Mapping a novel metric for Flash Flood Recovery using Interpretable Machine Learning

This data pertains to the paper titled "Mapping a novel metric for Flash Flood Recovery using Interpretable Machine Learning" published in the Journal of Hydrometeorology. In this paper we develop a new metric called Recoveriness to estimate the recovery potential of watersheds after flash floods. Using 78 years of historical flood data and advanced machine learning techniques, we provide probabilistic estimates of flash flood recoveriness across the conterminous United States. This approach models the recession limb of the hydrograph, which is essential in understanding post-flood recovery but has been less studied compared to the rising limb.