Description: Highlights • Shifts in WSBW properties to less dense varieties likely equate to less formation of WSBW. • The decline of WSBW volume ceased around 2005 and likely recovering after that. • Dense Shelf Waters drive and modulate the recent WSBW variability. • WSBW is composed by 71% of mWDW and 29% of Dense Shelf Waters. Abstract The role of Antarctic Bottom Water (AABW) in changing the ocean circulation and controlling climate variability is widely known. However, a comprehensive understanding of the relative contribution and variability of Antarctic regional deep water mass varieties that form AABW is still lacking. Using a high-quality dataset comprising three decades of observational shipboard surveys in the Weddell Sea (1984–2014), we updated the structure, composition and hydrographic properties variability of the Weddell Sea deep-layer, and quantified the contribution of the source waters composing Weddell Sea Bottom Water (WSBW) in its main formation zone. Shifts in WSBW hydrographic properties towards less dense varieties likely equate to less WSBW being produced over time. WSBW is primarily composed of 71 ± 4% of modified-Warm Deep Water (mWDW) and 29 ± 4% of Dense Shelf Waters, with the latter composed by ~ two-thirds (19 ± 2%) of High Salinity Shelf Water and ~ one-third (10 ± 6%) of Ice Shelf Water. Further, we show evidence that WSBW variability in the eastern Weddell Sea is driven by changes in the inflow of Dense Shelf Waters and bottom water from the Indian Sector of the Southern Ocean. This was observed through the rise of the WSBW contribution to the total mixture after 2005, following a twenty-year period (1984–2004) of decreasing contribution.

Description: In the Amundsen Sea, warm Circumpolar Deep Water (CDW) intrudes onto the continental shelf and flows into the ice shelf cavities of the West Antarctic Ice Sheet, resulting in high basal melt rates. However, none of the high resolution global models resolving all the small ice shelves around Antarctica can reproduce a realistic CDW flow onto the Amundsen Sea continental shelf, and previous studies show simulated bottom potential temperature at the Pine Island Ice Shelf front of about −1.8 °C. In this study, using the Finite-Element Sea ice–ice shelf-Ocean Model (FESOM), we reproduce warm CDW intrusions onto the Amundsen Sea continental shelf and realistic melt rates of the ice shelves in West Antarctica. To investigate the importance of horizontal resolution, forcing, horizontal diffusivity, and the effect of grounded icebergs, eight sensitivity experiments are conducted. To simulate the CDW intrusion realistically, a horizontal resolution of about 5 km or smaller is required. The choice of forcing is also important and the cold bias in the NCEP/NCAR reanalysis over the eastern Amundsen Sea prevents warm CDW from intruding onto the continental shelf. On the other hand, the CDW intrusion is not highly sensitive to the strength of horizontal diffusion. The effect of grounded icebergs located off Bear Peninsula is minor, but may act as a buffer to an anomalously cold year.

Description: Previous studies show accelerations of West Antarctic glaciers, implying that basal melt rates of these glaciers were previously small and increased in the middle of the 20th century. This enhanced melting is a likely source of the observed Ross Sea (RS) freshening, but its long-term impact on the Southern Ocean hydrography has never been investigated. Here, we conduct coupled sea-ice/ice-shelf/ocean simulations with different levels of ice shelf melting from West Antarctic glaciers. Freshening of RS shelf and bottom water is simulated with enhanced West Antarctic ice shelf melting, while no significant changes in shelf water properties are simulated when West Antarctic ice shelf melting is small. We further show that the freshening caused by glacial meltwater from ice shelves in the Amundsen and Bellingshausen Seas propagates further downstream along the East Antarctic coast into the Weddell Sea. Our experiments also show the timescales for the freshening signal to reach other regions around the Antarctic continent. The freshening signal propagates onto the RS continental shelf within a year of model simulation, while it takes roughly 5–10 years and 10–15 years to propagate into the region off Cape Darnley and into the Weddell Sea, respectively. This advection of freshening signal} possibly modulates the properties of dense shelf water and impacts the production of Antarctic Bottom Water.

Description: The deep basins of the Bransfield Strait (BS) are ventilated by Weddell Sea (WS) waters from different origins. Depending on the source and density, these water masses follow different routes across the complex topography near the tip of the Antarctic Peninsula and thus into the Bransfield Strait abyss. Using a global setup of the Finite Element Sea-ice Ocean Model (FESOM) we show that the WS waters found at the western WS continental shelf break have a higher influence on the short period variability of BS bottom waters than the waters present over the continental shelf. Adding passive tracers to the glacial melt water (GMW) from two different origins, Larsen Ice Shelf (LIS) and Filchner-Ronne Ice Shelf (FRIS), we show that the GMW from FRIS has a larger influence on BS bottom waters than the GMW from LIS. FRIS GMW has a higher concentration in the BS eastern basin, while LIS GMW is more abundant in the BS central basin. This duality mainly leads to the difference between BS central and eastern basins seen on the observations. This is a novel result and we believe is a significant contribution to the understanding of the BS-WS circulation and interactions.

Description: In the framework of the EU project Ice2sea we utilize a global finite element sea ice - ice shelf - ocean model (FESOM), focused on the Antarctic marginal seas, to assess projections of ice shelf basal melting in a warmer climate. Ice shelf - ocean interaction is described using a three-equation system with a diagnostic computation of temperature and salinity at the ice-ocean interface. A tetrahedral mesh with a minimum horizontal resolution of 4 minutes and hybrid vertical coordinates is used. Ice shelf draft, cavity geometry, and global ocean bathymetry have been derived from the RTopo-1 data set. The model is forced with the atmospheric output from two climate models: (1) the Hadley Centre Climate Model (HadCM3) and (2) Max Planck Institute’s ECHAM5/MPI-OM coupled climate model. Data from their 20th-century simulations are used to evaluate the modeled present-day ocean state. Sea-ice coverage is largely realistic in both simulations. Modeled ice shelf basal melt rates compare well with observations in both cases, but are consistently smaller for ECHAM5/MPI-OM. Projections for future ice shelf basal melting are computed using atmospheric output for IPCC scenarios E1 and A1B. Trends in sea ice coverage depend on the scenario chosen but are largely consistent between the two forcing models. In contrast to this, variations of ocean heat content and ice shelf basal melting are only moderate in simulations forced with ECHAM5/MPI-OM data, while a substantial shift towards a warmer regime is found in experiments forced with HadCM3 output. A strong sensitivity to salinity distribution at the continental shelf break is found for the Weddell Sea, where in the HadCM3-A1B experiment warm water starts to pulse onto the southern continental shelf during the 21st century. As these pulses reach deep into the Filchner-Ronne Ice Shelf (FRIS) cavity, basal melting increases by a factor of three to six compared to the present value of about 100 Gt/yr. By the middle of the 22nd century, FRIS becomes the largest contributor to total ice shelf basal mass loss in this simulation.

Description: Projection of the forthcoming Antarctic contribution to sea-level rise is seriously hampered by the poor ability of current ice sheet models to properly compute comprehensive dynamics of the grounding line. This is a a serious limitation as large sectors present a bedrock below sea level and marine ice sheet instability may occur with drastic inland retreat of the grounding line. In order to circumvent this restriction we prescribe the grounding line migration in the global ice sheet model GRISLI. All regions presenting a bedrock lying below sea level are considered as unstable and a range of plausible migration rates from 500 to 3000 m/yr are imposed. The resulting simulations of sea level change are moderated using projections of future ocean warming in individual regions of the ice sheet’s coast. These latter estimates are based on results from the FESOM high-resolution, finite element ocean circulation model forced by sea-surface boundary conditions based on HadCM3 and ECHAM5 simulations under the A1B scenario. The probability distribution of projected sea-level contribution is estimated by incorporating uncertainty in the rates of grounding line retreat, the areas vulnerable to such retreat and the magnitude of ocean warming likely to trigger retreat.

Description: In the framework of the EU project Ice2sea we utilize a global finite element sea ice - ice shelf - ocean model (FESOM), focused on the Antarctic marginal seas, to quantify heat and freshwater fluxes in the Antarctic ice shelf cavities and to assess ice shelf basal melting in a warmer climate. Ice shelf - ocean interaction is described using a three-equation system with a diagnostic computation of temperature and salinity at the ice-ocean interface. A tetrahedral mesh with a minimum horizontal resolution of 4 minutes and hybrid vertical coordinates is used. Ice shelf draft, cavity geometry, and global ocean bathymetry have been derived from the RTopo-1 data set. Additional simulations were carried out with the circumpolar coarse-scale finite-difference model developed as part of the Bremerhaven Regional Ice Ocean Simulations (BRIOS). Simulations for present-day climate were forced with the NCEP reanalysis product and the atmospheric output from 20th century simulations of the Hadley Centre Climate Model (HadCM3). The projections for the period 2000-2199 use the output of HadCM3 simulations for the IPCC scenarios A1B and E1. Results from both models indicate a strong sensitivity of basal melting to increased ocean temperatures for the ice shelves in Amundsen Sea. An even stronger impact is found for warm water starting to pulse onto the southern Weddell Sea continental shelf in the middle of the 21st century, originating from a redirected coastal current. As these pulses propagate far into the Filchner-Ronne Ice Shelf (FRIS) cavity, basal melting increases significantly compared to the present value of about 100 Gt/yr. At the end of the 21st / beginning of the 22nd century both models suggest a stabilization of FRIS basal mass loss on a high level.