Population Exposure to Rainfall-Driven Flooding from Hurricane Maria in Puerto Rico

Abstract

Climate change is making rainfall associated with tropical cyclones more extreme. Some of the places most affected by tropical cyclones are small islands, such as Puerto Rico in the Caribbean which was severely impacted by Hurricane Maria in 2017. However, we know very little about how sensitive flooding in small islands is to changing rainfall characteristics, or how population exposure to flooding might change in the future. This is due to the limited data availability necessary to produce high-resolution flood hazard and population exposure estimates for a wide range of possible scenarios. Using an island-scale (~9000km2) event-based rainfall-driven hydrodynamic flood model at 20m resolution for the island of Puerto Rico, we simulate a range of observed rainfall grids from Hurricane Maria across time and space (such as IMERG and NCEP Stage IV). We assess how the current population exposure to rainfall-driven flooding changes across the range of observation rainfall footprints to determine how sensitive the flood extent and population exposure is to different rainfall inputs. We also compare these outputs to flood extents produced using an event set of synthetic hurricane rainfall events that share similar rainfall and track characteristics to Hurricane Maria under current and future climate scenarios (1.5°C and 2°C). Additionally, we utilise high-resolution (90m) gridded estimates of future population in Puerto Rico (FuturePop), to determine how an event with the same extreme magnitude as Hurricane Maria would impact population exposure to flooding under different future Shared Socioeconomic Pathway scenarios. The results of this analysis aim to improve understanding regarding the range of plausible estimates of current and future population exposure to flooding in Puerto Rico. These results will help inform adaptation to more extreme flood risk in Puerto Rico under current and future climate change.