Investigating the influences of climate change on drought in a small data-scarce river catchment

Abstract

This study focuses on the influences of climate change on drought characteristics in the small, data-scarce Ndembera River catchment in Tanzania, using the Climatic Research Unit (CRU) datasets and six regional climate models (RCMs) of the CORDEX-Africa framework. The RCM simulations were validated against the CRU data, and their optimistic (RCP 2.6) and worst-case (RCP 8.5) scenarios were used in estimating ensemble simulations using the random forest regression. The ensemble simulations were used to compute standardized precipitation evapotranspiration indices, which characterized drought for the near-future, mid-future, and far-future. Finally, patterns and trends of changes in temperature and precipitation were compared to those in drought characteristics. The findings revealed substantial uncertainty in RCMs, although their ensembles fairly reproduced the historical climate in the catchment. Under the optimistic scenario, a slight decrease in precipitation and a small increase in maximum temperature are projected. Consequently, drought intensity is projected to rise just slightly but remain moderate throughout this century. These changes are expected to be more pronounced during the mid-future than the far-future. Under the worst-case scenario, the projected changes are expected to be more pronounced, especially during the end of the century. These sorts of insights are vital for enhancing adaptations to climate change in catchments. HIGHLIGHTS Investigation of the influences of climate change on drought in a small catchment.; Gridded observational data overcome the scarcity of historical data in the catchment.; Regional climate models fairly simulate the historical climate in the catchment.; Temperature change has longer and more direct influences on drought in the catchment.; Future climate change might not change the current moderate drought in the catchment.;

Keywords

• cordex-africa
• cru dataset
• random forest regression
• regional climate model
• skill score test
• spei