Suspended Sediment Estimation by Artificial Neural NetworkApproaches

Suspended sediment flux is a crucial parameter for managing hydraulic projects and monitoring soil erosionand ecological conditions in a catchment. Estimating the amount of suspended sediment with highresolution is a challenging task. Several empirically and physically-based models have been developed topredict the suspended sediment transport in a catchment. Empirical models estimate the flux of suspendedsediment by correlating it with catchment characteristics such as land cover, topography, climate, drainagearea, and deposition rates in ponds or reservoirs. Meanwhile, physically based models aim to representvariable spatial heterogeneity by dividing the catchment into grids and describing sediment transportprocesses between grids using simplified partial differential equations. The nonlinear black box model ofthe artificial neural network (ANN) approach appears to be a valuable alternative for modeling complexsuspended sediment series. In the presented study, the application of artificial neural networks for theestimation of suspended sediment was investigated.

کلیدواژه ها:

Artificial neural network ، Suspended sediment load ، Modeling