Exeter caspiche paper_in_economic_geology_2013

Introduction and Exploration HistoryTHE CASPICHE porphyry gold-copper deposit is the latest dis-covery in the Maricunga metallogenic belt, located betweenlatitudes 26° and 28°S in the high volcanic Andes of the Ata-cama Region, northern Chile (Fig. 1). The belt, some 200 kmlong and 30 km wide, is an N-trending alignment of latest

Oligocene to Miocene volcanic complexes and comagmaticsubvolcanic stocks associated with porphyry gold ± copperand high sulfidation epithermal gold ± silver deposits andprospects (Vila and Sillitoe, 1991; Mpodozis et al., 1995; Fig.2). The alignment of mineralized centers continues to thesouth, via the Caserones (formerly Regalito) porphyry copperdeposit (Perelló et al., 2003) and nearby prospects, to the ElIndio belt, where high-grade vein (e.g., El Indio) and bulk-tonnage (e.g., Pascua-Lama, Veladero), high sulfidation epi-thermal gold ± silver deposits predominate (e.g., Maksaev etal., 1984; Jannas et al., 1999; Chouinard et al., 2005; Fig. 1).Miocene porphyry gold prospects also occur farther east, inwesternmost Argentina, where they were generated in thebackarc to the Maricunga belt.

0361-0128/13/4108/585-20 585

Geology of the Caspiche Porphyry Gold-Copper Deposit, Maricunga Belt, Northern Chile*

RICHARD H. SILLITOE,1,† JUSTIN TOLMAN,2,** AND GLEN VAN KERKVOORT3,***1 27 West Hill Park, Highgate Village, London N6 6ND, England

2 Exeter Resource Corporation, Suite 1660 - 999 W. Hastings St., Vancouver, BC V6C 2W2, Canada 3 Exeter Resource Corporation, Suite 701, 121 Walker St., North Sydney, NSW 2060, Australia

AbstractThe Caspiche porphyry gold-copper deposit, part of the Maricunga gold-silver-copper belt of northern

Chile, was discovered in 2007 beneath postmineral cover by the third company to explore the property over a21-year period. This company, Exeter Resource Corporation, has announced a proven and probable mineralreserve of 1,091 million tonnes (Mt) averaging 0.55 g/t Au, all but 124 Mt of which also contain 0.23% Cu, fora total of 19.3 Moz of contained gold and 2.1 Mt of copper.

The deposit was formed in the latest Oligocene (~25 Ma) during the first of two volcanic and corre-sponding metallogenic epochs that define the Maricunga belt. The gold-copper mineralization is centeredon a composite diorite to quartz diorite porphyry stock, within which five outward-younging phases are rou-tinely distinguished. The centrally located, early diorite porphyry (phase 1) hosts the highest-grade ore, av-eraging ~1 g/t Au and 0.4% Cu. The subsequent porphyry phases are quartz dioritic in composition and char-acterized by progressively lower gold and copper tenors. Stock emplacement was both pre- and postdatedby the generation of large-volume, andesite-dominated breccias, with tuffaceous matrices, which are be-lieved to be shallow portions of diatremes. The deposit is characterized by a central gold-copper zone andpartially overlapped but noneconomic molybdenum halo. The gold-copper mineralization in the lower halfof the deposit accompanies quartz ± magnetite-veined, potassic-altered rocks, whereas the shallower min-eralization occurs within quartz-kaolinite–dominated, advanced argillic alteration. Upper parts of the ad-vanced argillic zone are characterized by siliceous ledges, some auriferous, composed of vuggy residualquartz and/or silicified rock. The chalcopyrite-pyrite mineralization in the potassic zone was partially trans-formed to high sulfidation-state sulfides and sulfosalts during the advanced argillic overprint, although theunderlying chalcopyrite-bornite assemblage was mainly too deep to be affected. The deposit terminatesdownward in a sulfide-deficient, potassic-calcic zone defined by K-feldspar, actinolite, and magnetite, whichformed at the expense of biotite. A relatively minor, shallowly inclined zone of intermediate sulfidation epi-thermal gold-zinc mineralization, comprising narrow veinlets and disseminations, abuts the late-mineral diatreme contact. Supergene sulfide oxidation throughout the deposit is relatively shallow, and chalcocite enrichment extremely minor.

The Caspiche deposit is thought to have been emplaced at relatively shallow paleodepths, within the south-ern, flat-bottomed part of the premineral diatreme vent. Earliest porphyry system development, probably northof the present deposit, appears to have been aborted by diatreme formation. Much of the gold and copper inthe Caspiche deposit was introduced during the potassic alteration stage, with the highly telescoped, advancedargillic overprint being responsible for only minor redistribution of the two metals, and the addition of arsenic.The late-mineral diatreme was emplaced west of the Caspiche deposit, and caused destruction of only its up-permost peripheral parts. The late-mineral diatreme was both pre- and postdated by advanced argillic alter-ation. Finally, the intermediate sulfidation epithermal gold-zinc zone was localized by the enhanced perme-ability provided by intense fracturing along the underside of the upward-flared, late-mineral diatreme contact.

† Corresponding author: e-mail, [email protected]*A digital supplement to this paper is available at http://economicgeology.

org/and at http://econgeol.geoscienceworld.org/.**Present address: New Gold Inc., 12200 E Briarwood Ave., Suite 165,

Centennial, CO 80112.***Present address: Rio Pison Gold Ltd., P.O. Box 758, Buderim, Queens-

land 4556, Australia.

©2013 Society of Economic Geologists, Inc.Economic Geology, v. 108, pp. 585–604

Submitted: February 12, 2012Accepted: September 15, 2012

The Maricunga belt was defined during the mid- to late1980s during a regional exploration program of areally exten-sive alteration zones for gold and silver conducted by MineraAnglo American Chile Ltda. and joint-venture partners.However, it was not until one of these zones, the host to theMarte deposit (Fig. 2), was systematically explored that thepresence of porphyry-type mineralization was first recognized(Vila and Sillitoe, 1991; Vila et al., 1991). The porphyry de-posits, including those at Marte, Lobo, Refugio, La Pepa, andVolcán (Fig. 2), typically contain gold as the only economi-cally extractable metal, whereas those at Caspiche and CerroCasale (Fig. 2) contain gold plus by-product copper potential.

Caspiche is located at 4,200 to 4,560 m above sea level, im-mediately east of the topographically and visually prominentSanta Cecilia alteration zone, from which it is separated by a250-m-wide valley (Fig. 3). Santa Cecilia was extensively

586 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 586

CASPICHE

Antofagasta

-30°

CASERONES (REGALITO)

PASCUA -LAMA

EL INDIO

VELADERO

Bolivia

Argentina

Chile

Santiago

-70°

-7

0°

-30° El Indio belt

Maricunga belt

200 km

Fig. 2

Copiapó

FIG. 1. Location of the Caspiche porphyry gold-copper deposit in theMaricunga belt with respect to its southern continuation—the El Indiobelt—and some key deposits. Area shown in Figure 2 is marked.

16 to 11 Ma volcanic rocks

26 to 17 Ma volcanic rocks

Mineral deposit

Reverse fault

Other fault

CERRO CASALE

10 km

CASPICHE

REFUGIO

Laguna del Negro Francisco

VOLCAN

LA PEPA

LOBO

MARTE

PANTANILLOS

Salar de Maricunga

LA COIPA PUREN

ESPERANZA

Potrerillos

-26°30’

-27°

-27°30’

-69°

30’

FIG. 2. Location of the Caspiche porphyry gold-copper deposit in theMaricunga belt, showing the two ages of mineralized volcanic rocks and themain porphyry and epithermal deposits. The underlined deposits accompa-nied the younger volcanism. Compiled from Kay et al. (1994).

investigated, but with only limited success, by Anglo Ameri-can from 1985 to 1990 under an option agreement with aChilean mining entrepreneur. At the same time, Anglo Amer-ican also claimed surrounding open ground, within which thecompany recognized and then explored the Caspicheprospect, with a focus on the prominently exposed, gold- andsilver-bearing, siliceous ledges of high sulfidation epithermaltype (Fig. 3). Anglo American collected 22 rock-chip samplesfrom the ledges during the 1985 to 1986 summer field season,obtaining gold values ranging from 0.5 to 4 g/t. Grid talus-fines geochemistry and more detailed rock-chip samplingwere followed by 12 50-m air-track holes during the 1986 to1987 season. Based on the reasonably encouraging resultsfrom two of the holes (including 48 m at 1.03 g/t Au), detailedgeologic mapping and six deeper (150–200 m) reverse-circu-lation holes were completed in the 1989 to 1990 season, butthe best intercept was 148 m at 0.49 g/t Au. A preliminary re-source of 6.3 million tonnes (Mt) averaging 0.45 g/t Au wasestimated, leading to all work at Caspiche being terminated.

The Caspiche prospect lay fallow for six years before MineraNewcrest Chile Ltda. entered into a purchase option agree-ment with Anglo American, with the aim of following up thelow-grade, disseminated gold intercepts obtained previouslyas well as testing the possibility that a “microdiorite” intrusion

present at depth alongside the main siliceous ledge mightcontain higher-grade, structurally controlled gold ore. Duringthe 1996 to 1997 season, Newcrest conducted detailed geo-logic mapping, rock and trench geochemical sampling, and aninduced polarization-resistivity survey over the Caspiche andnewly pegged contiguous claims, preparatory to drilling 12 re-verse-circulation holes to a nominal depth of 300 m atCaspiche. The results suggested the presence of bulk-ton-nage, high sulfidation epithermal gold mineralization grading0.3 to 0.6 g/t, but one of the holes, collared in an area of post-mineral cover between the siliceous ledges (Fig. 3), report-edly intersected potassic-altered “microdiorite porphyry” av-eraging 0.73 g/t Au and 0.23% Cu from 232 to 326 m, withvalues increasing near the bottom of the hole. Later in 1997,a district-wide, helicopter-borne aeromagnetic survey wasflown, resulting in definition of a prominent magnetic highcentered immediately south of the previously drilled area.

Newcrest geologists appreciated that the 94-m gold inter-cept was of porphyry rather than epithermal style, and rec-ommended follow-up core drilling to test the gold-coppermineralization beneath the covered area farther south andwest and at greater depth. Eventually, only two reverse-cir -culation holes, totaling 532 m, were drilled on the porphyrygold-copper target during the 1997 to 1998 season, both

GEOLOGY OF THE CASPICHE PORPHYRY GOLD-COPPER DEPOSIT, MARICUNGA BELT, NORTHERN CHILE 587

0361-0128/98/000/000-00 $6.00 587

Ledge

Surface projection of Caspiche Au-Cu deposit

Ledge

Santa Cecilia West

FIG. 3. View of Caspiche looking west toward the highest point of the contiguous Santa Cecilia alteration zone (~4,600 mabove sea level). The surface projection of the concealed porphyry gold-copper deposit, ~800 m north-south (red line), liesbetween two prominent silicified and vuggy residual quartz ledges (pale tones). The late-mineral diatreme lies beneath thevalley (midview) separating Caspiche and Santa Cecilia.

designed to test the magnetic high. The best results, 224 maveraging 0.14 g/t Au, were discouraging and led to the optionbeing terminated, but without it being appreciated that theholes had penetrated a weakly mineralized, intermineral por-phyry intrusion (see below).

Following an additional eight years of inactivity, the prop-erty was again optioned by Anglo American, this time to Ex-eter Resource Corporation, a Toronto-listed junior explorer.During the 2005 to 2006 field season, Exeter conducted geo-logic mapping, rock-chip sampling, and a controlled sourceaudio-frequency magnetotelluric (CSAMT) survey, followedin early 2007 by reverse-circulation drill testing of high sulfi-dation targets beyond the immediate Caspiche porphyryprospect. As documented more fully by Van Kerkvoort et al.(2009), only at the very end of the 2006 to 2007 season was itelected to drill the Caspiche prospect itself, with time remain-ing for just one hole, which returned 304 m averaging 0.90 g/tAu and 0.26% Cu. In early 2008, this discovery hole was fol-lowed up with a campaign of core drilling, with the third holereporting 793 m at 0.96 g/t Au and 0.40% Cu (Van Kerkvoortet al., 2009). By the beginning of 2012, ~63,000 m of coredrilling had been completed at Caspiche, the results of which,coupled with engineering studies, enabled estimation of aproven plus probable sulfide reserve of 967 Mt grading 0.57g/t Au and 0.23% Cu, using an economic cutoff formula basedon recovery, mining cost, and other parameters, but approxi-mating 0.3 g/t Au equiv. To this should be added an overlyingproven plus probable oxide reserve of 124 Mt averaging 0.38g/t Au, at a 0.18 g/t Au equiv cutoff. The reserves contain atotal of 19.3 million ounces of gold and 2.1 Mt of copper (Ex-eter Resource Corporation press release, 17 January 2012).

This paper describes the geologic setting, alteration-miner-alization characteristics, and evolutionary history of the com-pletely concealed Caspiche porphyry gold-copper deposit overits known 1,400-m vertical extent, with particular emphasis ona spectrum of variably mineralized breccias. The geologic de-scription and interpretation, based on logging of all drill core,reflects the latest version of the model that has guided the exploration program; it complements the detailed discoverycase history presented elsewhere (Van Kerkvoort et al., 2009).

Regional SettingThe Maricunga belt, situated along the southwestern edge

of the Altiplano-Puna plateau, spans the boundary betweenthe currently active Central Volcanic Zone of the Andes and,to the south, the nonvolcanic segment between 28° and33°S—a change caused by the transition from moderate-angle to flat-slab subduction (Thorpe et al., 1982; Cahill andIsacks, 1992; Kay and Mpodozis, 2002). Arc magmatism inthe Maricunga belt was initiated in the late Oligocene to earlyMiocene, at ~26 Ma, when reorganization of plate motionsbeneath the Pacific Ocean led to breakup of the Farallonplate into the Nazca and Cocos plates (Lonsdale, 2005).These processes were accompanied by increased oceaniccrust generation at the East Pacific Rise (Conrad and Lith-gow-Bertelloni, 2007) and acceleration of plate convergenceat the Peru-Chile trench (Pardo-Casas and Molnar, 1987).

The basement to the Maricunga belt comprises metasedi-mentary and felsic volcanic rocks and granitoids of late Pale-ozoic age, like those that form large tracts of the Cordillera

Frontal farther east in Argentina (Mpodozis and Ramos,1990). These rocks are capped by Triassic bimodal volcanicand sedimentary rocks deposited in extensional rift basins andJurassic to Early Cretaceous marine and continental sedi-mentary rocks that accumulated in a backarc basin developedthroughout northern Chile (Mercado, 1982; Mpodozis et al.,1995; Cornejo et al., 1998; Iriarte et al., 1999; Arriagada et al.,2006). Late Cretaceous to Eocene arc to backarc volcanicrocks and continental redbeds formed after the tectonic in-version of the basin and eastward migration of the Andean arc(Cornejo et al., 1998; Iriarte et al., 1999; Mpodozis andClavero, 2002). In the southern Maricunga belt, betweenCaspiche and Cerro Casale (Fig. 2), as well as farther east andsouth, the Paleocene volcanic units are covered by Oligocenecontinental sedimentary sequences, composed of red sand-stone, siltstone, and conglomerate, and including evaporitichorizons (Mpodozis and Kay, 2003). The siliciclastic redbeds,which crop out west of Caspiche and are a major host rock tothe porphyry deposit at depth (see below), are most likelyOligocene in age because, farther south, they unconformablyoverlie Eocene redbeds and volcanic rocks (Mpodozis et al.,1991; C. Mpodozis, written comm., 2011).

The late Oligocene and Miocene Maricunga volcanicrocks, predominantly medium- to high-K calc-alkaline an-desite to dacite in composition (Kay et al., 1994; McKee etal., 1994; Mpodozis et al., 1995), are chiefly the products ofextinct stratovolcanoes and volcanic dome complexes, whichdisplay varied degrees of erosional dissection (Vila and Silli-toe, 1991; Kay et al., 1994; Mpodozis et al., 1995). The por-phyry and high sulfidation epithermal deposits, closelylinked in places (e.g., Caspiche, La Pepa; Fig. 2), have beenshown by K-Ar and 40Ar/39Ar dating (Sillitoe et al., 1991;Mpodozis et al., 1995) to have been generated beneath boththese main volcanic landforms during two principal metallo-genic epochs: latest Oligocene to earliest Miocene (26–21Ma) and mid-Miocene (14–10 Ma). The Caspiche deposit isassigned to the older of these two epochs on the basis of anRe-Os age for molybdenite of 25.38 ± 0.09 Ma (H. Stein,unpub. report for Gold Fields Chile S.A., 2009; see onlineSupplementary Data), which was determined using themethodology described by Zimmerman et al. (2008).Caspiche is penecontemporaneous with and probably geneti-cally related to the contiguous Santa Cecilia alteration zone(Fig. 3), which is hosted by a stratified volcanic sequence, in-cluding ignimbrite flows, and at least one dacite dome. SantaCecilia is known to host minor, albeit locally high grade, highsulfidation epithermal gold mineralization, which returnedessentially identical K-Ar ages of 24.3 ± 0.7 and 24.1 ± 0.8 Mafor hydrothermal sericite and alunite, respectively (Sillitoe etal., 1991).

The latest Oligocene to earliest Miocene (26–21 Ma) vol-canic activity and precious and base metal mineralizationlikely took place in a neutral to weakly extensional tectonicsetting, which gave way to contraction consequent upon ini-tial slab flattening between 20 and 18 Ma (Mpodozis et al.,1991, 1995; Kay et al., 1994, 2008; Kay and Mpodozis, 2001).The contraction resulted in diminished volcanism, compres-sive deformation, and crustal thickening (Kay et al., 1994;Mpodozis et al., 1995). The gold-deficient Caserones por-phyry copper-molybdenum deposit and nearby prospects,

588 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 588

immediately south of the Maricunga belt (Fig. 1), were gen-erated during this compressive event (Mpodozis and Kay,2003; Perelló et al., 2003; Sillitoe and Perelló, 2005). Volcanicactivity under more extensional conditions was again wide-spread and voluminous between 17 and 12 Ma, an intervalthat concluded with the second precious and base metal min-eralization event. This was followed by renewed contractionaltectonism and marked geographic restriction of volcanism be-tween 11 and 5 Ma, when all volcanic activity ceased in theMaricunga belt prior to reestablishment at the site of the pre-sent-day magmatic arc, some 60 km farther east (Kay et al.,1994, 2008; Mpodozis et al., 1995). Throughout the Mari-cunga belt, the volcanic centers and associated mineralizationthat formed during both of the noncompressive epochs dis-play a clear control by steep, NW-, NE-, and E-striking faults,particularly the first of these (Mpodozis et al., 1995; Fig. 2);however, E-striking faults are most prominent in theCaspiche area (Mpodozis et al., 1991). Glaciation affected thehigher parts of the Maricunga belt during the Plio-Pleis-tocene, giving rise to the final exhumation stages of the lateOligocene to mid-Miocene alteration zones and any con-tained mineralization.

Deposit GeologyThe principal rock types defined at Caspiche may be as-

signed to four broad units: premineral sedimentary host

rocks; volcanic breccia, which also predates all the potentiallyeconomic mineralization; five discrete porphyry intrusions,spanning the alteration-mineralization sequence; and a late-mineral diatreme breccia (Figs. 4, 5). To these may be addedrelatively small volumes of hydrothermal breccia.

Sedimentary rocks

Sedimentary rocks surround the composite Caspiche por-phyry stock on all sides below elevations ranging from 3,870to 3,700 m above sea level, which is roughly 500 to 750 mbelow the surface (Figs. 4, 5). The rocks typically comprise amonotonous sequence of black, hornfelsed, and highly alteredsandstone and siltstone, which may display little obvious tex-tural variation over tens of meters. In places, however, shal-lowly dipping, relict bedding is observed, confirming theirsedimentary origin (Fig. 6a). If, as seems probable, these sili-ciclastic rocks are correlative with the nearby redbed se-quence of inferred Oligocene age (see above), the original diagenetic hematite was transformed to magnetite consequentupon the hornfelsing and pervasive potassic alteration, whichalso gave rise, in places, to a prominent, mottled texture pro-duced by biotite clots. Notwithstanding the ubiquitous pres-ence of magmatic-hydrothermal anhydrite throughout thedeeper parts of the Caspiche deposit (see below), concentra-tions of centimeter-sized anhydrite nodules in several thinsiltstone horizons may be evaporitic in origin, possibly formed


0361-0128/98/000/000-00 $6.00 589

A 470,

500

mE

471,

000

mE

471,

500

mE

6,937,000 mE

6,937,500 mE

A’ North

400 m

Late-mineral quartz diorite porphyry (phase 5)

Late-mineral diatreme breccia

Hydrothermal breccia


Late intermineral quartz diorite porphyry and associated breccia (phase 3)

Early intermineral quartz diorite porphyry and associ-ated breccia (phase 2)

Early diorite porphyry (phase 1)

Sandstone and siltstone

Fault

Drill hole on plan

mi-etLa

tdroyH

mi-etLa

mi-etLa

pha(yryhpporetioridztruaql arnemi

aicecbrl amrhe

aieccbremertaidl arnemi

spha(yryhpporetioridztruaql arnemi

ylraE

detaylraE

)4es

)5es

Fau

ndsaS

m400

)1espha(yyrhorppetioridy

)2eshap(aieccbrdyrhporpetioridztruaql arneimrten iy

tl

neotstilsdnaeontnds

m

oci-assndayyr

ociaass

ten ietLa

)3espha(aieccbrdetocia

ayryhporpetioridtzquarlarnemirte

nd a llirD

nalponeloh

FIG. 4. Geologic level plan at 3,700-m elevation of the Caspiche porphyry gold-copper deposit. Note that the porphyryphases tend to become progressively younger outward from the early phase 1 diorite porphyry, which constitutes the high-grade core to the system. Projections of drill holes also shown.

during diagenesis (e.g., Dean et al., 1975) in a former playa-lake environment (cf. Mpodozis and Kay, 2003).

The deeper parts of the sedimentary sequence, beneath ap-proximately 400 m of the sandstone and siltstone, include twosedimentary breccia horizons (Fig. 6b). The thicker (~100 m),upper horizon has a deposit-wide, ~2-m-thick bed of calcare-ous rock, transformed to garnet-diopside-epidote skarn, at itsbase. Correlation of these marker horizons over distances ofseveral hundred meters confirms that the sedimentary se-quence is essentially flat lying (Fig. 5). The breccias, transi-tional to conglomerate in places, contain abundant clasts ofdistinctive rhyolite porphyry, almost certainly derived from

the late Paleozoic and Triassic basement, in which such felsicvolcanic rock types are widespread (e.g., Mpodozis andRamos, 1990; see above). The lower sedimentary brecciahorizon is directly underlain by andesitic volcanic rocks,chiefly flows (Fig. 5).

Minor bodies of probable andesite porphyry, characterizedby centimeter-sized plagioclase phenocrysts in a black, fine-grained, and highly altered groundmass, cut the siliciclasticsedimentary rocks locally, particularly near their upper con-tact. The bodies are clearly intrusive because of the presenceof chilled margins, but it is still uncertain if they constitutesills, dikes, or a combination thereof.

590 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 590

SW

4,000 masl

3,500 masl

10,0

00 m

E

9,00

0 m

E

9,50

0 m

E

10,5

00 m

E

NE

Overburden



Hydrothermal breccia

Late intermineral quartz diorite porphyry and associated breccia (phase 3)

Early intermineral quartz diorite porphyry and associ-ated breccia (phase 2)


Volcanic breccia

Sedimentary breccia

Sandstone and siltstone

Fault

Drill hole on section

400 m

Andesitic volcanic rocks

dl arnemi-etLa

ql arnemi-etLa

rburdenevO

aieccbremertaid

)5espha(yryhpporetioridztrua

m400

eontndsaS

rtanemideS

eccbrcincaloV

neotstilsdna

aieccbryr

aiecc

orppetioridylraE

p(aieccbrdetaneimrteniyylraE

eccbrdetociaassnemirten ietLa

l amrhetdroyH

)1espha(yryhorp

)2eshapassndayryhporpetioridztruaql arne

)3espha(aieccnd ayryhporpetioridtzquarlar

aicecbr

oci-ass onelohllirD

tlFau

ovcitisdenA

onitsec

sckrocincalo

FIG. 5. Representative geologic section through the Caspiche porphyry gold-copper deposit, showing all the main rocktypes recognized. Section line shown in Figure 4.


0361-0128/98/000/000-00 $6.00 591

a b

d c

e f

1 cm

1 cm

1 cm

1 cm

1 cm

1 cm

Qz veinlet xenolith

a

b

c

d

m1 c

m1 c

e

f

m1 c

m1 c

m1 c

m1 c

FIG. 6. Selected rock and alteration types from the Caspiche porphyry gold-copper deposit. a. Intensely biotitized, silici-clastic sedimentary wall rock from below 3,800-m elevation. Stratification is highlighted by more intense biotitization of silt-stone than of sandstone beds. b. Sedimentary breccia, containing large rhyolite clast of probable late Paleozoic or Triassicage, from biotitized siliciclastic sedimentary wall-rock sequence at ~3,400-m elevation. Note that veinlets cut the breccia. c.Premineral volcanic breccia, inferred to be part of an early diatreme, displaying patchy texture and advanced argillic alter-ation. The patchy texture, including the amoeboid clast form, is inherited from the original fragmental rock texture. Clastsare altered to kaolinite and pyrite, whereas the matrix is pervasively silicified. d. Potassic-altered part of the same premineralvolcanic breccia unit as depicted in c, containing flattened and aligned andesite porphyry clasts (fiamme) interpreted as for-mer magma blobs or pumice. e. Potassic-altered, premineral volcanic breccia containing an A-type quartz (Qz) veinlet xeno-lith indicative of a preexisting porphyry system. Note the typical concentration of hydrothermal biotite (black) in the clastsand K-feldspar in the matrix. f. Early phase 1 diorite porphyry affected by intense K-feldspar-dominant potassic alterationand associated magnetite and A-type quartz veinlet stockwork. g. Potassic-altered, early intermineral phase 2 quartz dioriteporphyry cut by weakly developed A-type quartz veinlet stockwork, and containing quartz (Qz) veinlet xenolith derived fromnearby phase 1 diorite porphyry. Note prominent quartz (Qz) phenocryst. h. Potassic-altered, early intermineral phase 2quartz diorite porphyry intrusion breccia developed near contact of the quartz diorite porphyry with sedimentary wall rocks.Sandstone and siltstone clasts are biotitized, whereas the quartz diorite porphyry cement is K-feldspar rich. Note D-typepyrite veinlet at top of image. i. Late-mineral, polymict hydrothermal breccia from peripheral chlorite-sericite alterationzone. The breccia contains abundant silicified, veinlet quartz (Qz), and quartz-veined clasts in a silicified rock-flour matrix.j. Late-mineral, polymict hydrothermal breccia displaying advanced argillic alteration, and cemented by alunite (al) andkaolinite (ka). Note quartz (Qz)-veined clast. k. Late-mineral diatreme breccia containing polymict clast population, includ-ing vuggy residual quartz (Qz), silicified, and pyrite (py)-veined clasts. Matrix is dominated by dacitic tuff. l. Phase 2 quartzdiorite porphyry, from ~3,400-m elevation, affected by potassic-calcic alteration. The veinlet contains actinolite (Ac)-mag-netite (Mg) and quartz (Qz)-K-feldspar (Kf)-rich segments, and has a K-feldspar alteration halo.

Volcanic brecciaThroughout the prospect area, the siliciclastic sequence is

overlain by a 500- to 750-m-thick volcanic breccia (Fig. 5),which has an apparently restricted areal extent of approxi-mately 5 km2.

The breccia is polymict and mainly composed of roundedto subangular clasts surrounded by an obscure, highly altered,fine-grained, fragmental matrix, which is suspected to bedominated by a tuff component (Fig. 6c-e). The clasts aretypically 1 to 4 cm in size, with isolated examples attaining 10cm, but with no hint of bedding or size sorting. In parts, someof the clasts have an amoeboid shape defined by cuspate mar-gins (Fig. 6c), and, very locally, aligned, lenticular clasts (fi-amme in a descriptive sense; Bull and McPhie, 2007) areprominent, some up to 8 cm long (Fig. 6d). The fiamme areeither magma blobs or pieces of pumice flattened by com-paction and, along with the cuspate clasts, are considered

juvenile in origin. Intense alteration of the breccia precludescertain identification of most component clast rock types, butthe remanent textures and characteristic absence of mag-matic quartz grains suggest that andesite/diorite predomi-nates. Clasts of hornfelsed siliciclastic sedimentary rock and,very locally, the intrusive andesite porphyry are particularlyprominent in the lower parts of the breccia, within ~50 m ofcontacts with these rock types, although suspected sedimen-tary clasts also occur elsewhere in the breccia. The youngestzircon population in a representative breccia sample yieldeda U-Pb age of 24.7 ± 0.7 Ma (V. Valencia, unpub. report forExeter Resource Corporation, 2010; see online Supplemen-tary Data), determined using the laser ablation-inductivelycoupled plasma-mass spectrometry (LA-ICP-MS) methoddetailed by Chang et al. (2006). This age is essentially thesame, within experimental error (2σ), as the Caspiche molyb-denite age (25.38 ± 0.09 Ma; see above).

592 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 592

g h Qz veinlet xenolith Qz phenocryst D veinlet

1 cm 1 cm

Siltstone clast

Sandstone clast

g

h

m1 c

m1 c

i

Truncated qz veinlet

Qz veinlet clast

j

Al Ka Al

Truncated py veinlet

k

Silicified clast

l

Ac Mg

Qz Kf

1 cm

1 cm

1 cm 1 cm

Silicified pyritic clasts

Vuggy qz clast

Truncated qz veinlet

i

j

m1 c

k

l

m1 c

m1 c

m1 c

FIG. 6. (Cont.)

The breccia clasts typically display different alteration min-erals to the matrix (Fig. 6c-e; see below), thereby enhancingappreciation of the fragmental texture. Although most clastsare internally homogeneous, recognition of veinlets confinedto a few of them is important because it confirms that somehydrothermal activity took place prior to breccia formation.The isolated, clast-confined veinlets comprise hydrothermalbiotite, specularite, or granular A-type quartz (as defined byGustafson and Hunt, 1975). In addition, isolated clasts of ei-ther A-type veinlet quartz (Fig. 6e) or silicified dioritic/an-desitic material are also widespread, particularly in the lower-most 100 m or so of the breccia.

The origin of this volcanic breccia remains to be fully re-vealed, but it seems most likely to be part of a small, localizedvolcanic center that overlapped both spatially and temporallywith development of the Caspiche porphyry gold-copper de-posit. In combination, the textural homogeneity, lack of bed-ding and size sorting over several hundred vertical meters,and localized presence of mineralized material would appearto preclude accumulation of the breccia in stratovolcano orash-flow caldera settings. A local origin is also supported bythe appearance of clasts of the underlying rock types on ap-proach to its basal contact. The breccia throughout much ofthe deposit area overlies a low-relief surface (Fig. 5), whichgives way northward to a much steeper slope, as implied bythe presence of the breccia below 3,700-m elevation. This de-pressed area could be interpreted as one sector of an upward-flared vent, which fed a restricted basin, although an incisedpaleosurface cannot yet be discounted.

Porphyry intrusions

Two main porphyry intrusions, early and early intermineralin timing (phases 1 and 2), constitute the well-mineralizedpart of the composite Caspiche stock, which is abutted toboth east and west by a late intermineral phase 3 (Figs. 4, 5).In contrast, the two late-mineral phases 4 and 5 have so faronly been encountered farther west (Fig. 4). Together, thewell-mineralized phases 1 and 2 measure roughly 500 × 500m in plan view (Fig. 4), with little appreciable change in sizeover their defined 1,400-m vertical extent (Fig. 5).

The early phase 1 porphyry is dioritic in composition andcontains sparse, 1-mm-sized quartz phenocrysts. Larger pla-gioclase and biotite phenocrysts are abundant and accompa-nied by subordinate hornblende. The original texture of thephase 1 porphyry is partly obliterated by intense alterationand veining (Fig. 6f), the latter dominated by a multidirec-tional, A-type quartz veinlet stockwork, which, near the apexof the intrusion, constitutes >50% of rock volume.

The other four intrusions, phases 2 to 5, are quartz dioriteporphyries. The intermineral phases 2 and 3 are texturallybetter preserved and coarser grained than the dioritic phase1; they contain prominent quartz, besides plagioclase, biotite,and hornblende phenocrysts (Fig. 6g). The phase 3 porphyryis noticeably coarser grained and contains larger quartz phe-nocrysts than the better-mineralized phase 2. The latter is cutby abundant, relatively narrow (mostly <1 cm), A-type quartzveinlets (Fig. 6g), but truncates many of the A-type quartzveinlets in the phase 1 diorite porphyry, including all thosewith widths of 2 to 4 cm. In several places, the peripheral partsof the phase 2 porphyry body comprise intrusion breccia:

abundant phase 1 diorite porphyry or sedimentary wall-rockclasts and quartz veinlet fragments in a porphyry rock matrix(Figs. 4, 5, 6h). In contrast, the phase 3 quartz diorite por-phyry is only weakly veined, and displays low alteration inten-sity and partial preservation of magmatic biotite and mag-netite. The late-mineral phase 4 and 5 quartz dioriteporphyries are texturally well preserved and nearly veinletfree. The earlier phase 4 displays extremely weak potassic al-teration, whereas phase 5 is characterized by intense propy-litic alteration; both contain <0.02 g/t Au and <100 ppm Cu.

The composite porphyry stock has a highly irregular formin plan and is associated with a number of dikes, which to-gether suggest both northwest and northeast structural con-trols on its emplacement (Fig. 4). The phase 1 diorite porphyry,constituting the core of the stock, occurs as a ~300-m-long,sigmoidal body ranging from ~50 to 150 m wide (Fig. 4), butis blind and concealed beneath up to 150 m of premineral vol-canic breccia (Fig. 5). The phase 2 quartz diorite porphyry,comprising at least two discrete pulses, occurs as a discontin-uous shell around phase 1; however, the largest volume ofphase 2 occurs as an NE-trending, 400-m-long body along thesoutheastern side of phase 1 (Fig. 4). The internal position ofthe phase 1 porphyry relative to the enveloping phase 2 is un-usual and, in map pattern, might be taken to suggest that theformer is younger than the latter. Nonetheless, the reverse isamply confirmed by the higher veinlet intensity and metalcontent in phase 1 as well as the truncation of some A-typequartz veinlets by, and presence of A-type quartz veinlet frag-ments within, phase 2 (Fig. 6e). The form and diameter ofphase 1 is roughly the same irrespective of whether it is en-veloped by the sedimentary basement or phase 2 quartz dior-ite porphyry (Fig. 5), implying that little or no phase 1 por-phyry was assimilated or otherwise removed during phase 2emplacement. The phase 3 quartz diorite porphyry intrusionmeasures at least 400 m in a northeast direction, and is nestedinto the western margin of phase 2 (Fig. 4). The phase 4quartz diorite porphyry locally truncates the western side ofphase 2 and, in turn, is cut by phase 5 quartz diorite porphyry,which also occurs as large blocks in the late-mineral diatreme(Fig. 5).

Hydrothermal breccias

Several, volumetrically minor hydrothermal breccias arepresent at Caspiche, chiefly in the peripheral and shallowparts of the deposit (Figs. 4, 5). The best-defined breccias areclearly dike like and <10 m wide. All the breccias are polymictand contain a variety of veined and mineralized clasts, whichrange from angular to subrounded in form.

The peripheral breccias are mainly cemented by fine-grained quartz (Fig. 6i) or chlorite, whereas the shallow ex-amples, within the advanced argillic zone (see below), arecharacterized by alunite ± kaolinite cement accompanied lo-cally by minor pyrite and enargite (Fig. 6j). Most of the brec-cias are largely barren, reflecting their late timing, althoughthose that display the advanced argillic alteration can containminor copper and gold.

Late-mineral diatreme

A large breccia body, which dips west at approximately 30°to 40° and then steepens downward, truncates the western


0361-0128/98/000/000-00 $6.00 593

side of the Caspiche porphyry deposit at shallow levels (Figs.4, 5). The breccia extends for a distance of at least 1,100 m ina north-south direction and has a maximum drilled thicknessof 700 m. The east-west dimension remains undefined, al-though the breccia is anticipated to underlie the full width ofthe valley that separates Caspiche from the Santa Cecilia al-teration zone (Fig. 3).

The breccia is highly polymict, matrix supported, and con-tains abundant mineralized clasts in proximity to its contactwith the Caspiche porphyry deposit. Besides quartz-veinedclasts and quartz-veinlet fragments, there are also prominent

pieces of vuggy residual quartz (Fig. 6k), showing that brec-ciation postdated development of some of the high sulfidationepithermal ledges (see below). Away from the contact zone,tuffaceous material, apparently compositionally similar to thelate-mineral phase 5 quartz diorite porphyry, is readily recog-nizable as a matrix to texturally varied andesite and andesiteporphyry clasts. The breccia contains blocks of phase 5 por-phyry, up to ~50 m in size, some of them displaying intenseepidote-rich, propylitic alteration (Figs. 5, 7). Many of theblocks were also brecciated in situ, with the introduced ma-trix comprising fine-grained, hematite-impregnated quartz.

594 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 594

SW

4,000 masl

3,500 masl

10,0

00 m

E

9,00

0 m

E

9,50

0 m

E

10,5

00 m

E

NE

Advanced argillic: silicification, vuggy quartz ± alunite

Advanced argillic: quartz - kaolinite ± pyrophyllite ±dickite

Chlorite -sericite

Propylitic: epidote -chlorite

Potassic: K-feldspar -albite -biotite

Potassic: biotite

Base of leached capping

Sulfate front

Bornite front

Overburden

Fault

Drill hole on section

400 m

Potassic-calcic: actinolite - K-feldspar

EN

tiorlhC

etickidncvadA

ncvadAAd

eticirse-e

ropyp±etiniloka-ztruaq:clilirgadenc

quarygugv,onitcaatificilis:clilirgadenc

±etlliyhrop

etiunla±ztquar

ornB

fluS

aseB

ntorfetiorn

ntorfetaatf

ngippcadheeaclff oase

400

m

tassoP

tassoP

liyproP

tassoP

etitoib:citass

etitoib-etibla-radsplef-K:citass

etiorlhc-etodipe:citli

radsplef-K-etlionitac:ciclca-citass

llirD

Fau

evO

onitseconeloh

tl

rburdene

FIG. 7. Alteration section showing the spatial relationships between the principal alteration assemblages and the rocktypes depicted in Figure 5, Caspiche porphyry gold-copper deposit. The positions of the base of the leached capping (baseof supergene sulfide oxidation), sulfate front (base of anhydrite and gypsum removal), and bornite front (top of megascopi-cally visible chalcopyrite-bornite assemblage) are also shown. Note confinement of vuggy residual quartz ledges to shallowparts of the system.

Although much of the breccia is nonbedded and nonsorted,there are local intervals of normally graded, silt- to pebble-sized material, which are interpreted as surge deposits. Onesuch interval has abrupt contacts against the nonbedded brec-cia, suggesting an origin as a sunken block derived from sub-aerial tuff accumulations (cf. Sillitoe, 1985; Davies et al.,2008), whereas other bedded intervals may have formed inplace.

On the basis of these characteristics and close similarities tolate diatreme breccias in porphyry systems elsewhere (e.g.,Sillitoe, 1985), it seems clear that the breccia fills a diatremevent. To date, only the upward-flared, eastern contact hasbeen partly defined, with the bulk of the body, including thethroat of the vent, lying farther west. The diatreme is a late,but not the final manifestation of the Caspiche system be-cause it not only contains clasts of ledge material but is alsocut by a few residual vuggy quartz ledges, one at least 300 mlong.

Hypogene Alteration and Mineralization FeaturesFive alteration types are depicted in Figure 7: potassic,

propylitic, potassic-calcic, chlorite-sericite, and advancedargillic, with the first and last of these being by far the mostwidely developed and economically important. Because thealteration zones overprint one another, commencing with thepotassic and ending with the advanced argillic, combinationsof alteration types are commonplace, although, for mappingpurposes, the predominant assemblage everywhere takesprecedence.

Potassic

Potassic alteration becomes progressively more widespreaddownward in the Caspiche system, and is essentially the onlyalteration type present between the ~3,600- and 3,400-m el-evations (Fig. 7). The potassic zone comprises two distinct al-teration assemblages, the earlier biotite dominated and theyounger containing K-feldspar, albite, and subordinate biotite(Fig. 7). Both contain >5 vol % of hydrothermal magnetitewhere not appreciably overprinted by the later alterationtypes. The biotite-rich type dominates the hornfelsed silici-clastic sedimentary rocks (Fig. 6a) as well as affecting manyclasts in the volcanic breccia (Fig. 6e). The K-feldspar-richpotassic alteration is widely developed in the early and inter-mineral porphyries as well as in the matrix to the volcanicbreccia (Fig. 6d-h). K-feldspar flooding is far more abundantthan is its confinement to veinlet selvages. The potassic zoneis characterized by widespread quartz ± magnetite veinlets,the quartz being granular and, hence, typical of A-type vein-let generations (Gustafson and Hunt, 1975; Fig. 6f, g). Sul-fide-free, magnetite-only veinlets predate most of the A-typeveinlet generations. Banded quartz veinlets, typical of theMaricunga porphyry gold deposits (Vila and Sillitoe, 1991;Muntean and Einaudi, 2000), are absent.

The potassic alteration contains extremely fine grainedchalcopyrite and subordinate pyrite in its shallower parts,with the pyrite/chalcopyrite ratio decreasing gradually down-ward and increasing outward. At a depth of about 700 m inthe phase 1 diorite porphyry and its immediate wall rock, vis-ible pyrite disappears entirely from the potassic alteration andbornite becomes progressively more abundant, although

everywhere subordinate to chalcopyrite (chalcopyrite/bornite~7). The bornite front is marked in Figure 7. The sulfide min-erals occur as disseminated grains in the quartz ± magnetiteveinlets as well as throughout the surrounding rock. A widevariety of pyrite-dominated veinlets, some having sericitichalos (D-type; Gustafson and Hunt, 1975), cut the potassic al-teration and quartz ± magnetite veinlet stockwork, as do late-stage anhydrite veinlets, some containing a few grains ofpyrite, chalcopyrite, and/or molybdenite, but others com-pletely barren.

Propylitic

Propylitic alteration was encountered by drilling 600 mwest of the mineralized Caspiche stock (Fig. 7), and is ex-pected to completely encircle it at depth, but also occurs asthe earliest recognizable alteration type in the late-mineral,phase 5 quartz diorite porphyry and parts of the late-mineraldiatreme breccia (Fig. 7). The alteration is characterized bychlorite replacement of mafic minerals, epidote after plagio-clase phenocrysts, and several percent of magnetite. The epi-dote content is >5 vol %, with both veinlets and dispersedgrains being commonplace. The pyrite content also exceeds 5vol % in the propylitic halo, but attains only 1 to 2 vol % in thelate-mineral porphyry and diatreme breccia.

The peripheral propylitic alteration, most prominently thecontained epidote veinlets, overprints the fringe of the potas-sic zone, although the two alteration types are assumed to bebroadly contemporaneous. A similar situation was reportedfrom the Bajo de la Alumbrera porphyry copper-gold deposit,Argentina (Proffett, 2003).

Potassic-calcic

Potassic-calcic alteration, defined by the presence of K-feldspar and actinolite and largely lacking biotite, is mainlyconfined to the central parts of the early intermineral, phase2 quartz diorite porphyry and its host rocks below ~3,400-melevation, although it is not widely developed in the part ofthe deposit depicted in Figure 7. The downward change frompotassic to potassic-calcic assemblages takes place relativelyabruptly, over just 10 to 20 m. The potassic-calcic zone ischaracterized by actinolite intergrown with 10 vol % mag-netite as veinlets, clots, and disseminations. The wider vein-lets, up to 2 cm across, tend to be segmented, with alternat-ing actinolite-magnetite and granular quartz intervals (Fig.6l). The K-feldspar forms prominent veinlet halos, some of itas centimeter-sized crystals, as well as occurring as a ground-mass alteration product. Diopside, albite, calcite, anhydrite,chalcopyrite, and trace bornite are subsidiary veinlet compo-nents. Some of the calcite fills numerous fractures in coarse-grained diopside.

In the transition between the potassic-calcic and overlyingpotassic zones, the actinolite and magnetite are observed tooverprint and destroy preexisting biotite, a process that alsoeliminates most of the contained chalcopyrite and bornite, al-though minor amounts of these sulfide minerals are retainedin the wider actinolite-magnetite-quartz veinlets. Conse-quently, there is a marked drop in both total sulfide and cor-responding gold and copper contents on passing from thepotassic to potassic-calcic zones, with the latter averaging 0.13g/t Au and 0.06% Cu.


0361-0128/98/000/000-00 $6.00 595

Chlorite-sericiteChlorite-sericite alteration occurs marginal to the potassic

core of the system, beyond the gold-copper orebody, but in-ternal to the propylitic zone (Fig. 7). The late intermineral,phase 3 quartz diorite porphyry is particularly susceptible tothis alteration type. Mafic minerals are transformed to chlo-rite and plagioclase to illite and/or sericite (fine-grained mus-covite). The dominance of illite over sericite was confirmedusing a portable spectrometer. Pyrite was also introduced inboth veinlet and disseminated forms.

The upper parts of the potassic zone in both the volcanicbreccia and porphyries have been partially overprinted by illite± smectite, the latter causing the sawn drill core to take on ayellow stain after only a few months of storage (Fig. 6g). Themain effects of this overprint are to convert plagioclase to illiteand magnetite to hematite, the latter in the form of both mar-tite and accompanying specularite. Much of the biotite re-mains stable, although it is locally retrograded to chlorite. Car-bonate is also present in places. Only below the top of the zonecontaining unaltered magnetite is the K-feldspar-rich potassiczone relatively free from the effects of the illite overprint.

Where the chlorite-sericite/illite alteration is strongest, it ismapped separately (Fig. 7), although it is commonly exten-sively overprinted by kaolinite and, hence, included in the ad-vanced argillic zone. This progression is recorded by plagio-clase phenocrysts that have illite cores and irregular kaoliniterims.

Advanced argillicThe shallower parts of the Caspiche system, to an average

depth of 300 to 400 m, but a maximum depth of approximately1,000 m (3,400-m elevation) within steep fractures, are over-printed by advanced argillic alteration dominated by a quartz-kaolinite assemblage; however, it also contains widespread albeit subordinate pyrophyllite and dickite as well as trace diaspore (Fig. 7). Tiny (1 mm), disseminated rosettes of bluish-colored dumortierite occur as a widespread accessory mineralin the deeper parts of the advanced argillic zone, but give wayto black, centimeter-sized rosettes of black tourmaline (schorl)closely associated with kaolinite, sericite, chalcopyrite, andpyrite at the base of the central portion of the advancedargillic zone. The martite and specularite, products of theillite overprint, along with any remanent magnetite, are en-tirely converted to pyrite.

Where advanced argillic alteration affects the volcanicbreccia, patchy texture is developed as a result of replace-ment of the clasts by kaolinite and pyrite and the matrix by ex-tremely fine grained, gray-colored chalcedony (Fig. 6c; cf.Yanacocha; Gustafson et al., 2004). The patchy texture mim-ics the similar texture developed during the preceding potas-sic alteration, in which biotite-rich clasts are surrounded byK-feldspar-rich matrix (Fig. 6d, e; see above). Such inheri-tance of patchy-textured advanced argillic alteration from apreexisting alteration type affecting fragmental rock is alsodescribed from the Oyu Tolgoi porphyry copper-gold deposit,Mongolia (Khashgerel et al., 2008).

The advanced argillic alteration contains a high sulfidation-state sulfide assemblage in which chalcopyrite and pyrite arerimmed and, possibly, partially replaced by complex, micro-scopic intergrowths of enargite, luzonite, tennantite, bornite,

digenite, chalcocite-group minerals, and covellite (Fig. 8). Thechalcocite group and covellite are both hypogene and shouldnot be confused with the minor chalcocite resulting from su-pergene enrichment (see below). This dominantly dissemi-nated, high sulfidation sulfide assemblage is largely confined tothe mineralized porphyry stock and its immediate wall rocks.

Siliceous ledges occur within the quartz-kaolinite zone atshallow depths, generally above 4,000-m elevation (Fig. 7),where they represent the main upflow conduits within the ad-vanced argillic lithocap. The drilled ledges, up to 35 m wide,comprise silicification, vuggy residual quartz, and/or semi-massive pyrite and marcasite where unoxidized. Enargite, orpiment, and native sulfur can occur as paragenetically lateadditions in unoxidized parts of the ledges. Several ledges areauriferous whereas others are totally barren. Some of theledges have prominent, but typically narrow (<5 cm) alunitehalos, although these can coalesce locally into broader quartz-alunite zones where ledges are closely spaced along thesouthern margin of the system (see below). The alunite, likemuch of the kaolinite, replaces plagioclase phenocrysts. Dick-ite is most evident as a final infill of quartz-pyrite ± enargiteveins interpreted as the roots of the siliceous ledges.

The surface exposures (Fig. 3) and information from above4,200-m elevation suggest that the siliceous ledges are pref-erentially developed around the periphery of the Caspicheporphyry deposit, particularly to the south. However, it is sus-pected that ledge development simply penetrated moredeeply around the system margins, thereby leading to earliererosional removal of the ledges that once capped the centralparts of the system.

Advanced argillic alteration, including siliceous ledges, isdeveloped far beyond the Caspiche deposit, at least 3 km eastand 500 m north, and to depths of at least 400 m in places.Some of the ledges are irregularly mineralized with enargite,luzonite, and associated gold. Late-stage addition of orpiment

596 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 596

En

Py

Cv

Cv

Cp Py

Py

Te

0.05 mm

mm5 00.

FIG. 8. Copper mineralization from the lower part of the advanced argillicalteration zone, Caspiche deposit. Chalcopyrite (Cp) and associated pyrite(Py) grains, formed during potassic alteration, are rimmed by a high sulfida-tion-state sulfide assemblage, comprising covellite (Cv), enargite (En), andtennantite (Te). With a ×20 hand lens, the high sulfidation species appear asan indivisible black rim. Polished section under plane-polarized, reflectedlight.

is also widespread in some of these peripheral ledges, ac-counting for arsenic contents of several percent in places.

Late-stage gold-zinc mineralization

Scattered quartz ± calcite veinlets, containing straw-col-ored, low-iron sphalerite, galena, chalcopyrite, and pyrite, cuta spectrum of alteration types, ranging from pyrite-rich,quartz-kaolinite to pyrite-poor, illite-overprinted, potassic as-semblages. The veinlets lack megascopically observable asso-ciated alteration. Along with accompanying sulfide dissemi-nations, they are particularly concentrated and gold rich in ashallowly dipping zone that abuts the eastern flared contact ofthe late-mineral diatreme breccia. A few such veinlets also cutoverlying parts of the breccia, thereby confirming that thegold-zinc zone is the product of an end-stage hydrothermalevent.

Hypogene Metal Distribution

Porphyry deposit

The Caspiche deposit displays a classic hypogene metal-zoning pattern over its known 1,400-m vertical extent (Figs. 9,10). Gold and copper contents correlate well, with the high-est average tenors (~1 g/t Au, 0.4 % Cu) being largely con-fined to the phase 1 diorite porphyry intrusion and its imme-diate wall rocks (Fig. 9a, b). The phase 2 and 3 quartz dioriteporphyries are characterized by roughly 50 and 20%, respec-tively, of the phase 1 diorite porphyry grades. Notwithstand-ing the unusually high Au/Cu ratio at Caspiche, the closeAu/Cu correlation is typical of that observed in most gold-richporphyry copper deposits (e.g., Sillitoe, 2000, 2010). At depthin the phase 1 diorite porphyry intrusion, where bornite ac-companies chalcopyrite (Fig. 7), copper contents increase, as


0361-0128/98/000/000-00 $6.00 597

Au ppm 0.3 0.5 1.0

Cu % 0.1 0.2 0.3

As ppm 150 300 500

a b

c d

Zn ppm 500 750

1000

FIG. 9. Element distribution contour plots for the same Caspiche section shown in Figures 5 and 7. a. Gold. b. Copper.c. Arsenic. d. Zinc. Note the central, broadly coincident positions of gold and copper and shallow concentration of arsenicassociated with the advanced argillic alteration overprint. Appreciable zinc is confined to the gold-zinc zone immediately be-neath the eastward-flared, late-mineral diatreme contact. Background geologic outline taken from Figure 5.

expected, but gold grades tend to remain constant. Both goldand copper grades decrease more or less progressively out-ward to the inner edge of the propylitic halo, which containsonly low-order metal values, whereas, downward, they dimin-ish abruptly as the potassic gives way to sulfide-poor, potassic-calcic alteration.

Molybdenum at Caspiche is concentrated in a broad, annu-lar zone that surrounds and partially overlaps the gold-coppercore to the deposit (Fig. 10). The molybdenite responsible forthe outer parts of this annulus occurs in a variety of forms, in-cluding quartz-molybdenite-pyrite and molybdenite-only vein-lets and disseminated molybdenite flakes. The molybdenite isnot present in B-type quartz veinlets, the commonest site ofmolybdenum concentration in gold-poor porphyry copper de-posits (e.g., Gustafson and Hunt, 1975; Seedorff et al., 2005).

Arsenic distribution at Caspiche is closely related to the ad-vanced argillic overprint because of its concentration in theassociated high sulfidation-state sulfide assemblages, princi-pally as enargite and lesser luzonite and tennantite. These sul-fosalt minerals are commonest in the shallower parts of theadvanced argillic zone, above 3,800-m elevation, where thehighest arsenic contents occur (Fig. 9c), as observed in sev-eral other gold-rich porphyry copper deposits (e.g., Wafi-Golpu, Papua New Guinea; Sillitoe, 1999). Interestingly, theshallow parts of the phase 3 quartz diorite porphyry containonly low arsenic values (Figs. 5, 9c), suggesting that it was em-placed later than most of the arsenic introduction.

Peripheral gold-zinc zone

West of the main Caspiche deposit, the drilling defined anovoid, shallowly W dipping zone of elevated (locally >1 g/t)gold values, which coincides with and is enveloped by a simi-larly shaped, 500-m-long zone of high zinc values, typically inthe 500 to 1,000 ppm range, but with isolated values attaining

>2,000 ppm (Fig. 9d); lead and copper can also attain severalhundred ppm locally. This zone, lying immediately beneathand parallel to the W-dipping contact of the late-mineral diatreme breccia (Fig. 9a, d), coincides with the greatest concentration of the auriferous sphalerite-galena-chalcopy-rite mineralization noted above.

Supergene Effects

Sulfide oxidation

The supergene profile at Caspiche is relatively thin (Fig. 7),in common with those developed over most deposits in theMaricunga belt (Vila and Sillitoe, 1991). The leached cappingis typically thickest over the core of the system, where it ishosted by the uppermost parts of the advanced argillic zone,but tends to shallow progressively northward from 150 to 200m in the southeast to 40 m in the northwest. All pyrite andother sulfide minerals, including any contained copper, havebeen completely removed from the oxidized rock, whereas thegold remains largely in situ within derivative jarosite to consti-tute the oxide gold reserve cited above. The acidic solutionsgenerated by the pyrite oxidation produced an unquantifiedamount of supergene kaolinite, although most of this is impos-sible to distinguish from the preexisting hypogene component.

Most of the copper leached from the oxidized zone appearsto have been lost to the paleoground-water system, but a thin(≤30 m) zone of incipiently developed chalcocite enrichmentwas intersected by a few, centrally located drill holes.

Sulfate front

In common with most porphyry copper deposits, all the de-fined alteration zones at Caspiche formerly contained a fewpercent of hypogene anhydrite, much of it confined to the A-type quartz and late pyrite-bearing veinlets as well as occur-ring as the end-stage, anhydrite-only veinlets. The formerpresence of this anhydrite is shown by characteristic openspaces present in many of the veinlets.

The anhydrite was removed by ingress of cool ground waterafter erosion had exhumed the porphyry gold-copper miner-alization. Gypsum formed as an initial hydration product ofthe anhydrite, but was then completely dissolved. The resul-tant sulfate front, present at depths of 600 to 900 m below thesurface within the deposit but locally much shallower on themargins (Fig. 7), is the level beneath which the rocks retaintheir calcium sulfate content. This is in the form of supergenegypsum and residual hypogene anhydrite in proximity to thefront, but exclusively as anhydrite at depth. The sulfate frontis typically sharp, but uncommon patches of sulfate-cementedrock also remain stranded hundreds of meters above it.

Genetic ConsiderationsThe Caspiche porphyry gold-copper deposit shares many

geologic features with gold-rich porphyry deposits elsewherein the world (e.g., Sillitoe, 2000), with metal contents beingclosely similar to those of the nearby Cerro Casale deposit,where an extensive advanced argillic lithocap is also partiallypreserved (Vila and Sillitoe, 1991; Fig. 2). Interestingly, asnoted above, Cerro Casale is part of the mid-Miocene metal-logenic epoch in the Maricunga belt and, therefore, some 12m.y. younger than Caspiche.

598 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 598

Mo ppm 50 100

Au 0.5 ppm Cu 0.2 %

FIG. 10. Plan view of gold, copper, and molybdenum distributions at se-lected cutoff grades, 3,700-m elevation, Caspiche porphyry gold-copperdeposit. Note the peripheral location of molybdenum with respect to goldand copper. Background geologic outline and drill hole traces taken fromFigure 4.

Caspiche is unique among well-described porphyry de-posits in that intrusion of the composite diorite to quartz dior-ite porphyry stock and gold-copper mineralization are brack-eted by emplacement of large-volume vent breccias,interpreted here as diatremes (Fig. 11), the products ofphreatomagmatic (hydrovolcanic) eruptive activity caused byinteraction between subsurface magma and external water(e.g., Sheridan and Wohletz, 1983; Lorenz, 1986). Althoughlate-stage diatreme formation is a relatively widespread phe-nomenon in gold-rich porphyry copper deposits (Sillitoe,1985), premineral volcanic vents are uncommon, albeit welldocumented (e.g., MacDonald and Arnold, 1994; Braxton etal., 2008).

Premineral development

The up to 750-m-thick volcanic breccia that hosts the upperhalf of the Caspiche porphyry stock and its associated gold-copper mineralization is best interpreted, for the reasons out-lined above, as the southern manifestation of a diatreme vent,the center of which seems likely to be situated north ofCaspiche (Fig. 12a). The near synchronism of the U-Pb zir-con age of the volcanic breccia (24.7 ± 0.7 Ma) and Re-Os ageof Caspiche molybdenite (25.38 ± 0.09 Ma) adds furtherweight to a genetic connection between breccia accumulationand the porphyry system and, by the same token, arguesagainst an external source for this thick but apparently local-ized volcaniclastic accumulation.

This genetic connection is further strengthened by the evi-dence, in the form of veined and altered clasts (Fig. 6e; seeabove), for at least minor porphyry copper formation beforethe volcanic breccia was generated (Figs. 11, 12a). This evi-dence may further imply that initial porphyry copper forma-tion was aborted by the catastrophic brecciation event, only tobe resumed farther south at the site of the Caspiche depositfollowing completion of diatreme emplacement (Figs. 11,12b). This early porphyry copper system appears to have beenaborted relatively early in its development, judging by the absence of any pyrite-bearing clast material in the volcanicbreccia. The close temporal association between theCaspiche deposit and the volcanic breccia that hosts its upperhalf further suggests that the gold-copper mineralization took

place in a relatively shallow setting, possibly beneath only afew hundred meters of extant volcanic cover (Fig. 12b-d).

Perhaps the most enigmatic aspect of the volcanic breccia,in the context of a diatreme model, is its subhorizontal basethroughout much of the Caspiche deposit and its immediatesurroundings, which implies accumulation within a flat-bot-tomed, basin-like depression rather than an upward-flared,cone-shaped vent. However, evidence from several porphyrycopper-related diatremes elsewhere, including the late-mineralexamples at Lepanto-Far Southeast, Philippines (Garcia, 1991),and Río Blanco-Los Bronces, Chile (J.C. Toro, pers. comm.,2010; E. Wettke, pers. comm., 2011), as well as the prem-ineral example at Resolution, Arizona (G. Zulliger and R.H.Sillitoe, unpub. data, 2006; Hehnke et al., 2012), suggests thatthick, basin-like tuff and breccia accumulations above re-stricted and, in some cases, poorly documented vents may bemore common than previously thought. The shallow portionsof diatreme vents tend to approximate broad, bowl-shapedand flat-bottomed structures, rather than being steeply conicalin form, where enclosing wall rocks are relatively incompe-tent (e.g., Auer et al., 2007), which may have been the case ofthe fine-grained sedimentary strata of Oligocene age atCaspiche. Furthermore, such thick but areally restricted tuffand breccia accumulations could also represent the transitionsbetween true diatremes and small calderas (Lipman, 1997).

Porphyry gold-copper formation

The main stages of gold and copper introduction at Caspicheaccompanied the potassic alteration that spanned emplace-ment of phases 1 and 2 of the five readily recognizable por-phyry intrusions (Figs. 11, 12b, c). With only few exceptions,the highest-grade ore occurs within the centrally positioned,phase 1 diorite porphyry and its immediate wall rocks: thevolcanic breccia and, at depth, siliciclastic sedimentary se-quence. Nonetheless, the encircling phase 2 quartz dioriteporphyry, in particular its interior parts, also hosts a substan-tial part of the orebody, albeit generally with lower averagegrades. Minor metal introduction followed intrusion of thephase 3 quartz diorite porphyry, but had essentially ceased bythe time the late-mineral, phase 4 and 5 quartz diorite por-phyries were emplaced (Figs. 11, 12d, e).


0361-0128/98/000/000-00 $6.00 599

LATE EARLY

MINERALIZATION

ADVANCED ARGILLIC

ALTERATION POTASSIC

ALTERATION

PORPHYRIES

BRECCIAS

Au-Cu

Cumulative 1 g/t Au 0.6 g/t Au 0.2 g/t Au

IS Au-Zn

Diorite 1 QDP 2 QDP 3 QDP 4 QDP 5

Early diatreme Hydrothermal breccias

Late-mineral diatreme

Aborted system

FIG. 11. Schematic time line diagram to show the relative age relationships between the main brecciation, porphyry in-trusion, hydrothermal alteration, and metal introduction events at Caspiche. Note that the average gold grades in the twoearliest porphyries (phases 1 and 2) are cumulative rather than being the products of only the host intrusions. Widths of barsapproximate the importance of events. IS = intermediate sulfidation epithermal, QDP = quartz diorite porphyry.

600 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 600

a b

c d

e f

Late-stage gold-zinc zone


Late quartz diorite porphyries (phases 3, 4, & 5)

Intermineral quartz diorite porphyry (phase 2)


Early diatreme breccia

Aborted porphyry system

Siliciclastic sedimentary rocks

Present surface

Advanced argillic

Strong potassic

N S N S

E W E W

E W E W

FIG. 12. Schematized evolution of the Caspiche porphyry gold-copper deposit. a. Formation of the premineral diatreme,showing the possible prediatreme position of weakly developed porphyry-type mineralization. b. Emplacement of the earlyphase 1 diorite porphyry, with associated potassic and advanced argillic alteration, into the southern part of the premineraldiatreme breccia. c. Emplacement of the early intermineral phase 2 quartz diorite porphyry and continued potassic and ad-vanced argillic alteration. d. Emplacement of the three later quartz diorite porphyries, phases 3, 4, and 5, following appre-ciable erosion and initiation of telescoping. e. Continued erosion and telescoping resulting in massive overprinting of the ad-vanced argillic lithocap over the porphyry intrusions and potassic alteration, leading to high sulfidation-state gold-coppermineralization. f. Emplacement of the late-mineral diatreme, localized advanced argillic alteration within it, and intermedi-ate sulfidation gold-zinc mineralization beneath the flared contact. The diatreme breccia covered and helped to preserve thedeposit, with its lower contact inferred to have been subhorizontal. The blue line approximates the present erosion surface,at which the porphyry intrusions and potassic alteration are largely blind. Note that sections a and b are oriented north-south,with b offset to the south of a, whereas c-f are east-west. See text for further discussion.

There is a crude outward decrease, most pronounced to thewest, in the relative age of the Caspiche porphyry intrusionsfrom the phase 1 diorite to the propylitized, phase 5 quartzdiorite, the latter emplaced after potassic alteration had com-pletely ceased. This intrusion sequence is the opposite of thatobserved in many gold-rich porphyry copper deposits, in whichthe porphyry phases tend to become progressively youngerinward (e.g., Panguna, Papua New Guinea, Batu Hijau, In-donesia, and Bajo de la Alumbrera, Argentina; Clark, 1990;Clode et al., 1999; Proffett, 2003).

Notwithstanding the unusually high Au/Cu ratio at Caspichecompared to most gold-rich porphyry copper deposits, theobserved close correlation of the gold and copper contentsand the external but partly overlapped position of the mainmolybdenum concentration are typical features of the depositclass (Figs. 9a, b, 10). This lateral metal zoning is widely ob-served in gold-rich porphyry copper deposits, including Bajode la Alumbrera (Proffett, 2003), Esperanza, Chile (Perelló etal., 2004), and elsewhere (Sillitoe, 2000, 2010), and suggestscontrol by outward temperature decline and/or differentmetal complexing of molybdenum compared to that of cop-per and gold (e.g., Ulrich and Mavrogenes, 2008).

The bottom of the Caspiche ore zone is defined by theabrupt transition from potassic to potassic-calcic alteration,the latter being sulfide and gold deficient, albeit substantiallyricher in magnetite. The potassic-calcic alteration is tenta-tively considered to have followed emplacement of the phase2 quartz diorite porphyry intrusion, and to have caused biotiteand copper-bearing sulfide destruction. The downward ter-mination of the Caspiche ore zone as a result of increasedmagnetite at the expense of sulfide contents differs markedlyfrom that described from the Batu Hijau gold-rich porphyrycopper deposit, Indonesia, where downward grade decreasecorresponds to increased pyrite/chalcopyrite ratios (Setyand-haka et al., 2008). In the context of a magmatic versus exter-nal basinal brine origin for actinolite-rich alteration assem-blages in porphyry copper deposits (e.g., Seedorff et al., 2005;Sillitoe, 2010), the deep potassic-calcic assemblage at Caspichemight be assigned either to the former, in view of its confine-ment to the deep core of the system, or to the latter becauseof the sedimentary host rocks, which potentially could haveacted as a connate brine source; however, in view of the ele-vated potassium content, in the form of K-feldspar, the first ofthese alternatives is preferred.

The gold grade at Caspiche remains approximately constantover ~1,000 vertical meters, without any noticeable changewhere copper contents increase on passing from the chalcopy-rite-only to deeper bornite-bearing zones (Figs. 7, 9a). Thissituation is contrary to that predicted experimentally, wherebyhigher gold contents in the high-temperature precursor tobornite than in intermediate solid solution—the high-temper-ature chalcopyrite precursor—should be reflected by in-creased gold tenors with greater bornite abundance (Kesler etal., 2002). However, the vertical constancy of the gold gradecombined with the clear gold-copper correlation may betaken as further support for relatively shallow emplacementof the Caspiche deposit (Fig. 12), which would have favoredcoprecipitation of gold and copper from a buoyant, sulfur-richvapor plume as a result of depressurization (Murakami et al.,2010). The uncommonly high Au/Cu ratio at Caspiche was

likely dictated by the subjacent magma column, possibly as aresult of preferential copper sequestration by magmatic sul-fides (Simon et al., 2010; Audétat and Simon, 2012).

Late-stage evolution

Advanced argillic lithocap development in the shallowerparts of the Caspiche system is assumed to have commencedonce potassic alteration was initiated at depth (Figs. 11, 12b,c; Sillitoe, 2010). Nonetheless, much of the preserved ad-vanced argillic zone, which overprints and obliterates pre -existing potassic assemblages over the uppermost 300 to 400vertical meters (Fig. 7), formed later in system developmentduring extreme telescoping. This is believed to have been adirect consequence of progressive but rapid erosional degra-dation of the paleosurface (Figs. 11, 12d, e), leading to anequally rapid transition from ductile to brittle conditionswithin and around the porphyry stock. A similar situation ex-ists at the Marte porphyry gold deposit farther north in theMaricunga belt (Sillitoe, 1994; Fig. 2).

This progressive overprinting of potassic by advanced argillicalteration caused total to partial destruction of the intermedi-ate sulfidation-state chalcopyrite ± pyrite assemblages in theupper parts of the potassic zone by high sulfidation-state cop-per ± iron sulfides and copper-arsenic sulfosalts (Fig. 8). Thisreconstitution of the potassic zone and its contained mineral-ization was effective notwithstanding the relatively low temperature, probably not much greater than ~200°C(Hedenquist et al., 1996; Seedorff et al., 2005), implied by thedominant quartz-kaolinite alteration assemblage. The arsenic,absent from the pristine potassic zone, must have been intro-duced by the fluid responsible for the advanced argillic over-print, whereas much of the gold and copper appear to havebeen recycled from the preexisting sulfide assemblages (cf.Brimhall, 1979). The latter conclusion is supported by thegeneral coincidence between the highest-grade gold and cop-per mineralization in the advanced argillic zone and the phase1 diorite porphyry intrusion and its related high-intensityquartz veinlet stockwork. Nonetheless, several prominent ex-ceptions to this generalization appear to require at least localredistribution of gold and copper.

Hydrothermal breccias at Caspiche are typically small andcommonly of dike-like form, in common with those in manyother gold-rich porphyry copper deposits (Sillitoe, 2000). Thebrecciation, whether within (Figs. 5, 6j) or beneath (Fig. 5, 6i)the advanced argillic zone, took place relatively late in systemevolution (Fig. 11), as evidenced by the abundance of miner-alized clasts and paucity of crosscutting veinlets (Fig. 6i, j).The breccias are consistently barren, and ascribed to phreaticbrecciation processes instigated by either late-mineral por-phyry intrusion or, within the lithocap, by localized fluid over-pressuring (Sillitoe, 1985, 2010).

Development of the Caspiche system concluded with em-placement of the late-mineral diatreme along its western side.Although the diatreme breccia contains abundant wall-rockclasts, including previously mineralized material, the brecciain the western, deeper known parts of the vent has a more ob-vious juvenile tuffaceous matrix and, hence, is more clearlyphreatomagmatic (hydrovolcanic) in origin. The magma cham-ber from which the juvenile component was erupted is inferred to be that from which the phase 5 quartz diorite


0361-0128/98/000/000-00 $6.00 601

porphyry was also derived, a proposal supported by thepyrite-poor, propylitic alteration affecting both the brecciaand phase 5 porphyry. Groundwater ingress to this late-min-eral porphyry stock could have been the trigger for thephreatomagmatic activity (e.g., Sheridan and Wohletz, 1983).The upward- and eastward-decreasing angle of the diatremecontact suggests that it may have approached horizontal overthe Caspiche deposit, in common with the situation describedabove for the premineral diatreme (Fig. 12a, f). The subsidedblock of subaerial surge deposits recognized in the diatremebreccia implies that the vent may formerly have been sur-mounted by a tuff ring or tuff cone (cf. Lorenz, 1986).

Late-mineral diatreme formation overlapped in time withthe waning stages of advanced argillic lithocap development(Fig. 11), which resulted in fault-controlled, quartz-kaolinitealteration and localized siliceous ledge formation within thebreccia. The final hypogene event at Caspiche was introductionof the gold-zinc-(lead-copper) mineralization, the main con-centration of which was localized by the heavily fractured vol-canic breccia immediately beneath the eastward-flared, late-mineral diatreme contact. The relatively impermeable natureof the diatreme breccia may have acted as an aquitard (cf.Davies et al., 2008), thereby impeding upward fluid flow andfacilitating gold, zinc, and subordinate lead and copper precip-itation along the diatreme margin. The zinc, lead, and carbon-ate contents of the mineralization and low-iron compositionof the sphalerite combine to suggest that this end-stage min-eralization is of intermediate sulfidation epithermal affinity(Einaudi et al., 2003). Similar late-mineral diatreme contactsin gold-rich porphyry copper systems elsewhere also localizedgold-bearing mineralization of high and/or intermediate sulfi-dation epithermal type (e.g., Lepanto, Philippines, and Wafi-Golpu, Papua New Guinea; Garcia, 1991; Sillitoe, 1999).

ConclusionsCaspiche is the latest discovery in the Maricunga gold-silver-

copper belt of northern Chile, with its delayed recognitionbeing due to the concealment by postmineral cover. The de-posit provides an excellent example of the full vertical alter-ation-mineralization zoning sequence in a telescoped, gold-rich porphyry copper deposit, from the high sulfidation,advanced argillic top through the potassic central parts to themetal-deficient, potassic-calcic roots. The ore zone, largelyblind at the bedrock surface as well as concealed beneathpostmineral cover, extends over approximately 1,000 verticalmeters. Early- and late-stage diatremes generated little, ifany, mineralization, although part of the former did act as thehost rock for the upper half of the Caspiche deposit, and thelatter helped localize the end-stage, intermediate sulfidationepithermal gold-zinc mineralization.

A notable feature of Caspiche is the diversity of pre-, syn-,and postmineral fragmental rocks, which need to be correctlydistinguished in drill core in order to properly map, under-stand, and interpret system evolution. The earliest of theseare the thick, flat-lying sedimentary breccia horizons in thedeeper drilled parts of the siliciclastic wall-rock succession(Figs. 5, 6b), which were followed by the thick, monotonousvolcanic breccia package ascribed to early diatreme formation(Figs. 5, 6c-e). Breccias generated during deposit formationcomprise the intrusion breccia, developed locally in marginal

parts of the phase 2 quartz diorite porphyry (Fig. 6h), andseveral texturally and mineralogically distinct, late-stagehydrothermal breccias (Fig. 6i, j). Brecciation processes con-cluded with emplacement of the late-stage diatreme brecciaalong the western side of the deposit (Figs. 4, 5), within whicha variety of mineralized clasts are commonplace near its east-ern contact (Fig. 6k). To these may be added a number ofsteep zones of tectonic brecciation caused by relatively minorpostmineral faulting (Figs. 4, 5).

AcknowledgmentsThe unstinting support of Exeter Resource’s senior man-

agement—the instigators of this paper—throughout theCaspiche exploration program is greatly appreciated, as is thegeologic input provided at various times by Celena Ibáñez,Leandro Sastre, Oscar Hernández, Marcia Reyes, MauricioArce, Cristian Salgado, Felipe Jiménez, Matt Houston, andthe late Martín Rodríguez. Petrographic descriptions wereprovided by E. Tidy y Cía. Ltda. and Paul Ashley. Discussionswith Juan Carlos Toro and Enrique Viteri clarified historicaldetails. Influential reviews were provided by ConstantinoMpodozis and Pepe Perelló and, on behalf of Economic Geology, Pete Hollings, Jeremy Richards, and Brian Townley.

REFERENCESArriagada, C., Roperch, P., Mpodozis, C., and Fernández, R., 2006, Paleo-

magnetism and tectonics of the southern Atacama Desert (25–28°S), north-ern Chile: Tectonics, v. 25, 26 p.

Audétat, A., and Simon, A.C., 2012, Magmatic controls on porphyry Cu gen-esis: Society of Economic Geologists Special Publication 16, p. 553–572.

Auer, A., Martin, U., and Németh, K., 2007, The Fekete-hegy (Balaton High-lands Hungary) “soft-substrate” and “hard-substrate” maar volcanoes in analigned volcanic complex—implications for vent geometry, subsurfacestratigraphy and the palaeoenvironmental setting: Journal of Volcanologyand Geothermal Research, v. 159, p. 225–245.

Braxton, D.P., Cooke, D.R., Rye, R.O., and Petersen, E.U., 2008, Geologic andstable isotope evidence for submarine diatreme formation, and emplacementof the Boyongan and Bayugo porphyry copper-gold deposits, Philippines:Pacrim Congress 2008, Gold Coast, Queensland, 2008, Extended Abstracts:Melbourne, Australasian Institute of Mining and Metallurgy, p. 259–265.

Brimhall, Jr., G.H., 1979, Lithologic determination of mass transfer mecha-nisms of multiple-stage porphyry copper mineralization at Butte, Montana:Vein formation by hypogene leaching and enrichment of potassium-silicateprotore: ECONOMIC GEOLOGY, v. 74, p. 556–589.

Bull, K.F., and McPhie, J., 2007, Fiamme textures in volcanic successions:Flaming issues of definition and interpretation: Journal of Volcanology andGeothermal Research, v. 164, p. 205–216.

Cahill, T., and Isacks, B.L., 1992, Seismicity and shape of the subductedNazca plate: Journal of Geophysical Research, v. 97, p. 17,503–17,529.

Chang, Z., Vervoort, J.D., McClelland, W.C., and Knaack, C., 2006, U-Pbdating of zircon by LA-ICP-MS: Geochemistry, Geophysics, Geosystems, v.7, 14 p.

Chouinard, A., Williams-Jones, A.E., Leonardson, R.W., Hodgson, C.J.,Silva, P., Téllez, C., Vega, J., and Rojas, F., 2005, Geology and genesis of themultistage high-sulfidation epithermal Pascua Au-Ag-Cu deposit, Chileand Argentina: ECONOMIC GEOLOGY, v. 100, p. 463–490.

Clark, G.H., 1990, Panguna copper-gold deposit: Australasian Institute ofMining and Metallurgy Monograph 14, p. 1807–1816.

Clode, C., Proffett, J., Mitchell, P., and Munajat, I., 1999, Relationships of in-trusion, wall-rock alteration and mineralisation in the Batu Hijau copper-gold porphyry deposit: Pacrim ‘99 Congress, Bali, Indonesia, 1999, Pro-ceedings: Melbourne, Australasian Institute of Mining and Metallurgy, p.485–498.

Conrad, C.P., and Lithgow-Bertelloni, C., 2007, Faster seafloor spreading andlithosphere production during the mid-Cenozoic: Geology, v. 35, p. 29–32.

Cornejo, P., Mpodozis, C., and Tomlinson, A.J., 1998, Hoja Salar de Mari-cunga, Región de Atacama: Servicio Nacional de Geología y Minería[Chile], Mapas Geológicos 7, scale 1:100,000, 1 sheet.

602 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 602

Davies, A.G.S., Cooke, D.R., Gemmell, J.B., and Simpson, K.A., 2008, Dia-treme breccias at the Kelian gold mine, Kalimantan, Indonesia: Precursorsto epithermal gold mineralization: ECONOMIC GEOLOGY, v. 103, p. 689–716.

Dean, W.E., Davies, G.R., and Anderson, R.Y., 1975, Sedimentological sig-nificance of nodular and laminated anhydrite: Geology, v. 3, p. 367–372.

Einaudi, M.T., Hedenquist, J.W., and Inan, E.E., 2003, Sulfidation stateof fluids in active and extinct hydrothermal systems: Transitions from porphyry to epithermal environments: Society of Economic GeologistsSpecial Publication 10, p. 285–313.

Garcia, Jr., J.S., 1991, Geology and mineralization characteristics of theMankayan mineral district, Benguet, Philippines: Geological Survey ofJapan Report 277, p. 21–30.

Gustafson, L.B., and Hunt, J.P., 1975, The porphyry copper deposit at El Sal-vador, Chile: ECONOMIC GEOLOGY, v. 70, p. 857–912.

Gustafson, L.B., Vidal, C.E., Pinto, R., and Noble, D.C., 2004, Porphyry-epi-thermal transition, Cajamarca region, northern Peru: Society of EconomicGeologists Special Publication 11, p. 279–299.

Hedenquist, J.W., Izawa, E., Arribas, Jr., A., and White, N.C., 1996, Epither-mal gold deposits: Styles, characteristics, and exploration: Resource Geol-ogy Special Publication 1, 17 p.

Hehnke, C., Ballantyne, G., Martin, H., Hart, W., Schwarz, A., and Stein, H.,2012, Geology and exploration progress at the Resolution porphyry Cu-Modeposit, Arizona: Society of Economic Geologists Special Publication 16, p.147–166.

Iriarte, S., Arévalo, C., and Mpodozis, C., 1999, Hoja La Guardia, Región deAtacama: Servicio Nacional de Geología y Minería [Chile], Mapas Geoló-gicos 13, scale 1:100.000, 1 sheet.

Jannas, R.R., Bowers, T.S., Petersen, U., and Beane, R.E., 1999, High-sulfi-dation deposit types in the El Indio district, Chile: Society of EconomicGeologists Special Publication 7, p. 219–266.

Kay, S.M., and Mpodozis, C., 2001, Central Andean ore deposits linked toevolving shallow subduction systems and thickening crust: GSA Today, v.11, p. 4–9.

——2002, Magmatism as a probe to the Neogene shallowing of the Nazcaplate beneath the modern Chilean flat-slab: Journal of South AmericanEarth Sciences, v. 15, p. 39–57.

Kay, S.M., Mpodozis, C., Tittler, A., and Cornejo, P., 1994, Tertiary magmaticevolution of the Maricunga mineral belt in Chile: International GeologyReview, v. 36, p. 1079–1112.

Kay, S.M., Coira, B., and Mpodozis, C., 2008, Field trip guide: Neogene evo-lution of the central Andean Puna plateau and southern Central VolcanicZone: Geological Society of America Field Guide 13, p. 119–183.

Kesler, S.E., Chryssoulis, S.L., and Simon, G., 2002, Gold in porphyry copperdeposits: Its distribution and fate: Ore Geology Reviews, v. 21, p. 103–124.

Khashgerel, B.-E., Kavalieris, I., and Hayashi, K., 2008, Mineralogy, textures,and whole-rock geochemistry of advanced argillic alteration: Hugo Dum-mett porphyry Cu-Au deposit, Oyu Tolgoi mineral district, Mongolia: Min-eralium Deposita, v. 43, p. 913–932.

Lipman, P.W., 1997, Subsidence of ash-flow calderas: Relation to caldera sizeand magma chamber geometry: Bulletin of Volcanology, v. 59, p. 198–218.

Lonsdale, P., 2005, Creation of the Cocos and Nazca plates by fission of theFarallon plate: Tectonophysics, v. 404, p. 237–264.

Lorenz, V., 1986, On the growth of maars and diatremes and its relevance tothe formation of tuff rings: Bulletin of Volcanology, v. 48, p. 265–274.

MacDonald, G.D., and Arnold, L.C., 1994, Geological and geochemical zon-ing of the Grasberg Igneous Complex, Irian Jaya, Indonesia: Journal ofGeochemical Exploration, v. 50, p. 143–178.

Maksaev, V., Moscoso, R., Mpodozis, C., and Nasi, C., 1984, Las unidadesvolcánicas y plutónicas del Cenozoico Superior en la Alta Cordillera delNorte Chico (29°–31°S): Geología, alteración hidrotermal y mineralización:Revista Geológica de Chile, no. 21, p. 11–51.

McKee, E.H., Robinson, A.C., Rytuba, J.J., Cuitiño, L., and Moscoso, R.D.,1994, Age and Sr isotopic composition of volcanic rocks in the Maricungabelt, Chile: Implications for magma sources: Journal of South AmericanEarth Sciences, v. 7, p. 167–177.

Mercado, M., 1982, Hoja de Laguna Negro Francisco: Servicio Nacional deGeología y Minería, Carta Geológica de Chile 56, 73 p.

Mpodozis, C., and Clavero, J., 2002, Tertiary tectonic evolution of the south-western edge of the Puna Plateau: Cordillera Claudio Gay (26°–27°S),northern Chile: Institut de Recherche pour le Développement (IRD)(Paris, France) and the Paul Sabatier University of Toulouse, InternationalSymposium on Andean Geodynamics, 5th, Toulouse, 2002, Proceedings, p.445–448.

Mpodozis, C., and Kay, S.M., 2003, Neogene tectonics, ages and mineraliza-tion along the transition zone between the El Indio and Maricunga mineralbelts (Argentina and Chile 28-29°) [abs.]: Sociedad Geológica de Chile,Congreso Geológico Chileno, 10th, Concepción, 2003, CD-ROM, 1 p.

Mpodozis, C., and Ramos, V., 1990, The Andes of Chile and Argentina: Cir-cum-Pacific Council for Energy and Mineral Resources Earth Science Se-ries, v. 11, p. 59–90.

Mpodozis, C., Allmendinger, R., and Jordan, T., 1991, La zona del Nevadode Jotabeche y Laguna del Negro Francisco: Evolución tectónica y volcá-nica de la extremidad meridional del Altiplano chileno: Sociedad Geoló-gica de Chile, Congreso Geológico Chileno, 6th, Santiago, Actas, v. 1, p.91–95.

Mpodozis, C., Kay, S.M., Cornejo, P., and Tittler, A., 1995, La franja de Ma-ricunga: síntesis de la evolución del frente volcánico Oligoceno-Mioceno dela zona sur de los Andes centrales: Revista Geológica de Chile, v. 22, p.273–313.

Muntean, J.L., and Einaudi, M.T., 2000, Porphyry gold deposits of the Refu-gio district, Maricunga belt, northern Chile: ECONOMIC GEOLOGY, v. 95, p.1445–1472.

Murakami, H., Seo, J.H., and Heinrich, C.A., 2010, The relation betweenCu/Au ratio and formation depth of porphyry-style Cu-Au ± Mo deposits:Mineralium Deposita, v. 45, p. 11–21.

Pardo-Casas, F., and Molnar, P., 1987, Relative motion of the Nazca (Faral-lon) and South American plates since Late Cretaceous time: Tectonics, v. 6,p. 233–248.

Perelló, J., Mpodozis, C., and Urzúa, F., 2003, Late Oligocene-early Mioceneporphyry Cu-Mo and Cu-Au mineralization, Pulido cluster, Atacama Re-gion, northern Chile [abs.]: Sociedad Geológica de Chile, Congreso Ge-ológico Chileno, 10th, Concepción, 2003, CD-ROM, 1 p.

Perelló, J., Brockway, H., and Martini, R., 2004, Discovery and geology of theEsperanza porphyry copper-gold deposit, Antofagasta Region, northernChile: Society of Economic Geologists Special Publication 11, p. 167–186.

Proffett, J.M., 2003, Geology of the Bajo de la Alumbrera porphyry copper-gold deposit, Argentina: ECONOMIC GEOLOGY, v. 98, p. 1535–1574.

Seedorff, E., Dilles, J.H., Proffett, Jr., J.M., Einaudi, M.T., Zurcher, L.,Stavast, W.J.A., Johnson, D.A., and Barton, M.D., 2005, Porphyry deposits:Characteristics and origin of hypogene features: ECONOMIC GEOLOGY 100TH

ANNIVERSARY VOLUME, p. 251–298.Setyandhaka, D., Arif, J., and Proffett, J., 2008, Characteristics of the roots of

a classic copper-gold porphyry system—the Batu Hijau copper-gold por-phyry deposit, Indonesia: Pacrim Congress 2008, Gold Coast, Queensland,2008, Extended Abstracts: Melbourne, Australasian Institute of Mining andMetallurgy, p. 445–450.

Sheridan, M.F., and Wohletz, K.H., 1983, Hydrovolcanism: Basic considera-tions and review: Journal of Volcanology and Geothermal Research, v. 17,p. 1–29.

Sillitoe, R.H., 1985, Ore-related breccias in volcanoplutonic arcs: ECONOMICGEOLOGY, v. 80, p. 1467–1514.

——1994, Erosion and collapse of volcanoes: Causes of telescoping in intru-sion-centered ore deposits: Geology, v. 22, p. 945–948.

——1999, Styles of high-sulphidation gold, silver and copper mineralisationin porphyry and epithermal environments: Pacrim ‘99 Congress, Bali, In-donesia, 1999, Proceedings: Melbourne, Australasian Institute of Miningand Metallurgy, p. 29–44.

——2000, Gold-rich porphyry deposits: Descriptive and genetic models andtheir role in exploration and discovery: Reviews in Economic Geology, v.13, p. 315–345.

——2010, Porphyry copper systems: ECONOMIC GEOLOGY, v. 105, p. 3–41.Sillitoe, R.H., and Perelló, J., 2005, Andean copper province: Tectonomagmatic

settings, deposit types, metallogeny, exploration, and discovery: ECONOMICGEOLOGY 100TH ANNIVERSARY VOLUME, p. 845–890.

Sillitoe, R.H., McKee, E.H., and Vila, T., 1991, Reconnaissance K-Ar geo -chronology of the Maricunga gold-silver belt, northern Chile: ECONOMICGEOLOGY, v. 86, p. 1261–1270.

Simon, A., Muntean, J., and Aillaud, B., 2010, The formation of Au-rich por-phyry deposits: Where is the Cu? [abs.]: Geological Society of America Ab-stracts with Programs, v. 42, no. 5, p. 46.

Thorpe, R.S., Francis, P.W., Hammill, M., and Baker, M.C.W., 1982, TheAndes, in Thorpe, R.S., ed., Andesites: Orogenic andesites and relatedrocks: Chichester, John Wiley & Sons, p. 187–205.

Ulrich, T., and Mavrogenes, J., 2008, An experimental study of the solubilityof molybdenum in H2O and KCl-H2O solutions from 500°C to 800°C, and150 to 300 MPa: Geochimica et Cosmochimica Acta, v. 72, p. 2316–2330.


0361-0128/98/000/000-00 $6.00 603

Van Kerkvoort, G., Tolman, J., and Hernandez, O., 2009, Caspiche porphyrygold-copper deposit, Maricunga belt, Chile. Exploration, discovery and re-source development: NewGenGold 2009 Conference, Perth, 2009, Pro-ceedings: Perth, Paydirt Media Pty Ltd., p. 169–183.

Vila, T., and Sillitoe, R.H., 1991, Gold-rich porphyry systems in the Mari-cunga belt, northern Chile: ECONOMIC GEOLOGY, v. 86, p. 1238–1260.

Vila, T., Sillitoe, R.H., Betzhold, J., and Viteri, E., 1991, The porphyry golddeposit at Marte, northern Chile: ECONOMIC GEOLOGY, v. 86, p.1271–1286.

Zimmerman, A., Stein, H.J., Hannah, J.L., Koželj, D., Bogdanov, K., andBerza, T., 2008, Tectonic configuration of the Apuseni-Banat-Timok-Sred-nagorie belt, Balkans-South Carpathians, constrained by high precision Re-Os molybdenite ages: Mineralium Deposita, v. 43, p. 1–21.

604 SILLITOE ET AL.

0361-0128/98/000/000-00 $6.00 604

Exeter caspiche paper_in_economic_geology_2013

Education

Transcript of Exeter caspiche paper_in_economic_geology_2013