Landslide Magnitude Exceedance Probability Modelling for Ribeira Quente Valley (São Miguel Island, Azores-Portugal)

Landslide size is a strong predictor of runout distance across a wide range of landslide types and therefore represents a key parameter for hazard assessment. Within the conceptual risk framework, landslide hazard analysis requires estimating the probability of exceedance of landslide magnitude, in a manner analogous to approaches commonly applied to other natural hazards, such as earthquakes. Integrating magnitude–probability relationships into landslide hazard assessments enhances the robustness of potential impact characterization and supports informed risk-based decision-making.

Situated in the North Atlantic, the Azores archipelago comprises nine volcanic islands where numerous destructive landslide events have occurred over the past five centuries, triggered by seismic activity, volcanic eruptions, and intense rainfall. Within this context, this study focuses on the Ribeira Quente valley (Povoação Municipality, São Miguel Island), covering 9.15 km². The study area exhibits high susceptibility to landslide occurrence, characterized by very friable volcanic deposits and extremely steep slopes. Landslides frequently affect the only access road to Ribeira Quente village, leaving it isolated. Since 1900, 31 landslides events have affected Ribeira Quente parish, causing 32 fatalities. A major event on 31 October 1997 triggered nearly 1,000 shallow landslides, resulting in 29 fatalities, the destruction of 36 houses, and 114 people left homeless, while the village remained isolated for over 12 hours.

Three historical landslide inventories were compiled. The first inventory, based on 2004 data, included ~400 landslides. The second, from 2010, contained ~250 landslides. The third, compiled in 2025, identified ~260 landslides. Overall, the inventories include approximately910 landslides, mainly superficial translational slides and debris flows.

The main objective of this study is to propose and parameterize probability distributions specifically tailored to the study area. The landslide scar areas were used as the magnitude descriptor. A total of 65 theoretical probability distributions were fitted to the scar area data. Parameterization was performed using the maximum likelihood method, and goodness of fit was evaluated with the Kolmogorov–Smirnov (K-S) test. The best-fitting probability density function (PDF) was then selected, and exceedance probabilities for different magnitude scenarios were computed based on its complementary cumulative distribution function (1 − CDF).

This study provides a probabilistic approach for assessing landslide magnitudes, presenting valuable insights for land-use planning and civil protection. The derived magnitude–exceedance functions enhance hazard characterization and can guide the prioritization of risk mitigation actions and targeted geotechnical investigations. This research was supported by the INTERREG program through the PRISMAC project – “Análise, Mitigação e Gestão do Risco de Movimentos de Vertente Potenciados pelas Alterações Climáticas na Macaronésia” (Ref. 1/MAC/2/2.4/0112).