Laurent Testut (LEGOS, Toulouse, France); Fabien Durand (LEGOS, Toulouse, France); Stéphane Calmant (LEGOS, Toulouse, France); Yann Krien (LARGE, Université Antilles-Guyane, Pointe-à-Pitre, France); Jamal Khan (IWFM/BUET, Dhaka, Bangladesh); Damien Allain (LEGOS, Toulouse, France); Florent Lyard (LEGOS, Toulouse, France); Valérie Ballu (LIENSs, Université de La Rochelle, France); Sara Fleury (LEGOS, Toulouse, France)

The shoreline of the Bengal delta (Bangladesh and India) is a macrotidal area (over 4 m), with a broad (200 km) and shallow shelf. It is also home to marked variability of the water cycle, over a broad range of timescales, from a few hours (cyclonic surges, flash floods) to a few weeks or months (monsoonal floods in the rivers, mesoscale turbulence in the near-shore ocean). Despite profound implications of the water level variability on the society and economy of the 150 M people populating the near-shore region, the characteristics of the ocean tide is poorly observed and understood in this region. Numerical tidal models also do not perform well in this region, compared to the rest of the tropical oceans. This stems, among others, from the lack of knowledge of the bathymetry of the shelf region.
This poster presents an attempt to curb this lack of knowledge, by making use of along-track spaceborne altimetry. We consider standard products (GDRs) for the whole Jason1-2-3 archive, as well as re-processed PISTACH Jason-2 products. We apply harmonic analysis to these datasets, without any editing criterion prior to analysis. We compare our estimates to standard publicly available altimetric tidal constants (from CTOH) and from state-of-the-art numerical models (FES2014 and BAND-AID/SCHISM).
It is found that our analyses allow to extend shoreward the coverage of standard altimetric tidal constants by 5 km to 15 km (depending on the region considered, within the Bay of Bengal). This implies that the editing strategy applied routinely in the standard altimetric products, though probably suited for the offshore domain, is too stringent/conservative for the coastal ocean. It is also found that the choice of the set of altimetric corrections, is as much instrumental in the estimation of tidal constants in the coastal domain as the retracking algorithm can be.
The observational coverage of our altimetric tidal constants estimates, extended towards the shore, opens promising prospects for the tidal modeling community, as it corresponds to the coastal strip where the various tidal models solutions diverge from one another.