A SEASONAL SNOW COVER CLASSIFICATION-SYSTEM FOR LOCAL TO GLOBAL APPLICATIONS

A new classification system for seasonal snow covers is proposed. It h as six classes (tundra, taiga, alpine, maritime, prairie, and ephemera l), each class defined by a unique ensemble of textural and stratigrap hic characteristics including the sequence of snow layers, their thick ness, density, and the crystal morphology and grain characteristics wi thin each layer. The classes can also be derived using a binary system of three climate variables: wind, precipitation, and air temperature. Using this classification system, the Northern Hemisphere distributio n of the snow cover classes is mapped on a 0.5 degrees lat X 0.5 degre es long grid. These maps are compared to maps prepared from snow cover data collected in the former Soviet Union and Alaska. For these areas where both climatologically based and texturally based snow cover map s are available, there is 62% and 90% agreement, respectively. Five of the six snow classes are found in Alaska. From 1989 through 1992, hou rly measurements, consisting of 40 thermal and physical parameters, in cluding snow depth, the temperature distribution in the snow, and basa l heat flow, were made on four of these classes. In addition, snow str atigraphy and texture were measured every six weeks. Factor analysis i ndicates that the snow classes can be readily discriminated using four or more winter average thermal or physical parameters. Further, analy sis of hourly time series indicates that 84% of the time, spot measure ments of the parameters are sufficient to correctly differentiate the snow cover class. Using the new snow classification system, 1) classes can readily be distinguished using observations of simple thermal par ameters, 2) physical and thermal attributes of the snow can be inferre d, and 3) classes can be mapped from climate data for use in regional and global climate modeling.