Accurate estimation of the sediment volume carried by the rivers is important in water related projects and recognition and suggestion proper methods for estimating suspended sediment goals which should be conducted by related researches. Among the methods that have been recently used to model suspended sediment, machine learning based methods such as decision trees, support vector machine, and artificial neural networks are importance. In the present study, the applicability of such techniques in predicting suspended sediment load of Babaaman watershed in Bojnord, Iran has been evaluated. Input data for predicting Babaaman watershed’ suspended sediments in this project are: Debi, suspended load, raining and evaporation, which are related to the statistical period 1349 to 1380. In order to assess the accuracy and precision of the model results, statistical measures including R, RMSE, and MAE have been utilized. Consequently, the results of statistical value of R and RMSE for sediment rating curve method 0.80 and 55863.77, neural network 0.98 and 1.28, decision tree model 0.96 and 48881.56 and support vector machine 0.99 and 0.6998. The obtained values reveal that the support vector machine was more consistent with the measured values compared to the above-mentioned methods.
Rights and permissions | |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |