Combining magma flow models and deformation measurements to understand magma ascent at silicic volcanoes

"Volcanic deformation can be caused by a broad range of processes related to the movement and evolution of magma.The type of processes detectable depends on the temporal and spatial resolution of the measurement method. Satellite radar measurements have so far most often captured reservoir-related processes, with a modal source depth of ~ 5km (Ebmeier et al., 2018). GPS or tilt measurements made near the volcanic conduit are more likely to capture transient signals associated with magma transport, but have been often modelled and interpreted as inflation or deflation of a shallow magma reservoir (Hautmann et al., 2009). More recently, shear stress has been identified as an additional source of deformation (Neuberg et al.2018; Marsden et al. 2019) which takes magma flow in shallow, silicic conduit systems into account. Hence, deformation models can now describe a range of processes between the pressurization of deep reservoirs and shallow magma ascent. So far, most models treat magma as a liquid, with a volume change that can account for surface displacements.

In this project we intend to combine deformation models and magma flow models to incorporate three-phase fluid flow (melt, crystals, volatiles), pressurization through crystallization (second boiling), degassing and outgassing, and thermal boundary layers into models of magmatic processes (Marsden et al., 2019). These magmatic processes affect the possible range of viscosities at different depths, as well as the range of pressures that can act at reservoir depth and shallow conduit level.

By combining the modelling of shallow magma flow and magmatic processes with the deformation patterns originating at reservoir depths , this project will gain new insights into the large variety of cyclic behaviours of silicic volcanoes. Examples of volcanoes with well-documented cyclic behaviour and deformation on multiple spatial scales include Soufriere Hills in Montserrat, West Indies and Tungurahua, Ecuador. . For both volcanoes we have access to extensive data sets which will form the data base for the investigation, but further data sets such as seismicity and petrological data will be taken into account as well.