Modeling Issues of Snow and Land-Ice in a General Circulation Model

Snow cover is among the most spatially and temporally varying characteristics of the land surface, and has the ability to significantly affect both surface energy and water cycles. In addition, permanent land-ice constitutes a large portion of land-surface cover for Antarctica and Greenland. Recently, an assessment of snow modeling within land-surface schemes (LSSs) from the Project for the Intercomparison of Land-Surface Parameterization Schemes (PILPS) Phase 2(d) experiment demonstrated that a considerable disparity exists in the LSSs' simulation of snow for an 18-year period under identical atmospheric forcing. Moreover, the current treatment of land-ice in most general circulation models (GCMs) is rather simple and less understood. These issues raise a concern about the proper simulation of snow and land-ice variability in a GCM, as well as their effects on climate phenomenon, such as the Asian monsoon, and extra-tropical and high-latitude climate variability. The sensitivity of the treatment of snow and land-ice processes in a GCM is explored through a suite of experimental integrations conducted with the Center for Ocean-Land-Atmosphere Studies (COLA) GCM. These integrations aim to identify regions where the simulated climate's mean state and variability are sensitive to the treatment of snow and land-ice, as well as to provide insights for which processes are crucial toward the proper and efficient representation of snow and land-ice grid-points in a GCM.