Uncertainties in multi-model ensemble drought projections and implications on future drought risk

Future drought projection studies typically use multi-model ensemble climate and hydrological simulations. In particular, precipitation, soil moisture, and streamflow simulations are used to quantify the changes in meteorological, agricultural, and hydrological droughts under future climate. Many different drought indices have thus been developed and employed in these projections with different indices often leading to varying states of future droughts. Recently, terrestrial water storage (TWS) has also been used to examine future droughts considering integrated climatic and hydrologic impacts on water stores. This presentation will shed light on drought projections using precipitation, soil moisture, runoff, and TWS drought indices and highlight uncertainties in these projections, including those arising from differences in drought definition or the diversity in drought indices. The presentation will then discuss how the consideration of vulnerability alters drought risk projections, specifically by incorporating human development projections as a proxy of broad vulnerability. Results presented will be based on several dozen ensemble hydrological simulations that include multiple climate models, hydrological models, Representation Concentration Pathways (RCPs), and Shared Socioeconomic Pathways (SSPs). Emphasis will be placed on global scale analyses and regional projections over drought hotspots. The results have appeared in three recent publications.