A hybrid mechanism and ridge regression model to separate the effects of advection and resuspension on suspended sediment concentration

Abstract

Estuarine and coastal areas are land-sea interaction zones where sediment particles can easily move, leading to water quality and aquatic ecological issues. The suspended sediment concentration (SSC) is an important indicator of the movement of suspended particles and coastal water environments monitoring. However, traditional in situ SSC observations cannot directly separated contributions of advectionand sediment resuspension. In this study we aim to propose a hybrid mechanism and ridge regression model to improve the accuracy of instantaneous SSC estuarine and coastal waters. The new model is based on the application of hydrodynamic factors, functional principles and dimensional evaluation methods, which can analysis the effects of horizontal convective transport, vertical suspended upward movement on the SSC. It is compared with existing formulas (models) and mathematical model (Mike21) calculation results, using field data from single stations at the Yangtze River Estuarine. The comparison shows that the accuracy of the new SSC calculation model based on the equations in this paper is high. The Pearson correlation coefficient of the new model has increased by approximately 0.20, and the relative mean square error and the mean absolute errorhave decreased by approximately 0.26. In addition, the new model can explain the hysteresis of the sediment transport process and water flow change reasonably.