-
0361-0128/01/3247/573-20 $6.00 573
IntroductionSOME OF the worlds richest and largest known
Cenozoic cop-per, gold, and silver deposits occur in the Andean
orogen as-sociated with continental arc magmatism. Some of these
de-posits have been studied extensively, but very little
ispublished about others, and although regional-scale varia-tions
in metal sources have been documented, our under-standing of metal
source processes remains incomplete. Onecommon method of
determining the source of metals in anore deposit is to use lead
isotope ratios to evaluate the sourcesof lead in associated igneous
rocks, assuming that the igneousrocks represent the predominant
source of metals in the ores.This approach is facilitated by
extensive literature on thepetrology of continental arc magmas in
general and centralAndean magmas in particular (e.g., James, 1982;
Harmon etal., 1984; Hildreth and Moorbath, 1988; Aitcheson and
For-
rest, 1994; Aitcheson et. al, 1995; Kay et al., 1999).
Whilethese comparisons can provide clues to the source of ore
met-als in many deposits, they are not entirely satisfactory on
theirown. Most Andean metal deposits are formed by hydrother-mal
activity that postdated any significant isotopic evolutionof the
magma, and the processes that create the orebodies aredominantly
hydrothermal. Even when ore metals are derivedfrom associated
magmatic rocks, other elements in the ores,including strontium,
oxygen, and sulfur, may come from othersources (e.g., Macfarlane et
al., 1994). Other isotopic tracerssuch as Nd, Sr, and Os are only
of indirect use in determiningpotential ore metal sources.
Macfarlane et al. (1990) and Petersen et al. (1993) summa-rized
new and published lead isotope data for the ore depositsin the
central Andes and defined three large-scale lead iso-tope provinces
(Fig. 1, inset). Province I comprises thecoastal volcanic arc of
Per and Chile, province II includesthe Jurassic and Cretaceous
miogeosynclinal sedimentary belt
Sources of Lead in the San Cristobal, Pulacayo, and Potos Mining
Districts, Bolivia, and a Reevaluation of Regional Ore Lead Isotope
Provinces
GEORGE KAMENOV,Department of Geological Sciences, 24 Williamson
Hall, University of Florida, Gainesville, Florida 32611
ANDREW W. MACFARLANE,
Department of Earth Sciences, Florida International University,
Miami, Florida 33199
AND LEE RICIPUTIInorganic Geochemistry Division, Oak Ridge
National Laboratory, Oak Ridge, Tennessee 37831
AbstractNew lead isotope data on ores, crustal rocks, and
leachates of crustal rocks, combined with data in the liter-
ature, provide important new constraints on the sources of ore
metals in southwest to south-central Bolivia, in-cluding the very
large recently discovered silver-zinc deposit at San Cristobal, the
Pulacayo polymetallic dis-trict, and the giant Potos
silver-tin-base metal deposit.
Lead isotope ratios of ores and igneous rocks from the San
Cristobal deposit and from Paleozoic and Creta-ceous sedimentary
rocks are compared with published data on high-grade Middle
Proterozoic metamorphicbasement rocks. These data constrain the
major source of lead, and by inference of other ore metals, at
SanCristobal to be the metamorphic basement rocks. Leaching
experiments on samples of Paleozoic and Creta-ceous sedimentary
rocks show that the easily leachable lead from these rocks is much
less radiogenic than thewhole-rock compositions. However, lead
isotope ratios of both whole rocks and leachates of these upper
crustalrocks are too radiogenic for them to be major sources of ore
lead at San Cristobal.
Lead isotope ratios of ores from Pulacayo and Potos are similar
to each other and lie within the range of Pa-leozoic and Cretaceous
sedimentary whole-rock compositions. Leaching of Pb from the
sedimentary rocks can-not explain the isotopic compositions of the
Pulacayo and Potos ores, and the isotopic homogeneity of the Po-tos
ores also argues against mixing of lead from diverse sources in the
hydrothermal system. Lead from thesedimentary rocks may have been
incorporated by magmatic assimilation followed by extraction of ore
metalsfrom the resulting magma.
Lead isotope ratios of San Cristobal ores are different from
those of Pulacayo, Potos, and other deposits tothe east, but
resemble the compositions of ores and volcanic rocks in western
Bolivia. On this basis we identifya new ore lead isotope province
extending from San Cristobal northward across the eastern Altiplano
and intosouthern Per. This province is coincident with but smaller
than the extent of the proposed Arequipa-Anto-falla metamorphic
basement craton. The degree of incorporation of ore metals from the
metamorphic base-ment appears to depend on the timing and/or
location of the mineralizing event. Ore deposits in the
northernpart of province IV formed before the thickening of Andean
crust, beginning around 20 Ma, and incorporatedminor amounts of
metals from the metamorphic basement. Younger deposits farther to
the south contain majorto dominant components of basement lead.
Economic GeologyVol. 97, 2002, pp. 573592
Corresponding author: e-mail, [email protected]
-
that dominates the geology of the high Andes of Per, andprovince
III comprises the Eastern Cordillera of southeasternPer, central
Bolivia, and northwestern Argentina, includingthe Bolivian tin
belt. While the existence of these provincesand their general
characteristics are well established, impor-tant details of their
distribution and the roles of shallowcrustal rocks in their
formation remain unclear. Studies basedmainly on lead isotope
compositions of late Tertiary volcanicrocks, or combining volcanic
rock data with ore deposit data,have documented extensive
incorporation of basement leadinto late Tertiary magmas (Wrner et
al., 1992; Aitcheson etal., 1995). A major component of relatively
nonradiogenicmetamorphic basement lead was also documented in
oresand volcanic rocks from the Berenguela district in
westernBolivia (Tosdal et al., 1993; Tosdal, 1996). These
effortsdemonstrate the need for a revision of the lead isotope
province scheme of Petersen et al. (1993) and an evaluationof
the potential role of upper crustal Paleozoic and
Creta-ceous-Tertiary sedimentary rocks in creating the isotopic
sig-natures of province II and III ores.
The purpose of this study is to provide a clearer view of
thederivation of ore metals in major deposits of the centralAndes,
and of the relationship between the shallow crustalhost-rock
sequences and the ore deposits. We are especiallyinterested in
whether the ore metals represent new additionsto the shallow crust
from an enriched mantle source region atdepth, or, alternatively,
whether they represent remobiliza-tions of metals already enriched
in the upper crustal rocks.We examine the lead isotope
characteristics of crustal rocks insouth-central Bolivia and
compare them with the ores at Po-tos, Pulacayo, and San Cristobal
to determine the immediatesources of their ore metals, and we
consider briefly the
574 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 574
ChileArgentina
Bolivia
Pacific Ocean
Province I
ProvinceIII
19S
20S
21S
22S66W67W68W
SALAR DE UYUNI
UYUNI
TODOSSANTOS
TATASI
0 km 50km
100km
Quaternary alluvium
studyarea
PULACAYO
POTOSI
SAN CRISTOBAL
Tertiary volcanic rocks, chieflyandesitic to daciticTertiary
ignimbrites, dacitic
Tertiary nonmarine conglomerates,sandstones, shales and
mudstonesCretaceous marine and nonmarinesandstones, shales, marls
and limestonesPaleozoic sedimentary rocks, dominantlymarine shale
and sandstoneScale
Peru
Sample location
Province II
98M01398M014
98M01798M018
FIG. 1. Geologic map of the study area, simplified from Perez
(1996). Inset shows the location of the study area and
theapproximate boundaries of the lead isotope provinces proposed by
Petersen et al. (1993).
-
sources of magmatic lead in the San Cristobal district. Wealso
redefine the Andean lead isotope province map and ob-tain new
insights into the source processes that create thoseprovinces.
Metal Provenance for Hydrothermal Ores
Hydrothermal leaching of ore metals
Within a hydrothermal system, the potential sources ofmetals can
be evaluated by comparing the lead isotope ratiosof ores in the
district with associated magmatic rocks, andwith host rocks in the
system that may vary widely in age,lithology, and isotopic
characteristics. If the host rocks andmagmatic rocks have different
Pb isotope ratios, such com-parisons can suggest or rule out major
source rocks and revealmixing among metal sources. In other
geologic settings, hy-drothermal leaching of metals from
sedimentary and meta-morphic rocks is well documented, including
the UnitedStates Mississippi Valley-type ores (e.g., Crocetti et
al., 1988),the major Irish base metal deposits (Dixon et al.,
1991), andthe Wairakei magmatic-hydrothermal geothermal
system(Hedenquist and Gulson, 1992). However, few studies havebeen
made of Andean magmatic-hydrothermal systems. Suchstudies would
require a knowledge both of the bulk lead iso-tope composition of
the host rocks and of the lead that couldbe hydrothermally leached
from them. A very small numberof lead isotope ratios of Andean
crustal rocks have been pub-lished, and only a handful of
compositions of leachates.
Sources of magmatic metals
Sorting out metal sources in continental arc magmas ismuch more
complicated than determining where metalscame from in a given
hydrothermal system, because many dif-ferent processes and sources
are involved at different times.Potential metal sources for
continental arc magmas includethe subcontinental mantle, the
subducted oceanic plate, thecontinental crust through which the
magmas rise, and sub-ducted sediments of various kinds that may be
incorporatedinto the arc magma at its source. Subducted sediments
con-tain hydrothermal components from the ocean crust, pelagicand
continentally derived clastic sediments, and chemical
andbiochemical components from the seawater (Lin,
1992;Peucker-Ehrenbrink et al., 1994). Budgets of various metals,as
well as petrologic tracers like Sr, Nd, and Os, will varywidely
among these potential source components, so thatmodeling of metal
contributions from them is only approxi-mate at best.
It is well understood that Andean magmas generally, andAltiplano
igneous rocks specifically, are dominated by crustallead (Hildreth
and Moorbath, 1988; Stern, 1991; Aitchesonand Forrest, 1994;
Aitcheson et al., 1995; Davidson and deSilva, 1995). This lead is
derived from subducted marine sed-iments, from abraded and
subducted continental crust, andfrom the continental crust through
which the magma laterrises, or from a combination of those sources.
Because deepmarine sediments typically contain about 25 ppm Pb and
themantle probably less than 0.1 ppm, the addition of a smallamount
of sediment will rapidly swamp the Pb isotope ratiosof magmas
generated in a subduction-related mantle wedge
(Othman et al., 1989; Aitcheson and Forrest, 1994). Since
iso-topic differences between the mantle and sediments aremuch
greater for 207Pb/204Pb than for 206Pb/204Pb, the primaryeffect of
mixing will be to increase 207Pb/204Pb, which is thecase for many
island arc magmas (e.g., Kay et al., 1978; Milleret al., 1994).
Unlike subducted sediments, which are isotopically ho-mogenized
to some degree by the sedimentary processes thatcreate them and
then probably further mixed within the man-tle wedge, the
continental crust through which magmas rise isoften strongly
heterogeneous. Some Andean upper crustalrocks that experienced
high-grade metamorphism during thePrecambrian (e.g., the
Arequipa-Antofalla craton) containless radiogenic lead owing to
depletion of U and sometimesTh during metamorphism (Tilton and
Barreiro, 1980; Tosdal,1996). Other Precambrian terranes such as
the Guyana andAmazon cratons have been enriched in U and contain
highlyradiogenic lead (Montgomery and Hurley, 1978;
Macfarlane,1999) In either case, lead isotope ratios of Proterozoic
andArchean upper crustal rocks exposed at the surface, and
sed-imentary rocks derived from them, are often distinct fromthose
of lead in subduction-related igneous rocks, either pre-dicted by
models (Zartman and Doe, 1981) or observed(Mukasa and Tilton, 1985;
Macfarlane et al., 1990). This het-erogeneity can sometimes be used
to distinguish betweencontributions of lead from the upper crust
and from sub-ducted sediments (e.g., Mukasa and Tilton, 1985;
Macfarlaneand Petersen, 1990).
Geology of the study area
Our study area extends from the Potos deposit southwestacross
the Pulacayo and San Cristobal deposits (Fig. 1). Thestudy area
lies in a very poorly constrained area of the isotopicprovince map
of Petersen et al. (1993), and crosses their ten-tative province
II-III boundary (inset, Fig. 1). The southernBolivian Andes consist
of three morphotectonic provinces;from west to east, these are the
Cordillera Occidental, the Al-tiplano plateau and the Cordillera
Oriental. The CordilleraOccidental in the study area consists of
late Miocene to Re-cent volcanic rocks primarily of andesitic to
dacitic composi-tion, overlying Jurassic and Cretaceous sedimentary
and vol-canic rocks (Cunningham et al., 1991; Richter et al.,
1991).The Cordillera Oriental is a fold-and-thrust belt
consistingmainly of Paleozoic deep marine and platform
sedimentaryrocks and Cretaceous marine and nonmarine
sedimentaryrocks. These rocks were deposited mostly on
Precambrianbasement and were subsequently deformed and folded
dur-ing at least three orogenic cycles: Caledonian
(Ordovician),Hercynian (Devonian to Triassic), and Andean
(Cretaceous toCenozoic) (Sempere et al., 1990; Sempere, 1995). The
sourceof the Paleozoic sediments was the Precambrian
Brazilianshield to the northeast, which is composed of
granulitic,gneissic, and metasedimentary rocks, and the 2 Ga-old
Pro-terozoic Arequipa massif of PerChile to the west andsouthwest
(Litherland et al., 1989). The Altiplano plateau wasproduced mainly
by crustal shortening and magmatic addi-tion (Isacks, 1988; Lamb
and Hoke, 1997) and represents aseries of intermontane basins
filled with thick deposits of con-tinental sediments of Cretaceous
and Tertiary age (Kennan et
Pb SOURCES IN SAN CRISTOBAL, PULACAYO, AND POTOS MINING
DISTRICTS, BOLIVIA 575
0361-0128/98/000/000-00 $6.00 575
-
al., 1995; Sempere et al., 1997). These continental sedimen-tary
rocks unconformably overlie the folded Paleozoic sedi-mentary rocks
(Gagnier et al., 1996) and locally Precambrianmetamorphic rocks
(Lehmann, 1978). Many major and minorvolcanic centers, mainly with
andesitic to dacitic composi-tions, are found in the Altiplano
(Davidson and de Silva,1992). The San Cristobal and Pulacayo mining
districts areclosely associated with such volcanic centers (Fig.
1).
Ore geology of the Potos, San Cristobal, and Pulacayo
deposits
The famous Cerro Rico de Potos has been mined sinceprecolonial
times and is considered to be the worlds largestknown silver
deposit. Ores occur in veins and disseminationsrelated to north-
and northeast-trending fracture systemswhich cut a dacitic dome
complex and the surrounding hostrocks (Cunningham et al., 1991).
The dome was emplacedthrough the thick, deformed Ordovician black
shales of theCordillera Oriental. Early volcanic activity produced
surge
and breccia deposits surrounding the vent, and the dome
wasextruded onto those earlier deposits, plugging the vent. Thedome
was emplaced at 13.8 0.2 Ma (U-Pb dating of zircon;Zartman and
Cunningham, 1995). Mineralization in theupper part of the deposit
was dominantly silver rich, withbase-metal ores (cassiterite,
sphalerite, and galena) predomi-nating at lower levels.
Silver was discovered in the San Cristobal district in the1630s,
and intermittent, small-scale mining operations havecontinued since
that time (Fig. 2; Jacobson et al, 1969). Be-ginning in 1995,
exploration by Apex Silver Mines Ltd. delin-eated one of the worlds
largest open pit reserves of silver,comprising 240 million tonnes
(Mt) of ore grading 2.0 oz/t sil-ver, 1.67 percent zinc and 0.58
percent lead (Apex SilverMines Ltd. 1999 Annual report). Two large
open pits arebeing developed, centered roughly on the Tesorera
andHedionda mine areas (Fig. 2). The deposits in this area
areclustered near the center of a deeply eroded Tertiary an-desitic
to dacitic intrusive-extrusive complex (Richter et al.,
576 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 576
N
Animas mine
Toldos mine
San Cristobal
Hedionda Mine
Tesorera Mine
Scale (m)0 500 1000
Potoco Fm. (Eocene-Oligocene)
Quehua Fm. (Oligocene-Miocene)
Intrusive dacite porphyry(late Miocene)
Intrusive dacite porphyry breccia
Extrusive dacite porphyry
Colonmine
Bertha Mine
Trapiche Mine
Intrusive andesite
2106'
6712
'
Mine sites
Sample locations
99SC0999SC04
99SC01
91BSL016
91BSL015
91BSL022
99SC0399SC08
FIG. 2. Schematic surface geology of the San Cristobal area,
modified after Jacobson et al. (1969). Sample locations
areapproximate; latitude and longitude of district estimated from
locations in Long (1991).
-
1991). Two major types of subvolcanic intrusive rocks are
rec-ognized: andesite porphyry consisting of plagioclase and
alteredhornblende phenocrysts in a microcrystalline groundmass;and
dacite porphyry consisting of plagioclase, hornblende, bi-otite,
and occasional quartz phenocrysts in a micro- to cryp-tocrystalline
groundmass (Richter et al., 1991). Potassium-argon ages of 8.0 0.1
Ma and 8.5 0.3 Ma have beenreported on biotite from dacite samples
(Ludington et al.,1992). The igneous rocks were emplaced into
continental redbeds (sandstone, shale, and conglomerate) of the
earlyEocene to early Oligocene Potoco Formation and intoOligocene
to Miocene volcaniclastic rocks of the Upper Que-hua Formation.
Galena, sphalerite, pyrite, and chalcopyriteoccur in veins,
stockworks, and disseminations in hydrother-mally altered igneous
rocks and breccias and in some placesin sedimentary and
volcanoclastic rocks. Silver occurs as na-tive Ag or argentiferous
galena. Mineralization in the Toldosarea consists of
northeast-striking, steeply dipping veins con-taining hematite,
siderite, barite, quartz, native Ag,stromeyerite, and tetrahedrite,
with interspersed stockworksof silver-bearing pyrite veinlets
(Jacobson et al., 1969; Richteret al., 1991).
The large silver-rich polymetallic Pulacayo deposit islocated in
the western margin of the Cordillera Oriental. Pu-lacayo was
discovered in 1833 and mined until 1958 (Cun-ningham et al., 1991)
and is being explored for possible rede-velopment as a disseminated
gold mine. The deposit isassociated with a dacitic volcanic dome
intruded into Silurianshales and sandstones and overlain by
Tertiary continentalsedimentary and pyroclastic rocks (Lyons,
1963). Ore miner-alization occurs in veins and veinlets hosted by
the dome, ex-cept the large Tajo vein which extends into the
Tertiary sedi-mentary rocks (Cunningham et al., 1991). Ore minerals
arechiefly sphalerite, tetrahedrite, freibergite,
argentiferousgalena, and chalcopyrite associated with abundant
pyrite,quartz and barite (Pinto-Vasquez, 1993).
Analytical MethodsThe locations and brief descriptions of
samples are found in
Table 1 and are indicated in Figures 1 and 2. Most rock sam-ples
lack visible evidence of weathering or alteration of anykind. One
igneous rock sample analyzed from the San Cristo-bal district
(91BSL015) was hydrothermally altered. Whole-rock samples were sawn
into slabs, and slabs having evidenceof pervasive weathering,
secondary fractures, or veinlets werediscarded. Saw marks were
ground away using silicon-carbidesandpaper, and the slabs were
cleaned at least three times for5 min each in an ultrasonic cleaner
with deionized water.Cleaned slabs were then broken into small
chips, powderedin an alumina-lined shatterbox vessel; powders were
stored inacid-leached polyethylene screw-top bottles.
Leaching experiments
Elements in crystalline rocks are distributed heteroge-neously
among major, minor, and accessory minerals, or areadsorbed onto
mineral surfaces. Ion exchange at such sur-faces and dissolution of
surface layers supply most of the ele-ments mobilized at the
beginning of fluid-rock interaction(Mller and Giese, 1997). Tessier
et al. (1979) found thattrace metals (Cd, Co, Cu, Ni, Pb, Zn, Fe,
and Mn) could be
grouped into five fractions: exchangeable, bound to carbon-ates,
bound to Fe-Mn oxides, bound to organic matter, andresidual, and
that Fe-Mn oxides and organic matter phasesscavenge trace metals
far out of proportion to their own con-centration. We used NaCl and
HCl solutions to extract themost easily leached lead from our
sedimentary rock samples.
To determine which conditions leached the most lead(NaCl and HCl
concentrations, time, and fluid/rock ratio)preliminary experiments
were performed with sample98M013 (Table 2). The concentration of
lead in the leachateincreased with increasing NaCl and HCl
concentrations,fluid/rock ratio and time. Subsequent leaching runs
were con-ducted on 200-mg samples of each of five sedimentary
rocksamples using leachates of 2 ml 15 percent NaCl solution and2
ml 0.5N HCl. The 15 percent NaCl solution was preparedfrom Spex
ultrapure NaCl (lot 09981 E) dissolved in ultra-pure water. Rock
powders and leaching solutions were placedin Teflon-lined Parr acid
digestion bombs, and the bombswere tightly closed and placed in a
gravity convection oven at200C (the bombs were found to leak above
200C). Leach-ing of sample 98M018 by 0.5N HCl was conducted at
100C.NaCl leachates were heated for 24 h, and 0.5N HCl
leachateswere heated for 7 h. Bombs were allowed to cool to
roomtemperature before being opened. There was no visible lossof
solution in any of the reported experiments. One ml ofeach leachate
was carefully separated with micropipette, andhalf of the separated
leachate was transferred to a cleanTeflon beaker, dried down twice
with 0.5N HBr, and pre-pared for Pb isotope ratio measurement. The
other half of theleachate was spiked with 99.86 percent 208Pb and
99.91 per-cent 235U and diluted to 10 ml with 0.5N HNO3. Isotope
di-lutions were measured on a HP 4500 ICP-MS in the Depart-ment of
Chemistry at Florida International University.
Lead isotope ratio measurements
Chemical preparation of mineral separates and whole rocksfor Pb
isotope analysis were carried out in the Radiogenic Iso-tope clean
laboratory of the Department of Earth Sciences atFlorida
International University. Lead blanks in water, 0.5NHNO3 and 0.5N
HBr prepared in the lab are uniformly 2g/gram, and total chemistry
blanks are typically 500 g.Sample powders were dissolved in 3:2
HF-HBr mixture andevaporated in laminar flow boxes filtered at
0.3m. Sampleswere redissolved in 0.5N HBr and dried down twice.
Pyritesamples were dissolved in 1:1 HNO3 and diluted in 0.5N
HBr.Lead was then separated and purified with cation
exchangecolumns and an HBr medium, following the procedure ofManhes
et al. (1978). Hand-picked galena grains were dis-solved in warm 2N
HNO3. Pb was deposited electrolyticallyon clean Pt electrodes and
then redissolved in 2N HNO3. Aquantity of solution corresponding to
approximately 1 gmPb was analyzed.
All Pb samples were loaded in phosphoric acid on single
Refilaments with silica gel and run on a VG-354 multicollectormass
spectrometer in the Transuranium Research Laboratoryat Oak Ridge
National Lab. In-run precisions of 204Pb/208Pbvalues are better
than 0.05 percent, and those of 206Pb/208Pband 207Pb/208Pb are
better than 0.01 percent, except forleachates of sample 97H022.
Normalization of 7 analyses ofthe lead isotope ratio standard
SRM-981 against the average
Pb SOURCES IN SAN CRISTOBAL, PULACAYO, AND POTOS MINING
DISTRICTS, BOLIVIA 577
0361-0128/98/000/000-00 $6.00 577
-
of values published for that standard by Todt et al. (1984)
andHamelin et al. (1985) yielded correction factors of 0.149
per-cent/amu on 204Pb/208Pb, 0.079 percent on 206Pb/208Pb, and0.102
percent on 207Pb/208Pb. Standard deviations of SRM-981 analyses are
0.06 percent on 206Pb/204Pb, 0.12 percent on207Pb/204Pb, and 0.16
percent on 208Pb/204Pb; overall uncer-tainties of 0.05 percent per
mass unit are cited for all ratios.
Results
Lead isotope measurements
Isotopic analyses of rock and ore samples from the studyarea are
summarized in Table 3. Seven igneous rock samplesfrom the San
Cristobal mining district were analyzed. Fiveigneous rocks show low
lead isotope ratios relative to other
578 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 578
TABLE 1. Description of Samples
Sample Latitude Longitude Area Description
Igneous rocks91BSL013 2107'06" 6713'03" San Cristobal Dacite
dike, fine-grained gray matrix containing small feldspar and
biotite (mm size)
crystals91BSL015 2106'12" 6713'45" San Cristobal Strongly
hydrothermally altered white to yellow fine-grained aggregate
containing
small quartz crystals; altered dacite? 91BSL016 2106'30" 6712'
San Cristobal Dacite intrusion, light-colored fine-grained matrix
containing small amphibole and
altered feldspar crystals (mm size)91BSL018 2104'52" 6711'39"
San Cristobal Dacite, gray, fine-grained matrix containing some
feldspar and biotite crystals91BSL020 2110'04" 6716'51" San
Cristobal Basalt, black, fine-grained matrix, no visible
crystals91BSL021 2102'06" 6724'09" San Cristobal Andesite dike,
fine-grained dark gray matrix containing few euhedral amphibole
crystals up to 5mm 91BSL022 210607" 6712'04 San Cristobal
Andesitic pluton, reddish fine-grained matrix containing altered
feldspars and biotite
Sedimentary rocks97H016 1619'54" 6802'39" La Paz Fine-grained
dark (black) shale, no visible weathering or alteration197H017
1619'54" 6802'39" La Paz Fine-grained dark (black) shale, no
visible weathering or alteration197H018 1619'54" 6802'39" La Paz
Fine- grained dark (black) shale layers alternating with thin gray
to yellow siltstone
layers and some thin Fe oxide veinlets197H019 1842' 6535' Near
Ocuri Red siltstone consisting of alternating thin red
(finer-grained) and white (coarser-
grained) layers 97H020 1842' 6535' Near Ocuri Red siltstone
consisting of alternating thin red (finer-grained) and white
(coarser-
grained) layers97H022 1842' 6535' Near Ocuri Red siltstone
consisting of alternating thin dark red (finer-grained) and light
red
(coarser-grained) layers97H014 1856'50" 6522'46" Sucre
Fine-grained dark (black) shale layers alternating with thin gray
to yellow siltstone
layers and some thin Fe oxide veinlets197H015 1856'50" 6522'46"
Sucre Fine-grained dark (black) shale, no visible weathering or
alteration198M013 1938'45" 6548'09" Near Potos Fine-grained dark
(black) shale, no visible weathering or alteration198M014 1936'55"
6547'04" Near Potos Fine-grained dark (black) shale, no visible
weathering or alteration198M017 1952'49" 6542'09" Near Potos
Fine-grained dark (black) shale, no visible weathering or
alteration198M018 1952'49" 6542'09" Near Potos Fine-grained dark
(black) shale, no visible weathering or alteration198M022 2034'04"
6538'84" East of Pulacayo Fine-grained dark (black) shale layers
alternating with thin gray siltstone layers1
Ore samples98M024C 2025' 6642' Pulacayo Galena crystals from
coarse-grained pyrite aggregate containing some galena, spha-
lerite, and quartz98M024A 2025' 6642' Pulacayo Pyrite crystals
from coarse-grained pyrite aggregate containing some galena,
spha-
lerite, and quartz98M024B 2025' 6642' Pulacayo Pyrite crystals
from coarse-grained pyrite aggregate containing some galena,
spha-
lerite, and quartz98M024D 2025' 6642' Pulacayo Pyrite crystals
from coarse-grained pyrite aggregate containing some galena,
spha-
lerite, and quartz99SC01 2105'42" 6712'37" San Cristobal Galena,
fine-grained aggregate; core sample from 70.0 m99SC02 2105'55"
6712'29" San Cristobal Galena crystals from a small galena vein
hosted in hydrothermally altered volcanic
rock; core sample from 118.8 m99SC03 2105'50" 6712'26" San
Cristobal Galena crystals from ore aggregate containing galena,
sphalerite, barite, and kaolin;
core sample from 283.6 m99SC04 2105'19" 6712'44" San Cristobal
Galena aggregate;core sample from 155.4 m99SC08 2105'59" 6712'16"
San Cristobal Galena crystals from a vein associated with
chalcopyrite, sphalerite, and barite;
Trapiche mine adit99SC09 2105'22" 6712'11" San Cristobal Galena
crystals from coarse-grained aggregate consisting of galena and
bariteTD 07 2106'42" 6712'06" Toldos Toldos mine aditTD 09 2106'42"
6712'06" Toldos Toldos mine adit
1 Samples left a black carbon residue after dissolution
-
volcanic rocks in the central Andes, with 206Pb/204Pb = 17.579to
17.966, 207Pb/204Pb = 15.557 to 15.615, and 208Pb/204Pb =37.804 to
38.209. Sample 91BSL015 yielded an elevated207Pb/204Pb ratio.
Sample 91BSL021 has higher 206Pb/204Pbthan the other igneous rock
samples.
Six galena samples and one pyrrhotite from San Cristobalcontain
relatively nonradiogenic Pb, with 206Pb/204Pb = 17.766to 17.890,
207Pb/204Pb = 15.555 to 15.631 and 208Pb/204Pb =37.954 to 38.206.
Samples of sphalerite and hematite fromthe Toldos deposit, a small
polymetallic deposit near SanCristobal, have slightly higher
206Pb/204Pb. Three pyrites andone galena from the Tajo vein in the
Pulacayo mining districtvary from 18.600 to 18.746 in 206Pb/204Pb,
15.664 to 15.808 in207Pb/204Pb, and 38.850 to 39.212 in
208Pb/204Pb. These valuesare much more radiogenic than the values
for San Cristobalores.
Lead isotope analyses were performed on whole-rock pow-ders of
the most common sedimentary rocks in the region, 11Paleozoic black
shales and 3 Mesozoic red siltstones. The sed-imentary rocks are
more radiogenic than the San Cristobalvolcanic rocks, with
206Pb/204Pb = 18.337 to 19.226, 207Pb/204Pb= 15.640 to 15.814, and
208Pb/204Pb = 38.494 to 39.920.
Leaching experiments
Concentrations of Pb in NaCl leachate solutions increasedsharply
after 24 h of leaching, up to 1,492 ppb (Table 3; Fig.3). A similar
enrichment was observed after 7 hours with HClleaching. Lead
concentrations of HCl leachates are slightlyhigher than NaCl
leachates, with samples 91H017 and91H015 having very similar
concentrations. Leachates of Pa-leozoic black shales contained more
Pb (up to 2,704 ppb Pb)than the Cretaceous red siltstone leachates
(up to 266 ppb).The red siltstone differed from the black shales in
that littlePb was leached by NaCl solution, but almost all of the
Pb con-tained in the rock powder was leached out in 0.5N HCl.
Leadisotope ratios of leachates were consistently lower than
thewhole-rock compositions, to extents well in excess of
instru-mental error. Lead isotope ratios of leachates range
in206Pb/204Pb from 18.131 to 18.623, in 207Pb/204Pb from 15.621to
15.768, and in 208Pb/204Pb from 38.018 to 39.180. Residuesof two
leaching experiments were analyzed. The residue ofsample 98M018 was
more radiogenic than the bulk rock com-position, but both leachate
and residue analyses of sample97H022 were less radiogenic than the
whole-rock analysis;this is difficult to explain unless the
whole-rock powder wasnot well homogenized.
Uranium behaved differently from lead during the same
ex-periments. The NaCl leachates of samples 98M013 and98M018
contained less U than the blank level, indicating thatU was
scavenged by the rock powder. The NaCl leachate ofsample 91H017
contained U close to the blank level, and theNaCl leachate from
sample 97H015 contained relatively highU. Differences in U mobility
between these samples may bedue to the presence of U in different
phases, or differing
Pb SOURCES IN SAN CRISTOBAL, PULACAYO, AND POTOS MINING
DISTRICTS, BOLIVIA 579
0361-0128/98/000/000-00 $6.00 579
TABLE 2. Leaching Experiments with Sample 98M013 under Different
Conditions
Duration Solution Temp. (C) NaCl (%) HCl (N) U (ppb) Th (ppb) Pb
(ppb)
24 h 1 ml 200 5 5.04 13.92 2.424 h 1 ml 200 10 4.82 10.01 28.624
h 1 ml 200 15 4.6 11.07 12424 h 2 ml 200 15 14.4 1913 days 10 ml
200 3.5 4.6 6.15 2543 days 10 ml 200 10 4.7 5.26 59824 h 2 ml 200 1
36.8 145.6 4810
*All runs with 0.2 g rock powder
97H022*
10 100 1000Pb (ppb)
97H015**
97H017**
98M013**
98M018**
97H022*
0.1 1 10 100
U (ppb)
97H015**
97H017**
98M013**
98M018**
A
B
15% NaCl leachate0.5N HCl leachate
15%
NaC
l Bla
nk
6.2N
HCl
Bla
nk15
% N
aCl B
lank
* K-T red siltstone** Pz black shale
FIG. 3. Lead (A) and uranium (B) content in NaCl and HCl
leachates. Pbblank of 6.2 N HCl is less than 10 ppb in (A).
-
580 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 580
TABLE 3. Lead Isotope Composition of Rocks, Leachates, and Ores
from the Study Area
Sample Material Pb U 206Pb/204Pb 207Pb/204Pb 208Pb/204Pb
San Cristobal ppm ppm
91BSL013 Volcanic rock 101 2.0 1.40 17.966 15.615 38.20991BSL015
Volcanic rock 55 1.02 1.31 18.050 15.711 38.51891BSL016 Volcanic
rock 45.4 0.4 0.62 17.899 15.588 38.11591BSL018 Volcanic rock 14
0.8 4.04 17.579 15.557 37.80491BSL020 Volcanic rock 5.0 0.4 5.66
17.628 15.603 37.92191BSL021 Volcanic rock 24.7 1.8 5.15 18.528
15.655 38.66591BSL022 Volcanic rock 120 0.8 0.47 17.838 15.591
38.09699SC01 Galena 17.882 15.631 38.20699SC02 Galena 17.812 15.624
38.15799SC03 Galena 17.784 15.579 38.04999SC04 Galena 17.813 15.612
38.11899SC06 Pyrrhotite 17.890 15.577 38.06899SC08 Galena 17.772
15.588 38.04699SC09 Galena 17.766 15.555 37.954
Toldos99A07 Sphalerite 17.999 15.708 38.57399T11 Hematite 17.964
15.637 38.238
Pulacayo98M024A Pyrite 18.717 15.759 39.04198M024B Pyrite 18.746
15.808 39.21698M024C Galena 18.600 15.664 38.85098M024D Pyrite
18.620 15.698 38.913
Paleozoic and Cretaceous sedimentary rocks ppb ppb
97H013 Sedimentary whole rock 18.691 15.680 39.02497H014
Sedimentary whole rock 18.830 15.688 39.10097H015 Sedimentary whole
rock 18.831 15.814 39.55597H015 NaCl leachate 1,492 30.8 1.46
18.510 15.690 38.94497H015 HCl leachate 1,518 28.7 1.34 18.482
15.659 38.88997H016 Sedimentary whole rock 18.739 15.664
39.31397H017 Sedimentary whole rock 18.821 15.700 39.33797H017 NaCl
leachate 886 1.9 0.15 18.452 15.643 38.96197H017 HCl leachate 931
29.9 2.2 18.475 15.652 38.98897H018 Sedimentary whole rock 18.824
15.790 39.56597H019 Sedimentary whole rock 18.514 15.640
38.49497H020 Sedimentary whole rock 18.589 15.764 38.76497H020 NaCl
leachate 18.131 15.646 38.01997H022 Sedimentary whole rock 13,000
2000 11.0 18.757 15.802 39.17297H022 NaCl leachate 162 45.3 19.6
18.250 15.621 38.32197H022 HCl leachate 266 51.1 13.7 18.282 15.622
38.36897H022 HCl residue 18.685 15.645 38.86598M013 Sedimentary
whole rock 35,600 2200 4.4 18.712 15.802 39.32098M013 NaCl leachate
43.5 0.61 0.99 18.623 15.768 39.18098M013 HCl leachate 2,704 31.1
0.82 18.381 15.648 38.58498M014 Sedimentary whole rock 19.226
15.734 39.92098M017 Sedimentary whole rock 18.889 15.676
39.74698M018 Sedimentary whole rock 19.261 15.741 40.14698M018 NaCl
leachate 165 0.43 0.18 18.596 15.671 39.17998M018 HCl leachate 470
30.5 4.59 18.549 15.643 39.05098M018 HCl residue 20.137 15.778
42.16398M022 Sedimentary whole rock 18.337 15.743 38.610
Blank NaCl 25 0.9 2.55 22.138 17.056 42.811Blank 6.2N HCl 0.48
0.4 56.7
-
amounts of reduced carbon in the sediments, which couldhave
allowed U scavenging by the more reducing powders.However, C
content of sample powders was not measured.Sample 97H022, a red
siltstone considered to be free of or-ganic matter, yielded the
highest U content in the NaClleachate.
In contrast to experiments with 15 percent NaCl, the 0.5NHCl
solutions leached about 10 to 15 percent of U from theshales and
about 25 percent from the red siltstone. Shaleleachates have very
similar U concentrations, about 30 ppb;U in the red siltstone
leachate was 51 ppb. Whole-rock pow-ders of shale (98M013) and red
siltstone (97H022) have verysimilar U content, about 2 ppm (Table
3). Different U con-tent in leachates of shales and siltstone
probably reflect dif-ferences in the mineralogical siting of U
between the tworock types.
Discussion
Lead isotope ratios of the metamorphic basement
The oldest known basement in the central Andes is the
Are-quipa-Antofalla craton (Ramos, 1986). The northern part ofthis
craton, called the Arequipa massif, is exposed in southernPer and
has Early Proterozoic protolith ages (Dalmayrac etal., 1977). Lead
isotope ratios of high-grade gneisses from theArequipa massif are
shown in Figure 4. They have very low206Pb/204Pb (
-
1990; Wrner et al., 2000). It is also known from gneiss
andgranofels clasts and xenoliths in Tertiary sedimentary and
vol-canic rocks in western Bolivia, northwestern Argentina,
andnorthern Chile (Tosdal, 1996). Rocks from the southern
Are-quipa-Antofalla craton yield generally Middle
Proterozoicprotolith ages (Damm et al., 1990). Although these rocks
arenot as isotopically distinctive as the Arequipa massif, theyhave
low present-day 206Pb/204Pb (17.2518.00) and elevated207Pb/204Pb
and 208Pb/204Pb values (Fig. 4; Tosdal et al., 1993;Aitcheson et
al., 1995; Tosdal, 1996). Samples from CerroUyarani yield Early
Proterozoic protolith ages, but theirwhole-rock lead isotope ratios
overlap with those from thesouthern Arequipa-Antofalla craton. A
few have lower206Pb/204Pb, like the Arequipa massif (Wrner et al.,
2000).
Tosdal (1996) proposed that the entire Arequipa-Antofallacraton
was overprinted by Grenville age (ca 1.1 Ga) meta-morphism.
Granulite facies metamorphism in the northernpart of the basement
terrane produced depletion of U withrespect to Th, while lower
grade metamorphism in the south-ern part of the craton produced
less fractionation of U/Th andU/Pb values. The Belen complex
exposed in northern Chilewas apparently not affected by Grenvillian
metamorphismand instead shows two Paleozoic metamorphic
events.Wrner et al. (2000) interpret these differences to reflect
amajor terrane boundary between the Belen complex and therest of
the Arequipa-Antofalla craton; however, their whole-rock lead
isotope systematics are similar.
Lead isotope ratios of Paleozoic and Cretaceous sedimentary
rocks
The early Paleozoic section sampled in the study area
isdominated by a sequence of dark, deep-marine shales up to15 km
thick, formed in a large foreland basin, possibly at apassive-type
margin (Gagnier et al., 1996). Lead isotope ratiosof the Paleozoic
shales are slightly more radiogenic on aver-age in 207Pb/204Pb and
208Pb/204Pb than the present-day esti-mate of the bulk earth from
the Stacey-Kramers model (Fig.4), and are substantially more
radiogenic than both the meta-morphic basement of the Arequipa
massif (Tilton and Bar-reiro, 1980) and the Antofalla craton of
northern Chile(Wrner et al., 1992; Aitcheson, et al., 1995; Tosdal,
1996).Lead isotope ratios of the Ordovician shales are fairly
typicalof upper crustal continental sedimentary rock
compositions,and do not appear to have been affected by U and Pb
loss dueto metamorphism. The Arequipa-Antofalla craton is
thereforeunlikely to be the source of sediments in the Paleozoic
shalessampled for this study, which were most likely derived
fromthe Brazil Shield to the east.
Cretaceous-Tertiary sedimentary rocks in the study area
aremostly red beds. Compositions of nine K-T sedimentary
rocksamples cover a broad range of lead isotope compositions,
butcoincide generally with the compositions of the
underlyingPaleozoic shales (Aitcheson et al., 1995; this study);
theytherefore probably share the same general provenance as
thePaleozoic sedimentary rocks.
Lead isotope ratios of San Cristobal district igneous rocks
San Cristobal district igneous rocks are mostly high K
an-desites and dacites, except for one basalt and one sample
thatshows extensive hydrothermal alteration (Kamenov, 2000).
The unaltered samples have low 206Pb/204Pb values for
given207Pb/204Pb and 208Pb/204Pb and reflect the same source oflead
(Table 3; Fig. 5). On the plot of 207Pb/204Pb versus 206Pb/204Pb,
they lie well above the compositions of Nazca platebasalts and
Nazca plate metalliferous and pelagic sediments(Unruh and
Tatsumoto, 1976; Dasch, 1981; Hamelin et al.,1984), indicating a
crustal lead isotope signature. The mostmafic sample (91BSL020;
SiO2 = 48.3 wt %) contains leadthat is isotopically identical to
that in the more felsic sam-ples(Kamenov, 2000). Lead isotope
ratios of the igneous rocksdo not overlap with the K-T and
Paleozoic sedimentary rocks.Instead, Pb isotope ratios of the
igneous rocks resemble thoseof the Altiplano and the northern Chile
metamorphic basement.
Sample 91BSL021, an andesitic dike located northwest ofthe San
Cristobal district and intruded into the Potoco For-mation
(Kamenov, 2000), contains more radiogenic lead thanthe other
samples. The minor basaltic center, Chiar Kkollu,located north of
San Cristobal at 1926' S and 6723' W, is iso-topically similar to
sample 91BSL021, with 206Pb/204Pb =18.520, 207Pb/204Pb = 15.629,
and 208Pb/204Pb = 38.67. David-son and de Silva (1992) suggest that
these basalts representthe primitive magmas that fed the more
felsic magmatism.The lead isotope ratios of the Chiar Kkollu basalt
are alsonearly identical to the average of all Chilean ore
deposits(Puig, 1988), which are thought to represent the
compositionof the enriched sub-Andean mantle wedge (Macfarlane et
al.,1990).
The isotopic compositions of lead in unaltered igneousrocks from
the San Cristobal district lie near the radiogenicedge of the field
of Altiplano and northern Chile basementrock compositions (Fig. 6).
Igneous rocks in the San Cristobaldistrict may therefore have
acquired their lead isotope signa-ture by mixing between a high
206Pb/204Pb parental magmawith the metamorphic basement. If the
compositions of ChiarKkollu volcanic rocks and average province I
ores approxi-mate the parental composition, the proportions of lead
con-tributed by the parental magma and by the metamorphicbasement
may be estimated. Averages of published isotopiccompositions for
the two parts of the Arequipa-Antofalla cra-ton are shown in Figure
6, together with the average ofChilean ores (Puig, 1988). In this
model, the ores of the SanCristobal deposit, with an average
206Pb/204Pb = 17.82, con-tain about 25 percent lead from the
primary source and 75percent from the metamorphic basement.
Sample 91BSL015 has higher 207Pb/204Pb and slightlyhigher
206Pb/204Pb than the unaltered igneous rock samplesfrom San
Cristobal. This sample shows evidence of strong hy-drothermal
alteration, but its composition does not corre-spond to the
compositions of hydrothermal ores in the districtor to the
compositions of leachates from sedimentary hostrocks and remains
enigmatic.
Hydrothermal leaching of lead
Uranium is incompatible relative to lead in metamorphicfeldspars
and micas, so easily leachable parts of rocks such asgrain
boundaries and crystal defects acquire much higher (238U/204Pb)
than the bulk rock during metamorphism andproduce highly radiogenic
lead over time. Leaching of old,high-grade metamorphic rocks
produces much more radi-ogenic lead in solution than the whole-rock
composition
582 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 582
-
(Gray and Oversby, 1972). Leaching of younger, less
stronglymetamorphosed rocks is less well understood. Figure 7
shows208Pb/204Pb versus 206Pb/204Pb for 15 percent NaCl and 0.5NHCl
leachates of six samples of nearly unmetamorphosed Pa-leozoic and
Cretaceous sedimentary rocks from the studyarea. Plots of
207Pb/204Pb versus 206Pb/204Pb show the same re-lationships. Lead
isotope ratios of the easily leachable frac-tions of Paleozoic
sedimentary rocks are much lower thanthose of bulk, HF-soluble
samples, and the Pb isotope ratiosof the HCl and NaCl leachates of
four samples are within an-alytical error of each other. The
concentrations of lead in theHCl and NaCl leachates of each sample
are also similar, es-pecially for samples 97H015 and 97H017 (Fig.
3). The effectof leaching appears to be fairly insensitive to the
leachingmethodboth leachates extracted roughly the same
solublefraction of the samples. The NaCl leachate of sample
98M013(Fig. 7b) is more radiogenic than the HCl leachate,
nearlyidentical with the bulk sample, and lead concentrations
inthese two leachates are substantially lower than in the
othersamples (Table 3), suggesting that the NaCl and HCl
leachates attacked different lead-bearing materials in this
Pa-leozoic black shale. Most trace elements in sedimentary rocksare
contained in nonlithic particles such as Fe and Mn oxidesand
hydroxides, amorphous aluminosilicates, carbonates, sul-fides, and
organic matter (Martin et al., 1987). Evidently, theleachates of
our samples reflect the nonlithic fraction, whilethe whole-rock
data represent mixtures of the nonlithic andlithic fractions. The
analysis of the HCl residue of sample98M018 supports this idea
(Fig. 7a), indicating a proportionof about 60 percent lead in the
leachable fraction and 40 per-cent in the residue.
Various explanations for the isotopic difference betweenthe
leachable and HF-soluble fractions of the sedimentaryrock powders
deserve consideration. The Paleozoic samplesare deep-marine shales
formed in a foreland basin, possiblyalong a passive margin (Gagnier
et al., 1996). The sampleswere apparently not isotopically
homogenized, because re-gressions through leachate and whole-rock
analyses for indi-vidual samples yield ages of 1.2 to 4.2 Ga, far
older than thesedimentation age. The leachable and HF-soluble
fractions
Pb SOURCES IN SAN CRISTOBAL, PULACAYO, AND POTOS MINING
DISTRICTS, BOLIVIA 583
0361-0128/98/000/000-00 $6.00 583
38
39
40
41
42
15.4
15.5
15.6
15.7
15.8
15.5 16.5 17.5 18.5 19.5 20.5
206Pb/204Pb
207 P
b/20
4 Pb
208 P
b/20
4 Pb
Potos oresSan Cristobal oresSan Cristobal igneous rocksPulacayo
ores
Nazca Plate sedimentsNazca Plate MORBPz, K and T sedimentary
rocks
Arequipa Massif basement
Toldos ores
Altiplano/N. Chile Mid-Proterozoic basement
FIG. 5. Lead isotope ratios of ores and igneous rocks from the
San Cristobal, Potos, and Pulacayo mining districts of Bo-livia,
compared with potential metal sources. Uncertainties of ore and
igneous rock analyses are comparable to symbol sizes.Data sources
include those listed for figure 4, and Unruh and Tatsumoto (1976),
Dasch (1981), Tilton et al. (1981), Hamelinet al. (1984) and
Macfarlane et al. (1990).
-
may have been deposited with different initial lead isotope
ra-tios if hydrothermal activity in or near the basin contributedto
the leachable fraction, although we are not aware of evi-dence for
such activity. These rocks may also have been per-meated by
diagenetic or low-temperature metamorphic fluidswith low values. If
such fluids contained nonradiogenic Pbfrom the metamorphic
basement, they could have producedthe observed isotopic
systematics.
The nonradiogenic compositions of the shale leachates mayalso be
explained by their low values (Table 3), and the ob-servation that
some of the shales scavenged U from theleachate solution rather
than the reverse. During diagenesisor low-grade metamorphism, pore
fluids may have mobilizedlead into the leachable fraction of the
rock, but U remainedinsoluble due to the reducing nature of the
black shale. Thelow values in the leachable fractions would produce
nonra-diogenic Pb in this part of the rock over time. The failure
ofwhole-rock and leachate analyses to form sensible
isochronssuggests that at least one component (probably the
leachablefraction) was open to U and/or Pb exchange at various
times.This explanation could be valid for clay and organic-rich
sed-imentary rocks (i.e., the shales) but not for the red
siltstones.Sample 97H022 is a red siltstone and the NaCl and
HClleachates are also less radiogenic than the whole-rock
sample,but the values are relatively high (Table 3).
Sources of metals in the Potos, Pulacayo, and San Cristobal
deposits
Comparison of the lead isotope compositions of ore miner-als
from San Cristobal with those of the related igneous rocks,
the regional Paleozoic and Cretaceous sedimentary rocks,and the
underlying regional metamorphic basement yields apicture of the
likely provenance of metals in this deposit (Fig.8). The similarity
of ore lead isotopes with those in the ig-neous rocks in the
district indicates that the proximal sourceof ore metals in the
hydrothermal system was the igneous ac-tivity. Isotopic
compositions of leachates of the Paleozoic andCretaceous
sedimentary rocks are not consistent with those ofthe ore minerals,
so hydrothermal leaching of these rockscannot have supplied a
significant proportion of the oremetal. The igneous rocks
themselves appear to have acquiredtheir lead mainly from the
regional metamorphic basementwith a lesser component, probably a
province I-type well-mixed source derived from the sub-Andean
mantle wedge en-riched by subducted sediments. Ores from the Toldos
depositin the San Cristobal district have higher 206Pb/204Pb than
theother San Cristobal ores and igneous rocks. The significanceof
this difference is not clear, but may reflect the presence ofan
igneous body at depth with isotopic characteristics differ-ent from
those sampled so far.
The Pulacayo and Potos deposits are both related to felsicdomes
emplaced through and onto the Paleozoic sedimentarysequence. Some
veins in each deposit extend into the subjacentsedimentary rocks.
Hydrothermal circulation must have af-fected the sedimentary rocks,
and lead isotope ratios of Potosores are known to resemble those of
whole-rock Paleozoicshale samples (Macfarlane et al., 1990). Until
now, however, ithas not been clear whether the lead was
incorporated by mag-matic assimilation or derived from hydrothermal
leaching ofthe Paleozoic sedimentary rocks. Analyses of ore
minerals from
584 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 584
15.5
15.6
15.7
16.5 17 17.5 18 18.5 19
San Cristobal oresBerenguela oresToquepala oresCerro Verde
oresMarcona ores
Altiplano/N. Chile basement
25%
50%
75%
25%75%
Province Ia (Chile) oresArequipa Massif
Average Arequipa MassifEarly-Proterozoic basement
Average Province Ia
Average Altiplano/N. ChileMid-Proterozoic basement
206Pb/ 204Pb
207 P
b/20
4 Pb
Madrigal ores
FIG. 6. Comparison of lead isotopic compositions of the basement
beneath province IV, and ores from province Ia andprovince IV. Data
from Tilton and Barreiro (1980), Damm et al. (1990), Macfarlane et
al. (1990), Aitcheson et al. (1995), Tos-dal (1996), and Wrner et
al. (2000) and this study. Mixing trajectories are shown between
the average composition of provinceIa lead and lead from the
Arequipa massif metamorphic basement and from northern Chile
metamorphic basement.
-
the Tajo vein in the Pulacayo mining district and from the
Po-tos deposit are shown in Figure 8 with data for Paleozoic
andCretaceous sedimentary rocks. Ore minerals from Pulacayoand
Potos are isotopically very similar to each other, and bothresemble
the whole-rock compositions of the regional Paleo-zoic and
Cretaceous sedimentary rocks. The leachates havemuch lower
206Pb/204Pb than ores from the Potos and Pulacayodistricts (Fig.
8), which suggests that hydrothermal leaching of
the Paleozoic sedimentary rocks cannot have been a
significantsource of metals in the Potos and Pulacayo ores.
Instead, leadmust have been incorporated by magmatic assimilation
or ana-texis of sedimentary rocks, followed by partitioning of
thosemetals into hydrothermal fluids. Melting of the Paleozoic
sedi-mentary rocks is supported by the presence of at least two
agesof inherited zircon, one Mesozoic and one Proterozoic, in
thePotos dacite volcanic dome (Zartman and Cunningham, 1995).
Pb SOURCES IN SAN CRISTOBAL, PULACAYO, AND POTOS MINING
DISTRICTS, BOLIVIA 585
0361-0128/98/000/000-00 $6.00 585
38
40
42
44
18 19 2038
39
40
18 18.5 19
38
39
40
18 18.5 1938
39
40
18 18.5 19
38
39
40
18 18.5 1938
39
40
18 18.5 19
98M018Pz black shale
A B
C D
E F
206Pb/204Pb 206Pb/204Pb
208 P
b/20
4 Pb
208 P
b/20
4 Pb
208 P
b/20
4 Pb
98M013Pz black shale
97H015Pz black shale
97H020Pz black shale
97H017Pz black shale
97H022K-T red siltstone
FIG. 7. 208Pb/204Pb versus 206Pb/204Pb for Paleozoic and
Cretaceous sedimentary rocks from the study area, compared
toleachates and residua. Solid squares represent whole-rock
analyses, open squares are 15 percent NaCl solution leachates,open
diamonds are 0.5N HCl leachates and solid diamonds are residua of
HCl leaching experiments. Uncertainties of leadisotope measurements
are comparable to symbol sizes, except for leachates of 97H022,
which yielded low-intensity runs.
-
Reevaluation of Andean lead isotope provinces
Although there are very well documented differences be-tween the
predominant ore metals mined in different parts ofthe Andes, they
appear to be largely independent of the orelead isotope provinces,
and therefore the source regions ineach province. For example, both
gold-rich and gold-poordeposits in province I have the same lead
isotope characteris-tics, and in province III copper, silver, gold,
and base-metaldeposits all possess the same general isotopic
characteristics(Macfarlane et al., 1990). These isotopic provinces
illustratethe importance of long-lived, isotopically distinct
crustalreservoirs in providing metals to a variety of different
types ofcontinental arc ore deposits.
Province I ores have a restricted range of lead isotopic
ra-tios, with 206Pb/204Pb = 18.2 to 18.8, 207Pb/204Pb = 15.55
to15.69, and 208Pb/204Pb = 38.11 to 38.95 (Fig. 8). Macfarlane
etal. (1990) proposed that the lead in province I was derivedfrom
the sub-Andean mantle enriched by subducted sediments.The
relatively lead-rich subducted sediments would rapidlydominate the
lead isotope composition of the sub-Andeanmagma source region
(Aitcheson and Forrest, 1994; Mac-farlane, 1999). Province III lead
is much more isotopically
variable than province I lead, having 206Pb/204Pb = 17.9
to25.18, 207Pb/204Pb = 15.51 to 16.00, and 208Pb/204Pb = 37.7
to40.07 (Fig. 8). The Bolivian-Argentine section of province
III(referred to here as province IIIa) define a trend with
ele-vated 207Pb/204Pb and 208Pb/204Pb relative to 206Pb/204Pb,
indi-cating dominantly continental sources. Several authors
havesuggested that the Paleozoic sequence in the central Andescould
be the source of radiogenic province III type lead(Tilton et al.,
1981; Macfarlane et al., 1990; Tosdal et al., 1993).Province IIIa
deposits show a progression from Ordoviciandeposits at the
nonradiogenic end of the array, to Mesozoicdeposits in the middle
of the array, to Tertiary deposits at theradiogenic end (Macfarlane
et al., 1990). This progression isthought to reflect episodic
melting of the thick Paleozoic se-quence having 238U/204Pb about 10
and 232Th/204Pb about 45(Schuiling, 1967; Tilton et al., 1981;
Macfarlane et al., 1990).Derivation of late Cenozoic magmatism
related to provinceIIIa ores primarily from melting of crustal
materials is sup-ported by Nd values in the range of 2 to 10 for
these rocks(Miller and Harris, 1989).
Lead from province II ores have consistently higher208Pb/204Pb
and 207Pb/204Pb than the province I ores (Fig. 8).Province II ores
frequently show steep mixing arrays that
586 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 586
15.5
15.6
15.7
18.2 18.4 18.6 18.8 19
Province I or
es
207 P
b/20
4 Pb
206Pb/204Pb
Province III Pz-Ksedimentarywhole-rocks
38.0
38.2
38.4
38.6
38.8
39.0
39.2
15.8
208 P
b/20
4 Pb
ColquiJulcaniCasapalca
OrcopampaMilpo/Atacocha
Potos oresPulacayo ores
Sedimentaryrock leachates
Provinc
e I ores
Province III Pz-Ksedimentarywhole-rocks
Province II ores Province III
Province IIIa
ores
Province
IIIa ore
s
FIG. 8. Comparison of lead isotope ratios in province I and IIIa
ore deposits with those in province III sedimentary
rocks,sedimentary rock leachates, and province II ore deposits.
Data from compilations of Macfarlane et al. (1990) and
sourcestherein and of Aitcheson et al. (1995) and sources therein,
Macfarlane (1999) and this study. Province I includes ores
fromcentral and northern Chile, and some coastal areas of Per.
Province IIIa includes deposits in Bolivia and northwest Ar-gentina
to the east and south of the central Altiplano (cf. Fig. 9a).
-
originate within the province I field and extend to
composi-tions within the province III field. While steep arrays
canform on lead isotope plots due to uncorrected
instrumentalfractionation, and fractionation cannot be ruled out as
a causeof some of the spread in the province II arrays, other
factorsmitigate against their being mere artifacts of the analysis.
Thearrays are steeper than fractionation arrays on
208Pb/204Pb206Pb/204Pb diagrams and span a larger range of
compositionthan can be explained by typical analytical
uncertainties. Thearrays occur only in analyses by various authors
from a re-stricted area of the Peruvian cordillera, and not in data
fromdeposits in provinces I and III. The arrays therefore
probablyindicate mixing of province I lead with upper crustal
compo-nents having higher 207Pb/204Pb and 208Pb/204Pb typical
ofprovince III ores (Macfarlane, 1999). In principle, mixingcould
take place either by magmatic assimilation or hy-drothermal
scavenging of radiogenic lead from host Paleozoicand Mesozoic
sections. However, leachates of Paleozoic andCretaceous sedimentary
rock plot at 206Pb/204Pb values muchtoo low to serve as an
acceptable radiogenic endmember forany of the province II arrays
(Fig. 8). The province II mixingarrays are, therefore, more likely
caused by assimilation of ra-diogenic province III-type material by
relatively nonradi-ogenic, province I type magmas, followed by
extraction ofmetals from magmas that are not completely
homogenized.
The Potos, San Cristobal, and Pulacayo mining districtsspan the
tentative boundary between province II and III pro-posed by
Petersen et al. (1993). Only one analysis each hasbeen available
for San Cristobal and Pulacayo, so their posi-tion within the lead
isotope province scheme has not beenwell constrained. The Potos
deposit was included in the orig-inal lead isotope province map of
Macfarlane et al. (1990),and plotted within the range of the
Paleozoic sedimentaryrock lead isotope ratios and clearly
represents province IIImetal sources. The same is true of the
Pulcayo deposit.
Late Miocene San Cristobal ores have 206Pb/204Pb valuesbetween
17.766 and 17.882, lower than the least radiogenic(Ordovician) ores
from province III and far lower than anyTertiary province III ores
(Fig. 9c, d). Miocene polymetallicvolcanic-hosted veins from the
Berenguela district, north-northwest of San Cristobal in
west-central Bolivia, also con-tain relatively nonradiogenic lead,
interpreted by Tosdal et al.(1993) to reflect incorporation of Pb
from the Proterozoicmetamorphic terrane underlying the area. Most
of the vol-canic rocks in western Bolivia north of San Cristobal
alsoshow relatively low lead isotope ratios, interpreted by
manyauthors to reflect local basement influence (e.g., Tosdal et
al.,1993; Aitcheson et al., 1995). Farther northwest in southernPer
are the large Paleocene porphyry copper deposits atToquepala and
Cerro Verde, and the undated Pb-Zn-Ag de-posit at Madrigal, and the
late Jurassic volcanogenic mag-netite deposit at Marcona (Mukasa et
al., 1990; Fig. 9a). Al-though these deposits lie within the
coastal magmatic arcassigned to lead isotope province I by
Macfarlane et al.(1990), their ore lead isotope data are all
shifted toward lower206Pb/204Pb and higher 208Pb/204Pb values than
ores of theChilean section of province I. Macfarlane et al (1990)
attrib-uted this shift to incorporation of lead from the local
meta-morphic basement, and divided province I into three
sub-provinces as a result.
Based on these new and published data, we have defined anew ore
lead isotope province in the central Andes, extendingfrom the
coastal area of southern Per through southern Perand western
Bolivia to San Cristobal (Fig. 9a). Province IV isdivided into
subprovinces reflecting the relatively minorbasement imprint in the
southern Per segment (IVa) and themuch stronger basement imprint in
the Bolivian segment(IVb). The internal subdivision of province IV
is only approx-imate and may well be gradational; although the
southernpart of the Arequipa-Antofalla metamorphic basement
hasgenerally similar lead isotope characteristics, it appears to
beheterogeneous in age and may be a mosaic of smaller
terranes(Aitcheson et al., 1995; Wrner et al., 2000). The eastern
andsoutheastern boundary between province IV and provinceIII, which
passes roughly north-south through the center ofthe Altiplano, is
constrained by analyses of several small Ter-tiary polymetallic
deposits (Maria Luisa, Almacen, Escala,and Todos Santos), and the
Bi- and W-bearing Lipia-Galandeposit (Long, 1991; Aitcheson et al.,
1995), and by analysesof copper ores from the Corocoro district
south-southeast ofLake Titicaca (A.W. Macfarlane and H. Lechtman,
unpub.data). Each of these deposits has a typical Tertiary
provinceIII-type lead isotope signature and contrasts strongly
withSan Cristobal and with the volcanic rocks in the northern
Al-tiplano province of Aitcheson et al. (1995). The
southwesternboundary (west and southwest of San Cristobal) is still
poorlyconstrained due to a lack of ore deposit data. The
westernboundary between province I and province IV is loosely
de-fined by northern Chilean porphyry Cu deposits
includingChuquicamata and Tignamar in far northern Chile (Zentilli
etal., 1988; Puig, 1988).
Although province IVa and IVb ores both contain lead fromthe
local metamorphic basement, the relative proportions ofbasement
lead in their ores differ. In the San Cristobal de-posit, as we
have seen, about 75 percent of the lead in the oreappears to be
derived from the metamorphic basement (Fig.6). A similar estimate
of mixing between the average Are-quipa massif metamorphic basement
and a hypothetical pri-mary magma with a lead isotope composition
like that ofprovince I yields about 10 percent basement-derived
lead inthe Toquepala and Cerro Verde deposits, about 12 percent
inMadrigal ores and 18 percent in Marcona ores (Fig. 6).Berenguela
district ores have higher 208Pb/204Pb than the SanCristobal ores,
suggesting that some of the Berenguela orelead was assimilated from
higher-grade metamorphic base-ment with elevated Th/U like that of
the Arequipa massif. Iso-topic compositions of ores from the
Berenguela district arecompatible with incorporation of about 25
percent of high-grade metamorphic basement lead.
While these estimates are crude, the different proportionsof
basement lead in ores from provinces IVa and IVb must
besignificant. Deposits in the two parts of province IV differ
inage. The Peruvian deposits in province IVa, which contain
theleast basement lead, are late Paleocene, with Cerro Verdehaving
formed about 58.9 2.0 Ma (K-Ar date on biotite;Estrada, 1975) and
Toquepala about 58.7 1.9 Ma (K-Ar dateon biotite; Clark et al.,
1990). Ore formation farther south inthe Berenguela district, which
contains more basement lead,probably took place in mid- to late
Miocene (Wallace et al.,1992). San Cristobal, which contains the
most basement lead,
Pb SOURCES IN SAN CRISTOBAL, PULACAYO, AND POTOS MINING
DISTRICTS, BOLIVIA 587
0361-0128/98/000/000-00 $6.00 587
-
588 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 588
Toquepala
22
70
0 400Scale (km)
Cerro VerdeMadrigalMarcona
Deposits with Pb isotope data
22
1818
70
1414
66
6674
Berenguela
San Cristobal
Tignamar
A
Orcopampa
Julcani
IVa
IVb
Approx. limitof Altiplano
7050
22
70
Approx. depth toMoho (km) (James , 1971)
22
1818
70
1414
66
6674
Approx. margin of Arequipa-Antofalla craton (Todsal, 1996)
IVb
B
60
Julcani
IVa
Extent of province IV
19.018.518.0206 204
Pb/
Pb
207
204
15.5
15.6
15.7
15.8
38.5
39.0
Pb/
Pb
208
204
39.5
Province IVb Province IVa
38.0
San Cristobal ores
Berenguela ores
Province IProvince IIProvince IIIaProvince IIIb
Pb/ Pb
Province IVb
Province IVa
Province IVa ores
Province IV
Toldos ores
C
D17.5
FIG. 9. Ore lead isotope provinces of the central Andes. (A)
Geographic distribution of ore lead isotopic provinces; (B)position
of province IV relative to the margins of the Arequipa-Antofalla
terrane (Tosdal, 1996) and contours of crustal thick-ness (James,
1971); (C) and (D) lead isotope ratios of ores from each province.
Black dots indicate locations of deposits withfor which lead
isotope data are available. Data sources as in Figure 9, plus
Tosdal et al. (1993).
-
formed about 8.5 Ma (Ludington et al., 1992). These
secularchanges in the proportion of basement lead incorporationmay
correlate with the history of crustal thickening in the
Al-tiplano.
The crust beneath the modern Altiplano is 70 to 75 kmthick (Fig.
9b); thickening is thought to have begun about 40m.y. ago and
increased rapidly about 20 m.y. ago (Isacks,1988; Lamb and Hoke,
1997; James and Sacks, 1999). The 25Ma Chiar Kkollu basalts are not
extensively contaminated bymetamorphic basement lead, and are
thought to representthe primary magmas that feed the arc (Davidson
and de Silva,1992). Crustal thickening was accompanied by
increasing87Sr/86Sr in magmas erupted through the thickened
crustwithin the last 20 m.y. and by assimilation of old,
isotopicallyevolved metamorphic basement by late Cenozoic
magmaserupted in and near the study area (Davidson et al.,
1990;Feeley, 1993). Volcanic rocks in the western Altiplano
andCordillera Occidental containing dominantly metamorphicbasement
lead are younger than 20 Ma (Aitcheson et al.,1995). Major and
trace element data for the San Cristobal dis-trict igneous rocks
(Kamenov, 2000) indicate extensive horn-blende fractionation,
consistent with combined assimilationand fractional crystallization
under high total pressure andPH2O (Davidson and de Silva, 1995).
The isotopic homogene-ity of lead in igneous rocks from San
Cristobal indicates thatchemical differentiation of the magmas
occurred after thecontamination with basement material, and that
the magmaswere then emplaced into the upper crustal rocks without
sig-nificant further assimilation. Incorporation of lead from
themetamorphic basement by deep magmatic assimilation ap-pears to
have been enhanced in province IVb by the thicken-ing of the Andean
crust after 20 Ma.
The Arequipa-Antofalla metamorphic basement terrane isbelieved
to have a much greater extent than the province wecan define based
on ore lead isotope data (Fig. 9b; Tosdal,1996). The area of
basement influence on volcanic rock com-positions is also wider and
extends further to the south thanprovince IV (Wrner et al., 1992;
Aitcheson et al., 1995). Thebasement influence zones defined by
Aitcheson et al. (1995)occupy most of the width of the Altiplano in
Bolivia, corre-sponding roughly to the 60 km crustal thickness
contour onFig. 9b, though it also is not as wide as the proposed
Are-quipa-Antofalla craton. Aitcheson et al. (1995) divided
theirarea of basement influence into a northern Altiplano zonewhere
the basement was nonradiogenic, a transition zonewhere the basement
was of mixed radiogenic-nonradiogeniccharacter, and a southern
Altiplano zone with radiogenic leadisotope signatures. Our province
IV extends well into thetransition zone of Aitcheson et al. (1995),
where San Cristo-bal contains the least radiogenic ore lead in the
central Andes.
The significance of the much smaller area of province IVcompared
to that of the metamorphic basement that providedits distinctive
ore lead isotope signature, or to the isotopic do-mains defined
from volcanic rock compositions, is not clear.Province IV is
defined by a small number of samples com-pared to the basement
domains outlined by scores of volcanicrock analyses (Aitcheson et
al., 1995). Analyses of additionaldeposits will better define
province IV and may expand itsboundaries. However, the eastern
boundary of province IV isthe best constrained, and it indicates a
much narrower zone
of influence than that reflected in volcanic rocks. The
base-ment may be thinner in some areas than others, or it may
notoccur at the depth where magmatic assimilation took place inall
areas, and it probably contains isotopically heterogenousenclaves
that create different isotopic imprints on assimilatingmagmas. None
of these possibilities alone explain why vol-canic rocks and ores
in the same or nearby areas should showstrong isotopic
differences.
There is also an age bias between the sampling of volcanicrocks
and ores. Most Altiplano ore deposits are of Mioceneage and have
been exposed by some erosion; none of the leadisotope data are from
ore deposits known to be younger.Many of the volcanic rocks
analyzed by Aitcheson et al. (1995)are of Pleistocene and Holocene
age. Because the crust be-neath the Altiplano is thought to have
thickened significantlyin that interval (Lamb and Hoke, 1997), the
Miocene oresmay reflect basement influence only over a small area
wherethe crust was thickest at that time. The
Pleistocene-Holocenevolcanic rocks may reflect extensive basement
assimilationover a wider area of thickened crust.
ConclusionsLead isotope ratios of ores and igneous rocks from
the San
Cristobal district, of regional sedimentary rock units
andleachates of those sedimentary rocks, and of the
regionalmetamorphic basement provide a clear picture of the
sourceof metals in the San Cristobal Ag-Zn deposit. Ore Pb in
thehydrothermal system, and probably geochemically similarmetals
such as Cu, Zn, and Ag, were derived from the associ-ated igneous
rocks. Those igneous rocks in turn appear tohave assimilated about
75 percent of their lead from the Mid-dle Proterozoic metamorphic
basement (part of the Are-quipa-Antofalla craton) which underlies
this area. The re-maining, more radiogenic component was probably
derivedfrom greater depth. Ore lead in the Pulacayo and Potos
de-posits was derived by magmatic assimilation of the thick
EarlyPaleozoic sedimentary sequence that immediately underliesthose
deposits, perhaps because the sedimentary rocks arethicker to the
east beneath those deposits, or the Middle Pro-terozoic metamorphic
basement may be absent there.
Lead isotope ratios of leachates of samples from the thickEarly
Paleozoic and Cretaceous sedimentary sequences ofthe Eastern
Cordillera and Altiplano indicate that hydrother-mal leaching of
these rocks is probably not an importantmechanism for concentrating
ore metals in the central Andesgenerally. Instead, most central
Andean ore deposits inland ofthe coastal belt (province I) appear
to contain a minor to dom-inant proportion of lead derived by
magmatic assimilation ofthe crust. This implies that central Andean
ores form at leastpartly by reconcentration of metals already
resident in thecrust. Additions from a deeper, more radiogenic
source, likeprovince I lead, make up the rest of the metal budget,
andprobably represent new additions of metals to the crust.
Thecentral Andean crust has probably therefore been progres-sively
enriched in lead and perhaps other ore metals throughtime. This may
partly explain the extraordinary wealth ofmagmatic-hydrothermal
ores in the orogen. Although world-class deposits like Potos, San
Cristobal, and Chuquicamata(which is dominated by province I type
lead) can contain met-als from greatly different sources if the
magmatic-hydrothermal
Pb SOURCES IN SAN CRISTOBAL, PULACAYO, AND POTOS MINING
DISTRICTS, BOLIVIA 589
0361-0128/98/000/000-00 $6.00 589
-
processes are favorable, enrichment of the crust in metalsmay
have played a role in the genesis of all of them. ProvinceI crust
may have been enriched in metals over time, but if iso-topically
distinct metamorphic basement is absent, magmaticassimilation will
not be revealed by lead isotope analyses.
The new data in this paper and the review of published datahas
allowed an extensive revision of the lead isotope provincemap of
the central Andes and helped to define a new lead iso-tope
province. The presence of isotopically distinctive Earlyand Middle
Proterozoic crust (the Arequipa-Antofalla craton)beneath this
province provides a clear indication of the extentof midcrustal
magmatic assimilation by major ore-formingcontinental arc systems.
Ores in the northern part of thisprovince, designated province IVa,
contain a small to moder-ate component of lead from the highly
nonradiogenic Are-quipa massif rocks that underlie the region. Ores
in the south-ern part of this province, designated province IVb,
containthe majority of lead extracted from the underlying
metamor-phic basement. Province IVa ores are of Paleocene age
andformed prior to the late Tertiary thickening of the Andeancrust,
while later Miocene deposits to the south contain thelargest
proportions of basement-derived lead. Thicker crustin Miocene time
may have promoted ponding of magmas atintermediate levels in the
crust and extensive assimilation ofmetamorphic basement in province
IVb. Still younger Pleis-tocene-Holocene volcanic rocks in the
Altiplano reflectstrong basement influence over a broader area,
perhaps re-flecting a greater extent of thickened crust with time.
Furtherstudies of the depth of the Arequipa-Antofalla
metamorphicbasement combined with new measurements of the lead
iso-tope ratios and chronology of Altiplano ores should
provideimportant insights into the depth of assimilation and
metalsource processes in continental arcs.
AcknowledgmentsThis work grew out of the MS thesis of the senior
author at
Florida International University. We would like to thankRosemary
Hickey-Vargas for advice and instruction duringthe thesis work, and
Grenville Draper and K. Panneerselvamfor instructive comments. Yun
Cai and Alberto Sabucedo ofthe FIU Department of Chemistry provided
access to andsupport for our use of the ICP-MS laboratory. The
researchwas partially sponsored by the Division of Chemical
Sciences,Geosciences and Biosciences, Office of Basic Energy
Sciences,U.S. Department of Energy, under contract
DE-AC05-00OR22725 with Oak Ridge National Laboratory, managedand
operated by UT-Battelle, LLC. Eddie McBay providedtechnical support
for the lead isotope measurements made byAWM at ORNL. Johnny
Delgado, Mark West, Carlos Lozano,and Magdalena Luna of Andean
Silver helped us obtain sam-ples from San Cristobal and Pulacayo.
Heather Lechtman ofMIT collected some of the samples of Paleozoic
sedimentaryrocks and provided field support for the collection of
the rest.Steve Ludington of the U.S. Geological Survey
generouslyprovided igneous rock samples from San Cristobal and
gavepermission to cite his unpublished report on that deposit.This
manuscript was significantly improved by thoughtful andconstructive
reviews from Richard Tosdal, Daniel Kontak,John Dilles, and Mark
Hannington.July 1, 2001; January 21, 2002
REFERENCESAitcheson, S.J., and Forrest, A.H., 1994,
Quantification of crustal contami-
nation in open magmatic systems: Journal of Petrology, v. 35, p.
461488.Aitcheson, S.J., Harmon, R.S., Moorbath, S., Scneider, A.,
Soler, P., Soria-Es-
calante, E., Steele, G., Swainbank, I., and Wrner, G., 1995, Pb
isotopesdefine basement domains of the Altiplano, central Andes:
Geology, v. 23, p.555558.
Clark, A.H., Farrar, E., Kontak, D.J., Langridge, R.J., Arenas,
F., France,L.J., McBride, S.L., Woodman, P.L., Westeneys, H.A.,
Sandeman, H.A.,and Archibald, D.A., 1990, Geologic and
geochronologic constrains on themetallogenic evolution of the Andes
of southern Per: ECONOMIC GEOL-OGY, v. 85, p. 15201583.
Crocetti, C.A., Holland, H.D. and McKenna, L.W., 1988, Isotopic
composi-tion of lead in galenas from the Viburnum Trend, Missouri:
ECONOMIC GE-OLOGY, v. 83, p. 355376.
Cunningham, C.G., McNamee, J., Pinto-Vasquez, J., and Ericksen,
G.E.,1991, A model of volcanic dome-hosted precious metal deposits
in Bolivia:ECONOMIC GEOLOGY, v. 86, p. 415421.
Dalmayrac, B., Lancelot, J.R., and Leyreloup, A., 1977,
Two-billion-yeargranulites in the late Precambrian metamorphic
basement along the south-ern Peru coast: Science, v. 198, p.
4951.
Damm, K.-W., Pichowiak, S., Harmon, R.S., Todt, W., Kelley, S.,
Omarini, R.,and Niemaeyer, H., 1990, Pre-Mesozoic evolution of the
central Andes:The basement revisited: Geological Society of America
Special Paper 241,p. 101126.
Dasch, J., 1981, Lead isotopic composition of metalliferous
sediments fromNazca plate: Geological Society of America Memoir, v.
154, p. 199210.
Davidson, J.P., and de Silva, S.L., 1992, Volcanic rocks from
the Bolivian Al-tiplano: Insights into crustal structure,
contamination and magma genesisin the central Andes: Geology, v.
20, p. 11271130.
1995, Late Cenozoic magmatism of the Bolivian Altiplano:
Contribu-tions to Mineralogy and Petrology, v. 119, p. 387408.
Davidson, J.P., McMillan, N.J., Moorbath, S., Worner, G.,
Harmon, R.S., andLopez-Escobar, L., 1990, The Nevados de Payachata
volcanic region (18S,69 W, N. Chile) II. Evidence for widespread
crustal involvement in An-dean magmatism: Contributions to
Mineralogy and Petrology, v. 20, p.412432.
Dixon, P.R, Rye, D.M., and Janecky, D.R., 1991, Fluid flow
connections tobasement rocks below sedimentary basins: evidence
from the base metaldeposits in Ireland: Geophysical Research
Letters 18, p. 943946.
Estrada, Ch.F., 1975, Geologia de Quellaveco: Boletn de la
Sociedad Geo-logica del Per, v. 46, p. 6586.
Feeley, T., 1993, Crustal modification during subduction-zone
magmatism:Evidence from the southern Salar de Uyuni region (2022S),
centralAndes: Geology, v. 21, p. 10191022.
Gagnier, P.Y., Bliek, A., Emig, C.C., Sempere, T., Vachard, D.,
and Vangues-taine, M., 1996, New paleontological and geological
data on the Ordovicianand Silurian of Bolivia: Journal of South
American Earth Sciences, v. 9, p.329347.
Gray, C., and Oversby, V., 1972, The behavior of lead isotopes
during gran-ulite facies metamorphism: Geochimica et Cosmochimica
Acta, v. 36,p.939952.
Hamelin, B., Dupr, B., and Allgre, C., 1984, Lead-strontium
isotopic vari-ations along the East Pacific Rise and the
mid-Atlantic ridge: A compara-tive study: Earth and Planetary
Science Letters, v. 67, p. 340350.
Hamelin, B., Manhes, G., Alberede, F., and Allgre, C.G., 1985,
Precise leadisotope measurements by the double-spiked technique: A
reconsideration:Geochimica et Cosmochimica Acta, v. 49,
p.173182.
Harmon, R., Barreiro, B., Moorbath, S., Hoefs, J., Francis, P.,
Thorpe, R.,Deruelle, B., McHugh, J., and Viglino, J., 1984,
Regional O, Sr and Pb iso-tope relationships in late Cenozoic
calc-alkaline lavas of the AndeanCordillera: Journal of the
Geological Society of London, v. 141, p. 803822.
Hedenquist, J.W., and Gulson, B.L., 1992, Intrusive and basement
rockssources of lead in hydrothermal systems of the Taupo volcanic
zone, NewZealand: Geochimica et Cosmochimica Acta, v. 56, p.
28212829.
Hildreth, W., and Moorbath, S., 1988, Crustal contributions to
arc magma-tism in the Andes of central Chile: Contributions to
Mineralogy and Petrol-ogy, v. 98, p. 455489.
Isacks, B.L., 1988, Uplift of the central Andean plateau and
bending of theBolivian orocline: Journal of Geophysical Research,
v. 93, p. 32113231.
Jacobson, H.S., Murillo, G., Ruiz, L., Tapia, O., Zapata, H.,
Alargon, H., Del-gadillo, E., and Velasco, C., 1969, Geology and
mineral deposits of the San
590 KAMENOV ET AL.
0361-0128/98/000/000-00 $6.00 590
-
Cristobal district, Villa Martin province, Potos, Bolivia: U.S.
GeologicalSurvey Bulletin, v. 1273, 22 p.
James, D.E., 1971, Andean crust and upper mantle structure:
Journal ofGeophysical Research, v. 76, p. 32463271.
1982, A combined O, Sr, Nd and Pb istopic and trace element
study ofcrustal contamination in central Andean lavas. I. Local
geochemical varia-tions: Earth and Planetary Science Letters, v.
57, p. 4762.
James, D.E., and Sacks, I.S., 1999, Cenozoic formation of the
central Andes:Society of Economic Geologists Special Publication 7,
p. 126.
Kamenov, G.D., 2000, Sources of metals in San Cristobal and
Pulacayo min-ing districts, southwest Bolivia: A lead isotope
study: Unpublished MS the-sis, Miami, Florida International
University, 86 p.
Kay, R.W., Sun, S.S., and Lee-Hu, C., 1978, Pb and Sr isotopes
in volcanicrocks from the Aleutian Islands and Priblof Islands,
Alaska: Geochimica etCosmochimica Acta, v. 42, p. 263273.
Kay, S.M., Mpodozis, C., and Coira, B., 1999, Neogene magmatism,
tecton-ism and mineral deposits of the Central Andes (22 to 33
latitude): Soci-ety of Economic Geology Special Publication 7, p.
2759.
Kennan, L., Lamb, S., and Rundle, C., 1995, K-Ar dates from the
Alti-plano and Cordillera Oriental of Bolivia: Implications for
Cenozoicstratigraphy and tectonics: Journal of South American Earth
Sciences, v.8, p. 163186.
Lamb, S., and Hoke, L., 1997, Origin of the high plateau in the
centralAndes, Bolivia, South America: Tectonics, v. 16, p.
623649.
Lehmann, B., 1978, A Precambrian core sample from the Altiplano,
Bolivia:Geologisches Rundschau, v. 67, p. 270278.
Lin, P.-N., 1992, Trace element and isotopic characteristics of
western Pacificpelagic sediments: Implications for the origin of
island arc melts: Geochim-ica et Cosmochimica Acta, v. 56, p.
16411654.
Litherland, M., Annells, R.N., Darbyshire, D.P.F., Fletcher, G.,
Hawkins, M.,Klinck, B. A., Mitchell, W.I., OConnor, E., Pitfield,
P.E., Power, G., andWebb, B.C., 1989, The Proterozoic of eastern
Bolivia and its relationship tothe Andean mobile belt: Precambrian
Research, v. 43, p. 157174.
Long, K.R., 1991, Mines, prospects, and mineral occurrences,
Altiplano andCordillera Occidental, Bolivia: U. S. Geological
Survey Bulletin, v. 1975,Appendix A, p. 243272.
Ludington, S., McKee, E.H., and Shew, N., 1992, K-Ar ages of
Bolivian Ter-tiary polymetallic vein deposits:, U.S. Geological
Survey Bulletin, v. 2039,p. 8793.
Lyons, W.A., 1963, Structural geology of Pulacayo mine, Bolivia:
ECONOMICGEOLOGY, v. 58, p. 978987.
Macfarlane, A.W., 1999, Isotopic studies of northern Andean
crustal evolu-tion and ore metal sources: Society of Economic
Geologists Special Publi-cation 7, p. 195218.
Macfarlane, A.W., and Petersen, U., 1990, Pb isotopes of the
Hualgayoc area,northern Per: Implications for metal provenance and
genesis of acordilleran polymetallic mining district: ECONOMIC
GEOLOGY, v. 85, p.1303-1327.
Macfarlane, A.W., Marcet, P., LeHuray, A.P. and Petersen, U.,
1990, Leadisotope provinces of central Andes inferred from ores and
crustal rocks:ECONOMIC GEOLOGY, v. 85, p. 18571880.
Macfarlane, A.W., Prol-Ledesma, R.M., and Conrad, M.D., 1994,
Isotopicand fluid inclusion studies of the geological and
hydrothermal evolution ofthe Hualgayoc district, northern Per:
International Geology Review, v. 36,p. 945977.
Manhes, G., Minister, J.F., and Allegre, C.J., 1978, Comparative
uranium-thorium-lead and rubidium-strontium study of the Saint
Severing ampho-terite; consequences for early solar system
chronology: Earth and PlanetaryScience Letters, v. 39, p. 1424.
Martin, J. M., Nirel., P., and Thomas, A.J., 1987, Sequential
extraction tech-niques: promises and problems: Marine Chemistry, v.
22, p. 313341.
Miller, D.M., Goldstein, S. L., and Langmuir, C.H., 1994,
Cerium/lead andlead isotope ratios in arc magmas and the enrichment
of lead in the conti-nents: Nature, v. 368, p. 514520.
Miller, J., and Harris, N., 1989, Evolution of continental crust
in centralAndes: Constraints from Nd isotope systematics: Geology,
v. 17, p.615617.
Mller, P., and Giese, U., 1997, Determination of easily
accessible metal frac-tions in rocks by batch leaching with acid
cation-exchange resin: ChemicalGeology, v. 137, p. 4155.
Montgomery, C.W. and Hurley, P.M., 1978, Total rock U-Pb and
Rb-Sr sys-tematics in the Imataca Series, Guyana Shield, Venezuela:
Earth and Plan-etary Science Letters, v. 39, p. 281290.
Mukasa, S.B., and Tilton, G.R., 1985, Pb isotope systematics as
a guide tocrustal involvement in the generation of the Coastal
batholith, Per, inPitcher, W.S., Atherton, M.P., Cobbing, E.J., and
Beckinsale, R.D., eds.,Magmatism at a plate edge: The Peruvian
Andes: New York, John Wylieand Sons, p. 235238.
Mukasa, S.B., Vidal, C.E., and Injoque-Espinosa, J., 1990, Pb
isotope bear-ing on the metallogenesis of sulfide ore deposits in
Central and SouthernPeru: ECONOMIC GEOLOGY, v. 85, p.
1438-1446.
Othman, D.B., White, W.W., and Patchett, J., 1989, The
geochemistry of ma-rine sediments, island arc magma genesis and
crust-mantle recycling: Earthand Planetary Science Letters, v. 94,
p. 111.
Perez, H., 1996, Mapa Geologico de Bolivia: SERGEOMIN-YPFB,
Bolivia,scale 1:1000000.
Petersen, U., Macfarlane, A. W., and Danielsen, A., 1993, Lead
isotopicprovinces in Per, Bolivia and northern Chile: 2nd
International Sympo-sium on Andean Geodynamics, Oxford University,
England, ORSTOM, p.477480.
Peucker-Ehrenbrink, B., Hofman, A.W., and Hart, S. R., 1994,
Hydrother-mal lead transfer from mantle to continental crust: the
role of metalliferoussediments: Earth and Planetary Science
Letters, v. 125, p. 129142.
Pinto-Vasquez, J., 1993, Volcanic dome-associated precious and
base metalepithermal mineralization at Pulacayo, Bolivia: ECONOMIC
GEOLOGY, v. 88,p. 697700.
Puig, A., 1988, Geologic and metallogenic significance of the
isotopic com-position of lead in galenas of the Chilean Andes:
ECONOMIC GEOLOGY, v.83, p. 843858.
Ramos, V., 1986, The birth of southern South America: American
Scientist,v. 77, p. 444450.
Richter, D.H., Brooks, E., Ludington, S., Hinjosa-Velasco, A.,
Escobar-Diaz,A., McKee, H., and Shew, N., 1991, The San Cristobal
district: U.S. Geo-logical Survey Bulletin, v. 1975, p. 153156.
Schuiling, R.D., 1967, Tin belts on the continents around the
Atlantic ocean:ECONOMY GEOLOGY, v. 62, p. 540550.
Sempere, T., 1995, Phanerozoic evolution of Bolivia and adjacent
regions:American Association of Petroleum Geologists (AAPG) Memoir,
v. 62, p.207230.
Sempere, T., Hrail, G., Oller, J., and Bonhomme, M.G., 1990,
LateOligocene-Early Miocene major tectonic crisis and related
basins in Bo-livia: Geology, v. 18, p. 946949.
Sempere, T., Butler, R.F., Richards, D.R., Marshall, L.G.,
Sharp, W., andSwisher III, C.C., 1997, Stratigraphy and chronology
of upper Cretaceous-lower Paleogene strata in Bolivia and northern
Argentina: Geological Soci-ety of America v. 109, p. 709-727.
Stacey, J.W. and Kramers, J.D., 1975, Approximation of
terrestrial lead iso-tope evolution by a two-stage model: Earth and
Planetary Science Letters,v. 26, p.207221.
Stern, C.R., 1991, Role of subduction erosion in the generation
of Andeanmagmatism: Geology, v. 22, p. 261272.
Tessier, A., Campbell, P. and Bisson, M., 1979, Sequential
extraction proce-dure for the speciation of particulate trace
metals: Analytical Chemistry, v.51, p. 844851.
Tilton, G.R., and Barreiro, B.A., 1980, Origin of lead in Andean
calcalkalinelavas, southern Per: Science, v. 210, p. 12451247.
Tilton, G.R., Pollak, R.J., Clark, A., and Robertson, R.C.,
1981, Isotopic com-position of Pb in central Andean ore deposits:
Geological Society of Amer-ica Memoir, v. 154, p. 791816.
Todt, N., Cliff, R.A., Hanser, A., and Hofman. A.W., 1984, 202Pb
and 205Pbdouble spike for lead is