Skip to content

Snowflake Reserach

picturethis

 

More accurate storm forecasting begins with a single snowflake

The study of snow is nothing new. Avalanche forecasters have been analyzing the intricate crystals for decades and modern atmospheric scientists have been tracking the frozen precipitation for just as long. But it’s a dynamic and temperamental medium that changes so quickly that even the most accurate weather model isn’t always that accurate.

 

Case in point: “We’ve all experienced times when the weather forecast has made really erroneous predictions of how much snow will fall on the ground. This causes problems,” says Tim Garrett, professor of atmospheric science at the University of Utah. “Skiers may have their hopes up that they were expecting 12 inches at Alta, but turns out there was only two or three. Or alternatively the reverse happens, and the city is caught unaware in a very bad snowstorm. They’re not given sufficient warning, and the snowfall proves to be much more serious for drivers than anticipated. Weather models are struggling with this issue.”
So, armed with a custom-made, high-powered camera – known simply as the Multi Angle Snow Camera (MASC) – and a strategic partnership with Alta Ski Area, Garrett and a team of researchers set out to change the way meteorologists forecast winter weather and help avalanche pros assess snowpack stability. By logging images of snowflakes and relating them to local meteorology, the goal is to ultimately improve the accuracy of weather forecasts and snowpack conditions. Because while the technologies to track storms have greatly improved over the years, the formulas to predict them haven’t changed.

 

“Meteorologists are getting these forecasts wrong because they’re using formulas for predicting snowflake shape and fall speed that are based on measurements from the 1970s,” explains Garrett. “What’s been holding them back is there hasn’t been a formula available to make more accurate predictions.”

 

You see, today’s methods of snow forecasting are rooted in research that dates back to 1973, when a team of scientists from the University of Washington photographed and categorized snowflakes by hand in the state’s Northern Cascade Range. All together, the researchers collected a few hundred varieties of flakes, an impressive total given the painstakingly slow and laborious process. And while this research paved the way for modern snow forecasting as we know, the study was inherently riddled with damaged, flawed and inaccurate samples thanks to the delicate nature of snowflakes.

 

Nevertheless, the results gave meteorologists across the country a new platform for predicting winter storms, but given the faulty data, it also set the stage for four decades of less-than-accurate predictions. Additionally, the microwave rays transmitted in modern-day satellite and Doppler radar imaging are obstructed and scrambled when it snows, making it even more difficult to read and predict incoming storms.

 

Garrett, Shkurau and Fallgatter in their U of U lab

Garrett, Shkurau and Fallgatter in their U of U lab

And that, explains Garrett, is what he’s aiming to improve. Especially with regards to backcountry users, “The camera is helping to build a better system that relates weather to avalanche conditions. It’s building a deeper understanding of why these avalanches are happening because of a certain type of snowflake that fell at a certain time.”

- – - – - – - – -
The MASCteam is loosely comprised of five members: Garrett, who spearheaded the initial idea; Cale Fallgatter, the engineer and designer behind the camera; Konstantin Shkurao, a University of Utah Ph.D student tasked with developing the data collection system; Howie Howlett, assistant avalanche director at Alta; and Sandra Yuter, a professor of atmospheric science at North Carolina State University who shares her Alta-based radar weather system with the team. Given the existing snow safety infrastructure and massive quantities of annual snow at Alta, it was a natural fit for these snow enthusiasts to partner with the Little Cottonwood resort. “Alta has been a supporter of snow science for decades,” says Garrett. “Initial studies of snowflakes types were done at Alta back in the 50s and 60s. Being able to team up with them is a privilege.”


After the roughly three years that it took to obtain funding and build a prototype, the MASC made its official debut at the High Alpine Research Laboratory for Diversity in Snow (HARoLDS) in Alta’s Collins Gulch area on April 30, 2011, where it has since spent the winters snapping images of snowflakes in free-fall. Today, two separate cameras at Alta are equipped with three high-speed digital cameras and three LED flashes. The instrument has allowed the team to get a sense of snowflakes in 3D form, providing a much more accurate visual than a two-dimensional photo. And with a resolution 10 times finer than a human hair, even flakes measuring a whopping 100 micrometers long are detailed to a stunning degree.

 

The camera shoots on standby 24 hours a day, automatically triggered by a sensor that detects just a few falling flakes of snow. During any given storm, the camera collects images every half-second with a speedy shutter speed of 1/40,000. “Even a bad year at Alta means millions of snowflake pictures for me. There’s no shortage of snow to take pictures of,” he says.

 

But photographing the flakes is more than just a high-tech art project. Garrett’s team can cull specific data from a single photo, which gives insight into a flake’s aerodynamics. Statistics like fall speed, size, shape, orientation and aspect ratio can be used to determine what’s happening inside the storm cycle as a whole. And this is a key factor in making accurate weather forecasts.

Understandably, it’s a massive amount of data to sort through. If it’s true that no two snowflakes are alike, then the three million photos and accompanying data that Garrett collected last season are downright staggering. “I’ve been tearing my hair out restructuring the data,” he says, audibly overwhelmed with the amount of information.

Snowflakes2

 

- – - – - – - – -


Snowflakes
captured by MASC are, of course, specific to Utah climate, but thanks to all sorts of weather variables that occur through the season, the team has been able to collect and document an infinite range of snow patterns. From an aesthetic point of view, Garrett has discovered “some of the most bizarre, fantastical shapes you can imagine.”

 

(Surprisingly, very few flakes are the iconic, six-sided snowflake, known as a stellar dendrite, that represents winter’s unofficial logo. “The shapes that we’re familiar with are ridiculously rare,” he says. “I have to sift through thousands of photos to find one that looks like that. If you look at what falls on your jacket sleeve, it’s not a super simple shape, but we’ve clung onto this perception that snowflakes look like this.”)

 

But can the snow found in one specific climate illustrate what’s happening with snow found in different regions and climates? “I’ve heard that criticism before, and I think it’s unfair,” says Garrett. “We’ve gathered long-term measurements [at Alta] and have gotten a wide range of what’s out there,” further explaining that certain weather patterns create specific types of snow, regardless of location. “Skiers know the snow at Mammoth is different than the snow at Alta, but I suspect that there will be strong similarities across the board.”

 

His hypothesis will be put to the test this season as three organizations have commissioned Garrett’s team to build cameras for their own projects: one to the U.S. Army to be used in avalanche study at Mammoth Mountain, one to Vanderbilt University for a precipitation study in Greenland and one to the U.S. Department of Energy for a weather study in Alaska. “[The projects] seems like such a silly, little thing,” says Garrett, “but it has quite the impact on how people forecast large-scale storms.”

 

The professor admits that his interest in this project has less to do with skiing (“Actually,” he claims, “I’m a pretty terrible skier.”), and more to do with his love of science. “I think that snowflakes are a wonderful way to explore physics,” he says. “That’s my main driving interest. It’s immediately communicable to the general public. They’re beautiful, intricate and people are fascinated them. They present really basic physics questions that are widely interesting to a large number of people in a large number of fields.”

Snowflakes3

For example? “The department of transportation is very concerned about snowflakes,” explains Garrett. The quantity of snowflakes determines how many plow trucks the DOT needs to deploy on the freeways, and the quality of snowflakes play a role in determining the severity of winter road conditions. Just look at graupel pellets. The heavy, ball bearing traits of these frozen water droplets are the culprit for extra-slick driving conditions and result in a higher rate of car accidents than other types of snow. “It sounds overly dramatic, but snowflakes kill. There’s practical interest in them, and there’s a tangible benefit to revealing these snowflake shapes to the public.”

 

Those benefits couldn’t be any more gratifying for winter warriors. For starters, when you cash in a sick day in advance of the Storm of the Year, you’ll never be disappointed by overhyped weather reports because they’re down-to-the-inch accurate. Or when you’re melting a fresh coat of wax on the board, you’ll know to use highly glideable fluorocarbon wax because those rare, six-sided stellar dendrites that indeed fell last night are sharp and pointy and will otherwise slow you down thanks to a phenomenon called static friction – the scientific culprit for your sticky problem.

 

Or imagine, hypothetically, that these images become a direct line of resource for avalanche forecasters, ski patrollers and backcountry guides who have relied on subjective crystal cards for years – virtually the same inconsistent technique that the UW scientists used back in 1973. The factual data could more accurately pinpoint the weak layers in the snowpack or give insight into the rapid process of riming. Of course, the photos don’t take into account the variable weather conditions that can morph a snowflake’s characteristics once it’s on the ground, but the general principle is a valuable tool in the hands of a snow safety inspector.

 

For the rest of us, perhaps it’s just a matter of enjoying the simplicities of winter and watching artfully intricate snow crystals fall from the sky. With the MASC returning to Alta again this season, new pictures of snowflakes will be streamed through a live feed on the ski area’s website. Garrett mentions that there’s even talk of developing a smart phone app, which needs additional funding, that “lets people explore the information that goes along with the snowflakes. A user could look at the snowflake from multiple angles and see how fast it was falling and be able to relate these snowflakes to local meteorological conditions.”

 

To view the MASC’s live feed this winter or to learn more about Tim Garrett’s study, visit the Snowflake Showcase at alta.com.

 

 

Last Updated: 4/27/21