Int J Earth Sci (Geol Rundsch) (2005) 94: 231242 DOI 10.

1007/s00531-005-0464-5

O R I GI N A L P A P E R

Marc Polliand Urs Schaltegger Martin Frank Lluis Fontbote

Formation of intra-arc volcanosedimentary basins in the western ank of the central Peruvian Andes during Late Cretaceous oblique subduction: eld evidence and constraints from UPb ages and Hf isotopes
Received: 16 October 2003 / Accepted: 11 December 2004 / Published online: 16 February 2005 Springer-Verlag 2005

Abstract During late Early to Late Cretaceous, the Peruvian coastal margin underwent fast and oblique subduction and was characterized by important arc plutonism (the Peruvian Coastal Batholith) and formation of volcanosedimentary basins known as the Western Peruvian Trough (WPT). We present high-precision UPb ages and initial Hf isotopic compositions of zircon from conformable volcanic and crosscutting intrusive rocks within submarine volcanosedimentary strata of the WPT hosting the Perubar massive sulde deposit. Zircons extracted from both the volcanic and intrusive rocks yield concordant UPb ages ranging from 67.890.18 Ma to 69.710.18 Ma, indicating that basin subsidence, submarine volcanism and plutonic activity occurred in close spatial and temporal relationship within the Andean magmatic arc during the Late Cretaceous. Field observations, satellite image interpretation, and plate reconstructions, suggest that dextral wrenching movements along crustal lineaments were related to oblique subduction. Wrench tectonics is therefore considered to be the trigger for the formation of the WPT as a series of pull-apart basins and for the emplacement of the Coastal Batholith. The zircon initial Hf values of the dated magmatic rocks fall between 5.5 and 7.4, and indicate only very subordinate inuence of a sedimentary or continental component. The absence of inherited cores in the zircons suggest a complete lack
M. Polliand U. Schaltegger (&) L. Fontbote Department of Mineralogy, University of Geneva, Rue des Mara chers 13, 1205 Geneve, Switzerland E-mail: [email protected] E-mail: [email protected] Tel.: +41-22-3796638 Fax: +41-22-3793210 E-mail: [email protected] U. Schaltegger M. Frank Department of Isotope Geology and Mineral Resources, Federal Institute of Technology ETH, Sonneggstrasse 5, 8092 Zurich, Switzerland E-mail: [email protected]

of old basement below the WPT, in agreement with previous UPb and Sr isotopic data for batholithic rocks emplaced in the WPT area. This is supported by the presence of a most likely continuous block of dense ($3.0 g/cm3) material observed beneath the WPT area on gravimetric crustal cross sections. We suggest that this gravimetric anomaly may correspond to a piece of lithospheric mantle and/or oceanic crust inherited from a possible Late PermianTriassic rifting. Such young and mac crust was the most probable source for arc magmatism in the WPT area. Keywords Andean magmatism Continental arc UPb dating Hf isotopes Strike-slip basins

Introduction
During late-Early to Late Cretaceous the relative motion between the oceanic Farallon (previously Phoenix) and South American plates along the central Andean margin was characterized by a period of high convergence rate (e.g., Jaillard et al. 2000; Larson 1991; Soler and Bonhomme 1990) with a signicant oblique component (northerly to northnortheasterly motion) of the subducted lithosphere (e.g., Jaillard 1994; Gordon and Jurdy 1986; Duncan and Hargraves 1984) and a most likely absolute trenchward motion of the South American plate (e.g., Jaillard and Soler 1996; Jaillard 1994; P. Soler, unpublished PhD thesis) due to the denitive opening of the South Atlantic ocean at about 110 Ma (e.g., Scotese et al. 1988; Larson and Pitman 1972). Along the coastal margin of Peru, this period of fast and oblique subduction was characterized by important arc magmatism and formation of volcanosedimentary basins with high average subsidence rates, known as the Mesozoic Western Peruvian Trough (WPT; e.g., Cobbing 1978; Atherton et al. 1983; Fig. 1a). The development of the WPT was accompanied and followed by

232 Fig. 1 a Geologic map of the Peruvian Coastal Batholith and Mesozoic Western Peruvian Trough (WPT) with related volcanosedimentary basins hosting massive sulde deposits. Volcanosedimentary basin limits after Vidal (1987), Atherton et al. (1983), and Cobbing (1978). The yellow doted lines encompass the 3.0 g/cm3 wedge at 10-km depth. Distribution of the Cretaceous Peruvian Costal Batholith, Casma Group, and Early Cretaceous strata after the digital version of the geological map of Peru (INGEMMET 2000). b Main lineaments and structures based on Landsat-5 TM image (Red Band 7; Green Band 4; Blue Band 2) interpretation, downloaded from http:// www.zulu.ssc.nasa.gov/mrsid/. c Geologic interpretation of the deep sturctures along the 9S and 12S gravimetric, crustal and subcrustal cross sections (Couch et al. 1981; Jones 1981) of the continental margin of Peru

intrusion of numerous plutons that form the present-day Peruvian Coastal Batholith (Fig. 1), mostly emplaced between about 100 Ma and 30 Ma (e.g., Soler and Rotach-Toulhoat 1990a; Mukasa 1986a). Within the volcanosedimentary basins, several massive sulde deposits were formed (i.e., Aurora Augusta, Cerro Lindo, Maria Teresa, Palma, Perubar; Fig. 1a, b), amongst which the Perubar deposit has been exploited from 1978 to 2000 and is the best studied (Polliand 2005; Polliand and Fontbote 2000; Vidal 1987). The WPT in the central coastal margin of Peru, between about Trujillo and Pisco, has been subdivided into

two major basin units known as the Huarmey and the Canete basins (e.g., Atherton et al. 1983; Cobbing 1978; Fig. 1). According to several authors (e.g., Atherton and Aguirre 1992; Atherton and Webb 1989; Cobbing et al. 1981), both the Huarmey and Canete basins had their main period of subsidence in the Albian, during which up to 9,000 m of volcanic (mainly basalt ows), volcaniclastic, and subordinate sedimentary rocks were deposited. These dominantly submarine volcanic and volcaniclastic successions have been attributed in the literature to the Casma Group and been described as Albian to Cenomanian in age (e.g., Atherton and Webb

233

1989; Cobbing et al. 1981). The extensional tectonic setting during which the Casma Group and related volcanosedimentary basins were formed was essentially attributed either to an extensional marginal basin with incipient formation of oceanic crust (e.g., Atherton and Aguirre 1992; Atherton 1990; Atherton and Webb 1989; Atherton et al. 1985), or to the extensional tectonic (pull-apart?) subsidence within the volcanic arc (e.g., Jaillard 1994; P. Soler, unpublished PhD thesis; Soler 1991). The overall geologic, lithologic and geometric framework of the WPT and related volcanosedimentary basins remains, however, poorly known, except in some areas near Huarmey and Culebras (Webb 1976; Myers 1980), and more detailed work is necessary to reach more sound geodynamic conclusions. We present new UPb age determinations and initial Hf isotopic compositions of zircon from conformable volcanic and crosscutting intrusive rocks within submarine volcanosedimentary strata attributed so far to the Casma Group and hosting the Perubar massive sulde deposit in the northeastern part of the Canete basin (50 km east of Lima, 1155S 7634W; Fig. 1). These new data (1) demonstrate that the volcanosedimentary rocks of the Perubar area are much younger (Maastrichtian) than the Albian to Cenomanian age so far assumed for the Casma Group; (2) allow a better understanding of the temporal relationships between basin subsidence, subaqueous volcanism, hydrothermal massive sulde mineralization, and plutonism during the uppermost Cretaceous in this region of the Peruvian Andes, (3) provide evidence for a formation of the WPT during extensional tectonic subsidence within the volcanic arc, and (4) constrain the melt sources and the nature of the basement beneath the central Peruvian coastal margin. Finally, this paper proposes a new hypothesis for the continental growth of the Peruvian active margin during late Early to Late Cretaceous, when the tangential component between the converging South American and Farallon plates was large.

Geologic setting of the volcanosedimentary rocks hosting the Perubar deposit

The Late Cretaceous volcanosedimentary strata that host the Perubar massive sulde deposit are located in the northeastern part of the Canete basin (Fig. 1a). These strata are truncated to the east and north by younger intrusions (i.e., the Canchacaylla monzogranite pluton and the Inclinado quartz monzodiorite porphyry stock, Fig. 2) and overlapped to the south by more recent continental volcanic deposits (i.e., the Calipuy Group, Fig. 2). The Canchacaylla monzogranite yields an hornblende KAr age of 66.72 Ma (Vidal 1987). The smaller and moderately altered hornblende+plagioclase+biotite phenocryst-bearing Inclinado quartz monzodiorite porphyry (sample number 4 in Fig. 3a), that follows the northsouth-trending Inclinado fault over a distance of 1,400 m (Fig. 2), yields a concordant

zircon UPb age of 67.890.18 Ma (Table 1). The Ricardo Palma tonalite, delimiting the western part of the basin (Fig. 2), is overlain by the volcanosedimentary strata and most probably constituted the western bedrock shoulder of the basin (Fig. 2). According to Vidal (1987) and Mukasa (1986a), the Ricardo Palma tonalite yields a 82.02.3 Ma hornblende KAr age and concordant zircon UPb ages of 86.4 and 84.4 Ma, respectively. It was emplaced into older magmatic rocks of the Coastal Batholith and Early Cretaceous sedimentary strata such as those exposed further west of the study area along the coast line (Fig. 1a, b). Both the older and younger intrusions occurring in the Perubar area belong to the Cretaceous Peruvian Coastal Batholith. The stratigraphy of the studied volcanosedimentary succession at Perubar has been divided into four main lithological units dened, from bottom to top, as the Footwall, Prospective, Hangingwall and Upper Units, respectively. Their lithological description is given in Fig. 3a. These volcanic and sedimentary strata were erupted and deposited in a pull-apart marine basin environment (Fig. 3b; Polliand 2005) without important periods of interrupted sedimentation, since less than 2 Ma separate the end of the Footwall Unit sedimentation and the deposition of the Upper Unit (Fig. 3a). The Hangingwall and Upper Unit deposits (Fig. 3) are interpreted as the result of a tectonically controlled caldera collapse, where collapse and magmatic plumbing were inuenced by pre-existing faults (Polliand 2005). The Hangingwall Unit polymictic breccias represent debris avalanche deposits resulting from a gravitational fault-block subsidence, that occurred prior to a catastrophic eruption responsible for the deposition of the Upper Unit pyroclastic and/or volcaniclastic debris ow deposits (Polliand 2005). The latter are inferred to be genetically related to the ignimbrite that outcrops along the Inclinado fault (Fig. 2), interpreted as the synvolcanic inll of a major eruptive ssure delineated by that same fault (Fig. 3b; Polliand 2005). The massive sulde mineralization, situated within the Prospective Unit (Fig. 3), was formed at the seaoor during a period of coeval submarine volcanism including the formation of lava domes, hyaloclastites, volcaniclastic sediments, and the sedimentation of impure limestones (Polliand 2005). The mineralizing hydrothermal event caused the strong hydrothermal alteration of the host and footwall rocks. A strongly hydrothermally altered, devitried, perlitically and/or hydrothermally fractured, ow-banded rhyolite, showing relict feldspar crystals and situated in the upper portion of the Lower Unit (sample number 1 in Fig. 3a), and a strongly hydrothermally altered porphyritic rhyolite with plagioclase > quartz > K-feldspar (sample 2 in Fig. 3a) directly overlying the massive sulde mineralization, yield zircon UPb concordant ages of 69.710.18 Ma and 68.920.16 Ma, respectively (Table 1 and Fig. 4). In addition, moderately altered aplites and rhyodacitic dykes with plagioclase > K-feldspar phenocrysts

234 Fig. 2 Geologic map of the Perubar deposit area. See Fig. 3b for cross section. Coordinates are UTM

crosscut the Footwall and Prospective Units and possibly the whole stratigraphic column, although crosscutting relationships with the Hangingwall and Upper Units were not observable in the eld (Fig. 3). One of these rhyodacitic dykes was dated and yields a concordant zircon UPb age of 67.940.16 Ma (Table 1 and sample number 3 in Fig. 3a; Fig. 4).

porphyritic rhyolitic lava, directly overlying the massive sulde orebodies and representing the upper limit of the Hangingwall Unit; (3) a porphyritic rhyodacitic dyke, believed to crosscut the whole stratigraphy; (4) the Inclinado quartz monzodiorite porphyry. UPb and Hf isotope data are summarized in Tables 1 and 2 (analytical techniques follow those in Schaltegger et al. 2002). Flow-banded rhyolite (MPP657; Fig. 4a)

UPb and Hf isotopic determinations

Determinations of UPb ages and analysis of initial Hf isotopic compositions in zircon were carried out on four dierent volcanic, subvolcanic and intrusive lithologies from the Perubar deposit area (see Figs. 3a, 4), consisting of: (1) a ow-banded rhyolite situated in the upper portion of the Lower Unit; (2) a massive The zircons are transparent, mostly subhedral to prismatic, and have a uniform U concentration of 403 432 ppm. Three multigrain analysis (sample no. 13) yield reproducible concordant points that dene a mean 206 Pb/238U age of 69.710.18 Ma (Fig. 4). Amongst

235 Fig. 3 a Lithostratigraphic column of the Late Cretaceous volcanosedimentary strata hosting the Perubar deposit. Numbers 14 delineate dated rock samples (see Table 1). b Longitudinal section across the Perubar VHMS deposit (see Fig. 2 for location). Note the deeply subsided blocks along the steep synsedimentary faults. This section was constructed on the basis of drill core, supercial, and underground working informations processed with the Gemcom mining software

these three multigrain fractions, Hf isotopic data were determined for two of them (1 and 2) and yield uniform initial Hf of 7.40.2 and 7.00.3. Porphyritic rhyolitic lava (MPP594; Fig. 4b) The zircons of this rhyolite are transparent and most of them have a G-type euhedral morphology (according to Pupin 1980). The analysed fractions are U-rich, with relatively homogeneous concentrations ranging between 3,067 ppm and 3,244 ppm. Three multigrain analyses (no. 46) yield well reproducible 206Pb/238U ratios that dene a mean age of 68.920.16 Ma. The Hf values for zircon fractions 5 and 6 are 5.73.4 and 7.70.5. The Hf value measured for fraction 4 (i.e., 11.01.3) is believed to represent an analytical outlier, since it strongly deviates from the other thirteen Hf values centered around a value of 6 (Table 2).

Porphyritic rhyodacitic dyke (JG120; Fig. 4c) The zircon population consists of one type of transparent euhedral grains, which mostly have a G-type morphology. About half of the zircons show small transparent melt inclusions; they have relatively uniform U concentrations ranging between 332 ppm and 593 ppm. Five multigrain fractions (no. 711) were analyzed and yield similar, reproducible 206Pb/238U ratios dening a concordant age of 67.940.16 Ma. When removing the two analytical points 7 and 8, which yield signicantly larger uncertainties due to their higher common lead concentrations, a mean concordant 206 Pb/238U age of 67.910.17 Ma can be calculated. Since these two ages are the same within error, the age calculated on the basis of the ve analytical points (67.940.16 Ma) will be preferred. The Hf values obtained for all analyzed fractions are relatively uniform and fall between 5.50.3 and 6.90.5.

236

Table 1 UPb isotopic data Apparent ages (Ma)

d

No. Description Error 207/235 Error 207/206 2s (%) 2s (%) 0.39 0.37 0.53 0.39 0.38 0.35 0.73 1.56 0.41 0.47 0.34 0.40 0.43 0.45 0.07016 0.06890 0.06919 0.06943 0.06912 2.38 2.64 0.87 0.61 0.87 0.06898 1.99 0.06921 0.82 0.06946 0.65 0.07017 0.51 0.07032 0.41 0.07048 0.51 0.07101 0.91 0.07148 1.49 0.07129 1.51 0.04746 0.04476 0.04747 0.04732 0.04748 0.07458 0.04783 0.04708 0.04742 0.04759 0.04733 0.04739 0.04741 0.04753 0.68 1.36 1.40 0.36 0.20 0.34 2.27 2.91 0.75 0.46 0.75 1.84 0.70 0.49
c c, d

Weight Number Concentrations Atomic ratios (mg) of grains a U Pb rad Pb nonrad. Th/U 206/204b 206/238c, (ppm) (ppm) (pg) 432 419 403 3151 3244 3067 332 403 593 465 499 238 328 103 2.73 3.98 1.23 4.2 0.7 0.4 0.65 0.87 0.79 344 1378 3444 0.01057 0.01059 0.01060 3.98 4.63 7.07 5.40 5.83 11.0 2.2 1.3 0.6 1.3 0.74 0.78 0.79 0.68 0.72 182 357 1123 2229 1,129 0.01064 0.01062 0.01058 0.01059 0.01059 39.26 40.93 39.60 0.8 0.5 4.5 0.92 0.96 1.04 2910 11195 1494 0.01076 0.01074 0.01074 5.05 4.86 4.88 2.6 2.9 1.0 0.62 0.57 0.74 876 372 684 0.01085 0.01089 0.01089

Error 206/238 ) 207/235 207/206 2s (%) 69.58 69.85 69.84 68.96 68.87 68.89 68.22 68.07 67.86 67.94 67.92 67.78 67.89 67.96 69.65 70.11 69.92 68.86 69.01 69.16 68.85 67.66 67.93 68.16 67.86 67.73 67.95 68.18 72.28 78.85 72.84 65.31 73.60 78.43 90.93 53.12 70.56 75.84 65.83 65.78 69.77 76.15

Error corr.

0.0076 6 0.0036 6 0.0024 7

0.73 0.46 0.38 0.71 0.87 0.74 0.30 0.11 0.51 0.67 0.52 0.46 0.52 0.66

0.0011 4 0.0024 5 0.0032 4 6 6 5 5 5

MPP657 1 Prism to subround 2 Subhedral 3 Subhedral MPP594 4 G-types euh 5 G-types euh spr 6 Small abr JG120 7 Clear euh 8 Round clrls 9 Round to prism 10 Round 11 Prism MPP338 12 Round 13 spr 14 Prism lge

0.0081 0.0028 0.0036 0.0040 0.0044

0.0086 6 0.0046 5 0.0065 3

euh euhedral, clrls colourless, incl inclusions, prism prismatic, spr short-prismatic, subh subhedral, transp transparent, G-type zircons according to Pupin (1980); analyses 16 unabraded, 710 abraded a Calculated on the basis of radiogenic 208Pb/206Pb ratios, assuming concordancy b Corrected for fractionation and spike c Corrected for fractionation (.09%), spike, blank and common lead (Stacey and Kramers 1975) d Corrected for initial Th disequilibrium, using an estimated Th/U ratio of 4 for the melt

237 Fig. 4 UPb concordia diagrams for volcanic and intrusive lithologies at Perubar

Inclinado quartz monzodiorite porphyry (MPP338; Fig. 4d) Two zircon populations are present in this rock, the rst consisting of long prismatic transparent, the second of round to short-prismatic transparent grains. The short prismatic zircons have a U content of 238 and 328 ppm and the long prismatic of 103 ppm. Three multigrain analyses (no. 1214) yield reproducible concordant points that dene a mean 206Pb/238U age of 67.890.18 Ma. The Hf for the three analyzed fractions are relatively uniform and range between 5.90.4 and 6.60.5.

Tectonic setting
The main structural elements in the Perubar deposit area are dominantly N- to NNE- and NW-oriented synsedimentary and later reactivated faults (Figs. 2, 5). The synsedimentary faults can be recognized by changes of depositional facies (Fig. 3b; Polliand 2005). When removing Tertiary dextral normal-slip movements that occurred along the main Corte Ladrones and Split faults (Polliand 2005), the synsedimentary faults dene a typical rhomb-shaped structure (Fig. 5), suggesting a pullapart origin for the Perubar basin. At a regional scale, the Perubar deposit is situated between two major NWoriented crustal lineaments (recognized on a Landsat-5 image) extending to the coast and being reected in the

orientation of the coastline (Fig. 1b). Similar structures are visible along the Peruvian coast from about 6 to 13S (Fig. 1a). The strongly oblique northerly to northnortheasterly motion assumed for the subducting slab (Farallon plate) during Late Cretaceous times (e.g., Pardo-Casas and Molnar 1987; Gordon and Jurdy 1986; Duncan and Hargraves 1984), is in agreement with dextral wrenching movements along these NW-oriented crustal lineaments (Fig. 1b). They were also recognized and described in the $100 Ma to $63 Ma-old Haura plutonic complex north of Lima (Bussell 1983). The orientation and extensional direction of the pull-apart basin structure evidenced at Perubar are in full agreement with the large-scale structures (Figs. 1b, 5) and thus suggest that Late Cretaceous dextral wrench tectonics associated with oblique subduction may have generated the Perubar strike-slip basin and possibly the whole WPT.

Discussion of the U/Pb ages and Hf isotope data

Timing of plutonism and basin subsidence during the Late Cretaceous Sedimentation in the Perubar pull-apart basin most likely started in the early Maastrichtian, before the 69.710.18 Ma dated subaqueous ow-banded rhyolite was formed (no. 1, Fig. 3). The 1015-Ma older Ricardo Palma tonalite (Fig. 2) is inferred to have formed the

238 Table 2 Hf isotopic data Number MPP657 1 2 3 MPP594 4 5 6 JG120 7 8 9 10 11 MPP338 12 13 14

176

Hf/177Hf 176Hf/177Hf 2s Hf (0) (T)a (T)b 0.282930 0.282920 0.283035 0.282883 0.282940 0.282877 0.282881 0.282917 0.282902 0.282899 0.282889 0.282909 6 9 36 94 14 6 16 12 7 10 9 11 7.4 7.0

2s T(DM) (Ga)c 0.2 0.3 0.63 0.65 0.41 0.73 0.61 0.74 0.73 0.66 0.69 0.69 0.72 0.67

0.282937 0.282927 0.283041 0.282889 0.282936 0.282883 0.282887 0.282923 0.282908 0.282905 0.282895 0.282915

(11.0) 1.3 5.7 3.4 7.7 0.5 5.5 5.6 6.9 6.3 6.2 5.9 6.6 0.3 0.6 0.5 0.3 0.4 0.4 0.5

a The Hf fraction was isolated using Eichrom Ln-spec resin, and measured in static mode on a NuPlasma multi-collector ICP-MS using a MCN-6000 nebulizer for sample introduction. Zircons were corrected for in situ radiogenic ingrowth from 176 Lu using an estimated 176 Lu/177 Hf of 0.005. The Hf isotopic ratios were corrected for mass fractionation using a 179Hf/177Hf value of 0.7325 and normalized to 176Hf/177Hf=of 0.282160 of the JMC-475 standard (Blichert-Toft et al. 1997). Mean isotopic values are at the 95% condence level b Calculated with present-day CHUR values of 176Hf/177Hf= 0.282772, 176 Lu/177 Hf=0.0332 c Calculated with present-day DM values of 176Hf/177Hf=0.283252, 176 Lu/177 Hf=0.04145 and a crustal 176 Lu/177 Hf=0.017 (Blichert` Toft and Albarede 1997)

bedrock and the western shoulder of this subsiding basin. The upper, 540-m thick, volcanosedimentary strata were deposited between 69.710.18 Ma and 67.890.18 Ma, ages that correspond to the ow-banded rhyolite and the Inclinado monzonite porphyry, respectively (no. 1 and 4, Fig. 3), yielding a minimum average sedimentation rate of 0.3 mm/year. The massive sulde mineralization formed within the 0.8-Ma interval comprised between the 69.710.18 Ma ow-banded rhyolite and the 68.920.1 Ma porphyritic rhyolitic lava (no. 1 and 2, Fig. 3). Therefore, the studied volcanosedimentary strata in this part of the Canete basin cannot be considered as a part of the Casma Group but represent a distinct and younger basin sedimentation event, since they were emplaced in the uppermost Cretaceous (Maastrichtian) instead of the mid-Cretaceous (AlbianCenomanian) age previously dened for the Casma Group (Atherton and Webb 1989 and references therein). The deposition of the studied volcanosedimentary pile and related massive sulde deposit was quickly followed by the injection of porphyritic rhyodacitic dykes at 67.910.17 Ma (no. 3, Fig. 3) and the intrusion of both the Inclinado porphyry and the Canchacaylla pluton at 67.890.18 Ma (no. 4, Fig. 3) and 66.72 Ma (Vidal 1987; Fig. 2), respectively. Since these two intrusive rocks

Fig. 5 Model for Late Cretaceous pull-apart origin of the Perubar basin. The tectonic stress (r1) is inferred from relative plate motion reconstructions between the South American continent and the oceanic Farallon plate (e.g., Gordon and Jurdy 1986; Duncan and Hargraves 1984)

belong to the subduction-related arc magmatism of the Peruvian Coastal Batholith (e.g., P. Soler, unpublished PhD thesis; Soler and Bonhomme 1990; Mukasa 1986a), it is likely that the subaqueously erupted volcanic deposits at Perubar represent the eusive counterpart of the plutonic arc. Therefore, Late Cretaceous intra-arc strike-slip faulting, as observed at Perubar (Figs. 3b, 5), was most probably responsible for pull-apart basin subsidence and related subaqueous volcanism, and provided at the same time space in the crust for the emplacement of plutons. This interpretation is in agreement with mechanisms related to oblique convergence, the latter being typically

239

resolved into a strongly orthogonal compressional component at the trench and a strike-slip component expressed by structures in the forearc and arc itself (Fitch 1972). Melt sourcesevidence for lacking Precambrian basement beneath the WPT The Hf values determined for the zircon fractions of the four dierent analyzed lithologies at Perubar all fall between 5.50.3 and 7.70.2 (Table 2). This range of positive values indicates only very subordinate contamination by a sedimentary or continental component during emplacement of both the volcanic and intrusive rocks at Perubar. The absence of inherited Proterozoic ages in the zircons measured in both the four dated samples at Perubar and in the Lima segment of the Coastal Batholith in general (Mukasa 1986a; Fig. 1a) indicates the lack of old basement beneath the WPT area. This contrasts with the occurrence of inherited zircons from the Precambrian basement of the Arequipa craton in plutons from the Arequipa and Toquepala segments of the Coastal Batholith (Mukasa 1986a; Fig. 1a). Moreover, the Coastal Batholith in the Lima segment displays consistent lead and strontium isotopic ratios (206Pb/204Pb: 18.55420.803; 87Sr/86Sr: 0.70307 0.70516), which infer that the rocks are apparently uncontaminated by upper crustal radiogenic components (e.g., Haeberlin 2004; Macfarlane et al. 1990; Soler and Rotach-Toulhoat 1990a, b; Mukasa 1986b) and thus suggest a lack of old basement below the WPT area between Pisco and Chimbote (Fig. 1a). It is supported by the presence of a most likely continuous block of dense (3.0 to 3.04 g/cm3) material (oceanic crust and lithospheric mantle?) beneath the WPT area between Pisco and Trujillo according to the gravimetric crustal cross sections (Couch et al. 1981; Jones 1981) shown in Fig. 1a and c. At the latitude of the 12S prole, we extended the limit of the 3.0 g/cm3 gravity anomaly to the eastern boundary of the Canete basin (Fig. 1a), since no evidence of signicant upper-crustal contamination can be identied in the studied volcanic and plutonic rocks at Perubar. This zone, free of old basement underlying the WPT, would correspond to the lead province Ib dened in Macfarlane et al. (1990), the lead of which is believed to have been extracted from an enriched upper mantle reservoir. Similarly, Soler and Rotach-Toulhoat (1990a) suggested that the homogeneous strontium isotopic composition ($0.704) of the Coastal Batholith between Pisco and Chimbote is related to an undepleted mantle source (OIB-type or enriched sub-continental) that was contaminated by uids derived from the subducting slab and associated subducted sediments. The range of Hf values between +5.5 and +7.7 of the volcanic and intrusive rocks at Perubar are compatible with: (1) such an enriched upper mantle source reservoir, (2) with a depleted mantle contaminated by slab-derived uids carrying Hf (Woodhead

et al. 2001), or alternatively, (3) with a mixture of enriched mantle and reworked oceanic crust. Hf model ages evidence an average mantle extraction or residence age of 0.6-0.7 Ga. Implications of a pull-apart origin for the volcanosedimentary basins of the WPT Several authors (e.g., Atherton and Aguirre 1992; Atherton 1990; Atherton and Webb 1989; Atherton et al. 1985) consider the WPT as an aborted ocean oor extensional marginal basin that had its main period of extension and subsidence during the Albian. This model mainly relies on the interpretation of the gravimetric and seismic data along the 9S, 12S and Pisco crustal cross sections (Fig. 1a, c; Couch et al. 1981; Jones 1981), where the dense 3.0 g/cm3 gravity anomaly is attributed to mantle-derived material that oored and lled up this Albian marginal basin. Accordingly, the 2.672.8 g/cm3 blocks underlying the so-called Outer Shelf Higharea (Fig. 1a, c) are interpreted as pieces of rifted continental crust. However, we have demonstrated that the Albian marginal basin model cannot be applied to the volcanosedimentary strata exposed at Perubar, the latter being Maastrichtian in age and deposited within a pull-apart basin formed atop of the plutonic arc and related to the N to NNE motion assumed for the paleo-Pacic slab during Late Cretaceous times. Since a roughly northward motion of the subducting paleo-Pacic plate beneath the central Peruvian margin is also assumed for the preceeding Albian to Campanian period (e.g., Jaillard 2000; Jaillard 1994; Duncan and Hargraves 1984) combined with a high convergence rate (>10 cm/year; Jaillard and Soler 1996; Larson 1991), we suggest that wrench tectonics, strike-slip faulting and subsequent pull-apart basin formation also occurred at that time, in line with the proposition of Jaillard (1994) and P. Soler (unpublished PhD thesis, Soler 1991) that the Casma Group-hosting volcanosedimentary basins possibly relate to extensional pull-apart subsidence within the volcanic arc. Thus, the WPT volcanosedimentary basins might represent distinct intra-arc pull-apart basin generations progressively migrating eastward with the arc from Albian to Maastrichtian times. During the AlbianCampanian period, the relative age of the subducted lithosphere was rapidly decreasing (e.g., Jaillard and Soler 1996; Jaillard 1994; Soler and Bonhomme 1990), leading most likely to a decrease of the slab dip. This was possibly coupled to an absolute trenchward motion of the overriding South American plate (e.g., Jaillard and Soler 1996; Soler and Bonhomme 1990). Both phenomena could account for the eastward migration of the arc. The inferred intra-arc extensional pull-apart origin for the WPT volcanosedimentary basins implies a pre-late Early Cretaceous origin for the underlying dense 3.0 g/ cm3 structure (Fig. 1a, c). This 10 km thick block of dense material might represent a piece of lithospheric mantle

240

and/or oceanic crust inherited from a possible Late PermianTriassic rifting, believed to have developed in the Eastern Cordillera of Peru and extended into Bolivia in the Late TriassicMiddle Jurassic (Sempere et al. 2002 and references therein). Several authors (e.g., Haeberlin 2004; Ramos and Aleman 2000; Dalziel et al. 1994) suggested that this rifting event would have removed the northwest prolongation of the Arequipa craton of southern Peru, from which the detached slice or slices are thought to have migrated northwards and possibly represent the allochthonous Oaxaquia Terrane in Mexico. The Outer Shelf High, which, in this case, cannot be considered as a piece of rifted continental crust, may therefore be interpreted as an accretionary complex and/ or an accreted oceanic plateau (Nazca ridge?) and/or accreted island arc terranes previously formed on the paleoPacic plate (Fig. 1c). However, more work needs to be done prior to reach any rm conclusions on the origin and nature of the present-day deep structures found along the coastal margin of Peru between $7S and $14S. The correlation between oblique subduction, basin formation and magmatism The apparent correlation between plutonism, oblique subduction and associated strike-slip faulting observed in the study area suggest that the Cretaceous Peruvian Coastal Batholith might be related to episode(s) of oblique subduction, as already observed by Glazner (1991) for major episodes of Mesozoic plutonism in California. According to this author, major episodes of continental growth by addition of plutons to continental margins may occur only if there is a signicant oblique component at the trench, the magmatic room problem being solved by passive emplacement of plutons at releasing bends in the strike-slip faults and conservation of volume by thrusting at the trench. Therefore, the central Peruvian continental margin may have built its own new crust (i.e. the Peruvian Coastal Batholith), by similar processes with magma originating from an enriched mantle source. Numerous modern intra-arc fault-bounded basin analogues are observed in strike-slip zones related to oblique subduction, such as the Taupo volcanic zone in the Kermadec-Tonga intraoceanic arc (Cole and Lewis 1981), the Lake Managua pull-apart graben in the Nicaraguan Depression (Burkart and Self 1985), basins along the central Aleutian Ridge arc platform (Geist and Scholl 1992), and intra-arc depocenters along the central Salomon Trough (Ryan and Coleman 1992).

formation and magmatism. The example of the Perubar basin evolution, during which volcanic activity progressively increased to reach a tectonically controlled submarine caldera collapse stage that was quickly followed by the intrusion of porphyries and plutons, fully supports this interpretation. The zircon UPb ages of both volcanic and intrusive rocks at Perubar range between 67.890.18 Ma and 69.710.18 Ma, indicating that volcanosedimentary rocks of at least part of the Casma Group are not AlbianCenomanian in age but were in fact formed during Maastrichtian times. The Perubar volcanic-hosted massive sulde deposit was formed during Maastrichtian times, between 69.710.18 Ma and 68.920.16 Ma. Volcanic-hosted massive sulde (VHMS) deposits in the Peruvian segment of the Andes, such as the Perubar deposit, are generally located along the major NW-oriented crustal lineaments, where contemporaneous strikeslip faulting, pull-apart basin opening, and magma plumbing were most probably acting. These structures are likely to represent old inherited deep crustal discontinuities, e.g., transform faults from a Late Permian Triassic rifting event. The initial Hf values of the magmatic rocks fall between values of +5.5 and +7.4, and record subordinate contamination by a low-radiogenic Hf, possibly derived from slab sediments. These values may also represent an enriched mantle source and/or recycled oceanic crust as possible melt sources. The lack of inherited cores in zircons demonstrates that there is no old basement involved in the generation of volcanic and intrusive rocks at Perubar. We think that the data evidence the lack of old basement under the WPT, such as the Precambrian Arequipa massif in southern Peru (Wasteneys et al. 1995). Young and mac crust is the probable source for the arc magmatism within the Lima segment of the Coastal Batholith. Voluminous plutonism in an arc environment thus requires extension of relatively hot crust triggering remelting and emplacement in pull-apart structures. This process is suggested to be equally important in oceanic arcs (such as the Kohistan Arc complex; Schaltegger et al. 2002) as well as in continental active margin situations such as in the Andes.
Acknowledgements This research was supported by the Swiss National Science Foundation grants FN 200054150.98 and FN 2062000.00, and is part of a PhD thesis of the rst author at the University of Geneva. We gratefully acknowledge W. Mueller and L. Oldham who instigated this project and transmitted us their knowledge of the Perubar deposit area, the geologists and sta of the Perubar mine for their help during the eld work periods, especially R. Egoavil and C. Zumaran, and the geologists from Glencore Peru, M. Steinmann and S. Bureau, for logistical support during the project.

Conclusions
Basin subsidence, submarine volcanism and plutonic activity occurred in close spatial and temporal relationship within the Andean magmatic arc during the uppermost Cretaceous oblique subduction, indicating that wrench tectonics was the trigger of pull-apart basin

References
Atherton MP (1990) The Coastal Batholith of Peru; the product of rapid recycling of new crust formed within rifted continental margin. Geol J 25:337349

241 Atherton MP, Aguirre L (1992) Thermal and geotectonic setting of Cretaceous volcanic rocks near Ica, Peru, in relation to Andean crustal thinning. J S Am Earth Sci 5:4769 Atherton MP, Webb S (1989) Volcanic facies, structure, and geochemistry of the marginal basin rocks of central Peru. J S Am Earth Sci 2:241261 Atherton MP, Pitcher WS, Warden V (1983) The Mesozoic marginal basin of central Peru. Nature 305:303306 Atherton MP, Warden V, Sanderson LM (1985) The Mesozoic marginal basin of Central Peru: a geochemical study of withinplate-edge volcanism. In: Pitcher WS, Atherton MP, Cobbing EJ, Beckinsale RD (eds) Magmatism at a plate edge; the Peruvian Andes. Blackie, Glasgow, pp 4758 ` Blichert-Toft J, Albarede F (1997) The LuHf isotope geochemistry of chondrites and the evolution of the mantle-crust system. Earth Planet Sci Let 148:243258 ` Blichert-Toft J, Chauvel C, Albarede F (1997) Separation of Hf and Lu for high-precision isotope analysis of rock samples by magnetic sector-multiple collector ICP-MS. Contrib Mineral Petrol 127:248260 Burkart B, Self S (1985) Extension and rotation of crustal blocks in northern Central America and eect on the volcanic arc. Geology 13:2226 Bussell MA (1983) Timing of tectonic and magmatic events in the Central Andes of Peru. J Geol Soc Lond 140:279286 Cobbing EJ (1978) The Andean Geosyncline in Peru, and its distinction from Alpine geosynclines. J Geol Soc Lond 135:207 218 Cobbing EJ, Pitcher WS, Wilson JJ, Baldock JW, Taylor WP, McCourt W, Snelling NJ (1981) The geology of the Western Cordillera of northern Peru. Overseas Mem Inst Geol Sci, vol 5, pp 143 Cole JW, Lewis KB (1981) Evolution of the Taupo-Hikurangi subduction system. Tectonophysics 72:121 Couch RW, Whitsett RM, Huehn B, Briceno GL (1981) Structures of the continental margin of Peru and Chile. Geol Soc Am Mem 154:703726 Dalziel IWD, Dalla Salda L, Gahagan LM (1994) Paleozoic LaurentiaGondwana interaction and the origin of the Appalachian Andean mountain system. Geol Soc Am Bull 106:243252 Duncan RA, Hargraves RB (1984) Plate tectonic evolution of the Caribbean region in the mantle reference frame. Geol Soc Am Mem 162:8193 Fitch TJ (1972) Plate convergence, transcurrent faults, and internal deformation adjacent to southeast Asia and the Western Pacic. J Geophys Res 77:44324460 Geist EL, Scholl DW (1992) Application of continuum models to deformation of the Aleutian island arc. J Geophys Res 97:4953 4967 Glazner AF (1991) Plutonism, oblique subduction, and continental growth; an example from the Mesozoic of California. Geology 19:784786 Gordon RG, Jurdy DM (1986) Cenozoic global plate motions. J Geophys Res 91:12,38912,406 Haeberlin Y, Moritz M, Fontbote L, Cosca M (2004) Carboniferous orogenic gold deposits at Pataz, Eastern Andean Cordillera, Peru: geological and structural framework, paragenesis, alteration, and 40Ar/39Ar geochronology. Econ Geol 99:73112 INGEMMET (2000) Mapa geologico digitalizado del Peru (Escala 1:1000000). Lima, Instituto Geologico Minero y Metalurgico, CD-ROM Jaillard E (1994) Kimmeridgian to Paleocene tectonic and geodynamic evolution of the Peruvian (and Ecuadorian) margin. In: Salty JA (ed) Cretaceous tectonics of the Andes. Earth evolution sciences monograph series. Wiesbaden, Vieweg, pp 101 167 Jaillard E, Soler P (1996) Cretaceous to early Paleogene tectonic evolution of the northern Central Andes (018 degrees S) and its relations to geodynamics. Tectonophysics 259:4153 Jaillard E, Herail G, Monfret T, Diaz-Mart nez E, Baby P, Lavenu A, Dumon JF (2000) Tectonic evolution of the Andes of Ecuador, Peru, Bolivia and northernmost Chile. In: Cordani UG, Milani EJ, Thomaz Fihlo A, Campos DA (eds) Tectonic evolution of South America. In: 31st international geological congress, Rio de Janeiro, pp 481559 Jones PR (1981) Crustal structures of the Peru continental margin and adjacent Nazca Plate, 9 degrees S latitude. Geol Soc Am Mem 154:423443 Larson RL (1991) Latest pulse of Earth; evidence for a Mid-Cretaceous super plume. Geology 19:547550 Larson RL, Pitman WC (1972) World-wide correlation of Mesozoic magnetic anomalies, and its implications. Geol Soc Am Bull 83:36453661 Macfarlane AW, Marcet P, LeHuray AP, Petersen U (1990) Lead isotope provinces of the Central Andes inferred from ores and crustal rocks. Econ Geol 85:18571880 Mukasa SB (1986a) Zircon U-Pb ages of super-units in the Coastal Batholith, Peru; implications for magmatic and tectonic processes. Geol Soc Am Bull 97:241254 Mukasa SB (1986b) Common Pb isotopic compositions of the Lima, Arequipa and Toquepala segments in the Coastal Batholith, Peru; implications for magmagenesis. Geochim Cosmochim Acta 50:771782 Myers JS (1980) Geologia de los cuadrangulos de Huarmey y Huayllapampa; hojas 21-g y 21 h, vol 33. Instituto Geologico Minero y Metalurgico, Lima, Peru, pp 153 Pardo-Casas F, Molnar P (1987) Relative motion of the Nazca (Farallon) and South American plates since Late Cretaceous time. Tectonics 6:233248 Polliand M (2005) Genesis, evolution, and tectonic setting of the Upper Cretaceous Perubar Ba-PbZn volcanic-hosted massive sulde deposit. Terre and Environement, Thesis 3404, Section ` des Sciences de la Terre, Universite de Geneve, (in press) Polliand M, Fontbote L (2000) The Perubar Ba-Pb-Zn VHMS deposit, Central Peru. In: Sherlock RL, Logan MAV (eds) VMS deposits of Latin America. Geological Association of Canada, Mineral Deposits Division, Special Paper, Vancouver 2:439446 Pupin JP (1980) Zircon and granite petrology. Contrib Mineral Petrol 73:207220 Ramos VA, Aleman A (2000) Tectonic evolution of the Andes. In: Cordani UG, Milani EJ, Thomaz Fihlo A, Campos DA (eds) Tectonic evolution of South America. In: 31st international geological congress, Rio de Janeiro, pp 635685 Ryan HF, Coleman PJ (1992) Composite transform-convergent plate boundaries; description and discussion. Mar Petrol Geol 9:8997 Schaltegger U, Zeilinger G, Frank M, Burg JP (2002) Multiple mantle source during island arc magmatism: UPb and Hf isotopic evidence from the Kohistan arc complex, Pakistan. Terra Nova 14:461468 Scotese CR, Gahagan LM, Larson RL (1988) Plate tectonic reconstructions of the Cretaceous and Cenozoic ocean basins. Tectonophysics 155:2748 Sempere T, Carlier G, Soler P, Fornari M, Carlotto V, Jacay J, Arispe O, Neraudeau D, Cardenas J, Rosas S, Jimenez N (2002) Late PermianMiddle Jurassic lithospheric thinning in Peru and Bolivia, and its bearing on Andean-age tectonics. Tectonophysics 345:153181 Soler P (1991) El volcanismo Casma del Peru Central: cuenca marginal abortada o simple arco volcanico?. VII Congreso Peruano de Geolog a, Resumenes Extendidos, Soc Geol Peru, Lima 2, pp 659663 Soler P, Bonhomme MG (1990) Relation of magmatic activity to plate dynamics in central Peru from Late Cretaceous to present. Geol Soc Am Spec Pap 241:173192 Soler P, Rotach-Toulhoat N (1990a) Implications of the timedependent evolution of Pb- and Sr-isotopic compositions of Cretaceous and Cenozoic granitoids from the coastal region and the lower Pacic slope of the Andes of central Peru. Geol Soc Am Spec Pap 241:161172 Soler P, Rotach-Toulhoat N (1990b) Sr-Nd isotope compositions of Cenozoic granitoids along a traverse of the central Peruvian Andes. Geol J 25:351358

242 Stacey JS, Kramers JD (1975) Approximation of terrestrial lead isotope evolution by a two-stage model. Earth Planet Sci Lett 26:207221 Thornburg TM, Kulm LD (1981) Sedimentary basins of the Peru continental margin; structure, stratigraphy, and Cenozoic tectonics from 6 degrees S to 16 degrees S latitude. Geol Soc Am Mem 154:393422 Vidal CCE (1987) Kuroko-type deposits in the Middle-Cretaceous marginal basin of central Peru. Econ Geol 82:14091430 Wasteneys HA, Clark AH, Farrar E, Langridge RJ (1995) Grenvillian granulite-facies metamorphism in the Arequipa Massif, Peru; a Laurentia-Gondwana link. Earth Planet Sci Let 132:63 73 Webb SE (1976) The volcanic envelope of the coastal batholith in Lima and Ancash, Peru. PhD thesis, University of Liverpool Woodhead JD, Hergt JM, Davidson P, Eggins SM (2001) Hafnium isotope evidence for conservative element mobility during subduction zone processes. Earth Planet Sci Let 192:331346