A mountain goat on the cliffs is one of the most iconic sights in Glacier. What many people don’t realize is despite being a North American alpine species, mountain goats are extremely sensitive to a warming climate. They are high-elevation animals with little room to retreat upward when conditions change.
Since 2008, hundreds of volunteers have helped count mountain goats to measure long-term population changes. Over the last decade, fewer of these animals have been turning up in the places biologists have long looked for them.
A new study published in Ecosphere in early 2026 draws on that data to produce the first rigorous, park-wide population trend for Glacier’s native mountain goats. The findings, co-authored by researchers at the US Geological Survey and Glacier National Park, are sobering: between 2008 and 2019, the estimated median number of goats per site fell by 45 percent. That decline is large enough to meet the International Union for Conservation of Nature’s threshold for classifying a population as vulnerable.
It is also data that could not have been produced without the help of citizen scientists and the program that trained and organized them.
A Count Worth Keeping
Mountain goats are notoriously hard to monitor. They live in steep, inaccessible terrain, where the cliffs and ridges that protect them from predators also make them difficult for biologists to reach. Aerial surveys are rare because Glacier limits flights to maintain its wilderness values. For decades, estimates of the park’s goat population relied on broad extrapolations with wide margins of error.
In 2008, the park launched its High Country Citizen Science Program to close that gap. Volunteers were trained and deployed to 37 representative sites across the park, each selected to capture a grid of typical goat habitat. At each site, observers scan the viewshed for 1 hour with binoculars and spotting scopes, tracking individual goats to avoid double-counting. Every August, the program anchors around a “Goat Days” event that aims to survey every site within a 3-day window.
Over 12 years, volunteers and staff completed 1,961 surveys. Many returned year after year, and the new analysis shows that consistency mattered.
Reading 12 Years of Data
Counting goats is only part of the job. Any given survey is affected by weather, observer experience, the optics used, and whether the animals happened to be in the viewshed that day. Turning 12 years of raw counts into a trustworthy population trend required a statistical method that separates how many goats are actually at a site from how many a given survey is likely to spot. The approach, called a Bayesian N-mixture model, let the team pool information across sites and years.
Lead author Tabitha Graves, a researcher with the US Geological Survey’s Northern Rocky Mountain Science Center, and her co-authors tested environmental variables at every stage of the model, including snow water equivalent, winter and summer temperatures, early summer precipitation, the extent of permanent snow and glaciers, the presence of natural mineral licks, and how habituated the goats at each site were to human presence.
The length of the record is what made the trend visible. “Although the decline was large, the trend in the goat population was not obvious in the raw data, nor in single-year N-mixture model estimates,” Graves and her co-authors write. Only the full 12-year series, paired with the Bayesian framework, brought the pattern into view.
What the Numbers Show
The results are striking. Across the park, the median number of goats per site dropped from 77.8 in 2008 to 42.3 in 2019. The decline was steepest through 2015, after which numbers appeared to stabilize at a much lower level.
Of the climate variables tested, early summer precipitation during the neonatal period (roughly May 15 to June 15 of the prior year, when nannies are giving birth and nursing newborn kids) emerged as the strongest predictor of population growth.
The broader climate picture is not straightforward. Warmer temperatures alone could support growth in this model, but shrinking snowfields, shifting precipitation, and projected drought push the other way. “Projected reductions of permanent snow, increasing spring and summer temperatures, and insufficient and variable spring precipitation raise concerns for the future of native goats in this region,” the authors write.
Better Counting, Better Conservation
Alongside the trend itself, the study identified ways to sharpen future surveys. Detection rates climbed when observers had more years of experience, when surveys used binoculars, when they happened at cooler temperatures and earlier in the season, and at sites where goats were habituated to people. Those findings will shape how the citizen science program trains and deploys volunteers in the years ahead, improving the statistical power to detect future changes sooner.
The research also points toward what’s needed next: a park-wide population estimate using individual identification methods like fecal DNA sampling, a closer look at the genetic structure and diversity of the goat population, and projections of how continued warming and shrinking snowfields may affect the population’s long-term persistence.
The paper closes with a call grounded in the National Park Service’s core mission. “While numbers may have stabilized at relatively low levels, increased attention to this vulnerable population in the forms of both research and management may be needed to ensure the persistence of this population and meet the NPS mission to preserve and protect resources for the enjoyment of future generations,” the authors write.
Moving Forward
Glacier’s mountain goats are the largest non-hunted, native population of the species in the contiguous United States. Their decline in a fully protected park raises questions that reach well beyond the park’s boundaries. It is also precisely because the park is protected, and the monitoring has been consistent, that the trend could be measured at all.
Twelve years of observation has given biologists a clear baseline, a set of climate variables to track, and a practical roadmap for making the next decade of counts even sharper. Long-term monitoring like this is rare, slow, and expensive, and in Glacier it has been sustained by hundreds of volunteers and by the donors who fund the citizen science program that trains and organizes them.
How can you support mountain goat research in Glacier?
This project and many other critical projects and programs would not be possible without your donations to the Glacier Conservancy.
Donate now to support important work like this in Glacier National Park!
