Abstract:
Detail understanding of rainfall-runoff processes is tremendously important for a watershed with variable
streamflow generation. The streamflow of the Meki River fluctuates seasonally causing flooding on surrounding
agricultural land. This study adopted Hydrologic Engineering Center Hydrologic Modeling System (HEC-HMS) to
model the streamflow and predict flood in the Meki watershed, Rift valley basin. The model was calibrated with
observed streamflow data from (1987–2004) whereas the consecutive year’s data (2005–2010) was used for
model validation. The evaluation criteria namely: Nash Sutcliff Efficiency and coefficient of determination were
preferred to evaluate the model performance. The finding relieved that, the model can perform very well (NSE =
0.83, R2 = 0.91) during Calibration, and (NSE = 0.804, R2 = 0.89) at validation period. Moreover, the predicted
floods at 2, 10, 25, 50, and 100 years were 133.2, 178.1, 239.7, 313.2, and 346.19 m3
/s, respectively in the
watershed. The novelty of this study lies in evaluating the model results using both statistical parameters (NSE
and R2
) and the Generalized Extreme Value method). The finding of this study is vital to developing flood
mapping and designing flood mitigation measures in the study area. Further, the developed models can be
applied to other hydrology with similar hydrological conditions.
