Adrift like Shackleton: Robot float survives Antarctic ice
But here’s where the story gets unfolded: a small, resilient robot drifted under East Antarctica’s massive ice shelves, gathering temperature and salinity data from ocean regions never before sampled. Over two and a half years, an Argo float with oceanographic sensors logged nearly 200 profiles along a 300-kilometer sweep between the Denman and Shackleton ice shelves. At times it vanished beneath the ice, yet it endured and transmitted the first ocean transect beneath an East Antarctic ice shelf.
“Our luck held,” notes oceanographer Dr. Steve Rintoul from CSIRO, Australia’s national science agency, partnering with the Australian Antarctic Program Partnership at the University of Tasmania.
“The float drifted beneath the ice for eight months, recording profiles from the seafloor up to the ice base every five days.” These unprecedented measurements illuminate how vulnerable the ice shelves are to ocean forcing.
Key findings show that the Shackleton ice shelf—the northernmost in East Antarctica—currently remains insulated from warm water capable of melting it from below, suggesting it is less vulnerable for now. In contrast, the Denman Glacier sits at a tipping point: warm water can reach beneath it, and relatively small changes in the warm-water layer could trigger much higher melt rates, potentially driving unstable retreat and contributing more to global sea level rise.
Heat transfer from ocean to ice hinges on the ten-meter-thick boundary layer just below the ice shelf. “A major advantage of floats is their ability to measure the boundary-layer properties that govern melt rates,” explains Dr. Rintoul. The data from these floats will refine how such processes are represented in computer models, reducing uncertainty in future sea-level projections.
Expanding the network of floats along the Antarctic continental shelf could transform understanding of ice-shelf vulnerability to ocean changes and, by extension, shrink the largest uncertainty surrounding sea-level rise estimates.
As Prof Delphine Lannuzel, leader of the Australian Antarctic Program Partnership, remarked after sampling near the Denman and Shackleton shelves earlier this year: “Against the enormity of such a wild region, this is the amazing story of the little float that could. Under incredibly testing conditions, a relatively tiny instrument has delivered a wealth of invaluable information.”
Published in Science Advances: Rintoul, van Wijk, Herraiz-Borreguero, and Rosevear (2025) circled the circulation and ocean–ice shelf interaction beneath the Denman and Shackleton Ice Shelves. The research team includes CSIRO, the Australian Antarctic Program Partnership, and the Institute for Marine and Antarctic Studies at the University of Tasmania, with support from Australia’s Integrated Marine Observing System (IMOS), enabled by the National Collaborative Research Infrastructure Strategy (NCRIS).
BACKGROUND CONTEXT
Rising sea levels threaten hundreds of millions living along coasts, including low-lying islands, deltas, and cities. Determining how much Antarctica will contribute to future sea-level rise is the single largest source of uncertainty in projections. Some of the most vulnerable ice lies in East Antarctica, once thought insulated from warm ocean waters. New observations, however, reveal large ice volumes in East Antarctica that could be at risk.
The Antarctic Ice Sheet’s stability depends on floating ice shelves that buttress glacier flow into the ocean. Glaciers begin to float as they move from land to sea, forming shelves that resist inland ice from displacing into the ocean. If these shelves weaken or collapse, more ice can reach the sea, driving higher sea levels. The decisive factor is how much ocean heat reaches the base of these floating shelves.
Directly observing processes in ice shelf cavities is challenging. Ice shelves can be hundreds to thousands of meters thick. Traditional drilling to place sensors through the ice is costly and infrequent, so measurements remain limited. Drift-worthy floats, carried by currents and periodically surfacing to profile temperature and salinity, offer a practical alternative.
Dr. Rintoul reflected on the challenges: “The obstacle was communication—ice blocked the float’s surface path, so GPS and satellite transmission weren’t available.” The team solved this by correlating the float’s occasional contact with the ice to ice draft measurements, aligning these data with satellite estimates to reconstruct the float’s path beneath the shelf.
This study marks a milestone in understanding how ocean conditions influence ice-shelf melt and, ultimately, sea-level rise. The findings point toward a future where increased deployment of ocean-observing tools could sharpen predictions and inform climate policy discussions.
