Avalanche! How lasers and wind tunnels make mountains safer
Wired, October 2012
Avalanches travel at 130kph and kill 300 people a year. That’s why, deep in the Swiss Alps, scientists are turning to lasers, CT scans and wind tunnels to understand snow better
At 9.30 on the morning of 17 August, 2008, a British former pro-snowboarder turned cameraman was shooting footage in the Southern Alps of New Zealand near the 3,754-metre Mount Cook, the country’s highest peak. Johno Verity had been hired by a UK TV-production team making a show, Gethin Jones’ Danger Hunters, about extreme sports. His job that day was to film Austrian snowboarder Eric Themel in action, keeping pace beside him with a camera as Themel arced on his board through deep snow.
For the previous three days heavy snow had fallen, making a planned ascent into the mountains by helicopter impossible. But on the 17th, the team woke to better weather. “It was epic,” says Verity. “Blue skies, light breeze, perfect light snow. You couldn’t ask for better conditions. We were brimming with excitement.” Themel, Verity and a local mountain guide boarded a red-and-white Eurocopter AStar 350 and touched down on an unnamed peak deep in the range. Following standard back-country safety procedure, the guide shovelled a metre-deep hole in the snow, exposing the numerous layers of snow that had built up over the past days, and checked for any weak layer that signalled avalanche risk.
Verity noticed the guide was inspecting an aspect of the slope marginally less steep than the one he and Themel would be riding. He knew that even subtle variations in the conditions at which the Sun or wind hit the mountain could affect the stability of the snowpack. But the day was bright and he was eager to get moving.
Tracking Themel through the viewfinder of a high-definition camera, Verity rode beside him, keeping pace ten metres to his right. “Suddenly, I thought, ‘I’ve kicked up a massive amount of snow.’ I was annoyed because I’d almost lost Eric on camera in the snow. Then I realised it was something else.” From around him, Verity heard “a screechy, creaking sound”. He saw a slab of snow, at a width of about 20 metres either side of him, break away from the mountain and start to tumble down the slope in pieces. “I realised I was in a pretty big avalanche.”
As an experienced snowboarder, Verity knew the only form of escape lay in accelerating as fast as possible out of the path of the snow. That was what Themel, who’d also been initially caught in the avalanche, had done, cutting right on to stable terrain at the side of the mountain. But Verity had less momentum and he found it impossible to speed up. “I was a sitting duck. I was in a lot of trouble,” he says. Caught in the middle of the slope, he was being carried down the mountain on a tide of snow. What worried him most was that the slope dipped further ahead and then rose again, forming a natural bowl that would already be filling with snow up to ten metres deep. If he ended up in there, he probably wouldn’t make it out alive. But that’s exactly where he was heading. “I went straight into the bowl. Instantly I was flipped on to my front. Snow started piling on top of me. It started getting darker and darker.” Churned around, he tried to cup his hands over his mouth to create an air pocket, even as snow forced itself up his nose and into his lungs. Verity was buried deep. “I was very, very afraid,” he says. “The most afraid I’d ever been.”
But he hadn’t stopped moving. As the wave of snow surged up the side of the bowl, he was pushed out over the top, flying through the air, then landing and sliding 200 metres down the mountainside before coming to a stop. He was breathless, dazed, hacking furiously to clear his lungs of snow. But he was unhurt. Themel, who’d been watching helplessly from the side, rode down and grabbed Verity’s hand. “That was an amazing moment. It was pretty good to have that contact with someone. It felt good. It felt good being alive.”
Verity’s ordeal had lasted less than a minute. But during that time he was forced 800m down the mountain by a release of snow some 90m wide and 50cm deep. He was lucky to be alive. But others caught in similar circumstances are less fortunate. Around 300 people die annually in avalanches. The number of fatalities has remained unchanged for decades even though improved avalanche mapping has led to the construction of fewer buildings on high-risk land. The reason? Every year more and more of us are venturing into unmarked, off-piste territory in search of adventure. We’re taken there by snowmobiles, fat skis, snowboards and other high-grade equipment that can get us into — but not necessarily out of — hazardous situations. Today, 90 per cent of avalanche deaths around the world occur in accidents triggered by skiers and snowboarders themselves. The majority of victims are men in their 20s — a group that typically has an exaggerated faith in its own abilities, and a sometimes fatal under-appreciation of the risks of open country.
In the Alps there are around 100 victims a year. Switzerland, France, Austria, and Italy experience the most avalanches a year — from October 2010 to September 2011 there were 26 victims in Switzerland. Switzerland’s accident reports record a sobering litany of mishap: 7 December, 2011. Activity: Skiing on piste. One child completely buried and found (dead) after 45 minutes by probing. 8 February, 2012. Activity: Out-of-bounds skiing. One person completely buried and died. Found by avalanche dog. 12 February, 2012. Activity: Back-country skiing. One person caught and injured. Died one week later. 23 April, 2012. Activity: Back-country skiing. Two people caught. One swept over rock face. Fatally injured.
Switzerland’s worst season in modern times was 1951, the so-called “Winter of Terror”, when 98 people died in avalanches. In 1999, the country faced a series of major avalanches. Seventeen people died, thousands more were evacuated from their homes, and damage to infrastructure came to more than 600 million Swiss francs (£393 million) after more than 1,300 avalanches fell.
The ever-present risk of avalanche has spurred a national resolve to tackle the problem. Leading the effort is the Swiss Federal Institute for Snow and Avalanche Research (SLF). Based in Davos, the SLF is an applied research centre with more than 120 employees. Most are scientists and graduate students — physicists, geographers, geologists and engineers — who are engaged in the highest level work on the forecasting of avalanches and the dynamics of snow. The SLF’s mission is to help the people and businesses of Switzerland manage the risk of living and working in the avalanche-prone Alps.
To do this, it relies upon the most advanced technology — “Whatever is available and will work in the harsh alpine environment,” as Jurg Schweizer, SLF’s head of research, puts it. Its tools range from the use of X-ray CT scanners, to studies of microscopic variations in snow structure, to helicopter-mounted lasers that are fired down on to mountainsides to measure snow distribution.
The SLF also has its own wind tunnel, located 1,645 metres above sea level on the edge of a slope near Davos. Snow particles are blown along its 15 metres and bombarded with lasers, ultrasonics and other measurement apparatus to determine the penetration of near-surface wind into snow pack.
In a six-month project in 2009, SLF researchers joined scientists from Ecole Polytechnique Fédérale De Lausanne (EPFL) to erect an X-band radar on the mountains above Davos. Its two-metre-wide parabolic antenna scanned the skies for rain and snowfall. Both partners also developed one of the fastest high-resolution cameras in the world, using single-photon detector technology to create an exactingly precise prototype system for monitoring precipitation.
If you’ve ever been skiing in Switzerland, you’ll have come across the SLF’s output in the form of the avalanche-warning bulletins posted daily at mountain resorts throughout the country. But the range of its activities is broader, covering areas such as biodiversity, climate change and mountain ecosystems. There are other centres round the world devoted to similar tasks, chiefly in the US, but these operate on a regional level, in Colorado, say, or Alaska. By contrast, the SLF is a nationwide body and it functions at a scale of expertise and technological sophistication that far outstrips its peers. In avalanche research the SLF is comparable to MIT, a place where smart people from different disciplines gather in pursuit of new ways to understand the world.
Human beings have been killed in avalanches for as long as we have been venturing into the mountains. In 218 BC, Hannibal and an army of 38,000 troops were crossing the French Alps when an avalanche struck, sweeping 18,000 of his men and 2,000 horses and elephants to their deaths. In April of this year a huge avalanche engulfed a military complex in Siachen, northern Pakistan, burying 139 soldiers and civilians and leaving no survivors. And as recently as July, nine experienced climbers were killed after an avalanche hit Chamonix in the French Alps. Yet our understanding of the actual mechanics of avalanches remains relatively limited, even now. The SLF hopes to change that.
“Avalanches are rare, but often extreme events; they are characterised by their abruptness with no early warning and the potential for events of nearly any size to occur,” Schweizer says. “Studying them is challenging since they are difficult to observe due to their dangerous nature and since their occurrence is largely not predictable. Small changes in snow properties over a limited area — hidden to the human eye — may have catastrophic consequences.”
It’s impossible to predict when or where an avalanche will strike and their speed and ferocity make gathering data extremely difficult. And getting too close involves a high degree of danger. But in order to prevent a repeat of 1951 or 1999, SLF scientists have to turn avalanche chaser and actively seek out conditions at their most capricious and lethal. If the job doesn’t exactly require them to risk life and limb, it certainly demands a passion for wild nature that sometimes makes SLF staff sound more like adventurers than dispassionate scientists. As Betty Sovilla, a gregarious, Italian-born SLF scientist, puts it: “They are a mystery. Avalanches are beautiful, sexy and complex. I’m fascinated by them.”
Almost a hundred years ago, Thomas Mann made Davos famous as the setting for his novel The Magic Mountain. Today the town is more commonly known as the setting for the World Economic Forum, the annual global gathering of heavy hitters from finance, politics and business. Despite its reputation, and at 1,560m above sea level having the distinction of being one of Europe’s highest towns, it is an underwhelming place; largely bland and occasionally shabby.
The SLF is stationed on the outskirts of town in a sleek, three-floor office building clad in solar panelling on its southern face. Inside, the atmosphere is studious, with scientists and researchers working two or three to a room analysing data in offices that stretch down hushed corridors.
The SLF was founded in 1936. Its first building was a wooden barrack house located on the side of the Weissfluhjoch mountain that rises above Davos. The organisation remained there in gradually expanding quarters until 1996, when it moved down to its current base in the valley. “We were snowed in for nine months of the year,” says Charles Fierz, team leader at its snow and permafrost unit and an employee for 20 years. In the old location, he recalls, the only way to get to work in the morning was by cable car. “Sometimes, by the time you’d finished working, the cable car would be closed and you had to ski down.” That seems to have added a further thrill to the attraction of working in such an isolated environment. “I love mountains,” says Fierz. “So it was a sort of ideal location for me and many others.”
The elevated position of their site enabled scientists to set up a natural cold laboratory where they could study the characteristics of snow in temperatures that never rose above 0C. The organisation still maintains an outdoor test site on Weissfluhjoch, overseen by Fierz, who monitors an array of finely calibrated measurement devices that have been recording wind, temperature, snowfall and other data “on the same field in the same spot since 1936″. Being close to the mountains was also a powerful draw for Nicholas Dawes, co-ordinator of a multidisciplinary scientific project called the Swiss Experiment that deploys mobile sensing technology in the mountains to track natural hazards such as avalanches. Dawes, a lean, softly spoken Englishman, joined the SLF four years ago but his links to the Alps go back to early childhood. “I’ve got pictures of me in nappies in Davos,” he says. “My parents worked in the village next door during the winter seasons. My dad was a chef and my mum was a nanny.”
Dawes holds a master’s degree in applied digital-signal processing from Southampton University. He’s also an avid cross-country skier who competed at elite level for Great Britain in the 2012 Winter Triathlon European Cup, held in Oberstaufen, Germany. In the courtyard parking area outside the SLF building, he points at the racks of ski and snowboard equipment stacked there by staff. “People here spend a lot of time in the building. They work late because they’re interested in what they do. But they also love mountains and nature and doing things like skiing and ski touring.”
Dawes says this college-like atmosphere — part scholarly, part devoted to play — permeates Davos. As well as the SLF, the town is home to a number of other research institutes, including the world-leading Davos Solar Observatory and the Swiss Institute of Allergy and Asthma Research. “You see a lot of the same people from different institutes at sports clubs and events,” he says. “Everyone
knows each other and you end up hanging out together.”
Although the SLF’s core responsibilities of avalanche forecasting and snow research have remained largely unchanged for the past 76 years, the reach of its activities has expanded hugely. Today, it has a budget of 15 million Swiss francs (£9.9 million), more than half of which comes from its parent body, the Swiss Federal Institute for Forest, Snow and Landscape Research, with the rest coming from industry, academic funding and other sources. And the knowledge of SLF scientists is now exported round the world with staff advising the organisers of the Winter Olympics in Sochi, Russia, and teams tackling infrastructure and environmental projects in Chile, Iceland, Alaska and other locations. They are also conducting research on ski technology for winter-sports brands and a project to help Nestlé improve its ice cream.
The cornerstone of SLF efforts to study avalanche behaviour is the Vallée de la Sionne, a steep, narrow valley that lies on the eastern flank of the 2,700-metre-high Crêta Besse mountain. This is remote territory, deep in the Swiss Alps, in the Canton du Valais around 300km southwest of Davos. There are few signs of human presence. Fed with snow from Crêta Besse and its neighbouring peak, Pra Roua, the valley is particularly prone to avalanches. In 1998, the SLF arrived to establish what the organisation describes as a “full-scale avalanche dynamics test site”; essentially a place where researchers can watch snow come roaring down the mountain from the safety of a reinforced concrete bunker.
The SLF bunker is a two-storey structure that sits 50m above the opposite side of the valley floor from where the avalanches occur. Through its porthole windows observers can see the mark of previous avalanches on the slope above: deposits of scree and patches of mountainside bare of snow, the soil beneath exposed and gouged. It’s a demonstration of the power that’s released once a snowpack starts to slide.
“You have to see it to understand,” says Sovilla, who co-ordinates the SLF valley team, at her office in Davos. “You see the energy of this monster coming down. And, of course, they’re so huge. It’s overwhelming. They change continuously. They have structure. They create. They destroy. They are extremely alive.” An avalanche is defined as a mass of snow, ice and other materials, such as earth and rocks, that moves rapidly down a mountain slope. The phenomena are triggered when layers of snow that have settled on to a mountainside over days or weeks fail to bond securely together. A heavy fall of powder may set into a dense and icy layer of snow on the mountain. But if the layer beneath it is softer and more porous then it will be balancing on unstable foundations. It will only be a matter of time before the lower tier collapses and the upper layer sheers free to slide down the mountainside like a book sliding off a table.
A slab avalanche is what almost killed Johno Verity in New Zealand. They’re the most common and deadly form of snow slide and can reach speeds of 130kph within five seconds of forming. As it careers downhill it breaks up, uprooting trees and rocks, forming a mass of hundreds of cubic metres of snow, ice and debris, and sends up a cloud of powder as high as 100 metres. If you’re caught in the approaching surge the odds of survival are low. When an avalanche comes to a halt it sets like concrete, leaving anyone who’s trapped in a diminishing pocket of air. Ninety-three per cent of victims can be rescued alive if dug out in the first 15 minutes. The odds begin to drop precipitously from then on. After 45 minutes, only about 30 per cent of avalanche victims are still alive. Two thirds die from asphyxiation as the air supply around them turns to carbon dioxide. At two hours, the likelihood of a safe rescue will have fallen to ten per cent.
In order to observe avalanches at their full destructive strength, the valley at Sionne is rigged with more than 20 measurement technologies. Three pairs of Frequency Modulated Continuous Wave radars are buried beneath the ground at different points to determine soil erosion rates and flow heights of onrushing snow. And at 1,650 metres, three huge monitoring masts stand sentry. The largest of these is a 19-metre-high steel pylon laden with optical velocity sensors, impact pressure sensors, probes to measure density and other instruments to record air pressure and flow height.
During the winter, SLF scientists and engineers are placed on three-day warning in anticipation of an avalanche. When they get the word, airborne and ground teams scramble. A helicopter crew laser-scans the slopes to measure the pre-event distribution of the snowpack. In the lower level of the bunker, scientists huddle in quarters equipped with folding beds, a hot plate, food, water and red wine in case of a long wait. Computer screens hooked up to the measurement sensors track real-time data from the mountain. Above them, on the upper level, cameras stand ready to record, and doppler radars prepare to beam microwave signals up the slope and listen for their reflection to record velocity.
Yet, the reality of a full-scale snow slide can upend even the most meticulous preparations. The valley’s first year of operation coincided with the severe winter of 1999. The bunker, which stands in a supposedly safe position across the valley, was buried under six metres of snow. “There were five people in there at the time,” says Francois Dufour, the manager of the site. “We had four hours to rescue them before their air ran out. We broke ten shovels trying to dig them out. Eventually, we had to use an electric chainsaw to carve the snow.”
Despite such setbacks, Sovilla insists that findings from the valley have led to “an exponential increase in the quality of data” about avalanches. Among results gathered by her and her colleagues is a new understanding of the factors governing the mass and momentum of avalanche flows, a discovery that’s led to the overturning of models and theories in place since the 50s. “We understand a lot more about what’s going on,” says Sovilla.
Learning about avalanches has revealed the complexity of the variables that contribute to their structure and mechanical properties. Like earthquakes, avalanches are unknowable: you can warn of their imminence but there’s no guarantee they’ll turn up when predicted. The role of scientists in such circumstances is to try to describe their behaviour as closely as possible before they ever occur.
That means using digital technology to create simulations that map the effects of different terrain, temperature, wind and other conditions on sliding snow. Using data of the avalanche experiments in Vallée de la Sionne, SLF scientists developed 3D software to model the behaviour of natural hazards such as avalanches and debris flow. The algorithm, known as Rapid Mass Movements (RAMMS), simulates how an avalanche will move in a given terrain. Among other uses, RAMMS is deployed in risk management and hazard mapping for alpine villages or roads.
Before moving to Davos in 2008, Nicholas Dawes worked as an active sonar analyst for the UK Defence Science and Technology Laboratory. That turned out to be useful, if improbable, training for his role at the SLF, where he’s been creating a system to allow teams real-time access to research data from across the country. The Swiss Experiment platform is one of the developments of that system. It is a network of low-cost, high-resolution, wireless sensors that Dawes and his team have deployed across the Alps to supplement existing measurements of atmospheric and environmental conditions. The measured or calculated results can be mapped on to satellite imagery and cartographic maps for an even more accurate assessment. “This is higher resolution science,” says Dawes. “We’re bringing in technology to allow measurements and mapping at scales not previously viable.”
But spend time with the scientists in Davos and you soon start to understand that it’s not just avalanches that have complex behaviour patterns and structure — it’s the snow itself. The image of the six-pointed stellar flower is only part of the story. Snow crystals form when water vapour condenses directly into the air in the clouds. Most crystals start off as simple hexagonal prisms but their shape is determined by temperature and humidity, meaning that the same crystal can be a large six-sided plate at around -20C, a long thin needle at around -5C or our familiar star shape at -2C. They are entities in constant dynamic motion.
Martin Schneebeli has made the study of snow crystals his life’s work. The Swiss scientist is team leader of the SLF’s snow-physics group. “Snow is paradoxical,” says Schneebeli, 54. “You think of it as cold and inert. But from a scientific point of view snow is hot because it’s so close to the melting point. It’s extremely fast in changing and recrystallising.” He spends much of his time in the basement of the SLF headquarters in Davos, overseeing a dozen scientists, PhD students and technicians. Dressed in polar-grade jackets and snowsuits, they work in labs where temperatures can be adjusted to as low as -40°C. In these labs, his team is able to grow its own snow crystals and study them under conditions that simulate Sun, wind, temperature and other factors. To watch those processes as closely as possible, Schneebeli became, in 2001, the first person to use an X-ray computed tomography scanner, or microtomograph, for the purposes of snow science. More commonly employed in medicine to examine bone and body tissue, the machine is able to make readings down to a resolution of 20 micrometres. It has enabled Schneebeli to build up a 3D sequence of images revealing the evolution of snow structures in hitherto hidden detail.
In the winter of 2010, and again a year later, Schneebeli took part in expeditions to the Antarctic. Based out of the US-run McMurdo Station, he participated in a three-week trek to study layers of snowpack dating back thousands of years. “The Antarctic is a kind of desert,” he says. “Every year it only gets about 10cm of new snow. And because it’s so cold everything is preserved. Time moves slowly there.” As a consequence, digging into the snowpack is like opening a record book into the past. “Even at a depth of one metre you find snow that’s 20, 30, maybe 100 years old.” He and his expedition partners conducted experiments that included bombarding the snow with infrared light, to study grain size and density, and measuring its reflectivity with a spectrometer. Schneebeli also collected samples of polar ice and snow that he preserved at sub-zero temperatures and then had shipped back to Davos.
“The picture was completely different. What we’ve found is the snow is much more dynamic than we thought,” says Schneebeli. “We used to have a very static picture of snow structure and its mechanics. Now we’re looking at much less of a mystery and more of a physical process. And we can use those findings to calculate different properties. How elastic the snow is, how strong, how fast it recrystallises, what structures evolve.”
This kind of discovery means that experienced figures such as Schneebeli are constantly forced to question their beliefs. Bob Brown, a leading avalanche researcher, offers a perspective: “When I worked on the Apollo programme, I thought rocket science was the hardest form of physics. But snow science is even harder.”
SLF scientists haven’t found a way to predict when or where avalanches will strike, and the likelihood is they may never do so. But at least they’re making sure that, should Switzerland ever get hit by snowstorms as bad as 1999′s or 1951′s, the country will be as ready as it can be for the avalanches that will inevitably follow.