Antarctic Bottom Water (AABW) formation is a crucial process in the global ocean overturning circulation and climate regulation. AABW originates from the descent of dense shelf waters (DSW) formed through brine rejection during sea ice growth and ocean-ice shelf interactions. The Weddell and Ross Seas are primary AABW source regions, but recent research identifies other contributing areas like the Adélie Coast and Prydz Bay. While the general process of DSW overflow and descent is understood, the precise dynamical mechanisms influencing the descent speed and the resultant AABW properties remain unclear. Theoretically, Earth's rotation and bathymetry influence downslope flow, but observations show DSW reaching the deep ocean faster than expected by benthic Ekman transport alone. Previous studies suggested tides and TRWs as accelerating mechanisms, but their relative importance and combined effect on AABW formation across various Antarctic overflow regimes is not fully elucidated. This study aims to provide a comprehensive understanding of these dynamical mechanisms and their influence on AABW formation by integrating historical in situ data from the Weddell and Ross Seas with idealized high-resolution numerical model experiments that cover the range of overflow conditions across the Antarctic continental margin.

Existing literature establishes the Weddell and Ross Seas as major AABW source regions. Recent studies have expanded this understanding to include the Adélie Coast and Prydz Bay. A review of nearly 50 years of hydrographic observations reveals widespread dense overflows around the Antarctic continental margin, highlighting the need for a more generalized understanding of the controlling dynamics. Theoretically, steady downslope flows are expected to turn westward and descend slowly via Ekman transport. Topographic steering can facilitate downslope flow, but observations suggest additional accelerating mechanisms are involved. Tides and overflow-forced TRWs have been proposed as such mechanisms. However, the relative contribution of these two processes in the different Antarctic overflow regimes has not yet been fully explored. This existing literature serves as a foundation for this study which focuses on quantifying the individual and combined effects of tides and TRWs on the descent of DSW and the resulting properties of AABW.

This study uses a combination of observational data analysis and numerical modeling. For observational data, historical in situ measurements from the Weddell and Ross Seas were collected. These data include moored measurements of downslope flows, temperature, and salinity from multiple moorings deployed over various periods. In the Ross Sea, CTD sections were also used to analyse the vertical/cross-slope distributions of temperature and salinity. To validate the findings, additional mooring data from the Adélie coastal region were analyzed. The numerical model used was the Regional Ocean Modelling System (ROMS), chosen for its accuracy in representing flows over steep slopes. The model domain was an idealized embayment connected to a flat abyssal ocean via a linear slope with varying steepness. A stretched horizontal grid with high resolution near the trough mouth and increased vertical resolution near the sea floor were implemented. The model incorporated various physical parameterizations, including the Mellor-Yamada level 2.5 turbulence closure scheme for vertical viscosity and mixing. Tidal forcing was included in some experiments using data extracted from the TPXO7 tidal product for the K1 tidal constituent. Passive tracers were used to track the overflow. Different experiments were conducted, varying parameters such as slope steepness, tidal forcing, and DSW density to explore the dynamics across a wide range of Antarctic overflow conditions. Key diagnostics included zonally-integrated diapycnal tracer mass flux, pycnocline depth, and the isobath-weighted tracer center of mass to quantify the DSW descent pathway and entrainment. Tidal flow strength and excursion were calculated from the model and observed data.

Analysis of moored measurements from the Weddell and Ross Seas revealed significant downslope flows with periodic oscillations. The Ross Sea showed mainly diurnal tidal oscillations, while the Weddell Sea exhibited oscillations consistent with TRWs. In the Ross Sea, coldest waters emerged during spring tides, indicating tidal advection of DSW across the continental slope rather than tidal mixing as the primary mechanism. Numerical model experiments showed that in Ross Sea-like conditions, tides advected tracers to greater depths and reduced the rate of DSW mixing with overlying waters, leading to pycnocline deepening during ebb tides and the formation of a V-shaped front. In Weddell Sea-like conditions, TRWs were the dominant driver of DSW descent, also leading to the formation of a V-shaped front. Sensitivity experiments with varying slope steepness revealed four dynamical regimes: Mixing (steep slope, weak tides), Tidal (steep slope, strong tides), Wavy (moderate slope, TRWs dominant), and Eddying (gentle slope, eddy formation). The Mixing regime is least favorable for AABW formation, while the others are more favorable. Applying this classification to the entire Antarctic margin suggests that most regions are favorable for AABW formation, assuming DSW production. Analysis of mooring data from the Adélie Coast and Prydz Bay supported the model-predicted regimes for those locations. The study also found that varying DSW density did not significantly alter the overflow dynamics.

This study demonstrates that both tides and TRWs significantly influence AABW formation by accelerating DSW descent and reducing entrainment. Tides play a major role only on steep slopes where TRWs are suppressed, while TRWs dominate on gentler slopes. Both mechanisms facilitate the formation of a V-shaped hydrographic front, bringing DSW and surface waters into contact and influencing AABW properties. This highlights the potential importance of surface water changes in AABW production, particularly in the context of climate change. The study's limitations include the exclusion of the Antarctic Slope Front/Current system and the focus on idealized model simulations. Future research should address these aspects and investigate the impact of biogeochemical transport on abyssal storage and long-term climate change.

This study provides a comprehensive understanding of AABW formation mechanisms, identifying the crucial roles of tides and TRWs and characterizing four distinct dynamical regimes. The findings improve our knowledge of the global overturning circulation and highlight the need for improved representation of high-frequency processes in climate models. Future research should focus on more realistic model simulations incorporating the Antarctic Slope Front/Current and exploring the biogeochemical implications of tidal and TRW-mediated transport.

The study uses idealized numerical models, which may not fully capture the complexity of real-world Antarctic overflows. The impact of the Antarctic Slope Front/Current system and the potential effects of changes in surface water properties on AABW were not explicitly explored. The classification of dynamical regimes is sensitive to local slope steepness, and the model does not resolve all scales of variability. Further observational data are needed to refine the model's representation and to investigate small-scale dynamics.