Deconstructing the role of hydrothermal alteration in 3D volcanic dome collapse in La Soufrire de Guadeloupe

Volcanic domes are inherently unstable structures as they grow incrementally, with varied extrusion rates, material properties, and directions of flow. These instabilities can bring about volcanic dome collapse, leading to turbulent and hot avalanches of material that can devastate communities surrounding a volcano, as well as affecting the volcanos eruptive dynamics. The objective of this research is to investigate the effect of hydrothermal alteration on dome stability, where hydrothermal alteration typically results in mechanical weakening of volcanic rock. To achieve this goal, it is important to understand the spatial distribution of alteration within a dome. The internal structure of the La Soufrire de Guadeloupe dome was mapped by Heap et al. (2021), whereby electrical conductivity surveys were carried out to obtain the rock porosity and therefore the density variation within the dome. The density contrasts were correlated with mechanical parameters (i.e., uniaxial compressive strength of volcanic rock) to obtain a 3D internal strength map of the volcano. We designed a novel methodology to input this geophysical data into a new 3D particle-based Discrete Element Method model. This involves creating a digital elevation model from satellite data and interpolating the geophysical data to assign strengths to each modelled particle. We show here the results of alteration scenario testing. This involves varying the degree of alteration-induced weakening, the spatial extent, and the size of alteration zones. This allows us to make predictions on potential for collapse and direction of material flow, quantify collapse volumes, and explore small-scale to large-scale failures. In particular, we investigate the effect of ongoing alteration of the La Soufrire de Guadeloupe dome. Observations show pervasive hydrothermal alteration, particularly in an area in the south of the dome known as the bulge. This represents a potential detachment plane and thus is a focus for our collapse models. Thus far, key findings from our investigations suggest that even near-surface alteration can cause deep-seated deformation. We also show varied weakening scenarios for a bulk rock strength of 10% and 50% of the original strength, with a focus on the southern flank of the dome. To date, no 3D dynamic models of stability exist and therefore these models are key to forecasting volcanic hazards as a result of hydrothermal alteration.