A TEST OF THE RADIATIVE ENERGY BALANCE OF THE SHAW MODEL FOR SNOWCOVER

Abstract

Snow and ice present interesting challenges to hydrologists. Simulating the radiative balance over snow, which is an important part of surface–atmosphere interactions, is particularly challenging because of the decay in albedo over time and the difficulty in estimating surface temperature and incoming long-wave radiation fluxes. Few models are available that include a comprehensive energy and water balance for cold season conditions. The simultaneous heat and water model (SHAW) is a detailed, physical process model of a vertical, one-dimensional canopy–snow–residue–soil system which integrates the detailed physics of heat and water transfer through a plant canopy, snow, residue and soil into one simultaneous solution. Detailed provisions for metamorphosis of the snowpack are included. The SHAW model was applied to data for one winter/spring season (November to May) on a ploughed field in Minnesota without prior calibration to test the performance of the radiation components. Maximum snow depth during this period was 30 cm. For the nearly 100 days of snowcover, the model accounted for 69% of the variation in net solar radiation, 66% of the variation in incoming long-wave radiation, 87% of the variation in emitted long-wave radiation, 26% of the variation in net long-wave radiation and 55% of the variation in net radiation balance. Mean absolute error in simulated values ranged from 10 W m−2 for emitted long-wave radiation to 27 W m−2 for the entire net radiation balance. Mean bias error ranged from 8 W m−2 for emitted long-wave radiation to −16 W m−2 for the entire net radiation balance. When the entire 170 days of simulation, which included periods without snowcover, were included in the analysis, the variation in observed values increased greatly. As a result, the variation in observed values accounted for by the model increased to 97, 71, 93, 56 and 94%, respectively, while the mean absolute and mean bias errors in simulated values remained nearly the same. Model modifications and parameter adjustments necessary to improve winter-time simulation were investigated. Simulation results suggest that the SHAW model may be a useful tool in simulating the interactive influences of radiative transfer at the surface–atmosphere interface.