Like seeing the back of the moon for the first time
Beneath one of Antarctica's most vulnerable ice shelves, a machine traveled where no human can go and returned with a portrait of the unknown — teardrop-shaped formations and ice dunes carved by forces our best models cannot yet explain. The autonomous vehicle Ran mapped the underside of the Dotson ice shelf in 2022, completing a 1,000-kilometer survey that revealed a subglacial world far more complex and dynamic than science had imagined. What it found does not merely add detail to existing knowledge — it demands that glaciologists rebuild their understanding of how ice shelves melt, and with it, how confidently we can speak about the future of our coastlines.
- Ran surfaced with evidence that the Dotson ice shelf's underside is riddled with unexplained teardrop indentations and ice dunes — formations no existing glacial model predicted or can account for.
- The discovery creates urgent pressure on climate science: sea-level rise projections depend on melt models that are now known to be incomplete, potentially making current forecasts unreliable.
- Researchers are working to decode the subglacial water flows and fracture patterns Ran recorded, hoping to reconstruct the physical processes behind these enigmatic structures.
- In January 2024, Ran vanished beneath the ice on a follow-up dive, cutting the mission short and leaving critical data gaps that will take time and significant resources to fill.
- The work continues — the scientists are committed to replacing Ran and returning — because the ice is still melting, and the questions it raises are only growing more consequential.
When researchers sent an autonomous underwater vehicle called Ran beneath West Antarctica's Dotson ice shelf in 2022, they expected complexity — but not what they found. Over 27 days and more than 1,000 kilometers, the jointly operated American-British mission produced the first high-resolution map of the shelf's underside, and the terrain it revealed felt, to the scientists involved, like seeing the far side of the Moon for the first time.
The landscape beneath the ice was nothing like satellite data or ice cores had suggested. Peaks, valleys, plateaus, and formations resembling sand dunes covered the base — but the most confounding discovery was a series of teardrop-shaped indentations whose origins no existing model of glacial behavior could explain. Lead researcher Anna Wåhlin and co-author Karen Alley described a system far more dynamic and unstable than the field had assumed, with vertical fractures showing signs of intense melting driven by subglacial currents science had not yet characterized.
The stakes are high. Accurate predictions of sea-level rise depend on understanding exactly how Antarctic ice melts from below, and Ran's findings revealed that current models are missing something fundamental. The teardrop structures and melt patterns it recorded represent processes that forecasters cannot yet account for — a gap that matters enormously for coastal planning worldwide.
The urgency deepened in early 2024, when the team returned to extend Ran's surveys. After a single successful dive, the vehicle disappeared on its next run beneath the ice and was never recovered. The loss was a significant setback, though some data was retrieved. Replacing Ran will take time, but the researchers have not stepped back from the work — the mysteries carved into the ice are still there, and so is the need to understand them.
Imagine seeing something no human eye has ever witnessed before—the underside of an ice shelf the size of a small country, mapped in high resolution for the first time. That's what happened when researchers sent an autonomous underwater vehicle called Ran beneath the Dotson ice shelf in West Antarctica. The experience, according to the scientists involved, felt like gazing at the far side of the Moon.
Ran completed its mission in 2022, traveling over 1,000 kilometers in a 27-day survey operated jointly by American and British teams. What the vehicle discovered challenged everything glaciologists thought they understood about how ice shelves work. Beneath the shelf's surface lay a landscape far more complex than satellite data or ice cores had ever suggested—a terrain of peaks and valleys, plateaus and formations that resembled sand dunes, except they were made entirely of ice. But the most puzzling discovery was a series of teardrop-shaped indentations scattered across the underside, their origins and formation mechanisms completely unexplained by existing models of glacial behavior.
Anna Wåhlin, an oceanographer at the University of Gothenburg and lead author of the research published in Science Advances, described the moment of revelation. The high-resolution maps Ran produced showed something scientists had only glimpsed in fragments before: the full, intricate picture of what happens beneath an ice shelf as it melts. The teardrop structures, the team believes, are carved by water flow—but the exact processes remain enigmatic. Vertical fractures throughout the glacier showed signs of intense melting, suggesting that the underside of the shelf is far more dynamic and unstable than previous understanding had indicated.
Karen Alley, a glaciologist from the University of Manitoba and co-author of the study, emphasized the significance of what Ran had uncovered. Scientists had suspected for years that ice-shelf bases were more complex than models suggested, but Ran provided the first comprehensive evidence of just how intricate these systems truly are. The data revealed subglacial currents and melt patterns that will force researchers to rebuild their understanding of how Antarctic ice responds to warming oceans.
That understanding matters enormously. Predicting how fast Antarctic ice will melt—and therefore how much sea levels will rise—depends on accurate models of what happens beneath the ice. The teardrop structures and other formations Ran discovered represent processes that current models cannot account for. Getting those models right is essential for climate forecasting.
But there's a complication. In January 2024, the research team returned to the Dotson shelf to repeat and extend Ran's surveys. After completing just one dive, the vehicle disappeared on its next routine run beneath the ice. The loss was a setback—Wåhlin acknowledged that while the team retrieved valuable data from that final mission, they didn't get everything they had hoped for. Replacing Ran will require time and resources, but the researchers are committed to continuing the work. The mysteries beneath the ice shelf aren't going anywhere, and neither is the urgency of understanding them.
Citas Notables
We've had hints of how complex ice-shelf bases are, but Ran uncovered a more extensive and complete picture than ever before.— Karen Alley, glaciologist, University of Manitoba
Although we got valuable data back, we did not get all we had hoped for. We hope to be able to replace Ran and continue this important work.— Anna Wåhlin, oceanographer, University of Gothenburg
La Conversación del Hearth Otra perspectiva de la historia
What made these teardrop shapes so puzzling that they warranted calling them enigmatic?
Because they don't fit any existing theory of how water and ice interact beneath a shelf. We have models for how glaciers flow, how they melt, how currents move—but these indentations don't match any of those patterns. That's what makes them genuinely mysterious.
So the vehicle was doing something that satellites couldn't do?
Exactly. Satellites see the top of the ice, ice cores give you a thin vertical slice of history. But Ran could actually map the three-dimensional landscape where the ice meets the ocean. It's the difference between reading about a place and walking through it.
When the vehicle disappeared, was there any way to recover it?
Not really. Once it's gone beneath a kilometer of ice in the Antarctic, retrieval becomes nearly impossible. The team had to accept the loss and move forward with plans to replace it.
Does losing the vehicle change what they learned from the first mission?
No—the data Ran collected is still there, still valuable. But it means they can't answer the follow-up questions yet. They can't map seasonal changes, can't test hypotheses about what causes those teardrops. It's like having one photograph of a place you need to understand deeply.
Why does this matter for people who don't study glaciers?
Because Antarctic ice holds enough frozen water to raise sea levels by 58 meters if it all melted. Getting the melt rates right isn't academic—it determines whether coastal cities need to plan for one meter of rise or three. These teardrop structures might be part of why the ice is melting faster than models predict.