| Sumario: | Drought continues to be the worst natural hazard in the worldaffecting ecosystems, economies and overall human welfare withsevere consequences in developing countries. While drought is anunavoidable climatic phenomenon, actions can be made toimprove preparedness and mitigate the impacts upon properforecast on susceptibility. Drought susceptibility modeling plays akey role in determining how best to mitigate and adapt todrought occurrences. This research explores the application ofgeospatial techniques and Analytic Hierarchy Process (AHP) indrought susceptibility modeling for the drought-prone BilateRiver watershed, located in the central Rift Valley drylands ofEthiopia. A total of 15 parameters were used including rainfall,temperature, evapotranspiration, soil moisture, normalized differ-ence vegetation index, land surface temperature, soil texture, landuse-land cover, topographic wetness index, modified normalizeddifference water index, drainage density, slope, elevation, popula-tion density and aspect to model drought susceptibility. Findingsshowed that nearly 70.2% of the area falls under the moderatedrought category, followed by the severe (23.2%), mild drought(6.6%) and extreme (0.02%) drought categories in the watershed.Based on zonal administration, Wolayta Zone has a high spatialcoverage of severe drought with 54.1% (659.3 km2) while Hadiyahas a high spatial coverage of moderate drought with 24.6%(908.9 km2). The drought susceptibility model (DSM) receiver oper-ating characteristic (ROC) curve was then created, and the areaunder curve (AUC) was calculated. The results of the analysisshowed that the AUC is 0.701 (70.1%) indicating that the modelis reasonably good model. Hence, geospatial approaches in con-junction with the AHP model improved the drought susceptibilitymodeling’s reliability, which has significant implications fordrought adaptive management and preparedness planning.
|
|---|