Assessing the influence of climate variability drivers on seasonal rainfall extremes and flood risks in Katakwi District Teso Subregion, Uganda

Abstract

This study assessed the influence of climate variability drivers on seasonal rainfall extremes in Katakwi District, Eastern Uganda. The study was governed by specific objectives which included (i) assess temporal and spatial trends in rainfall extremes during the March April Mary (MAM) and September October November December (SOND) seasons, (ii) to characterize the temporal patterns of climate variability drivers (iii) to examine the relationship between climate variability drivers and rainfall extremes in relation to historical flood occurrences. The study aimed at determining how climate variability drivers (Indian Ocean Dipole (IOD), El Niño–Southern Oscillation (ENSO), and Madden–Julian Oscillation (MJO)) influence rainfall extremes during the March–May (MAM) and September–December (SOND) rainfall seasons. Six indices which included Rx1day (Maximum daily precipitation), Rx5day (Maximum consecutive 5-day precipitation), R95p (very wet days), R99p(extremely wet days), SDII(Mean precipitation during wet days), and PRCPTOT(Seasonal total rainfall) recommended by the Expert Team on Climate Change Detection and Indices (ETCCDI) were computed from Climate Hazards Group Infrared Precipitation with Station dataset (CHIRPS) daily rainfall data from 1990 to 2023. Temporal and spatial trend analyses of rainfall extremes were conducted using Mann-Kendall trend tests, Sen’s slope estimator, and Kriging interpolation in ArcMap10.4. The relationships between climate drivers and rainfall extremes were evaluated through Spearman correlation. A comparative analysis was then used to assess the relationship between known documented flood years, rainfall extreme indices and climate drivers. Objective 1 results revealed that SOND had statistically significant increasing trend in rainfall extremes especially in Annual total precipitation in wet days (p=0.0379). Spatial analysis of Katakwi district revealed significant increasing trends in seasonal rainfall extremes, indicating increased vulnerability to flood risk in both seasons. However, the entire Katakwi district is at risk, and the southeastern and central parts, with their sub-counties, emerge as the most vulnerable regions prone to flood risk. Objective 2 results showed that IOD emerged as the dominant and influential climate driver with statistically significant increasing trends in MAM (p=0.026) and near-significant trends in SOND (p=0.052). Objective 3 results reveled that ENSO and MJO showed weaker and insignificant correlations with rainfall extremes across both seasons. Analysis of historical flood years (1997, 2007, 2010, 2018, 2019, and 2023) revealed strong alignment with positive IOD phases and elevated Rx5day (Maximum consecutive 5-day precipitation) and PRCPTOT (Seasonal total rainfall) values, illustrating the dominance of multi-day rainfall events over single-day extreme triggering floods in the study area. The study therefore, recommends the integration of IOD informed early warning systems for rainfall forecasts to reduce increasing climate induced flood vulnerabilities in the district.