The importance of the western Weddell Sea to Weddell Sea Deep Water formation

The dense water flowing out from the Weddell Sea (WS), the Weddell Sea Deep Water (WSDW), significantly contributes to Antarctic Bottom Water (AABW) and plays an important role in the Meridional Overturning Circulation. However, the relative importance of the western Weddell Sea as a major source region remains unclear. Several studies hypothesized that the continental shelf off Larsen Ice Shelf (LIS) is important for deep and bottom water production, but the role of the Larsen Ice Shelf remains speculative. In this work the importance of the western WS including the LIS to the production of WSDW is investigated using in situ observations and results from numerical simulations. Measurements made during the Polarstern cruise ANT XXIX-3 (2013) in the northwestern WS add evidence to the importance of the western WS as a dense water source. An Optimum Multiparameter Analysis shows that the dense water found near the shelf break in front of the former Larsen A and B ice shelves, together with a very dense water observed off Larsen C Ice Shelf, increases the thickness and changes the _/S characteristics of WSDW that leaves the WS through gaps in the South Scotia Ridge to form AABW. A numerical experiment performed with the Finite Element Sea-ice Ocean Model (FESOM) was used to verify the hypothesis that the continental shelf of the western WS is important for dense water formation. The model results show the changes in the thermohaline properties of the WSDW flowing along the continental slope of the western WS, as well as an increase in the transport downstream. The variability along the continental slope can be explained by fluctuations of the large-scale circulation, namely theWeddell Gyre. In addition, there is no indication that dense waters are formed in the continental shelf of the western WS, and the exchanges between continental shelf and continental slope are small. These results suggest that the area is not important for WSDW formation as previously inferred from the sparse observations mainly along the continental slope. Instead, the western WS seems to be a region where the characteristics of WSDW are determined due to mixing of waters formed upstream. Two sensitivity experiments were designed to investigate whether LIS plays an indirect role in the dense water production: (1) Larsen B Ice Shelf was added to the grid, (2) Larsen C Ice Shelf was completely removed from the grid. The experiments show that LIS plays an important role for the waters on the continental shelf but has only minor importance for the WSDW. Given the disagreement between the hypothesis derived from the observations and the model results, more in situ data are needed to determine whether the western Weddell Sea is a region where dense water is formed or whether it only serves as a conduit for dense waters formed further upstream, which interact in the western WS before reaching the final WSDW characteristics.