Katharine Cashman

Understanding the mechanical evolution of magmatic systems requires careful assessment of their rheological characteristics, particularly in light of the growing evidence that magma reservoirs are dominated by magma‐mush surrounded by thermally‐altered host‐rock. To address this complexity, we develop models for volcano deformation based on a poroviscoelastic source within a thermo‐viscoelastic host. We use Finite Element modeling to investigate the rheological and mechanical response to melt injection in a Maxwell poroviscoelastic reservoir hosted in a temperature‐dependent standard linear solid viscoelastic (“thermo‐viscoelastic”) crust. Our models consider the competing roles of poroelastic diffusion, and viscoelastic creep and relaxation. All cause time‐dependent post‐injection deformation. Post‐injection deformation of a poroviscoelastic reservoir in an elastic crust is dominated by poroelastic diffusion, a consequence of the short relaxation timescale of a hot and relatively low viscosity mush. For cooler, more viscous mush, the magnitude of viscoelastic deformation increases, supplementing the deformation caused by post‐injection poroelastic diffusion. Thermo‐viscoelasticity of the crust amplifies the poroelastic deformation response of the magma‐mush, leading to increased time‐dependent deformation both during and after melt injection. The rate of post‐injection surface deformation decreases at a rate proportional to the reservoir temperature. Crucially, our model sensitivity analysis demonstrates that the wall rock thermo‐viscoelastic response contributes more to surface deformation than the viscous effect of the magma‐mush. For this reason, neglecting the viscoelastic properties of the host rock and the poroelastic properties of the reservoir in interpretations of surface deformation data could produce errors in inferred processes (e.g., injection duration) and subsurface characteristics (e.g., reservoir compressibility, shape and depth).