By JIM ROBBINS BOZEMAN, Montana - Not long ago, Ed Adams, a civil engineering professor, studied avalanches by setting them off with dynamite and studying their movement as they buried him, his instruments and his colleagues in a tiny shack.

Recently, though, Dr.Adams, a 58-yearold materials researcher, started a new and somewhat quieter phase of research, studying avalanches in the lab at Montana State University. A $2 million “cold lab”financed primarily by the National Science Foundation and the Murdock Charitable Trust and completed here in November allows Dr.Adams to replicate and control the uncontrollable field conditions of mountains in winter and understand in detail how snow behaves under widely varying conditions. The goal is to be better able to predict an avalanche.

“Snow seems simple, but it’s extraordinarily complex,”Dr.Adams said.“If I set a box of snow in the refrigerator and come back in an hour, it’s changed significantly. It’s almost always in a constant state of motion, and studying it is a moving target.”That is where the lab comes in, allowing researchers to vary the sky, sun and temperature to see how snow responds.

There have been 31 avalanche fatalities in North America this season through January 19, 16 in the United States and 15 in Canada. The record in the United States is 35 in the winter of 2001-02.

“The number of fatalities we have had shows they’re a difficult phenomenon for us to understand,”said Karl Birkeland, an avalanche scientist at the Forest Service’s National Avalanche Center here. “There’s definitely a need to better understand them.”

Montana State is well situated for the study of avalanches. There are four Class A avalanche zones? the most severe? at nearby skiing areas, and numerous backcountry locations for study.

For years, Dr.Adams and his colleagues set up their instruments in a small shack on a steep slope at Bridger Bowl, about 24 kilometers from the university, and sent another researcher up the slope to ignite a one-kilogram bomb that set off an avalanche.

As the wall of snow rumbled around or over the shack, Dr.Adams, bundled up against the cold, watched his laptop record information on velocity, depth, flow and temperature. He estimates he has survived dozens of such self-inflicted avalanches.

In the cold lab, however, where the temperature is 22 degrees below zero Celsius, the focus is on a one-square-meter panel, brilliantly lighted by an artificial sun and watched over by an icy artificial sky that can be widely varied to replicate different winter conditions. Wearing his puffy down jacket, wool hat and sunglasses, Dr.Adams shows how he can reproduce the wide range of conditions found on mountain slopes and create different types of snow.“We want to understand what conditions cause the change in the crystalline structure and the bonding between crystals,”he said.

Once he and his students and colleagues have created the snow crystals under certain conditions, they put them under the microscope to see what conditions made for the strongest or weakest layers.

The biggest cause of avalanches is a weak layer of snow on a slope covered by solid layers, Dr.Adams said.“The weak layers are faceted crystals, very smooth and unbonded to each other,”almost like ball bearings, he said. Strong layers have stronger bonds between crystals, which makes them more stable.

“It’s like a layer cake with very weak frosting,”Dr.Adams said. When something causes the weak layer, usually less than 2.5 centimeters thick, to give way, the strong layer or layers? there can be dozens, some of them a meter thick or more? go with it. Even skiing at low altitudes can fracture a weak layer and set off an avalanche far above. Contrary to conventional wisdom, sound, unless it is from an explosion, does not set off avalanches.

Based in the jagged mountains of the northern Rockies, the avalanche center at Montana State was founded by Charles Bradley and John Montagne, veterans of the Army’s Tenth Mountain Division who came here after World War II.

Other major avalanche centers include the Swiss Federal Institute for Snow and Avalanche Research in Davos, the world’s largest, and the Nagaoka Institute for Snow and Ice Studies in Japan.

The best way to survive an avalanche is still not clear. Some researchers say the most critical thing is to create a pocket in front of the face to breathe while waiting for rescue.

“I would swim, though,”Dr.Adams said.“Get prone in the snow and stay on top.”