Impact of Monthly Curve Number and Five-Day Antecedent Rainfall-Runoff Data Set on Performance of SCS-CN Method for Ozat Watershed in India – A Case Study

Abstract

The Soil Conservation Service Curve Number (SCS-CN) is a well-established and widely used loss-rate model
to estimate surface runoff. It combines watershed and climatic parameters in one entity curve number (CN). Much of the
variability in CN has been attributed to antecedent runoff condition (ARC). The (CN) also exhibits an inherent
seasonality beyond its spatial variability, which cannot be accounted for by the conventional methods.
In the present study, CN were determined by three different approaches, standard CN, monthly CN and CN based on five
day antecedent rainfall-runoff (ARR) data set using standard asymptotic fit and gauged rainfall-runoff data with an
objective to evaluate the impact of monthly CN and five days ARR data set on runoff estimation for Ozat watershed
(Gujarat State-India). The significant improvement in performance of SCS-CN method is found on application of CN
based on five day ARR data set as compare to monthly CN for Ozat watershed. Refined Willmott’s index (dr) and mean
absolute error (MAE) were used to assess and validate the performance of SCS-CN method. For the study region, the CN
determined based on five day ARR data set was judged to be more consistent with dr=0.58 and MAE=0.93 mm for
λ=0.05.