Late Cretaceous Subduction Initiation on the Eastern Margin of the Caribbean-Colombian Oceanic Plateau: One Great Arc of the Caribbean (?)

Home , Geology of Aruba, La Orchila, Leeward Antilles

Late Cretaceous subduction initiation on the eastern margin of the Caribbean-Colombian Oceanic Plateau: One Great Arc of the Caribbean (?)

James E. Wright and Sandra J. Wyld Department of Geology, University of Georgia, Athens, Georgia 30602, USA

ABSTRACT INTRODUCTION Grenada Basin represent an arc/remnant arc/ back arc basin triad formed during Late Paleo- Detailed mapping on the Leeward Antilles There is widespread agreement that the inte- cene–Eocene arc rifting and back arc basin for- islands of Aruba, Curaçao, Bonaire, and La rior of the Caribbean Plate (Fig. 1) is largely mation (Fig. 1; e.g., Bouysse, 1988). The Lesser Blanquilla has led to a reassessment of their underlain by Pacifi c oceanic crust that was sub- Antilles contain the modern arc produced by stratigraphic, magmatic, and structural sequently thickened by oceanic plateau magma- subduction of Atlantic Ocean fl oor beneath the evolution. In general, each island preserves tism. Bathymetry and seismic imaging indicate eastern Caribbean. its own distinct sequence of geologic events. that the Caribbean Plate is signifi cantly thicker, The Leeward Antilles are isolated subaerial The Cretaceous geology of Aruba and Cura- ~10–20 km in aggregate thickness, than normal exposures of a submarine ridge that extends çao consists of a mafi c igneous complex, oceanic crust (e.g., Diebold et al., 1981; Mauffret from Aruba on the west to La Blanquilla on the long interpreted to represent exposures of and Leroy, 1997; Driscoll and Diebold, 1998). east at the southern terminus of the Aves Ridge the Caribbean-Colombian Oceanic Plateau Geologic and geochemical studies from Deep (Fig. 1). These islands have played a pivotal role (CCOP), intruded by 89–86 Ma arc-related Sea Drilling Project (DSDP) sites, submersible in models for the tectonic evolution of the Carib- plutons and dikes. The rocks on both islands collections from the Beata Ridge (Fig. 1), and bean Plate because they contain onland expo- that are interpreted as remnants of the subaerial exposures of widely dispersed basaltic sures of CCOP rocks and Cretaceous magmatic CCOP underwent a period of subaerial ero- igneous rocks believed to be accreted fragments arc assemblages (Beets et al., 1984; White et al., sion in the Late Cretaceous, but subsequently of the Caribbean Plate, all indicate a within- 1999; Kerr et al., 2003; Thompson et al., 2004). their geologic histories diverge signifi cantly plate oceanic plateau origin (e.g., Donnelly, Many studies have implied or concluded that in terms of their stratigraphic and structural 1973; Donnelly et al., 1990; Kerr et al., 1996; Cretaceous arc rocks in the Greater Antilles, evolution. Mapping on Bonaire has resulted Sinton et al., 1998; White et al., 1999; Revillon Aves Ridge, and Leeward Antilles were origi- in a major revision to the Cretaceous bedrock et al., 2000; Kerr et al., 2009). Following Kerr nally continuous (Beets et al., 1984; Duncan geology. Instead of a single stratigraphic unit et al. (2003), we refer to exposures of the pla- and Hargraves, 1984; Bouysse, 1988; White (Washikemba Formation) the island con- teau rocks as the Caribbean-Colombian Oceanic et al., 1999; Kerr et al., 2003; Thompson et al., tains two stratigraphic units separated by a Plateau (CCOP) throughout the remainder of 2004; Jolly et al., 2006; Pindell et al., 2006) and northwest-trending fault. The southwest side the paper. Relatively recent geochronological formed a single “Great Arc” of the Caribbean of the fault consists of an arc-related Early to investigations (Alvarado et al., 1997; Sinton (Fig. 2; Burke, 1988). Late Cretaceous volcaniclastic section cut by et al., 1997, 1998; Lapierre et al., 1999; Walker The origin of the Caribbean Plate and its shallow level intrusions, whereas the north- et al., 1999; Revillon et al., 2000; Luzieux et al., fringing arc system is controversial. Most work- east side is composed of Early to Late Cre- 2006) suggest the bulk of the CCOP formed in ers interpret the Caribbean Plate and associated taceous epiclastic/hemipelagic strata that are a relatively short amount of time (92–88 Ma; “Great Arc” to have originally formed in the locally cut by small arc-related mafi c intru- Turonian–Coniacian according to the time Pacifi c. Ultimately, subduction of proto-Carib- sions. La Blanquilla represents the southern- scale of Gradstein et al. [2004]), although there bean seafl oor led to insertion of the Caribbean most exposure of the Aves Ridge which is a is evidence for younger magmatic additions Plate between the Americas. Some models sug- remnant arc separated from the modern arc (e.g., Revillon et al., 2000), and, as discussed gest that the plateau originated on oceanic crust of the Lesser Antilles by the Grenada back in more detail below, there is also evidence on subducting to the east beneath the “Great Arc,” arc basin. The bedrock geology consists of Curaçao for an older Albian period of plateau and that arrival of the plateau at the subduct- two Late Cretaceous arc-related plutons. The magmatism . ing boundary in the early Late Cretaceous led geologic evolution of the Leeward Antilles The Caribbean Plate is rimmed on the north, to polarity reversal, trapping the overthickened when combined within a broader context of east, and south by islands of the Greater, Lesser, plateau crust behind an east-facing arc super- Caribbean tectonics leads us to a tectonic and Leeward Antilles, respectively (Fig. 1). imposed on the older west-facing arc (Fig. model involving three distinct arcs rather Most of the Greater Antilles contain magmatic 2A; e.g., Burke et al., 1984; Duncan and Har- than a single “Great Arc” of the Caribbean arc assemblages of Early Cretaceous through graves, 1984; White et al., 1999; Kerr et al., as an explanation for the geodynamic evolu- Eocene age (Pindell et al., 2005, and references 2003; Thompson et al., 2004). In this model, tion of the CCOP and its fringing arc system. therein). The Lesser Antilles, Aves Ridge, and Late Cretaceous magmatism in the “Great Arc”

Geosphere; April 2011; v. 7; no. 2; p. 468–493; doi: 10.1130/GES00577.1; 22 ﬁ gures; 3 supplemental ﬁ les.

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

72.72°W

19.93°N N Greater

Antilles

Ridge

Antilles

Caribbean Plate Ridge

Basin Beata

Aruba Curaçao Aves

Bonaire Grenada Lesser GR LO LB Leeward T Antilles M 11.09°N

Venezuela 60.29°W

Figure 1. Shaded relief physiography of the Caribbean region, showing basic features, locations, and plate boundaries. Aruba, Curaçao, and Bonaire are located at the western end of the Leeward Antilles. Abbreviations as follows: GR—Gran Roque; LO—La Orchilla; LB—La Blanquilla; M—Margarita Island; T—Tobago. Image modiﬁ ed from Smith and Sandwell (1997).

following polarity reversal would have occurred different predictions for geologic relations in form boundary) at ca. 90 Ma, forming a Late on a composite basement consisting of Early the Leeward Antilles. In order to test these Cretaceous magmatic arc that developed inde- Cretaceous arc rocks and the margins of the models, we conducted a comprehensive study pendently of the Cretaceous arc of the Greater CCOP, with evidence (structural or otherwise) of the geology of the large western islands of Antilles. These results also provide a frame- for a change in the location of the plate bound- the Leeward Antilles, Aruba, Curaçao, and work that serves to link the southern Carib- ary in the early Late Cretaceous. Other stud- Bonaire (ABC islands), and the small island of bean Arc rocks to similar Late Cretaceous arc ies present several lines of evidence indicating La Blanquilla to the east (Fig. 1). Our analy- sequences exposed in Ecuador and Colombia. that subduction polarity reversal more likely sis is based on a combination of new detailed In the following sections, we describe geo- occurred in the Early Cretaceous, prior to devel- mapping, stratigraphic and structural stud- logic relations and new data from the ABC opment of the CCOP (e.g., Pindell et al., 2006; ies, geochemical analyses, and geochronol- islands and La Blanquilla. Our focus is on Jolly et al., 2006). In these models, there was ogy on all four islands, and is integrated with their Cretaceous to Paleocene evolution. For no collision, the Aptian–Albian and younger previously published data. The study reveals the most part we do not address the younger “Great Arc” was continuously east-facing, and pronounced differences in the Cretaceous to Paleogene to recent accretion and dispersal of the CCOP developed west of the arc and on the Paleocene geology of the islands that can- Caribbean terranes along the Venezuelan mar- same plate (Fig. 2B). There is also a body of not be explained by either of the prevailing gin. Throughout the text, we use the time scale literature that interprets the Caribbean Plate to end-member models for Caribbean evolution, of Gradstein et al. (2004) for the ages of Creta- have formed essentially in its present location and that are inconsistent with the concept of ceous and Paleogene stage boundaries. Analyti- between the Americas. This in situ model, most a long-lived and continuous “Great Arc” sys- cal methods and data for geochronology (U-Pb recently summarized by James (2009), has been tem. Instead, our analysis suggests that the zircon, 40Ar/39Ar) and geochemistry (major and challenged by numerous Caribbean research- Greater Antilles Arc was terminated at its east- trace elements) are presented as supplemental ers (e.g., Pindell and Kennan, 2009; Maresch ern edge by a transform boundary, that frag- ﬁ les. U-Pb zircon analyses (plutonic and detrital et al., 2009; Stanek et al., 2009; Kerr et al. 2009; ments of the Greater Antilles were translated samples) were obtained either with the sensitive Diebold, 2009; Kennan and Pindell, 2009). into what is now the southern Caribbean realm high resolution ion microprobe-reverse geom- The two Paciﬁ c end-member models for the along this boundary, and that subduction initi- etry (SHRIMP-RG) at the Stanford–U.S. Geo- evolution of the Caribbean Plate make very ated beneath the CCOP (at the former trans- logical Survey Micro-Analytical Center, or by

Geosphere, April 2011 469

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

North ca. 90 Ma North America ca. 85–80 Ma America

Farallon Early Cretaceous Proto-Caribbean plate Late Cretaceous proto-Caribbean “Great Ar

Collision zone c”

“Great Arc”

CCOP forms CCOP crust at ca. 90 Ma

South CCOP collides with South America great arc at ca. 85 America Ma, causing A subduction reversal

B North ca. 90 Ma North ca. 85–80 Ma America America

proto-Caribbean “Great Arc”Proto-Caribbean Trench at ca. 120 Ma) at 90 Ma “Great at Arc” 85 Ma at 110 Ma at 100 Ma at 90 Ma

CCOP forms CCOP crust at ca. 90 Ma

South South America America Trench reversal occurs in the There is no collision; the CCOP Aptian, prior to CCOP formation, crust rides passively in the and is followed by trench rollback upper plate behind the arc

Figure 2. Two prominent models for evolution of the Caribbean Plate, Caribbean-Colombian Oceanic Plateau (CCOP), and “Great Arc” during the Cretaceous. Both models show two time slices, ca. 90 Ma on the left and ca. 85–80 Ma on the right. (A) The “Great Arc” of the Antilles develops above an east dipping subduction zone in the Early Cretaceous. The CCOP mostly forms at ca. 90 Ma on the Farallon Plate which is subducting to the east beneath the “Great Arc.” When the thick plateau crust collided with the “Great Arc” in the Late Cretaceous ( ca. 85–80 Ma), subduction reversal occurred, and subsequent “Great Arc” magmatism occurred above a west-dipping subduction zone. Model based on Burke (1988), Kerr et al. (2003), and Thompson et al. (2004). (B) Subduction reversal occurs in the Aptian, at ca. 120 Ma, prior to development of the CCOP. Post–120 Ma magmatism in the “Great Arc” occurs above a west-dipping subduction zone. The CCOP forms in the upper plate of this subduction zone, behind the arc, at ca. 90 Ma. There is no collision. Model based on Pindell et al. (2006).

laser ablation-multicollector-inductively cou- cal analyses were performed by X-ray ﬂ uores- of detrital hornblende were obtained at the U.S. pled plasma-mass spectrometry (LA-MC-ICP- cence (XRF) and ICP-MS at the Washington Geological Survey Thermochronology Labora- MS) at the University of Arizona LaserChron State University GeoAnalytical Laboratory tory in Denver, Colorado (see Supplemental Center (see Supplemental File 11). Geochemi- (see Supplemental File 22). 40Ar/39Ar analyses File 33).

1Supplemental File 1. Excel fi le of U-Pb data. 2Supplemental File 2. Excel fi le of major- and trace- 3Supplemental File 3. Excel fi le of 40Ar/39Ar data. If you are viewing the PDF of this paper or read- element geochemistry data. If you are viewing the PDF If you are viewing the PDF of this paper or reading it offl ine, please visit http://dx.doi.org/10.1130/ of this paper or reading it offl ine, please visit http:// ing it offl ine, please visit http://dx.doi.org/10.1130/ GES00577.S1 or the full-text article on www.gsapubs dx.doi.org/10.1130/GES00577.S2 or the full-text arti- GES00577.S3 or the full-text article on www.gsapubs .org to view Supplemental File 1. cle on www.gsapubs.org to view Supplemental File 2. .org to view Supplemental File 3.

470 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

ARUBA mapable units (basalt, argillite, and diabase including metamorphic amphibole in outcrops units) named for the dominant rock types within near the Aruba batholith). The metamorphosed Bedrock geology of Aruba consists of each (Fig. 3). The basalt unit consists mostly of spheroidal weathering features are also locally the Late Cretaceous Aruba Lava Formation pillowed to massive basalt fl ows (Fig. 4A), with cut by dikes emanating from the Aruba batho- (ALF) and crosscutting Aruba batholith, over- only rare exposures of stratifi ed rocks includ- lith, and are found exclusively in lower ALF lain unconformably by Eocene limestone and ing basaltic tuff or volcaniclastic mudstone and rocks just below the unconformity. (3) Strata in younger strata (Fig. 3; Westermann, 1932; sandstone. The overlying argillite unit (Fig. 4B) the upper ALF appear to have been deposited in Monen, 1980; Beets et al., 1977, 1996; White consists mostly of thinly bedded, variably silty a subaerial setting: accretionary lapilli are gener- et al., 1999). The Aruba batholith forms most argillite, with less common volcaniclastic silt- ally believed to form in subaerial eruption col- of the basement exposure; the ALF is primar- stone and sandstone, pumiceous lapilli tuff, and umns (e.g., Schumacher and Schmincke, 1995), ily exposed in the central part of the island pebble conglomerate. Graded bedding is com- and sedimentary features of the conglomerate (Fig. 3A). The ALF, which consists largely of mon and indicative of deposition from turbid- unit, including rounded clasts, suggest deposi- weakly metamorphosed mafi c lavas, diabase ity currents. The diabase unit intrudes both the tion in a fl uvial setting and certainly indicate sur- intrusions, and associated volcaniclastic strata, basalt and argillite units (Fig. 3) and consists face erosional processes. (4) Finally, the unusual is widely interpreted to be an exposed part of of a complex assemblage of texturally variable local preservation and patchy outcrop pattern of the CCOP on the basis of geochemical and dikes, sills, and small intrusions. Most common the upper ALF (Fig. 3), in conjunction with the age similarities (Kerr et al., 1997; Sinton et al., are fi ne- to medium-grained, equigranular rocks evidence for subaerial accumulation, suggests 1998; Beets et al., 1984; White et al., 1999). In (Fig. 4C) with clinopyroxene as the dominant that it may have been deposited in canyons or particular, major element, trace element, and mafi c phase. river valleys, only remnants of which are now isotopic analyses (White et al., 1999) show that The upper ALF, which is only locally pres- preserved. We were unable to structure contour mafi c rocks of the ALF are similar to other ana- ent (Fig. 3B), consists of distinctive pyroclastic the contact between the upper and lower ALF lyzed examples of the CCOP (e.g., Sinton et al., and epiclastic strata that overlie units of the because there is not enough topographic relief 1998; Revillon et al., 2000) and are chemically lower ALF along an unconformity (Fig. 3). Two on the contact. This, along with the subsequent distinct from Caribbean arc magmatic associa- units can be distinguished: a locally preserved folding of both the upper and lower ALF, pre- tions. The dominantly tonalitic Aruba batho- basal unit of basaltic tuff and an overlying poly- vents construction of a meaningful cross section lith, in contrast, has geochemical similarities to mictic conglomerate (Fig. 3C). Where present, across the unconformity. subduction-related magmas (Beets et al., 1984; the tuff unit is depositional on the lower ALF; The age of the exposed part of the ALF is con- White et al., 1999; Kerr et al., 2003; Thomp- elsewhere, the conglomerate unit rests directly strained by the presence of ammonites collected son et al., 2004). Together, the ALF and Aruba on the lower ALF. In most places, the tuff unit from a pebbly mudstone within the argillite unit batholith have played a prominent role in tec- consists of well-bedded accretionary lapilli tuff of the lower ALF that have been interpreted as tonic models of the southern Caribbean (e.g., (Fig. 4D), although locally it is dominated by Turonian in age (MacDonald, 1968) and by the Beets et al., 1984; White et al., 1999; Thomp- tuff breccias with abundant lapilli and bombs 89 ± 1 Ma age of the Aruba batholith and associ- son et al., 2004). In order to better understand of basalt scoria. Accretionary lapilli show no ated dikes, which crosscut all units of the ALF the history recorded by these Cretaceous rocks, evidence of fragmentation or abrasion (Fig. (see below). Humphrey (2010) has obtained a we mapped the central part of the island (Fig. 4D). The conglomerate unit consists mostly of preliminary U-Pb micro-zircon date of 97.3 ± 3B) at a scale of 1:25,000. Our new work indi- pebble to cobble conglomerate, with only local 5.2 Ma on a gabbro from the lower ALF. This cates important differences from prior studies interbeds of fi ner-grained strata. Clasts in this date is in agreement with other data that indi- in how the Cretaceous geology of Aruba should unit are generally well-rounded and are derived cate a Late Cretaceous age for the lower ALF be interpreted. entirely from underlying units of the ALF (Fig. but does not help resolve a more precise age as it 4E) including the accretionary lapilli tuff unit. spans the Latest Albian to early Turonian within Aruba Lava Formation The unconformity between the lower and analytical uncertainty. upper ALF refl ects a period of exhumation and Previous mapping of the island (summarized erosion, and a transition from marine deposition Aruba Batholith by Beets et al., 1984, 1996) led to interpretation to subaerial conditions (Fig. 3C), as indicated of the ALF as an unbroken stratigraphic suc- by the following. (1) Diabase and gabbro clasts As mapped by Beets et al. (1996) and White cession consisting of interlayered basalt fl ows, are common in the upper ALF conglomerate et al. (1999), the Aruba batholith is a predomi- pyroclastic and volcaniclastic deposits, and epi- unit (Fig. 4E) and were clearly derived from nantly tonalitic intrusion that postdates deposi- clastic strata, intruded at all levels by diabase erosion of the underlying lower ALF diabase tion, eruption, and intrusion of the ALF. Our sills. Apparent interlayering of submarine and unit. This requires some amount of exhumation new mapping confi rms that the pluton, and subaerial strata formed the basis for conclud- and erosion. (2) There is abundant evidence of abundant associated dikes, cut discordantly ing that the ALF likely accumulated near sea subaerial exposure and weathering along the across all units of the ALF (Fig. 3). The batho- level, on or near the fl anks of an emergent vol- unconformity, based on the presence of meta- lith has been the subject of a number of geo- cano (Beets et al., 1984; White et al., 1999). In morphosed spheroidal weathering features in chronological investigations, including that contrast, our mapping indicates an important lower ALF basalts and diabase that immediately of White et al. (1999) who reported 40Ar/39Ar unconformity on the island that separates the underlie the contact (Fig. 4F). For comparison, ages suggesting intrusion during the interval ALF into two very different stratigraphic pack- modern spheroidal weathering of the diabase 85–82 Ma. We analyzed zircon separates from ages, the entirely submarine lower ALF and the unit on Aruba is shown in Figure 4G. Unlike a sample of the pluton (location shown by the subaerial upper ALF (Fig. 3C). the modern examples, the Cretaceous weathered star in Fig. 3B; data and methods presented in The lower ALF forms the majority of the rinds are composed of metamorphic minerals Supplemental File 1 [see footnote 1]). Based on exposure area and can be divided into three (pumpellyite and chlorite in most places, but also the weighted mean of nine individual analyses,

Geosphere, April 2011 471

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

70°03 Cenozoic cover

′ 1 s 8

u Aruba batholith .6 o

″ e e t

W c 0 5 km a a L 62 t e Aruba Lava Formation Fig. 3B r

ARUBA C 67

″N ′23 7 °3 12 68 83 A 49 72 31 74 76 Caribbean

46 69 area of Figure 6C Sea 70 65 61 66 70 76 26 55 64 44 78 65 73 N 83 76 60 78 61 58 42 56 74 80 63 64 73 52 65 83 71 58 85 82 86 82 72 76 64 64 79 56 67 70 79 42 80 47 73 62 60 74 52 52 52 53 56 63 59 0 500 1000 m 82 88 35 74 82 72 52 47 31 55 73 70 84 86 66 74 60 87 83 86 80 67 66 B 63 71 81 15 71

66 Strike and dip 71 Strike and dip Eastern limit of structure Zircon sample of foliation of bedding reorientation adjacent to locality (Figure 5) the Aruba batholith

Eocene Unconformity Aruba batholith Limestone and associated dikes 89 ± 1 Ma Locally-derived conglomerate Late Turonian regional deformation and metamorphism

F Sandstone L Conglomerate unit ~25 m A

r Tuff unit (thickness Basaltic sandstone, tuff, e exaggerated) p tuff breccia, hyaloclastite

p Subaerial weathering U ORMATION

F Argillite unit Accretionary lapilli tuff A t i V F n Argillite - variably u A L 1000

L silty or tuffaceous e

A Turonian s

r m a A e b Mafic intrusions a B w i o D U Basalt unit L R Massive to pillowed C A basalt flows

Figure 3. Geology of the island of Aruba (see Fig. 1 for location). (A) Simpliﬁ ed geologic map of the island (from Beets et al., 1996). (B) Detailed geologic map of the central part of the island (see location in A), based on this study. (C) Generalized time-stratigraphic column for Mesozoic to Eocene geology of Aruba, based on this study.

472 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

A B C

Figure 4. (A–F) Photographs of D E main units in the Aruba Lava For- mation (ALF). (A) Pillow lavas in the basalt unit of the lower ALF. (B) Thin bedded argillite and siltstone, weakly folded, in the argillite unit of the lower ALF. (C) Diabase dikes from the diabase unit of the lower ALF: ﬁ ner- grained dike (below knife) on left, and coarser-grained dike on right. (D) Bedded accretionary lapilli tuff from the tuff unit of the upper ALF. Surface is perpendicular to bedding and parallel to spaced cleavage (not evident in this view). (E) Pebble conglomerate from the conglomerate unit of the upper ALF. (F) Cretaceous spheroidal weathering features in lower ALF diabase from just below contact with the upper ALF. Cores of diabase (darker rock), surrounded by weathered rinds (lighter rock), are both recrystallized by Cretaceous F G metamorphic mineral assemblages and are consequently both equally resistant to modern weathering. (G) Example of modern spheroidal weathering of lower ALF diabase from nearby exposure, for comparison with F. Note how the weathered rinds in the modern example are much less resistant to erosion than the core stones.

Geosphere, April 2011 473

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

our U-Pb data (originally reported in Wright and 0.08 Wyld, 2004) indicate an emplacement age of KARB-1 89 ± 1 Ma (Fig. 5; Supplemental File 1 [see foot- 0.07 Weighted Mean = 89 ± 1.0 Ma note 1]). This is consistent with 87–90 Ma ages (40Ar/39Ar and U-Pb zircon) obtained recently from the batholith by van der Lelij (2008, 2010). 0.06

Collectively, the data indicate that the Aruba Pb

206

batholith is close in age to the ALF wall rocks. / Chemically the batholith shows similarities 0.05

Pb 120 110 100 90 80 to subduction-related magmas, including nega- 207 70 60 tive Nb anomalies, but isotopically the batholith 0.04 is similar to basalts and diabases of the CCOP, including samples from the ALF (White et al., 1999). On a K-Ca-Na plot, samples from the 0.03 batholith do not follow a typical calc-alkaline trend, but instead plot in the tonalite-trond- hjemite-granodiorite (TTG) fi eld (White et al., 0.02 50 60 70 80 90 100 110 1999). Elevated Sr/Y and modestly depleted heavy rare earth elements (HREE) in some 238U/206Pb samples also imply affi nity of the batholith to TTG as well as Cenozoic adakites (White Figure 5. Terra-Wasserburg plot of individual zircon analyses from et al., 1999). Models for batholith petrogenesis the Aruba batholith. See Figure 3 for sample location. include remelting of CCOP basaltic crust during injection of new plume-related magma, melting of the CCOP during incipient subduction sion of the Aruba batholith. Structures formed a span of <4 m.y. Coincident timing between this (White et al., 1999), or melting of the CCOP during this event include open to tight folds at remarkable vertical journey and the transition in by injection of subduction-generated magmas the outcrop and map scale (Figs. 6B and 6C), magma genesis on the island suggests that the (Thompson et al., 2004). As explained later, we and the foliation, which is axial planar to folds processes are likely related, as we explore in a favor a subduction model. and defi ned in part by preferred alignment of later section. subgreenschist facies metamorphic minerals. Deformation and Metamorphism These structures and metamorphic fabrics are CURAÇAO crosscut sharply by dikes associated with the Deformation and metamorphism of the ALF Aruba batholith (Fig. 3), overprinted by contact Bedrock geology of Curaçao consists of the is well-known and has long been associated metamorphism (hornblende hornfels facies) in Cretaceous Curaçao Lava Formation (CLF) with intrusion of the Aruba batholith (Wester- a narrow zone around the Aruba batholith (see and overlying sedimentary strata of the Late man, 1932; Monen, 1980; Beets et al., 1984, dashed line in Figs. 3B and 6C), and progres- Cretaceous to Early Paleocene (Danian) Knip 1996; White et al., 1999). The principal struc- sively reoriented toward parallelism with the Group and Midden-Curaçao Formation (Fig. 7; tures noted in these previous studies are gener- batholith contact in the thermal aureole (Figs. Beets, 1972; Klaver, 1976; Beets et al., 1977). ally E-W–striking faults (of unspecifi ed sense 6C and 6D). These relations require that batho- These rocks are overlain uncon formably of offset) and apparently spatially associated lith emplacement postdates and is unrelated to by Eocene limestone and younger strata zones of strong foliation (shear zones of White regional deformation of the ALF on Aruba. If (Fig. 7). The CLF, which consists mostly of et al., 1999). the interpreted age of the ammonite collected mafi c submarine lava fl ows and is at least 5 km Our detailed mapping reveals a signifi cantly from the lower ALF is correct, then the timing thick, is widely considered to be an exposed different structural and metamorphic history of deformation is constrained to the Turonian, part of the CCOP and to be correlative with the for the ALF. First, we found little evidence by the Turonian age of the deformed ALF and ALF on Aruba (Beets et al., 1984; Donnelly for any of the faults shown on previous maps; the 89 ± 1 Ma age of the post-tectonic Aruba et al., 1990; Kerr et al., 1996; Sinton et al., most apparent offset contacts on older maps are batholith. 1998; White et al., 1999). The Knip Group actually irregular stratigraphic or intrusive con- and Midden-Curaçao Formation, in contrast, tacts (Fig. 3B). Second, we found no evidence Aruba Summary have no counterpart on Aruba (compare Figs. that foliation in the ALF is restricted to narrow 3 and 7). Finally, Curaçao was not affected zones. Intensity of foliation development varies Results of our study indicate that the Late by regional metamorphism and deformation with rock type: basalt, diabase, and conglom- Cretaceous (Turonian–Coniacian) stratigraphic, until the Late Paleocene (Beets, 1972), at least erate exhibit a variably developed cleavage, structural, and magmatic evolution of Aruba 30 m.y. after the deformation on Aruba. whereas fi ne-grained clastic rocks and tuffs is much more dynamic than previously recog- In order to better understand and interpret the have a more penetrative foliation (Fig. 6A), nized. CCOP magmatism in the Turonian (ALF) Cretaceous evolution of Curaçao, especially in but the foliation is found in all ALF units and was replaced abruptly by arc magmatism at 89 ± relation to that of Aruba, we examined the geol- throughout the mapped area (Fig. 6C). 1 Ma (Aruba batholith). Simultaneously, Aruba ogy of the northwest part of the island, where Our structural studies indicate clear evi- went from a submarine environment to subaerial pre-Eocene units are best exposed, and collected dence for an episode of regional deformation conditions to depths associated with regional samples for geochronology. Our analysis builds and metamorphism that entirely predates intru- deformation and then batholith emplacement in on the detailed mapping of Beets (1972).

474 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation fold 1 axis = 057, 60° Calculated D = N62W, 76 SW = N62W, Average foliation Average foliation (n = 52) 1 of dashed line in map) gillite unit (bedding is horizontal gillite unit (bedding is horizontal Poles to S Poles to bedding (n = 56) and foliation (filled squares) Structural data from area beyond influence of Aruba batholith (east ne-grained strata of the argillite unit. Poles to bedding (filled circles) Structural data from area near Aruba batholith (west of dashed line in map) equal area stereonet projections of structural projections stereonet equal area D of fold. Red dashed line highlights bedding. 74 66 72 N N dip of 52 81 bedding 80 66 Strike and overturned bedding of upright of foliation Strike and dip 42 Strike and dip 52 71 79 86 73 66 79 82 no facing indicators Strike and dip of bedding - 47 71 83 86 82 73 76 80 63 56 74 0 500 m 67 0 500 m 85 64 82 87 61 74 84 52 64 foliation (lower map). Colors for lower ALF units are same as in Figure 3B. Upper ALF ALF 3B. Upper same as in Figure units are ALF lower map). Colors for foliation (lower 35 1 47 83 79 63 70 59 52 60 88 80 64 58 78 55 73 63 86 78 70 76 60 65 42 71 44 63 70 65 82 72 55 58 62 70 66 67 60 74 26 52 61 69 batholith 76 the Aruba ; diagonal from upper right to lower left). (B) Photograph of folded bedding in ﬁ right to lower upper ; diagonal from adjacent to to adjacent batholith 15 of structure the Aruba 1 adjacent to to adjacent Eastern limit Eastern of structure Eastern limit Eastern reorientation reorientation reorientation reorientation 53 52 56 73 83 C 64 46 A B developed cleavage (S developed (C) Expanded maps of the area outlined in Figure 3B, showing bedding (top map) and S outlined in Figure (C) Expanded maps of the area data from the ALF (data shown here includes data from the entire map area of Fig. 3B). Blue dashed line highlights axial trace map area the entire includes data from (data shown here ALF the data from units have been combined (pale orange color), and dikes of Aruba batholith have been removed, for clarity. (D) Lower hemisphere (D) Lower clarity. for Aruba batholith have been removed, units have been combined (pale orange color), and dikes of Figure 6. Figure emphasizing structural relations in the Aruba Lava Formation (ALF). (A) Photograph of thin bedded strata in ar in the emphasizing structural relations 6. Figure Figure in photo) crosscut by a well- in photo) crosscut

Geosphere, April 2011 475

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

12°23′ ″ B 34.1 N CURAÇAO 0 5 km

N W ″ CUR-17 ′10.9 A

69°09 Caribbean Sea

CUR-21 0 5 km CUR-22

N CUR-14

Midden-Curaçao Danian Formation

Lagoen Formation Maastrichtian Quaternary Campanian Sint Christoffel and Seroe Gracia Formations Eocene Turonian (to Knip Group Albian?) Curaçao Lava Formation Geochronology sample locality and number

C Eocene unconformity

younger dikes - limestone leucocratic diorite Jan Kok Member CUR-14 conglomerate,

Danian sandstone, shale MIDDEN- older dikes - shale, siltstone, quartz diorite CURACAO FM. sandstone

shale and siltstone micro-baddeleyite sample Koea Joeda Member volcaniclastic Fm. CUR-21 and 22 sandstone and tuff Lagoen 1000 igneous zircon cherty limestone sample m chert and cherty

KNIP GROUP KNIP mudstone detrital zircon sample breccia and Seroe Gracia Fms. Sint Christoffel Sint Christoffel

Campanian to Maastrichtian 0 basalt and detrital zircon and unconformity diabase hornblende sample 112.7 ± 86.2 ± 0.8 Ma 7.3 Ma CUR-17

detrital hornblende Albain LAVA FM. LAVA sample CURAÇAO

Figure 7. Geology of the island of Curaçao (see Fig. 1 for location). (A) Simplifi ed geologic map of the island (from Beets, 1972). (B) Detailed geologic map of the northwest part of the island (see location in A), modifi ed from Beets (1972), and showing the location of our geochronology samples. (C) Generalized time-stratigraphic column for Cre- taceous to Eocene geology of Curaçao, modifi ed from Beets (1972, 1977) and Klaver (1987), and showing the stratigraphic/crosscutting relations of our geochronology samples.

476 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

Curaçao Lava Formation (CLF) Diorite Intrusions Chemically, the dikes are quite similar to the Aruba batholith. Both produce nearly identi- Bedrock geology of Curaçao is dominated Dikes and sills of dioritic composition are cal patterns, with Ta/Nb anomalies, on a multi- by the CLF (Fig. 7), a sequence of mafi c, sub- locally common on Curaçao, but no larger element diagram normalized to primitive mantle marine lava fl ows, commonly pillowed, with intrusions are present (Beets, 1972). Two gen- (Fig. 9A), and both show moderate depletion minor intercalations of basaltic hyaloclastite eral groups can be distinguished. An older in HREE on a chondrite-normalized REE plot and some diabase sills. Compositionally, the group consists of semicircular dikes (necks of (Fig. 9B). Likewise, Sr/Y values of the dikes CLF consists of picrites to olivine-phyric tho- Beets, 1972) that are <50 m in diameter and and the batholith also overlap (Fig. 9C). REE leiites in the lower part of the formation and intrude the CLF in the northwest part of the patterns and Sr/Y values are similar to those of olivine-phyric tholeiites to plagioclase-clino- island. These have the composition of quartz adakites and suggest melting of a garnet-bearing pyroxene tholeiites in the upper part (Beets, diorite and contain variably abundant pheno- mafi c source. Our interpretation of the data from 1972). Only one nonvolcanic interval has been crysts of plagioclase and hornblende. Another Aruba and Curaçao is that this magmatic epi- found in the CLF, a thin succession of pelagic group of dikes and sills intrudes the Knip sode represents subduction initiation along the limestone and siliceous shale interlayered with Group and Midden-Curaçao Formation but not margin of the CCOP and that mantle-derived hyaloclastites, located in the upper part of the the overlying Eocene and younger strata. These magmas may have ponded in the lower crust of CLF in the southeast part of the island (Beets, younger intrusions consist mostly of deeply the CCOP. Resultant melting of the CCOP led to 1972; Klaver, 1987). weathered leucocratic diorite with hornblende the adakite-like magmatism recorded on Aruba The great thickness of the CLF and paucity phenocrysts. and Curaçao. Thus, Curaçao, like Aruba, records of intercalated or interpillow sediments sug- In an effort to determine whether the older a fundamental change from within-plate CCOP gests that CLF lavas were erupted in a short time dikes could be related to the magmatic event related magmatism to arc related magmatic interval (Beets, 1972; Klaver, 1987). Sinton which generated the nearby Aruba batholith, activity at ca. 89–86 Ma. et al. (1998) reported fi ve 40Ar/39Ar analyses of we collected samples for geochronology and We also collected a sample of one of the CLF basalts; two samples (one collected near geochemistry (for location, see CUR-17 on younger dikes for U-Pb zircon geochronology, the top of the unit and the other near the base) Fig. 7). U-Pb geochronology (data and methods but no zircons were found. Crosscutting rela- yielded plateau ages of 89.5 ± 1 Ma and 88.9 ± contained in Supplemental File 1 [see footnote tions, however, indicate that this group was most 0.8 Ma. These Turonian–Coniacian radio metric 1]) indicates that the older dikes are close in age likely emplaced in the Paleocene (65–52 Ma; dates confl ict with an Albian age defi ned by to the Aruba batholith. Based on the weighted see Beets, 1972; Fig. 7). ammonites from the pelagic interval (Wiedman, mean of 11 individual zircon analyses from 1978), leading Kerr et al. (1997) to suggest one of the dikes, our U-Pb data indicate an Knip Group and Midden- that the ammonites were either misidentifi ed intrusion age of 86.2 ± 0.8 Ma (Fig. 8; Supple- Curaçao Formation or reworked. Humphrey (2010) has obtained a mental File 1 [see footnote 1]). Trace-element 112.7 ± 7.3 Ma U-Pb micro-baddeleyite date data from dike samples (data and methods con- The Knip Group consists of up to 2 km of from a diabase intruding the Upper CLF. This tained in Supplemental File 2 [see footnote 2]) deep marine strata that were deposited uncon- new result is in confl ict with the 40Ar/39Ar ages are compared in Figure 9 with data from the formably on the CLF in the Campanian to and is in general agreement with the age call on Aruba batholith collected by White et al. (1999). Maastrichtian (Beets, 1972). It consists mostly the ammonites. The available age data indicate that plateau magmatism on Curaçao likely signifi cantly predates that on Aruba. 0.08 The CLF is unconformably overlain in most CUR-17 places by the Campanian to Maastrichtian Knip Weighted Mean = 86.2 ± 0.8 Ma Group (Fig. 7). As documented in detail by Beets 0.07 (1972), a variety of features indicate that this unconformity marks a period of uplift and sub- 0.06 aerial exposure of the CLF. First, weathering and

soil formation processes are indicated by discon- Pb 206 tinuous zones of fragmented and mineralized 0.05 CLF just below the contact. In these zones, CLF Pb/ 120 207 110 100 90 80 70 60 rocks display extensive brecciation with gaps between fragments ﬁ lled by carbonate, jasper, 0.04 and iron oxide; these “in-situ breccias,” whose

fragments ﬁ t together like a jigsaw puzzle , 0.03 extend downward and laterally into unbrecci- ated basalt (Beets, 1972). Second, the contact is also locally marked by thin, discontinuous lenses 0.02 of shallow marine limestone that separate the 50 60 70 80 90 100 110 underlying deep marine CLF from the overlying 238U/206Pb deep marine Knip Group (Beets, 1972). These limestones yield fossils of late Santonian to early Figure 8. Terra-Wasserburg plot of individual zircon analyses from Campanian age and most likely reﬂ ect renewed sample CUR-17, one of the quartz diorite dikes intruding the CLF in transgression following exposure. northeastern Curaçao (Beets, 1972). See Figure 7 for sample location.

Geosphere, April 2011 477

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

1000 A B 100 100

10 10

Rock/Chrondrite Rock/Primitive Mantle Rock/Primitive

1 1 CsRbBa Th U Nb Ta K La CePb Pr Sr P Nd Zr SmEu Ti Dy Y Yb Lu La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 60 Curaçao dikes - older group, quartz diorite 50 C Aruba batholith

Sr/Y Figure 9. Geochemical plots comparing the quartz diorite dikes 30 on Curaçao with the Aruba batholith. Aruba batholith data from White et al. (1999). In all ﬁ gures, green triangles are Aruba batholith data, and blue squares are Curaçao dike data. (A) Multielement 20 plot normalized to primitive mantle (Sun and McDonough, 1989). (B) Chondrite-normalized REE diagram (Sun and McDonough, 1989). (C) Y versus Sr/Y plot. 10 9 1113151719 Y (ppm)

of pelagic chert, cherty limestone, and argillite, The overlying Early Paleocene (Danian) crystals with less abundant continentally interspersed at various levels with epiclastic Midden-Curaçao Formation is a clastic suc- derived material such as muscovite, quartz, and volcanogenic deposits (Fig. 7C). The lat- cession, at least 1 km thick, of conglomerate, and quartz-rich metasedimentary lithics (Fig. ter include discontinuous wedges of coarse- sandstone, and shale (Beets, 1972). It reﬂ ects 10C). The geochronological results (Supple- grained breccias derived from erosion of the continued submarine fan deposition of clastics mental File 1 [see footnote 1] are similar for CLF at the base of the group, and turbidite with a mixed continental and arc provenance, both units and consistent with a dual prov- deposits with both continental and arc-derived but it lacks the pelagic and tuffaceous strata enance, one from the continental margin of detritus at higher stratigraphic levels (e.g., found in the underlying Knip Group. South America, as exhibited by early Meso- Lagoen Formation) along with local intervals Samples were collected for detrital zircon zoic, Paleozoic, and Precambrian grains, and of tuffaceous volcaniclastic sandstone (e.g., U-Pb geochronology from volcaniclastic sand- the other from a Cretaceous arc (Fig. 11). The Koea Joeda Member of Lagoen Formation) stones of the Lagoen Formation of the Knip intimate mixing of arc and continental detritus (Fig. 7C; Beets, 1972). Volcanogenic material Group (sample CUR-21) and from siliciclastic (Figs. 10C and 11) indicates a combined arc in the Knip Group is of andesitic composition sandstones of the Midden-Curaçao Formation and continental source region. As discussed and distinctly different from that found in the (sample CUR-14; see Fig. 7 for locations and more fully below, we suggest that the detritus CLF (e.g., hornblende is abundant in the clas- Supplemental File 1 [see footnote 1] for data was derived from the ca. 75 Ma collision zone tics). Siliciclastic detritus includes quartz and and methods). The Midden-Curaçao sample between an arc terrane constructed on the muscovite, plus quartzite, mica schist, and fel- was collected from the Jan Kok member CCOP and the Ecuadorian/Colombian conti- sic plutonic lithics. Based on the stratigraphic (Beets, 1972) of the formation (Fig. 10A) and nental margin of South America (Vallejo et al., analysis of Beets (1972), the Knip Group was contains abundant quartz and detrital mus- 2009; Cardona et al., 2008). deposited during a period of unrest on Curaçao, covite (Fig. 10B). The Lagoen sample was We also obtained 40Ar/39Ar geochronology leading to large thickness variations of units collected from the Koea Joeda member of on detrital hornblende separated from two from the southeast part of the island (very thin) the formation (Beets, 1972), which is a unit samples of the Koea Joeda Member: sample to the northwest (much thicker; Fig. 7), and dominated by immature clastics containing CUR-21, the same sample from which we also rapid lateral facies variations. an abundance of plagioclase and hornblende analyzed detrital zircon, and sample CUR-22,

478 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

CUR-14z A B Figure 10. (A) Field photograph of turbidites of the Jan Kok member of the Midden- Curaçao Formation. (B) Photo- micrograph of siliciclastic sandstone from the outcrop shown in A. Photo is of sample CUR-14 (see Fig. 7 for location). (C) Photo- 0.56 mm A micrograph of volcanic-rich sandstone of the Koea Joeda Member of the Lagoen Forma- tion (Knip Group) with promi- CUR-22 C nent hornblende and plagioclase P crystal fragments as well as siliciclastic continental detritus. Photo is of sample CUR-21 (see Fig. 7 for location). Abbre- H viations : S—siliciclastic lithic; H—hornblende; P—plagioclase. S 0.56 mm

80 CUR-14 and 21 Combined

60 35

30 50 25 Cretaceous ages Figure 11. Probability density plot with histo- 20 gram of detrital zircon ages from samples 40 15 CUR-21 (Knip Group) and CUR-14 (Midden-

Number Number Curaçao Formation) combined. Inset shows 30 10 Cretaceous ages only. See Supplemental File 1 5 (see footnote 1) for data and methods. 20 0 60 80 100 120 140

10 Detrital Zircon Age (Ma)

0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 Detrital Zircon Age (Ma)

which was collected from a separate but simi- Deformation and Metamorphism including the Midden-Curaçao Formation lar sandstone bed higher in the unit (Fig. 7; see (Beets, 1972). This Paleocene event is charac- Supplemental File 3 [see footnote 3] for data The deformation history of Curaçao differs terized by NW-trending folds at various scales, and methods). Both samples yield 40Ar/39Ar markedly from that of Aruba. There is no evi- from the outcrop to the map scale where they plateau and isochron ages that are ca. 74 Ma dence on Curaçao for the Turonian regional control the map pattern of bedrock units (Fig. 7; (Fig. 12; Supplemental File 3 [see footnote 3]), deformation that is so prominent on Aruba. Beets, 1972; Klaver, 1987). No tectonic folia- which corresponds closely to the main peak Instead, Curaçao was affected by a single phase tion was formed, but mineral assemblages indi- of Cretaceous ages in our detrital zircon data of regional deformation and metamorphism cate metamorphism at phrenite-pumpellyite to (Fig. 11, inset). that affected all units older than the Eocene, zeolite grade.

Geosphere, April 2011 479

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

Inverse-isotope correlation diagram 4

A 2 0.0032 CUR-21 C hornblende 0 C K/Ca B -2 0.0024 D E 160 -4 A

140 CUR-21 hornblende F 120 0.0016 J

100 Ar/ Ar 74.0 ± 0.6 Ma 36 40 80 F G H I J 60 A-E 0.0008 Isochron Age = 73.7 ± 0.2 Ma [40Ar/36Ar] = 304 ± 3 40 i I MSWD = 1.6 G

Apparent Age (Ma) ± 39 20 Plateau Age = 74.0 0.6 Ma Steps G through J with 92.3% of ArK H

0 0 0 20 40 60 80 100 0 0.010 0.020 0.030 39 39 40 Cumulative % ArK Released Ar/ Ar Inverse-isotope correlation diagram 0.20

0.15 B CUR-22 D 0.10 0.0030 K/Ca hornblende 0.05 B 0 C

160 -0.05 0.0020 D 140 CUR-22 I 120 hornblende E

Ar/ Ar A 100 36 40 73.0 ± 0.7 Ma 80 0.0010 A-E F G HI 60 Isochron Age = 73.13 ± 0.20 Ma 40 36 [ Ar/ Ar]i = 292 ± 2 H 40 MSWD = 15 F

Apparent Age (Ma) 39 20 ± Steps F through I with 97.6% of ArK Plateau Age = 73.0 0.7 Ma G 0 0 0 20 40 60 80 100 0 0.010 0.020 0.030 39 Cumulative % ArK Released 39Ar/40Ar

Figure 12. Analytical data from detrital hornblende samples CUR-21 and CUR-22 from Koea Joeda Member of the Lagoen Formation (Knip Group). See Figure 7 for location and Supplemental File 3 (see footnote 1) for data and methods. (A) and (B) show 40Ar/39Ar age spectra. (C) and (D) show inverse isotope correlation diagram and isochron ages for the two samples.

Curaçao Summary vide an important record of processes affecting BONAIRE the southern Caribbean during a time when there The older history recorded on Curaçao is is no information preserved on Aruba. Three Bedrock geology on Bonaire consists of similar to that on Aruba although the CLF signifi cant conclusions can be drawn from the Late Cretaceous volcaniclastic, sedimentary, does appear to be older than the ALF. CCOP available data. (1) Arc magmatism in proxim- and intrusive rocks that have been tradition- magmatism occurred in a deep marine setting ity to Curaçao began no earlier than ca. 91 Ma, ally grouped as part of a coherent stratigraphic in the Albian, followed by uplift and subaerial peaked at 74 Ma, and was over by ca. 66 Ma. succession named the “Washikemba Forma- exposure at the end of plateau magmatism, (2) Curaçao made a paleogeographic transition tion” (Pijpers, 1933; Klaver, 1976; Beets et al., and then an abrupt shift to arc magmatism by during the Late Cretaceous from a setting far 1977, 1984; Thompson, 2002; Thompson et al., 89–86 Ma. Unlike Aruba, however, Curaçao removed from any continental sediment infl ux 2004). These rocks, which are overlain uncon- was not affected by any regional deformation in the Albian (during CCOP magmatism) to a formably by Maastrichtian and younger sedi- event during the transition in magma genesis. setting that received sediment infl ux from the mentary strata, outcrop in two separate areas Instead, it experienced subsidence and accumu- South American margin by the latest Cretaceous of the island but are much better exposed and lation of a thick sequence of submarine strata (Lagoen Formation, Knip Group). (3) During the more extensively studied in the northwestern (Knip Group and Midden-Curaçao Formation) latest Cretaceous to Early Paleocene, the tectonic area (Fig. 13). The “Washikemba Formation,” in the latest Cretaceous to Early Paleocene. setting of Curaçao allowed sediment infl ux from as defi ned in prior studies, consists of vol- These Late Cretaceous strata on Curaçao pro- both an arc and the South American continent. canic rocks with magmatic arc geochemistry

480 Geosphere, April 2011

dacitic sills and dikes ″

mostly diorite sills and dikes 8.2

Eocene and younger strata

′

18 68°

locally-derived debris flow deposits

″

1 ′

volcanics (felsic) volcanics 1

°2

Group

8 6 Washikemba Washikemba

Washikemba Washikemba

ed map) that overlap partly in age, are ed map) that overlap partly in age, are

Albian to Cenomanian to Albian oniacian Neogene limestone

Maastrichtian Rincon Fm. C

Paleocene fluvial clastics undivided volcanic, intrusive and sedimentary strata ound only in the Matijs conglomerate unit 1976) (outside dashed boundary). Note our 1976) (outside dashed boundary). Note our FAULT

re locations where Beets et al. (1977) noted the locations where re Albian to to Albian SOEBI BLANCO FORMATION 1976; Thompson, 2002). Detailed geologic map of 1976; “WASHIKEMBA FORMATION” “WASHIKEMBA argillite unit (Aptian or older) chert unit mafic stocks (ca. 112 Ma) (ca. 112 Matijs Group Salinas and Sea Eocene limestone conglomerate unit (Coniacian) Bartol Fault 34 32 2 km 60 60 64 32

29 68 0

68 42 33

47 33 Bartol Fault 42 fault, dashed where inferred 17 48 45 39 46 38 29 36 61 57 28 strike and dip of bedding 38 47 63 N

Inoceramid ammonite area mapped in detail in mapped area fossil locality* U-Pb zircon locality nition of the “Washikemba Formation” (see inset) as consisting of two distinct units (Washikemba and Matijs Groups; see detail and Matijs Groups; Formation” (see inset) as consisting of two distinct units (Washikemba nition of the “Washikemba Figure 13. Geology of the island of Bonaire (see Fig. 1 for location). Inset shows generalized geology of island (from Klaver, Klaver, location). Inset shows generalized geology of island (from (see Fig. 1 for 13. Geology of the island Bonaire Figure ( mapping (inside dashed boundary), and by Beets (1972), et al. (1977), Klaver based on our northwest Bonaire redeﬁ (see area mapped in detail) whose location we can therefore infer outside of the area we mapped in detail. outside of the area infer mapped in detail) whose location we can therefore (see area separated by a fault, and are not part of a continuous stratigraphic succession. Circles in conglomerate unit of Matijs Group a in conglomerate unit of Matijs Group not part of a continuous stratigraphic succession. Circles separated by a fault, and are f are work demonstrates that these rocks with conglomerate and sandstone. Our beds interlayered of distinctive boulder presence

Geosphere, April 2011 481

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

and is thought to have accumulated over a time fragments in the tuffs include various propor- and Supplemental File 1 [see footnote 1] for span from at least the Albian to the Coniacian tions of pumice, felsic volcanic lithics, crystal data and methods). One sample (BN-45) was (Klaver, 1976; Beets et al., 1984; Thompson clasts of quartz and feldspar, and glass shards. collected from a felsic lapilli tuff in the upper et al., 2004). These rocks are widely viewed Sedimentary features indicative of submarine part of the WG; the other sample (BN-26) was as representative of the southern end of a long- deposition from turbidity currents and debris from the large block of rhyodacite in the con- lived Cretaceous “Great Arc” of the Carib- fl ows are well-defi ned, as noted by prior work- glomerate unit at the top of the WG (sampled bean, and have played a pivotal role in tectonic ers (Klaver, 1976; Thompson, 2002). These block shown in Fig. 15F). Both samples gave models of the southern Caribbean region (Beets include normal grading, parallel lamination, and Cenomanian ages: 98.2 ± 0.6 Ma for the tuff et al., 1984; Thompson et al., 2004). crossbedding in the fi ner-grained strata, and a and 94.6 ± 1.4 Ma for the rhyodacite block (Fig. We remapped much of the northwest part of combination of massive bedding to normal and 16; Supplemental File 1 [see footnote 1]). Col- the island and collected samples for geochrono- reverse grading in the coarser-grained strata. lectively, the available data indicate that the WG logic analyses in order to more quantitatively Abundance of pumice, poor sorting, and angu- accumulated over a period from the mid- or late evaluate existing models and for comparison larity of clasts suggests reworking of primary Albian to the Cenomanian. with data from the nearby islands of Curaçao pyroclastic deposits from a nearby arc volcano and Aruba (Fig. 1). Our results indicate that (see also Klaver, 1976; Thompson, 2002). Matijs Group (MG) the “Washikemba Formation” of previous stud- Volcaniclastic strata of the WG are intruded ies actually consists of two very different units by two different suites of hypabyssal rocks The Matijs Group, as defi ned here, can be separated by a fault, and that the Cretaceous (Figs. 13 and 14). Diorite dikes, sills, and small divided into three units, named for the domi- geology of Bonaire bears no relation to that of stocks (Figs. 15C and 15D) are prevalent in nant rock type in each (Fig. 13): an argillite unit Aruba or Curaçao. the lower part of the WG where they crosscut that is locally crosscut by diabase stocks; a con- Based on our mapping, we propose that the and/or surround selvages of volcaniclastic strata glomerate unit that appears to be in depositional term “Washikemba Formation” be abandoned (mostly too small to show at the scale of Figure contact above the argillite unit; and a chert unit and that the rocks formerly assigned to this for- 13). Folds of bedding are locally found adjacent whose stratigraphic relations to the other units mation be separated into two herein-named new to the intrusive contacts, suggesting that intru- is ambiguous. units: the Washikemba Group and the Matijs sion took place before the volcaniclastic strata The Matijs argillite occurs at the base of Group. The distribution of the two new units, were fully lithifi ed (see also Klaver, 1976; the group (Fig. 14). It consists of pelagic and and their stratigraphies, are shown in Figures Thompson, 2002). Elsewhere, particularly in hemipelagic strata, including argillite, sili- 13 and 14. The Matijs Group is dominated by the upper 1500 m of the WG, the volcaniclastic ceous argillite, argillaceous chert, and cherty pelagic and clastic strata and corresponds with strata are intruded by rhyodacite sills (Fig. 15E) limestone, with less common fine-grained the Salina Matijs assemblage of prior studies that typically contain phenocrysts of plagio- feldspathic sandstone and siltstone (Fig. 17A). (Klaver, 1976; Beets et al., 1977; Thompson, clase ± quartz. Strata are generally laminated and thin bedded. 2002). The Washikemba Group, in contrast, is In addition to the rocks described above, we Our mapping indicates that the diabase stocks, dominated by the intermediate to felsic com- also identifi ed a distinctive conglomerate unit at which have narrow contact aureoles and island position volcanogenic and hypabyssal rocks the top of the WG in the northwestern part of the arc trace-element characteristics (Supplemental most commonly associated with the “Washi- map area (Figs. 13 and 14), which was not spe- File 2 [see footnote 2]; Fig. 18), are found exclu- kemba Formation.” In the following sections, cifi cally recognized in prior studies. This unit sively in the argillite unit (Fig. 13). Preliminary we describe the geology of Bonaire using this consists mostly of very poorly sorted conglom- in situ secondary ion mass spectrometry (SIMS) new nomenclature. erate containing subrounded volcanic clasts up dating of seven individual microbaddeleyite to boulder size (5 m diameter; Fig. 15F) that grains from one of these intrusions has yielded a Washikemba Group (WG) appear to be derived entirely from underlying U-Pb age of 111.6 ± 5.1 Ma (Humphrey, 2010). rocks of the WG. The most common clast type The argillite unit is thus Albian or older based The WG comprises the Wecua, Slagbaai, and is a fl ow-banded rhyodacite (shown in Fig. 15F) upon the microbaddeleyite geochronology. Branderis assemblages of prior workers (Klaver, with phenocrysts of plagioclase ± quartz that is The overlying conglomerate unit is charac- 1976; Thompson, 2002). It consists of a thick similar in composition to the rhyodacite sills of terized by the presence of distinctive polymic- (up to 4 km) succession of felsic, submarine the upper WG. These conglomerates occur in tic conglomerate and breccia beds (referred to volcanogenic strata intruded to varying degrees thick (mostly >5 m) beds that are interlayered by prior workers as “boulder beds”; Klaver, by shallow level dikes and sills of intermediate with much less common silty argillite and sand- 1976; Thompson, 2002). Coarser clastics are to felsic composition. These rocks have long stone. Composition, sedimentary features, and commonly very poorly sorted and locally con- been identifi ed as an island arc assemblage map pattern suggest that the conglomerate unit tain boulders up to 2.5 m in length. Most of on the basis of lithology and facies as well as may be a channel deposit cut into and sourced the larger clasts, and some fi ner-grained ones, major, trace-element, and isotope geochemistry from the underlying WG. consist of rock types identical to those in the (Klaver, 1976; Beets et al., 1984; Thompson, Previous age constraints from strata that we underlying argillite unit, from which they were 2002; Thompson et al., 2004). now include in the WG are limited. Fossils from evidently derived (Figs. 17B and 17C). Some Volcanogenic strata throughout the WG are a pelagic interval in tuffs of the lower WG are limestone boulders contain a shallow marine of dacitic to rhyodacitic composition and con- mid- to late Albian (Beets et al., 1977), or ca. fauna including corals and algae (Klaver, sist mostly of interbedded tuff breccia, lapilli 108–100 Ma. Thompson et al. (2004) attempted 1976). The remainder of clasts, both large and tuff, and fi ner-grained ash tuffs (Figs. 15A and 40Ar/39Ar analyses on volcanic rocks and their small, consists of igneous detritus ranging 15B), with local horizons of radiolaria-rich best data suggest an age of ca. 96 ± 4 Ma. We from diabase and diorite to amygdular basalt pelagic strata in the lower part of the group collected two samples for U-Pb geochronol- and plagioclase-phyric volcanics. The upper (Klaver, 1976; Thompson, 2002). Constituent ogy from the WG (see Fig. 13 for location; part of the conglomerate unit also includes

482 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

Eocene

limestone

SOBEI Paleocene BLANCO FM. siliciclastic conglomerate

RINCON FM. Maastrichtian

low grade regional metamorphism, tilting, uplift and exposure biostratigraphic control based on inoceramid fauna

micro-baddeleyite sample chert unit bedding in chert chert unit is strongly folded

conglomerate unit Turonian - shale, Coniacian average bedding in sandstone, conglomerate unit conglomerate N47W, 30NE

111.6 ± 5.1 Ma Matijs Group

MATIJS GROUP MATIJS basalt

1000 argillite unit poles to bedding in strata of the m average bedding in diabase argillaceous chert, argillite, argillite unit stocks N46W, 51NE rare limestone 0

fault

94.6 ± 1.4 Ma average bedding (BN-26) volcanic debris flow, N71W, 41NE conglomerate and sandstone 98.2 ± 0.6 Ma (derived from underlying units) (BN-45)

dacitic intrusions

dioritic intrusions

1000 arc rocks m pelagic intervals N=26 tuffaceous felsic volcaniclastic strata 0 poles to bedding in volcanogenic WASHIKEMBA GROUP WASHIKEMBA zircon sample strata of the Washikemba Group Albian biostratigraphic control based on ammonites

Figure 14. Figure summarizing stratigraphic and structural relations on Bonaire. Relations shown are based primarily on this study and modiﬁ ed from Beets (1972, 1977), Klaver (1976), and Thompson (2002). Structural data is plotted on lower hemisphere equal-area diagrams. Poles to bedding are shown, along with the great circle deﬁ ned by average bedding (strike and dip of average bedding orientation noted) where this is possible to determine. Note that bedding orientations in Matijs Group strata are distinct from those in the Washikemba Group, and that the chert unit of the Matijs Group is strongly folded, unlike any other units (see photograph in Fig. 17D).

Geosphere, April 2011 483

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

A B

C D

Figure 15. Field photographs of the Washikemba Group. (A) Graded bedding in volcanic tuffs, from lapilli tuff in lower part of photograph to fi ner-grained tuff in upper part. (B) Well-bedded fi ne-grained tuffs. (C) Dikes intruding dikes in the diorite unit; dike under hammer is less weathered than dikes above and below, but all are of similar composition. (D) Dioritic dike (weathered rock on left side of photograph) intruding bedded tuffs (gray rocks beneath hammer; bedding orientation is nearly parallel to the base of the photograph). (E) Columnar-jointed dacitic sills; joint planes oriented from upper right to lower left of photograph. (F) Large boulder of fl ow-banded rhyodacite lava in the uppermost conglomerate unit.

484 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

0.08 in the WG (Klaver, 1976; Beets et al., 1977; A Thompson, 2002). This conclusion was based BN-45 in part on the generally NE dip of bedding in 0.07 Weighted Mean = 98.2 ± 0.6 Ma all units (Fig. 13) coupled with the presence of Albian strata in the southwest and Conia- 0.06 cian strata in the northeast. Our new mapping and geo chronol ogy, however, indicates that this contact is not stratigraphic: (1) there are 0.05 no obvious shared rock types between the two

Pb/ Pb 120 110 100 90 80 70 groups, nor is there any evidence of interlayer-

207 206 0.04 ing across the contact; (2) felsic intrusions of the upper WG are absent in the Matijs Group and maﬁ c intrusions in the Matijs Group are 0.03 not found in the WG; (3) conglomerates and breccias in the Matijs Group contain clasts indicative of erosion from the lower part of the 0.02 group, but are lacking in clasts consistent with 50 60 70 80 90 100 238 206 derivation from the underlying WG; (4) the U/ Pb thickness of the argillite unit in the lower Matijs Group varies substantially from north- 0.12 west to southeast, while the conglomerate unit BN-26 B in the underlying Washikemba Group is trun- cated along the contact (Fig. 13); (5) the aver- 0.10 Weighted Mean = 94.6 ± 1.4 Ma age orientation of bedding in the Matijs Group is actually distinctly different from that of the WG, as shown by stereonet plots of bedding 0.08 (Fig. 14); and (6) ﬁ nally, our new geochrono- logic results indicate that the lower part of the Matijs Group is actually older (mid-Albian or

Pb/ Pb 0.06 older) than the apparently underlying rocks

207 206 of the upper WG (Cenomanian). Collectively, these relations suggest that the contact between 120 110 100 90 80 70 0.04 the Washikemba and Matijs Groups is a fault that likely records signifi cant offset. We name this the Bartol fault, for the name of the bay 0.02 at its northwestern end (Fig. 13). More map- 50 60 70 80 90 100 ping is needed to determine what the sense of 238U/206Pb offset is along this fault. Figure 16. Terra-Wasserburg plot of individual zircon analyses from Maastrichtian to Eocene Overlap Sequence igneous rocks of the Washikemba Group. See Figure 13 for sample locations. (A) Sample BN-45; felsic lapilli tuff from volcaniclastic unit of the The Matijs Group is unconformably overlain Washikemba Group. (B) Sample BN-26; fl ow-banded dacite boulder in two localities in northwest Bonaire by shal- (see Fig. 15F) from upper debris fl ow unit of the Washikemba Group. low marine limestone and calcareous sandstone of the mid- to Late Maastrichtian Rincon Forma- tion (Fig. 13; Beets et al., 1977). Farther south- interlayered sandstone, siltstone, and argilla- bedded chert, radiolarion chert, argillaceous east on the island, rocks that have been mapped ceous to cherty limestone beds, and is locally chert, and minor shale (Fig. 17D). This unit is as “Washikemba Formation” (Thompson, 2002) depositional on diabase and pillow basalt. exposed only in Salina Matijs and has unclear and that appear to be similar to what we now Finer-grained strata in this part of the unit have contact relations with other units. It is tightly call the WG, are unconformably overlain by up fi gured prominently in prior studies because folded on the outcrop scale, unlike other mem- to 400 m of fl uvial conglomerate, sandstone, they contain a rich marine fauna including radio- bers of the Matijs Group, and may actually be a and shale of the Soebi Blanco Formation (Fig. laria, foramanifera, and inoceramids indicating large block in the conglomerate unit. 13, inset map; Beets, 1972; Beets et al., 1977). a Coniacian age (Beets et al., 1977). The appar- Clasts in the Soebi Blanco Formation include a ent disparity in age between the argillite unit Contact between the Washikemba and combination of volcanic lithics, likely derived (Albian or older) and overlying conglomeratic Matijs Groups from the WG, and lithics derived from an strata (Coniacian) suggest that the contact may external continental source (presumably north- be an erosional unconformity. Previous workers have considered the rocks ern South America), including felsic gneisses, The chert unit consists of pelagic strata we include in the Matijs Group to be in strati- schists, and quartzites (Beets et al., 1977). Priem including dark gray to tan, laminated to thin- graphic continuity with rocks we include et al. (1986) obtained a ca. 1 Ga U-Pb zircon

Geosphere, April 2011 485

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

A B

Figure 17. Field photographs of the Matijs Group. (A) Fine-grained argillite unit (beneath hammer head) overlain by the conglomerate unit (more resistant rock beneath hammer handle). (B) Typical dark cherty clast (beneath hammer handle) in conglomerate unit. (C) Typical gray limestone clast in conglomerate unit. (D) Folded chert beds from chert unit (axial trace from upper right to lower left).

date from one of the gneissic cobbles. The Soebi Deformation Formations. Offset along the Bartol fault must Blanco Formation is not directly dated, but it is be younger than the Coniacian age of the young- overlain by Eocene limestone and locally con- Unlike Aruba or Curaçao, the principal mani- est rocks of the Matijs Group and older than the tains clasts of the Maastrichtian Rincon Forma- festation of tectonism on Bonaire is tilting of Eocene strata that overlap it (Fig. 13). tion, suggesting that it was likely deposited in strata and faulting in the latest Cretaceous to the Paleocene (Fig. 14; Beets, 1972; Beets et al., Paleocene. Tilting and faulting must be younger Bonaire Summary 1977). Eocene limestone and other shallow than the Cenomanian and Coniacian ages of the marine strata also unconformably overlie the youngest affected rocks in each unit and older The Cretaceous geology of Bonaire is com- Washikemba and Matijs Groups in northwest than the unconformably overlying Maastrich- pletely unlike that of Aruba or Curaçao, and Bonaire (Fig. 13). tian and Paleocene(?) Rincon and Soebi Blanco there is no evidence to suggest that it originated

486 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

Rock/Primitive Mantle Sun and McDonough, 1989 garnet-bearing and plagioclase-free metabasalt 1000 (eclogite?) source. Figure 18. Multielement plot of Based on our study of La Blanquilla, the maﬁ c rocks from Bonaire nor- southern part of the Aves Ridge consists of a malized to primitive mantle (Sun latest Cretaceous to Paleocene arc. This geology 100 and McDonough, 1989). Green is very different from that found on the ABC triangles are from gabbros islands to the west. The only point of similarity in trud ing the Matijs argillite unit is the ca. 76 Ma age of the older pluton which as discussed in text. Blue squares is close to the ca. 74 Ma peak in detrital zircon are data from maﬁ c rocks of 10 ages and the ca. 74 Ma detrital hornblende ages the Matijs Group reported by from the Knip Group and Midden-Curaçao For- Thompson et al. (2004). mation on Curaçao.

1 SUMMARY AND DISCUSSION: Cs Rb Ba Th U Nb Ta K La Ce Pb Pr Sr P Nd Zr Sm Eu Ti Dy Y Yb Lu ABC ISLANDS AND AVES RIDGE (LA BLANQUILLA)

in proximity to the other islands. Instead, its with the islands of the Leeward Antilles. Prior Prevailing models for the Cretaceous evo- record of intermediate to felsic, island arc geochronology of rocks from the Aves Ridge is lution of the southern Caribbean Plate have magmatism spanning the Albian to the Ceno- very limited. K-Ar ages of 57, 58, 65, 67, 78, emphasized the presence of both CCOP rocks manian (WG) is more compatible with the his- and 89 Ma were reported for granitic samples and Early to Late Cretaceous volcanic arc rocks tory of arc magmatism in the Greater Antilles obtained in dredge hauls (Fox et al., 1971): on the islands of the Leeward Antilles, and infer from which we speculate that it was tectoni- although not explicitly stated, the analyses that the islands record interactions between cally derived. The paleogeographic setting of appear to have been performed on whole-rock the CCOP province and the “Great Arc” of the Bonaire changed signifi cantly in the latest samples and the actual data are not published. Caribbean (Fig. 1). Our study shows that there is Cretaceous to Paleocene(?) as the faulted and A U-Pb zircon age of 75.9 ± 0.7 Ma has also considerable diversity in the Cretaceous history tilted deep marine bedrock units were uncon- been determined on a granitoid dredged from of the islands, which is not adequately explained formably overlain by shallow marine strata of the ridge (Neill et al., 2011). by existing tectonic models. In this section we the Rincon Formation and fl uvial strata of the The island of La Blanquilla is located at summarize the key similarities and differences Soebi Blanco Formation. By this time, Bonaire the southern end of the Aves Ridge and con- between the islands and the limitations they was situated in a position to receive sediment tains the only accessible subaerial bedrock pose for geodynamic models of the southern from the South American continental margin, exposures of the ridge (Fig. 1). We therefore Caribbean region. whereas prior to this time it was in an oce- examined the geology of La Blanquilla for CCOP rocks are found on the islands of Aruba anic arc setting removed from the infl uence comparison with the ABC islands and to col- and Curaçao (Aruba and CLFs). Evidence on of continental sedimentation. Within available lect samples for geochronology and geochem- both islands for subaerial exposure and weather- age constraints the Soebi Blanco Formation istry. Previous studies indicated that the island ing at the end stages of magmatism provides a may be temporally equivalent to the Danian is underlain by intrusive rocks that were named new tie linking the formations and suggests the Midden-Curaçao Formation on Curaçao (Figs. the Garanton granodiorite by Maloney (1971) possibility of a regional episode of uplift (ther- 7 and 14), which also contains abundant conti- and later the Garanton tronjhemite by Schubert mal?) in the evolution of the CCOP province. nentally derived detritus. We speculate that the and Moticska (1973). Santamaria and Schubert There is no comparable early Late Cretaceous Soebi Blanco Formation may represent a fl uvial (1974) noted that the tronjhemitic rocks grade history recorded on Bonaire, neither CCOP facies of the Midden-Curaçao Formation, and into tonalitic rocks in the northwestern part magmatism nor subaerial exposure (Fig. 21). that Bonaire was situated in proximity to Cura- of the island and reported K/Ar biotite ages of Arc magmatism affected all three islands, çao by the earliest Paleocene. 64–66 Ma from the intrusive rocks. but at different times. Bonaire was active as an We revisited La Blanquilla and discovered arc from the mid-late Albian (or older) to the LA BLANQUILLA (AVES RIDGE) that it contains two plutons, an older biotite- late Cenomanian (ca. 95 Ma). As previously hornblende granodiorite, and a crosscutting discussed, we found no unequivocal evidence The Aves Ridge is a largely submarine fea- hornblende tonalite, exposed in the northwest- for younger volcanic activity on that island. In ture that forms the eastern boundary of the ern part of the island. Our U-Pb zircon analyses contrast, the fi rst sign of possible arc magmatic Caribbean seafl oor (Fig. 1). It is widely con- (Supplemental File 1 [see footnote 1]; Fig. 19) activity on the other two islands is at 89 Ma sidered to represent a subsided Cretaceous to indicate a crystallization age of 75.5 ± 0.9 Ma (Aruba batholith) and 86 Ma (dikes on Cura- Paleogene remnant arc that rifted in the early for the older granodiorite (Fig. 19A) and 58.7 ± çao). Detrital zircon ages, along with 40Ar/39Ar Paleogene as the extensional Grenada Basin 0.5 Ma for the younger tonalite (Fig. 19B). ages from Maastrichtian to Danian sandstones opened and the locus of arc magmatism shifted Multielement primitive mantle normalized on Curaçao, provide proxy data for the age of eastward to the Lesser Antilles (Fig. 1). In mod- plots indicate that both plutons were formed volcanic source rocks or active volcanism in els that invoke a long-lived “Great Arc” of the in an arc environment (Fig. 20; Supplemen- proximity to the island. These ages indicate a Caribbean (Fig. 2), the Aves Ridge is predicted tal File 2 [see footnote 2]). The younger plu- peak at 74 Ma, with only two grains with Creta- to be underlain by arc igneous rocks of both ton also has very depleted HREE, positive Eu ceous ages older than 91 Ma. From these data, Early and Late Cretaceous age, and to share anomalies, and extremely high Sr/Y ratios (Fig. we conclude that arc magmatism on and/or elements of a common Cretaceous evolution 20), which suggests an origin via melting of a near Aruba and Curaçao spanned the Turonian

Geosphere, April 2011 487

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

0.08 suggesting that it evolved elsewhere during this LBQ-3 time. It is, in fact, much more similar in its history to islands of the Greater Antilles where 0.07 Weighted Mean = 75.5 ±0.9 Ma Early Cretaceous arc volcanism is common and CCOP magmatic rocks are scarce or absent (e.g., 0.06 Pindell et al., 2005, and references therein).

b It was not until the latest Cretaceous to Paleo-

0 gene that the three islands began to merge into

b 0.05 a common tectonic history. During this time,

0 26 100 90 80 70 60 50 there was exhumation of batholithic rocks on Aruba, strata derived in part from South Amer- 0.04 ica were deposited on Bonaire and Curaçao, and all islands were unconformably overlain by 0.03 Eocene limestone (Fig. 21).

TECTONIC ANALYSIS 0.02 60 80 100 120 140 Models for the evolution of the Caribbean

0.08 region include the concept of a “Great Arc” that extended from the Greater Antilles, Aves Ridge, LBQ-5 through the Leeward Antilles and into Colom- 0.07 Weighted Mean = 58.7 ±0.5 Ma bia and Ecuador, which was active in both the Early and Late Cretaceous (Fig. 2; e.g., Burke, 1988). Models differ in their interpretation of 0.06 where the CCOP formed and on what tectonic

P plate, and whether collision of the CCOP with

2 / 0.05 a west-facing “Great Arc” caused subduction

0 reversal and entrapment of the plateau behind 26 100 90 80 70 60 50 a new east-facing portion of the “Great Arc” 0.04 (Fig. 2). These models are necessarily based on limited data because much of the CCOP lies

0.03 underwater and is inaccessible to detailed study. The islands of the Leeward Antilles are thus crucial to tectonic interpretations because they 0.02 contain on-land exposures of the CCOP. 60 80 100 120 140 Aruba, Curaçao, and Bonaire have been par- 238 U/206 Pb ticularly important because of their large size relative to other Leeward islands. Their geology Figure 19. Terra-Wasserburg plot of individual zircon analyses has been combined in tectonic models, but our from intrusive rocks on La Blanquilla. (A) Sample LBQ-3 and study indicates that this approach is only war- (B) Sample LBQ-5. ranted for Aruba and Curaçao; Bonaire has a very different history and likely evolved in a different part of the Caribbean during much of the to Maastrichtian, which entirely postdates arc instead buried by deep marine strata in the Cam- Cretaceous. magmatism on Bonaire. panian to Maastrichtian following a period of Based on our study and analysis, the Creta- There are also considerable differences exposure and/or shallow marine conditions (Fig. ceous evolution of Aruba and Curaçao is not in the depositional setting over time of the 21). Bonaire presents yet another history; con- compatible with prevailing tectonic models. three islands, and in their deformational histo- tinuing marine sedimentation through at least First, these islands contain no record of arc vol- ries. Following subaerial exposure at the end the Coniacian, followed by uplift and tilting, canism prior to the Turonian, and therefore do of CCOP magmatism, Aruba was buried to and resumption of marine deposition only in the not represent the southern end of an Early to depths suffi cient for regional deformation with mid to Late Maastrichtian (Fig. 21). Late Cretaceous “Great Arc” of the Caribbean. foliation development in all units of the ALF, In summary, Aruba and neighboring Cura- Second, models invoking collision of the pla- metamorphism, and batholith emplacement. çao share a similar history of CCOP and arc teau with a west-facing arc, followed by subduc- Available age data indicate that this change in magmatism; they diverge in terms of their Late tion fl ip and establishment of a new east-facing environment may have taken place over a very Cretaceous depositional setting and deforma- arc might appear consistent with the record of short interval of time (<4 m.y.) in the Turonian. tion history, but this can be explained by their Turonian deformation and structural burial Rapid structural burial appears to be the only different locations within an evolving conver- on Aruba, but they are diffi cult to reconcile possible explanation. This record is in contrast gent boundary as discussed in the next section. with the nearly coeval ages of CCOP and arc with that on Curaçao, which was unaffected Bonaire, in contrast, appears to share no com- magmatism on Aruba, or with the lack of any by any Late Cretaceous deformation and was mon Cretaceous history with the other islands, Turonian/Coniacian deformation on Curaçao.

488 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

1000 A B

100 100

Rock/Chondrites

Rock/Primitive Mantle 1

0.1 La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Cs Rb Ba Th U Nb Ta K La Ce Pb Pr Sr P Nd Zr Sm Eu Ti Dy Y Yb Lu

400 C 300 Figure 20. Geochemistry of intrusive rocks on La Blanquilla. Blue squares are from the older intrusion ( ca. 76 Ma) and green triangles are from the younger intrusion ( ca. 59 Ma). (A) Multielement 200

Sr/Y plot normalized to primitive mantle (Sun and McDonough, 1989). (B) Rare earth element plot normalized to chondrite (Sun and McDonough, 1989). (C) Sr/Y versus Y plot. 100

0 0 10203040506070 Y

Finally, there is no evidence on Aruba, Curaçao, arc granitoid intrudes rocks correlated with the Late Cretaceous Geodynamic Model Bonaire, or La Blanquilla for construction of a CCOP (Villagomez et al., 2008). It appears for Northwestern South America and Late Cretaceous arc on an amalgamated base- from these data subduction initiation along the the Caribbean ment of CCOP and Early Cretaceous arc rocks. CCOP in Ecuador and Colombia was tempo- In order to fully develop a geodynamic model rally equivalent to the same event described here We begin in the Late Albian (Fig. 22A) with for the southern Caribbean we fi rst review Late on Aruba and Curaçao (Fig. 22A). Thus Aruba a northwest-trending Greater Antilles arc ter- Cretaceous tectonics of Ecuador and Colombia. and Curaçao appear to be a northern continua- minated along its eastern boundary by a south- tion of the Ecuadorian and Colombian arc ter- west-trending Subduction Termination Edge Late Cretaceous Subduction Initiation rane that was constructed on the CCOP. In the Propagator (STEP; Govers and Wortel, 2005) and Arc-Continent Collision in Ecuador Campanian/Maastrichtian to early Paleogene fault that is also linked at a triple junction to and Colombia the arc collided with the Ecuadorian and Colom- spreading in the proto-Caribbean. The STEP bia margins (Vallejo et al., 2009; Cardona et al., lengthens to the north as rollback occurs along The model we develop for the geodynamic 2008). Following collision, subduction polar- the Greater Antilles subduction zone. We inter- evolution of the southern Caribbean is sig- ity reversed and the collided arc and its CCOP pret Bonaire as a part of the Greater Antilles nifi cantly infl uenced by recent investigations basement were partially subducted beneath the Island Arc. We also hypothesize that the Late in Ecuador and Colombia. In Ecuador the ca. South American margin, resulting in the estab- Cretaceous Peruvian trench transitioned from 85–72 Ma oceanic arc rocks of the Rio Cala lishment of a postcollisional latest Cretaceous/ a subduction boundary to a transform bound- Group overlie the equivalent of the CCOP which Paleogene magmatic arc on the South American ary to the northwest due to oblique conver- has been dated at ca. 88 Ma (Luzieux et al., continental margin (Cardona et al., 2009). The gence. The CCOP has already partly formed 2006; Vallejo et al., 2006, 2009). In addition, Aruba/Curaçao section of the arc escaped colli- by this time based upon the Albian ammonites the plateau rocks are intruded by arc related plu- sion as this arc segment was located to the north within the CLF as well as a 112.7 ± 7.3 Ma tonic rocks locally dated at ca. 85 Ma in Ecuador and east of the Ecuadorian and Colombian mar- U-Pb microbaddeleyite date from the CLF (Vallejo et al., 2009). In Colombia, a ca. 91 Ma gins in the proto-Caribbean seaway. (Humphrey , 2010).

Geosphere, April 2011 489

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

BONAIRE Age (Ma) ARUBA CURAÇAO Eo- Middle 48.6 cene Early 55.8 deformation Paleo- exhumation of cene 61.7 Aruba batholith continental Sobei Danian 65.5 M-C Fm. arc and detritus Blanco Fm. Maastri- continental chtian Knip detritus 70.6 Group Rincon Fm. Cam- panian 83.5 deformation Santonian Washikemba Group Matijs Group 85.8 Aruba batholith 86.2 ± 0.8 Ma dikes

Senonian Coniacian 89 ± 1 Ma 89.3 deformation Turonian upper subaerial

Late Cretaceous Late exposure

and weathering fault 93.5 expanded time scale deep 94.6 ± 1.4 Ma Cenomanian Fm. lower marine Curacao Lava Fm. 98.2 ± 0.6 Ma 99.6 Lava Aruba ? ? ? ? CCOP 89 Ma Albian 112.7 ± 7.3 Ma deep 89 Ma Early Creta-c eous 112.0 marine 111.6 ± 5.1 Ma Arc Intrusions Locally-derived conglom- Turbidites (v-pattern for ? ? ? ? erate, and accretionary volcanic component) Argillite lapilli tuff (basaltic) Locally-derived volcanic breccia Limestone Locally-derived conglomerate and Diabase breccia, siltstone, and limestone Arc Volcanics, and Hemipelagic Pillow Basalt Hypabyssal Intrusions chert, argillite Fluvial conglomerate 40Ar/39Ar plateau age (Sinton et al., 1998) U-Pb zircon sample detrital zircon sample micro-baddeleyite sample

Figure 21. Time stratigraphic columns comparing the geologic evolution of Aruba, Curaçao, and Bonaire from the late Early Cretaceous to the Eocene. Time scale is from Gradstein et al. (2004).

By ca. 89 Ma subduction was initiated along zone prior to subduction initiation in the Cam- involved in the collision (Fig. 22C). For exam- the southeastern margin of the CCOP (Fig. 22B). panian. We view the STEP/Greater Antilles arc ple, Curacao was a deep marine basin during the This event is the initiation of the Ecuadorian- intersection as a complicated zone from which collision where turbidites derived from a vol- Colombian-Leeward Antilles arc (ECLA) built fragments of the Greater Antilles like Bonaire canic arc and continental margin source were largely upon a basement of the CCOP. The ALF (Fig. 22B) were transferred to the proto-Carib- accumulating during the Campanian/Maas- of Aruba was located on the southeastern edge bean Plate. Thus Bonaire was originally located trichtian and Early Paleocene (see Figs. 7 and of the plateau in the vicinity of the STEP and near the subduction/transform boundary at the 10), as previously summarized. The occurrence was partially subducted, which accounts for Cre- eastern end of the Greater Antilles arc and subse- of abundant detrital grains in the 1.0–1.2 Ga taceous ductile deformation of the ALF. Eventu- quently transferred to the proto-Caribbean Plate interval and a signiﬁ cant number of Triassic and ally, the ALF was transferred to the hanging wall as the Greater Antilles migrated past the north- Late Paleozoic grains all indicate an Andean of the subduction zone where the Aruba batho- western South American margin (Fig. 22B). The source for these detrital components as rocks of lith was emplaced. On the other hand, the CLF presence of coarse ﬂ uvial continental detritus these ages are present in the Andes of Colombia of Curaçao was located away from the newly in the Soebi Blanco Formation indicates that (e.g., Kroonemberg, 1982; Restrepo-Pace et al., initiated subduction zone (Fig. 22B) and did not Bonaire became attached to continental South 1997; Cordani et al., 2005; Vinasco et al., 2006; undergo deformation associated with subduction America by the latest Cretaceous/early Paleo- Molina et al., 2006). Thus, we place Curaçao initiation. The Late Cretaceous arc intrusions on gene (Fig. 22C). in a position (Fig. 22C) to receive sediment Aruba and Curaçao represent the northernmost Oblique collision in the Campanian (ca. derived from the collision zone between the col- exposures of the ECLA arc, while the region 75 Ma), as previously discussed, led to the lided part of the ECLA arc and the continental between Aruba and Curaçao and the Greater demise of arc magmatism along the ECLA margin. Following collision of the Ecuadorian Antilles arc was maintained as a transform arc, and the arc and its basement (CCOP) were and Colombian part of the ECLA arc, oblique boundary at this time. The section of the STEP translated north along the oblique collision zone convergence along the collision boundary between the Greater Antilles arc and the ECLA (Fig. 22C). Aruba and Curaçao were located apparently resulted in subduction of the CCOP arc represents the ancestral Aves subduction too far north along the ECLA arc to have been beneath the South American margin which led

490 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

ALBIAN (ca.100 Ma) TURONIAN (ca. 89 Ma) proto- Caribbean proto-Caribbean

GA B GA B CCOP C S.A. S.A. A Figure 22. Geodynamic model for the C ECLA Cretaceous to Paleocene plate tectonic evolution of northern South America CCOP and the southern Caribbean in four time sketches: (A) Albian ( ca. 100 Ma), (B) Turonian ( ca. 89 Ma), (C) Cam- panian ( ca. 75 Ma), and (D) Late Paleocene ( ca. 59 Ma). Abbreviations: A—Aruba; B—Bonaire; C—Curaçao; A PA B PA CCOP—Caribbean-Colombian Oceanic P.A. Plateau; ECLA—ECLA arc; GA— Greater Antilles arc; LB—La Blan- quilla; MAA—Middle American arc, Panama, and Costa Rica; PA—Peru- CAMPANIAN (ca. 75 Ma) LATE PALEOCENE (ca. 59 Ma) vian arc; PCA—postcollisional arc; Collision with Bahama Banks S.A.—South America; STEP—STEP. Colors refer to the following: green— CCOP; red—active arc; blue—collided GA arc. Other symbols: closed triangles GA indicate hanging wall of subduction AVES AVES zone; open triangles indicate hanging LB ARC CCOP wall of postcollisional subduction zone. LB ARC CCOP See text for discussion. B C B AC A obliqueS.A. oblique S.A. collision convergence PCA

MAA S.A. S.A. MAA

C P.A.PA D P.A.PA

to the formation of a postcollisional arc (PCA, tons on La Blanquilla may approximate the We suggest that the Cretaceous Great Arc of Fig. 22D; e.g., Cardona et al., 2009). initiation of and termination of arc magmatism the Caribbean (e.g., Burke, 1988) may instead Finally, we hypothesize that following colli- during the Campanian and early Paleogene, have evolved as three arcs now represented by sion, subduction propagated to the north along respectively. The 59 Ma pluton on La Blanquilla the Greater Antilles, the Aves Ridge, and the the STEP producing the Aves Arc (Fig. 22D). was emplaced shortly before the opening of the ECLA arc. This geodynamic model is offered In the absence of additional data from the Aves Grenada Basin when the Aves arc became an as a testable framework in which to view Ridge, we hypothesize that the two dated plu- inactive remnant arc. the Late Cretaceous tectonic evolution of the

Geosphere, April 2011 491

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Wright and Wyld

southern Caribbean region. Finally, progress in Map of Aruba: Foundation for Scientifi c Research in Gran Roque: Implications for the temporal evolution interpreting the tectonic signifi cance of other the Caribbean Region, scale 1:50,000. of the Caribbean large igneous province (CLIP) and Bouysse, P., 1988, Opening of the Grenada back-arc Basin early arc magmatism [M.S. thesis]: Athens, University problematic southern Caribbean terranes such and evolution of the Caribbean Plate during the Meso- of Georgia, 86 p. as Tobago, the Villa de Cura Complex and Las zoic and early Paleogene: Tectonophysics, v. 149, James, K.H., 2009, In situ origin of the Caribbean: Dis- p. 121–143, doi: 10.1016/0040-1951(88)90122-9. cussion of data, in James, K.H., Lorente, M.A., and Hermanas volcanics of mainland Venezuela, and Burke, K., 1988, Tectonic evolution of the Caribbean: Annual Pindell, J.L., eds., The origin and evolution of the additional bedrock exposures of the Leeward Review of Earth and Planetary Sciences, v. 16, p. 201– Caribbean Plate: Geological Society [London] Special Antilles (Gran Roque and La Orchila; Fig. 1) 230, doi: 10.1146/annurev.ea.16.050188.001221. Publication 328, p. 77–125. Burke, K., Cooper, C., Dewey, J.F., Mann, P., and Pindell, Jolly, W.T., Lidiak, E.G., and Dickin, A.P., 2006, Role of might result when viewed within the context of J.L., 1984, Caribbean tectonics and relative plate crustal melting in petrogenesis of the Cretaceous Water the proposed model. motions, in Bonini, W.E., Hargraves, R.B., and Island Formation (Virgin Islands), northeast Antilles Shagam , R., eds., The Caribbean-South American Plate island arc: Geologica Acta, v. 4, p. 7–33. ACKNOWLEDGMENTS Boundary and Regional Tectonics: Boulder, Colorado, Kennan, L., and Pindell, J.L., 2009, Dextral shear, terrane Geological Society of America Memoir 162, p. 31–63. accretion and basin formation in the Northern Andes: This work was supported by NSF grants EAR Cardona, A., García-Casco, A., Ruiz, J., Valencia, V., Busta- Best explained by interaction with a Pacifi c-derived mante, C., Garzón, A., Saldarriaga, M., and Weber, Caribbean Plate?, in James, K.H., Lorente, M.A., and 0087361 and EAR 067533. This is a contribution to M., 2008, Late Cretaceous Caribbean–South America Pindell, J.L., eds., The Origin and evolution of the the Continental Dynamics BOLIVAR project. We par- interactions: Insights from the metamorphic record Caribbean Plate: Geological Society [London] Special ticularly acknowledge conversations with Jim Pindell of the NW Sierra Nevada de Santa Marta, Colombia, Publication 328, p. 487–531. who has patiently educated us on Caribbean geol- in Caribbean Geological Conference, 18th (Santo Kerr, A.C., Pearson, D.G., and Nowell, G.M., 2009, Magma ogy for many years. We also thank Dirk Beets (now Domingo, Republica Dominicana, March 24–29), source evolution beneath the Caribbean oceanic pla- deceased) and Gerard Klaver who generously gave Abstracts and Program: Sociedad Dominicana de Geo- teau: New insights from elemental and Sr-Nd-Pb-Hf of their knowledge of the ABC islands and supplied logia, p. 12. isotopic studies of ODP Leg 165 Site 1001 basalts, in us with theses from the University of Utrect. We are Cardona, A., Valencia, V., Bayona, G., Jaramillo, C., Ojeda, James, K.H., Lorente, M.A., and Pindell, J.L., eds., G., and Ruiz, J., 2009, U/Pb LA-MC-ICP-MS zircon The origin and evolution of the Caribbean Plate: Geo- also grateful to Pat Thompson who supplied us with geochronology and geochemistry from a postcolli- logical Society [London] Special Publication 328, pertinent unpublished portions of her Ph.D. disserta- sional biotite granite of the Baja Guajira basin, Colom- p. 809–827. tion on the island of Bonaire and to Rosalind White bia: Implications for Late Cretaceous and Neogene Kerr, A.C., Tarney, J., Marriner, G.F., Klaver, G.T., Saunders, who provided us with a copy of the geologic map of Caribbean-South American tectonics: Journal of Geol- A.D., and Thirlwall, M.F., 1996, The geochemistry and Aruba by Beets et al. (1996). Transportation, via ship, ogy, v. 117, p. 685–692. petrogenesis of the Late Cretaceous picrites and basalts to the Venezuelan military base on La Blanquilla was Cordani, U., Cardona, A., and Jiménez, D.M., 2005, Geo- of Curaçao, Netherlands Antilles: A remnant of an oce- arranged by the Fundacion Venezolana de Investiga- chronology of Proterozoic basement inliers in the anic plateau: Contributions to Mineralogy and Petrol- ciones Sismologicas (FUNVISIS) and provided by Colombian Andes: Tectonic history of remnants of a ogy, v. 124, p. 29–43, doi: 10.1007/s004100050171. fragmented Grenville belt, in Vaughan, A.P.M., Leat, Kerr, A.C., Tarney, J., Marriner, G.F., Nivia, A., and the Armada de Venezuela. Luis Melo (FUNVISIS) P.T., and Pankhurst, R.J., eds., Terrane Processes of the Saunders , A.D., 1997, The Caribbean-Colombian Cre- acted as liaison between us and the military person- Margins of Gondwana: Geological Society [London] taceous igneous complex: The internal anatomy of an nel onboard the ship and on La Blanquilla. Gustavo Special Publication 246, p. 329–346. oceanic plateau, in Mahoney, J.J., and Coffi n, M.F., Rodriguez, the Comandante of the La Blanquilla Diebold, J., 2009, Submarine volcanic stratigraphy and the eds., Large igneous provinces: Continental, oceanic, military base, provided lodging, island transportation, Caribbean LIP’s formational environment, in James, and planetary fl ood volcanism: American Geophysical and hospitality during our stay. The meeting hosted K.H., Lorente, M.A., and Pindell, J.L., eds., The Ori- Union Monograph 100, p. 123–144. by Andrew Kerr, Jim Pindell, Iain Neill, and Alan gin and evolution of the Caribbean Plate: Geological Kerr, A.C., White, R.V., Thompson, P.M.E., Tarney, J., and Hastie at Cardiff University, Wales in September of Society [London] Special Publication 328, p. 799–808. Saunders, A.D., 2003, No oceanic plateau—No Carib- Diebold, J.B., Stoffa, P.L., Buhl, P., and Truchan, M., 1981, bean Plate? The seminal role of an oceanic plateau in 2009 contributed signifi cantly to our knowledge of Venezuelan basin crustal structure: Journal of Geo- Caribbean Plate evolution, in Bartolini, C., Buffl er, Caribbean tectonics and in particular made us aware physical Research, v. 86, p. 7901–7923, doi: 10.1029/ R.T., and Blickwede, J.F., eds., The circum-Gulf of of research projects in Ecuador and Colombia that JB086iB09p07901. Mexico and Caribbean region; Hydrocarbon habitats, are pertinent to the tectonic model proposed in this Donnelly, T.W., 1973, Late Cretaceous basalts from the basin formation, and plate tectonics: Tulsa, Ameri- paper. Discussions with Roelant van der Lelij, Diego Caribbean, a possible fl ood basalt province of vast size: can Association of Petroleum Geologists Memoir 79, Villagomez, Iain Neill, Agustin Cardona, and Andrew Eos (Transactions, American Geophysical Union), p. 126–168. Kerr have signifi cantly infl uenced our ideas on Carib- v. 54, p. 1004. Klaver, G.T., 1976, The Washikemba Formation, Bonaire bean tectonics. Reviews by Art Snoke, Kevin Burke, Donnelly, T.W., Beets, D.J., Carr, J.J., Jackson, T., Klaver, [M.S. thesis]: University of Utrecht, The Netherlands, G.T., Lewis, J., Maury, R., Schellenkens, H., Smith, 128 p. Associate Editor Bob Hatcher, and particularly Jim A.L., Wadge, G., and Westercamp, D., 1990, His- Klaver, G.T., 1987, The CLF: An ophiolitic analogue of the Pindell, signifi cantly improved the manuscript. tory and tectonic setting of Caribbean magmatism, in anomalous thick layer 2B of the mid-Cretaceous oce- Dengo, G., and Case, J.E., eds., The Caribbean Region: anic plateaus in the western Pacifi c and central Carib- REFERENCES CITED Boulder, Colorado, Geological Society of America, bean [Ph.D. thesis]: University of Amsterdam, The The Geology of North America, v. H., p. 339–374. Netherlands, 168 p. Beets, D.J., 1972, Lithology and stratigraphy of the Cre- Driscoll, N.W., and Diebold, J.B., 1998, Deformation Kroonemberg, G.S., 1982, A Grenvillian granulite belt in taceous and Danian succession of Curaçao [Ph.D. of the Caribbean region: One plate or two?: Geol- the Colombian Andes and its relations to the Guyana thesis ]: University of Utrecht, The Netherlands, 153 p. ogy, v. 26, p. 1043–1046, doi: 10.1130/0091-7613 Shield: Geologie & Mijnbouw, v. 61, p. 325–333. Beets, D.J., 1977, Cretaceous and Early Tertiary of Cura- (1998)026<1043:DOTCRO>2.3.CO;2. Lapierre, H., Dupuis, V., de Lépinay, B.M., Bosch, D., çao, in 8th Caribbean Geological Conference; Guide to Duncan, R.A., and Hargraves, R.B., 1984, Plate tectonic Monié, P., Tardy, M., Maury, R.C., Hernandez, J., the Field Excursions on Curaçao, Bonaire and Aruba: evolution of the Caribbean region in the mantle ref- Polvé, M., Yeghicheyan, D., and Cotton, J., 1999, Late Amsterdam, GUA Papers of Geology, no. 10, p. 7–17. erence frame, in Bonini, W.E., Hargraves, R.B., and Jurassic oceanic crust and Upper Cretaceous Caribbean Beets, D.J., MacGillavry, H.J., and Klaver, G.T., 1977, Geol- Shagam , R., eds., The Caribbean-South American plate plateau picritic basalts exposed in the Duarte Igneous ogy of the Cretaceous and early Tertiary of Bonaire, boundary and regional tectonics: Boulder, Colorado, Complex, Hispaniola: The Journal of Geology, v. 107, in 8th Caribbean Geological Conference; Guide to Geological Society of America Memoir 162, p. 81–93. p. 193–207, doi: 10.1086/314341. the Field Excursions on Curaçao, Bonaire and Aruba: Fox, P.J., Schreiber, E., and Heezen, B.C., 1971, The Geol- Luzieux, L.D.A., Heller, F., Spikings, F., Vallejo, C.F., and Amsterdam, GUA Papers of Geology, no. 10, p. 18–28. ogy of the Caribbean crust: Tertiary sediments, granitic Winkler, W., 2006, Origin and Cretaceous tectonic his- Beets, D.J., Maresch, W.V., Klaver, G.T., Mottana, A., and basic rocks from the Aves Ridge: Tectonophysics, tory of the coastal Ecuadorian forearc between 1°N and Bocchio , R., Beunk, F.F., and Monen, H.P., 1984, v. 12, p. 89–109, doi: 10.1016/0040-1951(71)90011-4. 3°S: Paleomagnetic, radiometric and fossil evidence: Magmatic rock series and high-pressure metamor- Govers, M., and Wortel, M.J.R., 2005, Lithosphere tearing at Earth and Planetary Science Letters, v. 249, p. 400– phism as constraints on the tectonic history of the STEP faults: Response to edges of subduction zones: 414, doi: 10.1016/j.epsl.2006.07.008. southern Caribbean, in Bonini, W.E., Hargraves, R.B., Earth and Planetary Science Letters, v. 236, p. 505– MacDonald, W.D., 1968, Communication, in Status of geo- and Shagam, R., eds., The Caribbean-South Ameri- 523, doi: 10.1016/j.epsl.2005.03.022. logical research in the Caribbean 14, University of can plate boundary and regional tectonics: Boulder, Gradstein, F.M., Ogg, J.G., and Smith, A.G., 2004, A Geo- Puerto Rico, p. 40. Colorado, Geological Society of America Memoir 162, logic Time Scale 2004: Cambridge, UK, Cambridge Maloney, N.J., 1971, Geologia de la isla de La Blanquilla y p. 95–130. University Press, 589 p. notas sobre el archipielage de Los Hermanos, Venezuela Beets, D.J., Westerman, J.H., De Buisonje, P.H., Monen, Humphrey, C.G., 2010, In-situ U-Pb secondary ion mass oriental: Acta Cientifi ca Venezolana, v. 22, p. 6–10. H.P., Stienstra, P., Klaver, G.T., Ruiz, A.V., Curet, spectrometry (INSIMS) geochronology from the Lee- Maresch, W.V., Kluge, R., Baumann, A., Pindell, J.L., E.A., White, R.V., and Fouke, B.W., 1996, Geologic ward Antilles islands of Aruba, Curaçao, Bonaire, and Kruckhans-Lueder, G., and Stanek, K., 2009, The

492 Geosphere, April 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021 Late Cretaceous subduction initiation

occurrence and timing of high-pressure metamorphism Science Letters, v. 150, p. 427–441, doi: 10.1016/ Vallejo, C., Spikings, R.A., Winkler, W., Luzieux, L., Chew, on Margarita Island, Venezuela: A constraint on Carib- S0012-821X(97)00091-5. D., and Page, L., 2006, The early interaction between bean-South Anmerica interaction, in James, K.H., Revillon, S., Hallot, E., Arndt, N.T., Chauvel, C., and the Caribbean Plateau and the NW South American Lorente, M.A., and Pindell, J.L., eds., The Origin and Duncan , R.A., 2000, A complex history for the Carib- Plate: Terra Nova, v. 18, p. 264–269. evolution of the Caribbean Plate: Geological Society bean plateau: Petrology, geochemistry, and geochronol- Vallejo, C., Winkler, W., Spikings, R.A., Luzieux, L., Heller, [London] Special Publication 328, p. 705–741. ogy of the Beata Ridge, South Hispaniola: The Journal F., and Bussy, F., 2009, Mode and timing of terrane Mauffret, A., and Leroy, S., 1997, Seismic stratigraphy of Geology, v. 108, p. 641–661, doi: 10.1086/317953. accretion in the forearc of the Andes in Ecuador, in and structure of the Caribbean igneous province: Tec- Santamaria, F., and Schubert, C., 1974, Geochemistry and Kay, S.M., Ramos, V.A., and Dickinson, W.R., eds., tono physics, v. 283, p. 61–104, doi: 10.1016/S0040- geochronology of the southern Caribbean-northern Backbone of the Americas: Shallow Subduction, Pla- 1951(97)00103-0. Venezuela plate boundary: Geological Society of teau Uplift, and Ridge and Terrane Collision: Boulder, Molina, A.C., Cordani, U.G., and MacDonald, W., 2006, America Bulletin, v. 7, p. 185–198. Colorado, Geological Society of America Memoir 204, Tectonic correlations of pre-Mesozoic crust from the Schubert, C., and Moticska, P., 1973, Reconocimiento geo- p. 197–216. northern termination of the Colombian Andes, Carib- logico de las islas venezolanas en el Mar Caribe, entre van der Lelij, R., 2008, Thermochronology and tectonics of bean region: Journal of South American Earth Sci- Los Roques y Los Testigos (Dependencias Federales): the Dutch Antilles: Aruba and Bonaire [M.S. thesis]: ences, v. 21, p. 337–354. Acta Cientifi ca Venezolana, v. 23, p. 210–223. Universite De Geneve, Switzerland, 34 p. Monen, H.P., 1980, The Aruba Lava Formation [M.S. Schumacher, R., and Schmincke, H.-U., 1995, Models for van der Lelij, R., Spikings, R.A., Kerr, A., Kounov, A., thesis ]: Universiteit Amsterdam, 43 p. the origin of accretionary lapilli: Bulletin of Volcanol- Cosca, M., Chew, D., and Villagomez, D., 2010, Ther- Neill, I., Kerr, A., Hastie, A.R., Stanek, K.-P., and Millar, ogy, v. 56, p. 626–639. mochronology and tectonics of the Leeward Antilles: I.L., 2011, Origin of the Aves Ridge and Dutch-Vene- Sinton, C.W., Duncan, R.A., and Denyer, P., 1997, Nicoya Evolution of the Southern Caribbean Plate Boundary zuelan Antilles: Interaction of the Cretaceous “Great Peninsula, Costa Rica: A single suite of Caribbean oce- Zone: Tectonics, v. 29, 30 p. Arc” and Caribbean-Colombian Oceanic Plateau?: anic plateau magmas: Journal of Geophysical Research, Villagómez, D., Spikings, R.A., Seward, D., Magna, T., and Journal Geological Society of London, v.168, no. 2, v. 102, p. 15,507–15,520, doi: 10.1029/97JB00681. Winkler, W., 2008, New thermochronological con- doi: 10.1144/0016-76492010-067. Sinton, C.W., Duncan, R.A., Storey, M., Lewis, J., and straints on the tectonic history of western Colombia, Pijpers, P.J., 1933, Geology and paleontology of Bonaire Estrada, J.J., 1998, An oceanic fl ood basalt prov- in Garver, J.I., ed., 11th International Conference on (D.W.I.) [Ph.D. thesis]: University of Utrecht, The ince within the Caribbean Plate: Earth and Planetary Thermochronometry: Schenectady, New York, Union Netherlands, 103 p. Science Letters, v. 155, p. 221–235, doi: 10.1016/ College, p. 253–255. Pindell, J., Kennan, L., Maresch, W.V., Stanek, K.-P., S0012-821X(97)00214-8. Vinasco, C.J., Cordani, U.G., González, H., Weber, M., and Draper, G., and Higgs, R., 2005, Plate kinematics and Smith, W., and Sandwell, D., 1997, Measured and Estimated Pelaez, C., 2006, Geochronological, isotopic, and geo- crustal dynamics of circum-Caribbean arc-continent Seafl oor Topography, World Data Center for Geophys- chemical data from Permo-Triassic granitic gneisses interactions: Tectonic controls on basin development in ics & Marine Geology, Boulder Research Publication and granitoids of the Colombian Central Andes: Journal Proto-Caribbean margins, in Ave Lallemant, H.G., and RP-1, poster, 34′ × 53′. of South American Earth Sciences, v. 21, p. 355–371. Sisson, V.B., eds., Caribbean-South American Plate Stanek, K.P., Maresch, W.V., and Pindell, J.L., 2009, The Walker, R.J., Storey, M., Kerr, A.C., Tarney, J., and Arndt, interactions, Venezuela: Boulder, Colorado, Geological geotectonic story of the northwestern branch of the N.T., 1999, Implications of 187Os heterogeneities in Society of America Special Paper 394, p. 7–52. Caribbean Arc: Implications from structural and geo- a mantle plume: Evidence from Gorgona Island and Pindell, J., Kennan, L., Stanek, K.P., Maresch, W.V., and chronological data of Cuba, in James, K.H., Lorente, Curacao: Geochimica et Cosmochimica Acta, v. 63, Draper, G., 2006, Foundations of Gulf of Mexico and M.A., and Pindell, J.L., eds., The Origin and evolution p. 713–728, doi: 10.1016/S0016-7037(99)00041-1. Caribbean evolution: Eight controversies resolved: of the Caribbean Plate: Geological Society [London] Westermann, J.H., 1932. The geology of Aruba [Ph.D. Geologica Acta, v. 4, p. 303–341. Special Publication 328, p. 361–398. thesis ]: University of Utrecht, The Netherlands, 129 p. Pindell, J.L., and Kennan, L., 2009, Tectonic evolution of the Sun, S.-S., and McDonough, W.F., 1989, Chemical and iso- White, R.V., Tarney, J., Kerr, A.C., Saunders, A.D., Gulf of Mexico, Caribbean and northern South Amer- topic systematics of oceanic basalts: Implications for Kempton, P.D., Pringle, M.S., and Klaver, G.T., 1999, ica in the mantle reference frame: An update, in James, mantle composition and processes, in Saunders, A.D., Modifi cation of an oceanic plateau, Aruba, Dutch K.H., Lorente, M.A., and Pindell, J.L., eds., The Origin and Norry, M.J., eds., Magmatism in the Ocean Basins: Caribbean: Implications for the generation of con- and evolution of the Caribbean Plate: Geological Soci- Geological Society of London Special Publication 42, tinental crust: Lithos, v. 46, p. 43–68, doi: 10.1016/ ety [London] Special Publication 328, p. 1–55. p. 313–345. S0024-4937(98)00061-9. Priem, H.N.A., Beets, D.J., and Verdurmen, E.A.T., 1986, Thompson, P.M.E., 2002, Petrology and geochronology of Wiedmann, J., 1978, Ammonites from the CLF, Curaçao, Precambrian rocks in an early Tertiary conglomerate an arc sequence, Bonaire, Dutch Antilles, and its asso- Caribbean: Geologie & Mijnbouw, v. 57, p. 361–364. on Bonaire, Netherlands Antilles (southern Caribbean ciation with the Caribbean plateau [Ph.D. thesis]: Uni- Wright, J.E., and Wyld, S.J., 2004, Aruba and Curacao: borderland): Evidence for a 300 km eastward displace- versity of Leicester, United Kingdom, 322 p. Remnants of a collided Pacifi c oceanic plateau? Ini- ment relative to the South American mainland?: Geolo- Thompson, P.M.E., Kempton, P.D., White, R.V., Saun- tial geologic results from the BOLIVAR project: Eos gie & Mijnbouw, v. 65, p. 35–40. ders, A.D., Kerr, A.C., Tarney, J., and Pringle, M.S., (Transactions, American Geophysical Union), v. 85, Restrepo-Pace, P.A., Ruiz, J., and Gehrels, G., 1997, Geo- 2004, Elemental, Hf-Nd isotopic and geochronologi- no. 47, p. F1713. chronology and Nd isotopic data of Grenville-age cal constraints on an island arc sequence associated rocks in the Colombian Andes: New constraints for with the Cretaceous Caribbean plateau: Bonaire, MANUSCRIPT RECEIVED 6 JANUARY 2010 Late Proterozoic-Early Paleozoic paleocontinental Dutch Antilles: Lithos, v. 74, p. 91–116, doi: 10.1016/ REVISED MANUSCRIPT RECEIVED 9 JUNE 2010 reconstructions of the Americas: Earth and Planetary j.lithos.2004.01.004. MANUSCRIPT ACCEPTED 11 JUNE 2010

Geosphere, April 2011 493

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/2/468/3715354/468.pdf by guest on 28 September 2021