CLASSIC biweekly meeting: Snow cover heterogeneity and its impact on the Climate and Carbon cycle of Arctic regions

Abstract

This talk describes the preliminary work intended to serve as a basis for the development of a multilayer snow model adapted for the Arctic region in the Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC). Ten sites – including three Arctic sites – have been set to run the model in 1D simulations to assess the skills of the snow model’s physics. Three model configurations were used: (1) an older model version (CLASS v3.6) used in the latest Snow Model Intercomparison Project (SnowMIP), and (2, 3) the latest model version (CLASSIC v1.0) with a windless exchange coefficient activated or not in the calculation of the sensible heat fluxes. The new and old model versions demonstrate similar skills in simulating the snowpack at most of the sites, although the new version performs slightly better in simulating the albedo, soil temperatures, and snow cover duration. Activating the windless exchange coefficient allows a simulated snow surface temperature cold bias of about -2 °C to be alleviated but at the detriment of inducing an underestimation of the simulated snow depth and snow water equivalent (SWE) of about 10 cm and 25 to 50 kg m-2, respectively. Overall, the model demonstrates good skills in simulating the snowpack at most of the mid-latitude sites, with, however, a tendency to overestimate albedo at the end of the snow season, likely due to the non-consideration of the deposition of the light-absorbing particles on snow. At the Arctic sites, the model exhibits larger biases, especially in simulating the soil temperatures. Inherent uncertainties in snow modeling pose significant challenges in evaluating models, particularly in locations where snow distribution is influenced by blowing snow transport. Future work will consist of implementing a multilayer snowpack within CLASSIC, including consideration of blowing snow sublimation losses and adaptation to Arctic snowpack characteristics.

comments powered by Disqus