Bias from Rainfall Spatial Distribution in the Application of Areal Reduction Factor

Abstract

There are many different approaches for deriving Areal Reduction Factor (ARF); however, only three methods (Methods (1), (2) and (3)) have been developed for the application of ARF. In Method (1), the ARF estimated for the entire basin is applied to all the sub-basins, and in Method (2), the ARF of a sub-basin is independently estimated by considering its sub-basin area. Finally, in Method (3), the ARF is estimated by considering the accumulated area up to the point of determination of the design flood. In this study, these three methods are evaluated by their bias involved in the estimation of runoff peak flow. This evaluation is based on the design flood at the exit of each sub-basin and the entire basin. For this purpose, a total of 50 realistic rainfall spatial distributions were generated by considering the shape, orientation and center of a storm and applied to the estimation of realistic runoff peak flow. This study considered the Chungju Dam Basin as a study basin, which is located at the center of the Korean Peninsula. As a result, it was found that Method (3) is most reasonable, with the negative bias of just 5% or less.