This research project focuses on improving flood prediction in the Alps by analyzing compound climate events like rain-on-snow flooding. It aims to integrate statistical and machine learning methods to create a robust model that can predict flood scenarios and their probabilities. This model aims to support disaster preparedness and contribute to climate resilience by utilizing satellite data to explore new combinations of flood hazards and their impacts.
This project integrates:
The modelling framework:
Predicts from unusual or unseen combinations of events, like rain-on-snow. This is achieved through statistical advances to predict the compounding effect of events leading to flooding.
Estimates hazard frequency more effectively (e.g., the 100-year flood), emerging as a result of the model being informed of the events leading to the flooding. This results in more informed infrastructure design and better risk-informed decision making.
Explains the reason behind flooding by answering questions like: What paths in the event tree are most likely to lead to a 100-year flood? What is the probability of a 200-year flood, given a large snowpack this winter? This allows stakeholders to explore critical ‘what-if’ scenarios, making informed decisions in emergency response planning and resource management.
Provides a clearer picture of flooding under climate change from the improved understanding of events leading up to flooding.
Dr. Vincenzo Coia is an expert in environmental statistics and hydrological research, focusing on innovative approaches to modeling complex climate-related hazards. His work includes developing models for flood and drought prediction, integrating machine learning with traditional hydrological methods. Dr. Coia has contributed to projects such as the Nicola Valley Floodplain Mapping and Flooding under Climate Change in Western Canada, which have supported government initiatives and improved floodplain mapping. His development of the Probaverse R Packages and hydrotechnical machine learning models reflects his ongoing commitment to advancing scientific computing practices.
Dr. Carlo De Michele is a respected figure in hydrology and water resource engineering at the Politecnico di Milano. His work focuses on multivariate modeling of hydrologic and geophysical extreme events, contributing significantly to the understanding and management of these phenomena. Prof. De Michele has guided numerous students and researchers, fostering a collaborative environment that advances the field. His involvement in international collaborations and training schools highlights his commitment to education and research in climate-related events.
This research is supported by the European Space Agency’s Open Space Innovation Platform (OSIP) and the Politecnico di Milano. You can find the ESA’s project page here. We extend our gratitude to our collaborators and funding partners for their invaluable contributions.