Evaluating flood frequency analysis methods for sand dam projects: a case of Dire Dawa City Administration, Ethiopia

This study evaluates flood frequency analysis (FFA) to support sand dam design and planning in the Dechatu catchment of Dire Dawa, Ethiopia, an area vulnerable to flash floods and irregular rainfall. Rainfall–runoff modeling was performed using the Hydrologic Engineering Center's Hydrologic Modeling System (HEC-HMS) with the Soil Conservation Service Curve Number (SCS-CN) method, SCS Unit Hydrograph, and Muskingum routing. Curve numbers ranged from 78.5 to 83.07, while basin lag times varied between 8.69 and 57.34 h. Peak discharge rates fluctuated, with Kombolcha experiencing the highest at 214.4 m3/s. Annual runoff volumes ranged from 18,440.35 m3 in 1993 to 72,553.7 m3 in 2006, reflecting heavy wet-season rainfall. FFA tested multiple distributions, Log-Pearson III, generalized Pareto (GPA), generalized extreme value (GEV), and normal with Log-Pearson III estimating a 100-year peak flow of 165.36 m3/s, closely matching HEC-HMS results. In semi-arid regions, FFA is applied to optimize the design, planning, and implementation of sand dams. Sand dams are water storage structures built across seasonal streams to capture and store water during floods. The study underscores the importance of data-driven planning to improve sand dam resilience, water management, and flood preparedness, ensuring sustainable and safe water resources for vulnerable communities.