Geomorphological Parameters Based Watershed Sediment Yield Estimation

Abstract

Most models available for computation of sediment yield of watersheds use inter-correlated geomorphologic parameters of the watershed which is against the principle of regression analysis. Principal components which are transformed but unrelated were derived from the set of correlated geomorphological parameters of watersheds. In this study eleven watersheds were selected. Out of which the data of ten watersheds were utilized for the development of models and the data of one watershed were utilized for validation of the models. Twenty-two numbers of geomorphological parameters were quantified for each selected watershed. It was found that the geomorphological parameters are correlated among themselves. So these parameters were transformed into the scores of principal components to make them uncorrelated. Then multiple regression models between sediment production rate (SPR) and scores of principal components were developed using principal component regression in Statistical Analysis System (SAS) environment. The average absolute deviation of predicted SPR was found within 10.55% and 5.43% for different percent ranges of runoff and ranges of rainfall. Hence the developed model can be used for better prediction of SPR for other similar ungauged watershed.