| Abstrakt: |
The management of surface runoff is a very crucial component of integrated water resources planning. One of the most important methods of estimating runoff due to rainfall is the Soil Conservation Service Curve Number method. However, just like many other hydrologic equations which are either empirical or semi-empirical in nature, the universal applicability of this method in its native form remains questionable. This study presents modified SCS models for Oraimiriukwa river basin (ORB) and evaluates their performance in predicting ORB runoff. The performance of the traditional SCS-CN model was examined using NSE and PBIAS. The modified models used an optimization scheme to obtain the actual value of the ratio of initial interception to maximum potential retention (α), which in the traditional SCS-CN method is taken as 20%. The major thrust of this work is that the ratio of initial interception to maximum potential retention (α) is not constant, but rather a function of precipitation. Average α values of 21%, 20%, and 27%, corresponding to CNI, CNII, and CNIII, respectively, were obtained. The traditional SCS model yielded NSE and PBIAS values of − 11.07 and 55.10, respectively. The modified models yielded NSE and PBIAS values of − 0.19 and 2.13, − 0.33 and 3.27, and − 4.37 and − 26.04, for antecedent moisture conditions corresponding to CNIII, CNII, and CNI, respectively. Only CNII yielded models which are single-valued for any given magnitude of rainfall. In view of performance and practicality, the modified models based on CNII were recommended for runoff estimation in ORB. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink