Over the years, advances in satellite technology made it seem like Arctic snow cover was growing. In reality, the satellites were simply getting better at spotting the shrinking patches of snow.
For years, reports from the United Nations’ Intergovernmental Panel on Climate Change (IPCC) have helped shape our understanding of the planet’s changing climate. But new research from the University of Toronto (U of T) suggests that one widely used data set has been telling an incomplete and misleading story about what’s happening in the Arctic.
At the centre of the issue is long-running snow cover data from the U.S. National Oceanic and Atmospheric Administration (NOAA), which tracks how much of the Northern Hemisphere is covered by snow each autumn. These observations go back to the 1960s and have played an important role in IPCC assessments.
Snow matters because it acts like a giant mirror. While bare ground and vegetation absorb much of the sun’s energy, fresh snow reflects most of it back into space. When snow disappears, more heat is absorbed by the land, which leads to further warming and even more snow loss, a self-reinforcing loop known as the snow-albedo effect. The Arctic is warming faster than the rest of the world, partly due to this feedback.
“Snow loss doesn’t just respond to warming; it amplifies it,” says Aleksandra Elias Chereque, a PhD student in physics at U of T. “That’s a big part of why the Arctic is changing so quickly.”
For a long time, though, scientists have been puzzled by NOAA’s snow cover records, which appeared to show that Northern Hemisphere snow cover was actually increasing over time. This discovery contradicted other observations and sparked concerns within the climate science community.
Elias Chereque and her colleagues decided to take a closer look. In a detailed reanalysis of the NOAA data, they found that the apparent increase in snow cover wasn’t real at all. Instead, it resulted from gradual changes in satellites and measurement techniques over the decades. As the instruments improved, they became better at detecting very thin or patchy snow, creating the illusion that snow cover was expanding.
It’s comparable to regularly upgrading a pair of eyeglasses,” explains Elias Chereque. “You’re not suddenly encountering new objects in your field of vision; rather, your ability to see what is already there becomes sharper and more precise. Similarly, the satellites were not detecting an actual increase in snow cover; they were simply improving in their capacity to identify the remaining, increasingly scarce patches of snow across the Arctic.
Accounting for these technical shifts reverses the trend. Rather than growing, autumn snow cover in the Northern Hemisphere has been shrinking by about half a million square kilometres per decade.
The study, published in Science Advances, was co-authored by U of T atmospheric physicist Paul Kushner and researchers from Environment and Climate Change Canada. Their findings strengthen the evidence that snow cover has been declining throughout the year and help clarify how this loss contributes to Arctic warming.
By carefully untangling how changes in measurement methods distorted the data, the researchers say their work also significantly improves confidence in climate models and future projections, especially for better understanding long-term Arctic warming trends.
“Understanding where the data went wrong helps us use it more carefully,” says Elias Chereque. “That, in turn, helps us better understand the past, test our models, and make more reliable predictions about where the climate is headed.”
In short, the Arctic isn’t gaining snow; it’s losing it. Seeing this clearly helps us grasp how rapidly the region is warming and why those changes matter for the planet as a whole.
