Mobilisation of magmatic mushes: an experimental study with implications for large-scale volcanic eruptions

点击翻译按钮获取中文摘要

Volcanic eruptions constitute a major societal and economic hazard. Eruptions of silicic magmas are the result of magma ascent from crustal reservoirs to the surface. The counterparts of erupted silicic magmas are fully crystallized bodies of magma (or 'granitic plutons') that never reached the surface and crystallized totally in the crust. The knowledge of the destiny of crustal magma bodies is important because the largest eruptions observed on the Earth (e.g. Fish Canyon Tuff, Coloradao, volume>5000 km3) are the result of the re-mobilization of sub-volcanic magmatic bodies. Several giant volcanoes such as Yellowstone (USA) and Uturuncu (Bolivia) have been identified as potential candidates for large volume eruptions. During their residence in the crust magmas cool and, as a consequence, crystallize. The increase of crystals content has been demonstrated to increase magma viscosity (the resistance of a material to flow under the application of a stress). On the contrary, addition of heat and/or volatiles, such as H2O, related to injection of hot, volatile rich magma at the base of the cooling reservoir can lead to re-melting of the crystals, decrease of viscosity and re-mobilization of plutons. Despite the widespread occurrence of these processes in nature, and their undoubted importance for triggering major eruptions, no quantification of them is possible due to the lack of appropriate experimental datasets dedicated to this problem. Two main factors limit our ability to forecast the destiny of a magmatic body residing in the Earth crust: a) the lack of petrological experiments characterizing accurately the relative proportions of solid and liquid as function of temperature and water content and b) the lack of appropriate rheological (or 'flow') models describing the viscosity variations for mixtures of crystals with different sizes and shapes. This study aims to reconstruct the chemical evolution of magmas as functions of temperature and water content using high temperature and pressure apparatuses, characterize accurately the products of the experimental runs and reproduce their characteristic physical properties using analogue materials. The measurements of the variation of viscosity as function of crystal content and shape will link the chemical evolution of magmatic mushes with their rheological properties. The combination of petrological and rheological experiments will make it possible to understand how the rise of magmas from depth to the surface occurs and to quantify the amount of hot, volatile-rich magma that needs to be supplied before the remobilization of the magmatic mush occurs. This would be of extreme interest in volcanoes like Yellowstone and Uturuncu where the size and rate of the uplift give information on the amount and rate of magma supplied at the base of the magmatic reservoir. Additional experiments will be performed to relate the velocity of propagation of seismic waves through samples containing crystals in different proportions and shapes. These data can be used to invert the seismic data obtained from natural earthquakes to identify and characterize magmatic reservoirs present below active volcanic areas. The combination of the results obtained on propagation of compressional and shear waves with the petrological and rheological results will give the possibility to evaluate the risk associated with the presence of bodies of partially molten magma below active volcanic areas.