Quick Search:    advanced search

GSW Home   GeoRef Home   My GSW Alerts   Contact GSW   About GSW   Journals List   Help

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Quang, C. X. Articles by Hawkes, N. Search for Related Content GeoRef GeoRef Citation

Response of Supergene Processes to Episodic Cenozoic Uplift, Pediment Erosion, and Ignimbrite Eruption in the Porphyry Copper Province of Southern Perú

Chan X. Quang, Alan H. Clark and James K. W. Lee

Department of Geological Sciences and Geological Engineering, Queen’s University, Kingston, Ontario, Canada K7L 3N6

Nicholas Hawkes

Rio Tinto Mining and Exploration Limited, Manco Cápac 551 - Lima 18, Perú

Corresponding author: e-mail, charbon{at}geol.queensu.ca

	Abstract

Top Abstract Introduction Previous Model for Landform... Geochronology of Ignimbrite... Revised Landform Model Supergene Profiles of Porphyry... Integrated Model of Landscape... Conclusions APPENDIX References

 
The Jurassic to middle Eocene porphyry copper deposits and prospects exposed on the Pacific slopes of the central Andean Cordillera Occidental of southern Perú between latitudes 16°30' and 18° S record a protracted, ca. 30-m.y. history of supergene processes that were fundamentally controlled by the evolving local geomorphologic environment, itself a response to successive regional tectonic events, including the late Eocene Incaic, the late Oligocene to earliest Miocene Aymará, and the middle to late Miocene Quechuan events. Weathering of the porphyry centers also overlapped temporally with the local resumption of arc volcanism in southern Perú at 25.5 Ma following a 27-m.y. amagmatic interval, and supergene processes were variously interrupted or terminated by ignimbrite blanketing, although in several locations supergene profiles were preserved by such cover.

The landform chronology for the area surrounding the Cuajone, Quellaveco, and Toquepala deposits (ca. 17° S) is revised and extended northwestward through field mapping to the Cerro Verde-Santa Rosa district (ca. 16° 30' S). The ⁴⁰Ar-³⁹Ar incremental-heating dates of supergene alunite group minerals from the Angostura (38.1 and 38.8 Ma) and Posco (38.8 Ma) prospects and the Cerro Verde deposit (36.1–38.8 Ma) demonstrate that supergene processes were underway in the late Eocene beneath a subplanar topography resulting from uplift and erosion during the Incaic orogeny, now represented by a regional unconformity in the Cenozoic volcanic-sedimentary rock succession. Broadly contemporaneous supergene processes were probably active in the Cuajone-Quellaveco-Toquepala district. Slow erosion and the accumulation of clastic sediments through the tectonically quiescent early to mid-Oligocene are envisaged to have caused a rise in the water table and the widespread preservation of the Incaic supergene profiles. Aymará uplift subsequently led to the incision of the 23.8 to 24 Ma Altos de Camilaca and the 18.8 to 19.1 Ma Pampa Lagunas pediplains and their regional correlatives. The ensuing water-table lowering was associated with intense leaching and sulfide enrichment from the late Oligocene (24.4–28 Ma natroalunite at Cerro Verde, 26–27 Ma natroalunite at Santa Rosa, and 28.6 Ma jarosite at La Llave) to the early Miocene (23 Ma alunite and 21 Ma natroalunite at Cerro Verde, and 19.2 Ma jarosite at La Llave) and was plausibly responsible for much of the upgrading of the Cuajone and Toquepala deposits and thr Quellaveco prospect, which are intersected by both the Altos de Camilaca pediplain and erosional features representing upslope extensions of the Pampa Lagunas pediplain. The younger supergene profiles were widely superimposed on the remnants of those generated during the Incaic orogeny. Middle Miocene (<14.2 Ma biotite age) Chuntacala Formation flows protected the Cuajone supergene profile from destruction by erosion, but at 13.0 Ma interrupted supergene processes at Quellaveco. Revision of volcano-stratigraphic relationships in the latter area reveals that subsequent erosion of the Chuntacala Formation ignimbrites and part of the supergene profile took place prior to the deposition of a 10.1 Ma ash-flow tuff of the Asana Formation. Elsewhere, supergene activity persisted at the Cachuyito prospect through 11.4 Ma, and minor jarosite development occurred at least until 4.9 Ma both there and at Cerro Verde during and following the Multiple Pediment landform stage (ca. 7.9–15.0 Ma).

The occurrence of relics of late Eocene alunite group minerals within considerably younger late Oligocene to late Miocene supergene alteration profiles suggests that the overall physiographic configuration of the Pacific piedmont of southern Perú remained remarkably consistent from the late Eocene to the middle Miocene. Moreover, the new age data confirm that, as in northern Chile, semiarid climatic conditions prevailed along much of the plate boundary from the mid-Eocene until the late Miocene or early Pliocene onset of hyperaridity.

The local geomorphologic and volcanic conditions in southern Perú, however, conspired to generate more complex supergene profiles with lower aggregate enrichment factors relative to the strongly enriched profiles in the late Eocene to early Oligocene porphyry copper belt of northern Chile, which underwent supergene upgrading over relatively brief periods.

	Introduction

Top Abstract Introduction Previous Model for Landform... Geochronology of Ignimbrite... Revised Landform Model Supergene Profiles of Porphyry... Integrated Model of Landscape... Conclusions APPENDIX References

 
THE PACIFIC SLOPE of the central Andean Cordillera Occidental (western cordillera) in southern Perú and contiguous northern Chile is, in broad terms, an enormous monocline, the product of orogenic contraction, crustal thickening, uplift, and erosion initiated in the late Paleocene (ca. 58 Ma) in response to a change in the convergence direction of the South American and Farallón plates from north and/or north-northeast to northeast (Pardo-Casas and Molnar, 1987). Contractional tectonism intensified in this region in the late-middle Eocene (ca. 41 Ma; Farrar et al., 1988; Horton et al., 2001; McQuarrie and DeCelles, 2001) and, particularly, in the late Oligocene to Miocene (Isacks, 1988; Roeder, 1988) during the development of the Bolivian orocline. Since Steinmann’s (1929) classic studies, the late Eocene to early Oligocene tectonism has generally been assigned to the Incaic orogeny and the later Cenozoic events to the Quechuan. In a recent overview, Jaillard et al. (2000, p. 545) grouped all tectonic activity from 58 to 26 Ma as the "Incaic contraction tectonic phase," but we retain the original scheme, including the use of the term Aymará or, in Chile, Pehuenchean (Hartley et al., 2000) to describe the late Oligocene uplift at the outset of the Quechuan orogeny (Sandeman et al., 1995).

From the standpoint of mineral deposits geology, the tectonic and concomitant physiographic evolution of the oceanward flank of the cordillera since ca. 58 Ma assumes importance through its influence on the supergene modification of the major upper Paleocene to middle Eocene and upper Eocene to lower Oligocene porphyry copper deposits of the region. The former constitute a narrow metallogenic subprovince (Fig. 1) extending in southern Perú from the 61 to 62 Ma (Quang et al., 2003) Cerro Verde-Santa Rosa cluster in the north (lat. 16°33’ S) to the Cuajone (52.4 Ma; Clark et al., 1990a; Clark, 2003)-Quellaveco (54.5 Ma; Minera Quellaveco S.A., unpub. data)-Toquepala (55.0–55.5 Ma; Zweng and Clark, 1995; Clark, 2003) district at ca. 17° S. An almost uninterrupted Neogene ignimbrite cover mantles the southern extension of this belt in northernmost Chile, but its probable continuity is emphasized by the 58 Ma (Quirt et al., 1971) Mocha prospect (19°49’ S) and the 51.8 Ma Cerro Colorado deposit (20°03’ S), the latter exhibiting a thick and complex supergene profile (Bouzari and Clark, 2002). In southern Perú, Clark et al. (1990a), Quang (2003), and Quang et al. (in prep.) further record the emplacement in the Cordillera de la Costa of both Middle Jurassic and Upper Cretaceous porphyry copper and associated Cu-Au vein systems, similarly affected by middle to late Cenozoic supergene modification.

View larger version (21K): [in this window] [in a new window]   FIG. 1. Locations of the Cerro Verde-Santa Rosa, Cuajone, and Toquepala mines and other porphyry copper deposits and prospects of southern Perú and northern Chile cited in the text. Geochronologic data for hypogene mineralization: 1 = Quang et al. (2003); 2 = Clark et al. (1990b), 3 = Clark (2003); 4 = Minera Quellaveco, S.A. (unpub. data), 5 = Zweng and Clark (1995) and Clark (2003); 6 = Quirt et al. (1971), 7 = Bouzari and Clark (2002), 8 = Reynolds et al. (1998); 9 = Ballard et al. (2001), 10 = Rowland and Clark (2001), 11. A.H. Clark, unpublished report to Minera Escondida Ltda. (2001), and 12 = Gustafson et al. (2001).

A faceted, multiphase, Neogene landscape, dominated by pediments, is extensively preserved under the hyperarid climatic conditions now prevailing below ca. 3,000 m a.s.l. in this segment of the Atacama-Peruvian coastal desert. The Neogene landforms are particularly well represented in the areas surrounding the valleys of the Ríos Moquegua and Locumba (Tosdal, 1978; Tosdal et al., 1984), where they constitute the llanuras costaneras, or coastal slopes. In this area, moreover, the ages of formation of many of the pediments, and hence the wider landform chronology, may be established through geochronologic study of numerous superincumbent and broadly coeval ash-flow tuff units (Tosdal et al., 1981). These relationships provide a temporal framework for interpretation of the supergene profiles in the Mesozoic and Paleogene porphyry copper centers, although the strictly geomorphologic record extends back only to the latest Oligocene (ca. 23–24 Ma), and the earlier Incaic events must therefore be deduced from tectonostratigraphic relationships and geochronology. The model relating supergene processes and landforms (Tosdal, 1978; Clark et al., 1990b) followed that originally proposed for the Copiapó mining district of the southern Atacama Desert (Fig. 1) by Clark et al. (1967a- b), Sillitoe et al. (1968), and Mortimer (1973), who inferred that episodes of leaching, oxidation, and enrichment of Jurassic to Paleogene copper deposits occurred prior to the late Miocene, under semiarid climatic conditions beneath regionally extensive pediments, and were stimulated by drastic and abrupt descents in the water table caused by uplift and resultant pediment erosion (Fig. 2).

View larger version (35K): [in this window] [in a new window]   FIG. 2. Summary of previously proposed landform and supergene chronologies in northern Chile and southern Perú. References for northern Chile: A.H. Clark, unpub. short course notes for Rio Tinto Mining and Exploration Ltd.: Mendoza (1999), Nishiizumi et al. (1998), Mortimer (1973), Sillitoe et al. (1968), and Clark et al. (1967a-b), and for southern Peru: Tosdal (1978), Tosdal et al. (1981, 1984), and Clark et al. (1990b)

	Previous Model for Landform Chronology and Supergene Profile Evolution in Southern Perú

Top Abstract Introduction Previous Model for Landform... Geochronology of Ignimbrite... Revised Landform Model Supergene Profiles of Porphyry... Integrated Model of Landscape... Conclusions APPENDIX References

 
Tosdal (1978) and Tosdal et al. (1984) defined a three-stage Oligocene to Miocene landform chronology (Figs. 2–3) for the areas surrounding the Moquegua and Locumba valleys on the basis of geomorphologic mapping and K-Ar dating of ignimbrites. The oldest (stage 1) planar landform element, the Altos de Camilaca surface, was the product of protracted erosion (Fig. 2) coinciding with major uplift associated with the Aymará event and the resumption of arc volcanism following a ca. 25-m.y. amagmatic period (Clark et al., 1990a). It comprises a regionally preserved, southwest-dipping degradational surface in the precordillera and an aggradational facies represented by the upper member of the Moquegua Formation (Fig. 4), locally preserved in the llanuras costaneras and Cordillera de la Costa. The pediplain truncated Upper Cretaceous and Paleogene rocks and structures, including the Toquepala, Quellaveco, and Cuajone porphyry deposits, and is represented by beveled, subplanar ridgelines or an undulating surface buried by earliest Miocene Huaylillas Formation ignimbrites (Fig. 4). Ignimbrites intercalated with coarse clastic units in the upper Moquegua Formation (aggradational facies) yield K-Ar biotite ages of ca. 23 Ma (Tosdal, 1978; Tosdal et al., 1981), but R.J. Langridge (in Clark et al., 1990a) recorded ⁴⁰Ar-³⁹Ar total-fusion mineral dates as old as 29 Ma, which were subsequently interpreted as evidence for a late Oligocene onset of rapid uplift and erosion (Sandeman et al., 1995). K-Ar ages for Huaylillas Formation ignimbrites covering the erosional surface have a narrow range from 18.3 to 18.4 Ma (Tosdal, 1978; Tosdal et al., 1981), inferred to provide a minimum age for stage 1 erosion (Fig. 2).

View larger version (58K): [in this window] [in a new window]   FIG. 3. Major physiographic units of the Cordillera Occidental, southern Perú, showing the locations of dated tuffs documented in earlier studies and the distribution of Tertiary landform elements within the limits of geomorphologic mapping by Tosdal (1978; simplified after Tosdal et al., 1984). Dated minerals: bi = biotite, hb = hornblende, pl = plagioclase.

View larger version (59K): [in this window] [in a new window]   FIG. 4. Late Tertiary stratigraphy of southern Perú (Tosdal et al., 1981). Solid circles indicate the approximate locations of dated volcanic rocks, the ages of which are shown on the right (in Ma). Minerals dated: bi = biotite and pl = plagioclase. The K-Ar dates are from Tosdal et al. (1981). Relative positions of Moquegua landforms are indicated in the left-hand column. AdC = Altos de Camilaca, CdlC = Cerro de las Chulpas, CS = Cerro Sagollo, PL = Pampa Lagunas, PS = Pampa Sitana.

Tosdal (1978) and, in more detail, Clark et al. (1990b) applied the landform relationships described by Tosdal et al. (1984) in developing models for the evolution of supergene profiles in the neighboring variably enriched Cuajone, Quellaveco, and Toquepala porphyry copper deposits (Fig. 3). Chalcocite development at Cuajone was inferred to have occurred during stage 1 uplift and pediplanation in the latest Oligocene and to have been terminated at 22.8 Ma when blanketed by a thick Huaylillas Formation ignimbrite flow. Supergene enrichment at Quellaveco, which is more extensive than at Cuajone, is considered to have commenced beneath the developing Altos de Camilaca surface in the earliest Miocene but to have been interrupted by the deposition of an 18.1 Ma Huaylillas Formation ignimbrite. Following the incision of an open valley at the onset of the Multiple Pediment stage, the chalcocite horizon was deepened in the middle Miocene (Clark et al., 1990b), before accumulation of Chuntacala Formation (Fig. 4) ignimbrites and lahars at 13.1 Ma. At Toquepala, a vertically extensive, multilevel enrichment profile was developed (Anderson, 1982; Clark et al., 1990b), but Neogene ignimbrites are not preserved in the mine area, precluding assignment of direct age constraints for the different supergene episodes. However, the highest altitude chalcocite zones were inferred to have formed during the late Oligocene or early Miocene below the Altos de Camilaca surface, whereas the deepest and most important chalcocite horizon was tentatively correlated with the earliest part of the Multiple Pediment stage.

Two major episodes of supergene enrichment were therefore proposed by Clark et al. (1990b) for the Toquepala-Cuajone district, one in the late Oligocene to early Miocene, associated with the Altos de Camilaca surface, and the second in the middle Miocene, associated mainly with the Pampa Lagunas pediment and hence with the earlier erosional events of the Multiple Pediment stage (Fig. 2). No geochronologic data are available for supergene minerals in these deposits.

	Geochronology of Ignimbrite Flows

Top Abstract Introduction Previous Model for Landform... Geochronology of Ignimbrite... Revised Landform Model Supergene Profiles of Porphyry... Integrated Model of Landscape... Conclusions APPENDIX References

 
In order to provide a reliable template for geomorphologic and supergene relationships farther north, refinement of the Moquegua-Locumba landform chronology was required through evaluation of the existing geochronologic database for ash-flow tuff eruption. This comprises: conventional K-Ar biotite dates determined by Tosdal (1978) and Tosdal et al. (1981), prefixed herein as SP and SPT; and ⁴⁰Ar-³⁹Ar total-fusion and step-heating dates for phenocrystic biotite, plagioclase, and hornblende, carried out by R.J. Langridge (in Clark et al., 1990a), prefixed SPAM. Fourteen previously studied mineral samples (Table 1; Figs. 3, 5) were reanalyzed by laser-induced, incremental-heating ⁴⁰Ar-³⁹Ar techniques at the Geochronological Laboratory of the Department of Geological Sciences and Geological Engineering at Queen’s University, Kingston, Ontario. The analytical data are presented in the Appendix and all ages are reported with uncertainties of ±2. Most of the new age determinations are concordant, within 2 error, with the previous dates, but some are significantly older or younger, with implications for the ages of eruption and, hence, of erosional events.

View larger version (38K): [in this window] [in a new window]   FIG. 5. 40Ar-39Ar step-heating spectra for ignimbrite samples from the Moquegua-Ataspaca transect dated in earlier studies (Table 1). Min = minimum age, PA = plateau age.

View larger version (58K): [in this window] [in a new window]   FIG. 6. Major physiographic units of the Cordillera Occidental, southern Perú, showing the locations of the newly studied ash-flow tuffs and the landform profiles in Figure 10.

View larger version (34K): [in this window] [in a new window]   FIG. 8. (a). Simplified topographic map of the Quellaveco prospect area, showing locations of dated ignimbrite flows on the northern flank of the Río Asana valley. The northwest-southeast (A-A') line indicates the approximate location of the cross section. (b). Schematic northwest-southeast cross section along the axis of the Quellaveco orebody, illustrating the interrelationships of present topography, dated Miocene ash-flow tuffs, inferred Pampa Sitana terrace pediment, and the Quellaveco sulfide enrichment zone (modified after Clark et al., 1990b). Vertical scale is exaggerated.

View larger version (28K): [in this window] [in a new window]   FIG. 7. 40Ar-39Ar step-heating spectra for newly studied ash-flow tuffs (Table 2). Min = minimum age, PA = plateau age.

Four ash-flow tuffs in the precordillera gave ages slightly older than the reported K-Ar dates (Table 1). A welded tuff (SPT-127f) that lies unconformably on the eroded Upper Cretaceous-Paleogene arc terrane gave a plateau age of 24.0 ± 0.8 Ma (Fig. 5d). This flow is assigned to the Moquegua Formation because it falls in the age range of 23.1 to 25.5 Ma determined for ignimbrite flows intercalated with coarse clastics near the top of the upper member of that formation. Three semiwelded porphyritic tuffs (SP-85, SPT-124, SPT-305, Fig. 5) from a higher stratigraphic position relative to SPT-127f gave plateau ages of 18.8 to 19.1 Ma (Table 1). These three lower Miocene flows are assigned to the Huaylillas Formation (Fig. 4), which comprises regionally extensive ignimbrite sheets that blanket the eroded Upper Cretaceous to Paleogene strata in the precordillera (Tosdal et al., 1981) and overlie a sequence of conglomerates in the llanuras costaneras mapped as the Moquegua Formation by Bellido and Landa (1965).

Sample SPT-211, a semiwelded, pumiceous, porphyritic tuff deposited in a channel incised into the tilted Moquegua Formation (Fig. 3), gave a disturbed age spectrum (Fig. 5h) that shows partial resetting. The high atmospheric argon to radiogenic argon ratio (Fig. 5i) reflects the chloritization apparent in some of the biotite grains. Elsewhere, an unconsolidated tuff (SPT-330, Fig. 3) intercalated with alluvial gravels yielded a disturbed age spectrum showing recoil effects (Fig. 5j) for biotite with interlaminated chlorite.

Four middle Miocene tuffs intercalated with coarse gravels deposited in channels incised into the Moquegua and Huaylillas Formations (Tosdal et al., 1981) also gave ⁴⁰Ar-³⁹Ar ages that are slightly older than the previously reported K-Ar dates (Table 1). Sample SP-65, a semiwelded porphyritic tuff exposed ca. 5 km east of Quellaveco, gave a plateau age of 14.8 ± 0.5 Ma (Fig. 5k), whereas a slightly welded, pumiceous, porphyritic tuff (SP-73) in Quebrada Chuntacala gave a plateau age of 14.2 ± 0.2 Ma (Fig. 5l). These areally restricted Chuntacala Formation ignimbrites (Fig. 4) from the type locality define a narrow age range of 14.2 to 14.8 Ma. Two unconsolidated, porphyritic tuffs (SPT-323 and SPT 324) located, respectively, in a valley incised into Moquegua Formation sediments and intercalated with alluvial gravels on Pampa Sitana (Fig. 3), yielded identical ca. 10 Ma plateau ages (Fig. 5m-n). An unconsolidated tuff (SPT 308) yielded an integrated age of 9.6 ± 0.5 Ma from a disturbed age spectrum (Fig. 5o) for a biotite separate with inclusions of apatite and zircon and fragments of hornblende, plausibly responsible for the increased Ca/K ratios in the highest temperature steps (Fig. 5p). In addition, the ratio of atmospheric argon to radiogenic argon is significantly higher than for other samples. Thus, the 8.9 Ma K-Ar date reported by Tosdal et al. (1981) for this sample is considered to be unreliable.

View larger version (36K): [in this window] [in a new window]   FIG. 10. Schematic cross sections, showing the landform profiles for the selected areas of the Pacific slope between Moquegua and La Joya (see Fig. 6). (a). Moquegua valley. (b). Yaral prospect area. (c). Pampa Guaneros. (d). Desierto de Clemesí. (e). Pampa Colorada. (f). Cocachacra. (g). Pampa de La Joya. Pediments are assigned to stages I through V on the basis of the chronology established for the Moquegua area (Fig. 9). Actual or projected locations of mines and prospects are shown.

A semiwelded porphyritic tuff (S-16b) ca. 6 km east of Moquegua gave a four-step plateau age of 23.8 ± 0.5 Ma (Fig. 7a). The other four samples from this area are from a sequence of ignimbrites and conglomerates overlying the leached zone of the Quellaveco prospect (Fig. 8). An unwelded, pumiceous, porphyritic tuff (S-40) from near the base of the thick volcanic sequence that overlies the northern half of the deposit (Fig. 8b) yielded a seven-step plateau age of 13.0 ± 0.4 Ma (Fig. 7b). A five-step plateau age of 10.1 ± 0.3 Ma (Fig. 7c) was obtained from an unconsolidated ash-flow tuff (S-38: the "white tuff" of Clark et al., 1990b), exposed ca. 40 m lower than S-40 (Fig. 8b) and discontinuously overlying the Quellaveco Conglomerate, a local accumulation dominated by hematitic, leached cobbles.

Two unwelded, pumiceous, porphyritic tuffs (S-37 and S-41), probably representing a single flow that is discontinuously exposed (Fig. 8), gave identical plateau ages of 7.9 Ma (Fig. 7d-e). This volcanic flow is assigned to the Sencca Formation of Tosdal et al. (1981; Fig. 4).

Farther to the northwest, an unwelded, unconsolidated tuff (S-30) exposed in an arroyo between the Desierto de Clemesí and Quebrada Guaneros (Fig. 6) yielded a four-step plateau age of 10.5 ± 0.2 Ma (Fig. 7f) and is thus assigned to the Asana Formation (Fig. 4).

Three tuffs (S-2, S-59, S-60) from the wider La Joya area (Fig. 6) gave identical ca. 5 Ma plateau ages (Fig. 7g-i) and are assigned to the La Joya Formation (Fig. 4).

A revised stratigraphic column for the Oligocene to Pliocene volcanic and sedimentary units in littoral southern Perú is summarized in Figure 4 (cf. fig. 2 of Tosdal et al., 1981).

	Revised Landform Model

Top Abstract Introduction Previous Model for Landform... Geochronology of Ignimbrite... Revised Landform Model Supergene Profiles of Porphyry... Integrated Model of Landscape... Conclusions APPENDIX References

 
Moquegua-Locumba area
The new ⁴⁰Ar-³⁹Ar data indicate that, following a protracted middle Tertiary period of quiescence, ignimbrite eruption was initiated along the arc front of southern Perú at ca. 25.5 Ma (cf. Clark et al., 1990a; Sandeman et al., 1995). These uppermost Oligocene tuffs are generally intercalated with coarse clastic sediments of the upper member of the Moquegua Formation (Fig. 4; Tosdal et al., 1981). Resumption of arc magmatism, therefore, closely followed an episode of tectonic uplift that produced the slight angular unconformity between the uniformly bedded, fine- to medium-grained siltstones and sandstones of the lower member of the Moquegua Formation and the conglomeratic sandstones of the upper member (Fig. 4).

The landform chronology for littoral southern Perú requires revision in the light of these data and the improved understanding of the stratigraphic relationships. The proposed modifications mainly concern the earlier erosional events and are shown in Figure 9. Whereas Tosdal (1978) interpreted the Huaylillas Formation ignimbrites (i.e., SPT-124 and SPT-305) as providing a minimum age of 18.3 to 18.4 Ma (K-Ar dates; Table 1) for development of the Altos de Camilaca surface (I), the age of this major erosional feature is now temporally constrained by coeval dates of 23.8 Ma (S-16b) and 24.0 Ma (SPT-127f) for tuffs assigned to the Moquegua Formation (Table 1). The relationships between the Altos de Camilaca surface (I) and the Pampa Lagunas (II) and younger surfaces are shown in Figure 10a, in which ignimbrites (SP-85, SPT-124, and SPT-305) of the Huaylillas Formation (Fig. 4) dated at 18.8 to 19.1 Ma are correlated with erosion resulting in the development of the Pampa Lagunas surface (II; Fig. 10a). In contrast, Tosdal et al. (1984) inferred an age range of 11.2 to 14.2 Ma for the Pampa Lagunas erosional event on the basis of two K-Ar dates (SPT-211 and SPT-308, Table 1). Our analyses of biotite from these tuffs revealed disturbed age spectra (Fig. 5h, i, o, p) which indicate that neither date is reliable.

View larger version (23K): [in this window] [in a new window]   FIG. 9. Comparison of the revised landform chronology with that proposed by Tosdal et al. (1984).

View larger version (114K): [in this window] [in a new window]   FIG. 11. (a). Panorama looking south from Cerro Baúl. The Pampa Lagunas pediplain (PL) forms the prominent surface stretching across the llanuras costaneras, whereas the precordillera (on the extreme left and beyond the photo) is dominated by the Altos de Camilaca (AdC) pediplain. A discontinuous, residual bench just below the Pampa Lagunas pediplan represents the Cerro de las Chulpas (DdlC) pediment, which is better preserved in the lower llanuras costaneras. The younger Pampa Sitana (PS) pediment forms dissected ridge spurs below the Pampa Lagunas pediplain, and benches just above the Moquegua valley are remnants of the Cerro Sagollo (CS) pediment. The summits of the Cordillera de la Costa, scarcely visible in the far-right background, are remnants of the Altos de Camilaca (AdC) surface incised by the Pampa Lagunas pediplain. (b). Panorama looking southeast across Quebrada Guaneros (Fig. 6). (Pampa de Congas pediment = PdC; Pampa Guaneros pediment = PG) (c). Panorama looking northwest across the La Llave prospect. (La Caldera surface = LC; Pampa de La Joya = PLJ).

The revised landform chronology for the wider Moquegua region shown in Figures 9 and 10a emphasizes the elimination of an extended early Miocene hiatus between the Altos de Camilaca surface (I) and the Multiple Pediment stage (cf. Tosdal et al., 1984).

Ilo, Clemesí, Tambo, and La Joya areas
The erosional landforms of the Moquegua area have been traced continuously through field mapping southwest to the Ilo area, and northwestward across the Desierto de Clemesí and the Río Tambo valley to the La Joya area (Fig. 6). Geomorphologic relationships in this region are based on the relative positions of local landforms with respect to the mapped continuous surfaces and are illustrated in a series of cross sections (Fig. 10), the locations of which are shown in Figure 6. In each area, planar landforms are given local geographic names, which can be correlated with corresponding features in the Moquegua area (Table 3; Figs. 10a, 11a). The ca. 4500-km² region covered by Figure 12 experienced a consistent history of erosional events. For ease of comparison, the erosional features are assigned to stages I through V, corresponding to the Altos de Camilaca, Pampa Lagunas, Cerro de las Chulpas, Pampa Sitana, and Cerro Sagollo surfaces in the Moquegua area.

View larger version (109K): [in this window] [in a new window]   FIG. 12. Geomorphologic map, showing the distribution of Tertiary landforms in the study area. The precordillera and llanuras constaneras are dominated by, respectively, the Altos de Camilaca (dark gray) and Pampa Lagunas (light gray) pediplains and their correlatives (numbers in brackets refer to stages in Fig. 9). In contrast, the areally restricted landforms of the Multiple Pediment stage (Fig. 2), comprising the Cerro de las Chulpas, Pampa Sitana, and Cerro Sagollo pediments and their correlatives, are concentrated along throughgoing rivers. The relationships between the landforms of the Multiple Pediment stage and remnants of the Altos de Camilaca and Pampa Lagunas pediplains in the Moquegua-Pampa Guanero transect are shown in greater detail in an inset at the top right of the figure, whereas those in the area around the Cachuyo and Cachuyito prospects, north-northeast of Cocachacra, are shown in an inset at the top center.

The Cerro Sagollo surface extends west from the Moquegua valley to the base of Cerros Homo, the accordant summits of which are remnants of the Altos de Camilaca surface (Figs. 10c, 11b; Table 3). A prominent bench below the summit of Cerros Homo, immediately above the Cerro Sagollo surface, corresponds to the Pampa Sitana surface (Figs. 10c, 11b). The Pampa Guaneros and Pampa de Congas surfaces (Fig. 10c) are correlated, respectively, with the Cerro Sagollo and Pampa Sitana surfaces (Fig. 10a). A 10.5 ± 0.2 Ma ash flow (S-30) intercalated with clastic sediments in the aggradational facies of the Pampa de Congas surface demonstrates that erosion was active at that time (Fig. 10c).

A southwest-northeast cross section (Fig. 10d) of the Clemesí area is dominated by the well-preserved Desierto de Clemesí surface which, in agreement with Laharie (1976), we correlate with the Pampa Lagunas surface (Fig. 10a; cf. Tosdal et al., 1984). Remnants of the older Altos de Camilaca surface occur in the northeastern sector of the region and in the Cordillera de la Costa (Fig. 10d). The younger Quebrada Chayanto surface (Fig. 10d) is correlated with the Cerro de las Chulpas surface (Fig. 10a).

An east-west cross section (Fig. 10e) illustrates the landform relationships farther to the northwest, between the Clemesí area and the Río Tambo valley. The Pampa Colorada surface (Fig. 10e) is correlated with the Pampa Sitana surface (Fig. 10a). The Cerro de las Chulpas surface is here represented by eroded remnants (Fig. 10e). The Pampa Pacheco and Pampa Yamayo surfaces (Fig. 10e) are both correlated with the Cerro Sagollo surface (Fig. 10a). The summit of Cerro Buena Vista (Fig. 10e-f) is interpreted as a remnant of the Pampa Lagunas surface and is incised by benches that are correlated with the Pampa Sitana surface (Fig. 10a).

A west-northwest–east-southeast cross section (Fig. 10f) shows the landform relationships in the area between the western bank of the Río Tambo valley and the La Joya area. The rounded summit of Cerro Bronce (Fig. 10f) is correlated with the Altos de Camilaca surface and the areally extensive Pampa de La Joya surface (Fig. 10f) with the Pampa Lagunas surface (Fig. 10a). A minimum age for the Pampa de La Joya surface is delimited by three 4.9 to 5.0 Ma ash-flow tuffs (S-2, S-59, and S-60, Table 2). The La Caldera surface (Figs. 10g, 11c), represented by accordant summits in the precordillera (e.g., Cerro Verde, Cerro Negro, Cerro San Martín), as well as eroded remnants (e.g., Cerro Nieves) standing above the Pampa de La Joya surface, is correlated with the Altos de Camilaca surface (Fig. 10a).

Geomorphologic mapping therefore reveals that, as in the Moquegua-Locumba transect, upper Oligocene and lower Miocene surfaces are the dominant landform elements in the Ilo, Clemesí, Tambo, and La Joya areas (Fig. 12). The two older, regionally extensive, planar landforms, assigned to stages I and II, may thus be classified as pediplains, and it is evident that the most intense erosional dissection of the Pacific slope of the Cordillera Occidental was initiated at 25 to 26 Ma and completed by 18 Ma. Subsequently, erosion was focused around the exorheic river systems (Fig. 12). We therefore recommend the restriction of the Multiple Pediment stage (Tosdal et al., 1984) to the Cerro de las Chulpas (III), Pampa Sitana (IV), and Cerro Sagollo (V) pediments and their correlatives to the northwest (Table 3). The revised landform chronology (Fig. 9) and new geomorphologic map (Fig. 12) provide a context in which to examine the age and evolution of supergene processes in littoral southern Perú.

	Supergene Profiles of Porphyry Copper Deposits and Prospects

Top Abstract Introduction Previous Model for Landform... Geochronology of Ignimbrite... Revised Landform Model Supergene Profiles of Porphyry... Integrated Model of Landscape... Conclusions APPENDIX References

 
Supergene profiles in the Paleogene Cerro Verde-Santa Rosa and Cuajone-Quellaveco-Toquepala porphyry districts, as well as in the various Mesozoic copper prospects, were examined in order to document the age and extent of leaching and chalcocite development and to determine the local geomorphologic settings of supergene activity (Figs. 13–15). The results of the age determinations are summarized in Table 4 and Figure 13; full analytical data are listed in the Appendix. Descriptions of the supergene profiles along with ⁴⁰Ar-³⁹Ar age constraints are summarized here from northwest to southeast in Figure 15.

View larger version (55K): [in this window] [in a new window]   FIG. 13. Map showing the locations of dated supergene alunite group samples from Cu deposits and prospects in southern Perú (Table 4).

View larger version (63K): [in this window] [in a new window]   FIG. 15. Simplified schematic cross sections, showing the geomorphologic setting and age of supergene mineralization in selected profiles from the porphyry copper belt of southern Perú. Toquepala and Cerro Verde exhibit complex profiles with multiple chalcocite horizons, reflecting polystage histories of leaching, oxidation, and enrichment. Alunite group mineral dates in the Cerro Verde deposit and Angostura and Posco prospects indicate that supergene activity began in the late Eocene beneath a regional, subplanar topography. Late Oligocene supergene activity is recorded in the Cerro Verde-Santa Rosa district and the La Llave prospect beneath the La Caldera surface, a correlative of the 24 Ma Altos de Camilaca surface, which is considered to have controlled late Oligocene leaching and enrichment in the Cuajone-Toquepala district. Deepening of supergene profiles during the early Miocene is considered to have been controlled by the Pampa Lagunas pediplain and its correlatives. The development of the Multiple Pediment stage eroded preexisting profiles over wide areas but was responsible for continued supergene activity at the Cerro Verde and Toquepala deposits, prior to the onset of hyperaridity in the late Miocene.

Prospects from the La Joya to the Ilo area
Six porphyry-style centers that exhibit leached and oxidized profiles have recently been assessed in the La Joya, Río Tambo, and Ilo areas (Fig. 15).

Leached assemblages rich in hematite and with minor copper oxide and jarosite at the 68 Ma Angostura prospect are truncated by the Pampa de La Joya surface (Figs. 13, 15). Supergene jarosite (S-96) and natroalunite (S-98) from Angostura gave disturbed age spectra interpreted as evidence for leaching at 38.1 to 38.9 Ma (Fig. 14a-b).

View larger version (24K): [in this window] [in a new window]   FIG. 14. 40Ar-39Ar step-heating spectra for supergene alunite group minerals (Table 4). Min = minimum age, PA = plateau age, TPA = "technical plateau" age (comprising 3 or more steps but less than 50% of total 39Ar released).

Supergene mineralization in the pyrite-poor, 166 Ma, Cachuyo porphyry copper prospect (Fig. 13) comprises copper and iron oxides that extend vertically for over 100 m and overlie a weak chalcocite horizon (Fig. 15; Andrews et al., 2000). The oxide zone underlies a goethite-hematite leached zone preserved below a remnant of the Altos de Camilaca surface (Cerro Bronce, Fig. 10f). The Pampa de La Joya surface truncates both the mineralized oxide and leached zones (Fig. 15).

Five kilometers east-northeast of Cachuyo, copper oxides associated with a weakly mineralized stockwork in biotitized andesite (Liassic Chocolate Formation) crop out on the Pampa Colorada surface at the 160 Ma Cachuyito prospect (Fig. 13). Supergene natrojarosite (S-12) and jarosite (S-13) from a leached outcrop ca. 500 m to the southeast gave 11.4 and 4.9 Ma dates (Fig. 14f-g). The 11.4 Ma natrojarosite veins are interpreted to have formed beneath a correlative of the Cerro de las Chulpas surface in Figure 10e (e.g., Lomillas del Toro surface, Fig. 15). The 4.9 Ma jarosite veinlets are interpreted as recording local oxidation during the Valley and Terrace stage (VI).

The hematite-goethite leached zone of the undated Posco prospect (Fig. 13), exposed below the Pampa Colorada surface, exhibits abundant hematite-coated fractures and quartz stockwork with an alteration assemblage of quartz-sericite-clay and relict disseminated pyrite. Supergene alunite veins from the leached capping at Posco gave a plateau age of 38.8 Ma (Fig. 14c).

Supergene copper oxide mineralization in the 106 Ma Yaral stockwork of the Ilo area is incised by the Pampa Sitana surface, whereas a higher elevation, hematite-dominated, leached zone associated with minor copper oxides occurs below the Pampa Lagunas surface (Figs. 13, 15).

Cuajone-Quellaveco-Toquepala district
The models proposed by Clark et al. (1990b) relating supergene profile evolution in the Cuajone-Quellaveco-Toquepala district to local landform development and the extent of Neogene ignimbrite cover are reproduced in Figure 15, but with age revisions resulting from the new ⁴⁰Ar-³⁹Ar ignimbrite dates.

Sulfide enrichment at Cuajone was apparently confined to a single gently southwest-dipping blanket with an average thickness of ca. 20 m and an average grade of 1.5 percent Cu, underlain by a transitional zone characterized by coatings of sooty chalcocite on hypogene sulfides (Manrique and Plazolles, 1975; Clark et al., 1990b; A.H. Clark, unpub. report for Southern Perú Copper Corporation, 2001). The chalcocite zone passes upward into an oxide zone, which is itself overlain by a hematitic leached zone. Widespread remnants of supergene sulfides in the oxide and leached zones are evidence for the oxidation and leaching of preexisting chalcocite. The 23.8 to 24.0 Ma Altos de Camilaca surface truncates the leached zone on the southern benches of the Cuajone open pit (Fig. 15). Subsequently, renewed erosion incised the Altos de Camilaca surface, removing ignimbrites, the leached zone, and the upper part of the oxide zone. This episode of uplift and erosion was associated with the 18.8 to 19.1 Ma Pampa Lagunas pediment (Fig. 15). Development of a broad valley, correlated with the Cerro de las Chulpas surface, eroded the overlying ignimbrites and truncated the leached, oxide, and chalcocite enrichment zones in the northern part of the deposit. Deposition in this valley of the middle Miocene Chuntacala Formation ignimbrite prevented further erosion of the supergene profile (Fig. 15).

Supergene sulfide enrichment at Quellaveco reaches depths of 250 to 300 m below the present surface (Clark et al., 1990b). An irregular chalcocite blanket with a troughlike cross section and a thickness of 50 to 60 m comprises a thicker, upper zone of moderate to strong enrichment overlying a thinner, lower grade zone representing a downward transition to unaltered hypogene assemblages (cf. fig. 12 in Clark et al., 1990b). The enrichment blanket is overlain by a hematite-bearing leached-oxidized zone, extending from the surface to depths of up to 80 m. Middle Miocene Chuntacala Formation ignimbrites were deposited on a correlative of the Cerro de las Chulpas surface that had incised the leached zone (Figs. 8, 15), effectively terminating supergene activity. This sequence is beveled by a valley bench (Fig. 8), correlated with the Pampa Sitana surface, and overlain by an upper Miocene unwelded white tuff (S-40; Fig. 15). Incision of the supergene profile during the development of the presently active Asana valley, correlated with the Valley and Terrace stage, is inferred to have taken place after the deposition of a 7.9 Ma ignimbrite (Fig. 15).

The Toquepala deposit exhibits the most extensive record of supergene sulfide enrichment in the region, with well-defined hematitic and minor jarositic leached zones separating chalcocite horizons (Fig. 15; Clark et al., 1990b). The absence of ignimbrites in the vicinity of Toquepala probably contributed to the development of a thicker supergene profile but precluded direct age constraints for the landforms that controlled subjacent supergene activity. The highest occurrence of relict supergene chalcocite, however, is documented at the 3,535-m level, 40 to 70 m below the accordant summits of the precordillera, correlated with the Altos de Camilaca surface (I).

	Integrated Model of Landscape Evolution and Supergene Development

Top Abstract Introduction Previous Model for Landform... Geochronology of Ignimbrite... Revised Landform Model Supergene Profiles of Porphyry... Integrated Model of Landscape... Conclusions APPENDIX References

 
Late Eocene to earliest Oligocene
Our new ⁴⁰Ar-³⁹Ar age data for supergene alunite group minerals demonstrate that weak supergene alteration at some of the Mesozoic and Paleogene mineralized hydrothermal systems in littoral southern Perú took place prior to or during the late Eocene Incaic event, which resulted in widespread uplift, folding, and erosion in the (proto-) Cordillera Occidental (Clark et al., 1990a-b). This orogenic event, temporally constrained in south-central Perú (41–45 Ma; Noble et al., 1979; McKee and Noble, 1982) and northern Chile (38.5 Ma; Hammerschmidt et al., 1992), affected the entire central Andes (Jaillard et al., 2000). In the study area, Tosdal et al. (1984) inferred a late Eocene age for the sub-Moquegua Formation unconformity (Fig. 4) exposed in the Moquegua valley. ⁴⁰Ar-³⁹Ar dates for supergene minerals from the Cerro Verde porphyry copper deposit (36.1–38.8 Ma for alunite from 2,648 m a.s.l.; Quang et al., 2003) and from the Angostura (38.1 Ma for jarosite and 38.8 Ma for alunite from 1,780 m a.s.l.) and Posco (38.8 Ma for alunite from 773 m a.s.l.) prospects, now confirm that supergene processes had commenced by the late Eocene, almost certainly in response to the Incaic uplift (Fig. 16). Late Eocene ages for alunite and jarosite veins from Cerro Verde and Angostura exposed more than 250 m below the oldest preserved upper Oligocene pediplain, the La Caldera surface, record deep oxidation, presumably controlled by permeable structures (Figs. 15–16). In contrast, the upper Eocene alunite zone at the lower elevation Posco prospect (Fig. 16) is preserved below an unconformity between Mesozoic volcanic and intrusive rocks and Tertiary sedimentary cover, correlated with the sub-Moquegua Formation unconformity, and is itself truncated by the Pampa Pacheco surface (V; Fig. 15). The occurrence of these similar ages over a vertical interval of more than 1,800 m may indicate that a highly irregular topography controlled late Eocene supergene processes or, more probably, results from monoclinal tilting of the Pacific piedmont of southern Perú since the formation of the alunite group minerals. However, the preservation of these Eocene alunite group minerals at essentially the same elevations as considerably younger supergene alteration profiles suggests that the overall configuration of the piedmont has changed a little since the Eocene, despite a protracted history of later Tertiary uplift and arc deformation.

View larger version (21K): [in this window] [in a new window]   FIG. 16. Schematic northeast-southwest profile of the Pacific slope in the porphyry copper belt of southern Perú, showing the relative positions of the documented landform surfaces: the Altos de Camilaca (I) and Pampa Lagunas (II) pediplains and equivalents, the Cerro de las Chulpas (III), Pampa Sitana (IV), and Cerro Sagollo (V) pediments and equivalents. The landform chronology is constrained by 40Ar-39Ar dates for Tertiary ignimbrite ash flows, and the age of supergene mineralization is summarized by 40Ar-39Ar dates for alunite group minerals for mines (solid black square) or deposits and prospects (light gray square).

The above geochronologic database establishes a late Eocene to early Oligocene episode of leaching and enrichment that affected Mesozoic to middle Eocene deposits throughout the central-central Andes. The Incaic landscape and, presumably, much of its associated enrichment have been destroyed by subsequent uplift and erosion, as in the Cerro Verde-Santa Rosa and the Cuajone-Toquepala districts (Figs. 15–16). As argued by Bouzari and Clark (2002), semiarid conditions and deep water tables must have been imposed by 40 to 45 Ma, at least between 16° and 27° S. Such a climate would also have favored pediment erosion in an uplifting terrane. An even earlier episode of supergene activity was proposed by Clark et al. (1967a) and Sillitoe et al. (1968), who argued that enrichment of Mesozoic and Paleocene copper deposits in the Copiapó area (Fig. 1) took place beneath a postulated "Cumbre" (summit) surface. Sillitoe (1969) noted that thick enrichment zones are developed only in deposits now located 1 km or more from erosional remnants of the 53 Ma ignimbrites (Cerro La Peineta Formation) and inferred that their eruption terminated the enrichment. However, remapping by Arévalo et al. (1994) demonstrated that the Eocene ash flows were deposited on downdropped caldera floors (Hornitos hemigraben) and that the landforms below the ignimbrite do not therefore represent remnants of a regional erosion surface.

Late Oligocene
The tectonically quiescent post-Incaic interval (ca. 28–38 Ma; Sandeman et al., 1995) was presumably dominated by slow relief reduction as the Cretaceous-Paleogene arcs were further eroded to generate the forearc molassic sediments of the lower member of the Moquegua Formation in southern Perú (Marocco and Noblet, 1990). The mid-late Oligocene Aymará (ca. 25–28 Ma) tectonic event, widely recognized throughout the central Andes (Tosdal et al., 1984; Sébrier et al., 1988; Sébrier and Soler, 1991; Sandeman et al., 1995), was responsible for crustal thickening, uplift, and erosion across broad areas of southern Perú. In the study area, this is correlated (Fig. 4) with the unconformity between fine shales and sandstones of the lower member of the Moquegua Formation and the upper member comprising polymictic conglomerates interbedded with coarse sandstone (Bellido, 1979; Marocco and Noblet, 1990; this study) and coincided precisely with resumption of arc activity at 25.5 Ma after a hiatus of more than 20 m.y. These events overlap in age with, and may have been related to, oblique subduction ^<