Climate extremes including drought, floods, and high/low temperatures are recurrent in South Sudan and they are associated with widespread socio-economic miseries. This study aims at enhancing our understanding of past and present characteristics of climate extremes over South Sudan based on the temporal patterns of the past rainfall and temperature extremes. The datasets used in this study were monthly rainfall and temperature for period 1954 to 2013. Temperature records included both maximum and minimum values. Graphical methods and statistical techniques such as Mann Kendall statistics, Spectral analysis, Standardized Precipitation Index (SPI) and Gaussian kernel function were used to investigate the past and present characteristics of rainfall and temperature.

It was found that the rainfall over South Sudan is uni-modal with its peak in the months of July and August with June through to August (JJA) being the main rainy season and contributing more than 50% of annual rainfall. Highest maximum temperatures were recorded in April while the lowest minimum values were recorded in the months of December and January. Results from trend analysis showed a decline in the JJA seasonal rainfall trends in many locations. However, both the maximum and minimum temperature indicated a significant increasing trend for all locations considered in the study at 0.05 significance level although with some slight seasonal differences in the patterns of the increasing trends. The inter-annual patterns of rainfall indicated recurrent patterns of floods and droughts over all locations of South Sudan. Two distinct spectral peaks of rainfall and temperature were common in all locations centered at 3-5 to 7-9 years. The Probability distribution function showed a shift of the entire distribution towards the dry climate conditions for rainfall and warmer climate for maximum and minimum temperatures.

The findings of this study showed key indicators of climate change signals over South Sudan. These findings can help policymakers to mainstreaming climate related risks into national policies, plans, and development projects. Further work should be done to strengthen these findings by use of observed gridded dataset and projected future climate based on the Global /Regional Climate Model (G/RCMs).