A systemic approach to complex landslide risk reduction in coastal tourist areas: The case of Sirolo, central Italy

Coastal landslides in structurally complex and tourist-rich areas, such as the Monte Conero promontory in central Italy, pose significant challenges to environmental stability and public safety. In these settings, complex landslides often involve deep-seated translational movements evolving into sudden debris collapses at the footslope, generating significant hazards in beach areas heavily frequented by tourists. This study develops a conceptual, transferable model for complex landslide behavior in the Sirolo coastal sector in an integrated Hazard–Exposure–Vulnerability framework. The hazard component is investigated by systematically integrating multi-source data, including COSMO-SkyMed and Sentinel-1 Differential Interferometric Synthetic Aperture Radar (DInSAR) data, in-situ inclinometer and piezometric time series, and high-resolution geological and geomorphological field mapping. Rainfall analyses and an Extreme Rainfall Periodic Index (ERPI) were used to explore the correlation between precipitation distribution and kinematic responses of the slope, offering an empirical insight into seasonal hazard modulation. To assess exposure, we developed a deep learning model to estimate beach attendance using limited Google Earth imagery (5 useable acquisition dates, moderate resolution), calibrated with regional tourism statistics, enabling a spatially and temporally explicit assessment of human presence. Finally, to explore vulnerability, a face-to-face survey was carried out to document gaps in visitor awareness of risk and civil-protection procedures. This integrated approach offers enhanced predictive capacity and supports targeted mitigation and communication measures in coastal environments characterized by complex landslides and intense seasonal human activity. To the authors knowledge, this is the first study integrating geological monitoring, rainfall indices, satellite data, deep learning-based exposure, and tourist risk perception into a systemic coastal landslide risk framework.