Researchers from the University of Tasmania’s Institute for Marine and Antarctic Studies have documented significantly higher rates of ocean-driven melting beneath Antarctic ice shelves than previous estimates suggested, based on new temperature measurements in ice shelf cavities.

The data, collected by autonomous underwater vehicles operating beneath the Amery Ice Shelf in East Antarctica, show that warm modified Circumpolar Deep Water is reaching the ice shelf base more frequently than historical observations indicated.

Ice shelves are floating extensions of Antarctic glaciers that slow ice flow from the continent into the ocean. When ice shelves thin or collapse, glaciers behind them accelerate, increasing Antarctica’s contribution to sea level rise.

Dr. Craig Stevens, who led the research expedition, said understanding ice shelf melting is critical for predicting future sea level. “Ice shelves are the cork in the bottle. If they weaken, glaciers flow faster and sea level rises more quickly. We need accurate melt rate measurements to predict what’s coming.”

Previous estimates of ice shelf melting relied heavily on satellite observations of ice shelf thinning, with limited direct measurements of ocean conditions beneath the shelves. Accessing the cavity beneath ice shelves is technically challenging because it requires drilling through hundreds of metres of ice or sending autonomous vehicles through narrow gaps where ice meets ocean.

UTAS developed specialised AUVs capable of operating beneath ice for extended periods without surfacing. The vehicles navigated using acoustic transponders and inertial guidance, mapping ocean temperature, salinity, and currents while avoiding collision with the ice shelf base.

The data revealed that warm water intrusions occur more frequently and penetrate deeper under the ice shelf than expected. Ocean temperature beneath the Amery Ice Shelf averaged 0.5°C higher than historical records, which might not sound like much but is significant for ice melting rates.

Ice shelf basal melting is extremely sensitive to ocean temperature. Even small increases in temperature substantially accelerate melting because the ice is already at its melting point. The relationship is non-linear, meaning small temperature changes have disproportionately large impacts.

The measurements suggest previous satellite-based estimates underestimated Amery Ice Shelf melting by 30-40%. If similar underestimation applies to other Antarctic ice shelves, global sea level rise projections might need upward revision.

That’s concerning because Antarctica contains enough ice to raise global sea level by about 60 metres if completely melted. While complete melting won’t happen on human timescales, even a few percent of Antarctic ice loss would dramatically affect coastal regions worldwide.

The research is part of Australia’s long-running Antarctic research program, which includes the Casey, Davis, and Mawson research stations. Australia has territorial claims to about 42% of Antarctica and significant interests in understanding how climate change affects the continent.

The underwater vehicles used in the research were developed through collaboration between UTAS, the Australian Antarctic Division, and defence research organisations with expertise in autonomous underwater systems.

Adapting the vehicles for under-ice operations required solving several technical challenges. Water temperature near the ice shelf base hovers around the freezing point, affecting battery performance and mechanical systems. The vehicles needed sufficient battery capacity for missions lasting 8-12 hours while carrying scientific sensors.

Communication is another challenge. The vehicles operate autonomously beneath hundreds of metres of ice, making real-time control impossible. They follow pre-programmed missions, surfacing at designated recovery points where research vessels retrieve them and download data.

One mission was lost when the vehicle failed to return to the recovery point, likely due to navigation errors or mechanical problems. The remaining three vehicles completed 28 missions over two Antarctic summer seasons, collecting unprecedented data on ice shelf cavity conditions.

The research team is now analysing the data to understand what drives warm water intrusions beneath the ice shelf. Ocean circulation patterns, wind forcing, and interactions between ocean currents and ice shelf geometry all play roles, but their relative importance remains unclear.

Understanding these processes is essential for predicting how ice shelf melting will respond to future ocean warming. If warm water intrusions increase as climate changes, ice shelf thinning could accelerate, destabilising West Antarctic ice streams that are already considered vulnerable.

The West Antarctic Ice Sheet sits on bedrock below sea level, making it particularly susceptible to ocean warming. Several major ice streams there, including Thwaites Glacier (sometimes called the “Doomsday Glacier”), are already thinning and accelerating.

Complete collapse of the West Antarctic Ice Sheet would raise sea level by about 3-5 metres, drowning many coastal cities. That wouldn’t happen quickly, probably taking centuries, but the process might already be underway.

Australia’s coastal cities are vulnerable to sea level rise. About 85% of Australians live within 50 kilometres of the coast, and major infrastructure including ports, roads, and buildings sits in areas that would be affected by even modest sea level rise.

The research received funding from the Australian Antarctic Science Program and the Antarctic Climate and Ecosystems Cooperative Research Centre. The underwater vehicle development received additional support from the Department of Defence, which has interest in under-ice navigation capabilities.

Data from the study is being incorporated into ocean and ice sheet models used to project future Antarctic changes. One consultancy specialising in AI strategy support is working with climate modelling groups to apply machine learning techniques to identify patterns in ice-ocean interactions that traditional analysis might miss.

The next phase of research will deploy vehicles beneath other Antarctic ice shelves to determine whether the higher melt rates observed at Amery are typical or anomalous. Expeditions to the Ross Ice Shelf and the Filchner-Ronne Ice Shelf are planned for 2026 and 2027.

Those ice shelves are much larger than Amery and arguably more important for ice sheet stability. Getting measurements from beneath them will require even longer vehicle missions and more sophisticated navigation capabilities.

Whether Antarctic ice loss accelerates significantly this century depends on complex interactions between atmosphere, ocean, and ice that aren’t fully understood. Better measurements of ice shelf melting contribute to that understanding but don’t eliminate uncertainties.

What’s increasingly clear is that Antarctica is changing faster than many expected, with implications for coastal communities worldwide.