The Radical Plan to Refreeze Antarctica

It is difficult to ignore how frequently the climate debate returns to the same unsolvable question: what will happen if cutting emissions isn’t going to be sufficient? The majority of sincere people, including scientists, will tell you in private that it isn’t. A small group of British engineers in parkas drilling holes through 80 centimeters of Arctic ice, lowering pumps into the seawater below, and watching it freeze on the surface is one explanation that is becoming more and more plausible. The ice thickens. At least that part is true.

The team is part of Real Ice, a non-profit organization in the UK led by Andrea Ceccolini, a former software entrepreneur whose first business went global. Last winter, he traveled to Cambridge Bay in Arctic Canada, where temperatures were as low as minus 47 degrees Celsius. He lost feeling in his toes for weeks, and he told an Australian podcast that he thought he might actually die. The experiment covered an area the size of a soccer field with about 20 inches of ice. His crews increased the test plot tenfold by the end of 2024. Four inches of fresh ice had already formed after ten days of pumping.

Things become more intriguing and speculative at the Antarctic angle. As of right now, Real Ice has not disclosed any plans to conduct business at the South Pole. Last year, Ceccolini made similar remarks. The southern continent is structurally different from the Arctic in that it is a land mass encircled by ocean rather than an ocean ringed by land, and although Antarctic ice loss has accelerated, it hasn’t exactly followed the same vertical line. However, if the Arctic trials are valid, the reasoning behind the concept extends beyond one hemisphere. Antarctica becomes the next obvious frontier once you’ve determined that pumping seawater onto sea ice can buy time, and that buying time has quantifiable value. Simply put, it’s much more difficult.

The part that silently shatters your brain is the numbers. The wind-pump version of this idea was first proposed in 2017 by Arizona State physicist Steven Desch, who estimated the cost at $500 billion. Thickening ice over 386,000 square miles of the Arctic, an area more than twice the size of California, is Real Ice’s more modest goal, and it would cost between $5 and $6 billion annually. The company plans to use about 500,000 hydrogen-powered submersible drones, each about six and a half feet long, to pump water up by melting holes in the ice from below. You cannot accuse the plan of being timid, regardless of your opinion of it.

Naturally, there is a lot of skepticism. Late last year, Liz Bagshaw, a polar scientist at the University of Bristol, told CNN that Real Ice’s approach’s scalability was “extremely questionable” and that the larger category of interventions was “morally dubious at best.” In a recent report, dozens of researchers cautioned about the “grave unforeseen consequences” of polar geoengineering, including the peculiar new burden of spreading “an unprecedented level of human presence” throughout some of the planet’s last unspoiled ecosystems. Although sympathetic, Jennifer Francis of Woodwell Climate Research Center continues to have “serious doubts” about the approach’s ability to scale sufficiently to be significant.

As you watch this develop, you get the impression that the argument has changed in a subtle but significant way. Refreezing the poles was a novel idea five years ago, something climate scientists would have scoffed at over coffee. Today, “Refreeze the Arctic” is openly listed as one of the three institutional missions of the Centre for Climate Repair, which has its headquarters in the applied mathematics department of Cambridge, along with emissions reduction and greenhouse gas removal. Sir David King, the company’s founder, served as Tony Blair’s primary scientific advisor. It is difficult to write him off as a crank. One indication of how the Overton window has shifted is the fact that a reputable British university is currently organizing research into pumping seawater onto polar ice.

Observing this from a distance, what worries me the most is how much of the conversation still sounds like a financial dispute wrapped in glaciological terminology. Selling “cooling credits” would allow polluters to purchase refreezing in the same manner as they currently purchase carbon offsets, according to Real Ice. Investors are intended to feel reassured by the phrase. It also makes the trap visible. Someone must pay if governments don’t. In reality, that someone will be the same industries whose emissions caused the ice to melt in the first place. It mostly depends on who is handling the accounting whether that arrangement is climate repair or climate laundering.

In this way, Antarctica is more of a mirror than a destination. The more difficult question—whether civilizations can engineer their way out of consequences they engineered into existence—will remain unanswered at the bottom of the world, regardless of the outcome of the Arctic experiment. The answer’s likelihood of being liked by everyone is still unknown.