Full text: Download
The Amazon River is by far the largest river by volume of water in the world, representing around 17% of the global riverine discharge to the oceans. Recent studies suggested that its impact on sea level is potentially important at global and regional scales. This study uses a set of regional simulations based on the ocean model NEMO to quantify the influence of the Amazon runoff on sea level in the Tropical Atlantic Ocean. The model is forced at its boundaries with daily fields from the ocean reanalysis GLORYS2V4. Air-sea fluxes are computed using atmospheric variables from DFS5.2, which is a bias-corrected version of ERAinterim reanalysis. The particularity of this study is that interannual daily runoffs from the up-to-date ISBA-CTRIP land surface model are used. Firstly, mean state of sea level is investigated through a comparison between a simulation with an interannual river discharge and a simulation without any Amazon runoff. Then, the impact of the Amazon River on seasonal and interannual variability of sea level is examined. It was shown that the Amazon River has a local contribution to the mean state sea level at the river mouth but also a remote contribution of 3.3 cm around the whole Caribbean Archipelago, a region threatened by the actual sea level rise. This effect is mostly due to a halosteric sea level contribution for the upper 250 m of the ocean. This occurs in response to the large scale advection of the plume and the downward mixing of subsurface waters at winter time. The Amazon discharge also induces an indirect thermosteric sea level contribution. However, this contribution is of second order and tends to counterbalance the halosteric sea level contribution. Regional mass redistributions are also observed and consist in a 8 cm decrease of the sea level at the river mouth and a 4.5 increases on continental shelves of the Gulf of Mexico and Caribbean Sea. In terms of variability, simulations indicate that the Amazon discharge may contributes to 23% and 12% of the seasonal and interannual sea level variances in the Caribbean Archipelago area. These variances are first explained by the Amazon time mean discharge and show very weak sensitivity to the seasonal and interannual variability of the Amazon runoff.