2.1. Challenge in defining a vanished glacier

While it is important to account for vanished glaciers in glacier-change assessments (Linsbauer and others, Reference Linsbauer, Huss, Hodel, Bauder and Barandun2025), there appears to be no clear definition of when a glacier has vanished or become extinct; this problem is also revealed in the studies in this Collection. Some studies have declared a glacier no longer a glacier when it becomes too thin to flow or fails to exhibit signs of flow (e.g. references within Carlson and others, Reference Carlson, Nicolas, Thayne, Pappas, Molnar and Rood2025; Guðmundsson and others, Reference Guðmundsson, Magnússon, Belart, Hannesdóttir and Aðalgeirsdóttir2025; Pope, Reference Pope2025), whereas others use an area threshold or await the absence of ice before declaring a glacier vanished (e.g. Huss and Fischer, Reference Huss and Fischer2016; Andreassen and others, Reference Andreassen, Robson, Smith, Weber, Carrivick and Kjøllmoen2025b; Ugalde and others, Reference Ugalde2026). A common definition cited in this Collection is from the UNESCO/IACS glossary of glacier mass balance (Cogley and others, Reference Cogley2011), where a glacier is defined as a ‘perennial mass of ice, and possibly firn and snow, originating on the land surface by the recrystallization of snow or other forms of solid precipitation and showing evidence of past or present flow’. As pointed out by Pope (Reference Pope2025), this definition introduces a hysteresis whereby ice must start flowing to become a glacier, but it does not cease to be a glacier once it stops flowing.

2.2. Mapping vanishing and vanished glaciers

Mapping glacier extents can be challenging as glaciers approach extinction. For example, glacier remnants may be covered by debris (Fischer and others, Reference Fischer, Schwaizer, Seiser, Helfricht and Stocker-Waldhuber2021; Conzelmann and others, Reference Conzelmann2026), making them difficult to distinguish from ice-free terrain or formerly glacierized areas may entrap snow, making them appear glacierized (Kristensen and Andreassen, Reference Kristensen and Andreassen2026). Paul (Reference Paul2026) suggests encoding criteria for glacier disappearance in inventories, ranging from there being no ice left to the ice area being below a size threshold or not suitable for measurement. Implementing this suggestion would require an additional attribute field and manual work by the investigator, but would permit more robust calculation of area change rates and a more consistent accounting of vanished glaciers across studies.

The GGCL categories describing glacier status are not precisely defined, but ‘critically endangered’ describes a glacier that can be reliably estimated to disappear within the next 20 years or by 2050 (Howe and Boyer, Reference Howe and Boyer2025), while ‘almost disappeared’ describes a glacier whose disappearance is imminent (Dominic Boyer, personal communication, 3 December 2025). The designation of ‘critically endangered’ can be applied to many mountain glaciers worldwide due to recent significant mass loss and the committed loss due to temperature increase of the recent past (The GlaMBIE Team, 2025; Zekollari and others, Reference Zekollari2025). Carlson and others (Reference Carlson, Nicolas, Thayne, Pappas, Molnar and Rood2025) categorize 34 glaciers in Oregon, USA, according to their interpretation of the GCCL categories: they define a ‘disappeared’ glacier to be between a stagnant ice mass and one that is completely absent, an ‘almost disappeared’ glacier as one expected to reach disappeared status ‘any year now’, and a ‘critically endangered’ glacier as one expected to reach disappeared status by 2050. Classification of glacier state and determination of glacier disappearance can be useful, but some subjectivity is unavoidable. As highlighted by Pope (Reference Pope2025), the answer to the question of when a glacier is no longer a glacier depends on who is asking and why. This complication underlines the importance of local knowledge, context and expertise in determining what criteria to apply when identifying glaciers or declaring them vanished (Paul, Reference Paul2026).

In some regions, detailed field observations allow glaciers to be tracked every year until they are declared vanished (e.g. Huss and others, Reference Huss, Fischer, Linsbauer and Bauder2025), whereas other locations are rarely or never visited and therefore rely on remote sensing techniques to determine glacier status (e.g. Winsvold and others, Reference Winsvold, Andreassen and Kjøllmoen2026). Repeated regional and global inventories are powerful tools for tracking and mapping vanishing glaciers. Size thresholds can be used to determine whether a glacier is to be included or reported as extinct. The selection of a minimum size threshold depends on both the region and the spatial resolution of the data used to map the glaciers. When mapping with very high-resolution imagery (<1 m) with minimal seasonal snow cover, glaciers smaller than 0.01 km² can be clearly identified (e.g. Linsbauer and others, Reference Linsbauer, Huss, Hodel, Bauder and Barandun2025) and scoring systems can be applied to classify ‘certain’, ‘probable’ or ‘possible’ glaciers (Leigh and others, Reference Leigh2019). The global glacier inventory (Randolph Glacier Inventory) uses a minimum size threshold of 0.01 km² to define a glacier (Maussion and others, Reference Maussion2023), whereas GLIMS (Global Land Ice Measurements from Space), which tracks the extent of glaciers worldwide over time, allows the observer to decide what to include as a glacier and when a glacier can be registered as extinct (Raup and others, Reference Raup, Andreassen, Boyer, Howe, Pelto and Rabatel2025). GLIMS has been documenting ‘extinct’ glaciers since August 2023, with a published tally of 181 (Raup and others, Reference Raup, Andreassen, Boyer, Howe, Pelto and Rabatel2025) that has increased to 194 (Fig. 4; Andrew Barrett, GLIMS, pers. communication, 5 March 2026). As revealed in several papers in this Collection, more glaciers already qualify, or soon will, to be included in the GLIMS list of extinct glaciers (e.g. Guðmundsson and others, Reference Guðmundsson, Magnússon, Belart, Hannesdóttir and Aðalgeirsdóttir2025; Linsbauer and others, Reference Linsbauer, Huss, Hodel, Bauder and Barandun2025; López-Moreno and others, Reference López-Moreno, Revuelto, Izagirre, Alonso-González, Vidaller and Bonsoms2025).

Figure 4.

Location map showing the location of study glaciers in the Vanishing Glaciers Collection, divided into studies of one or a few individual glaciers (1–10 glaciers) and larger groups of investigated glaciers. The GLIMS list of extinct glaciers reported in Raup and others (Reference Raup, Andreassen, Boyer, Howe, Pelto and Rabatel2025) is shown separately. Global Terrestrial Network for Glaciers (GTN-G) glacier regions 1–20 (GTN-G, 2023) and glaciers outside the ice sheets according to the Randolph Glacier Inventory (RGI) are also shown (Maussion and others, Reference Maussion2023). The inset shows GTN-G regions 2, 8 and 11. See also Figure 1 in Pilø (Reference Pilø2026) for his map of archaeological locations. World base map from ©ESRI. Names on the map refer to glaciers, countries or regions studied and may be shortened. Abbreviations: OK, Okjökull; Hof, Hofsjökull eystri; Switz., Switzerland, Vor, Vorarlberg, Geltal, Geltalferner.

Figure 4 Long description

The map displays the location of study glaciers in the Vanishing Glaciers Collection, categorized by the number of glaciers and extinction status. Symbols indicate 1-10 glaciers (red dots), more than 10 glaciers (red stars) and GLIMS extinct glaciers (black triangles). GTN-G regions are outlined and numbered from 1 to 20. RGI glaciers are marked in blue and world countries are outlined. Insets provide detailed views of Western Canada and USA (region 2) and Central Europe (region 11). Specific labeled glaciers include Okjökull (OK), Hofsjökull eystri (Hof) and Zugspitzeplatt. The map highlights regions such as Scandinavia, Western Canada, USA and Central Europe, showing the distribution of glaciers and their status across these areas.

2.3. A need for defining vanishing glaciers terms

When registering a glacier to have vanished, particular care must be taken. Determining the absence of flow or evidence of past flow is not straightforward (e.g. Paul, Reference Paul2026), and since the glacier definition by Cogley and others (2011) includes past flow, it cannot be used to declare a glacier vanished or extinct when it no longer flows. Cogley and others (2011) also have the subcategory ‘glacieret’ defined as ‘a very small glacier, typically less than 0.25 km² in extent, with no marked flow pattern visible at the surface’. As mentioned by Pope (Reference Pope2025), the term glacieret has also been used to describe any small ice or possibly snow mass of indefinite shape (UNESCO, 1970). However, other terms such as ‘ice patch’ (e.g. Carlson and others, Reference Carlson, Nicolas, Thayne, Pappas, Molnar and Rood2025; Pilø, Reference Pilø2026), ‘very small glacier’ (e.g. Huss and others, Reference Huss, Fischer, Linsbauer and Bauder2025) or ‘very small glacieret’ (e.g. Ugalde and others, Reference Ugalde2026) are used in this Collection but have not been defined in Cogley and others (2011). We see a need for the community to define terms for vanishing glaciers, such as ‘ice patch’, ‘dead ice’ and ‘very small glacier’ now commonly used in the community and in many of the papers in this Collection.

3.1. Geographical distribution

Glaciers are vanishing in all parts of the world, and the studies presented in this Collection cover 9 of 20 glacier regions (GTN-G, 2023, Fig. 4). There is a regional focus on Europe (regions 6–8 and 11) and the Americas (regions 2, 16, 17), but Papua in Indonesia (16) and New Zealand (18) are also represented. Regions with high glacier concentrations, such as all Asian mountain ranges, such as the Himalaya, the Karakoram, the Tien Shan or the Caucasus (10, 12–15) are lacking (Fig. 4).

3.2. Vanishing glaciers in Europe

In Iceland, ∼70 small glaciers (0.1–3 km²) were reported to have vanished in the 2025 Iceland Glacier Newsletter, the most well-known being Okjökull (Fig. 2b), which was commemorated in August 2019 with a ceremonial event (Howe and Boyer, Reference Howe and Boyer2025). Five years later, Hofsjökull eystri was entered in the GGCL as a critically endangered glacier. The evolution and fate of these two glaciers from their Little Ice Age (LIA) maximum until today is traced by Guðmundsson and others (Reference Guðmundsson, Magnússon, Belart, Hannesdóttir and Aðalgeirsdóttir2025). In Switzerland, Linsbauer and others (Reference Linsbauer, Huss, Hodel, Bauder and Barandun2025) report that 40%, or over 1000, of all Swiss glaciers inventoried in 1973 had disappeared by 2016, representing 13% of the total glacier area in Switzerland. Measurements indicated a loss of 10% of the overall Swiss ice volume just within the two extreme years of 2022 and 2023 (GLAMOS, 2024). This rapid loss led to the fragmentation of many glaciers and the complete demise of several long-term monitoring sites. Five glaciers with 10–40 years of continuous mass-balance monitoring vanished in the last few years, including Pizolgletscher (Huss and others, Reference Huss, Fischer, Linsbauer and Bauder2025), which was commemorated in a ceremony attended by ∼250 hikers, scientists and journalists in September 2019. One of the two glaciers on Zugspitzplatt in Germany has already completely vanished, and the other is projected to vanish within a decade (Hagg and Mayer, Reference Hagg and Mayer2026). The glacier area in Vorarlberg, Austria, has decreased by 50% since the early 2000s. Five of 30 glaciers in Vorarlberg have vanished since 2017, with only half of them showing signs of motion (e.g. conspicuous crevasses) (Conzelmann and others, Reference Conzelmann2026). The three largest glaciers in the Pyrénées, Spain (Aneto, Monte Perdido and Ossoue), only covered between 0.2 and 0.3 km² in 2024, with the average remaining ice thickness being 10–15 m. According to López-Moreno and others (Reference López-Moreno, Revuelto, Izagirre, Alonso-González, Vidaller and Bonsoms2025), these glaciers will most likely be gone by 2034. In Norway, both the northernmost (Winsvold and others, Reference Winsvold, Andreassen and Kjøllmoen2026) and southernmost glaciers (Andreassen and others, Reference Andreassen, Robson, Smith, Weber, Carrivick and Kjøllmoen2025b) are about to vanish, and 20 glaciers are reported extinct (Andreassen, Reference Andreassen2022; Raup and others, Reference Raup, Andreassen, Boyer, Howe, Pelto and Rabatel2025). In Svalbard, Elsabreen and Ferdinandbreen rapidly retreated and thinned and are now isolated ice patches (McCerery and others, Reference McCerery2026). The mass-balance year of 2024 was extremely negative in northern Scandinavia and Svalbard, likely contributing to the rapid demise of these glaciers (Schuler and others, Reference Schuler2025; Andreassen and others, Reference Andreassen, Hagen, Kirchner, Moholdt and Schuler2025a).

3.3. Vanishing glaciers in North and South America

Much of the ice in North America is remote, so glacier loss in this region often goes unnoticed except as a contribution to global sea level (The GlaMBIE Team, 2025) or when hazards ensue (e.g. Higman and others, Reference Higman2018). Exceptions are in populated areas where glaciers hold disproportionate symbolic value as sentinels of climate change or have been the subjects of long-term monitoring. Since 2020, western Canada and the conterminous United States have lost 12% of their total ice volume, with a doubling of the 2010–20 mass-loss rate in 2021–24 (Menounos and others, Reference Menounos, Huss, Marshall, Ednie, Florentine and Hartl2025). In the Cascade Mountains of Oregon, USA, half of the named glaciers have disappeared or will disappear by the 2030s (Carlson and others, Reference Carlson, Nicolas, Thayne, Pappas, Molnar and Rood2025). The future of the alpine glacier below Sunlight Peak in northwest Wyoming is similar, with ablation rates expected to consume the remaining 5–20 m of ice in the next decade or two (Meng and others, Reference Meng2026). The compounded effects of climate change are on display as heatwaves directly accelerate glacier wastage, extended ablation seasons cause transient snowlines to rise (Bevington and Menounos, Reference Bevington and Menounos2025) and wildfires darken glacier surfaces, lowering albedo, thereby enhancing summer melt (e.g. Aubry‐Wake and others, Reference Aubry‐Wake, Bertoncini and Pomeroy2022; Menounos and others, Reference Menounos, Huss, Marshall, Ednie, Florentine and Hartl2025). In Canada, among the glaciers vulnerable to extinction are some that have supported decades of athlete development in ski camps (Monty and others, Reference Monty, Flowers, Crompton, Menounos and Mathias2026), and Helm Glacier (Fig. 5), which is one of Canada’s long-term World Glacier Monitoring Service (WGMS) monitoring sites and projected to disappear within the coming decade (Crompton and others, Reference Crompton, Menounos and Ednie2026). In South America, the Andes host glaciers that play a critical role in the hydrology and socio-economics of the surrounding countries. However, glaciers in the Andes are shrinking at an accelerating rate. Their decline is being tracked by a number of detailed in situ monitoring programmes that are themselves endangered by the rapid ice loss and imminent glacier disappearance (Bello and others, Reference Bello, Cashpa, Vilca, Cruz, Torres and Rabatel2026; McPhee and others, Reference McPhee2026). According to Malone and others (Reference Malone, Adrianzen and Broglie2026), 174 glaciers have disappeared from 1998 to 2024 in the Cordillera Real of Bolivia, and even more are at risk of vanishing; the total glacier area decreased by 33% (−1.3% a⁻¹) over this period. In the Northern and Central Andes of Chile, 321 very small (<0.01 km²) glacierets are reported to have vanished (Ugalde and others, Reference Ugalde2026). Also located in the Central Andes, the WGMS reference glacier Eucharren Norte experienced a ∼65% reduction in glacier area from 1955 to 2023 (McPhee and others, Reference McPhee2026).

Figure 5.

Helm Glacier, British Columbia, Canada. The aerial image (acquired 16 September 2024 from the Hakai-UNBC Airborne Coastal Observatory) reveals the glacier’s unhealthy state and impending fragmentation (area of separation denoted with yellow arrow). The glacier is projected to vanish within a decade (Crompton and others, Reference Crompton, Menounos and Ednie2026).

Figure 5 Long description

An annotated aerial map view of Helm Glacier, British Columbia, Canada, showing a broad ice-covered area surrounded by exposed terrain and scattered snow patches. A single arrow annotation points to a narrow linear feature on the glacier surface near the central area of the frame. A scale bar at the lower right is labeled “200 m” and shows a 0 to 200 m distance. A north arrow symbol appears at the lower right.

3.4. Vanishing glaciers in other regions

In Oceania, an investigation of whether snow deposition by wind and avalanches can help offset the increased summer melt of the small (∼0.1 km²) cirque glacier Rolleston Glacier in New Zealand (Purdie and others, Reference Purdie2025) finds that these processes contribute only 8% to the accumulation, insufficient to save the glacier under current climate conditions.

Mountains in the Tropics generally do not provide conditions favourable for glaciers, but for those that do, glacier disappearance is well underway. This includes glacier loss in Africa, which is expected to be the first continent to lose its glaciers because of global warming (Howe and Boyer, Reference Howe and Boyer2025). The remaining glaciers of Papua, Indonesia, have recently diminished from 0.51 km² in 2017 to 0.18 km² in 2024 (Paul, Reference Paul2026). In 2024, the Carihuairazo Ice Cap within the Ecuadorian Andes disappeared (Basantes-Serrano, Reference Basantes-Serrano2026) following retreat rates of 3% a⁻¹ since the 1980s. In the tropical Andes of Peru and Bolivia, glaciers shrank by 14.7% (at an average rate of 1.83% a⁻¹) between 2016 and 2024, with losses concentrated at lower elevation and south- to west-facing slopes (Pacheco-Ferrada and Seehaus, Reference Pacheco-Ferrada and Seehaus2026).

As mentioned, we have no direct studies from High Mountain Asia in the Collection, but the GGCL features stories of the Dagu glacier in China and the Yala glacier in Nepal, which are projected to disappear in the 2030s and 2040s, respectively (Boyer and Howe, Reference Boyer and Howe2024). According to a global-scale modelling study, High Mountain Asia is expected to see peak extinction rates of 1000–2000 glaciers per year ∼2050 (van Tricht and others, Reference van Tricht2025). A previous study comparing two glacier inventories in China also showed that 17% of the glaciers disappeared between the two surveyed periods 1950s–80s and 2000s–10s (Su and others, Reference Su2022).

4.1. Glacier archaeology as an emerging field

Glacial archaeology has emerged as a distinct new field in response to the warming climate and the melting of glaciers and ice patches (Kristensen and Andreassen, Reference Kristensen and Andreassen2026; Pilø, Reference Pilø2026). Thousands of artefacts and biological materials are being released from retreating ice, offering unique insights into past human activities (Fig. 6). Kristensen and Andreassen (Reference Kristensen and Andreassen2026) define a ‘ghost patch’ as an archaeological ice-patch site that no longer contains ice. The disappearance of ice compromises the preservation of artefacts. Researchers should therefore prioritize work where artefact preservation is threatened and learn to recognize ghost patches from the absence of vegetation and other physiographic anomalies (Kristensen and Andreassen, Reference Kristensen and Andreassen2026).

Figure 6.

Archaeological artefacts are released from vanishing glaciers and ice patches. (a) Arrowhead found in August 2019 in front of the ice remnants in (b), eastern Jotunheimen, Norway. Photos: Dag Inge Bakke. (c) New discoveries from the summer of 2024 are displayed at the Norwegian Mountain Center in Lom, Norway. Photo: Liss M. Andreassen.

Figure 6 Long description

The image A showing a close-up view of an arrowhead embedded in ice among small rocks and gravel. The image B showing a wide outdoor landscape with rocky ground and several patches of snow beneath a cloudy sky. The image C showing a side view of one woman leaning over a glass display case with multiple small objects arranged inside, with a wall display in the background.

4.2. Impact on biodiversity, ecosystems and societal consequences

Glacier retreat, and even more importantly, the disappearance of glacier ice in a mountain range, can impact the biodiversity in alpine streams and negatively affect ecosystems (e.g. Bosson and others, Reference Bosson2023). Valle and others (Reference Valle, D’Adda, Scotti, Gobbi and Caccianiga2025) show that the demise of what was once the largest glacier in the Berga Alps of Italy significantly changed the species distribution in the proglacial stream. While one species shifted uphill vertically by 30 m a⁻¹, the other species almost completely collapsed after the glacier meltwater contribution ceased. Dramatic glacier shrinking also impacts the landscape, subjecting formerly subglacial areas to paraglacial processes (McCerery and others, Reference McCerery2026).

Thirty semi-structured interviews with alpinists in the European Alps reveal that the awareness of glacier loss is leading to technical adaptation and avoidance of increasingly unstable areas (Salim, Reference Salim2026). Moreover, shrinking and disappearing glaciers also elicit emotional responses from alpinists who personally experience and adapt to the consequences of the rapidly changing environment (Salim, Reference Salim2026).

4.3. Implications for glacier monitoring

Long-term observations are crucial for capturing the effect of climate change on glaciers. However, the drastic area loss and fragmentation of glaciers observed in recent years critically endangers monitoring efforts in many regions globally. Examples of the imminent loss of valuable long-term monitoring series are documented in this Collection for the European Alps (Huss and others, Reference Huss, Fischer, Linsbauer and Bauder2025), North America (Crompton and others, Reference Crompton, Menounos and Ednie2026) and South America (Bello and others, Reference Bello, Cashpa, Vilca, Cruz, Torres and Rabatel2026; McPhee and others, Reference McPhee2026). Measured since 1975, Echaurren Norte in the Andes is one of two WGMS reference glaciers (with more than 30 years of ongoing glaciological mass-balance measurements) in the Southern Hemisphere (WGMS, Reference Zemp, Gärtner-Roer, Nussbaumer, Welty, Dussaillant and Bannwart2023). Echaurren Norte has therefore played a pivotal role in global-scale assessments of glacier mass change (Dussaillant and others, Reference Dussaillant2025) and must be replaced by expanding mass-balance measurements to nearby glaciers (McPhee and others, Reference McPhee2026).