Disentangled attribution of human flood impacts

During my month-long stay at the Weather and Climate Risks group at ETH Zurich, I worked on human flood impact (displacement, fatalities) attribution by disentangling the different drivers: hazard, exposure, and vulnerability.

The idea is, first, to create a model that could identify which predictors are the most important using a Bayesian hierarchical model, and then to perform the attribution based on a Shapley algorithm, using time series of factual and counterfactual data.

The STSM resulted in fruitful exchanges with Professor David Bresch and his group, along with scientists at ETH Zurich. I would like to thank the Weather and Climate Risks group for their amazing welcome and their openness to collaboration- this was impressive.

I would also like to thank COST Action CA24163 for their support and dedication to assessing climate impacts.