Research on the SCS-CN initial abstraction ratio using rainfall-runoff event analysis in the Three Gorges Area, China

Abstract

The Soil Conservation Service Curve Number (SCS-CN) method is widely used for predicting direct runoff from rainfall. The ratio of initial abstraction ([a) to maximum potential retention (S) was assumed in its original development to be equal to 0.2 in SCS-CN method. The constant initial abstraction ratio is the most ambiguous assumption and requires considerable refinement. The objectives of this study were (1) to determine the initial abstraction ratio, in an experimental watershed in the Three Gorges Area of China, by analyzing measured rainfall-runoff events; (2) to compare the performance of the traditional and modified la/S values with observed rainfall-runoff data. The dataset consisted of 6 years of rainfall and runoff measurements from the experimental watershed. The results indicated that the la/S values, using event rainfall-runoff data, varied from 0.010 to 0.154, with a median of 0.048. The average initial abstraction ratio of the watershed was equal to 0.053. The standard SCS-CN method underestimates large runoff events, yielded a slope of the regression line of 0.559 and an intercept of 0.301. The modified la/S value was about 0.05 that better predicted runoff depths with an R(2) of 0.804 and a linear regression slope of 0.834. It also improved model efficiency coefficient (E) to 0.768 compared with 0.482 for traditional la/S value. This la/S-adjusted SCS-CN method appears to be better appropriate for runoff prediction in the Three Gorges Area of China. (C) 2008 Elsevier B.V. All rights reserved.