Antarctic Bottom Water (AABW) formation and transport constitute a key component of the global ocean circulation. Direct observations suggest that AABW volumes and transport rates may be decreasing, but these observations are too temporally or spatially sparse to determine the cause. To address this problem, we develop a new method to reconstruct AABW transport variability using data from the GRACE (Gravity Recovery and Climate Experiment) satellite mission. We use an ocean general circulation model to investigate the relationship between ocean bottom pressure and AABW: we calculate both of these quantities in the model, and link them using a regularised linear regression. Our reconstruction from modelled ocean bottom pressure can capture 65-90% of modelled AABW transport variability, depending on the ocean basin. When realistic observational uncertainty values are added to the modelled ocean bottom pressure, the reconstruction can still capture 30-80% of AABW transport variability. Using the same regression values, the reconstruction skill is within the same range in a second, independent, general circulation model. We conclude that our reconstruction method is not unique to the model in which it was developed and can be applied to GRACE satellite observations of ocean bottom pressure. These advances allow us to create the first global reconstruction of AABW transport variability over the satellite era. Our reconstruction provides information on the interannual variability of AABW transport, but more accurate observations are needed to discern AABW transport trends.

Recent ice loss on the western Antarctic Peninsula has been driven by warming ocean waters on the continental shelf. However, due to the short observational record, our understanding of the dynamics and variability in this region remains poor. High-resolution ocean model simulations show that the temperature variability along the western Antarctic Peninsula is controlled by the rate of dense water formation in the Weddell Sea. Passive tracer advection reveals connectivity between the Weddell Sea and the coastline of the western Antarctic Peninsula and Bellingshausen Sea. During multi-year periods of weak Weddell dense water formation, dense overflow transport in the Weddell Sea decreases, while the transport of cold water around the tip of the Antarctic Peninsula strengthens, driving a temperature decrease of 0.4oC along the western Antarctic Peninsula. This mechanism implies that western Antarctic Peninsula coastal ocean temperature may cool in the future if Weddell Dense Shelf Water production slows down.

Antarctic Bottom Water (AABW) is a major component of the global overturning circulation, originating around the Antarctic continental margin. In recent decades AABW has both warmed and freshened, but there is also evidence of large interannual variability. The causes of this underlying variability are not yet fully understood, in part due to a lack of ocean and air-sea-ice flux measurements in the region. Here, we simulate the formation and export of AABW from 1958 to 2018 using a global, eddying ocean–sea-ice model in which the four AABW formation regions and transports agree reasonably well with observations. The simulated formation and export of AABW exhibits strong interannual variability which is not correlated between the different formation regions. Reservoirs of very dense waters at depth in the Weddell and Ross Seas following 1-2 years of strong surface water mass transformation can lead to higher AABW export for up to a decade. In Prydz Bay and at the Adélie Coast in contrast, dense water reservoirs do not persist beyond 1 year which we attribute to the narrower shelf extent in the East Antarctic AABW formation regions. The main factor controlling years of high AABW formation are weaker easterly winds, which reduce sea ice import into the AABW formation region, leaving increased areas of open water primed for air-sea buoyancy loss and convective overturning. Our study highlights the variability of simulated AABW formation in all four formation regions, with potential implications for interpreting trends in observational data using only limited duration and coverage.