Impact of topography and water load on magma propagation modelling

Coastal and submarine volcanoes are characterized by complex topographies, a significant portion of which lies below sea level, complicating efforts to fully quantify how surface geometry influences magma transport. Understanding the coupling between topography, stress fields, and magma propagation is essential for assessing volcanic hazards, including dike-fed eruptions and edifice instability. 

Conventional models of dike propagation commonly approximate volcanic edifices as simplified surface loads, thereby neglecting the spatially variable stress perturbations introduced by realistic topography and bathymetry. To overcome this limitation, we develop a two-dimensional Boundary Element Model for fluid-filled fractures that explicitly incorporates a discretized free surface. This approach enables direct coupling between detailed topography and magma-driven deformation, allowing magma pathways to dynamically respond to surface geometry.

We implement the model geometry in COMSOL Multiphysics to compute stress under four representative scenarios: (1) a flat surface with an imposed surface load, (2) a symmetric volcanic edifice, (3) an asymmetric edifice, and (4) an asymmetric edifice subjected to an additional water load, with gravitational forces included in all cases. These end-member configurations are designed to isolate the effects of topography and water loads on magma propagation.

Preliminary results indicate that incorporating realistic topography significantly alters dike trajectories, fracture geometries, and associated stress and displacement patterns compared to simplified surface-load models. The presence of asymmetric topography and water loads further enhances stress heterogeneity, with implications for both magma ascent pathways and slope stability. These findings highlight the importance of explicitly resolving topography and marine loading when interpreting deformation signals and assessing hazards in coastal and submarine volcanic systems.