Boundary and Regional
Soc. Am. Mem., 162: 189-212.
Tectonics. Geol. Soc. Am. Mem., 162:31-63.
Villasefior, A., Mufioz, M.I., Franke, M. and Gajardo, E., 1992.
Case, J.E., MacDonald, W.D. and Fox, EJ., 1990. Caribbean
Estudios de microsismicidad en el norte de Venezuela, 3. Zona
crustal provinces; seismic and gravity evidence. In: G. Dengo
nororiental. VI Congreso Venezolano de Geoffsica, pp. 529-
and J.E. Case (Editors), The Caribbean Region. The Geology
535.
of North America, H, Geological Society of America, Boulder,
Wilson, C.C., 1968. The Los Bajos fault. Transactions of the 4th
CO, pp. 5-36.
Caribbean Geological Conference, Port of Spain, Trinidad, pp.
Di Croce, J., 1995. Eastern Venezuela Basin: Sequence Stratigra-
phy and Structural Evolution. Unpublished Ph.D. Thesis, Rice 87-90.
University, Houston, TX, 225 pp.

Chapter 18
Structural and Sedimentary Development of a Neogene

Transpressional Plate Boundary between the Caribbean and
South America Plates in Trinidad and the Gulf of Paria
STEPHEN BABB and PAUL MANN
Trinidad, the Gulf of Paria, and Eastern Venezuela lie in a diffuse and complex zone of Cenozoic tectonic interaction between
the Caribbean and South America plates. Numerous models have being proposed to explain the complex tectono-stratigraphic
evolution of the area. In this paper, we interpret an integrated data base consisting of well logs and seismic reflection profiles to
document five sequences of the Gulf of Paria-Northern basin of Trinidad. Sequences 1 and 2 consist of a Late Jurassic-early
Valanginian carbonate megaplatform with overlying carbonate bank buildups that together form the basement to overlying
siliciclastic sedimentary sequences 3 through 5 of mainly Neogene age. Sequence 3 is a Late Miocene-Early Pliocene shallow
marine to brackish-water conglomerate and sandstone that represents a southward-fining and thinning and eastward-thickening
siliciclastic wedge derived from the Late Miocene uplift and erosion of Trinidad's Northern Range and records early activity along
the E1 Pilar strike-slip fault at the mountain front of that range. Sequence 3 and overlying sequences 4 and 5 fill in topographic
relief created by the growth of the carbonate banks of sequences 1 and 2 and by space created by tectonic extension in the Gulf
of Paria and Northern basin. Sequence 4 is an Early to Middle Pliocene inner neritic to shallow marine conglomerate, sandstone,
silt, and clay that represents a northward-fining and thinning siliciclastic wedge derived primarily from the Late Miocene-Early
Pliocene uplift and erosion of the Central Range with some input from the northwest and north. Sequence 5 is a Late Pliocene to
Pleistocene marine to brackish-water sand, silt, clay and minor conglomerate that represents continued siliciclastic deposition in
an increasingly restricted basin between the uplifted Northern and Central Ranges.
Comparison of two-way travel time structural contour maps of each boundary between sequences 3-5 (totaling 2-3 km in
thickness) with two-way travel time isochron maps of the seismic sequence immediately overlying each structural surface allows
a better visualization of the space available at tectonically significant times in the history of the Gulf of Paria-Northern basin
and the subsequent sedimentary infill onto that structural surface and sequence stratigraphic boundary. This comparison allows the
identification of three Neogene deformational phases that have affected the area and closely control the proposed sequence strati-
graphic boundaries: Phase one, Late Miocene-Early Pliocene strike-slip motion along the E1 Pilar fault and north-to-south filling
of the Gulf of Paria and Northern basins; Phase two, Middle to Late Pliocene strike-slip motion along the Warm Springs-Central
Range fault zone and south-to-north filling of the Gulf of Paria and Northern basins; and Phase three, Late Pliocene to Pleistocene
strike-slip motion along the Warm Springs-Central Range fault zone and continued filling of the Gulf of Paria and Northern basins.
Paleocurrents and environments of deposition of the Gulf of Paria and Northern basins have been closely controlled by
structural events during the three deformational phases. Paleocurrents for deformational phase one (deposition of sequence 3) are
oriented orthogonal (north-south) and parallel (east-west) to the axis of the basins and reflect strong control of the bounding E1
Pilar fault zone along the northern edge of the basin. Paleocurrents for deformational phases two (sequence 4) and three (sequence
5) become increasingly dominated by a west-to-east paleoflow along with a southeast-to-northwest flow in the area of the Goodrich
sub-basin (southern Gulf of Paria basin). Environments of deposition reflect the Late Miocene (~ 10 Ma) uplift and erosion of the
Northern Range during phase one followed by subsequent uplift and erosion of the Central Range and the gradual decrease in the
importance of the Northern Range siliciclastic source area during phases two and three. Environments generally remain proximal,
shallow, and brackish along the E1 Pilar fault zone on the northern edge of the basin and distal, deeper, and less restricted along
the central and southern edges of the basin. Through time the basin has become increasingly brackish and more restricted.
Seismic lines are used to illustrate the variation in the structural style and geomorphic expression of the E1 Pilar fault zone
along the northern edges of the Gulf of Paria-Northern basin. In the western Gulf of Paria, the fault exhibits a transpressional
structure consistent with its east-northeast strike. A slight change in strike to a more east-southeast strike in the eastern Gulf of

496 S. B A B B and E M A N N
Paria is consistent with the more transtensional structure of the fault and its association with the Puerto Grande sub-basin in the
northern Gulf of Paria basin. The trace of the E1 Pilar fault zone is poorly expressed onland in Trinidad and may reflect its waning
activity in recent time in an eastward direction and/or cultural obliteration of its tectonic geomorphology in this highly urbanized
part of Trinidad.
Seismic lines are used to illustrate the structural style of previously undescribed oblique-slip faults in the Gulf of Paria. These
styles include: (1) gravity-related detachment faults possibly localized along the unconformity between Cretaceous carbonate and
overlying siliciclastic rocks; and (2) transtensional faults formed in the stepover area between the E1 Pilar fault to the north and
the Warm Springs fault to the south.
Seismic lines are used to illustrate the previously undescribed structural styles and geomorphic expression of the Warm
Springs-Central Range-Caigual fault zone. In the Gulf of Paria, the Warm Springs fault zone exhibits a transtensional structure
consistent with its east-southeast strike. An abrupt change in strike to a more east-northwest strike near the western shoreline of
Trinidad is consistent with the more transpressional structure of the continuation of the Warm Springs fault zone in the Central
Range (Central Range-Caigual fault zones). We reinterpret the overall structure of the Central Range as a transpressional uplift
bounded by inwardly dipping reverse faults and bisected by the Central Range-Caigual strike-slip fault system rather than as the
southeast-verging fold-thrust belt proposed by previous workers.
Integration of data by previous workers in the Southern basin of Trinidad allows us to make a comparison between the style and
age of deformation in the Southern basin and that described by us in the Gulf of Paria-Northern basin. The style of deformation
in the Southern basin is predominantly south-vergent and older than the deformation in the Gulf of Paria and Northern basins
because Middle Miocene (~11.4 m.y. horizon) units onlap folded and faulted older Middle Miocene units. The unconformity
between the Late Miocene-Early Pliocene sequence 3 and the middle to Late Pliocene sequence 4 in the Northern-Gulf of Paria
basins correlates with a less prominent unconformity between the Upper Cruse and Lower Forest Formations of southern Trinidad.
The deformation in the Southern basin is therefore older than observed in the Gulf of Paria-Northern basin.
Compilation of regional geologic data and integration of the Southern basin data with our results from the Gulf of Paria-
Northern basins allows the following events to be constrained in the Middle Miocene through recent evolution of the Trinidad
region. Middle Miocene ('.~10 Ma): high-angle faulting affects the Southern basin; we do not regard this event as a regional
fold and thrusting event because we do not observe widespread fold-thrust sturctures in the pre-Late Miocene rocks of the Gulf
of Paria-Northern basin. Late Middle Miocene: uplift of the Central Range at this time may be related to the Middle Miocene
thrusting observed in the Southern basin. Early Late Miocene: during this period a lobe of conglomeratic sediment was shed
from the Northern Range into the Gulf of Paria-Northern basin and is interpreted to reflect the first significant lateral movement
along the right-lateral E1 Pilar fault zone. We propose that the fault was propagating from west to east during this time as the
accretionary wedge and forearc of the Lesser Antilles arc moved past the South American passive margin. The Los Bajos fault is
thought to have propagated southeastward at this time. Termination of that strike-slip fault on the Southern Range may have led to
a widening of the Neogene belt of transpressional thrusting during this time. Late Miocene-Early Pliocene: during this time, there
was a rapid spread of strike-slip and oblique-slip faulting to the south of the E1 Pilar fault zone along the Warm Springs fault zone
and within the Goodrich sub-basin. This southward shift is interpreted as recording the development of a pull-apart basin at the
stepover area between the E1 Pilar and Warm Springs fault zones. Formation of the stepover may have deactivated slip along the
eastern continuation of the E1 Pilar fault zone at the northern margin of the Northern basin. Early to middle Pliocene: this period
saw continued development of the subsidence and faults of the previous period in the stepover area between the E1 Pilar and Warm
Springs faults. Late Pliocene to Pleistocene: a major wedge of sediment derived from erosion of the Central Range and sources to
the north and northwest filled the Goodrich sub-basin and continued southeastward propagation of the Los Bajos fault offset fold
axes in the Southern basin by 10.5 km. The E1 Pilar fault zone appeared to undergo transpressional reactivation in the Northern
basin but most offset seems to have shifted to the Warm Springs-Central Range-Caigual fault zone in the Central Range.
We compare this sequence of events and the structural development of Trinidad with the events and structure of the Eastern
Venezuelan basin in order to infer large-scale tectonic controls on the broad zone of deformation between the South America and
Caribbean plates. The Eastern Venezuelan basin to the west and along-strike of Trinidad is a classic foreland basin associated
with an adjacent, southeast-verging fold-thrust belt (Serranfa del Interior) formed by ~ 4 5 - 9 0 km of shortening. Trinidad exhibits
a slightly younger record of fault-related sedimentation, lacks evidence for significant shortening of the Cretaceous-Pleistocene
sequences 1-5 documented in this paper, and exhibits transpressional-type structures with no uniform sense of vergence in the
Northern Range, Gulf of Paria-Northern basin and Central Range. In order to explain these differences between the pattern of
deformation in the Eastern Venezuelan basin and Trinidad, we propose that Precambrian crust of the Guyana shield adjacent to
the Eastern Venezuelan basin acted as an effective backstop and did not allow continued southeastward migration of the obliquely
colliding Caribbean crust. In contrast, Trinidad appears to occupy the site of a rifted passive margin whose thinner crust acted
as less of a backstop than the Precambrian crust underlying the Eastern Venezuelan basin. This less confined 'free face' to the
southeast of Trinidad allowed strike-slip movement of Caribbean and passive margin crust to the southeast in a direction roughly
parallel to the trend of the former passive margin.

DEVELOPMENT OF A NEOGENE TRANSPRESSIONAL PLATE BOUNDARY 497
INTRODUCTION ary zone in Trinidad and adjacent eastern Venezuela

has r e m a i n e d controversial and this controversy has
Tectonic problems in Trinidad led to n u m e r o u s p u b l i s h e d m o d e l s put forward to
explain the c o m p l e x tectono-stratigraphic evolution
The island of Trinidad lies in a diffuse and of the area. T h e s e tectonic m o d e l s predict differ-
c o m p l e x zone of N e o g e n e interaction b e t w e e n the ent late N e o g e n e tectonic and structural styles in
Caribbean and South A m e r i c a plates (Fig. 1A). The the Trinidad area that include oblique collision
geologic and tectonic evolution of the plate bound- and thrusting, right-lateral strike-slip faulting, and
Fig. 1. (A) Present-day plate structure of the Caribbean region. Directions and rates of plate motion relative to a fixed Caribbean plate
are from DeMets et al. (1994) and Dixon et al. (1998). (B) Main tectonic elements of the Caribbean plate and its margins include a Late
Cretaceous oceanic plateau beneath the Caribbean Sea and cropping out in uplifted areas at its margins; a Cretaceous-Recent island arc
or 'Great Arc of the Caribbean'; Late Cretaceous-Eocene back-arc basins; arrows indicate inferred direction of opening in the Yucatfin
and Grenada back-arc basins; and Late Jurassic-Cenozoic passive margins on rifted continental crust in northern South America, Chiapas
and the Yucatfin Peninsula and the Bahama platform. Tectonic features of boxed area are shown in Fig. 2.

498 S. BABB and E MANN
transpressional models involving both thrusting and effects of tectonic activity on sedimentation and
strike-slip. The Paleogene history of the island, that depositional processes. Specific sedimentary goals
is thought by most previous workers to represent of the study include: (1) integration of seismic and
a thermally subsiding passive margin, remains even well data to enhance and refine knowledge of the
less understood than the Neogene active margin his- depositional systems and stratigraphic architecture;
tory because of the strong late Neogene tectonic (2) determination of the structural style and age
overprint on these older rocks and because of the of major fault zones that act as important controls
limited surface exposure and well penetration of on basinal sedimentation; and (3) determination of
Paleogene sedimentary rocks. tectonic controls on the formation of major sequence
These conflicting tectonic models by previous boundaries that separate the five identified sequences
workers in Trinidad reflect varying emphasis on in the basins of northern Trinidad.
methods, data types, and local study areas in
Trinidad. For example, previous models have empha-
sized surface structural, stratigraphic and geochrono- TECTONIC AND GEOLOGIC SETTING OF TRINIDAD
logic data from older rocks exposed in the three
ranges of Trinidad, subsurface well and seismic re- Plate tectonic setting
flection data made available from oil companies
exploring the region, and earthquake data collected The Caribbean plate is presently bounded to the
from worldwide or local seismic networks. Surface east by the Lesser Antilles subduction zone, to the
geologic mapping has mainly focussed on the North- west by the Middle America subduction zone, to
ern Range which offers the most complete and best the north by left-lateral strike-slip faults, and to the
exposed section of Jurassic and Cretaceous age. south by fight-lateral strike-slip faults (Mann et al.,
Published industry data are most plentiful from the 1990) (Fig. 1A). Rates of relative motion of the
southern part of the island and its offshore area North and South America plates, relative to a fixed
where most hydrocarbons have been found to date Caribbean plate, range from about 1 to 2 cm/yr in
(Persad, 1984, 1985). a general east-west direction (DeMets et al., 1994;
Dixon et al., 1998).
Objectives of this study The crust of the Caribbean plate is made up of
three main components:
This study was undertaken to conduct a detailed (1) The Caribbean oceanic plateau formed about
and systematic regional analysis of the geologic evo- 88 Ma ago probably as the result of the ascent of a
lution of northern Trinidad within the Gulf of Paria mantle plume into oceanic crust of Jurassic to Early
and Northern basins. We used an integrated data Cretaceous age (Burke, 1988; Diebold and Driscoll,
base consisting of well logs and seismic reflection Chapter 19; Driscoll and Diebold, Chapter 20). The
profiles. These data provide a new set of geologic plateau forms a large area of the stable central part
constraints on the sedimentary and tectonic history of the plate beneath the Caribbean Sea (Fig. 1B).
of the area. (2) Passive margins formed above rifted crust
The Neogene basins of northern Trinidad were of Precambrian to Paleozoic age in the Chiapas-
chosen for investigation in this tectonically-oriented Yucatfin area of southern Mexico, the Bahamas plat-
study for the following reasons: (1) these basins form near Cuba and Hispaniola, and the northern
record the nature and timing of major tectonic events margin of South America in Colombia, Venezuela,
because they are adjacent to key tectonic features Trinidad and Guyana (Fig. 1B). These margins
of Trinidad including the Northern Range, the E1 reached their approximate present-day positions fol-
Pilar fault zone, the Warm Springs fault zone, the lowing the breakup of North and South America
Central Range-Caigual fault zone, and the Central during Late Jurassic and Early Cretaceous times
Range; (2) these basins are relatively undeformed (Burke, 1988; Pindell and Barrett, 1990). During the
and contain a fairly complete marine sedimentary Early to Late Cretaceous, these margins subsided
record which can be accurately dated using marine thermally and became the sites of both carbonate
microfauna and pollen; and (3) the basins cover a and siliciclastic sedimentation.
significant part of Trinidad and its offshore shelf (3) The Great Arc of the Caribbean of Burke
areas and can be used to test regional structural, (1988) ranges in age from Early Cretaceous to Re-
stratigraphic and tectonic predictions made by previ- cent in the northern, eastern and southern quadrants
ous workers for the Trinidad region. of the circum-Caribbean (Fig. 1B). The Great Arc
The general focus of the study is to constrain the generally moved from west to east during Cenozoic
timing of tectonic activity in the basins of northern time and formed arc-collision zones in areas where
Trinidad with emphasis on dating the movement the east to northeastward-facing arc collided with
on late Cenozoic basin bounding faults and the continental passive margins like those of the Ba-

hamas platform or northern South America (Pindell faults is released in a few large events, perhaps
and Barrett, 1990). separated by centuries, rather than the smaller more
(4) Prior to these collisions, the Great Arc ex- frequent events characteristic of most subduction
perienced a Paleogene back-arc rifting phase that settings (Mann et al., 1990). Known surface rupture
led to the formation of the Yucatfin (Rosencrantz, associated with an earthquake along the E1 Pilar fault
1990) and Grenada (Bird et al., Chapter 15) back-arc zone occurred in 1931 (Mann et al., 1990) and in
basins leaving remnant arcs at the Cayman and Aves 1996 (R.M. Russo et al., pers. comm., 1997). Perez
ridges (Fig. 1B). and Aggarwal (1981), Speed (1985), and Russo and
The only active segment of the arc is found in the Speed (1992) have pointed out that the low levels of
Lesser Antilles (Fig. 1B). This segment of the arc strike-slip earthquake activity on the E1 Pilar fault
did not encounter continental material and continues zone east of its intersection point with the Los Bajos
to subduct oceanic crust of the Atlantic Ocean. Trin- fault near the Paria Peninsula and the absence of
idad and eastern Venezuela lie within a complex large historical strike-slip events in the Trinidad area
transpressional plate boundary zone separating the are consistent with the interpretation that the E1 Pilar
Mesozoic passive margin of South America from fault zone in the Trinidad area has become inactive.
the southern ends of the active (Lesser Antilles) and
inactive (Aves Ridge) Great Arc of the Caribbean
(Fig. 1B). REGIONAL GEOLOGIC SETTING OF NORTHERN
SOUTH AMERICA AND TRINIDAD
Active tectonics and seismicity of the Trinidad
region Offsets of strike-slip faults in northern South
America and Trinidad
Earthquake epicenters at depths from 0 to 200
km are densely concentrated in the Trinidad area Right-lateral strike-slip faults form prominent
(Fig. 2A). Events greater than 20 km are generally morphologic features that can be divided into
related to the Benioff zone of the subducted oceanic two groups with difering fault strikes in north-
crust of the Atlantic Ocean that passes to a depth of ern Venezuela and Trinidad: (1) east-west-striking
about 150 km to the northwest of Trinidad (Perez faults in the coastal areas of northern Venezuela and
and Aggarwal, 1981; Wadge and Shepherd, 1984; northern Trinidad (Mor6n-E1 Pilar; E1 Coche-North
Russo et al., 1993) (Fig. 2B). The concentration of Coast); and 2) east-southeastward- to southeastward-
shallow to deep events is related to tear faulting striking faults in more inland areas of northern
at the junction between the subducting oceanic part Venezuela and in southern Trinidad (unnamed faults
of the South America plate and the unsubducted of the Foothills fold-thrust belt, Urica, San Fran-
continental part of the plate as shown in Fig. 2B cisco, E1 Soldado, Los Bajos, and Warm Springs-
(Molnar and Sykes, 1969, 1971). Central Range) (Fig. 3).
Russo et al. (1993) compiled earthquake focal As can be seen on the radar image in Fig. 3,
mechanisms from this region of tear faulting using the first group of faults have much longer fault
both local network results and worldwide databases. traces that are straighter and more continuous over
They grouped focal mechanisms into three main distance of hundreds of kilometers, while faults of
groups: (1) right-lateral strike-slip events within a the second group are more curved and extend over
60-km-wide east-west zone parallel to the Mor6n- distances less than 100 km. Known strike-slip offsets
E1 Pilar fault system; (2) shallow thrust faults of both groups of faults total from 10 to 70 km (cf.
with east-northeast striking fault planes between the compilation in Mann et al., 1990) (Fig. 4A).
Araya Peninsula of Venezuela and Gulf of Paria to
the east of Trinidad; and (3) shallow normal faulting East-west strike-slip faults in Venezuela and
events east and northeast of Trinidad which are in- Trinidad
terpreted to be expressions of bending of the oceanic
part of the South America plate as its subducted at M o r 6 n - E l Pilar fault zone
the Lesser Antilles subduction zone (Fig. 2B). This fault is traditionally identified as the
As in other subduction to strike-slip transition main plate boundary fault of the South America-
areas of the world, the amount of seismicity falls Caribbean plate boundary zone because of its geo-
off markedly from the Lesser Antilles subduction morphic prominence, its east-west straight charac-
zone to the strike-slip zone parallel to the Mor6n-E1 ter over distances of hundreds of kilometers, and
Pilar strike-slip faults (Fig. 2A). Earthquake gen- the strike-slip ruptures of the fault trace during the
eration and strain release is less efficient along earthquakes of 1931 and 1996 (Schubert, 1984; R.M.
strike-slip boundaries in comparison to subduction Russo, pers. commun., 1997) (Figs. 3, 4A). The turn-
boundaries. Stored seismic energy along strike-slip ing of the E1 Pilar fault zone northeast of Trinidad

Fig. 2. (A) Relationship of earthquakes to faults and lithologic belts along the Caribbean-South America plate boundary zone. Earthquake
epicenters have magnitudes between 4.0 and 6.2 and are from the International Seismological Centre (ISC) catalogue. Direction of South
America-Caribbean plate motion is from DeMets et al. (1994). Area of vertical slab of subducted Atlantic crust is from Wadge
and Shepherd (1984). (B) Block diagram showing a three-dimensional view of the tectonic features seen on the gravity map of the
southeastern Caribbean shown by Mann et al. (Chapter 1). Black lines on the North and South America plates are fracture zone highs that
pass under the accretionary wedge of the Barbados Ridge complex and are subducted at the Lesser Antilles subduction zone. The dense
concentration of intermediate and deep earthquakes at the junction of the transform and subduction boundaries near Trinidad are related
to a tear fault along which the South America plate is constantly torn as the Caribbean plate moves eastward over it.
into a northeasterly direction toward the lithospheric strands of the Mor6n and E1 Pilar fault zones over
trace of the Lesser Antilles subduction zone supports the past 2 million years, Schubert, 1984) (Fig. 4A).
its interpretation as a major plate boundary zone In the Trinidad area, Robertson and Burke (1989)
fault of the southeastern Caribbean plate (Robertson postulated 3 5 - 4 0 k m of offset on the E1 Pilar fault
and Burke, 1989; Figs. 2, 4). Right-lateral offsets by realigning the truncated Late M i o c e n e synclinal
of the M o r 6 n - E 1 Pilar fault zone based on field axis of the Northern basin and an u n n a m e d offshore,
mapping in the A r a y a - P a r i a area range from 20 Late Miocene synclinal basin to the east of Trinidad
k m (offset of B a r r e m i a n - A p t i a n rocks, Vierbuchen, (Fig. 4A). An offset of 3 5 - 4 0 km over the past 6 - 8
1984) to 70 k m (amount of displacement needed to m.y. would yield an average rate of ~ 5 m m / y r on
form the Cariaco pull-apart basin b e t w e e n the offset the E1 Pilar fault zone.

El Coche-North Coast fault zone (Fig. 4A). Onshore offset of the axis of the Erin-
This high-angle, submarine right-lateral strike- Siparia syncline indicates 10.5 km of fight-lateral
slip fault extends parallel to the coast of northern offset for the Los Bajos (Wilson, 1940). The Los
Venezuela and Trinidad and is best described from Bajos and Warm Springs faults strike southeast and
seismic reflection profiles presented by Robertson are oblique to the trend of the E1 Pilar fault zone
and Burke (1989) in the area north of Trinidad (Fig. 4A). Tyson (1989) mapped transpressional and
(Fig. 4A). Its strike-slip offset is unknown. transtensional features along the fault trace and
showed that they are closely dependent on fault
East-southeast- to southeastward-striking faults terminations and slight changes in the direction of
in Venezuela and Trinidad the fault trace. The Los Bajos fault zone curves to
the northeast in southern Trinidad with strike-slip
Faults in the Foothills fold-thrust belt offset converted to thrust offset on the Galeota Point
Two east-southeastward-striking right-lateral thrust fault (Fig. 4A) that bounds the steep, southern
faults offset the frontal thrust fault of this Paleo- coast of the island and is adjacent to the Columbus
gene fold-thrust belt in central Venezuela by 10- foredeep basin (Fig. 4B).
30 km (Fig. 4A). This fold-thrust belt formed mainly
in Eocene and Oligocene time and is responsible Warm Springs-Central Range fault zone
for the formation and subsidence of the Gu~irico The Central Range fault zone was originally
sub-basin of the Eastern Venezuelan basin (Parnaud mapped by Kugler (1953) as a northwest-dipping
et al., 1995) (Fig. 4B). thrust fault that bisects the topographically higher
elevations of the Central Range and bounds the
Urica fault zone oldest outcrops of the lower Cretaceous Cuche For-
This southeastward-striking fault zone exhibits 35 mation (Fig. 4A). Based on new data presented in
km of right-lateral offset and forms the western edge this paper, we propose that the Central Range fault
of the Serranfa del Interior (Fig. 4A). Munro and is the onshore continuation of the offshore Warm
Smith (1984) constrained the 35-km right-lateral off- Springs fault zone in the Gulf of Paria and that both
set amount using gravity data. This fold-thrust belt faults are sub-vertical right-lateral strike-slip faults.
formed mainly in Oliogocene-Miocene time and A component of transtension on the Warm Springs
was responsible for the formation and subsidence and transpression on the Central Range fault can
of the Maturfn sub-basin of the eastern Venezue- account for the differences in the structural and to-
lan foreland basin (Erlich and Barrett, 1992; Roure pographic character of the Gulf of Paria and Central
et al., 1995; Parnaud et al., 1995; di Croce et al., Range.
Chapter 16) (Fig. 4B). Shortening is now inactive as
suggested by the distribution of earthquake epicen- Cenozoic sedimentary basins of northern South
ters (Fig. 2A) and unconformity relationships (Roure America and Trinidad and their relationship to
et al., 1995). major strike-slip and thrust faults
San Francisco fault zone and unnamed Cenozoic basins of this area can be divided into
parallel faults two groups of similar shape and tectonic origin:
This fault exhibits 18-40 km of right-lateral off- (1) basins elongate in an east-west direction and
set that is readily apparent on geologic maps of controlled by roughly east-west strike-slip fault
the Cretaceous and Paleogene passive margin units zones in the coastal areas of northern Venezuela
in the Serranfa del Interior (e.g., Parnaud et al., and northern Trinidad (Mor6n-E1 Pilar; E1 Coche-
1995) (Fig. 4A). The radar image shown in Fig. 3 North Coast); and (2) foreland basins elongate in
shows three other unnamed faults between the San an east-northeast direction and bounded by roughly
Francisco and the Gulf of Paria (Nos. 4, 5, and east-northeast-striking frontal thrusts of the Foothills
6) with similar strikes, tectonic geomorphology, and belt and Serranfa del Interior (Gu~irico and Maturfn
possibly strike-slip offsets. sub-basins of the Eastern Venezuelan foreland basin)
(Fig. 4B). The first group of strike-slip basins are
El Soldado fault zone much smaller in area than the second group and have
This fault is present in the eastern Gulf of Paria much thinner sedimentary fills (1-3 km) than basins
and is believed to be a fight-lateral strike-slip fault of the second group (5-10 km).
(Fig. 4A). Its offset is unknown.
North Coast basin
Los Bajos fault zone This offshore basin filled with marine sedimen-
The Los Bajos fault zone is continuous with the tary rocks forms the northern flank of the Northern
northwestern part of the Warm Springs fault zone Range of Trinidad and was described by Robertson

502 S. BABB and R MANN
~ ~
o
~0 ~'~
~=~
= I1.1 ~
~-~ . ~
~~
.~"~
~) -,~
~ ~
{o
"~ ~
~~

and Burke (1989) (Fig. 4B). Basal units are Mid- Lithologic belts of northern South America and
dle to Late Miocene in age, and overlie Pliocene their extensions in the Trinidad area
and Pleistocene rocks that are mainly claystone and
sandstone. The total thickness of the basinal rocks is The South America-Caribbean plate boundary
about 2 km. The North Coast basin appears to have zone can be subdivided into several subparallel litho-
formed along the obliquely downthrown side of the logic belts defined on the basis of similar litholo-
North Coast-E1 Coche fault zone (Fig. 4A). gies and ages of deformation and metamorphism
(Fig. 4C). Some of these belts appear to extend from
eastern Venezuela into Trinidad. From north to south
Gulf of Paria-Northern basin the main features of these belts are summarized
This off- and onshore, 2- to 3-km-thick basin below.
occurs south of the Paria Peninsula and Northern
Range and forms the focus of this paper (Fig. 4B). Caribbean volcanic arc
Basin subsidence is closely related to down-to-the- These mildly deformed and metamorphosed rocks
south and lateral movements on the right-lateral E1 represent the leading edge of the exotic Caribbean
Pilaf fault zone and down-to-the-north and lateral plate that is presumably moving eastward from its
movements on the Warm Springs-Central Range original Cretaceous location in the eastern Pacific
fault zone (Fig. 4A). The Southern basin in south- Ocean (Burke, 1988; Pindell and Barrett, 1990)
ern Trinidad formed in a post-Early Miocene low (Fig. 1B). This belt contains well exposed Early
south of the Central Range and was affected by an Cretaceous to Paleogene arc rocks on the island of
early phase of Middle Miocene folding and reverse Tobago northeast of Trinidad (Snoke et al., 1990),
faulting (Dyer and Cosgrove, 1992) along with Late Cretaceous-Paleogene granitic and basaltic intrusive
Miocene to Recent transpression. rocks on Margarita Island (Av6 Lallemant, 1997),
and Cretaceous-Eocene volcanic rocks from wells
and outcrops in the Leeward Antilles (Speed et al.,
Columbus basin
1985) (Fig. 4C). The precise boundary between the
The Columbus basin is the offshore basin directly
Caribbean volcanic arc and the Cordillera de la
adjacent to and southeast of the Galeota Point thrust
Costa belt to the south is poorly defined because
fault, that terminates right-lateral slip along the Los
elements of both belts are found across a broad zone
Bajos fault (Fig. 4B). Michelson (1976), Leonard
on the island of Margarita (Av6 Lallemant, 1997).
(1983) and di Croce et al. (Chapter 16) document
a thick fill of sediments mainly derived from Late
Cordillera de la Costa belt
Miocene to Holocene progradation of the Orinoco
This highly deformed metamorphic belt includes
delta of eastern Venezuela (Prieto, 1987) into non-
lithologies mainly affiliated with both the Mesozoic
marine to shelf-slope environments of the offshore
passive margin of South America and exotic oceanic
area separating eastern Venezuela and Trinidad.
lithologies presumably obducted during the oblique
collision between the Caribbean volcanic arc and the
Maturin and Gufirico sub-basins of the passive margin (Fig. 4C). Av6 Lallemant (1997) sub-
Eastern Venezuelan basin divided this belt into the Coastal Range/Margarita
These sub-basins of the Eastern Venezuelan basin belt consisting of rocks of oceanic affinity and the
formed as foreland basins mainly in Oligocene Cordillera de la Costa belt sensu stricto consisting
to Pliocene by thrust movement along the frontal of continentally derived rocks. The former belt con-
thrusts of the Foothills belt and Serran/a del Interior, tains serpentinites and other ophiolitic rocks while
respectively (Gonzalez de Juana et al., 1980; Rohr, the latter belt contains mainly low-grade metamor-
1991; Erlich and Barrett, 1992; di Croce et al., Chap- phic rocks of passive margin affinity and a few
ter 16) (Fig. 4B). The Urica fault forms the boundary metaigneous and serpentinite bodies. The age of the
between the Matur/n and Gufirico sub-basins (Erlich protolith of the metasedimentary rocks is known
and Barrett, 1992) (Fig. 4A). from a few places like the Araya-Paria Peninsula of
Venezuela (Av6 Lallemant, 1997) and the Northern
Range of Trinidad (Potter, 1976) to be of Jurassic or
Morochito piggyback basin Early Cretaceous age and to have been deposited in
This basin formed on top of the Serrania del In- a passive margin setting (Fig. 4C).
terior fold-thrust belt in response to out-of-sequence Deformation of these rocks is complex with one
thrust movement on the Pirital thrust fault (Route et or two metamorphic phases of deformation and sev-
al., 1995) (Fig. 4B). It is filled with a tilted sequence eral brittle phases. Av6 Lallemant (1997) proposed
of shallow-water and continental deposits of Late that the high-grade part of the belt on the Araya
Miocene to Pliocene age. Peninsula experienced pure and simple shear in its

504 S. B A B B and R M A N N
Fig. 4. (A) Offsets of strike-slip faults and thrust belts in northern South America. Offsets of both thrust and strike-slip faults along the
plate boundary are controversial. Some tectonic models call for large offsets >500 km for the Caribbean plate north of the Mor6n-E1
Pilar fault system. Observed offsets are much smaller but may represent minimum offsets as some major strike-slip faults like the E1
Coche-North Coast fault zone are not well mapped. The amount of shortening in the Serranfa del Interior fold-thrust belt has been
estimated at 45-90 km based on balanced cross-section estimates of Parnaud et al. (1995). Right-lateral strike-slip faults that strike
east-southeast to southeast have well-mapped offsets ranging from 10 to 40 km. (B) Cenozoic sedimentary basins in northern South
America. Major basins are formed either by strike-slip faults of the right-lateral Mor6n-E1 Pilar fault system or as foredeeps along major
thrust faults at the southern limits of the fold-thrust belts. The Morochito basin studied by Parnaud et al. (1995) is a piggyback basin
formed by thrusting within the Serranfa del Interior thrust belt. (C) Lithologic belts of northern South America formed as the Caribbean
arc obliquely collided with the passive margin of northern South America. The age of deformation youngs from an Eocene-Oligocene
event that formed the Foothills fold-thrust belt and obducted crystalline rocks of the Villa de Cura belt, to an Oligocene-Pleistocene event
that formed the Serranfa del Interior fold-thrust belt, to the Trinidad belt where deformation began definitely in the Late Miocene and is
continuing to the present.

original trench setting off northwestern South Amer- SUMMARY OF PREVIOUS TECTONIC MODELS FOR
ica while the low-grade parts of the belt experienced TRINIDAD
a similar style of deformation at a shallower depth
in the trench (Fig. 4C). Structures mapped by Av6 Introduction
Lallemant (1997) indicate a protracted history of
northwest-dipping (present geometry) thrusts. The The tectonic interpretation of complex structural
syn-metamorphic thrusting and folding event affect- relationships in Trinidad spans the period since sys-
ing passive margin sedimentary rocks of the North- tematic and detailed outcrop mapping of the entire
ern Range of Trinidad has an opposite sense of dip island was completed at a scale of 1 : 100,000 by Ku-
and vergence to the Araya Peninsula and may indi- gler (1959). Onland mapping in Trinidad combined
cate deformation in a different tectonic setting from with the larger-scale recognition of the eastward
its present geographic location (Algar and Pindell, relative motion of the Caribbean plate (Molnar and
1993). Sykes, 1969; Fig. 2B) has led to a variety of predic-
tive tectonic models for the Trinidad region.
Caucagua-Paracotos-Villa de Cura belt Most of the models agree that the Mesozoic-
These rocks are mainly metasedimentary passive Paleogene passive margin of northeastern South
margin rocks overthrust by the large, ophiolitic Villa America converted to the zone of presently ac-
de Cura nappe (Fig. 4C). tive deformation sometime during the Neogene and
that this event was linked to the interaction between
Foothills belt the margin and the eastward or southeastward-mov-
This belt forms a narrow fold-thrust belt in Creta- ing Caribbean plate (e.g., Pindell and Barrett, 1990).
ceous and Paleogene sedimentary rocks south of the Differences in models center on three fundamen-
Caucagua-Paracotos-Villa de Cura belt (Fig. 4C). tal questions concerning the interaction of the exotic
Figueroa de Sanchez and Hern~indez (1990) docu- Caribbean arc and the passive margin of northeastern
South America:
ment late Eocene-Oligocene north to south over-
thrusting into the Gu~irico sub-basin adjacent to this (1) When did the Trinidad area of the northeastern
belt (Fig. 4B) using evidence from an exploration passive margin of South America convert from a
well. passive to active margin?
(2) What type of active margin was initiated
(strike-slip vs. thrust) by the oblique passage of the
Serrania del Interior belt
Caribbean arc along the passive margin?
This east-northeast-trending fold-thrust belt falls
(3) What were the main fault controls (normal
on strike with the deformed zone in Trinidad and
vs. strike-slip vs. thrust) on active margin sedimen-
has been well studied and dated during petroleum
tation?
exploration of the Eastern Venezuelan basin (Erlich
and Barrett, 1992; di Croce et al., Chapter 16; Flinch
Sampling of tectonic models for Trinidad
et al., Chapter 17) (Fig. 4C). Thrusting is south- to
southeast-vergent, involves unmetamorphosed Cre-
A sampling of four of the more recent and widely
taceous to Pliocene lithologies, and occurred in
cited tectonic models are illustrated in Fig. 5 and
Oligocene through Pliocene time (Parnaud et al.,
briefly summarized here.
1995; Roure et al., 1995). Based on balanced cross-
sections, the total north-south shortening across the
Perez and Aggarwal (1981)
belt is about 45-90 km (Roure et al., 1995).
They compile earthquake epicenter locations and
focal mechanisms of larger events to propose a
Guyana shield shift in the location of strike-slip faulting from the
The Guyana shield is a pre-Grenville age Pre- east-west-striking, right-lateral E1 Pilar fault zone to
cambrian craton that dips northward beneath the the northwest-striking, right-lateral Los Bajos and
Foothills and Serranfa del Interior fold-thrust belts E1 Soldado fault zones in the Gulf of Paria, which
(Fig. 4C). they inferred to have a cumulative right-lateral offset
of 22 km (Fig. 5A). In this model, these surfi-
Trinidad belt cial strike-slip faults act as transform or tear faults
The Trinidad belt is a zone of strike-slip and separating the subducted oceanic part of the South
thrust deformation of Middle Miocene to Recent age America plate from the unsubducted continental part
in Trinidad, the Gulf of Paria, and the eastern shelf of the plate (Fig. 2B). These authors proposed that
of Trinidad. This deformation forms the topic of this the north-to-south shift in plate motion from the E1
paper and is also discussed by Flinch et al. (Chapter Pilar fault zone to the Los Bajos-E1 Soldado sys-
17) and di Croce et al. (Chapter 16) (Fig. 4C). tem (Wilson, 1940) was a Pleistocene response to

Trinidad from a relatively minor offset (~30 km)

strike-slip fault on the Paria Peninsula (Fig. 4A).
Robertson and Burke (1989)

They present a regional strike-slip model based
on a grid of seismic lines tied to wells in the
North Coast basin off the north coast of Trinidad
and field results from Late Pleistocene outcrops of
the Northern basin south of the E1 Pilar fault zone
(Fig. 5C). They conclude that Trinidad and the North
Coast basin was a diffuse and active zone of fight-
lateral faulting related to the eastward motion of
the Caribbean plate past South America (Fig. 4A).
They proposed that the stratigraphy of Trinidad was
the result of the strike-slip juxtaposition of sev-
eral blocks with distinct depositional histories that
are now expressed morphologically as the North-
ern, Central and Southern Ranges and the Northern
and Southern basins. Observed deformation patterns
in Late Pleistocene rocks near the E1 Pilar fault
zone were attributed to a right-lateral simple shear
mechanism.
Erlich and Barrett (1992)

They followed the idea of a predominantly strike-
slip boundary but proposed normal throws on south-
east and northeast faults bounding the rifted mar-
gins of the triangular Gulf of Paria-Northern basin
(Fig. 5D). The mechanism for rifting in the Gulf of
Paria-Northern basin was attributed to simultaneous
fight-lateral motion on the E1 Pilar and Los Bajos
Fig. 5. Summary of previous tectonic models proposed for fault zones and transfer of slip from the E1 Pilar to
Trinidad and the easternmost Eastern Venezuelan basin. (A) Los Bajos. A similar rift-related model for the Gulf
Perez and Aggarwal (1981) emphasize strike-slip tectonics along of Paria-Northern basin was proposed by Salvador
the southern limit of the Benioff zone (cf. Fig. 2B). (B) Speed
and Stainforth (1968). A Miocene age of movement
(1985) and Russo and Speed (1992)emphasize a southeastward-
to south-southeastward-verging fold-thrust belt with little or no was inferred from previously published structural
associated strike-slip tectonics in Trinidad. (C) Robertson and and stratigraphic data along with some new industry
Burke (1989) emphasize a diffuse zone of east-west-striking data presented in their paper. The Warm Springs
right-lateral strike-slip faults. (D) Erlich and Barrett (1992) em- fault zone was interpreted as a post-Miocene horse-
phasize two, east-west-striking strike-slip faults and intervening
tail splay fault of the Los Bajos fault zone.
normal faults bounding a triangular-shaped rift zone in northern
Trinidad. See text for further discussion of each model.
Previous published studies using industry data
from Trinidad
the southward widening of the Benioff zone formed
by the subducted Atlantic slab beneath Trinidad Several of the above tectonic models were hand-
(Fig. 2B). icapped by the lack of subsurface data needed to
constrain the character of fault zones like the contro-
Speed (1985) and Russo and Speed (1992) versial E1 Pilar fault zone and the overall structural
Using outcrop, gravity and regional seismicity style and age of structures in the Gulf of Paria and
data, these workers proposed that Trinidad and east- Northern basins (Fig. 4B). While limited industry
ern Venezuela is a south to southeast-verging fold- seismic and well data had been published from the
thrust belt with an associated foreland basin formed Gulf of Paria (Eva et al., 1989), offshore Los Bajos
by thrust loading (Fig. 5B). In their model, the E1 fault zone (Tyson, 1989) and the Gulf of Paria-
Pilar fault in Trinidad is regarded as a north-dipping Northern basin (Payne, 1991), most studies in the
thrust along which the Northern Range has been Southern basin-Gulf of Paria-Northern basin region
uplifted. In this model, the main strike-slip plate were local in scale, did not present a large amount
boundary fault has not yet propagated eastwards into of systematic mapping using seismic grids tied to

D E V E L O P M E N T OF A N E O G E N E T R A N S P R E S S I O N A L PLATE B O U N D A R Y 507
wells, and consequently did not attempt a tectonic are derived from well and seismic reflection data and
synthesis for the entire area of northern Trinidad. presented later in this paper.
However, in the area of the Caribbean Sea north of
Trinidad, Robertson and Burke (1989) synthesized a Basin and sub-basin nomenclature used in this
large industry data set consisting of eleven wells and study
5000 km of seismic profiles.
Payne (1991) proposed large, late Neogene nor- Basins
mal movement on the E1 Pilar fault zone using seis- For this study, the term 'Gulf of Paria basin' is
mic reflection data that were reinterpreted by Algar used to describe all sedimentary rocks in the northern
and Pindell (1993) who reached the same interpreta- offshore Gulf of Paria area that overlie latest Early
tion of normal faulting. Payne (1991) also proposed Cretaceous rocks of the passive margin section of
that the Gulf of Paria is a typical rift basin charac- northern South America that is horizontally ruled on
terized by a horst and graben structure. He proposed Figs. 7 and 9. The 'Northern basin' or 'Caroni basin'
two types of faults depending on their strike: east- (cf. Robertson and Burke, 1989 for this second usage)
west or northeast-southwest-striking faults are trans- refers to the similar succession of sedimentary rocks
pressional while northwest-southeast-striking faults within this onshore area of Trinidad (Fig. 6A).
are transtensional to extensional. Data coverage in the Gulf of Paria basin extends
only to the Trinidad-Venezuela boundary that bi-
sects the Gulf of Paria in a roughly north-south
DATA USED IN THIS STUDY OF THE GULF OF PARIA direction (Fig. 6A). Data from both the Venezuelan
AND NORTHERN BASIN, TRINIDAD and Trinidadian part of the Gulf of Paria basin are
described by Flinch et al. (Chapter 17) and Di Croce
A map showing new seismic reflection and well et al. (Chapter 16).
data used in this paper is shown in Fig. 6A. The
data set includes approximately 1600 km of mi- Sub-basins
grated seismic reflection data. Some of these data Mapping of seismic sequences on seismic reflec-
were reprocessed by the Petroleum Company of tion lines tied to wells reveals the presence of four
Trinidad and Tobago (Petrotrin). These data are tied sub-basins within the Gulf of Paria basin and the
to 28 wells with electric logs and well completion western part of the Northern basin (Fig. 9A). These
reports. basins, formally named in this paper for the first
The seismic reflection data used in this study are time, include the following.
concentrated in the Gulf of Paria while the well (1) E1 Pilaf sub-basin with a total Late Miocene-
data are concentrated in the Northern basin and Recent thickness in two-way time of about 3 s. This
its nearshore area in the Gulf of Paria (Fig. 6A). semicircular basin is asymmetrical with its steep
Only one reflection line used in this study (Fig. 36) northern side bounded by the sub-vertical plane of
has been previously published but is differently a southern strand of the E1 Pilar fault zone and its
interpreted by Payne (1991) and Algar and Pindell more gently sloping, southwestern flank bounded by
(1993). the Gulf high, a remnant Cretaceous passive margin
carbonate bank (Fig. 9A).
(2) Puerto Grande sub-basin with a total Late
MAJOR SUB-BASINS AND HIGHS OF THE GULF OF Miocene-Recent thickness in two-way time of about
PARIA AND WESTERN PART OF THE NORTHERN 2.3 s. This rectangular basin is also asymmetrical
BASIN with its steep northern side bounded by an unnamed
fault south and parallel to the E1 Pilar fault zone. Its
Tectonic map and regional cross-section southwestern end is bounded by the Avocado high, a
remnant Cretaceous passive margin carbonate bank
A tectonic map of Trinidad showing the major (Fig. 9A).
onshore basins and ranges was adapted and simpli- (3) Goodrich sub-basin with a total Late Mio-
fied from the more detailed regional maps of Kugler cene-Recent thickness in two-way time of about
(1959) and Persad (1984) (Fig. 6A). This map is 3.3 s. This basin has a more complex and irregu-
keyed to a regional cross-section A-A' adapted from lar geometry than the E1 Pilar and Puerto Grande
section D-D' that accompanied the regional geo- sub-basins. Its northern and northwestern edges are
logic map of Persad (1984) (Fig. 7). Formation bounded by the Galf and Domoil highs (Fig. 9A).
names shown on the cross-section are from Persad Its northeastern and southwestern edges are bounded
(1984) and are also shown in tabular form in Fig. 8. by large oblique-slip faults between the E1 Pilar and
Fig. 6B provides an index map of more detailed Warm Springs fault zones and its southern boundary
maps of the Gulf of Paria and Northern basins that is bounded by the Warm Springs fault zone.


PREVIOUS LITHOSTRATIGRAPHIC STUDIES OF THE The Mesozoic section in the Northern Range
GULF OF PARIA AND NORTHERN BASIN exhibits large-scale northward-verging folds formed
either by large-scale syn-sedimentary slumping or
Introduction tectonic deformation along with four phases of sub-
sequent brittle faulting (Algar and Pindell, 1993)
Previous studies subdividing the lithostratigra- (Figs. 7 and 10A, B). The major structure of the
phy of outcrops and well data from the Northern Northern Range is a large, overturned antiform
basin and Gulf of Paria area include Kugler (1953), whose gently curved, east-west-trending axial trace
Barr and Saunders (1968), Higgins and Saunders roughly parallels the 500- to 925-m-high, topo-
(1968), and Payne (1991). Payne (1991) made the graphic crest of the range (Fig. 10A).
first attempt to subdivide the post-Middle Miocene, The Northern Range is a direct along-strike con-
subsurface sequences for the Northern basin using tinuation of the Coastal Fringe/Margarita belt of the
wells and seismic reflection surveys. He defined a Araya-Paria Peninsula of Venezuela (Gonzalez de
depositional sequence framework using differences Juana et al., 1968; Av6 Lallemant, 1997; Fig. 4C)
in seismic responses, provided relative ages of the which exhibits many of the same Mesozoic rock
sequences, and discussed stratigraphic correlation types and styles of deformation (Fig. 10A). Fis-
across the Northern basin. This study attempts to sion track ages by Algar and Pindell (1993) from
build on Payne's work by integrating the subsurface Mesozoic metamorphic rocks of the Northern Range
geology of the Gulf of Paria and Northern basins indicate that the main phase of shallow unroofing
and the outcrop geology of the Northern and Cen- and erosion through a temperature of 200o-250 ~ be-
tral Ranges into a single stratigraphic framework gan early Late Miocene about 11 m.y. ago. This
(Fig. 8). age of unroofing and initial erosion of Mesozoic
rocks in the Northern Range is consistent with the
Outcrop stratigraphy of the Northern and presence and composition of the siliciclastic wedge
Central Ranges adjacent to the Gulf of of Late Miocene age (Cunapo Formation) shed into
Paria-Northern basins the Northern basin (Fig. 8).
The isolated outcrop in the Lower Cretaceous
Structure of Jurassic-Cretaceous rocks in Cuche Formation in the Central Range does not
the Northern and Central Ranges exhibit either large-scale folding or metamorphism
Late Jurassic-Early Cretaceous carbonate, sili- but does show evidence of brittle faulting probably
ciclastic, and volcanic rocks in Trinidad crop out related to its position and late Neogene uplift his-
as a variety of mapped formations of low-grade tory within the Mount Harris push-up block of the
metasedimentary rocks that have been studied in Central Range fault zone (Fig. 6A).
detail in the Northern Range (Trechmann, 1935; Ku-
gler, 1959; Barr and Saunders, 1968; Furrer, 1968; Jurassic-Cretaceous outcrop stratigraphy in
Saunders, 1972; Potter, 1976; Wadge and Macdon- the Northern and Central Ranges
ald, 1985; Frey et al., 1988; Algar and Pindell, 1993) Although the lithostratigraphy of the Northern
(Fig. 10). Unmetamorphosed Barremian to Aptian- Range is well mapped, the formation names and
Albian black-gray shale (Cuche Formation) has been structural interpretations by Kugler (1959) were
described from deep wells drilled in the onshore questioned and subsequently modified by Algar
Northern basin and offshore Gulf of Paria basin and Pindell (1993). We have maintained the Ku-
as well as from outcrops within the Mount Harris gler (1959) interpretations as outlined in Fig. 10B
push-up block along the Central Range right-lateral because they are consistent with the main features of
strike-slip fault (Barr, 1952; Bartenstein et al., 1957, the Kugler (1959) geologic map that are schemati-
1966; Koutsoukos and Merrick, 1985; Algar and cally shown in Fig. 10A.
Pindell, 1993) (Fig. 6A). The Mesozoic section of the Northern Range con-
Fig. 6. (A) Tectonic map of Trinidad showing major fault systems and associated sedimentary basins. Inset shows direction and rate of
South America plate relative to the Caribbean plate from DeMets et al. (1994). Thin lines in the Gulf of Paria and Northern basins are
multi-channel seismic reflection lines provided by the Southern Basin Consortium and Petrotrin for use in this study. Wells are identified
by capital letters. Well C corresponds to the Couva Marine-1 well of Bray and Eva (1983). Transtensional strike-slip basins including the
E1 Pilar, Goodrich and North Soldado sub-basins of the Gulf of Paria are found mainly offshore, while basins affected by transpressional
tectonics like the Northern basin and Southern basin are found onshore. Outcrops of metamorphosed Mesozoic passive margin rocks are
restricted to the Northern Range and Paria Peninusla of Venezuela. One isolated unmetamorphosed outcrop of lower Cretaceous passive
margin rocks (Cuche Fm.) is found in the Mount Harris area. (B) Index map showing locations of more detailed maps presented in this
paper and line of cross-section A-A' shown in Fig. 7. Regional cross-section A-A', modified from Persad (1984) and Robertson and
Burke (1989), is shown in Fig. 7.

(D
sists of a latest Jurassic through Barremian section
of low-grade metamorphic rocks, mainly of a mixed
siliciclastic and carbonate origin. These metamor-
phic rocks are conformably overlain by metamor-
phosed sedimentary rocks of Late Cretaceous age
r
(D (Fig. 10B). Most workers agree that carbonate and
9
minor evaporitic Mesozoic rocks of the Northern
Range and in the Couva Marine-1 well of the Gulf
(D
of Paria basin (Bray and Eva, 1983; Eva et al.,
9
.,..,
1989) record an early passive margin phase in the
geologic history of Trinidad as shown on the chart in
Fig. 8.
(D Tertiary passive margin stratigraphy

o,..~
~D
In general, the Cenozoic stratigraphic succes-
r/3
sion of the Northern Basin-Gulf of Paria is in-
9
~D
dicative of a progressive shallowing and infill
from Paleogene-Early Miocene, deeper-water, finer-
Z
grained passive margin or 'flysch-type sediments'
of the Chaudiere, Pointe-a-Pierre, and Brasso For-
~0
~
mations to the Early Miocene-Late Miocene, shal-
.=. lower-water and coarser-grained active margin or
9
'molasse-type' sedimentary rocks of the Cunapo
and Manzanilla Formations (Tyson and Ali, 1990)
(Fig. 8). The Paleogene-Early Miocene deeper-
9
~
water rocks are unevenly distributed and generally
(D
thin as one moves from south to north across the
9
~D Central Range as seen on the regional cross-section
.,..~
in Fig. 7. For this reason, Paleogene-Early Miocene
deeper-water rocks are thin in the subsurface of the
Northern basin and are commonly difficult to distin-
r/3
guish on seismic reflection lines from the overlying,
(D Neogene shallower-water units.
9
?: Passive to active margin transition

I
Transitional units between the deeper, passive
9
.,.., margin Paleogene units and the shallower-water
~9
(D
active margin Neogene units include the Brasso
O and Tamana Formations (Fig. 8). During the Early
9
Miocene, calcareous clay, silt, and conglomerate
.,..~ of the Brasso Formation were deposited across
9
the Northern Basin and Gulf of Paria (Fig. 7).
Foraminiferal and macrofaunal assemblages in the
~
Brasso Formation indicate an outer neritic environ-
r
ment of deposition. The Tamana Formation is a
reefal limestone unit developed in the Central Range
~9
during the latest Early Miocene and middle part of
:<
I the Middle Miocene (Erlich et al., 1993) (Fig. 8).
<
The Tamana Formation is interpreted as a shallow-
~
ing-upward sequence, where outer shelf and slope
shales grade stratigraphically upsection into high-
energy shoreline and intertidal limestone and shale
(Erlich et al., 1993).
~
oK The occurrence of the Tamana limestone indicates
the first appearance of shallow-water conditions in
the Central Range area (Carr-Brown and Frampton,
1979). Robertson and Burke (1989) and Erlich et al.

S T R A T I G R A P H Y OF THE N O R T H E R N BASIN AND

GULF OF PARIA, TRINIDAD
TIME (Ma) I AGE" POL.',,,N FORMATION S LITHOLOGY DEP. ENVlR. SEQ. TECTONICS
O Z
...... ,,, Cedros Sst, Cly, Cgl Coastal terraces
~,?,8 . . . . .
~9 ~ 5
h" rr .~ Upper Talparo Sst, Cly Brackish water
1.65 9 - " - I I I
Central Range
n
~. I Lower Talparo Sst, Cly Open marine transpression,

9 Gulf of Paria
3.5- ~ ~. ~ Springvale Sst, Sits, Cly Inner neritic transtension
E = ~;;.%
%=%"
' Northem Basin
=i/.-,p= receives sediment
5.2 """'"
i~.~.~:~. "
Manzan,lla Cgl, Sst, Sh Brackish water 9 from Central Range;
Late ~ t..~..~_-,~_,.~, to W a r m Springs, El Pilar
M ioc. ~ ~~~~_~.=. inner neritic ii faults active and pull-
=apart initiates
10.2 - "~ I
Early- = Tamana _Carbonate_.--'Shallow marine ,~~ "~iFirstshedding of ---

, reefs , 'clastic rocks from--
Middle
Mioc. ' Brasso_ , Sst, Sh, , , Not .~north; thrusting i n ~
Nanva'--Cgl (rare)-- Deep neritic -included-Southern basin ---
252 -- I
Olig Cipero Clavs Ma. in - First tectonic --
. S a n F emando.., - - ==.,~1. C , Lst,i-...,.,Sst,
oh .-.. ==~m~-ffallow water~nentic -- --"
Eoc '
N.. .. ... .. .. . . . . . . . . . . .Sst
...... " Deep water "
this ~activity on passive_
Pointc-c~Pieffe study margin? --
Pal' Chaudiere Sst, Sh i D~ep water
66.5 , Guyaguayare ..... " Siltstone/Claystone: Deep wate, ~ L 1
L. C'ret. NaDadma H i l = - -
II
96- ,
Gautier
Cuche. ~
Sh, Sst
Lst~ Sst, Sh
~
~
Slope
,
~['
Early ' ' Lst, Evp, Sh ~Carbonate platform~ 2------Passive margin

Cret. '--Couva Evaporite----' Evaporite, Sh ,------Restricted , :subsidence
131 -
Late Lst, Evp, Carbonate
Jurassic Dol, Sh platform?
Fig. 8. Nomenclature, age, lithology, and depositional environments of formations in the Northern basin and Gulf of Paria and their
correlation to sequences defined in this study. Data are compiled from Barr and Saunders (1968) and this study. Lined pattern indicates
passive margin lithologies and is keyed to the formations with a lined pattern shown on the cross-section in Fig. 7. Tectonic interpretations
in right column are proposed in this study.
(1993) have inferred that the shallow-water condi- Gulf of Paria basins (Fig. 8). The maximum drilled
tions over the Central Range represent the surface thickness of the Manzanilla Formation approximates
expression of large-scale folds forming along the 6500 feet (1950 m) near its depocenter in the E1 Pilar
plate boundary zone. Termination of carbonate sed- sub-basin south of the E1 Pilar fault zone (Fig. 9A).
imentation of the Tamana Formation correlates with Sequence B consists of sandstone and clay of the
the rapid influx of shallow water siliciclastic sedi- Plio-Pleistocene Springvale and Talparo Formations
mentary rocks during the Late Miocene (Tyson et (Fig. 8). The thickness of the Springvale Formation
al., 1991; Erlich et al., 1993) (Fig. 8). varies widely and suggests changing tectonic and
sedimentary conditions during its deposition.
Stratigraphic nomenclature and age control of For age control of his sequences A and B, Payne
Payne (1991) for the Northern basin and Gulf of (1991) uses the palynological zonation of the North-
Paria ern Basin developed by E. Gonzales (contract pa-
lynologist for the Petroleum Company of Trinidad
Payne (1991) subdivided the Late Miocene- and Tobago). The zonation is based on statistical
Recent stratigraphic succession of the Northern pollen counts and the first appearances of distinctive
Basin-Gulf of Paria into two sequences, A and B. marker palynomorphs. The specific species upon
Sequence A consists of the massive, Late Miocene which this zonation was based were not provided
Cunapo conglomerate, which aggrades and pro- by Payne (1991) or in the internal reports available
grades southward from the southern flank of the for this study. The pollen zones identified by Gon-
Northern Range and interfingers with sandstone zales include: UM I, UM II pollen zones Upper
of the Manzanilla Formation of the Northern and Miocene; PI, PII, PIII pollen zones - - Pliocene; and

512 S. B ABB and E MANN
Fig. 9. (A) Major tectonic elements and total isochron map for Late Miocene through Pleistocene sequences 3, 4, and 5 in the Gulf of
Paria (contour interval is 1000 ms). The depocenters are mainly formed by transtensional fault movements associated with or between
the E1 Pilar and Warm Springs fault zones. The Domoil and Gulf highs are remnant, constructional carbonate topography inherited from
a Late Jurassic-Early Cretaceous carbonate passive margin. Well locations are given by letters. (B) Location map of regional seismic
sections and geologic sections based on well logs shown in Figs. 11-20.
a Pleistocene zone. These zones and their correlation A correlation b e t w e e n these pollen zones with the
to the lithostratigraphic subdivision of the Northern planktonic biochronozones shown on the Haq et al.
Basin by Payne (1991) are used in this study and are (1987) sea level chart was proposed by R. Liska
s u m m a r i z e d on the stratigraphic column in Fig. 8. (pers. commun., 1995): U M I is equivalent to lower

N 16; UM II is equivalent to N 17 and the upper part Sequence 5: Late Pliocene to Pleistocene marine
of N 16; the P zones are probably equivalent to N18 to brackish-water sand, silt, clay, and minor con-
and part of N19 (3.5-5.2 Ma). glomerate. These rocks are equivalent to the Talparo
and Cedros Formations known from outcrops in the
uplifted northern and southern flanks of the North-
SEISMIC REFLECTION DEFINITION AND G E O L O G I C ern basin (Persad, 1984, 1985) (Fig. 8). These two
CORRELATION OF LATE M I O C E N E TO PLEISTOCENE formations exhibit increasingly brackish-water envi-
SEQUENCES 3-5 IN THE GULF OF PARIA AND ronments probably related to the regional uplift and
NORTHERN BASINS shallowing of the basin floor during this period and
the constriction of the basin caused by uplift of the
Criteria for defining seismic sequences in the Northern and Central Ranges.
subsurface of the Gulf of Paria and Northern
basin
Seismic character and definition of sequence 3:
The definition of seismic sequences proposed in Late Miocene Manzanilla and Cunapo
this study was based on the recognition of onlap, Formations
downlap, toplap or truncation surfaces on seismic
reflection lines. The sequence boundaries were cor- Distribution of sequence 3
related on all seismic reflection lines and wells in Using the seismic lines and wells shown in
the seismic grid shown in Fig. 6A. Using these cri- Fig. 6A, sequence 3 was mapped in the E1 Pilar
teria, we identify five Late Jurassic-late Neogene and Goodrich sub-basins of the Gulf of Paria basin.
sequences in the subsurface of the Gulf of Paria and Sequence 3 is also known from wells and seismic
Northern basins and correlate them with previously lines to extend eastward and underlie much of the
proposed lithologic formations in Fig. 8 and on the Northern basin. Sequence 3 is thickest in the E1
cross-section of Fig. 7. The five proposed sequences Pilar sub-basin adjacent to the E1 Pilar right-lateral
include the following. strike-slip fault zone and thins to the south in the
Sequence 1: Late Jurassic-Early Valanginian Goodrich sub-basin (Fig. 9A).
carbonate megaplatform consisting of carbonate-
evaporitic lithologic cycles (Fig. 8). This sequence
is described in detail by Babb (1997) and is not Base reflector of sequence 3
discussed in this paper. The base of sequence 3 on a north-south line
Sequence 2: Middle Valanginian-middle Ap- crossing the Gulf high between the E1 Pilar and
tian restricted carbonate banks developed on the Goodrich sub-basins onlaps the top of sequence 2
megaplatform (Fig. 8). This sequence consists of formed by remnant topography of the Cretaceous
evaporite-sand-shale cycles formed in a restricted carbonate bank in the center of the Gulf of Paria
bank setting. This sequence is described in detail by basin (Fig. 11). On a north-south line crossing the
Babb (1997) and is also not discussed in this paper. Avocado high, another Cretaceous carbonate high to
Sequence 3: Late Miocene-Early Pliocene shal- the east of the Gulf high, sequence 3 can also be
low marine to brackish water conglomerate and observed onlapping sequence 2 (Fig. 12).
sandstone. These rocks are equivalent in part to In the south of the Gulf of Paria basin, sequence 3
the Cunapo and Manzanilla Formations known from rests unconformably on a northward-thinning packet
wells along the northern flank of the Northern basin of reflectors that is known from wells C in Fig. 12
(Persad, 1984, 1985) (Fig. 8). The Cunapo Forma- and V in Fig. 13 to be deep-water sandstone and
tion represents the first influx of coarse sedimentary shale of the Middle Miocene Brasso Formation
rocks from the Northern Range into the Northern (Fig. 8). The Brasso Formation along with the rest
basin and interfingers southwards with sandstone of of the underlying, deep-water early Tertiary 'passive
the Manzanilla Formation (Fig. 8). margin' section of Trinidad characteristically thins
Sequence 4: Early to middle Pliocene inner neritic northward onto Early Cretaceous carbonate bank
to shallow marine conglomerate, sandstone, silt and rocks of sequence 2 (Persad, 1984) (Fig. 7).
clay. These rocks are equivalent to the Springvale
Formation and the upper part of the Manzanilla
Formation known from outcrops in the uplifted Top of sequence 3
flanks of the Northern basin (Persad, 1984, 1985) The top of sequence 3 is defined on the regional
(Fig. 8). Water depths increased slightly during northwest-southeast line in the Goodrich sub-basin
deposition of sequence 4 (Springvale Formation) as by onlap, toplap, and truncation reflection termina-
shown by the retrogradational nature of the well logs tions as seen on the seismic lines shown in Figs. 11-
from this unit discussed later in this paper. 13.

514 S. B ABB and R M A N N
(D %...~ %...~
,.~ O 9
,..a
.,..~
..~ ~, .,...,
,..a
.,..~
. ,...~
. ,...,
Z ~
.,.~ O
~~
9 ,-..,
tt'3
~D
.,zZ~
9 t,.., ..~
~3 ~ .,.., .,..,
o .~
.,..a
+..a
~o~

Fig. 11. North-south seismic section showing representative example of seismic sequences 2, 3, 4, and 5 with sequence 3 onlapping
the Lower Cretaceous carbonate bank and the top of sequence 3 defined by toplap and truncation (line of section shown on map in
Fig. 9B). The Gulf high is inferred to be constructional carbonate morphology that is elongate in a northwest-southwest direction and
developed on relatively flat megacarbonate platform of sequence 1 (see Fig. 9A for map view of Gulf high). Steep sides of platform led
to gravitational slumping of sequence 3 at its edges. On the southern margin of the bank, low-angle normal faults sole out at the base of
sequence 3. At the base of the northern margin of the bank, the basal, mounded part of sequence 3 suggests gravity flows derived from
the Gulf high.
Seismic facies within sequence 3 northward direction (Fig. 12).

A variety of seismic facies are present within se- On the southern flank of the Avocado high
quence 3 that include the following types: mounded (Fig. 9A), sequence 3 exhibits variable-ampli-
to chaotic, wedge, hummocky, sub-parallel to hum- tude, continuous-discontinuous reflectors that show
mocky, and prograding. medium frequency and are subparallel to h u m m o c k y
Mounded and chaotic facies are present in de- (Fig. 12). This seismic section shows reflections
positional lows at the base of the Gulf high in the south of well C that are of lower frequency and
E1 Pilar sub-basin (Fig. 11). Low-angle faulting is slightly higher continuity than the equivalent section
associated with this facies on the southern flank of north of well C.
the Gulf high carbonate bank (Fig. 9A). In the eastern part of the Goodrich sub-basin,
Near the northern margin of the E1 Pilar sub- sequence 3 exhibits sigmoidal-oblique, high-low-
basin near the E1 Pilar strike-slip fault zone, reflec- amplitude reflection packets whose sense of progra-
tions within sequence 3 consist of predominantly dation is from northwest to southeast (Fig. 13). Well
high-amplitude, discontinuous reflectors that have V confirms that this facies consists of sandstone and
a wedge-shaped external form which tapers in a shale.

Fig. 12. North-south section showing sequence 3 overlying the Lower Cretaceous section to the north and the Lower Miocene Brasso
Formation to the south (line of section shown on map in Fig. 9B). Logs of wells B and C provide lithologic control on seismic facies.
Well B shows that sequences 3 and 4 are conglomeratic and well C shows that the equivalent section is more distal sandstone and
shale. This pattern of north-to-south fining is observed throughout the Gulf of Paria and Northern basins and is interpreted as coarse
sedimentation shed off the east-west-striking E1 Pilar fault zone.
Correlation of seismic facies of sequence 3 with mainly conglomeratic rocks of the Cunapo Forma-
lithologies from electric logs and cores from wells tion of Late Miocene to Early Pliocene age (Fig. 8).
Unit 1 of this conglomeratic section is an 80-m-thick
We have made stratigraphic cross-sections based aggradational stack of massive conglomerate with
on aligning wells in the Goodrich sub-basin and thin shale interbeds increasing towards the top of
Avocado high to constrain the lithology and sed- the unit (Fig. 14). Unit 1 is bounded at its base by
imentary facies of seismic sequences 3, 4 and 5. an unconformity surface above the Early Cretaceous
Cross-sections were made in both an approximately carbonate platform section of the northern Avocado
north-south (Fig. 14) and in an approximately east- high (Figs. 9, 12). The top of unit 1 is bounded by a
west direction (Figs. 15 and 16) to provide a 10.5-m-thick shale interval. Unit I appears to pinch
better understanding of how seismic/sedimentary out to the south. The character of this pinch out is
facies vary on a basin-wide scale (locations of demonstrated in the seismic line through sequence 3
aligned wells shown in Fig. 9B). Spacing between in Fig. 12. A conglomeratic section very similar to
wells along these sections varies from 5 to 16 unit I is present in well F within the E1 Pilar fault
km. zone and on the north flank of the Puerto Grande
sub-basin (Fig. 9A).
Lithologic variations of sequence 3 in a Unit II in well B begins with sandstone and
north-south direction conglomerate units which display a blocky-shaped
Sequence 3 can be divided into three lithologic electric log signature and are possibly channel fea-
units in wells B, C1, C2, and V which vary in tures (Fig. 14). The electric log profile suggests
character in a north-south direction (Fig. 14). Well aggradation of units I and II. At wells C 1 and C2, 10
B at the western end of the Puerto Grande sub-basin km to the southwest of well B, and well V 18 km to
adjacent to the E1 Pilar strike-slip fault contains the southwest of B, unit II shows a more distal fa-

Fig. 13. Northwest-southeast seismic section showing the three sequences mapped and defined in this study (line of section shown in
Fig. 9B). Sequence 3 is a southeastwardly prograding unit defined at its upper boundary by onlap, toplap, and minor truncation. During
the Late Miocene-Early Pliocene, a reversal in progradation direction is indicated by the northwestwardly downlap of sequence 4, which
thins to the north by onlap. Within the upper part of sequence 4, a lapout surface is interpreted as a maximum flooding surface. The
boundary between sequences 4 and 5 appears concordant except in the very southeasterly part of the line, where possible truncation of
sequence 4 occurs. Sequence 5 is characterized by distinct, high-frequency shingling of northwestwardly prograding units. Well V (cf.
log in Fig. 14) is projected onto this section and provides age and lithologic control.
cies of sandstone and shale with minor conglomerate varying amounts of white, metamorphic vein quartz
(Fig. 14). The sand is predominantly grayish white, and disseminated carbonaceous material. At well
very fine-grained to medium-grained, fair to poorly V, unit III consists of thinly bedded sandstone and
sorted, and contains minor amounts of carbonaceous shale. The sand is medium-grained with rare coarse
material. The shale is gray, commonly sericitic, and grains and pebbles.
locally carbonaceous.
Unit III in well B consists of an interbedded Geologic interpretation of sequence 3 in a
conglomerate-shale interval which exhibits several north-south direction
9- to 12-m-thick prograding units (Fig. 14). At wells In the seismic section of Fig. 12, sequence 3 con-
C1 and C2, unit III consists of interbedded sandstone tains a southward-prograding conglomerate facies to
and shale. The sandstone is fine- to coarse-grained, the north adjacent to the E1 Pilar strike-slip fault.
generally poorly sorted, rarely pebbly, and contains This unit would correspond to the conglomeratic
subangular to subrounded grains. The sand contains facies of units I and II in well B that is closest to the

Fig. 14. North-south stratigraphic correlation between electric logs of wells V and B showing the correlation of sequences 3, 4 and 5.
The datum is the base of the Durham sand at the base of sequence 5. The lithologies of units II, III, IV, and V are described in detail in
the text. SB denotes sequence boundary.
fault, and is interpreted as a fanglomerate deposited seen south of well C generally indicate sediments
within a in a shallow marine to brackish setting. prograding into shallow water.
Coarse grain sizes of the conglomerate are attributed Clinoform geometries of reflectors south of well
to local source areas in the Northern Range and B in Fig. 12 indicate a southward increase in water
steep topographic slopes formed by movement along depth. A general increase in the sorting of sand in
the E1 Pilar strike-slip fault zone (Fig. 9). Mounded units II and III of sequence 3 suggests a southward
reflectors at the base of sequence 3 on the line in increase in the degree of marine reworking south-
Fig. 11 suggest the occurrence of gravity flow pro- ward along this inferred north-to-south paleoslope.
cesses in more distal areas of the E1 Pilar sub-basin. Clinoform geometries in sequence 3 south of well
A shallow-water depositional setting is inferred for C (Fig. 12) and south of well V (Fig. 13) indicate
the reflectors north of well C in Fig. 12 because of a similar pattern of southward increase in water
their position on the Avocado high (Fig. 9A) and be- depth and related increase in the amount of marine
cause the type of subparallel to hummocky reflectors reworking.

WEST EAST
WELL E ~ ' I 6 KM WELL V 16.5 KM WELL J
I
R
~<~ SEQUENCE 5 I
~ 0
1-
w
_J
<
"r
03
I
, ,
DURHAM SAND
Z
.----=SB
MAXIMUM FLOODING SURFACE
UNIT V
I.U
7'
O
1-
03
a SEQUENCE 4 UNIT IV
z
r <
03
14.1
SB
c3
T a
BRACKISH mu.i UNIT III
I ~
n..
LU
I--
'SEQUENCE 3
z_
MIDDLE MIOCENE UNIT II

SHALE OF BRASSO FM.
DETAILEDLOG OF MIDDLE
MIOCENE UNCONFORMITYAT WELL E
SHELF
MIDDLE MIOCENE
UNCONFORMITY SURFACE
SB
L,=18GO
MIDDLE MIOCENE,
SHALE OF BRASSO FM.
Fig. 15. East-west stratigraphic correlation between electric logs of wells E and J showing the correlation of sequences 3, 4, and 5.
The datum is the base of the Durham sand at the base of sequence 5. The lithologies of units II, III, IV, and V are described in detail
in the text. SB denotes sequence boundary. Paleoenvironments based on benthic foraminifera are listed to the left. Inset shows detail of
electric log of well E. Paleodepth work in well E reveals a major environmental change from middle-outer shelf in the Middle Miocene
to brackish in the Late Miocene. Lithologic data from the well indicate that this abrupt transition is represented by an unconformity with
possible subaerial exposure indicated by red shale horizon.
Conglomerates of units I and II are correlated Manzanilla Formation into three, shallowing-upward
with the Cunapo Formation known mainly from drill- members: (1) the oldest San Jose member consists of
holes including well F of the Northern basin (Payne, dark, calcareous silts, rich in small molluscs and for-
1991; Tyson et al., 1991) (Fig. 9A). Equivalent and aminifera deposited in an inner neritic environment;
interfingering sandy units to the south are corre- (2) the Montserrat glauconitic sandstone overlies the
lated with the Manzanilla Formation which has been San Jose and was deposited on a shallow shelf; and (3)
well described from outcrops along both flanks of the Telemaque member records shallowing and infill-
the Northern basin (Saunders, 1968; Carr-Brown and ing of the Northern basin as indicated by arenaceous
Frampton, 1979). These workers have subdivided the foraminiferal faunas.

Fig. 16. East-west stratigraphic cross-section between wells B and S showing the correlation between sequences 3, 4, and 5. The datum
is the base of the Durham sand at the base of sequence 5. The lithology of unit IV is described in detail in the text. The top of the Lower
Cretaceous platform in wells B and S is indicated.
Lithologic variations of sequence 3 in an west at well B and in the east at well S. The
east-west direction conglomeratic wedge thickens eastward from 1230
Sequence 3 can be divided into units II and m at well B on the Avocado high to 2340 m at well
III described in detail above on two approximately S. The boundaries between conglomerate, sandstone,
east-west basin profiles using wells E, V and J in and shale units I, II and III of sequence 3 are not
the Goodrich sub-basin and the southern part of the well defined on this profile (Fig. 16).
Northern basin (Fig. 15) and wells B, H, T1, P, and
S on the Avocado high and Northern basin (Fig. 16).
On the first profile in Fig. 15, the lower boundary Geologic interpretation of sequence 3 in an
of sequence 3 is not penetrated in well V but is east-west direction
recognized in both wells E and J. Unit II of sequence The southward transition from massive conglom-
3 appears to be missing in well E, which is located erate along the E1 Pilar fault zone to sandstone and
in the Goodrich sub-basin (Fig. 9B). The east-west shale of the basin axis remains a relatively straight,
profile along the northern part of the Gulf of Paria east-west boundary. The conglomerate wedge ap-
and Northern basins using wells B, H, T1, P, and pears to have been about 20 km wide in the west and
S in Fig. 16 indicates two areas of predominantly narrowed to a width of 10 km or less in the Northern
conglomeratic fill of the Cunapo Formation in the basin.

Age of sequence 3 (Cunapo and Manzanilla Top of sequence 4

Formations) The upper boundary of this sequence is an ero-
sional truncation. The underlying sequence 3 is
The age of sequence 3 and its equivalent litho- disrupted by what appears to be listric normal faults
logic formations shown in chart form in Fig. 8 (Figs. 17 and 18).
is based on paleontology and palynology of well
cuttings tied to seismic lines. The source of these Seismic facies within sequence 4
biostratigraphic data was internal company reports A variety of seismic facies are present within
of the Petroleum Company of Trinidad and Tobago. sequence 4 including prograding oblique clinoforms
Sequence 3 is a Late Miocene-Early Pliocene, and associated mounds, low-amplitude continuous
mainly shallow-marine to brackish-water conglom- reflectors, and very low-amplitude, continuous re-
erate and sandstone unit whose age is based on flectors.
recognition of palynozones UMI and UMII in well East of the Goodrich sub-basin, high- to medium-
V (Fig. 15) which has been tied to the seismic line amplitude, relatively continuous, oblique clinoforms
shown in Fig. 13. prograde eastward and downlap apparent mounds at
Sequence 3 is bounded at its base by the post- the base of sequence 4 (Fig. 19).
Middle Miocene unconformity surface identified us- In the Goodrich sub-basin, sequence 4 is char-
ing well data. The age of the lower boundary of acterized by a zone of very low-amplitude and
the unconformity is based on recognition of the continuous reflectors (Fig. 20). The low-amplitude
Globorotalia mayeri zone (top Middle Miocene) in seismic facies appears transitional to higher-ampli-
well E (Fig. 15). Truncation beneath the Middle tude reflectors on the more proximal northern and
Miocene unconformity is not clear on lines shown southern ends of this northwest-southeast line.
in Figs. 12 and 13 because of poor seismic imaging In the E1 Pilar and Goodrich sub-basins, a zone of
at this depth. The base of the siliciclastic wedge of two to five, very high-amplitude continuous reflec-
sequence 3 (Cunapo Formation) was deposited in an tors occurs near the upper contact of unit 4 (Fig. 11).
outer neritic environment based on paleobathymetric This seismic facies is found in the northern and
data from well B that is tied to the seismic line southern parts of the Gulf of Paria basin.
shown in Fig. 12 (internal report of the Petroleum
Company of Trinidad and Tobago, 1991). Correlation of seismic facies of sequence 4
with lithologies from electric logs and cores
from wells
Seismic character and definition of sequence 4: Cross-sections were made in both an approxi-
Early to middle Pliocene Manzanilla and mately north-south (Fig. 14) and east-west direction
Springvale Formations (Figs. 15 and 16) to provide a better understanding
of how seismic/sedimentary facies of sequence 4
Distribution of sequence 4 vary on a basin-wide scale. Spacing between wells
Using the seismic lines and wells shown in along these sections varies from 5 to 16 km.
Fig. 6A, sequence 4 was mapped in detail in the
E1 Pilar, Puerto Grande, and Goodrich sub-basins of Lithologic variations of sequence 4 in a
the Gulf of Paria basin. Sequence 4 is also inferred to north-south direction
extend eastward and underlie much of the Northern Sequence 4 can be divided into two lithologic
basin (Fig. 7). Sequence 4 is thickest in the Goodrich units in wells B, C1, C2, and V which vary in
sub-basin adjacent to the Warm Springs right-lateral character in a north-south direction (Fig. 14). Well
strike-slip fault zone and thins to the north in the E1 B at the western end of the Puerto Grande sub-basin
Pilar and Puerto Grande sub-basins (Fig. 9A). adjacent to the E1 Pilar strike-slip fault contains
mainly conglomeratic and sandy rocks of the Cu-
napo Formation of Late Miocene to Early Pliocene
Base reflector of sequence 4 age (Fig. 8). Unit IV of this conglomeratic section is
The base of sequence 4 on a northwest-southeast a 50-m-thick retrogradational stack of massive con-
line crossing the Goodrich sub-basin (Fig. 9A) is glomerate, sandstone, and shale that is transitional
defined by downlap and onlap of a northwest- to equivalent sandstone and shale at wells C1 and
ward-directed sedimentary wedge south of well V C2 10 km to the south (Fig. 14). In well V, unit IV
(Fig. 13). Onlap of the base of sequence 4 across a consists of poorly organized sand overlain by a well
deformed sequence 3 is also observed on a north- developed shale interval interpreted as a maximum
south line in the E1 Pilar sub-basin (Fig. 18). Se- flooding surface. The position of this maximum
quence 4 undergoes pronounced thinning over these flooding surface is defined by the turnaround from
underlying structural highs. a retrogradational stacking pattern of unit IV to the

Fig. 17. North-south seismic section in northern Gulf of Paria basin showing the boundary between sequences 4 and 5 defined by
erosional truncation. See Fig. 9B for location.
progradational stacking pattern of unit V (Fig. 14). (Fig. 14). The very high-amplitude, continuous re-
The east-west cross-section of the basin using wells flectors in the upper part of sequence 4 in the north-
E, V, and J (Fig. 15) and B, H, T1, R and (Fig. 16) ern and southern parts of the basin are interpreted to
S shows a similar shale-dominated unit V above the represent very uniform, quiet sedimentary conditions
maximum flooding surface. during the period of maximum flooding recorded by
shaley horizons of unit V in Figs. 14 and 15. This
Geologic interpretation of sequence 4 in a maximum flooding surface indicates a change from
north-south direction retrogradation to progradation within sequence 4 and
Very low-amplitude continuous reflectors seen on defines a major reorganization in basin deposition.
a north-south line through the Goodrich sub-basin
are interpreted as a uniform shaley lithology in the Geologic interpretation of sequence 4 in an
central part of the Goodrich sub-basin (Fig. 20). east-west direction
This lithology is transitional with higher-amplitude Mounded reflectors at the base of sequence 4 on
reflectors to the north and south. These high-ampli- the line in Fig. 19 suggest the occurrence of gravity
tude reflectors at the basin edges are interpreted as flow processes and formation of lobes at the base of
representative of higher-impedance sand and con- a slope during eastward progradation of sequence 4.
glomerate at the northern and southern edge of the Clinoform geometry also seen in Fig. 19 indicates
basin (Fig. 12). an eastward increase in water depth.
Retrogradational stacking of unit IV in well B Seismic characteristics and age of sequence 4
indicates basin deepening and overall transgression indicate that it can be correlated with the Springvale

Fig. 18. North-south seismic section in northern Gulf of Paria basin showing onlap and thinning of sequence 4 onto folded and tilted
sequence 3. The section also illustrates syn-faulting growth along young faults affecting sequence 5. See Fig. 9B for location.
Formation, which is interpreted by Carr-Brown and narrowing of the Northern basin and consequent de-
Frampton (1985) to represent a marine incursion velopment of restricted, brackish water environments
into the Northern basin (Fig. 8). Faunas from the (Fig. 6A). Deepening in the central part of the basin
Springvale Formation indicate its deposition in an may have resulted in the transgressive and retrogra-
inner neritic environment. dational character of sequence 4. In the northern
Despite the deepening trend in sequence 4, brack- Gulf of Paria basin, deposition of sequence 4 was
ish water conditions persisted along the margins of localized in the Puerto Grande sub-basin (Fig. 9A)
the Gulf of Paria and Northern basins adjacent to which indicates an eastward shift in the depocenter
the Northern and Central Ranges (Fig. 6A). Pale- from the E1 Pilar sub-basin of sequence 3 time.
obathymetry data from well L in the southwestern
Northern basin (Fig. 6A) indicate that rocks of se- Age of sequence 4
quence 4 at this locality record a coastal environment The age of sequence 4 and its equivalent litho-
with marginal marine influence. Uplift of the Cen- logic formations is based on paleontology and paly-
tral Range during this time may have resulted in nology of well cuttings tied to seismic lines. The

Fig. 19. Northwest-southeast seismic section in Goodrich sub-basin showing eastward progradation and downlap of clinoforms of
sequence 4 over those of sequence 3 and mounded reflectors at base of sequence 4. See Fig. 9B for location.
source of these biostratigraphic data was inter- Base reflector of sequence 5

nal company files of the Petroleum Company of The base of sequence 5 is defined by an erosional
Trinidad and Tobago. Sequence 4 is a mid-Pliocene, unconformity in the eastern Gulf of Paria basin
mainly shallow-marine to brackish-water sandstone (Fig. 17). In other parts of the Gulf of Paria basin,
and shale unit whose age is based on recognition of the base of sequence 5 is conformable with sequence
palynozones PI and PII (Fig. 8). 4 (Figs. 18-20).
Seismic character and definition of sequence 5" Top of sequence 5

Late Pliocene-Pleistocene Talparo and Cedros For the purpose of this study, the top of se-
Formations quence 5 is assumed to be the seafloor in the Gulf
of Paria basin or the ground surface in the North-
Distribution of sequence 5 ern basin. Payne (1991) identified two additional
Using the seismic lines and wells shown in unconformity surfaces within the Early to Middle
Fig. 6A, sequence 5 was mapped in the E1 Pilar, Pleistocene of sequence 5 that are not recognized in
Puerto Grande, and Goodrich sub-basins of the Gulf this study.
of Paria basin. Sequence 5 is also inferred to extend
eastward and underlie much of the Northern basin Seismic facies within sequence 5
(Fig. 6A). Sequence 5 is thickest in the Goodrich Three main seismic facies are present within
sub-basins adjacent to the Warm Springs fight-lateral sequence 5: prograding oblique clinoforms and asso-
strike-slip fault zone and thins to the north in the E1 ciated mounds, low-amplitude continuous reflectors,
Pilar and Puerto Grande sub-basin (Fig. 9A). and continuous and parallel reflectors.

Fig. 20. Northwest-southeast seismic section in the Goodrich sub-basin showing low-amplitude seismic facies of sequence 4 in the
central part of the basin and clinoforms showing southeastward progradation in sequence 5. See Fig. 9B for location.
In the southeastern Goodrich sub-basin, high- derstanding of how seismic/sedimentary facies of

to medium-amplitude, relatively continuous, oblique sequence 5 vary on a basin-wide scale. Spacing be-
clinoforms of sequence 5 cyclically prograde north- tween wells along these sections varies from 5 to 16
westward (Fig. 13) and southeastward (Fig. 20). km.
These clinoforms are continuous with relatively par-
allel reflectors interpreted as basinal, deeper-water
equivalents of strata represented by the clinoforms.
Lithologic variations of sequence 5 in a
Correlation of seismic facies of sequence 5 north-south direction
with lithologies from electric logs and cores Sequence 5 can be correlated as a relatively
from wells uniform tabular body on wells in a north-south
Cross-sections were made in both an approxi- direction (Fig. 14). A 24-m-thick sandstone unit, the
mately north-south (Fig. 14) and in an east-west Durham sand, is present near the base of sequence 5
direction (Figs. 15, 16) to provide a better un- in all wells in the profile (Fig. 14).

Geologic interpretation of sequence 5 in a INTEGRATION OF SEQUENCE IDENTIFICATIONWITH

north-south direction STRUCTURAL MAPPING
Clinoforms in the southeastern part of the
Goodrich sub-basin indicate northwestward progra- Introduction
dation of sequence 5 from source areas in the Cen-
tral Range (Fig. 13). The presence of southeastward In order to best illustrate the tectonic controls on
progradation of sequence 5 (Fig. 20) into the same the three Neogene sequences defined above in the
basin suggests that the Goodrich sub-basin was a Gulf of Paria, we present structural contour maps
closed basin surrounded by source areas to the north of the boundaries between the three sequences. In
(Northern Range) and south (Central Range). addition, we present three isochron maps of the
Paleoenvironmental data (Internal company re- sequences overlying each of these surfaces. Com-
port, 1991) from wells B (Fig. 14) and E and V parison of the structural, contour and isochron maps
(Fig. 15) indicate a shallow-marine to brackish-water can reveal correlations between the geometry of the
depositional environment during the deposition of surface preceding the deposition of a sequence and
sequence 5 in the area of the Goodrich sub-basin and the thickness of that sequence. The comparison of
Avocado high (Fig. 9A). This type of environment these map pairs also allows better visualization of the
had been recognized in the Lower and Upper Talparo space available at tectonically significant times in the
Formations using outcrop samples from the northern history of the Gulf of Paria and the sedimentary in-
and southern edges of the Northern basin (Kugler, fill onto that sequence stratigraphic boundary. These
1953; Carr-Brown and Frampton, 1979) (Fig. 8). comparisons indicate source areas and paleogeogra-
phy of land areas surrounding the basin. Limits of
the approach are due to the effects of tectonic defor-
Geologic interpretation of sequence 5 in an
mation, subsidence, and differential compaction.
east-west direction
The area of thickest deposition of sequence 5
Neogene deformation phase one: Late
appears closely controlled by faults bounding the
Miocene-Early Pliocene strike-slip motion along
Goodrich sub-basin (Fig. 9A). For this reason, the
the El Pilar fault
sediment thickness decreases abruptly to the east in
the Northern basin as seen from the thinner sections
Structure contour map at top of sequence 2
of sequence 5 in wells T1, P, and S in the interior
(latest Early Cretaceous carbonate platform)
parts of the Northern basin (Fig. 16).
This map (Fig. 21A) reveals the following fea-
tures at the top of the Cretaceous platform horizon.
Age of sequence 5 (Talparo and Cedros (1) The Gulf Domoil and Avocado highs. These
Formations) highs formed by the morphology of the latest Early
The age of sequence 5 and its equivalent litho- Cretaceous bank margin that was later modified
logic formations is based on paleontology and pa- by Neogene faulting (Babb, 1997). The Cretaceous
lynology of well cuttings tied to seismic lines. The highs form the main east- and northwest-trending,
source of this biostratigraphic data was internal com- relief-forming features on the seafloor at the time
pany files of the Petroleum Company of Trinidad of initial deposition of the siliciclastic sediments in
and Tobago. Sequence 5 is a Late Pliocene to Pleis- sequence 3 (Fig. 21B).
tocene, mainly shallow-marine to brackish water (2) Lows. Prominent lows include the area north-
sandstone and shale unit (Fig. 8). east of the Gulf high and north and northwest of
The inferred environments of deposition for se- the Avocado high (El Pilar sub-basin) and an area of
quence 5 present in wells B, E, and V range from shal- several isolated lows in the area between the Gulf
low marine to brackish water (Figs. 15, 16). Sequence and Domoil highs (Goodrich sub-basin). These lows
5 is correlated with the Upper Talparo and Cedros in the Gulf of Paria appear to have formed along
Formations (Fig. 8). The Lower Talparo Formation normal faults related to pull-apart formation.
represents a marine environment while the Upper Tal- (3) Faults. Because faulting is predominantly of
paro is indicative of brackish water (Carr-Brown and Late Miocene-Pleistocene age in the Gulf of Paria
Frampton, 1979). In coastal exposures of the Talparo and Northern basins, the fault pattern on the top of
Formation along the east coast of Trinidad, Saunders the Cretaceous platform resembles that mapped in
(1968) identified lagoon, beach, and tidal fiat environ- the late Neogene sequences.
ments. The Pleistocene Cedros Formation is correla-
tive with the upper part of sequence 5 and consists Isochron map of sequence 3 (Late
of poorly consolidated clay, fine to coarse sandstone Miocene-Early Pliocene)
with fragments of leaves and other carbonaceous mat- The isochron map of sequence 3 (Fig. 21B)
ter (Kugler, 1953; Donovan, 1994). reveals two lobate sediment bodies composed mainly

Fig. 21. (A) Structural contour map in two-way travel time to top of sequence 2 (latest Early Cretaceous horizon). The location and
history of the latest Early Cretaceous edge of carbonate platform is indicated by heavy line. (B) Isochron map of Late Miocene-Early
Pliocene sequence 3 in the Gulf of Paria and location of lines used to constrain isochrons. Source of sequence 3 is from the E1 Pilar fault
zone and Northern Range.

of conglomerate of the Cunapo Formation. The Isochron map of sequence 4 (middle-Late

body is about 0.8 s in two-way time and fills the Pliocene)
low between the Gulf, Domoil and Avocado highs The isochron map of sequence 4 (Fig. 22B)
(El Pilar sub-basin). Southward thinning indicates reveals a major northwest-trending lobe that is con-
a northern source for sequence 3 in the Northern fined by the faults bounding the Goodrich sub-basin.
Range as observed in the north-south profile shown Northward thinning indicates a mainly southerly
in Fig. 14. source in the Central Range with additional sources
A second, smaller lobe of sequence 3 is present i n the west and northwest.
south of a narrow saddle separating the Gulf and Structural control on sedimentation during this
Avocado highs. This lobe appears to consist of phase is pronounced in the Goodrich sub-basin but
thinner, 0.2 s two-way time material from the larger less pronounced on more discontinuous faults in the
lobe to the north that was spilled southward through area south of the E1 Pilar fault zone (Fig. 22A). We
the saddle area. Because of this saddle, the Goodrich interpret this pattern as reflecting gradual abandon-
sub-basin remained partially isolated from an influx ment of the E1 Pilar fault zone and transfer of slip to
of sequence 3 from the northern source area during the Warm Springs fault zone across the oblique-slip
this period. faults in the Goodrich sub-basin.
Structural control on sedimentation within the
basin itself during this phase was not pronounced. Neogene deformation phase three: Late Pliocene
Gravity-driven coarse siliciclastic sedimentation ap- to Pleistocene strike-slip motion along the Warm
pears to have filled existing lows adjacent to topo- Springs-Central Range fault zone
graphic highs created along the Early Cretaceous
carbonate passive margin. The erosion of the North- Structural contour map at top of sequence 4
ern Range is interpreted here as reflecting down-to- (Late Pliocene horizon)
the-south, oblique-slip motions on the E1 Pilar fault This map (Fig. 23A) reveals the following fea-
zone. We interpret this Late Miocene vertical motion tures at the top of sequence 4.
as the first manifestation of strike-slip displace- (1) The Gulf Domoil and Avocado highs. The
ment on the E1 Pilar fault zone along which the carbonate banks still form prominent seafloor highs
Northern Range or equivalent rocks of the Coastal despite the filling of their intervening lows during
Fringe/Margarita lithologic belt of Margarita and structural phases one and two.
Trinidad move eastward along the northern margin (2) Lows. The prominent low during this period
of South America (Fig. 4C). remains the Goodrich sub-basin north of the Warm
Springs fault zone. Continued fault movement on the
Neogene deformation phase two: middle to Late oblique-slip faults may be have depressed the basin.
Pliocene strike-slip motion along the Warm (3) Faults. The fault pattern at the top of se-
Springs-Central Range fault zone quence 4 reflects increased oblique opening of the
Goodrich sub-basin. The faults of this structural
Structural contour map to top of sequence 3 phase affect a larger area and are wider spaced
(Early Pliocene horizon) than the faults of the early phase in the Goodrich
This map (Fig. 22A) reveals the following fea- sub-basin. This change in fault pattern suggests in-
tures at the top of sequence 3 (Early Pliocene creased opening of the basin. Movement on these
horizon). faults will act to create additional space to be filled
(1) The Gulf Domoil and Avocado highs. These by sediments of sequence 5 because of the increased
highs still formed prominent highs despite some rate of displacement along faults initiated in phase
filling of lows during structural phase one. two.
(2) Lows. The prominent low during this period
was within the Goodrich sub-basin north of the Isochron map of sequence 5 (Late
Warm Springs fault zone. Pliocene-Pleistocene)
(3) Faults. The fault pattern at the top of sequence The isochron map of sequence 5 (Fig. 23B)
3 reflects increased oblique opening of the Goodrich reveals a widening and lengthening of the major
sub-basin. Movement on these faults is generally northwest-trending lobe that is confined by the faults
down toward the northwest-trending axis of the basin bounding the Goodrich sub-basin. Northward thin-
and acted to create additional accommodation space ning indicates a continuation of the southern source
for the sediments of sequence 4. The deeper struc- in the Central Range and possibly the Orinoco delta
ture or topographic edges of the northwest-trending, that began in structural phase two (Fig. 22B).
underlying Cretaceous carbonate rocks may have Structural control on sedimentation during this
acted to control the faults of the Goodrich sub-basin phase was pronounced in the Goodrich basin but
that parallel the edges of the banks. not so obvious in the area south of the E1 Pilar

Fig. 22. (A) Structural contour map in two-way travel time to top of sequence 3 (Early Pliocene horizon). (B) Isochron map of
Middle-Late Pliocene sequence 4 in the Gulf of Paria and location of lines used to constrain isochrons. Source of sequence 4 is mainly
from the Warm Springs-Central Range fault zone and the Central Range, with additional sources from the west and northwest.

530 S. BABB and E M A N N
Fig. 23. (A) Structural contour map in two-way travel time to top of sequence 4 (Late Pliocene horizon). (B) Isochron map of Late
Pliocene-Pleistocene sequence 5 in the Gulf of Paria and location of lines used to constrain isochrons. Source of sequence 5 is mainly
from the Warm Springs-Central Range fault zone and the Central Range with possible input from the Orinoco River. Dark gray indicates
thickest accumulation of sequence 5.

fault zone. We interpret this pattern as reflecting STRUCTURAL STYLES AND GEOMORPHIC
gradual abandonment of the E1 Pilar fault zone and EXPRESSION OF THE EL PILAR FAULT ZONE
transfer of slip to the Warm Springs fault zone across
the oblique-slip faults in the Goodrich sub-basin. Introduction
These faults open the Goodrich pull-apart basin and
increases its accommodation space. Here we illustrate the structural styles and geo-
morphic expressions of the E1 Pilar fault zone. The
character, age of motion, and tectonic role of the
E1 Pilar fault zone has been controversial (Fig. 5).
Comparison of paleocurrents and This controversy has been in large part because there
paleogeography produced during structural has been little seismic data available across the fault
phases one through three trace to constrain its structure and age of movement.
We suggest on the basis of new seismic reflection
Introduction data presented here that the strike-slip style seen
Fig. 24 combines the isochron map of Figs. 21- along the E1 Pilar fault and its apparent inactivity
23 with paleocurrent data derived from seismic sec- in the Quaternary as one moves west along its trace
tions and from east-to-west and north-to-south strati- supports a general model for progressive west-to-
graphic sections based on alignments of wells (Figs. east abandonment of the E1 Pilar fault zone. This
14, 15, 16). idea is consistent with our interpretation of the struc-
tural and isochron maps summarized in Fig. 24A-C
in the previous section.
Paleocurrents
Paleocurrents for structural phase one (deposition Transpressional style of the El Pilar fault zone in
of sequence 3) are oriented parallel to and normal the Gulf of Paria
to the axis of the Gulf of Paria-Northern basin
(Fig. 24A). North to south paleocurrents interpreted Two lines across the E1 Pilar fault zone in the
from clinoforms and thinning in the Late Miocene northern Gulf of Paria both suggest that the fault
Cunapo conglomerate (Fig. 8) are consistent with is a high-angle strike-slip fault with a transpres-
the erosion of the Northern Range. Basin axis clino- sional component (both locations shown on map in
forms such as one near well B (Fig. 9A) indicate a Fig. 23). A seismic line east of Patos Island shows
west to east flow along the floor of the basin. Pale- four fault strands of which the central strand breaks
ocurrents for structure phases two (sequence 4) and the seafloor and appears active (Fig. 25). Down-to-
three (sequence 5) become increasingly dominated the-south components of movement on the southern
by a west to east paleoflow along with a southeast to strands have allowed a thick accumulation of se-
northwest flow in the area of the Goodrich sub-basin quences 3 and 4. The fault has an east-northeast
(Figs. 24B, C). strike in this area which appears to be a transpres-
sional fault orientation throughout Trinidad (Robert-
son and Burke, 1989; Payne, 1991; Tyson, 1989).
The upthrown northern side of the fault may have
Environments of deposition uplifted Patos Island to the west of the location of
Environments of deposition reflect the early uplift the line shown in Fig. 25. On the seismic line in
and erosion of the Northern Range during structure Fig. 25, overstepping of the southern two strands by
phase one followed by later uplift of the Central sequence 5 of Late Pliocene-Pleistocene age is con-
Range and the gradual decrease in the importance sistent with decreased activity (or at least focussing)
of the Northern Range and possibley Orinoco sili- of active movement on the E1 Pilar fault zone.
ciclastic sources during structure phases two and The line in Fig. 26 near the northeast coast of the
three. Environments remained shallow and brackish Gulf of Paria (Fig. 23) shows three fault strands that
along the E1 Pilar fault scarp and its fan apron are onlapped by sequences 3, 4 and 5. This onlap is
based on well samples from wells shown in Figs. consistent with the abandonment of the active traces.
14 and 16. To the south in the Gulf of Paria and Sequence 3 exhibits northward growth consistent
Northern basins, early environments in phase one with the isochron map in Fig. 2lB. Sequence 5 is
are brackish but become deeper water in phase two affected by a late period of transpressional reactiva-
and return to brackish conditions in phase three. The tion. As in the case of the line in Fig. 25 near Patos
deepest-water area in phases two and three remains Island, the fault trace has a slight east-northeast ori-
the fault-controlled Goodrich sub-basin (Fig. 24B, entation that might be responsible for the late phase
C). of transpressional reactivation.


Transtensional style in diffuse extensional zone top of the Early Cretaceous platform and the over-
south of the El Pilar fault zone in the Gulf of lying siliciclastic sediments is subjected to slightly
Paria transtensional opening at the time of sequences 4
and 5. Downslope slumping off of the carbonate
The line in Fig. 27 is located across the trend of passive margin highs may contribute to the rapid
the Puerto Grande sub-basin (location in Fig. 23). infilling of the Goodrich sub-basin as seen on the
The east-west-striking rift or negative flower struc- isochron maps of sequences 4 (Fig. 22B) and 5
ture reflects local extension responsible for the for- (Fig. 23B).
mation of this local depocenter and thickening seen
in sequence 5. The faults bounding this sub-basin Transtension between the Gulf and Domoil highs
cannot be traced as continuous features to the west
across the Gulf of Paria. This localized zone of The line in Fig. 30 illustrates transtension be-
extension may therefore represent a southward step tween these two highs with an early phase of
in the main trace of the E1 Pilar fault zone that transpression that has resulted in folding (Fig. 23).
places it in a more basinward position within the Thickening of sequence 5 shows that the later phase
onland Northern basin. This southward step would of movement has been oblique opening. This type
lead to a decrease in activity of the E1 Pilar fault of oblique opening and widening of the fault pattern
along the mountain front of the Northern Range can be seen by comparing the structural maps of
(Fig. 9A). sequences 4 (Fig. 22B) and 5 (Fig. 23B).
Onland geomorphic expression of the El Pilar

fault zone STRUCTURAL STYLES AND GEOMORPHIC
EXPRESSION OF THE WARM SPRINGS-CENTRAL
Fig. 28 is an aerial photograph showing the Arima RANGE FAULT ZONE
fault zone and onland extension of the E1 Pilar fault
zone. In this area, the onland projection of the E1 Warm Springs fault zone
Pilar fault zone is covered by alluvium and appears
to have no geomorphic expression. Some or all of its The seismic lines in Figs. 31 and 32 show the
geomorphic expression may have been lost during negative flower structure profile of the active, right-
the intensive development of this heavily populated lateral Warm Springs fault zone. The syn-tectonic
and urbanized area south of Port-of-Spain, the cap- thickening of Plio-Pleistocene sequence 5 north of
ital and largest city of Trinidad and Tobago. The the Warm Springs fault seen on the isochron map
Arima fault appears to have had mainly down-to- of Fig. 23B is also apparent on the seismic lines in
the-south oblique-normal throw followed by late Figs. 31 and 32.
up-to-the-south oblique-reverse reactivation as seen
on the line shown in Fig. 26 (Algar and Pindell, Transition between offshore Warm Springs and
1993). onshore Central Range fault zone
The geologic map in Fig. 33 and the aerial

STRUCTURAL STYLE OF OBLIQUE SLIP FAULTS IN photo in Fig. 34 shows the abrupt transition at
THE GULF OF PARIA the western coast of Trinidad between the off-
shore east-west-striking and transtensional Warm
Detachment off flank of Gulf high Springs fault zone and the onshore, northeast-strik-
ing and transpressional Central Range-Caigual fault
The line in Fig. 29 illustrates the style of de- zone. Thick Early Pliocene-Pleistocene rocks of
tachment of lower Tertiary or Upper Cretaceous to sequences 3-5 deposited by northward prograda-
Pleistocene siliciclastic rocks off the Gulf high in tion along the eastern margin of offshore Goodrich
the northern part of the Goodrich basin (Fig. 23). sub-basin are exposed by uplift and folding along
This structure is believed to result when the large the northern edge of the Warm Springs-Central
lithologic contrast across the unconformity at the Range fault zone (Fig. 34). Widening of the out-
Fig. 24. (A) Paleogeographic map of the Gulf of Paria and Northern basins and basin flanks during the Late Miocene-Early Pliocene
(sequence 3); (B) Paleogeographic map of the Gulf of Paria and Northern basins and basin flanks during the Early and Middle Pliocene
(sequence 4). (C) Paleogeographic map of the Gulf of Paria and Northern basins and basin flanks during the Late Pliocene-Pleistocene
(sequence 5). These maps were made by combining the isochron data shown in Figs. 21-23 with the well logs shown in Figs. 14-16 and
with outcrop data published by previous workers and discussed in the text.

Fig. 25. North-northeast-striking seismic line showing structure of east-northeast-striking, transpressional part of the E1 Pilar fault zone
in the marine strait (Boca Grande) east of Los Patos Island in the Gulf of Paria (see Fig. 23A for location). Note onlap of southern
and northern strands by sequence 5 of Late Pliocene-Pleistocene age and folding of reflectors between the fault strands. Seafloor scarp
indicates continued activity on the central fault strand.
crop pattern of these rocks along the edge of the Reinterpretation of the structure of the Central
Northern basin is consistent with the transpres- Range
sional nature of the uplift adjacent to the north-
east-trending Central Range and Caigual fault zones There are two possible interpretations of the
(Fig. 33). n o r t h - s o u t h section of the southwestern end of the

Fig. 26. North-south-striking seismic line from the northeastern Gulf of Paria basin showing structure of east-west segment of the E1
Pilar fault zone in the northeastern Gulf of Paria (see Fig. 23A for location). Note onlap of southern traces of the fault by sequences 3
and 4 of Early to Late Pliocene age and reactivation of the fault to form fold in sequence 5 of Late Pliocene-Pleistocene age.
Central Range adjacent to the Gulf of Paria (line the northern flank of the Central Range are prob-
of section shown in Fig. 6B). The first by Ku- lematic because younger rocks are thrust over older
gler (1953, 1959) and Speed (1985) proposed that rocks.
the Central Range is an asymmetric, south-south- A reinterpretation of the outcrop pattern is pro-
east-verging Late Neogene fold-thrust belt based on posed here based on offshore profiles of the Warm
outcrop mapping. In the Kugler (1953, 1959) inter- Springs fault zone shown in Figs. 31 and 32. In
pretation, shown in Fig. 35A, the thrust faults along this interpretation shown in Fig. 35B the Central

Fig. 27. North-south-striking seismic line showing structure of the discontinuous zone of transtensional faults south of the E1 Pilar fault
zone that form the Puerto Grande sub-basin in the northeastern part of the Northern basin (see Fig. 23A for location). Note growth of
Late Pliocene-Pleistocene sequence 5 along faults.
Range is a symmetrical, late Neogene restraining This reinterpretaton is consistent with the inter-
bend along the transpressional Central Range and pretation of the line shown in Fig. 36 from the
Caigual strike-slip faults. The topographic uplift of central part of the onland Northern basin (location
Mount Harris, the highest point of the Central Range shown in Fig. 33). In this line, the southeastern ex-
at 265 m, occurs at a slight left step between these tension of the Fishing Pond fault zone is expressed as
two transpressional faults (Fig. 33). The high-angle a 5-km-wide sub-vertical strike-slip fault or 'flower
Central Range-Caigual fault zone thrusts older rocks structure' along which the Guatapajaro anticline has
of the Central Range over younger rocks along the been folded as a transpressional feature. The dip on
southern flank of the Northern basin. The predicted the thrust faults beneath the anticline is southeast-
depth to the top of the Mesozoic carbonate platform ward and not northwestward as depicted by Kugler
is closer to the 3.5 km depth known from offshore (1953, 1959) and Speed (1985). This transpressional
and onshore seismic lines than the much greater interpretation of the Central and Southern Ranges
depth predicted in the Kugler section (Fig. 35B). uplifts bounded by thrusts and reverse faults dipping

Fig. 28. Air photograph of the Port-of-Spain area showing the scarp of the Arima fault separating the Mesozoic Chancellor and Lavantille
Formations along the southern edge of the Northern Range (see Fig. 23A for location). Note the onlap of alluvium on the lower mountain
front along the northern margin of the Northern basin. The projection of the offshore E1 Pilar fault zone seen on seismic lines projects
into the southernmost strand of the E1 Pilar fault zone shown in alluvium of Northern basin by Kugler (1959).

Fig. 29. Seismic line showing detachment surface between flat-lying Cretaceous rocks along the southwestern margin of the Gulf high
and undifferentiated Lower Tertiary and Upper Cretaceous rocks and overlying Miocene-Pleistocene rocks of sequences 3-5 (see Fig.
23A for location). Movement may be a gravitational response to Tertiary sediment loading or tectonic dip-slip movement along the Gulf
fault zone, a transtensional fault between the E1Pilar and Warm Springs strike-slip fault zones.
inward beneath the ranges is also consistent with the Fig. 38 are based on wells drilled in the Barrackpore
regional cross-sections of Persad (1984) (Fig. 7). and Penal oil fields in the Southern basin (Dyer and
Cosgrove, 1992) (see locations of lines on map in
Fig. 37). The thrust faulted margin in both areas is
STRUCTURAL STYLES AND AGE OF DEFORMATION similar to that interpreted in the section of Fig. 35B
IN THE SOUTHERN BASIN along the northern margin of the Central Range. A
major difference is that this Southern basin deforma-
Introduction tion is southward verging and begins earlier than that
of the Northern and Gulf of Paria basin deformation,
The Southern basin of Trinidad has been well because Middle Miocene ('~ 11.4 m.y. horizon Gg.7)
mapped during petroleum exploration and offers a units onlap onto slightly older Middle Miocene units
wealth of structural and stratigraphic information (Dyer and Cosgrove, 1992; Fig. 38).
that needs to be reconciled with any regional tec- The unconformity between Late Miocene-Early
tonic model (Fig. 37). The two cross-sections in Pliocene sequence 3 and middle to Late Pliocene

sequence 4 in the Northern and Gulf of Paria

basins correlates with a less prominent unconfor-
mity between the Upper Cruse and Lower Forest
Formations of southern Trinidad (Fig. 8). The two
cross-sections in Fig. 38 document the older Mid-
dle Miocene tectonic phase in the Southern basin.
However, this deformation does not appear to be
a regional, low-angle thrust event as envisioned by
Kugler (1953, 1959) and Speed (1985).
Correlation of unconformity at the base of

sequence 3 with an unconformity in the Southern
basin of Trinidad
Based on the similarity in age, the shallowing

event defining the base of sequence 3 is correlated
with the major Middle Miocene unconformity doc-
umented in Trinidad's Southern basin (Fig. 8). The
structural sections of Dyer and Cosgrove (1992)
show the onlap and thinning of Late Miocene-
Pliocene sediments on the Middle Miocene uncon-
formity (defined by the top of the Globorotalia
mayeri-Gg 7 biozone). The folding during the Mid-
dle Miocene has been interpreted to be the result
of convergence in response to arrival of the Lesser
Antilles arc and accretionary prism along this part
of the South American passive margin (Dyer and
Cosgrove, 1992; Algar and Pindell, 1993).
DISCUSSION
Summary of Middle to Late Miocene-Pleistocene

fault evolution in the Trinidad area
Fig. 30. Seismic line showing east-west-striking transtensional
A summary of the Middle Miocene to Late fault trough between the Gulf and Domoil fault zones (see Fig.
Pliocene-Pleistocene fault evolution of the Gulf of 23A for location). Note evidence for folding within sequences 3
and 4.
Paria and western Northern basin is given in Fig. 39.
The fault and depocenter maps are based on the
original structural contour and isochron maps shown megaplatform (cf. Figs. 11, 12; Babb, 1997). Some
in Figs. 21, 22, and 23 and summarized in Fig. 24. regional lines such as the one across the northwest-
ern edge of the Goodrich sub-basin do show thrust
Middle Miocene faulting (Fig. 13), but these faults appear localized
During this interval reverse faults and loading af- like the ones seen in the Southern basin (Fig. 38).
fected the area of the Southern basin (Figs. 38, 39A). Because of these observations, we conclude
This deformation is attributed to the earliest trans- that the Northern, Gulf of Paria and Southern
pressional effects related to the oblique collision of basins of Trinidad are unaffected by major contrac-
the Caribbean arc with the northern passive mar- tional deformation until the onset of transpressional
gin of South America (Pindell and Barrett, 1990). and transtensional faulting associated with discrete
We do not interpret this event as a regional folding strike-slip fault zones like the E1 Pilar and Warm
and thrusting event as envisioned by Kugler (1959), Springs in the Late Miocene. Prior to a relatively
Speed (1985), and, more recently, Flinch et al. subtle Middle Miocene folding and faulting event
(Chapter 17), because we do not observe pervasively (Fig. 38), Trinidad was characterized by passive mar-
folded and thrust rocks at the level of sequence 2 gin sedimentation and subsidence. Algar and Pindell
in the Gulf of Paria and Northern basins. Instead, (1993) have proposed that uplift of the Central and
we observe a relatively flat-lying basement (defined Southern Ranges in the Middle Miocene may have
at the reflector we take to be top of the carbonate been related to the peripheral bulge formed as the

Fig. 31. North-northwest-striking seismic line across the Warm Springs fault zone and across the southwestern flank of the Domoil high
in the southern Gulf of Paria basin (see Fig. 23A for location). The Warm Springs fault zone exhibits a negative flower structure zone
characteristic of a slightly transtensional strike-slip fault.
Caribbean arc approached the subducting passive the Central Range to sea level. From this time, the
margin of northeastern South America. We can offer Central Range divided the Northern and Southern
no new observations to support or contradict their basins which exhibit differing stratigraphies as seen
bulge hypothesis. on the regional cross-section in Fig. 7. The uplift
The Tamana limestone, a reefal unit formed at sea of the Central Range may be linked to an early
level in the Middle Miocene (Robertson and Burke, phase of folding seen by Middle Miocene time in the
1989; Erlich et al., 1993), marks the appearance of Southern basin (Dyer and Cosgrove, 1992; Fig. 38)

Fig. 32. North-south-striking seismic line across the Warm Springs fault zone exhibiting a negative flower structure profile characteristic
of a slightly transtensional strike-slip fault (see Fig. 23A for location)
and Northern basin (Fig. 13) and/or to movement passive margin (Fig. 1B). The Northern Range, a
along the Warm Springs-Central Range-Caigual complexly metamorphosed and folded block of pas-
strike-slip fault zone (Fig. 33). sive margin lithologies, was also moving across the
area at this time after being deformed and metamor-
Early Late Miocene phosed in a setting far to the west (Burke, 1988; Av6
During this period a lobe of conglomeratic sed- Lallemant, 1997). Basin topography during this time
iment was shed from the Northern Range and is was heavily influenced by the presence of remnant
interpreted to reflect significant lateral movement topography inherited from Early Cretaceous, north-
along the E1 Pilar fault zone (Fig. 39B). We concur east-facing banks formed during the passive margin
with the proposal of Robertson and Burke (1989) phase. Strike-slip faulting is largely concentrated on
that the fault is propagating from west to east during the E1 Pilar zone and has not spread southward
this time as the accretionary wedge and forearc of to the Warm Springs fault zone or Central Range
the Lesser Antilles moved past the South American (Fig. 39B).

542 S. BABB and P. MANN
Fig. 33. Geologic map of the Central Range and Warm Springs-Central Range-Caigual fault zones compiled from a variety of sources
discussed in text.
The Los Bajos fault is thought to have propagated sional widening of the Goodrich sub-basin and an
southeastward in Late Miocene time as marked by the influx of sequence 3 sediments into the rifted area.
subsidence of the North Soldado sub-basin at its junc-
tion with the Warm Springs fault zone (Radovsky and Early to middle Pliocene
Iqbal, 1979; Tyson et al., 1991). Termination of this This interval continued the subsidence and fault
fault by thrusting on the Southern Range widens the pattern of the previous interval as motion continued
zone of Neogene deformation. The Orinoco delta be- to be transferred from the E1 Pilar to the Warm
gins to fill the Southern basin and offshore Columbus Springs system (Fig. 39D). Faults became wider
basin of Trinidad at this time (Leonard, 1983). spaced and more extensive in the Goodrich sub-basin
and may reflect the increased extension related to
Late Miocene-Early Pliocene pull-apart opening. Subsidence of the North Soldado
During this interval of deposition of sequence 3 in basin (Radovsky and Iqbal, 1979; Tyson et al., 1991)
the Northern and Gulf of Paria basins, there was rapid reflected southeastward propagation of the Los Bajos
increase in strike-slip and oblique-slip faulting to the fault beyond its junction with the Warm Springs fault
south along the Warm Springs fault zone and in the zone.
Goodrich sub-basin (Fig. 39C). This southward shift We believe that the primary sediment supply to
is interpreted as the development of a pull-apart basin the Goodrich pull-apart basin during this interval
at the stepover area between the E1 Pilar and Warm is the Central Range because clinoforms generally
Springs fault zones in the Gulf of Paria area. Transten- dip to the northwest away from that feature and
sional opening of the pull-apart basin transfers slip because the area of south-central Trinidad formed
from the E1 Pilar to the Warm Springs fight-lateral a long-lived, broad high which restricts the inflow
faults and deactivates the previously active, eastern of south-derived Orinoco River sediments (Fig. 7).
continuation of the E1 Pilar in the Northern basin However, it is possible that the Orinoco River delta
(Fig. 39C). This conclusion differs from Robertson and sources to the north and northwest also are
and Burke's (1989) interpretation that the eastern E1 contributing to the infilling of the basin.
Pilar fault remains active and is possibly accelerat-
ing in slip rate. Transpressional uplift of the Central Late Pliocene-Pleistocene
Range accompanies increased displacement along the A major wedge of sediment derived from uplift
Warm Springs-Central Range-Caigual fault zones and erosion of the Central Range along with sed-
(Fig. 33). Increased displacement also led to transten- iment from the north, northwest, and possibly the

Fig. 34. (A) Aerial photograph of the Warm Springs fault zone area at the western coast of Trinidad (see Fig. 23A for location).
(B) Interpretation of photograph in (A) showing transition between offshore transtensional Warm Springs fault zone and onshore
transpressional Warm Springs-Central Range-Caigual fault zone. Thick Early Pliocene-Pleistocene rocks of sequences 3-5 deposited
by erosion and northward progradation along the eastern margin of offshore Goodrich sub-basin are exposed by monoclinal uplift and
folding along the northern edge of the Warm Springs-Central Range-Caigual fault zone.

Fig. 35. Two contrasting interpretations of a north-south cross-section of the southwestern end of the Central Range adjacent to the Gulf
of Paria. Key to numbered formations the same as shown on compilation map by Kugler (1953): 1 - 5 -- Cretaceous; 6 = Cretaceous
fragments; 7-12 = Tertiary. (A) Kugler (1953) proposed that the Central Range was an asymmetric, south-southeast-verging Late
Neogene fold-thrust belt based on outcrop mapping. Note that his interpretation of thrust faults along the northern flank of the Central
Range is problematic because younger rocks are thrust over older rocks. (B) Reinterpretation of the outcrop pattern based on offshore
profiles of the Warm Springs fault zone shown in Figs. 31 and 32 suggests that the Central Range is a symmetrical, late Neogene
restraining bend structure along transpressional strike-slip faults. Note that the high-angle Warm Springs fault zone thrusts older rocks
of the Central Range over younger rocks along the northern flank of the Northern basin and that the predicted depth to the top of the
Mesozoic carbonate platform is closer to the 3.5 km depth known from offshore and onshore seismic lines than the deeper depth depicted
in the Kugler (1953) section.
Orinoco River delta continues to infill the Goodrich transtensional as seen by the negative flower struc-
sub-basin (Fig. 39E). Tyson et al. (1991) suggested ture in Fig. 32. Continued offset on the Los Bajos
that the North Soldado sub-basin continued to be ac- fault (Wilson, 1940) and linked thrusting in the Cen-
tive and recorded southeastward propagation of the tral Range helped the deformation belt to widen in
Los Bajos fault which attained a total lateral offset of southern Trinidad.
10.5 k m on fold axes of the Southern basin (Fig. 37).
Activity on the E1 Pilar system to the east in the How similar is the geology of the Eastern
Northern basin consisted of transpressional reactiva- Venezuela basin and Trinidad?
tion (Fig. 26) perhaps related to the slightly north
of east strike of the fault in this area (Fig. 23A). A fundamental question about the tectonic evolu-
Activity on the part of the east-west-striking W a r m tion of Trinidad is whether its structure, stratigraphy
Springs fault in the offshore area continued to be and age of deformation is identical to that seen in

Fig. 36. North-south-striking, onland seismic line across the east-central part of the Northern basin (see Fig. 33 for location) showing a
sub-vertical strike-slip fault zone as postulated in the interpretation of Fig. 35B and transpression-related folding of Late Pleistocene-
Recent age that affects Late Miocene to Pleistocene sequences 3-5. Note evidence for Late Pleistocene-Recent transpressional
reactivation of the E1 Pilar fault zone along the northern edge of the basin. A similar style of transpressional reactivation of the E1 Pilar
fault zone is seen in the seismic line from the northern Gulf of Paria that is shown in Fig. 26.
the Gufirico and Maturfn sub-basins of the Eastern of thrust-related structures of the Maturfn sub-basin
Venezuelan basin to the west (Fig. 4B). This ques- extend to the east-northeast across the Gulf of Paria
tion was originally posed in the title of the 1965 and into Trinidad, albeit with a slightly younger age
paper by Salvador and Stainforth entitled 'Clues of deformation in Trinidad (Fig. 4A). Other workers
in Venezuela to the geology of Trinidad, and vice like Robertson and Burke (1989) and Erlich and Bar-
versa' and deserves to be reexamined in the light of rett (1992) have emphasized that faults in Trinidad
the new data presented in this paper. are dominantly strike-slip faults and therefore dif-
Some authors like Speed (1985) and Russo and fer in character from the predominant thrust faults
Speed (1992) have emphasized that the same set seen in the fold-thrust belts and associated foredeep

Fig. 37. Geologic map of the Southern Range compiled from Kugler (1953) and other sources discussed in the text.
basins of the Eastern Venezuelan basin (Erlich and rocks are not exposed, have not been encountered in
Barrett, 1992; Passalacqua et al., 1995; di Croce et wells, and have not been seen on seismic reflection
al., Chapter 16). lines (Fig. 40A). At depth beneath the carbonate
The purpose of this section is to point out some passive margin section of Trinidad seen on the re-
differences in the geologic character of the two areas gional lines of Figs. 11 and 12, basement is possibly
derived from comparison of regional cross-sections a thinned equivalent of the Precambrian crust crop-
shown in Fig. 40. The structure and stratigraphy of ping out in the Guyana shield (Fig. 40B).
the Maturfn sub-basin is summarized in the form
of a regional cross-section modified from Parnaud Late Jurassic rift phase
et al. (1995) (Fig. 40B) while the structure and Late Jurassic half-grabens are present at depths
stratigraphy of Trinidad is summarized in the form of 1-2 km in the zone of thin foreland sedimentary
of a regional cross-section modified from Persad cover near the Guyana shield (Fig. 40B). These rifts
(1984) and Robertson and Burke (1989) (Fig. 40A). record a phase of opening between South America
The horizontal and vertical cross-sections of the two and YucaUin Peninsula in Late Jurassic time (Pindell
sections are approximately the same to facilitate and Barrett, 1990; Mann, Chapter 1).
comparison. In Trinidad no rifts of Jurassic age have been
drilled or observed on seismic reflection lines
Comparison of the main stratigraphic features of (Burke, 1988; Fig. 40A). Rifts are possibly present
the Eastern Venezuelan basin and Trinidad area beneath the thick Early Cretaceous carbonate and
in cross-section siliciclastic rocks of the passive margin section de-
scribed by Babb (1997) beneath the Gulf of Paria
The regional cross-sections of eastern Venezuela and seen on the regional lines of Figs. 11 and 12.
(Fig. 40B) and Trinidad (Fig. 40A) reveal the fol-
lowing stratigraphic information about this part of Latest Jurassic-Paleogene passive margin
the passive margin of northern South America. phase
In the Eastern Venezuelan basin, a thick se-
Precambrian basement quence of passive margin rocks are seen on seismic
Basement rocks of the Precambrian Guyana lines dipping northward beneath the foreland basin
shield can be seen dipping northward beneath the (Fig. 40B). More distal equivalents of these Early
Maturin foreland basin and the Serranfa del Inte- Cretaceous-Eocene units have been imbricated in
rior fold-thrust belt (di Croce et al., Chapter 16) the Serranfa fold-thrust belt that is described in de-
(Fig. 40B). A peripheral bulge related to its flex- tail by Parnaud et al. (1995), Roure et al. (1995) and
ure beneath the overthrust belts is located near the di Croce and Bally (Chapter 16). In both Venezuela
northern margin of the Guyana shield (Fig. 40B). and Trinidad the passive margin history can be sub-
In Trinidad pre-Jurassic crystalline basement divided into two phases:

Fig. 38. Cross-section based on wells of the Barrackpore (A) and Penal (B) oil field within the Southern basin (from Dyer and Cosgrove,
1992; see Fig. 37 for location). Note the steepness of the faulted margin similar to that shown in the line of Fig. 29 along the northern
margin of the Central Range and also note onlapping of late Middle Miocene (~11.4 m.y. horizon Gg.7) onto slightly older Middle
Miocene units. The unconformity between Early Pliocene sequence 3 and middle to Late Pliocene sequence 4 in the Northern basin
would correlate to a less prominent unconformity below the top middle Cruse horizon.
(1) An early shallow carbonate bank phase from of this b a n k that forms the surface b e n e a t h the G u l f of
latest J u r a s s i c ? - M i d d l e Cretaceous. This shallow wa- Paria and Northern basins that was infilled by silici-
ter bank was d r o w n e d in both areas by the end of the clastic sediments during the C e n o z o i c (Fig. 3 9 B - E ) .
Early C r e t a c e o u s (Aptian in Trinidad and Albian in The shallow carbonate bank rocks are not well
eastern Venezuela; Babb, 1997). It is the irregular top e x p o s e d in eastern Venezuelan b e c a u s e they have

Fig. 39. Structural evolution of sub-basins in the Gulf of Paria based on sequence and structural mapping described in this study. (A)
Middle Miocene. (B) Early Late Miocene. (C) Late Miocene-Early Pliocene (sequence 3). (D) Early-middle Pliocene (sequence 4); (E)
Late Pliocene-Pleistocene (sequence 5). See text for discussion.
been overthrust in the Serranfa del Interior and are from Late Cretaceous through Paleogene character-
now deeply buried beneath the Maturfn basin (Figs. ized by both carbonate and siliciclastic units that
4B, 40B). Evaporitic rocks are present in the Cre- are generally thin. The northward thinning of these
taceous carbonate section beneath the Gulf of Paria units is shown on the regional cross-section of
basin of Trinidad (Bray and Eva, 1983; Eva et al., Trinidad in Fig. 40A because the overall amount
1989; Babb, 1997) and have been described in small of shortening related to transpressional faulting and
sections from the Araya-Paria Peninsula (Gonzalez uplifted ranges like the Central and Southern Ranges
de Juana et al., 1968). The passive margin section is small. In contrast, the large amount of shorten-
of Trinidad is locally within 1-3 km of the surface ing in the Serranfa del Interior make recognition
and is known through drilling and seismic reflec- of a similar pattern of northward thinning of pas-
tion (Fig. 40A). General rock types and depositional sive margin rocks impossible in eastern Venezuela
environments are compiled onto the stratigraphic (Fig. 40B).
columns shown in Fig. 41. The most obvious difference between the passive
(2) A later deeper-water passive margin phase margin sections of Trinidad and eastern Venezuela

~o
~ o
0
0
.0 =
~ .o
oz~
0 ,~ 0 .~
~ ~ .
> .~ "a
"~, ~ ~
0 0.-~
~o~
0 ~ ~"~
~9o ~ -~~
c~"~ 9=
~'~


Fig. 42. Map showing major structural features of the northern margin of South America in Venezuela and Trinidad and the ages of
convergent deformation along strike of the margin. The development of northwest-striking right-lateral strike-slip faults in the Trinidad
area may be related to oceanic free face or lack of a continental backstop between the eastward moving Caribbean plate and arc terranes
and the South American continent. See text for discussion.
is their degree of deformation (Fig. 40). In eastern Possible explanation for structural differences
Venezuela, the margin is intensely imbricated in between Trinidad and eastern Venezuela
a fold-thrust style with 4 5 - 9 0 km of horizontal One possible difference to explain the lack of
shortening while the strata in the Gulf of Paria are thrusting in Trinidad is the lack of Precambrian
generally horizontal in attitude and deformed mainly continental crust of the Guyana shield which acts
by late Neogene strike-slip faults. as a 'backstop' southeast of the fold-thrust belt
(Fig. 42). Instead, Trinidad appears to occupy the
Oblique collision and foreland basin phase site of a rifted passive margin characterized by the
In eastern Venezuela a foreland basin (Maturfn northwest-trending carbonate highs formed in latest
sub-basin; Fig. 4B) formed as a result of thrust- Jurassic to Early Cretaceous time (Fig. 21). This
ing in the Serranfa del Interior fold-thrust belt in thinner crust acts as less of a barrier or backstop
Late Oligocene through Pliocene time (Parnaud et and provides an unconfined or less confined 'free
al., 1995; Roure et al., 1995; di Croce and Bally, face' for the strike-slip movement of crustal mate-
Chapter 16). Foreland basin sedimentary rocks are rial to the southeast in a direction roughly parallel
mainly marine in the early stages and shallow ma- to the trend of the former, northeast-facing pas-
rine to non-marine in the later stages of thrusting as sive margin (Fig. 42). Transpression between the
the Orinoco River delta prograded eastward (Robert- Caribbean and South America plates was accom-
son and Burke, 1989; Erlich and Barrett, 1992) modated by now extinct thrusting in the confined
(Fig. 41). area to the west but now can be relieved through
In Trinidad the earliest evidence of folding, lim- southeastward propagation of strike-slip faults with-
ited thrusting and shallowing in the Central Range out thrusting over the continent-ocean transition
and Southern basin is Early Miocene in age. During in Trinidad and to the south in the Orinoco delta
the Early to Middle Miocene, the passive margin region and along the Guyana margin (Jankowsky
appears to convert to an active margin characterized and Schlapak, 1983). Northwest-oriented basement
by somewhat shallower depositional environments fracture zones in the Trinidad passive margin (possi-
and an increasing amount of structural control on bly reflected in northwest-trending carbonate banks
deposition (Figs. 39 and 41). as seen on the map of the carbonate platform in
Fig. 41. Chart comparing the Jurassic to Recent stratigraphy of the Northern basin, Gulf of Paria, and Southern basin of Trinidad
and the Eastern Venezuelan basin. Striped pattern on older units designates mainly neritic or deep-water marine units deposited in a
passive margin setting. Uncolored pattern on younger units designates mainly shallower water units deposited after the formation of an
oblique collisional boundary. The darker pattern in the Eastern Venezuelan basin designates the Carapita Formation, whose deep marine
environment is thought to have accompanied the flexural formation of the foreland basin (Parnaud et al., 1995). Note that Oligocene
foreland basin subsidence in the Eastern Venezuelan basin precedes the main pulse of Late Miocene to Recent subsidence in the Northern
basin and Gulf of Paria in Trinidad and is therefore consistent with the eastward motion of the Caribbean plate along the northern passive
margin of South America.

Fig. 21) may also act to reorient strike-slip faults the southeastward-moving thrust sheets (Fig. 4B).
from east-west to more east-southeast or southeast In contrast, basins of Trinidad are much less ex-
striking. Movement of material along these north- tensive and confined as steep-sided basins between
west-striking faults that possibly follow deep-seated high-angle strike-slip or transpressional fault zones
basement fracture zones results in the thick-skinned (Fig. 40).
deformational style of Trinidad that is distinctive The driving force for this southeastward move-
from the thin-skinned deformational style seen in ment of material along faults like the Warm Springs,
eastern Venezuela (Fig. 40). Los Bajos, and E1 Soldado (Fig. 4A) is the long-term
Thicker crust to the east fronting the Foothills eastward translation of the Caribbean arc along the
and Serranfa del Interior fold-thrust belts is a back- passive margin of South America (Fig. 43). This
stop that does not allow southeastward translation oblique collisional process now active in Trinidad
of crust to the southeast (Fig. 42). The subse- and active as recently as the Pleistocene in east-
quent space problem that resulted as the edge of ern Venezuela has progressively affected the passive
the Caribbean plate attempted to move southeast- margins of North and South America with Late
ward created the thin-skinned Foothills and Ser- Paleocene thrust deformation documented in Lake
ranfa del Interior fold-thrust belts in Oligocene to Maracaibo of western Venezuela (Lugo and Mann,
Pleistocene time. As a result, there is an exten- 1995) and Late Paleocene-Early Eocene deforma-
sive foreland basin in eastern Venezuela that formed tion affecting western Cuba (Gordon et al., 1997)
as the continental crust flexed under the load of (Fig. 43A-F).
Fig. 43. Map showing the west to east migration during Late Cretaceous to Recent time of the eastern margin of the Caribbean plate into
the space formed by the Mesozoic opening between North and South America. (A) Campanian-Maestrichtian. (B) Latest Paleocene-
earliest Eocene. (C)Latest Early Eocene. (D) Middle Eocene'Middle Miocene. (E) Middle Miocene. (F) Post-Middle Miocene. The
position of the first interaction of the Caribbean plate in the southeastern Caribbean is taken as Oligocene in the Eastern Venezuelan basin
and Late Miocene in Trinidad. See text for discussion.

Rocks of the Cordillera de la Costa belt west-to-east motion of the Caribbean plate along the
These rocks are present in both the Araya-Paria northern margin of Trinidad as shown in Fig. 42.
Peninsula north of the Serrania del Interior fold- Complications in the age of deformation along
thrust belt and in the Northern Range of Trinidad. the margin may relate to irregularities in the shape
The rocks of the Araya-Paria area exhibit both of the Jurassic-Cretaceous passive margin which we
high- and low-grade metamorphic grades and duc- have shown to have significant vertical relief in the
tile structures including southeast-verging isoclinal subsurface beneath the Gulf of Paria and Northern
folds and thrust faults (Fig. 40B). Av6 Lallemant basins. The age of deformation could be varying in
(1997) interpreted these structures to have formed in a complex way as the arc encounters such a margin
a trench setting far to the west of their present po- with variable topographic relief and structural trends.
sition (Fig. 43A). Rocks of the Northern Range are
all low metamorphic grade and exhibit northward
vergence (Algar and Pindell, 1993) that is in the CONCLUSIONS
opposite direction to that predicted by Speed (1985)
for late Neogene fold-thrust belts (Fig. 40A). As in The main conclusions of this study can be sum-
the case of the Araya-Paria Peninsula, it is likely marized as follows.
that the Northern Range was juxtaposed with the (1) An integrated data base consisting of well
areas of Trinidad's basins by strike-slip faulting and logs and seismic reflection profiles mainly from the
originally formed in a location far to the west of its Gulf of Paria and Northern basins was used to
present position as shown on the plate reconstruction document major tectonic and sedimentary events in
of Fig. 43. the Neogene geologic history of Trinidad.
(2) Previous models proposed to explain the
Tectonic implications of Venezuela-Trinidad tectono-stratigraphic evolution of the area can be
comparison divided into three categories: (a) models in which
oblique collision and thrusting are the dominant pro-
Differences in the age and style of deformation cess; (b) models in which strike-slip faulting and
seen in Trinidad and eastern Venezuela provide help- related normal faulting, folding, and thrusting are
ful constraints on the large-scale tectonic processes the dominant tectonic process; and (c) transpres-
affecting the plate boundary zone. sional models which combine elements of thrusting
and strike-slip. The conflicting tectonic models re-
Eastward movement of Caribbean plate flect varying emphasis on methods, data types, and
First, the transition from passive to active margin local areas of Trinidadian geology.
tectonics, marked by abrupt shallowing in sedimen- (3) For this study, the term 'Gulf of Paria basin'
tary environments and the formation of foreland is used to describe all the sedimentary rocks in the
basins flanking thrust belts (Fig. 4B), is about 17-20 offshore Gulf of Paria that overlie latest Early Creta-
m.y. later in Trinidad than in eastern Venezuela. ceous rocks of the passive margin of northern South
Units on the stratigraphic chart in Fig. 41 illustrat- America. The 'Northern basin' refers to the equiv-
ing the transition between passive and active margin alent sedimentary succession of sedimentary rocks
tectonics include the following: with the onshore area of northern Trinidad ('Caroni
(1) In Trinidad, the main transition is proposed basin' of Robertson and Burke, 1989). Mapping of
to occur during the deposition of the conglom- seismic sequences on seismic reflection lines tied
eratic wedge from the Northern Range (sequence to wells revealed the presence of four fault-con-
3, Late Miocene-Early Pliocene). While there are trolled sub-basins within the Gulf of Paria basin
other anomalous episodes of shallowing (e.g., Mid- and western Northern basin. These are the E1 Pi-
dle Miocene shallowing of the Central Range), the lar, Puerto Grande, and Goodrich sub-basins which
deposition of sequence 3 marks the first radical contain Neogene sedimentary sequences ranging in
departure from passive margin sedimentation on a two-way travel time from 2.0 to 3.3 s.
north-facing slope. (4) Five seismic sequences were recognized in
(2) In eastern Venezuela, the main transition the Gulf of Paria and Northern basins based on the
is proposed to occur during the deposition of recognition of onlap, toplap, downlap and truncation
marine shales in the foreland basin (Oligocene; surfaces, and seismic reflection characteristics and
Fig. 41). Foredeep subsidence continued through the the correlation of seismic reflection characteristics
Miocene and only becomes inactive in the Pleis- with well logs in the Gulf of Paria and Northern
tocene (Fig. 4B). basins. Sequences 1 and 2 consist of a Late Jurassic
The delay between the Late Miocene Trinidad to Middle Cretaceous carbonate megaplatform with
event and the Oligocene Eastern Venezuela event overlying carbonate banks that form the basement
is consistent with the slow (1-2 cm/yr) but steady to sequences 3 through 5 of Neogene age. Sequence

3 is a Late Miocene-Early Pliocene shallow-marine phase one followed by uplift and erosion of the Cen-
to brackish-water conglomerate and sandstone that tral Range and the gradual decrease in the Northern
represents a southward-fining and thinning and east- Range siliciclastic sources during phases two and
ward-thickening siliciclastic wedge derived from the three. Environments generally remained proximal,
Late Miocene uplift and erosion of the Northern shallow and brackish along the E1 Pilar fault zone
Range and activity along the E1 Pilar strike-slip fault on the northern edge of the basin and distal, deeper
close to the mountain front of that range. Sequence and less restricted along the central and southern
3 and the overlying sequences 4 and 5 fill in topo- edges of the basin. Through time the basin becomes
graphic relief created by the growth of the carbonate increasingly brackish and more restricted.
banks of sequences 1 and 2 that was later modified (7) Seismic lines were used to illustrate the vari-
by extensional structures formed in the Gulf of Paria ation in the structural style and geomorphic expres-
pull-apart basin. Sequence 4 is an Early to middle sion of the E1 Pilar fault zone along the northern
Pliocene inner neritic to shallow-marine conglomer- edges of the Gulf of Paria-Northern basin. In the
ate, sandstone, silt, and clay that represents a northwestern Gulf of Paria, the fault exhibits a trans-
ward-fining and thinning siliciclastic wedge derived pressional structure consistent with its local, slightly
from the Late Miocene-Early Pliocene uplift and east-northeast strike. A slight change in strike to
erosion of the Central Range with additional con- a more east-southeast strike in the eastern Gulf of
tributions from the north and west and possibly the Paria is consistent with the more transtensional struc-
Orinoco River delta. Sequence 5 is a Late Pliocene to ture of the fault and its association with the Puerto
Pleistocene marine to brackish-water sand, silt, clay Grande sub-basin. The trace of the E1 Pilar fault zone
and minor conglomerate that represents continued is poorly expressed onland in Trinidad and may re-
siliciclastic deposition in an increasingly restricted flect its waning activity in recent time in an eastward
basin between the Northern and Central Ranges. direction and/or the overprinting cultural effects on
(5) Comparison of structure contour maps of low scarps in this densely populated region.
each boundary between sequences 3-5 with isochron (8) Seismic lines were also used to illustrate the
maps of sequences 3-5 allowed a better visualiza- structural style of oblique-slip faults in the Gulf of
tion of the space available at tectonically signif- Paria. These styles include gravity-related detach-
icant times in the history of the Gulf of Paria- ment faults possibly localized along the unconfor-
Northern basin and the nature of the subsequent sed- mity between Cretaceous carbonate and overlying
imentary deposition onto those structural surfaces siliciclastic rocks and tectonic transtensional faults
and sequence stratigraphic boundaries. This cross- formed in the stepover area between the E1 Pilar
comparison identified three Neogene deformational fault to the north and the Warm Springs fault to the
phases that have affected the area: phase one, Late south.
Miocene-Early Pliocene strike-slip motion along the (9) Seismic lines were used to illustrate the struc-
E1 Pilar fault and north-to-south filling of the Gulf tural styles and geomorphic expression of the Warm
of Paria and Northern basins; phase two, middle Springs-Central Range-Caigual fault zone. In the
to Late Pliocene strike-slip motion along the Warm Gulf of Paria, the Warm Springs fault zone exhibits
Springs-Central Range fault zone and south-to-north a transtensional structure consistent with its slightly
filling of the Gulf of Paria and Northern basins; and east-southeast strike. An abrupt change in strike to a
phase three, Late Pliocene to Pleistocene strike-slip more east-northeast strike near the western shoreline
motion along the Warm Springs-Central Range fault of Trinidad is consistent with the more transpres-
zone and continued south-to-north filling of the Gulf sional structure of the continuation of the Warm
of Paria and Northern basins. Springs fault zone in the Central Range (Central
(6) Paleocurrents and environments of deposi- Range-Caigual fault zones). On the basis of seismic
tion of the Gulf of Paria and Northern basins were lines and existing geologic maps, we reinterpreted
closely controlled by structural events during the the overall structure of the Central Range as a
three deformational phases. Paleocurrents for defor- transpressional uplift bounded by inwardly dipping
mational phase one (deposition of sequence 3) are reverse faults and bisected by a strike-slip fault sys-
oriented orthogonal and parallel to the axis of the tem rather than as the southeast-verging fold-thrust
basins and reflect the strong control of the bounding belt proposed by previous workers.
E1 Pilar fault zone along the northern edge of the (10) Compilation of data by previous workers in
basin. Paleocurrents for deformational phases two the Southern basin of Trinidad allows us to make
(sequence 4) and three (sequence 5) become increas- a comparison between the style and age of defor-
ingly dominated by a west-to-east paleoflow along mation in the Southern basin and that described by
with a southeast-to-northwest flow in the area of the us in the Gulf of Paria-Northern basin. The style
Goodrich basin. Environments of deposition reflect of deformation in the Southern basin is dominantly
the uplift and erosion of the Northern Range during south-vergent and starts earlier than in the Gulf

of Paria-Northern basins because Middle Miocene delta filled the Goodrich basin and continued south-
(~ 11.4 m.y. horizon) units onlap slightly older Mid- eastward propagation of the Los Bajos fault offsets
dle Miocene units (Fig. 38). The unconformity be- fold axes in the Southern basin by 10.5 kin. The E1
tween the Late Miocene-Early Pliocene sequence 3 Pilar fault zone appears to undergo transpressional
and the middle to Late Pliocene sequence 4 in the reactivation in the Northern basin but most offset
Northern-Gulf of Paria basin correlates with a less seems to have shifted to the Warm Springs-Central
prominent unconformity between the Upper Cruse Range-Caigual fault zone in the Central Range.
and Lower Forest Formations of southern Trinidad. (12) We compared the above sequence of events
The deformation in the Southern basin begins earlier and structure of Trinidad with the events and struc-
than most of that observed in the Gulf of Paria- ture of the Eastern Venezuelan basin in order to
Northern basin. infer large-scale tectonic controls on regional defor-
(11) Compilation of regional geologic data and mation. The Eastern Venezuelan basin is a foreland
integration with our results from the Gulf of Paria- basin associated with an adjacent, southeast-verg-
Northern basins allows the following events to be ing fold-thrust belt (Serranfa del Interior) associated
constrained in the Middle Miocene through Recent with 45-90 km of shortening. Trinidad exhibits a
evolution of the area. Middle Miocene: high-angle slightly younger record of fault-related sedimenta-
faulting affects the Southern basin; we do not re- tion, lacks evidence for significant shortening of the
gard this event as a regional fold and thrusting Cretaceous-Pleistocene sequences 1-5 documented
event because we do not observe widespread fold- in this paper, and exhibits transpressional-type struc-
thrust structures in the pre-Late Miocene rocks of tures with no uniform sense of vergence. We pro-
the Gulf of Paria-Northern basin. Instead, we con- pose that Precambrian crust adjacent to the Eastern
clude that most deformation observed in Trinidad is Venezuelan basin acted as an effective backstop and
transpressional or transtensional in style, related to did not allow continued southeastward migration
movements on the E1 Pilar, Warm Springs-Central of obliquely colliding Caribbean crust. In contrast,
Range-Caigual, and Los Bajos fault zones, and is Trinidad appears to occupy the site of a rifted passive
Late Miocene to recent in age. Late Middle Miocene: margin whose thinner crust acts as less of a backstop
the uplift of the Central Range at this time may be than the Precambrian crust to the southwest. This
related to the Middle Miocene thrusting observed less confined 'free face' allowed strike-slip move-
in the Southern basin. Early Late Miocene: during ment of Caribbean and passive margin crust to the
this period a lobe of conglomeratic sediment is shed southeast in a direction roughly parallel to the trend
from the Northern Range into the Gulf of Paria- of the former passive margin.
Northern basins and is interpreted to reflect the first
significant lateral movement along the E1 Pilaf fault
zone. We suggest that the fault propagated from west ACKNOWLEDGEMENTS
to east during this time as the accretionary wedge
and forearc of the Lesser Antilles arc moved past the This paper is a partial summary of a Ph.D.
South American passive margin. The Los Bajos fault dissertation conducted by the first author from
is thought to have propagated south by this time. 1992 through 1997 at the University of Texas at
Termination of this fault on the Southern Range may Austin and supported by the Southern Basin Consor-
have led to a widening of the Neogene belt of trans- tium, Petroleum Company of Trinidad and Tobago
pressional thrusting. Late Miocene-Early Pliocene: (Petrotrin), and the Institute for Geophysics of the
during this time, there was a rapid spread of strike- University of Texas at Austin. We thank Petrotrin
slip and oblique-slip faulting to the south of the E1 for release of these data for the dissertation study.
Pilar fault zone along the Warm Springs fault zone Special thanks to Ph.D. committee members R. Buf-
and within the Goodrich sub-basin. This southward tier (co-chair), E. McBride (co-chair), W. Galloway,
shift is interpreted as the development of a pull-apart and N. Tyler for their suggestions on improving this
basin at the stepover area between the E1 Pilar and study. UTIG contribution number 1425.
Warm Springs fault zones. Formation of the stepover
may reduce slip along the eastern continuation of
the E1 Pilar fault zone at the northern margin of REFERENCES
the Northern basin. Early to middle Pliocene: this
interval saw the continued subsidence and fault pat- Algar, S.T. and Pindell, J., 1993. Structure and deformation
tern of the previous interval in the stepover area history of the Northern Range of Trinidad and adjacent areas.
between the two faults. Late Pliocene to Pleistocene: Tectonics, 12: 814-829.
Av6 Lallemant, H., 1997. Transpression, displacement partition-
a major wedge of sediment derived from erosion of ing, and exhumation in the Caribbean/South American plate
the Central Range with probable contributions from boundary zone. Tectonics, 16: 272-289.
the north and west and possibly from the Orinoco Babb, S., 1997. Tectonics and Sedimentation of the Gulf of Paria

556 S. BABB and R M A N N
and Northern Basin, Trinidad. Unpublished Ph.D. dissertation, Caribbean Geol. Conf., Port of Spain, pp. 25-29.
University of Texas at Austin, 200 pp. Gonzalez de Juana, C., Iturralde de Arozena, J.M. and Picard,
Barr, K.W., 1952. Limestone blocks in the Lower Cretaceous X., 1980. Geologia de Venezuela y de sus cuencas petroliferas.
Cuche Formation of the Central Range, Trinidad, B.W.I. Geol. Ediciones Foninves, Caracas, 1031 pp.
Mag., 89:417-425. Gordon, M.B., Mann, P., C~iceres, D. and Flores, R., 1997.
Barr, K.W. and Saunders, J.B., 1968. An outline of the geology Cenozoic tectonic history of the North America-Caribbean
of Trinidad. Trans. Fourth Caribbean Geol. Conf., Port of plate boundary zone in Cuba. J. Geophys. Res., 102: 10,055-
Spain, pp. 1-10. 10,082.
Bartenstein, H., Bettenstaedt, F. and Bolli, H.M., 1957. Die Haq, B.U., Hardenbol, J. and Vail, P.R., 1987. The chronology
foraminiferan der Unterkreide von Trinidad, B.W.I., Erster of fluctuating sea level since the Triassic. Science, 235:1156-
Teil: Cuche- und Toco Formation. Eclogae Geol. Helv., 50: 1167.
5-67. Higgins, G.E. and Saunders, J.B., 1968. Excursion number 4,
Bartenstein, H., Bettenstaedt, F. and Bolli, H.M., 1966. Die Western Central Range. Trans. Fourth Caribbean Geol. Conf.,
foraminiferen der Unterkreide von Trinidad, B.W.I., Zweiter Port of Spain, pp. 435-437.
Teil: Maridale Formation (Typlokalitat). Eclogae Geol. Helv., Jankowsky, W.J. and Schlapak, G., 1983. Guyana offshore.
59: 129-178. In: A. Bally (Editor), Atlas of Seismic Stratigraphy. Vol. 2.
Bray, R. and Eva, A., 1983. Age, depositional history, and American Association of Petroleum Geologists, Tulas, OK,
tectonic significance of the Couva Marine evaporite, offshore pp. 2.2.3-47.
Trinidad. Tenth Caribbean Geol. Conf., Cartagena, Colombia Koutsoukos, E.A.M. and Merrick, K.A., 1985. Foraminiferal
(Abstr. Vol.), p. 372. paleoenvironments from the Barremian to Maestrichtian of
Burke, K., 1988. Tectonic evolution of the Caribbean. Annu. Rev. Trinidad, W.I. Trans. First Geol. Conf. Geol. Soc. Trinidad and
Earth Planet. Sci., 16:201-230. Tobago, Port of Spain, pp. 85-101.
Carr-Brown, B. and Frampton, J., 1979. An outline of the stratig- Kugler, H.G., 1953. Jurassic to recent sedimentary environments
raphy of Trinidad. In: Field Guide to the Fourth Latin Ameri- in Trinidad. Bull. Assoc. Suisse Geol. Ing. Pet., 20: 27-60.
can Geological Congress, Port of Spain. Key Caribbean Publi- Kugler, H.G., 1959. Geological map and sections of Trinidad
cations, Trinidad, pp. 7-19. (scale 1:100,000). The Petroleum Association of Trinidad,
DeMets, C., Gordon, R.G., Argus, D.E and Stein, S., 1994. Port of Spain.
Effects of recent revisions to the geomagnetic time scale on Leonard, R., 1983. Geology and hydrocarbon accumulations,
estimates of current plate motions. Geophys. Res. Lett., 21: Columbus basin, offshore Trinidad. Bull. Am. Assoc. Pet.
2191-2194. Geol., 67: 1081-1093.
Dixon, T.H., Farina, E, DeMets, C., Jansma, R, Mann, E and Lugo, J. and Mann, P., 1995. Jurassic-Eocene tectonic evolution
Calais, E., 1998. Relative motion between the Caribbean and of Maracaibo basin, Venezuela. In: A.J. Tankard, R. Su~irez
North America plates and related boundary zone deformation and H.J. Welsink (Editors), Petroleum Basins of South Amer-
based on a decade of GPS observations. J. Geophys. Res., 103: ica. American Association of Petroleum Geologists, pp. 699-
15,157-15,182. 725.
Donovan, S.K., 1994. Trinidad. In: S.K. Donovan and T.A. Jack- Mann, P., Schubert, C. and Burke, K., 1990. Review of
son (Editors), Caribbean Geology: An Introduction. University Caribbean neotectonics. In: G. Dengo and J.E. Case (Edi-
of West Indies, Kingston, pp. 209-228. tors), The Caribbean Region. The Geology of North America,
Dyer, B.L. and Cosgrove, R, 1992. Penal/Barrackpore field Vol. H, Geological Society of America, Boulder, Colo., pp.
West Indies, South Trinidad basin, Trinidad. In: E.A. Beau- 307-338.
mont and N.H. Foster (Editors), Structural Traps VII. Amer- Michelson, J.E., 1976. Miocene deltaic oil habitat, Trinidad. Am.
ican Association of Petroleum Geologists, Tulsa, Okla., pp. Assoc. Pet. Geol. Bull., 60:1502-1519.
139-157. Molnar, P. and Sykes, L.R., 1969. Tectonics of the Caribbean and
Erlich, R.N. and Barrett, S.E, 1992. Petroleum geology of the Middle America regions from focal mechanisms and seismic-
Eastern Venezuelan foreland basin. In: R.W. Macqueen and ity. Geol. Soc. Am. Bull., 80: 1639-1684.
D.A. Leckie (Editors), Foreland Basins and Foldbelts. Amer- Molnar, P. and Sykes, L.R., 1971. Plate tectonics in the Hispan-
ican Association of Petroleum Geologists, Tulsa, Okla., pp. iola area: discussion. Geol. Soc. Am. Bull., 80:1123-1126.
341-362. Munro, S.E. and Smith, ED., Jr., 1984. The Utica fault zone,
Erlich, R.N., Farfan, EE and Hallock, E, 1993. Biostratigraphy, northeastern Venezuela. In: W.E. Bonini, R.B. Hargraves and
depositional environments, and diagenesis of the Tamana For- R. Shagam (Editors), The Caribbean-South American Plate
mation, Trinidad: A tectonic marker horizon. Sedimentology, Boundary and Regional Tectonics. Geol. Soc. Am. Mem., 162:
40: 743-768. 213-215.
Eva, A.N., Burke, K., Mann, R and Wadge, G., 1989. Four Parnaud, E, Gou, Y., Pascual, J.C., Truskowski, I., Gallango, O.
phase tectonostratigraphic development of the southeastern and Passalacqua, H., 1995. Petroleum geology of the central
Caribbean. Mar. Pet. Geol., 6: 9-21. part of the Eastern Venezuelan basin. In: A.J. Tankard, R.
Figueroa de Sanchez, L. and Hern~indez, L., 1990. Explo- Su~irez and H.J. Welsink (Editors), Petroleum Basins of South
raci6n geoffsica-geol6gica del ~irea de Guarumen. Trans. 5th America. Am. Assoc. Pet. Geol. Mem., 62: 741-756.
Venezuelan Geophys. Congr., pp. 219-227. Passalacqua, H., Fernandez, E, Gou, Y. and Roure, E, 1995.
Frey, M., Saunders, J. and Schwander, H., 1988. The mineral- Crustal architecture and strain partitioning in the Eastern
ogy and metamorphic geology of low-grade metasediments, Venezuelan Ranges. In: A.J. Tankard, R. Sugtrez and H.J.
Northern Range, Trinidad. J. Geol. Soc., London, 145: 563- Welsink (Editors), Petroleum Basins of South America. Am.
575. Assoc. Pet. Geol. Mem., 62: 667-679.
Furrer, M.A., 1968. Paleontology of some limestones and cal- Payne, N., 1991. An evaluation of post-Middle Miocene geolog-
cphyllites of the Northern Range of Trinidad, West Indies. ical sequences, offshore Trinidad. Trans. Second Geol. Conf.
Trans. Fourth Caribbean Geol. Conf., Port of Spain, pp. 21- Geol. Soc. Trinidad and Tobago, Port of Spain, pp. 70-87.
24. Perez, O.J. and Aggarwal, Y.R, 1981. Present-day tectonics of
Gonzalez de Juana, C., Munoz, N.G. and Vignali, M., 1968. On the southeastern Caribbean and northeastern Venezuela. J.
the geology of the eastern Paria (Venezuela). Trans. Fourth Geophys. Res., 86:10791-10804.

Persad, K.M., 1984. Tectonic map and geologic-tectonic sections Trans. Fourth Caribbean Geol. Conf., Port of Spain, pp. 427-
through Trinidad and Tobago (scale 1: 200,000). Robertson 429.
Research (U.S.), Houston, Texas. Saunders, J.B., 1972. Recent paleontological results from the
Persad, K.M., 1985. Outline of the geology of Trinidad. Fourth Northern Range of Trinidad. Trans. Sixth Caribbean Geol.
Latin Am. Geol. Congr., Port of Spain, pp. 738-758. Conf., Margarita Island, pp. 455-460.
Pindell, J.L. and Barrett, S.E, 1990. Geological evolution of the Schubert, C., 1984. Basin formation along the Bocono-Mor6n-
Caribbean region: a plate-tectonic perspective. In: G. Dengo E1 Pilar system, Venezuela. J. Geophys. Res., 89:5711-5718.
and J.E. Case (Editors), The Caribbean Region. The Geology Snoke, A.W., Yule, J.D., Rowe, D.W., Wadge, G. and Sharp,
of North America, Vol. H, Geological Society of America, W.D., 1990. Stratigraphic and structural relationships on To-
Boulder, Colo., pp. 405-431. bago, West Indies, and some tectonic implications. In: D.K.
Potter, H., 1976. Type sections of the Maraval, Maracas, and Larue and G. Draper (Editors), Transactions of the 12th
Chancellor Formations in the Caribbean Group of the Northern Caribbean Geological Conference, St. Croix, U.S. Virgin Is-
Range of Trinidad. Trans. Seventh Caribbean Geol. Conf., lands. Miami Geological Society, Coral Gables, Fla., pp. 389-
Guadeloupe, pp. 505-527. 402.
Prieto, R., 1987. Seismic Stratigraphy and Depositional Systems Speed, R.C., 1985. Cenozoic collision of the Lesser Antilles arc
of the Orinoco Platform Area, Northeastern Venezuela. Un- and continental South America and the origin of the E1 Pilar
published Ph.D. Dissertation, University of Texas at Austin, fault. Tectonics, 4:41-69.
144 pp. Trechmann, C.T., 1935. Fossils from the Northern Range of
Radovsky, B. and Iqbal, J., 1979. Geology of the North Soldado Trinidad. Geol. Mag., 72: 166-175.
field. Fourth Latin Am. Geol. Congr., Port of Spain, pp. 759- Tyson, L., 1989. Structural features associated with the Los
769. Bajos fault and their interpretation in light of current theories
Robertson, E and Burke, K., 1989. Evolution of southern of strike-slip tectonics. Trans. 12th Caribbean Geol. Conf., St.
Caribbean plate boundary, vicinity of Trinidad and Tobago. Croix, U.S. Virgin Islands, pp. 403-414.
Am. Assoc. Pet. Geol. Bull., 73: 490-509. Tyson, L. and Ali, W., 1990. Cretaceous to Middle Miocene
Rohr, G.M., 1991. Paleogeographic maps, Maturfn basin of East- sediments in Trinidad. Trans. Second Geol. Conf. Geol. Soc.
ern Venezuela and Trinidad. Trans. Second Geol. Conf. Geol. Trinidad and Tobago, Port of Spain, pp. 266-277.
Soc. Trinidad and Tobago, Port of Spain, pp. 88-105. Tyson, L., Babb, S. and Dyer, B., 1991. Middle Miocene tecton-
Rosencrantz, E., 1990. Structure and tectonics of the Yucatan ics and its effects on Late Miocene sedimentation in Trinidad.
basin, Caribbean Sea, as determined from seismic reflection Trans. Second Geol. Conf. Geol. Soc. Trinidad and Tobago,
studies. Tectonics, 9:1037-1059. Port of Spain, pp. 26-40.
Roure, F., Carnevali, J.O., Gou, Y. and Subieta, T., 1995. Ge- Vierbuchen, R.C., 1984. The geology of the E1 Pilar fault zone
ometry and kinematics of the North Monagas thrust belt and adjacent areas in northeastern Venezuela. In: W.E. Bonini,
(Venezuela). Mar. Pet. Geol., 11: 347-362. R.B. Hargraves and R. Shagam (Editors), The Caribbean-
Russo, R.M. and Speed, R.C., 1992. Oblique collision and tec- South American Plate Boundary and Regional Tectonics. Geol.
tonic wedging of the South American continent and Caribbean Soc. Am. Mere., 162: 189-212.
terranes. Geology, 20: 447-450. Wadge, G. and Macdonald, R., 1985. Cretaceous tholeiites of the
Russo, R.M., Speed, R.C., Okal, E.A., Shepherd, J. and Row- northern continental margin of South America: the Sans Souci
ley, K.C., 1993. Seismicity and tectonics of the southeastern Formation of Trinidad. J. Geol. Soc. London, 142: 297-308.
Caribbean. J. Geophys. Res., 98: 14299-14319. Wadge, G. and Shepherd, J.B., 1984. Segmentation of the Lesser
Salvador, A. and Stainforth, R.M., 1968. Clues in Venezuela Antilles subduction zone. Earth and Planetary Science Letters,
to the geology of Trinidad, and vice versa. Trans. Fourth 71: 297-304.
Caribbean Geol. Conf., Port of Spain, Trinidad, pp. 31-40. Wilson, C.C., 1940. The Los Bajos fault of south Trinidad,
Saunders, J.B., 1968. Excursion Number 1: Manzanilla coast. B.W.I. Am. Assoc. Pet. Geol. Bull., 24: 2102-2125.

Chapter 19
New Insights on the Formation of the Caribbean Basalt Province

Revealed by Multichannel Seismic Images of Volcanic Structures
in the Venezuelan Basin
J O H N D I E B O L D , N E A L D R I S C O L L and EW-9501 S C I E N C E T E A M 1
A regional multichannel seismic survey (EW-9501) was carried out in the Venezuelan Basin aboard R/V Ewing. The resulting
reflection profiles image previously unseen structures within the entire thickness of the Caribbean oceanic plateau basalts east of
Beata Ridge. Identifiable volcanic products include large amounts of extrusive material, which form two morphologically distinct,
vertically stacked sequences. The lower sequence consists of local highs and ridges, flanked by wedges made up of dipping
flows. The upper sequence is more homogeneous, featuring widespread flows which fill morphological and extensional lows in
the lower sequence. Dipping sequences of volcanic flows are seen with length scales varying from 20 km to over 100 km. In
contrast to seaward-dipping reflectors described elsewhere, these flows appear to be of submarine origin, and to have maintained
their primary sense of dip. The morphology and scales of the flows are controlled by the volume of source material. The top of
the upper 'high-volume' sequence forms the smooth B" horizon sampled by DSDP and ODP drilling. A more detailed survey was
made of the southeastern edge of the plateau which includes a 100-kin-wide sequence of flows forming gently dipping wedges
overlying thinned oceanic crust. The resulting images show clearly that the present edge is constructional, and that buried beneath
it is an earlier, possibly rifted edge. This two-phase development is characteristic of the plateau everywhere in the Venezuelan
Basin, and interpretations of the Caribbean's history must take this into account. It is not clear how much time elapsed between
the end of the first, and the onset of the later phases, but our results indicate that the later phase is the only one so far sampled by
Caribbean drilling.
INTRODUCTION the Venezuelan Basin seem to form linear N E -

S W trends, which correspond with those of minor
Two-ship refraction profiles carried out in the faults and grabens i m a g e d in reflection profiles and
1950s yielded velocity functions of a Caribbean crust m a p p e d by Case and H o l c o m b e (1980). A n u m b e r
that was anomalously thick (up to 20 km), with an of efforts (Christofferson, 1973; Ghosh et al., 1984)
unusual two-layer structure. Based on early analog have been m a d e to fit the magnetic 'lineations' in
single-channel seismic reflection data, two distinc- the Venezuelan and C o l o m b i a n Basins to known
tive horizons were designated A" and B", according seafloor spreading sequences, but the results have
to their resemblance to A and B in the Atlantic, been unpersuasive.
and A', B' in the Pacific. Sediment cores, carefully Drilling, during D S D P Leg 15, sampled B" for
located on the basis of those single-channel profiles, the first time, and found it to be the top of a Cre-
showed A" to correlate with an Eocene chert layer, taceous volcanic sequence. The igneous rocks of the
a result later confirmed by D S D P drilling, which Caribbean basalt province, as sampled by drilling at
indicated that layers of limestone also contributed five sites during D S D P Leg 15, Site 1001 during
to the reflective character of A". The makeup of O D P Leg 165, and in on-land occurrences, are virtu-
B" (stands for acoustic ' b a s e m e n t ' ) remained un- ally all basalt. Their character has been s u m m a r i z e d
known for a while longer. Magnetic anomalies in by Donnelly et al. (1990; Fig. 1). The bulk of the
basaltic material occurs in three petrochemical vari-
1Lewis Abrams, Peter Buhl, Thomas Donnelly, Edward Laine, eties: transitional tholeiitic basalt, highly magnesian
Sylvie Leroy, Adrienne Toy basalt, and mildly alkalic basalts. The transitional

562 J. DIEBOLD et al.
Fig. 1. EW-9501 track, with line numbers and sonobuoylocations on bathymetry; 500 m contours. DSDP and ODP sites where volcanic
basement was sampled by drilling are indicated.
tholeiitic basalts are the most widely distributed and also occur because it is difficult to assess the dura-
the most voluminous of the basalt types sampled in tion of the hiatus between basalt emplacement and
the Caribbean region. The highly magnesian basalts pelagic accumulation. Even worse, if the basalts are
have been termed 'komatiite' on Isla Gorgona by sills then the biostratigraphic age only yields a max-
Echeverria (1980) and in Colombia by Spadea et al. imum age of emplacement. Despite the numerous
(1989), 'picrite' on Curaqao by Beets et al. (1982) dating problems, there are several firm ages for the
and Klaver (1987), and 'low-Ti-basalts' in the Santa termination of magmatic activity, indicated by the
Elena Peninsula of Costa Rica by Wildberg (1984). locally youngest ages of the Cretaceous basalts. As
The mildly alkalic basalts are best known from the reviewed in Donnelly et al. (1990) these dates are
Beata Ridge (DSDP Site 151) and from the Du- mainly derived from dating sedimentary deposition
misseau Formation of southern Haiti. They are also on the basalt complex, including DSDP Leg 15 sites.
represented in the younger (latest Cretaceous or Pa- These ages cluster tightly at the latest Turonian, but
leogene) basaltic centers from Costa Rica (Wildberg, in a few cases could be as young as Santonian, and
1983; Frisch et al., 1992). These basalts are enriched even early Campanian (DSDP Site 152 and ODP
in K, Th, and several minor elements. They resem- Leg 165). Recently, a detailed radiometric study of
ble well-known, mildly alkalic basalts of Hawaii, basalts from Gorgona, Costa Rica, Haiti, and Cu-
and are characteristic of ocean island basalts. Where raqao has been undertaken by Sinton et al. (1998)
there is some local basis for stratigraphic position- using 4~ incremental heating methods, which
ing, they seem to occur among the later eruptive yield dates consistent with the sedimentary ages for
products. the termination of the event. The number and ge-
One of the most unsatisfactory aspects of the ographic extent of consistent age determinations at
previous studies of the Cretaceous flood basalts has 88-90 Ma suggest that a truly vast igneous event
been establishing ages of magmatic activity. Conven- occurred at this time. Nevertheless, there remain
tional K-Ar age determinations that are very numer- several circumstantial fossil occurrences of Albian
ous in the literature (Donnelly et al., 1990) can be and Aptian age associated with the Cretaceous basalt
shown in several instances to contradict clear strati- province. For example, intercalated within the Cu-
graphic constraints. In general, the ages ascertained raqao basalt complex is a sedimentary succession
from K-Ar are too young because of Ar loss or K with six genera of ammonites that indicate a late-
uptake during seafloor weathering or during burial middle Albian age (Weidmann, 1978), which is
metamorphism. Problems with biostratigraphic ages significantly older than the biostratigraphic and ra-

MULTICHANNEL SEISMIC IMAGES OF VOLCANIC STRUCTURES IN THE VENEZUELAN BASIN 563
diometric ages obtained elsewhere. There are two trate by seismic reflection. This is principally due
interpretations: (1) the ammonite ages are incorrect, to the high reflection coefficients of the sediment-
or (2) there was an earlier age of magmatic activity basement interface there, though in early work, large
of unknown extent, and the 88-90 Ma event was a water depths and weak profiling systems played a
terminal event for the main Caribbean Cretaceous part. Reflections beneath the smooth B" horizon
basalt province, but not the initial event. were detected from time to time in early MCS data
Early multichannel seismic (MCS) data acquired (Hopkins, 1973; Ladd and Watkins, 1980; Stoffa et
by IFP (France), UTIG (Texas) and L-DEO (Lam- al., 1981; Diebold et al., 1981). In some cases, these
ont) occasionally imaged internal layering beneath reflectors were observed to dip in relation to the B"
B", and showed that to the southeast, the volcanic surface, though their identification was often open
sequence had a distinct edge, beyond which lay to doubt. The strong basement reflection coefficient
a triangular area of deep, rough acoustic basement gives rise to strong intrabed ('peg-leg') multiples
(called 'rough B"') whose reflective character resem- that reverberate efficiently within the sedimentary
bled oceanic crust. This area was found, however, section. The fact that water depths were quite large,
to be practically devoid of magnetic signature (Don- compared to the length of receiving arrays used in
nelly, 1973), suggesting that the crust was either the early work, made it difficult to discriminate be-
created by seafloor spreading during a magnetically tween those multiples and primary arrivals. Intrabed
quiet period, or that the original magnetic signature multiples are still a problem, even with the 4 km
was degraded by thermal or tectonic modification. receiving array of R/V Ewing.
Early MCS surveys, carried out during the 1970s,
defined the approximate area of rough basement.
The distinction between the rough and smooth va- EW-9501
rieties of B" was first described by Talwani et al.
(1977) and a series of normal faults was shown to In February and March, 1995, 5200 km of
coincide with the rough-smooth boundary by Biju- MCS data, shot with a 20-airgun (8415 cubic inch)
Duval et al. (1978). The acquisition of additional source array, and a 4-km, 160-channel hydrophone
crossings of the boundary showed that it had a va- streamer, were acquired aboard R/V Ewing. Under-
riety of manifestations, ranging from imperceptibly way geophysics included measurements of gravity,
gradual to abrupt (Diebold et al., 1981). Lamont data magnetics, and swath bathymetry. 104 successful
(described by Diebold et al., 1981) provided five deployments of expendable sonobuoys were made.
crossings of the rough-smooth B" boundary, which A track map, identifying the seismic reflection lines
were not enough to characterize the exact nature of by number, with sonobuoy deployment locations, is
the transition. Additional crossings by French (IFP) shown as Fig. 1. The purpose of this NSF-funded
and US (UTIG) investigators all converged at the cruise was to better define the boundary between
boundary in the same area as Lamont line 108. smooth and rough B", to image structures within
The resulting coverage showed that the basement and beneath the Caribbean volcanic complex, and
faults controlling the rough-smooth boundary had to provide new data on the structure and develop-
strikes oblique to the general trend of the transition, ment of the Beata Ridge. The new data meet these
suggesting that the basement fabric pre-dated the goals, and also reveal several unexpected features
volcanic emplacement. MCS line 108 was unique in of the sedimentary column which provide important
that it showed clearly how depth to Moho increases, new constraints on the tectonic and oceanographic
in a step-wise fashion, northward across the rough- development of the Caribbean region.
smooth transition, leading Diebold et al. (1981) to The motivation behind the cruise was to obtain
interpret smooth B" as the top of Cretaceous flows deep reflection seismic data in the Venezuelan Basin
which had overrun and depressed rough B" crust. and on the Beata Ridge crest and its eastern flank.
Sonobuoy refraction lines shot in 1974 in con- In the Venezuelan Basin, the survey track was laid
junction with MCS data (Talwani et al., 1977) fol- out to complement MCS data previously acquired by
lowed by two-ship multichannel seismic expanding R/V CONRAD in the 1970s (Talwani et al., 1977;
spread profiles (ESPs) carried out in 1976 showed Stoffa et al., 1981; Diebold et al., 1981) specifically
that the rough basement horizon B" was the top where these lines crossed the rough-smooth B"
of what appeared to be anomalously thin oceanic boundary. The general NNW-SSE orientation of the
crust, and that the crust underlying the only ESP lines, and their northwestward extent, reflected the
located over smooth B" was, as expected, unusually desire to image deep structure (if any) related to
thick, and comprising (in a gross sense) two layers previously mapped basement topography (Case and
(Diebold et al., 1981). Holcombe, 1980) and magnetic lineations (Donnelly,
Though it was the first to be discovered, the area 1973) which trend ESE-WSW. In retrospect, the
of smooth B" has been the most difficult to pene- Venezuelan Basin lines should have extended farther

to the north-northwest, and more cross-lines should applied. The reflection data plots presented here
have been shot, but to do this extra work would have have been horizontally compressed so that they can
required an impossibly long cruise. be shown in their entirety. This has resulted in a
104 Navy-supplied expendable sonobuoys were vertical exaggeration of about 20.8 in the water col-
deployed during the survey. Data from most of these umn. A special dip-adaptive trace mixing process
were recorded to maximum offsets of between 30 was employed, to enhance the appearance of coher-
and 40 km, providing refraction and wide-angle ent reflectors, so that when only every fifth trace is
reflection data for crustal velocity analysis. One- plotted, the result is to retain most of the structural
dimensional velocity analysis, using interactive ray information in the data. All of the reflection profiles
trace modeling has been attempted for all of the are plotted with north and west to the left, south and
sonobuoys. Velocity functions were obtained for 85 east to the right. CDP numbers are in chronological
buoys. Results from this analysis, combined with sequence in each line. CDP spacing is 12.5 m, so
earlier wide-angle reflection and refraction work, 800 CDPs equals 10 km horizontal distance.
provide several important constraints on the struc-
ture and mode of emplacement of the Caribbean LIP
(Large Igneous Province). CRUSTAL ELEMENTS OF THE COLOMBIAN AND
The 4-km-long hydrophone array used in acquir- VENEZUELAN BASINS
ing the reflection profiles, along with the powerful
seismic source array of R/V Ewing, has produced Thin crust
the best MCS data ever shot in the deep-water basins
of the Caribbean. The streamer length, in partic- Despite 20 years accumulation of evidence to
ular, allows high resolution of hyperbolic stacking the contrary, the concept that the Colombian and
velocity, which, if accurately determined, results in Venezuelan Basins are uniformly capped by a Cre-
amazingly detailed images of what now can be seen taceous igneous body persists. In fact, the crustal
as intricate and convoluted sub-basement structures structure of these basins is much more complex, and
in the Venezuelan Basin. The other side of this coin is made up of at least four categories of lithologi-
is that stacking velocity analysis is a painstaking and cal elements: (1) thin ocean-like crust; (2) volcanic
time-consuming process. The high velocity contrasts extrusive sequences with dipping reflectors; (3) mas-
at the sediment-basement interface, and the large sive bodies of volcanic extrusives and intrusives; and
range of source-receiver offsets present in the data (4) extrusive volcanic mounds.
(and required, to ameliorate the effects of intrabed While it might be argued that the last three
multiples), cause unusually extreme lateral changes categories listed above are simply different mani-
in the stacking velocities, despite the large water festations of the same crust-thickening Cretaceous
depths of the Caribbean, which normally could be volcanic event, the Caribbean Sea contains a signif-
expected to minimize these changes. As a result, icant amount of crust whose thickness varies from
most of the reflection data we present here are in the normal (6-8 km) to abnormally thin (3-5 km).
form of 'brute' stacks. In processing, normal move- Although analysis of many of the early two-ship
out corrections were made based on a velocity model refraction profiles indicated the presence of anoma-
created by extrapolation from sonobuoy results, and lously thick crust, several, particularly west of Beata
the stacking velocity analyses manually carried out Ridge, detected crust of normal oceanic thickness
for line 1293, the only line whose stacking velocities (Ewing et al., 1960; Edgar et al., 1971). Ludwig et
have so far been completely determined. Therefore, al. (1975) described the basement reflector in the
the foldout section for line 1293 (Fig. 2) is very Colombian Basin as 'rough' in comparison to that
important in evaluating the discussions presented in the Venezuelan Basin. In 1977, Houtz and Lud-
below. While the other, brute-stack sections show wig mapped a number of areas in the Colombian
features never before seen, the 1293 profile gives a Basin where crust was deep, rough and thin. Bow-
feeling for the level of fine detail that will eventually land and Rosencrantz (1988) re-mapped a rough-
be imaged by these data. smooth boundary in one of the areas outlined by
Excepting velocity analysis, all of the profiles Houtz and Ludwig (1977) and mapped several oth-
were processed identically. The field data were fil- ers. To the east, Talwani et al. (1977) provided MCS
tered and decimated to 4-ms sample size. Water profiles showing rough basement in the Venezuelan
depth-dependent outer and inner mutes were ap- Basin. Simultaneously recorded sonobuoys showed
plied (the effect of outer muting is to make imaging that crust in those areas was thin, compared to
of sedimentary structures less sensitive to errors in normal oceanic crust. Biju-Duval et al. (1978) also
stacking velocities; the inner mute reduces the ef- observed rough, deep basement in the southeastern
fects of peg-leg and water column multiples). After comer of the Venezuelan Basin and suggested that it
stacking, the data were filtered again and AGC was corresponded to oceanic crust, upon which was su-

perimposed the volcanics to the northwest. Diebold have been imaged (though not always recognized
et al. (1981) interpreted additional two-ship veloc- for what they are) in the Colombian Basin (Lu and
ity profiles and obtained velocity-depth functions McMillen, 1982; Kolla et al., 1984; Bowland and
showing that the rough B" crust resembled unusually Rosencrantz, 1988; Bowland, 1993). The presence
thin oceanic crust. They also extended mapping of of these sediments appears to be diagnostic of deep,
the rough-smooth B" transition around the Venezue- rough, thin crust, and identifying them in older data
lan Basin and imaged Moho in a few places beneath can further extend the mapped areas of thin crust.
both rough and smooth crustal types. A compendium Several EW-9501 MCS profiles show that in a
map of crustal thickness, throughout the Caribbean, few places the anomalously thin rough B" crust of
was presented by Leroy (1995). the SE Venezuelan Basin produces low-angle, up-
The Moho reflection can be seen beneath rough ward-concave reflectors which appear to be the sur-
B" on all of the EW-9501 profiles, and on a few of faces of listric faults extending from the seafloor to
the older RC- 1904 and RC-2103 lines (Diebold et the Moho; see for example line 1293 (Fig. 2) CDPs
al., 1981). Interval two-way time of the rough crust 24,000-25,000. As a rule, these do not correlate with
is typically 1.2 s, with a corresponding sonobuoy- crustal offsets, even where the reflections are seen
derived thickness of 3800 m. In a few places, crustal to intersect the top of crust. Similar reflections have
thickness varies, thinning to as little as 0.9 s, and been imaged in several areas, including marginal
thickening to as much as 1.6 s. Velocity seems to NW Atlantic crust of Jurassic age (McCarthy et al.,
vary also, though no systematic changes have yet 1988; Mutter and Karson, 1992; Rosendahl et al.,
been detected. Our results agree with those of Tal- 1992; Zehnder Mutter, 1992) and are, apparently, the
wani et al. (1977) and Diebold et al. (1981). Upper products of some aspect of seafloor spreading. A
crustal velocities are commonly around 5.5 kin/s, recent study by Kent et al. (1997) has demonstrated,
though occasionally, a thin upper layer is detected however, that the same pattern of reflections may
with velocities from 4.2 to 4.9 km/s. Mid-crustal ve- arise due to sideswipe from a sub-parallel basement
locities of 6.6 km/s are typical, and often velocities ridge, and it is certainly possible that some of the
around 7.1-7.2 km are seen below these. 'Normal' 'faults' we see are artifacts of this kind. Although
Moho velocities are never seen, but instead, criti- velocity analysis (not yet complete) indicates that
cally refracted energy with velocities between 7.4 these are intra-crustal events, Kent et al. (1997) have
and 7.8 km/s appears, corresponding to rays turn- shown that streamer feathering can be the cause of
ing at depths below reflection Moho. These crustal high stacking velocities for sideswipe events, which
thicknesses are well below the normal range, as would otherwise be better imaged with lower veloci-
defined by White et al. (1992), who also demon- ties.
strated a historical bias towards the underestimation The question remains: was Venezuelan Basin
of oceanic crustal thickness. White et al. (1992) rough crust formed thin, or was it thinned by some
compared inversion results based on forward syntectonic process, presumably, extension. Unusually
thetic seismogram modeling to those obtained by thin oceanic crust has been associated with the devel-
traditional slope-intercept methods and found that opment of passive margins, e.g. the Iberian margin
they differed (typically) by 20%. Even if such a bias (Whitmarsh et al., 1990) and in the Labrador Sea
is present in our methods (neither slope-intercept, (Hinz et al., 1979; Srivastava and Keen, 1995) and
nor synthetics-based modeling) the thicknesses we occasionally in ocean basins (Jackson et al., 1982;
(and Talwani et al., 1977) determine for rough B" Muller et al., 1997) presumably created at slow-
crust are abnormally thin. spreading ridges (Reid and Jackson, 1981; Chen,
EW-9501 MCS profiles in the southeastern 1992; White et al., 1992). Mantle material beneath
Venezuelan Basin indicate that the seismically fast this thin crust often exhibits abnormally low ve-
(3.5-4 km/s) 'transparent' sediments identified by locity, presumably from serpentinization, resulting
Biju-Duval et al. (1978) are, in our view, lami- from invasion of seawater along fault planes. Similar
nated turbidite sequences directly overlying rough velocities are observed in the southeastern Venezue-
basement there. The stratal relationships that make lan Basin (Talwani et al., 1977). We note that the
this obvious (filling bathymetric lows and onlap- 'originally thin' oceanic crust of the Iberian mar-
ping highs) are particularly well represented in line gin forms a narrow strip (Whitmarsh et al., 1990),
1298 (Fig. 3). The interval two-way time between while the rough Venezuelan Basin crust is much
the easily identifiable A" and underlying 'basement' wider in all directions. The southeastern extent of
increases dramatically, from 0.2 s at CDP 14,500 to thin, rough B" crust is entirely unknown; it can be
about 0.5 s at CDP 12,500. The number of reflectors seen (for example in EW-9501 line 1319, Fig. 4)
within the A"-B" interval increases dramatically being subducted beneath the accretionary wedge of
across the rough-smooth B" transition, and the re- the Curacao Ridge. The destruction of the crust by
flections become more distinct. Similar sediments normal faulting, apparent in this profile, is typical of

that seen in other MCS profiles crossing this zone. rough B" crust. The strike of those faults which
The amount of thin crust consumed by subduction can be accurately mapped is ENE-WSW, and none
and obduction is difficult to estimate, and its original seems to extend for more than 100 km m most are
southern boundary is difficult to reconstruct. shorter. Their en-echelon pattern may be indicative
The Iberian margin also features a narrow zone of wrench faulting. Some of the extrusive fingers
(25-30 km wide) of greatly thinned continental can be seen as bright, smooth spots as far as 20-30
crust whose abnormally high basal velocities are km from the principal rough-smooth boundary (for
thought to result from underplating (Whitmarsh et example, line 1298, CDP 11,000; line 1293, CDP
al., 1990). Both of these Iberian scenarios (thinned 20,350). When successive flows filled and finally
continental crust, thin transitional oceanic crust) buried the valleys, overlying flows were able to spill
produce crust having a thickness similar to that of over into the next valley, thus spreading quickly to
the southeast Venezuelan Basin, and also having the east-northeast, more slowly south-southeast. In
little or no magnetic signature. The variation of several places, the tilting of the normally faulted
apparent velocities in rough Venezuelan Basin crust blocks continued after a thin veneer of volcanic
is such that seismic velocity alone cannot be used to material had been emplaced, creating the smooth-
make the distinction between crustal types described topped tilted blocks ('ski-jumps') characteristic of
by Whitmarsh et al. (1990). As in the case of its the Central Venezuelan Fault Zone (so named by
thin, early-stage oceanic crust, the radically thinned Biju-Duval et al., 1978), and seen here in Ewing
continental crust of the Iberian margin is found lines 1293 (Fig. 2, CDP 19,100), 1298 (Fig. 3, CDP
only in a narrow band. Rosendahl et al. (1992) 14,000), and 1300 (Fig. 5, CDP 15,000).
have determined a wider (not specified, but shown Lines 1300 and 1302, though only 35 km apart,
as at least 65 km) zone of attenuated continental show different manifestations of the rough-smooth
crust off the African margin in the Gulf of Guinea. B" transition. On line 1300 the distal edge of the
This crust has also, apparently, been thinned by smooth B" volcanics is tilted back into a 'ski-
faulting, which causes a rough 'pseudo-oceanic' jump' form, probably caused by post-emplacement
appearancepseudo-oceanic appearance in reflection extensional rotation of underlying blocks of rough
profiles. This crust is of comparable thinness to that basement. The fault system that controls the geome-
of Caribbean rough B", but this thinness is defined try and structure of the 'ski-jump' must be relatively
by an elevated 'equilibrium' Moho whose smooth surficial and antithetical to a larger fault system that
reflection character is quite dissimilar to that of dips north-northwest and controls the location and
Moho imaged anywhere in the Caribbean. Next, we geometry of the large divergent wedge observed be-
present evidence that crustal extension was taking neath B" (CDPs 9000-13,000; in line 1293, CDPs
place during the emplacement of Venezuelan Basin 10,500-19,000) and the large step in Moho topog-
smooth B" volcanics. It is likely that such extension raphy (CDP 15,000; in line 1293, CDP 19,000). In
may have reactivated preexisting faults in rough B" line 1302, the transition (Fig. 6, CDP 5200) is quite
crust. gradual, the flows having apparently run around the
end of line 1300's rotating block. In both cases,
Rough-smooth B" transition thin, upper flow units appear to have traveled in
and out of the profile, leaving thin edges facing
It is apparent from the EW-9501 data that at least both north (CDP 10,900) and south (CDP 6900).
some of the area of thin rough crust pre-dates the The rough-smooth transition as seen in line 1302
emplacement of smooth B" volcanic material, and is gradual, as the expression of the normal faults is
that the area of 'originally' rough crust is, therefore, only slight there. As a result, the rough B" surface
significantly larger than previously realized. Both the can be seen beneath the distal edge of the smooth B"
reflection profiles and the sonobuoy data show that flows. A pair of the previously described through-
the thin crust underlies the Cretaceous volcanics, and going listric crustal 'faults' characteristic of rough
is present at least 50 km northwest of the rough- B" crust can also be seen (CDPs 6500 and 7500).
smooth boundary along some profiles. Ewing pro- Once it is covered by smooth B" volcanics, the re-
files 1300-1316 provide 3-D coverage dense enough flection from the top of the thin crust is quite weak,
to show that the model proposed by Diebold et al. which is probably the result of three things, i.e. a
(1981) for the rough-smooth boundary as an over- small velocity contrast, a rough interface, and the
running of en-echelon normal fault blocks by lava absence of an intervening sedimentary layer. Some
flows encroaching from the northwest is basically sonobuoys from this and previous studies have pro-
correct. The uppermost, and presumably youngest, duced complete crustal sections, particularly in an
sequence of extrusives included highly mobile flows, area where sonobuoy records (e.g. Fig. 7) show that
which were able to spread in thin fingers, along the strong postcritically refracted energy is frequently
valleys separating gently tilted fault blocks in the returned from the Moho. This area is crossed by

oes 'ewp,/~eN~-OM±
0 O4
CO o~
0
0
0
0
0
0
0 0
0 0
0
cO CO
0 0
0 0
0
. .
0
0
0 0
0 0
0
¢0 ¢0
0 0
0 0
0
c~
,:,, .. • ~D
.::.. : s : ..
0 .'..~--:_ ". 0 o
0 ..
-~-..:~ : 0
0 o
2 - " - _- -.
.
.!'. f.-.: GO o
. . .
-<.:C.
.~ .: . . . . ..
-_ .
0
0 . ~ . N
.... .-
~.
..,- ::.
....

. ...
0 7._-.~. >.
. . . _-. _ .,--~
0 .- ....
.-..r
c~
0
0 o >
0 0
~o o
¢j
i
13.. T - ' r ' -
O
0 o *r
~0oO 8
0
E
,..0
O
.~ 0 0 uD
_1 0 0
~o o
0
1.0
~o o
LU 0
0 o
q
t
0 0
0 0
0 0
.o
0 0
0 0
0 0
¢0
©
-..~.
0
0 ~:~L.:-C~::ii~:~ ::i..<~. 0
0
©
0 ?
©
0 0
0 0
0
co co
0 ©
0
0
o~
©
0 0
®o 0
0
0
00 ° 0~1
03 ':.<~:Y~<:~i<:>:i:
0 0
0 , _:-: .:.!:i.) ::: J:.::":: . 0
0 0
C'-I
o4 0~1 ©
,..o
rm
. .
©
-~. 1 . . ~ - . •
0 t¢'3
0
0
CO ©
: , 04
0 :
iI .-; .:,-.-
!.:~.....
....... ..
0
0 0
0 0
. . . . . . . . ,. . . . . . ..
~!"
04 --..-., -. . . . , • i _- . 04
CO 0 1-- 04
. ,...,
08£ 'eLl.ill/~eM-OM.L
MULTICHANNEL SEISMIC IMAGES OF VOLCANIC STRUCTURES IN THE V E N E Z U E L A N BASIN 571
{D
. ,...~
. ,...~
y.
.,..~
+..a
r,.)
e~t)
....~
. ,...~
o
o
tt-)
.,..~
o ~" ~
t--,~

572 J. D I E B O L D et al.
o<
.,..~
~D
~[--,
~c5
~D
~D
~D ~D
.,..~
. ,...~ ~D
. ,...~
oo==~
bq
9 ,.~
9 O
r~
,.~ 0
d~ 0

lines 1300 (CDPs 5000-14,000) and 1302 (CDPs 13,500 to 0.56 s (1455 m) at CDP 1800, and has
6000-12,000). The velocity functions determined a velocity (varying from 5.0 to 5.2 km/s) which is
for these buoys consistently include a pair of deep distinctly lower than that of most underlying vol-
layers whose velocities (high 6 over low 7) are canic layers, in which velocities typically increase to
consistent with a 'buried' thin crustal layer. A final 5.4-5.6 km/s. This velocity contrast is responsible
argument for identifying rough B" beneath smooth for the clear break in slope of the first arrivals at
B" crust is the similarity of the Moho reflection 12 km of sonobuoy 33. The reduced velocities of
character, as seen on EW-9501 profiles throughout this upper layer may be the result of some com-
the Venezuelan Basin. positional or physical variation (vesicles or breccia-
North-northeast of the rough-smooth B" bound- tion, for example), weathering, or interbedded (but
ary, the reflectors immediately beneath the smooth non-reflective) sediments. Planke (1994) found sys-
B" basement exhibit a divergent pattern (Figs. 2, 3, tematic velocity variations within individual flows in
5 and 6) and have interval velocities on the order of seaward-dipping wedges, sampled at ODP Site 642
5 km/s. The divergent geometries and spatial distri- on the V~ring volcanic margin. Velocities decreased
bution of the reflector packages observed across the steadily from the interior to about 3 km/s at the up-
Venezuelan Basin suggest multiple volcanic extru- per and lower boundaries of all logged flows. If flows
sive events, with the earlier events being less volu- were thick enough (more than 9 m) velocities in the
minous and more spatially restricted. The multiple body of the flows would stabilize at about 5.2 km/s,
magmatic events inferred from the stratal geometry with an upper velocity gradient zone 7 m thick, and
suggests that some of the older dates for magmatic a bottom zone 2 m thick. The marginal velocity
activity (e.g. Aptian) may actually be recording real decreases are attributed by Planke to brecciation and
events and are not artifacts of the dating technique vesicularity. Similar patterns at a larger scale are
(Donnelly et al., 1990). The rotation and divergence seen in Colombia River basalt flows (Waters et al.,
of the onlapping seismic reflectors diminishes up 1981). Planke's observation predicts that the average
section and is indicative of differential subsidence velocity of a sequence of flows will depend on the
and block rotation (Fig. 2). Note that the spatial average thickness of individual component flows; a
extent of each successive divergent reflector package sequence with a larger number of thinner flows will
increases within the wedge. Symmetry of the dip- have an overall velocity lower than another sequence
ping packages is also often observed locally towards made up of a smaller number of thicker flows. It well
seamounts and away from highs in the underlying may be that the uppermost Caribbean flow sequence
volcanic sequence. is made of thinner flows than the lower sequence.
On the other hand, such systematic velocity layering
Smooth B"; extrusive volcanics might also produce an anisotropic substance, whose
true average velocity would not be determined from
EW-9501 seismic profiles show extensive layer- sonobuoy or MCS data.
ing within the volcanic extrusive sequences making In some places, a thin (ca. 40 m) layer can be
up smooth B" crust. Patterns of layering are consis- seen in the reflection profiles, forming the uppermost
tent from line to line in the Venezuelan Basin, and element of smooth B". EW-9501 acquisition param-
along with sonobuoy results, define several distin- eters (source and receiver towed at 7.5 m) dictate
guishable sequences. Distinctions can be most easily that this is the thinnest basaltic layer that can be
made in the southern half of the basin. An upper se- detected. Drilling of seaward-dipping reflector se-
quence thickens dramatically in the direction of the quences has rarely sampled a flow thicker than 10
rough-smooth B" boundary. Sub-units of this upper m, but we note that the morphology of the smooth
sequence also thicken towards the southeast, and B" sequences is more reminiscent of continental
seem to be individually thick and voluminous. Dis- flood basalts, in which individual flows may have
tinct and characteristic refracted arrivals in sonobuoy thicknesses of 100 m or more (Waters et al., 1981).
records show that velocity within individual layers The observation of simple, edge-like terminations
tends to increase gradationally. Sonobuoy 33 (Fig. 7) of these thin upper layers suggests that they might
deployed at the location of CDP 10,100 of line 1300 in fact be individual flows. An example, referred to
(Fig. 5) is an excellent example. Refracted arrivals above, is seen at CDP 6900 of line 1302 (Fig. 6).
from the volcanic sequence are seen to emerge from Similar flow sequences form the thin, mobile 'fin-
the seafloor reflection at 8.2-km source-receiver off- gers' filling basement lows southeast of the principal
set and 7 s, reduced traveltime. This record also rough-smooth B" boundary. Their identification as
features a strong Moho critical reflection. flows, and not turbidite lenses is based on the strong
The base of the uppermost layer is clearly im- reflection amplitude of their upper surfaces, their
aged as a reflection in MCS line 1293. The layer distribution over thin B" crust, and on detailed anal-
increases in thickness from 0.3 s (780 m) at CDP ysis of stacking velocities in a few areas. Many

tt~
~D
~D
~D
o,..,
I
~D
~D
~D
9
E
o,..,
O
o,..,
o~
O
9
O
~D
tt')
d0
o,..,
cr
~D
~
~D
r162
9
eq
r
~D
o,,.~
~5

Fig. 7. Sonobuoy33, located along line 1300 (Fig. 5) returned an unusually fine record. Precritically reflected, postcritically reflected and
refracted arrivals were all used to find a detailed 1-D solution.
sonobuoy records contain weak, wispy arrivals, ap- reflector when more of the lines are re-stacked after
parently refracted in thin flows (or sills) separated detailed velocity analysis.
from the underlying body of volcanics by sedi- The upper layer described in the preceding para-
mentary layers. These observations suggest that the graphs is the uppermost member of a sequence of
reduction in overall upper-sequence velocity arises similar layers. Like the uppermost layer, these un-
from the inclusion of intercalated sediments. Such derlying layers thicken towards the rough-smooth
sediments (marbleized limestones) were sampled at B" boundary. Between CDPs 11,000 and 16,000 of
DSDP Site 152 (Edgar et al., 1971). A sonobuoy line 1293 (Fig. 2), the base of the resulting wedge
deployed as closely as possible to the location of is marked by a bright reflector, whose continuity is
DSDP Site 150 features an early refracted arrival broken only by scattering from a structural high in
characteristic of a thin (ca. 100 m) sill (or sequence the upper smooth B" surface at CDP 14,400. In-
of sills) whose top is about 300 m above the massive terval two-way time for this entire wedge increases
volcanic body. Similar events have been observed in from about 150 ms at CDP 11,000 to 1700 ms at
sonobuoy records elsewhere in the Venezuelan Basin CDP 16,000. Sonobuoy-derived velocities indicate
and in the western Pacific (Abrams et al., 1993). that this represents a thickening from 400 m to 5
In many other sonobuoy records from the zone of km, which takes place across a distance of 62.5
smooth B", the refracted arrivals similarly show that km. Similar bright reflectors can be seen in the
what appears in the reflection profiles to be the up- less-painstakingly processed sections, and are partic-
per surface of the LIP is in fact a sill (or flows), ularly apparent around CDP 17,500 of line 1298 and
or a sequence of sills (or flows) and sedimentary CDP 11,500 of line 1300 (Fig. 5). The strength of
layers, whose overall thickness varies from zero to these reflectors suggests three characteristics of the
300 m. Similar sills were drilled during DSDP Leg corresponding horizon: smoothness, a single inter-
15. The bottom of this uppermost layer (or sequence face with extremely high impedance contrast, and/or
of layers) is marked in much of line 1293, by a superposition of a sequence of thin layers with
visible reflector, indicating the transition to material moderately high impedance contrast. The sonobuoy
with higher seismic velocity. The fact that this layer, results indicate a slight overall increase in seismic
with its characteristic slightly lower velocity, is seen velocity, and therefore, impedance, but the result-
in many sonobuoy records, suggests that it exists in ing contrast alone is not sufficient to explain the
most of the area, and that its base will be seen as a reflection strengths. The smooth appearance of the

reflections may result from the inclusion of an appre- more or less constant at 0.8 s, or ca. 2 km thickness.
ciable amount of metamorphosed sedimentary layers Moving farther west, to lines 1304-1317, the up-
between volcanic extrusives, which would produce per-sequence thickness increases, and begins to vary,
reflection-enhancing fine-scale layering. Noting also as two sub-basement highs begin to appear (Figs. 8
the patterns of less-strong but still clear reflectors and 9). The upper-sequence thickness at CDP 5000
between this basal horizon and B" it is worthwhile to is about 4 km, nearly equal to the maximum 5 km
re-examine the sequences mapped in southern Haiti of the wedge of line 1293. In general, the upper
by Maurrasse et al. (1979) who interpreted them as surface of the upper sequence is far smoother than
the products of intermittent but "extensive magmatic its bottom. The lower surface also tends to dip away
outpouring - - at least throughout the late Creta- from local basement highs. Such dips are strikingly
ceous, and possibly throughout the early Cretaceous apparent between CDPs 11,000 and 16,000, line
as well." This ca. 2000-m-thick series of diabase and 1293 (Fig. 2), but they are seen at various places on
basalt flows and sills, some featuring columnar joint- all of the other lines as well.
ing, and others pillows, typically separated by layers In one area, sonobuoy records consistently show
of pelagic sediments, would provide a reasonable that this upper sequence includes at its base a
model for parts of the superior wedge of Venezuelan low-velocity zone with locally substantial thickness.
Basin volcanics. The most persuasive example of the seismic arrivals
The most remarkable feature of the upper se- characteristic of this zone are seen in sonobuoy 65.
quence is its massive, wedge-like southeastward Located on N-S line 1320, near the W S W - E N E
thickening toward the rough-smooth B" boundary. line of volcanic edifices whose magnetic anomalies
In the eastern part of the basin, as exemplified in were previously mapped, this record (Fig. 8) shows
line 1293 (Fig. 2) the wedge is about 100 km long, clear evidence for a significant low-velocity zone
but north of the narrow point of the wedge (CDP within the volcanic sequence. Based on analysis of
11,000) the upper sequence is relatively thin, though the adjacent sonobuoys, this zone thins rapidly to the
its thickness seems to increase in the vicinity of the north and south and is, we judge, likely to comprise
seamounts at CDPs 2100 and 6000. Examining the volcaniclastics from eruptive events at the nearby
profiles, one by one, progressively to the west, the volcanic mounds, similar to those which have also
flat part of the upper sequence is (usually) clearly been imaged along the northern and northwestern
defined, and can be seen to thicken steadily. For- edges of the Venezuelan Basin.
tunate velocity contrasts perhaps coupled with the Directly beneath the bright reflector marking
presence of intervening layers of sediments, make the base of this extensive, smooth-surfaced upper,
the basal reflector of the upper sequence quite ap- wedge-forming sequence is another sequence of dip-
parent between CDPs 300 and 7000 on line 1300 ping reflectors whose character is entirely different.
(Fig. 5). Interval two-way time of the sequence is As it appears between CDPs 11,500 and 13,000 of
Fig. 8. Sonobuoy65, though not yielding as high a quality signal, nor lasting as long as sonobuoy 33, returned a remarkable diagnostic
pattern of arrivals defining a prominent low-velocityzone beneath the upper volcanic layer.

line 1293, this sequence resembles the 'seaward- end just at (apparently) the crest of the ridge-like
dipping reflectors' (SDRs) observed at many pas- high. Brute-stack processing of the connecting lines
sive continental margins (Hinz, 1981; Mutter et al., 1299 and 1303 shows that there is little relief along
1982) but also those seen in the Kerguelen Plateau this ridge, and that Moho remains flat beneath it.
(Ramsay et al., 1986). In comparison to the major The southern ridge, though well developed at
reflectors of the upper wedge, which are typically CDP 12,500 on line 1319, becomes less distinct,
well separated, and some of which can be followed and possibly bifurcates as it is mapped to the E N E -
continuously for 75-100 kin, these 'SDR-like' re- WSW, along crossings of Conrad MCS lines 119
flections are downward-concave and are densely and 137 (Diebold et al., 1981) and EW-9501 lines
spaced. Although the elements of the upper wedge 1304, 1302, and 1300 (Fig. 11). In line 1300 (Fig. 5)
can be clearly seen to terminate at and overflow the the basement high is actually a shoulder, at CDP
tops of rotated normal fault blocks, the basal extent 9000, but a small but perceptible high can be clearly
of these 'SDR' reflectors cannot (so far) be reliably seen in the lower volcanic sequence there. Sixty-five
determined. In nearly every case, dipping reflectors km farther east-northeast, no high can be seen; the
of both kinds dip away from a basement high. This magnetic lineation has disappeared, and the upper,
kind of high is best developed at CDP 10,000, line dipping volcanic sequence leading to the rough-
1293. Here, as on some other lines, wedges of re- smooth B" transition has shifted NNW-SSE.
flectors can be seen dipping away from the highs in
both directions, though in every case, the larger and Volcanic mounds
most clearly developed set dips toward the southeast
and the rough-smooth B" boundary. Volcanic edifices, some completely covered by
sediment, others seen as protruding seamounts, are
Massive volcanic bodies numerous in the Venezuelan and Colombian Basins.
Some of these appear to fall along linear trends.
Between the zones of dipping reflectors, we find The seamount shown in fig. 4 of Ladd and Watkins
basement highs, in the form of gentle domes or (1980) is one of an E-W-trending chain, which in-
ridges. These highs are generally about 30-40 km cludes the large seamount imaged at CDP 2100,
across and feature few, if any, internal reflectors. line 1293. Other volcanic cones are found parallel
In the Venezuelan Basin, the principal highs form to the NE-SW-trending magnetic anomalies mapped
two linear trends, essentially coincident with lin- by Donnelly (1973) and Watkins and Cavanaugh
ear magnetic anomalies mapped in the region by (1974). As in the case of the seamount at CDP 5900,
Donnelly (1973). The two lineations are somewhat line 1293, many of these cones (the shape of this
oblique, and trends mapped by Case and Holcombe one is well determined by multibeam echo sound-
(1980) indicate that they may intersect near CDP ing and side scan sonar) appear to be surrounded
10,000 of line 1293 (Fig. 2). This might be a factor by a now submerged moat, or perhaps more accu-
in producing the prominence of the high in this rately, a thickening of the upper volcanic sequence
particular crossing. The dipping reflector sequences described above. This thickening might be responsi-
appear to originate in these highs which suggests ble for some of the elevation of the northeastward
that they are the locus of volcanic sources. Small basement high. The fact that most of the sonobuoys
knolls are evident at a few of the highs, made along this N E - S W trend show clear evidence of
particularly apparent in the horizontally compressed a low-velocity zone (Fig. 11) suggests that these
sections shown here. Examples can be seen at CDP low velocities correspond to volcaniclastic material
1200, line 1298 (Fig. 3) and CDP 13,000, line 1302 causing the thickening of the upper sequence. These
(Fig. 6). These may be the peaks of (or the flanks of low-velocity zones are seen only along this trend,
nearby) extrusive mounds. and they are always found just below the upper, slow
The northernmost of the two ridge-like lineations extrusive layer. A possible explanation is that the
is crossed completely by only two reflection profiles: culminative volcanic sequence began with produc-
1293 (Fig. 2) and 1320 (Fig. 10). ODP Site 150 tion of volcaniclastics, followed by extensive, mobile
was located near CDP 5500 of line 1320, on the flows, whose final expression is the thin flows seen
flank of this high, whose relief, in conjunction to overlie smooth and rough B" basement alike.
with ocean currents, has produced a dramatically Fig. 12 shows the SE end of EW-9501 line 1323,
thinned sedimentary section there (see Driscoll and on the flank of the Beata Ridge. Completely buried
Diebold, Chapter 20). In both profiles, the Moho by sediments, a feature we interpret as a volcanic
can be seen continuously beneath the high, roughly mound is located at CDP 23,800. The upper volcanic
maintaining a constant two-way time, indicating sequence is clearly defined by horizon B" and the
that the ridges are supported by a thickened crust. distinctive basal reflector. A deeper (7.5 s, two-way
Between these two profiles, four EW-9501 profiles time) bright reflector, truncated around CDP 19,700,

o
ca
o
~D
.o
,.r
o
9
T
,.r
~D
~D
~D
~
"O
cD
O
r162
~D
c5
o,I
,,..~
9 ,..-,
o
rz~-
r162 , ~
r162 ~D
r~
~D
ca
tt3
G'~ er

M U L T I C H A N N E L SEISMIC IMAGES OF V O L C A N I C STRUCTURES IN THE V E N E Z U E L A N BASIN 579
Fig. 10. MCS line 1320, which runs N-S, through DSDP Sites 146/149 and 150. A broad basement high is seen at CDP 5000, and a
volcanic edifice appears at CDP 12,500. Sonobuoy 65 (Fig. 8) was deployed near DSDP Site 146/149.
Fig. 11. Map of basement highs, dipping reflectors, the rough-smooth B" boundary and related normal faults. Locations of sonobuoys
whose records show low-velocity zones (cf. sonobuoy 65, Fig. 8).

Fig. 12. ESE end of MCS line 1323, Beata Ridge flank. The upper volcanic sequence thickens dramatically toward what is interpreted as
a buried seamount. To the west-northwest, flexural rebound has elevated the formerly flat B" horizon.
is interpreted as the reflection from the top of a DISCUSSION

sill. The post-volcanic deformation and uplift seen
in the WNW end of the figure is related to the An increasingly important part of the puzzle of
extensional flexure of the Beata Ridge flank (Driscoll the origin of the Caribbean plate is the zone of thin,
and Diebold, Chapter 20). The sill-like reflector rough basement, described by Talwani et al. (1977).
is, essentially, parallel to overlying B", and would The EW-9501 profiles show clearly that this crust
therefore (according to our reconstruction) have been pre-dates at least the voluminous late-stage volcan-
horizontal when originally emplaced. The thickness ism that characterizes the Caribbean oceanic plateau,
of the upper volcanic sequence doubles towards the and may, therefore, typify 'original' Caribbean ocean
volcanic mound. This pattern is frequently observed crust. Rough mapping of this zone (Diebold et al.,
throughout the Venezuelan Basin, suggesting that 1981) was improved by Leroy (1995) and continues
the eruptions feeding the mounds started before, or in this paper. EW-9501 results indicate that this thin,
simultaneously with, the flows forming the upper rough basement extends beneath smooth B" flows at
sequence. The association of these volcanic features least as far as the leading edge of the underlying
with the underlying divergent reflectors indicates early-stage series of dipping reflectors (Figs. 13 and
that these features are long-lived and not just a 14). This interpretation is based on the continuity of
late-stage phenomenon sourced from small degree reflections from the top of the rough crust, on the
partial melts associated with the extension (e.g. identifiable character of reflections from the base of
decompressional melting) or differentiated magma the upper volcanic sequence, and on sonobuoy ve-
from the solidifying basalt province. Plateau rifting locities. Sonobuoy velocity analyses do not rule out
and volcanoes are features of other oceanic LIPs the possibility that thin, rough crust is sandwiched
(Manihiki, Ontong Java, Kerguelen) and may be an between the Moho and all of the volcanics of the
integral part of the evolution of oceanic plateaus. Venezuelan Basin, but neither do they prove it.
Determination of the age and petrologic character The new profiles have also led to an important
of these events would distinguish between late-stage reinterpretation of the sediments immediately over-
parasitic volcanism versus a more long-lived source lying rough B". The thickness of the A"-B" interval
of volcanism throughout the development of the increases markedly in the zone of rough basement,
basalt province. particularly to the south and east. This increase is

5000
6000
9~ seafloor
o 7000
~ horizon A"
E
d 8000
E J
early stage dippingreflectors->'~. ~ to~can~,_ / thin, rougn B"crust
~ 9000
10000
11000 -
f
12000 -" ' ' ' I . . . . I . . . . ! . . . . I ' ' ' ' I ' '
EW9501 MCS line 1293
0 -
5000 -
10000 -
/
15000 -
"~" ~ V.E. 5.0

20000 ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' '
0 5000 10000 15000 20000 25000

CDP#
Fig. 13. Line drawings showing our interpretation of the major horizons in MCS line 1293 (Fig. 2) in two-way time (top) and converted
to thickness (bottom) using averaged sonobuoy velocities.
due to the presence of a body of horizontally layered the Colombian Basin also reveals elevated sequences
sediments that is not seen above the higher-stand- of dipping reflectors adjacent to this rough crust,
ing smooth B" volcanic basement. Biju-Duval et al. similar to the relationships we see at the rough-
(1978) identified these sediments as a 'transparent' smooth B" transition in the Venezuelan Basin.
sequence not seen northwestward, in the smooth B" It is not easy to explain the large area of abnor-
zone, and otherwise characterized by relatively high mally thin oceanic crust in the Caribbean. A zone
seismic velocities. This well-laminated deeper se- of thin (3.3 km) crust in the Labrador Sea was
quence fills basement lows and onlaps highs, appear- associated with slow spreading rates (6 mm/year)
ing to be the result of turbiditic, rather than pelagic by Srivastava and Roest (1995). This thin crust
deposition. In agreement with earlier results, we de- is broken into tilted blocks, exhibiting a basement
termine an acoustic velocity for the turbidites, ca. 3.2 topography much rougher than that seen in the
km/s, which is distinctively higher than that of the Venezuelan Basin, however. Similar, thin crust ob-
overlying pelagic sediments. This sequence reaches served at the passive Iberian margin exists only in
thicknesses as great as 800-1000 m, and thus rep- a narrow zone, whereas the thin, rough crust we
resents a significant volume of material. Sediments see in the Caribbean occupies an extensive area.
with similar onlapping relationships have been ob- A scenario like the one presented by Rosendahl
served to the west, in the Colombian Basin (Lu and et al. (1992) whereby a passive continental margin
McMillen, 1982; Kolla et al., 1984; Bowland and is flanked by a broad zone of thinned 'oceanized'
Rosencrantz, 1988; Bowland, 1993) and interpreted continental crust is also possible. The interpreted
there as volcaniclastic turbidites. The stratal geome- composite section presented as that paper's fig. 2
tries and areal extent (at least 50,000 km 2) of these also includes a wedge of seaward-dipping reflectors
sediments in the Venezuelan Basin, however, sug- over a necked-down crustal section at the seaward
gests strongly that they are of terrigenous origin (see boundary of this thinned crustal zone. One of the ob-
Driscoll and Diebold, Chapter 20). It is likely that servations that supports their notion of an elevated,
the rough crust underlying these similar sediments 'equilibrium' Moho is that some of the diamond-
in the southern Colombian Basin is also thin. Exam- pattern apparent faults continue through and below
ination of better-quality 1970's vintage MCS data on reflection Moho. If many of the 'faults' were ac-

6000 =
seafloor
7000
8ooo horizon A"
9000 "
.. thin, rough B" crust
10000 ......" " . j ~

Moho 9 " .....:!::::::.....i " "
11000 ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I '
EW9501 MCS line 1300
5000 -
~. 10000
15000
V. E. 5.0
20000 ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I '
0 5000 10000 15000 20000

CDP #
Fig. 14. Line drawing interpretation of major horizons in MCS line 1300 (Fig. 5) as picked in two-way time (top) and converted to
thickness (bottom).
tually sideswipe reflections from line-oblique base- crust to have been created at a much slower rate
ment ridges (and this kind of geometry could be (such as the 0.5 cm/year, full rate, determined for
expected from most of the ship tracks in the data the Arctic mid-ocean ridge) would exacerbate this
set of Rosendahl et al., 1992) then the amount of problem of the lack of pelagic sediments, significant
thinning required would be greatly increased. The amounts of which cannot be detected between the
absence of such Moho-penetrating reflectors from rough B" crust and the overlying turbidites. A third
EW-9501 data could be explained in this way. On possibility is that the crust, once of normal thickness,
the other hand, this amount of extension of the was pervasively thinned by extension, without any
thinned oceanic crust should have generated signif- associated volcanism. It is difficult to document such
icant decompression melting and underplating, for a process elsewhere, however.
which there is no evidence. In fact, the Rosendahl The thin, rough crust has a tendency for brittle
et al. (1992) model involves a volcanic-extrusive failure along E - W trends, as evidenced in its de-
transition to pure seafloor spreading, which requires struction by normal faulting beneath the Curaqao
a seaward smoothing of the crustal surface. The Ridge (Fig. 4). Elsewhere, extensional features are
transition to thin crust in the Caribbean shows the most concentrated near the rough-smooth B" tran-
opposite trend. Even thinner crust (3-4 km) has been sition. Tilted blocks, typical of normal faulting, are
recorded at slow spreading ridges in the Arctic (Keen seen from place to place, and EW-9501 results
and Barrett, 1972; Jackson et al., 1982). We deter- confirm that the en-echelon, ENE-WSW-trending
mine that minimum dimensions for the Venezuelan pattern determined for the 'Central Venezuela Fault
Basin thin crust are roughly 300 km (E-W) and Zone' by Biju-Duval et al. (1978) is characteristic
350 km (N-S). At a 1 cm/year half-spreading rate of the entire rough-smooth boundary. The faulting
(and neglecting subsequent extension) it would take of the rough basement is not restricted only to the
at least 30 m.y. to create this much crust, which is CVFB. These faults appear to be the result of ex-
unlikely, given the constraint of reasonable rates of tensional processes which may exploit pre-existing
pelagic sedimentation (1-2 cm/ka). For the thin B" faults which were, perhaps, formed at the time of

crustal creation. We suggest that the extensional di- of faulting and of the rough-smooth boundary sug-
rection is somewhat oblique to that of the strike gest some change in spreading direction between the
of the preexisting faults (and hence, perhaps, to an time that rough crust was created and when it was
original seafloor spreading direction) resulting in an extended and covered by volcanics. The extension
en-echelon pattern of ' wrench' -like tilted blocks. and volcanism are definitely contemporaneous, but
This oblique relationship, which allows the Creta- the oceanic-like crust seems to have existed for at
ceous flows to run around the ends of the en-echelon least a short time. Second, we see, in line 1293,
fault segments, entirely explains the variations seen an entire rough-smooth boundary sequence, some
in the character of the rough-smooth B" transition. 25 km northwest of the current boundary, buried
In several locations at which the flows appear to be beneath at least 5 km of volcanics forming the
dammed by a rough basement fault block, the distal subsequent sequence.
edges of the upper flows are tilted back, forming The EW-9501 profiles show that the volcanic
characteristic 'ski-jumps'. The likely mechanism for structures imaged beneath the smooth B" horizon
these is that the underlying block continued to rotate form two sequences, distinguished by morphology
after being covered by a thin Cretaceous flow de- and separated by a prominent reflecting surface.
posit, tipping it back towards the northwest. In some The upper and lower volcanic bodies both con-
cases, subsequent flows filled in the resulting trough, tain dipping reflectors, but these too have distinct
and in line 1293, an entire buried 'ski-jump' se- morphologies. The uppermost dipping reflectors be-
quence can be seen between CDPs 17,000-18,000, neath smooth B", first identified by Hopkins (1973),
at 8.5 s two-way time. This process suggests that Ladd and Watkins (1980), Stoffa et al. (1981), and
extension and emplacement of the upper volcanic Diebold et al. (1981) form systematic sequences that
sequence were penecontemporaneous. That some are quite different than those observed elsewhere
tilted blocks appear to be capped by (presumably at passive continental margins (Hinz, 1981; Mutter
originally sub-horizontal) thin flows requires that et al., 1982; Planke and Eldholm, 1994). Dipping
tilting and extension continued, at least briefly, after reflectors in the deeper, older Caribbean sequence,
volcanic production in the area had stopped. however, resemble these 'classic' seaward-dipping
North-northwest of the present-day rough- reflectors (SDRs) much more closely. Similar fea-
smooth boundary, an even more deeply buried series tures are found at some oceanic plateaus, particularly
of dipping reflectors is seen, exhibiting the smaller- Kerguelen (Ramsay et al., 1986; Coffin et al., 1986;
scale, steeper dips and concave-downward curvature Colwell et al., 1988; Schlich et al., 1993), a plateau
characteristic of typical SDRs imaged at the Vcring whose size is comparable with that of the Caribbean
Plateau escarpment and elsewhere. The top surface Plateau. As in the case at Kerguelen, the sources
of the rough, thin basement exposed to the south- of the older dipping sequences in the Venezuelan
east can be followed almost continuously from the Basin appear to be basement highs, beneath which
rough-smooth transition to the area in which this reflectors, if any, are horizontal. Similar, too, is the
'typical' sequence is seen. All of these features sug- organization of the basement highs, which form sev-
gest (or, rather, require) that at least the southeastern eral roughly parallel ridges, which lie perpendicular
75 km of Venezuelan Basin volcanics were deposited to the direction of extension. The plate tectonic
as a sequence of flows, with little, if any, sill forma- setting of Kerguelen's origin is much clearer than
tion. It appears that volcanic sources were 'uphill', that of the Caribbean, however, since it is flanked
to the northwest, and that the southeastward-thick- by reliably identified seafloor spreading magnetic
ening of the individual flows was the joint result anomalies in the southeast Indian Ocean.
of local relief, the source area having been uplifted, Unlike the examples just cited, the Caribbean
perhaps as White and McKenzie (1995) suggest, by Plateau features an overlying thicker, more extensive
the thermal effects of a mantle plume, and of the set of flows, each apparently more voluminous and
steady subsidence of the underlying weak, thin crust. mobile than the individual flows of the underlying
It is conceivable that the small-scale dipping re- 'classic' sequence. We are prompted, therefore, to
flectors of the early volcanic sequence, as seen near envision a two-phase volcanic history, in which an
CDP 12,500 on line 1293 (Fig. 2), mark the initi- initial period of slow extension and crustal thicken-
ation of a seafloor spreading episode during which ing, typified by the construction of local volcanic
rough B" crust was formed. It is thus tempting to highs, with flanking dipping flows, is superseded by
correlate the ski-jumps with landward-dipping 'D' a more vigorous episode of extension and magmatic
reflectors observed by Keen and Potter (1995) mark- emplacement. This second phase could also be sub-
ing the transition between a broad seaward-dipping divided, as indicated by the low-velocity zones found
sequence and oceanic crust southeast of Nova Scotia. within it, and by low seismic velocities and high flu-
At least two observations argue against (but do not idity evident in the uppermost flow sequences. The
preclude) this scenario. First, the en-echelon patterns upper series of flows can be seen to thicken towards

the west, and this thickening is particularly apparent veloped a volcano-tectonic model for the magmatic
between the underlying highs, which appear to be and tectonic development of the crust that floors the
sources of the early-stage dipping reflectors. This Venezuelan Basin (Fig. 15). This model is consistent
trend continues in the EW-9501 reflection profiles with the available age, compositional, and geophys-
on the flank of the Beata Ridge, indicating that the ical data, and is compatible with the 'mobilist'
volume of second-stage volcanics increases toward model for the development of the Caribbean. We as-
the west. An upper sequence of volcanics has also sume that the proto-Caribbean crust was formed by
been imaged on the Ontong Java Plateau (Hagen et seafloor spreading in Late Jurassic/Early Cretaceous
al., 1993) but those data unfortunately lacked the time in the eastern Pacific. An unresolved issue is
penetration to reveal underlying structure. Bercovici the initial thickness of the proto-Caribbean crust. We
and Mahoney (1994) proposed, based on Ar-Ar age have assumed normal oceanic crust thicknesses of
dating, that the Ontong Java Plateau was emplaced approximately 6 km for the proto-Caribbean because
in two phases, the first from 122 to 120 Ma, and a the seismic reflection data indicate that faulting and
second from 90 to 88 Ma. Therefore, it is conceiv- extension were concomitant with, and subsequent to,
able that the upper volcanic sequence there is 30-34 the magmatic activity. However, we appreciate the
m.y. older than the underlying material. Tarduno et fact that the initial crustal thickness could be sub-
al. (1991), however, hypothesized that the Ontong stantially less if formed at an ultra-slow spreading
Java Plateau was emplaced over only 3 m.y., during ridge, as discussed above.
the early Aptian. Prior to the Senonian, crustal intrusion and erup-
Northwest of the rough-smooth boundary, NE- tion of basaltic flows began (Fig. 15a). The fact that
SW-trending grabens and monoclinic normal faults, the sub-basement highs we feel were created dur-
previously mapped by Case and Holcombe (1980) ing this process are relatively transparent in seismic
and others, may be the result of post-volcanic ex- reflection profiles may result from their being com-
tension, or perhaps are simply the result of thermal posed principally of vertical dikes, which cannot be
subsidence. It is our interpretation that sonobuoy- imaged with conventional profiling techniques. This
derived low-velocity zones (Fig. 11) near the flanks magmatism was accompanied, at least in its late
of the volcanic cones situated along the northwest- stages, by extension and crustal thinning, producing
erly edge of our Venezuelan Basin seismic grid some of the structure imaged in the upper surface of
are volcaniclastic aprons of those volcanic edifices, this sequence. The extent and volume of magmatic
which were produced early in the final stage of products imaged in the EW-9501 data are consistent
extrusive production, and emplaced on the upper, with a hotspot or mantle plume source. It is difficult
perhaps sediment-covered surface of the previous to determine whether extensional deformation pre-
sequence. Low-velocity zones just landward of sea- dated the impact of the plume head and associated
ward-dipping reflectors along North Atlantic rifted magmatic activity (Galapagos hotspot; Duncan and
margins are shown in the synthesis by Eldholm and Hargraves, 1984) or vice versa. Numerous hypothe-
Grue (1994), but their origin was not identified, and ses concerning the interaction of plume heads with
no nearby volcanic mounds were imaged. both active and passive rifting of the overlying litho-
It is frequently observed that around these vol- sphere have been advanced (Coffin and Eldholm,
canic edifices (either seen as seamounts or sub- 1994). One school of thought is that the impact of
merged beneath sediments) the upper surface of the the plume head on the mechanical boundary layer at
lower sequence is depressed. Since it is unlikely the base of the lithosphere causes large-scale melt-
that the sheer weight of these relatively minor vol- ing and the transfer of stress from the plume to the
canic cones is sufficient to result in crustal flexure, overlying lithospheric plate (i.e. active rifting). Mod-
we conclude that while they have subsided, as the ifications on this theme have been put forth by White
consequence of a combination of thermal contrac- and McKenzie (1995) in which they propose that
tion and the withdrawal of the underlying magmatic the magmatic activity is a response to the thermal
source, it is also likely that the volcanic cones, and uplift and associated decompression melting. The al-
other sources of late-stage volcanism, were located ternative hypothesis is that regions of the lithosphere
in topographic lows between the highs typifying deformed by plate reorganizations act as foci or es-
early-stage crustal thickening. caping points for hot regions of the mantle, allowing
melts to breach the lithosphere (Anderson et al.,
Magmatic and tectonic development of the 1992). In this scenario, the crustal thinning would
Caribbean be related to subduction along the eastern plate and
the associated back-arc extension. Because of the
On the basis of the new MCS data acquired extensive overprinting by the magmatic activity, it
in the eastern Caribbean (Fig. 1), together with is difficult to ascertain whether the plume head or
previous collected data from the region, we have de- lithospheric extension was the cause or response.

Fig. 15. Cartoon showing the volcanic and tectonic evolution of the Venezuelan Basin. (a) Original, normal-thickness oceanic crust (right)
is thickened by intrusion (vertical dikes) with mound-building extrusion producing sloping flanks, and through underplating by depleted
source material. Original oceanic crust is indicated by gray shading. (b) A second phase of volcanism is marked by increased extension,
re-intrusion by sills, production of massive extrusives forming the smooth B" horizon, and additional underplating. (c) Further extension
is focused in the west, resulting in rifting of the entire thickened crust and flexural uplift of Beata Ridge.
The structure of the Caribbean basalt province, dipping fault system (Driscoll and Diebold, Chapter
as imaged by EW-9501 data, indicates at least two 2O).
episodes of magmatic activity which is consistent The dipping reflector packages observed in both
with the age dating information (Donnelly et al., the early- and late-stage volcanic sequences in the
1990). The early stage was more localized with Venezuelan Basin exhibit structural similarities to
steeper-dipping reflectors flanking basement highs seaward-dipping reflectors (SDRs) observed along
(Fig. 15a); the late stage is regionally more extensive other volcanic margins (Mutter et al., 1982; Coffin
with localized brittle deformation (Fig. 15b). Based and Eldholm, 1994). The commonly cited model
on seismic reflection and refraction, the plate was for emplacement of the SDRs is that thermal dy-
thickened further during this second phase by extru- namic support uplifts the magmatic source above
sion of lava flows over a vast area, intrusion of dikes sea level with extrusions flowing away from the
and sills, and underplating by residual mantle from source region (Mutter et al., 1982). With time, the
the melting event. The rough-smooth B" boundary source region evolves into a spreading center and
imaged in seismic profiles is, in this model, the when the anomalous heat dissipates, the plate cools,
edge of the basalt province overlying the older plate contracts, and subsides. The primary dip of the vol-
(Fig. 15b). The edge of the basalt province is also canics is reversed because of the subsidence and
coincident with an abrupt shoaling of Moho (Figs. 2 the SDRs diverge, or thicken, toward the extinct
and 13). The divergent wedge that is well imaged source. In the case of the eastern Caribbean, how-
below B" in line 1293 (Fig. 2) and the abrupt shoal- ever, we know that the adjacent rough crust was
ing of Moho may be controlled by a northwestward present at the time of the late-stage magmatic em-

placement forming the dipping wedges northeast of CONCLUSIONS

the rough-smooth B" boundary, because of the con-
temporaneous faulting and overlap with tongues of The concept of the 'Caribbean Plateau' as a
smooth basement. The thickness and the distribution monolithic allochthon of crust thickened by Creta-
of the dipping reflectors varies between the early- ceous flood basalts is laid to rest by multichannel
and late-stage basalt flows. We propose that these seismic data acquired aboard the NSF research ves-
observed variations record changes in the volume sel Ewing in 1995. In the search to define the
of magma supplied to the region through time. The location and structure of the edge of the Caribbean
internal geometry of the divergent volcanic wedges volcanics, we found two edges, whose stratigraphic
and the intercalated limestones indicate that the dip- relationships show that the volcanism occurred in at
ping divergent wedge was emplaced in the marine least two distinct phases. The upper series of vol-
environment. Consequently, we propose that the for- canics are the only ones which have been sampled, at
mation of ' seaward' -dipping reflectors can occur in least in the Venezuelan Basin. The underlying phase
the marine environment, and that changes in the must certainly be chronologically earlier, though
flow morphology record variations in magma pro- how much earlier is difficult to quantify. Some of the
duction. These effects may explain the variety of crustal elements and their stratigraphic relationships,
scales at which continental margin SDRs have more described above, provide some constraints on the
recently been observed (e.g. Skogseid and Eldholm, relative sequence of events. Some, if not all, of the
1989). thin, rough crust now seen as basement in the south-
Due to the segmented nature of the NNE fault east part of the basin must have existed before the
systems forming the rough-smooth boundary in the emplacement of the upper volcanic sequence. This is
Venezuelan Basin, tongues of basalt flows have clear, because thin stringers of the uppermost flows
flowed subparallel to the down-dropped hanging can be followed from where they overlie 'smooth'
walls, overrunning rough basement towards the areas to where they are seen as a thin covering of
southeast, generating a complex but predictable pat- what is in every other respect thin, 'rough' crust.
tern between the rough and smooth B" basement. Where that covering is seen, it is evident that an
Our hypothesis, that the source of the basalts is intervening layer of sediment, if any, is very thin,
either along-strike or up-dip, is consistent with new implying that the age difference is not great. That the
models for the evolution of similar dipping re- thin crust was not being created simultaneously (say,
flectors on the Kerguelen Plateau (Schlich et al., in a rifting margin) is ruled out by the evidence that
1993). Furthermore, recent studies from the Pacific the thin crust was being mildly extended, via block
indicate that as in the case of some continental faulting, while the upper sequence of volcanics was
extrusives (Self et al., 1996) marine basalt flows spilling onto it. Similar relationships between thin,
develop an outer carapace (chill margin) that allows rough crust and the lower sequence of volcanics,
the flows to be insulated and consequently flow over whose southeasterly edge forms structures very sim-
great distances (Clague et al., 1990; Abrams et al., ilar to the wedges of seaward-dipping reflectors
1993). The earlier models proposed that the flows observed at many passive margins, are not clear in
were subaerial because they could not reconcile the the data. It is certainly possible that the generation of
vast spatial distribution of the volcanics with in- the thin crust and these wedges of dipping reflectors
ferred high-viscosity submarine flows (Mutter et al., was coincident. We hypothesize that the southeast-
1982). ern edge of the early-stage volcanic sequence in the
The location and N E - S W distribution of the Venezuelan Basin, with its 'seaward-dipping' reflec-
seamounts appear to have a structural control. Most tors is the rifted margin, either of a pre-existing
of the seamounts that have been mapped are in the submarine plateau of modest thickness, or perhaps
northern half of the Venezuelan Basin (Matthews and even of a re-rifted segment of normal oceanic crust.
Holcombe, 1976) but as the EW-9501 data show (cf. The thin, rough crust would then have been formed,
Fig. 12) many seamounts in the southern half of the perhaps with normal thickness, by seafloor spreading
basin have been buried by sediments and are, thus, away from this margin. If at this point the crust in
difficult to map. We hypothesize that seamounts question was still in the Pacific Ocean, the conjugate
there tend to lie along the same E N E - W S W trends margin would later have been subducted beneath the
as the basement highs typifying the lower volcanic western edge of South America.
sequence. Late-stage faulting across the region with At some time thereafter, the final phase of vol-
the same trends would provide conduits for any canic extrusion began. Although re-extension was
molten material. Nevertheless the underlying dip- taking place in the new, rough crust, volcanic sources
ping reflectors beneath the volcanic constructs are were located further to the northwest, supporting the
indicative of their longevity and preclude a simple notion that the upper (and only sampled) volcanic se-
late-stage parasitic origin for the seamounts. quence stemmed from the drifting Caribbean plate's

crossing of the Galapagos hotspot (Duncan and Har- ACKNOWLEDGEMENTS

graves, 1984). At some short time after volcanic
extrusion had ended (at least in the SE Venezuelan This geophysical investigation in the Caribbean
Basin) the Caribbean plate became close enough to a region was supported by the National Science Foun-
source of voluminous turbidites that the thin, rough dation under grant OCE-93-02578. The manuscript
crust, along with the lower parts of what is now benefited greatly from reviews by R. Duncan and M.
the r o u g h - s m o o t h B" boundary became covered by Coffin. Lamont-Doherty Earth Observatory Contri-
these low-filling, stratified sediments. bution no. 5683.
The facts that in places, turbidite layering down-
laps on tilted, smooth B"-covered blocks, and that
both the turbidites and the thin stringers of smooth REFERENCES
B" material seem to directly overlie rough B", pro-
vide important constraints on the timing of the Abrams, L.J., Larson, R.L., Shipley, T.H. and Lancelot, Y., 1993.
Cretaceous volcanic sequences and Jurassic oceanic crust in
formation of thin rough B" basement, its exten-
the East Mariana and Pigafetta basins of the western Pacific.
sional wrench faulting, its overspreading by smooth AGU Geophys. Monogr., 77: 77-101.
B" volcanics, and the subsequent infilling of bathy- Anderson, D.L., Zhang, Y.-S. and Tanimoto, T., 1992. Plume
metric lows by (presumably terrigenous) turbidites. heads, continental lithosphere, flood basalts, and tomography.
Coring, by deep-sea drilling, will provide the sed- In: B.C. Storey, T. Alabaster and R.J. Pankhurst (Editors),
Magmatism and the Causes of Continental Break-Up. Geol.
iment and basement samples needed to answer the
Soc. London Spec. Publ., 68: 99-124.
few fundamental questions that stand between us Beets, D.J., Klaver, G., Beunk, F.F., Kieft, C. and Maaskant, E,
and full understanding of the tectonic history of the 1982. Picrites as parental magma of MORB-type tholeiites.
Caribbean basins. What is the age and origin of thin, Nature, 296:341-343.
rough B" basement? Was it formed thin, or has it Bercovici, D. and Mahoney, J., 1994. Double flood basalts and
plume head separation at the 600-kilometer discontinuity. Sci-
been subsequently extended? What is the age and
ence, 266: 1367-1369.
provenance of the sediments directly overlying it, i.e. Biju-Duval, B., Mascle, A., Montadert, L. and Wanneson, J.,
was rough B" formed in an open Pacific Ocean envi- 1978. Seismic investigation in the Colombia, Venezuela and
ronment? Could it have been produced in a back-arc Granada Basins, and on the Barbados Ridge for future IPOD
basin of the proto-Caribbean? What sediments are drilling. Geol. Mijnbouw, 57:105-116.
Bowland, C.L., 1993. Depositional history of the western Colom-
sandwiched between the smooth B" stringers and
bian Basin, Caribbean Sea, revealed by seismic stratigraphy.
the underlying rough B" crust? What are the ages Geol. Soc. Am. Bull., 105: 1321-1345.
and provenances of the thick turbiditic layers that Bowland, C.L. and Rosencrantz, E., 1988. Upper crustal structure
overran thin, rough B" and appear to have crossed of the western Colombian Basin, Caribbean Sea. Geol. Soc.
the r o u g h - s m o o t h B" boundary at some points? Our Am. Bull., 100: 534-546.
Case, J.E. and Holcombe, T.L., 1980. Geologic-tectonic map of
interpretation of the seismic reflection data predicts
the Caribbean region, scale 1:2,500,000. In: Miscellaneous
that the volcanic thickening of the Venezuelan Basin Investigations, Ser. Map I-1100, U.S. Geol Survey.
has been due primarily to the emplacement of ex- Chen, Y.J., 1992. Oceanic crustal thickness versus spreading rate.
trusives (therefore, in bottom-up sequence) and that Geophys. Res. Lett., 19: 753-756.
the final, youngest extrusive event had an extremely Christofferson, E., 1973. Linear magnetic anomalies in the
Colombia Basin, central Caribbean Sea. Geol. Soc. Am. Bull.,
large areal extent, spreading as thin, mobile flows
84: 3217-3230.
even out over the previously uninvaded rough B" Christofferson, E., 1976. Colombian Basin magmatism and
crust. Caribbean plate tectonics. Geol. Soc. Am. Bull., 87: 1255-
Testing of this hypothesis, and the refinement of 1258.
it, can be accomplished by drilling and sampling Clague, D.A., Holcombe, R.T., Sinton, J.M., Detrick, R.S. and
Torresan, M.E., 1990. Pliocene and Pleistocene alkalic flood
the easily accessed upper layer, but this must be
basalts on the seafloor north of the Hawaiian islands. Earth
done in several places. Much more of a puzzle, and Planet. Sci. Lett., 98: 175-191.
likely to remain so, is the nature and age of the Coffin, M.E and Eldholm, O., 1994. Large igneous provinces:
early-stage volcanics, imaged as a clearly definable crustal structure, dimensions and external consequences. Rev.
structural sequence buried by several kilometers of Geophys., 32: 1-36.
Coffin, M.F., Davies, H.L. and Haxby, W.F., 1986. Structure
late-stage extrusives and separated from it by sed-
of the Kerguelen Plateau province from Seasat altimetry and
imentary and/or volcaniclastic material, whose age seismic reflection data. Nature, 324:134-136.
and makeup is entirely unknown. Since the upper se- Colwell, J.B., Coffin, M.F., Pigram, C.J., Davies, H.L., Stagg,
quence appears to thicken westward, sampling (i.e. H.M.J. and Hill, EJ., 1988. Seismic stratigraphy and evolution
drilling) of the earlier structures might be more ef- of the Raggatt Basin, southern Kerguelen Plateau. Mar. Pet.
Geol., 5: 75-88.
fectively carried out to the east, far from any locale
Diebold, J.B., Stoffa, EL., Buhl, E and Truchan, M., 1981.
previously sampled by DSDP or ODP. Venezuelan Basin crustal structure. J. Geophys. Res., 86:
7901-7923.
Donnelly, T.W., Melson, W., Kay, R. and Rogers, J.J.W., 1973.

Basalts and dolerites of Late Cretaceous age from the Central Caribbean. GUA (Univ. of Amsterdam) Papers of Geology,
Caribbean. Init. Rep. DSDP, 15:989-1011. Ser. 1, No. 27, 168 pp.
Donnelly, T.W. and 10 others, 1990. History and tectonic setting Kolla, V., Buffler, R.T. and Ladd, J.W., 1984. Seismic stratigra-
of Caribbean magmatism. Decade of North American Geol- phy and sedimentation of Magdalena fan, southern Colombian
ogy, Vol. H, Geological Society of America, Boulder, CO, pp. Basin, Caribbean Sea. Am. Assoc. Pet. Geol. Bull., 68: 316-
339-374. 332.
Driscoll, N.W., Hogg, J.R., Karner, G.D. and Christie-Blick, Ladd, J.W. and Watkins, J.S., 1980. Seismic stratigraphy of the
N., 1995. Extensional tectonics in the Jeanne d'Arc basin: western Venezuela Basin. Mar. Geol., 35:21-41.
implications for the timing of break-up between Grand Banks Leroy, S., 1995. Structure et origine de la plaque Caraibe; im-
and Iberia. In: R.A. Scrutton, M.S. Stoker, G.B. Shimmield plications Geodynamiques. PhD Thesis, Univ. Paris IV, pp.
and A.W. Tudhope (Editors), The Tectonics, Sedimentation 95-32.
and Palaeoceanography of the North Atlantic Region. Geol. Lu, R.S. and McMillen, K.J., 1982. Multichannel seismic survey
Soc. London Spec. Publ., 90: 1-28. of the Colombia Basin and adjacent margins. In: J.S. Watkins
Duncan, R.A. and Hargraves, R.B., 1984. Plate tectonic evolution and C.L. Drake (Editors), Studies in Continental Margin Geol-
of the Caribbean region in the mantle reference frame. Geol. ogy. Am. Assoc. Pet. Geol. Mem., 34: 395-410.
Soc. Am. Mem., 162:81-94. Ludwig, W.J., Houtz, R.E. and Ewing, J.I., 1975. Profiler-
Echeverria, L.M., 1980. Tertiary or Mesozoic komatiites from sonobuoy measurements in Colombian and Venezuelan Basins,
Gorgona Island, Colombia: field relations and geochemistry. Caribbean Sea. Am. Assoc. Pet. Geol. Bull., 59:115-123.
Contrib. Mineral. Petrol., 73: 253-266. Matthews, J.E. and Holcombe, T.L., 1976. Regional
Edgar, N.T., Ewing, J.I. and Hennion, J., 1971. Seismic refraction geological/geophysical study of the Caribbean Sea (Navy
and reflection in Caribbean Sea. Am. Assoc. Pet. Geol. Bull., ocean area NA 9). Naval Oceanographic Office ref. Publ.
55: 833-870. RP 3, Washington, DC.
Eldholm, O. and Grue, K., 1994. North Atlantic volcanic mar- Maurrasse, E, Husler, J., Georges, G., Schmitt, R. and Damond,
gins: dimensions and production rates. J. Geophys. Res., 99: E, 1979. Upraised Caribbean sea floor below acoustic reflector
2955-2968. Btt at the southern peninsula of Haiti. Geol. Mijnbouw, 58:
Ewing, J., Antoine, J. and Ewing, M., 1960. Geophysical mea- 71-83.
surements in the western Caribbean Sea and in the Gulf of McCarthy, J., Mutter, J.C., Morton, J.L., Sleep, N.H. and Thomp-
Mexico. J. Geophys. Res., 65: 4087-4126. son, G.A., 1988. Relic magma chamber structures preserved
Frisch, W., Meschede, M. and Sick, M., 1992. Origin of the within Mesozoic North Atlantic crust? Geol. Soc. Am. Bull.,
Central American ophiolites: evidence from paleomagnetic 100: 1423-1436.
results. Bull. Geol. Soc. Am., 104:1301-1314. Muller, M.R., Robinson, C.J., Minshull, T.A., White, R.S. and
Ghosh, N., Hall, S.A. and Casey, J.E, 1984. Seafloor spreading Bickle, M.J., 1997. Thin crust beneath ocean drilling program
anomalies in the Venezuelan Basin. In: W.E. Bonini, R.B. borehole 735B at the Southwest Indian Ridge? Earth Planet.
Hargraves and R. Shagam (Editors), The Caribbean-South Sci. Lett., 148: 93-107.
American Plate Boundary and Regional Tectonics. Geol. Soc. Mutter, J.C. and Karson, J.A., 1992. Structural processes at
Am. Mem., 162: 65-80. slow-spreading ridges. Science, 257: 627-634.
Hagen, R.A., Mayer, L.A., Mosher, D.C., Kroenke, L.W., Ship- Mutter, J.C., Talwani, M. and Stoffa, E, 1982. Origin of seaward-
ley, T.H. and Winterer, E.L., 1993. Basement structure of the dipping reflectors in oceanic crust off the Norwegian margin
northern Ontong Java Plateau. Proc. ODE Sci. Results, 130: by 'sub-aerial seafloor spreading'. Geology, 10:131-134.
23-31. Planke, S., 1994. Geophysical response of flood basalts from
Hinz, K., 1981. A hypothesis on terrestrial catastrophes. Wedges analysis of wire line logs: Ocean Drilling Program site 642,
of very thick oceanward dipping layers beneath passive conti- Vcring volcanic margin. J. Geophys. Res., 99: 9279-9296.
nental margins. Geol. Jahrb., E22: 3-28. Planke, S. and Eldholm, O., 1994. Seismic response and con-
Hinz, K., Schluter, H.-U., Grant, A.C., Srivastava, S.E, Umpleby, struction of seaward dipping wedges of flood basalts: VCring
D. and Woodside, J., 1979. Geophysical transects of the margin. J. Geophys. Res., 99: 9263-9278.
Labrador Sea: Labrador to southwest Greenland. Tectono- Ramsay, D.C., Colwell, J.B., Coffin, M.E, Davies, H.L., Hill,
physics, 59:151-183. EJ., Pigram, C.J. and Stagg, H.M.J., 1986. New findings from
Hopkins, H.R., 1973. Geology of the Aruba Gap Abyssal Plain the Kerguelen Plateau. Geology, 14: 589-593.
near DSDP site 153. Init. Rep. DSDP, 15: 1039-1050. Reid, I. and Jackson, H.R., 1981. Oceanic spreading rate and
Houtz, R.E. and Ludwig, W.J., 1977. Structure of Colombia crustal thickness. Mar. Geophys. Res., 5: 165-172.
Basin, Caribbean Sea, from profiler-sonobuoy measurements. Rosendahl, B.R., Meyers, J., Groschel, H. and Scott, D., 1992.
J. Geophys. Res., 82: 4861-4867. Nature of the transition from continental to oceanic crust and
Jackson, H.R., Reid, I. and Falconer, R.K.H., 1982. Crustal the meaning of reflection Moho. Geology, 20:721-724.
structure near the Arctic mid-ocean ridge. J. Geophys. Res., Schlich, R., Rotstein, Y. and Schaming, M., 1993. Dipping
87: 1773-1783. basement reflectors along volcanic passive margins new
Keen, C.E. and Barrett, D.L., 1972. Seismic refractions studies insight using data from the Kerguelen Plateau. Terra Nova, 5:
in Baffin Bay: and example of a developing ocean basin. 157-163.
Geophys. J. R. Astron. Soc., 30: 253-271. Self, S., Thordarson, T., Keszthelyi, L., Walker, G.EL., Hon,
Keen, C.E. and Potter, D.E, 1995. The transition from a volcanic K., Murphy, M.T., Long, E and Finnemore, S., 1996. A new
to a nonvolcanic rifted margin off eastern Canada. Tectonics, model for the emplacement of Colombia River basalts as
14: 359-371. large, inflated pahoehoe lava flow fields. Geophys. Res. Lett.,
Kent, G.M., Detrick, R.S., Swift, S.A. and Collins, J.A., 1997. 23: 2689-2692.
Evidence from Hole 504B for the origin of dipping events Sinton, C.W., Duncan, R.A., Storey, M., Lewis, J. and Estrada,
in oceanic crustal reflection profiles as out-of-plane scattering J.J., 1998. An oceanic flood basalt province within the
from basement topography. Geology, 25: 131-134. Caribbean plate. Earth Planet. Sci. Lett., 155:221-235.
Klaver, G.T., 1987. The Curacao lava formation: an ophiolitic Skogseid, J. and Eldholm, O., 1989. Vcring Plateau continen-
analogue of the anomalous thick layer 2B of the mid-Cre- tal margin: seismic interpretation, stratigraphy, and vertical
taceous oceanic plateaus in the western Pacific and central movements. Proc. ODE Sci. Results, 104: 993-1030.

M U L T I C H A N N E L SEISMIC I M A G E S OF V O L C A N I C S T R U C T U R E S IN THE V E N E Z U E L A N BASIN 589
Spadea, E, Espinosa, A. and Orrego, A., 1989. High-Mg extru- Provinces in Other Parts of the World. Geol. Soc. India Mem.,
sive rocks from the Romeral zone ophiolites in the southwest- 3: 19-44.
ern Colombian Andes. Chem. Geol., 77: 303-321. Watkins, J. and Cavanaugh, T., 1974. Implications of magnetic
Srivastava, S.E and Keen, C.E., 1995. A deep seismic reflection anomalies in the Venezuelan Basin. Transactions of the 7th
profile across the extinct Mid-Labrador Sea spreading center. Caribbean Geological Conference, Guadalupe, pp. 129-138.
Tectonics, 14: 372-389. Weidmann, J., 1978. Ammonites from the Curacao Lava Forma-
Srivastava, S.E and Roest, W.R., 1995. Nature of thin crust tion, Curacao, Caribbean. Geol. Mijnbouw, 57:361-364.
across the southwest Greenland margin and its bearing on Wildberg, H., 1983. Die Magmatite des Nicoya-Komplexes,
the location of the ocean-continent boundary. In: E. Banda, Costa Rica, Zentralamerika. Dissertation, Wilhelms-Univ.,
M. Torne and M. Talwani (Editors), Rifted Ocean-Continent Munster, 168 pp.
Boundaries. NATO ASI Ser. C, Vol. 463, pp. 95-120. Wildberg, H., 1984. Die Nicoya-Komplex, Costa Rica, Zen-
Stoffa, EL., Mauffret, A., Truchan, M. and Buhl, E, 1981. tralamerika: Magmatismus und Genese eines polygenetischen
Sub-B tl layering in the southern Caribbean: the Aruba Gap Ophiolith-Komplexes. Munster. Forsch. Geol. Paleontol., 62:
and Venezuela Basin. Earth Planet. Sci. Lett., 53:131-146. 123.
Talwani, M., Windisch, C., Stoffa, EL., Buhl, E and Houtz, R.E., White, R.S. and McKenzie, D., 1995. Mantle plumes and flood
1977. Multichannel seismic study in the Venezuelan Basin and basalts. J. Geophys. Res., 100: 17,543-17,585.
the Curacao Ridge. In: Island Arcs, Deep Sea Trenches and White, R.S., McKenzie, D. and O'Nions, R.K., 1992. Oceanic
Back-Arc Basins. AGU, Maurice Ewing Series, I, pp. 83-98. crustal thickness from seismic measurements of rare earth
Tarduno, J.A., Sliter, W.V., Kroenke, L., Leckie, M., Mayer, H., element inversions. J. Geophys. Res., 97: 19,683-19,715.
Mahoney, J.J., Musgrave, R., Storey, M. and Winterer, E.L., Whitmarsh, R.B., Miles, ER. and Mauffret, A., 1990. The
1991. Rapid formation of Ontong Java Plateau by Aptian ocean-continent boundary off the western continental mar-
mantle plume volcanism. Science, 254: 399-403. gin of Iberia, I. Crustal structure at 40~ Geophys. J. Int.,
Waters, A.C., Myers, C.W., Brown, D.J. and Ledgerwood, R.K., 103: 509-531.
1981. Colombia Plateau with special emphasis on the basalt Zehnder Mutter, C., 1992. DSDP Hole 504B: are gabbros in
stratigraphy of the Pasco Basin. In: K.V. Subbaro and R.N. sight? JOI/USSAC Newsl., 5: 4, 5, 12.
Sukheswala (Editors), Deccan Volcanism and Related Basalt

Chapter 20
Tectonic and Stratigraphic Development of the Eastern Caribbean"

New Constraints from Multichannel Seismic Data
N E A L W. D R I S C O L L and J O H N B. D I E B O L D
New high-resolution multichannel seismic reflection data collected in the eastern Caribbean during R/V Ewing cruise 9501
imaged both the crustal structure and overlying stratigraphic successions. On the basis of these new MCS data, we define the
geologic development of the Beata Ridge and Venezuelan Basin. The proto-Caribbean crust was formed by seafloor spreading
in Late Jurassic-Early Cretaceous time. Prior to the Senonian, widespread and rapid eruption of basaltic flows began in concert
with extensional deformation of the proto-Caribbean crust. Large divergent volcanic wedges observed along the rough-smooth B"
boundary are coincident with the abrupt shoaling of Moho and appear to be bounded by a large northwest-dipping fault system.
The location of the major extensional deformation migrated through time from the Venezuelan Basin to the western flank of
the Beata Ridge. Extensional unloading of the Beata Ridge footwall caused uplift and rotation of the ridge and collapse of its
hangingwall (i.e., Hess Escarpment). Sediment thicknesses and stratal geometry observed across the Venezuelan Basin and Beata
Ridge suggest that the majority of the deformation in this region occurred soon after the emplacement of the volcanics. Minor
fault reactivation in the Neogene along the eastern flank of the Beata Ridge is associated with an accommodation zone (i.e.,
tear fault) that records a change in the deformation style from bending and subduction of the Caribbean Plate along the Muertos
Trough south of Puerto Rico to compressional deformation and obduction of the Caribbean Plate south of Hispaniola. We propose
that this difference in deformational style is, in part, a consequence of the thicker crust on the Beata Ridge, which is more resistant
to subduction.
Coincident with the rough-smooth crustal boundary in the Venezuelan Basin is a marked change in sediment thickness. The
stratal geometry and spatial distribution of the basal sequence suggest that the Late Cretaceous to Early Miocene sediments are
terrigenous deposits derived from South America. We propose that during the Late Cretaceous to Early Miocene the dominant
drainage in South America was a northward-flowing axially parallel fluvial system that drained the Maracaibo-Peruvian foreland
basin east of the Andes and supplied sediment to the Venezuelan Basin. The middle Tertiary uplift and deformation along the
northern South America Plate boundary blocked this axial-parallel fluvial network. As a result of the blockage, the axial-parallel
rivers were dammed or diverted eastward. This blockage, together with the renewed uplift of the Central and Eastern Cordillera,
supplied abundant sediment to the developing foreland basin and other paleo-structural lows to the east. The consequent regrading
of the fluvial systems as the foreland basin was filled allowed drainage systems to flow east across the South American continent
(e.g., Amazon and Orinoco) and deliver sediment to the Atlantic Ocean.
The seismic reflection data also imaged an Eocene-Early Miocene current-controlled drift deposit which reflects the
movement of bottom currents from the eastern Pacific to the Caribbean during this period. The gradual shoaling of the Central
American Isthmus in Late Oligocene-Early Miocene time closing the gateway is consistent with the diminished occurrence of
current-controlled features upsection and paleo- deep and intermediate water geochemistry in DSDP cores.
INTRODUCTION N o r t h and S o u t h A m e r i c a ( W i l s o n , 1966; M a l f a i t

and D i n k e l m a n , 1972; P i n d e l l and D e w e y , 1982;
T h e h i s t o r y o f the C a r i b b e a n ' P l a t e ' is m u c h de- B u r k e et al., 1984; D u n c a n and H a r g r a v e s , 1984;
b a t e d w i t h the m a i n c o n t r o v e r s y c e n t e r i n g a r o u n d P i n d e l l and Barrett, 1990). In the 'fixist' m o d e l , the
the t e c t o n i c r e l a t i o n s h i p o f the C a r i b b e a n to the p r e s e n t plate d e v e l o p e d m a i n l y in situ and experi-
s u r r o u n d i n g N o r t h A m e r i c a n and S o u t h A m e r i c a n e n c e d o n l y m i n o r C e n o z o i c m o t i o n w i t h r e s p e c t to
plates. In the ' m o b i l i s t ' m o d e l , the C a r i b b e a n P l a t e the A m e r i c a s ( K l i t g o r d and S c h o u t e n , 1987; D o n -
is p u r p o r t e d to h a v e f o r m e d in the e a s t e r n Pacific nelly, 1989). T h e d e b a t e c o n t i n u e s , in large part,
and s u b s e q u e n t l y m o v e d e a s t w a r d w i t h r e s p e c t to b e c a u s e o f the structural c o m p l e x i t i e s b o t h o n s h o r e

592 N.W. DRISCOLL and J.B. DIEBOLD
and offshore, the limited high-quality marine data account for both the origin of the faulted and highly
(e.g., seismic, drilling, and magnetic data), and, most attenuated crust in the Venezuelan Basin and the
importantly, the fact that few studies have integrated deformation across the Beata Ridge and surrounding
the wealth of onshore and offshore data to establish region.
an internally consistent model for the development Because of the limited constraints, the timing and
of the Caribbean region. origin of the tectonic deformation across the Beata
A related and equally puzzling problem is the Ridge and surrounding region also remains contro-
crustal thickness variations observed across the versial. Recent studies have proposed that a renewed
Caribbean Plate, ranging from 4 km in the Venezue- phase of deformation occurred in the Miocene and
lan Basin to 12 km across the Beata Ridge. The ori- that deformation is still ongoing in the region of the
gin of the thickened oceanic crust in the Caribbean Beata Ridge (Mauffret and Leroy, 1997; Mauffret
and its relation to the tectonic development of the and Leroy, Chapter 21). In the Mauffret and Leroy
Caribbean Plate remains poorly understood. Duncan model, the Colombian Plate is presently overthrust-
and Hargraves (1984), using a plate tectonic model ing the Venezuelan Plate with the thrust front being
tied to the fixed hotspot reference frame, proposed located along the eastern flank of the Beata Ridge
that the Caribbean Plate was located over the initia- (Mauffret and Leroy, 1997). Such a two-plate kine-
tion of the Galapagos mantle plume during the Late matic model for the Caribbean was first proposed
Cretaceous. The excess volcanism associated with by Dewey and Pindell (1985), and requires differen-
the mantle plume increased the oceanic crustal thick- tial motion between the Venezuelan and Colombian
ness and formed the Caribbean Cretaceous basalt basins. This plate kinematic model predicts that de-
province. Subsequent northeastern movement of the formation across the Beata Ridge accommodated the
Farallon Plate brought the resultant oceanic plateau differential motion between the eastern and west-
from the Pacific into the Caribbean. In this model, ern Caribbean plates (Fig. 1). Earlier studies also
the remnant hotspot trail was consumed by plate proposed that the Beata Ridge was uplifted after
subduction at the Central American arc. However, the Eocene (i.e., post-A") on the basis of seismic
although the Galapagos hotspot scenario explains reflection data (Fox et al., 1970). An alternative hy-
the unusual thick crust in the Caribbean, it fails to pothesis, the single-plate model for the Caribbean,
Fig. 1. Map of the Caribbean region showing the plate boundaries with North and South America and the basalt provinces onshore
and offshore. The dark pattern in the Caribbean Sea corresponds to the mapped extent of seismic horizon B" in the eastern half
(Venezuelan Basin) with a presumed western extension (Hess Escarpment and southern Nicaraguan Rise). Black patches on land are
obducted fragments of the basalt province. For example, Nicoya (Costa Rica), Isla Gorgona (Colombia), Dumisseau Formation (Haiti),
and Curaqao. Drilled Sites 146, 150, 151, 152, and 153 of DSDP Leg 15 (solid circles) and ODP Site 999 and 1001 are also shown.

TECTONIC AND STRATIGRAPHIC DEVELOPMENT OF THE EASTERN CARIBBEAN 593
Fig. 2. Track map for EW9501 MCS data in the eastern Caribbean. Gray circles denote locations of sonobuoysalong track.
purports that the majority of the deformation ob- boundary. Because the high-quality seismic reflec-
served across the Beata Ridge and Venezuelan Basin tion data imaged the preserved stratal geometry of
occurred early in the history of the Caribbean (i.e., the basal stratigraphic sequences, we were able for
Late Cretaceous) prior to large sediment accumula- the first time to decipher the depositional processes
tion. Minor fault reactivation along the eastern Beata and sediment sources responsible for their forma-
Ridge and Venezuelan Basin is inferred to have tion. Using these new insights, we re-interpreted the
been caused by the different styles of deformation older seismic reflection data, in which the basal se-
in response to the north-south compressional stress quences are either transparent or highly distorted, to
(Holcombe et al., 1990; Driscoll et al., 1995). establish a more regional framework for the deposi-
In this paper, we use recently acquired high-res- tional systems in the Venezuelan Basin (Biju-Duval
olution multichannel seismic reflection data (MCS) et al., 1978; Ladd and Watkins, 1980; Stoffa et al.,
collected onboard the R/V Ewing from the eastern 1981; Diebold et al., 1981).
Caribbean (Fig. 2), together with onshore geologic
information from South America, to constrain the
timing and nature of the tectonic deformation across GEOLOGIC DEVELOPMENT OF THE VENEZUELAN
the Beata Ridge and Venezuelan Basin. This analy- BASIN
sis provides critical constraints on the history of the
relative motion of the Caribbean Plate and enables The Venezuelan Basin is bounded to the east and
us to discern whether the azimuthal and deformation west by the Aves Rise and Beata Ridge, respectively
data can be better explained by a one- or two-plate (Figs. 1 and 2). The internal northeasterly trending
model. The emplacement history of the Caribbean grain of the Venezuelan Basin, defined by appar-
Cretaceous basalt province and its relationship to ently linear magnetic anomalies (Donnelly, 1973),
the Caribbean tectonic deformation is discussed in basement faulting, and a marked boundary between
more detail in Diebold et al., Chapter 19. Using two distinct crustal types (smooth and rough) with
this data set, we also are able to define the strati- little to no gravity signature (Case and Holcombe,
graphic evolution of the Venezuelan Basin and its 1980; Case et al., 1990a), is sub-parallel to the Beata
relationship to the Caribbean-South American Plate Ridge. Early seismic refraction results (Officer et

,.~ azl
.,..a .,.a
.,...
O
z~ .,..a
r.tz
. ,....~ 0.~
r.gz
~<
,.t::
r ,.t::
c~o
. ,....~ r.~
. ,.... 0.)
,...-
0.,)
e~o
.,.~
r~
a9 ;~
~ ,...~
r
. ,...~
.,..a
e,.9
"" ,x:

al., 1957, 1959) showed that crustal structure in the correlates with the interface between unconsolidated
Venezuelan Basin is unusual. It comprises two dis- Miocene and Eocene oozes and consolidated Lower
tinct layers, and is thicker than normal oceanic crust Eocene cherts and chalks (Edgar et al., 1973a,b;
but thinner than typical continental crust. Later work Holcombe et al., 1990). However, the DSDP and
(Ewing et al., 1971; Edgar et al., 1971) revealed sim- ODP sites are located on structural highs and thus
ilar, but even thicker, crustal structure in the Colom- failed to sample the basal onlapping sedimentary
bian Basin. Analysis of closely spaced airgun- sequence between A" and B". Furthermore, the
sonobuoy wide-angle reflection and refraction pro- transparent acoustic character of this basal sequence
files suggested that the base layer in both basins was in the existing seismic reflection data across the
geophysically similar to 'normal' oceanic crust, but Venezuelan Basin makes it difficult to infer the de-
that the outer rind of the upper layer had a much positional processes responsible for its formation.
smoother appearance than oceanic crust imaged in The A" to B" sediment interval is thickest in the
other reflection profiles (Ludwig et al., 1975; Houtz southeastern portion of the Venezuelan Basin and
and Ludwig, 1977). This horizon was named B" and systematically decreases toward the east, north, and
was inferred to be the top of a volcanic sequence. west. Similar basal onlapping packages also have
The source of the B" has been attributed to the been observed in structural lows in the Colombian
Galapagos hotspot (Duncan and Hargraves, 1984). Basin (Edgar et al., 1971; Lu and McMillen, 1982;
By the early 1970s, DSDP drilling in the central Kolla et al., 1984; Bowland and Rosencrai~tz, 1988;
Caribbean had penetrated B" and sampled Creta- Bowland, 1993). Similar to the A" to B" sediment
ceous intrusives (Edgar et al., 1973a,b; Donnelly, interval, the stratigraphic sequence overlying A" in
1973). Despite the numerous dating problems, there the Venezuelan Basin thickens dramatically toward
are several firm ages for the locally youngest age of the southeast displaying a pronounced increase in
the Cretaceous basalts and thus, the termination of thickness across the rough-smooth B" basement
magmatic activity. As reviewed in Donnelly et al. boundary. The acoustic character changes markedly
(1990) these dates are mainly derived from dating across horizon A" in the Venezuelan Basin: from
sedimentary units within and on the basalt complex, laminated below the horizon to more chaotic and
including DSDP Leg 15 sites. These ages cluster hummocky above it. This change in acoustic char-
at the latest Turonian, but in a few cases could acter is attributed to the onset of current-controlled
be as young as Santonian, and even early Campa- deposition in the Caribbean (Driscoll et al., 1995).
nian (DSDP Site 152 and ODP Leg 165). Detailed In light of the complex tectonic and stratigraphic
radiometric study of basalts from Gorgona, Costa history of the Beata Ridge and Venezuelan Basin and
Rica, Haiti, and Curacao undertaken recently us- the possibility that more than one magmatic event
ing 4~ incremental heating methods, yielded was responsible for the thickened crust, we collected
dates consistent with the sedimentary ages for the a series of MCS lines across and along the rough-
termination of the event (Sinton et al., 1993). The smooth B" boundary to determine the tectonic, mag-
number and geographic extent of these age determi- matic, and stratigraphic development of the region
nations at 90-88 Ma suggests that a truly vast ig- and to define the complex interplay between South
neous event occurred at this time (i.e., the Caribbean American deformation, evolving drainage patterns,
Cretaceous basalt province). and the consequent sediment supply to the Caribbean
Velocity analysis and laboratory experiments sug- Plate through time. During cruise 9501 of R/V Ew-
gested that the sub-B" interval observed in the multi- ing in February and March, 1995, we collected over
channel seismic data is principally igneous (Ludwig 5700 km of MCS data and deployed 104 successful
et al., 1975). Estimates of sub-B" velocities are vari- sonobuoys in the Caribbean (Fig. 2). A 20-air-
able from place to place, a variation that may be gun, 8415 in 3 array and a 160-channel hydrophone
indicative of intercalated volcanic clastic sediments streamer were used to acquire the MCS data.
(e.g., Abrams et al., 1993). Smooth B" is not ubiq-
uitous in the Caribbean and the boundary between
rough and smooth B" crust in the Venezuelan Basin RESULTS
was discovered by MCS profiling (Biju-Duval et al.,
1978; Diebold et al., 1981; Stoffa et al., 1981). In Acoustic b a s e m e n t
more cases than not, the boundary is an abrupt one
in this basin, accompanied by changes in basement We will present the MCS data from the Venezue-
elevation and, sometimes, apparent faulting. lan Basin and Beata Ridge from east to west and
A highly reflective horizon observed within the discuss the observations beginning with basement
sedimentary packages overlying B" was named hori- structure for each cross-section.
zon A". Examination of the seismic reflection and Seismic line 1294, located along the eastern edge
DSDP and ODP drill data indicates that horizon A" of the Venezuelan Basin, trends sub-parallel to the

Aves Ridge (Fig. 3). The rough-mooth B" basement boundary is not marked by an escarpment on line
boundary, which also delineates the abrupt increase 1302. In this region, the smooth B" basement sur-
in sediment thickness below reflector A", occurs face is made up of several en-echelon reflectors
near CDP 9000 (Fig. 3). Continuing farther west, that commonly downlap onto the underlying surface,
line 1293 again illustrates the abrupt boundary be- suggesting multiple flows (Diebold et al., Chapter
tween the rough and smooth B" acoustic basement 19). Along composite line 1304/1317 the rough-
(Fig. 4). The rough-smooth B" boundary occurs at smooth B" boundary occurs across a pronounced
approximately CDP 19,000 and is associated with a basement high (Fig. 7), a high which appears to be
reversal of dip near the boundary and a sharp shoal- a volcanic edifice similar to those observed along
ing of Moho (see fig. 2, Diebold et al., Chapter 19). line 1293 (Fig. 4). This high, however, is not asso-
The change in dip is local (~10 km, Fig. 4). Brit- ciated with underlying divergent reflectors. Graben
tle deformation and fault rotation near the rough- structures that offset smooth B" basement also are
smooth B" basement boundary is the most likely observed in this region.
cause for this reversal of dip. The fault system that Sub-B" divergent reflectors beneath the struc-
controls the geometry and structure of the dip re- tural high near DSDP Site 150 indicate that the
versal must be relatively surficial and antithetic to basalts recovered at Site 150 are only representative
a larger normal fault system that dips north-north- of the late-stage evolution of the Cretaceous basalt
west and controls the location and geometry of the province. Correlation of seismic reflection line 1320
large divergent wedge observed beneath B" (CDPs to DSDP Sites 146 and 150 suggests that B" corre-
10,500-19,000) and the large step in Moho topogra- lates with Late Cretaceous dolerites (Fig. 9; Edgar
phy (CDP ~ 19,000). et al., 1973a,b). The dolerite recovered at DSDP
The divergent wedge beneath the smooth B" base- Site 146 is a sill within Turonian limestone. Beneath
ment has interval velocities on the order of 5 km/s the Turonian limestone another 7 m of dolerite was
(Fig. 4; Diebold et al., Chapter 19). Similar reflector penetrated with fairly high recovery (Edgar et al.,
geometries and velocities on numerous other rifted 1973a,b). At DSDP Site 150, diabase also was re-
margins have been attributed to large igneous events covered at the base of the hole, with approximately 4
at, or immediately prior to, the cessation of con- m of penetration and again good recovery. Rough B"
tinental rifting and the onset of seafloor spreading basement has yet to be sampled and its relationship
and are termed seaward-dipping reflectors (SDRs; to smooth B" basement remains conjectural (Diebold
Hinz, 1981; Mutter et al., 1982; Schlich et al., 1993). et al., 1981).
The divergent geometries and spatial distribution of Analyses of basalts recovered at ODP Site 1001
the reflector packages across the Venezuelan Basin on the Hess Escarpment indicate that the Caribbean
suggest multiple igneous events. The rotation and oceanic plateau was active until at least 77-76
divergence of the onlapping seismic reflectors di- Ma (mid-Campanian). Benthic foraminiferal assem-
minishes upsection and is indicative of differential blages together with the vesicular nature of the
subsidence and block rotation (Fig. 4). The spatial basalts suggests that emplacement on the Hess Es-
extent of each successive divergent reflector package carpment occurred at shallow water depths followed
increases within the wedge. by rapid subsidence (Sigurdsson et al., 1997). The
The rough-smooth B" basement boundary is also mid-Campanian volcanism is consistent with the
delineated by a local reversal in dip on line 1300 Duncan and Hargraves (1984) model that volcan-
(Fig. 6). Horizon B" is offset by a number of small ism associated with the Galapagos hotspot continued
graben structures centered around CDP 1500 and from 100 to 75 m.y.B.P, in the Caribbean. Be-
9500 (Fig. 6). Graben formation appears to have cause Site 1001 failed to recover Senonian and older
occurred soon after the emplacement of the basalt basalts (88 Ma), the results from the site cannot
flows and prior to the onset of significant pelagic be used to determine whether or not rapid outpour-
sedimentation. Although the overlying pelagic de- ing across the Caribbean Plateau occurred during
posits mirror the underlying topography, there is Senonian time (90-88 Ma).
no evidence of growth faulting observed with these Farther west, seismic line 1321 that crosses the
structures. Similar to line 1293, the late-stage di- western Venezuelan Basin and the Beata Ridge il-
vergent reflectors are located in close proximity to lustrates the highly faulted nature of the region
the rough-smooth B" basement boundary, the lo- (Fig. 10). The western flank of the Beata Ridge is
cal reversal of dip (i.e., fault escarpment), and the bounded by a large westward-dipping normal fault.
shoaling of Moho topography. The basement offset across the normal fault is ap-
The rough-smooth B" basement boundary varies proximately 5 s two-way travel time (TWT; ~3750
dramatically along strike from a well defined escarp- m). The Beata Ridge is markedly asymmetric with
ment to a more subdued gradual boundary (see the western scarp having more offset and a steeper
Diebold et al., Chapter 19). The rough-smooth slope. The uplift, rotation and wavelength of the

6000 -
7000 -
'= 8000 -
9000 -
10000" I I i I i
1293
5000 - 1700 17(~ 1700 17001700 1700 1700 1700 1700 17oo, ,, IToo , I~oo
nAo I~9 +Mo I ~ o Ioo0 leSO 1o8o IOOO leOO 1o8o 18~o 1880 1880 I_~)
'ooo 2000
9 2000 2000 2000 20002000 2000 2000 2000 2000
~300 2000 2000 2000
24O0
6000 -
A ll
300 ~ 3 3 0 0 _2___ 5 ~.,,. 0 0 33003300 3300 3300
311100 Ik~OO
S300
7000 -
S300
~ 8000 -
63O0
9000 -
10000 -- ! ~ ! I !
16000 21000 213000
GDP #
Fig. 5. Velocity model derived from sonobuoy data collected during EW9501 used to process MCS line 1293. The lateral velocity
gradient observed within the B" to A" interval coincides with the region where the basal terrigenous sediments pinch out by onlap against
the escarpment.
Beata Ridge topography suggest that extensional un- sections and their occurrence diminishes upsection
loading and the consequent flexural rebound of the (Fig. 10).
lithosphere play an important role in its formation. Extension within what is now the Beata Ridge
Along the eastern flank of the Beata Ridge are a se- and along its western flank is focused in the north,
ries of half-graben structures and the basal reflectors and broadens to the south. Neglecting the possible
in the graben dip and thicken toward the west. Note effects of N-S compression along the northern edge
the small seamounts/mounds on top of B" toward of the plate (Muertos Trough, southern Hispaniola)
the southeastern portion of the line. The normal the distribution and number of faults as well as
fault located at CDP ~ 12,000 is spatially coincident the heave across any one given fault explains the
with a small mound suggesting that these volcanic architecture of Beata Ridge: high and narrow in
constructions might be zones of weakness that focus the north, wide and low in the south (Fig. 2). In
the extensional deformation (Fig. 10). The onlap of fact, toward the south, the eastern flank of the ridge
B" to A" sequence onto the rotated and deformed is more dissected by N-S-trending grabens. Crust
hangingwall blocks indicates that deformation in this at the foot of the Beata Ridge escarpment, in the
region occurred soon after magmatic emplacement Haiti Basin, appears to have been thinned by normal
and prior to substantial sediment accumulation. To- faulting and block tilting (Fig. 10), a thinning which
gether with the basal onlap, the differential thickness has reached its maximum immediately south of Haiti
of the B" to A" sediment interval across the faults (Fig. 2).
attests to the existence of the fault at the time of
deposition (Fig. 10). In regions where the subbottom B" to A" sedimentary sequence
reflectors are not distorted by the rough seafloor,
sub-B" reflectors are well imaged along the eastern The sedimentary succession overlying B" is sep-
flanks of the Beata Ridge (e.g., CDP 21,000). West arated by two prominent reflectors, horizon A"
of the Beata Ridge are thick ponded turbidites with and eM. Where sampled, horizon A" correlates
minor growth faults offsetting the lower stratigraphic with the interface between unconsolidated Miocene

to Eocene oozes and consolidated Lower Eocene formation appears to have been complete soon after
cherts and chalks and horizon eM correlates with the emplacement of the basalt flows and prior to the
the boundary between Early Miocene radiolarian onset of significant pelagic sedimentation. Although,
ooze and Early to Middle Miocene calcareous ooze the overlying pelagic deposits mirror the underlying
(Edgar et al., 1973a,b; Holcombe et al., 1990; Sig- topography, there is no evidence of growth fault-
urdsson et al., 1997). A marked change in acoustic ing observed with these structures. The stratigraphy
character also occurs across horizon A", from well overlying B" in line 1300 exhibits the same evolution
laminated below the horizon to more lenticular and from a well laminated interval between horizons B"
hummocky immediately above it. The hummocky and A" to a more acoustically hummocky character
character persists throughout the overlying sequence above A".
(A"-eM), culminating just above horizon eM. The B" to A" sediment thickness diminishes to-
Seismic line 1294, located along the eastern edge ward the west (Figs. 4, 6 and 7). In addition, along
of the Venezuelan Basin, trends sub-parallel to the the southward-dipping slope separating smooth and
Aves Ridge (Fig. 3). The abrupt increase in sediment rough B" basement, the B" to A" sediment inter-
thickness below reflector A", near CDP 9000, is spa- val pinches out completely near CDP 9500 on line
tially coincident with the rough-mooth B" basement 1304/1317 with thicknesses increasing away from
boundary (Fig. 3). Farther west, the thickness of the this region to both the south and north. The unit
B" to A" sediment interval systematically increases appears to thin predominantly by downlap and onlap
toward the south along line 1293 with a marked rather than by truncation suggesting that the hia-
increase also occurring across the rough-smooth tus is recording non-deposition rather than erosion
boundary (Fig. 4). Fig. 5 shows the smoothed in- (Fig. 7). Farther west on line 1320 (Fig. 2), DSDP
terval velocity field derived from sonobuoy data Site 150 is located at CDP 4900 in an area where
collected during EW9501 and was used in con- the B" to A" sediment interval thins updip across
junction with stacking velocities to process MCS the structural high by onlap, again suggesting that
line 1293. The lateral velocity gradient within the the hiatus in the B" to A" sequence is recording
B" to A" interval coincides with the region where non-deposition rather than erosion (Fig. 8).
the basal terrigenous sediments pinch out by on- At DSDP Site 146, radiolarian sandy turbidites
lap against the escarpment separating smooth and occur in the Campanian and Maastrichtian and ap-
rough B" basement. On the basis of velocity infor- pear to be locally derived. Late Paleocene to Early
mation, the maximum thickness of this sequence is Eocene zeolitic brown clays and marls recovered at
approximately 600 to 800 m and its distribution is DSDP Site 150 show characteristics of terrigenous
roughly coincident with the area of rough basement mineralogy. For example, the mica, kaolin-chlorite
in the Venezuelan Basin. This lateral velocity gra- and minor corroded plagioclase could be derived ei-
dient displayed in Fig. 5 was also recognized by ther from South America or from an increase in ash
Talwani et al. (1977). The velocity increase from input during this time (Edgar et al., 1973a,b). Mi-
2.4 km/s to 3.2 km/s that accompanies the lateral nor amounts of Paleogene clays were also recovered
change in thickness also implies the existence of from ODP Site 1001 located along the eastern flank
a facies boundary between pelagic and terrigenous of the Hess Escarpment, which were interpreted to
sediments (Fig. 5). The higher velocities were not be of terrigenous origin on the basis of chromium
consistent with the velocities determined for A" by (Cr) and titanium (Ti) (Sigurdsson et al., 1997).
DSDP Leg 15 drilling (Edgar et al., 1973a,b), which Furthermore, Cr and Ti analysis of the Site 1001
forced Talwani et al. (1977) to identify a new re- cores suggests that there is a quantifiable change
flector, 'horizon X'. Despite the velocity difference, in sediment provenance from the Paleocene-Eocene
they did recognize that horizon X might actually and the Miocene and Pleistocene (Sigurdsson et al.,
be chronostratigraphically equivalent to horizon A". 1997).
Nevertheless, at this early stage of interpreting seis-
mic reflection data, they placed more weight on A" to eM sedimentary sequence
correlating surfaces with similar velocities than try-
ing to determine why certain seismic units exhibited The hummocky acoustic character above A",
lateral velocity variations (Talwani et al., 1977). which has been recognized by previous researchers
Compaction, small readjustments caused by sed- and mapped across the Venezuelan Basin (Talwani
iment loading, and minor fault reactivation have et al., 1977), is a time transgressive boundary that
caused horizon A" to be offset across the rough- progressively shoals through time from south to
smooth B" basement boundary in seismic line 1300 north. For example in line 1294, the boundary be-
(Fig. 6). A number of small graben structures are ob- tween well laminated horizons and the hummocky
served across the region, two of which are centered acoustic character occurs at approximately 6.8 s
around CDP 1500 and 9500 (Fig. 6). The graben two-way travel time at CDP 2000 in the south, is

Fig. 6. Interpreted seismic reflection line 1300 showing that the abrupt shoaling of Moho is spatially coincident with the rough-smooth boundary. Northwest is loward the left. See Diebold et al., Chapter 19 for the uninterpreted
section.

Fig. 7. Interpreted composite seismic reflection lines 1304 and 1317 illustrating the complex nature of the rough-smooth boundary in this region. Note that the B" to A ~' sedimentary sequence is thinning towards the west away from
lines 1293 and 1300. Northwest is toward the left.

Fig. 8. Interpreted seismic reflection line 1320 illustrating the style of sediment thinning (truncation versus onlap-downlap) from Site 146 to Site 150. Note the sub-B" dipping reflectors near Site 150. North is toward the left.

EW9501 Line 1320 CDP #

Site 146/149 9000 8000 7000 6000
9_ - - _ - _ . . - -
._-_--_---
..--.-
- - - -
- I I I I I I
.--_.-._-- - , - - ~
._.- -.-_-- - - - -
. --_-_--..- . , - - - .-
. -_-_-- - . - - -
, - -- ---- - . - - . - truncation
' -
, -
--
-
-,.-
-...-
..- -
-..-.-.-
-_-
Site 150
. . . . 1- " l r '1- . r
. . . . . 9r T T ' r
. . . . /-.I-.I-.I-
. . . . . - r " r - r "Ir
. . . . 91- '1- '1- ~r
9 9 9 9 I''1-'I'-i-
- : - - : - ' : " ~ '1" '1- " r 'T
LA..A...A. 1- ' I " "1" '1"
. . . . '1" '1" " r ' ~
I~K_A..A, "r "r "r T
. . . .
9 . .
" r " r - r ~r
, '1-'l-'1r'~r
-6
~ 1--r -r "r
9' r " r " r "It
I...,_.A_~ 1" "It" " r - r A A A
~..A..A_.~
- r - r " r ~r
' r "1" " r "t"
91" " r '1- ~r
A -1r
^ ^ ^ ^ , , ^ ,
91- - r - r - r
,_.A_,_.~ 1" " r "1" T
-r "r "It -r
L_,~.I_.~ 1- " r T " r
"r "r -r "f
A Ak &k
,& ,A ,~
-7
-8
~ limestone r ~ oozeSilice~ ~ clay
chalk ~ calcareous D basalt

ooze
Fig. 9. Correlation of seismic reflection line 1320 to DSDP Sites 146 and 150 indicates that Btt correlates with the top of Late Cretaceous
dolerites. Horizon Att correlates with the interface between unconsolidated Miocene to Eocene oozes and consolidated Lower Eocene
cherts and chalks and horizon eM correlates with the boundary between Early Miocene radiolarian ooze below and Early to Middle
Miocene calcareous ooze above. The change from siliceous to carbonate ooze, together with the diminished effects of currents on the
depositional regime after the Early Miocene, are consistent with the shoaling of the Panamanian Isthmus.
exposed at the seafloor near CDP 9200, and becomes to 1304) and exhibits a more hummocky acoustic
quite dramatic by CDP 11,500 in the north (Fig. 3). character (Fig. 6). The A" to eM sediment inter-
Consequently, this horizon has no time stratigraphic val in line 1300 continuously thickens toward the
significance and is, in fact, only recording the com- southeast, in contrast to lines 1293 and 1294, where
plex interplay between gravity flows and geostrophic this interval shows signs of thinning both towards
bottom currents. This change in acoustic character the northwest and southeast. The eastern Caribbean
from laminated below to acoustically hummocky paleo-structure during Eocene time appears to have
above is also observed in the Colombian Basin controlled, to some degree, the thickness variations
(Lu and McMillen, 1982). Similar observations have of the A" to eM interval. For example, the thinning
been made in other ocean basins where gravity towards the southeast Venezuelan Basin observed in
flows interact with abyssal circulation (Driscoll and lines 1293, 1294, and 1296 might reflect basement
Laine, 1996). In the region where the hummocky relief at the time of deposition (Fig. 2). Along line
acoustic character is observed on the seafloor in 1297 (Fig. 2), which trends northeast-southwest,
the MCS data, the 3.5 kHz precision depth records both the B" to A" and the A" to eM sediment inter-
indicate that this acoustic character correlates with vals thicken markedly away from the basement high
sediment waves having amplitudes of tens to hun- (i.e., western edge of the Aves Rise). As previously
dreds of meters and wavelengths on the order of a mentioned, where the hummocky acoustic character
few kilometers (Fig. 3). As in line 1294 toward the outcrops on the seafloor (i.e., north of CDP 8500),
east, the acoustically hummocky chaotic sequence is 3.5 kHz precision depth recordings indicate that the
observed above horizon eM in line 1293 (Fig. 4). acoustic character is the result of sediment waves
The sequence above horizon A" is systematically (Damuth, 1980). In fact, individual sediment waves
less well laminated toward the west (from lines 1293 can be identified in the MCS data (Fig. 6).

612 N.W. D R I S C O L L and J.B. D I E B O L D
EW9501 Line 1321 - Beata Ridge - CDP #

25000 20000 15000 10000 5000
~!~:~~ _d,.,~,~
=.))~.'~r +~:7.'..":r;{c'.: :: ;1 "G3~7~-L'I'.~<-"I=~L,:(! ,]r .:,~17".7'":. :. :L,~,>~ii~:~'..<;s;7~."f,"g;~iq-_'eL 3~1~(.~.s163 ,7 :! ~;.D.'..f<7.:'.:+%.i:,.)'.'..:"~7,~,:&+.,'~!:.<.,7,1

:.ji~,iii:-~i;~i:~)~'!~.~.,:,~ (";.7L.,.::~E.;~J;iLI;(::~,:( ;.:,7.~.?~'1::.
:!~ii)~i;:1!i:7{ ,g;!i~:7: ~ :i~ i:~i~: :~:
EW9501 Line 1321 - Beata Ridge - CDP #

25000 20000 15000 10000 5000
sE
thickness
I ~ ~ ~ , onlapoing
5
~ i ~ ; ~ ~~. _
;.... ~]..~ 'Q ,~:~ - ! .- ~ :, __.
package
~ ~'~- '5
! - -
I 6 I
.a=
>,
I .7 I
~ '~ , .... !'I~4!~:~'
ii~ '~:'~' '~ ~" "..~:.-.~ . . . . . . . . . . ~ " .... , ' ~ ' ~
8 "8
9 "9
lO "10
11 "11
Fig. 10. Interpreted and uninterpreted seismic reflection line 1321 illustrating the highly faulted nature of the Beata Ridge. The
western flank of the Beata Ridge is bounded by a large westward-dipping normal fault. The basement offset across the normal fault is
approximately 5 s two-way travel time (~3750 m). The Beata Ridge is markedly asymmetric with the western scarp having more offset
and a steeper slope. The sediment thickness and stratal geometry observed across the Venezuelan Basin and Beata Ridge suggest that the
majority of the deformation in this region occurred soon after the emplacement of the volcanics.
The A" to eM sediment interval displays a sim- nondeposition, DSDP Site 146 (CDP 10,200) sam-
ilar pattern of thinning as the underlying B" to A" pled a more complete sedimentary section. In fact, the
sequence in the region. Nevertheless, in the area of erosional and non-depositional unconformities have
maximum thinning of the B" to A" sequence, this coalesced between CDPs 4200 and 6000, maximiz-
overlying younger unit maintains a finite thickness ing the duration of the hiatus in the region. The up-
(Fig. 7). Line 1320 traverses one of the few areas in dip contemporaneous pelagic units also display lateral
the study region where the A" to eM sequence shows thickness variations across the volcanic constructions
signs of truncation (Figs. 8 and 9). DSDP Site 150 located at CDP 12,500 and 14,000.
is located at CDP 4900 in an area where the max- At DSDP Site 146, iron manganese nodules (up to
imum hiatus occurs due to erosion of the A" to eM 5 cm in diameter) occur at the Early Miocene-Late
sequence and non-deposition within the B" to A" unit. Eocene unconformity at Site 150 suggesting slow ac-
Farther to the north away from this region of erosion- cumulation rates (i.e., current-controlled deposition;

Driscoll and Laine, 1996). Because the paleo-water rough-smooth B" boundary this thinning is much
depths between DSDP Sites 146 and 150 are not less pronounced than the thinning of the two un-
demonstrably different and some of the thinning in derlying sequences (e.g., B" to A", and A" to eM).
the A" to eM sequence occurs by truncation, Edgar The change from siliceous to carbonate ooze across
et al. (1973a,b) suggested that dissolution associated horizon eM, together with the diminished effects of
with the CCD could not account for the increased currents on the depositional regime after the Early
hiatus toward DSDP Site 150 (Fig. 9). They postu- Miocene, are consistent with the shoaling of the
lated that abyssal currents might explain the hiatus Panamanian Isthmus in the Early Miocene which
and the occurrence of manganese nodules. Similar would have closed the gateways for the deep cur-
to the hiatus observed at DSDP Site 150, there is a rents.
large hiatus ('-~30 Ma) that separates Early-Middle
Eocene chalk from Middle Miocene calcareous ooze
along the Hess Escarpment at ODP Site 1001 (Sig- DISCUSSION
urdsson et al., 1997).
Tectonic development of the Caribbean
eM to present sedimentary sequence
On the basis of the new MCS data acquired
All of the sedimentary wedges observed in the in the eastern Caribbean (Fig. 2), together with
southern portion of the Venezuelan Basin thin to- previous collected data from the region, we have
ward the north and interfinger with pelagic deposits developed the following tectonic model to illustrate
on the structural highs that were accumulating sed- the geologic development of the crust that floors the
iment concomitantly. However, this lateral thinning Venezuelan Basin (Fig. 11). Details of the magmatic
updip is less pronounced in the upper wedge (Early emplacement and building of the Caribbean oceanic
Miocene-Holocene; Fig. 4). These pelagic sedi- plateau can be found in Diebold et al., Chapter 19.
ments uniformly drape the existing topography with Our tectonic model supports the 'mobilist' model
thickness variations only occurring across steep to- for the development of the Caribbean (Pindell and
pographic features (e.g., seamounts and fault scarps). Barrett, 1990). We assume that the proto-Caribbean
The height where the pelagic sediment pinches out crust was formed by seafloor spreading in Late
against seamounts varies dramatically from one side Jurassic-Early Cretaceous time in the eastern Pacific
to the other (e.g., line 1293, CDP 2200; Fig. 4). In (Fig. 11), with an initial oceanic crust thickness of
this example, a marked change in seafloor character approximately 6 km (Diebold et al., Chapter 19).
accompanies this variation in sediment elevation on Prior to the Senonian, widespread and rapid eruption
either side of the seamount with sediment waves of basaltic flows began in concert with extension and
located toward the southeast and smoother seafloor thinning of the 'old' plate. It is difficult to deter-
occurring toward the northwest (Fig. 4). Because of mine whether extensional deformation pre-dated or
the different sedimentary styles, the seamount ap- post-dated the impact of the plume head and associ-
pears asymmetric in the 1 from the area, with a large ated magmatic activity (Galapagos hotspot; Duncan
slope to the north and small slope to the south. South and Hargraves, 1984). Nevertheless, extensional de-
of the seamount, the seafloor rises gently, and is formation had minimal effect on the thickness and
characterized by sediment waves with amplitudes of distribution of the early-stage basalt flows (Fig. 11).
tens to hundreds of meters and wavelengths of about Numerous hypotheses concerning the interaction of
1-2 km. plume heads with both active and passive rifting
The acoustic character above horizon eM system- of the overlying lithosphere have been advanced
atically becomes more laminated toward the west (Anderson et al., 1992; Coffin and Eldholm, 1994;
(from lines 1293 to 1304). Above horizon eM, thin- White and McKenzie, 1995) and are discussed in
ning along the northern slope still occurs, but the lat- more detail in Diebold et al., Chapter 19).
eral thickness variations are much less pronounced. The observed structure of the Caribbean basalt
The more acoustically laminated unit immediately province indicates at least two episodes of magmatic
overlying horizon eM thins both toward the south activity (Diebold et al., Chapter 19), which is consis-
and north away from the depocenter (Fig. 6, CDP tent with the age information reported by Donnelly
14,500) and is onlapped by the overlying sedimen- (1990). The early stage was more localized with
tary package. The upper sequence, the flat-lying steeper-dipping reflectors; the late stage is regionally
well-laminated package, thickens markedly towards more extensive with its eastern boundary being fault-
the south and also systematically thickens towards controlled. The rough-smooth B" basement bound-
the west (Fig. 7). ary imaged in seismic profiles is, in this model, the
Even though the eM to Present stratigraphic in- edge of the basalt province overlying the older plate
terval thins across the escarpment delineating the (Fig. 11). The divergent wedge imaged below B" in

Fig. 11. Schematic illustrating the geologic development of the Caribbean crust. (1) Proto-Caribbean oceanic crust formed by seafloor
spreading in Late Jurassic-Early Cretaceous time in the eastern Pacific. (2) Widespread and rapid eruption of basaltic flows in concert
with extension and thinning of the 'old' plate. The plate was thickened by at least two stages of basalt flows. The large divergent volcanic
wedge observed along the rough-smooth B't boundary, are coincident with the abrupt shoaling of Moho, and appear to be bounded
by a large northwest-dipping fault system. (3) Minor extensional deformation across the Venezuelan Basin continued after magmatic
thickening of the crust as evidenced by faulted and rotated basalt flows. The location of the major extensional deformation migrated
through time from the Venezuelan Basin to the western flank of the Beata Ridge. The extensional unloading of the footwall caused uplift
and rotation of the Beata Ridge and collapse of the hangingwall (i.e., Hess Escarpment).
line 1293 (Fig. 4) and the abrupt shoaling of Moho stratal geometry observed across evolving rift basins
are inferred to be controlled by a northwestward- (Driscoll et al., 1995). If the emplacements of the
dipping fault system. This fault system was active volcanics are similar to those of syn-rift sediments,
during the late stage of basalt flows. The high veloc- then this implies that the source of the basalts is
ities (,-~5 k m / s ; Diebold et al., Chapter 19) of this either along strike or toward the northwest, that is
divergent wedge suggest basaltic material with only structurally updip, and the volcanics progressively
minor intercalated volcaniclastics. The dips of the infill the deforming basin (Fig. 11).
reflectors that comprise the divergent wedge dimin- The dipping reflector packages observed in the
ish upsection and this evolution is very similar to the Venezuelan Basin exhibit many structural similari-

ties to seaward-dipping reflectors (SDRs) observed of deformation occurred in the Miocene and that
along other volcanic margins (Mutter et al., 1982; deformation is still ongoing. They propose that this
Coffin and Eldholm, 1994). The commonly cited deformation is associated with the overthrusting of
model for emplacement of the SDRs is that ther- the Colombian Plate onto the Venezuelan Plate along
mal dynamic support uplifts the magmatic source the eastern flank of the Beata Ridge (Fig. 10, CDP
above sea level with extrusions flowing away from 13,000). In our alternative hypothesis, the one-plate
the source region (Mutter et al., 1982). When the model, the majority of the deformation occurred
anomalous heat dissipates the plate cools, contracts, early on in the history of the Caribbean with mi-
and subsides, the primary dip of the volcanics is nor reactivation being caused by the north-south
reversed and the SDRs dip toward the extinct source. bending of the plate. The onlap above the rotated
In the case of the eastern Caribbean, the fault-con- block, which is the inferred thrust fault of Mauffret
trolled eastern extent of the late-stage flows together and Leroy (1997), indicates that the deformation
with the intercalated sills and marine limestones in- occurred prior to sedimentation in this region. In
dicate that the divergent wedge was formed in the summary, much of the topography associated with
marine environment. Details concerning the source the formation of the Venezuelan Basin and Beata
of the basalts flows and their inferred low viscosity Ridge pre-dated substantial sediment input into the
is discussed in Diebold et al., Chapter 19. region (Diebold et al., 1995a,b). We propose that
Minor extension continued after magmatic thick- the minor fault reactivation in the Neogene along
ening of the Venezuelan Basin crust as evidenced the eastern flank of the Beata Ridge is associated
by faulted and rotated tongues of smooth basement with the shortening between the North and South
(Fig. 11). However, the location of the major ex- American Plates which began in the Eocene (Pindell
tensional deformation migrated through time from and Barrett, 1990). The deformation style changes
the western portion of the Venezuelan Basin to the along strike from the Muertos Trough to the east and
western flank of the Beata Ridge. Strain hardening the Hispaniola margin toward the west. A north-
of the lithosphere in the Venezuelan Basin might south-trending transfer zone (i.e., tear fault) accom-
have caused the deformation to migrate from an modates this change in the deformation style from
area with thin crust to a region with thicker crust. bending and subduction of the Caribbean Plate along
The deformation across the Beata Ridge is the con- the Muertos Trough to the east and compressional
sequence of extension across a westward-dipping deformation and obduction of the Caribbean Plate
fault, with the Hess Escarpment being the collapsed along Hispaniola to the west (Fig. 2). Onshore
hangingwall block. Antithetic faults along the east- Hispaniola at the intersection of the island with
ern Hess Escarpment result in a graben-like structure the Beata Ridge, NNE-trending normal faults occur
between the Beata Ridge (footwall block) and the along the eastern margin of the Sierra Bahoruco
Hess Escarpment (hangingwall block). Furthermore, (Pindell et al., 1988). Toward the west of the nor-
the regional uplift of the Beata Ridge caused minor mal faults, the Beata Ridge becomes subaerial (i.e.,
faulting across the plateau. The onlap of the B" to Sierra Bahoruco) as a result of the collisional defor-
A" sediment interval onto the rotated and deformed mation with the Central Cordillera arc of Hispaniola
blocks suggests that deformation occurred soon after (Biju-Duval et al., 1982; Heubeck and Mann, 1991).
magmatic emplacement. Furthermore, the current- Our hypothesis explains the north-south trend of the
controlled depositional features, which are enhanced faults and predicts that fault displacement should in-
around structural highs, suggests that the majority crease northward in concert with the relief between
of the deformation responsible for the formation of the downgoing and obducted slab. We propose that
Beata Ridge was completed prior to any substantial this difference in deformational style along this por-
sediment accumulation. This timing of the deformation of the Caribbean Plate is, in part, a consequence
tion is also consistent with the stratal patterns of the of the thicker crust on the Beata Ridge, which is
turbidites that infill the graben between Beata Ridge more resistant to subduction.
and the Hess Escarpment and the basal turbidites that
infill the Venezuelan Basin. Decompression melting Late Cretaceous to Eocene turbidites deposition
as a result of the extensional deformation may have
led to partial melting of the depleted residual mantle, Coincident with the rough-smooth B" boundary
generating some new crust formation in Campanian is a marked change in the thickness of the A" to
time (DSDP Site 152 and ODP Site 1001 basalts). B" sediment interval (Figs. 3, 4 and 6). Correla-
The stratal geometry and basement structure are tion from existing DSDP Sites 146 and 150 in the
not consistent with the two-plate hypothesis used Caribbean constrains the age of this interval to be
to explain the origin of the observed deformation older than Middle Eocene (~50 Ma) and younger
across the Beata Ridge and Venezuelan Basin. Mauf- than Senonian (~88 Ma; Fig. 9). However, because
fret and Leroy (1997) proposed that a renewed phase the DSDP and ODP sites are located on structural

Fig. 12. Plate reconstructions of the Caribbean during the Late Cretaceous and early Tertiary illustrating the sediment delivery from
the axial-parallel fluvial systems draining the Maracaibo-Peruvian foreland basin to the Venezuelan Basin (modified from Pindell and
Barrett, 1990 and Pindell and Tabbutt, 1995). (Top) Uplift and deformation of the Andes that began in the Late Cretaceous led to the
formation of the Maracaibo-Peruvian foreland. The northern margin of South America was relatively undeformed at this time allowing
the axial-parallel fluvial system access to the Venezuelan Basin. Obduction of the arc and related terranes occurred along the eastward
verging Romeral suture-thrust in the Late Cretaceous. (Bottom) Progressive infilling of the Maracaibo-Peruvian foreland basin occurred
from south to north from Late Cretaceous to early Tertiary time.

highs, they have failed to sample the basal onlapping more in-depth discussion on the tectonic history of
sedimentary sequence (Figs. 8 and 9). The A" to South America the reader is referred to Tankard et
B" sediment interval is thickest in the southeastern al. (1995).
portion of the Venezuelan Basin and systematically Several phases of tectonic uplift and deforma-
decreases toward the east, north, and west. Similar tion have been documented for the Andes between
basal onlapping packages also have been observed in the Late Cretaceous and Pleistocene (Hoorn et al.,
structural lows in the Colombian Basin (Edgar et al., 1995). The Maracaibo-Peruvian Trough, which as
1971; Lu and McMillen, 1982; Kolla et al., 1984; the name implies extends from southern Peru to
Bowland and Rosencrantz, 1988; Bowland, 1993). the Gulf of Maracaibo, Venezuela, was first iso-
However, their regional extent is much more limited lated from the open ocean to the west by a vol-
than the onlapping packages infilling the Venezuelan canic barrier in the Central Cordillera of Colombia
Basin, suggesting that they either have a local source in the Late Triassic-Early Jurassic (Browning and
or represent the distal reaches of larger more con- Walper, 1982; Macelli, 1988; Pindell and Tabbutt,
tinuous proximal deposits. The onlapping packages 1995). During the Late Jurassic and Early Creta-
observed in the Colombian Basin may be derived ceous several marine incursions of the Maracaibo-
from local fiver networks draining the western flanks Peruvian Trough occurred from the north and pene-
of the Central Cordillera (Fig. 12). The thinning trated southward to at least southern Peru (Browning
of the sequence in the Venezuelan Basin occurs by and Walper, 1982). In Colombia, metamorphism and
onlap onto pre-existing structural highs. Their well- uplift of the Amaime-Chaucha Terrane occurred
laminated acoustic character together with the onlap by an eastward-vergent thrust system (i.e., Romeral
pattern suggest that these deposits are the result of Suture) onto the Central Cordillera, which was an
gravity flows infilling topographic lows (Driscoll et active margin during Campanian time (Bellizzia and
al., 1995). Dengo, 1990; Case et al., 1990a,b; Pindell and Bar-
On the basis of velocity information (Fig. 5), the rett, 1990; Kellogg and Vega, 1995; Pindell and
maximum thickness of the B" to A" sequence is Tabbutt, 1995). Fission-track data from the region,
approximately 600 to 800 m and its distribution is which are a proxy for denudation, indicate that up-
roughly coincident with the area of rough basement lift and erosion began in Late Cretaceous-Paleocene
in the Venezuelan Basin (Diebold et al., 1981). We time, but accelerated in the late Tertiary (Kroo-
interpret the velocity increase from 2.4 km/s to nenberg et al., 1990). Escalante (1990) proposes
3.2 km/s that accompanies the change in thickness compression across the Romeral fault zone during
as evidence for a facies boundary between pelagic the Late Cretaceous, resulting in the obduction and
sediments to the north and terrigenous sediments to emergence of Pacific Ocean crust (i.e., the Choc6
the south (more silt-sand prone). The thickness and Block) against the South American Craton. The
distribution of this sedimentary succession suggests tectonic loading associated with the thrusting caused
that it represents the distal portion of a terrige- subsidence in the foreland basin to the east of the up-
nous depositional system. The proximal portions of lift. The tectonic subsidence was augmented by the
the deposit to the south may underlie the Curaqao contemporaneous long-term eustatic sea level rise
Ridge (Fig. 15; Edgar et al., 1971) and infill a (Fig. 12). During the Late Cretaceous through early
marginal basin, which may explain the large nega- Tertiary, the Maracaibo-Peruvian foreland basin was
tive gravity anomaly (Bowin, 1976) observed in that progressively infilled with fluvially dominated de-
region. If our interpretation is correct and these flat- posits from south to north (Fig. 12 Browning and
lying, acoustically laminated sediments are the distal Walper, 1982; Macelli, 1988; Pindell and Tabbutt,
portion of a large terrigenous system, then these sed- 1995). The sediments infilling the basin were de-
iments will provide valuable information about the rived from the Guyana and Central Brazilian shields
evolution of the Caribbean-South American Plate to the east and the incipient Andes to the west
boundary during the Late Cretaceous and early Ter- (Fig. 14; Browning and Walper, 1982). The west-
tiary. Based on paleo-reconstructions of the South erly flowing fluvial systems draining the Guyana and
American Plate, together with the spatial distribution Central Brazilian shields exhibited a radial pattern
of these deposits, we propose that the source for away from the structural highs with no preferred
these terrigenous sediments is a northward-flowing orientation to the drainage networks. The drainage
axial-parallel drainage system developed east of the from the Andes and the shields were tributaries to a
uplifting and deforming Andean Cordillera during large-scale northerly flowing drainage system, which
the Late Cretaceous and early Tertiary (Fig. 11). was axial parallel to the Maracaibo-Peruvian fore-
We will briefly discuss the tectonic and stratigraphic land basin. In actively deforming regions, be they
development of South America, the establishment extensional or compressional, the dominant drainage
of drainage systems, and their relationship to the networks are usually axial parallel to the tectoni-
onlapping packages in the Venezuelan Basin. For a cally induced topography (Driscoll et al., 1995). The

northern margin of South America was relatively subsidence appears to have occurred in the Late
undeformed during this period and thus allowed the Paleocene to Early Eocene. Obduction of island-arc,
axial-parallel fluvial networks access to the Venezue- frontal basin, and oceanic crust over the north-
lan Basin (Figs. 12G and 14B). The high-standing dipping continental crust along the Frontal Thrust
Guyana and Central Brazilian shields, as well as segmented the larger Maracaibo-Peruvian foreland
the structural barriers (e.g., De Purus and Iquitos basin into a series of smaller basins and associated
arches) across the Marfijo and Solim6es basins, in- topographic highs, which led to the formation of
hibited drainage to the east (i.e., towards the Atlantic the Venezuelan Caribbean Mountains and Venezue-
Ocean) away from the Maracaibo-Peruvian foreland lan foredeep (Kasper and Larue, 1986). Kasper and
basin (Browning and Walper, 1982; Hoorn et al., Larue (1986) proposed that the northern-trending
1995; Pindell and Tabbutt, 1995). Likewise, the An- Andes foredeep (i.e., Maracaibo-Peruvian foreland
dean Central Cordillera prevented drainage from the basin) and the northeast-trending Venezuelan fore-
Maracaibo-Peruvian foreland basin toward the west deep acted as conduits for sediment transport to the
into the Pacific Ocean (Fig. 12). Through time, the Caribbean. Furthermore, terrigenous sediment trans-
fluvial networks became graded and the established port through the Venezuelan foredeep might explain
equilibrium profiles change in concert with base the existence of Paleogene sandstones on the Island
level variations (Schumm, 1991). This equilibrium of Barbados that appear to be derived from the
profile was perturbed by relative sea level changes Cordillera and Shield regions as well as the Mi-
and tectonic uplift of the hinterland. During the Late rador Formation sandstones that are interpreted as
Cretaceous and early Tertiary, the long-term eustatic the fluvial deposits of a large Paleogene deltaic com-
sea level was falling, which further entrenched the plex in Colombia and western Venezuela (Kasper
axial-parallel drainage system in the Maracaibo- and Larue, 1986). Northeast progradation over the
Peruvian foreland basin (Fig. 12). The transition present-day Lake Maracaibo region (e.g., Misoa For-
from shallow marine to predominantly continen- mation sandstones) in the Early to Middle Eocene
tal conditions in the Maracaibo-Peruvian foreland is additional evidence for a southwestwardly derived
basin throughout the Late Cretaceous and early Ter- sediment source at this time.
tiary records some combination of the following: (1) Increased resistance to obduction along the
a decrease in tectonic subsidence; (2) an increase in Frontal Thrust may have produced a back thrust
eustatic sea level fall; and (3) an increase in sediment to accommodate the north-south strain between the
supply to the evolving basin. On the basis of the ex- Caribbean and South American plates (Bosch and
isting data, it is difficult to determine which of the Rodriguez, 1992). Along the northern boundary of
above was the main cause for the regional regression the Venezuelan Caribbean Mountains a transform
across the area. The newly acquired MCS data from zone developed between the Caribbean Plate to
the Venezuelan Basin for the first time imaged the the north and the obducted terranes to the south,
stratal geometry of the basal sequences and allowed which accommodated the large strike-slip motion.
us to develop a consistent model for the onshore de- The east-west Oca, Bocon6, and Bocon6-Mor6n-
formation, the development of the South American E1 Pilar fault systems exhibit major fight-lateral dis-
drainage systems, and the delivery of sediments to placement since the Late Oligocene (Fig. 13). The
the Caribbean. Because the preservation potential in two fault systems that partition the strain, the South
the oceans is much greater than onshore, combining Caribbean marginal fault and the North Venezuelan
both data types provides for a more complete recon- fault belt, define the northern and southern extent of
struction of the tectonic and stratigraphic evolution the Bonaire Block (Bosch and Rodriguez, 1992).
of the Caribbean-South American Plate boundary. Continued uplift and deformation along the north-
Eastward migration of the Caribbean Plate and ern South America Plate boundary (e.g., Venezuelan
the onset of compression between the North and Caribbean Mountains; Bellizzia and Dengo, 1990)
South American plates in the early Tertiary (Pindell affected the axial-parallel fluvial networks flowing
and Barrett, 1990) caused progressive deformation into the Venezuelan Basin. Because the uplift first
along the Caribbean-South American Plate bound- occurred along the western portion of the margin,
ary (Fig. 13). Nappes were emplaced southeastwards the drainage systems were progressively deflected
onto the Venezuelan margin since the Paleocene towards the east (Figs. 13 and 14A). The consequent
with the age of the deformation becoming system- increase in the overall river length (this evolving
atically younger towards the east (Pindell et al., drainage system will become the Orinoco River)
1988; Pindell and Barrett, 1990; Bosch and Ro- caused an increase in base level and deposition
driguez, 1992). Although, there is controversy over along the upper reaches of the drainage network.
the vergence direction in some of the northern basins The uplift of the Venezuelan Caribbean Mountains
(e.g., Maracaibo; Lugo and Mann, 1995), the on- that dammed and diverted the fiver systems along
set of the deformation and increased uplift and the northern boundary of South America in the

TECTONIC AND STRATIGRAPHIC D E V E L O P M E N T OF THE EASTERN C A R I B B E A N 619
Fig. 13. Plate reconstructions of the Caribbean during the Eocene and Early Miocene illustrating the progressive deformation of the
northern South American Plate boundary (modified from Pindell and Barrett, 1990 and Pindell and Tabbutt, 1995). (Top) Onset of
deformation along the northern boundary of South America during the Paleocene and Eocene, segmenting the larger Maracaibo-Peruvian
foreland basin into a series of smaller basins and associated topographic highs. (Bottom) Continued deformation along the northern
margin uplifted the Venezuelan Caribbean Mountains along the southward verging Frontal Thrust, which blocked the northward-flowing
axial-parallel fluvial systems. This late-stage deformation, in large part, reorganized the drainage of South America and led to the
development of the Orinoco and Amazon drainage systems. The east-west Oca, Bocon6, and Bocon6-Mor6n-el Pilar fault systems
along northern South America exhibit major right-lateral displacement.

Fig. 14. (A) Topographic map of northern South America showing present-day drainage systems. Note that the Orinoco River has been
diverted east and flows north along the western boundary of the Guyana Shield. Black bars denote barriers in the Marajo (MB) and
Solim6es (SB) basins. WC -- Western Cordillera; CC = Central Cordillera; EC = Eastern Cordillera; MA = M6rida Andes; M = Lake
Maracaibo; VCM = Venezuelan Caribbean Mountains. (B) Paleogeographic map for Late Cretaceous to early Tertiary illustrating the
development of the Central Cordillera (CC) and the Maracaibo-Peruvian foreland basin. Radial drainage from the Guyana and Brazilian
shields as well as the Central Cordillera are tributaries to the northward-flowing axial-parallel drainage.
Early Miocene (Figs. 13 and 14A), together with thickness and spatial distribution of the onlapping
a renewed phase of tectonic activity in the Cen- basal turbidites and overlying current-controlled de-
tral and Eastern Cordillera, dramatically changed the posits indicate a southem source for the terrigenous
drainage pattern of South America. Prior to these sediments (Fig. 15), which persisted from Late Cre-
events, the major drainage was to the north and taceous to Early Miocene time (Figs. 4, 6 and 7).
supplied sediment to the Venezuelan Basin. The A dramatic change in the sediment supply occurred

-76" -72 ~ -68 ~ -64"

20"
O
(7 O
16"
12"
Fig. 15. Regional isopach map for the eastern Caribbean region (modified and updated from Edgar et al., 1971) with the inferred current
axis superposed. Note the thick depocenter (> 14 km) beneath the Curacao Ridge. Given the eastern migration of the Caribbean Plate
with respect to the South American Plate, during the Late Cretaceous and early Tertiary the depocenter would have been farther west,
located roughly adjacent to the northern extension of Lake Maracaibo (see Fig. 12). The contour current appears to flow along and
parallel to the 4000 m isobath. Current activity began in the Middle Eocene, above reflector A", and the current intensity started to wane
toward the Early-Middle Miocene.
in Early Miocene time; the only major river system and Miocene Andean tectonic activity (Fig. 13),
still disgorging sediment from South America to sufficient sediment was supplied to the developing
the Caribbean Plate is the Magdalena River. Note foreland basin to infill the paleo-structural lows east
that, like the previous drainage system, the Mag- of the north-south-trending foreland basin (e.g., the
dalena River is an axial-parallel river system with Marfijo and Solim6es basins; Fig. 14). As a con-
one branch being sourced from the basin between the sequence of this regrading, fluvial systems flowed
Western and Central Cordillera and the other sourced to the east across the South American continent
from the basin separating the Central from the East- and delivered sediments to the Atlantic Ocean that
ern Cordillera (Hoorn et al., 1995; Fig. 14A). This led to the development of the Amazon delta and
change in sediment source region is consistent with fan. On the basis of limited biostratigraphy, it is
the inferred sediment provenance change from the postulated that most of the Amazon fan was de-
Paleocene-Eocene and the Miocene and Pleistocene posited after the Middle Miocene in response to
recorded at ODP Site 1001 (Sigurdsson et al., 1997). Andean uplift (Damuth and Flood, 1984). Never-
Onshore sediment provenance studies indicate that theless, because the Andes were high-standing re-
the early Tertiary sediments sampled in Barbados gions since the Late Cretaceous, the initiation of
and Venezuela were sourced from the Cordillera sys- the Amazon fan is not simply a consequence of
tem and Guyana Shield (Kasper and Lame, 1986; Middle Miocene uplift. We propose that two im-
Baldwin et al., 1986). Furthermore, fission-track portant events were responsible for modifying the
analysis of these samples yields age ranges from 80 drainage of South America and forming the present-
to 15 Ma, which is consistent with the erosion of the day Amazon and Orinoco drainage systems. First,
high Cordillera mountain system beginning in the and most important, incipient uplift and deformation
Late Cretaceous (Baldwin et al., 1986). of the northern Andes (Central Cordillera, Eastern
As a result of the northward damming and/or Cordillera, Santander Massif, Santa Marta Massif,
diverting of the rivers and the renewed Oligocene Sierra de Perija), Venezuelan Andes (Mdrida An-

des), and the Venezuelan Caribbean Mountains in calized within the central Venezuelan Basin. The
Late Oligocene to Early Miocene time dammed or hemipelagic-pelagic sediments which comprise the
diverted the existing axial-parallel drainage system post-A" sequence thin markedly towards the north-
of the Maracaibo-Peruvian foreland basin (Figs. 12- west away from the depocenter predominantly by
14A). Either damming or increasing the length of onlap, suggesting that the hiatus results from non-
the fluvial system would raise base level and cause deposition rather than erosion. However, near DSDP
deposition along the upper reaches of the fluvial Site 150 (Fig. 8), the sedimentary section thins by
profile (i.e., the equilibrium profile; Schumm, 1991). truncation suggesting that, in part, the hiatus also is
Renewed tectonic activity in the Central and East- due to erosion. Farther north away from the abyssal
ern Cordillera caused additional uplift, but also in- current axis, DSDP Site 146 sampled a more com-
creased tectonic subsidence in the foreland basin plete sedimentary section (Fig. 9). We attribute this
east of the Cordillera due to flexure loading of the pattern of thickening, thinning, and truncation to
footwall block. Second, through time, the evolving the flow of geostrophic bottom currents around the
drainage systems infilled the paleo-structural lows Venezuelan Basin. Geostrophic flows usually par-
because the axial-parallel drainage systems were allel the contours because the pressure gradient is
forced east by the tectonic dam to the north. Upon balanced by the coriolis force. The core of bottom
infilling the structural lows and regrading of the water flow appears to have been concentrated near
fluvial profiles by erosion, deposition, and isostasy, the 4000 m isobath and given the topography and
the fluvial systems were able to bypass the structural the northern latitude, the circulation pattern is coun-
barriers across the Marfijo and Solim6es basins and terclockwise. The onset of this current-controlled
deliver sediments to the Atlantic Ocean. deposition above reflector A" began in the Mid-
The dramatic change in sediment accumulation dle Eocene and waned in the Early Miocene (~20
and distribution that we observe in the MCS data Ma; Fig. 13). This resulted in more subdued cur-
from the Venezuelan Basin indicates that axial-par- rent-controlled features in post-Early Miocene time.
allel drainage networks delivered sediment from the Even though current intensity has waned the circu-
Late Cretaceous to Early Miocene. Our results are lation pattern appears to have been maintained to
consistent with early Tertiary paleo-flow directions the present (Fig. 4). Analysis of 3.5 kHz echograms
based on sediment provenance studies in Barbados suggests that the finer portions of the turbidites en-
and Venezuela (Kasper and Lame, 1986). These tering the Venezuelan abyssal plain in recent times
studies indicated that the major drainage of the are entrained by the bottom currents and deposited in
South American Craton was towards the north in blankets of sediment waves mantling the underlying
the Late Cretaceous and early Tertiary. The present deposits.
Orinoco and Amazon drainage systems developed as Given that the acoustic character and distribution
a result of tectonic deformation along the northern of the A" to eM is diagnostic of current-controlled
South American margin (e.g., Venezuelan Caribbean deposition, the cogent question becomes, what is
Mountains; Fig. 14). Even though the Late Creta- the source for the bottom water and through which
ceous to Early Miocene sediments in the Venezuelan gateways did it gain access to the Caribbean. There
Basin appear to be derived from South America, are no present-day gateways through the northern
the deposition was predominantly by gravity flows and eastern Caribbean structural barrier (Greater
prior to the Middle Eocene and by current-control and Lesser Antilles) that are deep enough to allow
after the Middle Eocene (i.e., above and below A"). lower North Atlantic Deep Water (NADW) and/or
A change in climatic conditions during the Late Antarctic Bottom Water (AABW) access to the
Eocene and the onset of abyssal currents, together Caribbean. The sill depth of the deepest gateway
with the acoustic character of the sediments, suggest along the Antilles, the Anegada Passage, is 1960
that the deposition of the A" to eM sequence was m (Worthington, 1966) and it only allows Antarc-
controlled by abyssal current circulation (Figs. 4 and tic Intermediate Water (AAIW) and upper NADW
6; McCave and Tucholke, 1986; Westall et al., 1993; access to the Caribbean (Worthington and Wright,
Driscoll and Laine, 1996). 1970; Wright and Worthington, 1970; Wunsch and
Grant, 1982). Even so, the configuration of the east-
Late Eocene to Early Miocene current-controlled ern Caribbean structural barrier might have been
deposits quite different during the Middle Eocene to Early
Miocene. Towards understanding the paleo-gateways
Similar to the B" to A" succession, the sedi- to the Caribbean, Donnelly (1990) analyzed pelagic
ment drift sequence above A" thickens dramatically sediments collected by DSDP drilling in the At-
toward the southeast displaying a pronounced in- lantic, Caribbean, and Pacific. Geochemical analysis
crease in thickness across the rough-smooth bound- of the pelagic sediments, used to infer intermediate
ary (Fig. 4). The depocenter of this deposit is lo- and bottom water chemistry, indicated that a barrier

between the Caribbean and western North Atlantic hypothesis is correct, then surface circulation could
was established at least by Middle to Late Eocene have been more vigorous when the Panamanian
time (,~40 Ma). If the Caribbean intermediate and Isthmus was open than it is today. The increased
deep waters were in communication with the west- intensity of the surface current might explain the
ern Atlantic, then the excess silica in the sediments different styles of the current-controlled deposition
of both basins should vary in concert. However, the before and after the Early Miocene.
relatively precipitous Late Eocene disappearance of The current-controlled deposit observed in
excess silica in the western Atlantic was not ob- EW9501 seismic data appears to be a detached
served in either the Caribbean or Pacific samples. drift with a morphology similar to the Greater An-
Siliceous sedimentation in the Caribbean continued tilles, Blake, and Eirik drifts (McCave and Tucholke,
until Early Miocene time suggesting that a structural 1986). Within the A" to eM sediment interval, the
barrier isolating the Caribbean from the Atlantic ex- laminated character systematically decreases west-
isted since at ~40 Ma. The barrier is presumed to be ward away from line 1293 (Figs. 4 and 6). In
the Aves Ridge, or the Lesser Antilles, or both. The addition, the acoustic character diminishes slightly
reduction of excess silica in Caribbean sediments in toward the east away from line 1293 (Fig. 2). Us-
the Early Miocene is interpreted to record the grading the hummocky acoustic character to define the
ual shoaling of the Panamanian Isthmus (Donnelly, region where the swiftest-flowing ribbons of current
1990). Consequently, we propose that the gateway occurred, has allowed us to reconstruct the paleo-cir-
for the deep Caribbean circulation is the Panamanian culation (Figs. 13 and 15). Given that much of the
Isthmus between Central and South America, which structuring of the present-day Caribbean Plate had
allowed abyssal water (AABW??) in the eastern Pa- occurred by Eocene time (i.e., Venezuelan Basin,
cific access into the Caribbean. The gradual shoaling Beata Ridge, and Hess Escarpment) or was well
of the Central American Isthmus closed the gateway underway (Caribbean-South American Plate bound-
in Late Oligocene-Early Miocene time, consistent ary), we propose that the increase in hummocky
with the diminished occurrence of current-controlled character within A" to eM toward the west records
features since the Early Miocene observed in the a velocity increase associated with the basin con-
seismic sections (i.e., the more uniform thickness off striction around Beata Ridge (Fig. 15). Likewise, the
the sequence overlying horizon eM). regions with hummocky acoustic character immedi-
We interpret the truncation and non-deposition ately above eM are structurally shallower and are
between A" and eM horizons observed in the seis- closer to the axis of the current at the time of depo-
mic reflection data to record the counterclockwise sition (Figs. 3 and 4). Conversely, the regions with
flow of bottom currents from the eastern Pacific well laminated sequences overlying eM are struc-
around the Venezuelan Basin (Figs. 4 and 6-8). A turally deeper and basinward of the main current
core of bottom water flow appears to be concen- axis at the time of deposition (Figs. 6 and 7).
trated near the present-day 4000 m isobath (Figs. 13
and 15). This ribbon of flow would explain the ero-
sion and non-deposition observed along line 1320 CONCLUSIONS
(Fig. 8). DSDP Site 150 is located in an area
where the maximum hiatus occurs due to erosion The newly acquired high-resolution MCS data
and non-deposition. Farther north, DSDP Site 146 during EW9501 clearly imaged the crustal structure,
(Fig. 9) sampled the sedimentary section farther crustal thickness, and the character of the overlying
away from the ribbon of swift current and recovered sedimentary successions. These data coupled with
a more complete sedimentary section. The onset of previous collected onshore and offshore data, have
current-controlled deposition above horizon A" be- allowed us to develop the following model for the
gan in the Middle Eocene and waned in the Early geologic evolution of the Venezuelan Basin.
Miocene (~-,20 Ma). Even though the dominant cur- (1) The proto-Caribbean crust was formed by
rent-controlled features diminished during the Early seafloor spreading in Late Jurassic-Early Creta-
Miocene, current-controlled deposition has contin- ceous time. We have assumed normal oceanic crust
ued to the Holocene. The present-day circulation thicknesses of approximately 6 km for the proto-
could be the result of AAIW or upper NADW en- Caribbean crust because the seismic reflection data
tering the basin through the Anegada and Cayman indicates that faulting and extension were concomi-
passages and causing a sluggish cyclonic circula- tant with, and subsequent to, the magmatic activity.
tion around the basin (~5 cm/s). An alternative (2) Prior to the Senonian, widespread and rapid
hypothesis, is that Guyana surface current entering eruption of basaltic flows began in concert with exten-
the Caribbean and flowing across the Caribbean into sion and thinning of the 'old' plate. We are not able to
the Gulf of Mexico excites a deep water circulation determine whether extensional deformation pre-dated
in the Venezuelan Basin (Fig. 15). If this alternative or post-dated the onset of magmatic activity.

(3) The plate was thickened by at least two if drainage was a north-flowing axial-parallel system
not more magmatic events. The excess thickness was that supplied sediment to the Venezuelan Basin. The
generated by the extrusion of lava flows over a vast middle Tertiary uplift and deformation along the
area, intrusion of dikes and sills, and underplating by northern South America Plate boundary blocked the
residual mantle from the melting event. The rough- axial-parallel fluvial networks. As a result of the
smooth B" boundary imaged in seismic profiles is northward damming of the rivers and the renewed
the edge of the basalt province created by these Oligocene and Miocene Andean tectonic activity,
magmatic events over the older deforming plate. abundant sediment was supplied to the developing
(4) The large divergent volcanic wedges observed foreland basin east of the Andes that infilled the
along the rough-smooth B" boundary and the abrupt paleo-structural lows in the Mar~jo and Solim6es
shoaling of Moho appear to be controlled by a north- basins. The consequent regrading of the fluvial sys-
westward-dipping fault system. The dip of the reflec- tems allowed drainage systems to flow east across
tors that comprise the divergent wedge diminish up- the South American continent (e.g., Amazon and
section and resemble the stratal geometry observed Orinoco) and deliver sediment to the Atlantic Ocean.
in rift basins. In contrast to the common model (8) Finally, the seismic reflection data also imaged
cited for the formation of these volcanic wedges an Eocene-Early Miocene current-controlled drift
(seaward-dipping reflectors, SDRs), we propose that deposit which might reflect the movement of eastern
the source of the basalt flows is either along strike Pacific bottom currents into the Caribbean during
or toward the northwest, structurally updip, and that this period. The gradual shoaling of the Central
they progressively infill the deforming basin. American Isthmus in Late Oligocene-Early Miocene
(5) Extension continued after magmatic thicken- time closed the gateway. Such closure is consistent
ing of the Venezuelan crust as evidenced by faulted with the diminished occurrence of current-controlled
and rotated tongues of smooth basement toward the features since the Early Miocene observed in the
east. The location of the major extensional defor- seismic sections and Caribbean paleo-deep and in-
mation migrated through time from the Venezuelan termediate water geochemistry of the sediments.
Basin to the western flank of the Beata Ridge. The
extensional unloading of the footwall caused uplift
and rotation of the Beata Ridge and collapse of the ACKNOWLEDGEMENTS
hangingwall (i.e., Hess Escarpment). The sediment
thickness and stratal geometry of the overlying sed- We thank the crew and scientific staff of the R/V
imentary successions across the Venezuelan Basin Ewing for making cruise EW9501 in the Caribbean
and Beata Ridge suggest that the majority of the ob- such a great success. This manuscript benefited
served deformation in this region occurred soon after from numerous discussions with Lewis Abrams,
the emplacement of the volcanics. Minor fault reac- Peter Buhl, Thomas Donnelly, Bob Duncan, Ed-
tivation in the Neogene along the eastern flank of the ward Laine, Sylvie Leroy, Garry Karner and Elazar
Beata Ridge is associated with an accommodation Uchupi. James Kellogg, Paul Mann, Walter Pitman,
zone (i.e., tear fault) that records a change in the James Pindell, and Elazar Uchupi critically read this
deformation style from subduction of the Caribbean manuscript and their comments are greatly appreci-
Plate along the Muertos Trough to obduction of the ated. Garry Karner compiled the topographic data
Caribbean Plate along Hispaniola. shown in Fig. 14. Support for this research was
(6) Coincident with the rough-smooth boundary provided by the National Science Foundation grant
is a marked increase in the thickness of the A" to OCE-93-02578. Woods Hole Oceanographic Insti-
B" sediment interval. Correlation with DSDP and tution contribution #9503. Lamont-Doherty Earth
ODP sites in the Caribbean constrains the age of Observatory #5684.
this interval to be younger than Senonian (,~88
Ma) and older than Middle Eocene (~50 Ma). We
conclude that the Late Cretaceous to Early Miocene
REFERENCES
sediments observed in the Venezuelan Basin are
terrigenous deposits transported northward by an Abrams, L.J., Larson, R.L., Shipley, T.H. and Lancelot, Y., 1993.
axial-parallel fluvial system draining the Maracaibo- Cretaceous volcanic sequences and Jurassic oceanic crust in
Peruvian foreland basin of South America. the east Mariana and Pigafetta Basins of the western Pacific.
(7) The initiation of the present-day South Amer- The Mesozoic Pacific: geology, tectonics, and volcanism. Geo-
ican drainage patterns is not simply a consequence phys. Monogr., 77: 77-101.
of Middle Miocene uplift of the Andes, as the Andes Anderson, D.L., Zhang, Y.-S. and Tanimoto, T., 1992. Plume
heads, continental lithosphere, flood basalts, and tomography.
are the result of several tectonic episodes begin- In: B.C. Storey, T. Alabaster and R.J. Pankhurst (Editors),
ning in the Late Cretaceous. We propose that during Magmatism and the Causes of Continental Break-Up. Geol.
the Late Cretaceous to Early Miocene the dominant Soc. London Spec. Publ., 68: 99-124.

T E C T O N I C AND STRATIGRAPHIC D E V E L O P M E N T OF THE E A S T E R N C A R I B B E A N 625
Baldwin, S.L., Harrison, T.M. and Burke, K., 1986. Fission track Venezuelan Basin and Beata Ridge; results from a new re-
evidence for the source of accreted sandstones, Barbados. gional geophysical survey. Geological Society of America,
Tectonics, 5: 457-468. Special Program: International Division and IGCP, Project
Bellizzia, A. and Dengo, G., 1990. The Caribbean mountain 364: Caribbean Arcs and Ophiolites: Geology and Tectonics
system, northern south America: a summary. In: G. Dengo of the Caribbean Region.
and J.E. Case (Editors), The Geology of North America: The Donnelly, T.W., 1973. Magnetic anomaly observations in the
Caribbean Region. Decade of North American Geology, Vol. eastern Caribbean Sea. Init. Rep. DSDP, 15:1023-1030.
H, Geological Society of America, Boulder, CO, pp. 167-175. Donnelly, T.W., 1989. Geologic history of the Caribbean and
Biju-Duval, B., Mascle, A., Montadert, L. and Wanneson, J., Central America. In: A.B. Balley and A.R. Palmer (Editors),
1978. Seismic investigation in the Colombia, Venezuela and The Geology of North America, Decade of North American
Granada Basins, and on the Barbados Ridge for future IPOD Geology, Vol. A, Geological Society of America, Boulder, CO,
drilling. Geol. Mijnbouw, 57:105-116. pp. 299-321.
Biju-Duval, B., Bison, G., Mascle, A. and Muller, C., 1982. Donnelly, T.W., 1990. Pelagic sediment, deep water chemistry,
Active margin processes: field observations in southern His- and tectonics: an application of the history of biological sed-
paniola. Am. Assoc. Pet. Geol. Bull., 67: 242-270. iment accumulation on the tectonic history of the Caribbean.
Bosch, M. and Rodriguez, I., 1992. North Venezuelan collisional Riv. Ital. Paleontol. Stratigr., 96 (2-3): 143-164.
crustal block: the boundary between the Caribbean and South Donnelly, T.W. and 10 others, 1990. History and tectonic set-
American plates. J. S. Am. Earth Sci., 6: 133-143. ting of Caribbean magmatism. In: G. Dengo and J.E. Case
Bowin, C., 1976. Caribbean gravity field and plate tectonics. (Editors), The Geology of North America: The Caribbean Re-
Geol. Soc. Am. Spec. Pap., 169: 1-79. gion. Decade of North American Geology, Vol. H, Geological
Bowland, C.L., 1993. Depositional history of the western Colom- Society of America, Boulder, CO, pp. 339-374.
bian Basin, Caribbean Sea, revealed by seismic stratigraphy. Driscoll, N.W. and Laine, E.E, 1996. Abyssal current influence
Geol. Soc. Am. Bull., 105: 1321-1345. on the southwest Bermuda Rise and surrounding region. Mar.
Bowland, C.L. and Rosencrantz, E., 1988. Upper crustal structure Geol., 130:231-263.
of the western Colombian Basin, Caribbean Sea. Geol. Soc. Driscoll, N.W., Hogg, J.R., Karner, G.D. and Christie-Blick, N.,
Am. Bull., 100: 534-546. 1995. Extensional tectonics in the Jeanne d'Arc basin: impli-
Browning, J.M. and Walper, J.L., 1982. A plate tectonic model cations for the timing of break-up between Grand Banks and
for the evolution of the Maracaibo-Peruvian Trough. Pac. Iberia. In: R.A. Scrutton, M.S. Stoker, G.B. Shimmield and
Geol., 16: 15-38. A.W. Tudhope (Editors), The Tectonics, Sedimentation and
Burke, K., Cooper, C., Dewey, J.F., Mann, J.E and Pindell, Palaeoceanography of the North Atlantic Region. Geological
J., 1984. Caribbean tectonics and relative plate motions. In: Society London Special Publ., 90: 1-28.
W.E. Bonini, R.B. Hargraves and R. Shagam (Editors), The Duncan, R.A. and Hargraves, R.B., 1984. Plate tectonic evolu-
Caribbean-South American Plate Boundary and Regional Tec- tion of the Caribbean region in the mantle reference frame.
tonics. Geol. Soc. Am. Mere., 162:31-64. In: W.E. Bonini, R.B. Hargraves and R. Shagam (Editors),
Case, J.E. and Holcombe, T.L., 1980. Geologic-tectonic Map The Caribbean-South American Plate Boundary and Regional
of the Caribbean region. U.S. Geol. Surv. Misc. Invest. Map Tectonics. Geol. Soc. Am. Mem., 162:81-94.
I- 1100, scale 1 : 2,500,000. Edgar, N.T., Ewing, J.I. and Hennion, J., 1971. Seismic refraction
Case, J.E., Shagam, R. and Giegengack, R.F., 1990a. Geology of and reflection in the Caribbean Sea. Am. Assoc. Pet. Geol.
the northern Andes; an overview. In: G. Dengo and J.E. Case Bull., 55: 833-870.
(Editors), The Geology of North America: The Caribbean Re- Edgar, N.T., Saunders, J.B., et al., 1973a. Initial Reports of the
gion. Decade of North American Geology, Vol. H, Geological Deep Sea Drilling Project, Vol. 15. Washington U.S. Govern-
Society of America, Boulder, CO, pp. 177-200. ment Printing Office, Washington, DC, 1137 pp.
Case, J.E., MacDonald, W.D. and Fox, P.J., 1990b. Caribbean Edgar, N.T., Holcombe, T. and Ewing, J., 1973b. Sedimentary
crustal provinces; seismic and gravity evidence. In: G. Dengo hiatuses in the Venezuelan Basin. Init. Rep. DSDP, 15:1051-
and J.E. Case (Editors), The Geology of North America: The 1062.
Caribbean Region. Decade of North American Geology, Vol. Escalante, G., 1990. The geology of southern Central America
H, Geological Society of America, Boulder, CO, pp. 15-36. and western Colombia. In: G. Dengo and J.E. Case (Edi-
Coffin, M.F. and Eldholm, O., 1994. Large igneous provinces: tors), The Geology of North America: The Caribbean Region.
crustal structure, dimensions and external consequences. Rev. Decade of North American Geology, Vol. H, Geological Soci-
Geophys., 32: 1-36. ety of America, Boulder, CO, pp. 201-230.
Damuth, J.E., 1980. Use of high frequency (3.5-12 kHz) Ewing, J., Edgar, N.T. and Antoine, J.W., 1971. Structure of the
echograms in the study of near-bottom sedimentation pro- Gulf of Mexico and Caribbean Sea. In: Maxwell (Editor), The
cesses in the deep-sea: a review. Mar. Geol., 38:51-75. Sea, Vol. 4, part 2. Wiley, New York, pp. 321-358.
Damuth, J.E. and Flood, R.D., 1984. Morphology, sedimentation Fox, EJ., Ruddiman, W.F., Ryan, W.B.F. and Heezen, B.C.,
processes, and growth pattern on the Amazon deep-sea fan. 1970. The geology of the Caribbean crust, I. Beata Ridge.
Geo-Mar. Lett., 3:109-117. Tectonophysics, 10:495-513.
Dewey, J.F. and Pindell, J.L., 1985. Neogene block tectonics of Heubeck, C. and Mann, E, 1991. Geologic evaluation of plate
Turkey and northern South America: continental applications kinematic models for the North-American-Caribbean plate
of the finite difference method. Tectonics, 4: 71-73. boundary zone. Tectonophysics, 191: 1-26.
Diebold, J.B., Stoffa, EL., Buhl, E and Truchan, M., 1981. Hinz, K., 1981. A hypothesis on terrestrial catastrophes. Wedges
Venezuelan Basin crustal structure. J. Geophys. Res., 86: of very thick oceanward dipping layers beneath passive conti-
7901-7923. nental margins. Geol. Jahrb., E22: 3-28.
Diebold, J.B., Driscoll, N.W., Abrams, L., Donnelly, T.W., Laine, Holcombe, T.L., Ladd, J., Westbrook, G., Edgar, N.T. and Bow-
E.E and Leroy, S., 1995a. Tectonic model for the origin and land, C., 1990. Caribbean marine geology; ridges and basins
evolution of the rough/smooth basement in the Venezuelan of the plate interior. In: G. Dengo and J.E. Case (Editors), The
Basin. EOS, 76: F615. Geology of North America: The Caribbean Region. Decade
Diebold, J.B., Driscoll, N.W., Abrams, L., Donnelly, T.W., Laine, of North American Geology, Vol. H, Geological Society of
E.E and Leroy, S., 1995b. Stratigraphy and tectonics of the America, Boulder, CO, pp. 231-260.

Hoorn, C., Guerrero, J., Sarmiento, G.A. and Lorente, M.A., Officer, C.B., Ewing, J.I., Hennion, J.E, Harkrider, D.G. and
1995. Andean tectonics as a cause for changing drainage Miller, D.E., 1959. Geophysical Investigations in the East-
patterns in Miocene northern South America. Geology, 23: ern Caribbean: summary of 1955 and 1956 cruises. In: L.H.
237-240. Ahrens et al. (Editors), Physics and Chemistry of the Earth,
Houtz, R.E. and Ludwig, W.J., 1977. Structure of Colombia Vol. 3. Pergamon Press, London, pp. 17-109.
Basin, Caribbean Sea, from profiler-sonobuoy measurements. Pindell, J.L. and Barrett, S.E, 1990. Geological evolution of the
J. Geophys. Res., 82: 4861-4867. Caribbean region: a plate-tectonic perspective. In: G. Dengo
Kasper, D.C. and Larue, D.K., 1986. Paleogeographic and tec- and J.E. Case (Editors), The Geology of North America: The
tonic implications of quartzose sandstones of Barbados. Tec- Caribbean Region. Decade of North American Geology, Vol.
tonics, 5: 837-854. H, Geological Society of America, Boulder, CO, pp. 405-432.
Kellogg, J.N. and Vega, V., 1995. Tectonic development of Pindell, J.L. and Dewey, J.E, 1982. Permo-Triassic reconstruc-
Panama, Costa Rica, and the Colombian Andes: constraints tion of western Pangea and the evolution of the Gulf of
from Global Positioning System geodetic studies and gravity. Mexico/Caribbean region. Tectonics, 1: 179-212.
In: P. Mann (Editor), Geologic and Tectonic Development of Pindell, J.L. and Tabbutt, K.D., 1995. Mesozoic-Cenozoic An-
the Caribbean Plate Boundary in Southern Central America. dean Paleogeography and regional controls on hydrocarbon
Geol. Soc. Am. Spec. Pap., 295: 75-90. systems. In: A.J. Tankard, R.S. Soruco and H.J. Welsink (Ed-
Klitgord, K. and Schouten, H., 1987. Plate kinematics of the itors), Petroleum Basins of South America. Am. Assoc. Pet.
Central Atlantic. In: B. Tucholke and P. Vogt (Editors), The Geol. Mem., 62: 101-128.
Geology of North America: The Western Atlantic Region. Pindell, J.L., and 6 others, 1988. A plate-kinematic framework
Decade of North American Geology, Vol. M, Geological Soci- for models of Caribbean evolution. Tectonophysics, 155: 121-
ety of America, Boulder, CO, pp. 351-378. 138.
Kolla, V., Buffler, R.T. and Ladd, J.W., 1984. Seismic stratigra- Schlich, R., Rotstein, Y. and Schaming, M., 1993. Dipping
phy and sedimentation of Magdalena fan, southern Colombian basement reflectors along volcanic passive margins - - new
Basin, Caribbean Sea. Am. Assoc. Pet. Geol. Bull., 68: 316- insight using data from the Kerguelen Plateau. Terra Nova, 5:
332. 157-163.
Kroonenberg, S.B., Bakker, J.G.M. and van de Wiel, M., 1990. Schumm, S.A., 1991. River response to baselevel change: impli-
Late Cenozoic uplift and paleogeography of the Colombian cations for sequence stratigraphy. J. Geol., 101: 279-294.
Andes: constraints on the development of high-Andean biota. Sigurdsson, H., Leckie, R.M., Acton, G.D., et al., 1997. Pro-
Geol. Mijnbouw, 69: 279-290. ceedings of the Ocean Drilling Program, Initial Report, 165.
Ladd, J.W. and Watkins, J.S., 1980. Seismic stratigraphy of the College Station, TX, 865 pp.
western Venezuela Basin. Mar. Geol., 35:21-41. Sinton, C.W., Duncan, R.A. and Storey, M., 1993. 4~
Lu, R.S. and McMillen, K.J., 1982. Multichannel seismic survey ages from Gorgona Island, Colombia, and the Nicoya Penin-
of the Colombia Basin and adjacent margins. In: J.S. Watkins sula, Costa Rica. EOS, 74: 553.
and C.L. Drake (Editors), Studies in Continental Margin Geol- Stoffa, EL., Mauffret, A., Truchan, M. and Buhl, E, 1981.
ogy. Am. Assoc. Pet. Geol. Mem., 34: 395-410. Sub-B" layering in the southern Caribbean: the Aruba Gap
Ludwig, W.J., Houtz, R.E. and Ewing, J.I., 1975. Profiler- and Venezuela Basin. Earth Planet. Sci. Lett., 53: 131-146.
sonobuoy measurements in Colombian and Venezuelan Basins, Talwani, M., Windisch, C., Stoffa., EL., Buhl, E and Houtz,
Caribbean Sea. Am. Assoc. Pet. Geol. Bull., 59:115-123. R.E., 1977. Multichannel seismic study in the Venezuelan
Lugo, J. and Mann, P., 1995. Jurassic-Eocene tectonic evolu- Basin and the Curacao Ridge. In: M. Talwani and W.C. Pitman
tion of Maracaibo Basin, Venezuela. In: A.J. Tankard, R.S. III (Editors), Island Arcs, Deep Sea Trenches and Back-Arc
Soruco and H.J. Welsink (Editors), Petroleum Basins of South Basins. Am. Geophys. Union, M. Ewing Symp. Ser., 1: 83-98.
America. Am. Assoc. Pet. Geol. Mem., 62: 699-726. Tankard, A.J., Soruco, R.S. and Welsink, H.J., 1995. Petroleum
Macelli, C.E., 1988. Cretaceous paleogeography and depositional Basins of South America. Am. Assoc. Pet. Geol. Mem., 62,
cycles of western South America. J. S. Am. Earth Sci., 1: 373- 761 pp.
418. Westall, E, Rossi, S. and Mascle, A., 1993. Current-controlled
Malfait, B.T. and Dinkelman, M.G., 1972. Circum-Caribbean sedimentation in the Equatorial Atlantic: examples from the
tectonic and igneous activity and the evolution of the southern margin of the Guinea Plateau and the Romanche
Caribbean plate. Geol. Soc. Am. Bull., 83:251-272. Fracture Zone. Sediment. Geol., 82: 157-171.
Mauffret, A. and Leroy, S., 1997. Seismic stratigraphy and struc- White, R.S. and McKenzie, D., 1995. Mantle plumes and flood
ture of the Caribbean Sea. Tectonophysics, 283:61-104. basalts. J. Geophys. Res., 100: 17,543-17585.
McCave, I.N. and Tucholke, B.E., 1986. Current controlled sed- Wilson, J.T., 1966. Are the structures of the Caribbean and
imentation in the Western North Atlantic. In: P. Vogt and Scotia arcs analogous to ice rafting? Earth Planet. Sci. Lett., 1:
B. Tucholke (Editors), The Geology of North America: The 335-338.
Western Atlantic Region. Decade of North American Geol- Worthington, L.V., 1966. Recent oceanographic measurements in
ogy, Vol. M, Geological Society of America, Boulder, CO, pp. the Caribbean Sea. Deep-Sea Res., 13: 731-739.
451-478. Worthington, L.V. and Wright, W.R., 1970. North Atlantic ocean
Mutter, J.C., Talwani, M. and Stoffa, P.L., 1982. Origin of atlas of potential temperature and salinity in deep water. The
seaward-dipping reflectors in oceanic crust off the Norwegian Woods Hole Oceanographic Atlas Series Volume II.
margin by subaerial sea-floor spreading. Geology, 10: 353- Wright, W.R. and Worthington, L.B., 1970. The water masses of
357. the North Atlantic ocean: a volumetric census of temperature
Officer, C.B., Ewing, J.I., Edwards, R.S. and Johnson, H.R., and salinity. Serial Atlas of the Marine Environment, Folio 19.
1957. Geophysical investigations in the Eastern Caribbean: American Geographical Society, New York, 8 pp. and 7 plates.
Venezuelan Basin, Antilles island arc, and Puerto Rico trench. Wunsch, C. and Grant, B., 1982. Towards the general circulation
Bull. Geol. Soc. Am., 68: 359-378. of the North Atlantic Ocean. Prog. Oceanogr., 11: 1-59.

C h a p t e r 21
Neogene Intraplate Deformation of the Caribbean Plate at the

Beata Ridge
A L A I N M A U F F R E T and S Y L V I E L E R O Y
We have studied all the seismic profiles existing in the Beata ridge area, in addition to Seabeam maps, to determine the
tectonics of this structure and its relationships with the adjacent areas. The basis of this work is the multichannel Casis seismic
survey carried out by the R/V Nadir in 1992. These seismic lines are migrated and clearly show evidence of compression and
transpression. The Presqu'~le du Sud d'Hispaniola is an uplifted part of the volcanic igneous province that formed the Caribbean
plate during Cretaceous time. This region was initially a part of the thick Beata volcanic plateau that collided with the central part
of Hispaniola. A Seabeam map and single-channel seismic lines from the Seacarib 1 cruise show the active collision between the
northeastern tip of Beata ridge and the western termination of the Muertos trough. Structural analysis indicates that the Muertos
trough is an Eocene feature that has been reactivated in Recent times. The progressive emergence of the Muertos prism and its
onland extension results from vertical stacking by thrusting of different parts of the prism. Several compressional structures can be
seen on the eastern flank of the Beata ridge. The importance of these structures decreases systematically towards the south. One of
these structures, the Taino ridge, was surveyed in detail during the Casis cruise. We describe reverse faults, pop-up and strike-slip
faults. These tectonic features are compatible with a NE-SW compressive stress. A detailed site survey of the Aruba Gap was also
performed during the Casis cruise. We show again compressional and wrench faults and an increase of the deformation towards
the north. In contrast, we found no evidence of any compressional deformation on the western side of the Beata ridge. Here, steep
NE-SW scarps shown by Seabeam maps, are predominant. We conclude from this tectonic framework that the Beata ridge has
been deformed by compression and strike-slip faulting since the Early Miocene (23 Ma) by a NE-SW-oriented compressive stress.
The Beata ridge is progressively uplifted from the south to the north up to the emergence of the Presqu'~le du Sud. However,
the Beata ridge is a Cretaceous plateau and initial topography must be taken into account. The Beata ridge is placed in the
regional tectonic framework and we show that the compression of the ridge is probably connected to the Sinu subduction zone
in Colombia. We distinguish between the Colombian and the Venezuelan microplates separated by the Beata compressional zone.
The former drifts towards the northeast faster than the latter. From our structural analysis we deduce 0.9 cm/yr of relative motion
between the two plates.
INTRODUCTION and Watkins, 1978; L a d d et al., 1981, 1984, 1990)

north of South A m e r i c a ( C o l o m b i a n and Venezue-
The Caribbean plate is a large volcanic province lan d e f o r m e d belts, Fig. 1) and south of Puerto
probably formed, during the Cretaceous, in the Pa- Rico (Muertos trench, Fig. 1) m a y result from this
cific O c e a n (Duncan and Hargraves, 1984). This c o m p r e s s i o n which increases towards the west.
plate is m o v i n g eastwards relative to the North The D S D P results of Leg 15 (Edgar et al., 1973b)
A m e r i c a n ( N O A M ) and South A m e r i c a n ( S O A M ) and the O D P results of L e g 165 (Scientific Party, Leg
plates (Pindell and Barrett, 1990). The C a r i b b e a n 165, 1996) indicated that the volcanic b a s e m e n t of
plate is d e l i m i t e d (Fig. 1) to the north by a left-lat- the C a r i b b e a n plate was f o r m e d during a short period
eral strike-slip fault zone ( C a y m a n - P u e r t o Rico fault of the Cretaceous time (late Turonian to Campanian,
system) and to the south by a c o m p l e x set of right- 8 8 - 7 4 Ma). The same volcanic rocks outcrop in the
lateral faults. In addition, the N O A M and S O A M Presqu'~le du Sud and B a h o r u c o Peninsula of His-
plates are slowly converging with a present pole of paniola (Maurasse et al., 1979) and Curaqao Island
rotation located near the M i d - A t l a n t i c ridge (Pindell (Klaver, 1987) and these areas are considered to
and Barrett, 1990; Mtiller and Smith, 1993; Mtiller be uplifted pieces of the C a r i b b e a n plateau. In the
et al., 1996). S e d i m e n t a r y d e f o r m e d belts (Ladd central part of the C a r i b b e a n Sea, the Beata ridge,

628 A. MAUFFRET and S. LEROY
Fig. 1. Plate tectonic frameworkof the Caribbean plate. The Gonave microplate (Rosencrantz and Mann, 1991) is indicated.
2-4 km deep, has a triangular shape and lies (Fig. 2) BATHYMETRY OF THE CENTRAL CARIBBEAN BASIN
between the Colombian basin to the west and the
Venezuelan basin to the east. The northern parts of Since 1980 (Case and Holcombe, 1980) no new
the Colombian and Venezuelan basins have been bathymetric map of this region has been published.
named the Haiti and Dominican sub-basins, respec- In 1985 a Seabeam survey was performed during
tively, and we gave new names (from Indian tribes) the French Seacarib cruise (1985, R/V Charcot).
to the main ridges that composed the Beata ridge. Some data were published (Mercier de Lepinay et
The Bahoruco Peninsula is the northern prolongation al., 1988; Jany et al., 1990; Mauffret and Jany,
of the Beata ridge, whereas this ridge is separated 1990) but the surveys of the Beata ridge as a
from the deformed margin of the South American whole are presented for the first time in this paper
plate by the Aruba Gap. (Fig. 3). These data are integrated in the new map
Many speculative models have been proposed for presented here (Fig. 3) that includes a compilation of
the formation of the Beata ridge: normal faulting previous data provided by the Marine Geophysical
(Fox et al., 1970; Fox and Heezen, 1975; Holcombe Data Center (Leroy, 1995).
et al., 1990); a buoyant thick oceanic plateau which The Muertos trough lies in the northern part of
resists subduction (Burke et al., 1978) or reverse the central Caribbean region between 5 and 4 km
faulting related to a transpressive motion (Vitali, deep. This depression is delimited towards the north
1985; Mauffret et al., 1994). New multichannel seis- by a deformed slope. Towards the west the Muertos
mic profiles, acquired during the Casis cruise per- trough undergoes a prominent bend then disappears
formed in 1992 on the R/V Nadir in the Caribbean near Hispaniola. The Bahoruco Peninsula is bounded
Sea (Fig. 2), demonstrate strong transpressive tec- towards the east by a steep scarp from the coast to
tonics of a former volcanic plateau (Mauffret et al., 3 km deep. The deep part of the scarp and the
1994; Leroy, 1995; Leroy and Mauffret, 1996). continental rise, below 3 km deep, are interrupted
The Beata ridge was studied in the framework by several seamounts that trend north-south. The
of a comprehensive work on the geophysical and southern tip of the Bahoruco Peninsula extends
geological data of the Caribbean Sea (Leroy, 1995) southward by a spur from the coast to 2 km depth.
to promote an ODP Leg on the deep structure of the The Beata ridge is bounded towards the northwest
Caribbean igneous province. by a steep scarp that trends NE-SW. At the foot of

NEOGENE INTRAPLATE DEFORMATION OF THE CARIBBEAN PLATE AT THE BEATA RIDGE 629
Tairona Ridge, uertos
Dominicansub basin
31 TainoRidge
DSDP 151
153
Warao Rise ta Platea~''~
Colombia Basin
Fig. 2. The northern parts of the Colombian and Venezuelan basins are renamed the Haiti and Dominican sub-basins, respectively. The
Warao rise is a buried feature that bounds the Haiti sub-basin to the south. The backbone of the Beata ridge is formed by the DSDP 151
and Tairona ridges. Warao, Taino and Tairona are the names of Indian tribes. The position of the DSDP Sites (Legs 4 and 13) and ODP
Sites (Leg 165) are indicated. The seismic tracks of the Casis cruise is also shown.
the scarp lies the Haiti sub-basin that is surrounded American deformed belt and forms a sill between
by the 4.2 km bathymetric contour. This basin is the Colombian and Venezuelan basins.
delimited towards the west by the Hess escarpment
and towards the north by the steep continental slope
of Haiti and the Haiti plateau that trends N W - S E . SEISMIC CONTROLS
The eastern boundary of the Beata ridge with the
Venezuelan basin is subdued, but two prominent In addition to the Casis cruise we examined all
features, the Taino ridge 3 to 4 km deep, and the the MCS (multichannel seismic profiles) and SCS
Beata plateau surrounded by the 4-km-bathymetric (single-channel seismic profiles): from the Univer-
contour outline the southeastern limit of the ridge. sity of Texas at Austin, Shell, Institut Franqais du
The central part of the Beata ridge, delineated by Pdtrole for the MCS profiles; from the Lamont
the 3-km-bathymetric contour, is formed by two Doherty Earth Sciences Observatory (Vema, R / V
prominent features: Tairona and DSDP 151 ridges Conrad data and the monitors of the Ewing 9501
that trend north-south. The Aruba Gap, 4 km deep, cruise); Texas A & M (Alamino cruise) and the
is located between the Beata ridge and the South Seacarib 1 cruise (SCS profiles) (Fig. 4). All these

Fig. 3. New bathymetric map of the central part of the Caribbean Sea (from Leroy, 1995, modified). The Seacarib 1 Seabeam surveys
(position indicated) were incorporated to the conventionalbathymetric data base (NGDC).
profiles were digitized and converted in depth with base of this interval. The third seismic interval cor-
a velocity curve derived from DSDP results and responds to lithified chalks, cherts, limestones and
velocity analysis of the Casis data. black shales which rest upon Santonian to Coniacian
basalts. Horizon B" correlates with these basalts. In
addition, we have identified in the Aruba Gap area a
SEISMIC STRATIGRAPHY lower sedimentary unit between an equivalent of B"
and a deeper horizon (V, Fig. 5B). In the volcanic
The seismic stratigraphy of the central Venezue- crust several reflectors have been identified (sub-B"
lan basin was defined in the early work of Ladd R reflectors and Moho, Fig. 5A), but a study of
and Watkins (1980). The DSDP results reveal the the deep crust is beyond the scope of this paper
presence of four main seismic intervals (Fig. 5). An (Mauffret and Leroy, 1997).
upper seismic interval in Hole 31 (Bader et al., 1970) The upper seismic interval has a variable thick-
and Holes 146 and 153 (Edgar et al., 1973b), corre- ness, relatively thin in the flank of the Beata ridge,
sponds to lithic unit 1, a chalk marl ooze and clay of and increasing to 2 km thick in the southern part of
Early Miocene to Recent age (eM, 23 Ma; Fig. 5). the Venezuelan and Colombian basins where it fills a
The second seismic interval corresponds to a Middle trench related to the South American deformed belt
Eocene to Early Miocene radiolarian chalk and ra- (Talwani et al., 1977; Biju Duval et al., 1982b). In
diolarian chalk unit. Horizon A" (Fig. 5) forms the the Colombian basin, the layer of Early Miocene to

Fig. 4. Seismic tracks used in this study.
recent sediments is thick in relation with the Mag- of several hiatuses in the DSDP holes that have
dalena deep-sea fan (Kolla et al., 1984). The DSDP been related to strong Early Miocene currents which
Sites and the seismic profiles indicate that this inter- were active as the Caribbean Sea opened towards
val consists of turbiditic sediments in basins, but a the Pacific Ocean (Edgar et al., 1973a; Holcombe
pelagic composition is inferred on the Beata ridge. and Moore, 1977). The hummocky aspect of the
The second seismic interval corresponds to a Early Miocene reflector is widespread and the unit
pelagic unit. The upper part, and in some places has been described as a prominent horizon by Houtz
the entire interval (Fig. 5B), is chaotic. This layer and Ludwig (1977). The Early Miocene to Middle
is current-controlled, as confirmed by the presence Eocene seismic interval dips toward the South Amer-

t,~
>,

Fig. 5. Seismic stratigraphy tied with the DSDP results. Except for (B) all the seismic profiles cross exactly the DSDP Site area. The main reflectors are Early Miocene (eM), Eocene (A ~I) and Santonian to
Coniacian (B") in age, respectively. Note the thickening of the A"-B 'I interval from west (A, B) to east (C, D). The deep reflectors (V, sub-B", R and Moho) are described in another paper (Mauffret and
-]
Leroy, 1997). The location of profiles is indicated in inset.
t-
7z
ican deformed belt and this disposition indicates that ner in the Aruba Gap area where it was penetrated
compressional deformation along South America oc- during drilling at DSDP Site 153 (Fig. 5A and B).
curred since the Early Miocene epoch. At this time A prominent hard ground is correlated with a late
the floor of the Caribbean Sea was completely de- Maastrichtian hiatus at the DSDP Site 153. A similar
formed, with a general tilting towards the south hard ground was described at DSDP Site 151, but
related to the formation of trenches (Biju Duval et here the Paleocene directly overlies the Santonian
al., 1982b). The structure of the young accretionary sediments and basalts (Edgar et al., 1973b).
prism indicates a large amount of offscraping (Ladd In a basin located to the north of the DSDP Site
et al., 1984) and tomography data indicate a long 153, the A " - B " interval correlates in thickness and
slab extending beneath northwest South America seismic facies with units encountered at the DSDP
(Hilst and Mann, 1994). However, the south dip Site 153 (Fig. 5B). However, the acoustic basement
of the Venezuelan basin crust is partly the result (V, Fig. 5) is very different from the typical smooth
of original construction of the Cretaceous volcanic B" reflector drilled at DSDP Site 153, and is overlain
plateau that thins from north to south (Diebold et al., by a sedimentary layer with a compressional velocity
1999). The current-controlled layer cannot be identi- of 3.9 k m / s (Fig. 5B).
fied on the top of the Beata ridge and the transparent
facies of the layer overlying the top of the ridge
suggests a pelagic environment (Fig. 5E). NORTHERN BOUNDARY OF THE CENTRAL
The A"-B" interval is evident in the Venezuelan CARIBBEAN REGION
basin and in the western part of the Colombian
basin (Ladd and Watkins, 1980; Bowland, 1993). The present-day northern limit of the Caribbean
However, horizon A", which correlates with Middle plate is the Oriente fault Fig. 1), the extension of the
Eocene chert, is not clearly identified in the eastern Oriente fault in Hispaniola that lies in the Cibao val-
Colombian basin and on the Beata ridge (Fig. 5E). ley (Fig. 6; Calais et al., 1992; Russo and Villasenor,
The A " - B " interval is thick in the Venezuelan 1995) and the Puerto Rico trench (Masson and Scan-
(Fig. 5D) and Colombian basins and also on the lon, 1991). A strain partitioning may occur and a
eastern flank of the Beata ridge (Fig. 5C), but thin- component of compression may be absorbed in the
Fig. 6. Tectonic framework of Hispaniola. The main boundary between the Caribbean plate and the North American plate is located along
the Oriente fault and the Cibao valley. However, a strain partitioning with a compressional boundary is possible north of Hispaniola. The
central part of Hispaniola is occupied by island arc crust (Central Cordillera), deformed sedimentary belts (Peralta and Neiba) and basins
(San Juan and Enriquillo). The Enriquillo basin bounds the Presqu'~le du Sud that is formed by Cretaceous volcanic rocks identical
to that of the Caribbean basement (Maurasse et al., 1979). The Presqu'~le du Sud is split into two parts by the left-lateral strike-slip
Enriquillo fault. This fault is related to the Navassa pull-apart basin (Mann et al., 1995). The southern offshore part of the Presqu'~le du
Sud is severely deformed by compression and transpression (Bien-Aime Momplaisir, 1986). The Haiti sub-basin has probably a quasi
oceanic crust. East of the Beata ridge the Seacarib Seabeam survey is indicated. The Muertos prism can be divided into three parts by
recent reverse faults. The location of Figs. 7 and 8 is shown.

r
t-
Fig. 7. The Haiti sub-basin has a thin crust as shown by the refraction data (refraction line 36W, Ewing et al., 1960). The sedimentary layers onlap a wedge at the base of the slope indicating old
deformation. The bottom of the wedge is indicated by the white line and arrows (flat reflector). In contrast the upper slope is actively deformed with the Ile-a-Vache anticline and a deep syncline bounded by
reverse faults (Bien-Aime Momplaisir, 1986). The location of the seismic profile is indicated in Fig. 6.
sedimentary deformed belt that lies north of Hispan- this wedge was formed during old compressional
iola (Fig. 6; Austin, 1984; Dillon et al., 1992). The tectonics and the present deformation is restricted
Caribbean plate boundary has been located along to the upper margin. Nevertheless the sedimentary
the north of Hispaniola since the Miocene and be- wedge may also be formed by slope breccias at the
fore this boundary was probably placed along the base of the scarp and the chaotic aspect of the upper
Oriente fault, the Central Cordillera and Muertos sedimentary layers suggests slumped sediments and
trough (Mercier de Lepinay, 1987; Pindell and Bar- erosional products.
rett, 1990). Hispaniola can be divided into four main The presence of dipping reflectors (Fig. 8) into
blocks. A Septentrional block separated from the the acoustic basement of the Beata plateau indicates
Eastern and Central Cordilleras by the Cibao valley. the volcanic formation of this feature. The deepest
These cordilleras underwent a Mesozoic island arc sedimentary unit onlaps the basement and this seis-
tectonism (Mercier de Lepinay, 1987; Lebron and mic configuration suggests that the basement relief is
Perfit, 1994). A central area, including the Peralta old. This basement and the lower sedimentary layers
flysch, the San Juan basin, the Sierra de Neiba that are tilted towards the NNE, whereas the thickness
is an accretionary prism since the Eocene up to the of the recent layers increases in the same direction
Pliocene (Mann and Lawrence, 1991; Mann et al., (Fig. 8). A wedge of deformed sediments is evident
1991a,b, 1995). This third block is delimited by the at the base of the western slope of the Sierra de Ba-
Enriquillo depression (Fig. 6). The southern block horuco. Thickening of the recent sedimentary layers
is formed by the Presqu'ile du Sud d'Haiti that is and tilting of the lower layer in the basin, reverse
an uplifted portion of the Cretaceous Caribbean ig- fault and d6collement suggest a compressional ori-
neous province (Maurasse et al., 1979). This block gin for the wedge. This compression is recent but
collided with the northern block recently (Mercier de probably inactive at the present day as shown by
Lepinay et al., 1988). The Early Pliocene strata are other seismic profiles crossing the southern part of
folded in the western part of the Enriquillo basin and the deformed wedge (see later). Consequently the
the collisional process is probably presently active Haiti plateau and the Haiti sub-basin had an east-
Vila et al., 1990). The Presqu'ile du Sud is cut by wards motion relative to the Presqu'ile du Sud, but
the Enriquillo fault that extends towards the west this motion is presently nonexistent.
to Jamaica and the Cayman spreading center (Sykes The Muertos trough has been described several
et al., 1982; Rosencrantz and Mann, 1991). The times (Ladd and Watkins, 1978; Biju Duval et al.,
Navassa trough (Fig. 6) is a pull-apart basin located 1982a; Ladd et al., 1981, 1990). The seismicity
along the left-lateral Enriquillo fault (Mann et al., shows a steep Benioff zone dipping to 125 km depth
1995). (Bryne et al., 1985; Russo and Villasenor, 1997). A
The margins of the Presqu'ile du Sud were stud- Seabeam survey (Mercier de Lepinay et al., 1988;
ied in detail by Bien-Aime Momplaisir (1986) and Ja W, 1989; Mauffret and Jany, 1990; Vila et al.,
she showed that the northern and southern margins 1990) of the western part of the Muertos trough
are affected by a compressional or/and transpres- shows the relationships between the front of the
sional tectonics governed by a N40-45 compressive accretionary prism and the Sierra de Martin Garcia
stress. The Presqu'*le du Sud can be assimilated (Fig. 9A). However, this front is connected to the
to a giant positive flower structure. A seismic pro- Peralta flysch belt in some publications (Mann and
file (Fig. 7) shows the deformation of the upper Lawrence, 1991; Mann et al., 199 la,b, 1995). In
margin with a broad syncline and a thrust of the the Ocoa Bay (Fig. 9B) the interpreted boundary of
sedimentary cover of this basin on the Tle-~-Vache the Muertos accretionary prism is located between
structure. A narrow anticline is located at the top a (piggyback?) basin and a deformed zone (writ-
of this feature. At the base of the steep slope lies ten communication of S.R. Lawrence, co-author of
the Haiti sub-basin. The correlation of the seismic Chapter 12) that in fact is located in the inner part
profile with refraction results (Ewing et al., 1960) of the Muertos prism. Moreover the 50-km contour
indicates that this basin has a thin oceanic crust. The of the Muertos Benioff zone is located near Ocoa
Haiti plateau is a thick block related to the Creta- Bay (Russo and Villasenor, 1997). The Muertos
ceous igneous province, whereas the thin crust of accretionary prism is divided into three parts: the
the Haiti sub-basin seems to be not affected by the San Pedro basin, an upper prism and a lower prism
Cretaceous volcanic event. However, the presence (Fig. 9A). Two out of sequence thrust zones sepa-
of some intra basement reflectors (sub-B"?; Fig. 7) rate the three units. A detailed study (Biju Duval et
may suggest a weak volcanic contamination of the al., 1982a) of the San Cristobal basin, an onshore
thin crust in the Haiti sub-basin. A small sedimen- extension of the San Pedro basin, indicates that the
tary wedge lies above a flat reflector (d6collement?) Muertos trough is active since the Eocene, but the
at the base of the slope. However, the sedimentary uplift of the San Cristobal basin is related to a Late
layers of the basin onlap this wedge. Consequently Miocene-Recent compressional event. The Eocene

636 A. M A U F F R E T and S. L E R O Y
Z'-6
~,,..~
9 ,.~
o "~
'-~ O
~ ~zZ
~;_v,
~ ~ ~
~,...~
o
"~ ~ 9
,-~ .,. ~
~''-d .

paleoprism (Peralta belt and Sierra de Neiba) has crantz and Mann, 1991; Mann et al., 1995). The
been shortened and uplifted by this recent event Enriquillo depression connects this microplate to the
that is related to the collision of the Beata ridge Dominican sub-basin. We propose that the Gonave
with the northern block of Hispaniola (Biju Duval et microplate belongs to the Venezuelan plate separated
al., 1982a). The uplift is presently occurring by the from the Colombian plate by the Enriquillo fault and
thrusting of the San Pedro basin over the upper prism the Beata fault system (see later).
that is underthrust by the lower prism (Fig. 10A, Biju
Duval et al., 1982a). North of 18~ the lower prism
disappears and the upper prism is directly adjacent EASTERN BOUNDARY OF THE BEATA RIDGE
to the undeformed basin (Figs. 9A, 10C). The slope
off Bahoruco Peninsula is very steep from the coast South of the Muertos-Beata collision zone pre-
to 2 km depth; then several hills, that trend north- viously described several hills trend north-south
south, form an extension of the Beata ridge. The (Fig. 9A). On a seismic profile (Fig. 11A), already
magnetic map contoured during the Seacarib cruise published in Ladd et al. (1981), the two western
(Fig. 9C) shows that some of these hills are made of seamounts are conical but the third seamount is
volcanic rocks. These seamounts are separated from asymmetrical. The thinning of the A"-B" interval
the prism by the narrow Muertos trough (Fig. 10A, (old sedimentary fill, Fig. 11) away from structures
B). However, the Beata seamounts are never in con- suggest that these features are contemporaneous with
tact with the prism and the easternmost seamount the formation of the volcanic plateau. The Beata
disappears abruptly (Fig. 10C). On the other hand ridge has a thick crust that has been underplated
the magnetic map (Fig. 9C) shows that N E - S W (Mauffret and Leroy, 1997). This underplating in-
magnetic lineations of the Caribbean basement are duced an uplift and a coeval rifting (Diebold et al.,
visible beneath the prism and a positive anomaly 1999; Driscoll and Diebold, 1999). However, the
(431 nanoteslas, Fig. 9C) may be correlated with sedimentary cover of the central basin is upturned
a buried seamount. The northern tip of the Beata and is now on the top of the structure (Fig. l lA).
ridge may be cut by a right-lateral strike-slip fault This configuration is abnormal for a tilted block
(Fig. 9C). The Enriquillo basin that is the onshore resulting from a normal faulting. The configuration
extension of the Muertos trough is a ramp valley of the sedimentary layers suggests an old feature
(Mann et al., 1991 a,b) bounded by two facing thrusts reactivated and recently uplifted (1 km, Fig. 11A).
(Fig. 6). It is probable that the Enriquillo depression Although this profile is not migrated we suggest that
is floored by the Cretaceous volcanic basement of a transpression is the cause of the uplift. The struc-
the Presqu'~le du Sud because volcanic Cretaceous ture previously described forms a north-south ridge
basalts have been observed northwest of the En- that is offset to the south (Fig. 11D). In the second
riquillo basin (Pierre Payen anticline, Fig. 6; (Vila seismic profile (Fig. 11B) an asymmetrical structure
et al., 1988). Moreover, the magnetic map (Fig. 9C) separates two regions of the Dominican sub-basin
shows the Caribbean basement below the Muertos with a different seismic stratigraphy. The western
prism. Consequently, the Cretaceous volcanic base- region shows several sub-B" reflectors, whereas the
ment may collapse by normal faulting produced by basement of the eastern region is void of these reflec-
flexural effects found in the bulge related to under- tions. Such a prominent contrast between the eastern
thrusting and this basement finally subducts beneath and the western part suggests a lateral displacement.
the Neiba and Muertos prisms. On land these normal The western region seems to be transported from the
faults have been recently described along the north- south where the sub-B" reflections are also promi-
eastern flank of the Sierra de Bahoruco (Pubellier nent (Fig. 11C), implying a right-lateral component
et al., 1999). The Seacarib profiles (Fig. 10) are of a transpressional fault. The vertical uplift of the
not migrated and we cannot define if the faults that fault is estimated to be 0.6 km. In addition this
bound the Beata seamounts are reverse or normal. profile shows an inversion of basin related to a com-
However, the southwards extension of these faults is pressional or transpressional event. The third profile
clearly reverse (see next paragraph). In conclusion, is a regional seismic section (Ewing 9501 cruise,
the northern boundary of the Presqu'~le du Sud- profile 1321; Fig. 11C). A pop-up with a probable
Beata ridge may be a reverse fault, but the subduc- reverse fault dipping towards the NNW is identified.
tion of this block implies a final normal faulting. Moreover, westwards steep dipping reflectors can
These normal faults and the strike-slip faults transfer be seen in the crust. Although the vertical throw
large fragments of the Cretaceous volcanic plateau of the pop-up structure is only 0.25 km the prob-
to the Gonave microplate (Fig. 1). This microplate able reverse fault affects the whole crust. These
was defined between the Cayman spreading center, transpressional features were also seen on several
the Oriente fault, the Plantain Garden-Enriquillo single-channel seismic lines. In conclusion, we iden-
fault and the western coast of Hispaniola (Rosen- tify a transpressional zone with a strong component

Fig. 9. Relationship between the Muertos prism and the geologic features of Hispaniola. (A) Seacarib Seabeam survey completed by conventional bathymetric data. The lower prism of Muertos is formed by
elongated anticlines. The toe of the prism turns abruptly towards the north when the first seamounts of the Beata ridge appear and the Muertos trough is very narrow. Then the lower prism disappears as well
as the Beata seamounts. The slope of the prism is steep and the toe is related to the recent deformed Sierra de Martin Garcia. The location of the seismic profiles illustrated in Figs. 10 and 11A is shown. (B)
The toe of the Muertos prism cannot be located in the Ocoa Bay as presumed by Mann et al. (1991a,b, 1995), Mann and Lawrence (1991). The Muertos prism and its forearc (San Pedro and San Cristobal)
were formed during the Eocene (Biju Duval et al., 1982a) but were reactivated recently by the collision of the Beata ridge. The San Pedro basin overthrusts the upper prism and the lower prism disappears
beneath the upper prism. This successive stacking generates the uplift and the emergence of the prism. (C) Magnetic map performed during the Seacarib 1 cruise. The magnetic basement of the Caribbean
basin can be seen beneath the Muertos prism. The southern Beata seamount shows a correlation with a positive magnetic anomaly, but the northern seamount is orthogonal to the magnetic grain. We cannot
exclude a rotation of this seamount that is parallel to the Muertos trough (see A). A prominent magnetic anomaly (431 nanoteslas) is located below a reentrant of the prism. This anomaly maybe correlated
with a seamount that subducts. A strike-slip fault may transport the tip of the Beata ridge beneath the Muertos prism.
Fig. 10. Single-channel seismic profile from the Seacarib 1 cruise. (A) The upper prism overthrusts the lower prism, the Muertos trough
is narrow and deformed and is bounded by a Beata seamount. (B) The folds of the lower prism are particularly evident. (C) The lower
prism disappears as well as the easternmost Beata seamount. This seamount may have subducted; see the magnetic map in Fig. 9C. The
locations of the seismic profiles are indicated in Fig. 9A).
of compression. This transpression is presently ac- BEATA PLATEAU

tive. The height of the d e f o r m e d features increases
towards the north where the collision of the Beata This 0 . 8 - k m - h i g h structure, n a m e d by Case and
ridge is in process. H o l c o m b e (1980), trends N N W - S S E and is isolated


Fig. 12. Seismic profile crossing the Beata Plateau. The Eocene A" horizon has been piston-cored (Talwani et al., 1966; Edgar et al.,
1971). Although this profile is unmigrated the reverse motion of the faults is clear. See the inset map for profile location.
f r o m the B e a t a ridge (Fig. 12B). A s e i s m i c profile TAINO RIDGE

shows two steps that are p r o b a b l y reverse. H o r i z o n
A" is e x p o s e d on the h i g h e s t scarp. E a r l y to Mid- T h e Taino ridge is l o c a t e d on the e a s t e r n flank
dle E o c e n e was i n d e e d piston c o r e d on this scarp of the B e a t a ridge (Fig. 13). T h e D S D P Site 31
(Talwani et al., 1966; E d g a r et al., 1971). 11 k m of ( B a d e r et al., 1970) allows us to c a l i b r a t e the s e i s m i c
s h o r t e n i n g (20%) is e s t i m a t e d for this 4 8 - k m - w i d e profiles and w e c o r r e l a t e the M i o c e n e - O l i g o c e n e
structure if w e a s s u m e that the B" reflector was b o u n d a r y with a p r o m i n e n t reflector (eM, 23 Ma;
initially flat. Fig. 5C and Fig. 14). A d e t a i l e d survey was per-
f o r m e d d u r i n g the Casis cruise. T h e s e i s m i c profiles
Fig. 11. (A) This seismic profile, already published by Ladd et al. (1981), shows two conical seamounts and a third western seamount
that is asymmetrical. The sedimentary cover of the adjacent basin are tilted and uplifted on the top of the seamount that indicates a
transpression. (B) UTIG (University of Texas at Austin) profile cruise courteously provided by J. Austin. The two basins separated by
an asymmetrical seamount show a different seismic stratigraphy. The eastern basin is underlain by a basement with several internal
reflections (sub-B"). If we compare this profile with the seismic profile illustrated in (C), this basin may be transported from the south
along a right-lateral strike-slip fault. (C) Seismic profile shot during the Ewing 9501 cruise courteously provided by J. Diebold. The
complete regional seismic profile is also shown in Driscoll and Diebold (1999). Pop-up related to a transpression. Observe the steep
dipping reflector that may correspond to a reverse fault at a crustal scale. The Moho reflection is shown at 10 s two-way travel time. The
thinning of the old sedimentary fill away from the structures (A, B) indicates that these features were formed during the construction of
the Cretaceous volcanic plateau, but a recent reactivation is evident. The uplift of the hills can be estimated to be 0.25 km in the south and
1 km in the north. See the inset map (D) for profile location. This depth to basement (B") map shows the eastern flank of the Beata ridge.
The north-south trend of the structures is evidenced by the seismic profiles presented and several other multichannel and single-channel
seismic lines. These structures are offset in a dextral sense between the seismic profiles presented in (B) and (A), respectively.

72~ 72"W 71~

16~ [I I ~, ~, ~ I I _k'~/t/ JI I,....-"' I 116~
15~ 15ON
p ]34
DSDP 31
B3
B2
B1
g
0 9
72~ 72W~' 71~

Depth to bucmeat
contour iaterv~: 0.2 km
Fig. 13. Depth to basement (Bf~) in the Taino ridge area. The Casis survey and the Figs. 14-20 are indicated. Reverse faults, pop-up and
strike-slip faults are compatible with a NE-SW compressive stress. DSDP Site 31 results (Bader et al., 1970) help to calibrate the seismic
profiles.
previously presented were unmigrated, whereas all the reverse fault indicates recent activity of this fault.
the Casis profiles are migrated. The southern part The throw of this fault is evaluated to 0.6 km. The
of the Taino ridge is relatively large (Fig. 13). It is second part of the Casis B01 (Fig. 15) seismic pro-
delimited on the western and eastern side by reverse file shows a flower structure and reverse faults with
faults with opposite vergence. On the western flank an eastern vergence that bound the Taino ridge to
a clear reverse fault (see the seismic trace, Fig. 14) the east. The strike-slip fault probably delimits to the
offsets the B fl, A I~ Early Miocene reflectors and the north the large plateau that forms the southern part
sea floor. The identical thickness west and east of of the Taino ridge (Fig. 13). A new flower structure

Fig. 14. Reverse fault in the southern part of the Taino ridge. The trace of the fault is visible on the profile. The throw of the fault is
estimated to be 0.6 km. See the inset map (depth to basement: B") and Fig. 13 for profile location.
is identified on the next profile (Fig. 16). This struc- was reactivated by recent faulting. The thickening of
ture is visible very deep in the crust and offsets a the crust below the Taino ridge is estimated to be 0.5
10-km-deep horizon (R, Fig. 16). We propose that km. To the north the Taino ridge loses its elevation.
this major strike-slip fault connects the Taino ridge A strike-slip fault may be responsible for the tilt of
to the transpressive structure described in the Do- a block (Fig. 19). A positive magnetic anomaly is
minican sub-basin (Fig. 11). To the north the Taino related to the block that could be a Cretaceous vol-
ridge is narrow and is divided into two hills (Fig. 17) cano. Another volcanic ridge lies west of this block
that are offset (Fig. 13). The transverse valley is (Fig. 20). We failed to detect any recent deformation
formed by a strike-slip fault (Fig. 5C). The northern of this structure that is probably an initial volcanic
part of the Taino ridge is a typical pop-up delimited feature of the Beata igneous province.
by reverse faults with opposed vergence (Fig. 18A). The shortening of the northern Taino ridge can be
A 15-km-deep ddcollement merges with the western estimated to be 4.6 km for a 21.5-km-wide structure
fault (Fig. 18B). The B " - A " and the A " - e M inter- (20%; from 8100 to 8500, Fig. 18A). If we assume
vals are thicker on the west part of the structure than that the northern Taino ridge and the southern Taino
on the top. This difference in thickness indicates that ridge (Fig. 13) had the same width initially, we es-
this part of the Taino ridge was an old seamount that timate the shortening to be 39 km per degree. The

4~
4~
>

t-
Fig. 15. A flower structure indicates the presence of a strike-slip fault. See the inset map (depth to basement: B ' ) and Fig. 13 for profile location.
==
..s=
rae~
,.IZZ
..s=
c.tz
.,..~
.,..o
.X:Z
. ,...~
,..o
. ,...,
r.gz
c~
, 9
e~ ,.o
~ m
m ~
m r,~
. ,...~ ~
~ r
~ .,-~
< =
r~

~ ,...~
k~
.,:..;
,.t:Z
~D
,....:.
r
r,.)
tt~
. ,...~
;.r..,
e~t)
~
[...,
~ ,,,,.~
. ,...~
9
.=_,
+,.a
xz
9 t.l:Z

..c:
,.o ~D
..o
.x::
~D ,.o
(D 9....~
.,.~
-,~
~'~
""~ o
~~
.,.~ ~-~
~ ~."
.,.~
~.~ 9
.,~
.~~
.~ b
<

Fig. 19. A tilted block maybe related to a strike-slip fault. The magnetic profile suggests that this block is a volcano. Note the prominent
reflector R visible also in Figs. 16 and 18. See the inset map (depth to basement: B") and Fig. 13 for profile location.
Fig. 20. A seamount has a probable volcanic origin as shown by the magnetic curve. This seamount does not seem to be affected by any
deformation. See the inset map (depth to basement: B") and Fig. 13 for profile location.

Fig. 21. Depth to basement map (B'~; modified from Leroy and Mauffret, 1996). This figure shows the contact between the Colombian
basin and the Beata ridge. The seismic profiles are calibrated by the DSDP Site 153 results (Hopkins, 1973). The locations of the seismic
profiles shown in Figs. 22 and 23 are indicated.
B" reflector is 4.3 km deep on the western side of wards regional dip, evident for the B" reflector and
the Taino ridge, 3.7 km on the top and 4.5 km on the sea floor, and normal faults are initial features
the eastern side (Fig. 18B). The uplift of the Taino formed during the construction of the Beata volcanic
ridge is about 1 km if we neglect a probable ini- plateau (Diebold et al., 1999; Driscoll and Diebold,
tial topography. The western fault is apparently the 1999). However, this dip has been probably accen-
master fault, but the cumulate throw of the eastern tuated during the Miocene to Present compressional
reverse faults is larger than the western fault and event. The results of the DSDP Site 31 indicate
the B" reflector is 0.2 km higher on the western approximately 1 km of uplift during the Miocene
side than on the eastern side (4.3 and 4.5 km, re- (Benson et al., 1970). This uplift of the Beata ridge
spectively, Fig. 18B). The master fault is probably is regional and unrelated to the Taino ridge because
a former normal listric fault with east-facing throw the DSDP Site 31 is located at the base of this
because we cannot explain the eastwards deepening structure (Fig. 18A). This regional uplift probably
of the B" reflector with the observed motion of the occurs step by step from the undisturbed Venezuelan
reverse faults (Fig. 18C). We conclude that the east- basin to the Beata ridge. The Taino ridge is one

Fig. 22. Seismic profile showing the Pecos fault zone (modified from Leroy and Mauffret, 1996). This is a pop-up structure delimited
by reverse faults. Note the difference in level of the reflectors A" and B" between the Colombian and the Venezuelan basins. The
Early Miocene (eM) is deformed, whereas the recent sedimentary layers onlap the structure. The sub-B" reflector (Stoffa et al., 1981) is
prominent. The compression began during Miocene time, but the deformation of the sea floor indicates that this compression is yet active.
See the inset map and Fig. 21 for profile location.
of these steps but other reverse faults exist east of the Beata ridge and the South American deformed
the ridge (Figs. 11 and 12). The dominant motion belt and forms a sill between the Colombian and
is probably a thrust of the Beata ridge over the Venezuelan basins (Fig. 1). A Casis seismic de-
Venezuelan basin, but our survey is not enough ex- tailed survey in addition to previous seismic profiles
tended to the east. However, the crustal fault shown (Fig. 21) allow us to contour a detailed map of the
in Fig. 11C has a clear westwards dip. The com- acoustic basement (Fig. 21). The acoustic basement
bination of N - S to N N W - S S W thrusts, N E - S W of the eastern Colombian basin is relatively smooth
fight-lateral and N W - S E left-lateral strike-slip faults and we can identify the typical B" reflector in the
(Fig. 13) indicates that the stress is oriented N E - eastern Colombian basin (Fig. 22), whereas the true
SW. oceanic rough crust is localized in the western part
of the Colombian basin (Bowland and Rosencrantz,
1988). DSDP Site 153 (Edgar et al., 1973b) allows
ARUBA GAP calibration of the seismic profiles (Hopkins, 1973)
and identification (Fig. 22) of a prominent Middle
This area, that was already studied (Hopkins, Miocene horizon, an Early Miocene marker at the
1973; Stoffa et al., 1981; Leroy, 1995; Leroy et al., top of a disturbed layer, the Middle Eocene A"
1996; Leroy and Mauffret, 1996), is located between reflector, and the top of the Cretaceous volcanic

N E O G E N E INTRAPLATE D E F O R M A T I O N OF THE C A R I B B E A N PLATE AT THE BEATA RIDGE 651
plateau (B" reflector). In addition an internal respectively. The main stress is oriented N E - S W in
flector named sub-B" was identified (Stoffa et al., the two interpretations (Fig. 25). This stress is com-
1981; Leroy et al., 1996). The acoustic basement patible with a compression along the Pecos fault
of the Colombian and Venezuelan basins dips to zone and left-lateral faults that trend E - W (Fig. 25).
the south towards the South American accretionary We estimated the shortening of the Pecos structure
prism, which has been active since the Early to to 40 km per degree of latitude (Leroy and Mauffret,
Middle Miocene (Biju Duval et al., 1982b). 1996).
The Pecos fault zone that trends N W - S W sepa-
rates the Beata from the Colombian basin (Figs. 21
and 22). Reverse faults, with a throw of 100-200 DSDP 151 RIDGE
m, are clearly displayed on the Casis seismic lines
(Fig. 22). The reverse faults are facing towards the This ridge forms the backbone of the Beata ridge
northeast and southwest and the Pecos fault zone (Figs. 2 and 3). In the Colombian basin the seis-
is clearly a compressive feature. Flower structures mic profile Cas-C03 (Fig. 26) crosses the southern
(Fig. 23) indicate also the presence of strike-slip extension of the DSDP 151 ridge. This part of the
faults. The identification, by seismic correlation with ridge is not reactivated, the top of the ridge is 4.7
the western Colombian basin, of the early Miocene km deep and its high relative to the adjacent base-
reflector in the eastern Colombian basin at the base ment of the Colombian basin is 1.9 km (Fig. 26C).
of a well layered seismic unit (Fig. 22), indicates The 5.6 km depth of the top of the ridge on the
that the main tectonic event occurred during the depth to basement map (Fig. 26B) is a mistake due
Early Miocene (23 Ma) because the post-Early to the smoothing effect of the automatic contour-
Miocene layers of the Colombian basin onlap the ing. The buried ridge is clearly offset fight-laterally
flank of the Pecos fault zone (Fig. 22), but some relatively to the northern ridge by faults that trend
recent reactivations (Fig. 23) are evident and the N E - S W (Fig. 26B). A prominent seamount, 1.8 km
sea floor is deformed. In the southern part of the deep (2.4 km on the depth to basement, Fig. 26B)
study area, near and below the accretionary prism, is located in the southern part of the DSDP 151
the deformation is weak and the Pecos structure ridge. This ridge is 130 km long and flanked by
is low (less of 0.4 km of throw relative to the a 4.4-km-deep depression to the east (Fig. 27A).
Colombian basin; Cas A10, Fig. 24). Towards the The seismic profiles recorded during the Seacarib 1
north the deformation increases (Cas A07, Fig. 22; cruise allow a good control of the shape of the ridge.
Cas A08, Fig. 24) and a new area of deformation The Seabeam map (Fig. 27B) shows the northern
appears at the base of the Pecos fault zone (Cas end of the ridge where is located the DSDP Site
A06, Fig. 24; Figs. 23 and 25). The top of the 151 (Fig. 28B). The trend of the DSDP 151 ridge
Pecos fault zone is 1 km higher than the floor of the that is north-south changes abruptly and is delim-
Colombian basin (5.4 km and 6.4 km deep, respec- ited by a N E - S W steep scarp (Figs. 27 and 28B).
tively; Fig. 21). Nevertheless, we do not know the This scarp bounds a basin that trends also N E - S W
initial topography in this area. The Early Miocene- (Fig. 27).
A" interval (Fig. 22) is much thicker (800 m) in
the Colombian basin than in the Venezuelan basin
(300 m) and this observation indicates that some W E S T E R N BOUNDARY OF THE BEATA RIDGE
topography (500 m?) existed prior to the Miocene
deformation. The B"-sub-B" interval and the crust A steep scarp delimits the Beata ridge and
are thinner in the Colombian basin than in the the Haiti sub-basin (Fig. 29). The Seabeam map
Venezuelan basin and this is again an argument to (Fig. 29B) clearly displays the steepness of the slope
suppose a different initial level between the two oriented NE-SW. However, the Tairona ridge, 1 km
basins. The 1 km step between the Pecos fault zone deep, shows a N-S trend. A deep southern terrace
and the Colombian basin results from an initial (between 3 and 4 km, Fig. 29B) is progressively
topography (500 m?) and a Miocene reactivation accreted to the slope towards the north and finally
(500 m?). It is clear that the deformation along disappears. The upper part of the slope is less abrupt
the NW-SE-trending Pecos fault zone increases to- and is covered by slumped sediments (Fig. 30), but
wards the north and consequently is not related to this profile does not show the true dip because it is
the subduction of the volcanic plateau beneath the oblique to the slope. The lower part of sedimentary
South American deformed belt (Burke et al., 1978). cover of the Haiti sub-basin is deformed (Fig. 30).
A first-motion solution for an earthquake in the We think that this deformation is related to recent
southern Beata ridge was reinterpreted from Molnar compression, although presently inactive, but the
and Sykes (1969) by Kafka and Weidner (1981). quality of the seismic profile is too poor to see the
Strike-slip and reverse motions were determined re- wedge displayed in Fig. 8.

Fig. 23. Flower structure indicating the presence of strike-slip fault with a compressional component (transpression). See the inset map
and Fig. 21 for profile location.

Fig. 24. Depth sections (modified from Leroy and Mauffret, 1996) showing the evolution of the Pecos fault zone from south to north. In
the south (Cas A10) the structure is low and affected by few reverse faults; the sea floor is undeformed. The Cas A08 profile shows the
rising of the structure and the formation of a pop-up. The northern section (Cas A06) shows a deformed sea floor and the formation of a
second zone of deformation at the base of the Pecos fault zone. See the inset map and Fig. 22 for profile location.
DISCUSSION and the Hess escarpment were supposed to have a

normal component related to rifting of the Haiti sub-
The western and eastern flanks of the Beata ridge basin (Donnelly et al., 1995; Driscoll and Diebold,
are completely different. On the eastern flank the 1999). However, this basin is closed to the south by
compressive structures are predominant, but these the Warao rise (Fig. 31) that formed a high connect-
structures are offset by right-lateral strike-slip faults ing the Beata ridge to the Nicaragua rise (Mauffret
that trend NE-SW. On the western flank the faults and Leroy, 1997). This rise cannot be formed by a
that trend N E - S W are the main structural features rifting of the Haiti sub-basin and we suggest that
and we failed to identify any compressional struc- this basin is an initial feature surrounded by volcanic
ture (Fig. 31). However, our seismic profiles do not plateaus (Mauffret and Leroy, 1997).
have a good quality and the DSDP 151 ridge may The structural activity that affected the southern
have been reactivated and uplifted by compression. part and the Beata plateau has been related to litho-
This ridge and the Tairona ridge trend N - S and spheric flexure accompanying Miocene to Recent
were probably the same initial feature; consequently, convergence along the Curaqao ridge (Holcombe et
the faults that trend N E - S W may have a strong al., 1990). A lithospheric flexure to the north (Muer-
fight-lateral strike-slip component. The basin that tos trench) and to the south (Curaqao ridge) is indeed
delimits the DSDP 151 ridge to the north could be evident in the Venezuelan basin and is surrounded
a pull-apart basin that trends NE-SW. Recent visual by the 4-km depth contour (Fig. 2). However, the
observations from a submersible survey (Mauffret et normal faults formed by these flexures trend E - W
al., in prep.) indicate that the steep scarp that bounds (Silver et al., 1975; Diebold et al., 1999), whereas
the Haiti sub-basin is divided in short segments of the structures described in this paper (Taino ridge,
north-south thrusts and northeast-southwest strike- Beata plateau, Pecos fault zone) are oriented N N W -
slip faults. The fault that bounds the Haiti sub-basin SSE. Consequently these structures that show evi-

Fig. 25. Structural interpretation of the Aruba Gap area (modified from Leroy and Mauffret, 1996). The Pecos fault zone is a pop-up
structure delimited by reverse faults. A second zone of deformation exists at the base of the northern part of the Pecos fault zone. The
shape of this deformed zone and the flower structure shown in Fig. 23 suggest left-lateral strike-slip faults. These faults and the Pecos
fault zone are compatible with a compressive stress oriented NE-SW. Such a stress is also determined by earthquake focal mechanism
shown in the top of the figure (Kafka and Weidner, 1981). The locations of the depth sections shown Fig. 24 are indicated.
d e n c e o f c o m p r e s s i o n and strike-slip faulting c a n n o t T h e c o m p r e s s i v e stress, a c c o r d i n g to our struc-

be r e l a t e d to the l i t h o s p h e r i c flexure of the C a r i b b e a n tural study and focal m e c h a n i s m analysis ( K a f k a and
plate. Weidner, 1981) is o r i e n t e d N E - S W . T h e s a m e off-

.,..,
,.i=:
.,..~
.,,o
. ,...,
.,..,
.xZ
b ~
go
. ,...,
~o
r.~ ,.o
9 t".-
c,.) ,4
,.lzZ
9~ .xz
tth 9
r~ ..xZ
,..o
,..o
~D
,..o
ra~
.,..~
eq~

73~ 73~ 73"15W 730W 72054W

15045N - I 15"45N
SEAeARIB' BEATA
IM ER.CATORprojection
isobathlcontour: 100 m
, ~lid line: 5(X)m ~ ii i
v/
15~ 15030N
15015N, 15"15N
Rg. 28 B
it
>
Fig. 28 A
15~ 150N
DSDP
site 151
14054N.
73~ 7:3~
,o -
7:3~ 7:30W
,14054N
72054W
B
~.2~
150N
14~
74~ 73~
A
Fig. 27. (A) Depth to basement map: B 't. (B) Seabeam map of the Seacarib 1 cruise. The position of the survey is indicated in Fig.
3. The northern tip of the DSDP 151 ridge, that trends north-south, terminates abruptly towards the north and a steep scarp shows a
northeast-southwest orientation. The positions of the seismic profiles illustrated in Fig. 28 are shown.

Fig. 28. (A) Seacarib 1 seismic profile (see also Fig. 5E) showing the location of the DSDP Site 151 where an hard ground separates
Paleocene sediments from the Santonian sandstone marl and chalk (Edgar et al., 1973b). (B) Seacarib 1 seismic profile showing the steep
scarp that trends NE-SW. For location see Fig. 27.
entation (N45 ~ of compression has been described (Fig. 13) and in Aruba Gap (Fig. 24). The shortening
in the Presqu'~le du Sud d'Hispaniola (Bien-Aime of the Beata ridge induces a shallowing towards the
Momplaisir, 1986). This stress increases towards the north. It is evident that the present topography results
north as shown by the structures described in the from an uplift linked to the compression, but it is
Venezuelan basin (Fig. 11) along the Taino ridge difficult to evaluate the initial topography. We know

658 A. M A U F F R E T and S. LEROY
72~ 72~ 72~ 71~ 71~
17~
t i
:~~:~e~RIE ~BEATAI~. : I
[
17~
i MERCAT(,R :l)rojeet~ I
eorlour:lOOm I
:Solid line: !
17~ 9:,#j/~ ~. . . . .,,.~" 17~
17~ 7~
4000
17~ ,17ON
2000
116~
16~ 16~
a
72~ 72~ 72~ 71~ 71~
q 17~
16~
A
73~ 72~
Fig. 29. (A) Depth to basement map (B"). (B) Seabeam map of the Seacarib 1 cruise showing the steep scarp that forms the boundary
between the Beata ridge and the Haiti sub-basin. The position of the survey is indicated in Fig. 3. The position of the seismic profile
illustrated in Fig. 30 is shown.

Fig. 30. Seacarib 1 seismic profile showing the steep scarp that bounds the Beata ridge. This is not the steepest portion of the scarp and
the profile is not normal to the structure. The upper slope shows evidence of slumping and the basement outcrops in the lower slope
as shown by a diving survey (Mauffret et al., in prep.). A recent deformation can be noted in the Haiti sub-basin (see also Fig. 8). For
location see Fig. 29.
that the Beata ridge is a 20-km-thick Cretaceous vol- topography and 0.5 km of uplift was estimated for
canic plateau. This great thickness is mainly caused the Pecos fault zone (Leroy and Mauffret, 1996).
by a volcanic underplating that initiated the uplift We suppose that the initial topography was the same
and rifting of the Beata ridge. Although a thicken- and the reactivation is consequently 0.8 km high.
ing by compression may have occurred during the This estimation is close to the 1-km-high regional
Miocene, we think that the initial Cretaceous thick- uplift deduced from the DSDP results (Benson et
ness was important. The Colombian and Venezuelan al., 1970). The buffed part of the DSDP 151 ridge
basins subduct in a normal way below the South (Fig. 32A) was initially deeper than the northern
American deformed belt. In contrast the northern part but the shape of the ridge is conserved (1.9 km
part of the Beata ridge collides with the central part high relative to the adjacent basement) except for the
of Hispaniola in relation with the eastwards drift seamount located on the southern tip of the ridge
of the Caribbean plate relative to Hispaniola. This (2.2 km, Fig. 32B). This seamount was probably
collision results from the buoyancy of thick volcanic uplifted (0.3 km) by the N E - S W fault that delimits
crust that resists subduction beneath Hispaniola this feature to the south (Fig. 26). The DSDP 151
(Burke et al., 1978; Mercier de Lepinay et al., 1988). ridge-Taino ridge area do not show any appreciable
The seismic profile displayed in Fig. 27 indicates topographic step, but the region located between the
that a non-reactivated part of the DSDP 151 ridge is DSDP 151 and Tairona ridges is 1.8 km uplifted
buried in the Colombian basin. This basin and the (from 4 km to 2.2 km, Fig. 32G). The northern part
Haiti sub-basin have a thin crust and were conse- of the Beata ridge, near the Bahoruco Peninsula, is
quently much deeper than the initial Beata volcanic 1 km uplifted relative to the south. The northern-
plateau. The Colombian basin is depressed by the most uplift is actually onland where the Sierra de
loading of the thick sedimentary cover and the sub- Bahoruco is 2 km high. We conclude that 6.8 km
duction below the South American deformed belt of uplift may occur between the southern tip of the
(7.4 km, Fig. 32G). The 1.3-km-high step shown Beata ridge and Hispaniola. However, the results of
in Fig. 32G represents the boundary between the a submersible survey (Mauffret et al., in prep.) indi-
Beata ridge and the Colombian basin. This boundary cate that the Beata ridge was shallow during the Late
extends from the Pecos fault zone to the southern tip Cretaceous after the volcanic event and the 6.8 km
of the DSDP 151 ridge (Fig. 31). 0.5 km of initial uplift results from the constructional history of the

Fig. 31. Depth to basement (B") of the central part of the Caribbean Sea. The eastern part of the Beata ridge is characterized by reverse
faulting offset by some NE-SW faults. N-S highs dominate in the central part of the structure. The NE-SW faults are predominant in the
western part of the Beata ridge.
volcanic plateau and a recent compressional event the Marie Aimee ridge and the Bahoruco Peninsula,
from the Miocene to the Present. the Presqu'~le du Sud should be separated from the
We estimated about 40 km of shortening per central Hispaniola by a 240-km gap (Fig. 33). The
degree of longitude. Thus, 170 km of shortening boomerang shape of the Presqu'~le du Sud-Beata
may have occurred at 18~ (Bahoruco Peninsula). ridge may be due to the progressive collision of
If we assume that the Sylvie ridge was in strike with the Beata plateau and the collage of fragments of

i/
0.91kin
-2000 Haiti Rateau
-4000
-6OO0
F
-74.00 __~ 0.00
1.1 km
-2000
Haiti sub basin L ~ Dominican
-40O0 / sub basin
-6O0O
-74.00 / -7:3.00 I -70.00
20001 NiRicsearagua / 2.4kin

1.7 km
-4000
-6000
o
D
-74.00 ~2~ -Tz00 / -70.00
19~
-2000
~ ud
' ~ / . 7 km
-4000 |
"O
~ " I VenezuelaBasin
18~
-6000
C ~"
r'a
03
-74.00 -72.00 -70.00

Haiti sub basin ) 17~
1.8 km
-2000
WaraoRise TainoRidge BeataPlateauu
2.2 km
-4000
._, s ~ ~ 16~
- ~ BeataRidge -600O
_
15~
3.1o -74.00 -72.00 -70.00
4~
14~
-2000
4.7 km
"1
tl
-4000 ColombiaBasin ~L Pecosfault VenezuelaBasin
-6000 1.9km -
...... ColombiaBasin \ 3ON
7.4 km G -74.00 -72.00 -70.00
Vertticalexaggeration:X40
Fig. 32. East-west cross-sections (A to F) and north-south cross section (G) of the Beata ridge. The depth to basement (B') is relative to
the sea level. Except for section (B) (2.2 km) the DSDP 151 ridge has the same height relative to the adjacent basement (1.7 to 19 km),
but we can see the rising towards the north of the ridge (G). For location see the inset map (depth to basement: B').

Fig. 33. The DSDP 151 ridge the Tairona ridge and the Presqu'~le du Sud have been displaced to form a north-south structure. This
reconstruction shows a 240-km-wide gap between the Presqu'~le du Sud and the central part of Hispaniola. The Presqu'ile du Sud may
have underwent a counter-clockwise rotation (Mercier de Lepinay et al., 1988; Van Fossen and Channell, 1988; Jany, 1989).
this plateau (Mercier de Lepinay et al., 1988; Jany, microplate and the latter overthrusts the former. We
1989). Therefore, the present east-west orientation suggest that the eastern boundary of the Colombian
of the Presqu'~le du Sud may be a neotectonic microplate lies along the easternmost thrust (Fig. 11)
feature and a substantial counter clockwise-rotation and the Beata plateau. The southern boundary of
of the Presqu'~le may have occurred (Van Fossen this microplate is the Colombian deformed belt. The
and Channell, 1988). northern part of this belt trends east-west and is
If we assume that the compression occurred since narrow (Vitali, 1985; Vitali et al., 1985). North of
the Early Miocene (23 Ma) then the motion rate the Santa-Marta Massif the deformed belt is seg-
is between 0.74 cm/yr (170 km) and 1.04 cm/yr mented with N-S and E N E - W S W orientation of the
(240 km). The eastern boundary is diffuse with toe of the prism (Fig. 34). After a new prominent
an intraplate deformation, particularly along the bend (Vernette et al., 1992) the Colombian prism
Taino ridge. However, we do not have good seis- merges with the onshore Sinu belt (Duque-Caro,
mic profiles to illustrate the easternmost deformation 1979, 1984; Vitali, 1985; Vitali et al., 1985; Toto
(Fig. l lA, B) and a deep thrust, dipping to the and Kellogg, 1992) that is 10 km thick (Lehner
west, is suggested (Fig. 11C). The several thrusts ob- et al., 1984). This prism was formed in the Early
served on the eastern side of the Beata ridge imply Miocene (Duque-Caro, 1979) along the buried Sinu
a deep d6collement level. Fig. 18C shows that this trench that trends north-south. This active margin is
d6collement is deeper than 15 km. confirmed by the seismicity that defines a 180-km-
It is clear that the Beata ridge forms a boundary deep Benioff zone. The length of the subducted crust
between a Venezuelan microplate and a Colombian is 350 km (Malav6 and Suarez, 1995). This Benioff

zone belongs to the Bucaramanga segment (Penning- Therefore, a 0.5 cm/yr of additional motion can be
ton, 1981) and cannot be related to the Nazca Plate estimated along the Pedro fault. The velocity vec-
as suggested by Van der Hilst and Mann (1994). We tor diagram (Fig. 35A) suggests a velocity as high
suggest that the Sinu subduction zone is related to as 2.6 cm/yr of the Colombian microplate relative
the Beata deformed zone by fight-lateral strike-slip the North American plate. GPS measurements in
faults and short north-south segments of subduction the Presqu'~le du Sud determine 2.3 cm/yr for the
zones (Fig. 34). The northern part of the Sinu belt relative velocity between this area and the North
is clearly (Vernette et al., 1992) offset by a strike- American plate (Farina et al., 1995). The relative
slip fault zone and the South Caribbean marginal motion between the central block of Hispaniola and
fault illustrated by fig. 13 in Kellogg and Bonini the North American plate is 1.5 cm/yr according
(1982) is probably a transpressive feature. West- to the previous study. Therefore, the relative motion
wards, the Colombian microplate is delimited by the between the central block of Hispaniola and the
Panamanian subduction zone (Adamek et al., 1988; Presqu'~le du Sud is 0.8 cm/yr, a value very close
Protti and Schwartz, 1994). This Panamanian block to our estimation (0.9 cm/yr). The plate motion of
is delimited to the north by a strike-slip fault that the southern Caribbean region is documented by the
is connected to the Middle American trench south GPS studies (Freymueller et al., 1993; Drewes et al.,
of Nicoya Peninsula (Fisher et al., 1994; Marshall 1995; Kellogg and Vega, 1995). The Colombian mi-
and Anderson, 1995). This peninsula may belong to croplate subducts beneath the Panama prism with a
the Colombian microplate and a fault north of the rate of 1 cm/yr and the rate of convergence between
peninsula may connect the Middle American trench the Colombian microplate and the North Andes
and the Hess escarpment (Dengo, 1985). The west- block (Sinu trench) can be evaluated to be 1.7 cm/yr
ern part of the Hess escarpment is active (Bowland, (Kellogg and Bonini, 1982). The differential motion
1993), but the eastern part of this feature does not between the two plates may result from the greater
show clear evidence of recent faulting. On the con- convergence of the NOAM-SOAM in the western
trary, the Pedro escarpment, that delimits the upper part of the Caribbean zone than in the eastern part
Nicaragua rise, shows much evidence of recent fault- (Mtiller et al., 1996) or from the influence of the
ing (Mascle et al., 1985; Holcombe et al., 1990). large convergent motion of the Cocos plate. In the
Moreover, the lower part of the Nicaragua rise eastern Colombian basin intraplate deformation has
has the same volcanic basement as the Colombian been observed northeast of the Panamanian prism
basin (Mauffret and Leroy, 1997) and we placed the (Bowland, 1993) and the results of the DSDP Site
northwestern boundary of the Colombian microplate 154 (Edgar et al., 1973b) indicate that the uplift of
along the Pedro escarpment (Fig. 34). The northern the deformed structure is related to an Early Pliocene
limit of the microplate is the Enriquillo strike-slip reverse fault with a southwest vergence clearly dis-
fault and the Bahoruco thrust. played in the seismic profiles shown by Bowland
The Venezuelan microplate is delimited to the (1993). The negative buoyancy of the young Cocos
east by the Lesser Antilles subduction zone and to oceanic crust may induce a compressional stress in
the south by the Venezuelan deformed belt. The the overriding Caribbean plate (England and Wor-
northern boundary lies along the Anegada Passage tel, 1980; Meijer, 1992). We cannot decipher the
(Jany et al., 1990) and Muertos trough. A Gonave dominant process (Atlantic or Pacific) to produce
microplate was defined (Rosencrantz and Mann, a higher displacement of the Colombian microplate
1991; Mann et al., 1995) between the Cayman than the Venezuelan plate. We observe that the
spreading center, the Oriente fault to the north, chrons 7 (25 Ma) and 8 (26 Ma) correspond to an in-
the Walton-Plantain Garden-Enriquillo fault to the crease of convergence between the Nazca plate and
south and the western coast of Hispaniola. We sug- the South American plate (Pargo-Casas and Molnar,
gest that this microplate is related to the Venezuelan 1987) and between the North and South American
microplate through the narrow Enriquillo basin. plates (Mtiller et al., 1996), respectively. At this time
A recent recompilation of the magnetic data the Cayman spreading center was also reorganized
(Leroy, 1995) in the Cayman trough determined (Leroy, 1995) and Hispaniola began to drift away
0.9 cm/yr of half-rate spreading between chron 8 from Cuba. A global reorganization of plate mo-
(26 Ma) and chron 1 (Fig. 35A). The motion along tion in the Pacific and Atlantic Oceans may have
the Walton-Plantain Garden fault is evaluated in influenced the internal stress of the Caribbean plate.
Jamaica Island to 0.4 cm/yr during the Miocene There is much controversy about the pole position
to Quaternary (Rosencrantz and Mann, 1991). How- of the North America-Caribbean plate either to the
ever, the slip rate of the Enriquillo fault in Hispaniola north (Sykes et al., 1982; Deng and Sykes, 1995)
is evaluated to 0.8 cm/yr (Mocquet and Aggarwal, or the south (Stein et al., 1988; Calais and Mercier
1983) and we suggest a 0.9 cm/yr motion towards de Lepinay, 1993; Lundgren and Russo, 1996). We
the northwest of the Colombian microplate (Fig. 35). agree with Heubeck and Mann (1991), who placed

7~
-]
Fig. 34. (A) The Colombian and Venezuela-Gonave microplates have been differentiated with two patterns. (B) Sketch that shows the relationship between the Sinu trench and the compression along the
eastern flank of the Beata ridge. Refer to the text for explanation.
Fig. 35. (A) Early Miocene reconstruction. Refer to the text for explanation. (B) Sketch (from Leroy and Mauffret, in prep., modified)
inspired from Heubeck and Mann (1991). We suggest two different poles for the North American plate-Venezuelan microplate and
the North American plate-Colombian microplate motions. A pole for Venezuelan-Colombian microplates motion is proposed near the
southern tip of the Beata ridge where the compressional deformation is weak.
a pole for the NOAM-Colombian microplate in CONCLUSIONS

the south and a pole for the NOAM-Venezuelan
microplate in the north (Fig. 35B). The pole for This study shows how detailed seismic surveys,
the Venezuelan-Colombian microplate must be very together with a regional coverage, can be useful
close to the southern tip of the Beata ridge (Fig. 35B) to investigate a complex tectonic area. This is par-
where the compressional deformation is weak. ticularly true for the Taino ridge and Pecos fault

zone where we showed reverse faults and pop-up stress can be applied to the Caribbean plate by the
offset by strike-slip faults. The tectonic framework Nazca and the Cocos plates.
of these areas is coherent with a N E - S W stress that
increases towards the north. The orientation of this
stress is incompatible with a simple squeeze of the ACKNOWLEDGEMENTS
Caribbean plate between the North and South Amer-
ican plates (Burke et al., 1978) although a westwards This work was supported by grants INSU ATP
increase of convergence between the two major 733 and 780. We thank the officers and crew of the
plates may have influenced the faster migration of R / V Nadir for assistance in this project. We are es-
the Colombian microplate towards the east than the pecially indebted to the technical crew of GENAVIR
Venezuelan microplate. A clear boundary between who greatly helped in the acquisition of multichan-
the Colombian and Venezuelan microplates cannot nel data. These data were processed at Institut de
be traced, although we do not have enough seismic Physique du Globe de Strasbourg and we thank R.
information to exclude a major structure along the Schlich and M. Schaming who have facilitated ac-
easternmost thrust. However, the deformation seem cess to the processing center and helped us to use the
to be diffuse and localized along several structures. Geovector software. We thank J. Diebold from La-
These structures are probably pre-existent but it is mont-Doherty Earth Sciences Observatory, E Mann
very difficult to differentiate between the initial to- and J. Austin from the University of Texas at Austin,
pography and the observed topography that results A. Mascle from the Institut du P6trole, and E Lehner
from the Miocene to Recent compression. Our short- from Shell to have provided several seismic lines
ening estimations, that are as high as 20%, can that completed our seismic coverage. We thank E
be biased by this poor knowledge of the pre-exis- Mann, T. Holcombe, N. Donnelly and E. Calais for
tent topography. The displacement of the Colombian their constructive reviews and helpful suggestions.
microplate relative to the Gonave-Venezuelan mi- Contribution of URA 1759.
croplate is evaluated between 170 and 240 km since
the Early Miocene. This is much greater than the 50
km of offset along the Enriquillo fault in Presqu'~le REFERENCES
du Sud (Mann et al., 1995), although an additional
motion may occur south of the Presqu'~le du Sud. Adamek, S., Frohlich, C. and Pennington, W., 1988. Seismicity
However, our rate of 0.9 cm/yr is compatible with of the Caribbean-Nazca boundary: constraints on micro-plate
the recent and preliminary GPS results (Farina et tectonics of the Panama region. J. Geophys. Res., 93, 2053-
2075.
al., 1995). The present Caribbean plate is a com-
Austin, J.A., 1984. Overthrusting in a deep-water carbonate ter-
posite, partly formed by continental crust (upper rane. In: A.W. Bally (Editor), Seismic Expression of Structural
Nicaragua rise), island arc crust (Jamaica and north Styles. Am. Assoc. Pet. Geol., Stud. Geol., 15: 3.4.2-167-
Hispaniola), but the bulk of this plate is the Cre- 3.4.2-172.
taceous igneous province. The boundary migrates Bader, R.G., Gerard, R.D., et al. (Editors), 1970. Initial Reports
to the south and fragments of the former plate are of the Deep Sea Drilling Project. U.S. Government Printing
Office, Washington, D.C., Leg 4.
abandoned and integrated into the North American Benson, W.E., Gerard, R.D. and Hay, W.W., 1970. Summary and
plate (Mann et al., 1995). In addition we propose conclusions. In: R.G. Bader, R.D. Gerard et al. (Editors), Ini-
internal boundaries that divide the Caribbean plate tial Reports of the Deep Sea Drilling Project. U.S. Government
in several blocks: upper Nicaragua block, lower Printing Office, Washington, D.C., Leg 4, pp. 659-673.
Nicaragua block, Colombian and Venezuelan mi- Bien-Aime Momplaisir, R., 1986. Contribution ~ l'6tude
g6ologique de la partie orientale du massif de la Hotte
croplates, northern Hispaniola. In such a complex
(Presqu'~le du sud d'Ha'fti). Synthbse structurale des marges
tectonic framework it is impossible to study just the de la Presqu'~le h partir de donn6es sismiques. Thbse de
main boundary, i.e. the Cayman spreading center- l'Universit6 Paris 6, 223 pp.
Puerto Rico trench-Lesser Antilles subduction zone, Biju Duval, B., Bison, G., Mascle, A. and Muller, C., 1982a.
to deduce the motion between North America and Active margin processes: field observations in Southern His-
the Caribbean plate and we suggest that the pole for paniola. In: J. Watkins and C.L. Drake (Editors), Studies in
Continental Margins. Am. Assoc. Pet. Geol. Mem., 34: 325-
the North America-Venezuelan microplate motion is 344.
different from that of the North America-Colombian Biju Duval, B., Mascle, A., Rosales, H. and Young, G., 1982b.
microplate motion. The South American plate under- Episutural Oligo-Miocene basins along the north Venezuela
goes an internal stress due to the rapid subduction of margin. In: J.S. Watkins and C.L. Drake (Editors), Studies in
the young oceanic crust of the Nazca plate (England Continental Margins. Am. Assoc. Pet. Geol. Bull., 34: 347-
358.
and Wortel, 1980; Meijer, 1992). In the northern part
Bowland, C.L., 1993. Depositional history Ofthe western Colom-
of South America, the North Andes block undergoes bian basin, Caribbean Sea, revealed by seismic stratigraphy.
an E - W stress and migrates towards the north rel- Geol. Soc. Am. Bull., 105: 1321-1345.
ative to the South American main plate. A similar Bowland, C.L. and Rosencrantz, E., 1988. Upper crustal structure

N E O G E N E I N T R A P L A T E D E F O R M A T I O N OF THE C A R I B B E A N PLATE AT THE BEATA R I D G E 667
of the western Colombian basin. Geol. Soc. Am. Bull., 100: Edgar, T.N., Ewing, J. and Hennion, J., 1971. Seismic refraction
534-546. and reflection in the Caribbean Sea. Am. Assoc. Pet. Geol.
Bryne, D., Suarez, G. and McCann, W.R., 1985. Muertos trough Bull., 55, 833-870.
subduction; microplate tectonics in the northern Caribbean. Edgar, T.N., Holcombe, T., Ewing, J. and Johnson, W., 1973a.
Nature, 317: 420-421. Sedimentary hiatuses in the Venezuela basin. In: T.N. Edgar
Burke, K., Fox, E and Seng6r, A.M.C., 1978. Buoyant ocean and J.B. Saunders et al. (Editors), Initial Report of the Deep
floor and the evolution of the Caribbean. J. Geophys. Res., 83, Sea Drilling Project. U.S. Government Printing Office, Wash-
3949-3954. ington, D.C., Leg 15, pp. 1051-1062.
Calais, E. and Mercier de Lepinay, B., 1993. Semiquantitative Edgar, T.N., Saunders, J.B. et al. (Editors), 1973b. Initial Report
modeling of strain and kinematics along the Caribbean/North of the Deep Sea Drilling Project. U.S. Government Printing
America strike-slip plate boundary zone. J. Geophys. Res., 98: Office, Washington, D.C., Leg 15.
8293-8308. England, E and Wortel, R., 1980. Some consequences of the
Calais, E., Mercier de Lepinay, B., Saint-Marc, E, Butterlin, subduction of young slab. Earth Planet. Sci. Lett., 47: 403-
J. and Schaaf, A., 1992. La limite de plaques d6crochante 415.
nord carfffbe en Hispaniola: 6volution paldog6ographique et Ewing, J., Antoine, J. and Ewing, M., 1960. Geophysical Mea-
structure c6nozoique. Bull. Soc. G6ol. Fr., 163" 309-324. surements in the Western Caribbean Sea and in the Gulf of
Case, J.E. and Holcombe, T.L., 1980. Geologic-tectonic map Mexico. J. Geophys. Res., 65: 4087-4125.
of the Caribbean region. Scale 1:2,500,000. U.S. Geol. Surv. Farina, F., Calais, E., DeMets, C., Dixon, T., Jansma, E and
Misc. Invest. Ser., Map I-II00. Mann, E, 1995. GPS measurements across the northern
Deng, J. and Sykes, L.R., 1995. Determination of Euler pole Caribbean plate boundary zone: preliminary results. Eos 76
for contemporary motion of Caribbean and North American (17): $94.
plates using slip vectors of interplate earthquakes. Tectonics, Fisher, D.M., Gardner, T.W., Marshall, J.S. and Montero, W.,
14: 39-53. 1994. Kinematics associated with late Cenozoic deformation
Dengo, G., 1985. Mid America: tectonic setting for the Pacific in Central Costa Rica: western boundary of the Panama mi-
Margin from southern Mexico to northwestern Colombia. In: croplate. Geology, 22, 263-266.
A.E.M. Nairn, EG. Stehli et al. (Editors), The Ocean Margin Fox, EJ. and Heezen, B.C., 1975. Geology of the Caribbean
and Basin, 7A. The Pacific Ocean. Plenum Press, New York, crust. In: A.E.M. Nairn and T.G. Stehli (Editors), The Ocean
pp. 123-180. Basins and Margins, Vol. 3. The Gulf of Mexico and the
Diebold, J.D., Driscoll, N.W. and EW-9501 Science Team, Caribbean. Plenum Press, New York, pp. 421-466.
1999. New insights on the formation of the Caribbean basalt Fox, P.J., Ruddiman, W.F., Ryan, W.B.F. and Heezen, B.C.,
province revealed by multichannel seismic images of vol- 1970. The geology of the Caribbean crust, 1. Beata ridge.
canic structures in the Venezuelan Basin. In: E Mann (Edi- Tectonophysics, 10:495-513.
tor), Caribbean Sedimentary Basins. Elsevier, Amsterdam, pp. Freymueller, J.T., Kellogg, J.N. and Vega, V., 1993. Plate mo-
561-589. tions in the North Andean Region. J. Geophys. Res., 98:
Dillon, W.E, Austin, J.A., Scanlon, K.M., Edgar, N.T. and Par- 21853-21863.
son, L.M., 1992. Accretionary margin of north-western His- Heubeck, C. and Mann, E, 1991. Geologic evaluation of plate
paniola: morphology, structure and development of part of the kinematic models for the North American-Caribbean plate
northern Caribbean plate boundary. Mar. Pet. Geol., 19: 70- boundary zone. Tectonophysics, 191: 1-26.
88. Hilst, R. and Mann, E, 1994, Tectonic implications of tomo-
Donnelly, T.W., Duncan, R.A., Sinton, C., Diebold, J., Driscoll, graphic images of subducted lithosphere beneath northwestern
N., Abrams, L. and Mauffret, A., 1995. Proposal for drilling a South America. Geology, 22:451-454.
major large igneous province: the Caribbean Cretaceous basalt Holcombe, T.H. and Moore, W.S., 1977. Paleocurrents in the
province. Proposal ODP 480, 53 pp. eastern Caribbean: geologic evidence and implications. Mar.
Drewes, H.D., Kaniuth, K., Stber, K., Tremel, H., Kahle, H.G., Geol., 23: 35-56.
Straub, C., Hernandez, N., Hoyer, M. and Wildermann, E., Holcombe, T.L., Ladd, J.W., Westbrook, G., Edgar, T. and Bow-
1995. The Casa 93 GPS campaign for crustal deformation land, C.L., 1990. Caribbean marine geology and basins of
research along the south Caribbean plate boundary. J. Geodyn., the plate interior. In: G. Dengo and J.E. Case (Editors), The
20:129-144. Geology of North America, Vol. H. The Caribbean Region.
Driscoll, N.W. and Diebold, J.D., 1999, Tectonic and strati- A Decade of North American Geology, Geological Society of
graphic development of the Caribbean: new constraints on America, Boulder, Colo., pp. 231-260.
the Caribbean-South American plate. In: E Mann (Edi- Hopkins, H.R., 1973. Geology of the Aruba Gap abyssal plain
tor), Caribbean Sedimentary Basins. Elsevier, Amsterdam, pp. near DSDP Site 153. In: T.N. Edgar, J.B. Saunders et al.
591-626. (Editors), Initial Reports of the Deep Sea Drilling Project.
Duncan, R.A. and Hargraves, R.B., 1984, Plate tectonic evolu- Washington, D.C., Leg 15, pp. 1039-1050.
tion of the Caribbean region in the mantle reference frame. Houtz, R.Z. and Ludwig, W.J., 1977. Structure of the Colombia
In: W.D. Bonini, R.B. Hargraves and R. Shagam (Editors), basin, Caribbean Sea, from profiler sonobuoy measurements.
The Caribbean-South American Plate Boundary and Regional J. Geophys. Res., 82:4861-4867.
Tectonics. Geological Society of America, pp. 81-94. Jany, I., 1989. Ndotectonique au sud des Grandes Antilles, colli-
Duque-Caro, H., 1979. Major structural elements and evolution sion (ride de Beata, Presqu'~le de Bahoruco)-subduction (fosse
of northwestern Colombia. In: J.S. Watkins, L. Montadert et de Muertos)-transtension (passage d'Anegada). Th6se, Univer-
al. (Editors), Geological and Geophysical Investigations of sit6 E et M. Curie, Paris 6, 306 pp.
Continental Margins. Am. Assoc. Pet. Geol. Mem., 29: 329- Jany, I., Scanlon, K.M. and Mauffret, A., 1990. Geological
351. interpretation of combined Seabeam, Gloria and seismic data
Duque-Caro, H., 1984. Structural style, diapirism, and accre- from Anegada Passage (Virgin Islands, north Caribbean). Mar.
tionary episodes of the Sinu-San Jacinto terrane, southwestern Geophys. Res., 12: 173-196.
Caribbean borderland. In: W.E. Bonini, R.B. Hargraves et al. Kafka, A.L. and Weidner, D.J., 1981. Earthquake focal mecha-
(Editors), The Caribbean-South American Plate Boundary and nism and tectonic processes along the southern boundary of
Regional Tectonics. Geol. Soc. Am. Mem., 162:303-316. the Caribbean plate. J. Geophys. Res., 86: 2877-2888.

Kellogg, J.N. and Bonini, W.E., 1982. Subduction of the of the Azua and Enriquillon basins, Dominican Republic;
Caribbean Plate and basements uplifts in the overriding South 2, Structure and Tectonics. In: Mann, E, Draper, G. et al.
America Plate. Tectonics, 1 (3): 251-276. (Editors), Geologic and Tectonic Development of the North
Kellogg, J.N. and Vega, V., 1995. Tectonic development of America-Caribbean Plate Boundary in Hispaniola. Geol. Soc.
Panama, Costa Rica, and the Colombian Andes: constraints Am., Spec. Pap., 262: 367-389.
from global Positioning system geodetic studies and gravity. Mann, E, Taylor, EW., Lawrence Edwards, R. and Teh-Lung Ku,
In: P. Mann (Editor), Geological and Tectonic Development 1995. Actively evolving microplate formation by oblique col-
of the Caribbean Plate Boundary in southern Central America. lision and sideways motion along strike-slip faults: an example
Geological Society of America, Boulder, Colo., pp. 75-90. from the northeastern Caribbean plate margin. Tectonophysics,
Klaver, G.T., 1987. The Curagao lava formation: an ophiolitic 246: 1-69.
analogue of the anomalous thick layer 2B of the mid-Cre- Marshall, J.S. and Anderson, R.S., 1995. Quaternary uplift and
taceous oceanic plateaus in the western Pacific and central seismic cycle deformation, Peninsula de Nicoya, Costa Rica.
Caribbean. GUA (Univ. Amsterdam) Pap. Geol., Ser. 1, 27: Geol. Soc. Am. Bull., 107: 463-473.
128. Mascle, A., Cazes, M. and Le Quellec, E, 1985. Structure des
Kolla, V., Buffler, R.T. and Ladd, J.W., 1984. Seismic stratigra- marges et bassins cara'fbes: une revue. In: A. Mascle (Editor),
phy and sedimentation of Colombian basin. Am. Assoc. Pet. Symposium sur la g6odynamique des Cara'ibes. Technip, Paris,
Geol. Bull., 68:316-332. pp. 1-20.
Ladd, J.W. and Watkins, J.S., 1978. Active marine structure Masson, D.G. and Scanlon, K.M., 1991. The neotectonic setting
within the north slope of the Muertos Trench. Geol. Mijnbouw, of Puerto-Rico. Geol. Soc. Am. Bull., 103: 144-154.
57: 255-260. Mauffret, A. and Jany, I., 1990. Collision et tectonique
Ladd, J.W. and Watkins, J.S., 1980. Seismic stratigraphy of the d'expulsion le long de la fronti~re nord-Cara'fbe. Oceanol.
western Venezuela basin. Mar. Geol., 35:21-41. Act., 10: 97-116.
Ladd, J.W., Shih, T.C. and Tsai, C.J., 1981. Cenozoic tectonics Mauffret, A. and Leroy, S., 1997. Seismic stratigraphy and struc-
of central Hispaniola and adjacent Caribbean Sea. Am. Assoc. ture of the Caribbean Sea. Tectonophysics (in press).
Pet. Geol. Bull., 65: 675-689. Mauffret, A., Mercier de Lepinay, B., Leroy, S., Vila, J.M.,
Ladd, J.W., Truchan, M., Talwani, M., Stoffa, E, Bulh, E, Houtz, Campan, A., Green, C., Gorini, C., Marton, G. and Reyes,
R., Mauffret, A. and Westbrook, G., 1984. Seismic reflections J.R., 1994. Premiers r6sultats de la campagne Casis dans le
profiles across the southern margin of the Caribbean. In: W.E. bassin Cara'fbe. C.R. Acad. Sci. Paris, 318: 1379-1386.
Bonini, R.B. Hargraves et al. (Editors), The Caribbean-South Mauffret, A., Hallot, E., Leroy, S., Mercier de Lepinay, B. and
American Plate Boundary and Regional Tectonics. Geol. Soc. Vila, J.M., in prep. First results of a submersible survey of the
Am. Mem., 162: 153-159. Beata ridge.
Ladd, J., Holcombe, T.L., Westbrook, G.K. and Edgar, N.T., Maurasse, E, Husler, G., Georges, G., Schmitt, R. and Damond,
1990. Caribbean marine geology; active margins of the plate P., 1979. Upraised Caribbean Sea floor below acoustic reflector
boundary. In: G. Dengo and J.E. Case (Editors), The Geology B :t and the Southern Peninsula of Ha'fti. Geol. Mijnbouw, 8:
of North America, Vol. H. The Caribbean Region. A Decade 71-83.
of North American Geology, Geological Society of America, Meijer, P.T., 1992. The dynamics of motion of the South Ameri-
Boulder, Colo., pp. 261-290. can plate. J. Geophys. Res., 97: 11915-11931.
Lebron, M.C. and Perfit, M.R., 1994. Petrochemistry and tec- Mercier de Lepinay, B., 1987. Evolution de la bordure nord-
tonic significance of Cretaceous island-arc rocks, Cordillera cara'fbe: l'exemple de la transversale d'Hispaniola. Th6se de
Oriental, Dominican Republic. Tectonophysics, 229: 69-100. Doctorat d'Etat, Universit6 Paris 6, 325 pp.
Lehner, E, Doust, H., Bakker, G., Allenbach, E and Gueneau, J., Mercier de Lepinay, B., Mauffret, A., Jany, I., Bouysse, P.,
1984. Active Caribbean margin of South America. In: Bally, Mascle, A., Renard, V., Stephan, J.E and Hernandez, E., 1988.
A.W. (Editor), Seismic Expression of Structural Styles. Am. Une collision oblique sur la bordure nord-cara'fbe h la jonction
Assoc. Pet. Geol., Stud. Geol., 15:3.4.1.111-3.4.2.128. entre la ride de Beata et la fosse de Muertos. C.R. Acad. Sci.
Leroy, S., 1995. Structure et origine de la plaque cara'ibe. Im- Paris, 307:1289-1296.
plications g6odynamiques. Thbse de l'Universit6 Paris 6, 240 Mocquet, A. and Aggarwal, Y.P., 1983. Seismic slip rates in the
PP. Greater and Lesser Antilles; Implications for the present-day
Leroy, S. and Mauffret, A., 1996. Intraplate deformation in the motion of the Caribbean plate relative to North America. Eos,
Caribbean region. J. Geodyn., 21: 113-122. 64: 832.
Leroy, S., Bitri, A. and Mauffret, A., 1996. Migration velocity Molnar, P. and Sykes, L., 1969. Tectonics of the Caribbean and
analysis based on common shot depth migration applied to Middle America regions from focal mechanism and seismicity.
the seismic data of the Caribbean oceanic plateau. Geophys. J. Bull. Geol. Soc. Am., 80:1639-1684.
Int., 125: 199-213. Mtiller, R.D. and Smith, W.H.E, 1993. Deformation of the
Lundgren, ER. and Russo, R.M., 1996. Finite element modeling oceanic crust between the North American and South Ameri-
of crustal deformation in the North America-Caribbean plate can plates. J. Geophys. Res., 98: 8275-8291.
boundary zone. J. Geophys. Res., 101: 11317-11327. Mtiller, R.D., Cande, S.C., Royer, J.Y., Roest, W.R. and Mass-
Malav6, G. and Suarez, G., 1995. Intermediate-depth seismicity chenkov, S., 1996. A quantitative analysis of North American
in northern Colombia and western Venezuela and its relation- and South American plate motions since chron 34 (83 Ma).
ship to Caribbean plate subduction. Tectonics, 14 (3): 617- In: P. Mann (Editor), Sedimentary Basins of the World, vol. 4,
628. Caribbean Basins. Elsevier, Amsterdam.
Mann, P. and Lawrence, S.R., 1991. Petroleum potential of Pargo-Casas, E and Molnar, P., 1987. Relative motion of the
Southern Hispaniola. Jour. of Pet. Geol., 14, 291-308. Nazca (Farallon) and south American plates since Late Creta-
Mann, P., Draper, G. and Lewis, J.E, 1991a. An overview of ceous time. Tectonics, 3 (3): 233-248.
the geologic and tectonic development of Hispaniola. In: P. Pennington, W.D., 1981. Subduction of the Eastern Panama
Mann, G. Draper et al. (Editors), Geologic and Tectonic De- basin and seismotectonics of Northwestern South America. J.
velopment of the North America-Caribbean Plate Boundary in Geophys. Res., 86: 753-770.
Hispaniola. Geol. Soc. Am., Spec. Pap., 262, 1-28. Pindell, J.L. and Barrett, S.E, 1990. Geological evolution of the
Mann, P., McLaughin, P.P. and Cooper, J.C., 1991b. Geology Caribbean Region. In: G. Dengo and J.E. Case (Editors), The

N E O G E N E INTRAPLATE D E F O R M A T I O N OF THE C A R I B B E A N PLATE AT THE BEATA R I D G E 669
Geology of North America, Vol. H. The Caribbean Region. Talwani, M., Windisch, C.C., Stoffa, EL., Buhl, E and Houtz,
A Decade of North American Geology, Geological Society of R.E., 1977. Multichannel seismic study in the Venezuelan
America, Boulder, Colo., pp. 405-432. basin and Curacao Ridge. In: M. Talwani and W.C.I. Pitman
Protti, S. and Schwartz, Y., 1994. Mechanics of back arc defor- (Editors), Islands Arcs, Deep Sea Trenches, and Back-Arc
mation in Costa Rica: evidence from an aftershock study of Basins. Am. Geophys. Union, Maurice Ewing Ser., 1: 83-98.
the April 22, 1991, Valle de la Estrella, Costa Rica, earthquake Toto, A. and Kellogg, J.N., 1992. Structure of the Sinu-San
(M = 7.7). Tectonics, 13: 1093-1107. Jacinto fold belt-- an active accretionary prism in northern
Pubellier, M., Mauffret, A., Leroy, S. and Vila, J., 1999. Plate Colombia. J. South Am. Earth Sci., 5, 211-222.
boundary readjustment in oblique convergence: example of the Van der Hilst, R. and Mann, E, 1994. Tectonic implications of
Neogene of Hispaniola, Greater Antilles. Tectonics (in press). tomographic images of subducted lithosphere beneath north-
Rosencrantz, E. and Mann, E, 1991. SeaMarc II mapping of western South America. Geology, 22:451-454.
transform faults in the Cayman Trough, Caribbean Sea. Geol- Van Fossen, M.C. and Channell, J.E.T., 1988. Paleomagnetism of
ogy, 19: 600-693. Late Cretaceous and Eocene limestones and chalks from Haiti:
Russo, R.M. and Villasenor, A., 1995. The 1946 Hispan- tectonic interpretations. Tectonics, 7:601-612.
iola earthquakes and the tectonics of the North Americas- Vernette, G., Mauffret, A., Bobier, C., Briceno, L. and Gayet,
Caribbean plate boundary zone northeastern Hispaniola. J. J., 1992. Mud diapirism, fan sedimentation and strike-slip
Geophys. Res., 100: 6265-6280. faulting, Caribbean Colombian Margin. Tectonophysics, 202:
Russo, R.M. and Villasenor, A., 1997. Reply. J. Geophys. Res., 335-349.
102: 793-802. Vila, J.-M., Pubellier, M., Jean-Poix, C., Feinberg, H., Butter-
Scientific Party, Leg 165, 1996. Deep sea cores from the lin, J., Boisson, D., Amilcar, H. and Amilcar, H.C., 1988.
Caribbean reveal history of volcanism, tectonic and oceanic D6finition de la limite entre les blocs m6ridional et septen-
changes. Eos, 77 (31): 291. trional d'Hispaniola: d6couverte d'un t6moin de la nappe
Silver, E.A., Case, J.E. and MacGillavry, H.J., 1975. Geophysical de Macaya dans l'anticlinal de Pierre Payen (centre d'Ha'fti,
study of the Venezuelan borderland. Geol. Soc. Am. Bull., 86: cha~ne des Matheux, Grandes Antilles); implications g6ody-
213-226. namiques: C.R. Acad. Sci. Paris, 307: 603-608.
Stein, S., DeMets, C., Gordon, R.G., Brodholt, J., Argus, D., Vila, J.M., Jany, I., Lepvrier, C., Feinberg, H. and Mauffret, A.,
Engeln, J.E, Lundgen, E, Stein, C., Wiens, D.A. and Woods, 1990. Mise en evidence de l'~ge post-plioc~ne inf6rieur de
D.E, 1988. A test of alternative Caribbean Plate relative mo- la collision entre la ride de Beata et l'orog~ne nord-cara'l"oe
tion models. J. Geophys. Res., 93: 3041-3050. (Grandes Antilles). C.R. Acad. Sci. Paris, 311: 1359-1366.
Stoffa, E, Mauffret, A., Truchan, M. and Buhl, E, 1981. Vitali, C., 1985. Etude morphostructurale des prismes de
Sub-B" layering in the southern Caribbean: the Aruba Gap Panama, de Colombie et du Venezuela: leurs relations avec
and Venezuela basin. Earth Planet. Sci. Lett., 53: 131-146. les domaines oc6aniques et continentaux proches. Th~se de
Sykes, L.R., McCann, W.R. and Kafka, A.L., 1982. Motion of 3me cycle, Universit6 Paris 6, 225 pp.
Caribbean Plate during last 7 millions years and implications Vitali, C., Mauffret, A., Kenyon, N., Renard, V. and Belder-
for earlier Cenozoic movements. J. Geophys. Res., 87: 10656- son, B., 1985. Deformed belts off Panama and Colombia
10676. (Caribbean Sea) and plate tectonics in Panama area. In:
Talwani, M., Ewing, J., Ewing, M. and Saito, T., 1966. Geo- A. Mascle (Editor), Symposium sur la G6odynamique des
logical and geophysical studies of the submarine escarpments. Caraibes. Technip, Paris, pp. 451-461.
Geol. Soc. Am. Spec. Pap., 101 : 217-218.

C h a p t e r 22
Caribbean Modelled After Archipelago Orogenesis"

Coda in Esperanto by the Series Editor
KENNETH J. HSU
A colleague recently declined our invitation to Spoken languages evolved naturally, but Esperanto is
serve as a candidate for a volume editorship of the an arbitrary, logical construction. Spoken languages
Sedimentary Basins of the World, because he would have their beauty. The one merit of Esperanto is
not accept the agreement between the publisher and its uniformity. The apparent differences between the
myself on the role of the Series Editor. I sensed expressions in different languages are minimized.
the resentment of one or two previous volume edi- The volumes of this series have been collective ef-
tors that their hard work should be concluded by a forts, but the series does not consist of philately
poorly informed maverick with his wise cracks, but albums. The series will not be merely collections
they were too much of a gentleman to complain. of random observations, interpreted idiosyncratically
The most recent episode compels me, however, to by individual prima donnas. There is the attempt
give an explanation on the rationale of the agree- to see a parallelism in the evolution of sedimentary
ment. basins, in China, in South Pacific, in Africa, and in
Science is a language. As I indicated in the Intro- the Caribbean, and this attempt is expressed by this
duction to the Series, each article of a volume may Coda in Esperanto.
"speak" a different dialect, and each volume of the In my Introduction to the Series, I emphasized
series may "speak" a different language. The series the indispensability of classification, and the need
editor cannot possibly request revisions of individual of having all-inclusive and mutually exclusive cate-
contributions, this is the task of the volume editor. gories. I suggested two sets of criteria: isostasy, and
However, the series editor has to bring a certain orientations of the principal stresses. The classifica-
uniformity to the structuring of the various volumes. tion of basins adopted by the volume editor for the
There are the problems of different approaches of Caribbean are:
writing, of different working hypotheses for inter- Strike-slip basins
pretation, of different standards of judgment, etc. In Island-arc basins
recommending the volume editors to the publisher, Collisional basins
I have expressed my full confidence in their ability Rift basins
to do their best. The past volume editors have not Inverted basins
disappointed me, and the editor of this Caribbean The first four are recognized by the classifica-
volume has done a great job, far beyond my original tion recommended for the series. They are, respec-
expectations. I am most impressed by the assembly tively, transcurrent (transform or strike-slip) pull-
of the many contributions, especially those of his apart basins (1.2), rifted basins on active margins or
own. The duty of the series editor is to impart a back-arc basins (1.1.4), compressional basins (2.1),
continuity to the series, by expressing his under- and rift basins in continental interior (1.1.1). The
standing of the volume in a common language. This fifth category is a combination of two simple types,
is the explanation as to why the subtitle of this such as one on passive margin (1.1.3) converted later
article is called Coda in Esperanto. English gave us into a compressional (2.1) and/or transcurrent basin
Shakespeare, German gave us Goethe, and Chinese (1.2). This Caribbean volume "speaks" English with
gave us Li Bei. Esperanto is not a literary language, a strong North American accent, and the vocabulary
but is an attempt to use one and the same language. is easily translated into the Esperanto.

674 K.J. HSU
The vocabulary on the sections on tectonics has and oceanic basins came into existence in Jurassic
more of a local flavor. Having been a geologist on after Africa and South America were split off from
three continents, an Esperanto in tectonics is more North America (Chapter 1). The fragmented terranes
necessary than one in sedimentology. I have devel- underlain by continental crust include the Bahamas,
oped during the last 20 years the concept of Tectonic Yucatan/W Cuba, the Maya and Chortfs blocks.
Facies. I shall attempt in this short contribution to The deep-sea basins of the archipelago were the
express my understanding of the Caribbean geology incipient Gulf of Mexico and the proto-Caribbean.
with an Esperanto vocabulary constructed on the Thus portrayed, the Mesozoic Caribbean differed
basis of the tectonic facies concept. from the Indonesian Archipelago. The first small
oceanic basins of the Caribbean owed their origin
from regional extension, related to the fragmentation
MAJOR TECTONIC FEATURES of Pangea and displacement of Africa and South
America. The Indonesian Archipelago, on the other
The tectonic setting of the Caribbean has to be hand, resulted from back-arc seafloor spreading; that
described in terms of plate-tectonics. The volume ed- archipelago is bounded by an island-arc on the active
itor proposed the major crustal provinces as follows: plate margin. There was, however, arc-volcanism in
Chortfs block the Trans-Mexican Volcanic belt. Could the Gulf of
Great Arc of the Caribbean Mexico and the proto-Caribbean have been basins
Back-arc basins associated with the Great Arc formed by back-arc seafloor spreading? Could Cuba,
Caribbean ocean plateau Yucatan, Maya and Chortfs block have been remnant
How are these tectonic units compared to those of arcs behind the frontal arc of the Trans-Mexican
China, of the South Pacific, or of Africa? Volcanic Belt? The volume editor and his authors
In a common language to describe the tectonics of presented good evidence against such an alterna-
all regions of the world, I suggested all-inclusive and tive interpretation. The series editor accepts their
mutually exclusive categories of mountains. They working hypothesis that the Mesozoic Caribbean
are classified on the basis of the orientations of the Archipelago owed its origin to the displacement of
orientations of the principal stresses: Africa and South America from North America.
(1) Germanic Type: extensional, e.g. Black Forest
and Vosges on the side of Rhine
(2) Alpine Type: compressional, e.g. the Alps THE RHAETIDES, CELTIDES, AND ALAMANIDES OF
(3) Californian Type: transpressional, e.g. Ceno- CUBA, CHORTiS, AND HISPANIOLA
zoic California Coast Ranges
The tectonics of the Chortfs block belongs to the The sequence of events indicated by the geology
Californian Type. The other three can be interpreted of Central America and the Greater Antilles included
in terms of the tectonic facies of the Alpine Type Mesozoic rifting and subsidence, followed by Cre-
of mountains (Hsti, 1994, 1999). The three Alpine taceous and/or early Tertiary collision. The Jurassic
tectonic facies are: and Lower Cretaceous strata in areas underlain by
(a) Rhaetide: overriding block in a collision, com- continental crust are mainly carbonate strata, and
monly underlain by rigid basement. they have been considered passive margin deposits
(b) Celtide: collision zone of subduction deforma- (Chapters 4, 5, 6 and 7). The presence of arc-vol-
tion and metamorphism. canics and foredeep sediments indicates, however,
(c) Alamanide: Decollement folding and thrust of the change from a passive margin to a magmatic arc
the thin-skinned cover of the subducted block during late Cretaceous and/or early Tertiary time.
in a collision. The geology of Cuba gives evidence of a collision
These three tectonic facies have been recognized between the Greater Antilles Arc and the North
in the Caribbean region. American margin (Chapter 4). In western Cuba,
the sedimentary cover of the subducted arc, has
been stripped off to form the southerly vergent
THE CARIBBEAN ARCHIPELAGO foreland fold-and-thrust belt (alemanide facies). The
overriding block was the Bahamas Platform. The
If James Hall had devoted, like Willem van den suture zone of the collision is manifested by the
Bold, his life to a study of the Caribbean geology, ophiolite-melange (celtide facies) of Cuba.
he would not have come up with a theory of geosyn- The collision structures at the western end of the
cline. Where are the geosynclinal precursors of the Great Arc of the Caribbean are northerly vergent.
Caribbean mountains? Only the theory of plate-tec- The Chortfs basement (rhaetide) was thrust north-
tonics provides us with a more comfortable answer. ward upon the Maya block during late Cretaceous
An archipelago of islands, marine carbonate banks, and early Tertiary (Chapter 8). The suture zone is

CARIBBEAN MODELLED AFTER ARCHIPELAGO OROGENESIS 675
marked by a zone of serpentinites and metamorphic AN EXOTIC PACIFIC COMPONENT IN THE

rocks (celtide facies) of Roatfin and Barbareta is- CARIBBEAN
lands. The collision-structures are cut by the Tertiary
transcurrent faulting. Whereas the Yucatan Basin Discovery of thick Cretaceous basalt by DSDP
owed its origin to back-arc seafloor spreading, the Leg XV in 1970 gave the first indications of the un-
Cayman trough is an 1100-km long and narrow pull- usual nature of the Caribbean basins. The Venezuela
apart basin of Cenozoic age (Chapters 1 and 2).The Basin and the Colombia Basin are underlain by a
northerly vergent structure south of the Cayman Mesozoic ocean crust and a thick sequence of basalt
Trough continued eastward to northern Hispaniola, of an oceanic plateau. The basins were not formed
where the collision between the North American and by Cenozoic seafloor spreading behind a magmatic
the Caribbean Plates is well documented (Chapters arc. The Caribbean Plate, according to the mobilitic
9, 10 and 11). The active margin of the Caribbean working hypothesis adopted by the authors of this
Plate has been defined by the Hispaniola volcanic volume, "was originally an area of eastern Pacific
arc, which ceased to be active during the Eocene. Ocean floor and oceanic plateau" (Chapter 1). The
The subduction complex under the arc consists of Mesozoic ocean has become a Cenozoic basin, after
serpentinite, gabbros, blue schists, and submarine an island arc was constructed on the active plate
volcanics (celtide facies). The overriding rhaetide is margin. This conclusion is now verified by a wealth
a granodiorite, which was the root of the magmatic of geological, geophysical, and ocean-drilling evi-
arc (rhaetide facies). The collision-structures on His- dences (Chapters 19 and 20). The Beata Ridge sep-
paniola and nearby islands are also modified by arating the Venezuela and Colombia basins is not a
post-Eocene strike-slip faulting, related to the east- remnant arc such as those of the Philippine Sea. The
ward movement of the Caribbean Plate (Chapters 11, ridge was a part of the Mesozoic oceanic-plateau
12 and 13). and has been deformed by compression during the
Neogene (Chapter 21).
THE ISLAND-ARC MARGIN OF THE LESSER

ANTILLES CARIBBEAN MODEL OF ARCHIPELAGO OROGENESIS
The Cenozoic tectonic evolution of the Caribbean The geologic history of the Indonesian Archi-
island-arc margin has been very well investigated pelago has inspired me to formulate a model of
by geophysics, oceanography, and deep-sea drilling archipelago orogenesis (Hsti, 1994, 1999). My read-
(Chapters 14 and 15). There is the accretionary ing of the Caribbean volume has been a revelation. I
complex, the oldest of which has been uplifted and have come to realize that not all archipelago basins
is exposed on Barbados. There is the volcanic arc owe their origin to seafloor spreading behind the
of the Lesser Antilles. There is the Grenada Basin, island arc of an active margin. Nor are all the
formed by back-arc seafloor spreading. And there is mountains on the islands related to back-arc basin
the Aves Ridge, a remnant arc. Those tectonic units collapses. The Wilsonian processes of rifting, sub-
serve as analogues of deformed arcs and basins in duction, and collision are still the paradigm. The
orogenic belts. geology of the Indonesian Archipelago illustrates,
however, only a continent-ocean interaction. The
history of the Caribbean Archipelago started out as
AN ON-GOING A R C - C O N T I N E N T COLLISION an ocean-ocean interaction, before the collisions of
the Caribbean ocean plate with the continents.
The tectonic evolution of the Trinidad/eastern I was uncomfortable with my interpretation of
Venezuela area is distinguished by a sequence of the late Precambrian and early Paleozoic geology of
four phases: (1) pre-Jurassic, pre-rift phase, (2) a China (Hsti, 1999). The North and South China have
Jurassic rift phase related to the separation of North the sizes of microcontinents, and I had difficulty
and South Americas, (3) a Cretaceous to Oligocene in considering them remnant arcs, split off during
thermal subsidence of the South American rifted the back-arc seafloor-spreading processes. The in-
margin, and (4) a Neogene foredeep phase related terpretation of the Tarim Basin as a relict back-arc
to an arc-continent collision. Superposed on the basin also invited criticism. After my reading of
collision-deformation is a Neogene tectonics domi- the Mann Volume, I see the need of revising my
nated by strike-slip faulting, related to the eastward Geologic Atlas of China even before it is published.
movement of the Caribbean Plate relative to South North and South China were more likely microcon-
America. The Chapters, 16, 17 and 18 are written tinents, formed when the late Precambrian Pangea
in the standard "language" of plate-tectonics and was fragmented. The Tarim and Qaidam basins were
require no translation into the "Esperanto." more likely fragments of an oceanic plateau in a

676 K.J. HSU
proto-Pacific Ocean. They have survived, like the REFERENCES

Colombia and Venezuela basins, as rigid blocks
during the Phanerozoic deformations. Hsti, K.J., 1994. The Geology of Switzerland: an Introduction to
The fate of the Caribbean basins is predictable. Tectonic Facies. Princeton Univ. Press, Princeton, NJ, 250 pp.
Hsti, K.J., 1999. The Geologic Atlas of China. Elsevier, Amster-
With the sediments brought down by the Missis- dam (in press).
sippi, the Magdalena, and other rivers of North and
South America, the Caribbean Sea is destined to be
converted, in 100 or 200 million years of time, into a
desert like the Taklimakan.

Author Index *
Abbot, EL. 282, 286 Audemard, E 474, 476 Bassoullet, J.E 161,164
Aboud, N. 474 Audemard, EE. 421,423,435, 437, 455, Bateson, J.H. 80, 90
Abrams, L.J. 575, 586, 587, 597, 624 473 Bauman, E 473
Acton, G.D. 626 Austin, J.A. 90, 191,635, 666, 667 Baumgartner, EO. 113, 118, 128, 130,
Adamek, S. 51, 57, 663, 666 Austin, J.A., Jr. 29, 90, 167, 168, 148
Adams, A.I. 216 191-193, 364 Bayes, J. 365
Adatte, T. 136, 148 Av6 Lallemant, H.G. 5, 6, 13, 15, 21, Bazhenov, M.L. 109, 115, 118
Aden, L.J. 425,473, 476 29, 206, 211, 213, 217, 435, 474, Beach, D.K. 171,191
Aggarwal, Y.E 493, 494, 499, 506, 556, 503,505, 509, 541,553, 555 Beall, R. 253, 285
668 Avedik, F. 387 Beard, L.S. 285
Agterberg, EE 58, 118 Aves, H.S. 217, 223, 235 Beaudouin, T. 118
Aiello, I.W. 105, 118 Azavache, A. 424, 474 Beaumont, C. 435, 474
Aita, Y. 126, 148 Azema, J. 152, 164 Beck, C. 31, 57-59, 388, 476
Alba, J.A. 162, 164 Azpiritxaga, I. 473 Beck, M.E., Jr. 213,217
Alberding, H. 476 Beckmann, J.E 474
Aldrich, J. 217 Babb, S. 14, 21, 23, 425, 474, 513, 526, Beebe, W. 148
Aldrich, M.J., Jr. 215, 216 539, 546-548, 555, 557 Beets, D. 30
Algar, S. 478, 479, 489, 492, 494 Baby, E 385, 386, 386, 388 Beets, D.J. 562, 587
Algar, S.T. 55, 57, 57, 505, 507, 509, Bader, R.G. 630, 641,642, 666 Behrens, G.K. 351,353,364
533, 539, 553,555 Bajo, Bejarano, C. 420, 424, 474
Ali, W. 510, 557 Bakker, G. 668 Belderson, B. 669
Allenbach, E 668 Bakker, J.G.M. 626 Belderson, R.H. 372, 387
Ambeh, W.A. 388 Baldwin, S.L. 621,625 Bellizia, G.A. 435,474
Amilcar, H. 669 Bale, E 388 Bellizzia, A. 477, 494, 617, 618, 625
Amilcar, H.C. 669 Ball, M. 193 Benedetti, M. 387
Andersen, B. 192 Ball, M.M. 12, 29, 167, 169, 171, 191, Benjamini, C. 112, 118
Anderson, D.L. 584, 587, 613, 624 478, 494 Benson, W.E. 649, 659, 666
Anderson, R.S. 663, 668 Bally, A.W. 435,474, 476 Bercovici, D. 587
Anderson, T.H. 108, 112, 118, 124, 148, Bandy, W.L. 394, 415 Berggren, W.A. 305, 341
152, 163, 164, 164 Bangs, N. 371,385, 386, 388 Bermudez, EJ. 253, 285
Andreieff, E 349, 351-353, 355, 364 Banks, C.J. 489, 494 Bernier, E 164
Angelier, J. 208, 211,216, 217 Banks, L.M. 437, 474 Bernoulli, D. 191
Angstadt, D. 91 Banks, N.G. 222, 235 Berryman, K.R. 217
Angstadt, D.M. 65, 74, 90 Banks, EO. 217 Berthon, J.L. 388
Anselmetti, F. 191 Banner, ET. 159, 164 Best, D.M. 415
Anselmetti, ES. 171,191 Bargar, K.E. 217 B6thoux, N. 58
Antoine, J. 90, 588, 667 Barker, E 394, 415 Bettenstaedt, F. 556
Antoine, J.W. 120, 191,625 Barr, K.W. 425, 474, 509, 511,556 Beunk, EE 587
Aponte, A. 474 Barrett, D.L. 582, 588 Bibee, L.D. 397, 415, 416
Applegate, A.V. 88, 90 Barrett, S.E 6, 14, 16-19, 21, 23, 31, Bickle, M.J. 588
Applin, E.R. 159, 164 34, 52, 59, 108, 111, 115, 116, 120, Bien-Aime Momplaisir, R. 633-635,
Arden, D.D. 197, 199, 217 152, 163, 165, 168, 169, 188, 192, 657, 666
Areces, A. and 118 219, 235, 389-391, 411, 412, 414, Bierley, R.E. 425,473
Argus, D. 29, 218, 236, 669 415, 416, 423, 474, 475, 498, 501, Biju-Duval, B. 51, 57, 164, 370-372,
Argus, D.E 58, 217, 235, 285, 341,387, 503, 505, 506, 539, 545, 546, 551, 378, 379, 387, 388, 476, 563-566,
556 556, 557, 591, 613, 615, 616, 618, 581, 582, 587, 593, 597, 615, 625,
Arkell, W.J. 126, 141,148 619, 626, 627, 635, 668 630, 633, 635, 637, 638, 651,666
Arnstein, R. 423,473 Barros, J.A. 6, 12, 23, 30, 168, 169, 191, Bilham, R. 385,387
Arozena, J. 30, 475 192 Bird, D.E. 6, 7, 13, 28, 29, 389, 390,
Artoni, A. 385, 386, 386 Bartenstein, H. 509, 556 392, 397, 398, 415,415
Atwood, M.G. 155, 164, 165, 217 Bartok, E 423, 474 Birsh, ES. 371,387
Aubry, M.-E 341 Barton, J. 286 Bitri, A. 668
Barton, R. 152, 157, 165, 198, 218 Bizon, G. 625, 666
* Page references to text are in Roman Bass, M.N. 200-203, 205, 206, 208, Bizon, J.J. 164
type, to bibliography in italics. 213,218, 219, 223, 235 Blanc G. 387

678 AUTHOR INDEX
Blanchet, R. 31, 57, 59, 365, 388, 476 Buhl, E 58, 388, 587, 589, 625, 626, Channell, J.E.T. 662, 669
Blarez, E. 58 668, 669 Chauvin, A. 115, 118
Blome, C . D . 1 2 4 , 126-130, 141, Bullard, E.C. 124, 148 Chen, Y.J. 565, 587
148-150 Bullard, T.E 286 Cheng, Y. 126, 128, 130, 150
Blondeau, A. 285 Burbank, D.W. 375, 387 Chennouf, T. 387
Blow, W.H. 233, 235, 459, 461,474 Burckhardt, C. 126, 127, 130, 133-135, Chermak, A. 282, 286
Bobier, C. 387, 669 137-139, 146, 147, 148 Chevalier, Y. 423, 435, 439, 442, 446,
Bock, W.D. 29, 191, 494 Burkart, B. 6, 9, 29, 152, 164, 199, 200, 447, 474
Boisseau, M. 243, 245 213,217 Chiari, M. 105, 118
Boisson, D. 669 Burke, K. 4-8, 13, 17, 29, 30, 51-53, Childs, J.R. 364
Bold, W.A. 292, 297, 304, 305, 307, 55, 56, 57-59, 168, 169, 188, Chin, A. 244, 246
309, 342 191, 192, 200, 201, 217, 218, 235, Chou, G.T. 388
Bolli, H.M. 159, 164, 346, 364, 466, 289, 341, 342, 344, 360, 363, 364, Chou, T.-A. 217
474, 556 365, 388, 477, 478, 494, 498, 500, Chowns, T.M. 91
Bonet, E 154, 159, 164 501, 503, 506, 507, 509, 510, 531, Christie-Blick, N. 192, 588, 625
Bonini, W.E. 663, 667 540-542, 545, 546, 551, 553, 556, Christofferson, E. 561,587
Borgois, J. 476 557, 591, 625, 628, 651, 659, 666, Cita, M.B. 338, 341,342
Bosch, M. 618, 625 667 Clague, D.A. 586, 587
Bosselini, A. 333, 341 Burr, G. 286 Clark, G.S. 217, 235
Bott, M.H.E 415 Butterlin, J. 285, 667, 669 Clark, T.E 415
Bougault, H. 403,415 Byme, D.B. 50, 57 Cloetingh, S. 23, 31
Boulbgue, J. 387 Cluff, L.S. 218
Bourgeois, E 192 Cabrera, E. 473 Coates, A.G. 29
Bourgois, J. 31, 57, 59, 242, 246, 258, Cabrera, S. 474 Cobbold, ER. 7, 29
285, 388 Caceres Avila, E 215, 217, 219, 223, Cobiella-Reguera, J.L. 97, 100, 118
Bouwman, S.A. 476 235 Coffin, M.E 5, 29, 583-585, 587, 588,
Bouysse, E 6, 13, 14, 29, 31, 57, 59, C~iceres, D. 30, 118, 148, 556 613, 615, 625
364, 365, 370, 387, 388, 389-391, Calais, E. 29, 31, 50, 57-59, 199, 217, Cole, J.T. 414, 416
397, 398, 411-415, 415, 438, 474, 248, 249, 253, 257, 279, 281-284, Coleman, M.L. 111, 113, 118
476, 668 285, 341, 364, 388, 476, 556, 633, Colletta B. 386, 388
Bowin, C. 617, 625 663, 667 Collette, B.J. 36, 48, 58, 59
Bowin, C.O. 247, 249, 279, 285, 292, Calassou, S. 371,387 Collins, J.A. 588
296, 341 Calkins, EC. 286 Collins, L.S. 9, 29
Bowland, C. 625 Callomon, J.H. 150 Colwell, J.B. 583,587, 588
Bowland, C.L. 5, 16, 29, 30, 564, 565, Camargo, Z.A. 119 Condit, D.D. 286
581, 587, 597, 617, 625, 633, 650, Campa, M.E 130, 145, 148 Coney, EJ. 124, 130, 145, 148
663, 666, 667 Campan, A. 668 Conkin, B.M. 157, 164
Bowles, R.M. 29, 191 Cande, S.C. 36, 37, 39, 41, 42, 51, 52, Conkin, J.E. 157, 164
Boynton, ~.H. 398, 409, 415 56, 58, 59, 120, 218, 416, 668 Conners, C. 388
Boynton, W.V. 119 Cantfi-Chapa, A. 120, 135, 140, 141, Conrad, M.A. 164
Bradley, D.C. 117, 118, 235, 365 143, 148 Contreras-Montero, B. 136, 148
Brakenridge, G.R. 285 Carey, S.W. 124, 148 Coogan, A.H. 159, 162, 164
Bralower, T.J. 6, 12, 28, 29, 52, 57, 97, Carfantan, J.C. 31, 57, 59, 164, 388, 476 Cook, H.E. 282, 285
108, 117, 118 Carlson, R.L. 415 Cooke, W. 286
Bray, R. 509, 510, 548, 556 Carnevali, J.O. 423,474, 476, 557 Cooper, C. 29, 191,217, 364, 494, 625
Breen, N. 192 Carpenter, G. 387 Cooper, J.C. 317, 341,342, 668
Breen, N.A. 388 Carpenter, R.H. 153, 155, 164 Cooper, M.A. 31
Breyer, J.A. 235 Carr, M.J. 30 Coriano, M. 474
Briceno, L. 669 Carr-Brown, B. 425, 474, 510, 519, 523, Corrigan, J. 56, 58
Britton, J.C. 235 526, 556 Corrigan, J.D. 9, 29
Brodholt, J. 218, 236, 669 Carrillo-Bravo, J. 138, 148 Corso, W. 82, 90, 192
Br6nnimann, E 141,148 Case, J.E. 4, 5, 29, 50, 58, 94, 118, Cosgrove, E 503, 538-540, 547, 556
Brooks, D.A. 394, 397, 414, 415, 415 197-199, 201, 217, 360, 364, 435, Covey, M. 108, 118
Brown, D.J. 589 474, 477, 494, 561, 563, 584, 587, Coward, E.L. 29
Brown, K.M. 371,372, 378, 382, 387 593, 617, 625, 628, 639, 667, 669 Coward, L. 191
Brown, N. 90 Casero, E 385, 386, 386, 388 Cowper, S. 365
Brown, N.K. 191 Casey, J.E 29, 415, 416, 588 Cox, A. 58
Browning, J.M. 617, 618, 625 Cashman, S.M. 213, 217 Cox, EG. 475
Bryan, G.M. 193 Castrec, M. 387 Craig, L.E. 475
Bryant, W. 65, 90 Castro-Mora, M. 450, 474 Cramez, C. 476
Bryant, W.R. 167, 191 Caus, E. 165 Cramez, C.D. 471,474
Bryne, D. 635, 667 Cavanaugh, T. 577, 589 Crosby, J.T. 193
Bueno Salazar, R. 16, 29 Cazes, M. 668 Cross, T.A. 394, 415
Buffler, R.T. 6, 8, 18, 19, 21, 29, 30, Cederstrom, D.J. 345, 347, 349, 354, Crowell, J.C. 185, 191
64, 65, 67-72, 74, 82, 87, 89, 90, 91, 358, 364 Crux, J. 424, 474
108-110, 112, 119, 124, 148, 168, Chalaron, E. 379, 382, 385, 387 Curray, J.R. 397, 415
183, 191, 289, 342, 474, 588, 626, Chang, T. 35, 38-40, 58, 59 Curry, R.E 164
668 Channel, J.E. 475 Curth, EJ. 364

AUTHOR INDEX 679
Cushman, J.A. 159, 164, 347, 364 Dillon, W.E 363, 364, 437, 446, 474, Eisner, EN. 476
Cutten, H.N.C. 217 635, 667 Eldholm, O. 583-586, 587, 588, 613,
Dinkelman, M.G. 6, 16, 30, 168, 192, 615, 625
Dahlen, E 388 591,626 Elsasser, W.M. 53, 58
Dahlen EA. 382, 387, 388 Dix, C.H. 405, 415 Embley, R.W. 387
Dallmeyer, R.D. 69, 74, 90 Dixon, T. 285, 288, 289, 297, 302, 304, Emmel, EJ. 415
Dallmus, K.E 476 341,667 Emmet, EA. 153-155, 157, 164
Damond, E 588, 668 Dixon, T.D. 388 Endignoux, L. 373, 374, 385, 386, 387,
Damuth, J.E. 371,385, 387, 625 Dixon, T.H. 29, 51, 58, 362, 364, 497, 388
Daniels, D.L. 91 498, 556 Engebretson, D. 52, 58
Danilewski, D. 120 Dmitriev, L. 415 Engebretson, D.C. 52, 58
Davies, H.L. 587, 588 Dobson, L.M. 67, 82, 87, 88, 90 Engelen, J.E 218
Davila-Alcocer, V. 126, 148 Dodd, J.E. 29, 191 Engeln, J.E 236, 388, 669
Davis, D. 382, 385, 387 Dohm, C.E 253, 285 England, E 51, 58, 663, 665, 667
Davis, D.M. 388 Dolan, J. 30, 58, 192, 251, 285, 286, Eppler, D. 217
Davy, E 29 341,365 Erben, H.K. 138, 139, 148
Daza, J. 436, 437, 461,474 Dolan, J.E 6-8, 12, 13, 29, 30, 49, 55, Erdman, C.E 217
de Albear, J.E 96, 98, 100, 119, 120 58, 248, 249, 255, 274, 275, 279, Erikson, J.E 423, 424, 447, 474
de Graziansky, EC. 475 281,283,285, 290, 291,295,341 Erjavec, J.L. 285
de la Torre, A. 98, 105, 106, 108, 118 Donnelly, T.W. 5, 6, 30, 112, 115, 118, Erlich, R.N. 423, 474, 501, 503, 505,
De Leon, R. 292, 296, 341 152, 153, 155, 164, 197, 199, 200, 506, 510, 511,545, 546, 551,556
De Lepinay, B. 388 213, 217, 218, 234, 235, 358, 364, Escalante, G. 617, 625
De Lepinay, B.M. 476 561-563, 573, 577, 585, 587, 588, Escandon, M. 475
de Urreiztieta, M. 29 591, 593, 597, 613, 622, 623, 625, Escobar, C. 216, 217
de Zoeten, R. 23, 29, 248, 249, 251, 653,667 Espinosa, A. 589
253-255, 257-260, 266, 267, 273, Donovan, S.K. 526, 556 Estrada, J.J. 588
283-285, 285, 341 Dooley, T. 492, 493, 494 Eugster, H.E 334, 341,342
Dean, B.W. 200, 217 Doppelhammer, S.H. 120 Eva, A. 509, 510, 548, 556
DeBalko, D.A. 67, 82, 87, 88, 90 Douglas, R.G. 126, 148 Eva, A.N. 19, 30, 51, 56, 58, 506, 556
DeCelles, EG. 313,341 Doust, H. 668 Evans, C.C. 288, 234, 335, 340, 341
Decima, A. 342 Drake, C.L. 200, 217, 416 Evans, R. 341
Deloffre, R. 164 Draper, G. 19, 28, 30, 31, 94, 119, 135, Everett, J.E. 148
DeMets, C. 3, 4, 29, 53, 58, 59, 197, 141, 149, 169, 191, 239, 241-243, Everett, J.R. 157, 164
198, 217, 218, 220, 223, 235, 236, 246, 248, 249, 255, 257, 272, 273, EW-9501 Science Team 667
285, 288, 341, 364, 370, 371, 380, 279, 281-283, 285, 286, 342, 668 Ewing, J. 387, 564, 588, 597, 625, 634,
385, 387, 497, 498, 500, 509, 556, Drewes, H.D. 663,667 635, 667, 669
667, 669 Driscoll, N.W. 5, 16, 19, 21, 28, 29, Ewing, J.I. 397, 409, 415, 416, 588, 625,
Deng, J. 200, 217, 663, 667 580, 581, 588, 593, 597, 611, 613, 626
Dengo, G. 152, 163, 164, 197, 199, 217, 614, 617, 622, 625, 637, 641, 653, Ewing, M. 365, 371,387, 588, 667, 669
477, 494, 617-619, 625, 663, 667 667
Denny, W. 169, 171,191 Driver, E.S. 437, 474 Fahlquist, D.A. 120
Denny, W.M., III 12, 29 Drobne, K. 165 Falconer, R.K.H. 588
Denyer, E 31 Duffield, W. 217 Farfan, EE 556
Dercourt, J. 31, 58, 59, 388, 423, 441, Duncan, R.A. 17, 30, 31, 389, 415, 584, Farina, E 29, 285, 341, 364, 556, 665,
474, 476 587, 588, 591, 592, 597, 600, 613, 667
Detrick, R.S. 587, 588 625, 626, 627, 667 Faug6res, J.C. 370, 372, 385, 386, 387
Dewey, J.E 29, 35, 59, 108, 115, 118, Duque-Caro, H. 662, 667 Faulkner, B. 476
120, 168, 191, 192, 217, 218, 364, Dyer, B. 557 Feinberg, H. 285, 669
394, 415, 415, 416, 423, 475, 494, Dyer, B.L. 503, 538-540, 547, 556 Feo-Codecido, G. 439, 441,474
591,592, 625, 626 Dziewonski, A.M. 217 Feray, D.E. 165, 218
Dewey, J.W. 4, 29 Ferguson, R.C. 285
Di Croce, J. 15, 18, 19, 21, 23,420, 441, Eberle, W. 249, 253,255, 258, 259, 272, Fernandez, E 475, 556
474, 477, 494 273, 285 Fern~indez Carmona, J. 118, 119
Di Giacomo, E. 474 Eberli, G. 168, 171,188-190, 191,192 Fern~indez, J. 98, 101, 112, 118, 121
Dia, N.A. 382, 387 Eberli, G.P. 18, 19, 21, 23, 191 Fern~indez Rodriguez, G. 119
Diallo, M.C. 476 Echeverria, L.M. 562, 588 Figueroa de Sanchez, L. 505, 556
Dfaz, M.L. 118 Edgar, N.T. 29, 30, 58, 241, 246, 364, Finch, R.C. 5, 15, 18, 19, 21, 118,
Dickinson, W.R. 124, 148, 275, 277, 415, 564, 575, 588, 597, 600, 602, 152-157, 162, 163, 164, 165, 199,
279, 282, 285 617, 621,625, 667, 668 217, 235
Diebold, J.B. 5, 16, 19, 21, 28, 29, 563, Edgar, T. 667 Fink, L.K. 398, 416
565, 566, 577, 580, 581, 583, 587, Edgar, T.N. 627, 630, 631, 641, 650, Finnemore, S. 588
593, 597, 613-615, 617, 625, 633, 663, 667 Fisher, D.M. 663, 667
637, 641,649, 649, 653, 667 Edwards, L. 58, 286 Fitch, T.J. 213, 217
Dieni, I. 162, 164 Edwards, R.L. 31,192, 342 Flemings, EB. 435,474
Dietz, R.S. 124, 148 Edwards, R.S. 415, 416, 626 Flinch, J. 474
Dill, R.E 357, 364 EEZ Scan Scientific Staff 360, 364 Flinch, J.E 23,479, 494
Dillon, ES. 168, 191 Ego, E 51, 58 Flood, R.D. 625
Dillon, W. 12, 29 Eiras, J.E 441,474 Flores, G. 474

680 AUTHOR INDEX
Flores, R. 30, 118, 148, 556 Ginsburg, R.N. 167, 168, 171, 188-190, Hamoui, M. 162, 165
Flores, W. 217 191,192 Hampton, M.A. 274, 286
FRiegel, C.V. 192 Gla~on, G. 285 Hancock, J.M. 475
Fontas, E 371,384, 387 Gleason, R.J. 91 Haq, B.U. 81, 90, 163, 165, 284, 286,
Ford, R.L. 286 Glover, L. 91 334, 335, 341, 346, 356, 365, 446,
Foreman, H.E 241,246 Goff, E 217 448-450, 455, 457, 459, 461-463,
Foucher, J.E 387 Gomberg, D.M. 199, 213, 217 466, 471-473,475, 512, 556
Fourcade, E. 152, 164 Gomez-Luna, M. 148 Hardenbol, J. 58, 90, 118, 165, 286, 341,
Fox, E 57, 667 Gonthier, E. 387, 387 365, 475, 556
Fox, EJ. 29, 118, 217, 398, 416, 441, Gonzales, G. 424, 474, 475 Hardie, L.A. 334, 341
447, 474, 494, 592, 625, 628, 667 Gonzales-Leon, C. 153, 157, 159, 165 Harding, T.E 185, 191,192, 217, 366
Foye, W.G. 219, 235 Gonzalez de Juana, C. 6, 30, 419-421, Hardy, N.C. 31
Frampton, J. 425, 474, 510, 519, 523, 421, 423, 439, 441, 447, 450, 461, Hargraves, R.B. 17, 30, 120, 389, 415,
526, 556 463, 475, 478, 494, 503, 509, 548, 584, 587, 588, 591, 592, 597, 600,
Franke, M. 494 556 613,625, 627, 667
Frankel, A. 360, 364 Goodell, H.G. 168, 191 Harkrider, D.G. 416, 626
Freeman-Lynde, R.E 441,475 Gordon, M. 249, 286 Harland, W.B. 466, 475
French, R.B. 150 Gordon, M.B. 4-6, 13, 18, 21, 23, 30, Harms, EJ. 317, 341
Frey, M. 509, 556 117, 118, 144, 147, 148, 150, 152, Harrison, C.G.A. 36, 58, 494
Freymueller J. 387 153, 155, 164, 198, 199, 201, 211, Harrison, T.M. 213, 218, 625
Freymueller, J.T. 663, 667 217, 219, 223, 225, 234, 235, 552, Harry, D.L. 415
Friedman, G.M. 288, 291,336, 341,342 556 Hatten, C.W. 31, 93, 95, 97, 98, 101,
Friend, EE 8, 31, 369, 370, 385, 388, Gordon, R. 29 106, 107, 112, 115, 117, 118, 119,
489, 494 Gordon, R.G. 53, 58, 59, 217, 218, 235, 143, 148, 167, 168, 192
Frisch, W. 16, 17, 30, 218, 562, 588 236, 285, 341,387, 556, 669 Haxby, W. 59, 120, 218, 416
Frohlich, C. 666 Gorini, C. 668 Haxby, W.E 58, 587
Frost, S.H. 165, 351-353, 356, 364, 365 Gose, W.A. 23, 30, 69, 91, 152, 156, Hay, W.W. 666
Fucugauchi, J.U. 120 164, 165, 198, 217 Hayes, D.E. 397, 416
Fundora Granda, M. 120 Gou, Y. 475, 476, 556, 557 Hayward, A.B. 342
Funes, D. 474 Gouyet, S. 441,475 Heath, R.E 31
Funkhouser, H.J. 421, 437, 463, 474, Gradstein, EM. 51, 52, 58, 118 Hedberg, H.D. 421,463,474, 475
475 Graham, E.A. 365 Heezen, B.C. 59, 398, 416, 441, 474,
Furrazola-Bermtidez, G. 98, 103, 118, Graham, R.H. 342 475, 625, 628, 667
119 Graham, S.A. 275,285 Heiken, G. 199, 217
Furrer, M. 90 Grant, A.C. 588 Hellinger, S.J. 38, 39, 58
Furrer, M.A. 191, 443, 448, 450, 457, Grant, B. 622, 626 Hempton, M.R. 6, 12, 23, 30, 168, 192,
474, 509, 556 Grant-Mackie, J.A. 126, 148 235, 365
Green, C. 668 Hennion, J. 588, 625, 667
Gahagan, L.M. 29, 59, 192 Griboulard, R. 370, 371,379, 380, 384, Hennion, J.E 416, 626
Gajardo, E. 494 387 Henry, M. 415
Galea-Alvarez, E 474 Grimm, J.E 286 Henry, E 372, 387
Gallango, O. 423,474, 475, 556 Grindlay, N. 248, 286 Hernandez, E. 668
Gallo, J. 154, 164 Grindlay, N.R. 12, 30, 365 Hernandez, G. 476
Gapais, D. 29 Gripp, A.E. 53, 58 Hernandez, M. 249, 258, 286
Garcia, A. 120 Grippi, J. 57 Hernandez, N. 667
Garda, S. 291,292, 341 Grodzicki, J. 120 Hern~indez, L. 556
Gardner, T.W. 667 Groetsch, G.J. 284, 286 Herrera, N.H. 96, 97, 119
Garman, K. 168, 191 Groschel, H. 588 Herrera, N.M. 143, 148
Gayet, J. 669 Grue, K. 584, 588 Herring, J.R. 246
Gaylord, M. 192 Gruszczyfiski, M. 118 Hess, H.H. 360, 365
Gealey, W.K. 28, 30, 168, 191 Guellec, S. 388 Heubeck, C. 6, 29, 30, 50, 55, 57, 58,
Gefell, M.J. 208, 218 Gtiendel, E 31 214, 217, 220, 225, 235, 285, 295,
Geist, E.L. 357, 364 Gueneau, J. 668 341,342, 615, 625, 663, 665, 667
George, R.E, Jr. 440, 474 Guerra Pena, E 289, 341 Heubeck, C.E. 284, 286
Georges, G. 588, 668 Guerrero, J. 626 Hickey-Vargas, R. 31
Gerard, R.D. 666 Gursky, H.J. 150 Higgins, G.E. 371,378, 387, 509, 556
Gerhard, L.C. 345, 347, 349, 351-355, Guth, L.R. 213,217 Hilde, T.W.C. 394, 415, 416
364, 365 Guti6rrez, G. 30, 285 Hildebrand, A.R. 107, 119
Ghosh, N. 389, 416, 561,588 Hill, I.A. 394, 415
Gibbs, A.D. 185, 191 Haczewski, G. 87, 90, 98, 99, 111, 112, Hill, EJ. 587, 588
Giegengack, R.E 625 118, 120, 135, 141,148 Hillhouse, J. 149
Giffuni, G. 450, 474 Hagen, R.A. 584, 588 Hilst, R. 633, 667
Giffuni, R. 474 Hagstrum, J.T. 127, 128, 148 Hine, A.C. 171,192
Gilbert, L. 29, 191 Hall, R. 150 Hinz, K. 565, 577, 583, 588, 600, 625
Gilbert, L.E. 474 Hall, S.A. 29, 77, 90, 415, 416, 588 Hirdes, W. 285
Gill, I.E 13, 344, 345, 348, 350, 351, Hallam, J.M. 471,475 Hiscott, R. 286
353, 354, 357, 365 Hallock, E 556 Hobart, M.A. 387
Gillett, M. 217, 235 Hallot, E. 668 Hogg, J.R. 588, 625

AUTHOR INDEX 681
Holcombe, R.T. 587 Jacobsen, S.B. 119 Knepp, R.A. 285

Holcombe, T. 625, 667 Jaffrezo, M. 164 Kolarsky, R.A. 8, 9, 12, 15, 30
Holcombe, T.H. 631,667 Jankowsky, W.J. 447, 475, 551,556 Kolla, V. 565, 581, 588, 597, 617, 626,
Holcombe, T.L. 6, 16, 30, 58, 118, Jansma, E 29, 285, 341, 362, 364, 556, 631,668
197-201, 217, 218, 357, 364, 365, 667 Komara, S. 476
474, 561, 563, 584, 587, 588, 593, Jany, I. 13, 30, 361-363, 365, 628, 635, Korgen, B.J. 415
597, 602, 625, 628, 639, 653, 663, 662, 663, 667-669 Kouroma, S. 476
667, 668 Jaquin, T. 475 Koutsoukos, E.A.M. 509, 556
Holden, R.C. 124, 148 Jarrard, R.D. 416 Kozuch, M.J. 155, 165, 198, 199, 201,
Hon, K. 588 Jean-Poix, C. 669 214, 217, 222, 235
Hook, S.C. 388 Johnson, G.L. 474 Krantz, R.W. 185, 188, 192
Hoorn, C. 617, 618, 621,626 Johnson, H.R. 415, 416, 626 Krijnen, J. 244, 246
Hopkins, H.R. 563, 583, 588, 649, 650, Johnson, W. 667 Kring, D.A. 119
667 Jones, D. 119 Kroenke, L. 589
Hopson, C.A. 127, 128, 130, 134, 148, Jones, D.L. 124, 148, 149 Kroenke, L.W. 588
149 Jones, EC. 120 Kroonenberg, S.B. 617, 626
Hor~i6ek, J. 120 Joran, J.L. 415 Krop~i6ek, V. 120
Horne, G.S. 118, 155, 164, 165, 198, Jordan, T.E. 435, 474, 475 Kruse, S.E. 385, 386, 387
199, 217, 219, 224, 235 Jordan, T.H. 220, 235, 370, 387 Ku, T. 58, 192, 286
Hottinger, L. 159, 165 Joyce, J. 248, 249, 286, 360, 365 Ku, T.L. 31,342, 668
Houlgatte, E. 344, 362, 365 Judoley, K.M. 98, 103, 119 Kugler, H.G. 425, 475, 483, 488,
Hou~a, V. 101, 103, 105, 119 Jurgens, A. 192 494, 501, 505, 507, 509, 514, 526,
Houtz, R. 58, 668 535-537, 539, 544, 556
Houtz, R.E. 416, 564, 588, 589, 597, Kafka, A.L. 59, 236, 286, 388, 651,654, Kulstad, R. 342
626, 669 667, 669 Kutek, J. 98, 101, 103, 119, 141, 143,
Houtz, R.Z. 631,667 Kahle, H.G. 667 149
Howell, D. 94, 119 Kanamori, H. 200, 217, 394, 416
Howell, D.G. 144, 148, 149, 388 Kaneps, A.G. 246 Labaume, E 385, 387
Hoyer, M. 667 Kaniuth, K. 667 LaBrecque, J. 59, 218, 416,
Hsti, K.J. 289, 338, 341,675, 676 Karig, D.E. 394, 397, 416 LaBrecque, J.L. 58, 120, 475
Huang, T. 475 Karner, G.D. 394, 415, 416, 588, 625 Ladd, J. 193, 625, 668
Huang, Z. 58, 118 Karson, J.A. 565,588 Ladd, J.W. 6, 7, 13-16, 30, 35, 50, 51,
Hubbard, D. 365 Kasper, D.C. 618, 621,622, 626 58, 167, 168, 171, 183, 188, 192,
Hubbard, D.K. 345, 351,354, 357, 365 Kay, R. 587 388, 476, 563, 577, 583, 588, 593,
Huggett, Q. 474 Kearey, E 398, 403,413,416 626, 627, 630, 633, 635, 641, 667,
Hugh, K.E. 151,165, 218 Keen, C.E. 565, 582, 583, 588, 589 668
Hull, D. 150 Keleba, E 476 Laine, E.E 611, 613, 622, 625
Hull, D.M. 120, 127-129, 134, 148-150 Kelldorf, M. 120 Lallemant, S. 387
Husler, G. 668 Kelldorf, M.E. 120, 150 Lallemant, S.J.C. 372, 387
Husler, J. 588 Kellogg, J.N. 4, 16, 23, 30, 617, 626, Lamar, M.E. 325, 330, 341
Hussong, D.M. 397, 416 662, 663, 667-669 Lancelot, Y. 587, 624
Hutson, E 19, 28, 30, 57, 111, 112, 118, Kelsey, H.M. 217 Land, L. 217
119 Kendall, A.C. 289, 291, 334, 335, 337, Land, L.S. 365
Huyghe, E 8, 14, 370, 371, 377-379, 338, 341 Langford, R.E 341
387 Kent, D.V. 36, 51, 52, 58, 341 Langseth, M. 387
Kent, G.M. 565,588 Langseth, M.G. 371,372, 387
Imlay, R.W. 98, 103, 111,119, 126, 127, Kenyon, N. 669 Lara, M.E. 12, 23, 30
130, 133-137, 139-141, 143, 146, Kenyon, N.H. 387 Larroque, C. 387
147, 149 Kerr, A.C. 5, 16, 19, 21, 28, 30 Larson K. 387
Ingersoll, R.V. 116, 119, 275, 285, 286 Kesler, S. 279, 286 Larson, R.L. 168, 192, 587, 624
Ingle, J.C., Jr. 29 Keszthelyi, L. 588 Larue, D. 476
Inman, K.E 285 Khain, V.E. 121 Larue, D.K. 360, 362, 365, 618, 621,
Iqbal, J. 542, 557 Khudoley, K.M. 93, 97, 106, 112, 119, 622, 626
Irving, E.M. 155, 165, 219, 235 141,149, 358, 365 Latreille, M. 473
Isea, A. 473 Kidd, W.S.E 117, 118 Lawrence, D.E 213, 217
Iturralde de Arozena, J.M. 556 Kieckhefer, R. 415 Lawrence Edwards, R. 668
Iturralde, J. 494 Kieft, C. 587 Lawrence, S.R. 635, 638, 668
Iturralde-Vinent, M. 57, 118, 119, 168, Kiessling, W. 126, 149 Lawver, L.A. 29, 59, 415
169, 192 King, A.E 165, 217 Le Pichon, X. 380, 382, 387
Iturralde-Vinent, M.A. 6, 12, 28, 29, 81, Kinoshita, E.M. 441,474 Le Quellec, P. 387, 388, 668
83, 87, 88, 90, 93, 94, 96, 97, 100, Kirkland, D.W. 288, 341 Lebr6n, M. 342
104-106, 108-110, 112, 116, 117, Klaus, A. 58 L6bron, M.C. 19, 28, 30, 635, 668
118, 119, 144, 147, 149, 277, 286 Klaver, G. 30, 587 Leckie, M. 589
Ivey, M.L. 219, 223, 235 Klaver, G.T. 562, 588, 627, 668 Leckie, R.M. 626
Klenk, C.D. 217, 235 Ledgerwood, R.K. 589
Jackson, H.R. 565, 582, 588 Klitgord, K. 591,626 Lee, C.S. 397, 414, 416
Jackson, J.B.C. 29 Klitgord, K.D. 36, 41, 42, 58, 108, 112, Lee, T.-Y. 29
Jackson, T. 30 119, 168, 183, 187, 188, 191,192 Lehner, E 662, 668

682 AUTHOR INDEX
Leonard, R. 371, 387, 425, 475, 503, Makino, M. 416 Mauk, EJ. 150
542, 556 Malav6, G. 662, 668 Maurasse, E 627, 633, 635, 668
Lepvrier, C. 669 Malavielle, J. 387 Maurrasse, E 31,576, 588
Leroy, S. 8, 13, 16, 29, 30, 43, 51, 53, Malfait, B.T. 6, 16, 30, 168, 192, 591, Maury, R. 30
56, 199, 218, 565, 580, 588, 592, 626 Mayer, L.A. 193, 588, 589
615, 625, 626, 628, 630, 632, 637, Maloney, N.J. 494 Maync, W. 165
649-651, 653, 654, 659, 663, 665, Mandl, G. 185, 192 McArthur, J.M. 233,235
667-669 Mann, J.E 625 McBirney, A.R. 157, 165, 199-203, 205,
Letouzey, E 219, 235, 435, 475 Mann, E 3, 4, 6-8, 12-16, 18, 19, 21, 206, 208, 213, 217, 218, 219, 223,
Leturmy, E 386, 387, 388 23, 28, 29-31, 50-53, 55-57, 57-59, 235,236
Lewis, J. 30, 286, 588 94, 115, 118, 119, 148, 170, 188, McCabe, R.J. 394, 414, 416
Lewis, J.A. 149, 246 189, 192, 200, 201, 214, 217, 218, McCaffrey, R. 213,218
Lewis, J.E 30, 31, 94, 119, 135, 141, 219, 220, 222, 225, 234, 235, 241, McCann, W. 236
149, 168, 169, 192, 246, 285, 342, 245, 246, 247-249, 251, 253-255, McCann, W.R. 4, 14, 31, 57, 59, 286,
668 257, 259, 267, 273, 279, 281-284, 360, 363, 364, 365, 388, 403, 413,
Liddle, R.A. 421,475 285, 286, 289-297, 299-302, 307, 416, 667, 669
Lidz, B.H. 345, 347, 349-354, 356, 357, 311, 317, 320, 341, 342, 360, 362, McCarthy, J. 565,588
365 363, 364, 365, 423, 455, 474, 475, McCave, I.N. 622, 623, 626
Lillie, J.R. 385, 387 477, 494, 498, 499, 552, 556, 615, McClay, K. 379, 385, 386, 388, 492,
Lilliu, A.G. 435, 437, 439, 461,475 618, 625, 626, 628, 633, 635, 637, 493,494
Linares Cala, E. 120 638, 663, 665, 666, 667-669 McCrevey, J.A. 120
Linares, E. 31 Manton, R.S. 224, 235 McDougall, I. 213, 218
Lindberg, EA. 285 Manton, W.I. 5, 6, 8, 13, 18, 19, 21, 31, McGrew, EO. 199, 218
Lindh, T. 58 198, 199, 201, 218, 219, 222-226, McKenzie, D. 53, 58, 583, 584, 589,
Link, M.H. 283, 286, 493,494 235 613, 626
Lithgow, C. 360, 363, 365 Mao, A. 51, 58 McLaughlin, EP. 345-347, 350, 352,
Llinas, R. 246 Mariner, R.H. 342 353, 355, 356, 365, 668
Llinas, R.A. 289, 292, 296, 311,341 Marinho, M. 58 McLaughlin, EE, Jr. 288, 289, 291,295,
Logan, B.W. 289, 333, 337, 341 Mariotti, A. 387 297, 301, 304, 305, 307, 311, 317,
Long, E 588 Marriner, G.E 30 320, 330, 334, 335, 338, 342
Long, R.E. 415 Marshall, J.S. 663,667, 668 McMillen, K.J. 565, 581,588, 597, 611,
Longoria, J.E 120, 124, 126, 140, 141, Marshall, M.C. 150 617, 626
145, 149, 150 Martin, C. 120 McNally, K. 31
L6pez Quintero, J.O. 119 Martin, C.B. 133, 135, 137, 149 McNulty, C.L. 192
L6pez Rivera, J.G. 97, 98, 119 Martin, R.G. 29, 58, 118, 191,364, 474 McWilliams, M.O. 149
L6pez-Casillas, A. 162 Martfn-Bellizia, C. 439, 475 Meijer, ET. 51, 58, 663, 665, 668
L6pez-Ramos, E. 109, 111, 118, 119, Martinez, R. 474 Melia, EJ. 415
146, 149, 164, 217, 235 Martfnez, D. 95, 96, 119 Melillo, A. 192
Lorente, M.A. 626 Martinez-Cortez, A. 148 Melson, W. 587
Lowenstein, T.K. 289, 337, 341 Marton, G. 64, 67, 68, 77, 82, 83, 89, Mendoza, V. 439, 475
Lowrie, W. 475 90, 108-110, 112, 119, 668 Menendez V., A. 118
Lozej, G.E 153, 154, 159, 165 Marton, G.L. 65, 67, 68, 74, 77, 82, 83, Meng, X. 120, 133, 134, 137, 149
Lu, R.S. 415, 416, 565, 581, 588, 597, 89, 90 Mercier de L6pinay, B. 30, 31, 57, 58,
611,617, 626 Masaferro, J. 18, 19, 21, 23, 168, 59, 199, 218, 217, 248, 279, 281,
Ludwig, W.J. 397, 405, 407, 411, 416, 188-190, 192 283, 285, 365, 628, 635, 659, 662,
564, 588, 597, 626, 631,667 Maschenkov, S. 35, 36, 58 663, 668, 667, 668
Lugo, J. 8, 15, 19,21,30, 115,119,423, Mascle, A. 57, 364, 365, 370-372, 374, Mercier, J. 388
455,473, 475, 552, 556, 618, 626 377, 387, 388, 435, 475, 476, 587, Merrick, K.A. 509, 556
Lundgen, P. 669 625, 626, 663, 666, 668 Meschede, M. 213, 218, 588
Lundgren, E 218, 236 Mascle, G. 387 Meschede, W. 30
Lundgren, ER. 663, 668 Mascle, J. 35, 58 Meyerhoff, A.A. 90, 93, 97, 98, 106,
Luyendyk, B.E 148 Masclr, A.J. 164 112, 119, 120, 141, 149, 167, 168,
Lynch, L.L. 388 Masschenkov, S. 668 191,192, 365
Lynts, G.W. 167, 192 Mass6, L. 387 Meyerhoff, H.A. 357, 358, 365
Masson, D. 360-362, 365 Meyers, J. 588
Maaskant, R 587 Masson, D.G. 12, 31,357, 365, 633,668 Miall, A.D. 435,475
Macdonald, R. 509, 557 Masvall, J. 476 Michaud, F. 152, 164
MacDonald, K.C. 200, 218 Mata, S. 474, 475 Michelson, J.E. 503,556
MacDonald, W.D. 23, 29, 30, 118, 217, Mathieu, Y. 364 Middleton, G.V. 274, 286
220, 235, 494, 625 Matta, S. 474 Miles, ER. 589
Macellari, C. 477, 494 Matthews, J.E. 217, 586, 588 Millan, G. 31
Macelli, C.E. 617, 626 Mattinson, J.M. 31,148 Milhin, E. 121
MacGillavry, H.J. 669 Mattson, EH. 168, 192 Milhin, G. 93, 111, 119
Mackenzie, G. 192 Mauffret, A. 16, 29, 30, 31, 43, 51, 53, Millegan, ES. 29, 415
MacLeod, N. 150 56, 57-59, 218, 362, 365, 388, 476, Miller, D.E. 416, 626
MacPhee, R.D.E. 277, 286 589, 592, 615, 626, 628, 630, 632, Mills, R.A. 151-153, 157, 165, 198,
Mahoney, J. 587 635, 637, 649-651, 653, 654, 659, 199, 218
Mahoney, J.J. 589 663, 665, 667-669 Minshull, T.A. 588

AUTHOR INDEX 683
Minster, J.B. 220, 235 Nafe, J.E. 416 Peralta-Villar, J. 249, 272, 286
Mitacchione, V. 474 Nagle, E 242, 243, 246, 248, 249, 255, Perch-Nielsen, K. 164
Mitchum, R.M. 464, 475, 476 272, 273, 279, 281,285, 286 Pereira, J.G. 474
Mocquet, A. 668 Najmuddin, I.J. 77, 90 P6rez Lazo, J. 109, 110, 115, 120
Moiola, R.J. 267, 282, 286 Nakatsuka, T. 416 P6rez, O. 493, 494
Molnar, E 3, 31, 38, 52, 58, 59, 200, Neathery, T.L. 91 Perez, O.J. 499, 506, 556
218, 477, 494, 499, 505, 556, 651, Needham, H.D. 415 Perez-Cruz, G. 476
668 Nely, G. 388 Perfit, M.R. 19, 28, 30, 59, 200, 218,
Monechi, S. 30, 58, 285, 286, 341 Nemec, M.C. 358, 365 635, 668
Montadert, L. 587, 625 Nestell, M.K. 139, 149 Persad, K.M. 425, 475, 498, 507, 509,
Montero, W. 30, 667 Nettleton, L.L. 175, 192 513, 538, 546, 549, 556, 557
Montgomery, H. 120, 149, 241-243, Neumann, A.C. 171,192 Pessagno, E.A., Jr. 18, 19, 28, 103, 109,
245, 246 Neumann, M. 159, 162, 165 111, 120, 124, 126-130, 133-135,
Montgomery, H.A. 18, 19, 28, 124, 128, Newport, R.L. 128, 149 137, 140, 141, 144, 148-150, 245,
144, 147, 149 Newton, M.S. 494 246
Moore, C. 356, 365 Nilsen, T.H. 282, 286 Peter, G. 371,379, 388, 416
Moore, C.H. 356, 365 Nivia, A. 30 Petit, J.E 208, 218
Moore, C.H., Jr. 365 Norconsult 289, 296 Petroconsultants 440, 446, 447, 475
Moore, D.G. 415 Norell, M.A. 100, 119 Petruccione, L. 192
Moore, G.F 386 Normark, W.R. 267, 286 Phair, R.L. 74, 82, 90, 91
Moore, J.C. 371,385,387, 476 Pia, J. 159, 165
Moore, R.E 415 O'Nions, R.K. 589 Picard, X. 494, 556
Moore, W.S. 631,667 Obando, J.A. 29 Picard-Cadillat, X. 30, 475
Morijini, R. 416 Odehnal, M. 475 Pickering, K. 256, 257, 264, 266, 286
Morin, K.M. 241,246 Officer, C.B. 398, 405, 412, 415, 416, Pieri, M. 386, 388
Morley, C.K. 82, 90 593, 626
Pigram, C.J. 587, 588
Morton, J.L. 588 Ogawa, Y. 58
Pilger, R.H., Jr. 394, 415
Mosher, D.C. 588 Ogden, J.C. 365
Pilkington, M. 119
Mossakovskiy, A. 96, 119 Ogg, J.G. 58, 118, 120, 130, 149
Pimentel, M.N. 474
Mountain, G.S. 171,192 Okal, E.A. 388, 557
Pindell, J. 505, 507, 509, 533, 539, 553,
Mrozowski, C.L. 416 Okuma, S. 397, 416
555, 625
Muehlberger, W.R. 30, 152, 164, 201, Olivet, J.-L. 31, 57, 59, 388, 476
Pindell, J.L. 6, 7, 14, 16-19, 21, 23, 29,
217, 219, 223, 225, 235 O16ritz, F. 150
31, 34, 35, 46, 49-52, 55, 56, 57, 59,
Muff, R. 249, 258, 285, 286 Olson, E.C. 218
108, 111, 115, 116, 118, 120, 124,
Mugnier, J.L. 373, 374, 379, 385, 386, Onstott, T.C. 31
149, 150, 152, 163, 165, 168, 169,
387, 388 Orange, D. 192
188, 191, 192, 197, 200, 217, 218,
Muir, J.M. 146, 149 Ori, G.G. 8, 31,369, 370, 385,388, 489,
219, 235, 239, 241-243, 246, 248,
Mukhopadhyay, M. 390, 414, 416 494
249, 255, 257, 273, 279, 281-283,
Muller, C. 364, 625, 666 Orrego, A. 589
286, 364, 389-391, 412, 411, 414,
Muller, M.R. 565, 588 Ortega-Guti6rrez, F. 150
415, 416, 423, 424, 447, 474, 475,
Miiller, R.D. 3, 18, 19, 21, 23, 37, 39, Osburn, W.L. 183, 192
494, 498, 503, 539, 546, 557, 591,
48, 49, 52, 55, 56, 59, 627, 663, 668 Osiecki, ES. 201,218
592, 613, 615-618, 625, 626, 627,
Mullins, H. 193, 249, 285, 286, 341
635, 668
Mullins, H.T. 29, 167, 168, 183, 188, Palacas, J.G. 90
Pinet, ER. 198, 199, 201,215, 218, 219,
190, 192, 282, 285 Palmer, H.C. 248, 281,284, 286
222, 223,234, 235
Multer, H.G. 345, 347, 353, 354, 365 Palmer, V.W. 463,475
Piotrowska, K. 94-97, 120
Mulugetta, G. 382, 388 Pardo, G. 93, 96, 120, 169, 192
Piotrowski, J. 112, 120
Mufioz, I. 149 Pargo-Casas, E 668
Mufioz, I.M. 148, 149, 246 Paris, E 476 Pitman, W.C. 218
Mufioz, M.I. 494 Parnaud, F. 423, 455, 474, 475, 501, Pitman, W.C., III 59, 120, 416
Munoz, N.G. 556 504, 505, 546, 549, 551,556 Plafker, G. 9, 31, 51, 59, 218
Munro, S.E. 478, 479, 492, 494, 501, Parnell, J. 313,342 Planke, S. 573, 583, 588
556 Parra, M. 387 Plumley, EW. 365
Murany, E.E. 437, 475 Parson, L.M. 29, 364, 667 Poehls, K.A. 394, 416
Murchey, B.L. 127, 128, 148 Pascal, J.C. 475, 556 Pogrebitsky, Y. 35, 36, 58
Murchison, R.R. 217 Passalacqua, H. 423,435,439, 475, 546, Pons, J.C. 387
Murphy, A.J. 364 556 Popenoe, E 119, 192
Murphy, M.T. 588 Patterson, J.M. 476 Potter, D.E 583, 588
Murray, G.E. 146, 149 Paul, C.K. 474 Potter, H. 509, 557
Musgrave, J. 217 Paulus, EJ. 167, 192 Powell, C.M. 31
Musgrave, R. 589 Payne, N. 506, 507, 509, 511, 512, 519, Powers, S. 219, 235
Mutter, J.C. 565, 577, 583, 585, 585, 524, 531,556 Prasetyo, H. 388
586, 588, 600, 615, 626 Pelaez, M. 285 Prentice, C. 286
Mutti, E. 267, 282, 286 Pelton, C.D. 387 Price, E. 388
Myczyfiski, R. 93, 98-103, 105, 111, Pena, L. 286 Priest, S. 217
113, 114, 118-120, 135, 141, 143, Penfield, G.T. 119 Prieto, R. 425, 436, 437, 450, 461,463,
144, 149 Pennington, W. 666 474, 476, 503, 557
Myers, C.W. 589 Pennington, W.D. 4, 14, 31,663,668 Project Idylhim members 387

AUTHOR INDEX
684
29, 31, 35, 43, 52, 55, 56, 59, 94, 95, Schaming, M. 588, 626
Protti, M. 12, 3I
109, 111, 121, 198, 200, 211, 218, Schellekens, H. 30
Protti, S. 663, 669
219, 222, 234, 235, 248, 283, 286, Schellekens, J.H. 149, 246, 365
Prud'Homme, R. 387
Pszcz6tkowski, A. 19, 83, 87, 88, 90, 360, 363, 365, 499, 557, 564, 565, Schermer, E. 119
93-108, 111-117, 120118-120, 135, 581, 587, 617, 625, 628, 635, 637, Schermer, E.R. 148
141,143, 147, 149, 150 650, 663, 666, 669 Schilling, J.G. 415
Rosendahl, B.R. 81, 90, 565, 566, 581, Schlager, W. 64, 65, 68, 69, 74, 77, 82,
Pubellier, M. 30, 218, 637, 669
582, 588 90, 91, 167, 168, 188, 190, 192, 282,
Pugaczewska, H. 99, 100, 120
Rosenfeld, J.H. 12, 31, 52, 59 286
Pujana, I. 126, 150
Ross, C.P. 286 Schlapak, G. 447,475, 551,556
Pujol, C. 387
Ross, M. 168, 183, 188, 189, 192 Schlich, R. 583, 586, 588, 600, 626
Puscharovsky, Yu. 96, 120
Ross, M.I. 16, 31, 34, 52, 59, 108, 111, Schluter, H.-U. 588
Pushcar, P. 217
115, 116, 121, 218, 235, 286, 389, Schmidt, V.A. 108, 112, 118, 124, 148,
Pushkar, P. 217, 235
416, 423,476 152, 163, 164, 164
Pyle, T. 90
Rossello, E.A. 29 Schmitt, R. 588, 668
Pyle, T.E. 109, 120, 191
Rossi, S. 626 Scholl, D.W. 364
Pyre, A. 421,475
Rossi, T. 435, 447, 476 Schoomaker, J.E. 476
Rotstein, Y. 588, 626 Schouten, H. 36, 37, 41, 42, 58, 59, 119,
Racheboeuf, P.R. 476
Roure, E 369, 388, 420, 435, 475, 476, 591,626
Radovsky, B. 542, 557
501,503, 505, 546, 551,556, 557 Schreiber, B.C. 289, 291,311,330, 333,
Raineri, R. 159, 165
Rowe, D.W. 557 336-338, 342
Raitt, R.W. 415
Rowett, C.L. 124, 150 Schreiber, E. 342
Rambaran, V. 479, 494
Rowley, D.B. 59, 120, 218, 416 Schroeder, R. 159, 162, 165
Ramos, N. 217
Rowley, K.C. 388, 557 Schubert, C. 30, 31, 58, 477, 494, 499,
Ramsay, D.C. 577, 583, 588
Royden, L.H. 55, 59, 385, 386, 387 556, 557
Ratschbacher, L. 218
Royer, J.-Y. 35, 38-40, 59 Schumm, S.A. 618, 622, 626
Rayhorn, J.E. 415
Royer, J.Y. 668 Schwander, H. 556
Raymond, C.A. 58
Ruddiman, W.E 625, 667 Schwartz, D.P. 200, 218
Redmond, B.T. 253, 266, 277, 286
Russo, R. 388 Schwartz, R.K. 341
Reed, D.L. 15, 31
Russo, R.M. 371, 380, 388, 499, 506, Schwartz, Y. 663, 669
Reed, K.M. 128, 129, 148
545, 557, 633, 635, 663, 668, 669 Scientist Party of Leg ODP 387
Reid, H. 360, 365
Russomano, F. 473 Sclater, J.G. 31, 59, 200, 218, 235, 286
Reid, I. 565,588
Ryabukhin, A.G. 112, 121 Scotese, C. 192
Reid, J.A. 365
Ryan, W.B.E 338,341,342, 625, 667 Scotese, C.R. 31, 34, 52, 59, 108, 111,
Remane, J. 148
Ryberg, P.T. 285 115, 116, 121, 168, 183, 188, 189,
Renard, V. 365, 387, 668, 669
Sadybakasov, E. 29 192, 389, 416, 423,476
Renne, P. 19, 30, 31, 57, 115, 118, 119
Sager, W. 416 Scott, D. 588
Renne, P.R. 111,120
Saint-Marc, P. 162, 165, 285, 667 Scott, G.R. 120
Renz, H.H. 421,476
Saito, T. 669 Scott, R.W. 5, 18, 19, 21, 155, 157, 159,
Renz, 0. 192
Saleh, J. 388 162, 163, 165
Reyes, J.R. 668
Salvador, A. 73, 83, 87, 88, 90, 124, Sebrier, M. 58
Ricchi Lucchi, F. 330, 335, 336, 342
139, 150, 421, 476, 478, 479, 494, Sedlock, R.L. 131,144, 145, 150
Ricci Lucchi, E 257, 282, 286
506, 557 Seely, D.R. 366
Richard, P. 185, 188, 192
Sams, R.H. 440, 450, 474, 476 Segura Soto, R. 96, 121
Richards, M.L. 222, 235
Sanchez, H. 473 Self, S. 6, 9, 29, 586, 588
Ricou, L.E. 474
Sandoval, J. 126, 150 Sen, G. 5, 31
Rigassi-Studer, D. 93, 95, 96, 120
Sandwell, D.T. 4, 8, 9, 31, 37, 38, 41, Sen Gupta, B.K. 305, 307, 342
Ritchie, A. 217
42, 59 Seng6r, A.M.C. 53, 57, 289, 341,667
Ritchie, A.W. 155, 164, 165
Santiago, N. 474, 475 Seng6r, C. 57
Roberts, M.T. 494
Sapper, K. 219, 235 Shaffer, B.L. 165
Roberts, R.J. 155, 165, 219, 235
Sares, S.W. 286 Shafiquallah, M. 235
Robertson, P. 371, 388, 478, 494, 500,
Sarg, J.E 333, 342 Shafiqullah, M. 217
501, 506, 507, 510, 531, 540, 541,
Sarmiento, G.A. 626 Shagam, R. 118, 625
545, 546, 551,553, 557
Sarzalejo, S. 474 Shanmugan, G. 267, 282, 286
Robertson, R.P. 388
Sass, L.C. 475 Sharp, W.D. 557
Robinson, C.J. 588
Sass, L.S. 474 Shaub, F.J. 65, 74, 83, 91,148
Robir, R.M. 476
Sassi, W. 369, 388 Shaw, P.R. 37, 39, 41, 42, 59
Rodriguez, I. 618, 625
Saunders, A.D. 30 Shepard, F.P. 476
Rodriguez, J. 342
Saunders, J. 556 Shepherd, J. 557
Rodriguez, P. 95, 121
Saunders, J.B. 164, 346, 364, 371, 378, Shepherd, J.B. 499, 500, 557
Roest, W.R. 37, 48, 58, 59, 589, 668
387, 425, 474, 476, 509, 511, 519, Sheridan, R. 190, 193
Rogers, J.J.W. 587
526, 556, 557, 625, 667 Sheridan, R.E. 167, 171, 183, 188, 192,
Rogers, R.D. 152, 153, 155, 157,
Scanlon, K.M. 12, 29-31,357, 360-363, 193
161-163, 165
364, 365, 474, 633, 667, 668 Shih, T.C. 668
Rohr, G.M. 503, 557
Scanlon, N. 191 Shih, T.T. 371,388
Rona, P.A. 36, 59
Scasso, R. 126, 149 Shipboard Scientific Party 90
Roque-Marrero, E 168, 192
Schaaf, A. 285, 667 Shipley, T.H. 171, 193, 386, 587, 588,
Rosales, H. 57, 666
Rosencrantz, E. 5-7, 12, 13, 16, 23, 28, Schaefer, C.T. 307,342 624

AUTHOR INDEX 685
Shipunov, S.V. 118 Stewart, G.S. 200, 217 Toto, A. 662, 669
Shiroma, J. 29, 285, 341 Stinnesbeck, W. 148 Tournon, J. 31, 57, 59, 388, 476
Shor, G. 416 Stock, J. 38, 52, 58, 59 Trechmann, C.T. 509, 557
Shor, G.G. 415 Stoffa, E 588, 650, 651,668, 669 Tremel, H. 667
Shouten, H. 192 Stoffa, EL. 58, 193, 563, 583, 587, 589, Truchan, M. 58, 587, 589, 625, 626,
Shurbet, G.L. 345, 347, 365 593, 597, 625, 626, 669 668, 669
Sick, M. 30, 588 Storey, M. 588, 589, 626 Truskowski, I. 475, 556
Sigurdsson, H. 600, 602, 613, 621,626 Storti, F. 379, 385, 386, 388 Tsai, C.J. 668
Silberling, N.J. 149 Stow, D. 286 Tubb, S.G. 415
Silver, E.A. 15, 31,372, 388, 653,669 Straub, C. 667 Tucholke, B.E. 37, 59, 171, 192, 622,
Simmons, M.D. 164 Stride, A.H. 387 623, 626
Simonson, B.M. 155, 165 Suarez, G. 57, 662, 667, 668 Turn~ek, D. 162, 164
Sinton, C. 667 Su~ez, G. 4, 29 Twiss, R.J. 208, 218
Sinton, C.W. 5, 17, 21, 28, 31,562, 588, Suayah, I.B. 165 Tyburski, S.A. 199, 201,218, 235
597, 626 Subieta, T. 476, 557 Tyson, L. 501,506, 510, 511,519, 531,
Sinton, J.M. 587 Suchanek, T.H. 365 542, 544, 557
Six, W.M. 150 Suczek, C.A. 279, 282, 285
Six, W.M., Jr. 150 Supko, ER. 494 Umpleby, D. 588
Skogseid, J. 586, 588 Suppe, J. 375, 387, 388, 388 Unternehr, E 475
Skupinski, A. 120 Surdam, R.C. 313,342 Uyeda, S. 394, 397, 416
Sleep, N.H. 394, 416, 588 Sutter, J. 286
Sliter, W.V. 57, 118, 589 Swart, E 191 Vai, G.B. 330, 335, 336, 342
Slootweg, A.E 58 Swift, S.A. 588 Vail, ER. 90, 165, 286, 341, 365, 464,
Smith, A.G. 148, 475 Swolfs, H.C. 165, 218 466, 471,474-476, 556
Smith, A.L. 30 Sykes, L. 3, 31,668 Valastro, S., Jr. 342
Smith, D.G. 475 Sykes, L.R. 53, 59, 200, 217, 218, 220, Valentine, EG. 474
Smith, ED. 474 236, 247, 279, 282, 286, 371, 388, Valery, E 379, 387, 388
Smith, ED., Jr. 478, 479, 492, 494, 501, 403, 413, 416, 477, 494, 499, 505, van de Wiel, M. 626
556 556, 635, 651,663, 667, 669 van den Berghe, B. 295, 342
Smith, J. 192 Sykes, M.A. 476 van den Bold, W.A. 341,347, 350, 358,
Smith, M.J. 372, 388 Sylvester, R.E. 29, 191 365, 364
Smith, E 150 Van Buren, H.M. 167, 190, 192
Smith, EL. 31, 126, 150 Tabbutt, K.D. 616-619, 626 Van Fossen, M.C. 662, 669
Smith, W.H.F. 8, 9, 31, 37, 38, 41, 42, Tabor, S. 360, 365 van Gestel, J.P. 363, 365
48, 49, 55, 59, 627, 668 Tajima, E 57 Van der Hilst, R. 55, 57, 59, 663, 669
Snelson, S. 435,474 Talwani, M. 58, 563-565, 580, 588, 589, Van der Hilst, R.D. 4, 16, 31
Snoke, A.W. 503, 557 602, 626, 630, 641,668, 669 Van der Voo, R. 124, 150
Sobolev, A. 415 Tamaki, K. 392, 394, 397, 416 van Veen, E 58, 118
Somin, M. 31 Tanimoto, T. 587, 624 Van Wagoner, J.C. 154, 164, 464, 475
Soruco, R.S. 626 Tankard, A.J. 435,476, 617, 626 Vann, I.R. 296, 342
Soulas, J.E 477, 494 Tappmeyer, D.M. 217, 235 Vaughan, A.E 168, 193
Southernwood, R. 157, 165, 226, 236 Tarduno, J.A. 584, 589 Vaughan, T.W. 253, 286
Southernwood, S. 198, 199, 218 Tardy, M. 31, 57, 59, 164, 388, 476 V~izquez, M. 95, 96, 119
Spadea, E 562, 589 Tarney, J. 30 Veeken, EC.H. 447, 476
Spano, E 435,474 Tavares, I. 246, 285 Vega, V. 4, 16, 23, 30, 617, 626, 663,
Speck, R. 286 Taylor, B. 394, 415,416 667, 668
Speed, R.C. 6, 14, 31, 150, 345, 360, Taylor, D. 29 Verg6s, J. 375, 387
362, 365, 370, 380, 388, 394, 397, Taylor, D.E. 191 Verhoef, J. 58
398, 405, 409, 412, 413, 416, 435, Taylor, D.G. 124, 126, 150 Verma, H.M. 135, 139, 150
438, 476, 478, 494, 499, 506, 535, Taylor, F. 58, 286 Vernette, G. 387, 662, 663, 669
536, 539, 545, 553, 557 Taylor, F.W. 31, 192, 289, 292, 299, Vierbuchen, R.C. 192, 478, 494, 500,
Sprague, A.R. 448, 476 315,342, 668 557
Srivastava, S.E 565, 581,588, 589 Tchekhovich, V.D. 121 Vignali, M. 556
Stagg, H.M.J. 587, 588 Testarmata, M.M. 69, 91 Vila, J. 285, 669
Stainforth, R.M. 421, 476, 478, 479, Thery, J.M. 31, 57, 59, 388, 476 Vila, J.M. 31, 57, 59, 246, 285, 388,
494, 506, 557 Thierry, J. 58, 118, 475 476, 635, 637, 668, 669
Stber, K. 667 Thomas, J.C. 29 Villaroel, V. 420, 437, 476
Stein, C. 218, 236, 669 Thomas, W.A. 80, 91, 168, 183, 191 Villasenor, A. 633, 635, 669
Stein, S. 29, 58, 217, 218, 220, 235, Thompson, G.A. 588 Villasefior, A. 493,494
236, 285, 341, 370, 387, 388, 476, Thompson, S., III 476 Villeneuve, M. 441,476
556, 663, 669 Thordarson, T. 588 Viniegra, O.E 109, 121
Steiner, M.B. 199, 218 Tiley, G.J. 388 Vinour, E 388
Steinmetz, J.C. 192 Tipper, H.W. 124, 150 Vitali, C. 628, 662, 669
Stephan, E 365 Tomblin, J.E 389-392, 416 Vivas, V. 447, 476
Stephan, J.E 360-363, 365, 370, 388, Tondji Biyo, J.J. 29 Vogt, E 387
423, 424, 476, 668 Torresan, M.E. 587 Vogt, ER. 59, 217
St6phan, J.E 7, 31, 34, 56, 57, St6phan, Torrini, R., Jr. 31 Von der Hoya, H.A. 199, 201, 214, 215,
J.E 59 Toskoz, M.N. 394, 416 218, 223,224, 236

686 AUTHOR INDEX
von Hillebrandt, A. 126, 150 Wellner, R. 192 Windisch, C.C. 669

Von Huene, R. 12, 31 Welsink, H.J. 626 Winker, C.D. 289, 342
Vrielynck, B. 474 Westall, E 622, 626 Winston, G.O. 90
Westbrook, G. 58, 625, 667, 668 Winterer, E.L. 193, 588, 589
Wadge, G. 30, 51, 56, 58, 59, 201,218, Westbrook, G.K. 6, 14, 16, 30, 31, 58, Wohletz, K. 217
307, 342, 499, 500, 509, 556, 557 370-372, 378, 379, 382, 386-388, Wooden, J.L. 201,218, 307, 342
Waggoner, D.G. 31 394, 397, 398, 405, 409, 412, 413, Woodhouse, J.H. 217
Walcott, R.I. 213, 218 415, 416, 476, 668 Woodring, W.P. 286
Wald, D. 285, 341 Westercamp, D. 30, 364, 370, 387, 438, Woods, D.E 218, 236, 669
Wald, D.J. 29 474 Woodside, J. 588
Walker, G.EL. 588 Westerfield, J.R. 365 Wornardt, W.W. 464, 466, 476
Walker, J.D. 199, 218 Westermann, G.E.G. 126, 135, 139, 150 Wortel, R. 23, 31, 51, 58, 663, 665,
Walker, R.G. 282, 286 Whalen, EA. 126, 150 667
Walles, EE. 168, 169, 171,174, 193 Whetten, J.T. 344, 358, 360, 366 Worthington, L.B. 626
Walper, J. 124, 150 White, A.H. 313, 342 Worthington, L.V. 622, 626
Walper, J.L. 617, 618, 625 White, R.A. 200, 201,218 Worzel, J.L. 148, 365
Wanneson, J. 587, 625 White, R.J. 164 Wright, A. 371,388
Warburton, J. 489, 494 White, R.S. 565, 583, 584, 588, 589, Wright, W.R. 622, 626
Ward, S.N. 51, 59 613, 626 Wu, S. 437, 446, 476
Warner, A.J. 358, 366 Whitmarsh, R.B. 565, 566, 589
Wunsch, C. 622, 626
Warner, A.J., Jr. 398, 416 Whittaker, J.E. 164
Warren, J.K. 288, 337, 338,342 Wiens, D. 218
Yang, Q. 150
Watanabe, T. 416 Wiens, D.A. 57, 236, 388, 669
Yeh, K. 126, 128, 130, 150
Waters, A.C. 573,589 Wierzbowski, A. 98, 100, 101, 111, 119,
Yeh, K.Y. 150
Watkins, J. 589 121, 143, 149, 150
Yepes, H. 58
Watkins, J.S. 148, 563, 577, 583, 588, Wilcox, R.E. 357, 366
Young, G. 57, 666
593, 626, 627, 630, 633, 635, 668 Wildberg, H. 562, 589
Young, K. 153, 155, 164
Watson, M. 286 Wildermann, E. 667
Williams, C.A. 382, 388 Youngs, B.C. 313,342
Weaver, C.E. 155, 165
Webb, E 388 Williams, G.D. 8, 31 Yule, J.D. 557
Weber, J. 388 Williams, H. 157, 165, 199, 218, 219,
Weber, J.C. 371,388 236 Zehnder Mutter, C. 565, 589
Weibord, N.E. 476 Wilson, C.C. 489, 494, 501, 505, 544, Zhang, Y.-S. 587, 624
Weidmann, J. 562, 589 557 Zhao, W.L. 382, 388
Weidner, D.J. 651,654, 667 Wilson, H.H. 213, 218 Zoetemeijer, R. 369, 388
Weiland, T.J. 151,155, 157, 161,165 Wilson, J.T. 16, 31, 591,626 Zonenshein, L.P. 121
Weissel, J.K. 394, 397, 415,416 Windisch, C. 387, 589, 626 Zubieta, D. 386

Subject Index
3-Dimensional ellipsoids 40 Andean Cordillera 617 Atima Formation 151, 153, 156, 157,
3-Dimensional finite rotation 44 Andes 126, 591 159, 159, 163
Andros Bank 188 Atima limestone 163
40Ar/39Ar measurements 69 Anegada fault zone 13,362, 363 Atlantic basin 245
At~ B" horizon561 Anegada Passage 343, 357, 358, 360, Atlantic floor 391
A~t-B~ interval 633 362, 363, 622, 623, 663 Atlantic fracture zones 18
A-subduction zones 435 Angostura block 301,302 Atlantic Ocean 14, 18, 168, 291,471,
Absolute plate motion 53 Angostura Formation 302, 304, 305, 499, 591,621,624
Absolute plate motion model 52 315, 336, 340 Atlantic oceanic crust 18, 439
Abuillot Formation 253 Anguilla 354, 358 Atlantic passive margin 437, 439
Abyssal current circulation 622 Anguilla/Saba Bank 358 Atlantic passive margin of South
Acapulco-Guatemala megashear 164 Antarctic Bottom Water 622 America 435
Active margin stage 93, 117 Antarctic Intermediate Water 622, 623 Atlantic Plate 423
Africa 18, 19, 35, 183 Antarctica 52, 111, 126 Atlantic-Indian hotspot reference
Africa plate 14 Antillean island arc 144, 147, 435 system 33, 56
African margin 566 Apalachicola basin 83, 88 Atlantic-Indian hotspots 56
Afro-South America plate 65 Apennines 336 Atlantic-Indian mantle reference system
Agalteca quadrangle 153, 154 Appalachian foldbelt 80 57
Aggradational configuration 459 Appalachian-Ouachita orogenic belt 19 Atlantic-type fracture zones 37
Agua Frfa Formation 155 Aptian-Albian shallow marine Atlantic-type passive margin 420, 424
Agu~in 211,220, 222 limestone 5 Atlantis Fracture Zone 33, 35
Agu~in beds 222, 227, 234 Apure thrust fault 441 Atrato-San Juan basin 16
Agu~in fault 201, 219 Arabian shelf 163 Australia 313, 333
Agu~in Valley 219, 220, 222, 224, 226, Araya Peninsula 499, 503 Aves Ridge 6, 13, 14, 16, 389, 391,397,
227, 228, 233 Araya-Paria area 500 398, 403, 405,409, 499, 600, 602,
Airport/Penitentiary outcrop 352, 356 Araya-Paria Peninsula 503,509, 548, 623
Airport/Penitentiary roadcut 351 553 Aves Ridge remnant arc 13
Airport/Penitentiary section 349-352 Arc terranes 4 Aves Rise 593, 611
Alamino cruise 629 Arctic 582 Aves Swell 391,397, 398
Alaska 126, 141 Areo Formation 450, 473 Avocado high 507, 513, 515, 516, 518,
Alluvial-deltaic phase 435 Areo shales 450 520, 526, 528
Altamira 253, 257, 259, 260, 273 Argentina 126, 126 Avocado-Couva High 482, 487, 488,
Altamira block 249, 253,257, 272, 283 Arima fault 492 491,492
Altamira facies 253 Arroyo Barranca 327, 328 Azua basin 295, 296, 305, 317
Altamira fault zone 251,259, 270 Arroyo Barranca section 333, 336, 337 Azua-Barahona highway 331
Altamira Formation 247, 251,253, 255, Arroyo Barrero 325, 328 Azucar Member 143
257, 258, 259, 261,263, 264, 266, Arroyo Bermejo 237, 238 Azuero-Son~i fault zone 15
267, 270, 272, 274-277, 279, Arroyo Bermejo chert 243
281-284 Arroyo Berraco Blanco stratigraphic Bt~ horizon 561
Alturas de Pizarras del Sur 95 section 260 Back Rfo Grande 244
Amaime-Chaucha Terrane 617 Arroyo Blanco Formation 287, 296, 300, Back-arc basins 4, 6, 499
Amaro Formation 107, 116 305, 317, 333-336, 340 Back-arc rifting 499
Amazon 591,624 Arroyo Boca de los Guiros 331 Backstepping behavior 459
Amazon delta 621 Arroyo de la Salvia 323 Backstepping transgressive phase 471
Amazon drainage system 621,622 Arroyo del Pozo 317 Backstop 551,555
Amazon fan 621 Arroyo Guanabano 271 Backstripping 335
Ammonite-bearing limestones 93 Arroyo Honduras 331 Backthrusts 489
Ammonites 99 Arroyo Honduras-Facolina 331 Bahfa de Neiba 292, 301
Amoco Production Co. 152 Arroyo Las Lavas 267, 269 Bahfa de Neiba block 300
Anaco fault 420 Arroyo Las Lavas section 271 Bahfa Honda composite terrane 96
Anaco inversion structure 437 Arroyo Seco Formation 287, 317, 320, Bahfa Honda quadrangle 241
Anaco trend 450 326, 340 Bahfa Honda segment of the volcanic
Anc6n Formation 107, 108, 117 Artemisa Formation 100, 102, 103, 109, arc 93
Andaman basin 390, 391,394, 397, 414 143 Bahfa Honda terrane 94, 116
Andean Central Cordillera 618 Aruba Gap 627-630, 650, 657 Bahama carbonate platform 247, 249

688 SUBJECT INDEX
Bahama collision 283 Bermeja Complex 128 Canyon San Matias 137
Bahama Platform 13, 22, 52, 241,245, Best fit reconstruction 39 Capas de San Pedro 135, 137
279, 282-285 Bimini Bank 188, 190 Caracas Group 477
Bahamas 50 Bismark basin 394 Carapita Formation 424, 489, 491
Bahamas archipelago 167, 168, 169, 188 Blake Bahama Basin 130, 147 Carapita overpressure shale belt 489
Bahamas carbonate platform 168, 291 Blake drift 623 Caratas Formation 448
Bahamas Fracture Zone 65, 188 Blake Plateau 437, 446 Carbonate buildups 63
Bahamas passive margin 114, 117 Blessing Formation 343, 344, 353 Carbonate megaplatform 495, 513
Bahamas Platform 6, 18, 21, 22, 57, 93, Blue Mountains 237, 238, 243 Cariaco pull-apart basin 15, 500
111, 116, 167, 168, 189, 336, 498, Blue Mountains block 243, 244 Cariaco Trough 477, 478
499 Blue Mountains inlier 244 Caribbean 6, 18, 124, 313, 592
Bahamas Platform margin 108 Boca Wampti 159 Caribbean arc 503, 539, 552
Bahoruco fault 301 Bocon6 50, 618 Caribbean basalt province 561,585, 592,
Bahoruco fault zone 300, 304 Bocon6 right-lateral fault 15 593, 599, 613
Bahoruco Peninsula 627, 628, 637, 659, Bocon6-eastern Andean right-lateral Caribbean crust 561
660 strike-slip fault system 4 Caribbean evolution 16
Bahoruco thrust 663 Bocond-Mordn-E1 Pilar fault systems Caribbean large igneous province 564
Baja California Sur 126 618 Caribbean oceanic plateau 5, 13, 15,
Bakersfield 126 Bodo Verde 38 498, 561,580, 600, 613
Banda Sea basin 391,394, 397 Bonacca Ridge 197, 201,206, 215, 222 Caribbean oceanic plateau crust 27
Bank margins 18 Bonaire Block 618 Caribbean Plate 3, 4, 14, 28, 29, 35, 115,
Barahona 312 Bonito Oriental 220, 231 144, 145, 147, 152, 189, 211,219,
Barbados 618, 621,622 Boreal ammonites 137 241,245, 247, 248, 279, 282, 289,
Barbados accretionary wedge 438, 477, Boreal assemblage 126 291,304, 360, 362, 363, 391,403,
489 Bouguer anomalies 55, 48, 389 423,425, 435, 450, 455, 463, 471,
Barbados Ridge accretionary complex Bove Basin 441 477, 495-498, 505, 552, 580, 586,
14 Brasso Formation 487, 488, 510, 513 591,592, 615, 623, 627, 663
Barbados Ridge complex 8 Brasso shales 491 Caribbean Plate south of Hispaniola 591
Barbareta 197, 201,203 Brazil 19, 441,473 Caribbean Plateau 583, 586, 600, 627
Barrackpore oil field 538 Breakup unconformity 435 Caribbean Province 152, 153, 159, 161,
Barracuda Ridge 14, 48, 51, 55,403 Brute stacks 564 162, 164
Barranca section 326, 331 Bucaramanga segment 663 Caribbean Sea 241,249, 287, 291,295,
Barranquin section 447 Bulk density 69 301,498, 564, 628, 631,633
Barrero section 320, 328 Caribbean slab 53
Basal foredeep 424 C. Victoria, Tamps. 123 Caribbean Tertiary reef deposits 353
Basal foredeep unconformity 419, 435, Cabritos anticline 299 Caribbean-American collision 188
438, 447, 449, 457, 466, 471,472 Cabritos- 1 well 302 Caribbean-South American Plate
Basement sites 68 Cacarajfcara Formation 106, 107, 116, boundary 477, 593, 617, 618, 623
Basin classification 7 118 Carmita Formation 106, 115
Basin inversions 6 Cahuasas Formation 139, 141 Caroni Basin 477, 489, 492, 493, 507,
Basin sites 68 Caigual fault zone 533 553
Basin-bounding faults 477 Calabaza measured section 259, 264 Carrizal clastics 441
Basin-center evaporite deposits 287 Calabaza section 260, 264, 266, 269, Carrizal Formation 441
Basin-edge evaporite deposits 287 282 Carrizal-2X well 441
Bass Collier Country 12-2 88 Calc-alkaline volcanoes 14 Casanay fault 482, 488, 489, 492
Bath 243 California 7, 128 Casanay-Arima 477
Bay Islands of Honduras 197, 206, 213, California Coast Ranges 126, 128, 144, Cas-C03 651
219, 222 147 Casis cruise 627, 628, 641
Bayano-Chucanqu6 basin 15 Camajuanf belt 111, 116, 117 Casis data 630
Beata compressional zone 627 Camajuanf succession 113 Casis seismic survey 627
Beata deformed zone 663 Campana High 487, 491 Castana No. 1 223
Beata fault system 637 Campeche 145 Castilla No. 1 223
Beata fault zone 300 Campeche Bank 74 Catoche Grid 65
Beata igneous province 643 Campeche Escarpment 64, 65, 69, 89 Catoche Knoll 65, 74
Beata plateau 629, 635, 639, 653, 662 Camfi fault zone 253, 257, 272, 273, 282 Catoche Tongue 65, 69, 74, 83
Beata Ridge 16, 28, 29, 33, 50, 51, 53, Canada Bonita 258, 271 Caucagua-Paracotos-Villa de Cura belt
561,564, 584, 591,592, 593, 599, Canada Bonita anticline 263 505
600, 615, 623, 624, 627, 628, 629, Canada Bonita Member 258, 259, 260, Cay Sal 189
633, 637, 639, 641,649, 650, 651, 261,263, 264, 266 Cay Sal Bank 168, 190
657, 659, 665 Canary-Bahamas Transect 35, 36 Cayman passage 623
Beata Ridge flank 580 Cangre belt 93, 95, 109, 113, 117 Cayman pull-apart 477
Beata Ridge footwall 591 Cangre tectonic unit 116 Cayman Ridge 13, 499
Beata seamounts 637 Cangrejal 224 Cayman Rise 13
Beata volcanic plateau 627, 649, 659 Cangrejal River 220, 224 Cayman spreading center 222, 635, 637,
Bel6n Vigoa tectonic unit 113 Canoa Formation 447 663
Belize 80, 145, 199 Cantarranas Formation 153, 153 Cayman spreading center-Puerto Rico
Benioff zone 499, 635 Cation San Matias 135 trench-Lesser Antilles subduction
Bermeja accretionary complex 245 Canyon fill 463 zone 666

SUBJECT INDEX 689
Cayman transform 152 Cinco de Mayo 145 Cordillera Septentrional 247, 248, 249,
Cayman Trough 13, 33, 43, 52, 56, 197, Cinco Pesos 101 251,267, 273, 279, 282, 283, 284
199, 201,222, 234, 248, 279, 283, Cinco Pesos area 99 Cordillera system 621
663 Cinco Pesos tectonic unit 107, 117 Coriolis force 622
Cayman-Puerto Rico fault system 627 Cipero Formation 487, 488 Corpoven S.A. 420
Cayo Coco area 168 Circum-Pacific margin 126 Costa Rica 5, 9, 599
CCC Test Well 39 349 Ciudad Victoria (Tamps.) 139 Costa Rica-Panama Arc 51, 56
CCC Test Well 45a 356 Civilian Conservation Corps 345 Cotton Valley clastics 88
CCC Test Well C-26 354 Classical breakup unconformity 419 Cotton Valley sequences 63
C6baco basin complex 15 Coahuila 146 Couva evaporite 482, 488, 491
Cedros Formation 513, 524, 526 Coahuiltecana terrane 130, 133, 144 Couva Formation 487, 488
Celebes basin 397, 414 Coahuiltecano terrane 123, 130, 145, Couva Marine- 1 well 510
Cenozoic strike-slip phase 19 146 Covariance matrix 38, 40
Central America 12, 112, 249, 623 Coast of Hispaniola 12 Cretaceous Caribbean igneous province
Central American arc 592 Coast Range Ophiolite 127, 128 635
Central American Isthmus 591,623, 624 Coast Ranges425 Cretaceous Caribbean oceanic plateau 6,
Central Atlantic 188 Coastal Fringe/Margarita belt 509, 528 29
Central block of Hispaniola 663 Coastal Range/Margarita belt 503 Cretaceous Great Arc subduction
Central Brazilian Shield 617 Cochinos Sound 174 complexes 245
Central Caribbean Plate 16 Cocos oceanic crust 663 Cretaceous igneous province 666
Central Cordillera 242, 591, 617, 620, Cocos plate 3, 22, 51, 56, 663, 666 Cretaceous passive margin phase 19
621,635 Cocos Ridge 9, 12, 24 Cretaceous volcanic plateau 633
Central Cordillera arc of Hispaniola 615 Collages 6 Cruise RC 1904 405
Central Cordillera of Colombia 617 Collantes Formation 110 Crustal elements 564
Central Cuba 6, 12, 93 Collision 4 Crustal provinces 3
Central fracture valleys 37 Collisional basins 8 Crustal structure 564
Central Mexico 109, 123, 144 Colombia 4, 16, 498, 617 Crystalline basement 64, 74, 89
Central Range of Trinidad 477, 488, Colombia River basalt flows 573 Cserna Megashear 126
489, 491,492, 493, 495, 496, 498, Colombian accretionary complex and Cuba 28, 64, 93, 167, 169, 189, 241,
501,506, 509, 510, 523, 526, 528, forearc basin 16 242, 247, 498, 663
531,535, 536, 541,542, 544, 548, Colombian basin 5, 16, 28, 33, 56, 564, Cuba Fracture Zone 188
551,554, 555 565, 577, 581,599, 611,617, 650, Cuban arc 168
Central Range fault 493 651,659, 663 Cuban orogen 74, 167
Central Range fault zone 501,533 Colombian deformed belt 627, 662 Cuban orogenic belt 168
Central Range-Caigual fault zone 496, Colombian microplate 51, 52, 627, 662, Cuban tectonostratigraphic terranes 94
498, 536, 554 663,666 Cuban-Bahamas collision 189
Central Tethyan Province 126, 141 Colombian Plate 33, 56, 592, 615, 637 Cuche Formation 501,509
Central Tethyan successions 147 Colombian Plate margin 57 Cuche shale 488
Central Texas 159 Colombian prism 662 Cul-de-Sac basin 291,292
Central Venezuelan Basin 622 Colombian trench 22 Cunapo Conglomerate 482, 487, 488,
Central Venezuelan Fault Zone 566, 582 Columbia University 392 492, 493, 511, 531
Central Yucat~in basin 13 Columbus Basin 425, 503, 542 Cunapo Formation 487, 509, 510, 513,
Cerro de Cabras tectonic unit 95 Columbus channel 437 516, 519, 520, 521,528
Cerro de la Virgen Limestone 213 Columbus foredeep 501 Curwao 599, 627
Cerro La Penita 226, 227, 228 Combined rotation model 52 Curwao Ridge 565, 582, 617, 653
Cerro Las Lomas 228 Common isochron segment 39 Current-controlled drift deposit 591
Cerro Panteon quarry unit 2 135, 136, Compressional satellite ('piggyback') Cusps 41
137 basins 455
Cerro Wamp6 157 Confluencia Rfos Patuca y Wamp6 Dajabon 237
Chapulhuac~in 137 quadrangle 155 Dajabon area 237
Chapulhuac~in Limestone 136, 137, 141 Conjugate isochrons 39 Dajabon cherts 238, 245
Charco Largo-1 well 289, 296, 297, 301, Conrad 577 Dajabon rocks 242
304, 305, 307, 315, 316, 337, 339, Continent-ocean transition 551 Danlf 155
340 Continental bridge 89 De Purus arch 618
Chaudiere Formation 510 Continental encroachment cycle 471 Dead Sea of Israel 337
Chevron Corporation 65 Continental fragment 4 Decompression melting 584
Chiapas 145 Continental lithospheric A-subduction Deep eastern Gulf 82
Chiapas-YucaUin area 498 435 Deep-water (flysch) phase 435
Chicxulub structure of Yucat~in 107 Continental shelf of Venezuela 407 Deep-water canyon cutting event 463
Chimana Formation 447 Continental slope of Venezuela 413 Deformation front 14
Choc6 Block 617 Cooperaci6n Dominicana de Empresas Degrees of freedom 39
Chocoy No. 2 146 Estatales 296, 312 Delaware basin, Texas 337
Chorotega block 152 Cordillera Central 287, 291,292, 295, Demerara Plateau 441,447
Chortfs block 4, 6, 8, 19, 21, 151,155, 296, 301,307, 317 Depth to basement calculations 6, 28
159, 197-201,211,213, 215 Cordillera de Guaniguanico 83, 87, 88 Devil's River 238, 243
Cibao 249 Cordillera de la Costa belt 503 Difference vectors 35
Cibao basin 249, 251,288, 291,340 Cordillera de la Costa Nappe 477 Dipping reflectors 564
Cibao Valley 253, 334, 633, 635 Cordillera Nombre de Dios 220, 223 Dispersion 39

690 SUBJECT INDEX
Dix formula 405 Eastern Caribbean structural barrier 622 E1 Soldado fault 552
DNAG timescale 51 Eastern Cordillera 591,620, 621,635 E1 Soldado fault zone 505
Doldrums 37 Eastern Cuba 93 E1 Tambor Formation 211
Doldrums fracture zone 41 Eastern Guatemala 112 E1 Tambor Group 200, 211
Dolomita Principale peritidal complex Eastern Maturfn sub-basin 424 E1 Verde Member 138
333 Eastern Mexico 199 Elk Point basin 289
Dolomite Mountains of Italy 333 Eastern Pacific 14, 503, 584, 591, 613, Ellipsoids 40, 44
Dominican Cartographic Institute 296 623 England 162
Dominican Republic 128, 237, 238, 242, Eastern Pacific Ocean floor 16 Enriquillo 249
245, 247, 284, 287, 288, 292, 300, Eastern passive margin of Yucat~in 117 Enriquillo basin 287, 288, 289, 291,292,
334 Eastern Venezuela 22, 33, 57, 421,495 295, 304, 309, 336, 337, 340, 635,
Dominican sub-basin 628, 637 Eastern Venezuela foredeep 471 637
Domoil High 477, 482, 487, 492, 526, Eastern Venezuela Geosyncline 421 Enriquillo depression 635, 637
528, 533 Eastern Venezuela offshore 471 Enriquillo fault 635, 637, 663, 666
Domoil-Gupe High 488 Eastern Venezuelan Basin 15, 419, 421, Enriquillo Valley 287, 296, 313, 315,
Doubloon Saxon 1 well 168, 171, 174 423,425,441,491,501,544, 545 334, 337, 340
Drift stage 93, 117 Eastern Venezuelan fold belt 491 Enriquillo-Plantain Garden fault zone
Drifting stage 65 Eastern Venezuelan fold-and-thrust belt 13, 291,300, 301,302, 304
DSDP 587 477 Equatorial Atlantic 35
DSDP 151 ridge 629, 651,653, 659 Eastern Venezuelan foreland basin 501 Erin Basin 489, 491
DSDP 151 ridge-Taino ridge area 659 Eastern Yucat~in 123 Erin-Siparia syncline 501
DSDP cores 591 Eastern Yucat~in block 69 ERS- 1 altimetry data 36
DSDP dated drill samples 27 Easternmost Honduras 151 ERS- 1 data 41
DSDP drilling 561 Edge-driven plate tectonic interactions Escambray terrane 94, 109, 110, 112
DSDP drilling results 63 53 Escape of the Caribbean 53
DSDP Hole 538A 83 Eirik drift 623 Esperanza belt 103
DSDP holes 631 E1 Abra Formation 161' 162 Espino Graben 441
DSDP Leg 15 561 E1 Abra Limestone 146 Esqufas 155
DSDP Leg 15 drilling 602 E1 Aguacate 245 Esqufas Formation 151, 155, 163
DSDP Leg 15 sites 599 E1 Americano Member 88, 101, 101, Estate Work and Rest 349, 355
DSDP Leg 15, Site 1001 561 143 Etang Saumatre 292, 292
DSDP Leg 77 65 E1Cacheal tufts 249, 253, 281,283 Eugenia Formation 126
DSDP Site A 241 E1 Cantil Formation 447 Evans Highway 351,353, 356
DSDP Site 4 241 E1 Carbon 220, 226, 227, 231,234 Evaporites 83, 287
DSDP Site 31 641,649 E1 Coche-North Coast 499, 501 Ewing 9501 cruise 629, 637
DSDP Site 95 241 E1 Coche-North Coast fault zone 501 EW-9501 profiles 565
DSDP Site 144 447 E1 Furrial area 450 Exogeosyncline 421
DSDP Site 146 241,600, 602, 612, 622, E1 Furrial fields 423 Extrusive volcanic mounds 564
623 E1 Furial-Carito oil fields 457, 473 Exuma Sound 171
DSDP Site 150 575, 600, 602, 612, 622, E1 Furrial-Carito trend 450
623 E1 Granado 321 Fairplain 351
DSDP Site 151 633 E1 Granado section 320, 321,325, 326 Fairplain fault 353, 356
DSDP Site 152 575, 599, 615 E1 Limon 267, 270 Falcon-Aruba 477
DSDP Site 535 68, 80, 88, 89 E1 Limon Member 267, 269-272, 282 Farallon Plate 592, 108, 168
DSDP Site 536 68, 89 E1 Mamey 253, 257, 258, 282 Fault A 300, 301,302
DSDP Site 537 68, 69, 77, 88 E1 Mamey Formation 253 Fault B 301,302, 304
DSDP Site 538 68, 77 E1 Mamey Group 247, 255, 257, 273, Fault C 301,301
DSDP Site 540 68, 69, 80, 89 274, 275, 277, 282, 283, 284 Fault zone D 301
DSDP sites 82, 599, 615 E1 Mamey measured section 260 Fault-angle depressions 7
DSDP well control 82 E1 Mamey section 259, 260, 267, 282 Fault-wedge basins 7
DSDP wells 89 E1 Mochito 154, 159 Felicidades 94
DSRV Alvin 357 E1 Pastor Member 138 Ferro-di-lancia 336
Duarte Complex 128, 242 E1 Pilar 51 Fiji Plateau 415
Duarte highway 240 E1Pilar fault 15, 477, 478, 492, 495, Finite motion poles 35
Durango 130 496, 506, 517, 554 Finite plate motion poles 38
Durham sand 525 E1 Pilaf fault scarp 531 Finite rotation poles 44
E1 Pilar fault zone 15, 495, 496, 498, Finite rotations 36, 40
Early Cretaceous post-rift sequence 64, 499, 505, 506, 507, 511,515, 518, First-motion studies 3
89 528, 539, 541,542, 554, 555 First-order cycle 472
Early shallow carbonate bank phase 546 E1 Pilar sub-basin 507, 513, 518, 521, Five Corners 349, 350
Earthquakes 3 523, 524, 553 Flexural basins 435
East End 344 E1 Plan Formation 155 Florida 441
East End Range 353 E1 Progreso 220 Florida block 63, 83
East Pacific Rise 12 E1 Sfibalo Formation 99, 100, 109, 111, Florida Escarpment 64, 69, 82, 89
East Panama deformed belt 15 112, 113 Florida Plain 74, 77
East-west shortening 56 E1 S~ibalo/Artemisa boundary 100 Florida platform 89
East-central Oregon 128 E1 Salvador 9 Florida scarp 437
Eastern Caribbean 413, 585, 591, 611 E1 Soldado 499 Florida-Bahamas area 187

SUBJECT INDEX 691
Florida-Bahamas block 65 GPS-based geodetic studies 4, 28, 29 515, 526, 528, 533
Flow lines 37, 41, 56 Graben 63, 81, 89 Gulf of Guinea 566
Flysch-type sediments 510 Gran Mangle 253 Gulf of Honduras 199
Foothills 503 Gran Mangle series 281 Gulf of Maracaibo 617
Foothills belt 505 Graphitic schist 5 Gulf of Mexico 6, 18, 124, 144, 145,
Foothills fold-thrust belt 499, 501 Gravitational collapse 437 153, 159, 168, 188, 289, 291,437,
Foraminifers 99 Gravity maps 8 623
Forearc basement 27 Great Abaco fracture zone 167 Gulf of Mexico basin 63
Foredeep 419, 421,425, 435 Great Arc of the Caribbean 6, 7, 12, 13, Gulf of Paria 14, 419, 477, 478, 487,
Foredeep clastic wedge 435 17, 18, 27, 28, 200, 241,245, 498 488, 489, 491,495
Foredeep phase 419, 457, 465 Great Bahama Bank 167, 168, 174, 187, Gulf of Paria basin 495, 498, 507, 524,
Foredeep sequence 435 188, 191 539, 553, 554
Foredeep sequence regime 438 Great circle segments 39 Gulf of Paria fault zone 501
Foredeep subsidence 553 Great Valley Supergroup 126 Gulf of Paria-Northern basin 495, 503
Foreland 435 Greater Antilles 50, 55, 128, 168, 247, Gulf Tectonics 65
Foreland basin 12, 419, 425 343, 391,397, 622 GULFREX-Gulf 65
Forestepping regressive phase 471 Greater Antilles Arc 93, 108, 111, 115, Guyana 19, 498
Four-North fracture zone 37, 41 118, 168, 248 Guyana Basin 447
Fourth-order sequence 464 Greater Antilles drift 623 Guyana Craton 441
Fourth-order unconformity 435 Greenland 109, 144, 147 Guyana margin 551
Fracture zones 8, 36 Grenada basin 6, 7, 13, 28, 389, 392, Guyana offshore 447, 448
Franciscan Complex 128 397, 398, 403,405, 407, 413,414, Guyana offshore basin 447
Francisco Formation 87, 100, 101, 111, 415, 499 Guyana passive margin 14, 440
113, 141 Grenville crustal age province 19 Guyana Shield 15, 19, 439, 446, 455,
Fredensburg Quarry 349 Growth fault zone 463 466, 496, 505, 546, 551,617, 621
Fredericksburg Group 159 Growth-faults 437
Fredericksted 353 GSA Decade of North American Haiti 291,292, 599, 629
Free-air gravity 8 Geology 7 Haiti Basin 601
Free face 28, 551,555 Guacamaya Formation 139 Haiti plateau 629, 635
Freites Formation 424, 461 Guachichil 144 Haiti sub-basin 628, 629, 635, 651,653
French Seacarib cruise 628 Guachichil terrane 145 Haitian border 309
Frontal thrust 618 Guajaib6n Formation 96 Half-grabens 8, 13, 63, 437, 443, 546
Full-graben setting 81 Guajaib6n-Sierra Azul belt 95, 96, 114 Harvard focal mechanism catalogue 3
Full-grabens 8 Guajaib6n-Sierra Azul unit 94 Hato Viejo Formation 441
Fusulinacea 99 Guanahacabibes Peninsula 95, 109 Havre basin 391,394, 397
Guanaja 197, 201,222 Haynesville Formation 87
Galapagos hotspot 9, 17, 584, 587, 599, Guanaja Island 203, 205, 216 Haynesville sequences 63
600, 613 Guananico measured section 260 Heat-flow measurements 28
Galapagos hotspot scenario 592 Guananico section 259, 260, 263, 267, Hess Escarpment 16, 28, 591,600, 613,
Galapagos mantle plume 592 282 615, 623, 624, 629, 653, 663
Galapagos rift 12, 22 Guane 95, 96 Hess Oil refinery 353
Galapagos seafloor 12 Guaniguanico belts 113 Hidalgo 123, 147
Galeota Point thrust fault 501,503 Guaniguanico foreland basin 108 Higher Boreal paleolatitudes 124
Galice Formation 128 Guaniguanico nappe 97 Hildago 137
Garcfa Member 447 Guaniguanico tectonostratigraphic belts Himalayas 111
Gateway 591 96 Hispaniola 6, 7, 12, 13, 16, 24, 57, 241,
Gazzi-Dickinson point-count method Guaniguanico tectonostratigraphic unit 247, 248, 275, 283, 284, 287, 289,
275 93 291,336, 413,498, 624, 627, 635,
Geanticlinal welt 421 Guaniguanico terrane 93, 94, 96, 97, 637, 659, 663
Geosat altimetry data 35, 36 103, 107, 109, 111, 112, 113, 116, Hispaniola arc 247, 279, 284
Geosat data 41 117, 118, 144, 147 Hispaniola margin 615
Geosat gravity data 18 Guanoco area 477, 492 Hispaniola restraining bend 13
Geostrophic flows 622 Guare Member 155, 162 Hispaniola terranes 241
Glacio-eustatic sea-level fluctuations 464 Gu~irico 15 Hole 146 630
Global sea level lowstand 284 Gu~irico sub-basin 419, 420, 424, 439, Hole 153 630
GLORIA data 48 501,503, 505, 544 Hole 31 630
GLORIA imagery 357 Guasasa Formation 101,102, 143, 144, Honduran shelf break 222
Gonave microplate 57, 637, 663 147 Honduras 5, 9, 110, 154, 163, 197, 199,
Gonave-Venezuelan microplate 666 Guatapajaro Anticline 489, 536 200, 219
Gondwana crust 145 Guatemala 9, 80, 110, 145, 153, 199, Honduras and Guatemala 216
Goodrich basin 533, 554, 555 200, 222, 477 Honduras Group 155
Goodrich pull-apart basin 529, 542 Guatemala-Belize coastline 234 Honduras-Facolina section 320
Goodrich sub-basin 495, 496, 507, 513, Guerrero 126 Horizon A" 630
515, 516, 520, 521,522, 524, 525, Guerrero block 152, 163 Horizon B" 630
526, 528, 531,539, 542, 553, 555 Guiamas River 225 Horst 81, 89
Gopa High 477, 482 Gulf Coast 154, 157 Hotspot trace 9
Gopa-Posa Highs 488 Gulf Coast region 159 Huayacocotla Anticlinorium 138, 146
Gorgona 599 Gulf High 482, 487, 488, 492, 507, 513, Huayacocotla Formation 138, 139, 140

692 SUBJECT I N D E X
Huayacocotla remnant 129, 139, 141, Kerguelen 583 257, 260, 264, 267, 269, 270, 271,
143, 144, 145, 147 Kerguelen Plateau 577, 583, 586 272, 274-277, 279, 282, 284
Huayacocotla segment 123, 145, 147 Kingshill basin 343, 346, 353, 354, 363 Las Lavas measured section 269
Huayacocotla terrane 145 Kingshill graben 347, 357 Las Lavas section 267, 271
Hyde Formation 129 Kingshill Limestone 343, 344, 345, 347, Las Mangas tonalite 224
Hydrocarbon deposits 28 358, 363 Las Piedras Formation 424
Hydrocarbon occurrence 288 Kingshill strata 354 Las Salinas 296
Hydrocarbon source rocks 288 Klamath Mountains 128, 144, 147 Las Salinas fault zone 311
Hydrocarbons 498 Knowles ramp 88 Las Salinas Formation 287, 296, 302,
Krause Lagoon 356 304, 307, 311,315, 316, 328, 336
Iberian margin 565, 566, 581 Krausirpi beds 151,157, 159, 161,162, Late Berriasian unconformity 80
Ilama Formation 157 163 Late Cretaceous-Recent arc-passive
ile-fi-Vache structure 635 Krausirpi quadrangle 159, 162 margin collisional phase 19
Imbert 243 Kroonvlag data 36 Late Jurassic rift phase 18
Imbert Formation 242, 243, 281 Kurile basin 397 Late Jurassic rift sequence 64, 89
Inclined subducted slabs 4 Late Jurassic syn-rift 74, 89
Indian ocean hotspot tracks 52 La Caja Formation 128, 135, 137, 139, Late Jurassic-Early Cretaceous seaway
Infierno Member 143 146 89
Inner arc setting 281 La Caja Unit E 138 Later deeper-water passive margin phase
Inner forearc (Los Hidalgos Formation) La Casita 135 548
283 La Casita Formation 135 Latitude, longitude, and rotation angle
Institute Fran~ais du P6trole 392, 629 La Ceiba 211,220, 222, 223, 224 space 40
Institute for Geophysics 65 La Ceiba fault 201,215, 223 Latitude-longitude sphere 40
Integral constraints 39, 41 La Cumbre Ridge 273 Lau basin 391,394, 415
Intermediate unconformity (SB-2) 438 La D6sirade 128, 245, 391,398 Lebanon 162
Internal plate deformation 6 La Esperanza 211 Leeward Antilles 6, 391,503
Inversion method 35, 38 La Esperanza belt 93, 95, 105, 112, 114, Left-lateral Santa Marta-Bucaramanga
Inverted basins 8 117 fault 15
Inverted Caribbean sedimentary basins 8 La Gloria Formation 133 Leg 77 volume 74
Iquitos arch 618 La Gtiira Member 107 Lesser Antilles 3, 391,392, 397, 398,
Isla Cabritos 301 La Isla Formation 281 407, 499, 622, 623
Isla de la Juventud 93 La Legua Member 107 Lesser Antilles Arc 3, 4, 8, 13, 14, 48,
Island-arc basins 7 La Palma 96, 112 55, 343, 360, 397, 389, 403,405,
Island-arc terranes 249 La Pefia 136 409, 423,539, 555
Israel 162 La Pica Formation 424, 462 Lesser Antilles Seismic Project 409
Isthmus of Panama 15 La Pocilguita del Limon 267 Lesser Antilles subduction zone 14, 48,
Isthmus of Tehuantepec 145 La Pocilguita Member 267, 270, 271, 498, 499, 500, 663
Italy 335 272 Limestone Caribbees 14
Izee terrane 128, 129, 141 La Quinta Formation 441 Lithospheric trace 14
La Reine Member 343, 349, 350, 351, Llanada and Point Sal remnants of the
Jacaguas Formation 143 353, 354, 356 CRO 130
Jagua Clara M61ange 243 La Toca block 249, 251,253, 255, 272, Llanos syncline measured section 266
Jagua Formation 87, 100, 101, 111, 113, 273, 283 Llanos syncline section 260, 266, 267,
143 La Toca Formation 242, 251,253, 255, 282
Jagua Vieja Member 100, 143 272-275, 277, 279, 282, 283, 284 Local structural rotation 24
Jaitique 155 La Zarza Member 88, 102, 103, 143 Loma de Sal y Yeso 287, 289, 296, 302,
Jaitique Formation 151, 155, 156, 162, Labrador Sea 565, 581 307, 309, 312, 315, 340
163 Lac Assal, Djoubouti, Persian Gulf 337 Loma del Muerto tectonic unit 112
Jamaica 13, 16, 52, 56, 153, 237, 243, Lago Enriquillo 292 Los Bajos 499
245, 247, 477, 635, 663, 666 Lago Enriquillo area 297 Los Bajos fault 478, 488, 489, 492, 496,
Japan 126 Lago Enriquillo block 302 499, 503, 541,552, 555
Jealousy 345 Lagoven S.A. 420 Los Bajos fault zone 501,505, 506, 555
Jealousy Formation 343, 344, 345, 347, Laguna Rinc6n 299 Los Cayos Member 107
353, 354, 358, 363 Lake Macleod, Western Australia 289, Los Guiros syncline 317, 320
Jealousy/Kingshill boundary 345 333, 337 Los Hidalgos Formation 247, 249, 253,
Jimanf 307 Lake Maracaibo 22, 552 257, 258, 260, 266, 270, 281,284
Jimanf Formation 304, 307, 336 Lake Maracaibo region 618 Los Hidalgos Pass 260
Jordan 65 Lake Yojoa 151, 152, 154, 155, 159, Los Jabillos clastics 450
Jordan Knoll 89 162, 163 Los Jabillos Formation 473
Josephine Ophiolite 128, 129 Lakes 9 Los Organos 94
Judith's Fancy 354 Lamont Doherty Earth Sciences Louann salt 87, 289
Jurassic half-grabens 441 Observatory 629 Louann-Campeche salt 83
Jurassic rifting event 473 Lamont-Doherty Geological Observatory Lower Carapita Formation 457, 466
Jurassic-Cretaceous boundary 69 392 Lower Cretaceous carbonate margins 64
Jutiapa-Trujillo road 220 Landward-dipping 'D' reflectors 583 Lower Cretaceous post-rift sequence 74
Large igenous province 575 Lower Forest Formation 538, 555
Kallinago basin 14 Large-scale plate-tectonic rotation 24 Lower La Pica Formation 487, 488
Kane Fracture Zone 33, 35, 37 Las Lavas Formation 247, 251,255, Lower Merecure Formation 457

SUBJECT INDEX 693
Lower Nicaragua block 666 459, 461 Montafia de Santa B~irbara 159
Lower Oficina Formation 466 Maya block 152, 197, 199, 200, 211, Montafias de Col6n 151,153, 155, 157
Lower Talparo Formation 526 213,215 Montafias de Col6n fold belt 152
Lower Tethyan paleolatitudes 124 Maya terrane 144, 145 Montafias de Col6n region 157
Lower Valle de Angeles Group 151 Mazapil 123, 130, 135, 138, 147, 146 Montafias de Col6n-Rfo Wampti 152
Lowstand 333 Mazapil remnant 137, 138, 139, 147 Monte Cristi 267
Lowstand fan 284 Mazapil succession 123, 147 Montserrat glauconitic sandstone 519
Lowstand prograding wedge 461 Median back-arc basin 8, 9 Moreno depocenter 93, 115, 116, 118
Lucas Formation 105 Median-Nicaraguan back-arc basin 7, 9 Moreno Formation 93, 106, 115, 117
Luperon facies 253 Mediterranean 289, 336, 337 Morningstar sections 349
Luperon Formation 255, 281 Mella block 301 Morochito piggyback basin 503
Lutgarda Formation 116 Mella wells 296, 297, 301 Mor6n-E1 Pilar fault zone 499, 500, 501
Melvin Evans Highway 349 Mor6n fault 478
Macropaleontologic data 27 Mercurius 37 Morro area 488
Magadi-type chert 313 Mercurius fracture zones 41 Morro northward-vergent imbricates 489
Magante 240 Merecure Formation 457 Motagua 477
Magdalena deep-sea fan 631 Merecure Group 450 Motagua fault 197, 200
Magdalena River 621 Merecure type section 450 Motagua fault zone 197, 199, 200, 211,
Magnetic anomaly 35 M6rida Andes 622 213,216
Magnetic anomaly and fracture zone Mesa Formation 424 Motagua Valley 211,222
date sets 35 Mesozoic Caribbean Arc 6 Motagua/Chixoy-Polochic fault zones
Magnetic data 55 Mesozoic passive margin 472 152
Magua Formation 242, 281 Messinian 334, 463 Motagua-Polochfc system 9
Maimon Bay 239, 242 Messinian event 356 Motion vectors 35
Maimon-Amina schists complex 242 Messinian sea-level lowering 419 Mount Harris 535
Mamey group 242 Mestanza tectonic unit 95
Mount Harris push-up block 509
Manacas Formation 108, 116 Metamorphic protolith rocks 4
Mt. Eagle Group 344, 357
Managua Lake 9 Mexico 124, 126, 146, 153, 159, 161,
Mt. Eagle Series 349
Manihiki 5 211
Muertos prism 635, 637
Mannings Hill 351,352 Mexico-Marathon-Ouachita-
Muertos trench 13, 16, 627, 653
Mannings Bay Member 343, 349, 351, Appalachian structural belt 124
Muertos Trough 33, 50, 55, 56, 360,
352, 356 Microcontinent 18
363, 591,601,615, 624, 627, 628,
Mantle plume 583 Micropaleontologic data 27
635, 637, 663
Mantle plume source 584 Mid-Atlantic ridge 627
Multichannel seismic reflection data 591
Mantua 96, 112 Middle America arc 3, 4, 6
Multifold seismic data 63
Manzanilla Formation 482, 487, 488, Middle America subduction zone 3,498
492, 510, 511,513, 519, 521 Middle America Trench 234, 663
Nafe-Drake curve 405
Maparito faults 437 Middle America Trench subduction zone
Naranjo tectonic unit 113
Maracaibo 15 152
Naricual clastics 450
Maracaibo basin 15, 419 Middle America volcanic arc and trench
Maracaibo block 4, 15, 24, 51, 56 Naricual Formation 473
9
Maracaibo-Peruvian foreland basin 591, Middle Cretaceous Sequence Boundary Nariva fold-and-thrust belt of southern
617, 618, 624 69, 74, 82 Trinidad 489
Maracaibo-Peruvian Trough 617 Middle East 159 Nariva turbidites 491
Marajo basin 618, 621,622, 624 Middle Fork of Smith River 128 Navarette 258, 267
Marathon 37, 41 Mid-Las Salinas horizon 304 Navarette measured section 271
Margarita 503 Mid-Las Salinas reflector 299, 301,304 Navarette section 267, 271
Margarita Island 503 Minas de Matahambre 112 Navassa trough 635
Mariana basin 397 Minas de Matahambre-La Palma area Nazca plate 3, 16, 22, 663, 666
Marie Aimee ridge 660 96 NE Yucat~in coast 93
Marine Geophysical Data Center 628 Misfits 39 Neiba prism 637
Marine heat-flow measurements 6 Mississippi Canyon 65 Neiba-Plaisance Formation 297
Marine seaway 89 Mississippi Fan turbidite plain (Florida Neogene foredeep 472
Marine seismic profiles 6 Plain) 64 Netherlands-Antilles 391
Martfn Mesa 1 well 96 Mobil 296, 297, 304 Nevadian island arc 144, 147
Martfn Mesa tectonic window 96 Mochito area 154 New Zealand 7, 111, 126
Martin Vaz 38 Mochito Mine 154 Nicaragua 9, 152
Matagalpa Formation 157 Mochito shale 151, 154 Nicaragua Lake 9
Matanzas Province 93 Mogote zone 95 Nicaraguan Rise 16, 152, 197, 199, 200,
Maturfn 15, 440 Moho 55 663
Maturfn Basin 437, 441,450, 547 Moho reflection 565 Nicaraguan Rise-Greater Antilles Arc
Maturfn foredeep 437 Moho topography 566, 600 93, 115, 117
Maturfn foreland basin 492, 546 Moho uplift 48 Nicely Formation 129
Maturfn sub-basin 420, 424, 448, 501, Moho-penetrating reflectors 582 Nicoya Peninsula 663
503, 544, 545, 546, 551 Mojave Sonora Megashear 145 Normal-polarity intervals 36
Maude Formation 126 Molasse-type sedimentary rocks 510 Norphlet Formation 87
Maximum flooding 459 Mona Canyon 360 North America 3, 18, 108, 126, 183,
Maximum flooding surface 419, 443, Monagas foothills 423, 424, 455, 466 423, 591

694 SUBJECT INDEX
North America plate 3, 35, 65, 189, 247, Northern Mexico 153, 155, 162 Olistostromes 284
248, 289, 291,304, 360, 362, 363, Northern province 145 Omoa 220
403, 498, 615, 627, 666 Northern Range of Trinidad 477, 488, Ontong Java 5
North America-Caribbean 43 489, 492, 495, 498, 501,503, 505, Ontong Java Plateau 584
North America-Caribbean foreland 506, 509, 510, 518, 523, 526, 531, Ophiolites 27, 52
basin 12 533, 541,553, 554 Oriente fault 633, 635, 637, 663
North America-Caribbean oblique-slip Northern Rosario 93 Oriente transform fault zone 199
plate boundary zone 287 Northern Rosario belt 93, 95, 99, 100, Orinoco 14, 591, 618, 624
North America-Caribbean plate 663 102, 103, 105, 106, 108, 112, 113, Orinoco delta 14, 425, 435,437, 441,
North America-Caribbean plate 114, 116, 117 464, 465, 503, 528, 542, 551,554,
boundary 247 Northern South America 13, 16, 63, 112, 555
North America-Caribbean plate 499 Orinoco drainage system 621,622
boundary zone 197, 279, 284, 343, Northern Taino ridge 643 Orinoco Geosyncline 421
345 Northern Tethyan Province 109, 126, Orinoco offshore 425
North America-Caribbean plate motions 137, 138, 139, 144 Orinoco Platform 419, 420, 425, 435,
43 Northern Transtensional Basin 480, 482, 437, 439, 447
North America-Caribbean strike-slip 487, 488, 493 Orinoco River sediments 542
zone 9 Northern Transtensional Basin fill 480 Orinoco shelf 448
North America-South America breakup Northern Trinidad 473, 477 Orinoco tar belt 419
7 Northern Venezuela 5, 397, 425 Orogenic grains 19
North America-South America motion Northern Venezuelan transpressional Orthogeosynclinal phase 421
18, 33 folded belts 455 Outer Apalachicola basin 87
North America-Venezuelan microplate Northern Yucat~in Straits 64 Outer arc-trench assemblage 281
motion 666 Northside Range 344, 349, 353, 354, Outer forearc-trench assemblage
North Andes block 663, 666 357 (Imbert Formation) 283
North Atlantic 19 Northwest Bahamas 190 Outer forearc-trench setting 281
North Atlantic Deep Water 622 Northwestern margin of South America Overlaps 44
North Atlantic Province 144 21
North Atlantic rifted margins 584 Northwestern Mexico 157 Pacific Ocean 586, 592, 587, 617, 618,
North Coast basin 501,506 Northwestern South America 18, 22, 24 627, 631
North Coast fault zone 478 Northwestern Venezuela 115 Pacific peninsula of Costa Rica 9
North Coast-E1 Coche fault zone 503 Nova Scotia 583 Pacific peninsula of Panama 9
North Fiji basin 391 Nuvel-lA plate 3 Pacific side of Mexico 163
North Panama deformed belt 15, 33, 50, NW Atlantic crust 565 Pacific source 147
51, 55, 57 Padre Miguel Group 157
North Soldado basin 542 Oaxaca 126 Padre Miguel volcanics 234
North Soldado sub-basin 541,542 Oaxaca area 19 Paleogene back-arc basins 6
North Venezuela margin 51 Oca 50, 618 Paleomagnetic reference frame 52
North-South America plate motions 50 Ocean Drilling Program 6 Paleozoic(?) Pre-rift rocks 64, 74, 89
North-South America relative motions Oceanic large igneous provinces Palma Picada 258
46 m Java 580 Palma Picada area 253
Northeast Providence Channel 171 J Kerguelen 580 Palma Picada Formation 258
Northeastern coast of Cuba 12 - Manihiki 580
- Palma Picada intrusions 258, 260
Northeastern Great Bahama Bank 171 - Ontong 580
-
Palma Picada intrusive rocks 249, 284
Northeastern margin of South America Oceanic plateau 5 Palma Picada porphyritic rocks 281
21 Oceanic plateau province 4 Palma Picada rocks 257
Northeastern Oaxaca 145 Oceanic plateau terrane 249 Palmarito Formation 112
Northern Andes 621 Ocoa Bay 635 Palo Alto-1 well 300
Northern Apennines 323 Ocoa Group 284 Palo Blanco Formation 140, 141
Northern Atlantic Ocean 471 ODP 587, 628 Pan de Azticar 100
Northern basin of Trinidad 495, 498, ODP cores 5 Pan de Azficar Member 100
500, 506, 507, 513, 520, 523, 526, ODP dated drill samples 27 Pan-African 69
539, 540, 553 ODP drilling 561 Pan-African crustal age province 19
Northern basin sidewall 492 ODP Leg 165 561,599 Panama 5, 241
Northern Boreal Province 126 ODP Site 1001 600, 602, 613, 615, 621 Panama Arc 4, 16, 22
Northern Canada Bonita section 259, ODP Site 642 573 Panama Arc collision 51
260, 263, 266, 282 ODP sites 599, 615 Panama Isthmus 15, 613, 623
Northern Central America 12, 16, 18 Offshore eastern Venezuela 472 Panama prism 663
Northern coast of Cuba 12 Offshore French Guyana 441 Panama seaway 24
Northern Cuba 168 Offshore northeastern Gulf 82 Panamanian block 663
Northern Grenada basin 389 Offshore Orinoco platform 443 Panamanian subduction zone 663
Northern Guarapiche province 477 Oficina Formation 424, 457 Pangea 19, 124, 144, 145, 168, 423,441
Northern Guatemala 145 Okinawa basin 397 Panuco No. 82 146
Northern Haiti 281,283 Olanchito 220, 222 Parece-Vela basin 391,397
Northern Hispaniola 279, 666 Olancho district 159 Paria Peninsula 477, 487, 488, 489, 492,
Northern Honduras 13 Old Bahama Channel 174, 175 499
Northern Italy 336 Oligocene passive margin section 435 Parral 145
Northern margin of South America 498 Oligocene-Miocene boundary 57 Parral terrane 130, 145

SUBJECT INDEX 695
Parras 133 Placetas belt 93, 111, 114-117 Puerto Rico-Hispaniola microplate 13
Partial uncertainty rotations 38 Plantain Garden-Enriquillo fault 637 Puerto Rico-Virgin Islands terrane 360
Passive margin 7, 419, 437, 438 Plataforma Deltana 420 Pull-apart basin 7, 13, 15
Passive margin of Guyana 420 Plate circuit 35 Purial area 242
Passive margin of North America 12 Plate circuit path 35 Push-ups 7
Passive margin phase 419 Plate hierarchy 52
Passive margin sequence 419, 435 Plate motions models 35 Quartzite 5
Passive margin successions 94 Plate reconstructions 3, 19 Quebrada Juana Leandra 224
Patos Island 531 Platform sequence 435 Queen Charlotte Islands 126
Patuca 153 Plume head 613 Querecual Formation 447, 448
Peak transgression 461 Point Sal 128 Quifiones 94
Pearl Islands basin 15 Pointe-a-Pierre Formation 510 Quifiones Formation 241
Pecos fault 651 Polier Formation 103, 105 Quifiones tectonic unit 106
Pecos fault zone 651,653, 659, 665 Pons 97, 106 Quintana Roo 145
Pedernales 491 Pons Formation 105, 115 Quita Coraza Formation 305, 317, 320,
Pedernales area 489, 492 Porosity 69 327, 328, 334, 335, 337
Pedernales oil field 492 Port-of-Spain 533
Pedernales region 477 Posa area 492 R/V Charcot 628
Pedernales shale ridge 489 Posa field 488 R/V Conrad 629
Pedro escarpment 663 Posa High 487, 492 R/V Ewing 561,564, 566, 593, 599
Pedro fault 663 Posa oil field 492 R/V Ewing cruise 9501 591
Pedro Garcfa 272, 273 Positive inversions 477 R/V Glomar Challenger 68
Pedro Garcfa anticline 251 Post-Kingshill limestones 351 R/V Nadir 627, 628
Pedro Garcfa Formation 249, 251,272, Post-rift sequence 63 Radiolarian assemblage 127
273 Post-rift stages 63 Radiolarian microfacies 93
Peg-leg and water column multiples 564 Post-rift unconformity 65 Rail Cabin Formation 129
Penal oil field 538 Precambrian crystalline basement 447 Ramp basin structure 338
Pefialver Formation 107, 116 Pre-rift 63, 89 Ramp or 'push-down' basins 7
Pefias Formation 106, 116 Pre-rift phase 18, 419 Ranchete 258
Peralta and Rfo Ocoa sediment groups Pre-rift rocks 63 Ranchete Member 258, 259, 260, 266,
50 Pre-rift section 73, 79 281
Peralta belt 295, 637 Pre-rift unconformity 435 Rattan Hill area 349
Peralta flysch belt 635 Presqu'~le du Sud 627, 635, 635, 637, Rattan/Belvedere 349
Peregrina Canyon 123, 139, 146 657, 660, 663, 666 Razorback Ridge 307, 311, 312, 315
Peripheral bulge 8, 435 Presqu'~le du Sud-Beata ridge 637, 660 Red cherts 27
Perisutural basins 435 Progradational forestepping 459 Red ribbon chert 245
Permian basin 291 Prograding (transitional) phase 435 Reflector A" 622
Permian Castile Formation 337 Promax TM 168 Refraction 564
Permian Delaware basin 289 Proto-Antillean Arc 6 Regional flooding events 419
Permian Salado Formation 333 Proto-Caribbean 55, 63, 83 Regional opening model 89
Peru 617 Proto-Caribbean basin 93, 108, 110, Relative motion path 18
Petroleos de Venezuela S.A. 420 111,113-117 Relative plate motion vectors 35
Petroleum Company of Trinidad and Proto-Caribbean crust 109, 168, 591, Remnant arc 13
Tobago 507, 521,524, 526 613,623 Reprocessing 67
Petroleum reserves 419 Proto-Caribbean Ocean 18 Researcher Ridge 48
Petrotrin 507 Proto-Caribbean oceanic crust 27 Restraining bend 7, 284
Phase shift angles 36 Proto-Caribbean passive margin 83 Restraining bend tectonics 284, 285
Philippine plate 414 Proto-Caribbean Sea 108, 113, 116 Ridge push 53
Philippines 130 Proto-Caribbean seaway 112 Rift 7, 8, 63
Phyllitic schist 5 Proto-Caribbean spreading ridge 21 Rift basins 8
Pica Pica Member 108 Providence Channel 167, 168, 188 Rift regime 438
Pico Bonito 223, 224 Pseudo-oceanic appearance 566 Rifted passive margin 551
Piedras Negras 224 Puebla 123, 147 Rifts of the Canal area 15
Pierre Payen anticline 637 Puerto Cortez 220 Right-steps 13
Piggyback basins 8, 14 Puerto Grande sub-basin 496, 507, 516, Rio Bajabonico fault zone 251
Pimienta Formation 141 521,523, 524, 531,553, 554 Rio Grande 38, 273
Pimienta Member 100, 101,143 Puerto Plata 239 Rio Grande fault 253
Pinalilla Formation 106, 115 Puerto Plata area 242 Rfo Grande fault zone 249, 257, 258,
Pinar 1 well 97 Puerto Plata basement complex 237, 269, 272, 273, 274, 283, 284
Pinar del Rfo 65 238, 240, 241,242, 249, 273 Rfo Jacagua measured section 271
Pinar del Rfo geology 93 Puerto Rico 6, 7, 12, 14, 22, 241,245, Rio Jacagua section 267, 269, 271
Pinar del Rio Knoll 89 247, 248, 275, 283, 354, 357, 358, Rfo Lenin graben 201, 215
Pinar del Rio Province 123, 144 360, 362, 364, 413, 591,627 Rfo Patuca 152, 153, 155, 157, 159, 162
Pinar fault 95 Puerto Rico microplate 360 Rfo Perez measured section 264
Pino Solo tectonic units 95 Puerto Rico platelet 362 Rio Perez section 264, 266, 267, 282
Pinos terrane 94, 109, 112 Puerto Rico platform 343, 360 Rfo Perez syncline section 260, 282
Pirital thrust fault 503 Puerto Rico terrane 360 Rio San Juan 238, 243
Pitman Fracture Zone 51 Puerto Rico trench 12, 199, 248, 663 Rfo San Juan area 272

Rio San Juan complex 237, 242, 243, San Marcos unit 239 Seismic velocities 4
249, 272, 273 San Pedro basin 635, 637 Septentrional block 635
Rfo San Juan mudstone 245 San Pedro del Gallo 128, 130, 131,133, Septentrional fault zone 248, 251,257,
Rfo San Juan-Puerto Plata-Pedro 136, 137, 138, 146 267, 269, 272
Garcfa disrupted terrane 273 San Pedro del Gallo area 127 Sepur clastics 115
Rfo Sutawala 155, 159, 162 San Pedro del Gallo Fault 145 Sepur foreland basin 12, 21
Rfo Viejo 201 San Pedro del Gallo remnant 133, 137, Sepur Formation 116
Rfo Viejo fault 211,223 138, 139, 145, 147 Sequence boundaries 443
Rfo Wampfi 152, 153 San Pedro del Gallo terrane 109, 111, Sequence boundary SB- 1 441
Rfo Wampti area 155 123, 124, 129, 130, 133, 141, Sequence boundary SB-2 441,443
Rio Wampfi- Rfo Patuca area 157 144-147 Serranfa del Interior 419, 420, 423, 425,
Rfo Wampfi-Montafias de Col6n 153 San Pedro units 136 437, 439, 442, 447, 448, 450, 455,
Rfo Yaque del Sur 287, 292, 300, 317, San Pedro Zacapa 159 465, 466, 472, 473, 477, 487, 494,
327, 340 San Salvador 168 496, 501,503, 547, 548, 555
Rfo Yaque section 333 San Vicente Member 87, 101,102, 113, Serranfa del Interior belt 505, 546, 551,
Rio Yaroa 273, 274 143, 144, 147 552
RoaUin Island 197, 201,203, 205, 208, Sanarate limestone 213 Shale diapirs 14
210, 213, 214, 215, 222 Sandino forearc basin 9 Shallow subduction 9
Roble Member 103 Santa Ana field 450 Shallow water-deep basin 338
Rogue Formation 128 Santa Barbara County 128 Shelf-break 463
Romeral Suture 617 Santa B~irbara quadrangle 159 Shelf-margin wedges 333
Rosario belts 96, 98, 105, 108, 113, Santa Elena fault-Hess escarpment 152 Shell 629
115-118 Santa Marta Massif 621 Shikoku basin 391,394, 397
Rosario North 94 Santa Marta-Bucaramanga fault 15 Sicily 330, 333, 334, 336, 337
Rosario South 94 Santa Rita 220 Sico River 220, 227, 231,232, 234
Rotation parameters 42 Santa Rosa group 80 Sideswipe 565
Rotation uncertainties 35, 40 Santa Teresa 241 Sierra Bahoruco 615
Rough B t~ basement 600 Santa Teresa Formation 105, 106, 114 Sierra Bermeja 245
Rough B" crust 565 Santa-Marta Massif 662 Sierra Cadnelaria 130, 146
Rough-smooth B" basement boundary Santana 321,325 Sierra Chiquita tectonic unit 116
600, 624 Santana-E1 Granado road 323 Sierra de Bahoruco 295, 296, 300, 302,
Rough-smooth B" transition 565 Santander Massif 621 309, 635, 637, 659
Rough-smooth boundary 12 Santaren Channel 188, 189 Sierra de Catorce 123, 130, 135, 138,
Royal Trough 48 Santiago 239, 257, 267 139, 145, 146
Santiago Formation 140, 143 Sierra de Catorce remnant 138, 139, 147
Saba 220, 354 Santiago-Altamira highway 253 Sierra de Escambray 93
Saba Bank 357, 358, 363, 364, 397, 398, Santiago-Puerto Plata highway 258, Sierra de la Caja 123, 130, 135, 138, 146
413 259, 267, 271,282 Sierra de los Organos 83, 87, 88, 95, 97,
Sag basins 78 Santo Domingo 296, 312 103, 105, 107, 109, 113, 123, 141,
Saline giant 287, 289, 291 Santonian time 5 143, 144, 147
Salt River Valley 350 Sarasota 87 Sierra de los Organos area 98
Samana Peninsula 249, 284 Sarasota arch 88 Sierra de los Organos belt 93, 95,
Sambfi basin 15 Satellite altimetry 33, 34, 55 97-102, 105-108, 111-117
Sambfi fault 15 Scotland 313 Sierra de los Organos remnants 111
San Andres 16 Sea of Japan basin 391,392, 397 Sierra de los Organos succession 112
San Andres Limestone 143, 144, 147 Seabeam map 627, 651 Sierra de Martin Garcia 635
San Andres trough 16 Seabeam survey 628 Sierra de Neiba 295, 305, 317, 635, 637
San Antonio Formation 448 Seacarib 1 651 Sierra de Omoa 220
San Bias forearc basin 15 Seacarib 1 cruise 627, 629 Sierra de Parras 135
San Cayetano basin 93, 99, 111, 117 Seacarib cruise 637 Sierra de Perija 621
San Cayetano deltaic sediments 117 Seacarib profiles 637 Sierra de Zuloaga 123, 138
San Cayetano Formation 83, 87, 98, 99, Seafloor bathymetry 8 Sierra del Abra 146
100, 109, 110, 111,112, 141, 143 Sea-level fall 463 Sierra del Rosario 87, 88, 95, 100, 103,
San Cristobal basin 635 Seasat altimetry data 35, 36 109, 110, 123, 141,143, 144, 147
San Esteban 219, 220 Seaward-dipping reflectors 577, 583, Sierra del Rosario belt 111
San Fernando Bay 488 585, 600, 624 Sierra del Rosario meridional 83
San Fernando High 488 Sebastopol Complex 477 Sierra del Rosario remnants 111
San Francisco 499 Secondary basal foredeep unconformity Sierra del Rosario septentrional 83
San Francisco fault zone 501 463 Sierra del Rosario succession 111
San Francisco-Quiriquire faults 493 Second-order sequence boundaries 419 Sierra Jimulco 135, 146
San Jose member 519 Second-order T/R packages 472 Sierra la Gloria 133
San Juan basin 295, 317, 635 Second-order transgressive-regressive Sierra Madre Oriental 123, 133, 145,
San Juan Formation 448 cycle wedge 461 146, 147
San Juan graben 477, 492 Second-order transgressive-regressive Sierra Madre Oriental terrane 123, 130,
San Juan-Azua basin 249, 291 cycles 419 145
San Juancito 155 Sedimentary accretionary wedges 8 Sierra Martfn Garcfa 295, 304, 305, 317
San Luis Potosi 123, 137, 147 Sedimentary basins 3 Sierra Martfn Garcfa anticline 300
San Marcos Formation 237, 238, 242 Seed points 41 Sierra Nevada 141

SUBJECT INDEX 697
Sierra Nombre de Dios 220 Southern Boreal paleolatitudes 137 Sumidero Member 88, 102, 103, 143
Sierra Ramirez 130, 146 Southern Boreal Province 109, 123, 126, Superior Oil Company 296, 297, 304
Sierra Santa Rosa 135, 137, 138, 147 127, 137, 139, 144 Superterrane 111
Sierra Sombreretillo 133 Southern Boreal/Northern Tethyan Sutawala Valley 153, 162
Sierra Sombretillo 130, 146 faunas 103, 109, 147 Swan Islands 197, 201,211,219, 222,
Sierra Zuloaga 130, 146 Southern Canada Bonita section 259, 223
Siete Cabezas basalt, 237, 243 260, 263, 266, 267, 282 Swan Islands fault 197, 201
Siete Cabezas Formation 240 Southern Central America 24 Swan Islands fault zone 197, 199, 200,
Siliciclastic wedge 495 Southern Compressional Zone 480, 482, 208, 214, 215, 216
Sinu belt 662 487, 488, 489, 493 Sylvie ridge 660
Sinu subduction zone 627, 663 Southern Cuba 281,283 Symmetric plate accretion 41
Sinu trench 662, 663 Southern Great Bahama Bank 171, 182, Symon 123, 130, 146
Ski-jumps 566, 583 190 Symon remnant 145
Slab pull 53 Southern Haiti 576 Syn-rift phase 419
Slow-spreading ridges 565 Southern Hispaniola 5, 52 Syn-rift sequence 63,435
Smackover carbonates 63 Southern Mexico 19, 22, 498 Syn-rift stage 65, 93, 117
Smackover Formation 87 Southern Middle America trench 9
Small- and medium-offset fracture zones Southern Peru 617 Tabasco 145
37 Southern platform 83 Tacutu Graben 441,473
Smith River subterrane 129 Southern province 145 Tacutu rift system 441
Smooth BI~basement 600 Southern Range of Trinidad 477, 478, Taino ridge 627, 629, 641,642, 643,
Smoothness 41 488, 489, 491,493, 496, 506, 536, 649, 653, 662, 665
Snowshoe Formation 128, 141 548, 555 Tairona ridge 629, 653,659
Soldado High 488 Southern Rosario belt 95, 96, 98, 99, Talanga 153
Solim6es basin 618, 621,622, 624 102, 103, 105-108, 113, 114, 117 Talparo Formation 488, 511,513, 524,
Sombrerito Formation 295, 298, 301, Southern Senegal 441 526
304 Southern Taino ridge 643 Taman Formation 140, 141, 146
Sonic velocity 69 Southern Trinidad 425,437 Taman, S.L.E 140
Sonobuoys 564 Southern United States 153 Tamana Formation 510
Soroa 96 Southern Yucatan margin 115 Tamana limestone 540
South America 3, 13, 18, 35, 93, 108, Southwest Pacific 415 Taman-Tamazunchale 137
111, 112, 423, 546, 586, 591,593, Southwestern Oregon 128 Tamaulipas 145
623 Southwestern Panama 15 Tamaulipas Formation 146
South America plate 3, 35, 391,403, Southwestern Puerto Rico 128 Tamayo 331
423, 425,435,450, 463, 471,477, Spears of iron 336 Tamayo-Vuelta Grande road 327, 328,
489, 495-498, 615, 617, 627, 663, Spreading ridge 13 330
666 Springvale Formation 482, 487, 488, Tampa embayment 83, 87, 88
South America plate boundary 624 492, 511,513, 523 Tampico 146
South America-Caribbean 43 St. Croix 13,343, 345, 347, 353, 354, Tampico Embayment 146
South America-Caribbean plate 357, 358, 360, 362, 363, 364 Tampico-Ciudad Valles line 145
boundary zone 499 St. Croix basement rocks 357 Taninul quarry 161
South American Craton 617, 622 St. Croix basin 357, 362 Taraises Formation 136
South American deformed belt 630, 650, St. Croix Ridge 357, 360 Taulab6 area 162
651,659 St. John 354, 358 Tavera Group 284
South American passive margin 539, St. John' s/Judith Fancy area 349 Tectonic evolution 16
555 St. Thomas 357, 358 Tectono-paleogeographic maps 82
South American platform 398 Stage poles 42 Tectono-stratigraphic megasequence 441
South American Precambrian craton 472 Stage vectors 44 Tectonostratigraphic terrane 144
South Atlantic 36, 289 Stanley Mountain 127, 128 Tegucigalpa 153, 155
South Australia 337 Stanley Mountain cherts 129 Tela 13, 219, 220
South Caribbean margin fault 16 Statistical parameters 40 Tela Basin 197, 201, 213-216, 219, 234
South Caribbean marginal fault 15, 16 Step 7 Telemaque member 519
South Caribbean Plate boundary 477 Stoneyford 128 Temblador Group (well 'H') 447
South Coast fault 478 Straits of Andros 188 Tepehuafio 144
South Fiji basin 394, 415 Straits of Florida 65, 74, 168, 188, 190 Tepehuafio terrane 145
South Florida platform 88 Stratigraphic and metamorphic terranes Tepexic Limestone 140
South Fork Member 129 94 Terranes 94
South Martfn Garcfa fault zone 300 Stress inversion technique 208 Test hole M 10 354
South Pacific 56 Strike-slip basins 6, 7 Tethyan ammonites 137, 139
South Sandwich 391 Stuart City Formation 162 Tethyan Realm 123, 126
South Sandwich basin 394 Submarine canyons 462, 463 Tethys 161
Southeast Indian Ocean 583 Subsatellite basement topography Texaco, Inc 168
Southeastern Gulf of Mexico 21, 63, 73 profiles 37 Texas 159, 162, 241
Southeastern Trinidad offshore 425 Subsidence mechanisms 6 Texas megashear 124
Southern Basin of Trinidad 448, 489, Sula graben 201 Thermal subsidence 89
496, 506, 538, 539, 540, 542, 544, Sula Islands 111,126 Third-order cycles 419
551,555 Sula Valley 220, 223 Third-order sequences 464
Southern Basin sidewall 492 Sulu basin 397, 414 Third-order unconformity 435

Three-plate system 18 Urica 499 Volcanic plateau 628

Tiburon Ridge 48, 49, 55 Urica fault 493, 503 VCring Plateau 583
Tiburon Rise 49, 55 Urica fault system 420 Vcring volcanic margin 573
Tigre Formation 447 Urica fault zone 501 Vuelta Grande 331
Time migration 68 Utila 197, 201,223, 224
Tirisne Cliffs 153, 155, 157, 159, 162 Utila Island 201, 215 Walkers Cay 190
Tobago 14 Walper Megashear 109, 111, 123, 124,
Tobago trough 14 Valanginian Tumbitas Member 143 130, 131, 133, 144-147
Todos Santos Formation 155 Valle de Angeles 159 Walton-Plantain Garden fault 663
Toe thrusts 437 Valle de Angeles Group 155, 156, 157, Walton-Plantain Garden-Enriquillo
Tonga Trench 415 159, 161,162, 163, 223 fault 663
Tongue of the Ocean 167, 171, 174, 185, Valle de Angeles redbeds 156 Wampti-Patuca region 161
191 Valle de Pons tectonic unit 97, 106 Wampusirpi quadrangle 159, 162
Tonosf basin 15 Valle Formation 126 Warao rise 653
Top basement unconformity 435, 438 Vector diagram 18 Warm Springs fault 477, 480, 482, 487,
Top Berriasian 69 Vectorial closure condition 18 488, 489, 492, 496, 552
Top evaporites/Angostura reflector 301 Vema 629 Warm Springs fault system 488, 492
Top Sombrerito reflector 301 Vena del Gesso basin 323 Warm Springs fault zone 496, 498, 501,
Trans-Atlantic Geotraverse 36 Venezuela 4, 6, 50, 241,419, 472, 477, 506, 507, 521,524, 539, 541,542,
Transfer faults 8 498, 621,622 555
Transform-ridge intersections 39 Venezuelan abyssal plain 622 Warm Springs Member of the Snowshoe
Transgressive-regressive cycles 443, Venezuelan Andes 621 Formation 129
472 Venezuelan Atlantic offshore 473 Warm Springs-Central Range 499
Transitional crust 74 Venezuelan Basin 5, 13, 16, 28, 33, 50, Warm Springs-Central Range fault zone
Transpression 41, 50, 551 51, 56, 241,398, 561,564, 565, 573, 495, 501,503, 533, 554
Transpressional folded belt 472 576, 577, 580, 581,582, 584, 586, Warm Springs-Central Range-Caigual
Transpressional foredeep 420 587, 591,592, 593, 599, 602, 611, fault zone 496, 542, 540, 555
Transtension 41 613, 614, 615, 617, 618, 622, 623, Washita Group 159
Transtensional basins 477 624, 649, 650, 651,657, 659 Well M 1 350, 356
Trench suction 53 Venezuelan Basin rough crust 565
Well M2 350
Trinchera Formation 295, 296, 300, 304, Venezuelan Caribbean Mountains 618,
Well M4 352, 356
305, 317, 334, 336 622
Well M5 356
Trinidad 6, 14, 419, 421,423,425, 435, Venezuelan Coastal Ranges 424
Well M10 349, 350
439, 447, 465, 472, 473, 487, 488, Venezuelan Cordillera 477
West Fiji basin 397
492, 495,497, 498, 503 Venezuelan crust 624
West Florida 446
Trinidad and Tobago 533 Venezuelan deformed belt 627, 663
West Indies 49
Trinidad area 24 Venezuelan fold-and-thrust belt 482
West Texas 289, 333
Trinidad belt 505 Venezuelan foredeep 471, 618
West YucaUin basin 13
Trinidad-Venezuela boundary 507 Venezuelan microplate 51, 52, 627, 662,
Trujillo 219, 220, 223, 227, 234 West-central Honduras 151
663, 666
Tucutunemo 15ormation 112 West-central Mexico 130
Venezuelan offshore 465, 448
Tulip-type cross-sectional fault Venezuelan oil industry 6 Western Atlantic 623
structures 185 Venezuelan Plate 33, 56, 57, 592, 615, Western Brazil basin 241
Tumbadero Member 88, 101,102, 143 637 Western Canada 289
Tumbitas Member 88, 101,102 Venezuelan shelf 15 Western Chortfs block 8
Venezuelan-Colombian microplate 665 Western coast of Hispaniola 637
U.S. 161 Veracruz 123, 145, 145, 147 Western Cordillera 51, 56
U.S. Gulf Coast 155, 163 Vicente Noble block 300 Western Cuba 6, 12, 22, 52, 83, 87, 88,
U.S. Virgin Islands 343 Victoria segment 145 93, 94, 130, 138, 141,144, 147, 552
Uncertainties 35 Vidofio Formation 448 Western Mediterranean 289
Uncertainty ellipses 56 Vieja Member 108 Western Pacific 575
Unconformity surfaces 63 Vieques 357, 358, 360 Western Pangea 63, 83
Unit II 443 Villa de Cura nappe 505 Western Straits of Florida 64
Unit III 443 Villa La Reine 349, 350, 354 Western Venezuela 22
Unit IV 443 Villa La Reine type section 351 Wichita megashear 124
United Nations 219 Villa Trina Formation 247, 251,257, Wide-angle reflection data 564
University of Texas at Austin 629 272, 279, 282-285 Wilbur Springs 128
University of Texas at Austin Institute Villa Vasquez series 253 Wild Cane complex 244
for Geophysics 392 Vifiales Limestone 143 Wood River Formation 88
University of Texas seismic data base 65 Virgin Islands 6, 14 Wrangellia terrane 124
Upper Atima Formation 162 Virgin Islands basin 357, 360, 362, 363, Wrench models 477
Upper Cruse Formation 538, 555 364 Wyoming 313
Upper La Pica Formation 487 Virgin Islands basin/Anegada Passage
Upper Merecure Formation 466 area 360 Yojoa Group 151,153, 155, 157
Upper Nicaragua block 666 Virgin Islands platform 354, 360, 362, Yoro 220
Upper Nicaragua rise 666 363, 364 Yucat~in 63, 87, 89, 93, 108, 145
Upper Talparo Formation 526 Virgin Islands Trough 343 YucaUin basin 7, 13, 22, 28, 52, 95, 392,
Upper Valle de Angeles Group 151, 155 Vizcaino Peninsula 126 499

SUBJECT I N D E X 699
Yucatan block 18, 19, 63, 65, 80, 83, Yucatan platform 89, 93, 109, 109, 144, Zarza Member 143, 143
109, 111,114 145, 147 Zaza terrane 11 l, 115
Yucatan block edge 96 Yucatan terrace 69, 79, 80, 81, 89 Zaza volcanic arc 115
Yucatan borderland 95, 109, 111 Yucat~in-Florida Straits 114 Zero edge 465
Yucatan margin 111, 114-118 Zuloaga Limestone 133, 137, 138
Yucatan Peninsula 12, 13, 22, 52, 144, Zacar~as Member 100
199, 546 Zacatecas area 145

Caribbean Basins. Volume 4. Sedimentary Basins of The World

Uploaded by

Copyright:

Available Formats

You might also like

Caribbean Basins. Volume 4. Sedimentary Basins of The World

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Caribbean Basins. Volume 4. Sedimentary Basins of The World

Uploaded by

Copyright:

Available Formats

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

SEDIMENTARY BASINS OF THE WORLD

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Zurich, Switzerland, January, 1999 KENNETH J. HSU

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

W.A. van den Bold

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

SELECTED PUBLICATION LIST W.A. VAN DEN BOLD

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Akad. Wetensch. Proc., ser. B., 64 (5), 627-639.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

ontology 18 (4), 410-442.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

1http://www.elsevier.nl/locate/caribas/with mirror sites: http://www.elsevier.com/locate/caribas/and

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Lewis Abrams Frederick Hutson

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

W. ALl 17 J.E CASEY 15

G. AVI~ LALLEMANT 8 Y. DENIAUD 14

* Superior ciphers refer to the chapter number.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

J-C. FAUGERES 14 G. HERNANDEZ a7

R.C. FINCH 6 K.J. HSO 22

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Caribbean Sedimentary Basins" Classification and Tectonic

INTRODUCTION Plate rates

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Fig. 3 summarizes the nomenclature I use in this Island-arc basins

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

north from rough Cocos lithosphere created at the

BASINS AND MAJOR TECTONIC FEATURES

North America-Caribbean foreland basin and

A semi-continuous foreland basin recording the

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Yucat~in back-arc basin Muertos trough and 'forearc' basin

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Riffs of the Canal area

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.

Colombian accretionary complex and forearc Beata Ridge

PROPERTY INTERNATIONAL GEOPHYSICAL TECHNOLOGY,S.L.