William Wergin did not set out to be the king of snow. He always thought of himself as more of a botanist and a nematode guy, studying how parasitic soil-dwelling worms interact with crop plants. Then in December 1993, he and Eric Erbe, a colleague at the U.S. Department of Agriculture’s Electron Microscopy Unit in Beltsville, Maryland, started experimenting with a newly configured low-temperature scanning electron microscope.
The custom-built device keeps samples chilled to –320 degrees Fahrenheit, making it possible to flash-freeze nematodes or insects and then magnify them enormously to observe their behavior at a single moment in time. Wergin and Erbe couldn’t get their hands on a suitable agricultural sample, so they decided to try their new toy on some of the snowflakes falling outside.
“We had nothing else to image,” Wergin says. The two researchers collected flakes on a copper plate and brought the crystals indoors to their microscope. Fellow USDA researcher Al Rango, now at the department’s Jornada Experimental Range in Las Cruces, New Mexico, stopped by the lab and was stunned by the results. “I’d seen a lot of snow-crystal imagery, but I had never seen crystals this way before,” he says.
He realized the views could lead to better understanding of winter snowpacks. In time, other researchers came calling, clamoring for help with everything from studies of glacial ice structure to modeling carbon dioxide frost on Mars. Wergin found himself spending more of his spare hours learning about the intricacies of snow. In late 2000 he retired from the USDA, but he still divides his time between his old work and his frosty sideline.
Since that first glimpse of what the low-temperature scanning electron microscope can do, Erbe has developed procedures for harvesting and preserving snow crystals. The sequence goes like this: Get some copper sample plates, each about the size of a penny. Coat the surfaces with an adhesive such as Tissue-Tek, sticky gunk that biologists use to affix cells to microscope plates, and chill the plates to freezing. Let snow fall onto the plate or scrape some onto it and quickly plunge it into a Styrofoam tray filled with liquid nitrogen. Slide the specimens into a brass storage tube and put it in a thermoslike container in the lab in Beltsville, where cryogenic containers can keep crystals intact for a decade or longer. Finally, prep the crystals with a layer of platinum less than a millionth of an inch thick, which clarifies and intensifies the image produced by the microscope.
Wergin and Erbe have imaged tens of thousands of snow crystals from as nearby as the parking lot outside their Beltsville laboratory to as far as away as Alaska. On the pages that follow is a sampling of the staggering range of forms, including grape-shaped blobs, six-sided plates, shattered rods, and pockmarked stars. All illustrate the fantastically complex, protean nature of water.
Crystal shapes are as diverse as the environments that created them. Crystals that form in two temperature ranges — from 27°F to 32°F and from –13°F to 18°F — are shaped like hexagonal plates.Air closer to 32°F yields columns and needles. The familiar six-armed snow crystals, technically known asdendrites, grow in vapor-rich air between 3°F and 10°F. Detailed views of such crystals clarify the relationship between atmospheric conditions and the snow crystals they produce. Extreme magnification, well beyond what is shown here, reveals airborne particles trapped in the crystals, which could provide a new way of studying acid snow and air pollution.
Despite its fame, the six-armed crystal is just one of the many dozens of identified crystal forms. It became the archetypal image of snow largely because of the work of Wilson Bentley, a Vermont farmer who spent the winters between 1885 and 1931 outside his house photographing snow through a light microscope. Due to the low resolution of this technique, and also to Bentley’s aesthetic preferences, he mostly photographed the large, symmetrical star-shaped flakes—the ones that are now reproduced ad infinitum on greeting cards and in newspaper ads. The sixfold symmetry reflects the way that water molecules link up into a hexagonal lattice as they freeze, a molecular pattern blown up to macroscopic scale.
Deformed snowflakes contain a miniature history of weather conditions from the point of their formation to the place where they land. Wergin and Erbe have used their low-temperature scanning electron microscope to infer what happens when falling ice crystals run into fogs of supercooled water droplets on their way down, a common occurrence.
Some of the droplets attach to the falling crystals and freeze there like flattened little pimples, just thousandths of an inch in diameter. At first, local air currents cause these accretions to stick to the trailing face of the crystals as they descend through a region of cloud droplets. If the crystals receive only a moderate coating—so that they remain recognizable as needles, columns, plates, or dendrites—the combination of the ice crystals and their blobby accoutrements is known as rime.
During its passage through the atmosphere, one side of an ice crystal often gets thicker and more heavily rimed than the other. That side can become more elongated than the cleaner side. The coating changes the crystal’s aerodynamics, causing it to flip. Further riming unfolds on the new trailing edge, leading to a thicker coating with more elongated topography. If the crystal encounters additional water droplets on the way down, riming can continue to the point at which the entire crystal is enveloped in an icy armor. These reshaped particles take on a mossy form known graupel.
Transfigured crystals arise from changes that occur after snowflakes land. Fluctuating temperatures, pressures, air currents, and sunlight can resculpture a crystal, disintegrate it, or reconfigure it into an entirely different shape. Over a span of days to months, the hexagonal geometry of a freshly fallen crystal might repeatedly deform, partially melt, refreeze, and fuse with nearby crystals, resulting in a minuscule ice carnation. More extreme environmental effects yield a shape akin to a cluster of grapes. One type of metamorphosis is of particular interest because it can produce the conditions that enable avalanches to occur.
When the temperature beneath a layer of snow crystals is significantly higher than the temperature above, ice from crystals lower in the snowpack sublimes—that is, vaporizes directly without melting—and then refreezes onto overlying crystals. In time, this redistribution of mass leads to large and blocky crystals known as depth hoar. A layer of depth hoar tends to make the snowpack unstable. When safety managers in ski areas find such layers in snow pits dug during routine inspections, they issue warnings, close off vulnerable areas, and sometimes fire mortars into the snowpack to provoke an avalanche before it can catch anyone off guard.
The labyrinthine interiors of depth hoar crystals also cause problems for researchers like the U.S. Department of Agriculture’s Al Rango, who uses microwave-sensing satellites to measure the amount of water locked away in the winter snow cover. Tiny passageways within the crystals are great at scattering microwaves, thereby fooling satellite-based sensing systems into reading six inches of snow as six feet of snow. Low-temperature electron microscopy images of depth hoar are leading to better models for converting the raw satellite readings into accurate measurements of snowfall.
Among the thousands of snowflakes examined by Wergin and Erbe over the past decade are many unusual specimens that do not fit into any of the standard classifications. Often the bizarre forms result from the exotic ways that frozen water interacts with living things.
Ice worms: The surface of a glacier contains a willy-nilly assemblage of irregularly shaped ice grains joined together by necks of ice. Threadlike ice worms often make a home in the crevices between the grains. Erbe remembers searching all day for such worms on the South Cascade Glacier in Washington State. Finally, at quitting time, he noticed small moving black lines on the thermal pad he had been lying on. “They were attracted by the heat,” he says. This quarter-inch-long worm is typical of the specimens he found there.
Frost: On a night when the temperature dipped below the frost point, this blade of grass from Bearden Mountain, West Virginia, served as an organic post onto which ice crystals nucleated and grew. The result was a tightly clustered bunch of needle-shaped crystals.
Red snow: When snow repeatedly melts partially and then refreezes, the stage is set for a red-pigmented alga, Chlamydomonas nivalis, to take up residence in the thin films of water around the snow particles. This fractured and strangely eroded sample of summertime snow, collected at Loveland Pass in Colorado, contains several spherical algal cells, including two that have been split apart.
Red Planet snow: To explore the conditions that future Mars probes might encounter, Wergin, Erbe, and a group of NASA researchers created imitation Martian snow. They caused carbon dioxide gas to condense onto a sample plate cooled to a Mars-like –240°F. Images from the low-temperature scanning electron microscope show the resulting carbon dioxide frost consists largely of octahedral crystals, quite unlike those of water ice. “These crystals are believed to be similar to those in the seasonal polar caps of Mars,” Wergin says. The snows there contain a mix of water and carbon dioxide crystals. Such studies will help researchers make better models of the climate cycle on Mars.