Abstract Fault zones that cut Paleoproterozoic Birimian Supergroup sedimentary and mafic volcanic rocks in southwestern Ghana, west Africa, host numerous gold deposits that form one of the richest mesothermal lode gold provinces in the world. The Ashanti gold deposit is the largest discovered to date in west Africa, with past production and current reserves exceeding ~1,200 tonnes (t) of gold. A complex multiphase deformation history is evident in the Birimian sedimentary rocks that host the deposit. The prominent northeast-striking structural grain and fold-thrust belt architecture that characterizes the Paleoprotero zoic rocks of southwestern Ghana was established during regional-scale southeast-directed shortening (D 2 ) after development of a widespread bedding-parallel cleavage (S 1 ). A further minor episode of southeast-directed shortening (D 3 ) overprints D 2 . Structures associated with D 1 -D 3 are folded around 300- to 500-m- scale upright folds (F 4 ) that plunge to the northeast and have axial planes that strike ~east-west and dip 50° to 80° N. Upright folding was followed by development of north-striking, small-displacement, sinistral strike-slip faults (D 5 ) and local sinistral reactivation of some older D 2 thrust faults. Disseminated auriferous arsenopyrite grains in rocks adjacent to the mineralized faults are either localized on or cut the crenulation cleavage associated with the F 4 folds, which implies that gold mineralization occurred towards the end of, or after, F 4 . Mineralization along the faults themselves is hosted in quartz vein arrays that commonly have sinistral asymmetries at scales ranging from a few centimeters to several hundred meters, implying that the main gold event occurred during D 5 . Mineralized faults locally cut across F 4 folds without deflection, again implying that ore deposition occurred after F 4 folding. Ore shoots within the Ashanti deposit and adjacent satellite deposits are predominantly structurally controlled and are located in the following: Dilatant and subordinate compressional sites where mineralized shear zones step left and right, respectively, across F 4 kink folds and reactivated D 2 transfer faults; In pressure shadows associated with volcanic units, felsic and granitoid intrusions within the sedimentary sequence; At the intersections of major structures that were active during mineralization. The Ashanti deposit as a whole occupies an ~8-km-long segment of an otherwise unmineralized northeast-striking D 2 thrust fault known as the Obuasi/Main Reef fissure. Sinistral reactivation of this specific fault segment during the D 5 mineralization event occurred in response to movement on the younger north-striking Ashanti fissure, which merges with the Obuasi/Main Reef fissure at the northern end of the Ashanti deposit. The southern end of the mine is marked by a sharp right-hand flexure in the Obuasi fissure where it steps across a D 2 transfer zone. Recognition of these structural controls on mineralization allowed extensions to ore shoots within the Ashanti deposit to be targeted with a greater degree of confidence and has led to delineation of significant additional resources. Similar structural sites were targeted during exploration of the surrounding area using “integrated” geologic maps that combined the results of geologic mapping, airborne geophysical surveys, soil geochemical data, aerial and satellite photography, and local costeaning. Detection of mineralized faults was best achieved with a combination of geologic mapping, soil geochemical surveys, and costeaning. Routine recognition of structural sites similar to those noted above is probably only possible with geologic mapping at scales larger than 1:50,000. Attempting to remotely detect 200- to 400-m-long bends in poorly exposed faults was the most difficult aspect of this program. However, the detailed understanding of the timing and structural controls on mineralization gained in the mine area is a powerful exploration tool in its own right, which allows the significance of scattered structural observations to be appreciated and incorporated into a robust targeting strategy.

Abstract Obuasi, with a total mineral resource plus past production of 70 Moz, is the largest gold deposit in West Africa, and one of the largest in the world. It is hosted by ~2135 Ma siliciclastic rocks of the Eburnean Kumasi Basin, which were obliquely shortened along an inverted boundary with the older Eoeburnean Ashanti belt to the east. Greenschist facies metamorphism was coeval with mineralization and related alteration at ~2095 Ma. The steeply dipping, ENE-plunging lodes extend over an 8-km strike length and to depths of >2.5 km. They include paragenetically complex gold-rich quartz veins surrounded by refractory auriferous arsenopyrite and closely associated carbonate-muscovite alteration halos in deformed carbonaceous phyllites and subordinate metaigneous host rocks. Gold and arsenic were initially precipitated during deformation-assisted interaction with reduced host rocks at ~350°C and 100 to 200 MPa. The mineralizing fluids were derived primarily from deeper, As-rich metasedimentary sources by basinal fluid expulsion and metamorphic devolatilization triggered by inversion and shortening, followed by transpression. Continued fluid injection during and after the metamorphic peak produced changes in gold fineness, sulfide assemblages, repeated dissolution (stylolites) and reprecipitation of mineralized veins, and a change from early deformed shear-related, sulfide-rich lodes to later quartz-rich lodes that plunge down or across the axes of younger transpressional folds. Channelized fluid flow due to reactivation of basin-edge transfer structures, and/or irregularly distributed gold source rocks, may explain the variation in gold endowment along the former basin boundary.

Abstract In May 2003, AngloGold Ashanti began greenfields exploration in Colombia with a team of four geologists. By 2007, the program employed 127 field geologists covering about 10.5 million hectares (Mha) with systematic reconnaissance exploration. To date, the result of this work is the discovery of several gold deposits, the most important being La Colosa, containing an initial resource of 381.4 million metric tons (Mt), grading 1.00 g/t Au or 381.4 t Au, using a 0.3 g/t cut-off grade. The La Colosa deposit is a gold-only porphyry system related to a late Miocene multiphase porphyritic diorite-granodiorite complex. Gold grades exceeding 1 g/t are associated with early dioritic phases that are altered to potassic and sodic-calcic mineral assemblages. Potassic and sodic-calcic alteration also affects later diorite porphyries, but gold grades are, on average, <0.4 g/t. A late granodiorite porphyry is mostly barren, with only erratic anomalous gold grades, which are all <0.4 g/t, and weak to moderate propylitic and intermediate argillic alteration. The deposit contains >5 vol percent magnetite. Pyrite content varies between 3 and 5 vol percent. Gold is mainly contained within pyrite. Copper and molybdenum contents are generally at background values for diorite. The La Colosa deposit is a grassroots discovery. It was made by AngloGold Ashanti geologists only 18 months after the initiation of a regional exploration program in the defined Mariquita target region. Discovery is the result of systematic regional data synthesis, conceptual target generation, and disciplined, multiphase, field-based, follow-up, which included stream sediment geochemistry, prospecting, rock chip sampling, and drilling. Early target generation work by AngloGold Ashanti, undertaken at a northern Andean scale between 2000 and 2003, focused field activities into the most prospective regions of Colombia, based not only on geology and mineral potential but also upon factors that would lead to the discovery and eventual development and operation of a successful, socially, and environmentally responsible mining operation. The exploration strategy has maintained a systematic methodology that includes conceptualization, reconnaissance stream sediment surveys and related prospecting, target definition, target drilling and conceptual economic study, and finally prefeasibility and feasibility studies. Clear decision points were established at the end of each work phase, always keeping in mind the company’s minimum economic target criteria. The key factors that led to the discovery of La Colosa included the execution of a well-planned business and exploration model, with recognition of the “first mover” advantage; the acquisition of a large land position with respect to legal exploration tenure, covering essentially all of the deemed prospective areas; implementation of exploration from the regional scale, working down to the prospect scale, instead of vice versa; maintaining a disciplined and systematic field-focused approach; use of a skilled exploration team; and maintaining a long-term (>5 yrs), adequately funded view to exploring in frontier mineral exploration regions.

Abstract The region of the gold-rich Ashanti belt in southern Ghana was chosen as the subject for a detailed regional thermal modelling study. Geological studies, in addition to laboratory measurements of thermal properties and heat-production rates, allow us to constrain a finite-element thermal modelling. Scenarios intergrating variations of the structure of the crust and various chronological settings were examined. We calculated the thermal regime before and after the thrust tectonism that affected the region during the Eburnean orogeny (2130–2095 Ma), just before ore deposit formation. This gives a new insight into the regional thermal state of the crust before the mineralizing events. To satisfy the thermobarometric observations, the most probable mantle heat flow must be 60 mW m −2 , which is at least three times greater than the present-day value. At shallow depths, our results also indicate anomalies of lateral heat flow reaching 25 mW m −2 , focused on the margins of each lithological unit, including the Ashanti belt. These anomalies are related to the distortion of the isotherms in the first few kilometres that can be explained mostly by lateral contrasts in thermal conductivity. Such anomalies could be of importance for the mineralizing events, as they would favour fluid circulation locally.

Abstract The Geita mine is operated by AngloGold Ashanti and currently comprises four gold deposits mined as open pits and underground operations in the Geita greenstone belt, Tanzania. The mine produces ~0.5 Moz of gold a year and has produced ~8.3 Moz since 2000, with current resources estimated at ~6.5 Moz, using a lower cut-off of 0.5 g/t. The geologic history of the Geita greenstone belt involved three tectonic stages: (I) early (2820–2700 Ma) extension (D 1 ) and formation of the greenstone sequence in an oceanic plateau environment; (II) shortening of the greenstone sequence (2700–2660 Ma) involving ductile folding (D 2–5 ) and brittle-ductile shearing (D 6 ), coincident with long-lived igneous activity concentrated in five intrusive centers; and (III) renewed extension (2660–2620 Ma) involving strike-slip and normal faulting (D 7–8 ), basin formation, and potassic magmatism. Major gold deposits in the Geita greenstone belt formed late in the history of the greenstone belt, during D 8 normal faulting at ~2640 Ma, and the structural framework, mineral paragenesis, and timing of gold precipitation is essentially the same in all major deposits. Gold is hosted in iron-rich lithologies along contacts between folded metaironstone beds and tonalite-trondhjemite-granodiorite (TTG) intrusions, particularly where the contacts were sheared and fractured during D 6–7 faulting. The faults, together with damage zones created along D 3 fold hinges and D 2–3 hydrothermal breccia zones near intrusions, formed microfracture networks that were reactivated during D 8 . The fracture networks served as conduits for gold-bearing fluids; i.e., lithologies and structures that trap gold formed early, but gold was introduced late. Fluids carried gold as Au bisulfide complexes and interacted with Fe-rich wall rocks to precipitate gold. Fluid-rock interaction and mineralization were enhanced as a result of D 8 extension, and localized hydrofracturing formed high-grade breccia ores. Gold is contained in electrum and gold-bearing tellurides that occur in the matrix and as inclusions in pyrrhotite and pyrite. The gold mineralization is spatially linked to long-lived, near-stationary intrusive centers. Critical factors in forming the deposits include the (syn-D 2–6 ) formation of damage zones in lithologies that enhance gold precipitation (Fe-rich lithologies); late tectonic reactivation of the damage zones during extensional (D 8 ) faulting with the introduction of an S-rich, gold-bearing fluid; and efficient fluid-rock interaction in zones that were structurally well prepared.