Abstract:
Extreme weather events, from river flooding to droughts and tropical cyclones, are likely to become both more severe and more common in the coming years as a result of climate change and over-exploitation of natural resources, further worsening the potential impacts. The damages caused by these extreme events will be felt across all sectors of society, particularly in the form of economic loss to productivity and physical assets. In the face of this threat, policy and decision-makers are increasingly calling for approaches and tools able to support risk management and climate adaptation pathways that can capture and deal with the full extent of multi-sectoral damages.
In the frame of this thesis, building on the state-of-the-art research in the field of Machine Learning, a GIS-based Bayesian Network (BN) approach was developed exploiting damage data collected for the agricultural, residential, and industrial sectors from the 2014 Secchia river flooding event. The designed BN approach was able to capture and model multi-sectoral damages under future ‘what-if’ scenarios, standing for potential changes in the vulnerability (e.g. land use change) and hazard patterns (i.e. flooding scenarios under different return periods) within the investigated case.
The methodology will provide valuable support for disaster risk management and reduction against river flooding events, giving a sound picture on their multi-sectoral impacts, in line with the objectives of relevant EU’s acquis.
