Variability and trends in past, current and future climate in East Africa

Abstract

This study evaluates the climate variability and change in past, current and future climate in East Africa (EA) using high resolution regional models under the CO-ordinated Regional climate Downscaling Experiment (CORDEX). Trends of mean variability and extreme events was determined using graphical, regression and Mann-Kendall test approaches for the past (1971-2000), current (2001-2014) and projected future climate change scenarios (Representative Concentration Pathway [RCPs] 4.5 wm-2 and 8.5 wm-2) for mid (2016-2045) and end (2071-2100) century. Intra seasonal characteristics of rainfall (onset, duration and cessation of the wet season, seasonal rainfall totals, mean rainfall intensity, mean duration of the spell) was also computed. All the CORDEX models performed well in simulating rainfall over EA with RMSE of less 5mm and predicted higher increase in minimum temperature compared to maximum temperature in EA for both RCP 4.5wm-2 and RCP 8.5wm-2. However, temperature increase was found to be higher during Long Rainy (LR) season while increase in rainfall was found to be higher during Short Rainy (SR) season. Maximum temperature extremes increased while minimum temperature extremes decreased with a statistically significant rise in the number of hot days and warm nights and a decrease in number of cool days and cold nights. Variability in rainfall, maximum and minimum temperature indicated that extreme values were changing with increasing frequency and magnitude for both past and projected climate. onset, duration and cessation analysis of rainfall indicated that there existed organized progression of rainfall onset especially over western Kenya region with the long rains showing a southerly progression while the short rains showed a south-westerly progression. Cessation of rainfall for both seasons show strong localized influences, mainly surrounding Lake Victoria and forested areas, including orographic features. For stations with long length of growing season, the length varied more than the onset date..