Glacial lakes speed up ice loss, but climate and terrain still dominate
Glaciers that end in lakes thinned considerably faster than those ending on land or in the sea between 2000 and 2020, according to a worldwide mapping study in Nature Climate Change. The international team, which includes Tobias Bolch of the Institute of Geodesy at Graz University of Technology, concludes that climate and glacier geometry are the main controls on ice loss, with lakes adding a smaller but measurable boost.
Mapping every glacial lake
The researchers analysed satellite images from 2000 and 2020 to chart glacial lakes across the planet. Artificial intelligence was used to detect the lakes, and the results were then checked manually. The team separated lakes in direct contact with glacier ice from those that have since become detached from it, a distinction that matters because only touching lakes can act on the ice directly.
Water in contact with a glacier front can melt the ice from below, and blocks of ice can break away and fall into the lake. Until now, how much these processes matter compared with other drivers had not been settled at a global scale.
Faster loss at lake-terminating glaciers
Glaciers ending in lakes lost on average 66 centimetres of water equivalent per year over the two decades. Glaciers ending on dry land lost 40 centimetres per year, and those ending in the sea lost 36 centimetres. Water equivalent expresses ice loss as the thickness of liquid water it would produce.
The number of lakes touching glaciers rose by more than 3,500, or about 35%, over 20 years. Their combined area grew by roughly 600 square kilometres and their estimated water volume by about 66 cubic kilometres. The authors report that the growth was especially marked in Alaska and the southern Andes.
How much do lakes really matter?
To weigh the different influences, the team used machine learning models to relate glacier mass balance, the net gain or loss of ice, to a range of factors. Within the model, about 46% of the explained influence came from topography, such as glacier altitude and slope. Climatic variables, including temperature and precipitation, accounted for around 43%. Glacial lakes contributed about 11%.
These shares describe the explanatory weight of each factor in the model, not a direct count of ice lost to lakes. They also indicate that no single factor acts alone. As Bolch puts it, lakes "significantly exacerbate" mass loss, while climate and topography remain the leading influences.
Why it matters for research and planning
Many projections of glacier change treat lake interaction crudely or ignore it. A global inventory showing where lakes touch ice, and how much faster those glaciers thin, offers a basis for including the effect in models of future glacier retreat. That is relevant to mountain regions and their forelands, where meltwater is an important water source.
The findings are also relevant to hazard assessment, since growing glacial lakes can pose risks if they burst their natural dams. According to Bolch, the results should help in estimating such risks. The study draws on satellite data from two time points and a statistical model, so further work will be needed to separate the physical processes at individual glaciers and to test how well the patterns hold in regional case studies.
The paper is titled "Climate and glacier geometry dominate global glacier–lake interactions" and is led by Shaochun Zhao and colleagues (DOI: 10.1038/s41558-026-02808-5).
Source: Global analysis reveals how lakes amplify glacier ice loss (Phys.org, 21 November 2026).
Image: AI-generated illustration.
Working on this topic? Find a journal for your research or join the discussion on Scholars Post.


