Assessment of Meteorological Drought Characteristics in North Eastern Counties of Kenya

Abstract

Climate has an influence on the day-to-day socio-economic development. Agriculture is a vital socio-economic sector in the North Eastern region of Kenya. In this study, the trend of past, present and future rainfall characteristics and severity of drought conditions over North Eastern Region Counties were assessed. Observed data included total monthly precipitations over the three synoptic stations spread in North Eastern counties of Wajir, Garissa, and Mandera and obtained from the Kenya Meteorological Department. Downscaled rainfall ensemble data from 8 Coordinated Regional Climate Downscaling Experiment (CORDEX) Regional Climate Models (RCMs)were also used.The spatial and temporal characteristics of rainfall were assessed based on time series analysis through determination of the trend and seasonality components of the time series. Drought severity was analyzed using Standardized Precipitation Index (SPI) for both past (1971-2015) and future (2015-2045) period. The frequency, probability and persistence of drought occurrence were also analyzed. The periodicity analysis to determine drought recurrence was done using the Single Series Fourier (Spectral) Analysis Rainfall was noted to be bi-modal marked by March to May (MAM) and October to December (OND). Based on the observed and ensemble CORDEX models, rainfall was noted to have the highest temporal variability and affirmed by the coefficient of variability values ranging from 1.96 to 3.02 indicating variability increases significantly in RCP 8.5 to values more than the past and RCP 4.5 Wm-2. Projected precipitation variability was higher in RCP 8.5Wm-2, compared to the past and RCP 4.5 Wm-2. SPI analysis indicated two prevailing drought conditions, namely, the mild drought with values ranging from -0.01 to – 0.99 and moderately dry condition with SPI values ranging from -1.0 to -1.49. There was a high probability of mild drought both in MAM and OND seasons in the three selected stations in the North-eastern region with Wajir recording the highest probability while the Probability for moderate drought was low at all stations. The individual drought category had a low probability for higher runs In both seasons, the probability of occurrence of a given drought characteristic decreased as the number of runs increased. The probability of occurrence of a mild drought of only a single run recorded the highest frequency in the MAM season in Wajir. In Mandera, the probability of occurrence of a mild drought of a single run recorded the highest frequency in OND season. Moderate drought conditions only recorded a single run for MAM season and no runs recorded for OND season except in Mandera. In Garissa, the probability of occurrence of a mild drought of a single run recorded the highest vi frequency in OND season. The cycle observed from the spectral analysis could be grouped into 2-3 years, 2.5-3.5years, 4.5-7years, and 8-12years SPI analysis of projected seasonal rainfall indicated drought of varying intensity based on RCPs scenarios. There was a high probability of moderate drought both in MAM and OND seasons based on RCP 4.5 and RCP 8.5 scenarios. However, Wajir recorded higher probability of moderate drought compared to the other two stations. Projected (2015-2045) drought persistence indicated that individual drought category had a low probability for higher runs. Computed conditional probability of drought occurrence indicated higher probability values based on lower run values during the MAM season compared to OND for both RCP 4.5 and RCP 4.5 scenario with the higher runs indicating lower conditional probability values. The periodicity of projected drought conditions based on RCP 4.5 could be grouped into 2 to 3 years, 4.5 to 7 years and 8 to 12 years in Garissa, Wajir and Mandera. For lower periodicity (< 3 years), Garissa and Wajir indicated a higher magnitude of the drought conditions compared to Mandera, which showed higher magnitudes for drought conditions for the periodicity of between 4 and seven years. The periodicity of projected drought conditions based on RCP 8.5 could be grouped into 2.5 to 4 years, 6 to 10 years and 10 to 15 years. Higher magnitudes were recorded for periodicity values of between 3.5 and 6 years. In general, spectral analysis based on RCP 4.5 and 8.5 indicated distinct groups of 2-3 years, 4.5-7years, and 8-12 years and thus could be attributed to a different scale of motions. The drought analyses over the area of study varied between mild drought and moderately dry condition. Given that these are rainfall seasons and as such the conditions might worsen during other dry seasons, reducing the risks and therefore the impacts of drought are of utmost importance.