Marine and Petroleum Geology 19 (2002) 1047–1071

www.elsevier.com/locate/marpetgeo

The role of depositional setting and diagenesis on the reservoir

quality of Devonian sandstones from the Solimões Basin,
Brazilian Amazonia
Rodrigo Dias Lima1, Luiz Fernando De Ros*
Instituto de Geociências, Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves, 9500, 91501-970 Porto Alegre, RS, Brazil
Received 31 July 2002; received in revised form 3 December 2002; accepted 7 December 2002

Abstract
Devonian sandstones of the Uerê Formation are important oil exploration targets in the Solimões Basin, western Brazilian Amazonia. The
basin fill comprises Ordovician to Permian sedimentary successions, Triassic diabase dykes and sills, and Cretaceous to Recent continental
deposits. This study deals with the Upper Devonian interval, which consists of sharp-based, progradational sandstones, attributed to a storm-
dominated shelf complex formed during an overall transgressive system tract, overlain by Frasnian-Famennian black shales. In spite of their
large lateral extent, the exploration of these sandstones is complicated by intense diagenesis, which strongly affected reservoir quality. The
main processes of porosity reduction are mechanical and chemical compaction and cementation by quartz overgrowths, carbonates (siderite
and dolomite) and fibrous illite. Porosity (up to 28%) was preserved by the inhibition of quartz overgrowth cementation and pressure
dissolution by grain-rimming, eogenetic, microcrystalline quartz and associated chalcedony, or smectite. Early diagenetic silica precipitation
is related to the dissolution of sponge spicules, which were concentrated in storm-reworked hybrid arenites and in interbedded spiculite
deposits. Locally, massive quartz cementation and recrystallisation occurred as a consequence of hot fluids circulation related to Triassic
magmatism.
q 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Stratigraphy; Sandstone diagenesis; Reservoir quality

1. Introduction Although the vast Solimões Basin has been explored

During 70s and 80s, diagenetic studies on clastic
reservoirs quality concentrated in the mechanisms of
secondary porosity generation (Franks & Forester, 1984;
Giles & Marshall, 1986; Schmidt & McDonald, 1979;
Surdam, Boese, & Crossey, 1984). On the other hand, last
decade research dealt mostly with the diagenetic mechan-
isms of porosity preservation (Aase, Bjørkum, & Nadeau,
1996; Bloch, Lander, & Bonell, 2002; Jahren & Ramm,
2000; Pittman, Larese, & Heald, 1992). The objective of this
study is to unravel the controls on the observed preservation
of porosity and variation of quality in the Devonian
sandstone reservoirs of the Solimões Basin (western
Brazilian Amazonia).
* Corresponding author. Fax: þ 55-51-3316-7047.
E-mail address: lfderos@inf.ufrgs.brs (L.F. De Ros).
1
National Petroleum Agency Grantee; present address: PETROBRAS
Corporate University, Rio de Janeiro, RJ, Brazil.
0264-8172/02/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0264-8172(03)00002-3
throughout the past three decades, little is known about the
reservoir quality controls of the Devonian sandstones of the
Uerê Formation. Most of the exploration efforts were
concentrated in the Carboniferous sandstones of the Juruá
Formation, which contain the largest gas accumulations of
Brazil (close to 200 £ 109 m3 of gas in place, and
11 £ 106 m3 of associated oil in place). The remote
geographic location of the Solimões Basin and the
extremely high costs of gas production and transportation
in the Amazonian forest has recently increased the interest
in the exploration for the Devonian sandstones, because
they contain mainly oil, occur throughout most of the basin,
and are closely associated to the major basin source rocks,
the Jandiatuba shales (Mello, Koutsoukos, Mohriak, &
Bacoccoli, 1994). However, exploration of the Uerê
sandstones is complicated by the heterogeneous quality of
these reservoirs, which range from highly porous (up to
28%) to extremely tight.
1048 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Therefore, this study aims to understand the strati-
graphic, depositional and petrologic controls on the quality
of the Uerê reservoirs. The data, parameters and patterns
generated by this research will be applied in the construction
of predictive models to be used in the systematic exploration
of the Devonian reservoirs.
2. Geological setting
The Solimões Basin is a large cratonic sag that covers
around 600 000 km2 of northwestern of Brazil, containing
up to 4.5 km of marine to continental Palaeozoic deposits
covered by Cretaceous to Tertiary continental deposits.
Eohercynian extension caused normal faulting and uplift of
the source areas of Brazilian Shield to the south, the
Guyanas Shield to the north, and the Carauari Arch, which
divided the Solimões Basin into the Jandiatuba and Juruá
sub-basins. The Carauari Arch exerted an important control
over the pre-Pennsylvanian sedimentation. Late Triassic to
Early Jurassic diabase sills and dykes were intruded in the
Devonian, Carboniferous and Permian sequences around
210 and 220 Ma ago, according to Ar/Ar dating (Szatmari,
1996). The thermal pulse related to this basic magmatism
played an important role in hydrocarbon generation,
expulsion and migration from the Devonian source
rocks to Devonian and Carboniferous reservoirs (Caputo
& Silva, 1990). The extensional period was followed by
Fig. 1. Location map of the Solimões Basin, with inset showing the São Mateus Field. Sampled wells shown as black dots. Wells with only log data in white.
Jurassic –Cretaceous transpressional tectonism, related to
the Andean orogenic events, which gave the area its present
configuration (Caputo & Silva, 1990; Fig. 1). Large
compressional structures, in the form of reverse faults and
asymmetrical folds constitute the main hydrocarbon traps.
The basin was filled by six second-order depositional
sequences: Ordovician, Silurian-Devonian, Devonian-Car-
boniferous, Carboniferous-Permian, Cretaceous and Ter-
tiary (Eiras et al., 1994; Fig. 2). The Devonian-Carboniferous
Sequence represents a transgressive-regressive cycle com-
posed of four successions, separated by regionally extensive
sequence boundaries, identified by miospore biostratigraphy
(Grahn, 1992; Loboziak, Melo, Quadros, Daemon, &
Barrilari, 1994; Silva, 1987; Fig. 2). The most basal
succession is comprised of prograding Lower Devonian
deposits (Fig. 2) that occur only in the deeper Jandiatuba
Sub-Basin and western portion of Juruá Sub-Basin. Over a
possibly transgressive surface (S2; Fig. 2), retro- to
aggradational, Eifelian-Early Givetian deposits occur at the
depocentre of Juruá Sub-Basin and eastern Jandiatuba Sub-
Basin. Early Givetian (?)-Frasnian sandstones and cherts
(informally herein named Uerê sandstone) lay uncomform-
ably on Middle Devonian deposits and Precambrian
metasediments and granitoids. The Uerê sandstones rep-
resent a progradational to aggradational succession limited at
the base by a regional erosional surface (S3; Fig. 2).
Radioactive Frasnian Jandiatuba shales regionally cover
the Uerê sandstones along the Devonian maximum flooding
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071 1049

Fig. 2. Simplified stratigraphic chart of the Solimões Basin (modified from Eiras et al., 1994) and detailed Devonian Sequence stratigraphy within the São
Mateus Field area (well NSM-1). Lithostratigraphic nomenclature informally adopted herein. LST, lowstand system tract; TST, transgressive system tract;
HST, highstand system tract. Two second-order sequence discontinuities (S1 and S5) bound the Devonian Sequence. S3 is a regionally correlatable surface,
corresponding to a transgressive erosion below the Uerê sandstones.

surface (MFS; Fig. 2), constituting the main source rock of
the Basin (TOC av. 1.0%, up to 8.25%; Eiras, 1998; Grahn,
1992; Mello et al., 1994; Silva, 1987). Towards the eastern
margin of the basin, the Jandiatuba shales are downlapped
by a 50 m thick, progradational succession of Famennian
sandstones rich in siliciceous sponge spicules interbedded
with shales and minor conglomerates of mud and chert
intraclasts reworked from the underlying Late Devonian
deposits, informally named Jandiatuba sandstones (Fig. 2).
A marked erosive boundary (S4; Fig. 2) cuts the
Frasnian-Famennian shales and sandstones, and is covered
by latest Famennian-Tournaisian glacial diamictites
(Caplan & Bustin, 1999; Caputo & Crowell, 1985; Johnson,
Klapper, & Sandberg, 1985; Streel, Caputo, Loboziak, &
Melo, 2000), corresponding to a worldwide glacio-eustatic
regression known as the Hangenberg Event (Caplan &
Bustin, 1999; Streel et al., 2000). A regional erosive surface
(S5; Figs. 2 and 3(A)–(C)) separates the Carboniferous-
Devonian Sequence from thick (, 800 –1200 m) marginal
marine evaporites, minor carbonate and siliciclastic deposits
of the Carboniferous-Permian Sequence (Eiras et al., 1994).
3. Depositional facies of the Uerê sandstones
The Uerê reservoir sandstones extend throughout the
Devonian southern palaeomargin of the Solimões Basin
(Figs. 1 and 3(A) – (C)). The entire succession is 20 –60 m
thick, and is characterised by sharp-based, aggradationally to
progradationally stacked sandstone deposits. The presence
1050 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071

of a basal erosional discontinuity (S3 surface) is suggested overlying Jandiatuba shales. Three facies associations
by (1) truncation of underlying Lower to Middle Devonian were identified in the Uerê sandstones, which are arranged
deposits and (2) lateral continuity of the individual facies in a coarsening-upward trend from the muddy facies
contained between the basal discontinuity surface and association 3 to sandy facies association 1. The following

Fig. 3. Well-log (gamma ray) cross-section of the Devonian Sequence through the São Mateus Field (A) and (B), and southern Solimões Basin (C). Correlation
surfaces are defined for the São Mateus Field area, with approximate correspondence to the regional stratigraphic boundaries in Fig. 2: SB1, to S1; TES, to S3;
SB2, to S4, and SB3, to S5. Datum is set at the MFS, which corresponds to the Jandiatuba shales. See Fig. 1 for location of wells and text for further
explanation.
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Fig. 3 (continued )
interpretations are based on comparisons with facies and
processes of modern shelves (Snedden & Nummedal,
1991).
Facies association 1 is composed of amalgamated, sharp-
based sets of massive, parallel to low-angle cross-stratified,
fine to coarse sandstones (Fig. 4(A)). The individual sets are
0.05 –0.25 m thick, some with thin basal conglomeratic
lags. Bioturbation is rare in this facies association,
consisting mostly of isolated Skolithos and escape burrows.
The setting of facies association 1 is interpreted as
amalgamated sands that were deposited in shallow (above
fair-weather wave base) marine shelf environment during
intermittent storms.
Facies association 2 is composed of sharp-based,
thinning-upwarding sandstones interbedded with chert and
shale layers. Sandstones are fine- to medium-grained, and
massive or parallel- to low-angle (swaley?) cross-stratified
(Fig. 4(E)). Coarse, sharp-based sandstone beds include
imbricated mud intraclasts (Fig. 4(B)). The thickness of
individual beds varies from 0.1 to 1.5 m. The sandstones are
locally bioturbated and soft-sediment-deformed (Fig. 4(D)).
Chert layers are 0.05 –0.2 m thick, and contain unbroken
monoaxon and rare triaxon, randomly oriented sponge
spicules that have been cemented and replaced by
chalcedony and microquartz (Fig. 6(A)). Centimetre-thick
siderite-cemented intervals occur in sandstones that are
interbedded with shales. Facies association 2 is interpreted
to represent sands deposited in middle to lowermost shelf
1051
environment during storms with high-energy, long period
waves. Sponge spicule layers were derived from the
disintegration of individual or small colonies of soft-bodied
demosponges. Random spicule orientation and lack of
breakage suggest limited transport. Bioturbation on the top
of storm sand layers and escape structures within laminated
beds were produced by organisms that colonised the storm-
deposited sands during fair weather conditions (MacEa-
chern & Pemberton, 1992).
Facies association 3 is characterised by laminated black
shale interbedded with 0.1 –0.5 m thick, burrowed, massive-
or parallel-laminated, fine to very-fine sandstones, and
0.05 – 0.15 m thick chert layers (Fig. 4(C)). The sandstone
beds display sharp bases covered by imbricated mud
intraclasts (Fig. 4(B)). Facies associations 2 and 3 display
a low-diversity assemblage of trace fossils, dominated by
small (, 10 mm diameter) Planolites, Teichchinus, Sko-
lithos, and locally robust Thalassinoides and Cylindrichnus
(Fig. 4(D)). Escape burrows are also commonly observed.
The high mud content and intense bioturbation of
facies association 3 indicate deposition under deeper
water, in which low-energy basinal conditions prevailed.
The deposition of the thin intercalated sandstone beds is
interpreted to have occurred below storm wave base, by
episodic influx of shoreface sediments.
The entire succession reflects the progradation of storm-
influenced shelf deposits, based on its coarsening-upward
character, the presence of thin, non-amalgamated storm beds,
1052 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071

Fig. 4. Core photographs illustrating the main facies recognised in the Uerê sandstones. (A) Facies Association 1: tabular cross-stratified to massive sandstones.
Note mottled aspect due to carbonate cementation. (B) Facies Association 2: sharp-based, parallel- to low-angle inclined cross-stratified sandstones and sponge
spiculites (SS). Note mud intraclast lag. (C) Facies Association 3: sponge spiculites interbedded with shales. (D) Strongly bioturbated hybrid arenites from
Facies Association 2, with a Teichchinus (Te) trace within a Cruziana ichnofabric. (E) Stacking pattern of Facies Association 2, displaying an overall
coarsening-upward trend composed of sharp-based (arrows), thinning-upward, cm-thick cycles with low-angle cross-stratified, bioturbated to massive
sandstones and shales.

slightly reworked soft-bodied sponge remains and glaucony
ooids and the widespread proximal Cruziana assemblage.
4. Stratigraphic framework
The Devonian section was deposited in a very flat,
shallow and wide intracratonic depression, without a
distinct shelf/slope break. Therefore, the extremely low
relief of the basin implies that a change of a few meters in
relative sea level would cause the flooding of large areas.
There is a great lateral continuity (< 200 km) of the
facies associations contained between the discontinuity
surface S3 and the Jandiatuba shales (MFS; Fig. 3(A) –(C)).
In their distal (NW) edge, there is a dominance of facies
associations 2 and 3.
The general progradational pattern of the Uerê deposits
was overlain by a MFS, represented by the transgressive
Jandiatuba shales (Silva, 1987). Two mechanisms could be
invoked to explain the deposition of these organic-rich
shales: (1) encroachment of the epicontinental sea by an
oxygen-minimum zone during late Famennian flooding
(Savoy, 1992); (2) upwelling (Caplan & Bustin, 1996;
Savoy, 1992). High rates of primary productivity at the sea
floor are evidenced by the abundance of sponges, a common
constituent of siliceous deposits related to ancient upwelling
cells (Caplan & Bustin, 1999; Parrish, 1982; Savoy, 1992).
The possible stratigraphic settings for the progradation of
storm-influenced shelf deposits underlain by erosional
surfaces are those of forced-regressive (falling relative sea-
level stage) and lowstand complexes. Both are allocyclic,
overlying genetically unrelated deposits and basally bounded
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
by sequence-stratigraphically important discontinuities
(Hunt & Tucker, 1992; Posamentier, 2002; Posamentier,
Allen, James, & Tesson, 1992; Walker & Bergman, 1993;
Walker & Wiseman, 1995). The evidence to reject a forced-
regressive scenario for the Uerê sandstones is that, in such
settings, the submarine component underlying the shallow
marine deposits extends only as far seaward as storm wave
base reach, extending below that only as non-erosional
correlative conformities (MacEachern, Zaitlin, & Pember-
ton, 1999). Conversely, the base of Uerê sandstones is
markedly erosional throughout the basin. Additionally, it
must be noted that in the distal areas (e.g. western Juruá Sub-
Basin) Uerê sandstones basal boundaries (S3) remains
erosional even where facies association 3 was deposited
below fair-weather wave base. In the trangressively incised
shoreface scenario the basal discontinuity is cut during
relative sea-level stillstand prior to progradation, being after
modified by wave ravinement during transgression (Down-
ing & Walker, 1988). The resulting basal surface is erosive
and correlatable basinward, as observed in Uerê sandstones
(MacEachern et al., 1999). Glaucony ooids that occur
dispersed in the Uerê sandstones (Fig. 6(B)) are also
consistent with a transgressive scenario.
5. Sampling and methods
In this study we examined 50 m of cores taken from eight
oil wells. Thin sections prepared from 90 blue epoxy-
impregnated samples were examined with a petrographic
microscope. The amounts of detrital and diagenetic com-
ponents, and pore types, as well as the textural modal grain
size and sorting parameters, were determined by counting 300
points in each of 65 selected representative thin sections.
Sorting was estimated by comparison with the standard charts
of Beard and Weyl (1973). Packing proximity index was
determined by following Kahn (1956) procedures. Carbonate
cements were stained for identification with an acid solution
of alizarin red and potassium ferrocyanide (Dickson, 1965).
Microporosity was determined as the difference between
petrophysical porosity and porosity measured by petrographic
modal analysis. The percentage of types of detrital and
diagenetic constituents and of pores is expressed in relation to
bulk rock volume. Standard petrophysical nitrogen porosity
and air permeability values were acquired from 38 horizontal
plugs corresponding to part of the studied thin sections.
The morphology and the textural relationships
among minerals were examined in 11 gold-coated samples
with a JEOL JSM 5800 scanning electron microscope
(SEM) equipped with a Noran energy-dispersive spec-
trometer (EDS), using an accelerating voltage of 10 kV.
In order to identify the clay minerals present in the
sandstones X-ray diffraction analyses of the , 2 mm
fraction were performed in a Rigaku RU 200 diffractometer
in 16 oriented samples. The samples were air-dried,
ethylene glycol-saturated and heated at 490 8C for 4 h.
1053
Twenty-five thin-sections were carbon-coated and ana-
lysed with a Cameca Camebax SX50 electron microprobe
(EMP) equipped with four spectrometers and a back-
scattered electron detector (BSE). The operating conditions
were an acceleration voltage of 15 kV, beam diameter of
5 mm, beam current of 8 nA for the carbonates and, 12 nA
for the silicates. The standards used were anorthite (Ca, Al,
Si), olivine (Mg, Fe, Mn), microcline (K), barite (Ba),
strontianite (Sr) and jadeite (Na).
For the stable-isotope analyses of diagenetic carbonates,
the bulk sample was powdered (, 200 mesh) and reacted
with 100% phosphoric acid after 1 h at 25 8C for calcite and
at 50 8C for dolomite, and after 6 days at 50 8C for siderite.
The evolved CO2 gas was analysed for carbon and oxygen
isotopes on a SIRA-12 mass spectrometer. The phosphoric
acid fractionation factors used were 1.01025 for calcite at
25 8C (Friedman & O’Neil, 1977) and 1.009082 for siderite
(Rosenbaum & Sheppard, 1986). The carbonate isotope data
are presented in the normal d notation relative to PDB
(Craig, 1957).
In this paper, the terms eo-, meso- and telodiagenesis are
applied for the diagenetic stages sensu Schmidt and
McDonald (1979).
6. Sandstone texture and composition
The Uerê sandstones range from very fine- to coarse-
grained, with predominance of fine to medium grain size. The
sorting is usually moderate. There is a marked mixture of
well-rounded and angular grains in these sandstones,
indicating a mixture of recycled and first-cycle sediments.
The sandstones are essentially subarkoses, and rarely
quartzarenites (Folk, 1968). The minor dissolution of
feldspars has not significantly modified the original average
detrital composition from Q83F16.4L0.6 to present-day
Q85.4F14L0.6 (Fig. 5). Original composition was reconstructed
Fig. 5. Detrital composition of 65 representative Uerê sandstones (plotted in
the upper half of Folk (1968) diagram).
1054 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
from modal petrographic values of grain-replacive diagenetic
constituents and grain-dissolution porosity.
Quartz grains are manly monocrystalline (av. 59%; max.
78%) with slightly undulose to abrupt extinction, whereas
polycrystalline quartz occurs as trace to 4% (av. 0.6%).
Detrital potassium feldspar (av. 5%; up to 10%) dominates
over plagioclase, which is totally albitised. Microcline and
perthite are the most common types of potassium feldspar
(av. 3%, up to 7%, and av. 2%, up to 5%, respectively). The
main types of rock fragments are of fine micaceous
metamorphic, granitic-gneissic plutonic, strongly illitised
felsic volcanic and mudrocks.
Mud intraclasts occur in trace amounts (up to 3%)
normally compacted to pseudomatrix (av. 0.7%; up to 4%)
and/or silicified (av. 1%; up to 11%). Trace amounts of
intraclasts of sponge spiculites occur in lag layers.
Clay ooids occur dispersed in the sandstones (av. 0.4%;
up to 10%; Fig. 6(B)). The ooids show an approximately
glauconitic composition according to microprobe analyses,
and typically contain nuclei of detrital heavy minerals,
quartz, feldspar, or expanded mica.
Bioturbation is widespread, and responsible for the mud
matrix with high organic matter content present in some of
the sandstones. Palinomacerals analysis reveals abundance
of land plant fragments, as well as chitinozoans, scoleco-
donts and acritarchs (Loboziak et al., 1994). This bioturba-
tion matrix is commonly silicified, replaced by framboidal
pyrite or, rarely, by sphalerite. Heavy minerals are locally
abundant (av. 1%; up to 6%), comprising mainly zircon,
tourmaline, sphene, and Fe– Ti oxides, the latter commonly
replaced by diagenetic TiO2.
The detrital composition of Uerê sandstones is consistent
with the intracratonic setting of the Solimões Basin, and with
the erosion of granitic and high-grade metamorphic Precam-
brian rocks continental blocks of Guyana and Brazilian shields
and recycling of pre-Middle Devonian sedimentary rocks.
7. Diagenetic constituents
The main diagenetic processes affecting the Uerê
sandstones are mechanical and chemical compaction,
authigenesis of various forms of silica, carbonates, and
clay minerals, and dissolution and replacement of detrital
grains by albite, carbonates, clay minerals, pyrite and TiO2.
7.1. Silica
The authigenesis of silica developed several habits in
the Uerê sandstones, including (1) microquartz rims;
(2) intergranular pore-filling microquartz; (3) isopachous
chalcedony rims; (4) intergranular pore-filling displacive
chalcedony; (5) syntaxial quartz overgrowths and (6)
prismatic quartz outgrowths. The two latter habits are
widespread, whereas the first four habits are restricted to
sandstones (originally) rich in sponge spicules.
Microquartz rims. Microquartz occurs as isopachous
rims less than 20 mm thick, continuously covering the
detrital grains (Fig. 6(C) – (E)). In the sandstones were they
occur, such silica rims average 4% (up to 11%), but their
general average in the unit is 2% (Table 1). The rims show a
layered structure of cryptocrystalline quartz (, 1 mm thick)
underlying a layer of randomly to parallel-oriented,
euhedral microquartz (1 –5 mm; Fig. 6(D) and (E)). Fibrous
or ribbon-like illite and honeycombed mixed-layer illite-
smectite are intimately mixed with the microcrystalline
quartz rims in nearly 35% of microquartz-cemented
sandstones (av. 1%; up to 5%; Fig. 6(E) and (F)).
Intergranular pore-filling microquartz. Pore-filling
microquartz cement occurs as aggregates of euhedral,
randomly oriented microquartz crystals (av. 2%; up to
8%), similar in size and habit to grain-coating microquartz
(Fig. 6(D)), being associated with small amounts of fibrous
illite and ribbon-like illite-smectite.
Isopachous chalcedony rims. Chalcedony rims occur in a
few sandstones containing silica/clay grain-coatings as 20–
80 mm thick continuous rims, averaging 1% (up to 13%; av.
0.6%). Chalcedony surrounds mouldic pores after the
dissolution of sponge spicules (20 – 80 mm long and 5–
25 mm wide) (Fig. 6(A)). Locally, microquartz rim covers
isopachous chalcedony.
Intergranular pore-filling displacive chalcedony. This
cement occurs only in the hybrid arenites (Zuffa, 1980) with
sponge spicules or in pure sponge spiculites (av. 2%; up to
34%), in which detrital grains comprise less than 10% of
volume. Floating bioclasts (present-day mouldic pores) and
siliciclastic grains and the high IGV values (< 40%)
indicate the displacive character of this chalcedony cement.
Pore-filling chalcedony shows a spherulitic texture and
contains abundant spherical micropores (10 – 20 mm), inter-
preted as mouldic pores after opal-CT lepispheres
(Fig. 7(A); Maliva & Siever, 1988). Spherulitic pore-filling
chalcedony covers (hence post-dates) isopachous chalced-
ony rims that line detrital grains and mouldic pores after
sponge spicules.
Quartz overgrowths. The volume of syntaxial quartz
overgrowths (av. 4%; up to 10%) do not increase with depth,
but rather vary widely from well to well. Quartz overgrowths
are abundant both in shallow-buried, basin-margin sand-
stones (av. 5%, up to 9% in RCA-1 at , 1940 m depth), and
in deeply buried, basin-centre sandstones (av. 5%, up to 10%
in NSM-1 at , 3280 m depth; Fig. 7(B)). The occurrence of
poorly developed and discontinuous syntaxial quartz over-
growths in early silica/clay-cemented sandstones (av. 2%; up
to 8%; Fig. 7(C)) suggests that overgrowth distribution is
controlled by microquartz and authigenic clay. Overgrowths
are mostly post-compactional, as evidenced by their absence
along intergranular contacts. In tightly compacted sand-
stones quartz overgrowths occlude most of the remaining
intergranular porosity, and eventually engulf of fibrous illite
(Fig. 7(D)). The scarcity and tiny size of fluid inclusions in
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071 1055

Fig. 6. (A) Optical photomicrograph of a sponge spiculite with mouldic pores after sponge spicules, rimmed and partially filled by chalcedony; plane light.
(B) Backscattered electrons (BSE) image of a glaucony ooid. (C) Optical photomicrograph of a sandstone with thin microcrystalline quartz rims. (D) Secondary
scanning electrons microscopy (SEM) image of a sandstone cemented by microcrystalline quartz rims. (E) SEM image detail showing rims of oriented
microquartz crystals. (F) SEM image of a grain surface rimmed by microcrystalline quartz and clay and minor fibrous illite (IS); note cryptocrystalline quartz
(Qc), as well as tiny quartz crystals grown on top of the clay coating (arrow); Qo: quartz outgrowth.

the quartz overgrowths rendered the acquisition of homo- microquartz, commonly engulfing fibrous illite as scat-
genisation temperature impossible. tered crystals, extending into mouldic, intra- and inter-
Quartz outgrowths. Prismatic macroquartz outgrowths granular pores mostly in sandstones cemented by early
occur on quartz grains discontinuously coated by silica/clay (av. 1%; up to 6%; Fig. 7(C)).
1056 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071

Table 1
Statistical summary of the petrographic and petrophysical parameters of the Uerê sandstones. Parameters that are distinctive among petrofacies are italicised

Petrofacies; number of samples Petrofacies A; n ¼ 29 Petrofacies B; Petrofacies C; Whole unit; n ¼ 59

n ¼ 19 n ¼ 11

Total Average Max Average Max Average Max Average Max

Detrital quartz 50.8 63.0 69.4 79.3 64.1 75.0 59.4 79.3
Quartz monocrystalline 50.3 62.7 68.6 78.3 63.6 74.3 58.8 78.3
Quartz polycrystalline 0.5 1.7 0.8 4.3 0.5 1.3 0.6 4.3
Quartz in plutonic r.f. 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.7
Detrital feldspar 5.2 11.0 5.4 10.0 5.8 9.0 5.4 11.0
Detrital K-feldspar 5.2 10.3 5.3 10.0 5.7 9.0 5.3 10.3
Orthoclase 0.2 2.0 0.1 0.7 0.1 1.0 0.1 2.0
Microcline 3.1 5.0 3.0 6.0 3.5 6.7 3.2 6.7
Perthite 1.9 5.3 2.2 4.0 2.0 3.3 2.0 5.3
K-feldspar in plutonic r.f. 0.1 2.0 0.1 0.7 0.1 0.7 0.1 2.0
Detrital plagioclase 0.0 0.7 0.0 0.3 0.0 0.3 0.0 0.7
Plagioclase monocrystalline 0.0 0.3 0.0 0.3 0.0 0.3 0.0 0.3
Plagioclase in plutonic r.f. 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.7
Plutonic r.f. 0.2 2.7 0.1 0.7 0.1 0.7 0.1 2.7
Total fine-crystalline lithics 0.3 1.0 0.4 1.3 0.1 1.0 0.3 1.3
Volcanic r.f. 0.0 0.3 0.0 0.3 0.1 0.7 0.0 0.7
Chert r.f. 0.1 0.7 0.1 0.7 0.0 0.0 0.1 0.7
Mudrock fragment 0.2 1.0 0.2 1.0 0.0 0.3 0.2 1.0
Metamorphic rock fragment 0.0 0.3 0.1 0.7 0.0 0.0 0.0 0.7
Mica 0.5 3.7 0.1 0.7 0.1 1.0 0.3 3.7
Heavy minerals 1.1 3.0 0.8 4.3 1.1 6.0 1.0 6.0
Mud intraclast 0.2 2.7 0.0 0.0 0.0 0.3 0.1 2.7
Silica intraclast (spiculite) 0.1 2.0 0.0 0.3 0.0 0.0 0.0 2.0
Clay ooid 0.1 2.0 0.8 10.0 0.5 4.0 0.4 10.0
Silica bioclast 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.7
Mud pseudomatrix þ bioturbation matrix 0.8 4.0 0.6 4.3 0.4 3.0 0.7 4.3
Diagenetics total 26.1 49.1 16.4 21.7 20.3 28.0 21.8 49.1
Quartz overgrowth 2.3 7.7 3.6 6.7 7.6 10.3 3.7 10.3
Quartz outgrowth 1.3 6.3 0.3 2.7 0.1 0.7 0.7 6.3
Quartz filling moldic pore 0.2 3.3 0.0 0.0 0.0 0.0 0.1 3.3
Chalcedony displacive cement 1.8 33.8 0.0 0.0 0.0 0.0 0.9 33.8
Chalcedony rims 1.2 13.3 0.0 0.3 0.0 0.0 0.6 13.3
Chalcedony spherullite 0.1 1.3 0.1 1.0 0.0 0.0 0.1 1.3
Microquartz rims 4.1 10.7 0.0 0.0 0.0 0.0 2.0 10.7
Microquartz pore-filling 2.0 8.3 0.0 0.0 0.0 0.0 0.9 8.3
Clay coatings 0.9 4.7 0.0 0.0 0.0 0.0 0.4 4.7
Silicified secondary matrix 1.9 11.0 0.1 1.0 0.2 2.0 1.0 11.0
Anhydrite coarse intergranular 0.5 2.7 0.3 1.0 0.1 0.3 0.3 2.7
Anhydrite replacing feldspar grain 0.0 0.3 0.0 0.0 0.1 0.7 0.0 0.7
Anhydrite replac. quartz grain 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.3
Dolomite replacive intergranular 1.4 6.3 2.3 5.3 1.1 4.7 1.6 6.3
Dolomite replacing feldspar grain 0.2 1.0 0.4 2.0 0.3 1.3 0.3 2.0
Dolomite replacing quartz grain 0.3 1.3 0.3 1.3 0.1 0.3 0.3 1.3
Dolomite replacing secondary matrix 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Siderite intergranular 0.1 2.3 0.0 0.0 0.0 0.0 0.0 2.3
Siderite replacing grain 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.7
Illite intergranular fibrous 0.7 2.3 1.9 5.7 0.8 3.3 1.1 5.7
Illite coatings 0.4 2.3 1.1 4.3 0.3 2.3 0.6 4.3
Illite in feldspar grain 0.1 1.3 0.2 0.7 0.1 0.7 0.1 1.3
Illite after kaolinite 0.3 4.0 0.0 0.0 0.0 0.3 0.2 4.0
Illite in mica 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.3
TiO2 intergranular 0.5 1.7 0.6 2.7 1.1 3.7 0.7 3.7
TiO2 replacing grain 0.1 1.0 0.1 1.0 0.3 0.7 0.2 1.0
Pyrite framboidal intergranular 0.8 2.3 0.1 1.0 0.8 3.0 0.5 3.0
Pyrite coarse intergranular 0.1 0.7 0.1 0.7 0.1 0.7 0.1 0.7
Pyrite replacing grain 0.3 2.7 0.1 1.0 0.3 2.0 0.3 2.7
Pyrite framboidal in feldspar 0.2 0.7 0.1 0.7 0.2 1.0 0.1 1.0
Albite replacing plagioclase 2.4 5.0 1.8 3.3 2.7 5.0 2.2 5.0
Albite replacing K-feldspar 1.8 4.0 1.7 4.0 3.1 6.0 2.0 6.0
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071 1057

Table 1 (continued)
Petrofacies; number of samples Petrofacies A; n ¼ 29 Petrofacies B; Petrofacies C; Whole unit; n ¼ 59
n ¼ 19 n ¼ 11

Total Average Max Average Max Average Max Average Max

Albite overgrowth 0.1 1.0 0.4 1.3 0.7 1.7 0.3 1.7
Bitumen 0.0 0.3 0.4 1.3 0.3 1.0 0.2 1.3
Macroporosity 14.9 25.0 6.4 11.7 7.6 13.3 10.7 25.0
Intergranular 9.8 16.3 4.0 8.3 5.3 9.7 7.0 16.3
Intragranular in feldspar 2.1 4.3 1.7 3.0 1.4 3.3 1.9 4.3
Intragranular in quartz grain 0.8 3.0 0.2 1.0 0.2 1.3 0.5 3.0
Intragranular in mica 0.0 0.7 0.0 0.0 0.0 0.3 0.0 0.7
Intragranular in heavy mineral 0.1 0.7 0.1 0.7 0.0 0.3 0.1 0.7
Dissolution of pseudomatrix 0.5 5.0 0.0 0.7 0.2 2.0 0.3 5.0
Dissolution of cement 0.1 2.6 0.0 0.3 0.1 1.3 0.1 2.6
Mouldic 1.1 6.3 0.0 0.0 0.0 0.0 0.5 6.3
Fracture 0.1 1.0 0.2 1.3 0.0 0.3 0.1 1.3
Oversized 0.3 3.3 0.3 1.3 0.3 1.0 0.3 3.3
Petrophysical porosity 15.3 28.0 9.2 13.8 12.3 17.8 12.8 28.0
Petrophysical permeability 67.6 1153.6 4.7 29.1 7.2 20.9 35.6 1153.6
Microporosity 15.3 25.0 6.4 11.7 7.2 13.3 10.8 25.0
Secondary porosity 5.1 10.1 2.4 5.0 2.3 8.0 3.7 10.1
Intergranular volume 31.6 45.1 16.0 21.0 19.2 26.0 24.1 45.1
Grain volume 68.4 78.7 84.0 90.7 80.8 84.3 75.9 90.7
Cement total 18.5 41.9 11.6 17.7 13.0 19.3 15.2 41.9
Carbonate total 9.5 77.0 2.9 6.0 1.5 6.0 5.9 77.0
Silica total 11.1 24.0 4.1 7.3 7.6 11.0 8.1 24.0
Grain replacement total 7.7 19.0 8.6 14.3 8.8 14.0 8.2 19.0
Modal grain size (mm) 0.2 0.4 0.3 0.5 0.2 0.4 0.2 0.5
Sorting 0.9 2.0 0.8 2.0 0.7 1.0 0.8 2.0
Packing index (Pp) (G/G) 15.1 26.0 40.4 80.0 37.4 54.0 27.6 80.0
Original porosity (Beard & Weyl, 1973) 33.3 40.2 33.9 39.7 34.4 38.8 33.7 40.2
Compactacional porosity loss (%) COPL 18.5 26.0 28.3 32.4 27.2 32.6 23.4 32.6
Cementational porosity loss (%) CEPL 8.8 25.7 4.0 6.2 4.8 6.8 6.5 25.7

7.2. Clay minerals
Authigenic clay minerals (trace to 9%) include mainly
illite and mixed-layer illite-smectite that occur as pore-
lining rims and pore-filling cement, and as replacement
of mud matrix and detrital feldspar. Illite was formed
both by direct neoformation in the pores and by
replacement of kaolinite and smectite precursors. Total
Illitic clay contents average 2%, of which 0.6% (up to
4%) occurs as discontinuous rims on detrital grains
(Fig. 7(E)), 0.2% (up to 1%) occurs as replacement of
detrital feldspars, and 1% (up to 6%) occurs as
intergranular neoformed cement. Unsilicified pseudoma-
trix was replaced by illite. Illite replacement of kaolinite
(av. 0.2%; up to 4%) is pseudomorphic, preserving the
vermicular and booklet habits of the kaolinite aggregates.
Such illitised kaolinite occurs adjacent to secondary
pores after feldspar dissolution (Fig. 7(F)). Illite replaces
the feldspar grains along cleavage boundaries and
twinning planes.
As determined by XRD analysis, mixed-layer illite-
smectite clays are well-ordered, with 75– 85% illite, and
were formed by the transformation of pre-compactional
smectite coatings intimately intergrown with the microquartz
rims (Fig. 7(D)). Such eogenetic smectite is abundant around
mouldic pores after sponge spicules and feldspar grains.
Illite-smectite and illite also formed by the transformation of
detrital smectite from bioturbation matrix or pseudomatrix
formed by the compaction of mud intraclasts.
Chlorite occurs as interlocked pseudohexagonal crystals
in the shallow-buried quartz-cemented sandstones
(, 1940 m depth) in well RCA-1, and as thin and
discontinuous rims (< 5 mm) covered by quartz over-
growths (Fig. 8(A)). Bulk XRD data of these sandstones
revealed that clay minerals average 5% (trace to 8%), in
which chlorite comprises 85% and I/S 15%.
7.3. Carbonates
Ferroan dolomite and ankerite dominate over siderite and
calcite (Fig. 9). Fe-dolomite/ankerite is widespread as
scattered spots, which give a mottled appearance to the
sandstones (Fig. 4(A)). Locally, it occurs as partial to
pervasive strata-bound (0.01 – 0.1 m thick) cement in the
proximity of organic matter-rich shale layers. Fe-dolomite/
ankerite shows a blocky to poikilotopic habit (0.1 – 10 mm)
habit, occurring as intergranular, locally replacive and
displacive cement in sandstones with high IGV values
1058 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071

Fig. 7. (A) SEM image of microquartz-rimmed spherical pores interpreted as opal-CT lepisphere moulds. (B) Optical photomicrograph of a sandstone
massively cemented by quartz overgrowths. (C) SEM image of a sandstone with grains partially rimmed by microcrystalline quartz, with discontinuous
overgrowths and outgrowths. (D) SEM image of a sandstone cemented by quartz overgrowths engulfing illite-smectite clays. (E) BSE image showing
well-developed fibrous illite rims. (F) BSE image showing illitised kaolinite engulfing Fe-dolomite (white) in a microquartz-rimmed sandstone.

(32 –37%; Fig. 8(B)). Oversized patches of dolomite are the
product of complete replacement of spicules, identified in
BSE images as darker, slightly magnesium-enriched ‘ghosts’
(Fig. 8(C)). Commonly, Fe-dolomite/ankerite poikilotopic
patches shows oxidised borders. Fe-dolomite/ankerite is
absent within mouldic pores after feldspar dissolution.
Ankerite occurs locally in tightly compacted to
recrystallised sandstones and metasandstones from
SMT-2 and SMT-3 wells area. Ankerite is anhedral to
subhedral in habit with curved crystal faces and undulose
extinction (saddle ankerite) occurring as veins cross-
cutting the metasandstones. Vein-filling saddle ankerite
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071 1059

Fig. 8. (A) SEM image showing chlorite rims partially engulfed by quartz overgrowths. (B) BSE image showing poikilotopic Fe-dolomite cement replacing
microquartz rims. (C) BSE image of a sandstone cemented by Fe-dolomite with ‘ghosts’ of replaced sponge spicules. (D) Optical photomicrograph showing a
metasandstone with mica and ankerite cementation along fractures. (E) BSE image showing Fe-dolomite (medium gray) engulfing siderite (white). (F) Optical
photomicrograph of a strongly pressure-dissolved sandstone with thin illite coatings (petrofacies B).

(1.0 – 2.5 mm thick) is often associated with stylolites along
mica/clay-rich laminae, replacing considerable portion of
these rocks as irregular patches (Fig. 8(D)). Saddle ankerite is
non-stoichiometric [Ca1.15(Fe0.42, Mg0.37, Mn0.12)(CO3)2].
Siderite cement occurs in loose-packed sandstones
(packing proximity index , 20%; IGV , 32%) as scat-
tered rhombs (5.0 –60 mm), commonly associated with
framboidal pyrite and engulfed by Fe-dolomite/ankerite
(Fig. 8(E)). Locally, siderite forms spherulitic aggregates
(20 – 60 mm), and replaces quartz, feldspar and mud
intraclasts. Many Uerê siderites are extremely Mg-rich
(MgCO3 av. 18 mol%; up to 32 mol%; sideroplesites,
Deer, Howie, & Zussman, 1962). Siderite crystals are
irregularly zoned in terms of Fe and Mg contents.
1060 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Fig. 9. Microprobe composition of representative carbonate cements in
Uerê sandstones.
7.4. Albite
Diagenetic albite with a typical pure end-member
composition is common (av. 2%; up to 6%), occurring as
replacement of, and as overgrowths on feldspar grains. Both
detrital K-feldspar and plagioclase were albitised, with
replacement occurring after partial grain dissolution, as
evidenced by the variable amounts of dissolution voids
within these grains. Albitised K-feldspar and plagioclase
were distinguished according to the petrographic and
chemical criteria described by Morad, Bergan, Knarud and
Nystuen (1990). Albitised K-feldspars are untwinned, with
irregular extinction and turbid appearance due to
the presence of numerous, small fluid inclusions and
microporosity. The albitised feldspars are composed of a
large number of small (10 –50 mm), lath-like albite crystals.
The albitised plagioclase grains show irregular polysyn-
thetic twins, due to their replacement by numerous
elongated albite crystals (0.02 – 0.1 mm) arranged parallel
to the twinning and cleavage planes of the host grains.
In addition to the replacive habit, albite also occurs as
overgrowths on plagioclase grains (av. 0.3%; up to 1.7%). In
some cases, the overgrowths surround intragranular and
mouldic pores that resulted from the post-overgrowth
dissolution of detrital plagioclase cores. Albite also occur as
small (2–5 mm) discrete crystals associated with illitic clays.
7.5. Other diagenetic constituents
Pyrite is widespread and occurs as intergranular
framboids (, 8 mm; av. 0.7%; up to 3%) and, rarely, as
coarse euhedral crystals (8 –20 mm; av. 0.3%; up to 0.7%).
Grain-replacive pyrite is also common, occurring along
feldspar fractures and cleavages (av. 0.3%; up to 1%), and
within mud intraclasts, heavy minerals, and glauconitic
ooids and concentrated along stylolites (av. 0.3%; up to 3%).
Authigenic titanium oxides occur mostly as intergranular
aggregates of scattered euhedral crystals of rutile
(2 –10 mm), replacing grains and filling secondary pores
after detrital Ti – Fe heavy minerals.
Anhydrite occurs as minor (av. 0.4%; up to 3%), coarsely
crystalline (10 –25 mm), post-compactional intergranular
cement that partially replaces feldspar and quartz grains and
overgrowths.
Bitumen occurs either infilling pores or as dark brown
coatings on the surface of quartz overgrowths and smectite-
illite rims (av. 0.2%; up to 1%).
8. Compaction and porosity
Conspicuous evidence of mechanical compaction
includes the fracturing of quartz and feldspar grains
(commonly healed by authigenic quartz and albite,
respectively) and by the deformation of ductile grains
(mica, mud intraclasts, glauconitic ooids). Chemical
compaction occurred by pressure dissolution both along
intergranular contacts and stylolites (Fig. 8(F)). Compaction
is limited in sandstones cemented by abundant eogenetic
silica or carbonate cementation. Locally, in sandstones
cemented by microquartz rims, pressure dissolution is
limited to intergranular sutured and concave-convex con-
tacts with rare, low-amplitude (, 0.1 mm), pyrite-lined
stylolites along laminations enriched in detrital heavy
minerals and micas. Sandstones devoid of early microquartz
cement display pervasive chemical compaction, represented
by abundant stylolites (0.25 –2 mm amplitude), in places
amalgamated in dissolution seams a few mm thick.
Thin section porosity (av. 11%; up to 25%) varies widely
with cement type. In sandstones cemented by eogenetic
microquartz, macroporosity averages 15%, whereas in
sandstones devoid of such cement and with scarce (, 5%)
or abundant (5 – 10%) quartz overgrowth, macroporosity
averages 6 and 8%, respectively (Table 1). About two-thirds
of the macroporosity is primary intergranular, with a
maximum value of 16% (av. 7%). Secondary porosity
comprises dominantly mouldic pores generated by the
dissolution of detrital feldspar (av. 2%; up to 4%), and
sponge spicules (av. 0.6%; up to 6%). The latter type is
volumetrically important in hybrid arenites in the form of
elongate (20 – 80 mm long and 5 – 25 mm wide) pores
rimmed by microquartz, chalcedony or smectitic clay.
Other secondary pore types include those generated by
the dissolution of pseudomatrix (av. 0.3%; up to 5%), silica
cement (chalcedony, intergranular microquartz aggregates;
av. 0.1%; up to 3%), and oversized pores (av. 0.3%; up to
3%) formed by the dissolution of detrital quartz and feldspar
grains. Microporosity is widespread in the sandstones,
averaging 3% (largest values, up to 11%, in microquartz/
chalcedony-cemented sandstones).
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
9. Discussion
9.1. Reservoir petrofacies
Three petrofacies were recognised in the sandstones,
based on their packing, porosity, and cementation (Table 1):
(1) petrofacies A: porous (. 15%) sandstones with micro-
quartz grain-rim; (2) petrofacies B: tight (, 10%), strongly
compacted to conspicuously quartz-cemented sandstones;
(3) petrofacies C: moderately porous (10 –15%), partially
quartz-cemented sandstones. These petrofacies are thus
defined based on petrographic and petrophysical character-
istics of the sandstones, which are in turn strongly controlled
by diagenesis, without stratigraphic or environmental
connotations. The three petrofacies are easily recognised
on a plot of intergranular volume (IGV) versus volume of
silica cement (including microquartz grain-coating and
syntaxial quartz overgrowths; Fig. 10). Petrofacies A is thus
characterised by IGV . 20% and various amounts of
microcrystalline silica cement. Petrofacies B an C are
characterised by IGV , 20%, and are distinguished based
on an amount of quartz overgrowth cement of , 5% for
petrofacies B and between 5 and 10% for petrofacies C.
The petrophysical characteristics of each petrofacies are
also consistent with the above-mentioned definition
(Fig. 11). Strongly compacted and quartz-cemented
petrofacies B and C show low porosity values (, 15%)
compared to porous petrofacies A sandstones (15 – 28%),
which also display better permeabilities.
The diagenetic evolution pathways of the three defined
petrofacies are schematically represented in Fig. 12, which
depicts the major diagenetic conditions encountered by the
Uerê sandstones, discussed below, and the resulting
reservoir quality products.
Fig. 10. Plot of the silica cement volume (chalcedony, microcrystalline and
macrocrystalline quartz) versus intergranular volume, showing a clear
distinction between the three reservoir petrofacies.
1061
Fig. 11. Plot of petrophysical helium porosity versus horizontal air
permeability for the three reservoir petrofacies. The highest porosity and
permeability values correspond to sandstones cemented by microcrystalline
silica from petrofacies A and the lowest values, to tightly compacted
petrofacies B sandstones. Quartz cemented petrofacies B sandstones show
intermediate porosity and permeability values.
9.2. Porosity reduction processes
A plot of cement volume versus intergranular volume
(Houseknecht, 1987; Fig. 13(A)) reveals that the reduction
of porosity occurred through an interplay of compaction and
cementation. In petrofacies A sandstones with abundant,
pre-compactional eogenetic silica cement, porosity
reduction was dominantly due to cementation. Conversely,
porosity reduction was dominantly due to chemical
compaction in petrofacies B (Fig. 8(F)), as well as in
petrofacies C sandstones with post-compactional quartz
overgrowths (Fig. 7(B)).
However, by plotting calculated indices that take into
consideration the reduction in bulk rock volume due to
chemical compaction (Lundegard, 1992), a more realistic
evaluation of the relative roles of compaction and
cementation in porosity reduction is obtained (Fig. 13(B)).
A plot of corrected compactional versus ‘cementational’
porosity reduction indices reveals that compaction was
overall more important than cementation in reducing
porosity in the whole unit, including in petrofacies A
sandstones, with the exception of a few samples
(Fig. 13(B)). The ‘original depositional porosity’ values
were obtained from Beard and Weyl (1973) parameters.
The assumed average original porosity of petrofacies A is
33.5% (i.e. moderately to well-sorted sandstones with
0.2 mm average modal grain size). These sandstones show
average intergranular volume of 27.3% and average
cement volume of 18%. According to the calculated
compactional and cementational porosity loss indices,
these silica-cemented sandstones have lost on average
19.4% porosity (around 57% of the original porosity) due
to compaction and 8.5% porosity (around 25% of the
original porosity) due to cementation.
1062 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Fig. 12. Diagenetic evolution pathways of Uerê sandstones for the defined reservoir petrofacies.
9.3. Marine eodiagenesis
Marine eodiagenesis was strongly influenced by the
presence of siliceous sponge spicules. Siliceous sponge
spicules were originally abundant in the Late Devonian
sediments of the Solimões Basin, but have been
extensively dissolved, recrystallised or replaced by
microcrystalline quartz during eodiagenesis and shallow
mesodiagenesis. Mouldic pores after dissolved spicules
are common within microquartz- and chalcedony-cemen-
ted sandstones of petrofacies A, attesting to a major
redistribution of silica from biogenic sources to eogenetic
cements (Fig. 14(A)). Thermodynamic and kinetic
models indicate that progressive silica transformations
in marine deposits occur from the phase of highest
entropy (amorphous opal) to the phase of lowest entropy
(quartz) in a dissolution-reprecipitation pathway. The
generalised diagenetic sequence is: opal-A (biogenic
silica) ! disordered opal-CT ! ordered opal-CT !
cryptocrystalline quartz or chalcedony ! microcrystalline
quartz (Williams & Crerar, 1985; Williams, Parks, &
Crerar, 1985). The process starts with biogenic silica
dissolution (point 1 in Fig. 14(A)) and supersaturation of
pore fluids with respect to opal-CT and quartz. Relative
rates of opal-A and opal-CT nucleation govern the extent
to which silica activity is buffered near point 2 in
Fig. 14(A), and hence the net increase in surface area.
From the point when further changes in specific surface
area have negligible effect on opal-A solubility, opal-CT
of the surface area # that of point 3 in Fig. 14(A) can
form. With progressively less opal-A remaining, silica
activity starts to decline and opal-CT nucleation becomes
subordinate to growth. Opal-CT crystals with lower
surface area form at the expense of smaller and less
ordered ones. Such competitive crystal growth process is
known as Ostwald ripening.
A specific surface area of 150 m2/g was assumed for
the sponge spicules in the Uerê sandstones, what is
significantly lower than the specific surface area of
microporous radiolaria and diatoms tests (, 250 m2/g)
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Fig. 13. (A) Plot of cement volume versus intergranular volume (House-
knecht, 1987) and (B) plot of compactional porosity loss (COPL) versus
cementational porosity loss (CEPL) (Lundegard, 1992) for 65 representa-
tive Uerê sandstones. See text for comments.
assumed by Williams et al. (1985), and higher than that
of opal-CT lepispheres. Evidence of intermediate opal-
CT precipitation in the Uerê sandstones includes the
presence of spheroidal pores within microquartz and
chalcedony cements that are probably moulds of
dissolved opal-CT lepispheres (Fig. 7(A); Astin, 1987;
Hendry & Trewin, 1995). The occurrence of micro-
quartz or chalcedony as rims around such opal-CT
moulds suggests that locally petrofacies A sandstones
followed the pathway 1 ! 2 ! 3 during opal-A/opal-CT
transition (Fig. 14(A)). Precipitation of cryptocrystalline
quartz began once the thermodynamic drive for Ostwald
ripening of the opal-CT became insignificant (point 4 in
Fig. 14(A)). Relative rates of opal-CT dissolution with
1063
Fig. 14. Diagrammatic interpretation of the genetic relationships between
diagenetic silica polymorphs with the evolution of pore-fluid silica activity
and specific surface area (Williams et al., 1985): (A) commonly observed
silica polymorphs evolution, and (B) evolution of silica polymorphs in clay-
bearing sandstones.
respect to quartz nucleation and growth buffer the silica
activity until it falls back towards the quartz solubility
line with decreasing pore-water silica saturation. As the
system evolves down this line, cryptocrystalline quartz
is gradually replaced by epitaxial microcrystalline
quartz (point 5 in Fig. 14(A)). The dominance of
cryptocrystalline and microcrystalline quartz suggests
that the dissolution rate of precursor phases (i.e.
biogenic silica, opal-A and/or opal-CT) was fast enough
to sustain silica saturation at the microquartz saturation
level, resulting in numerous crystals instead of larger
ones (Jahren & Ramm, 2000). The common occurrence
of 1 – 5 mm crystals could be explained as the relatively
low microquartz solubility have kept the driving force
for crystal growth by Ostwald ripening at minimal
levels. This eogenetic pathway showed by sponge
spicules-rich sandstones is similar to the diagenetic
transformation of deep ocean siliceous oozes (Aase
et al., 1996; Knauth, 1994).
However, this idealised sequence of diagenetic trans-
formations does not correspond precisely to what is observed
in petrofacies A sandstones, where different silica poly-
morphs commonly coexisted. Opal-A and chalcedony have
coexisted, allowing the preservation of sponge spicules
1064 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
moulds, as well as opal-CT and microquartz, allowing the
preservation of moulds of lepispheres. Such coexistence may
have been allowed by the activity of Mg2þ in solution, as
well as by the presence of Al, Mg, Fe and Mn hydroxides and
alkali cations adsorbed onto grain surfaces. These factors are
considered to significantly affect silica reactivity and the rate
of opal-A ! opal-CT reaction (Dove & Crerar, 1990;
Kastner, Keene, & Gieskes, 1977 and references therein).
The original amount and shape of the sponge spicules is
preserved within areas cemented by eogenetic dolomite or
Mg-rich siderite (Fig. 8(C)). The consumption of Mg2þ and
increase in alkalinity during carbonate precipitation would
thus facilitate opal-CT and microquartz nucleation.
Such differences in the evolution pathway of eogenetic
silica precipitation in petrofacies A sandstones may be
related to the common occurrence of eogenetic clay
coatings. The presence of clay is known to speed up the
replacement of opal-CT by quartz (Chang & Yortsos, 1994;
Siever & Woodford, 1973; Williams et al., 1985), because
clay adsorbs silica, and thus inhibits opal-CT nucleation.
The observed clay coatings are composed of illite-smectite,
probably evolved from eogenetic smectite. Eogenetic
dissolution of biogenic silica promotes the precipitation
of K-rich smectite in estuarine and marine environments
(Michalopoulos & Aller, 1995; Michalopoulos, Aller, &
Reeder, 2000; Wollast & de Broeu, 1971). Clay mineral
precipitation and concomitant dissolution of biogenic silica
exert control on dissolved Si concentration, and thus on the
diagenetic evolution of silica polymorphs in marine
sediments. Additionally, such clays constitute a sink of Si
and K (and perhaps Mg and Fe) in the pore-water fluids,
and so may further influence early silica diagenesis
(Michalopoulos et al., 2000).
The presence of eogenetic clay coatings in sandstones of
petrofacies A may have induced the evolution pathway
1 ! 4 in Fig. 14(B). Detrital clays that adsorbed silica at
higher pore-water silica concentrations at points 1 –4 in
Fig. 14(B) were at points 4 and 5 out of equilibrium with the
new pore-water concentrations and began to desorb silica
(Chaika & Williams, 2000). This process buffered pore
water at low silica concentrations (, 20 mg/kg), which
favoured microquartz precipitation at shallow burial. There-
fore, the precipitation of eogenetic clays may have favoured
the precipitation of quartz relative to opal-CT both by
decreasing concentration of dissolved silica, thus
competing by the silica with opal-CT, and by inhibiting
opal-A ! opal-CT transformation and early silica nuclea-
tion along coated grain surfaces.
Variations in the amounts of early silica cements in
petrofacies A depend thus largely on the relative amount of
detrital clay, biogenic silica, and above all, authigenic clay
coatings. In the presence of abundant detrital clays from
bioturbation or mud intraclasts, or mostly eogenetic
smectite coatings, opal-A to opal-CT transformation was
inhibited and microquartz formed directly from biogenic
silica dissolution. However, thicker smectite coatings or
large amounts of bioturbation matrix inhibited silica
cementation.
Marine eodiagenesis included the precipitation of
coarsely crystalline Fe-dolomite/ankerite and siderite.
High content of Mg showing in most siderites are
consistent with precipitation from marine pore-water
(Mozley, 1989). Locally, spherulitic siderite aggregates
suggest that precipitation was triggered by bacterial
activity (Huggett, Dennis, & Gale, 2000). Fe-dolomite is
a common eodiagenetic carbonate in organic-rich marine
sediments if sulfate is rapidly depleted prior to major
compaction (Curtis & Coleman, 1986; Morad, 1998).
9.4. Meteoric eodiagenesis
The observed feldspar dissolution and precipitation of
authigenic kaolinite (Fig. 7(F)) was probably caused by
meteoric flushing related to the late Pennsylvanian-Permian
uplift (Fig. 2). Although for the good lateral connectivity of
the sandstones, the extent of kaolinite precipitation varies
widely, apparently due to the patterns of meteoric flow from
basin margin recharge areas. Kaolinite rarely replaces
directly the feldspar grains, but instead fills intergranular
and feldspar dissolution pores. This suggests a certain
degree of aluminium mobilisation with bulk porosity
enhancement in the sandstones. Widespread oxidation of
the earlier Fe-dolomite/ankerite poikilotopic cement, even
in tight compacted sandstones of petrofacies B also suggests
an environment of oxidizing fluids previous to effective
burial, compaction and quartz cementation in the
sandstones.
9.5. Mesodiagenesis
The precipitation of quartz overgrowths and outgrowths
occurred after effective burial and compaction. This is
indicated by (1) the intergrowth of quartz with fibrous illite
(Fig. 7(D)); (2) the relatively small intergranular volume
and quartz overgrowths volume of (petrofacies B: quartz
overgrowth av. 3.6% and IGV av. 16.7%; petrofacies C:
quartz overgrowths av. 7.6% and IGV av. 19.7%); (3) the
partial replacement of post-compactional, poikilotopic
anhydrite by quartz.
From the silica sources commonly considered for
quartz cementation (McBride, 1989), five could be
invoked as potential sources for the precipitation of
quartz overgrowths in Uerê sandstones: (1) the pressure
dissolution of detrital quartz along intergranular contacts
and stylolites; (2) the dissolution of opaline skeletons and
spicules of sponges; (3) the dissolution or alteration of
detrital silicates, mainly of feldspars; (4) the illitisation of
detrital and authigenic smectite or kaolinite; (5) the
convection of hot fluids related to magmatism.
Pressure dissolution was an important source of silica
for quartz overgrowth cementation in petrofacies B
sandstones, as indicated by the common occurrence of
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
sutured intergranular contacts and discrete stylolites.
Such features are however, rare to absent in pervasively
quartz-cemented petrofacies C sandstones, indicating that
internal pressure dissolution was not the only source for
mesogenetic quartz cementation.
The dissolution of sponge spicules was the main source
for the microcrystalline eogenetic silica cements, but was an
unlikely source for mesogenetic quartz overgrowths, since
spicule remnants were preserved from eogenetic dissolution
only in areas pervasively cemented by eogenetic carbonates
or silica. Likewise, the dissolution of detrital feldspars was
not an important silica source for mesogenetic quartz,
considering the limited volume of feldspars affected, and
the fact that the main episode of feldspar dissolution took
place during meteoric eodiagenesis, and not during
mesodiagenesis.
The illitisation of eogenetic smectite and kaolinite may
have partially supplied silica to mesogenetic quartz
precipitation. Assuming K-feldspar as the most likely
source of potassium, the transformation of smectite in illite
can be described as follows:
KAlSi3 O8 þ 2K0:3 Al1:9 Si4 O10 ðOHÞ2
K-feldspar smectite
! 2K0:8 Al1:9 ðAl0:5 Si3:5 ÞO10 ðOHÞ2 þ 4SiO2ðaqÞ
illite quartz
According to the quantified amounts of illitic clays, the
transformation of an average of 1.8 vol% of smectite would
produce , 0.6 vol% of quartz.
Furthermore, illitisation of kaolinite may also supply
silica for mesogenetic quartz precipitation:
Al2 Si2 O5 ðOHÞ4 þ KAlSi3 O8 ! Kal3 Si3 O10 ðOHÞ2
kaolinite K-feldspar illite
þ 2SiO2ðaqÞ þ H2 O
quartz
However, this reaction is of limited importance since
illitised kaolinite occurs significantly only in eogenetic
silica-cemented petrofacies A. According to Eq. (2), the
alteration of 0.2 vol% kaolinite would produce , 0.1 vol%
quartz. Thus, the intergrowth of illite and quartz cements
may presumably reflect: (1) release of silica due to the
reaction of kaolinite and smectite to form illite, and (2) co-
precipitation from high-temperature ($ 100 –140 8C), Si –
K – Al-charged fluids (Ehrenberg & Nadeau, 1989).
Therefore, internal silica sources seem to be insufficient
to explain the observed volumes of mesogenetic quartz
cementation. Quartz precipitation must have thus involved
the subsurface circulation of Si-charged fluids. The
relatively small mud/sand ratio of the Devonian sequence
(Figs. 2 and 3) suggests that the illitisation of smectites in
the Devonian shales (Boles & Franks, 1979) was of
limited importance as source for mesogenetic quartz
cements. The precipitation of the observed quartz volumes
would therefore, involve enhanced, probably convective,
1065
fluid circulation, to which the role of intrabasinal
magmatism may have been essential. There is a similarity
between the K/Ar and Ar/Ar ages of authigenic illite in
the Carboniferous sandstones (180 – 220 Ma; Mizusaki
et al., 1990) and those of the Triassic diabase intrusions
(Szatmari, 1996). This suggests that the widespread
precipitation of illite and co-precipitated quartz over-
growths, both in the Carboniferous and Devonian
sandstones, may have occurred in connection with the
circulation of fluids related to the rapid subsidence and
high heat flow promoted by the magmatism, Such
enhanced thermal regime would induce active fluid
convection through faults and permeable beds, such as
what is interpreted for quartz cementation in other basins
(De Ros, Morad, Broman, Césero, & Gomez-Gras, 2000;
Girard, Deynoux, & Nahon, 1989; Gluyas, Grant, &
Robinson, 1993; Summer & Verosub, 1992).
The effects of magmatism on the alteration of the
Devonian sandstones may have been locally intensified. A
seismic anomaly within the São Mateus field has long been
described as a ‘basement high’. Cores taken at , 3200 m in
the SMT-2 and SMT-3 wells have recovered metasand-
stones with granoblastic texture, extensively recrystallised
quartz phengitic mica and breccias cemented by replacive
ð1Þ saddle ankerite (Fig. 8(D)). However, the similarity of the
gamma-ray logs of these rocks with those of the Uerê
sandstones (Fig. 3(A)), and petrographic analyses suggest
that these intervals correspond to hydrothermally altered
Uerê sandstones. Such focused circulation of hydrothermal
fluids, presumably connected with the Triassic magmatism,
promoted extensive quartz recrystallisation, as well as mica
and ankerite formation, similarly to what is observed in
several present and ancient geothermal systems influenced
by advective (hydrothermal) fluid flow related to magmatic
activity (McDowell & Paces, 1985; Pitman, Henry, &
ð2Þ Seyler, 1998; Schiffman, Bird, & Elders, 1985; Searl, 1994
and references therein).
The albitisation of feldspars and precipitation of albite
overgrowths, anhydrite and minor ankerite are interpreted to
be deep mesogenetic phases that post-date the Triassic
magmatism. The origin and chemical composition of fluids
from which anhydrite and albite precipitated cannot be
unravelled from the available data. However, the mesoge-
netic anhydrite in the Uerê sandstones was probably derived
from the dissolution of anhydrite in the Carboniferous and
Permian evaporitic sequences (, 850 – 1100 m thick; Eiras
et al., 1994), which are in contact with the Devonian through
large reversed faults. During Jurassic to Early Cretaceous
Solimões Basin, inversion and extensive uplift promoted
meteoric flushing of the overlying evaporitic sequences, as
is characteristic of many other occurrences of late-stage
sulphate cements in deeply buried sandstones (Dworkin &
Land, 1994; Gluyas, Jolley, & Primmer, 1997). Evaporite-
sourced brines may also account for the observed feldspar
albitisation and precipitation of albite overgrowths.
Additionally to external Na supply, albitisation of feldspar
1066 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071

(mostly of K-feldspar) occurs when the aNaþ/aHþ ratio
increase relative to aKþ/aHþ ratio in pore fluids, due to
transformation of smectite to illite and to authigenesis of
illite (Morad, 1986). Both vitrinite reflectance (av. , 1%
Ro) and illite transformation reaction suggest that Uerê
sandstones were subject to maximum temperatures about
120 8C, which are within the temperature range for albite
authigenesis (Morad et al., 1990).
9.6. Paragenetic sequence
The sequence of diagenetic processes in the sandstones is
presented in Fig. 15. Due to the complex paragenetic
relationships among the authigenic constituents and to
the limitation of geothermometric and geochronologic data,
only a schematic representation of the diagenetic evolution
could be achieved. The simplified paragenetic diagram

Fig. 15. Diagram of the paragenetic sequence and burial history of the Uerê sandstones in the São Mateus oil field. See text for explanation.
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
depicts the general evolution of Uerê sandstones, and should
be examined together with the diagenetic evolution path-
ways of each petrofacies illustrated in Fig. 12. The
paragenetic evolution is framed within a burial history of
São Mateus Field reservoirs, representative for Uerê
Formation along basin centre.
Four major stages of diagenetic evolution were recog-
nised, including (Fig. 15): (1) marine eodiagenesis,
dominated by carbonate cementation, dissolution of sponge
spicule and silica pore-lining cementation, the latter two
processes being restricted to petrofacies A; (2) meteoric
eodiagenesis, which induced heterogeneous feldspar dissol-
ution and kaolinite precipitation; (3) shallow mesodiagen-
esis (1900 – 2600 m depth), marked by heterogeneous
chemical compaction and quartz cementation; and (4)
deep mesodiagenesis (. 2600 m depth), marked by illite
authigenesis and further quartz precipitation, probably
connected to Triassic magmatism. Thermal maturation of
Devonian shales and hydrocarbon emplacement on reser-
voirs took place at this time (Fig. 15; Mello et al., 1994).
Extensive eogenetic silica authigenesis, represented by
microquartz and chalcedony cementation, silicification of
pseudomatrix, plus bioturbation matrix, and of smectite
coatings, was promoted significant dissolution of sponge
spicules, which took place soon after deposition. The
precipitation of pyrite, magnesian siderite and Fe-dolomi-
te/ankerite occurred subsequent to silica cementation.
Sponge spicules were replaced by eogenetic carbonate
cement, leaving ‘ghosts’ discernible only in BSE images
(Fig. 8(C)), which indicate the large original amount of
sponge spicules within petrofacies A sandstones, which
would characterize them as hybrid arenites. The absence of
eogenetic carbonate cement within mouldic pores after
dissolved feldspar grains and the oxidised borders of the
poikilotopic carbonate cement spots is consistent with the
eogenetic precipitation of the carbonates prior to meteoric
water influx.
Meteoric eodiagenesis was responsible for feldspar
dissolution and minor kaolinite precipitation. Rapid burial
due to deposition of the thick Carboniferous-Permian
sequence promoted heterogeneous mechanical and chemi-
cal compaction, which affected mostly petrofacies B and C
sandstones devoid of eogenetic cementation. During
subsequent shallow mesodiagenesis widespread, syn- to
pos-compactional quartz overgrowths were precipitated
within petrofacies B and C, while restricted quartz out-
growths were formed in petrofacies A.
The occurrence of quartz outgrowths engulfing fibrous
illite, which in turn cover earlier overgrowths in petrofacies
B and C, suggests a recurrence of quartz precipitation during
mesodiagenesis. Feldspar albitisation and minor albite
overgrowths, as well as precipitation of anhydrite engulfing
late quartz outgrowths and illite coatings are probably
connected to deep mesogenetic fluids derived from the
overlying Carboniferous evaporites.
1067
9.7. Porosity preservation
Two major mechanisms of porosity reduction have
been recognised within the Uerê sandstone: (1) mechan-
ical and chemical compaction (2) mesogenetic quartz
cementation. The rate of porosity reduction due to quartz
cementation is a function of temperature and availability
of clean quartz grains surfaces for the growth of quartz
overgrowth cement (Walderhaug, 2000). Two mechan-
isms may have prevented a more extensive compaction
and quartz cementation in the studied reservoirs: (1)
early hydrocarbon emplacement (2) eogenetic micro-
quartz grain-rimming.
It is generally accepted that the displacement of aqueous
pore fluids by hydrocarbons can inhibit the progress of
diagenetic reactions. The overall lack of mesogenetic quartz
cementation of petrofacies A sandstones could be in part
attributed to early emplacement oil. However, the Uerê
sadstones are immediately overlain by the Jandiatuba oil-
generator shales, and their burial history at basin depocentre
shows that hydrocarbon had migrated to reservoirs during
deep (, 3400 m) burial (Fig. 15). Additionally, the similar
volumes of quartz cementation and porosity above and
below oil/water contacts suggest that hydrocarbon
migration has not halted diagenesis. Therefore, other factors
than early hydrocarbon emplacement probably control the
porosity preservation and distribution in the Uerê reservoirs.
There is a direct correspondence between high values of
porosity and permeability in the reservoirs, and the presence
of microcrystalline quartz rims (Fig. 16). The action of
microquartz rims in inhibiting the precipitation
of quartz overgrowths is known in a series of occurrences
(Aase et al., 1996; Bloch et al., 2002; Osborne & Swarbrick,
1999; Ramm, Forsberg, & Jahren, 1997). Additionally,
the presence microquartz rims helps to stabilize the
intergranular contacts, thus increasing the resistance of the
rock to pressure dissolution and preserving porosity.
Therefore, it is concluded that the major control on
porosity evolution in the Uerê sandstones was the eogenetic
silica cementation and its inhibitory role on quartz
cementation and pressure dissolution.
9.8. Exploration significance
The main mechanism of porosity preservation in the
Uerê sandstones is the inhibition of quartz overgrowth and
pressure dissolution by microquartz rims. Thus, the most
important factor to constrain reservoir quality in an oil
exploration approach is the distribution of this eogenetic
silica cementation.
Eogenetic silica cements (grain-rimming and pore-filling
microquartz and chalcedony) and associated smectitic clay
coatings occur specifically in sandstones interpreted as
storm deposits. These deposits were originally hybrid
arenites (Zuffa, 1980) rich in sponge spicules which, upon
dissolution of the spicules both within the arenites and in
1068 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Fig. 16. Plot of thin section macroporosity versus horizontal air
permeability, with symbols representing the amounts of microcrystalline
silica rims in sandstones from petrofacies A. Some samples devoid of
microquartz present relatively high macroporosity values due to the presence
of clay coatings, which preserved porosity from quartz overgrowths,
however, decreasing permeability.
interbedded spiculites (now spiculitic cherts), became
porous, microquartz-cemented sandstones. Therefore, the
areas with high frequency of storm reworking are prone to
have more levels of spicule-rich deposits, and thus of
derived porous microquartz cemented sandstones. The
growth of the siliciceous sponges biostromes is interpreted
to have occurred in nearshore settings, where reworking by
storms allowed the deposition of hybrid sands between the
fair-weather and storm wave base (i.e. facies association 2).
There is a direct relationship between the volume of
microquartz rims, of preserved porosity and of per-
meability values (Fig. 16). It must be considered, however,
that there is an optimum thickness for the microquartz rims
effectively preserve the porosity without seriously com-
promising the permeability (Aase et al., 1996; Bloch et al.,
2002; Ramm et al., 1997). Thick rims obstruct the pore
throats, and extremely thin rims are not efficient in
preventing overgrowths and pressure dissolution. In the
Uerê sandstones, the optimum rim thickness is between
5 and 10 mm, and the optimum volume is 4 – 6%, which is
apparently related to sands with less than 10 vol% of
original spicules contents. Hybrid sands extremely rich in
spicules were cemented by thick rims and/or pore-filling
microquartz and/or chalcedony, and typically did not
evolved into good reservoirs (Fig. 12).
10. Conclusions
The study of Upper Devonian Uerê sandstones in eastern
Solimões Basin yielded important clues to the depositional
and diagenetic controls on the preservation of porosity and
quality evolution of these shallow-marine reservoirs:
(1) The stratigraphy and facies associations suggest
deposition within an overall progradational regime,
followed by a major transgression. The storm reworking
of nearshore siliciceous sponge biostromes allowed the
deposition of spiculites and spicule-rich (hybrid) arenites
bellow the fair-weather wave base depths.
(2) The main diagenetic processes affecting the Uerê
sandstones are the authigenesis of various forms of silica,
particularly microquartz rim cements, which occur orig-
inally in spicule-rich, hybrid arenites, and mechanical and
chemical compaction.
(3) Three reservoir petrofacies were defined, based on the
packing, porosity, and types of cementation: (i) petrofacies
A: porous sandstones (. 15%) with microquartz rims; (ii)
petrofacies B: tight (, 10% porosity), strongly compacted,
moderately quartz-cemented (, 5%) sandstones; (iii) pet-
rofacies C: moderately porous (10 – 15%), conspicuously
quartz-cemented (. 5%) sandstones.
(4) Four major diagenetic evolution stages were
recognised: (1) marine eodiagenesis, dominated by the
dissolution of sponge spicules and precipitation of silica
(restricted to petrofacies A), and siderite/dolomite;
(2) meteoric eodiagenesis, responsible for heterogeneous
feldspar dissolution and kaolinite precipitation; (3) shallow
mesodiagenesis (1900 –2600 m depth), marked by hetero-
geneous chemical compaction and quartz overgrowth
cementation and (4) deep mesodiagenesis (. 2600 m
depth), marked by illite authigenesis and further quartz
precipitation, and probably connected to the convection of
hot fluids related with Triassic magmatism. Thermal
maturation of Devonian shales and hydrocarbon emplace-
ment in reservoirs took place at this time.
(5) Mouldic pores after spicules, which are widespread
within petrofacies A microquartz- and chalcedony-cemen-
ted sandstones, attest to a major redistribution of silica from
biogenic sources to eogenetic cements. A thermodynamic-
kinetic model explains the progressive diagenetic
silica transformation from the phase of highest entropy
(amorphous opal) to the phase of lowest entropy (quartz) in
a dissolution-reprecipitation pathway. The scarcity of opal-
CT remnants is probably related to the occurrence of
eogenetic smectitic clay coatings, which accelerated
the opal-CT/quartz transformation and/or competitively
adsorbed silica, thus inhibiting opal-CT nucleation.
(6) The distribution of mesogenetic quartz overgrowth
cementation is widespread but extremely heterogeneous
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
from thin section to layer scale. This is attributed to
inhibition of overgrowth by the presence of eogenetic
microcrystalline quartz rims, and to the convection of hot
fluids connected to Triassic magmatism, which co-precipi-
tated late quartz and illite.
(7) The major mechanism of porosity preservation is the
inhibition of quartz overgrowth cementation and pressure
dissolution by eogenetic microquartz rims. The distribution of
the eogenetic silica cements (grain-rimming and pore-filling
microquartz, chalcedony) and associated clay coatings is
directly related to storm layers rich in sponge spicules.
Acknowledgements
We gratefully thank PETROBRAS, in special Humberto
Pampolha Lima and R. Nonato M. Cunha for access to
samples, data, information, resources, and for the license to
publish this work. Special acknowledgements are paid to the
support of the Brazilian National Petroleum Agency—ANP
(grant and research funds to R.D. Lima) and the National
Research Council—CNPq (grant to L.F. De Ros). We
acknowledge the use of the support and analytical facilities
of the Institute of Geosciences of Rio Grande do Sul Federal
University. We thank Dr S. Morad, from Uppsala University,
for his revision on a previous version of the manuscript.
References
Aase, N. E., Bjørkum, P. A., & Nadeau, P. H. (1996). The effect of grain-
coating microquartz on preservation of reservoir porosity. AAPG
Bulletin, 80, 1654–1673.
Astin, T. R. (1987). Petrology (including fluorescence microscopy) of
cherts from the Portlandian of Wiltshire, UK—evidence of an episode
of meteoric water circulation. In J. D. Marshall (Ed.), Diagenesis of
sedimentary sequences. Geological Society Special Publication 36,
Oxford: Blackwell, pp. 73–84.
Beard, D. C., & Weyl, P. K. (1973). Influence of texture on porosity and
permeability of unconsolidated sand. AAPG Bulletin, 57, 349 –369.
Bloch, S., Lander, R. H., & Bonell, L. (2002). Anomalously high porosity
and permeability in deeply buried sandstones reservoirs: Origin and
predictability. AAPG Bulletin, 86, 301–328.
Boles, J. R., & Franks, S. G. (1979). Clay diagenesis in Wilcox sandstones
of southwest Texas: Implications of smectite diagenesis on sandstone
cementation. Journal of Sedimentary Petroleum, 49, 55 –70.
Caplan, M. L., & Bustin, R. M. (1996). Factors governing organic matter
preservation potential in marine petroleum source rocks from
palaeocontinental margins: Evidence from Upper Devonian to Lower
Carboniferous Exshaw Formation. Canadian Petroleum Geology
Bulletin, 44, 474 –494.
Caplan, M. L., & Bustin, R. M. (1999). Devonian-Carboniferous
Hangenberg mass extinction event, widespread organic-rich mudrock
and anoxia: Causes and consequences. Palaeogeology, Palaeoclimatol-
ogy, Palaeoecology, 148, 187 –207.
Caputo, M. V., & Crowell, J. C. (1985). Migration of glacial centers across
Gondwana during the Paleozoic Era. Geological Society of American
Bulletin, 96, 1020–1036.
Caputo, M. V., & Silva, O. B. (1990). Sedimentação e tectônica da Bacia do
Solimões. In G. P. Raja Gabaglia, & E. J. Milani (Eds.), Origem e
1069
Evolução das Bacias Sedimentares (pp. 169 –193). Rio de Janeiro:
PETROBRAS.
Chaika, C., & Williams, L. A. (2000). Density and mineralogy variations as
a function of porosity in Miocene Monterey Formation oil and gas
reservoirs in California. AAPG Bulletin, 85, 149 –167.
Chang, J., & Yortsos, Y. C. (1994). Lamination during silica diagenesis-
effects of clay content and Ostwald ripening. American Journal of
Science, 294, 137–172.
Craig, H. (1957). Isotopic standards for carbon and oxygen correction
factors for mass spectrometric analysis of carbon dioxide. Geochimica
Cosmochimica Acta, 12, 133 –149.
Curtis, C. D., & Coleman, M. L. (1986). Controls on the precipitation of
early diagenetic calcite, dolomite and siderite concretions in complex
depositional sequences. In D. L. Gautier (Ed.), Roles of organic matter
in sediment diagenesis (pp. 23–33). SEPM Special Publication 38,
Society of Economic Paleontologists and Mineralogists.
Deer, W. A., Howie, R. A., & Zussman, J. (1962) (Vol. 5). Non-silicates.
Rock-forming minerals, London: Longman, p. 371.
De Ros, L. F., Morad, S., Broman, C., Césero, P., & Gomez-Gras, D.
(2000). Influence of uplift and magmatism on distribution of quartz and
illite cementation: Evidence from Siluro-Devonian sandstones of the
Paraná Basin, Brazil. In R. Worden, & S. Morad (Eds.), Quartz
cementation in sandstones (26) (pp. 231 –252). Special Publications of
International Association of Sedimentologists, Oxford: Blackwell.
Dickson, J. A. D. (1965). A modified staining technique for carbonates in
thin section. Nature, 205, 587.
Dove, P. M., & Crerar, D. A. (1990). Kinetics of quartz dissolution in
electrolyte solutions using a hydrothermal mixed flow reactor.
Geochimica Cosmochimica Acta, 54, 1609–1625.
Downing, K. P., & Walker, R. G. (1988). Viking formation, Joffre Field,
Alberta: Shoreface origin of long, narrow sand body encased in marine
mudstones. AAPG Bulletin, 72, 1212–1228.
Dworkin, S. I., & Land, L. S. (1994). Petrographic and geochemical
constraints on the formation and diagenesis of anhydrite cements,
Smackover sandstones, Gulf of Mexico. Journal of Sedimentary
Research, A64, 339–348.
Ehrenberg, S. N., & Nadeau, P. H. (1989). Formation of diagenetic illite in
sandstones of the Garn Formation, Haltenbanken area, mid-Norwegian
continental shelf. Clay and Minerals, 24, 233–253.
Eiras, J. F (1998). Geology and petroleum system of the Solimões Basin,
Brazil . 1998 AAPG International Conference and Exibition (pp. 446).
Rio de Janeiro.
Eiras, J. F., Becker, C. R., Souza, E. M., Gonzaga, F. G., Silva, J. G. F.,
Daniel, L. M. F., Matsuda, N. S., & Feijó, F. J. (1994). Bacia do
Solimões. Boletim de Geociências da PETROBRAS, 8, 17–45.
Folk, R. L. (1968). Petrology of sedimentary rocks. Austin, TX: Hemphill,
107 p.
Franks, S. G., & Forester, R. W. (1984). Relationships among secondary
porosity, pore fluid chemistry and carbon dioxide, Texas Gulf Coast. In
D. A. McDonald, & R. C. Surdam (Eds.), Clastic diagenesis. AAPG
Memoir 37, American Association of Petroleum Geologists.
Friedman, I., & O’Neil, J. R. (1977). Compilation of stable isotopic
fractionation factors of geochemical interest. In M. Fleischer (Ed.),
Data of geochemistry (pp. 12USGS Professional Paper 440-KK, United
States Geological Survey.
Giles, M. R., & Marshall, J. D. (1986). Constraints on the development of
secondary porosity in the subsurface: Re-evaluation of processes.
Marine and Petroleum Geology, 3, 243 –255.
Girard, J.-P., Deynoux, M., & Nahon, D. (1989). Diagenesis of the Upper
Proterozoic siliciclastic sediments of the Taoudeni Basin (West Africa)
and relation to diabase emplacement. Journal of Sedimentary
Petroleum, 59, 233 –248.
Gluyas, J. G., Grant, S. M., & Robinson, A. G. (1993). Geochemical
evidence for a temporal control on sandstone cementation. In A.
Horbury, & A. Robinson (Eds.), Diagenesis and basin development
(pp. 23–33). AAPG Studies in Geology, The American Association of
Petroleum Geologists.
1070 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Gluyas, J., Jolley, L., & Primmer, T. J. (1997). Element mobility during
diagenesis; sulfate cementation of Rotliegent sandstones, southern
North Sea. Marine and Petroleum Geology, 14, 1001–1011.
Grahn, Y. (1992). Revision of Silurian and Devonian strata of Brazil.
Palinology, 16, 35 –61.
Hendry, J. P., & Trewin, N. H. (1995). Authigenic quartz microfabrics in
Cretaceous turbidites: Evidence for silica transformation processes in
sandstones. Journal of Sedimentary Research, A65, 380 –392.
Houseknecht, D. W. (1987). Assessing the relative importance of
compaction processes and cementation to reduction of porosity in
sandstones. AAPG Bulletin, 71, 633–642.
Huggett, J., Dennis, P., & Gale, A. (2000). Geochemistry of early siderite
cements from the Eocene succession of Whiteclif Bay, Hampshire
Basin, UK. Journal of Sedimentary Research, 70, 1107–1117.
Hunt, D., & Tucker, M. E. (1992). Stranded parasequences and the forced
regressive wedge systems tract: Deposition during base-level fall.
Sedimentary Geology, 81, 1– 9.
Jahren, J., & Ramm, M. (2000). The porosity-preserving effects of
microcrystalline quartz coatings in arenitic sandstones: Examples from
the Norwegian continental shelf. In R. H. Worden, & S. Morad (Eds.),
Quartz cementation in sandstones (29) (pp. 271 – 280). Special
Publications of International Association of Sedimentologists, Oxford:
Blackwell.
Johnson, J. G., Klapper, G., & Sandberg, C. A. (1985). Devonian eustatic
fluctuations in Euramerica. Geological Society of American Bulletin,
96, 567 –587.
Kahn, J. S. (1956). The analysis and distribution of the properties of
packing in sand-size sediments. 1. On the measurement of packing in
sandstones. Journal of Geology, 64, 385–395.
Kastner, M., Keene, J. B., & Gieskes, J. M. (1977). Diagenesis of siliceous
oozes. I. Chemical controls on the rate of opal-A to opal-CT
transformation—an experimental study. Geochimica Cosmochimica
Acta, 41, 1041–1059.
Knauth, L. P. (1994). Petrogenesis of chert. In P. J. Heaney, C. T. Prewitt, &
G. V. Gibbs (Eds.), Silica: physical behavior, geochemistry and
materials applications (pp. 233–258). Reviews in Mineralogy, Miner-
alogica Society of America.
Loboziak, S., Melo, J. H. G., Quadros, L. P., Daemon, R. F., & Barrilari,
I. M. R (1994). Devonian-Dinantian Miospore Biostratigraphy of the
Solimões and Parnaı́ba Basins (with considerations on the Devonian of
the Parná Basin). Internal report. PETROBRAS/CENPES/DIVEX/
SEBIPE.
Lundegard, P. D. (1992). Sandstone porosity loss—a big picture view of the
importance of compaction. Journal of Sedimentary Petroleum, 62,
250– 260.
MacEachern, J. A., & Pemberton, S. G. (1992). Ichnological aspects of
Cretaceous shoreface succession and shoreface variability in the
Western Interior Seaway of North America. In S. G. Pemberton (Ed.),
Applications of ichnology to petroleum exploration, a core workshop
(pp. 57–84). SEPM, Core Workshop, Society of Economic Paleontol-
ogists and Mineralogists.
MacEachern, J. A., Zaitlin, B. A., & Pemberton, S. G. (1999). A sharp-
based sandstone of the Viking Formation. Journal of Sedimentary
Research, 69, 876 –892.
Maliva, R. G., & Siever, R. (1988). Pre-Cenozoic nodular cherts: Evidence
for Opal-CT precursor and direct quartz replacement. American Journal
of Science, 288, 798 –809.
McBride, E. F. (1989). Quartz cement in sandstones: A review. Earth
Science Reviews, 26, 69–112.
McDowell, S. D., & Paces, J. B. (1985). Carbonate alteration minerals in
the Salton Sea geothermal system, California, USA. Minerals
Magazine, 49(352), 469–479.
Mello, M. R., Koutsoukos, E. A. M., Mohriak, W. U., & Bacoccoli, G.
(1994). Selected petroleum systems in Brazil. In L. B. Magoon, &
W. G. Dow (Eds.), The petroleum system—from source to trap (pp.
499– 512). AAPG Memoir 60, The American Association of Petroleum
Geologists.
Michalopoulos, P., & Aller, R. (1995). Rapid clay mineral formation in
Amazon Delta sediments—reverse weathering and oceanic elemental
cycles. Science, 270, 614–617.
Michalopoulos, P., Aller, R. C., & Reeder, R. J. (2000). Conversion of
diatoms to clay during early diagenesis in tropical, continental shelf
muds. Geology, 28, 1095– 1098.
Mizusaki, A. M. P., Anjos, S. M. C., Wanderley, J. W., Filho Silva, O. B.,
Costa, M. G. F., Lima, M. P., & Kawashita, K. (1990). Datação K/Ar de
ilitas diagenéticas. Boletim de Geociências da PETROBRAS, 4, 237–252.
Morad, S. (1986). Albitization of K-feldspar grains in Proterozoic arkoses
and greywackes from southern Sweden. Neues Jahrbuch for Miner-
alogie Monatshefte, 4, 145– 156.
Morad, S. (1998). Carbonate cementation in sandstones: Distribution
patterns and geochemical evolution. In S. Morad (Ed.), Carbonate
cementation in sandstones (26) (pp. 1–26). Special Publications of
International Association of Sedimentologists, Oxford: Blackwell.
Morad, S., Bergan, M., Knarud, R., & Nystuen, J. P. (1990). Albitization of
detrital plagioclase in Triassic reservoir sandstones from the Snorre Field,
Norwegian North Sea. Journal of Sedimentary Petroleum, 60, 411 –425.
Mozley, P. S. (1989). Relation between depositional environment and the
elemental composition of early diagenetic siderite. Geology, 17, 704–706.
Osborne, M. J., & Swarbrick, R. E. (1999). Diagenesis in North Sea HPHT
clastic reservoirs—consequences for porosity and overpressure pre-
dicition. Marine and Petroleum Geology, 16, 337–353.
Parrish, J. T. (1982). Upwelling and petroleum source beds, with reference
to the Paleozoic. AAPG Bulletin, 66, 750– 774.
Pitman, J. K., Henry, M., & Seyler, B. (1998). Reservoir quality and
diagenetic evolution of Upper Mississippian rocks in the Illinois Basin:
influence of a regional hydrothermal fluid flow event during late
diagenesis. Geological Survey Professional Paper, 1597, Washington:
United States Government Printing Office, p. 24.
Pittman, E. D., Larese, R. E., & Heald, M. T. (1992). Clay coats:
Occurrence and relevance to preservation of porosity in sandstones. In
D. W. Houseknecht, & E. D. Pittman (Eds.), Origin, diagenesis, and
petrophysics of clay minerals in sandstones (pp. 241–264). SEPM
Special Publication 47, Society of Economic Paleontologists and
Mineralogists.
Posamentier, H. W. (2002). Ancient shelf ridges—a potentially significant
component of the transgressive systems tract: Case study from offshore
northwest Java. AAPG Bulletin, 86, 76–106.
Posamentier, H. W., Allen, G. P., James, D. P., & Tesson, M. (1992). Forced
regressions in a sequence stratigraphic framework: Concepts, examples,
and exploration significance. AAPG Bulletin, 76, 1687–1709.
Ramm, M., Forsberg, A. W., & Jahren, J. S. (1997). Porosity-depth trends in
deeply buried Upper Jurassic reservoirs in the Norwegian Central
Graben: An example of porosity preservation beneath the normal
economic basement by grain-coating microquartz. In J. A. Kupecz, J. G.
Gluyas, & S. Bloch (Eds.), Reservoir quality prediction in sandstones
and carbonates (pp. 177 –200). AAPG Memoir 69, The American
Association of Petroleum Geologists.
Rosenbaum, J. M., & Sheppard, S. M. F. (1986). An isotopic study of
siderites, dolomites and ankerites at high temperatures. Geochimica
Cosmochimica Acta, 50, 1147–1150.
Savoy, L. (1992). Environmental record of Devonian-Mississippian
carbonate and low-oxygen facies transitions, southernmost Canadian
rocky mountains and northernmost Montana. Geological Society of
American Bulletin, 104, 1412–1432.
Schiffman, P., Bird, D. K., & Elders, W. A. (1985). Hydrothermal
mineralogy of calcareous sandstones from the Colorado River delta in
the Cerro Prieto geothermal system, Baja California, Mexico. Minerals
Magazine, 49(Part 3), 435 –449.
Schmidt, V., & McDonald, D. A. (1979). The role of secondary porosity in
the course of sandstone diagenesis. In P. A. Scholle, & P. R. Schluger
(Eds.), Aspects of diagenesis (pp. 175 –207). SEPM Special Publi-
cation 29, Society of Economic Paleontologists and Mineralogists.
Searl, A. (1994). Diagenetic destruction of reservoir potential in shallow
marine sandstones of the Broadford Beds (Lower Jurassic), north-west
 R.D. Lima, L.F. De Ros / Marine and Petroleum Geology 19 (2002) 1047–1071
Scotland: Depositional versus burial and thermal history controls on
porosity destruction. Marine and Petroleum Geology, 11, 131–147.
Siever, R., & Woodford, N. (1973). Sorption of silica by clay minerals.
Geochimica Cosmochimica Acta, 37, 1851–1880.
Silva, O. B (1987). Análise da Bacia do Solimões (revisão litoestrati-
gráfica, magmatismo e geoquı́mica). MSc Thesis, Universidade Federal
de Ouro Preto, 158 pp.
Snedden, J. W., & Nummedal, D. (1991). Origin and geometry of storm-
deposited beds in modern sediments of the Texas continental shelf.
In D. J. P. Swift, G. F. Oertel, R. W. Tillman, & J. A. Thorne (Eds.),
Shelf sand and sandstone bodies: geometry, facies, and sequence
stratigraphy (14) (pp. 283–308). Special Publications of International
Association of Sedimentologists, Oxford: Blackwell.
Streel, M., Caputo, M. V., Loboziak, S., & Melo, J. H. G. (2000). Late Frasnian-
Famennian climates based on palynomorphs analysis and the question of
the Late Devonian glaciation. Earth Science Reviews, 52, 121–172.
Summer, N. S., & Verosub, K. L. (1992). Diagenesis and organic maturation
of sedimentary rocks under volcanic strata, Oregon. AAPG Bulletin, 76,
1190–1199.
Surdam, R. C., Boese, S. W., & Crossey, L. J. (1984). The chemistry of
secondary porosity. In R. C. Surdam, & D. A. McDonald (Eds.), Clastic
diagenesis (pp. 127 –149). AAPG Memoir 37, American Association of
Petroleum Geologists.
1071
Szatmari, P (1996). Datação 40Ar/39Ar do vulcanismo Mesozóico nas
bacias do Solimões e Amazonas. Internal Report. PETROBRAS/
CENPES/DIVEX /SEMBA.
Walderhaug, O. (2000). Modeling quartz cementation and porosity in
Middle Jurassic Brent Group sandstones of the Kvitebjorn Field,
Northern North Sea. AAPG Bulletin, 84, 1325–1339.
Walker, R. G., & Bergman, K. M. (1993). Shannon Sandstone in Wyoming:
A shelf-ridge complex reinterpreted as lowstand shoreface deposits.
Journal of Sedimentary Petroleum, 63, 839 –851.
Walker, R. G., & Wiseman, T. (1995). Lowstand shorefaces, transgressive
incised shorefaces, and forced regressions: Examples from the Viking
Formation, Joarcam area, Alberta. Journal of Sedimentary Research, 1,
132– 142.
Williams, A. W., & Crerar, D. A. (1985). Silica diagenesis. II. General
mechanisms. Journal of Sedimentary Petroleum, 55, 312–321.
Williams, A. W., Parks, G. A., & Crerar, D. A. (1985). Silica
diagenesis. I. Solubility controls. Journal of Sedimentary Petroleum,
55, 301–311.
Wollast, R., & de Broeu, F. (1971). Study of the behavior of dissolved silica
in the estuary of the Scheldt. Geochimica Cosmochimica Acta, 35,
613– 620.
Zuffa, G. G. (1980). Hybrid arenites: Their composition and classification.
Journal of Sedimentary Petroleum, 50, 21 –29.