Modelling Freshwater Plume in the Bay of Bengal with Artificial Neural Networks

In spite of advanced modelling techniques, the current prediction of ocean parameters along the ocean coasts remains a formidable challenge. The traditional methods of using mass and momentum equations to solve the physics of flow have helped us understand the oceans better, but their accuracy remains a problem. This article examines the ability of Delft3D to study freshwater plumes along the northern Bay of Bengal (BoB). Whereas the near shelf is primarily driven by tides and local winds, the far shelf is influenced by the freshwater-driven density circulation and monsoonal ocean currents. The prediction of far shelf waters is well represented by employing an artificial neural network. By tuning the parameters properly, we can better pre­dict the freshwater currents in the BoB with a correlation of 0.957 and 0.986 for u and v velocities respectively