Antarctica’s Ice Loss Slowed- Scientists Traced the Cause to the Tropics

Antarctica has been losing enormous amounts of ice for decades, adding water to the oceans and making the continent one of the biggest uncertainties in future sea-level projections. Something unusual happened between 2021 and 2023, i.e. the Antarctic Ice Sheet briefly gained enough mass to slow its long-running decline.

A new study titled 'Multiyear tropical warm pool warming drives slowdown in Antarctic mass loss,' published in the journal Nature, argues that this surprising change was largely driven by climate variability thousands of kilometres away. Persistent warming in the tropical warm pool altered atmospheric circulation, steering moisture towards East Antarctica and producing unusually heavy snowfall.

The finding matters because the temporary ice gain could easily be mistaken for evidence that a warming climate has begun producing enough extra Antarctic snowfall to counter ice loss. The researchers reach a different conclusion, i.e. the recent gain was probably a temporary climate-driven event rather than the beginning of a lasting recovery.

East Antarctica gained an extraordinary amount of ice

West Antarctica has experienced sustained losses, largely because relatively warm ocean water reaches ice shelves and contributes to glacier retreat and faster ice discharge. East Antarctica, which holds nearly 80% of Earth's land ice, is much more sensitive to changes in snowfall, meaning a few unusually wet years can noticeably change its mass balance.

Satellite measurements from the GRACE and GRACE-FO missions show that Antarctica lost mass at an average rate of about 140.5 gigatonnes per year between 2003 and 2024. That decline was dramatically interrupted from July 2021 to April 2023, when the ice sheet gained roughly 695 gigatonnes in just 22 months, the largest 22-month increase observed during the roughly two-decade satellite record.

Most of that gain came from Queen Mary Land and Wilkes Land in East Antarctica, referred to by the researchers as the QW sector. This region accumulated about 470 gigatonnes, representing roughly 68% of the continent-wide increase. Measurements of precipitation and surface mass balance closely matched the satellite-observed ice changes, giving the researchers strong evidence that exceptional snowfall was responsible.

The central question was why so much snow suddenly fell there.

Scientists already expect global warming to increase atmospheric moisture and shift Southern Hemisphere storm tracks towards Antarctica, which could raise snowfall over parts of the continent. Yet the researchers found that the 2021–2023 event did not look like this gradual human-driven trend. Previous work had also pointed to atmospheric rivers, the unusually prolonged La Niña, circulation changes and reduced sea ice as possible contributors.

Climate-model simulations representing the anthropogenic response produced only a small fraction of the observed precipitation increase in the QW region. The atmosphere during the event also developed a distinctive high-pressure pattern over East Antarctica that differed markedly from the long-term circulation response expected from increasing greenhouse gases.

Warm tropical waters changed the atmosphere over Antarctica

The researchers traced the unusual circulation back towards the tropical warm pool, a vast region of very warm ocean water around the Maritime Continent and western tropical Pacific. Sea-surface temperatures there remained unusually warm during 2021–2023. That heating affected the upper atmosphere and generated a chain of large-scale atmospheric disturbances known as Rossby waves. The wave train travelled towards the high southern latitudes and helped establish a distinctive pressure pattern: low pressure south of Australia and high pressure over coastal East Antarctica.

That high-pressure anomaly effectively changed the routes available to storms and moisture. Atmospheric rivers — long, narrow streams capable of carrying huge quantities of water vapour — became more favourable for transporting moisture towards Queen Mary Land and Wilkes Land. The researchers did more than identify a statistical relationship and tested the proposed mechanism with two atmospheric climate models, ECHAM5 and CAM5. When they artificially warmed the tropical warm pool, the models generated a pressure pattern and increased East Antarctic precipitation resembling what had actually happened.

Separate experiments tested other tropical conditions seen during the same period. La Niña-like cooling, a negative Indian Ocean Dipole pattern and warming around the South Pacific Convergence Zone could not independently reproduce the East Antarctic pressure dipole and precipitation increase. They may have strengthened or modified the tropical warm-pool signal, but the experiments identify the localised warm-pool anomaly as the more direct trigger.

Changes in the westerly winds altered short-lived weather systems, whose feedbacks reinforced the high-pressure anomaly. The result was a circulation configuration that could remain in place long enough to produce several seasons of exceptional snowfall.

Using a model that tagged moisture from 54 regions worldwide, the researchers found that much of East Antarctica's normal precipitation comes from the mid-latitude Indian Ocean. During 2021–2023, precipitation in the QW sector increased by about 85 millimetres per year, with three important source regions accounting for around 45% of that increase.

Crucially, those source regions were not simply evaporating much more water than usual. The atmospheric circulation had become better at delivering existing ocean moisture to Antarctica. In other words, the event was primarily about changing transport routes rather than a sudden increase in the amount of nearby moisture available for evaporation.

The ice gain probably will not last

The 2021–2023 episode was unusual, but the underlying climate pattern is not unprecedented. Using observations extending back to 1950, the researchers estimated about 9.2 prolonged tropical warm-pool warming events per century. A 1,800-year pre-industrial climate simulation produced about 10.6 events per century, while historical simulations gave a similar rate. Taken together, the evidence suggests that strong multiyear warm-pool episodes capable of influencing East Antarctica occur roughly once a decade.

A weaker version of the teleconnection appeared around 2000–2002, when QW precipitation also increased. The decade from 2011 to 2020 showed almost the opposite behaviour: the tropical warm pool was cooler, atmospheric circulation reversed, moisture transport weakened, and the region experienced a prolonged precipitation deficit.

One unresolved question is what causes tropical warm-pool warming to persist for several years. La Niña, the Indian Ocean Dipole, the Interdecadal Pacific Oscillation, and interactions with the tropical Atlantic may all play roles, although the researchers cannot yet identify a single mechanism responsible for initiating and sustaining these episodes.

There is also uncertainty about how this tropical-Antarctic connection will behave as the planet becomes warmer. A warmer atmosphere can carry more moisture, atmospheric rivers may intensify, and storm tracks may change, potentially modifying or amplifying the teleconnection. Climate-model experiments suggest that Antarctica responds particularly strongly to localised temperature contrasts in the tropical warm pool, rather than uniform warming across the tropics.

West Antarctica continues to lose ice through ocean-driven melting and dynamic glacier thinning, while vulnerable parts of East Antarctica, including the Totten Glacier region, remain exposed to oceanic change. The recent snowfall simply compensated for enough of that loss to create a temporary continent-wide slowdown.

The study reveals how a relatively concentrated patch of tropical ocean warming can set off an atmospheric chain reaction spanning thousands of kilometres, reorganise storm paths around Antarctica and temporarily alter the mass of the world's largest ice sheet. Understanding that connection will be important for separating short-lived natural fluctuations from long-term climate trends and for making more reliable projections of future sea-level rise.