Taking pretty pictures is only half of the story. The real puzzle is how a snowflake forms. Libbrecht reaches behind his desk, his face flushing with enthusiasm. At 45, he still looks like a farm boy. He grabs a small piece of stone, so silver it seems as if it were painted.
It's a crystal of silicon, he says, a gift from his wife, and the reason for society's preoccupation with crystals. Valued for their electrical properties, crystals are a multibillion-dollar enterprise, and although none of that money goes toward snowflake studies, Libbrecht hopes his work with snowflakes will one day help explain how crystals such as silicon are created.
Libbrecht leans back in his chair to describe the dynamics behind snowflake formation. "The basic principle to remember is this: Ice grows and water shrinks."
The physics professor you always wish you had, he holds up his hands, palms wide to show a cloud, then two fists -- one expanding like an ice molecule and the other contracting like a water molecule.
"Ice nucleates on the crevice of a speck of dust," he explains. "It then grows out and soon envelops the dust ... " A photograph of one flake, taken with blue and white lights, hangs on the wall above his cluttered desk-- the image fragile, aqueous and strangely indelible.
As Libbrecht describes the atomic intensity inside a winter storm, it's easy to feel a chill and imagine a soup of microscopic particles -- water, dust and air -- tumbling and turning around him in the air.
"At first," he continues, "the ice is a hexagonal prism, only about a 30th of the diameter of a human hair. Then, in most cases, the sides begin to grow, turning the prism into a thin, flat plate."
One snowflake is born. Then another. Then millions. A cold front churning at 4 degrees Fahrenheit shoulders its way beneath warm pockets of air and shoots them aloft, condensing more and more moisture off the ocean and the land. It clocks down the coast -- Seattle, Portland, Crescent City -- and skiers and snowboarders in Southern California start packing their cars.
"As the crystal grows, the plate branches and blossoms. It gets heavier and heavier and then it falls from the sky."
Libbrecht drops his hands. For the moment the room is still.
Across the hall from Libbrecht's office is the laboratory, where a 4-foot-high, 2-foot-diameter vacuum tank, wrapped in white insulation, sits on a raised platform. An engine hoist holds aloft its lid, from which curls a skein of electrical umbilicals that run to a pair of television screens.
Inside this tank, Libbrecht creates his own snowstorms and studies the growth velocity of snowflakes. So far he has learned that a flake's growth rate depends on temperature and the presence of other gases in the air, findings that have implications for skiers and climatologists alike.
There are those who might argue that dissecting a snowflake is for scientific spoilsports, akin to reading a sonnet into an oscilloscope.
Libbrecht responds to that slur with a quote from his Cal Tech protege, Richard Feynman: "It does not do harm to the mystery to know a little about it. For far more marvelous is the truth than any artists of the past imagined it."
