Investigating Drought Frequency in the Eastern Region of Rwanda during the Last Two Decades

Abstract

Climate has an influence on socio-economic development. Agriculture is a vital socioeconomic sector in the Eastern region of Rwanda. Climate variability, thermodynamic processes, small and large scale weather patterns, land and atmosphere feedbacks influence droughts significantly. In this study, the trend of past rainfall characteristics and severity of drought conditions over Eastern Region of Rwanda were assessed. Monthly rainfall data of ENACTS dataset which combines ground data observations (Station data) and satellite data, which make blended dataset over four stations spread in Eastern region of Rwanda districties Nyagatare, Kawangire, Kibungo and Bugesera and obtained from Rwanda Meteorological Agency were used. The spatial and temporal characteristics of rainfall were investigated based on time series analysis bringing to an end the trend and seasonality components of the time series. This study investigated annual and seasonal drought characteristic for the period spanning from 1981 to 2016 for Eastern region of Rwanda. The component of drought investigated in this study includes severity, Frequency and probability of occurrence. Precipitation deficit was quantified by the use of The Standardized Precipitation Index (SPI) computed over run 3 for March-April-May (MAM) season, October-November-December (OND) season and annual(12 months). Frequencies of diverse drought categories were computed and used to determine the probability of occurrence of diverse drought episodes. Time series analysis for the parameters was done to determine the trend and the temporal variability of drought characteristic. The SPI analysis done on the two rainy seasons returned four categories of drought namely the mild drought with values ranging between -0.01 to – 0.99, moderate drought ranging from -1 to -1.49, severe drought with values ranging between 1.5 to -1.99 and extreme drought condition with SPI values less or equal to -2. There is high probability of mild drought in both seasons in all stations that were studied with Bugesera recording the highest probability during MAM season while Kawangire record the highest probability in OND season. Droughts of varying intensity were noted to last for several seasons and individual drought category had low probability for higher drought intensity.