Antarctica Is Gaining Ice. It Is Also Losing More Ice Than at Any Point in the Satellite Record. Both Are True, and Neither Means What You Have Been Told.

ISSUE #050 · MYTHS & MISINFORMATION · AUGUST 2026

Almost every argument about Antarctic ice is really two arguments happening at once, conducted by people using the same word for two entirely different things.

So before anything else: there are two kinds of ice down there.

Sea ice is frozen ocean. It floats, it forms every winter and largely disappears every summer, and when it melts it changes sea level by essentially nothing — the same reason a melting ice cube does not overflow your glass.

The ice sheet is different. It sits on rock. It averages around a mile and a half thick and holds roughly sixty per cent of all the fresh water on the planet. When it loses mass, the sea rises.

These two things can, and currently do, move in opposite directions. Almost every misleading claim about Antarctica — from either direction — depends on someone quietly swapping one for the other.

Here is the fact that ought to be much better known than it is.

From the start of satellite gravimetry in 2002 until around 2020, the Antarctic ice sheet lost mass every year, and the losses accelerated: roughly 74 gigatonnes a year in the 2000s, nearly doubling to about 142 gigatonnes a year through the 2010s.

Then it stopped. Since 2016 the losses slowed, and from 2020 the sheet began putting on mass — a net gain of about 68 gigatonnes a year across 2020 to 2024, according to work published in Communications Earth & Environment in early 2026. A separate Chinese analysis of the same satellite data found an even sharper spike, around 108 gigatonnes a year for 2021 to 2023, briefly cancelling about 0.3 millimetres of annual sea level rise.

As of late 2025 there was no clear sign of a return to net loss.

This is not a rounding error, and it is not a fringe claim. It is a five-year trend in the best measurement system we have. Anyone who tells you Antarctica is simply melting, full stop, is not describing the last five years accurately.

The obvious question is what changed. The answer is the part that gets left out.

The ice sheet is not gaining because Antarctica got colder. It is gaining because it is snowing considerably more, and the extra snow is arriving for three linked reasons: more frequent atmospheric rivers carrying moisture south, stronger westerly winds, and reduced sea ice around the continent.

That last one deserves a moment. Sea ice acts as a lid on the Southern Ocean. Take the lid off and you expose open water, open water evaporates, and that moisture has to come down somewhere. A good deal of it comes down as snow on the ice sheet.

There is a second mechanism working the same way. Warmer air holds more water. A warmer atmosphere over the Southern Ocean is a wetter one, and a wetter one delivers heavier snowfall to the interior.

So the growth of the ice sheet is being driven, in part, by the loss of the sea ice and by the warming of the air above it. The single most-cited piece of evidence that the climate is not changing is partly a product of the climate changing.

Now the detail that makes the whole story turn, and which almost no coverage on either side mentions.

Antarctica is not losing less ice. It is losing more.

Ice leaves the continent by two routes. Snow falls and adds mass at the top. Glaciers flow to the coast and calve icebergs into the ocean, removing mass at the edges. Those are separate books, and they are both moving — in opposite directions, both getting worse.

Across 2020 to 2024, discharge from Antarctica’s outlet glaciers ran nearly 100 gigatonnes a year higher than it had over the previous two decades. The glaciers accelerated. The calving increased. The warm ocean water reaching the undersides of the ice shelves did not go away.

The sheet gained mass anyway, because the snowfall on top happened to outrun the ice loss at the edges. The balance flipped positive while both sides of the ledger deteriorated.

A bank account can grow while you spend more than ever, provided your income grows faster. Nobody would call that account healthy without looking at both columns.

The recent gain is real. It is also small against what came before.

Taken across the whole satellite record from 2002 to 2025, Antarctica has shed roughly 135 gigatonnes of ice a year on average, contributing about 0.4 millimetres a year to global sea level. The cumulative contribution peaked in early 2020 at around six millimetres before easing back slightly.

Five good years do not repay two bad decades. Plotted on a graph, the recent gain is a small upward kick at the end of a long descent — visible, genuine, and nowhere near the size of the drop.

The honest summary is that the trend reversed. Not that the damage did.

The idea that Antarctica is gaining ice did not start in 2026. It started in 2015, with a NASA glaciologist named Jay Zwally.

Zwally and colleagues published a paper in the Journal of Glaciology arguing that snowfall gains in East Antarctica exceeded the losses in the west, meaning the continent as a whole was gaining. It was picked up instantly and enthusiastically by climate sceptics and parts of the conservative press, under headlines announcing that global warming had been disproved.

The paper was genuinely contested, and on technical grounds rather than political ones. It relied on satellite altimetry — measuring the surface height of the ice — which struggles to distinguish fresh snow from compacted snow, and to handle how far a radar pulse sinks into the snowpack. It also depended heavily on a correction for the rock beneath the ice, which slowly rebounds or sinks under the changing weight above it. That correction alone has introduced uncertainty of up to 130 gigatonnes a year into estimates of this kind.

When an international reconciliation exercise pooled the available methods, its figure for East Antarctica came out at less than half of Zwally’s: his estimate was about 2.3 times the reconciled value.

So the situation now is genuinely awkward for everybody. The measurement Zwally was arguing about has since moved in his direction, for reasons that have very little to do with his argument — and that, on inspection, support the case he was arguing against.

Meanwhile, the other ice — the floating kind — has done something dramatic, and it runs the other way.

Antarctic sea ice grew slightly from the start of the satellite record in 1979 until about 2015, which was itself a favourite talking point at the time. Then it fell off a cliff. February 2023 set the record low. Four consecutive years saw the summer minimum drop below two million square kilometres, a threshold never previously crossed.

This year brought a partial recovery. The 2026 minimum came in at 2.58 million square kilometres on 26 February — sixteenth lowest in the 48-year record, much closer to normal than any of the preceding four years, though still 260,000 square kilometres below the 1981–2010 average.

Sea ice swings hard from year to year, driven by winds and currents, so a single season proves little in either direction. But note the structure of the argument: the same people who cited rising sea ice before 2015 as evidence against warming have not generally cited its collapse after 2015 as evidence for it. A measure that only counts when it points one way is not a measure. It is a search.

If you want the part that matters for the next century, it is not the annual mass balance. It is the ice shelves.

The shelves are the floating margins where the ice sheet meets the sea. They do not raise sea level themselves — they are already afloat — but they act as buttresses, holding back the grounded ice behind them. Thin them, break them, and the glaciers they were restraining speed up.

That is the mechanism that turns a slow problem into a fast one, and warm ocean water is currently working on it from below. The four East Antarctic glacier basins that saw the sharpest snowfall gains recently — Totten, Denman, Moscow University and Vincennes Bay — sit above enough ice to raise global sea level by more than seven metres if they were ever to go entirely.

Nobody is forecasting that soon. The point is that the number people argue about on social media is not the number that decides the outcome.

Recent mass gain and its causes: Marlen Kolbe, Eveline C. van der Linden and colleagues, “Atmospheric rivers and winter sea ice drive recent reversal in Antarctic ice mass loss,” Communications Earth & Environment vol. 7, art. 255 (2026); European Space Agency summary, “Why is Antarctica’s mass increasing?”, including the figures on the 2020–2024 gain of roughly 68 Gt/yr and the ~100 Gt/yr rise in discharge.

The 2021–2023 spike: Wei Wang and Yunzhong Shen et al., Tongji University, Science China Earth Sciences (2025), reporting −73.79±56.27 Gt/yr for 2002–2010, −142.06±56.12 Gt/yr for 2011–2020, and +107.79±74.90 Gt/yr for 2021–2023; coverage and independent comment in Live Science and Space.com, including Tom Slater, Northumbria University.

Long-run mass balance: NASA JPL GRACE and GRACE-FO mission data, “Antarctic Ice Mass Loss 2002–2025” (NASA Scientific Visualization Studio), giving an average of approximately −135 Gt/yr and 0.4 mm/yr of sea level contribution.

Sea ice: National Snow and Ice Data Center, “Antarctic sea ice extent arrives at a near-average minimum” (March 2026) and “Antarctic sea ice minimum hits a near-record low, again” (March 2025); UK Met Office, “Briefing on Arctic and Antarctic sea ice — June 2026”; commentary from Ted Scambos (CIRES) and Walt Meier (NASA NSIDC DAAC).

The 2015 controversy: H. Jay Zwally et al., “Mass gains of the Antarctic ice sheet exceed losses,” Journal of Glaciology (2015), and the subsequent Comment by T. Scambos and C. Shuman with Zwally’s reply; Washington Post and Big Think coverage of the study’s reception; the IMBIE reconciliation figures for East Antarctica; discussion of glacial isostatic adjustment uncertainty of up to 130 Gt/yr.

Context on ice shelves, buttressing and the four Wilkes Land–Queen Mary Land basins: Antarctic and Southern Ocean Coalition, “Ice Gain in Antarctica? Cool Story, Still a Hot Planet”; AntarcticGlaciers.org on Antarctic surface mass balance.

ONE MORE THING

You can read our MANY newsletters for free at our website:

Read our archive for free at OneWildWorld.com →