Flash flood impacts and vulnerability mapping at catchment scale: Insights from southern Apennines

Flash floods are frequent natural hazard events in many parts of the world. Generally, they occur in small catchments drained by torrential streams that feed alluvial fans or fan deltas. In the Mediterranean region, these phenomena are particularly common during the spring and autumn seasons, often causing significant damage to buildings, infrastructures, agriculture, and sometimes resulting in fatalities and injuries. To better understand and manage the potential consequences of these events on physical systems, probabilistic damage quantification is essential. Fragility functions, which describe the probability of reaching or exceeding a specific damage state based on an intensity measure, are valuable tools for assessing damage conditioned on the intensity of a natural hazard. While such curves are widely reported and extensively applied, there is a notable lack of interdisciplinary methodologies for their development and integration into broader risk management frameworks. This gap often leaves initiatives such as flood insurance premium planning, probabilistic loss estimation, and flood risk management reliant on uninformed or generic tools. This study proposes an interdisciplinary approach to developing flood fragility functions using post-event flash flood damage data. The event that occurred on 14–15 October 2015 in Solopaca – Paupisi area (Benevento, Italy) is adopted as the case study. The reactivation of alluvial fan lobes is analyzed together with the recorded rainfalls. It is based on the processing of post-event field data acquired with classical and remote sensing technologies such as UAV imagery. Impact mapping is then conducted to depict the spatial extent of the flash flood. The event is then characterized in terms of inundation depth and thickness of mobilized material and grain size distribution. The area of the event and the thickness of the deposits are considered to estimate the transported solid volumes. Finally, the damage incurred to buildings and respective inundation depth is assembled to construct flash flood fragility functions. The outcomes of this study can be used in numerical flow model calibration and validation as well as flash flood risk assessment and management initiatives. The fragility functions developed in this study can serve as a tool for loss assessment, resilient construction prioritization, and insurance premium planning. The interdisciplinary approach developed and implemented in this study will be insightful to many other regions across the world in terms of flash flood mitigation planning