Extreme Rainfall Wet Events under Changing Climate and Impact on Future Flood Risk over Lake Tanganyika Basin (East Africa)

Abstract

Extreme rainfall events leading to floods are among the most severe threats facing communities today in the Lake Tanganyika Basin. This study analyses the spatial and temporal distribution of extreme wet events, detects their trends, and predicts the risk of rare wet conditions that could trigger floods. We used Climate Hazards group Infrared Precipitation with Stations (CHIRPS) data over the reference period 1991-2020 and simulations from ten selected Coupled Model Intercomparison Project Phase six (CMIP6) models over the period 1991-2050. Future projections under two Shared Socio-economic Pathway scenarios (SSP2-4.5 and SSP5-8.5) were analysed in quartiles for 2021-2050. The Standardized Precipitation Index (SPI) and the modified Mann-Kendall (TFPWMK) test were used to characterise extreme wet periods and trends, respectively, while extreme value theory was applied to estimate the probability of rare wet events and flood risk. The findings reveal flood-prone conditions in November, December, January, March, and April. Future changes in consecutive wet days (CWD), heavy precipitation days (R10mm), and very heavy precipitation days (R20mm) are projected to increase flood probability across the entire basin shared by Burundi, the Democratic Republic of the Congo (DRC), Rwanda, Tanzania, and Zambia. The flood risk ratio (FRR) for CWD is projected to reach 1.84 under SSP2-4.5 and 2.18 under SSP5-8.5 in Rwanda’s share, indicating the highest flood risk among the partner states. Flood risk due to maximum 1-day precipitation (Rx1day) shows no marked change (FRR  1.00 – 1.04) under either scenario by 2050. The findings can support efforts to minimise the loss of life, crop damage, and damage to infrastructure in the Lake Tanganyika Basin.