A first-of-its-kind study looking at surface meltwater lakes around the East Antarctic Ice Sheet across a seven-year period has found that the area and volume of these lakes is highly variable year-to-year, and offers new insights into the potential impact of recent climatic change on the ‘Frozen Continent’.

The research involving researchers from Lancaster University and led by Durham University (UK), used over 2000 satellite images from around the edge of the East Antarctic Ice Sheet to determine the size and volume of lakes on the ice surface, also known as supraglacial lakes, across seven consecutive years between 2014 and 2020.

The study, which involved Newcastle University and the Georgia Institute of Technology, showed that lake volume varied year-to-year by as much as 200% on individual ice shelves (floating extensions of the main Antarctic ice sheet), and by around 72% overall.

Lakes were also found to be deeper and larger in warmer melt seasons and formed on some potentially vulnerable ice shelves.

This research, published today in Nature Communications, is the first time that meltwater lakes have been studied over consecutive melt seasons across the whole ice sheet, enabling the controls on their development to be explored. The study therefore provides vital insight into why and where lakes grow, and will help experts understand which ice shelves may be most at risk of breaking up as a consequence of surface melting.

Lead researcher, PhD student Jennifer Arthur, Department of Geography, Durham University, said; “We knew that supraglacial lakes were more extensive than previously thought around the East Antarctic Ice Sheet, but until now only had snapshots of these in some years.

“Our study reveals these lakes change in scale far more than we originally suspected. We were surprised at how much lakes can change year-to-year between ice shelves.

“We explored the potential reasons for this and found that warmer summer air temperatures in Antarctica correlated with more extensive lakes.”

Lancaster University’s Dr Amber Leeson, who studies supraglacial lakes on both the Antarctic and Greenland ice and is co-author of the study, said: “Supraglacial lakes form from accumulated snowmelt and have been associated with ice shelf disintegration in West Antarctica. It is important that we closely monitor their evolution here and in other parts of the ice sheet because when ice shelves collapse, we see an associated increase in global sea level.

“This study shows that ice shelves in East Antarctica, which is colder and more snowy than the West, also host persistent supraglacial lakes and that their abundance is very sensitive to variations in climate. As Antarctic temperatures increase with global warming, we can probably expect more melting on East Antarctic ice shelves, a subsequent growth in their supraglacial lake populations and potentially even more collapse events.”

The East Antarctic Ice Sheet is the world’s largest ice mass and holds enough ice to raise global sea levels by around 52 meters.

The loss of ice shelves fringing an ice sheet allows ice further inland to flow faster into the ocean, contributing to global sea-level rise.

Until now, observations of supraglacial lakes on the East Antarctica Ice Sheet were relatively scarce and the year-to-year variability was largely unknown, making it difficult to assess whether some ice shelves are close to meltwater-driven break up under climate change.

This study will help experts understand supraglacial lake formation, climatic impacts on this and predict which ice shelves may be most at risk of collapse.

Understanding the climatic conditions controlling meltwater lake variability will also improve the accuracy of regional climate models used to replicate observations and predict future ice sheet change in Antarctica.

The study used images from the Landsat 8 satellite. Work on this study was funded through a UKRI Natural Environment Research Council doctoral studentship and individual author grants from the Natural Environment Research Council.

DOI: http://dx.doi.org/10.1038/s41467-022-29385-3