It came silently in the night. Cloaked in a midnight windstorm, a wall of snow taller than a five-story apartment complex picked up speed as it mangled everything in its path. Adrienne Lindholm never heard or saw it on that night in March 2022.
Living outside Anchorage, Alaska, Lindholm had witnessed the power of avalanches before. She’d seen the massive debris fields in the surrounding Chugach Mountains and watched ski patrol bring down slides at nearby Alyeska, about 40 miles from Anchorage. Every time she stepped out in the backcountry with her skis, she knew there was a chance she could trigger an avalanche of her own. But this was different. As the mass of snow engulfed the driveway of her family home she wondered, “How could this be happening, here?”
Lindholm and more than 100 neighbors woke up on an island. Hiland Road, the artery that connects the collection of family homes to town, was buried under 80 feet of snow and debris.
“It was absolutely terrifying when we were able to appreciate the enormity and the power (of the slide) — it moved you to tears,” says Lindholm.
The avalanche had somehow shot the gap, missing a series of residential structures, including the house Lindholm shares with her husband, J.T., and their two daughters, by less than a hundred feet. Had it come down a little further in either direction, Lindholm’s story could have ended much differently.
“We’re talking about a matter of feet that could have been the difference between a fatality or several fatalities in the runout of that slide,” says Gabriel Wolken, the manager of the Climate and Cryosphere Hazards Program at the Alaska Division of Geological and Geophysical Surveys and a leader in snow science and avalanche behavior in Alaska.
Brad Meiklejohn, an Eagle River resident and former avalanche forecaster, knew the backcountry area up valley was susceptible to slides, but when he awoke to find a wall of snow cutting off his view of the Lindholm’s house, his heart sank.
You see, avalanches, by their very nature, are anomalous, particular beasts.
“I was in my ski gear with my beacon on and my shovel in my hand in about three minutes,” he says. “I was heading out the door prepared for one of the worst days of my life.”
Though Meiklejohn didn’t encounter the worst that day, he considers the Hiland Road slide “a dress rehearsal.” As our planet’s climate continues to change, recognizable weather and snow patterns have become less predictable, putting humans in the crosshairs of unprecedented climactic events. Hindered by a lack of reliable historical data, avalanche forecasters, snow scientists and climatologists are using current observations and future modeling to try and learn more about what winter — and its dangers — will look like in years to come.
The universal truths
For decades scientists have found links between the evolution of natural disasters and climate change. An elongated wildfire season, more powerful hurricanes and increased storm surges are just some of the devastating impacts that have been associated with a warming planet in recent years. Winter is bound to change as we know it, too.
Jim Steenburgh, a professor of atmospheric science at the University of Utah and author of “Secrets of the Greatest Snow on Earth: Weather, Climate Change and Finding Deep Powder in Utah’s Wasatch Mountains and around the World,” says he expects winter storms in Utah to become more intense, but that they will deviate from the traditional snowstorms of years past, with some coming as snow, others as rain, and many as a mix of the two — even at higher elevations. The gap between those large storms may increase as well, signaling that winter snowpack may further decrease in years to come — especially at lower elevations.
That estimation aligns with what has already become a significant drop in snowpack throughout the West. According to a July 2022 National Oceanic and Atmospheric Administration report, spring snowpack across the West has declined by nearly 23 percent on average between 1955 and 2022, and by significantly more in areas like portions of the Sierra Nevada in California.
McKenzie Skiles, an assistant professor and climate scientist at the University of Utah, says that field observations near Alta, Utah, suggest that same increased gap between major storms starting in early fall and lasting throughout the winter, but also point toward some pertinent implications for skiers and snowboarders in the backcountry. While on the surface those data-based conclusions may imply a drop in precipitation overall, they also signal an alarming shift in storm patterns that comes with a new set of consequences.
“Any time we get early season snow events and then precipitation shuts off for a while, we get a weak layer building near the ground,” says Skiles. “That’s the perfect storm for an avalanche and something we normally associate with Colorado snowpack. Now we’re seeing that in the Wasatch.”
A similar gap preceded the avalanche near Eagle River, where an uncharacteristically early September storm followed by a long period of high pressure formed a persistent weak layer that Meiklejohn says turned the Anchorage area’s snowpack into a ticking time bomb.
While avalanches may seem mysterious on the surface, they tend to follow a set of universal truths. Avalanches usually occur in terrain in between 30 degrees and 55 degrees of inclination and are the product of snow instability. Throughout the winter, snowfall adds up incrementally, each storm event forming its own unique layer. Some of these layers bond together to form cohesive, unified layers, but others, influenced by temperature, wind, time in between storms or other factors like dust or sand blown in by wind currents never fully adhere, creating a weak layer. When more snow loads on top of this weak layer, instead of bonding the snow layers together, it acts like old ball bearings between two panes of glass, releasing its new payload and rolling downhill, sometimes at an average speed of 80 miles per hour. While victim-related slides often require a trigger like a human or an animal, most release naturally because of wind loading or falling debris like trees, rocks and ice.
That’s what happened in the Eagle River slide, and scientists worry that those instances could increase as climate fluctuation creates new and atypical snowpacks at home, opening the door for the kind of conditions that form weak layers and cause snow to release destructively. But, as temperatures rise, glaciers melt and the threat of shortening winters nears on the horizon, the relationship between climate change and avalanche behavior is not as binary as it seems, says Steenburgh.
“There are so many different kinds of avalanches, so it’s hard to generalize and provide a simple answer,” he says.
You see, avalanches, by their very nature, are anomalous, particular beasts. Their behaviors change almost completely from one snowpack or slope to another — a few degrees, a few angles of inclination or a fraction of moisture content make all the difference between snow staying plastered to a hillside or hurtling down to the valley below. According to Steenburgh, that same variability is what makes it difficult for scientists and forecasters to concretely establish the broader effect of climate change on avalanche behavior.
Complicating the process is an inconsistent set of avalanche data over a large and varied collection of U.S. mountain ranges. Where European Alpine nations like Switzerland have centralized and federally funded snow science centers like SLF Davos, the U.S. has a handful of individual institutes and experts working with wide-ranging historical records in vastly different mountains. Snowfall data in Salt Lake City, for example, goes all the way back to 1874, but just up the road in Alta, Utah, consistent snowfall records only date back to the 1940s.
“I was in my ski gear with my beacon on and my shovel in my hand in about three minutes, heading out the door prepared for one of the worst days of my life.”
Even with the numbers experts do have, most avalanche reports are written by mitigation crews or resorts rather than witnessed by scientists during a natural slide cycle. As a result, many U.S. climatologists rely on modeling instead of observation when it comes to monitoring changes in avalanche behavior, constructing scenarios off of limited historical records. While these models have helped scientists explore a series of changing snowpack scenarios, experts like Steenburgh say it may require decades of data to make any absolute calls on the correlation between climate change and avalanche behavior.
The human factor
Hundreds of years ago, communities in the Alps blamed avalanches on witchcraft, burning young women they believed to have conjured snow to come down on their villages. Over a millennia earlier, the illustrious Carthaginian general Hannibal had led nearly 18,000 soldiers to their death in a matter of days, marching 100,000 soldiers through the terrain traps of the Italian Alps as slides picked them off by the hundreds.
Natives around Washington’s Mount Rainier used to warn of Tatoosh, the Thunder Bird whose mighty wings could bring down massive avalanches and kept most populations well below snow level. But white settlers in the Northwest pushed beyond these spiritual beliefs and deeper into the mountains, sometimes at great consequence. In fact, the most deadly avalanche in U.S. history occurred just over 60 miles north in Stevens Pass, Washington, when a 1910 slide destroyed a Seattle-bound train, killing 96 and injuring 23.
“Humans are the biggest part of the equation, the biggest contributors to the issue,” says Steenburgh.
At its most simple, when humans are not in avalanche terrain, they are not triggering slides and are not in avalanche danger. But as we move deeper into avalanche areas, whether that be for transportation, agrarian and residential development, or recreating in the backcountry, the natural scales begin to tip against us. We have worked to gain the upper hand, establishing state- and resort-funded avalanche mitigation teams — specialists meant to bring down and channel avalanches with explosives during periods of volatile weather or snow conditions. But those human-activated slides do little to help us understand avalanches or any evolution of avalanche behavior, and actually may be detrimental to fully grasping the changes occurring in the mountains around us.
“A lot of the (avalanche) observations in the U.S. come from avalanche programs designated to mitigate rather than just observe,” explains Erich Peitzsch, a physical scientist at the U.S. Geological Survey Northern Rocky Mountain Science Center. “It’s a little tricky because you’re triggering the avalanches, and that might not mimic natural avalanches.”
That’s part of what makes for this complicated stew of what we see, what we experience, what we believe, and what we know when it comes to the future of winter. And why scientists have to continue to get creative when it comes to collecting data.
Peitzsch, for one, uses tree ring observations in avalanche runout zones to monitor changes in avalanche activity around northern Montana. Overall, he says the Montana observations suggest that the season of typical winter storms in the area has shrunk starting in the ’80s and early ’90s, while “buffer” seasons — traditionally warmer and wetter fall and spring seasons — have grown.
Warming trends point toward less snow at lower elevations, which some say could actually lead to fewer avalanches overall. But Steenburgh says it also may come with a rise in extremely powerful storms, particularly arriving as rain-on-snow events in higher elevations. A product of uncharacteristic warming in areas like Utah’s Wasatch Mountains that typically receive consistent cold temperatures and precipitation, these events could act as potential accelerators for more destructive avalanches.
Hazard maps for the future
The familiarity heuristic is one of six mental “traps” avalanche researcher Ian McCammon identified as problematic for backcountry users in a 2004 article in Avalanche News. According to McCammon, this human propensity to feel comfortable in recognizable settings can prevent people from accurately assessing danger, and can lead to fatal avalanche scenarios.
Dallas Glass, the forecast deputy director for the Northwest Avalanche Center, has taught about heuristic traps for over a decade and has worked in avalanche forecasting and education since 2006. From late November through April, Glass is out in the field, assessing snowpack around Mount Rainier and the Cascades and writing up detailed conditions reports that help backcountry users select their terrain that day. He has also seen the familiarity heuristic play out in real time, where people were buried in avalanches in terrain they had skied, in some cases many times, before.
But he also notes that a changing climate may throw a deeper heuristic curveball. As weather gets increasingly unpredictable, zones that may have been traditionally stable or safe at a certain part of the year need to be looked at through a renewed lens.
“If the situation is morphing under climate change, then we do not have any real experience with that new situation, and the (familiarity) heuristic that we are applying is invalid,” he says. “If global warming is producing a weather pattern that in our 20 years of riding, we’ve never experienced, then your decision-making process is set up for failure at that point.”
He adds that that isn’t just for backcountry users. Avalanche forecasters often rely on pattern recognition, so when unrecognizable trends arise in familiar places, “it can be hard for us to be right,” says Glass.
Oftentimes it’s these human tragedies that tend to drive home understanding for scientists and governments, as well as the general public.
This summer, the Pacific Northwest received an unusual pulse of winter weather in May and June. Late snowstorms coincided with climbing season on Mount Rainier, and Glass was leading a group near the Emmons Glacier when a trio of massive avalanches released. Though he wasn’t near the action, one slide wiped out a popular climbing route up the Ingraham Glacier hours after clients and guides passed through the area. Operations immediately put trips — and the climbing industry on Rainier — on hold for an entire week.
A month after the incident on Rainier, 11 climbers were killed in northeastern Italy after a chunk of the Marmolada Glacier collapsed and kicked off an avalanche in the Dolomites. The Italian government was quick to attribute the natural disaster to climate change, as the failure came just a day after the glacier recorded its highest temperature ever during a scorching heat wave sweeping through southern Europe.
Wolken feels that oftentimes it’s these human tragedies that tend to drive home understanding for scientists and governments, as well as the general public. He doesn’t believe knowledge should come at the price of human life but adds that currently, development is moving faster than avalanche research, putting more and more humans into areas that may become, or may already be within, avalanche areas.
In order to combat that unchecked development, Wolken is working to develop large-scale hazard maps to form a baseline of future avalanche hazards. Made in conjunction with the team over at SLF Davos in Switzerland, the interactive maps will incorporate terrain and simulate changes in climate and conditions to help individuals, developers and municipalities assess avalanche behavior in specific areas. “We want to be able to use science to help guide our decisions in the future,” says Wolken.
As scientists work to fully comprehend the impact of our changing climate on the mountains around us, a major share of responsibility will be placed on backcountry travelers themselves.
“As skiers, we have to be aware of that possibility, that we are going to be seeing snowpack situations and weather situations that we don’t have a lot of experience with,” explains Steenburgh. “Places where this was unusual in the past, they are going to become more frequent.”
Glass says avalanche forecasters often reference the adage, “Weird weather leads to weird avalanches.” On its surface, it’s a vague proclamation, one meant to ask for patience from a community built on instant gratification. But as we step further into a climactic unknown, Glass thinks recognizing that when we find ourselves in uncharted territory, the first step we need to take may not be in the direction we’re used to.
“These are situations where we have a high level of uncertainty,” he says. “That means we’re all going to need to take a big step back.”