Multi-natural hazard mapping for critical infrastructures in complex territorial contexts: proposal of a novel...

This study presents a novel methodological approach for multi-hazard assessment (MHA) aimed at evaluating the exposure of critical infrastructures to natural hazards in complex territorial contexts. The proposed approach integrates harmonized hazard maps into a unified geospatial framework including five natural hazards relevant to the Sicilian region: seismic, geomorphological (landslides), hydraulic (floods), volcanic, and tsunami. The assessment relies only on those natural hazards for which authoritative and regionally coherent datasets are available. Other natural and non-natural hazard categories were not considered in this study. Three different weighting strategies for composing the multi-hazard index (MHI) from the five hazards were tested: i. equal weighting - strategy 1 (S1); ii. presence-based dynamic weighting - strategy 2 (S2); iii. interaction-based dynamic weighting - strategy 3 - (S3), which accounts for hazard interdependencies through adapted interaction matrices. In this study, wastewater treatment plants (WWTPs) were selected as representative critical infrastructures. The proposed methodological approach was tested in Sicily (Italy), a region where multiple hazards coexist and interact. The results demonstrated that the choice of weighting strategy significantly influenced the spatial distribution and classification of hazard levels. While equal weighting (S1) led to a conservative classification, the presence-based approach (S2) produced a more responsive distribution. The interaction-based approach (S3), which incorporates compound and cascading effects among hazards, emerged as the most balanced, reducing the risk of over- or underestimation of the single hazard. The application of the MHA to Sicily's WWTPs based on S3 revealed that approximately 80 % of facilities are located in areas classified as high or very high hazard. This underscored the critical vulnerability of these infrastructures and highlights the need for hazard-based planning. Future developments should aim to integrate infrastructure-specific vulnerability data to further advance multi-risk modelling frameworks.