HYDROLOGICAL DROUGHT PROPAGATION AND VULNERABILITY IN ETHIOPIA

Abstract

In Ethiopia, precipitation variability and extremes, often associated with large-scale climate anomalies, have led to frequent devastating droughts that impact livelihoods. The major energy production of the country relies on hydropower, and frequent droughts affect energy production due to the decline of streamflow. Therefore, understanding drought severity, duration, and spatial extent is crucial to mitigating drought-related impacts. The research investigates temporal and spatial precipitation extremes and their connection to climate indices. A total of eight extreme precipitation indices were selected to investigate precipitation variability, trends and change point. Wavelet coherence and correlation coefficient were used to identify the relationship between precipitation extremes and climate indices. Most precipitation extreme indices show an increasing trend over the south and southwest regions and a change in precipitation time series between 1990 and 2012. The multiscale analysis presents strong coherence between precipitation anomaly and Nino 3.4 and IOD over the south and southeast regions. Similarly, spatial correlation shows IOD and Nino 3.4 to have a positive correlation to most precipitation extreme indices over the country's south, southwest, and southeast parts. The Northern Atlantic Oscillation (NAO) climate index negatively correlates with most precipitation extremes. We further investigated the duration, severity, and propagation time of meteorological and hydrological drought. Our results reveal that the long-term pattern of meteorological and hydrological drought is increasing in most parts of Ethiopia, with the southern part experiencing a particularly noticeable trend. The spatial analysis indicates that basin characteristics highly influence the drought propagation time from meteorological to hydrological. Highland areas take less time (1 to 6 months) to propagate from meteorological to hydrological drought compared to lowland areas, which take an extended period (9 to 12 months). Furthermore, the research examines the impact of drought on streamflow in the selected sub-basins, providing insights into the recovery period of rivers after drought events. The study investigates meteorological and hydrological drought using multiscale standardized and anomaly drought indices in four sub-basins: Ethiopia's Genale, Tekeze, Awash, and Baro basins. Two Archimedean copulas (Clayton and Gumbel) were used to identify the joint return period between drought duration and severity. We computed the streamflow required to recover from hydrological drought for selected sub-basins. Results revealed that drought frequency has increased over most sub-basins over the last two decades. Tekeze and Baro sub-basins required more than 4BCM and 2.5BCM streamflow to recover from the fifteen-month drought duration. The shortest drought duration occurred for three months in the Awash sub-basin, requiring only 0.21BCM streamflow for drought recovery. Finally, our study addresses the challenge of limited spatial data by using remote sensing and reanalysis data to assess hydrological drought vulnerability in Ethiopia. The analytical hierarchy process (AHP) is employed to assign weightage values to various influencing factors, namely, Land use/land cover, soil texture, population density, elevation, drainage density, aridity index, slope, rainfall departure, and water storage deficit index. The study results categorized hydrological drought-vulnerable areas into six categories. Most of the highly vulnerable regions are in the northeast (lower Awash) and southeast lowlands (Wabi-Shebelle and Genale-Dawa basins). The southwest (most Baro-Akobo basin) part of the country falls under low to no vulnerability zones of hydrological drought, covering only 8% of the country's area.