Predicting Longitudinal Dispersion Coefficient in Natural Streams Using M5′ Model Tree

Abstract

The longitudinal dispersion coefficient is a key parameter in determining the distribution of pollution concentration, especially in temporally time-varying source cases after full cross-sectional mixing has occurred. Several studies have been performed to present simple formulas to predict it. However, they may not always result in an accurate prediction because of the complexity of the phenomenon. In this study, a M5′ model tree was used to develop a new model for predicting the longitudinal dispersion coefficient. The main advantages of the model trees are that (1) they provide transparent formulas and offer more insight into the obtained formulas and (2) they are more convenient to develop and employ compared with other soft computing methods. To develop the model tree, extensive field data sets consisting of hydraulic and geometrical characteristics of different rivers were used. By using error measures, the performance of the model was also compared with the performance of other existing equations. Overall, the results showed that the developed model outperforms the existing formulas and can serve as a valuable tool for predicting of the longitudinal dispersion coefficient.