By Ryan Campbell, Kristina Grolmusova, Lydia Kakampakou, Jeongjin Lee<\/p>\n\n\n\n
Abstract<\/strong><\/p>\n\n\n\n Motivated by the EVA 2025 Data Challenge<\/strong>, we address the problem of predicting extreme rainfall<\/strong> in the eastern United States using data from a large ensemble of climate model runs<\/strong>. The challenge focuses on three quantities of interest<\/strong> related to the spatial extent and\/or temporal duration of extreme rainfall, each requiring extrapolation<\/strong>. To tackle these questions, we adopt the recently developed geometric framework<\/strong> for extreme-value analysis, offering substantial flexibility for capturing complex extremal dependence structures and enabling extrapolation across the entire multivariate tail. In this work, we focus on the spatial geometric framework<\/strong> for analysing the spatial extent<\/strong> and consider a sampling procedure that retains the temporal information<\/strong> in the data, thereby enabling estimation of the duration of extreme rainfall events. We also account for the non-stationary behaviour<\/strong>, arising from topographical and seasonal effects, that commonly characterises extreme weather events in both space and time. Using diagnostic metrics, we demonstrate that the proposed model is appropriate for inferring extreme events on this dataset and apply it to estimate target quantities of interest.<\/p>\n\n\n\n\n