SNOWMIP2: An evaluation of forest snow process simulations

Abstract

Models of terrestrial snow cover, or snow modules within land surface models, are used in many meteorological, hydrological, and ecological applications. Such models were developed first, and have achieved their greatest sophistication, for snow in open areas; however, huge tracts of the Northern Hemisphere both have seasonal snow cover and are forested (Fig. 1). Forests have large influences on snow dynamics, and many snow models have been developed or modified in recent years to include vegetation canopies (e.g., Hellstrom 2000, Koivusalo and Kokkonen 2002; Niu and Yang 2004; Bartlett et al. 2006). Despite this, snow processes have been identified as an area of continuing weakness in global land surface models (Dirmeyer et al. 2006), and the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report concluded that "Large discrepancies remain in albedo for forested areas under snowy conditions, due to difficulties in determining the extent of masking of snow by vegetation" in climate models (Randall et al. 2007; Roesch 2006). This paper presents an overview of the results obtained in an intercomparison project evaluating the performance of a large number of models at several forested and open locations with seasonal snow cover.

Citation

Essery, Richard; Rutter, Nick; Pomeroy, John; Baxter, Robert; Stahli, Manfred; Gustafsson, David; Barr, Alan; Bartlett, Paul; Elder, Kelly. 2009. SNOWMIP2: An evaluation of forest snow process simulations. Bulletin of the American Meteorological Society. 90(8): 1120-1135.