SnowMIP2 Snow Model Intercomparison Project for forest snow processes

Motivation for a forest snow process model intercomparison

Boreal forests and seasonal snow cover large areas of the Northern Hemisphere land surface. Snow falling on forest canopies is partitioned into interception by the canopy and throughfall to the forest floor. Intercepted snow has a large exposed surface area and can be subject to high sublimation rates in dry continental environments, but it can also be removed from the canopy by direct unloading and dripping of melt water. Snow on the ground beneath the canopy, in contrast with snow in open areas, is sheltered from wind and solar radiation but receives increased thermal radiation from warm canopies. Forest canopies strongly modify interactions between snow-covered surfaces and the atmosphere, and boreal forests have thus been found to have important influences on weather and climate on hemispheric scales. Changes in sublimation and melt rates influence the quantity and timing of runoff from forested catchments, and radiation absorbed and emitted by forest canopies complicates the remote sensing of snow.

Because of their importance in hydrological modelling, numerical weather forecasting, climate prediction and remote-sensing, many land-surface models contain at least some representation of canopy snow processes, but these models have yet to be evaluated in a wide range of forested environments. SnowMIP1 showed that models can produce widely differing results even for the comparatively simple case of simulating snow accumulation and ablation at sites without exposed vegetation, and equally good snow mass simulations from different models can partition the surface energy balance very differently. These problems are likely to be even more significant in SnowMIP2, with additional uncertainty arising from energy and mass partitioning in the canopy and at the ground surface. To evaluate and interpret the performance of models, SnowMIP2 simulations will be compared with observations of a range of variables, including snow mass at paired forested and clear sites, surface temperatures, and radiative and turbulent fluxes above and below canopies.