Storm-Wise sediment yield prediction using rainfall and runoff variables

Document Type : Research Article

Authors

• L. Gholami ¹
• H.R. Sadeghi ¹
• A. Khaledi Darvishan ¹
• A. Telvari ²

¹ College of Natuaral Resources and Marine Sciences, Tarbiat Modares Univ., Noor

² Islamic Azad University

https://doi.org/10.22067/jsw.v0i22.1022

Abstract

Abstract
Application of empirical models is a must owing to complexity of process, different features, spatial and temporal variation of soil erosion and non-existence or lack of pertaining data. In this study, the efficiency of rainfall and runoff variables of 11 storms during winter 2006 and spring 2007 in explanation of storm-wise sediment yield in Chehelgazi watershed of Gheshlagh Dam basin in Kurdistan province was evaluated with the help of bivariate and multivariate regression models by using different transformed data. The models’ efficacy was then assessed by using coefficient of determination, error of estimation and verification. The results showed that bivariate regression models, using different transformed data with determination coefficient of beyond 66%, and respective error of estimation and verification of below 40 and 30%, had a better efficiency in estimation of storm-wise sediment yield than multivariate regression models. The results also verified that the rainfall variables could explain storm-wise sediment yield variations better than runoff relating factors with overall contribution of some 80%.

Key words: Sediment yield, regression models, Gheshlagh dam, Chehelgazi watershed, Kurdistan province, Iran