Science1 publisher3 min readPublished
A satellite record reaching back to 1972 assigns five-sixths of ice-sheet loss to warm ocean water
An ESA-led team pooled 45 surveys from 27 satellites to weigh Greenland and Antarctica back to the 1970s. It finds 11.3 trillion metric tons of ice gone since 1979, five-sixths of it to water working from below.
The Scientist · Science desk
What happened
- A major scientific report finds that glaciers in Greenland and Antarctica have lost about 11.3 trillion metric tons of ice since 1979, enough water to lay 1.5 meters of ice across the continental United States.
- The study, published in Scientific Data, attributes five-sixths of that melt to warmer water eating at the ice sheets from below and the sides as glaciers flow into the oceans. Hotter air is not the main driver.
- The lost ice raised global sea levels by about 3.1 centimeters since 1979, on top of other climate factors that are also pushing ocean heights up.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Shepherd said the sheets are responding decades behind the atmosphere, so the loss continues for decades even in a world that stops warming.
- exposure The rise has already happened and the people exposed to it are already exposed: at Otosaka's rate of 2 to 3 million people per centimeter, the 3.1 centimeters logged since 1979 corresponds to something like 6 to 9 million people newly flooded at least once a year.
- capability Pushing the baseline back before the acceleration gives modellers a stable reference decade, so future acceleration claims can be tested against measured quiet years. A record starting in the 1990s could not supply that.
- contradiction Two outside scientists interpret the same dominance of dynamic loss differently: Holland sees the signature of an instability collapse, while Scambos places the danger many decades out.
Five-sixths of 11.3 trillion metric tons is about 9.4 trillion tons of ice, and it left the sheets by flowing out to sea [21]. Warmer air did not melt it in place. Ines Otosaka of Northumbria University, the lead author, said the biggest changes the team saw are in what they call dynamic processes, which are more abrupt and active than the slow steady melting from the air, and prone to trigger runaway tipping points [16].
The value of extending the record backwards is that the acceleration now has a quiet period to be measured against. The collaboration had previously only looked at ice sheet melt since the 1990s, using European satellite data; adding NASA satellites pushed Antarctica back to 1979 and Greenland to 1972 [8]. Through the 1970s, 1980s and into the 1990s the sheets were relatively stable, Otosaka said, and the melt really kicked in during the 2010s [10]. The team assembled the record from 45 independent surveys and 27 different satellites [18]. David Holland, an ice scientist at New York University who was not part of the study, said "The use of relatively independent methods gives me greater confidence in the result," and added, "The upward trend is real." [19]
The sea-level number is small. The exposure number attached to it runs into the millions. The melt raised global sea levels about 3.1 centimeters since 1979, on top of other climate factors raising ocean heights [5]. Otosaka put the human cost at 2 to 3 million more people flooded at least once a year for every centimeter of rise [6]. Apply her rate straight across 3.1 centimeters and you get roughly 6 to 9 million people [22]. That multiplication assumes the relationship holds evenly over the whole range, which a single quoted rate does not establish, and the ice sheets are only one of the contributors to those centimeters [5].
Andrew Shepherd, also of Northumbria and a co-author, said the changes are "tracking decades behind global warming, which means we can expect them to continue for decades more" [11]. The record also contains an apparent pause from 2020 to 2023. The authors put it down to short-term weather, with the underlying trend intact [12]. Much of that Antarctic steadiness came from record snowfall in East Antarctica, offsetting rapid melting in the less stable West Antarctic Ice Sheet [14]. Co-author Eric Rignot of the University of California, Irvine called the three-year hiatus "just a bump in the record" [15]. Otosaka said the study stopped with 2023 data, and that measurements she has looked at since show the accelerating trend picked back up [13].
The record ends in 2023, so what it delivers is an account of which process did the damage [13]. The process it names is also the hardest one to predict. Holland said most of the ice loss is dynamical "and that is exactly what you would expect in an ice sheet instability collapse scenario" [17]. Ted Scambos of the University of Colorado, also not part of the study, called Antarctic mass loss a "clear and present danger, but many decades in the future" [20]. Otosaka said Greenland's Jakobshavn glacier is now retreating up to 50 meters a day [7].
What to watch
- Whether the post-2023 measurements Otosaka says she has examined show the acceleration resuming once they enter the published record.
- Whether East Antarctic snowfall keeps offsetting West Antarctic losses, since the Antarctic net total moves quickly if it fades.
- Whether the dominance of dynamic loss in this record changes the uncertainty ranges used in the next round of sea-level assessments.