ETH Zurich and VUB Launch Global Glacier Extinction Explorer Mapping Every Glacier

Researchers at ETH Zurich and Vrije Universiteit Brussel launched the Global Glacier Extinction Explorer on September 11, 2026, an interactive website that projects whether and approximately when every individual glacier worldwide will disappear under +1.5C, +2.0C, +2.7C and +4.0C of global warming. The platform, available at www.glacierextinction.com, lets anyone search for glaciers around the world and see how their futures could change under different levels of warming. Rather than reporting only the total mass or area of ice that the planet stands to lose, the tool zooms in on single glaciers, from the European Alps to the Himalayas, and assigns each one an approximate disappearance window.

The explorer builds on a study published in Nature Climate Change, led by researchers from ETH Zurich and Vrije Universiteit Brussel (VUB) together with an international team. According to the VUB press release, the study provided the first global assessment of the disappearance of individual glaciers. The website was developed by lead author Lander Van Tricht, who is affiliated with both ETH Zurich and VUB, together with Kristof Van Tricht of VITO Remote Sensing. The underlying paper is titled Peak glacier extinction in the mid-21st century and appeared in Nature Climate Change volume 16, pages 143 to 147.

The researchers argue that conventional glacier metrics are not enough. The explorer's methodology notes that glacier changes are traditionally quantified in terms of total regional mass and area loss, but that such numbers do not fully capture the disappearance of individual glaciers. "Global numbers such as glacier mass or area loss can feel very abstract," Van Tricht said. "With this website, people can zoom into their own valley or mountain region and see what different levels of global warming could mean for the glaciers they know."

The launch lands in a difficult year for mountain ice. Preliminary end-of-summer observations suggest that 2026 could be one of the worst years for glaciers in the European Alps since systematic measurements began, driven by low winter snow accumulation and several summer heatwaves. Bella Tola Glacier near Saint-Luc, Switzerland, was officially declared extinct in August 2026, closely matching the study's projection that it would disappear between 2026 and 2029. Urirotstock Glacier was also declared extinct in summer 2026.

Key Facts

To decide whether a glacier has effectively vanished, the explorer applies two complementary criteria. A glacier is considered extinct when its area drops below 0.01 square kilometers, consistent with commonly used inventory thresholds, or when its remaining volume falls below 1% of its initial value, indicating that only a negligible ice body persists. Volume and area trajectories for each glacier are derived by combining three global glacier models, GloGEM, OGGM and PyGEM, with multiple General Circulation Models and emission scenarios corresponding to projected temperature increases of +1.5C, +2.0C, +2.7C and +4.0C. The trajectories are merged to determine median extinction years as well as Q25 and Q75 ranges, the 25% and 75% probability intervals. Extinction is defined relative to initial glacier outlines from the Randolph Glacier Inventory version 6.0, and glacier fragmentation during retreat is not accounted for.

Phys.org reported on September 12, 2026 that at peak extinction around 2,000 glaciers per year are projected to disappear globally under +1.5C of warming, a number that roughly doubles under +4C. India Today reported on September 12, 2026 that glacier disappearance could peak around the middle of this century, with about 2,000 glaciers vanishing each year around 2041 under 1.5C of warming and a peak of about 4,000 glaciers a year in the mid-2050s under 4C. Muser Press reported on September 14, 2026 that the researchers project glacier extinction will peak between 2041 and 2055.

Vrije Universiteit Brussel announced on September 11, 2026 that by 2100 nearly half of today's glaciers could remain at +1.5C of warming, compared with about 20% at +2.7C and fewer than 10% at +4C. The squeeze is sharpest in the European Alps, where only around 110 glaciers may remain under +2.7C of warming and as few as 20 under +4C. Van Tricht framed the difference in stark terms: "Every fraction of a degree matters. Limiting warming to 1.5C could preserve more than twice as many glaciers as a 2.7C trajectory by 2100."

The dates shown on the website are projections, not fixed deadlines, and depend on a glacier's size, elevation, shape and local climate, India Today reported on September 12, 2026. The same article noted the August 26 collapse of glacier ice and rock near the Nepal-China border, which triggered floods that killed more than 1,300 people across Nepal and Tibet and which scientists linked to a warming and increasingly unstable Himalayan environment. Co-author Harry Zekollari of VUB stressed that the timing reflects decades of change rather than one hot season. "The disappearance of a glacier is not the result of a single warm summer, but the culmination of decades of cumulative mass loss," he said. "Extreme melt years such as 2022 or 2026 can, however, accelerate this final stage."

Analysis

What this really means is that the public conversation about glaciers is shifting from an abstract global balance sheet to a searchable, place-by-place forecast. Until now, most climate communication about ice has been built on aggregate figures, such as the total mass lost per year or the total area shrinking across a mountain range. Those numbers are scientifically essential, but they are hard to feel. The new explorer reverses the direction of the question: instead of asking how much ice the world will lose, it asks whether a particular glacier, in a particular valley, still exists at a given level of warming.

The bigger picture here is that the tool turns a long-running scientific debate about model uncertainty into something a visitor can inspect directly. Every glacier carries a median extinction year plus Q25 and Q75 ranges, so the uncertainty is visible rather than hidden. The projections rest on three independent global glacier models, GloGEM, OGGM and PyGEM, combined with multiple General Circulation Models, and they are tied to initial outlines from the Randolph Glacier Inventory version 6.0. That is a defensible design, but it also has limits that the methodology page states plainly: fragmentation during retreat is not accounted for, so a shrinking glacier that splits into several small ice bodies is treated as a single entity.

The Alps provide the clearest near-term test of the projections. Bella Tola Glacier was expected to disappear between 2026 and 2029, and it was declared extinct in August 2026. Urirotstock Glacier followed the same summer. Van Tricht used precisely that overlap as his central argument: "The fact that some of the glaciers projected to disappear in our study are already vanishing today shows that glacier extinction is not a distant, end-of-century problem. It is happening now." When a forecast and an observed extinction line up within a three-year window, the credibility of the wider dataset rises.

Still, the headline numbers are brutal enough on their own. A peak of about 2,000 glacier extinctions per year under +1.5C, doubling to roughly 4,000 a year under +4C, describes an extinction pulse concentrated between 2041 and 2055 rather than a slow, uniform decline. Nearly half of today's glaciers surviving to 2100 at +1.5C, against fewer than 10% at +4C, is the difference between losing most of the world's glaciers and losing almost all of them.

Why It Matters

For communities that live near ice, the explorer converts a global statistic into a local one. The Nepal flood of August 26, which killed more than 1,300 people across Nepal and Tibet, showed how quickly a warming and increasingly unstable Himalayan environment can turn into a direct threat to life. A tool that lets people look up the glaciers above their own valley gives them a way to connect warming trajectories to the hazards and water supplies they actually depend on.

The choice of warming levels is also a policy statement. The explorer presents +1.5C, +2.0C, +2.7C and +4.0C side by side, which makes the gap between current pledges and safer pathways visible. Van Tricht's point that limiting warming to 1.5C could preserve more than twice as many glaciers as a 2.7C trajectory by 2100 is the kind of comparison that a single map can now deliver for any region a user chooses.

Zekollari's caution matters as well. A glacier does not die in a single hot summer; it dies after decades of cumulative mass loss, with extreme years such as 2022 or 2026 accelerating the end. That framing pushes attention toward sustained emissions reductions rather than isolated bad seasons.

Next Up

The researchers have archived the regional and glacier-specific results publicly on Zenodo, and the web application was built with open-source tools including MapLibre GL JS, PMTiles, GeoPandas, NumPy and GDAL. That combination means the projections can be reused, checked and extended by other scientists as new observations arrive, and the explorer is designed to keep presenting median extinction years alongside their Q25 and Q75 ranges.

The immediate test will be the next round of end-of-summer measurements. If 2026 confirms as one of the worst years for Alpine glaciers since systematic measurements began, the gap between projections and reality will keep narrowing, and the roughly 110 glaciers the study expects to survive in the Alps under +2.7C, or the 20 under +4C, will become easier to count.