Submarine vanishes in Antarctic depths and uncovers a hidden side of the South Pole

Antarctic fieldwork can be both revelatory and unforgiving. The University of Gothenburg’s autonomous submersible Ran mapped the hidden base of West Antarctica’s Dotson Ice Shelf—and, on a later mission under Thwaites Glacier, was itself lost to the ice.¹

Diving beneath the ice to reveal the unseen

In one of the most ambitious Antarctic missions to date, an international team led by the University of Gothenburg sent Ran into the cavity beneath the Dotson Ice Shelf. Over 27 days, the AUV travelled more than 1,000 km and reached 17 km into the ice-shelf cavity, collecting the first high-resolution multibeam maps of the ice underside—data long missing from models that mostly rely on satellite altimetry and sparse boreholes.²

Did you know?
Under-ice sorties must thread narrow, shifting passages; a single navigation or comms failure can strand a vehicle until batteries die—one reason mission risk is high even for experienced teams.

A secret world beneath the glacier

The maps reveal an otherworldly landscape: terraces, steep fracture walls with enhanced melt, and previously unknown teardrop-shaped cavities (≈20–300 m long) carved upward in high-melt zones. These patterns point to diverse basal-melt mechanisms driven chiefly by ocean current speed, heat content, and interactions with basal fractures inside the cavity—rather than simple “underground rivers.”³

Did you know?
Dotson has thinned in recent decades primarily due to basal melting—warm, salty water accessing the shelf’s underside and eroding it from below.⁴

Rewriting the models

The Dotson results challenge oversimplified melt prescriptions still common in some ice–ocean models. Because basal melt can vary by orders of magnitude under a single shelf, process-resolving observations like Ran’s are now being used to calibrate and constrain simulations that feed into sea-level projections. How shelves thin and lose buttressing governs the acceleration of inland glaciers—and thus future sea-level rise.⁵

A game-changer for climate science

Detailed under-ice maps don’t replace satellites; they explain them. By tying geometry to local currents, temperature and salinity, these surveys help decode what remote sensing sees from above and plug a critical gap in prediction. Better modelling means planners can work with tighter ranges when designing coastal defences and setting adaptation timelines.

Looking further into the frozen unknown

The Dotson campaign shows why under-ice AUVs matter: they deliver co-located ice geometry and ocean measurements that satellites can’t reach. Teams are already planning deeper penetrations and comparative surveys at other shelves to test how general these melt patterns are. Antarctica still guards many secrets—but each mission makes our sea-level forecasts a little less uncertain.

Footnotes

1. University of Gothenburg — Ran AUV page (loss confirmed January 2024): https://www.gu.se/en/skagerak/ran-autonomous-underwater-vehicle-auv

2. University of Gothenburg — “A whole new view on glacier melting in Antarctica” (Dotson mission; 27 days; >1,000 km; 17 km): https://www.gu.se/en/news/a-whole-new-view-on-glacier-melting-in-antarctica

3. Science Advances — Wåhlin et al. (2024), “Ice base melt revealed by multibeam imagery of an Antarctic ice shelf” (Dotson maps; melt mechanisms): https://www.science.org/doi/10.1126/sciadv.adn9188

4. NSIDC — “Why Ice Shelves Matter” (basal melt, buttressing context): https://nsidc.org/learn/parts-cryosphere/ice-shelves/why-ice-shelves-matter

5. NASA Sea Level — “How Ice-Shelf Loss Drives Sea Level Rise” (buttressing and projections): https://sealevel.nasa.gov/news/266/how-ice-shelf-loss-drives-sea-level-rise/