Improving the spatial mapping of historical climate impacts by integrating hazard and exposure layers

Historical impacts of hydro-climate extremes collected in existing global disaster databases are typically recorded at the country or subnational administrative level. Such coarse spatial resolution strongly masks the spatial variability of phenomena and limits the assessment of the potential underlying environmental and human drivers. Here, we develop a new global spatially explicit database of impacts of hydro-climate extremes by integrating hazard and exposure layers. We focus on fatalities and economic damage caused by heatwaves, cold waves, droughts, and floods occurred over the 1981-2019 period. Impact records following the occurrence of hydro-climate extremes are initially derived from existing disaster databases. For each reported impact we identify those grid cells, within the administrative unit under consideration, that experienced a hydro-climate hazard at the time of the recorded event. Spatiotemporal dynamics of hydro-climate hazards are derived using the flood-fill algorithm applied to ETCCDI indicators retrieved from ERA5-Land reanalysis data. This allows us to identify spatially coherent patterns of hydro-climate extreme conditions within a three-dimensional data cube (space-time). The reported impact is finally distributed across grid cells subject to hydro-climate hazard and using local GDP and population density as weights retrieved from high resolution global products. Results are confronted with independent observational and modeled assessments of hydro-climate impacts. This new database offers a unique contribution to improving the quantitative estimation of global socioeconomic vulnerabilities to hydro-climate extremes and the consequent risks associated with climate change.