Ultramarine Glacier in 1986 and 2026 Landsat image, red arrow 1986 terminus and yellow arrow 2026 terminus. The glacier has retreated 2.2 km and in 2026 has no retaied snow cover. (NJ=Nellie Juan Glacier)
Ultramarine Glacier is an outlet glacier of the Sargent Icefield. The terminus extends northwest from the glacier and icefield. To the south is Princeton Glacier and the north is Nellie Juan Glacier. The glacier terminus has been in an expanding proglacial lake, but appers to be retreating from this lake. The glacier has a low elevation accumulaiton between 300 and 525 m, with a terminus tongue extending down to 100 m.
The glacier has retreated 2.2 km since 1986, similar to the 2.8 km retreat of Nellie Juan Glacier. In 2018, 2019, 2020, 2021, 2022, 2024 and 2026 the glacier lost all of its snow cover. The firn layers are evident in the Landsat image above and the Sentinel images below, these layers are not visible when snow cover exists. The ring pattern marks the ice divides where ice flows either to the Nellie Juan, Ultramarine or Princeton Glacier. A glacier without a persistent accumulation zone cannot survive. This is the situation for Ultramarine Glacier, with an area of 17 km2 it will endure for several decades. The Contact Glacier to the north that was connected to the Nellie Juan Glacier as well, is in the same situation, but with an area of 2.5 km2 will disappear sooner. The glacier fits well the regional pattern including in Kenai Fjords National Park (Black and Kurtz, 2021) and Excelsior Glacier on the south side of the icefield.
Ultramarine Glacier with no snow cover in 2018 and 2026. The firn layers are evident which also highlight the flow divides of the glacier. (NJ=Nellie Juan Glacier, PR+=Princeton Glacier)
Retreat of Sheridan Glacier and expansion of Sheridan Lake from 2002-2026 in Landsat images. Lake has expanded from 3.8 to 13.5 km2, due to 2.25 km retreat.
.Sheridan Glacier in the Chugach Mountains of Alaska flows southwest out of the mountains onto a terminus plain where it has gouged out a lake basin that expands as the glacier retreats. Sheridan Lake is a proglacial lake at the terminus that drains into the Sheridan River that 12 km later reaches tidewater. From 1950 to 2000 Sheridan Glacier experienced modest retreat, with a fringing proglacial Sheridan Lake persisting. Then a rapid retreat accompanied a terminus disintegration from 2000-2016. (Shugar et al 2018). This led to lake expansion that is ongoing in 2026.
In 2002 the proglacial lake had an area of 3.8 km2, with the terminus crossing one small island in the lake. The snowline was at 750 m. In 2013 there is a 4 km2 terminus area that has disintegrated, the terminus has retreated off of the island. By 2016 the terminus has retreated north of the glacier stream from Sherman Glacier entering from the east, though much of the lake is still filled with an iceberg melange. The snowline is at 925 m. In 2020 the Sheridan Lake area has expanded to 11.2 km2, with the snowline at 970 m. In 2026 the lake has expanded to an area of 13.5 km2,.epresenting a retreat of 9.7 km2 since 2002. In 2026 the terminus tongue is not stable, unbuttressed on its lateral margins for 2 km, and retreat will continue.
Sheridan Glacier retreat and lake expansion 2002-2020. Snowline indicated by blue dots on these Landsat images.Sheridan Glacier retreat and lake expansion 2013-2016. Snowline indicated by blue dots on these Landsat images.
The Skykomish River in the North Cascades is partially glacier fed. It joins with the Snoqualmie River to form the Snohomish River, flowing into the Puget Sound near Everett. Columbia Glacier is seen on the left of the painting, with the beautiful Blanca Lake beneath. This is one of three glaciers remaining in this region, the other six have recently disappeared. On the right side is the Alpine Lakes Wilderness, two glaciers in this complex remain. The large bar graph in this piece shows Skykomish River discharge (the volume of water) from mid July to late September in 2023. The decrease in water over the dry, hot summer period is evident. The line graph that makes up the mountain above the bars shows the temperature of the river at this same time, highlighting heat wave events. When the temperature spikes, and when the river levels drop, a critical threshold is reached for salmon and other species. This highlights the importance of glaciers in acting as a buffer during drought and heat events.
In 1958 the Skykomish River Basin was home to nineteen glaciers or perennial snow and ice bodies exceeding 0.1 km2 in area, with a combined area of 5.7 km2 (Post and others, 1971). It was the third most glaciated west side basin in the North Cascades behind the Nooksack and Skagit River. These glaciers provided significant runoff to the basin in late summer (Fountain and Tangborn, 1980). We have monitored the mass balance of two glaciers in this basin from 1984-2026, and the changing area and dynamic characteristics of six other glaciers during this period. The Skykomish River Basin has experienced rapid glacier loss leading to the loss of six of its nine glaciers, and over a 40% area reduction of the remaining three.
The North Cascade Glacier Climate Project has maintained an annual mass balance monitoring on the two largest glaciers in the basin from 1984-2026; Columbia and Lynch Glacier. The mean annual mass balance on these two glaciers was –0.4 m/year from 1984-2013, decreasing to –1.5 m/year from 2014 to 2025. This represents the average water equivalent thickness of ice lost from the glacier.
Columbia Glacier in 2003 and 2026 illustrating both the thinning, formation of a new lake and significant area decline.
In the field between 1984 and 1986 we identified nine glaciers in the basin that had active crevassing and sufficient area to be classified as a glacier. During the 2005-2023 period we observed six of these glaciers to have disappeared. Lynch and Columbia Glacier now comprise 88% of the total glaciated area in the basin in 2023, up from 37% of total glaciated area in 1958. Foss Glacier, the third remaining glacier in the basin, has lost 85% of its area since 1958 and has almost disappeared. We will check on this glacier later this summer to see if it still is a glacier. The basin is on a trajectory to lose all its remaining glaciers in the next 2-3 decades; this transition to a non-glacier fed watershed will lead to further reductions in summer discharge and increased summer water temperatures.
In 2026 the approach to Columbia Glacier via the Blanca Lake trailhead was lenghtened by bridge and road washouts from the December atmospheric river event. In 2026 the Columbia Glacier will thin by at least 2 m. The lack of avalanche snowpack indicates the warm winter conditions that prevented buildup of snowpack in the 5000-6500 foot range that feeds this glacier. Lynch Glacier had substantial snowpack remaining in its upper basin above 7200 feet. This indicated that the December atmospheric river event deposited snow not rain on the upper Lynch. This was partly due to the embedded cold air east of the divide that Lynch Glacier sits on. Our approach and exit from Mount Daniel and Lynch Glacier was impacted by smoke from the Three Queens fire.
Foss Glacier continues to fragement and disappear.
In glaciated watersheds glacier runoff is of particular importance to aquatic life in late summer and early fall when other water sources are at a minimum, increasing minimum flow and reducing maximum temperatures, thus providing crucial drought buffering capacity (Ultee and others, 2022). In the Skykomish River Chinook salmon are a threatened species with the fall runs being sensitive to lower flows and higher temperatures.
Stream discharge and temperature are observed at the USGS Skykomish River gaging station at Gold Bar. This gaging station is downstream of the confluence of the North and South Fork of the Skykomish River. The basin above Gold Bar has an area of 1386 km2, the average elevation of the basin is 1050 m. A key threshold of in-stream flow levels considered insufficient to maintain short term survival of fish stocks is below 10% of the mean annual flow, for Skykomish River that threshold is 14 m3s-1 (~490 cfs). In the Skykomish River from 1958 to 2023 there were 390 melt season days with discharge below 14 m3s-1. Of these only three occurred before 1985, and 74% have occurred since 2000. The loss of significant glacier runoff is a key reason for the increase of low flow days. Of more concern for aquatic life is the occurrence of extended periods (7+ days) of low flow-below 14 m3s-1 during the melt season, 1986, 1987, 1992, 1998, 2003, 2005, 2006, 2007, 2015, 2017, 2019, 2021, 2022, 2023, 2024, 2025 and 2026.
These low flows are associated with higher stream temperatures.The Skykomish River was listed as having a water quality standard impairment for water temperature in a 2008 303(d) listing and proposed 2010 303(d) listing, under the Clean Water Act[11]. Several segments of the Skykomish River as well as its tributaries consistently exceed water quality temperature standards today. The TMDL(total maximum daily load) for the Skykomish River indicates the maximum temperature should not exceed 16oC for seven consecutive days.
A water temperature sensor became operational in early July 2022 at the USGS Gold Bar site. The temperature exceeded 18oC for the first time on July 27. It surpassed 18oC diurnally on 37 of 42 days from July 27-September 7. In 2022 the 16oC threshold was exceeded continuously from July 26th to August 4th, August 14th-August 28th, and August 29th to September 4th. In 2023 the 16oC threshold was exceeded continuously from July 27th to August 21st, coinciding with the period of low streamflow from August 11to August 21. In 2025 stream temperature remained above 16C from August 22nd to Sept. 14th.
In 2026 stream temperature rose above 16C on Aug. 5th and as of Aug. 22nd has remained above this threshold. During this same interval discharge has reamined below the adequate flow threshold declining to 360 CFS. The low flow and high stream temperatures will continue to stress salmon and limit recreational uses of the river.
Lynch Glacier from across Pea Soup Lake. The lake has lost its pea soup color and is now trending toward azure. The west side of the glacier is now crisscrossed by rock bands and no longer contributes to flow.Skykomish River discharge and temperature at the USGS Gold Bar station in summer 2023 indicates temperature rising above the 16 C threshold and discharge falling below the 14 m3/second threshold.
Jill Pelto: Field painting of new large bedrock exposures in the big icefall on Easton Glacier, Mt. Baker. The 2026 field season of the NCGCP brought dramatic changes. While we expected most of these, it doesn’t make them any easier to physically navigate or emotionally experience. Even at 9,000 ft, the glaciers of Mt. Baker are thinning, with new bedrock emerging. Above about 7,000 ft, the snowpack was decent, which will slightly help to mitigate the losses from lack of snowpack at lower elevations. The atmospheric river carved a story across our field sites. From this big shift in snowpack up high, to washed out roads, and altered riverbeds. It was a year of fascinating, sad, and occasionally hopeful shifts.
Abby Connor: Looking down at the terminus of the Lower Curtis Glacier on Mt. Shuksan, noticeably thinner than when I last visited. I grew up in Bothell, in a family that spends its weekends outdoors, assuming that massive and seemingly unchanging glaciers are a fact of life. But as we approached our first field site on the Lower Curtis Glacier on Mt. Shuksan, I struggled to reconcile the view of the noticeably less massive terminus of the glacier with my memory from my last visit eight years ago. When we reached our break spot and looked far down at the glacier, Mauri described the wall that would have towered 100 feet over where we sat, before the glacier had lost its impressive thickness. Throughout the 2026 field season, I was continually awed with the scale of the glaciers we still have in the North Cascades while surrounded by reminders of how quickly they’re disappearing, fast enough that we are on track to lose almost all of our glaciers in the state within my lifetime. In Western Washington, a place defined by the mountains and Sound, glaciers are the link between the two. They are no longer an unchanging fact of life.
Ava Reynolds:Pink monkey flower and fireweed bloom in the rocky moraine of the Columbia Glacier – one of the glaciers in the 2026 field season that looked the most impacted by warm temperatures and a reduced snowpack, with avalanche bands that feed the glacier now shutoff, rockfall evident across the entire glacier, and a growing lake of floating ice at the terminus.This field season was visually striking, highlighting the impact of our warming climate on the glaciers and surrounding ecosystem. Flowers were blooming weeks ahead of schedule, sections of new bedrock were exposed, wildfires dotted the horizon at every field site, and residual evidence from the December 2025 storm system was prevalent in downed trees and washed out riverbeds.
Shaina Dotson:Amidst the icefall of Easton Glacier. Joining the NCGCP this field season was equal parts breathtaking and sobering. Each glacier was stunning and wholly its own — ancient and irreplaceable — yet we were immersed in the evidence of a warming planet. Wildfire plumes lined the horizon, ice thinned and retreated where it once held firm, and every day I found myself torn between the beauty of this world and its unraveling.
Danielle Schlunegger-Warner: Red Lens over Rainbow Glacier with Ice Alge sampled from Rainbow Glacier in the microscope. As an artist invited out to attend the 43rd field season with NCCGP, I spent my time in immersed with looking and observing. My attention drew not just to the fantastical forms of the ice on the glaciers, but the way they change the landscape around them. I saw ice algae and tiny black worms dotting the ice and snow, listened to the life bubbling up in the streams at their edges.The focus shifted between the expansive ancient glaciers, microorganisms, and the yearly changes diligently measured by the NCCGP with somber results. This experience illuminated for me not just that these glaciers the heartbeat of the North Cascades, but that their impact ripples out into our own lives in ways that are worth paying more attention to.
Rikki Held: Rainbow Glacier cave with Mauri for scale. The 43rd field season of the NCGCP was filled with both devastating and beautiful moments. We traversed landscapes with fresh glacier-fed streams, rugged mountains, wildflowers, mountain goats, and glaciers with unique formations and characteristics, while also witnessing heartbreaking sites including the large cave hollowing out Rainbow Glacier’s terminus on Mt. Baker and wildfire smoke plumes throughout the field season. Mauri and Jill shared their multi-year observations noting vegetation changes, newly exposed bedrock, and how far the glaciers have receded since they first visited them – e.g., Easton Glacier has retreated around 750m since 1990. This project highlights the importance of long-term data collection, science communication, and personal narratives as we all work to reduce the impacts of anthropogenic climate change and protect the environmental systems we depend on.
Mauri Pelto: Rainbow Glacier cave at the terminus with Abby and Rikki in foreground, viewed from an area where a previous cave had collapsed. Large ice caves are features of retreating glaciers where warm air circulates expanding the cave until it collapses. We observed several of these at the terminus of glaciers on Mount Baker this summer. The largest was at the terminus of Rainbow Glacier. The cave is large enough to hold a cape style house. It will collapse by the end of summer driving further retreat. The size of this cave and the roar of the water was quite intimidating.
Hannah Contreras: In a melted feature looking up at two newly exposed, widening bedrock areas in the Easton Glacier icefall. As bedrock becomes visible, it’s darker shade absorbs heat and melts the ice around it even faster, creating a vicious cycle. In between measuring crevasse and snow depth, this field season we observed newly exposed bedrock in the icefall, again at ~9000 ft just off the popular Easton climber’s route, and finally at the top of the famed Roman Wall (~10,000 ft). Seeing growing bands of melted out bedrock at these elevations and features highlights how threatened the Easton truly is. As a mountaineer, I have spent quite a bit of time both traveling on and marveling at glaciers, but never like this. Experiencing these landscapes always evokes a strong sense of awe, appreciation, and inspiration, but when faced with the magnitude, speed, and decisiveness of their retreat, these feelings were countered by an equally powerful sense of dread and loss.
Figure 1. Global mean annual balance (bars) and cumulative balance (line) from the WGMS data set.
Each year since 2008 I have been responsible for summarizing the global alpine glacier status for each specific year in the Bulletin of the American Meteorological Society State of the Climate report. The 2025 report came out in early August. Below is the report with several added figures.
The 2025 World Glacier Monitoring Service (WGMS) dataset of annual glaciological mass balance observations includes 162 glaciers from six continents and 26 nations, 158 with a negative balance and four with a positive balance. In 2025, the mean annual mass balance of the 59 reporting global reference glaciers was –1.34 m w.e., and -1.34 m w.e. for all 162 reporting glaciers. The 2025 regionalized global average annual balance of –1.09 m w.e., is the fourth consecutive year with mass balance loss exceeding 1.0 m w.e., 2023 had the most negative -1.23 m w.e and 2024 was at 1.04 m w.e. Since 1976, mountain glaciers and ice caps have lost 9179+621 Gt of water, contributing 25.3+1.7 mm to the global mean sea level rise, with 41% of this loss occurring in the last decade (Dussaillant et al. 2025).
The mean annual balance of global reference glaciers was -1.34 m w.e., a similar result to 2023 and 2024 with a mean annual balance for global reference glaciers of -1.60 m w.e. and -1.37 m w.e., respectively. This makes 2025 the 38th consecutive year with a global alpine mass balance loss, the 16th consecutive year with a regionalized global mass balance below –0.5 m w.e. The significant acceleration of mass balance loss is evident in glaciological observations, geodetic observations, altimetry observations and gravimetric observations (The GlaMBIE Team, 2025). The acceleration of mass balance loss indicates glaciers are not approaching equilibrium and increasingly are disappearing. The Global Land Ice Measurement from Space inventory of global glaciers has now added an extinct glacier layer to record glaciers that have disappeared, that is populated by 181 glaciers that had existed in a previous inventory (Raup et al. 2025). In conjunction with 2025 being the UN International Year for Glaciers’ Preservation the Global Glacier Casualty List has been created to chronicle the story of specific glaciers being lost from each glaciated region (Boyer and Howe, 2025).
Hofsjokull eystri, Iceland was snow free by mid-August, leading to extensive ice cap thinning.
In the European Alps all 54 glaciers observed had negative balances, with 45 of the glaciers losing over 1 .0 m w.e. In Iceland (6) and Svalbard (9) all 15 glaciers observed lost over 1.0 m w.e. In Norway 8 of 10 glaciers had a negative balance, with two glaciers having a positive balance. In Sweden 3 of 4 glaciers had a negative balance.
High snow line on Durung Drung Glacier, India near end of summer monsoon season.
In Asia all 23 glaciers across 7 nations had negative balances averaging 0.97 m w.e. In New Zealand the glaciers observed had a negative mass balance.
Agua Negra Glacier, Argentina with minimal snow cover at end of summer melt season.
In the Andes Mountains of South America nine of ten, reporting from four nations, had negative balances, with an average balance of -0.92 m w.e.
All 14 glaciers in Western North America, outside of Alaska, had mass balance below -1.0 m w.e. with an average of -2.24 m w.e. Peyto Glacier, Canada exemplifies the limited retained snow cover on alpine glaciers in Western North America in 2025, a glacier should be 50-60% snow covered at the end of summer to be in equilibrium (Fig X2). It is evident Peyto Glacier is representative of adjacent glaciers that also have limited retained snow cover. In Alaska 3 of 4 glaciers had negative balances. The five glaciers in Arctic Canada all had small negative balances averaging -0.16 m w.e.
Snow cover was minimal in the Yoho-Peyto Glacier regions, Canada at the end of the melt season.
Alpine annual mass balance glaciological observations are reported to the WGMS by National representatives with a December 1 annual submission deadline. WGMS reference glaciers (60) have at least thirty continuous years of mass balance observation. Benchmark glaciers (23) have at least a ten-year mass balance record and are in regions that lack sufficient reference glaciers. The combination of benchmark and reference glaciers are used to generate regional averages (WGMS, 2023). Global values are calculated using a single averaged value for each of 19 mountain regions, limiting bias from observed regions (WGMS, 2023). Expansion of this data set leads to reanalysis and updated values.
References
Boyer, D. and Howe, C. 2025. Communication efforts to educate the public about vanishing glaciers, 1958–2025. Annals of Glaciology, 67: e2. doi:10.1017/aog.2025.10033
Dussaillant, I., Hugonnet, R., Huss, M., Berthier, E., Bannwart, J., Paul, F., and Zemp, M. 2025: Annual mass change of the world’s glaciers from 1976 to 2024 by temporal downscaling of satellite data with in-situ observations. Earth System Science Data 17(5): 1977-2006, https://essd.copernicus.org/articles/17/1977/2025/.
Raup, B., Andreassen, L., Boyer, D., Howe, C., Pelto, M., and Rabatel, A. 2025: Tracking extinct glaciers in GLIMS. Annals of Glaciology, 66, e35, 1–6. https://doi.org/10.1017/aog.2025.10027
Hovgard Kystland Glacier in Sentinel image from July 4, 2026 with margin indicated by black dots. The 2020 margin seen below is indicated by green dots.
Hovgard Kystland Glacier is an outlet glacier in West Greenland between Alison and Hayes Glacier. Alison Glacier had the highest retreat rate from 1976-2021 losing 14.3 km in length and 59.4 km2 in terminus area (Black and Joughin, 2026). Hayes Glacier lost 2.7 km in length and 10.8 km2 in area (Black and Joughin, 2026).. They indicate that Hovgard Kystland Glacier retreated 5.4 km and lost 21.3 km2 during this interval.
Here we examine the acceleration of retreat from 2020 to 2026. In 2020 the central tongue of the glacier extended west beyond the main front. This central tongue collapsed by July 2024 leaving a generally north/south calving front, the glacier had lost 5.7 km2 of terminus area. From July 2024 to July 2026 an embayment formed generating a concave calving front. The glacier terminus lost another 4.8 km of area. Retreat of the calving front was 2.5 km during the 2021-2026 period. The glacier has lost an area that is 50% of the area lost from 1976-2021, in the last five years. The embayment is poised to further expand, though this summer an extensive packed melange is currently in place which typically limits calving (Meng et al. 2025).
Hovgard Kystland Glacier in Sentinel image from July 27, 2020. The 2020 margin seen below is indicated by green dots and the 2026 margin with black dots.Hovgard Kystland Glacier in July 4, 2024 Sentinel image. The calving front indicated by red dots for 2024 and black dots for 2026.
View from Heather Meadows to Table Mt on May 2, 2026 (Jill Pelto photo)
Since 1984 we have measured glacier mass balance on North Cascade glaciers every year. Mass balance is the difference between accumulation (income) and melt (expenses). The accumulation season typicallys ends around May 1. In 2026 it ended on April 19.
In early December snowpack was average above 1500 m (5000 feet) and limited below. The region was then impacted by a historic atmospheric river from Dec. 8-12. that led to flooding and landslides closing I-90 and Highway 2. Snowpack was completely lost at most locations below 1500 m including the Mount Baker station in Heather Meadows (1285 m). Above 1500 m at Lyman Lake (1825 m) SWE increased by 15 cm, which was 50% of the 30 cm of precipitation that fell in this period. The last half of December snow depth increased from 0 to 1.5 m at Heather Meadows. Snowpack built to 1.2 m w.e. at Lyman Lake by March 15, and then hovered around this peak until mid-April. At Heather Meadows snowpack depth peaked at 2.8 m on March 15th and was at 2.7 m on April 15th. This April 15th represented the maximum snowpack for elevations at 1500 m and above. The snowpack was less than 50% of normal below 1800 m. Above 2000 m warm wet winter events did deliver snowfall that was rain below and the situation is likely better. Working with Snowgoat Skimo as they prepare for the Kulshan Randoneee has noted snowpack of over 6 m at the 2000 m level.
The transition to melt season was rapid with a particularly warm period from April 26-May 4 has led to rapid melt off of snowpack. At Lyman Lake a third of the snowpack was lost by May 8th. At Heather Meadow 50% of snowpack depth was lost by May 8th.
Mount Baker snow depth station indicating loss of snow in December atmospheric river, then rapid development and finally the unusually rapid drop in late April. WSCO graph.
View toward lower Heather Meadows and Baker ski area-snowline above 1200 m on 5-2-26.-where the snowpack is already thin. (Jill Pelto photo)
Sentinel Image from 5-7-2020 a normal snow year-snow line 850 m.
Sentinel image from 5-4-2026 illustrating regional snow line at 1200 m. The limited winter snowpack indicates that in 2026 we have a low accumulation/income for glaciers and melt/expenses already underway. Coleman Glacier terminus (E) is already exposed ice the first week in May. We will be in the field this summer to measure the details. We are expecting more bare ice and rockfall during our work.
Sentinel image from 5-6-2025 snow cover- snow line 975 m.
Sentinel image from 5-9-2024 with snow line at 975 m.
Looking from Austin Pass across to Shuskan on 5-3-26 (Jill Pelto photo)
The outlook is poor for glaciers and snowpack in the Pacific Northwest in 2026. How this measures up with the snow drought of 2015 that we reported on in a project with NASA will be important to observe.
March 21, 2026 was World Day for Glaciers, telling this story is an ongoing four decades long project for us. The North Cascade Glacier Climate Project began in 1984 to identify the response of Washington’s North Cascade glaciers to climate change. The 2026 field season will mark our 43rd year of observations.
I co-direct the project as the science director, with Jill Pelto as the art director. We strive to create a diverse and collaborative portfolio of impactful observations connecting people to science through art and data publication. Our theory of change is that art has the power to inspire people by including the emotional context behind the science research. We seek collaborators who are passionate about contributing to and sharing this work. Here we reflect on some of the stories we have shared in the last 12 months that chronicle the rapidly melting reality of glaciers.
In April, 2025 the “Shaped by Ice” exhibit coordinated by Jill Pelto opened at the Slip Gallery in Seattle. This featured work from nine artists, eight of whom had been in the field with us and was the focus of a Seattle Magazine article.
In May we worked with Protect Our Winters to explore what the loss of glaciers, “Beyond the Ice” means for specific regions.
In June, High Country News focused on the use of art to preserve the legacy of glaciers, given the glaciers themselves are not being preserved.
In March through July we worked with the Global Glacier Casualty List to the tell the story of specific glaciers in the United States that are disappearing, including Ice Worm Glacier in the North Cascades, Darwin Glacier in the Sierra Nevada, Twins Glacier in Wyoming and Burroughs Glacier in Alaska.
In August our field season coincided with a Seattle Time report, on Melting Mountains in the Pacific Northwest means, that we collaborated on.
Joining us in the field on the Lower Curtis Glacier, Mount Shuksan was KING5 NBC-Seattle affiliate, led by Meteorologist Leah Pezzetti who backpacked with us into her first glacier. The logistical preparation for this by KING5 was thorough, and we provided guidance and support to them. The resulting feature was released while we were still in the field.
Upon returning to the trailhead from Lower Curtis Glacier a CBS Sunday Morning News team met us and travelled into the field with us to Sholes Glacier, Mount Baker. The overnight trip into the backcountry was a first for this news team and required extensive collaborative support. The results were shared just two weeks after the filming, and right after our field season.
We worked for a second year with Dan MacComb on a short film for the UN International Year of Glaciers’ Preservation called “Shaped by Ice”. This film published by Protect our Winters has been featured in tfour film festivals during 2025.
In January through February a second “Shaped by Ice” gallery event was held in at the Confluence Gallery in Twisp WA. This was spearheaded by Jill Pelto, Margaret Kingston, and Claire Waichler.
A poor winter snowpack will certainly lead to another year of extensive glacier loss across the Pacific Northwest and we will be there to observe it.
The retreat of the main outlet glaciers of Cook Ice Cap is evident in the Landsat images from 2001 (red arrows) to 2025 (yellow dots). Dumont D’Urville Glacier=-1.5 km, Vallot Glacier=-2.4 km, Naumann Glacier=4.0 km, Ampere Glacier=-3.3 km and Diosaz Glacier=-0.8 km.
Kerguelen Island is at the edge of the furious fifties in the southern Indian Ocean. The island is host to many glaciers, the largest being the Cook Ice Cap at 400 km2. A comparison of aerial images from 1963 and 2001 by Berthier et al (2009) indicated the ice cap lost 21 % of its area in that period. The east side of the Cook Ice Cap outlet glaciers tetreat has led to formation and expansion of a new group of lakes (Pelto, 2016). Here we examine the changes from 2001-2025 along using Landsat and Sentinel imagery.
Ampere Glacier main terminus retreated 800 meters from 2001-2011. Here the terminus has pulled back from the tip of the peninsula on the west side of the terminus and is currently at a narrow point. The eastern terminus has retreated to its junction with the main Ampere Glacier a distance of 1400 m. From 2011 to 2025 the retreat accelerated totalling 3.3 km by 2025. The glacier had two nunataks L and N in 2001, by 2025 L has emerged as a marginal mountain and N is barely surrounded by ice.
Diosaz Glacier retreated 0.8 km from 2001-2025 leading to a small new lake basin developing.
Naumann Glacier retreat has been 4.0 km from 2001-2025 creating an new alpine lake. With the glacier no longer terminating in the lake retreat should slow.
Vallot Glacier has retreated 2.4 km creating a new alpine lake at its southern terminus. The glacier will have just one terminus in the near future and the two lake basins could merge.
Dumon d”Urville Glacier has retreated 1.5 km with the lake it terminates in developing an expanding southern embayment.
The east side of the Cook Ice Cap on Kerguelen Island outlet glaciers retreatn and expanding group of lakes illustrates widespread ice cap thinning (Pelto, 2016). Verfaillie et al (2016) identified that the accelerating glacier wastage on Kerguelen Island was due to reduced net accumulation and resulting rise in the transient snowline since the 1970s, when a significant warming began.
Cook Ice Cap in 2011 Landsat and 2017 Sentinel image. Red arrows indicate 2001 terminus positions and orange dots the 2011 terminus position.
Glaciar Mayo terminus change from November 2025 to February 2026 illustrated in Sentinel images. The yellow dots indicate the margin, which has retreated on both sides forming a melange (M) and new icebergs (I).
Glaciar Mayo, Argentna in Los Glaciares National Park is an eastern outlet of the Southern Patagonia Ice Cap. The glacier has terminated on the northern shore of a glacial lake for the last 35 years. The mass balance from 1975-2011 was identified as slightly positive by Schaefer et al (2015). This enabled the glacier to terminate on the northern shore of a glacial lake, an arm of Lago Argentino from 1984-2020. From 2000-2019 Minowa et al (2021) noted that Glaciar Mayo had transitioned to a negative balance and overall thinning. This thinning is what has led to the terminus beginning to collapse into the lake.
The terminus tongue projecting into the lake had been 2.5 km2, had now declined to 1.4 km2. The terminus along the northern shore had been 2000 m wide and is now 1150 m wide. There is further crevassing/rifting that suggests the glacier tongue is not done thinning in 2026. This continues to be an active year for calving retreat in Patagonia as seen at Upsala and Jorge Montt Glacier, see below.
Glaciar Mayo terminus change from November 2025 to February 2026 illustrated in Landsat images. The yellow dots indicate the margin, which has retreated on both sides forming a melange (M) and new icebergs (I).Jorge Montt Glacier retreat from 2021-2026 with a particularly extensive and packed melanage in 2026, observed in Sentinel images.Tyndall Glacier 2 km retreat from 2022-2025 continues with heavily crevassed terminus front indicating an active calving retreat will continue in these Sentinel images. The terminus front is noticably higher above the water indicating thicker ice and likely deeper water.Upsala Glacier had a burst of calving in Feb. 2026. The comparison to 2021 highlights both the retreat, separation from Bertachhi Glacier and substantial drainage of Lago Guillermo.
Novosilski Glacier (N) on the west coast, Risting Glacier (R) in Drygalski fjord, Salomon Glacier (S) on the south shore, Twitcher Glacier (T) in Twitcher Bay and Hindle (Hi) and Ross Glacier (RO) in Royal Bay eachexperienced a detachment (D) from a tributary glacier due to glacier retreat from 2016 to 2016, as noted in these Landsat images.
South Georgia Isalnd is a notoriously cloudy location. A remarkably clear Landsat image from Jan. 21, 2026 provides a good snapshot of glacier change since a similarly clear day on February 19, 2016. Here we examine several glacier near the southern tip of the island that have experienced retreat leading to glacier detachment. Pelto (2017) documented the retreat of 11 of these glaciers during the 1989-2015 period. I worked with NASA Earth to document the retreat of some east coast glaciers evident in Landsat images at that time. BAS map provides context on wildlife populations and glacier retreat.
Novosilski Glacier (N) in 2016 is still connected to Tributary 1. By 2026 tributary 1 is separated from the main glacier due to a 1.3 km retreat.
Novosilski Glacier is a large tidewater outlet glacier on the west (cloudier) coast of South Georgia terminating in Novosilski Bay It shares a divide with the rapidly retreating Ross and Hindle Glacier on the east coast. The glacier retreated 1.3 km from 2020 to 2026 leading to Tributary 1 no longer connecting to the main glacier.
From 2016 to 2026 Ross Glacier retreated 2.5 km and Hindle Glacier 1.6 km. The Hindle retreat led to Tributary 1, 2 and 3 all detaching from one another.
For Ross and Hindle Glacier in 1989 the glaciers joined 2.5 km from the terminus spanning Royal Bay with a 3.2 km wide calving front. By 2001 the glacier front had retreated 800 m, but was still a single joined calving front. By 2009 the glaciers had separated due to an additional retreat of 1.4 km. The Hindle Glacier front was now retreating south up opening a new separate fjord from Ross Glacier. The calving front in 2009 was 1.6 km wide. By 2015 a 1.6 km retreat led to the separation of Hindle from Tributary 4. From 2016 to 2026 a further 1.6 km retreat to the approximate head of the fjord led to separation of Tributary 1, 2 and 3. Ross Glacier retreated 2.5 km from 2016-2026 without separating from any tributary.
A 1 km retreat of Risting Glacier (R) led to Tributary 2 detaching from the Tributary 2 and 3 between 2016 and 2026,
Risting Glacier terminates in Drygalski Fjord on the southeast coast of South Georgia. Risting and Jenkins Glacier were joined until the 1980’s. Cook et al (2010) note the glacier had a relatively uniform retreat rate from 1955-1999 of 40 to 50 meters/year, with retreat increasing after 2000. From 2002-2016 Risting Glacier retreated 1100 m, a rate of ~80 m/year twice the 1955-1999 rate. From2016-2026 the glacier retreated another 800 m, continuing at the same rate. This led to detachment of tributary 1 from the rest of the glacier.
Twitcher Glacier (T) retreated 3.7 km from 2016 to 2026 leading to detachment of Tributary 1.
Twitcher Glacier is the next glacier south of Herz Glacier (H) on the east coast of South Georgia. Until 1989 the glacier ended at the tip of a peninsula that separates it from Herz Glacier., the ensuing retreat has led to the opening of a new fjord. By 2015 the glacier has retreated 2.2 km. An accelerated retreat from 2016-2026 of 3.7 km led to separation from Tributary 1.
The story is the same from glacier to glacier with extensive retreat leading to tributaries detaching from each other. These glaciers still maintain snow cover across a significant area of the glacier and can survive current climate.
Videla Glacier, Chile is an outlet glacier of the Cordillera Darwin Icefield. The glacier has a number (Point 1-6) of terminus lobes where retreat has led to proglacial lake development between 1997 and 2025, as seen in these annotated Landsat images.
Videla Glacier is a land terminating glacier in the northwest portion of the Cordillera Darwin Icefield (CDI) in Tierra del Fuego, Chile. The glacier has terminates in several expanding proglacial lakes each in front of a different tongue of the glacier. The glacier flows northwest from Cerro Ambience towards Fiordo Profundo. Meier et al (2018) identified area change of Patagonia glaciers from 1870-2016 with a ~16% area loss of CDI, with more than half of the loss occurring since 1985. They also noted that CDI glaciers were retreating fastest between 1986 and 2005. Izagirre et al (2025) identified a 124% increase in glacier lake area from retreat between 1945 and 2024. The retreat has been largest on tidewater glaciers such as Marinelli Glacier and Ventisquero Grande Glacier.
In 1997 of Videla Glacier’s six main terminus lobes, five did not exhibit a proglacial lake, only the two northern most lobes (Point 4, 5, and 6) ending in a fringing yet to develop proglacial lake. The terminus lobes at Point 2 and 3 were joined. By 2019 lobes 1 and 4 had developed significant proglacial lakes, while the main terminus at Point 5 and 6 had opened up two halves of the same proglacial lake. The terminus lobes at Point 2 and 3 had separated. A rib (yellow arrow) was developing upglacier of the main terminus indicating thinning and reduced flow. A new lake had developed just downstream of this rib.
In 2025 the terminus at Point 1 had receded 950 m creating a 0.75 km2 proglacial lake. Terminus Lobe 2 and 3 had separated by 400 m. At Point 4 a 0.5 km2 proglacial lake had formed with the 1050 m retreat. The main terminus at Point 5 and 6 extends across the lake basin in a narrow 350 m wide tongue. The lake has grown to 3 km2, with 1.5 km of recession from Point 6 and 1.8 km from Point 5. This narrow tongue may well break off this coming summer.
Videla Glacier, Chile ongoing retreat and proglacial lake growth at terminus lobes (1-6) illustrated by Landsat images from 2019 and 2022.
