JOHN E. WALSH1
In terms of its interactions with the atmosphere, snow cover has received the most attention through the albedo-temperature associations that have been detected over daily to monthly time scales. For example, snow cover has been shown to contribute, albeit modestly, to the interannual variability of monthly surface air temperatures over land. On decadal and longer time scales, the more important role of snow cover may be its link with the hydrologic cycle over land. Changes in winter precipitation and/or temperature can, through snow cover, result in substantial temporal and spatial changes of soil moisture and runoff to the oceans. Model results are rather consistent in depicting an alteration of the seasonal cycle of temperature when the timing and amount of spring snow melt are modified. Runoff-induced effects on the ocean stratification in the middle and high latitudes have also been hypothesized, although such arguments need to be substantiated by observational data. Greenhouse-induced changes of high-latitude precipitation, especially snowfall, play significant roles in the scenarios of temperature change projected by global climate models. Data on observed snow cover and temperature variations of recent decades are, in some respects, consistent with the projections of the global models.
Snow cover has long been regarded as an indicator and as a possible agent of climate variability over a range of time scales. Over periods of several days to several weeks, the largest changes in the earth's surface properties result from variations in snow cover on land and on sea ice. Snow cover clearly influences the local values of near-surface atmospheric variables over these time scales. Over time scales of a thousand years or longer, the advance and retreat of the ice sheets depend ultimately on changes in the rates at which snow accumulates and ablates over the continents. However, over the interannual-to-century time scales on which this workshop will focus, the climatic roles of snow cover are poorly understood. Aside from the areal snow coverage of the past 15 to 20 years, the variability of snow cover over decadal time scales is not well documented. Yet model results are sufficiently suggestive of climatic roles of snow cover that one cannot ignore snow cover in projections of climate change over decade-to-century time scales.