KEY FACTORS IN THE GENESIS OF NORILSK, SUDBURY, JINCHUAN, VOISEY BAY AND OTHER WORLD-CLASS NI-CU-PGE DEPOSITS - IMPLICATIONS FOR EXPLORATION

Magmatic Ni-Cu sulfide deposits form as the result of the segregation and concentration of droplets of liquid sulfide from mafic or ultramaf ic magma and the partitioning of chalcophile elements into these from the silicate melt. Sulfide saturation of a magma is not enough in itse lf to produce a massive concentration of magmatic sulfide. The appropr iate physical environment is required to concentrate sulfides from a l arge mass of magma into one place. The deposits of the Noril'sk region have developed within flat, elongate bodies (15 X 2 X 0.2 km) that in trude argillites, evaporites and coal measures, adjacent to a major, t ranscrustal fault and immediately below the centre of a 3.5 km-thick v olcanic basin. An anticlinal axis that transects the axis of the basin at a high angle has brought these intrusions to surface to give rise to the two major ore junctions, Noril'sk and Talnakh. Studies of the o verlying basalts have shown that basalts forming a 500 m-thick sequenc e have lost 75% of their Cu and Ni and more than 90% of their PGE. Bas alts above this show a gradual recovery in their chalcophile element c oncentrations to reach 'normal' values 500 m above the highly depleted zone. Two groups of mineralised bodies have been identified as correl ative with these basalts: the poorly mineralised Lower Talnakh-type bo dies, which resemble the highly depleted basalts; and the ore-bearing Noril'sk-type intrusions which correlate with the overlying, essential ly undepleted basalts. The high proportion of sulfide (2-10 wt%) assoc iated with the Noril'sk-type intrusions, the high PGE content of the o res, the extensive metamorphic aureole (100-400 m around the bodies), and the heavy sulfur isotopic composition of the ores (+8 to +12 delta (34)S) are explicable if the ore-bearing bodies are exit conduits from high-level intrusions, along which magma has flowed en route to surfa ce. The Lower Talnakh bodies are interpreted as intrusions along which magma flow stopped earlier than along those of the Noril'sk type. The first magma to enter the high-level intrusion reacted with much evapo ritic sulfur, at a low 'R' value and thus gave rise to sulfides with l ow metal tenors. Successive pulses of magma through the system progres sively enriched the sulfides in the conduits, losing progressively les s of their chalcophile metals, and thus accounting for the upward incr ease in metals in successive lava flows. The flow direction along the conduits is shown by the direction in which the tenor of disseminated sulfides decreases. Sulfides have settled from the moving magma to for m separate injections of liquid sulfide, up to 3.5 X 1.5 X 0.05 km in size. Recent reflection seismic studies at Sudbury have shown the Sudb ury Igneous complex to have been much more extensive than originally s upposed. Nd and Sr isotopic studies on rocks of the Sudbury Igneous Co mplex and Re-Os studies on the ores have indicated the incorporation o f much country rock gneiss in the complex. Debate has centred around w hether the Sudbury Igneous Complex is entirely an impact melt or the c onsequence of the mixing of primary magma with 50 or more wt% impact m elt; the most recent evidence favours the latter hypothesis. The Jinch uan deposit of north-central China occurs within a 6 km-long dyke-like body of peridotite. The compositions of olivine within the dyke, the igneous rocks themselves, and the ore are all inconsistent with deriva tion of the body from ultramafic magma, as originally supposed, and in dicate that the structure forms the keel of a much larger intrusion of magnesian basalt. Flow of magma into the intrusion has resulted in ol ivine and sulfide being retained within this keel. The Voisey's Bay de posit lies partly within a 30-100 m-thick sheet of troctolite, which i s interpreted as a flat-lying part of a feeder for an adjacent intrusi on, and partly at the base of the intrusion, where the feeder adjoins it. Ore types range from disseminated sulfides in troctolite, which in crease downward, grading into massive ore. The latter is underlain by a breccia composed of fragments of gneiss, unmineralised troctolite an d peridotite in a troctolitic matrix that, in places, contains appreci able sulfide. When most major Ni-Cu sulfide deposits, including those at Kambalda, Western Australia, are viewed in the light of studies at Noril'sk, Sudbury, Jinchuan and Voisey's Bay, three factors become app arent: (i) the concentration of sulfides in channels or conduits throu gh which much ma,oma has flowed (feeder conduits for intrusions are mu ch more prospective targets for exploration than the base of the intru sions themselves); (ii) the interaction of the source magma with count ry rocks, either leading to the incorporation of sulfur, or the felsif ication of the magma in question; and (iii) fractional crystallisation of sulfide liquid giving rise to Cu-rich ores which may be far remove d from the 'source' ore.