LEUKOGRANITES OF THE HIMALAYA KARAKORAM - IMPLICATIONS FOR MAGMATIC EVOLUTION WITHIN COLLISIONAL BELTS AND THE STUDY OF COLLISION-RELATED LEUKOGRANITE PETROGENESIS

Two types of leucogranite have been identified in a collisional setting in the Himalaya. Type-a (e.g., Manaslu, High Himalaya) are syn-tectonic crustal melts related to "hot over cold" thrusting and subsequent fluid fluxing. Type-b(e.g., Baltoro, Karakoram) is related to lower crustal melting during/following crustal thickening and thermal re-equilibration, followed by crystal fractionation. For type-b, melting is triggered/promoted in the lower crust by input of heat and material from the sub-continental lithospheric mantle which is contemporaneously melting due to removal of the slab input and downward displacement. Both these granite types can be explained in terms of recent thermal models for continental collision, and so might be expected to be found in other orogenic terranes. These models also show that extension and thinning will occur as a result of the crust becoming gravitationally unstable, due to thickening, some time after collision. Thinning will lead to upwelling and melting of asthenosphere which could then intrude the base of the crust and promote further melting. This could explain the common occurrence of A-type granitoids as final plutonic events in orogenic terranes following intrusion of the collision-related type-a and type-b granites. Existing genetic classification schemes only classify granites in terms of source-type or broad environmental affinity (e.g., collisional). To further subdivide such broad environmental fields it is considered that processes and a general evolutionary model must be used. Source-type is not suitable due to the number of source components that are often involved and the overlap of these components from one environment to another. One or two selected features of a granite are not, however, sufficiently exclusive to identify different petrogenetic histories. We propose that an evaluation of field style, mineralogy, isotope and elemental geochemistry should be used to subdivide the collisional granite field into type-a and type-b (or Manaslu and Baltoro-types) respectively.