1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

The Cabo de la Vela Mac Ultramac Complex, Northeastern Colombian Caribbean region: a record of Multistage evolution of a late Cretaceous intra-oceanic arc
M. B. I. WEBER1*, A. CARDONA2,3, F. PANIAGUA1, U. CORDANI4, LVEDA1 & R. WILSON4 L. SEPU
1

Universidad Nacional de Colombia, Calle 65 No. 78-28, Facultad de Minas, n, Colombia M1-324, Medell

n, Panama ECOPETROL, Smithsonian Tropical Research Institute, Balboa, Anco Piedecuesta, Colombia
3

ria, Institute of Geoscience, USP, Rua do Lago 562, Cidade Universita o Paulo, Brazil 05508-080 Sa *Corresponding author (e-mail: mweber@unalmed.edu.co)

University of Leicester, Department of Geology, University Road, Leicester LE1 7RH, UK

Abstract: Ophiolite-related rocks accreted to Caribbean plate margins provide insights into complex intra-oceanic evolution of the plate and its interaction with continental margins of the Americas. Petrologic, geochemical and isotope (K Ar, Sr and Nd) data were obtained in poorly known serpentinites, gabbros and andesite dykes of the Cabo de la Vela MacUltramac Complex from the Guajira Peninsula, north Colombia. Field relations, metasomatic alteration patterns and whole rock mineral geochemistry combined with juvenile isotope signatures of the different units suggest that gabbros and serpentinites formed in a slow-spreading supra-subduction zone that was brought to shallower depths and subsequently evolved to an arc setting where andesitic rocks formed with little sediment input. The tectonomagmatic evolution of the Cabo de la Vela Mac Ultramac Complex involved an intra-oceanic arc that evolved from pre-Campanian time to 74 Ma. Relationships with other units from the Guajira Peninsula show either the existence of a mature arc basement or a series of coalesced allocthonous arcs, juxtaposed before accretion onto the passive continental margin of South American in pre-Eocene times. Correlations with ophiolite-related rocks of the Southern Caribbean and diachronism in the accretionary process are compatible with a WSWESE advancing Caribbean plate front.

Ophiolitic rocks are fundamental to understanding of the complex dynamics and evolution of oceanic crust and its interaction with continental margins to form so-called Cordilleran-type orogens (Shervais 2001; Beccaluva et al. 2004). In the Caribbean area, studies of ophiolitic complexes have identied remnants of rift margins, ocean plateaux and arc systems formed in oceanic domains that evolved since the Mesozoic, following the break-up of Pangea (Giunta et al. 2002, 2006). However, different models show lack of concensus. Testing of regional models through local relationships has often not been successful, possibly due to lack of detailed information (Iturralde-Vinent & Lidiak 2006 for discusions). A few ophiolite-related rocks have been described from the Colombian Caribbean (MacDonald 1964; Lockwood 1965; Alvarez 1967); however, their formation setting and their

implications for the geotectonic models of the Caribbean have not been discussed. We present integrated petrological, mineral chemistry, whole rock geochemistry, Nd and Sr isotopes and K Ar geochronology data from the Cabo de la Vela Mac Ultramac Complex of the southwestern Caribbean. They allow reconstruction of a Late Cretaceous, arc-related magmatic system that was accreted to the continental margin of South America in pre-Eocene times. The implication of this and other available regional data is discussed within the framework of the Caribbean plate tectonic evolution.

Geological setting
The Meso-Cenozoic history of the Colombian Andes in the Caribbean region is characterized by interaction between the northwestern border of

From: JAMES , K. H., LORENTE , M. A. & PINDELL , J. L. (eds) The Origin and Evolution of the Caribbean Plate. Geological Society, London, Special Publications, 328, 547566. DOI: 10.1144/SP328.22 0305-8719/09/$15.00 # The Geological Society of London 2009.

548

M. B. I. WEBER ET AL.

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116

South America and the Caribbean and Nazca plates (Fig. 1). Multiple plate boundaries and MesoCenozoic transpressive tectonics controlled accretion of oceanic terranes to the continental margin. Discrete tectonic blocks record this complex evolution (Pindell 1993; Toussaint 1996; Montes et al. 2005). The Guajira Peninsula, in northeasternmost Colombia (Fig. 2), is characterized by several isolated massifs with correlatable geology, surrounded by broader at lands and Cenozoic basins (MacDonald 1964; Lockwood 1965; Alvarez 1967). Within these massifs at least three main lithotectonic belts can be identied. From oldest to most recent they include the following elements (Fig. 2): 1. A composite, Late Mesoproterozoic and Palaeozoic metamorphic domain, which includes medium and high-grade units. This is intruded by Jurassic magmatism, similar to the parautochthonous basement of the Andes (Cordani et al. 2005; Cardona-Molina et al. (2006). 2. A weakly deformed belt of Mesozoic sedimentary rocks with the same depositional

characteristics and ages as the autochthonous South American margin (Villamil 1999). 3. A sequence of two different, Cretaceous, low-grade meta-volcano-sedimentary metamorphic units. The northernmost unit has intercalated mac and ultramac plutonic rocks, intruded by Eocene magmatism (MacDonald 1964; Lockwood 1965; Alvarez 1967; Pindell 1993). Fragments of high-pressure metamac and metasedimetary rocks in Miocene conglomerates on the northwestern fringe of the central massif (Fig. 2) may represent a probably mid-Cretaceous exhumed subduction-accretion complex (Green et al. 1968; Zapata et al. 2005). An association of mac and ultramac rocks occurs in the Cabo de la Vela region, isolated from the main massifs and close to the zone of CaribbeanSouth American plate interaction. The following sections describe the geological and geochemical characteristics of this mac ultramac complex and discuss their signicance to interpretations of tectonomagmatic evolution.

NORTH AMERICAN PLATE

Mexico

CVC

Colombian Guajira COCOS PLATE

Colombia Galpagos SOUTH AMERICAN PLATE NAZCA PLATE 400 km

Fig. 1. Tectonic Framework and location of the Caribbean region. The Guajira Peninsula and the CVC are shown. The distribuiton of ophiolite rocks within the Caribbean region is included (data from: Lewis et al. 2006; Guinta et al. 2002).

THE CABO DE LA VELA MAFIC COMPLEX

549

117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174

Fig. 2. Simplied Geological map of the northern Guajira Peninsula, showing major litostratigraphic units.

Cabo de la Vela Mac Ultramac Complex

The Cabo de la Vela consists of a series of isolated small hills (150 m) exposing different mac and ultramac units (Alvarez 1967). In general they include a sequence of serpentinites, gabbros and mac volcanic units. The serpentinite ultramac unit contains gabbro and leucograbbros as lenses and pods, cross-cut by basalt dykes (Fig. 3). Miocene sediments locally overlie this sequence (Alvarez 1967). To the SE, a lower and at-land region of Cenozoic sediments extends for 5 km, separating the Cabo de la Vela from the massifs of the Guajira region (MacDonald 1964; Lockwood 1965; Alvarez 1967; Fig. 2). Similarities between the plutonic mac and ultramac rocks of the Cabo de la Vela region and the Cretaceous units of the massifs suggest a geological link between them (review in Alvarez 1967). The Bouguer gravity map of northern Colombia (Fig. 4) shows a positive anomaly centred on the Cabo de la Vela, extending 100 km offshore to the northwest and 30 km onshore to the southeast

(Kellogg et al. 1991). This anomaly is consistent with a NWSE elongated dense body, with a narrow and gently sloping northwestern face and a broad and steep southeastern ank, and suggests wide extension of this mac ultramac rock complex (Nieto & Ojeda, pers. comm., 2006). The following section introduces the informal stratigraphic term Cabo de la Vela Mac Ultramac Complex (CVC) and describes the eld and petrographic elements of the constituent units.

Ultramac rocks
Ultramac rocks of the Cabo de la Vela area are mainly serpentinites. The only exposure of a partially serpentinized rock (sample FP-34C) shows meso-scale foliation, with original grains of olivine, spinel and clinopyroxene. Bastite minerals suggest the former presence of another pyroxene (orthopyroxene?), which indicates that the original ultramac rock was a wherlite. Relict textures in this rock are characterized by larger grains surrounded by ne-grained material and by deformed

550

M. B. I. WEBER ET AL.
y = 1,846.000

175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232

x = 880.000

Cenozoic cover Tertiary sedimentary rocks Cabo de la Vela Complex (CVC) Basaltic Dykes (exaggerated scale)
m 75

Caribbean Sea

CABO DE LA VELA

N
NW NE

SW

SE

SCALE
600 0 600 1200 km

Fig. 3. Geology of the Cabo the Vela area.

73W
50

72W
0

50

50

100

12N
0

0
50

50

Fig. 4. Bouguer gravity map of the Guajira area (Kellogg et al. 1991).

THE CABO DE LA VELA MAFIC COMPLEX

551

233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290

cleavages, indicating that the ultramac protolith was a mantle tectonite (Fig. 5) (Nicolas & Rabinowicz 1984). The serpentinites are mainly greenish to brownish rocks, in many cases mottled by the silkylooking bastite minerals. Microscopic analysis shows pseudomorph textures, such as antigorite

mesh and lizardite antigorite hourglass. Nonpseudomorphic textures include interpenetrating and intergrowth textures, both in the same sample and both dened by antigorite. Other local minerals formed during serpentinzation are bastite, magnetite, magnesite and brucite. Relict minerals are brown spinel and occasional amphibole. The

Colour online= colour hardcopy

Q4

Fig. 5. (a) Gabbro intrusions in serpentinite. (b) Continuous basaltic andesite dyke within serpentinite. The arrow indicates the direction of the dyke. (c) Serpentinised wehrlite (sample FP-34C) with relict olivine (Ol). Note the presence of brown bastite (Bst), as replacement of pyroxenes. (d) Gabbro tectonite. Note the brown amphibole inclusions in the central pyroxene (Cpx) crystal, and secondary amphibole on the recrystallised pyroxene grain margins (Hbl). The two upper pyroxenes and plagioclase (plg) have been completely recrystallised and have a granoblastic polygonal texture. (e) Basaltic Andesite with twinned pyroxene (Cpx) phenocryst in a hornblende (Anf) and saussuritisized plagioclase (Sauss) matrix.

552

M. B. I. WEBER ET AL.

291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348

presence of amphibole indicates a hydrated mantle protolith. Bastite may replace clinopyroxene, with opaques dening relict bend cleavages, and maybe orthopyroxene without opaques. The presence of two pyroxenes and the foliated texture suggests that the protolith was also a wehrlitic tectonite. In addition, there are at least four generations of veining, formed by either perpendicular or parallel to the wall opening precipitation of crysotile.

Gabbros and hornblendites

Gabbroic rocks occur in the ultramates as small, discontinuous, irregular bodies, pods and lenses with maximum dimensions of 1 m 100 m. They are coarse-grained to pegmatitic, and contain dark green pyroxene (diopside) and plagioclase (labradorite) crystals up to 15 cm in size. These rocks generally show well-dened mineral banding and lineation. Most show evidence of deformation and in thin section the overall texture is granoblastic polygonal due to deformation recrystallization of large crystals to smaller polygonal crystals at the edges (Fig. 5). The various degrees of recrystallization always involve more plagioclase than pyroxene, which is sometimes preserved in large crystals. These relict grains have numerous oriented inclusions of small brown amphibole blebs that impart a characteristic lustre to hand-specimen. The euhedral shape of the blebs, their colour and the fact that they formed before high temperature deformation all indicate magmatic origin (Coogan et al. 2001). Evidence of deformation of the larger grains before recrystallization is conspicuous, marked by bent and kinked cleavages and by alignment of the small amphibole blebs that disappear with recrystallization. The fact that pyroxene and plagioclase ductile deformation and recrystallization annealed with granoblastic textures, without changes in the original mineralogy assemblage, is indicative of deformation at granuliteamphbolite facies conditions (Seyler et al. 1998). Hornblendites are found generally on the edges of the gabbro dykes and sometimes as small metresized patches within the serpentinites. They are made up exclusively of coarse-grained dark brown and less commonly bright green amphibole. Field observations show that these rocks formed by complete replacement of pyroxene and plagioclase of the gabbros and possibly the ultramacs. Several stages of replacement are present. Early stages are seen as coronas around pyroxenes but at more advanced stages the reaction front replaces all of the minerals (Fig. 5). Petrographically these rocks show decussate textures and the nature of the replacement of these hornblendites indicates that they

formed through metasomatism during annealing and therefore after deformation. Gabbros and hornblendites have been rodingitized throughout the whole complex, and transitional rocks are common. Massive rodingites comprise Ca-rich minerals such as chlorite + hydrogrossular + vesuvianite + epidote + albite + tremolite-actinolite + prehnite. Some rodingites are zoned, with a chlorite margin, in which the relicts of the original pyroxene can still be identied. The mineral assemblages suggest that these rocks formed at greenschist facies and prehnite ska 1995; pumpellyte facies conditions (Dubin h-Green et al. 1996). Frost 1975; Fru The transition in facies, as well as different styles of complex superimposed deformation and fracturing during uid penetration, indicate that rodingitization occurred from ductile to brittle conditions.

Mac volcanic dykes

Dykes are continuous and undeformed and reach up to 2 m in width, cutting the previously described lithologies. They constitute ne-grained, phaneritic to aphanitic, green to grey-green rocks, generally with porphyritic texture. Plagioclase (andesine) and clinopyroxene (diopside) with intergranular to subophiitic texture are seen in thin section (Fig. 5). Subhedral crystals of clinopyroxene and plagioclase comprise the phenocrystal phase. Several of these rocks also contain brown amphibole as a matrix mineral and as uralitic replacement of pyroxene. The more felsic rocks contain quartz in the matrix. All minerals are generally altered to saussurite and chlorite and pumpellyte is common. The suggested crystallization order would be pyroxene and plagioclase opaquesamphibole quartz, and petrographically may resemble basalts or andesites.

Analytical techniques Whole rock geochemistry

Eleven samples were analysed for major and trace elements by X-ray uorescence (XRF) and inductively coupled mass spectrometry (ICP-MS) at the Mineralogy and Petrology Department of the Institute of Geosciences of the University of Sao Paulo and at ACME analytical laboratories in Canada. Samples were crushed with an iron steel crusher and pulverised in an agate mill. Sample preparation for XRF included microreduction to obtain pressed powder pellets, and fused glass discs for major and trace element determination. Major and selected

THE CABO DE LA VELA MAFIC COMPLEX

553

349 trace element analysis were carried out in a 350 wavelength dispersive Philips PW 2400 XRF spec351 trometer with detection limits generally of the order 352 of 110 ppm for trace elements, following the 353 methodology described by Mori et al. (1999). For 354 other trace elements including rare earths (REE), 355 the analyses were carried out by ICP-MS at 356 ACME analytical laboratories (Canada), after 357 lithium metraborate/tetraborate fusion and nitric 358 acid digestion of a 0.2 g sample. 359 Mineral chemistry 360 361 Mineral analyses were obtained from carbon-coated 362 polished thin sections using a Jeol 8600S electron 363 microprobe at the Department of Geology, Univer364 sity of Leicester, UK, using an accelerating 365 voltage of 15 kV and a probe current of 30 nA 366 with a beam diameter of 5 10 mm. Quantitative 367 background-corrected results were standardized 368 against a combination of synthetic materials and 369 well-characterized natural minerals and corrected 370 for matrix effects using a ZAF correction procedure. 371 Minimum detection limits under the analytical 372 conditions used range from 0.01 wt% for Na2O 373 to 0.04 wt% for FeO. 374 375 K Ar geochronology 376 377 Three whole rock samples were analysed by the 378 K Ar method at the Centre of Geochronological 379 o Paulo Research of the University of Sa 380 (CPGeo-USP). Two aliquots from the same 381 sample were separated for the K and Ar analysis. 382 Potassium analyses of each pulverized sample 383 were carried out in duplicate, coupled to an ultra384 vacuum system. A spike of 38Ar was added and 385 the gas was puried in titanium and copper ovens. 386 Final argon determinations were carried out in a 387 Reynold-type gas spectrometer. Analytical pre388 cision for K is of 5% whereas for Ar it is around 389 0.5%. Decay constants for calculation are after 390 Steiger & Jager (1977). 391 392 Nd Sr isotopes 393 394 Six whole rock samples were analysed by SmNd 395 and Rb Sr methods at the Centre for Geochronolo o Paulo 396 gical Research of the University of Sa 397 (CPGeo-USP). For the Sm Nd method the 398 Q1 analytical procedures followed Sato et al. (1995). 399 Isotopic ratios 143Nd/144Nd were obtained in a 400 multi-collector mass spectrometer Finnegan Mat, 401 with analytical precision of 0.0014% (2s). Exper402 imental error for the 147Sm 144Nd ratios is of 403 the order of 0.1%. La Jolla and BCR-1 standards 404 yielded 143Nd/144Nd 0.511849 + 0.000025 (1s) 405 and 0.512662 + 0.000027 (1s) respectively 406 during the period in which the analyses were

performed. The 1Nd were calculated following De Paolo (1988), and the constants used include 143Nd/144Nd (CHUR) 0.512638 and 147 Sm 144Nd (CHUR)0 0.1967. Rb Sr analyses followed procedures presented by Tassinari et al. (1996). Rb and Sr values were obtained by X-ray uorescence, and 87Sr/86Sr ratios were done with mass spectrometer VG-sector mass spectrometer and corrected for isotopic fractionation during thermal ionization with a 87 Sr/86Sr 01194.

Geochemistry
Whole rock geochemistry and mineral chemistry data were obtained from selected samples in order to understand its tectonic setting of the CVC.

Q2

Mineral chemistry
Four samples, a partially serpentinized a peridotite, a serpentinite, a gabbro and a basalt, were selected for mineral chemical analysis. Analysed minerals were spinel, olivine, pyroxene, plagioclase and amphibole.

Spinel
Dark brown spinel in the partially serpentinized sample (FP-34C) and in serpentinites is armoured by magnetite grains, possibly formed during the serpentinization process. Original spinel has an allotriomorphic, interstitial texture. Preserved spinel cores have Cr/Cr Al composition ranges from 0.5 to 0.6 and Fe/Fe Mg around 0.5. TiO2 compositions vary from 0.07 to 0.14 wt% and Al2O3 cluster around 25 wt%. In the Cr/Cr Al v. Mg/Mg Fe diagram (Fig. 6a) the CVC data do not fall within the abyssal spinel peridotite data of Dick & Bullen (1984) but do overlap the spinel data from the Mariana fore-arc of hole 780, ODP-Leg 125 from Parkinson & Pearce (1998).

Pyroxene
Pyroxene is seldom preserved in the ultramac rocks. Only relict pyroxenes were found in sample FP-34C, where composition is Fe-rich diopside (Fig. 6b) and TiO2 was found to be very low. Clinopyroxene from an analysed gabbroic sample is mainly diopside, falling within the same compositional area as the relict pyroxenes from the serpentinites (Morimoto et al. 1988) with high Mg no. between 0.90 0.91, high SiO2 and extremely low TiO2. Compositional patterns of magmatic clinopyroxene are consistently used to infer possible tectonic

554

M. B. I. WEBER ET AL.
(a)
0.80

407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464

Mariana fore-arc

0.60

(b)
Spinel Cr/(Cr+Al)

Wo
Ultramafic rock

0.40

Gabbro Basaltic Andesites

Abyssal Peridotites
Diopside Hedenbergite

0.20 Augite Serpentinite Sample FP-34C 0.00 0.00

Clinoenstatite

Pigeonite
Clinoferrosillite

1.00

0.80

0.60

0.40

0.20

En

Fs

Spinel Mg/(Mg + Fe)

(c) 0.3
0.25

(d)
TiO2
MORB
IAT

Ultramafic rock Gabbro MORB Basaltic Andesite

0.2

AI (t)

0.15

0.1

BON

0.05

BON

IAT

0.06

0.12 Ti

0.18

0.24

SiO2 /100

Na2O

Fig. 6. (a) Cr# [Cr/(Cr Al)] versus Mg# [Mg/(Mg Fe)] relation in spinels from a completely serpentinised and one partially serpentinised rock (sample FP-34C) of the CVC. For comparison, the spinel composition from peridotites with different refractory character are also plotted. Field for spinels from abyssal spinel peridotites (from Dick & Bullen 1984) and IzuBoninMariana forearc (hole 780, ODP-Leg 125, Parkinson & Pearce 1998) are shown; open circle subduction-related Oman ophiolite (Lippard et al. 1986). (b) Clinopyroxene classication for the CVC ultramacs, gabbros and basaltic dykes. (c) Co-variation diagram of Al(t) versus Ti (atomic ratios) of studied pyroxenes in ultramac rocks, indicating their tectonic settings. (d) TiO2-Na2O-SiO2/100 in clinopyroxene (wt %) (IAT, Island Arc Tholeiite; BON, Boninite; MORB, Mid-ocean ridge basalt; after Beccaluva 1989).

settings of ophiolitic mac and ultramac rocks (e.g. Nisbet & Pearce 1977; Hebert & Laurent 1987; Capedri & Venturelli 1979; Beccaluva et al. 1989). In bivariant diagrams that use TiO2 in clinopyroxene as a marker of the degree of fusion and

different tectonic settings, the Cabo de La Vela gabbroic rocks plot within island arc elds (Fig. 6b, c). Together with the low TiO2 content of pyroxene from ultramac rocks this suggests a supra-subduction setting.

465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
LS-22 LS-26 LS-34 LS-44 LS-45 LS-51 LS-53 LS-59A LS-62 45.87 0.302 14.75 3.1 0.053 14.61 14.58 1.19 0.02 0.006 4.37 98.85 GBR BA BA BA 52.91 0.732 14.95 11.21 0.181 5.37 8.17 4.33 0.79 0.095 1.89 100.63 BA 52.49 0.892 16.15 10.89 0.19 4.88 8.34 4.74 0.58 0.144 1.39 100.69 BA 52.03 0.384 16.6 6.5 0.11 7.83 10.39 4.41 0.08 0.01 1.98 100.32 GBR 52.67 0.521 13.23 9.22 0.183 11.5 11.11 2.19 0.03 0.012 0 100.67 GBR 53.64 0.372 4.29 3.7 0.07 22.12 12.42 0.7 0.01 0.007 2.17 99.5 HBL Q6
THE CABO DE LA VELA MAFIC COMPLEX 555

Table 1. Major and trace element analyses

Sample

LS-6

LS-18

SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total Type 166 9.1 1.9 269.6 0 12.3 0 0 0 3.1 1810 6.3 0 0 132 68.6 32.3 38 149 2 0 3 66.6 0.1 351 6558 272.6 11 219.5 0.7 17.7 0.6 0.5 1.1 36.9 4388 17.5 0.4 0.2 38 16.9 51.1 37 324 117.3 0.8 65 511.4 0.9 160 4815 553.8 8.4 293.9 0.4 19.2 0.3 1.3 1.8 51.2 5348 19.3 1.2 0.5 40 17.1 42.8 31 259 91.3 1.3 57 199.7 0.4 23 264.7 3.4 188.7 0.2 15.9 14.2 13.8 1.7 45.4 23.7 0.4 0.3 0 7.9 88.2 308 108 2.6 54 3671.9 5 0 506.9 8.9 283.7 0.3 18.9 0.3 1.3 1.4 51.7 20 1.7 0.6 0 16.9 43.5 262 90.7 1.1 55 212.1 0.4 0 664 18.6 1 240.6 0.3 17.7 0.5 0 0 5.9 2302 9.4 0 0 74 73.4 55.1 35 200 51.1 0 10 392.6 0.4 732 539.7 7.8 272.2 0.2 16.6 1.1 1.4 1.7 48.2 22 1.2 0.3 0 13.2 81.1 366 114.9 0.5 63 333.4 0.6 0 249 7 0 111.7 0 13.3 0.9 0 0 10.4 3123 13.8 0 0 502 53.2 83.4 40 233 30.7 0 4 901.3 0.7 279

49.03 1.011 7.31 7.65 0.116 18.95 11.69 1.03 0.06 0.016 2.5 99.36 HBL

47.61 1.092 10.09 5.83 0.071 18.79 12.04 1.44 0.02 0.007 2.21 99.2 HBL

K Ba Rb Sr Cs Ga Ta Nb Hf Zr Ti Y Th U Cr Ni Co Sc V Cu Pb Zn W Mo Cl

498 3.9 0.7 14.6 0 12.7 0.4 2.2 1.4 36.6 6061 24.9 0.7 0 733 129.1 71.1 41 277 0.6 0 7 118.1 0.3 1425

166 2.1 0 29.8 0 13 0.1 0 0 9.4 6547 18.8 0 0 390 91.9 69 60 422 3.4 0 4 154.1 0.2 548

83 0 0.6 5.1 0 4.4 0.5 0 0.5 9.3 2230 10.9 0 0 5054 26.5 52.5 37 261 0.4 0 3 651.8 0.5 342
(Continued )

556

M. B. I. WEBER ET AL.

523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580

LS-62

0.5 1.7 0.31 1.9 0.9 0.42 1.43 0.28 1.6 0.38 1.1 0.19 1.16 0.15

Other minerals
Relict olivine is preserved within mesh cells from sample FC-34C. The composition is Fo90 91. Plagioclase compositions for the gabbros were determined petrographically, and comprise mainly labradorite andesine. Chemical analyses for feldspar in some volcanic samples indicate spilitization for these rocks, as plagioclase is replaced by albite. Brown amphibole of metasomatic origin, which replaces minerals in the gabbros, and amphiboles from one of the mac dykes were analysed, both being magnesiohornblendes (Hawthorne 1981).

LS-59A LS-53

0.5 1.4 0.26 1.3 0.8 0.55 1.1 0.25 1.4 0.39 1.07 0.15 1.09 0.14

4 10.1 1.49 7.2 2.6 0.77 3.53 0.59 3.09 0.84 2.18 0.33 2.38 0.35

0 1.4 0.3 2.2 1 0.58 1.65 0.34 2.18 0.56 1.41 0.24 1.55 0.2

LS-51

Major and trace element geochemistry

Five volcanic rocks, together with three gabbros and three hornblendites representative of the main rock types of the CVC, were selected for major and trace element analyses. Element data are presented in Table 2. Element analyses for LILES in the gabbros are in general below detection limits. SiO2 values range from 45.87 to 52.67 wt% for gabbros, from 47.61 to 53.64 wt% for hornblendites, and cluster around 52.5 wt% for the mac dykes. MgO wt% values are variable but are highest in the hornblendite samples (18.79 22.12%) and lowest for the mac rocks (4.88 5.37%). According to the inmobile elements (Nb, Y, Zr and Ti) classication of Winchester & Floyd (1977) (not shown), all these rocks are of mac intermediate composition and plot in the andesite eld. Since most of the samples show signs of alteration, chemical interpretation is based on high eld strength elements and transitional metals considered to be immobile (Pearce & Cann 1973; Winchester & Floyd 1977). In the V v. Ti/1000 of Shervais (1982), all samples plot towards the arc-related rocks eld, whereas in the Ti v. Zr plot after Pearce (1982), the basaltic andesite dyke samples fall within the volcanic arc eld (gures not shown). Gabbros and hornblendites have particularly low Zr. Additional constraints on the geotectonic setting can be obtained from REE and multi-element diagrams presented in Figure 7b, c. Gabbros show a strongly depleted LREE pattern compared with HREE when plotted on the chondrite-normalized REE diagram (Fig. 7b), with Nd Lu values ranging from 6.67 to 11.00. They have a notable positive Eu anomaly, indicating that plagioclase fractionation was an important factor in the genesis of these rocks. Gabbro REE patterns roughly resemble modern MORB-type plutonic rocks. The multi-element diagram shows negative Zr and Ti anomalies as well as positive Sr and Ba

LS-45 LS-44 LS-34 LS-26

0 1 0.15 0.8 0.4 0.28 0.69 0.15 1.09 0.23 0.64 0.1 0.64 0.11 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 2.5 9.3 1.55 7.8 2.7 0.81 3.67 0.73 4.43 0.92 2.25 0.41 2.95 0.43 0 1.2 0.34 1.6 1.3 0.39 2.21 0.41 3.06 0.69 1.84 0.25 1.96 0.24

2.9 7.6 1.24 6.6 2.1 0.66 2.81 0.48 2.86 0.69 1.7 0.27 1.96 0.32
HBL hornblendite, GBR gabbro, BA basaltic andesite.

Table 1. Continued

Sample

LS-6

LS-18

LS-22

6.8 14.2 1.95 9.3 2.9 0.86 3.14 0.55 3.32 0.73 1.78 0.26 2.16 0.31

3.8 8.6 1.44 7.3 2.5 0.75 3.63 0.6 3.59 0.88 2.08 0.36 2.53 0.36

6.4 13.5 2.03 10.3 2.5 0.91 3.28 0.53 3.09 0.72 1.92 0.29 2.12 0.29

581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638
Olivine Sample SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO NiO K2O Total Si Al Al Fe(iii) Cr Ti Fe(ii) Mn Mg Ca Na K Al tot Total Wo En Fs Ac 48.68 49.96 0.66 0.70 1.94 0.06 0.00 0.03 0.05 0.00 2 0.01 0.00 0.97 0.95 0.01 0.00 0.06 4.09 1.96 0.04 0.03 0.06 0.00 0.00 0.04 0.00 0.91 0.94 0.03 0.00 0.00 4.02 47.48 45.84 5.10 1.57 99.56 99.75 99.92 1.98 0.02 0.04 0.06 0.00 0.00 0.34 0.01 0.53 0.96 0.07 0.00 0.00 4.02 48.74 26.93 20.98 3.35
557

Table 2. Representative mineral analyses Pyroxene FP-34C PER centre 53.26 0.00 1.50 0.36 1.88 0.06 17.95 24.34 0.19 0.00 SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO NiO K2O NiO Total Si Ti Al Cr Fe2 Mn Mg Ca Na K Ni Total 53.51 0.08 1.59 3.23 0.00 0.10 16.70 24.07 0.44 0.00 51.74 0.08 1.47 12.59 0.00 0.40 9.31 23.45 0.89 0.00 Sample FP-34C PER LS22 GBR LS11 AND Sample LS18 HBL LS22 GBR Amphibole LS11 AND

Spinel FP-34C PER border

Sample Type

LS-32 SER

FP-34C PER

SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO NiO K2O SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO NiO K2O NiO Total Si Ti Al Cr Fe(ii) Mn Mg Ni Ca Total 0.99 0.00 0.00 0.00 0.18 0.00 1.83 0.01 0.00 3.01 0.99 0.00 0.00 0.00 0.19 0.00 1.82 0.01 0.00 3.01

0.00 0.11 25.52 39.05 23.36 0.32 10.86 0.00 0.11 0.00

0.03 0.05 26.28 41.75 22.39 0.13 11.13 0.00

THE CABO DE LA VELA MAFIC COMPLEX

Total

99.32

101.80

40.16 0.00 0.01 0.00 8.69 0.11 49.88 0.02 0.05 0.00 0.42 99.33

40.48 0.00 0.00 0.00 9.37 0.11 50.03 0.03 0.01 0.02 0.43 100.48

47.46 1.04 11.55 0.02 5.15 0.11 18.46 12.39 2.06 0.06 0.08 98.37 6.65 0.11 1.91 0.00 0.60 0.01 3.85 1.86 0.56 0.01 0.01 15.57

49.40 0.70 9.63 0.04 5.08 0.06 18.78 12.85 1.75 0.05 0.11 98.44 6.90 0.07 1.58 0.00 0.59 0.01 3.91 1.92 0.47 0.01 0.01 15.48

51.51 0.42 5.63 0.00 12.18 0.40 7.11 18.74 2.58 0.01 0.00 98.58 7.53 0.05 0.97 0.00 1.49 0.05 1.55 2.94 0.73 0.00 0.00 15.30

Si Ti Al Cr V Fe(iii) Fe(ii) Mn Mg Ca Zn Total

0.00 0.02 7.44 7.64 0.00 0.88 3.95 0.07 4.00 0.00 0.00 24.00

0.01 0.01 7.47 7.96 0.00 0.53 3.99 0.03 4.00 0.00 0.00 24.00

Cr/Cr 1 Al Fe/Fe 1 Mg Mg/Mg 1 Fe Cr/Cr 1 Al Fo Fa Tp 91.00 8.89 0.11

0.51 0.55 0.45 0.51

0.52 0.53 0.47 0.52

90.39 9.50 0.12

SER Serpentinite, PER serpentinised peridotite, HBL hornblendite, GBR gabbro, AND andesite.

558

M. B. I. WEBER ET AL.

639 (a) 100 (b) 100 640 641 642 643 Q5 10 644 645 10 646 647 1 648 649 650 651 0.1 1 La Pr Eu Tb Ho Tm Lu 652 Eu Tb Ho Tm Lu La Pr Ce Nd Sm Gd Dy Er Yb 653 Ce Nd Sm Gd Dy Er Yb 654 1000 1000 655 656 657 100 100 658 659 10 10 660 661 1 1 662 663 664 0.1 0.1 665 666 0.01 0.01 667 Cs Ba U Nb Ce Nd Sm Ti Er Lu Cs Ba U Nb Ce Nd Sm Ti Er Lu 668 Rb Th K La Sr Zr Eu Y Yb Rb Th K La Sr Zr Eu Y Yb 669 670 Fig. 7. Geochemistry of gabbros from the CVC. Normalizing values are from Nakamura (1977) and Pearce (1983). 671 (a) REE patterns normalized to chondrite and multi-element patterns normalized to MORB for CVC gabbros. Open circles represent hornblendites, closed circles represent gabbros. (b) REE patterns normalized to chondrite and multi672 element patterns normalized to MORB for CVC basaltic andesite dykes. The grey elds outline the compositional 673 range of the southern Mariana Trough (Gribble et al. 1996). 674 675 676 spikes, indicative of a subduction-related com- which together with positive Sr and Ba anomalies 677 ponent in the source. and the already mentioned pattern, characterize 678 Volcanic dykes show at patterns, with a small subduction-related magmas. 679 negative Eu anomaly, when plotted on the Hawkesworth et al. (1993a, b), subdivided 680 chondrite-normalized REE diagram. Apparent island arc basalts into two groups on the basis of 681 682 slight variations in the LREE, with LaLu ranging LREE/HREE, using La Yb ratios to discriminate from 9.06 to 21.94, indicate small degrees of differ- between predominantly intra-oceanic arcs (La 683 entiation. This is in agreement with the petrographic Yb , 5) and arcs developed near continental 684 observations, where the LREE enriched samples margins (LaYb . 5). The CVC basaltic rocks 685 contain amphibole and quartz in addition to pyrox- fall within the low LaYb island arc group and 686 687 ene and plagioclase. overlap the data from the Mariana arc. The multi-element diagram for basaltic andesites 688 shows a decrease from LILE-enriched to HFSE689 depleted and contrasts with the gabbros. The LILE K Ar geochronology 690 enrichment is generally attributed to element mobi691 692 lity during alteration; however, the increased con- Previous age constraints from this region were tents of relatively immobile elements like Th and restricted to stratigraphical relationships with 693 La suggest that this pattern is inherited from the Miocene sediments. Three andesitic dykes were 694 original magmatic source. The basaltic rocks also dated by the K Ar whole rock method. Obtained 695 show Nb, Zr and a weak Ti negative anomaly, ages (Table 3a) overlap within error. Their 696
Sample/Nakamura, 1977 Sample/MORB Sample/MORB Sample/Nakamura, 1977

THE CABO DE LA VELA MAFIC COMPLEX

559

697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754

differences may be related to minor hydrothermal alteration shown by saussuritization of plagioclase. We consider c.74 Ma to be the age of dyke intrusion into the macultramac unit based on the more precise analytical quality. Low K contents in the hornblendites preclude reliable K Ar data to constrain the pre-dyke intrusion history. However, cross-cutting relations and depth of formation indicate that the gabbroic and ultramac rocks must have been exhumed before intrusion of the dykes, and therefore a signicant lapse of time must have passed between the formation of these two different units.

Sr Nd isotopes
Andesitic rocks and gabbros were analysed for Rb Sr and SmNd isotopic ratios. Initial ratios were calculated for the 74 Ma K Ar age. Results are presented in Table 3b and 3c and Figure 9. The 1Nd and initial 87Sr/86Sr values for the basaltic andesite dykes range between 4.1 and 7.3 and 0.7038 and 0.7041 respectively. Gabbros have 1Nd of 9.9 and initial 87Sr/86Sr of 0.7029 and 0.7031. When compared with the basaltic dykes, the gabbros overlap the MORB eld (Fig. 9), indicating derivation from a depleted source, whereas the basaltic dykes plot similar to primitive island arc

rocks found in other suites in the circum-Caribbean realm. Isotopic variation in the basalt dykes could be explained by different input of sedimentary components into the subduction-zone magmas. Interestingly, Mariana back-arc samples plot between the gabbros and the basaltic dykes from the CVC, which could be interpreted as a higher input of the sedimentary component for the basaltic dykes in the CVC. This is further seen by slight off-set towards higher Sr ratios of the CVC and the Caribbean plutons when compared with the Mariana back-arc basalts. When compared with the highly negative 1Nd data from Pre-Mesozoic basement and Jurassic plutons from the Guajira region (Cordani et al. 2005; Cardona-Molina et al. 2006), the data fall away from older crustal signatures. Therefore assimilation of older crust is precluded. Unpublished isotopic data from the Eocene continental-arc magmatic rocks intruding Cretaceous metamorphic complexes of the Guajira region, possibly related to the CVC, show mixing with older basement rocks or sediments, but this is not seen in the CVC rocks. These isotopic characteristics clearly indicate a primitive mantle source and a minor sediment input in the subduction-related magma source of the basaltic rocks. The differences between the isotopic ratios of gabbros and basaltic dykes may be

Table 3. Geochronological and isotopical data K-Ar Sample Ar40 Rad ccSTP/g ( 1026) 1.46 0.47 1.58
147

Rock

%K

K error (%) 0.5000 0.5000 1.7499 Nd (ppm) 1.459 0.761 8.892 7.340 6.057 Sr (ppm) 254.29 228.79 272.99 166.68 211.38

Ar40Atm (%) 71.94 68.79 77.84 Error 9 2 7 7 Error 1 125 7 4 12

Age (Ma) 73.5 69.5 77.3

143

+ (Ma)

LS-34 LS-44 LS-53 Sm-Nd Sample LS-51 LS-22 LS-34 LS-44 LS-26 Rb-Sr Sample LS-51 LS-22 LS-34 LS-44 LS-26

Basalt Basalt Basalt Rock Gabro Gabro Basalt Basalt Basalt Rock

0.5013 0.1711 0.5162 Sm (ppm) 0.661 0.680 2.462 1.986 Rb (ppm)

4.2 4.9 5.4 Error 17 12 41 8 13 Error 6 7 10 7 1 fSm/Nd 0.39 2 1.00 2 0.76 0.03 0.01
87

144

Sm/ Nd

144

Nd/ Nd

1 (74Ma) 9.90 4.05 6.02 7.25

0.2739 0.0462 0.2028 0.1983 Rb87/Sr86 (X) 0.0087 0.1932 0.0861 0.0520 0.1448

0.513183 0.513228 0.512774 0.512950 0.513011 Sr87/Sr86 (Y) 0.70291 0.70338 0.70386 0.70396 0.70406

Sr/86Sr 74 Ma 0.7029 0.7032 0.7038 0.7039 0.7041

Gabro 0.76 Gabro 15.28 Basalt 8.12 Basalt 3.00 Basalt 10.58

560

M. B. I. WEBER ET AL.

755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812

30 Basaltic Andesites Grenada 25 Gabbros

a/Y b= 5

20

La

15 Aleutians Is. 10

Mariana.

Sandwich Is.
N-MORB

0 0 1 2 3 4 5

Yb
Fig. 8. La v. Yb in basalts (SiO2 , 55%) in intraoceanic and continental margin island arc basalts, after Hawkesworth et al. (1993a, b). Other data elds from Jolly et al. (2006) and Taylor & Martinez (2003). Q6

explained by different input at source. The gabbros formed by partial fusion of mantle in a suprasubduction-zone environment, with limited input from the subducted slab. Input from the
15.0 MORB

subducting plate was greater in basaltic andesites and shifts the isotopic ratios towards lower 1Nd and higher Sr, probably reecting a changing tectonic conguration.

Basaltic Andesites Gabbros Caribbean PIA

10.0

5.0

Mariana arc

eNd
0.0 Guajira Eocene arc

5.0 Gabbros Basalts 10.0 0.701 0.702 0.703 0.704 0.705

87

Ocean Sediments 0.706 0.707 0.708 0.709 0.71

Sr/86Sr

Fig. 9. 1Nd v. 87Sr/86Sr for gabbros and basalts of the CVC. Other elds depicted are from the Mariana arc (Gribble et al. 1996 and sources listed therein), MORB (White & Hofmann 1982) and Caribbean PIA (Jolly et al. 2006). Also shown is the unpublished data eld for the Eocene arc magmatism.

THE CABO DE LA VELA MAFIC COMPLEX

561

813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870

Tectonomagmatic setting of the Cabo de la Vela Complex

Based on the lack of units typical of complete ophiolite sequences and the unclear relation with the margin where it was accreted, the Cabo de la Vela complex represents a Cordilleran-type ophiolite, as do many other Caribbean ophiolites (Beccaluva et al. 1996, 2004). Field and petrologic evidence indicate that the Cabo the la Vela rocks followed a complex succession of events, recording the dynamic oceanic tectonic cycles common of ophiolite rocks (Shervais 2001). The oldest unit comprises mantle rock of wherlitic composition. Coarse-grained gabbros and troctolites intrude this peridotite, recording continuous uplift in a slow spreading setting. Both units show evidence of correlatable high temperature deformation as well as hydration. Mineral chemistry of spinel and pyroxene (Figs 6a, b) from peridotites and gabbros indicates that these units correspond to a tectonite formed in a supra-subduction zone environment (Kenemetsky et al. 2001; Okamura et al. 2006). The MORB geochemical and more juvenile isotopic signature from the gabbros combined with the supra-subduction signature from the mineral chemistry are more akin to a back-arc tectonic setting where both MORB or subduction zone signatures are common (Saunders & Tarney 1984). The presence of hornblendites and different generations of pervasive serpentinization and rodingitization events are indicative of several hydrothermal overprints ranging from deep to shallow crustal depths and recording a continuous tectonic exhumation history in an oceanic setting. This, together with lateral heterogeneity of hydrothermal alteration, may be taken as evidence of a slow spreading ridge environment (Cannat 1996; Cannat et al. 1992). Absence of the other typical ophiolitic units, like cumulate gabbros and intrusive basaltic sheeted dykes, may be explained by erosion or tectonic removal after emplacement onto the margin, but nevertheless ts nicely into the evidence of a slow spreading ridge environment for these rocks. The history of the andesitic dykes differs from the previously discussed ultramacs gabbro unit. The dykes show neither deformation nor pervasive hydrothermal alteration, and must therefore correspond to another stage of the evolution history. Geochemical data show that these rocks may represent an intraoceanic island arc with poor sediment input or older crust contamination. Sr and Nd isotopic comparison between the andesites and the gabbros neatly shows the differences between magma sources and conrms that the CVC plutonic and volcanic rocks are two separate units, formed at

different stages within an intraoceanic environment. A possible explanation for such magmatic variations could include the presence of different mantle sources (MORB and supra-subduction) related to different phases of migration of the arc in a long term subduction environment (Stern 2002). These events are constrained by the 74 Ma K Ar crystallization age obtained for the volcanic dykes and imply that the ultramac rocks and the gabbros of the CVC were emplaced before the Campanian. Modern-day analogues for slow-spreading supra-subduction zone environments are the Mariana Trough and the Lau back-arc (Gribble et al. 1996; Taylor & Martinez 2003). Basalts from both back-arc basins include arc-like components and MORB-like end-members. Mantle ow and convection induce mixing of previously depleted mantle sources and produce a range of compositions that can vary through time between end-members (Taylor & Martinez 2003). This could well be the case in the CVC, whereby the ultramac and gabbroic units represent of a more MORB-like end-member source and the andesitic dykes unit a later, more subduction-related endmember source of the same arc. As described in the geological setting, two different Cretaceous low-grade vulcano-sedimentary metamorphic units are exposed in the Guajira Peninsula, the Jarara Formation to the SE and the Etpana Formation to the NW, described in great detail by Alvarez (1967) and Lookwood (1965). These are temporally and geologically linked to the Cabo de la Vela Mac Ultramac Complex of this paper (Fig. 2). The lithostratigraphic characteristics of the metamorphic protoliths from these two units include a sequence of mainly siliclastic sediments (pelites to rudites) with intercalations of mac tuffs and lavas indicative of a mature arc-related geological setting. Differences between the units are that the Jarara Fm. includes marbles and more abundant volcanic rocks, whereas the Etpana Fm. has intermixed serpentinites and gabbros. These metasediments are part of the same basin, with the Etpana Fm. representing the deeper sedimentary environment (Lockwood 1965). Fossil ages range from Turonian to Maastrichtian. Nearby Cretaceous units of the autochthonous margin of South America lack the volcanic and predominant silicicalstic components of the Etpana and Jarara Fms. (McDonald 1964; Villamil 1999). It is therefore also possible to assume that the Jarara Etpana sequence formed in an allochtonous arc position. The contact between these Cretaceous units and basement rocks has been described as a shear-zone (MacDonald 1964; Lockwood 1965; Alvarez 1967) that we interpret as the possible suture

562

M. B. I. WEBER ET AL.

871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928

between authoctonous South America and the allochtonous Jarara Etpana arc. The presence of undeformed Eocene continental arc plutonism in this region and the regionally correlatable plutons of the Santa Marta massif, that clearly intrude the South American margin (Tschanz et al. 1974), also suggest that accretion occurred before this magmatic event. The relation between the Jarara Etpana sequence and the CVC is hidden. However, as Alvarez (1967) pointed out, the ultramac and gabbroic rocks found in the CVC resemble the intercalated mac and ultramac rocks of the Etpana Fm. Reconnaissance eld and petrographic observations of the Etpana Formation indicate a strong similarity to the CVC in the nature and distribution of serpentinization and rodingization (Arredondo et al. 2005). The map-pattern shows that the expected tectonic pattern of intermixing of these units

(clearly aligned lens shapes in the low grade rocks) is not evident, and therefore we suggest that the vulcano-sedimentary protoliths of the Etpana Fm. could have been deposited over a substratum correlatable with the CVC ultramac and gabbroic units. This interpretation has two corollaries. The rst is that the mac dykes from the CVC may be contemporaneous with the deposition of both Jarara and Etpana Fms. The second is that deformation may be related to a subsequent tectonic episode. A tentative model for the early Cretaceous to Eocene evolution of the CVC derived from the above data is presented in Figure 10. It includes three main stages: 1. Initiation of an ocean ocean subduction zone, forming a slow-spreading back-arc basin in pre-Campanian times. This back-arc,

Fig. 10. Three-stage model for the formation of the Cabo de la Vela Mac-Ultramac Complex. SAM represents the South American margin (Modied from Giunta et al. 2006).

THE CABO DE LA VELA MAFIC COMPLEX

563

929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986

2.

3.

represented by the CVC ultramac and gabbroic units, was progressively exhumed as a consequence of the slow spreading dynamics. Subduction zone conguration changed and arc-like magmatism developed upon a migrating spreading centre, represented by the andesitic dykes of the CVC. Possibly simultaneously the arc evolved and sedimentary deposits formed. The distal sedimentary sequence, Eptana Fm., developed upon the previously rifted basin, whereas the more shelflike deposits of the Jarara Fm. formed as sediments near the arc. Continuous ocean subduction carried the arc towards the passive continental margin of South America. The units were accreted onto the South American margin. Following the accretion the Caribbean plate began to subduct under South America. This event is constrained by the age of Eocene magmatism that intruded the already deformed Etpana and Jarara Formations at 47 Ma (Lockwood 1965). CardonaMolina et al. (2006) deduced that this could have occurred between the late Cretaceous and the Palaeocene, based on Ar/Ar spectra of basement rocks.

Caribbean realm
The tectonic evolution of the Caribbean is related to Jurassic Early Cretaceous formation of oceanic Proto-Caribbean crust following the separation of North and South America. Subsequent development of a multistage intraoceanic-arc or several intraoceanic arcs in either a near mid American or a Pacic position happened from late Cretaceous to Recent times. Thickened oceanic plateau crust migrated from the west between the Americas, leaving fragments on the continental margin (Pindell 1993; Pindell & Kennan 2001; reviews in Giunta et al. 2002, 2006; James 2006). The Cabo de la Vela Mac Ultramac Complex and the associated Etpana Jarara Fms of the Guajira region record the tectonic evolution of an island arc of Campanian and older age (whole rock K Ar age of 74 Ma for the basaltic andesites, and the older ultramac and gabbroic units). This formed on already mature arc basement or as part of coalesced arcs juxtaposed before their accretion onto a passive continental margin in pre-Eocene times. High-pressure rocks in Miocene conglomerates that show notable similarities to rocks of the Venezuelan Cordillera de la Costa and Margarita, dated at 90 110 Ma (Sisson et al. 1997; Stockhert et al. 1995; Zapata et al. 2005), quartz and granitoid clasts in metaconglomerates from the Jarara Fm. may be vestiges of earlier subduction. This implies

prolonged intra-oceanic plate convergence for this intra-oceanic arc setting. Other remnants of magmatic arcs and subduction complexes recording ocean ocean convergence in the southern Caribbean are the early Cretaceous Villa de Cura and Dos Hermanas units from Venezuela, the Santa Marta Schists of Colombia and the Washikemba Formation in Bonaire (MacDonald et al. 1971; Beccaluva et al. 1996; Giunta et al. 2002; Thompson et al. 2004). Petrographical and geochemical comparison, as well as age constrains, show similarities which indicate that the Dos Hermanas unit or the Washikemba Formation (Giunta et al. 2002; Thompson et al. 2004) could be correlated to the rocks of the Jarara Etpana formations and the CVC. Ages similar to those obtained for the CVC have been recorded throughout the Dutch Antilles. In Curac ao, Sinton et al. (1998) report a c. 76 Ma Ar/Ar age for a dolerite sill that intrudes the Albian Turonian Curac ao lava succession. Late Cretaceous turbidites overlying this sequence contain a signicant population of euhedral zircons in the range of 7087 Ma (Wright 2004). In Aruba, Priem et al. (1986), based on various Rb Sr and K Ar age determinations, suggest that a c. 72 Ma thermal event was responsible for the isotopic resetting of the Aruba Batholith. These data suggest an important c. 70 76 Ma magmatic event that correlates with the CVC. The segmented distribution of correlatable Cretaceous arc-related and other intra-oceanic units (including the CVC) along the continental margin of Northern South America and the Caribbean, and SW- to east- younging accretion, becoming younger from Ecuador to Colombia and Venezuela, are compatible with the northeasterly migration of the Caribbean plate with its arc fronts between the Americas, and both the accretion and subsequent disruption within an oblique continental margin (Pindell 1993; Toussaint -Lallement & Sisson 2005; Vallejo 1996; Ave et al. 2006).
This project was partially funded by the National University of Colombia, Medellin. The authors wish to acknowledge the contributions of Jorge Gomez from the Colombian Geological Survey (INGEOMINAS) and the staff of the Geochronological Research Centre (CP-Geo) and geochemical laboratories of the University of Sao Paulo. We also thank G. Y. Ojeda and M. Nieto from the ICP for their help with gravimetric data and C. Jaramillo for constructive comments on the initial draft of the paper. We are also grateful to K. H. James for inviting and encouraging us to present this paper and G. Giunta for the constructive review and suggestions to improve the manuscript.

564

M. B. I. WEBER ET AL. D ICK , H. J. B. & B ULLEN , T. 1984. Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology, 86, 54 76. SKA , E. 1995. Rodingites of the Eastern Part of the D UBIN w Gogolo w Serpentinite Massif, Lower Jordano Silesia, Poland. The Canadian Mineralogist, 33, 585608. F ROST , B. R. 1975. Contact metamorphism of serpentinites: chloritic black-wall and rodingite at Paddy-goeasy, Central Cascades, Washington. Journal of Petrology, 16, 272313. CUYER , C. 1996. H -G REEN , G. L., P LASS , A. & L E F RU Petrologic and stable isotope constraints on hydrothermal alteration and serpentinization of the EPR shallow VEL , C. ET AL . mantle at Hess Depp (Site 895). In: M E (eds) Proceedings of the Ocean Drilling Program, Scientic Results, 147. Ocean Drilling Program, Texas A&M University, 255291. G IUNTA , G., B ECCALUVA , L., C OLTORTI , M., S IENA , F. & V ACCARO , C. 2002. The southern margin of the Caribbean Plate in Venezuela: tectono-magmatic setting of the ophiolite units and kinematic evolution. Lithos, 63, 19 40. G IUNTA , G., B ECCALUVA , L. & S IENA , F. 2006. Caribbean Plate margin evolution: constraints and current problems. Geologica Acta, 4, 265 277. G REEN , D. H., L OCKWOOD , J. P. & K ISS , E. 1968. Eclogite and almandine jadeitequartz rock from the Guajira Peninsula, Colombia, South America. American Mineralogist, 53, 13201335. BEN , G RIBBLE , R. F., S TERN , R. J., B LOOMER , S. H., S TU D., OH EARN , T. & N EWMAN , S. 1996. MORB mantle subduction components interact to generate basalts in the southern Mariana Trough back-arc basin. Geochimica et Cosmchimica Acta, 60, 2153 2166. H AWKESWORTH , C. J., G ALLAGHER , K., H ERGT , J. M. & M C D ERMOTT , F. 1993a. Mantle slab contributions in arc magmas. Annual Reiews in Earth and Planetary Science, 21, 175 204. H AWKESWORTH , C. J., G ALLAGHER , K., H ERGT , J. M. & M C D ERMOTT , F. 1993b. Trace element fractionation processes in the generation of island arc basalts. In: T ARNEY , J., P ICKERING , K. T., K NIPE , R. J. & D EWEY , J. F. (eds) Melting and Melt Movement in the Earth. Clarendon Press/The Royal Society, London, 393405. H AWTHORNE , F. C. 1981. Crystal chemistry of the amphiboles. Reviews in Mineralogy and Geochemistry, 9A, 1 102. H EBERT , R. & L AURENT , R. 1987. Mineral chemistry of the plutonic section of the Troodos ophiolite: new constraints for genesis of arc-related ophiolites. In: M ALPAS , G. J. & M OORES , E. M. (eds). Troodos 1987 Ophiolites and Oceanic Lithosphere, 149 163. I TURRALDE -V INENT , M. A. & L IDIAK , E. G. (eds) 2006. Caribbean plate tectonics: stratigraphic, magmatic, metamorphic and tectonic events (UNESCO/IUGS IGCP Project 433). Geologica Acta, 4, 1 341. J AMES , K. H. 2006. Arguments for and against the Pacic origin of the Caribbean Plate: discussion, nding for an inter-American origin. Geologica Acta, 4, 279302.

987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044

References
A LVAREZ , W. 1967. Geology of the Simarua and Carpintero area, Guajira Peninsula, Colombia. PhD thesis, Princeton University. a A RREDONDO , L. F., W EBER , M. ET AL . 2005. Petrograf a de Jarara, de las ultramatas y rodingitas de la Serran nsula de la Guajira-Colombia. X Congreso ColomPen a Simposio de Geolog a Regional. biano de Geolog -L ALLEMENT , H. G. & S ISSON , V. B. 2005. ExhumaA VE tion of eclogites and blueschist in northern Venezuela: constraints from kinematic analysis of deformation L ALLEMENT , J. G. & S ISSON , structures. In: A VE V. B. (eds) CaribbeanSouth American Plate Interactions, Venezuela. Geological Society of America, Boulder, CO, Special Papers, 394, 193306. B ECCALUVA , L., C OLTORTI , M. ET AL . 1996. Cross sections through the ophiolitic units of th Southern and Northern Margins of the Caribbean Plate, in Venezuela (Northern Cordilleras) and Central Cuba. Ooliti, 21, 85103. B ECCALUVA , L., C OLTORTI , M., G IUNTA , G. & S IENA , F. 2004. Tethyan vs. Cordilleran ophiolites: a reappraisal of distinctive tectono-magmatic features of suprasubduction complexes in relation to the subduction mode. Tectonophysics, 393, 163174. B ECCALUVA , L., M ACCCIOTTA , G., P ICCARDO , G. B. & Z EDA , O. 1989. Clinopyroxene composition of ophiolite basalts as petrogenetic indicator. Chemical Geology, 77, 165 182. C ANNAT , M. 1996. How thick is the magmatic crust at show spreading oceanic ridges? Journal of Geophysical Research, 101, 28472857. C ANNAT , M., B IDEAU , D. & B OGAULT , H. 1992. Serpentnized peridotites and gabbros in the Mid-Atlantic Ridge axial valley at 158370 N and 168520 N. Earth and Planetary Science Letters, 109, 87 106. C APEDRI , S. & V ENTURELLI , G. 1979. Clinopyroxene composition of ophiolitic metabasalts in the Mediterranean area. Earth and Planetary Science Letters, 43, 6173. C ARDONA -M OLINA , A., C ORDANI , U. & M AC D ONALD , W. 2006. Tectonic correlations of pre-Mesozoic crust from the northern termination of the Colombian Andes, Caribbean region. Journal of South American Earth Sciences, 21, 337 354. C OOGAN , L. A., M AC L EOD , C. J. ET AL . 2001. Wholerock geochemistry of gabbros from the Southwest Indian Ridge: constraints on geochemical fractionations between the upper and lower oceanic crust and magma chamber processes at (very) slow-spreading ridges. Chemical Geology, 178, 122. C ORDANI , U. G., C ARDONA , A., J IMENEZ , D., L IU , D. & N UTMAN , A. P. 2005. Geochronology of Proterozoic basement inliers from the Colombian Andes: tectonic history of remmants from a fragmented grenville belt. In: V AUGHAN , A. P. M., L EAT , P. T. & P ANKHURST , R. J. (eds) Terrane Processes at the Margins of Gondwana. Geological Society, London, Special Publications, 246, 329346. D E P AOLO , D. J. 1988. Age dependence of the composition of continental crust: evidence from Nd isotopic variations in granitic rocks. Earth and Planetary Science Letters, 90, 263 271.

THE CABO DE LA VELA MAFIC COMPLEX

565

1045 1046 1047 1048 Q7 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 Q8 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 Q9 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 Q3 1094 1095 1096 1097 1098 1099 1100 1101 1102

J OLLY , W. T., L IDIAK , E. G. & D ICKIN , A. P. 2006. Cretaceous to Mid-Eocene pelagic sediment budget in Puerto Rico and the Virgin Islands (northeast Antilles Island arc). Geologica Acta, 4, 35 62. K ELLOGG , J. N., G ODLEY , V. M., R OPAIN , C., B ERMU DEZ , A. & A IKEN , C. L. V. 1991. Gravity Field of Colombia, Eastern Panama and Adjacent Marine Areas. The Geological Society of America, Boulder, CO, Map and Chart Series MCH 070. K ENEMETSKY , V. S., C RAWFORD , A. J. & M EFFRE , S. 2001. Factors controlling chemistry of magmatic spinel: an empirical stud of associated olivine, Cr-spinel and melt inclusions from primitive rocks. Journal of Petrology, 42, 345 352. L EWIS , J. F., D RAPER , G., P ROENZA , J. A., E SPAILLAT , NEZ 2006. Ophiolite-related ultramac J. & J IME rocks (serpentinites) in the caribbean region: a review of their occurrence, composition, origin, emplacement and ni-laterite soil formation. Geologica Acta, 4, 237263. L IPPARD , S. J., S HELTON , A. W. & G ASS , I. G. 1986. The Ophiolite of Northern Oman. Geological Society Memoirs, 11. Blackwell, Oxford, 178. a de Jarara L OCKWOOD , J. P. 1965. Geology of the Serran nsula, Colombia. PhD thesis, area, Guajira Pen Princeton University. a de M AC D ONALD , W. D. 1964. Geology of the Serran nsula, Colombia. PhD Macuira area, Guajira Pen thesis, Princeton University. M AC D ONALD , W., D OOLAN , B. & C ORDANI , U. 1971. Cretaceous Early Tertiary metamorphic age values from the South Caribbean. Geological Society of America Bulletin, 82, 13811388. M ESCHEDE , M. 1986. A method of discriminating between different types of mid-ocean ridge basalts and continental tholeiites with the Nb Zr Y diagrams. Chemical Geology, 56, 207 218. M ONTES , C., H ATCHER , R. D. & R ESTREPO -P ACE , P. 2005. Tectonic reconstruction of the northern Andean blocks: oblique convergence and rotations derived from the kinematics of the Piedras Girardot area, Colombia. Tectonophysics, 399, 221 250. M ORI , P. E., R EEVES , S., T EIXEIRA , C. & H AUKKA , M. 1999. Development of a fused glass disc XRF facility and comparison with the pressed powder pellet o Paulo technique at Instituto de Geociencias, Sa ncias, 29, University. Revista Brasileira de Geocie 441446. M ORIMOTO , N., F ABRIES , J. ET AL . 1988. Nomenclature of pyroxenes. American Mineralogist, 62, 53 62. N AKAMURA , N. 1974. Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and ordinary chondrites. Geochimica et Cosmochimica Acta, 38, 75 775. N ICOLAS , A. & R ABINOWICZ , M. 1984. Mantle ow pattern at oceanic spreading centres: relation with ophiolitic and oceanic structures. In: G ASS , I. G., L IPPARD , S. J. & S HELTON , A. W. (eds) Ophiolites and Oceanic Lithosphere. Geological Society, London, Special Publications, 13, 147 151. N ISBET , E. G. & P EARCE , J. A. 1977. Clinopyroxene composition in mac lavas from different tectonic settings. Contributions to Mineralogy and Petrology, 63, 149160.

O KAMURA , H., A RAI , S. & K IM , Y.-U. 2006. Petrology of forearc peridotite from the Hahajima Seamount, the Izu-Bonin arc, with special reference to chemical characteristics of chromian spinel. Mineralogical Magazine, 70, 15 26. P ARKINSON , I. J. & P EARCE , J. A. 1998. Peridotites from the Izu Bonin Mariana forearc (ODP Leg 125): evidence for mantle melting and meltmantle interaction in a supra-subduction zone setting. Journal of Petrology, 39, 1577 1618. P EARCE , J. A. 1982. Trace element characteristics of lavas from destructive plate boundaries. In: T HORPE , R. S. (ed.) Andesites. Wiley, Chichester, 525 548. P EARCE , J. A. & C ANN , J. R. 1973. Tectonic setting of basic volcanic rocks determined using trace element analyses. Earth and Planetary Science Letters, 19, 290 300. P INDELL , J. L. 1993. Evolution of the Gulf of Mexico and the Caribbean. In: D ONOVAN , S. K. & J ACKSON , T. A. (eds) Caribbean Geology: An Introduction. University of the West Indies Publishers Association, Kingston, 1339. P INDELL , J. L. & K ENNAN , L. 2001. Kinematic evolution of the Gulf of Mexico and Caribbean. Transactions, Petroleum Systems of Deep-water Basins: Global and Gulf of Mexico Experience. GCSSEPM 1st Annual Research Conference, Houston, TX. Gulf Coast Section Society of Economic Paleontologists and Mineralogists Foundation, 159 192. P INDELL , J. L., K ENNAN , L., S TANEK , K. P., M ARESCH , W. V. & D RAPER , G. 2006. Foundations of Gulf of Mexico and Caribbean evolution: eight controversies resolved. Geologica Acta, 4, 303341. Q10 P RIEM , H. N. A., B EETS , D. J., B OELRIJK , N. A. I. M. & H EBEDA , E. H., 1986. On the age of the late Cretaceous tonalitic/gabbroic batholith on Aruba. Netherlands Antilles (southern Cribbean borderland). Geologie en Mijnbouw, 65, 247 256. S AUNDERS , A. D. & T ARNEY , J. 1984. Geochemical characteristics of basaltic volcanism within back-arc basin. In: K OKELAAR , B. P. & H OWELLS , M. F. (eds) Marginal Basin Geology: Volcanic Associated Sedimentary and Tectonic Processes in Modern and Ancient Marginal Basins. Geological Society, London, 16, 5976. S EYLER , M., P AQUETTE , J.-L., C EULENEER , G., K IENAST , J.-K. & L OUBET , M. 1998. Magmatic underplating, metamorphic evolution, and ductile shearing in a Mesozoic lower crustalupper mantle unit (Tinaquillo, Venezuela) of the Caribbean Belt. Journal of Geology, 106, 3558. S HERVAIS , J. W. 1982. Ti V plots and the petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters, 59, 101 118. S HERVAIS , J. W. 2001. Birth, death, and resurrection: the life cycle of suprasubduction zone ophiolites. Geochemistry, Geophysics, Geosystems, 2 (paper number 2000GC000080). S INTON , C. W., D UNCAN , R. A., S TOREY , M., L EWIS , J. & E STARADA , J. J. 1998. An oceanic ood basalt province within the Caribbean plate. Earth and Planetary Science Letters, 155, 221 235. L ALLEMANT , S ISSON , V. B., E RTAN , I. F. & A VE H. G. 1997. High pressure ( 2000 MPa)

566

M. B. I. WEBER ET AL. deformation, southern Central America and northwestern South America. CASA GPS Observations, 15, 157171. Q11 T SCHANZ , C., M ARVIN , R., C RUZ , J., M ENNERT , H. & C EBULA , E. 1974. Geologic evolution of the Sierra Nevada de Santa Marta. Geological Society of America Bulletin, 85, 269 276. V ALLEJO , C., S PIKINGS , R. A., L UZIEUX , L., W INKLER , W., C HEW , D. & P AGE , L. 2006. The early interaction between the Caribbean Plateau and the NW South American Plate. Terra Nova, 18, 264 269. V ILLAMIL , T. 1999. CampanianMiocene tectonostratigraphy, depocenter evolution and basin development of Colombia and Western Venezuela. Paloegeography, Palaeoclimatology, Paleoecology, 153, 239 275. W INCHESTER , J. A. & F LOYD , P. A. 1977. Geochemical discrimination of different magma series and their differentiation products using immobile elements. Chemical Geology, 20, 325343. ao: remnants of a W RIGHT , J. E. 2004. Aruba and Curac collided Pacic oceanic plateau? Initial geologic results from the BOLIVAR Project. American Geophysical Union, Fall Meeting 2004, abstract no. T33B-1367. lisis petrogra Z APATA , G., W EBER , M. ET AL . 2005. Ana n de la Serran a de Jarara, co de las rocas de alta presio nicas. X Congreso La Guajira y sus implicaciones tecto a Simposio de Geolog a Colombiano de Geolog Regional, 68 69.

1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160

glaucophane-bearing pelitic schist and eclogite from Cordillera de la Costa belt, Venezuela. Journal of Petrology, 38, 6583. S TEIGER , R. H. & J AGER , E. 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359362. S TERN , R. J. 2002. Subduction zones. Reviews of Geophysics, 10; doi: 1029/2001RG000108. CKHERT , B., M ARESCH , W. V. ET AL . 1995. Crustal S TO History of Margarita Island (Venezuela) in detail: constraint on the Caribbean plate-tectonic scenario. Geology, 23, 787 790. T ASSINARI , C. C. G., M EDINA , J. G. C. & P INTO , M. C. S. 1996. Rb Sr and Sm Nd geochronology and isotope geochemistry of central Iberian metassedimentary rocks (Portugal). Geologie en Mijnbouw, 75, 6979. T AYLOR , B. & M ARTINEZ , F. 2003. Back-arc basin basalt systematics. Earth and Planetary Science Letters, 210, 280 297. T HOMPSON , P. M. E., K EMPTON , P. D. ET AL . 2004. Elemental, HfNd isotopic and Geochronological constraints on an island arc sequence associated with the Cretaceous Caribbean plateau: Bonaire, Dutch Antilles. Lithos, 74, 91 116. n geolo gica de ColomT OUSSAINT , J. F. 1996. Evolucio cico. Universidad Nacional de Colombia, bia. Creta n. Medell T RENKAMP , R., K ELLOGG , J. N., F REYMUELLER , J. T. & M ORA , H. P. 2002. Wide plate margin