Find in Library

Sedimentation on the rivers beds causes many difficulties not only for an orderly and sustainable flow of water, but also by altering the rivers' hydraulics, which in certain cases drastically changes the flow direction by opening new channels that inundate farm fields, municipalities and infrastructures. As various models developed for the suspended load (SL) estimation are often site specific, therefore, it is prudent to test and calibrate them for the intended river. As the Hirmand River is the lifeblood of Sistan, therefore, the 70 km terminal reach, the Sistan River, was selected for this project. The monthly and seasonally discharge of the river and its SL contents for the period 1970–2015 were analyzed statistically using six models, namely: linear, double-linear, mean value, FAO, parametric factor and non-linear. As the determination of the break point is very important in the application of the double linear method, the genetic optimization algorithm was used in this process.Two criteria for choosing the correct method for estimation of the SL were used: relative mean error and root mean square error. It is concluded that the double linear method proposed in this research gives the best result in all forms without any delineation, seasonal and monthly classification with the relative mean error of 154.46, 142.7 and 116.15, respectively. Monthly classification of data increased accuracy and reduced the root mean square error in the Sistan River.