Signature of Indian Ocean Dipole on the western boundary current of the Bay of Bengal

摘要

This study uses an unprecedented collection of 27 years of repeated eXpendable Bathy Thermograph (XBT) sections crossing the western and north-western boundaries of the Bay of Bengal (BoB). Our objective is to analyse the variability of the boundary current that flows there, known as the East India Coastal Current (EICC). In the western BoB, in line with the past observational and modelling studies, our dataset confirms that the EICC seasonally flows poleward from February to July (with a peak transport of 5 Sv), then decays and reverses to equatorward towards the equator from October to December (with a peak transport of 3 Sv), reversing again to poleward in December. In the north-western BoB, the seasonal EICC prominently flows north-eastward, with a peak transport of 7 Sv in March. Over the rest of the climatological year, the transport remains north-westward and weak (of order 2 Sv at most). Beyond the seasonal climatology, the timespan of our dataset allows us to put a special emphasis on the departures from the seasonal cycle of the EICC velocity and transport. It is observed that this non-seasonal variability is actually larger than the seasonal climatology, so that the seasonal cycle may be completely distorted in any given year. This is true in the western boundary region as well as further offshore in the central BoB and concerns the surface as well as the subsurface layers. Indian Ocean Dipole (IOD) events influence EICC variability, supposedly through remote forcing from the equatorial Indian Ocean and generate northward (southward) anomalous transport typically reaching 5 Sv (7 Sv) in winter during positive (negative) IOD events. In addition to IOD events, most of the variability observed at inter-annual timescales seems to be driven by ocean turbulence. A comparison of our observed current with a suite of state-of-the-art ocean reanalyses and model products (SODA, ORAS4, MERCATOR-ORCA12) confirms this hypothesis, with non-eddy resolving models overestimating the wind-driven IOD influence on EICC variability. Our results emphasise the benefit of a sustained long-term monitoring programme of the EICC, spanning the entire continental slope region up to its offshore edge, associated with a modelling approach that would be capable of accounting for the oceanic turbulence, to decipher the various processes forcing the variability of the western boundary current (WBC) of the Bay of Bengal and their inter-play.