TY - JOUR ID - 102680 TI - Effectiveness Assessment of Suspended Sediment Load Estimation Methods in the Ghar Chai River JO - Environment and Water Engineering JA - EWE LA - en SN - AU - Moradinejad, Amir AU - Davod Maghami, Davod AU - Moradi, Mojtaba AD - Assist. Professor, Soil Conservation and Watershed Management Research Department, Markazi Agricultural and Natural Resources Research and Education Center, AREEO,, Arak, Iran AD - Assist. Professor, Department of Water Engineering, Faculty of Agriculture, University of Arak, Arak, Iran AD - Ph.D. Scholar, Department of Watershed Management, Faculty of Agriculture, Tarbiat Modares University, Tehran, Iran Y1 - 2020 PY - 2020 VL - 5 IS - 4 SP - 328 EP - 338 KW - Artificial Neural Network KW - Doab Bridge KW - Neuro-Fuzzy KW - Sediment rating curve KW - Suspended sediments DO - 10.22034/jewe.2020.211925.1341 N2 - Estimation of sediment load in rivers is one of the most important and applicable issues in studies and design of river and water engineering projects. Therefore, identification and suggestion of suitable methods for estimating suspended sediment load should be done based on relevant research. These methods include artificial neural network models, neurophysics, sedimentation rate curve as well as multivariate regression model. In this study, the efficiency of these methods was investigated in predicting discharge rate of suspended sediments of Gharachai River watershed. Daily time series data of measured flow discharge and sedimentation of bridge station of Gharachai River were used. The statistical period used in this model was two years (2016-2017). Independent variables used to enter the network include runoff and suspended load at the Gharachai River Doab Bridge station. The dependent variable, which is the network output, was the suspended load. After modeling with each compound and calculating RMSE and R2 values, the best combination was selected. The results showed that the neurophasic method based on discharge and sediment inputs and artificial neural network models based on discharge inputs were more accurate than multivariate regression and sedimentation rate curve. For values ​​higher the long-run mean of the statistical series, the values ​​simulated by the ANN model and for values less the mean and annual sediment load were also closer to the observed values. While for maximum values, no significant difference was found between ANN models, neurophysics, and linear regression. UR - http://www.jewe.ir/article_102680.html L1 - http://www.jewe.ir/article_102680_7698f80b7009a4986fd6ab8e2e738d6b.pdf ER -