Simulation of Tidal Morpho-dynamics in the Hooghly Estuary using CMS Flow and Artificial Neural Network Models

Abstract

To investigate the tidal hydrodynamics, sediment transport and morphology variation in the Hooghly Estuary, India, a two-dimensional model was developed in the CMS Flow platform. Application of numerical modeling techniques to sediment transport problems in rivers require detailed study of the river reach characteristics and development of the appropriate mathematical model. The characteristics of the river vary considerably starting from its origin to its mouth where it joins the sea, the flow and sediment transport processes also vary accordingly. This makes modeling of sediment transport for the whole reach very complex. Flow and sediment transport processes in estuaries are more complicated because of factors like two-way tidal flow, and widely varying morphology. Numerical models for estuaries are usually extensive and require high computational time because of the complexities in the solution algorithms and large time-space solution domain of the problem which also require a wide variety of data. It becomes very time-consuming to perform these simulations repeatedly, for analyzing changes in sediment transport for variations in boundary conditions, local changes in bathymetry, presence of structures or for long-term simulations. Artificial Neural Networks models when used in combination with numerical models, to simulate flow and sediment transport reduces the computational time considerably. In this work a combined ANN-numerical model is developed for simulation sediment transport and bathymetry change in the Hooghly estuary.