Himalayan leucogranites are hotter than previously thought
Cenozoic leucogranites in the Himalayan orogenic belt which was developed in response to the collision of Indian and Asian plates at ca.60 Ma are generally accepted as typical examples of low-temperature and pure crustal melts.The leucogranites provide a window to decipher the nature of anatectic sources and physicochemical conditions of partial melting at depth.However,problems are encountered in determining the temperature of crustal anatexis for the leucogranite magmatism.Most of them were previously considered as originating from fluid-absent partial melting of metasedimentary rocks by muscovite dehydration at temperature lower than 800℃.Cenozoic Himalayan leucogranites have been generally regarded as low-temperature magmas(＜800℃),and previous models mostly considered that they were produced by intracrustal heating through shearing heating associated with the movements of the Main Central Thrust and the South Tibetan Detachment System,decompression melting caused by the exhumation of the Higher Himalayan Crystallization Sequence,and radiogenic heating as a result of the radioactive decay of U,Th and K.However,previous approaches including experimental petrology and accessory mineral thermometers probably have underestimated the actual crustal anatectic temperatures.Recent studies using phase equilibrium modeling indicate that many granulites sampled from a wide range of the Himalayan orogen experienced partial melting at T≥800℃ in the Miocene.We argue that the maximum temperatures recorded by leucogranites should most closely approximate the true melting temperatures in the source.In this study,three separate approaches have been used to constrain the initial temperature of leucogranites,and these results show that the leucogranites are actually formed at higher temperatures(＞800℃).Firstly,through a comparison of experimental melts derived from metapelites and Himalayan leucogranites,it is demonstrated that high-T melts(820-900℃)have A/CNK(Al2O3/(CaO+Na2O+K2O),molar ratio)values consistent with the majority of granites(1.0-1.3),but the low-T melts(750-800℃)are more peraluminous(1.3-1.4),although both of the high-T and low-T melts can be comparable in all major elements with the leucogranites.Secondly,this study's newly analyzed data,combined with existing literature,demonstrates that zircon saturation temperatures are lower than many of the individual Ti-in-zircon temperatures recorded in the samples.This feature indicates that the primary magma is actually undersaturated with zircon and the zircon saturation is achieved by early crystallization of Zr-poor minerals.The maximum Ti-in-zircon temperatures in several samples exceed 800℃.Thirdly,pseudosection modeling using an internally consistent thermodynamic dataset and appropriate activity-composition(a-x)models was carried out on two representative two-mica granite samples.The liquidus temperatures at a crystallization pressure of 5 kbar is ca.800℃,and the liquidus temperature increases at P＞5 kbar.Therefore,if leucogranitic magmas were formed at deeper crust(8-10 kbar),their formation temperatures are even higher.Consequently,we conclude that the high liquidus temperatures reveal the Himalayan leucogranites were hotter initially than thought based on zircon temperatures.Such high temperatures are difficult to explain purely by the internal heating of the thickened orogenic crust.Instead,they require an extra heat source,which would be probably provided by upwelling of asthenospheric mantle subsequent to thinning of the orogenic lithospheric mantle by foundering along the convergent plate boundary.Therefore,the Himalayan leucogranites of Miocene age would be derived from partial melting of the metasedimentary rocks in the post-collisional stage.The formation of leucogranites by crustal anatexis could facilitate the onset of exhumation of the Higher Himalayan Crystallization Sequence and the initial movements of the South Tibetan Detachment System and the Main Central Thrust.
NETL
NSTL
万方数据
