How much ash to sink a vessel? Development of vulnerability functions for modelling capsizing risk under volcanic ashfall

Volcanic ashfall has only recently been recognized as a destabilizing hazard for small craft, as evidenced by vessel capsizing during the 2011–2012 Cordón Caulle eruption in Argentine Patagonia. However, the ash loads required to compromise vessel stability remain poorly constrained. This study presents the first operational methodology to quantify capsizing risk under volcanic ash loading through an analytical vulnerability framework. Using the Maxsurf software suite for numerical hydrostatic and stability modelling, large-angle static and dynamic simulations were conducted for three vessel types representative of those operating in Patagonian lakes: a rigid inflatable boat, a motor sailboat, and a small passenger vessel. Simulations combined three Beaufort-scale environmental forcing conditions with fourteen deck-accumulation thicknesses, two bulk densities, and an additional snow-loading scenario documented during the motivating case studies. Reductions in key stability indices were quantified and used to derive vessel-specific vulnerability curves. Additionally, three damage states (DS0-DS2) were defined by translating international intact-stability criteria into discrete categories of progressive stability loss, enabling identification of ashfall thresholds for each state. The resulting vulnerability model was then evaluated for applicability and predictive performance against documented incidents from the 2011–2012 Cordón Caulle eruption, using a combined metric-based and spatially informed retrospective validation approach. Predicted impacts show strong agreement with observed damage, demonstrating that documented ashfall accumulations exceeding 150 mm can plausibly induce vessel capsizing under realistic combinations of ash-density, snow-loading, and sea-state conditions. Minor biases were identified, including slight under- and overestimation for larger and smaller vessels, respectively, while intermediate damage states (DS1) remained less constrained because of rapid transitions from safe conditions (DS0) to failure (DS2). By resolving the coupled controls on vessel stability through systematic pre- and post-processing, the framework yielded ashfall thresholds broadly consistent with previously proposed generic limits, demonstrating its potential as a transferable basis for interpreting past incidents and informing risk assessment for vessels exposed to volcanic ashfall.