IGNEOUS PETROLOGY of the COLOSA GOLD-RICH PORPHYRY SYSTEM (TOLIMA, COLOMBIA) by Javier Gil-Rodríguez a Prepublication Manuscrip

Home , Chiles (volcano), Nevado del Huila, Nevado del Tolima, Romeral (volcano), Tolima Department

IGNEOUS PETROLOGY OF THE COLOSA GOLD-RICH PORPHYRY SYSTEM (TOLIMA, COLOMBIA)

Javier Gil-Rodríguez

A Prepublication Manuscript Submitted to the Faculty of the

DEPARTMENT OF GEOSCIENCES

In Partial Fulfillment of the Requirements for the Degree of

PROFESSIONAL SCIENCE MASTER

In the Graduate College THE UNIVERSITY OF ARIZONA 2010

Igneous Petrology of the Colosa Gold-Rich Porphyry System (Tolima, Colombia)

JAVIER GIL-RODRIGUEZ Department of Geosciences, University of Arizona, 1040 E. Fourth Street, Tucson Arizona, 85721-0077 AngloGold Ashanti Colombia S. A., Calle 116 # 7-15, Piso 8, Bogotá, Colombia [email protected], [email protected]

ABSTRACT

This work reports a petrologic study of the Colosa porphyry gold deposit based on its petrographic, geochemical, and isotopic features. The Colosa porphyry gold deposit is a large new discovery (Inferred Mineral Resource of 470 Mtonnes @ ~0.9 g/t Au) and a new member of this relatively poorly known class of porphyry deposits. This porphyry deposit (~8 Ma) is located in the Miocene calc-alkaline volcano-plutonic arc of the Central Cordillera of Colombia, and it consists of hypabyssal rocks of intermediate composition (andesine + hornblende + quartz ± orthoclase ± biotite) that intruded Paleozoic low-grade metamorphic rocks of the Cajamarca Complex.

The early intrusions of the system are fine- to medium-grained diorites and porphyry diorites with 46- 54 percent phenocrysts, and variable amounts of potassic and sodic-calcic alteration. The intermineral units are fine- to medium-grained diorite porphyries with 34-46 percent phenocrysts, and intermediate argillic, and propylitic alteration. The late intrusive bodies are fine- to coarse-grained diorite, quartz diorite, and dacite porphyries with 42-57 percent phenocrysts, and local propylitic and sericitic alteration.

Whole-rock analyses confirm a medium- to high-K calc-alkaline tendency for the rocks from Colosa. These rock exhibit an enrichment of LILE and LREE relative to HFSE and HREE, respectively. Present isotopic compositions of 143Nd/144Nd (0.512759-0.512870) and 87Sr/86Sr (0.704265-0.704510) from different units of Colosa indicate a source derived from the mantle.

The model proposed for Colosa consists of fluids from the dehydration of the subducted Nazca plate and fluids and melts from the subducted sediments that generated partial melting of the mantle wedge. These basaltic melts ascended to the mantle-crust boundary where they were retained due to density differences and began to produce processes of melting, assimilation, storage, and homogenization (MASH zone). At this depth (~35-50 km), differentiation processes began to produce more felsic magmas that were able to ascend through the crust and be emplaced at the boundary between the continental basement and the Cajamarca Complex (~15-20). At this site, the basaltic andesite magma began to produce more differentiated products (andesitic and dacitic magmas) that ascended across the Cajamarca Complex and were emplaced at depths of ~3-4 km.

INTRODUCTION

Gold-rich porphyries have been described and studied since the 1970s (Kesler, 1973; Sillitoe, 1979), being arbitrary defined by Sillitoe (1979) as those porphyry copper deposits that contain over 0.4 g/t Au; examples such as Bajo de la Alumbrera (Ulrich and Heinrich, 2002; Proffett, 2003), Bingham (Lanier et al., 1978; Moore, 1978; Cunningham et al., 2004), and Grasberg (MacDonald and Arnold, 1994; Pollard et al., 2005) are well known. On the other hand, porphyry gold systems were not reported before the 1990s (Vila et al., 1991; Vila and Sillitoe, 1991), and have been considered to be the end-member in the continuum of porphyry systems from Cu-Au, through Au-Cu, to the Au-rich and Cu-poor counterpart (Sillitoe, 1979; Sillitoe, 2000; Seedorff et al., 2005). Porphyry gold systems are not a very common type of gold deposit. Deposits assigned to this group include a few porphyries in the Chilean Maricunga belt (Vila and Sillitoe, 1991; Vila et al., 1991), and some prospects in the western United States (Canby et al., 1993; Fig. 1). These porphyries are mostly associated with the intermediate and mafic end of the broad range of rock compositions (55 to 78 wt % SiO2) characteristic of all porphyry systems (Seedorff et al., 2005).

The Colosa porphyry, located in the middle part of the Cordillera Central of Colombia (Fig. 1), is a new member of the porphyry Au class of Seedorff et al. (2005). This porphyry is part of the Miocene calc- alkaline volcano-plutonic arc of Colombia (Sillitoe, 2008), and consists of diorites, quartz diorites, and dacites. The Colosa system was built by early, intermineral, and late intrusions, each one carrying different amounts of gold. The early intrusions exhibit high-temperature potassic alteration and high grades of gold; the intermineral units have lower grades of gold than the early intrusions and a medium- temperature intermediate argillic alteration; and the late porphyries have a low-temperature propylitic alteration and are almost barren of gold. These intrusions formed a cluster of igneous bodies that intruded into Paleozoic low-grade metamorphic rocks of the Cajamarca Complex (Maya, 1992) and are surrounded by narrow zones of hornfels and breccia.

All the porphyry gold systems discovered in the world show similar gold-grades, rock compositions, and alteration types (Sillitoe, 2000; Seedorff et al., 2005; Sillitoe, 2008). The late Miocene age (Sillitoe, 2008), metamorphic wall-rock type, pyrrhotite mineralization, and isolated location in the northern Andes, makes the Colosa prospect distinctive from other existing porphyry gold systems. As a new discovery, the geology of the Colosa porphyry deposit is just beginning to be described and studied.

This work describes petrologic aspects of the Colosa porphyry deposit, emphasizing its petrographic, geochemical, and isotopic characteristics. This study indicates that the magma that formed the Colosa intrusions had a large mantle contribution and less from the crust; the magma was formed by melting of the mantle wedge above the subducted Nazca Plate and then ascended through the South American plate, 3 having a low crustal contamination, to be finally emplaced into the shallow crust in form of small intrusions with different textures and compositions and carrying anomalous contents of gold.

BACKGROUND

Previous Work

The Colosa prospect was discovered in 2006 by greenfields exploration programs of AngloGold Ashanti in Colombia (Fig. 1). Sillitoe (2008) was the first to mention the existence of this prospect, including it in the Middle Cauca belt in the Central Cordillera of Colombia. Two theses in preparation also mention Colosa porphyry: Leal-Mejía (in preparation) and García-Bernal (in preparation); the first relates Colosa to the metallogenesis of Colombia, and the second makes a comparison between Colosa and the porphyry gold systems of the Maricunga belt.

The Colosa porphyry shows similar characteristics to the Lobo, Marte, Verde, Pancho, and Cavancha (La Pepa) porphyry gold deposits of the Maricunga belt in Chile (Vila and Sillitoe, 1991; Vila et al., 1991; Muntean and Einaudi, 2000, 2001; Table 1). These systems belong to the calc-alkaline volcano-plutonic arc of Miocene age in the Andes (Fig. 1) and share similar rock compositions (diorites, quartz diorites, and dacites) with porphyritic textures. The alteration is also very similar in the porphyries of both locations, being mainly potassic with stockworks of quartz veinlets but with strong overprinting of intermediate argillic alteration (chlorite-sericite-clay) in some systems of the Maricunga belt (Vila and Sillitoe, 1991; Vila et al., 1991; Muntean and Einaudi, 2000, 2001). The presence of a propylitic halo surrounding the high and mid-temperature alteration zones is also common for all these porphyry systems.

The Zule prospect in Sierra County, California (Fig. 1), also has rocks of intermediate composition with a potassic (feldspar) alteration with quartz veinlets and magnetite. This potassic core is surrounded by an argillic halo, and this one in turn by a propylitic zone (Canby et al., 1993).

Although the Colosa porphyry shows similarities with the rest of the porphyry gold systems in the world, it also exhibits certain differences (Table 1). Colosa has a late Miocene age (8.3 Ma), compared with the early (24-23 Ma) and mid-Miocene (14-13 Ma) ages of the porphyries of the Maricunga belt (Sillitoe et al., 1991; Muntean and Einaudi, 2001), and the early Pliocene age (4.4 Ma) of Zule (Canby et al., 1993). Another important feature is the wall-rock type; Colosa has metamorphic wall rocks (schists and quartzites) but the porphyries of the Maricunga belt and Zule have volcanic wall rocks (Vila and Sillitoe, 1991; Canby et al., 1993). The isolated location of Colosa in the Northern Andes and the

4 presence of pyrrhotite in its mineralization assemblage are also distinctive features of this porphyry gold system (Table 1).

The tonnage and gold grade are variable in the Maricunga belt, such as ~80 Mt at 1.6 g/t for Lobo (Vila and Sillitoe, 1991), 46 Mt at 1.43 g/t for Marte (Vila et al., 1991), a mineable reserve of 101 Mt at 1.02 g/t for Verde (Brown and Rayment, 1991), an inferred mineral resource of 68 Mt at 0.96 g/t for Pancho (Brown and Rayment, 1991), and a mineral resource of 187 Mt at 0.56 g/t for La Pepa (Yamana Gold, 2009). The Colosa porphyry has inferred resources of 468.8 Mt at 0.86 g/t (AngloGold Ashanti, 2008) which makes it the porphyry gold system with the biggest tonnage of material and contained Au discovered to date (Fig. 2).

Regional Geologic Setting

The Central Cordillera of Colombia has a complex history of collision, accretion, faulting, magmatism and subduction (McCourt et al. 1984; Aspden et al., 1987; Restrepo and Toussaint, 1988; Taboada et al., 2000; Kennan and Pindell, 2009); its litho-tectonic and morpho-structural expression is a result of the Meso-Cenozoic Northern Andean orogeny and of orogenic events that predate Northern Andean-phase orogenic activity (Cediel et al., 2003; Fig. 3). At present, the Central Cordillera is bound on the east by the Magdalena River valley, and on the west by the Cauca River valley (Cauca-Patía graben; Aspden et al., 1987; Taboada et al., 2000). The Romeral fault system, with high-angle NNW- to NNE- trending reverse faults with a right-lateral strike-slip component, divides the cordillera into two domains, an eastern continental and a western oceanic domain (McCourt et al., 1984; McDonald et al., 1996; Taboada et al., 2000).

The eastern flank of the cordillera, included in the Central Continental Sub-Plate Realm of Cediel & Caceres (2000), consists of pre-Mesozoic polymetamorphic basement intruded by Mesozoic batholiths and stocks related to subduction (Taboada et al., 2000). During the early Paleozoic, the suprascrustal sequences of the western margin of the South American plate underwent Cordilleran-type orogenic deformation and regional metamorphism, as evidenced by the fragments of ophiolite and accretionary prism exposed in the Cajamarca-Valdivia terrane (Cediel et al., 2003). This terrane was sutured to the Guiana Shield of South America along the trace of the paleo-Palestina fault system (McCourt et al., 1984; Cediel et al., 2003; Kennan and Pindell, 2009). The Bolivar aulacogen, from the Late Paleozoic to the Early Cretaceous (Cediel and Caceres, 2000), developed as an intercontinental rift with deposition of marine and continental strata and also was affected by the emplacement of many intrusions of the same age (Aspden et al., 1987; Cediel et al., 2003). The opening of the Valle Alto rift during the Early Cretaceous facilitated the invasion of the epicontinental seaway that resulted in the deposition of marine and continental sequences of variable thicknesses over extensive areas of the Central Continental Sub- 5

Plate Realm; the extensional regime finished by the Aptian-Albian, followed by a compressive (transpressive) regime up to the Recent (Cediel et al., 2003).

The western part of the cordillera, included in the Western Tectonic Realm of Cediel et al. (2003), was formed by the accretion of many oceanic terranes. In the mid-Cretaceous the Farallon and South American plates reorganized and changed their drift direction and velocity. The resulting Meso-Cenozoic oblique collisions, subduction and obduction, and the birth of new oceanic plates (Caribbean and Nazca- Cocos system; Pindell and Kennan, 2001, 2009; Kennan and Pindell, 2009), are some of the features that characterize the Northern Andean orogeny (Cediel et al., 2003). Accretional episodes of many oceanic terranes have occurred since mid-Cretaceous along the Romeral-Cauca (McCourt et al., 1984; McDonald et al., 1996), Garrapatas-Dabeiba (Restrepo and Toussaint, 1988; Cediel et al., 2003), and Buenaventura (Cediel et al., 2003) fault zones. These terranes formed the western part of the Central Cordillera and the Western Cordillera and have been affected by periods of uplift, erosion, and subduction-related calc- alkaline magmatism (Cediel et al., 2003).

An eastward migration of the magmatic focus to the Central Cordillera in the Miocene (Toussaint and Restrepo, 1982; Restrepo et al., 1985; Restrepo and Toussaint, 1990; Ordóñez and Pimentel, 2001; Cediel et al., 2003) and the continued faulting of the late Miocene intrusions and disruption of their associated Cu-Au-Ag-Zn (Mo, Pb) mineralization, are also observed (Cediel et al., 2003). Important stress relief occurred in the late Miocene-Pliocene transpressive pop-up of the Eastern Cordillera (Cediel et al., 2003).

Local Geologic Setting

The Colosa porphyry system is located over the middle part of the Central Cordillera of Colombia (over 3000 m.a.s.l.), at 8 km northwest of the town of Cajamarca and at 30 km at the west of Ibagué, the capital of the Tolima Department (Fig. 3). Sillitoe (2008) included Colosa in the Middle Cauca belt of the Central Cordillera of Colombia, and assigned it a late Miocene age. This porphyry system intruded the metamorphic rocks of the Cajamarca Complex and, at present, is partially covered by dacitic pyroclastic material of the Cerro Machín volcano.

The Cajamarca Complex (Maya and González, 1995) is a suite of low-grade metamorphic rocks that formed the core of the Central Cordillera and that was initially named by Nelson (1962) as the Cajamarca Group. The unit crops out at the Cajamarca-Alto La Línea section and is composed of quartz-sericite schists, green schists, phyllites, quartzites, and marbles (Maya and González, 1995); a marine volcano- sedimentary protolith for the unit has been proposed by Nunez (2001). Different authors have dated the unit by the K-Ar method (Restrepo y Toussaint, 1978; Núñez et al., 1979; Millward et al., 1982; McCourt et al., 1984), obtaining ages from Paleozoic to Paleogene; whereby the oldest ages could correspond to

6 the regional metamorphic event and the youngest ages to superimposed dynamothermal events (McCourt et al., 1984).

The Cerro Machín volcano is located in the eastern part of the Central Cordillera, 7 km northeast of Cajamarca. This volcano belongs to the Ruiz-Tolima Massif (Thouret et al., 1995) or Complex (Nuñez, 2001) and is characterized by highly explosive eruptions; thick sequences (up to 400 m) of pyroclastic material from this volcano cover many kilometers around it (Thouret et al., 1995; Rueda et al., 2005). The Holocene volcanic material is usually ash and lapilli of dacitic composition (Thouret et al., 1995; Rueda et al., 2005; Dominguez et al., 2003) that belongs to the calc-alkaline trend (Dominguez et al., 2003).

Many subvolcanic bodies have been mapped near Cajamarca and in the entire Tolima Department, grouped under the denomination of Hypabyssal Bodies and assigned a Neogene age (Pulido, 1988). These intrusions have a rounded to elliptic form in map view and an irregular distribution; they intrude Precambrian to Jurassic units and are of intermediate composition (andesine + quartz + hornblende ± biotite) with aphanitic and porphyritic textures. Some of these hypabyssal units are related to gold occurrences (Pulido, 1988; Núñez, 2001).

The Colosa porphyry has been divided into early, intermineral, and late units based on the texture and composition of the rocks, type and intensity of the alteration, and Au grade (Fig. 4). Each of the geologic units is generally known by an acronym, e.g., NDE1, as defined below. Features such as chilled zones and displacement and density of veinlets have been very useful to separate similar units; such observations have been made during core logging by geologists of AngloGold Ashanti and have been used to distinguish between units. The presence of intrusive breccias (NBXE1, NBXE2, and NBXI) is also important in the system; they occupy a significant volume of the porphyry and also have important gold contents.

The early units consist of six dioritic bodies with high Au grades and a pervasive potassic alteration (biotite ± potassic feldspar) and patchy sodic-calcic alteration (actinolite + albite). The earliest unit (NDE1) is a fine to medium-grained diorite, followed by a fine- to medium-grained diorite breccia (NBXE1), and two fine- to coarse-grained diorite porphyries (NDE2 and NDE3). The intermineral units are four dioritic bodies with moderate Au grades and weak intermediate argillic (sericite + chlorite + illite) and propylitic alteration (chlorite + epidote ± calcite), which in certain cases are superimposed to the high-temperature potassic alteration. These diorites (NDI1, NDI2, NBXI, and NDI3) are fine- to medium-grained porphyries that differ slightly in grain size, whereas NBXI is an intrusive breccia. The late units are almost barren of gold and include porphyries of quartz dioritic composition (NDA and NQD; NDA is mapped as a dacite porphyry), and dioritic to quartz dioritic dikes (NDQ and NDL; NDQ is mapped as a quartz diorite dike); both groups have weak to moderate sericitic and propylitic alteration. 7

The Colosa fault is the main structure of the zone and has a N18-20°W trend with steep dips to the northeast (Pulido, 1988). This fault is located southeast of the porphyry system, and its trace is followed by Colosa Creek. Small faults with a NW-SE trend seem to be related to the Colosa fault and are cutting some of the intrusions. Other small inferred faults with the same trend also could be also offsetting the porphyry (Fig. 4).

After twelve months of exploration drilling at Colosa, AngloGold Ashanti (2008) announced the completion of an initial JORC-compliant resource estimate, defining a porphyry style mineralization at a >0.3 g/t Au-grade extending over a strike length in excess of 1,500 m (north-south) and a width of 600 m (east-west). The Inferred Resource estimated was 468.8 Mt of material at a grade of 0.86 g/t Au, for a total of 12.9 M contained oz Au.

To date, 59 holes have been drilled at the prospect, totaling 18,000 m. This drilling program has been developed on the western side of the prospect. Material remains to be tested material both along strike to the north and south, as well as the eastern part of the project.

METHODS

About eighty-six polished thin sections were studied for the petrographic characterization of the different units that form the Colosa porphyry system. Sections that showed a very low intensity of alteration were used for petrographic classification of rocks based on the modal concentration of primary minerals (Streckeisen, 1976); this process was developed by the counting 500 points per each thin section, and alteration minerals were attributed to the primary mineral from which they formed.

Whole-rock analyses were made of nineteen samples (with weak alteration) at the laboratories of Actlabs in Canada using a Thermo Jarrell Ash Enviro II simultaneous and sequential ICP with a detection limit between 0.001 and 0.01% for major elements, 0.002 and 0.05 ppm for REE, and 0.01 to 20 ppm for other trace elements.

In the same laboratory, five samples were analyzed for isotopic composition of 87Sr/86Sr and 143Nd/144Nd using a Triton multi-collector Mass-Spectrometer (TIMS) in static mode. Rubidium and strontium were separated using conventional cation-exchange techniques. During the period of work, the weighted average of 15 SRM-987 Sr-standard runs yielded 0.710257±07 (2s) for 87Sr/86Sr. Samarium and neodymium were separated by extraction-chromatography on HDEHP covered teflon powder. The ratios of 143Nd/144Nd are reported relative to a value of 0.511860 for the La Jolla standard.

A Cameca SX50 Electron Microprobe equipped with four wavelength dispersive spectrometers capable of full quantitative analysis of all elements from Be to U, located at the University of Arizona, was used for the measurement of the chemical composition of selected amphiboles and feldspars in different polished thin sections. Analyses were made at 15.0 kV accelerating voltage, using a beam current of 20 nA and a spot size of 5 μm for amphiboles; a beam current of 15.0 nA was used for feldspars. The detection limit is lower than 0.1% for all elements analyzed.

IGNEOUS PETROGRAPHY

The igneous petrography of the different units of the Colosa porphyry was based mainly on those samples that showed weak alteration (Table 2). A total of twenty-one thin sections (out of eighty-six) were used for point counting. The modal analyses were classified according to the QAP triangle for plutonic and silica-saturated rocks of Streckeisen (1976; Fig. 5). The composition of plagioclase was estimated using the extinction angle of the albite twins on eight different crystals of each thin section (Michel- Levy’s method).

Early Units

The early units consist of six dioritic bodies that contain high Au grades (NDE1, NBXE1, NDE2, NBXE2, NDE3, and NDE4) and that exhibit pervasive potassic alteration (biotite ± potassic feldspar) and patchy sodic-calcic alteration (actinolite + albite). Ore minerals (pyrite, magnetite, chalcopyrite, molybdenite, and pyrrhotite) are disseminated in the rock and in many types of veinlets, including those of the A, B, M, and EB-types (e.g., Sillitoe, 2000; Seedorff et al., 2005).

The youngest body of the entire system is the NDE1 unit; this intrusive is a fine- to medium-grained diorite (0.1-4.1mm) with hypidiomorphic texture formed by subhedral to euhedral crystals of plagioclase, amphibole, and quartz; nonetheless, orthoclase is mainly anhedral (Fig. 6A). Very fine crystals of apatite, zircon, sphene, biotite, and ilmenite are present as trace minerals in the rock. Locally a slightly porphyritic texture is found as well as poikilitic textures between plagioclase and amphibole. Plagioclase is mainly andesine (An30 to An52, averaging An42) and is slightly altered to kaolinite, sericite, and epidote. The amphibole is moderately altered to chlorite, biotite, and locally to sphene. Quartz is mainly fine- grained, anhedral, and sometimes interstitial. Orthoclase is fine-grained, interstitial, and is very weakly altered to sericite.

The second unit of the system is NBXE1, an intrusive breccia with fragments of NDE1. The matrix of this rock is a very fine- to medium-grained diorite (0.05-4.0mm) with anhedral to subhedral crystals of 9 plagioclase, amphibole, orthoclase, and quartz. Apatite, zircon, sphene, and ilmenite are found as trace minerals. The plagioclase is andesine (An33-An40, averaging An35); it is moderately altered to kaolinite, biotite, chlorite after biotite, and epidote. Amphiboles are strongly altered to chlorite, actinolite, sphene, and biotite.

The NDE2 unit is a diorite porphyry with a very fine-grained allotriomorphic matrix (0.01-0.1mm) composed by anhedral to subhedral crystals of plagioclase, amphibole, and quartz (Fig. 6B); euhedral crystals of apatite and zircon are also present in the matrix. The phenocrysts (46-54% of the rock) are fine to coarse-grained (0.5-7mm) plagioclase, amphibole, and quartz crystals with euhedral to subhedral shape. The plagioclase is mainly andesine (An26-An52, averaging An40), and is very weakly altered to sericite, kaolinite, biotite, chlorite, and epidote. Amphiboles are strongly altered to biotite, and less to chlorite (after biotite) and actinolite. Quartz crystals have rounded faces (resorbed). Poikilitic textures are found between plagioclase and amphibole, and the rock presents seriate texture from matrix to phenocrysts. Fine crystals of primary biotite are also present in phenocrysts.

A second intrusive breccia unit is NBXE2 that contains fragments of NDE2. The matrix of this unit is a porphyry diorite with a very fine-grained allotriomorphic matrix (0.01-0.1mm) composed of anhedral crystals of plagioclase, amphibole, and quartz. Apatite, zircon, and sphene are present in the matrix as trace minerals. The phenocryst phases comprise about 47 percent of this rock and consist of fine to coarse subhedral crystals of plagioclase and amphibole. The plagioclase is andesine (An38-An50, averaging An45), and is moderately altered to kaolinite, sericite, biotite, and epidote. Amphibole is strongly altered to biotite, chlorite, and sphene. A seriate texture is observed in this rock from matrix to phenocrysts.

The NDE3 unit is a diorite porphyry with a very fine-grained allotriomorphic matrix (0.01-0.1mm) formed by anhedral to subhedral crystals of plagioclase, amphibole, and quartz (Fig. 6C). Euhedral crystals of apatite are found in the matrix, as well as zircon and sphene as trace minerals. Phenocrysts occupy 46 to 53 percent of the rock and consist in fine- to medium-grained (0.2-3.5mm) plagioclase and amphibole with subhedral to euhedral shape. The plagioclase type is mainly andesine (An35-An52, averaging An44) and is very weakly altered to sericite, biotite, and epidote. The amphibole phenocrysts have been almost entirely destroyed by the alteration, and the amphibole of the matrix is replaced by biotite and chlorite. This rock presents a seriate texture from matrix to phenocryst size.

The last of the early units of the system is the NDE4 diorite porphyry that has a very fine-grained and allotriomorphic matrix (0.01-0.1mm) composed of anhedral crystals of plagioclase, amphibole, and quartz. Apatite, zircon, sphene, and ilmenite are found as trace minerals in the matrix. Phenocrysts form 50 percent of the rock and consist of fine- to medium-grained (0.2-2.0mm) plagioclase and amphibole with subhedral to euhedral shape. Plagioclase is mainly andesine (An41-An60, averaging An48), and is 10 moderately altered to biotite, kaolinite, and sericite. Amphibole is strongly altered to biotite, chlorite, and calcite. A seriate texture from matrix to phenocryst size is observed, as well as a poikilitic texture locally between plagioclase and amphibole.

Intermineral Units

The intermineral units are four dioritic bodies (NDI1, NDI2, NBXI, and NDI3) with medium Au grades and a weak intermediate argillic (sericite + chlorite + illite), and propylitic (chlorite + epidote ± calcite) alteration that in certain places overprints higher temperature types of alteration. The mineralization is mainly pyrite ± chalcopyrite ± pyrrhotite, occurring in veinlets and as disseminations.

The youngest intermineral unit is NDI1, a diorite porphyry with a very fine-grained and allotriomorphic matrix (0.01-0.1mm) composed of anhedral to subhedral crystals of plagioclase, amphibole, and quartz (Fig. 6D). Apatite, zircon, sphene, and ilmenite are present in the matrix as trace minerals. The phenocryst phases (34-39% of the rock) are euhedral to subhedral crystals of plagioclase, amphibole, and quartz of fine to medium grain size (0.2-5.0mm). The plagioclase of this unit is andesine

(An37-An50, averaging An45) and is very weakly altered to sericite and epidote. Amphibole crystals are moderately altered to actinolite, biotite, chlorite, and sphene. Quartz phenocrysts have rounded faces (resorbed). A seriate texture is observed in this rock from matrix to phenocryst size; also a poikilitic texture is observed locally between plagioclase and amphibole.

The NDI2 unit is a diorite porphyry with a very fine-grained allotriomorphic matrix (0.01-0.1mm) formed by anhedral to subhedral crystals of plagioclase, amphibole, and quartz (Fig. 6E); there are also crystals of apatite, zircon, and sphene in the matrix. Fine- to medium-grained (0.2-4.5mm) phenocrysts of plagioclase, amphibole, and quartz are present with subhedral to euhedral shapes; they occupy 38 to 39 percent of the rock. The plagioclase is mainly andesine (An32-An57, averaging An44) and is very weakly altered to kaolinite, sericite, biotite, and epidote. Amphibole is weakly altered to biotite, chlorite, epidote, sphene, and calcite. Quartz phenocrysts have rounded faces (resorbed). Poikilitic textures are found locally between plagioclase and amphibole, and the rock has a seriate texture from matrix to phenocryst size.

The intrusive breccia NBXI has fragments of NDI1. The matrix of this unit is a diorite porphyry with a very fine- to fine-grained allotriomorphic matrix composed of anhedral crystals of plagioclase, amphibole, and quartz; apatite, sphene, zircon and ilmenite are also present in the matrix (Fig. 6F). The phenocrysts (46 percent of the rock) are fine- to coarse-grained (0.2-6.0mm) plagioclase, amphibole, and quartz crystals with subhedral to euhedral shape. Plagioclase is mainly andesine (An40-An54, averaging at

An47) and is weakly altered to kaolinite, sericite, biotite, and epidote. Amphibole is weakly altered to 11 chlorite, biotite, sphene, epidote, and actinolite. The rock has a seriate texture, and locally, a poikilitic texture between plagioclase and amphibole.

The last intermineral unit is the NDI3 diorite porphyry that has a very fine-grained and allotriomorphic matrix (0.01-0.1mm) formed by anhedral to subhedral crystals of plagioclase, amphibole, quartz, and biotite. Apatite and zircon are found also in the matrix as trace minerals. The phenocryst phase (41% of the rock) is composed of plagioclase, amphibole, quartz, and biotite of fine to medium size (0.2-

3mm) with subhedral to anhedral shapes. Plagioclase is mainly andesine (An27-An47, averaging An36); plagioclase is weakly altered to kaolinite, sericite, and epidote. Amphibole is entirely altered to chlorite and sphene. Quartz phenocrysts have rounded faces (resorbed). Biotite is moderately altered to chlorite and sphene. The rock has a seriate texture from matrix to phenocryst size.

Late Units

The late units are almost barren in gold and include porphyry intrusions of quartz dioritic composition (NDA and NQD) and dioritic to quartz dioritic dikes (NDQ and NDL); both groups have weak to moderate sericitic and propylitic (chlorite + epidote ± calcite) alteration. Pyrite is disseminated through the rocks and also in veinlets.

The NDA unit is the biggest body in the porphyry system (diameter of ~1.6 km), consisting of a quartz diorite porphyry with a very fine-grained allotriomorphic matrix (0.01-0.1mm) composed by anhedral to subhedral crystals of plagioclase, amphibole, biotite, and quartz (Fig. 6G). Euhedral crystals of apatite are also present in the matrix as well as zircon, sphene, and ilmenite as trace minerals. The phenocrysts (54-57% of the rock) are plagioclase, quartz, biotite, and amphibole, fine to coarse grained

(0.2-6mm) and with subhedral to euhedral shapes. Plagioclase is andesine (An30-An40, averaging An36) and is weakly altered to kaolinite, sericite, and epidote. Amphibole is strongly altered to chlorite, calcite, sphene, epidote, and biotite. Quartz phenocrysts have rounded faces (resorbed). A seriate texture is present from matrix to phenocrysts, and poikilitic textures are common between plagioclase and amphibole. This rock grades to tonalite with the increase of quartz content as observed macroscopically in drill core; nevertheless, the rock has historically been termed dacite porphyry.

The NQD unit is located at the east of the porphyry system, and it is a quartz diorite porphyry with a very fine-grained (0.01-0.1mm) allotriomorphic matrix composed of anhedral crystals of plagioclase, amphibole, and quartz. Subhedral crystals of sphene, zircon, apatite, and ilmenite are also found in the matrix as trace minerals. Phenocrysts of plagioclase, amphibole, and quartz are present (46% of the rock), and have fine to coarse sizes (0.2-6mm) and euhedral to subhedral shapes. Plagioclase is andesine (An29-

An43, averaging An34), and is moderately altered to sericite and sphene. Amphibole is strongly altered to 12 chlorite, sericite, and sphene. Quartz phenocrysts have rounded faces (resorbed). A seriate texture is observed from matrix to phenocrysts, and locally, plagioclase phenocrysts enclose fine crystals of amphibole (poikilitic texture).

The NDQ unit is a late, northwest-trending dike. This rock is a diorite porphyry with a very fine- grained and allotriomorphic matrix (0.01-0.1mm) composed of anhedral to subhedral crysts of plagioclase, amphibole, quartz, and biotite (Fig. 6I). Euhedral apatite crystals are also found in the matrix, as well as zircons, sphene, and ilmenite as trace minerals. Phenocrysts of plagioclase, amphibole, quartz, and biotite are present (42-47% of the rock), having fine to coarse sizes (0.2-6mm) and euhedral to subhedral shapes. Plagioclase is andesine in composition the type (An30-An46, averaging An36), and is very weakly altered to epidote. Amphibole is weakly altered to chlorite, epidote, sphene, and biotite. Quartz phenocrysts have rounded faces (resorbed). A seriate texture is observed from matrix to phenocryst size, and locally, plagioclase phenocrysts enclose fine crystals of amphibole (poikilitic texture; Fig. 6H). This rock grades to quartz diorite with the increase of quartz content as seen macroscopically in many drill cores; the project geologists refer to this unit as a quartz diorite.

The NDL dike is located at the southern end of the porphyry system (Fig 4). This unit is a diorite porphyry with a very fine-grained matrix (0.01-0.1mm), and is composed of plagioclase, amphibole, and quartz crystals of subhedral to anhedral shape (Fig. 6J). Apatite, zircon, sphene, and ilmenite are also found in the matrix as trace minerals. The phenocryst phases (45% of the rock) are formed by fine to medium-grained (0.2-2mm) plagioclase and amphibole crystals of subhedral to euhedral shape. The plagioclase is mainly andesine (An41-An54, averaging An49), and is very weakly altered to epidote. Amphibole is very weakly altered to epidote and sphene. The rock has a seriate texture from matrix to phenocryst size.

GEOCHEMISTRY

The least altered rocks were geochemically analyzed to characterize the compositions of the various igneous units of the Colosa porphyry system and to distinguish geochemical patterns through time.

Whole-Rock Analyses

Weakly altered rocks of the main lithologic units of the Colosa porphyry were selected for whole- rock analyses. Prior to generating the diagrams, major element oxides were normalized to 100% on an anhydrous basis, and the normative mineralogy was calculated on an anhydrous basis using the GCDkit software (Janousek et al., 2006; Table 3). 13

Based on the content of major elements, a geochemical classification of the samples was made from the TAS diagram (Wilson, 1989; Fig. 7A) and the R1-R2 plot (De la Roche et al., 1980; Fig. 7B). In these diagrams, the early and intermineral units are classified mainly as diorites (weakly differentiated rocks), and the late porphyries and dikes are classified as quartz diorites, granodiorites, and tonalites; these classifications are petrographically reflected in the higher content of quartz and biotite of the late units (more strongly differentiated rocks). The TAS diagram is also useful to confirm the subalkaline tendency of rocks at Colosa (Fig. 7A).

Harker variation diagrams that plot major element oxides against SiO2 show the evolution of different elements within the magmatic system of Colosa (Fig. 8). The distribution of the samples in the Harker diagrams is generally scattered, especially for Al2O3 and K2O, but clear trends are observed for the other oxides. A negative correlation between SiO2 and P2O5, MgO, CaO, TiO2, and FeOt is clear. These oxides are incorporated in the structure of mafic minerals (olivine, pyroxene, and calcic plagioclase) during the first stages of crystallization; in the case of Colosa, these oxides are being incorporated in the structure of early plagioclase (most calcic), amphibole, and apatite. For this reason, the highest concentrations of these oxides are found in the least differentiated rocks (early and intermineral units), and the lowest values in the most differentiated rocks of the system (late units).

On the other hand, a clear positive correlation is observed between SiO2 and Na2O, and, though not as clear, perhaps also for Al2O3 and K2O (Fig. 8); these oxides are incorporated in felsic minerals (K-feldspar and sodic plagioclase) that crystallize during the later magmatic stages.

Major and trace elements are used to establish the geotectonic setting of Colosa. The AFM (Irvine and

Baragar, 1971) and SiO2 vs K2O (Peccerillo and Taylor, 1976) plots indicate the calc-alkaline tendency (medium- to high-K) of the porphyry system within the subalkaline trend (Fig. 9A and B).

The incompatible element diagram of Pearce et al. (1984) and Gorton and Shandl (2000) are consistent with the geologic observation that the Colosa porphyry system formed in a volcanic arc at an active continental margin (Fig. 9C and D). This information is consistent with the spider diagrams, where the negative anomalies of Nb, Ta, and Ti are typical for rocks from active continental margins related to subduction processes (Wilson, 1989; Rollinson, 1993; Fig. 8A). All samples show a similar pattern concerning the incompatible elements, relatively rich in mobile elements (LIL) and depleted in immobile elements (HFS).

All samples also have the same REE pattern, relatively rich in LREE and poor in HREE, where the most differentiated rocks (late porphyries and late dikes) have the lowest values (Fig. 10B). The fractionation degree for the different units can be estimated with the ratio of LREE/HREE, for which the

14 ratios of La/Yb and Ce/Yb are commonly used (Table 3). The La/Yb ratio for all the units of Colosa ranges between 5 and 20, and varies between 9-12 for early units; 5-12 for intermineral units; 16-20 for late porphyries; and 11-17 for late dikes. The Ce/Yb for Colosa is about 9-38, ranging between 17-22 for early units; 9-22 for intermineral units; 29-38 for late porphyries; and 22-30 for late dikes.

Another important ratio is the Ba/La ratio, which is used to measure the fractionation of LILE respect to HFSE (including REE) and could indicate insights into the magma source (Table 3). This ratio ranges between 11 and 78 for all rocks of Colosa, and between 17-76 for early units, 28-58 for intermineral units, 37-77 for late porphyries, and 11-78 for late dikes.

Isotopic Analyses

Isotopic analyses of 143Nd/144Nd and 87Sr/86Sr were made on some of the lithologic units of the Colosa Porphyry (NDE1, NDI2, NDI3, NDA, and NQD). All samples have similar present values of both neodymium (0.512759-0.512870) and strontium (0.704265-0.704510) isotopic compositions. All isotopic ratios are located in the mantle array field (Table 4; Fig. 11).

Microprobe Analyses

The electron microprobe was used to analyze silicate minerals, especially feldspars and amphiboles, to measure composition of these minerals as the porphyry system evolved (Table 5 and Table 6). These analyses also revealed the presence of very fine orthoclase and quartz in the matrix of most units (Fig. 12A-B).

Reverse zoning is observed within plagioclase grains of different units, such that a calcic plagioclase rims grew around cores of more sodic composition (Fig. 12B).

Amphiboles were classified based on the 13eCNK normalization of Schumacher (1997) and the amphibole classification diagrams of Leake et al. (1997). All amphiboles are calcic amphiboles and are mostly iron-rich respect to magnesium. They are classified mainly as ferropargasites, ferro-edenites, and ferrotschermakites (Fig. 13); only one amphibole of the NDI2 unit was classified as pargasite (relatively magnesium-rich respect to iron).

Thermobarometry

The chemical compositions of amphiboles were also used to estimate the conditions of temperature and pressure of crystallization following Ernst and Liu (1998). This process uses the contents of Al2O3 and TiO2 to estimate pressure and temperature, respectively (Fig. 14).

The entire Colosa system has estimated crystallization pressures that range between 0.1 and 2.1 GPa and temperatures between 630 and 880 °C. Ranges of temperature and pressure were also established for different units of the Colosa porphyry, where the early units vary between 0.5-1.5 GPa and 680-840 °C; intermineral units vary between 0.3-2.1 GPa and 630-850 °C; late porphyries between 0.4-0.8 GPa and 660-760 °C; NDQ late dike between 0.4-1.2 GPA and 680-820 °C; and NDL late dike between 0.1-0.5 GPa and 790-880 °C. Late units show very clear patterns of decreasing temperature with decreasing pressure during crystallization (Fig. 14).

Three main stages of crystallization are inferred from pressure, lithology, and temperature differences of groups of analyses (Fig. 14). Stage I has a pressure range of 1.3-2.1 GPa and temperatures of 630-780 °C, where an important episode of crystallization occurred at ~1.4 GPa (~40-50 km deep). Stage II has pressures of 0.3-1.2 GPa, temperatures of 660-850 °C, and important crystallization episode at ~0.5 GPa (~15-20 km deep). Stage III has pressure of 0.1-0.5 GPa, temperatures of 790-880 °C, and the shallowest pressures of crystallization at ~0.1 GPa (~3-4km deep).

IGNEOUS PETROLOGY

Source of the magma

As indicated by the Nd and Sr isotopic compositions, the main source of the magma for the Colosa porphyry is the mantle. The dehydration of the subducted Nazca plate and probably of the subducted sediments produced fluids that ascended through the mantle wedge causing partial melting and formation of a magma (Fig. 15).

The primary mantle wedge-derived magmas formed in subduction settings are commonly basaltic in composition (Gill, 1981; Thorpe et al., 1984; Wilson, 1989; Best, 2003), although generation of more siliceous magmas from the metasomatized mantle wedge remains a distinct possibility (Wilson, 1989). Crustal rocks can, therefore, serve as an effective density filter blocking ascent of denser mafic, mantle- derived magmas (Green, 1982; Best, 2003). Probably a zone of melting, assimilation, storage, and homogenization (MASH) was formed in the lowermost crust or mantle-crust transition, where basaltic magmas that ascended from the mantle wedge become neutrally buoyant (Thorpe et al., 1984; Hildreth and Moorbath, 1988). Crystal fractionation of such magmas, combined with crustal contamination at deep levels, can then account for the spectrum of more evolved rock types observed at shallow depths (Wilson, 1989), as seen at Colosa.

The most and least differentiated rocks of Colosa show a very narrow compositional trend in isotope 16 ratios and are always located in the mantle array (Fig. 11). This argues that there has been just a small degree of crustal contamination that could have slightly increased the Sr content and the 87Sr/86Sr ratio (Best, 2003), but not enough to relocate the rocks in the crust field. High ratios of Ba/La, as those of Colosa (11-78), could also suggest the presence of crustal contamination in the melt (Hildreth and Moorbath, 1988), although some authors (Kay, 1980; McBirney, 2007) relate these ratios to the presence of subducted-sediment melts in the magma. The high Rb/Cs ratios (23-160) from Colosa’s rocks also could indicate contamination from crust rather than from subducted sediments (Hildreth and Moorbath, 1988).

Some authors (Gill, 1981; Harmon et al., 1981; Thorpe et al., 1984) suggest some degree of influx of components from the subducted oceanic crust and lack of a significant crustal component for the genesis of Andean andesites. The contribution of LIL elements is thought to come from dehydration of the subducted slab that would also initiate the mantle partial melting process (Hawkesworth et al., 1979).

The negative anomalies of Nb, Ta, and Ti found in the rocks of Colosa are typical of arc-related rocks and could be explained by retention of a refractory phase in the source in which these elements are highly compatible and where LIL elements are incompatible (Wilson, 1989; Pearce and Peate, 1995; Best, 2003); fractional crystallization of HFSE minerals could also occur within the lower crust causing the depletion of such elements in the melt (Thorpe et al., 1980). Hildreth and Moorbath (1988) suggest that there are large crustal HFSE contributions and that their source is dominantly intracrustal rather than subduction- derived.

The high-K content of some of the rocks of Colosa could be explained by crustal contamination, a process that occurs in volcanic arcs when continental crust tends to be thick (Hildreth and Moorbath, 1988). Nevertheless, Kay (1980) and Gill (1981) suggests that the K component of arc magmas corresponds to recycling of altered oceanic crust and subducted sediments.

Crystal fractionation of phenocryst phases from basalt, usually of plagioclase + orthopyroxene/olivine + augite + magnetite (POAM), is by far the most common and extensive process, supplemented to an unknown extent by magma mixing, selective interaction with the crust, and vapor fractionation (Gill, 1981). However, the normal contents of Sr and Eu as shown in the spider and REE diagrams of Colosa (Fig. 10) could suggest a lack of or low content of plagioclase in the source and/or minimal fractionation of this mineral in early crystallization stages (Best, 2003). The presence of H2O in melts could decrease the pressure and temperature conditions of formation of plagioclase and increase the crystallization field of amphibole (Gill, 1981; Green, 1982); this is supported by the important presence of amphibole (5.1- 37.7%) in all rocks from Colosa. The presence of plagioclase as the main phenocryst phase at Colosa indicates that the fractionation of this mineral is important at least in the latter crystallization stages of the 17 system; this is also supported by the common presence of poikilitic textures, where plagioclase grains enclose amphiboles crystals, and not the opposite.

Partial melts of most crustal rocks would be rhyolitic to dacitic in composition and would normally raise LREE contents and Ce/Yb ratios of basaltic magmas assimilating them (Hildreth and Moorbath, 1988). In addition, as amphibole has partition coefficients of ~1.5-3.0 for HREE in melts of basaltic andesite to andesite composition (Rollinson, 1993), amphibole fractionation is seemingly a feasible HREE suppressant (Gill, 1981). The presence of amphibole phenocrysts as the main mafic phase through all the units of Colosa, especially in late and intermineral units, could support this idea for explaining the depletion of HREE in the rocks (Fig. 10B). Garnet is not observed in any of the samples, but its presence in the early stages of crystallization could have been another important factor in the depletion of HREE (Hildreth and Moorbath, 1988; Best, 2003).

Evolution of the system

Similar patterns of incompatible elements between all Colosa units, as seen in the spider and REE diagrams (Fig. 10), suggest that the same magma chamber generated them. Crystal fractionation and differentiation processes controlled the final composition of the rocks, and different cooling stages controlled the textures of each unit. The compositional trend observed in the Colosa suite is from older mafic rocks (diorites of early and intermineral units) to younger, somewhat more silicic rocks (quartz diorites to dacites of late porphyries and dikes).

According to the data provided by the thermobarometry on amphiboles (Fig. 14), the main magma chamber or MASH zone (Hildreth and Moorbath, 1988) could have been ~35-50 km deep, where an important crystallization stage began within the magma. This depth correlates with the estimated depth for the Moho under the Colombian Andes (Meissnar et al., 1976; Thorpe et al., 1984; Taboada et al., 2000; Cediel and Cáceres, 2000; Cediel et al., 2003). The location of this chamber could suggest a crustal contamination at lower crustal, rather than upper crustal, levels. At this depth, fractionation and differentiation of a probable basaltic magma occurred, where mafic phases could have crystallized producing less mafic and less dense magmas (basaltic andesite). These fractionates could have ascended through the crust and accumulated at a new and shallower magmatic chamber, located at depths of ~15-20 km according to the thermobarometric results and coinciding with the boundary between the Cajamarca Complex and the continental basement (Proterozoic rocks; Restrepo-Pace, 1992; Cediel and Cáceres, 2000). At this shallower depth, crystallization and fractionation processes continued, creating andesitic magmas (early and intermineral porphyries) that where emplaced at even shallower depths (~3-4 km deep; Fig. 16A-B). The age of the earliest intrusion (NDE1) is about 8.3±0.2 Ma (AngloGold Ashanti database), which that gives an initial age for the evolution of the system. 18

Early and intermineral units have similar mineralogic and chemical compositions (Tables 2, 3), but the textural features of each group are quite different. The first two early units are phaneritic (fine- to medium-grained), suggesting very stable conditions during their crystallization. The last early units are porphyritic, indicating two substages of crystallization, in which both were equally important as inferred from the relative volumes of phenocrysts (46-54 vol%) and groundmass.

All intermineral units have porphyritic textures, suggesting two substages of crystallization during their formation. The substages of phenocryst crystallization were less important than the matrix-formation stages, as indicated by the lower volume of phenocrysts (34-46 vol%) with respect to groundmass. The higher content of phenocrysts of the early porphyritic units respect to the intermineral units suggests a long stage (probably deeper) of phenocryst formation for the early units and a long stage (probably shallow) of groundmass formation for the intermineral units.

The continuous fractionation of magma at ~15-20 km deep generated dacitic magmas and more differentiated rocks (late porphyries and dikes) that were emplaced at the sides of the slightly older dioritic bodies (Fig. 16C). Differentiation processes of magmatic systems at these depths could last for a few thousands of years according to Turner et al. (2000).

The late units have phenocryst contents of 42-57 volume percent, and the most silicic compositions of all the units of Colosa. The phenocryst contents indicate a first long-stage of crystallization (probably deep) and a second short-stage of crystallization (probably shallow) of a more differentiated magma.

The crystallization processes in all units were somehow continuous during the travel of the magma from stage to stage, as suggested by the seriate texture (from matrix-size to phenocryst-size) observed in all the porphyritic units. Disequilibrium among phenocrysts is common in many of the rocks, such as reverse zoning in plagioclase (Fig. 12C) and resorption textures in quartz (Fig. 6A); these features could indicate mixing between magmas of similar composition, local rise in temperature that may have been caused by underplating and/or recharge of a chamber by a more mafic magma, and/or the ascent of an anhydrous magma (Gill, 1980).

The increase in the temperature of formation of amphiboles with decreasing pressure, as observed in the thermobarometric plot (Fig. 14), could be due to changing conditions during the crystallization process. When magma is emplaced into a relatively cold rock, it is forced to crystallize very fast at high temperatures; an extreme case is the crystallization of minerals of volcanic rocks at surface temperatures, where high-temperature minerals are formed, although generally only in the groundmass. Probably some mineral phases in rocks from Colosa, including amphiboles, crystallized very fast when they were emplaced at shallower depths and into colder rocks.

The age of the late dacite unit (NDA) is about 7.6±0.2 Ma and of a late dike about 7.3±0.2 Ma (AngloGold Ashanti database). Taken at face value, these ages indicate a lifetime of approximately a million years for the entire magmatic system. Erosion then removed the top of the system and any evidence of possible volcanic and epithermal systems related to the hypabyssal rocks now exposed at Colosa (Fig. 16D).

DISCUSSION

Comparisons with Miocene-Recent igneous rocks in Colombia

The tectono-magmatic setting of the Central Cordillera of Colombia for the late Miocene-Pliocene is very similar to the present state, where the most outstanding feature is the magmatism, mainly of andesitic to dacitic composition, related to the subduction of the Nazca plate under the South America plate (Marriner and Millward, 1984; McCourt et al., 1984; Aspden et al., 1987; Restrepo and Toussaint, 1990; Tistl and Salazar, 1994; Cediel and Cáceres, 2000; Taboada et al., 2000; Cediel et al. 2003; Pindell and Kennan, 2001; Ramos, 2009). According to some authors (Toussaint and Restrepo, 1990; Restrepo-Pace, 1992; Cediel and Caceres, 2000; Cediel et al., 2003; Vargas et al., 2005), there is at present a thin wedge of continental basement (Proterozoic rocks) beneath the Cajamarca Complex and associated rocks (batholiths and stocks); this indicates that the probable crustal contamination that the Colosa magma experienced could have been from old Precambrian rocks and/or from the pre-Mesozoic Cajamarca Complex. In both scenarios, the contamination must have been minimal due to the low 87Sr/86Sr ratios of rocks from Colosa (0.704265-0.704368) and the very high 87Sr/86Sr contents of metamorphic rocks in Colombia that are similar to those of the Cajamarca Complex (0.735108 for the Arquía Complex; Tassinari et al., 2008) and Proterozoic rocks of the Central Cordillera (0.74861-0.77792 for El Vapor mylonitic gneisses; Ordoñez et al., 2006).

Rocks from Colosa show similar patterns on spider and REE diagrams to Recent volcanic rocks of the Northeastern and Southwestern zones of the Colombian volcanic arc. Rocks from volcanoes of the Northeastern zone, including Nevado del Ruíz, Nevado del Tolima, Cerro Machín, and Nevado del Huila, belong to the calc-alkaline series (medium-K to high-K) and ranging compositionally from basaltic andesite to dacite (Marriner and Millward, 1984; Thouret et al., 1995; Correa et al., 2000; Dominguez et al., 2003; Borrero et al., 2009). Based on isotopic compositions of 87Sr/86Sr (0.7042-0.7044), 144Nd/143Nd 18 (0.5127-0.5129), and δ O (+6.8-+7.4%O), James (1982) and James and Murcia (1984) suggest that the andesites from the Nevado del Ruíz volcano have source contamination (subducted slab) and crustal contamination during their magmatic ascent. Marriner and Millward (1984) obtained similar ratios of

87Sr/86Sr for Nevado del Ruíz (0.70432-0.70480) and also observed enrichment in LILE and LREE, depletion in HFS and HREE, negative anomalies of Nb and Ta, absence of Eu anomalies, and high ratios of Ce/Yb (9-17). Ordoñez and Pimentel (2001) obtained isotopic compositions of ƐNd (+2.09-+4.08) and 87Sr/86Sr (0.70431-0.70445) for Nevado del Ruíz and Nevado del Tolima volcanoes, suggesting mantle magmatism with crustal contamination. The Cerro Machín volcano is located at 11 km to the ENE of Colosa and is also emplaced into the Cajamarca Complex. The material of this volcano is mainly ash and lapilli of dacitic composition (oligoclase + amphibole + biotite + quartz) and high-K calc-alkaline affinity (Dominguez et al., 2003).

Volcanoes from the Southwestern zone, including Puracé, Doña Juana, Galeras, Azufral, Cumbal, and Chiles, have rocks with calc-alkaline affinity (medium-K to high-K) and compositions ranging from basaltic andesite to dacite (Marriner and Millward, 1984; Cepeda, 1985; Druox and Delaloye, 1996; Calvache et al., 1997). Rocks from Galeras and Purace are enriched in LILE and LREE, depleted in HFS and HREE, and have 87Sr/86Sr ratios of 0.70410-0.70448 and 0.70435-0.70445, respectively (Marriner and Millward, 1984). Based on 87Sr/86Sr (0.7041-0.7048), 144Nd/143Nd (0.5127-0.5129) and δ18O (+6.5-

+7.9%O) isotopic compositions, James (1980; 1982) and James and Murcia (1984) suggest that the andesites from the Galeras and Purace volcanoes have subducted slab and crustal contamination during the magmatic ascent. Cepeda (1985) suggests that the rocks of the Galeras volcano were formed by crystal fractionation of magmas from the mantle wedge and oceanic crust (Nazca plate) without important crustal contamination. Based on relative enrichment of LREE compared to HREE, the La/Yb ratio (8.9- 13.7), and the enrichment in LILE, Calvache et al. (1997) suggest a source from the partial melting of a hydrated mantle wedge with the addition of highly soluble elements from the subducted sediments, and probably an important additional contribution from the continental crust; these authors also suggest crystal fractionation at depths of 36 km of a hydrated tholeiitic basalt melt that re-equilibrates at higher levels (10 km deep). Druox and Delaloye (1996) obtained on rocks from the Purace, Doña Juana, Galeras, Azufral, Cumbal, and Chiles volcanoes, enrichment of LREE, depletion of HREE, and enrichment of LILE, interpreting the genesis of the rock from melting of the mantle by fluids derived from the dehydration and/or partial melting of the subducted oceanic slab, followed by fractional crystallization of the primitive magma, mixing with lower crust material and possibly crustal contamination during the uprising of the magma through the crust. Ordoñez and Pimentel (2001) obtained isotopic compositions of 87 86 ƐNd (+0.51-+4.68) and Sr/ Sr (0.70407-0.70454) for Doña Juana, Galeras, and Azufral volcanoes, and consider this magmatism derived from mantle with crustal contamination.

The REE abundances and patterns of andesites of the Andean northern volcanic zone are similar to those in northern Chile, but experienced less fractional crystallization and negligible crustal contamination during their ascent (Thorpe and Francis, 1979). Thorpe et al. (1984) suggest that the recent 21 lavas from Colombia may be derived from a mantle enriched in radiogenic Sr derived from the subducted slab but with minor contamination by continental crust.

As seen above, recent rocks from different Colombian volcanoes have chemical characteristics similar to those of Colosa (Fig. 17). For these young volcanic rocks, the recurrent model proposed is a source originated by partial melting of the mantle wedge by fluids and/or melts from the subducted slab and/or sediments, followed by processes of crystal fractionation and crustal contamination, which also fits with the model proposed for the Colosa porphyry. Although chemical compositions are similar between rocks from Colosa and recent Colombian volcanoes, the recent volcanoes tend to show more mafic phases (olivine, clinopyroxene, and orthopyroxene) than rocks from Colosa (Calvache et al., 1997; Correa et al., 2000).

Other Miocene intrusions related to gold deposits, as in the case of Marmato (low-sulfidation system) 87 86 where andesite-dacite porphyries of 6.7±0.1 Ma have measured Sr/ Sr of 0.70440-0.70460 and ƐNd of +2.2-+3.2 (Tassinari et al., 2008), share very similar compositions than those of Colosa and a comparable genesis could also be inferred. These similar features between Colosa, contemporaneous rocks, and recent volcanic rocks suggest similar mechanisms for generating magmatism since the late Miocene in the Colombian volcano-plutonic arc of the Central Cordillera.

Petrological models for porphyry systems

The model proposed for the origin of the Colosa porphyry fits with recent petrogenetic models proposed for porphyry systems by Richards (2003) and Candela and Piccoli (2005). The petrological model of Richards (2003) is based on hydrous, relatively oxidized, and sulfur-rich mafic magmas (predominantly basalts) that are generated in the metasomatized mantle wedge by solute-rich fluids (rich in LILE but not in Ti, Nb, and Ta) above a subducting oceanic slab. These magmas rise buoyantly to the base of the overlying crust, where they stall due to density contrasts and mix with crustal melts (MASH), generating evolved (andesitic to dacitic), volatile-rich, metalliferous, hybrid magmas. The latter, are of sufficiently low density to rise through the crust. Magma ascent is driven primarily by buoyancy forces and is dominantly a fracture-controlled phenomenon where translithospheric, orogen-parallel, strike-slip structures may serve as a primary control on magma emplacement in many volcanic arcs worldwide. Having delivered a sufficient volume of evolved, fertile arc magma to a focused position in the upper crust, magmatic fractionation, recharge, and volatile exsolution lead to the development of ore-forming magmatic-hydrothermal systems. Richards (2009) adds a new connotation to the MASH model for postsubduction magmatism settings, by calling on remelting of small amounts of residual sulfide, left in the deep lithosphere by arc magmatism, to explain the high Au content.

Candela and Piccoli (2005) consider that elements such as B, Pb, As, and Sb, as well as Sr, LILE, and other elements, appear to be mobilized efficiently from the subducted slab; H2O, Cl, and S may be sourced partially from sea water contained in the oceanic slab. They also suggest that ore metals in arc magmas probably have diverse origins, including the mantle wedge, the lower continental crust, and the subducted lithosphere. Seedorff et al. (2005) think that the varying proportions of these sources (also including the upper crust) likely lead to the observed compositional diversity of ore-forming magmas for porphyry systems. Sillitoe (2000) proposes for the origin of gold-rich porphyries a metal-bearing hydrous melt derived from dehydration of the subducted slab that causes partial melting of the oxidized and metasomatized mantle wedge where it produces a variety of calc-alkaline magma that transports the volatiles and metals with varying amounts of crustal interaction and assimilation.

Comparison with porphyry gold systems in the Maricunga belt

As mentioned earlier, porphyry gold deposits other than Colosa are mostly located in the Maricunga belt in Chile, including the Lobo, Marte, Verde, Pancho, and Cavancha deposits. Many features from these porphyries are similar to Colosa, but others are different (Table 1).

The porphyry gold systems of the Maricunga belt have an early- to middle Miocene ages (Sillitoe et al., 1991), in contrast to the late-Miocene age of Colosa, notwithstanding both are composed by diorite to quartz diorite porphyries of the calc-alkaline trend (medium- to high-K; Kay et al., 1994; Muntean and Einaudi, 2000). At the present levels of erosion, Colosa is emplaced within low-grade metamorphic rocks, but the Chilean porphyries are emplaced into coeval volcanic rocks of intermediate composition (Vila and Sillitoe, 1991; Kay et al., 1994; Muntean and Einaudi, 2000). The presence of late dacites and intrusive breccias (diorite porphyry xenoliths in diorite matrix) is also a common feature of both (Vila and Sillitoe, 1991; Muntean and Einaudi, 2000). The rocks of the Maricunga belt are fine- to coarse-grained with phenocrysts of plagioclase (An20-56, zoned, and seriate), biotite, hornblende, quartz (rounded and resorbed), and subordinate pyroxene; the aphanitic matrix (<0.05mm) is usually more abundant than phenocrysts and is formed by microlites of plagioclase, mafic minerals, quartz, K-feldspar, ilmenite, magnetite, zircon, apatite, sphene, and rutile (Vila and Sillitoe, 1991; Vila et al., 1991; Flores, 1993; Muntean and Einaudi, 2000). The rocks of Colosa are very similar to those described above, but they lack pyroxene, primary magnetite, and rutile. Amphiboles from the Verde and Pancho porphyry gold deposits (Muntean, 1998) have higher Mg contents than the amphiboles from Colosa and could be classified as magnesiohastingsite, magnesiohornblende, and tschermakite. Using the pressure-temperature diagram of Ernst and Liu (1998) for the same amphiboles, they show temperatures (740-870°C) and pressures (0.1- 1.0 GPa) similar to those of stages II and III of Colosa; this indicates that both localities experimented the same or similar conditions of temperature and pressure during their final episodes of crystallization.

Porphyry gold deposits of the Maricunga belt were generated only at 600 to 2,000 m beneath the early or middle Miocene volcanic paleosurface and less than four intrusive units are recognized (Vila and Sillitoe, 1991; Muntean and Einaudi, 2000). In contrast, the rocks of Colosa could have been emplaced at depths near to 3-4 km and are composed by over twelve different intrusions.

Rocks from the porphyry gold deposits of the Maricunga belt have SiO2 contents of 59-64% (Muntean and Einaudi, 2000) and trace element ratios of 25-40 for Ba/La, 13-18 for La/Yb, and 26-35 for Ce/Yb (Kay et al., 1994). These values are contained within the values of Colosa for the same ratios and 87 86 SiO2 contents (58-68% SiO2, 11-78 Ba/La, 5-20 La/Yb, and 9-38 Ce/Yb). The Sr/ Sr (0.704838-

0.70616) and ƐNd (-0.4 - -4.3) isotopic compositions for other types of porphyry systems and volcanic rocks within the Maricunga belt are mainly located in the continental crust field (Robinson and McKee, 1993; Kay et al., 1994; McKee et al. 1994), whereas rocks from Colosa rocks are located in the mantle array field (Fig. 17).

The gold-rich porphyry deposits in the Maricunga belt were emplaced at a continental margin characterized by thick (35-50 km) continental crust and tectonic changes that occurred as the subducting Nazca plate changed shape and caused modifications in the overlying crust and lithospheric mantle (Vila and Sillitoe, 1991; Kay et al., 1994). Large depletions in the MREE and a decrease in the size of Eu anomalies indicate an increasing role for amphibole and a decreasing importance of residual feldspar (Kay et al., 1994). Mafic lower crustal magma sources are suggested from Sr isochrons (Robinson and McKee; McKee et al. 1994), but Hildreth and Moorbath (1988) indicate a source of for central Chilean magmas related to the partial melting of the mantle wedge by fluids from the dehydration of the subducted slab and with crustal contamination at the MASH zone. At Colosa, the crust could have had similar thickness, but amphiboles rather than feldspars seem to have been important fractionating minerals in the early stages of crystallization. The source of the Colosa magma correlates with the model of Richards (2003), which in turn is based on the model of Hildreth and Moorbath (1988) but more complete and applicable to porphyry systems.

CONCLUSIONS

The Colosa porphyry was constructed by hypabyssal rocks of intermediate composition formed by andesine + amphibole + quartz ± biotite ± orthoclase, and apatite + zircon + sphene as trace minerals. The early units of the system are diorites and diorite porphyries with 46-54 percent phenocrysts; the intermineral units are diorite porphyries with 34-46 percent phenocrysts; and the late units are dacite porphyries and diorite to quartz diorite dikes with 42-57 percent phenocrysts. The amphiboles were

24 classified as ferropargasites, ferro-edenites, and ferrotschermakites.

Probably all the units of Colosa were generated by the same magma chamber as shown by the similar patterns in spider and REE diagrams; fractional crystallization and differentiation processes controlled the final composition of the rocks, and cooling stages determined the texture. The rocks of Colosa belong to a medium- to high-K calc-alkaline suite and are related to the late Miocene volcanic-plutonic arc of Colombia, generated by the subduction of the Nazca plate under the South America plate.

All samples exhibit an enrichment of LILE and LREE relative to HFSE and HREE, respectively. These features are attributed to enrichment of LILE from the source and retention of HFSE (mainly Nb, Ta, and Ti) by refractory phases within the same source. The depletion of HREE is explain by fractionation of mineral phases that have a high partition coefficients for these elements, especially amphiboles, the major mafic phase in the rocks.

Isotopic compositions of Colosa indicate a source derived from the mantle, and high contents of LILE and high ratios of Ba/La suggest contamination with melts from the subducted sediments and probably from the continental crust. The model proposed for Colosa consists of fluids from the dehydration of the subducted slab (Nazca plate) and fluids and melts from the subducted sediments that generated partial melting of the mantle wedge. These basaltic melts ascended to the mantle-crust boundary where they were retained due to density differences and began to produce processes of melting, assimilation, storage, and homogenization (MASH zone). At this depth (~35-50 km), fractional crystallization and differentiation processes began to produce more felsic magmas that were able to ascend through the crust and be emplaced at the boundary between the continental basement and the Cajamarca Complex (~15- 20). At this site, the basaltic andesite magma began to produce more differentiated products (andesitic and dacitic magmas) that ascended across the Cajamarca Complex and were emplaced at depths of ~3-4 km; it is here where the rocks completely crystallize and where all the hydrothermal activity took place.

Although Colosa shares common features with the porphyry gold deposits of the Maricunga belt, such as similar rock composition (diorites, quartz diorites, and dacites), and alteration type (potassic, intermediate argillic, and propylitic), there are some characteristics that are different, such as amphibole composition (more magnesian in Chile), isotopic compositions (mantle array for Colosa and continental crust array for Maricunga), wall-rock type (coeval volcanic rocks for the Maricunga belt and low-grade metamorphic rocks for Colosa), mineralization assemblage (presence of pyrrhotite in Colosa), and age (early to middle Miocene for Maricunga and late Miocene for Colosa). These characteristics, and its enormous size (12.9 Moz of contained Au), make the Colosa porphyry unique within the porphyry gold systems discovered to date.

AKNOWLEDGMENTS

Special thanks to Eric Seedorff, Germán González, Jennifer Betancourt, AngloGold Ashanti Colombia S. A., and The Lowell Program in Economic Geology of the University of Arizona.

REFERENCES

AngloGold Ashanti, 2008, New Release AGA 12.08, 06 May 2008., http://www.anglogoldashanti.com/ resources/20080605_colosa.pdf

Aspden, J. A., McCourt, W. J. and Brook, M., 1987, Geometrical control of subduction-related magmatism: The Mesozoic and Cenozoic plutonic history of western Colombia: Journal of the Geological Society of London, v. 144, p. 893-905.

Barrick Gold, 2009, Annual Report 2008, 158 p., http://www.barrick.com/Theme/Barrick/files/docs_ annual/2008%20Annual%20Report%20-%20English.pdf

Best, M. G., 2003, Igneous and Metamorphic Petrology: Turin, Blackwell Publishing, 729 p.

Borrero, C., Toro, L. M., Alvarán, M., and Castillo, H., 2009, Geochemistry and tectonic controls of the effusive activity related with the ancestral Nevado del Ruíz volcano, Colombia: Geofísica Internacional, v. 48, issue 1, p. 149-169.

Brown, A. J., and Rayment, B., 1991, Refugio gold project, Chile: Mining Magazine, v. 165, p. 306-312.

Calvache, M. L., and Williams, S. N., 1997, Geochemistry and petrology of the Galeras volcanic complex, Colombia: Journal of Volcanology and Geothermal Research, v. 77, p. 21-38.

Canby, V. M., Noble, D. C., and McKee, E. H., 1993, Porphyry-type gold mineralization of late Neogene age at the Zule volcanic center, Sierra County, California [abs]: Geological Society of America Abstracts with Programs, v. 25, no. 5, p. 17.

Candela, P. A., and Piccoli, P. M., 2005, Magmatic processes in the development of porphyry-type ore

systems: ECONOMIC GEOLOGY 100TH ANNIVERSARY VOLUME, p. 25-37.

Cediel, F., and Cáceres, C., 2000, Geologic map of Colombia, Geotec Ltd., Third Edition, Digital Format with Legend and Tectono-Stratigraphic Chart.

Cediel, F., Shaw, R. P., and Cáceres, C., 2003, Tectonic assembly of the Northern Andean block, in Bartolini, C., Buffler, R. T., and Blickwedw, J., eds., The Circum-Gulf of Mexico and Caribbean: Hydrocarbon habitats, basin formation, and plate tectonics: American Association of Petroleum Geologists Memoir 79, p. 815-848.

Cepeda, H., 1985, Anotaciones acerca de la geología del Volcán Galeras (Colombia, S. A.): Sexto Congreso Latinoamericano de Geología, p. 338-383.

Cooke, D. R., Hollings, P., and Walshe, J. L., 2005, Giant porphyry deposits: Characteristics, distribution,

and tectonic controls: ECONOMIC GEOLOGY, v. 100, p. 801-818.

Correa, A. M., Cepeda, H., Pulgarín, B., and Ancochea, E., 2000, El Volcán Nevado del Huila (Colombia): Rasgos generales y caracterización composicional: Geogaceta, v. 27, p. 51-54.

Cunningham, C. G., Austin, G. W., Naeser, C. W., Rye, R. O., Ballantyne, G. H., Stamm, R. G., and Barker, C. E., 2004, Formation of a paleothermal anomaly and disseminated gold deposits associated with the

Bingham Canyon porphyry Cu-Au-Mo system, Utah: ECONOMIC GEOLOGY, v. 99, p. 789-806.

De La Roche, H., Leterrier, J., Grandclaude, P., and Marchal, M., 1980, A classification of volcanic and plutonic rocks using R1R2- diagram and major element analyses - Its Relationships with current nomenclature: Chemical Geology, v. 29, p. 183-210.

Domínguez, L., Concha, A. E., and Cepeda, H., 2003, Volcán Cerro Machín: Mecanismo eruptivo, transporte y sedimentación (Tolima, Colombia): Geología Colombiana, v. 28, p. 3-21.

Droux, A., and Delaloye, M., 1996, Petrography and geochemistry of Plio-Quaternary calc-alkaline volcanoes of southwestern Colombia: Journal of South American Earth Sciences, v. 9, p. 27-41.

Ernst, W. G., and Liu, J., 1998, Experimental phase equilibrium study of Al- and Ti-contents of calcic amphibole in MORB-A semiquantitative thermobarometer: American Mineralogist, v. 83, p. 952-969.

Exeter Resource, 2009, Management’s discussion and analysis for the nine months ended September 30 - 2009, 16 p., http://www.exeterresource.com/pdf/financials/2009/exeter_financials_Q3_2009b.pdf

Flores, R., 1993, Geología del pórfido aurífero Verde, proyecto Refugio, Tercera Región, Chile: Revista Geológica de Chile, v. 20, p. 57-69.

Gill, J. B., 1981, Orogenic Andesites and Plate Tectonics: New York, Springer-Verlag, 390 p.

Gómez, J., Nivia, A., Montes, N., Jiménez, D., Sepúlveda, M., Gaona, T., Osorio, J., Diederix, H., Mora, 27

M., and Velásquez, M., 2007, Atlas Geológico de Colombia, Plancha 5-09, escala 1:500,000,

INGEOMINAS.

Gorton, M. P., and Shandl, E. S., 2000, From continents to island arcs: A geochemical index for tectonic setting for arc-related and within-plate felsic to intermediate volcanic rocks: Canadian Mineralogist, v. 38, p. 1065-1073.

Green, T. H., 1982, Anatexis of mafic crust and pressure crystallization of andesite, in Thorpe, R. S., ed., Andesites: Orogenic andesites and related rocks: Chicester, John Wiley & Sons, p. 465-487.

Harmon, R. S., Thorpe, R. S., and Francis, P. W., 1981, Petrogenesis of Andean andesites from combined O-Sr isotope relationships: Nature, v. 290, p. 396-399.

Hawkesworth, C. J., Norry, M. J., Roddick, J. C., Baker, P. E., Francis, P. W., and Thorpe, R. S., 1979, 143Nd/144Nd, 87Sr/86Sr, and incompatible element variations in calc-alkaline andesites and plateau lavas from South America: Earth and Planetary Science Letters, v. 42, p. 45-57.

Hildreth, W., and Moorbath, S., 1988, Crustal contributions to arc magmatism in the Andes of central Chile: Contributions to Mineralogy and Petrology, v. 98, p. 455-489.

Irvine, T. M., and Baragar, W. R., 1971, A Guide to the chemical classification of common volcanic rocks: Canadian Journal of Earth Sciences, v. 8, p. 523-548.

James, D. E., 1980, Oxygen and strontium isotopic composition of Quaternary andesites from the Andes of SW Colombia: Carnegie Institution of Washington Year Book, v. 79, p. 452-455.

James, D. E., 1982, Isotopic evidence for the petrogenesis of modern andesites of the northwestern Andes: Carnegie Institution of Washington Year Book, v. 81, p. 485-489.

James, D. E., and Murcia, L. A., 1984, Crustal contamination in northern Andean volcanic: Journal of the Geological Society of London, v. 141, p. 823-830.

Janousek, V., Farrow, C. M., and Erban, V., 2006, Interpretation of whole-rock geochemical data in igneous geochemistry: Introducing geochemical data toolkit (GCDkit): Journal of Petrology, v. 47, no. 6, p. 1255–1259.

Kay, R. W., 1980, Volcanic arc magmas: Implications of a melting-mixing model for element recycling in the crust-upper mantle system: Journal of Geology, v. 88, no. 5, p. 497-522.

Kay, S. M., Mpodozis, C., Tittler, A., and Cornejo, P., 1994, Tertiary magmatic evolution of the

Maricunga mineral belt in Chile: International Geology Reviews, v. 36, p. 1079-1112.

Kennan, L., and Pindell, J., 2009, Dextral shear, terrane accretion and basin formation in the northern Andes: Best explained by interaction with a Pacific-derived Caribbean plate?, in James, K. H., Lorente, M. A., and Pindell, J. L., eds., The Origin and Evolution of the Caribbean Plate: Geological Society of London, Special Publications, v. 328, p. 487-532.

Kesler, S. E., 1973, Copper, molybdenum and gold abundances in porphyry copper deposits: ECONOMIC

GEOLOGY, v. 68, p. 106-112.

Lanier, G., John, E. C., Swensen, A. J., Reid, J., Bard, C. E., Caddey, S. W., and Wilson, J. C., 1978,

General geology of the Bingham mine, Bingham Canyon, Utah: ECONOMIC GEOLOGY, v. 73, p. 1228-1241.

Leake, B. E., Woolley, a. R., Arps, C. E. S., Birch, W. D., Gilbert, M. C., Grice, J. D., Hawthorne, F. C., Kato, A., Kisch, H. J., Krivovichev, V. G., Linthout, K., Laird, J., Mandarino, J., A., Maresh., W. V., Nickel, E. H., Schumacher, J. C., Smith, D., C., Stephenson, N. C. N., Ungaretti, L., Whittaker, E. J. W., and Youzhi, G., 1997, Nomenclature of amphiboles: Report of the Subcommittee on Amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names: Canadian Mineralogist, v. 35, p. 219-246.

MacDonald, G. D., and Arnold, L. C., 1994, Geological and geochemical zoning of the Grasberg igneous complex, Irian Jaya, Indonesia: Journal of Geochemical Exploration, v. 50, p. 143-178.

MacDonald, W. D., Estrada, J. J., Sierra, G. M., and, Gonzalez, H., 1996, Late Cenozoic tectonics and paleomagnetism of North Cauca Basin intrusions, Colombian Andes: Dual rotation modes: Tectonophysics, v. 261, p. 277-289.

Marriner, G. F., and Millward, D., 1984, The petrology and geochemistry of Cretaceous to Recent volcanism in Colombia: The magmatic history of an accretionary plate margin: Journal of the Geological Society of London, v. 141, p. 473-486.

Maya, M., 1992, Catalogo de Dataciones Isotópicas en Colombia: Boletín Geológico Ingeominas, v. 32, no. 1-3, p. 127-188.

Maya, M., and González, H., 1995, Unidades Litodémicas en la Cordillera Central de Colombia: Boletín

Geológico INGEOMINAS, v. 35, no. 2-3, p. 43-57.

McCourt, W. J., Aspden, J. A., and Brook, M., 1984, New geological and geochronological data from the 29

Colombian Andes: Continental growth by multiple accretion: Journal of the Geological Society of London, v. 141, p. 831-845.

McBirney, A. R., 2007, Igneous Petrology: Sudbury, Jones and Bartlett Publishers, 550 p.

McKee, E. H., Robinson, A. C., Rybuta, J. J., Cuitiño, L., and Moscoso, R. D., 1994, Age and Sr isotopic composition of volcanic rocks in the Maricunga belt, Chile: Implications for magma sources: Journal of South American Earth Sciences, v. 7, p. 167-177.

Meissnar, R. O., Flueh, E. R., Stibane, F., and Berg, E., 1976, Dynamics of the active plate boundary in southwest Colombia according to recent geophysical measurements: Tectonophysics, v. 35, p. 115- 136.

Millward, D., Marriner, G. F., and Verdugo, G., 1982, La geoquímica del Grupo Diabásico en la Cordillera Occidental al Oeste de Cali, Valle Colombia: Geología Norandina, v. 5, p. 39-45.

Moore, W. J., 1978, Chemical characteristics of hydrothermal alteration at Bingham, Utah: ECONOMIC

GEOLOGY, v. 73, p. 1260-1269.

Muntean, J. L., 1998, Magmatic-hydrothermal gold deposits of the Maricunga belt, northern Chile, PhD dissertation, Stanford University, 400 p.

Muntean, J. L., and Einaudi, M. T., 2000, Porphyry gold deposits of the Refugio district, Maricunga belt,

northern Chile: ECONOMIC GEOLOGY, v. 95, p. 1445-1472.

Muntean, J. L., and Einaudi, M. T., 2001, Porphyry-epithermal transition: Maricunga belt, northern Chile:

ECONOMIC GEOLOGY, v. 96, p. 743-772.

Nakamura, N., 1974, Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and ordinary chondrites: Geochimica et Cosmochimica Acta, v. 38, p. 757–775.

Nelson, H. W., 1962, Contribución al conocimiento de la Cordillera Central de Colombia. Sección entre Ibagué y Armenia: Boletín Geológico Servicio Geológico Nacional, v. 10, no. 1-3, p. 161-202.

Newcrest Mining, 2010, Annual Report 2009, 118 p., http://newcrest.ice4.interactiveinvestor.com.au/ Newcrest0901/Annual%20Report%202009/EN/body.aspx?z=1&p=-1&v=2&uid=

Newmont, 2009, Annual Report 2008, 204 p., http://files.shareholder.com/downloads/NEM/826999486x0 x285595/61C5A9C3-D895-4D3E-9678-8C3217C6E072/Newmont_Annual_Report_2008.pdf

Northgate Exploration, 2004, Annual Report 2003, 42 p., http://www.northgateminerals.com/Theme/ Northgate/files/pdf/2003Annual_Report.pdf

Núñez, A., 2001, Memoria Explicativa Mapa Geológico del Departamento del Tolima, INGEOMINAS, 101 p.

Núñez, A., González, H., and Linares, E., 1979, Nuevas Edades Radiométricas de los Esquistos Verdes del Grupo Cajamarca. Publicación Especial de Geología, Universidad Nacional de Colombia, v. 23, p. 1-18.

Oceana Gold, 2009, Annual Report 2008, 76 p., http://www.oceanagold.com/images/documents/files/ogc _2008_annual_report.pdf

Ordoñez, O., and Pimentel, M., 2001, Geoquímica Isotópica del Magmatismo Reciente (< 11Ma), en los Andes Colombianos: Memoirs of the VIII Congreso Colombiano de Geología (in CD), 4 p.

Ordoñez-Carmona, O., Restrepo, J. J., and Pimentel, M., M., 2006, Geochronological and isotopical review of pre-Devonian crustal basement of the Colombian Andes: Journal of South American Earth Sciences, v. 21, p. 372-382.

Pearce, J. A., Harris, N. B. W., and Tindle, A. G., 1984, Trace element discrimination diagrams for the tectonic interpretation of granitic rocks: Journal of Petrology, v. 25, part 4, p. 956-983.

Pearce, J. A., and Peate, D. W., 1995, Tectonic implications of the composition of volcanic arc magmas: Annual Reviews of Earth and Planetary Sciences, v. 23, p. 251-285.

Peccerillo, A., and Taylor, S. R., 1976, Geochemistry of Eocene calc-alkaline volcanic rocks from the Kastamonu area, northern Turkey: Contributions to Mineralogy and Petrology, v. 58, p. 63-81.

Pindell, J. L., and Kennan, L., 2001, Kinematic evolution of the Gulf of Mexico and Caribbean: Gulf Coast Section Society of Economic Paleontologists and Mineralogists Foundation, 21st Annual Research Conference Transactions, Petroleum Systems of Deep-Water Basins, p. 193-220.

Pindell, J. L., and Kennan, L., 2009, Tectonic evolution of the Gulf of Mexico, Caribbean and Northern South America in the mantle reference frame: An update, in James, K. H., Lorente, M. A., and Pindell, J. L., eds., The Origin and Evolution of the Caribbean Plate: Geological Society of London, Special Publications, v. 328, p. 1-55.

Pollard, P. J., Taylor, R. G., and Peters, L., 2005, Ages of intrusion, alteration, and mineralization at the

Grasberg Cu-Au deposit, Papua, Indonesia: ECONOMIC GEOLOGY, v. 100, p. 1005-1020.

Proffett, J. M., 2003, Geology of the Bajo de la Alumbrera porphyry copper-gold deposit, Argentina:

ECONOMIC GEOLOGY, v. 98, p. 1535-1574.

Pulido, O. H., 1988, Geología y geoquímica del área de San Antonio, Cajamarca, Tolima: Boletín

Geológico INGEOMINAS, v. 29, no. 2, p. 37-84.

Ramos, V. A., 2009, Anatomy and global context of the Andes: Main geologic features and the Andean orogenic cycle, in Kay, S. M., Ramos, V. A., and Dickinson, W. R., eds., Backbone of the Americas: Shallow Subduction, Plateau Uplift, and Ridge and Terrane Collision: Geologic Society of America Memoir 204, p. 31-65.

Restrepo, J. J., and Toussaint, J. F., 1978, Datación de una metadiabasa del Grupo Cajamarca: Publicación Especial de Geología, Universidad Nacional de Colombia, sede Medellin, v. 16, p. 4.

Restrepo, J. J., and Toussaint, J. F., 1988, Terranes and continental accretion in the Colombian Andes:

EPISODES, v. 11, No. 3, p. 189-193.

Restrepo, J. J., and Toussaint, J. F., 1990, Cenozoic arc magmatism of northwestern Colombia, in Kay, S. M., and Rapela, C. W., eds., Plutonism from Antarctica to Alaska: Geological Society of America Special Paper 241, p. 205-212.

Restrepo, J. J., and Toussaint, J. F., Hall, M., Pimentel, N., Cordani, U., Kawashits, K., and Linares, E., 1985, Evolución magmática de los Andes septentrionales: Comunicaciones, v. 35, p. 201-208.

Restrepo-Pace, P. A., 1992, Petrotectonic characterization of the Central Andean Terrane, Colombia: Journal of South American Earth Sciences, v. 5, no. 1, p. 97-116.

Richards, J. P., 2003, Tectono-magmatic precursors for porphyry Cu-(Mo-Au) deposit formation:

ECONOMIC GEOLOGY, v. 98, p. 1515-1533.

Richards, J. P., 2009, Postsubduction porphyry Cu-Au and epithermal Au deposits: Products of remelting of subduction-modified lithosphere: Geology, v. 37, p. 247-250.

Robinson, A. C., and McKee, E. H., 1993, Sr isotopic composition of the Miocene Volcanic Maricunga Belt, 26-28S, Chilean Andes: Evidence for contiguous Middle and Late Proterozoic lower crustal magma sources [abs.]: EOS (Transactions American Geophysical Union), v. 74, no. 43, p. 662-663.

Rollinson, H. R., 1993, Using geochemical data: Evaluation, presentation, interpretation: London, 32

Longman, 352 p.

Rueda, H., Macias, J. L., Siebe, C., Cepeda, H., Mendez, R., and Cortes, G., 2005, Holocene eruptions of Machin volcano: Stratigraphy and eruptive dynamics [abs]: EOS (Transactions of the American Geophysical Union), v. 86, issue 52, Supplement Abstract, V53B-1550, 1340h poster.

Schumacher, J. C., 1997, Appendix 2. The estimation of the proportion of ferric iron in the electron- microprobe analysis of amphiboles, in Leake, B. E. et al., 1997, Nomenclature of Amphiboles: Report of the Subcommittee on Amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names: Canadian Mineralogist, v. 35, p. 238-246.

Seedorff, E., Dilles, J. H., Proffett, J. M., Einaudi, M. T., Zurcher, L., Stavast, W. J. A., Johnson, D. A., and Barton, M. D., 2005, Porphyry Deposits: Characteristics and Origin of Hypogene Features:

ECONOMIC GEOLOGY 100TH ANNIVERSARY VOLUME, p. 251-298.

Sillitoe, R. H., 1979, Some thoughts on gold-rich porphyry copper deposits: Mineralium Deposita, v. 14, p. 161-174.

Sillitoe, R. H., 2000, Gold-rich porphyry deposits: Descriptive and genetic models and their role in exploration and discovery, in Hagemann, S. G., and Brown, P. E., eds., Gold in 2000: Reviews in Economic Geology, v. 13, p. 315-345.

Sillitoe, R. H., 2008, Major gold deposits and belts of the North and South American Cordillera:

Distribution, tectonomagmatic settings, and metallogenic considerations: ECONOMIC GEOLOGY, v. 103, p. 663-687.

Sillitoe, R. H., McKee, E. H., and Vila, T., 1991, Reconnaissance K-Ar geochronology of the Maricunga

gold-silver belt, northern Chile: ECONOMIC GEOLOGY, v. 86, p. 1261-1270.

Streckeisen, A. L., 1976, To each plutonic rock its proper name: Earth-Science Reviews, v. 12, p. 1-33.

Taboada, A., Rivera, L. A., Fuenzalida, A., Cisternas, A., Philip, H., Bijwaard, H., Olaya, J., and Rivera, C., 2000, Geodynamics of the Northern Andes: Subductions and intracontinental deformation (Colombia): Tectonics, v., 19, no. 5, p. 787-813.

Taseko Mines, 2009, Technical Report on the 344 Million Tonne increase in mineral Reserves at the Prospery Gold-Copper Project, 167 p., British Columbia, Canada. http://www.sedar.com/GetFile.do? lang=EN&docClass=24&issuerNo=00003212&fileName=/csfsprod/data103/filings/01492653/00000 002/C%3A%5CSEDAR%5CFILINGS%5CTKO09ProsperityTechRpt.pdf 33

Tassinari, C., C., G., Díaz, F., and Buenaventura, J., 2008, Age and sources of gold mineralization in the Marmato mining district, NW Colombia: A Miocene-Pliocene epizonal gold deposit: Ore Geology Reviews, v. 33, p. 505-518.

Terrane Metals, 2009, Technical Report: Feasibility Update Mt. Milligan Property, Northern BC, 481 p., http://www.terranemetals.com/i/pdf/2009-10-23-Mt-Milligan-Technical-Report-2009.pdf

Thompson, R. N., 1982, British Tertiary province: Scottish Journal of Geology, v. 18, p. 49-107.

Thorpe, R. S., and Francis, P. W., 1979, Variations in Andean andesite compositions and their petrogenetic significance: Tectonophysics, v. 57, p. 53-70.

Thorpe, R. S., Francis, P. W., and Harmon, R.S., 1980, Andean andesites and crustal growth: Revista Geológica de Chile, v. 10, p. 55-73.

Thorpe, R. S., Francis, P. W., and O’Callaghan, L., 1984, Relative roles of source composition, fractional crystallization and crustal contamination in the petrogenesis of Andean volcanic rocks: Philosophical Transactions of the Royal Society of London, series A, v. 310, p. 675-692.

Thouret, J. C., Cantagrel, J. M., Robin, C., Murcia, A., Salinas, R., and Cepeda, H., 1995, Quaternary eruptive history and hazard-zone model at Nevado del Tolima and Cerro Machín volcanoes, Colombia: Journal of Volcanology and Geothermal Research, v. 66, p. 397-426.

Tistl, M., and Salazar, G., 1994, The tectono-magmatic evolution of northwestern South America: Zentralbaltt fuer Geologie und Palaeontologie Teil I, issue 1/2, p. 439-453.

Toussaint, J. F., and Restrepo, J. J., 1982, Magmatic evolution of the northwestern Andes of Colombia: Earth-Sciences Reviews, v. 18, p. 205-213.

Toussaint, J. F., and Restrepo, J. J., 1990, Cronología de las acreciones de terrenos alóctonos en los Andes colombianos: Comunicaciones, Universidad de Chile, v. 41, p. 85-88.

Turner, S. P., George, R. M. M., Evans, P. J., Hawkesworth, C. J., and Zellmer, G. F., 2000, Time-scales of magma formation, ascent and storage beneath subduction-zone volcanoes: Philosophical Transactions of the Royal Society of London, series A, v. 358, p. 1443-1464.

Ulrich, T., and Heinrich, C. A., 2002, Geology and alteration geochemistry of the porphyry Cu-Au deposit

at Bajo de la Alumbrera, Argentina: ECONOMIC GEOLOGY, v. 97, p. 1865-1888.

Vargas, C. A., Kammer, A., Valdes, M., Rodríguez, C. E., Caneva, A., Sánchez, J. J., Arias, E., Cortes, C. 34

A., and Mora, H., 2005, New geological and geophysical contribution in the section Ibagué-Armenia, Central Cordillera-Colombia: Earth Sciences Research Journal, V. 9, no. 2, p. 99-109.

Vila, T., and Sillitoe, R. H., 1991, Gold-rich porphyry systems in the Maricunga belt, northern Chile:

ECONOMIC GEOLOGY, v. 86, p. 1238-1260.

Vila, T., Sillitoe, R. H., Betzhold, J., and Viteri, E., 1991, The porphyry gold deposit at Marte, northern

Chile: ECONOMIC GEOLOGY, v. 86, p. 1271-1286.

Wilson, M., 1989, Igneous Petrogenesis: A Global Tectonic Approach: Dordrecht Springer, 466 p.

Yamana Gold, 2009, Annual Report 2008, 130 p., http://www.yamana.com/Theme/Yamana/files/YAM 08%20FF_mar31.pdf

FIGURE CAPTIONS

FIG. 1. Distribution of porphyry gold systems, including prospects. Worldwide occurrences are only in the western Americas.

FIG. 2. Grade-tonnage diagram for gold in selected gold-rich porphyries worldwide. Porphyry gold systems are enclosed by the dashed line, where Colosa shows high tonnage and Au content respect to the other porphyry gold systems. Light circles stand for Au-Cu porphyries and dark circles for Cu-Au porphyries. The tonnage data is based on the sum of Inferred, Indicated, and Measured Resources, and Probable and Proven Reserves. See text for references for porphyry gold systems. Gold-rich porphyries names and references: 1 = Cadia Hill, Australia (Newcrest Mining, 2010); 2 = Didipio, Philippines (Oceana Gold, 2009); 3 = Kemess South, Canada (Northgate Exploration, 2004); 4 = Mount Milligan, Canada (Terrane Metals, 2009); 5 = Prospery, Canada (Taseko Mines, 2009); 6 = Conga, Peru (Newmont, 2009), 7 = Cerro Casale, Chile (Barrick Gold, 2009); 8 = Caspiche, Chile (Exeter Resource, 2009); 9 = Panguna, Papua New Guinea; 10 = Ok Tedi, Papua New Guinea; 11 = Grasberg, Indonesia; 12 = Cadia, NSW, 13 = Lepanto, Philippines, 14 = Bingham, Utah; 15 = Bajo de la Alumbrera, Argentina; 16 = Kal’makyr, Uzbekistan; Cooke et al., (2005) from 9 to 16. Diagram based on Seedorff et al. (2005). 36

FIG. 3. Tectonic map of Colombia and regional location of the Colosa porphyry deposit. WC = Western Cordillera; CC = Central Cordillera; EC = Eastern Cordillera. Modified from Taboada et al. (2000) and Gómez et al. (2007).

FIG. 4. Geologic map of the Colosa porphyry. Mapping by geologists of AngloGold Ashanti Colombia S.A. Elevation as m.a.s.l.

Fig. 5. Modal classification diagram of Streckeisen (1976) for plutonic silica-saturated and oversaturated rocks. The Colosa samples are classified as diorites and quartz diorites.

Fig. 6. Photomicrographs with crossed nicols of different units of the Colosa porphyry deposit, all at the same scale. A. Early unit NDE1. B. Early unit NDE2. C. Early unit NDE3. D. Intermineral unit NDI1. E. Intermineral unit NDI2. F. Intermineral unit NBXI. G. Late unit NDA. H. Late dike NDQ. I. Late dike NDQ. J. Late dike NDDL. Note the difference of texture and crystal size in the various units. Amph = amphibole, Apa = apatite, Bi = biotite, Orth = orthoclase, Pg = plagioclase, Qz = quartz, Sph = sphene. 39

Fig. 7. Geochemical classification of Colosa rocks based on major elements. A. TAS classification diagram of Wilson (1989). The Colosa samples are located in the diorite and quartz diorite fields of the subalkaline series. B. R1-R2 classification diagram of De la Roche et al. (1980). Colosa samples are located in the granodiorite, tonalite, diorite, and gabbro-diorite fields. Samples belong to the units indicated in Table 2.

Fig. 8. Major elements variation diagrams. There is a clear inverse correlation between SiO2 and P2O5, MgO, CaO, TiO2, and FeOt; whereas Al2O3, NaO2, and K2O have a more scattered pattern where a slightly positive correlation with SiO2 is observed. Symbols as in Fig. 7.

Fig. 9. Geotectonic diagrams for Colosa samples. A. AFM diagram from Irvine and Baragar (1971) indicating the calc-alkaline tendency of the Colosa porphyry. B. SiO2 vs K2O diagram from Peccerillo and Taylor (1976) showing the medium- to high-K location of Colosa within the calc-alkaline series, assuming no K metasomatism. C. Incompatible elements diagrams from Pearce et al. (1984), locating rocks from Colosa in the VAG field (VAG=Volcanic-Arc Granites, ORG=Ocean-Ridge Granites, WPG=Within-Plate Granite, and syn-COLG=syn-Collisional Granites). D. Incompatible elements diagrams from Gorton and Schandl (2000) with Colosa rocks in the field of Active Continental Margins. Symbols as in Fig. 7.

Fig. 10. Diagrams for incompatible elements of Colosa’s samples normalized to chondrite. A. Spider diagram normalized to chondritic values of Thompson (1982). Note the negative anomalies of Nb, Ta, and Ti. B. REE diagram normalized to chondritic values of Nakamura (1974). Symbols as in Fig. 7.

Fig. 11. 143/Nd/144Nd vs 87Sr/86Sr isotope correlation diagram for the rocks of Colosa. Present values of all the samples are located in the mantle array field.

Fig. 12. Analyses of microprobe on feldspars of the Colosa porphyry. A. Microprobe picture showing the mineralogic composition of the matrix of the early unit NDE3. B. Microprobe picture showing the mineralogic composition of the matrix of the intermineral unit NDI1. C. Photomicrograph with crossed nicols showing the reverse zoning in plagioclase from unit NDE3. Composition determined with microprobe analyses. Amph=amphibole, Apa=apatite, Orth=orthoclase, Pg=plagioclase, Qz=quartz, Sph=sphene.

Fig. 13. Classification diagram for amphiboles from different units of the Colosa porphyry. A. Calcic amphiboles (CaB≥1.50) with (Na+K)A≥0.50 and Ti<0.50. Most amphiboles are relatively iron-rich and classified as Ferropargasite and Ferro-endenite. B. Calcic amphiboles (CaB≥1.50) with (Na+K)A<0.50 and CaA<0.50. All amphiboles are relatively iron-rich and classified as Ferrotschermakite. Classification diagrams from Leake et al. (1997). Symbols as in Fig. 14.

Fig. 14. Temperature vs pressure diagram based on contents of Al2O3 and TiO2 from amphiboles of the Colosa porphyry. The polygons represent the ranges of temperature and pressure of early porphyries (red), intermineral porphyries (blue), late porphyries (yellow), NDQ late dike (orange), and NDL late dike (dark yellow). All points are grouped in three broad stages of crystallization (I, II, and III). Diagram modified from Ernst and Liu (1998).

Fig. 15. Schematic model for the magma origin of the Colosa porphyry gold system. Fluids from the dehydration of the Nazca plate and from dehydration and probable melting of subducted sediments ascend through the mantle wedge, causing partial melting and producing a basaltic magma that accumulates at the mantle/crust boundary. At this point the magma experiences processes of melting, assimilation, storage, and homogenization (MASH zone, after Hildreth and Moorbath, 1988). Crystal fractionation of the magma produces more evolved magmas that are emplaced at shallow depths within the South American plate. Modified from Richards (2003).

Fig. 16. Magmatic evolution of the Colosa porphyry system. A. Emplacement of early units (diorites) into the metamorphic rocks of the Cajamarca Complex in the late Miocene (~8.3 Ma). This andesitic magma comes from differentiation processes occurred in the basaltic magma chamber located under the Cajamarca Complex/Continental basement boundary. B. Emplacement of intermineral units (diorites) at the sides of the early units (8.3-7.6 Ma). C. Emplacement of late porphyries (~7.6 Ma) and dikes (~7.3 Ma). This dacitic magma comes from higher degrees of fractional crystallization and differentiation processes occurred at the basaltic magma chamber. D. Present-day state of the Colosa porphyry with the top removing any evidence for possible high- level volcanic and/or epithermal. Densities from Thorpe et al. (1984) and Restrepo-Pace (1992). Depth in kilometers below paleosurface. Textures on porphyries and magmatic chambers symbolize crystallization. 45

Fig. 17. Isotopic compositions of 87Sr/86Sr and 144Nd/143Nd for rocks from Colosa, recent volcanoes of Colombia and Marmato, and porphyries and volcanic rocks from the Maricunga belt. Note the similarity between the isotopic compositions of Colosa and Recent volcanic rocks from Colombia (James,1982; James and Murcia, 1984; Marriner and Millward, 1984; Ordoñez and Pimentel, 2001; Tassinari et al., 2008), and the difference between these such rocks and rocks from the Maricunga belt (Robinson and McKee, 1993; Kay et al., 1994; McKee et al., 1994).

TABLES

TABLE 1. Principal Features of the Known Porphyry Gold Systems in the World. Deposit Location Age (Ma)1 Host Intrusion Wall-rock Main Alteration Mineralization References Intermediate argillic Maricunga Diorite PY+CP+MT+ Lobo 13 Andesite lava (QZ veinlets Vila and Sillitoe (1991) belt, Chile porphyries HM+BOR+MO +CT+SE+Clay) Intermediate argillic Maricunga Diorite PY+CP+MT+ Vila et al. (1991); Vila Marte 13-14 Andesite lava (QZ veinlets belt, Chile porphyries HM+ BN+MO and Sillitoe (1991) +CT+SE+Clay) Refugio Vila and Sillitoe (1991); district, Diorite + dacite Andesite to Chloritic (QZ MT+PY+CP+ Verde 23 Flores (1993); Muntean Maricunga porphyries dacite lava veinlets+CT+AB) BN and Einaudi (2000) belt, Chile Refugio Diorite + Potassic (QZ Vila and Sillitoe (1991); district, Andesite to MT+HM+PY+ Pancho 23 quartz diorite veinlets Muntean and Einaudi Maricunga dacite lava CP porphyries +KFD+OG+BI) (2000) belt, Chile La Pepa Andesite to Cavancha district, Quartz diorite Potassic (Qz Muntean and Einaudi 24 dacite MT+PY+CP (La Pepa) Maricunga porphyries veinlets +BI) (2001) volcanoclastic belt, Chile Intermediate- Intermediate- Potassic (QZ MT+HM+ Zule California 4.4 composition composition Canby et al. (1993) veinlets + feldspar) Sulfides porphyry lavas Middle Diorites + Schists and Potassic (QZ PY+PO+MT+ Colosa Cauca belt, 8.0 quartz diorites Sillitoe (2008), this work quartzites veinlets +BI+KFD) CP+MO Colombia porphyries Abbreviations for minerals: BI = biotite, BN = bornite, CP = chalcopyrite, CT = chlorite, HM = hematite, KFD = potassium feldspar, MO = molybdenite, MT = magnetite, OG = oligoclase, PO = pyrrhotite, PY = pyrite, QZ = quartz, SE = sericite. 1Ages from Sillitoe et al. (1991), and Muntean and Einaudi (2001), except ages for Zule (Canby et al., 1993).

Table 2. Percentage of Primary Minerals in the Different Units of the Colosa Porphyry System. Unit Sample Pg Qz Orth Amph Bi Apa Zir Sph Ilm Classification1 NDE1 E1a 66.5 3.5 2.5 27.2 --- 0.2 0.2 trace trace Diorite NDE1 E1b 63.4 3.4 1.4 31.4 trace 0.2 0.2 trace trace Diorite NBXE1 EBX1a 80.4 1.1 2.0 16.5 --- trace trace trace trace Diorite NDE2 E2a 67.1 2.3 --- 30.4 trace trace 0.2 trace --- Diorite NDE2 E2b 72.0 2.8 --- 25.0 trace 0.2 trace trace --- Diorite NBXE2 EBX2a 74.5 0.8 --- 24.7 --- trace trace trace --- Diorite NDE3 E3a 76.4 0.8 --- 22.6 --- 0.2 trace trace --- Diorite NDE3 E3b 90.9 3.2 --- 5.6 --- 0.2 trace trace --- Diorite NDE4 E4a 89.7 2.5 --- 7.8 --- trace trace trace trace Diorite NDI1 I1a 69.6 1.9 --- 28.5 --- trace trace trace trace Diorite NDI1 I1b 76.4 1.7 --- 21.7 trace 0.2 trace trace trace Diorite NDI2 I2a 67.3 1.0 --- 31.3 0.2 0.2 trace trace --- Diorite NDI2 I2b 64.2 1.2 --- 34.3 --- 0.2 trace trace --- Diorite NBXI IBXa 61.2 0.8 --- 37.7 --- 0.2 trace trace trace Diorite NDI3 NDI3a 84.1 2.1 --- 13.2 0.6 trace trace trace --- Diorite NDA DAa 80.5 8.6 --- 5.1 5.5 0.2 trace trace trace Quartz Diorite NDA DAb 77.9 7.2 --- 9.6 5.1 0.2 trace trace trace Quartz Diorite NQD NQDa 83.2 6.7 --- 9.3 0.9 trace trace trace trace Quartz Diorite NDQ DADa 83.0 0.8 --- 16.1 0.8 0.2 trace trace trace Diorite NDQ DADb 83.8 2.6 --- 12.8 0.6 0.2 trace trace trace Diorite NDL DID 71.0 1.7 --- 27.3 --- trace trace trace trace Diorite Abbreviations for minerals: Amph = amphibole, Apa = apatite, Bi = biotite, Ilm = ilmenite, Orth = orthoclase, Pg = plagioclase, Qz = quartz, Sph = sphene, Zir = zircon. 1Modal classification according to Streckeisen (1976).

Table 3. Whole-Rock Chemical Analysis of Different Units of the Colosa Porphyry1 Unit NDE1 NDE1 NDE2 NDE2 NDE3 NDE3 NDE4 NDI1 NDI1 NDI2 NDI3 NDI3 NDA NDA NQD NDQ NDL 3NDQS 3NDQE Sample E1a E1b E2a E2b E3a E3b E4a I1a I1b I2a I3a I3b DAa DAb NQD DADa DID NDQS NDQE FeO 5.28 4.99 4.51 3.45 2.9 3.01 5.31 4.44 3.77 3.8 1.53 1.53 2.19 3.23 4.7 3.62 2.57 1.95 2.05

SiO2 55.2 55.91 61.77 59.43 61.24 61.09 60.08 59.43 59.81 61.16 65.79 65.79 65.33 66.07 60.54 63.55 62.1 62.28 62.62

Al2O3 16.85 17.01 17.2 16.61 17.26 17.18 16.9 16.7 16.96 16.97 18.28 18.28 16.01 16.8 16.15 16.15 18.28 15.88 17.74

Fe2O3 0.46 1.45 0.81 2.32 2.47 1.43 0.38 1.13 2.09 2.14 0.67 0.67 1.2 0.05 1.92 0.25 1.52 3.08 2.42 MnO 0.063 0.066 0.017 0.033 0.022 0.008 0.02 0.042 0.119 0.108 0.018 0.018 0.034 0.017 0.447 0.049 0.056 0.054 0.017 MgO 3.49 3.47 2.61 2.5 1.82 1.83 1.75 2.61 2.8 2.78 2.04 2.04 1.07 1.14 1.7 1.44 1.86 3.07 1.99 CaO 6.75 6.69 4.59 4.7 4.29 3.74 4.57 5.84 6.66 6.51 1.3 1.3 3.88 2.35 2.11 3.99 4.01 3.01 2.58

Na2O 3.63 3.57 4.33 4.24 3.86 4.08 4.28 4.38 4.06 4.03 4.24 4.24 4.74 4.56 3.85 4.46 4.78 3.96 3.57

K2O 2.45 2.29 1.95 1.66 3.57 2.65 1.56 1.86 1.55 1.5 2.63 2.63 2.04 2.19 2.08 3.28 1.33 2.25 2.52

TiO2 0.805 0.799 0.589 0.814 0.498 0.487 0.483 0.554 0.58 0.589 0.568 0.568 0.414 0.411 0.48 0.484 0.602 0.654 0.551

P2O5 0.25 0.27 0.26 0.25 0.23 0.2 0.21 0.24 0.26 0.25 0.23 0.23 0.17 0.14 0.19 0.19 0.22 0.31 0.2 LOI 3.10 3.08 2.26 3.32 1.94 3.75 2.89 1.69 1.29 1.06 2.31 2.31 2.21 2.81 5.41 1.79 2.17 2.35 4.24 TOTAL 98.33 99.59 100.9 99.33 100.1 99.45 98.43 98.92 99.95 100.9 99.61 99.61 99.29 99.77 99.58 99.25 99.5 98.85 100.5 Sc 23 23 15 16 11 11 11 15 15 15 12 12 5 6 9 7 10 15 11 Be 1 1 1 2 1 1 1 1 1 1 2 2 2 2 2 2 2 2 1 V 229 232 137 142 99 98 97 126 135 136 117 117 69 72 100 88 116 134 111 Cr < 20 < 20 30 30 < 20 < 20 20 50 60 50 30 30 20 < 20 < 20 < 20 < 20 140 20 Co 10 40 47 9 9 5 7 11 13 8 5 5 < 1 9 14 5 6 19 10 Ni < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 30 < 20 Cu 190 160 150 250 310 60 410 240 20 30 30 30 40 90 40 90 40 50 40 Zn 60 60 30 70 40 40 40 30 50 < 30 80 80 30 50 100 50 60 80 60 Ga 20 20 19 20 19 20 19 19 19 18 22 22 22 22 20 21 24 21 21 Ge 2.4 2.2 2 0.7 2.2 2.5 1.3 1.7 1.3 1.5 2.2 2.2 1.4 1.7 1.7 2.4 1.9 1.7 1.5 As 10 10 6 6 6 7 < 5 < 5 5 < 5 < 5 < 5 < 5 7 41 10 < 5 5 11 Rb 49 54 94 68 96 72 68 57 33 34 129 129 46 56 59 64 45 48 51 Sr 520 533 589 557 468 427 437 567 660 669 451 451 650 602 523 579 666 640 521 Y 19.4 18.3 16.4 17.9 18.9 19.3 19.1 17.1 17.5 18 30.7 30.7 9.7 8.5 13.2 11.9 15.3 13.6 16 Zr 85 97 111 113 103 103 138 101 119 88 124 124 165 113 108 117 129 116 126 Nb 4.7 4.6 4.6 7.9 4.5 4.3 4.5 4.3 4.4 4.7 5.5 5.5 6.5 4.6 4.9 4.8 5 6.3 5.4 Mo < 2 < 2 < 2 < 2 7 < 2 23 6 2 < 2 3 3 < 2 < 2 < 2 9 < 2 4 3 Ag < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 In 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 Sn 2 1 < 1 1 < 1 2 < 1 < 1 < 1 < 1 2 2 < 1 < 1 1 < 1 < 1 < 1 3 Sb < 0.2 < 0.2 < 0.2 0.2 < 0.2 < 0.2 0.9 4.1 3.8 1 6.7 6.7 0.4 < 0.2 < 0.2 < 0.2 < 0.2 6.5 5.4 Cs 0.7 1.4 3.5 2.9 2.4 2 2 1.3 0.6 1.2 1.1 1.1 1.5 2.3 1.5 0.7 0.7 0.5 0.6 Ba 1333 1112 344 664 1104 835 383 909 964 985 561 561 1158 1224 761 1395 291 1179 1051 La 17.5 16.3 17.9 18.7 22.5 19.8 21.9 17.4 16.7 20.7 12.7 12.7 15.5 16 20.4 17.8 27.6 16.4 17.2 Ce 34.1 31.3 33.2 33.6 38.5 36.2 39.9 34.5 33 38.6 23.3 23.3 30.1 29.7 36.6 33.2 47.2 32.1 33.8 Pr 4.59 4.17 4.28 4.46 4.74 4.51 5.18 4.02 3.92 5.15 2.57 2.57 3.9 3.82 4.31 4.26 5.45 4.04 4.02 Nd 17.2 15.8 15.5 15.9 16.3 15.7 17.8 15.7 15.3 18.4 10.4 10.4 14.5 13.5 14.7 14.8 18 15.6 15.3 Sm 3.98 3.74 3.39 3.4 3.58 3.53 3.54 3.62 3.66 3.78 2.57 2.57 3.13 2.89 2.91 3.28 3.52 3.48 3.51 Eu 1.38 1.2 0.9 1.01 0.906 0.95 1.16 1.07 1.13 1.15 0.794 0.794 0.85 0.867 1.03 0.929 1.22 1.08 1.23 Gd 3.69 3.5 2.95 3.12 3.22 3.04 3.17 3.3 3.25 3.39 3.45 3.45 2.36 2.25 2.58 2.8 2.99 3.03 3.15 Tb 0.68 0.63 0.54 0.55 0.6 0.59 0.56 0.51 0.52 0.56 0.69 0.69 0.36 0.34 0.43 0.46 0.53 0.44 0.5 Dy 3.52 3.28 2.98 2.96 3.24 3.1 2.97 2.88 2.97 3.05 4.44 4.44 1.65 1.57 2.29 2.22 2.79 2.34 2.81 Ho 0.72 0.69 0.61 0.6 0.69 0.7 0.62 0.56 0.59 0.6 0.95 0.95 0.26 0.3 0.48 0.43 0.57 0.42 0.54 Er 2.08 1.97 1.83 1.84 2.01 1.97 1.89 1.66 1.75 1.84 2.91 2.91 0.91 0.8 1.37 1.13 1.68 1.19 1.56 Tm 0.352 0.3 0.316 0.287 0.304 0.323 0.292 0.25 0.265 0.285 0.433 0.433 0.127 0.139 0.212 0.174 0.263 0.172 0.235 Yb 2.02 1.87 1.75 1.85 2.03 2.03 1.85 1.61 1.68 1.77 2.72 2.72 0.84 0.79 1.25 1.06 1.59 1.1 1.54 Lu 0.294 0.279 0.251 0.272 0.289 0.279 0.291 0.255 0.269 0.267 0.398 0.398 0.127 0.108 0.205 0.147 0.22 0.161 0.231 Hf 2.6 2.8 3.3 3 2.8 2.9 3.7 2.6 2.9 2.7 3 3 4.4 3.4 2.9 3.4 3.6 2.9 3 Ta 0.29 0.3 0.29 0.59 0.3 0.29 0.31 0.31 0.32 0.31 0.41 0.41 0.36 0.33 0.35 0.33 0.35 0.44 0.41 W 5.6 3.4 3.4 10.8 4.4 14.4 4.5 6.6 1.7 1.1 5.6 5.6 6.3 6.6 4.4 2.7 2.7 0.8 6.8 Tl 0.45 0.55 0.65 0.63 0.51 0.41 0.45 0.42 0.38 0.11 1.1 1.1 0.25 0.92 0.64 0.32 0.35 0.54 0.64 Pb 5 7 7 < 5 < 5 < 5 < 5 < 5 < 5 < 5 < 5 < 5 < 5 8 10 < 5 < 5 5 < 5 Bi 1.8 2.1 0.6 0.1 0.6 0.5 < 0.1 < 0.1 0.1 < 0.1 < 0.1 < 0.1 0.3 0.7 3.1 7.5 0.5 < 0.1 0.2 Th 3.05 3.15 3.62 4.04 3.33 3.01 3.33 4.38 4.38 4.21 4.4 4.4 3.61 3.17 3.19 3.45 2.83 3.61 4.28 U 1.47 1.59 1.83 1.71 1.64 1.73 1.64 1.45 1.72 1.71 2.16 2.16 1.63 1.52 1.59 1.59 1.03 1.82 1.52 La/Yb 9 9 10 10 11 10 12 11 10 12 8 5 18 20 16 17 17 15 11 Ce/Yb 17 17 19 18 19 18 22 21 20 22 16 9 36 38 29 31 30 29 22 Ba/La 76 68 19 36 49 42 17 52 58 48 28 44 75 77 37 78 11 72 61 Rb/Cs 70 39 27 23 40 36 34 44 55 28 160 117 31 24 39 91 64 96 85 2Q 4.3 6.5 13.3 14.5 13.1 16.1 14.1 10.0 12.4 14.1 26.0 26.0 20.4 23.1 21.7 13.5 17.8 21.2 24.8 48

C 0.0 0.0 0.2 0.0 0.0 1.3 0.4 0.0 0.0 0.0 6.8 6.8 0.0 3.1 4.4 0.0 2.3 2.3 5.1 Or 15.2 14.0 11.7 10.2 21.5 16.4 9.6 11.3 9.3 8.9 16.0 16.0 12.4 13.3 13.1 19.9 8.1 13.8 15.5 Ab 32.3 31.3 37.1 37.4 33.3 36.1 37.9 38.1 34.8 34.2 36.9 36.9 41.3 39.8 34.6 38.7 41.6 34.7 31.4 An 23.6 24.5 21.4 22.3 19.6 18.0 22.3 21.0 23.8 23.8 5.1 5.1 16.9 11.1 9.8 14.7 19.0 13.4 11.9 Di 8.2 6.6 0.0 0.3 0.5 0.0 0.0 5.9 6.5 5.6 0.0 0.0 1.5 0.0 0.0 3.6 0.0 0.0 0.0 Hy 13.6 12.7 13.4 9.6 6.9 8.5 13.6 10.3 8.4 8.6 6.6 6.6 4.5 8.3 12.0 7.7 7.4 8.0 6.1 Mt 0.7 2.2 1.2 3.5 3.6 2.2 0.6 1.7 3.1 3.1 1.0 1.0 1.8 0.1 3.0 0.4 2.3 4.6 3.6 Il 1.6 1.6 1.1 1.6 1.0 1.0 1.0 1.1 1.1 1.1 1.1 1.1 0.8 0.8 1.0 0.9 1.2 1.3 1.1 Ap 0.6 0.7 0.6 0.6 0.6 0.5 0.5 0.6 0.6 0.6 0.6 0.6 0.4 0.3 0.5 0.5 0.5 0.8 0.5 1Major element-oxides and normative minerals in percent; trace elements in ppm. 2Normative minerals: Q=quartz, C=corundum, Or=orthoclase, Ab=albite, An=anorthite, Di=diopside, Hy=hypersthene, Mt=magnetite, Il=ilmenite, and Ap=apatite. 3NDQS and NDQE are other late dikes in the Colosa system.

Table 4. 143Nd/144Nd and 87Sr/86Sr Isotopic Compositions of Rocks of the Colosa Porphyry System Unit 143Nd/144Nd Nd (ppm) Sm/Nd 87Sr/86Sr Sr (ppm) Rb/Sr NDE1 0.512870 15.8 0.24 0.704289 533 0.10 NDI2 0.512842 18.4 0.21 0.704265 669 0.05 NDI3 0.512797 16.5 0.23 0.704510 547 0.04 NDA 0.512797 13.5 0.21 0.704279 602 0.09 NQD 0.512759 14.7 0.20 0.704368 523 0.11

Table 5. Compositions of Feldspars from the Colosa Porphyry System 1 Unit Feldspar SiO2 Al2O3 Na2O F K2O MgO CaO MnO FeO Cr2O3 TiO2 Total Composition

NDE1 Pg 1 core 58.23 26.72 6.43 0.23 0.35 0.00 8.40 0.03 0.23 0.00 0.00 100.62 An41

NDE1 Pg 1 rim 65.40 22.42 9.76 0.00 0.45 0.00 2.85 0.00 0.06 0.00 0.03 100.97 An14

NDE2 Pg 2 57.68 26.91 6.59 0.17 0.23 0.00 8.55 0.03 0.16 0.02 0.00 100.33 An42

NDE3 Pg 3 a 53.66 29.33 4.89 0.11 0.23 0.00 11.56 0.00 0.22 0.00 0.00 99.99 An57

NDE3 Pg 3 b 56.50 27.83 5.95 0.23 0.24 0.00 9.56 0.01 0.27 0.01 0.00 100.60 An47

NDE3 Pg 3 c 57.81 26.73 6.61 0.00 0.27 0.00 8.34 0.00 0.26 0.00 0.00 100.02 An41

NDE3 Pg 3 d 57.50 26.86 6.54 0.27 0.30 0.00 8.49 0.00 0.26 0.00 0.03 100.25 An42

NDE3 Pg 3 e 54.44 29.35 5.11 0.00 0.16 0.00 11.13 0.00 0.22 0.00 0.02 100.43 An55

NDE3 Pg 3 f 53.48 29.98 4.68 0.05 0.10 0.00 11.98 0.01 0.26 0.00 0.00 100.54 An59

NDI1 Pg 4 core 55.18 28.25 5.86 0.00 0.14 0.00 10.37 0.00 0.14 0.00 0.00 99.95 An43

NDI1 Pg 4 rim 57.65 26.86 6.51 0.00 0.17 0.00 8.80 0.00 0.11 0.00 0.03 100.13 An50

NDI2 Pg 5 57.05 26.73 6.47 0.00 0.17 0.00 8.74 0.03 0.15 0.05 0.05 99.45 An42

NDA Pg 6 core 58.92 26.13 7.19 0.22 0.17 0.02 7.56 0.00 0.19 0.00 0.03 100.44 An37

NDA Pg 6 rim 57.87 26.50 7.10 0.00 0.23 0.00 8.14 0.00 0.14 0.04 0.05 100.07 An39

NDQ Pg 7 59.06 25.42 6.98 0.33 0.41 0.00 7.35 0.00 0.14 0.00 0.00 99.70 An36

NDL Pg 8 55.04 27.78 5.61 0.00 0.18 0.00 10.17 0.00 0.26 0.02 0.06 99.12 An50

NDE1 Orth 1 64.17 18.84 0.45 0.05 15.47 0.00 0.00 0.00 0.04 0.01 0.03 99.07 Or96

NDI2 Orth 2 62.33 18.62 0.91 0.17 13.48 0.21 0.15 0.01 0.50 0.00 0.10 96.48 Or90 1Pg = plagioclase; Orth = orthoclase.

Table 6. Compositions of Amphiboles from the Colosa Porphyry System Unit NDE1 NDE1 NDE1 NDE1 NDE2 NDE2 NDE2 NDE2 NDE2 NDE2 NDE3 NDE3 NDE3 NDE3

SiO2 46.91 49.85 52.78 53.81 40.81 41.70 40.34 43.23 42.41 42.06 43.93 46.35 47.46 42.76 MgO 14.57 13.79 15.94 16.57 9.97 10.62 9.99 12.02 11.95 10.78 14.69 14.24 14.09 12.80

Na2O 1.71 0.88 0.52 0.32 2.24 2.05 2.33 1.87 2.02 2.16 2.28 1.58 1.45 2.08

K2O 0.53 0.31 0.19 0.09 0.70 0.61 0.57 0.67 0.83 0.56 0.69 0.78 0.33 0.89

Al2O3 7.48 4.91 3.22 2.34 14.60 13.35 14.50 12.90 11.96 13.40 11.52 9.33 6.96 12.33 CaO 11.44 11.92 12.24 12.31 10.87 10.92 11.23 11.13 11.05 10.86 11.51 11.21 11.13 11.37 MnO 0.39 0.47 0.25 0.33 0.38 0.37 0.34 0.31 0.39 0.41 0.14 0.24 0.52 0.25 FeO 12.09 13.67 11.96 10.97 16.42 15.90 16.58 13.52 14.65 15.85 11.60 12.95 12.94 13.03

Cr2O3 0.00 0.02 0.01 0.03 0.04 0.01 0.03 0.03 0.04 0.00 0.09 0.04 0.00 0.06

TiO2 1.97 0.88 0.49 0.28 1.21 1.33 1.30 0.85 1.74 1.17 1.84 0.90 1.36 1.63 Total 97.23 96.69 97.62 97.04 97.25 96.87 97.22 96.54 97.03 97.24 98.29 97.62 96.25 97.20 Formula coefficients based on 13eCNK

Si (IV) 6.82 7.31 7.55 7.69 6.04 6.17 6.00 6.34 6.25 6.19 6.30 6.67 6.94 6.26 Al (IV) 1.18 0.69 0.45 0.31 1.96 1.83 2.00 1.66 1.75 1.81 1.70 1.33 1.06 1.74 Al 0.11 0.15 0.09 0.09 0.58 0.50 0.53 0.58 0.33 0.51 0.24 0.25 0.13 0.39 Mg 3.16 3.01 3.40 3.53 2.20 2.34 2.21 2.63 2.62 2.36 3.14 3.06 3.07 2.79 Fe 2+ 0.98 1.38 1.10 1.03 1.15 1.09 1.24 0.93 0.99 1.05 0.64 0.73 0.91 0.94 Fe 3+ 0.49 0.29 0.33 0.29 0.89 0.87 0.82 0.73 0.81 0.90 0.75 0.83 0.67 0.65 Mn 0.05 0.06 0.03 0.04 0.05 0.05 0.04 0.04 0.05 0.05 0.02 0.03 0.06 0.03 Ti 0.22 0.10 0.05 0.03 0.13 0.15 0.15 0.09 0.19 0.13 0.20 0.10 0.15 0.18 Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.01 Ca 1.78 1.87 1.87 1.88 1.72 1.73 1.79 1.75 1.74 1.71 1.77 1.73 1.74 1.78 Na 0.49 0.25 0.15 0.09 0.64 0.59 0.67 0.53 0.58 0.62 0.63 0.44 0.41 0.59 K 0.10 0.06 0.04 0.02 0.13 0.12 0.11 0.12 0.16 0.11 0.13 0.14 0.06 0.17 [ ] 0.64 0.82 0.95 0.98 0.50 0.57 0.43 0.59 0.52 0.57 0.47 0.69 0.79 0.46 OH 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 Classification1 F-e F-e F-e F-e F-p F-p F-t F-p F-p F-p F-t F-e F-e F-t Abbreviations: F-e = ferro-edenite, F-p = ferropargasite, F-t = ferrotschermakite, P = pargasite. 1Classification based on 13eCNK normalization (Schumacher, 1997) and on diagrams of Leake et al. (1997). Table 6. Compositions of Amphiboles from the Colosa Porphyry System Unit NDI1 NDI1 NDI1 NDI1 NDI1 NDI2 NDI2 NDI2 NDI2 NDI2 NDI2 NDI2 NDA NDA

SiO2 44.47 41.26 44.39 41.43 42.33 40.95 45.31 42.09 42.22 42.10 41.46 40.69 43.28 46.16 MgO 11.62 9.44 15.78 9.45 10.39 9.45 17.11 9.96 10.14 11.19 8.96 9.09 10.42 11.81

Na2O 1.61 2.22 2.00 2.05 2.09 1.85 2.40 1.73 1.85 2.03 1.75 1.85 1.63 1.33

K2O 0.47 0.62 0.93 0.63 0.80 0.73 0.87 0.67 0.56 0.68 0.66 0.79 0.73 0.62

Al2O3 11.40 14.55 10.84 14.15 13.60 14.46 9.94 13.09 14.07 14.03 14.06 14.88 10.54 8.17 CaO 11.27 10.79 11.55 10.94 10.91 10.78 10.76 10.87 10.52 10.68 10.82 10.91 11.30 11.56 MnO 0.34 0.33 0.17 0.33 0.23 0.39 0.08 0.39 0.60 0.27 0.40 0.41 0.48 0.43 FeO 14.64 16.05 10.16 16.53 15.23 17.20 9.08 17.10 16.30 14.52 16.78 17.50 16.95 16.02

Cr2O3 0.15 0.00 0.18 0.00 0.00 0.00 0.00 0.00 0.10 0.00 0.00 0.02 0.00 0.02

TiO2 0.91 1.85 1.51 0.59 1.02 1.15 1.46 0.91 0.63 1.68 0.87 0.99 1.11 0.83 Total 96.86 97.12 97.49 96.11 96.59 96.94 97.03 96.81 96.98 97.18 95.75 97.13 96.43 96.95 Formula Coefficients based on 13eCNK

Si (IV) 6.53 6.12 6.37 6.21 6.29 6.07 6.45 6.24 6.19 6.15 6.24 6.04 6.49 6.83 Al (IV) 1.47 1.88 1.63 1.79 1.71 1.93 1.55 1.76 1.81 1.85 1.76 1.96 1.51 1.17 Al 0.50 0.67 0.20 0.71 0.67 0.60 0.12 0.53 0.63 0.57 0.73 0.65 0.36 0.26 Mg 2.54 2.09 3.37 2.11 2.30 2.09 3.63 2.20 2.22 2.44 2.01 2.01 2.33 2.61 Fe 2+ 1.14 1.38 0.42 1.35 1.30 1.16 0.07 1.17 0.91 0.91 1.40 1.25 1.47 1.43 Fe 3+ 0.66 0.61 0.80 0.72 0.60 0.97 1.01 0.95 1.09 0.86 0.71 0.93 0.65 0.55 Mn 0.04 0.04 0.02 0.04 0.03 0.05 0.01 0.05 0.07 0.03 0.05 0.05 0.06 0.05 Ti 0.10 0.21 0.16 0.07 0.11 0.13 0.16 0.10 0.07 0.19 0.10 0.11 0.13 0.09

Cr 0.02 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 Ca 1.77 1.72 1.78 1.76 1.74 1.71 1.64 1.73 1.65 1.67 1.74 1.74 1.82 1.83 Na 0.46 0.64 0.55 0.59 0.60 0.54 0.66 0.49 0.53 0.57 0.51 0.53 0.47 0.38 K 0.09 0.12 0.17 0.12 0.15 0.14 0.16 0.13 0.10 0.13 0.13 0.15 0.14 0.12 [ ] 0.68 0.53 0.50 0.53 0.51 0.62 0.54 0.65 0.72 0.63 0.62 0.58 0.57 0.67 OH 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 Classification F-e F-p F-p F-p F-p F-p P F-p F-p F-p F-p F-p F-p F-e

Table 6. Compositions of Amphiboles from the Colosa Porphyry System Unit NDA NDQ NDQ NDQ NDQ NDQ NDQ NDL NDL NDL NDL NDL NDL

SiO2 42.25 45.59 43.49 45.78 43.73 40.58 42.05 45.15 43.58 44.74 43.11 42.81 41.86 MgO 9.44 11.89 10.53 12.47 10.64 9.82 9.32 12.76 12.63 12.57 12.21 12.82 10.79

Na2O 1.89 1.50 1.58 1.42 1.71 2.36 1.75 1.94 2.11 2.20 2.39 2.18 2.33

K2O 0.98 0.49 0.85 0.47 0.68 0.61 0.99 0.67 0.62 0.48 0.59 0.60 0.72

Al2O3 11.60 8.91 10.12 8.49 10.39 13.95 11.47 9.09 10.65 9.14 10.92 11.57 12.64 CaO 11.18 11.28 11.24 11.36 11.20 11.31 11.20 11.09 11.31 11.24 11.26 11.42 11.25 MnO 0.55 0.43 0.53 0.50 0.55 0.26 0.55 0.49 0.40 0.55 0.38 0.34 0.38 FeO 18.16 15.37 17.76 15.04 16.92 16.61 18.64 14.31 13.98 14.19 13.91 12.54 14.66

Cr2O3 0.05 0.02 0.04 0.03 0.01 0.03 0.00 0.01 0.02 0.03 0.00 0.02 0.06

TiO2 1.22 0.98 1.19 0.91 1.19 1.58 1.19 1.59 1.94 1.88 2.30 2.06 1.63 Total 97.30 96.47 97.31 96.48 96.99 97.09 97.14 97.08 97.23 97.02 97.06 96.33 96.32 Formula Coefficients based on 13eCNK

Si (IV) 6.34 6.75 6.48 6.76 6.51 6.07 6.33 6.64 6.41 6.61 6.38 6.33 6.28 Al (IV) 1.66 1.25 1.52 1.24 1.49 1.93 1.67 1.36 1.59 1.39 1.62 1.67 1.72 Al 0.39 0.31 0.25 0.23 0.33 0.52 0.36 0.21 0.25 0.20 0.28 0.35 0.51 Mg 2.11 2.62 2.33 2.74 2.36 2.19 2.09 2.80 2.77 2.77 2.69 2.83 2.41 Fe 2+ 1.63 1.28 1.41 1.14 1.41 1.44 1.61 1.13 1.08 1.26 1.26 1.06 1.43 Fe 3+ 0.65 0.62 0.81 0.71 0.70 0.63 0.74 0.63 0.63 0.49 0.46 0.50 0.41 Mn 0.07 0.05 0.07 0.06 0.07 0.03 0.07 0.06 0.05 0.07 0.05 0.04 0.05 Ti 0.14 0.11 0.13 0.10 0.13 0.18 0.13 0.18 0.21 0.21 0.26 0.23 0.18 Cr 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 Ca 1.80 1.79 1.79 1.80 1.79 1.81 1.80 1.75 1.78 1.78 1.79 1.81 1.81 Na 0.55 0.43 0.46 0.41 0.49 0.68 0.51 0.55 0.60 0.63 0.68 0.63 0.67 K 0.19 0.09 0.16 0.09 0.13 0.12 0.19 0.13 0.12 0.09 0.11 0.11 0.14 [ ] 0.46 0.69 0.59 0.71 0.59 0.39 0.50 0.58 0.50 0.50 0.42 0.45 0.38 OH 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 Classification F-t F-e F-p F-e F-e F-t F-p F-e F-p F-e F-t F-t F-t