Manikandan Mathur

We investigate two-dimensional advective stirring by satellite-derived ocean surface currents in the north Bay of Bengal during December 2015–March 2016 using the framework of Lagrangian Coherent Structures (LCS). The forward and backward finite-time Lyapunov exponent (fFTLE and bFTLE, respectively) fields indicate that freshwater parcels from the river run-off are subjected to intense stirring by the ocean surface currents, which contain both geostrophic and wind-forced Ekman currents. Large-scale thermal fronts present in the satellite-based sea surface temperature (SST) fields are then shown to closely resemble attracting LCS (aLCS) that govern horizontal advective stirring on subseasonal time scales. Statistical estimates of the angle between the thermal fronts (of all spatial scales present in the SST data) and the aLCS from the entire north Bay of Bengal show a preference to align with each other. Significantly, in the presence of sufficiently strong aLCS in their neighbourhood, thermal fronts are shown to have a strong preference to align with them. Finite-time advection of fluid particles over around a week to compute the bFTLE fields results in the strongest alignment between the thermal fronts and neighbouring aLCS. The robustness of our results are established by presenting analyses of (i) geostrophic currents from altimeter data, and (ii) a reanalysis-based fine-scale salinity field. Attracting LCS, identified as sufficiently strong ridges in the bFTLE fields, are argued to represent reasonable estimates of the location and orientation of fronts (thermal or salinity) formed by advective stirring over subseasonal time scales.

A very high-resolution model for the Bay of Bengal is configured to simulate submesoscale flow features, which are otherwise absent in coarse resolution models. In this manuscript, we present the response of this model to simulate summer monsoon freshening in the northern Bay of Bengal when forced with two different winds, one based on weather forecast model output from National Centre for medium range weather forecast (NCMRWF) and the other based on satellite-derived SCATSAT ocean surface winds for 2017 as a case study. Our results reveal that the spatio-temporal evolution of lateral freshwater dispersal in the northern Bay of Bengal is sensitive to wind vorticity and eddy formation/evolution. A reasonably realistic decrease of 2–3 psu in the sea surface salinity at the mooring location of 18oN, 90oE towards the end of September 2017 is simulated in the model when forced with the SCATSAT winds, whereas significantly weaker freshening occurs at the mooring location when the model is forced with the NCMRWF winds. A filament of strong positive vorticity connecting two cyclonic eddies in the north central Bay in the NCMRWF wind forced simulations prevents the direct southward advection of freshwater, resulting in higher salinity at the mooring location. A similar mechanism that would prevent southward advection of fresh water is absent in the SCATSAT forced model run. Westward moving Rossby waves in the SCATSAT forced model run are found to play an important role in the evolution of freshening events at various locations in the northern Bay of Bengal. Lagrangian Coherent Structures (LCS) derived from the model currents suggest that the background conditions were favourable to advect freshwater from Ganges-Brahmaputra towards the mooring location in the northern Bay of Bengal in the SCATSAT forced simulations.