In the basement of CRREL headquarters, Sam Colbeck, senior research scientist, scurries through a catacomb of cold lab rooms. A slightly rumpled white-haired physicist dressed in an olive-drab military-issue parka, Colbeck, 59, is a former petroleum engineer who decided to devote himself to the study of frozen water molecules; he earned a Ph.D. in geophysics, with a focus on snow and ice, from the University of Washington in 1970, and has been with CRREL ever since.
Nearly everyone in the avalanche business speaks Colbeck's language: In 1990, he was instrumental in establishing the standard nomenclature now used to assess snow. As Don Bachman, executive director of the American Association of Avalanche Professionals, says, "We're standing on Sam's shoulders when we get into studies of snow."
But even Colbeck, an avid skier who occasionally teaches avalanche-awareness courses, says that it is "horrendously difficult" to make the leap from understanding an ice crystal to predicting what a snow slope will do when a person steps onto it. Part of the problem of linking science with praxis is that Colbeck himself recently found that one of the basic assumptions of snow science was wrong. He discovered that the shape of snow crystals, which forecasters rely on to make their hazard assessments, is not the best indicator of the strength of the snowpack. Instead, he says, it is the way that crystals bond to one another that matters. Surprisingly, until Colbeck decided to study bonds under his microscope, no one had ever accurately described the process. Colbeck filled a void: He found that snow crystals bond by way of a distinct "grain boundary groove" that forms an obtuse 145-degree angle, not a smooth surface, as scientists previously had thought.
In general terms, Colbeck explains, big, round crystals bond strongly together and are less likely to avalanche, whereas sharp, angular crystals don't bond well and are therefore more likely to create a weak layer of snow that will fail. While the shape of the crystal will still tell a forecaster something about how the snow is bonding, Colbeck is experimenting with ways for those in the field to look directly at the strength of the bond itself. He admits that his revelation is a small piece in the much larger puzzle of why a snow slope avalanches, but he believes that forecasters examining these bonds under a magnifying lens when they dig their snow pits will gain a more precise grasp of why and when avalanches occur.
Dan Howlett, who heard Colbeck present his discovery at the biennial International Snow Science Workshop in Sunriver, Oregon, in 1998, was intrigued by the insights. "That work may help us to understand deep slab instability," he says, but the pure science does not immediately translate into changes in his tried-and-true avalanche control techniques. Colbeck is changing the way experts think, but he hasn't yet changed the way they work.
The contradiction doesn't bother Colbeck. "They're damn good at what they do," he says, admitting that guides and forecasters, though not always scientifically rigorous, have a good track record of predicting high avalanche hazard. "Do they have accidents? Yeah," he says. "It's an incredibly dangerous business. But we take risks. That's part of the thrill of skiing and climbing. If you take away the element of danger, you'd take away some of the pleasure."
Lately, Colbeck has been worried about how avalanches are changing in response to the way human beings are transforming the environment through pollution and development. "As the high alpine environment changes, the nature of avalanches changes," he says. He raises the specter that avalanches have become like drug-resistant viruses, mutating to overcome new obstacles that are put in their way. "Pollution," Colbeck notes, "can destroy vegetation in high alpine environments. If you destroy trees in an alpine environment, you are removing natural avalanche defense mechanisms in potential avalanche starting zones. Then you are creating starting zones where they haven't existed in modern history."
"We are doing what the Europeans did many years ago, expanding and developing into the mountains," he says. "As we do more of that, we will have more avalanche deaths."