Ensemble Drought Exposure Projection for Multifactorial Interactive Effects of Climate Change and Population Dynamics: Application to the Pearl River Basin.

With a changing climate, drought has become a common natural disaster. In our study, population exposure to drought over the Pearl River Basin (PRB) under climate change is investigated. Drought frequency is evaluated through the Standardized Precipitation Evapotranspiration Index (SPEI). The data needed for SPEI calculation are obtained based on the ensemble of multiple global climate model (GCM) outputs. Population exposure to drought for the future is assessed by combining drought frequency under two emission scenarios (RCP4.5 and RCP8.5) with three downscaled population scenarios (SSP1, SSP2, and SSP3). Moreover, the main contribution (and their interactions) of GCM, RCP, and SSP to the sources of uncertainty on population exposure projections is explored through multilevel factorial analysis. Results indicate that the temperature and precipitation would continually increase for the future, and the increase in drought frequency is more substantial during the 2080s than in the 2050s. Meanwhile, population exposure to drought accounts for 21.60% of the total population in 1976–2005 over the PRB area. During the 2050s, it would decrease to 11.98–12.28% under RCP4.5 and 14.15–14.40% under RCP8.5, respectively. By the 2080s, population exposure would slightly reduce under RCP4.5 and increase to 28.86–29.44% under RCP8.5. GCM is the primary uncertainty source of drought exposure in the 2050s, with contribution rates of 72.41%, 57.47%, 51.10%, and 78.71% to the four responses. In comparison, by the 2080s, RCP is the primary contributing factor, with contribution rates of 53.91%, 44.92%, 64.20%, and 48.00%, respectively. Plain Language Summary: Climate change has an extensive impact on future drought conditions in the Pearl River Basin. Therefore, the projection of population exposure to drought is essential for making informed decisions toward climate change mitigation and adaptation. This study aims at exploring the changes in population exposure to droughts (e.g., moderate, severe, and extreme) under multiple emission‐population scenarios and quantifying the main contribution (and their interactions) of global climate model (GCM), RCP, and SSP to the sources of uncertainty on population exposure. The temperature and precipitation would continually increase for the future, and the increase in drought frequency is more substantial during the 2080s than in the 2050s. As a result, future population exposure to moderate, severe, and extreme drought would decrease in the 2050s, under two RCPs. While during the 2080s, there is a slight decrease under RCP4.5 and an apparent increase under RCP8.5. GCM is the primary uncertainty source of drought exposure in the 2050s to the four responses. In comparison, by the 2080s, RCP is the major contributing factor. Key Points: Population exposure to drought for the future is projected through the combination of multiple emission‐population scenariosPopulation exposure to drought would increase apparently in the 2080s under RCP8.5Global climate model (GCM) and RCP are the primary uncertainty sources of drought exposure in the 2050s and 2080s, respectively [ABSTRACT FROM AUTHOR]