Magma viscosity and its ascent rate are key factors in controlling eruption style. Shear stress exerted on the conduit walls as magma ascends pulls up the surrounding edifice, whilst overpressure pushes the edifice outwards.
Magma fractures if shear stress exceeds its shear strength, triggering low-frequency seismicity. Shear stress is proportional to both the viscosity of magma and its ascent velocity. Hence, it provides an important link between ascent dynamics and both deformation and seismicity that can be recorded at the surface. Tiltmeters measure changes in inclination, and both shear stress and pressure have been linked conceptually to changes in tilt recorded close to the conduit. However, how much shear stress and pressure are produced as magma ascends, and the relative contribution of each to the tilt, has not previously been quantified.

Firstly, flow and deformation modelling are combined using COMSOL Multiphysics to quantitatively link magma ascent and tilt. Despite shear stress being several orders of magnitude smaller than overpressure at most depths, shear stress generally dominates the tilt signal. Next, I systematically investigate how topography influences tilt, showing how topography controls both the amplitude and polarity of the tilt, and thus the relative contribution of shear stress and pressure. 3D deformation modelling is performed including real volcanic topography to show how a tiltmeter can be strategically deployed at the location most sensitive to changes in source stress. Finally, time-dependent flow modelling is used to show how magma ascent dynamics, and thus shear stress and overpressure, evolve through time due to transient volcanic processes. The growth of a lava dome exerts an increasing loading pressure at the conduit vent that impedes magma ascent, and can cause it to stall even if conditions at depth remain unchanged. By unloading, a full or partial dome collapse can therefore cause an eruption to recommence.

By quantitatively linking magma ascent and deformation, and examining how ascent evolves through time, this work shows the importance of combining flow and deformation modelling in retrospectively investigating what drives temporal variations in seismicity and deformation. This is an important step towards being able to develop a combined forecasting tool using both seismicity and deformation that can be used to detect critical changes in ascent dynamics.