RECONSTRUCTION OF FLOW REGIMES OF BORKENA CATCHMENT FOR IRRIGATION DEVELOPMENT USING RAINFALL - RUNOFF MODELING

Abstract

The main objective of the research was to reconstruct flow regimes of Borkena catchment in ungauged locations for irrigation development using rainfall-runoff modeling. Four black-box-type rainfall-runoff models, namely, the Simple Linear Model, the seasonally based Linear Perturbation Model, the wetness-index­ based Linearly Varying Gain Factor Model, and the Artificial Neural Network Model, along with the conceptual Soil Moisture Accounting and Routing Model, were used to test the hydrological response of the Borkena catchment. The performance of these hydrological models for the study area was tested and the best candidate model for the catchment response prediction was selected. Artificial neural network model was selected as a robust rainfall-runoff model to obtain the estimated flow of the gauged station which is the basis to transfer flow data to ungauged sites on the catchment. The R 2 during calibration and verification was 97.57% and 91.30% respectively. On the basis of the selected model 15 days 75% dependable flow derived from the flow duration was used as the basis of estimating dependable low flows at ungauged locations of the catchment using area ratio method of transferring flows.