Snow albedo and its parameterization for natural systems and climate modeling

Abstract

The physical factors having influence on albedo of snow cover, as well as the main methods for its parameterization in models of natural systems, are considered. Numerous studies by various authors have shown that the most important characteristics determining the snow albedo in the near infrared range  (hereinafter referred to as NIR) is the size of snow grains and crystals, and in the visible and UV ranges –  the presence of impurities, primarily dust and soot. We have proposed the new scheme for parameterizing the albedo of snow cover, taking into account most of the processes and factors important for the metamorphism of snow and changes in its stratification and microstructure, namely: the influence of weather conditions during snowfall, its age, density and rate of background pollution, air temperature and solar radiation intensity, as well as the height of the Sun (angle of the Sun above the horizon). The proposed parameterization scheme is introduced into the LSM SPONSOR model. A new scheme for parameterizing snow albedo as part of the LSM SPONSOR model was tested using long-term observational data. Observational data were obtained for four ESM-SnowMIP project sites located in the mountainous regions of Europe and North America: Col-de Porte (France), Weissfluhjoch (Switzerland), Senator Beck and Swamp Angel (USA, Colorado). The series of observational data on the surface noon albedo are 20 years long for the first two sites, and 10 years long for the rest. When compared with the old scheme for parameterizing the albedo of snow cover in the LSM SPONSOR model, based on the dependence of the albedo only on the age of the snow, the new scheme showed a significant increase in the quality of albedo calculations: the correlation coefficients between the observed data and the calculation results are 0.78–0.83, which gives determination coefficients of 0.61–0.69. The new scheme makes it possible to obtain unbiased albedo estimates with statistical distribution characteristics that practically coincide with those obtained for observational data. The set of test sites covers the specific conditions of snow formation in the mountains, both in forested and treeless zones, so the scheme can be recommended for calculating albedo in a wide range of mountain landscapes. The quality of the scheme is also confirmed by the fact that the calculations were carried out with the same values of all model parameters  and coefficients for all four test sites located in different climatic conditions.