In the Library of Congress there is a colour print of the Rhône Glacier from around 1900. The Hôtel Belvédère sits on its hairpin bend on the Furka road, and the glacier fills the valley just below it, a broad tongue of ice close enough that guests could walk out to it after breakfast. The people who bought that print as a souvenir were not documenting anything; they likely just wanted a pretty picture of a holiday in the Alps. Now that picture serves as a measurement.

Stand at that same bend today and the ice is gone from the frame. The Rhône has pulled back roughly 1,300 metres up its valley over the last century, and lost around 60 percent of its volume since 1850 (Swiss glacier monitoring). The hotel closed years ago. The tongue (frontmost tip of a glacier) that once filled the postcard now sits far up the slope, and in summer the Swiss partly wrapped the glacier white blankets to slow the melt.

Almost none of the great historic glacier photographs were taken as science. They were postcards, tourist snapshots at the ice grotto, mountaineering trophies, the occasional survey plate. A century later they have become the most honest long-term visual record of the ice we have, and repeat photography is the tool that turns them back into data.

The Alps hold one of the longest photographic glacier records anywhere. The Rhône was easy to reach and drew tourists from the 1870s. The Mer de Glace above Chamonix and the glaciers of Mont Blanc were sketched, painted, and then photographed from the middle of the 18th century onward. Those scattered images have since been pulled together into real archives. The Glacier Photograph Collection alone holds something like 25,000 images, some dating to the mid-1800s (NSIDC). This is not a fringe method, either. The US Geological Survey has been re-photographing Glacier National Park since 1997, matching modern frames to originals shot between 1887 and 1943, and the resulting pairs became some of the most widely shared climate images in the country (USGS).

The history and the exercise we are undergoing on this sabbatical reveal an additional manmade twist on capturing and sharing these glaciers - copyright law. The photographs we can freely use are the old ones. In the United States that now means roughly anything published before 1931, and across the Alps copyright usually runs until seventy years after the photographer dies. That leaves a gap right where it hurts, across much of the twentieth century, including the decades when the melt truly began to accelerate. To use those images, an independent researcher, a teacher, or a curious student often has to track down the rights holder, ask permission, sometimes pay, and sometimes just be told no (if you can get a hold of the owner at all). It is a strange barrier to build around a shared record of a shared loss, and a big reason we wanted our archive to live in the public domain from the start.

A historic image is only useful if you can pin down three things: when it was taken, exactly what it shows, and where the camera sat (Euro-Climhist, University of Bern). That last one is the real hunt. Vantage points do not stay put. A spot that was open pasture in 1900 is now forest. A hotel terrace becomes a car park. And sometimes the original photographer was standing on the glacier itself, which means you need a drone to suspend the camera in the air. A classic photo of Robert F. Kennedy taken on the Rhone glacier in 1938 with the massif towering in the background is just such a photo, unrepeatable without a drone to suspend the camera 30+ feet in the air. I'd love to share it here, but as mentioned using this photo requires express permission to reproduce online... so I'll just share the link while I wait to hear back from the JFK Library on permission to reshare it here. (https://jfk.blogs.archives.gov/2017/09/08/the-rhone-glacier-evidential-value-of-a-kennedy-family-photograph/)

A couple of the images on our shortlist show why the numbers land harder as pictures. The Rhône is the obvious one. That Belvédère viewpoint gives a clean before and after, more than a century apart, from almost the same patch of road.

The Mer de Glace gives you something even more literal. Every year a crew carves an ice cave into the glacier for visitors, and every year the staircase down to it has to grow longer, because the ice keeps dropping away beneath it. It reached 580 steps by the early 2020s. Here is what those steps are measuring: beneath the Montenvers station the ice was about 250 metres thick around 1915, and at cave level in 2023 it was roughly 30 (Luc Moreau, Train du Montenvers). Each new flight is an accidental ruler, bolted on one summer at a time. The glacier has also given up close to two kilometres of its length since 1850.

So over the next few weeks, that is the work. Pull the best historic frames we can find for "then," work out where the camera stood, then go stand there and shoot "now." Some of those old vantage points will have vanished, and we will have to get as close as the changed ground allows. That gap, between where they stood and where we can, is part of the record too.

PS. It has been really fun practicing German, French, and Italian (and even using a little bit of Spanish) while traveling around Switzerland on this project.

In the fall of 2024, Switzerland and Italy agreed to move their shared border. The reason was not political. The glacier that had defined the frontier high on the Matterhorn massif for more than a century had melted enough to shift the watershed beneath it, so the old line no longer matched the ice. The two countries sat down and redrew the map.

We tend to picture climate change as something measured in centuries (or at least decades) and global averages, abstract enough to argue about or to think “well someone will figure that out.” Glaciers remove that abstraction. They keep a plain physical ledger of every warm summer and thin winter, and right now the ledger is grim.

Since 1976, the world's glaciers, setting aside the Greenland and Antarctic ice sheets, have lost more than 9,000 billion metric tonnes of ice. That is roughly a quarter to a third of all the sea level rise measured, a contribution that now rivals that of rapidly melting Greenland (World Glacier Monitoring Service). About 40 percent of that loss has come in the last decade alone. For three straight years, 2022 through 2024, every glaciated region on Earth lost mass, and 2023 set a record near 600 billion tonnes, the worst single year in five decades (WMO). The 2025 figures, around 408 billion tonnes, offer no real relief.

Numbers on that scale stop meaning much, so here is why they matter to people who will never set foot on a glacier. Mountains supply 55 to 60 percent of the world's annual freshwater flow, and close to two billion people draw their water from them. Two-thirds of the planet's irrigated farmland depends on mountain runoff (UN World Water Development Report, 2025). In the Andes, 85 percent of hydropower comes from mountain catchments. A glacier is a savings account that pays out in the hot, dry months when the rain does not come. As the ice shrinks, the account first overpays, with more meltwater and more flooding, and then it runs dry. Communities from Peru to Pakistan are already living through the early version of that arc. I don’t want to take away from the power of images like the polar bear on the tiny iceberg, but glaciers are more than a symbol, they are a critical source of food and water security in both the developed and developing world.

The Alps, where we will spend most of this project, are the clearest preview we know of. Even if global emissions and warming had stopped completely in 2022, the Alps are already committed to losing at least a third of their ice by mid-century, and on the trend of the last twenty years it is closer to half (Cook et al., 2023). Swiss glaciers gave up about a tenth of their entire volume in the two hot summers of 2022 and 2023. A study published in December 2025 went further and put dates on the disappearances: under the climate policies in place today, perhaps 110 glaciers would survive in central Europe to 2100, about 3 percent of the 3600+ today (Van Tricht et al., 2025). The great Aletsch breaks into pieces, the Rhône becomes a remnant, and stories like the village of Blatten being destroyed in moments by a collapsing glacier will become regular news .

So, why build our glacier archive? We decided to step away for a season to build something we have talked about building for years but have never had the time (or AI enabling technology) to do. The United Nations made 2025 the International Year of Glaciers' Preservation. The attention is welcome but the translation into something a non-specialist can see and feel is still missing.The Glacier Archive Project is an independent effort to document and communicate the changing state of glaciers, using repeat photography, open data, and a set of AI tools we are building as we go. The plan is to put the result into the public domain, an archive anyone can use, because the science here is excellent and well funded, yet most of it lives in journals that ordinary people never open or photographers charge license fees to use.

Repeat photography is the heart of it, and it is beautifully simple. You find where a photographer stood five, ten, fifty, or a hundred years ago, you stand in the same spot, you take the same frame, and then even a kindergartener could play “Spot the Difference.”

Over the next twelve weeks we will move through archival research, fieldwork in the mountains, and a final synthesis, and we will post as we go, including the parts that do not work! Next time we will get into the history of glacier photography and the specific images we hope to match.

Glaciers keep an honest record. Our aim for this project is to help a few more people learn to read it.

The Glacier Archive website is now live.

This site will serve as the public home for the Glacier Archive Project, an effort to document glacier change through repeat photography and field observations. Over the coming months we plan to revisit historic glacier viewpoints and recreate photographs taken decades or even more than a century ago.

The goal is simple: build a visual record showing how glaciers have changed through time and make those comparisons accessible to the public.

The project is currently pending final approval for the upcoming field expedition. Once confirmed, this site will host journal updates from the field, photo comparisons, and documentation of the locations visited.

More updates will follow soon as planning progresses.