Journal of Range and Watershed Managment

Abstract Estimation of suspended sediment basins of data and information is readily available but for basins without statistics, other methods require. The experimental method known as sediment rating curves or relationship "discharge sediment load" is provided. This study seeks to provide better separation of data and the preparation and suspended sediment curve refine hydrometric stations Maraveh Tape, Houtan, Qazanqayeh on the Atrak River in the Atrak watershed is located in Golestan province. To increase the accuracy of the classification scale monthly, quarterly, all samples, wet and dry periods and classifieds sediment discharge with frequency curve fitting techniques to fit a linear, linear fit by adjusting FAO, fitting combination (multi-line ), graphical approach (maximum concentration) and is fitted intermediate categories. For this purpose, statistical indicators were used to select the top. The results showed that sediment rating curve model by fitting a line to the correct application of FAO data Maraveh Tape station, method fitting a line dividing adjusted by applying FAO data on dry and wet period in the station Qazanqayeh and method Fitting a line segmentation classification data on discharges in Houtan Station Among the models tested with the least amount of mean square error and the predictability of the study is to estimate precipitation stations. Selected on the basis of the average amount of sediment suspended for watershed Atrak to Maraveh Tape station, Qazanqayeh and Houtan respectively 705, 683 and 781 Ton/Km2.day was estimated.

Abstract Understanding of suspended sediment rate is one of the fundamental problems in water projects which water engineers consistently have involved with it. Wrong estimations in sediment transport cause incorrect design and destruction of hydraulic systems. Due to the difficulty of suspended sediment measurements, sediment rating curves is considered as the most common method for estimating the suspended sediment load. The main purpose of this research is the capability challenge of this method in comparison to some state of the art models. In this study, we selected some computational intelligence models (i.e. K-nearest neighbor (KNN), artificial neural networks (ANN), Gaussian processes (GP), decision trees of M5, support vector machine (SVM) and evolutionary support vector machine (ESVM)) and compared them with their sediment rating model in 8 basins located in Gilan province. Daily sediment and discharge data considered as the input data for 30-years. Evaluation of the results indicated that the Gaussian process model has the lowest residual sum of squares (RMSE) and the highest correlation coefficient (r) than the other models.

Abstract Suspended sediment load estimation is one of the most important and complicated debates on sediment transport and river engineering. Therefore, the variety of methods developed to estimate suspended sediment load in rivers. According to the empirical nature of these methods, the results have low accuracy and vary so widely from one method to another. Recently, with advances in computer science, various algorithms such as tree based methods have been developed. In this study, sediment rating curve method along with new algorithms such as M5 and Genetic Programing (GP) are used for estimating suspended sediment load in rivers. Flow and sediment discharge data at five hydrometric stations, Behbahan and Cham Nezam on Maroon River, Jow Kanak on Allah River and Moshrageh and Shadegan on Jarahi River are used in present study. The efficiency of tree models, in all stations, was greater than sediment rating curve method, and the RMSE performance of M5 method is 7 to 41 percent superior to sediment rating curve method. The results of this study indicate the close proximity of both M5 and GP efficiency, which according to the simple and conceptual structure of M5, this method is proposed to estimate suspended sediment load in river streams.

Abstract Suspended sediment load is the biggest non-point pollution source and a major factor of degradation of surface water quality. Because of hydraulic models of sediment transport can not predict the suspended sediment load, sediment rating curves as usual hydrological methods are utilized spread for this goal. Cause of regression equations of rating curve have a lot of bias due to logarithmic convert, correction factors in optimization of sediment rating curve were used for eliminating of logarithmic conversion effect and bias of extrapolation in 20 hydrometric stations in up streams and major rivers of Sefidrood watershed. Comparing of 9 rating curve methods as one-linear, one-linear with correction factors as CF1, CF2, FAO, two-linear, mean loads within discharge classes, mean loads within discharge classes with correction factors as CF1, CF2 and
FAO was conducted by RMSE and NASH criteria. Results showed that mean loads within discharge classes, mean loads within discharge classes with CF1 and CF2 correction factors have the most fitting to Sefidrood watershed stations. Our findings illustrated that CF1 and CF2 correction factors in majority of stations have compensated underestimation of rating curves and increased efficiency of models. Power of equation between sediment load and area was more than of one. According to results 30 million ton suspended sediment load enter to reservior of Sefidrood dam annually.
Key words: sediment rating curve, Sefidrood, model efficiency, logarithmic conversion, NASH criteria.

Abstract Suspended sediment estimation is an important factor from different aspects including, farming, soil conservation, dams, aquatic life, as well as various aspects of the research. There are different methods for suspended sediment estimation. This study aims to estimate suspended sediment using feed forward neural network with error back propagation with Levenberg-Marquardt back propagation algorithm and compare the results with best sediment rating curves among commonly used sediment rating curves, including: linear, seasonal, monthly and Mean load within discharge classes. To attain this, the sediment discharge and the corresponding water discharge data for ten hydrometric stations of Lorestan province of Iran were used. In next step different methods of sediment rating curves along with different correction factors, a total of 20 methods were applied to data. Results showed among examined methods; monthly rating curve with MUVE correction factor has been selected as best, based on Nash and Sutcliffe index and accuracy index. Then results of estimating sediment load by using selected sediment rating curve were compared with the results of the neural network. Mean-square error and Nash and Sutcliffe index were applied to select more appropriate method. The results showed the suitability of the feed forward neural network error propagation in compare with sediment rating curves.  

Abstract Correct estimation of suspended sediment transported by a river is an important practice in water structure design, environmental problems and water quality issues. Conventionally, sediment rating curve used for suspended sediment estimation in rivers. In this method discharge and sediment discharge or concentration related using regression relation that generally is exponential model. Respect to uncertainty and nonlinear relation between discharge and sediment concentration, sediment rating curve has not enough efficiency for this purpose. In this study using Artificial Intelligent (Fuzzy Logic and Artificial Neural Network), suspended sediment in Karaj River was estimated. First, various neural network and fuzzy logic models established. For neural network and fuzzy logic, models with four neuron in hidden layers and FIS (Fuzzy Inference System) with four Gaussian membership functions, respectively were selected as the best structure. Finally, the results showed that fuzzy logic estimates the suspended sediment loud better than the other techniques and therefore is suggested for estimation of suspended sediment load.