Stepwise clustering ensemble downscaling for future drought prediction under climate change: A case study of the Yangtze River Basin

Abstract

Drought is a highly destructive natural disaster that can even lead to other complex and volatile extreme weather events in the Yangtze River Basin (YRB). Researching potential changes in future drought occurrences is crucial for developing effective climate change adaptation strategies under a warming climate. In this study, a stepwise clustered ensemble drought downscaling (SCEDD) model has been proposed, and two internationally recognized drought indices (the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI)) are used to evaluate drought events in YRB. The validation results show that the proposed model can be used to generate robust climate projections (daily precipitation) based on simulation. On this basis, SPI and SPEI are calculated from historical and future forecasts (Tmean and precipitation). The study suggests that predicted precipitation will increase dramatically in the future (2075–––2099) at all stations under the SSP245 and SSP585 scenarios. Drought events under SPI increased by 1.5% and decreased by 1.2% during 2075–––2099 under SSP245 and SSP585, respectively. In contrast, drought events in SPEI increased by 16.5% and 33.7%, respectively. In addition, drought events under SPI present an increase in spring, summer, and fall, with a decrease in winter under both SSPs. For comparison, drought events under SPEI will increase substantially in all seasons. In particular, drought predictions made using SPEI are more accurate and trustworthy because SPEI takes temperature into account. The results obtained can provide decision-makers in the water resources management sector with a reliable forecast of drought and minimize drought-related losses.