Antarctic Bottom Water: How it Contributes to the Global Ocean Circulation

• Around Antarctica, cold, salty water sinks and acts like a global ocean pump, moving water around the planet on a multi-centennial scale [1, 2, 5, 11, 12]. This is known as Antarctic Bottom Water (AABW). 
• This massive deep-ocean circulation system helps regulate Earth’s climate, oxygen levels, and carbon storage [2, 6, 10, 12].
• This circulation is slowing as meltwater from increased ice shelf basal melting freshens and lightens the water on the Antarctic shelf, inhibiting the formation of Bottom Water (though distinguishing a slowdown from changes in abyssal water properties remains challenging). While subject to natural variability, this process is projected to accelerate significantly under high-emissions scenarios and anthropogenic forcing [6, 7, 8, 10, 12, 13].
• A weaker ocean circulation system means less carbon absorbed by the deep ocean, lower oxygen levels in deep waters, and more unstable climate/weather patterns worldwide as heat sequestration is disrupted [3, 6, 9, 10, 12].
• What happens in Antarctica affects every country through impacts on weather and rainfall patterns, sea level rise, fisheries, and future generations, which is why action this decade matters [3, 8, 10, 12].

Fig 1: Graphic summarising the observed and projected changes of Antarctic Bottom Water.

The world’s oceans are connected by a global circulation system driven partly by differences in water density. Around Antarctica, extremely cold and salty water becomes dense enough to sink to the ocean floor and spread through the deep ocean [1, 2, 12]. This process, known as AABW formation, helps drive the global ocean circulation and heat distribution [5, 11, 12].

AABW fills around 30–40% of the deep ocean and plays a critical role in regulating Earth’s climate [1, 2, 12]. It transports oxygen to the deep sea, stores carbon away from the atmosphere for centuries, and redistributes heat between ocean basins [6, 10, 12]. 

Fig 2: Schematic of important Antarctic processes that play a role in Antarctic Bottom Water formation (Image credits: Antarctic Science Platform)

AABW forms in four primary regions, the Weddell Sea, Ross Sea, Adélie Land, and Cape Darnley, each characterised by unique physical drivers [6, 11, 12]. While all four regions rely on sea-ice production and brine rejection in coastal polynyas, the Weddell and Ross systems are also strongly shaped by large continental embayments and ice-shelf interactions that contribute to building water density [2, 11, 12]. Understanding these diverse regional mechanisms is essential for improving global climate models, which have historically struggled to accurately replicate complex Antarctic shelf dynamics and overflow processes [9, 10, 12]. Ultimately, expanded research across all four sites will enable scientists to provide the high-resolution data needed for more reliable predictions of how global ocean circulation will respond to future climate forcing [1, 2, 7, 10, 12].

Fig 3: Map of annual sea ice production (in meters per year) and generation of Antarctic Bottom Water. The sea ice area is shown in white, and you can see that there is a high amount of new sea ice produced near the coasts, in coastal polynyas. The coastal polynyas circled in blue dots are those that make enough sea ice to produce AABW.

Observed changes in AABW properties provide direct evidence that this global circulation is weakening [2, 6, 12]. Since the mid-1980s, the AABW layer has warmed, freshened, and thinned across multiple basins around Antarctica, indicating a widespread reduction in the volume of the densest waters [12]. Basin-specific observations support this trend, with a contraction of 50–120 meters recently measured in the Indian sector and a 4.0 Sverdrup (Sv; a unit of ocean flow equal to one million cubic metres of water per second) drop in AABW volume transport observed in the Australian Antarctic Basin between 1994 and 2009 [6]. 

While natural climate variability can drive localised recoveries, the net abyssal overturning has slowed at a rate of 0.8 Sv per decade over the past 30 years [6, 12]. Under high-emissions scenarios, climate models project this decline will accelerate significantly, reaching a 40–42% reduction in circulation strength by 2050 as freshening from Antarctic meltwater continues to suppress dense water formation [10, 12]. 

Fig 4: Schematic of AABW transport from Antarctic shelves to the Australian deep ocean basins from 1990-2017 [6], with 2017-2050 predictions calculated based on findings in Li et.al, 2023 [10] (42% reduction applied to the 1994 peak transport of 5.9 Sv identified in [6])

Why these studies matter for policy

• Global Climate and Carbon Regulation: AABW drives the lower limb of the global overturning circulation, redistributing heat, carbon, oxygen, and nutrients throughout every major ocean basin [2, 5, 10, 12]. As AABW accounts for 30–40% of the total global ocean volume, it also acts as a critical conduit for transferring anthropogenic heat and CO2 into the deep ocean, where it can remain sequestered for centuries [1, 2, 12]. Any weakening of this system would both destabilise Earth’s climate regulation and limit the ocean’s capacity to buffer global warming [2, 3, 6, 10, 12].
• Deep Ocean Ecosystems: As the primary source of oxygen for the abyssal ocean, AABW is vital for sustaining life in deep-sea environments [2, 6, 12]. Recent observations show that a slowdown in overturning is already driving abyssal deoxygenation and the thinning of well-ventilated water layers, placing significant stress on ecosystems already facing climate pressures [3, 6, 12].
• Sea Ice and Freshwater Disruptions: AABW production is fundamentally dependent on the formation of Dense Shelf Water (DSW), created through intense sea-ice production and brine rejection in coastal polynyas [4, 11, 13]. Ongoing Antarctic sea-ice loss and increased freshening from melting ice shelves threaten to disrupt these “sea-ice factories,” reducing the density of sinking water and weakening the global pump [2, 7, 10, 12, 13].
• Improving Climate Projections: Historically, global climate models have struggled to accurately replicate the complex Antarctic shelf processes and regional water mass variants that drive AABW formation [4, 8, 9, 12]. By better identifying and representing regional mechanisms, such as specific bottom water types or coastal polynyas, scientists can enhance the accuracy of global models, providing more reliable data for future environmental planning and policy [2, 7, 9, 10, 12].

Fig 4: Schematic of AABW transport from Antarctic shelves to the Australian deep ocean basins from 1990-2017 [6], with 2017-2050 predictions calculated based on findings in Li et.al, 2023 [10] (42% reduction applied to the 1994 peak transport of 5.9 Sv identified in [6])

Key uncertainties and research gaps

• Sea-Ice Measurements: There is a critical lack of reliable, year-round, circumpolar data on sea-ice thickness, which is essential for predicting how quickly ice will retreat [2, 3, 12, 13]. 
• Deep Ocean Baseline: Scientists need more high-resolution, long-term observations to establish a baseline, making it easier to identify how the deep ocean responds to extreme events [2, 3, 12]. 
• Model Accuracy: Current climate models struggle to accurately simulate complex shelf processes and often do not account for dynamic melting of the Antarctic ice sheet [8, 9, 10, 12].
• Regional Variability: More research is needed to understand why different formation regions (such as the Weddell vs. Ross Seas) are changing for different reasons and at different speeds [1, 6, 12]. 
• Cape Darnley Pathways: Despite its importance, Cape Darnley remains the least-studied site; we still need to accurately measure exactly how much water is exported through its canyons [2, 4, 6, 7].
• Ecosystem Impacts: The capacity for Antarctic wildlife to adapt to rapid shifts in sea ice and thinning of well-ventilated (oxygen-rich) water layers remains largely unknown [3, 6, 12].

Watch the short explainer video above for a visual overview of how Antarctic Bottom Water forms and why it matters for global climate and ocean health, or see the list of key references below for further peer reviewed reading.

Other relevant portal pages

• Antarctic Sea Ice #4: Record lows between 2022 and 2025  
• Antarctic Sea Ice #3: Trends and Future Projections  
• The warming of the Southern Ocean and its significance – Antarctic Environments Portal 
• Ten scientific messages on risks and opportunities for life in the Antarctic  
• Antarctica’s uncertain contribution to sea level rise – Antarctic Environments Portal

This article and the accompanying videos were created by Sienna Blanckensee (University of Queensland, Australia) as part of the Ice Narratives Fellowship.

Reach out to Sienna at [email protected].