Modelling sediment transfer in rivers: Challenges in modelling sediment matters

Rainfall and runoff induced erosion and sediment transport in hydrological watersheds are complex processes. This process has great importance in scientific research studies and engineering practice. The amount of sediment transported within the watershed is needed for hydrological and environmental problems. Sediment transport over a watershed can be estimated by time series analysis, empirical or mechanistic equations, monitoring, sampling, surveying, remote sensing or geographical information systems. As monitoring and sampling sediment transport process are costly and not easy to implement yet, modelling has become an alternating tool used for estimating sediment transport. Data-based empirical models as well as process-based hydrological models are available for this purpose, yet modelling is difficult and challenging. Challenges encountered in the modelling are the variability in the estimate of sediment calculated by each model, data requirement for the calibration of model parameters, complexity in the calibration and validation stages of the process-based models, uncertainty in the transport capacity approach used in model construction, etc. In this chapter, these challenges related to the modelling sediment matters are discussed with an emphasis on the process-based sediment transport models. A case study on Buyukcekmece dam reservoir in the greater municipality region of Istanbul, Turkey shows that order of magnitude different outputs are obtained when data-based models are used for estimating sediment transport in hydrological watersheds. Process-based models were paid particular attention on their microtopographical structure, parameterization and data requirement.