Numerical modelling of coastal changes between Chennai and Ennore, India

Abstract

In this report a study is presented concerning the morphological impact of breakwater structures perpendicular to a coast. This study was initially associated with the Indian coast between Chennai (formerly known as Madras) and Ennore in India. Like many ports, the harbour of Chennai had, and still has a considerable morphological impact on the surrounding coast. By obstructing the longshore transport, the presence of the breakwaters caused accretion and erosion. Construction of a new harbour near Ennore, 20 km north of Chennai, started in 1997. During construction of the harbour similar effects like near the Chennai Port are visible. Morphological consequences of these harbours can be simulated and predicted by a morphodynamic model. The objective of this study is to design a morphological model of the coast between Chennai and Ennore, and calibrate this model by means of the recorded coastline advancement near the Chennai Port. The time period for these morphological simulations is set to 25 years. Use is made of DELFT2D-MOR, a software package developed by WL I Delft Hydraulics. This numerical program offers a wide range of tools for implementing physical processes that determine coastal morphology. The objective mentioned above could not be met within the timeframe of this study. In the course of this study, the implementation of breakwater structures appeared to be more complicated than expected. Hence a greatly simplified representation of the real situation has been modelled: simulations have been carried out for a breakwater, perpendicular to the shore where only waves determine the morphological behaviour. The following conclusions can be drawn: a) The present definition of structures like breakwaters in the computational area of DELFT2D-MOR is not suitable for morphological calculations. The breakwaters do not obstruct the longshore sediment transport properly. Most of the incoming sediment does not settle, but "disappears" near the breakwater. A large portion of the longshore sediment drift therefore does not contribute to beach accretion. These structures have to be implemented as closed boundaries within the computational area. b) If the breakwaters are implemented as closed boundaries, DELFT2D-MOR is able to simulate the accretion process due to the presence of a breakwater. The numerical model responds well to input with respect to varying particle size, and varying bottom roughness. c) The erosion process in DELFT2D-MOR is not simulated properly. Although coastline retreat is taking place, the erosion mainly generates unrealistic steep beach slopes. d) The single line theory of Pelnard-Considere yields the same results with respect to coastline advancement near the breakwater, compared to DELFT2D-MOR. The similarity with the computational results only holds for the accreting coastline.