Earthquake-induced avalanches

By Brad Carpenter Contributor

The sound of mice scurrying in the walls woke me. For a moment, I wondered what they were doing. Then the walls started moving.

As the employee housing building shook, my bed began to sway and bounce. I grabbed the sides and held on. Was it time to run for the door? Get in the closet?

Then just as suddenly, things settled back to normal.

The magnitude 7.1 earthquake that shook the Canterbury region of New Zealand on Sept. 4, 2010, was the start of over six terrifying months of seismic activity on the country’s South Island. The cycle peaked on Feb. 22, 2011, when a magnitude 6.3 earthquake claimed 181 lives and caused billions in damage to the Christchurch area, creating a ghost town of the downtown business district in a city of over 360,000 people. The aftershocks still continue to this day, with a magnitude 4.9 recorded in late August 2011.

The epicenter of the Sept. 4 tremor was near the Southern Alps, just 45 km from Porters Ski Area, in the Craigieburn Mountains. I’ve worked for four seasons on the ski patrol at Porters, and I’m now the assistant snow safety director there. While the quake caused very little damage to Porters infrastructure and no in-bounds avalanches, it caused an extensive earthquake-induced avalanche event in the surrounding mountains.

Craigieburn Range. Photo by Luke Armstrong:
Craigieburn Range _stitch_luke photo

Earthquake-triggered avalanche events are rare, so I was excited to have witnessed one. Only a couple dozen have ever been fully recorded, and the physics of these occurrences have seen little study. The aftermath of this particular cycle brought up several questions.

At Porters Ski Area, avalanche control work from several weeks prior had mitigated weak layers in the snowpack that otherwise could have been unstable; it made sense that no in-bounds slides had occurred. Just outside the boundaries, however, we saw multiple avalanches that had run thousands of feet from ridge line to valley floor. Nearby ski areas reported similar observations. We’d witnessed one of the largest natural avalanche cycles ever recorded in New Zealand.

As aftershocks continued to hit the region in the following few weeks, management at Porters began to ask an obvious, but challenging question of the snow safety department: Could we forecast for an earthquake-induced avalanche event, if we were to start seeing a more avalanche-prone snowpack? We weren’t sure, and I began to look around for any history of such an event. There didn’t seem to be anything.

Formula for an avalanche

Understanding avalanches and snowpack is crucial to understanding an earthquake-induced avalanche event. For an avalanche to occur, there must be certain layers within the snow:

A sliding layer, or bed surface—tends to be firmer snow

A weak layer on top of the bed surface—usually either weak, poorly bonded, sugary snow; or feathery surface hoar crystals, which are similar to dew, but in the winter. these are snow crystals formed by the accretion of water vapor to the surface of the snowpack and usually form with cool temperatures, clear skies, and very light winds

A slab—this a settled and cohesive layer of snow sitting above the weak layer

For the snow to slide downhill in an avalanche, there must be a trigger. In some circumstances, the weight of a single skier can trigger an avalanche. Other times, large explosives are necessary to affect the weak layers in a snowpack. The majority of avalanches that occur at ski areas are triggered on purpose, either by explosives or by ski cuts, in which the weight of a ski patroller is used to promote failure of a slab or new snow.

Luckily, most of the aftershocks were minor, and another earthquake induced avalanche cycle never occurred. But in October, as the snowpack melted, we found it had been affected in a very big way. Spider webs of long, deep, disconcerting cracks traversed most of the big faces in the backcountry, and even some of our in-bounds terrain.

These cracks were unlike anything I’d ever seen, and were evidence that the earthquake had affected our snowpack by essentially shattering it into pieces. It hadn’t caused any deeper layers in our snowpack to avalanche, because within our deeper, older snowpack there were no significant weak layers for snow to slide upon. With signs that the September earthquake had indeed disturbed our inbounds pack, I really started to wonder if we could actually forecast for earthquake-induced avalanche events.

During winter in the northern hemisphere, I’m the snow safety director at Moonlight Basin. Southwest Montana is also a highly active seismic area, so the earthquake-induced avalanche concept began to enter my thoughts. While we cannot forecast for seismological activity, could we, in theory, predict how a seismic event would affect our snowpack? This kind of avalanche forecasting is an entirely different game than what most avalanche practitioners are used to, so I kept looking for answers.

In a stroke of luck, two of the best, and only, scientific papers on earthquake-induced avalanches were published in 2010. In one, a group of Russian and Japanese scientists created a series of mini snowpack models in a cold lab and then simulated an earthquake. As simple as this sounds, it had never been done before. The goal was to better understand how the Earth’s vibrations affect snow and avalanches. The scientists determined that a magnitude 4 or 5 is necessary to cause avalanches, and it tends to affect an area within 20 to 40 miles of the seismic epicenter.

They also found that smaller earthquakes (as low as magnitude 1.9) can cause avalanches, and that even when the snow doesn’t avalanche, the layers of the snowpack almost always fracture into pieces and break apart—as evidenced during melt-out in New Zealand. Furthermore, they determined this type of avalanche can be triggered up to several dozen miles from an earthquake’s epicenter.

Carpenter doing avalanche control work at Porters Ski Area in New Zealand. Photo by Luke Armstrong

Earthquakes in Montana

Many of Southwest Montana’s beautiful mountain ranges were formed by hundreds of thousands of years of shifting faults and concurrent earthquakes. Today, hundreds of small tremors are recorded annually in Montana. Most of these earthquakes are too small for humans to detect, but are monitored by seismographs.

In the 20th century several major tremors have hit this part of Montana. In 1925, a magnitude 6.7 earthquake in the Gallatin Valley caused extensive damage to unreinforced masonry buildings in Manhattan, Logan, Three Forks and Lombard. 10 years later, a series of severe earthquakes struck the Helena area causing four deaths and millions in property damage.

The magnitude 7.5 that occurred in 1959 north of West Yellowstone was the largest earthquake recorded in Montana history and the most well known. The subsequent landslide killed 22 people and dammed the upper Madison River, creating Quake Lake.

Although the potential for earthquake‐triggered avalanches in Southwest Montana is significant, some key variables would need to align. First, seismic activity would have to occur between November and May, when snow is present in the mountains. And second, it would have to happen close enough to avalanche terrain to affect a weak layer in the snowpack.

Seismically speaking, areas closer to Yellowstone National Park and south of Bozeman might see more earthquake activity than those further afield. Popular backcountry ski touring areas like Bacon Rind, Hebgen Lake, the Lionshead region, and Cooke City could be more in the firing line of earthquake-induced avalanches.

To truly predict avalanches caused by earthquakes, we would need to be able to predict actual earthquakes, which so far has proved impossible for scientists. While we cannot predict seismic activity, we’ve seen that a significant earthquake-induced avalanche cycle like the one in Canterbury, New Zealand, can be planned for, and is another factor that could be considered in the avalanche prediction equation.

Does this mean that a seismometer is the next piece of forecasting equipment for traveling or working in avalanche terrain? Not likely, but knowing earthquake-induced avalanches are possible in Southwest Montana will only improve our knowledge and preparation for them.

Brad Carpenter really likes to ski and spends most of his time doing so. This story was first published in the Winter 2011/12 issue of Mountain Outlaw magazine.