The presence of snow and ice adds complexity to surface energy and water balance calculations due to changes in surface albedo and roughness and the energies involved in phase changes and heat transfer within the snow/soil profile (Slater et al., 1998a; Viterbo et al., 1999). The parametrizations of snow processes have received significant attention since the SAR and more complex snow schemes are now used in some climate models (Loth et al., 1993; Verseghy et al., 1993; Lynch-Stieglitz, 1994). These models include advanced albedo calculations based on snow age or temperature and may explicitly model the metamorphism of snow as well as representing liquid water storage and wind-blown snow. Douville et al. (1995), Yang et al. (1997), Loth et al. (1993) and Slater et al. (1998b) examined the ability of snow modules within specific land-surface schemes to simulate snow cover. An offline evaluation of many schemes by Schlosser et al. (2000) focused on how successfully current land-surface schemes simulated snow over an 18 year period; they found considerable scatter in the simulation of snow and no evidence that the ability to simulate cold climate hydrology was related to scheme complexity.
Permafrost, defined as any soil/rock material that remains frozen throughout two or more consecutive years, underlies almost 25% of the exposed land surface in the Northern Hemisphere (Zhang et al., 1999b). The uppermost layer of ground above permafrost, which experiences seasonal thawing, is called the active layer. The most distinct feature of land-atmosphere interactions in permafrost regions is that mass exchange is usually limited to this relatively shallow active layer, with complex transfers of heat by conduction and percolation across the ice/water interface. Recent modelling studies indicate that by the middle of the 21st century, climatic warming may result in a 12 to 15% reduction of the near-surface permafrost area and a 15 to 30% increase of the active layer thickness (Anisimov and Nelson, 1996, 1997; Anisimov et al., 1997). Because of the latent heat involved, thawing of ice-rich permafrost under the changing climatic conditions will be slow, while the reaction of the active layer will be very fast.
There are two major longer term feedbacks between climate and permafrost: release of greenhouse gases from thawing permafrost (Goulden et al., 1998; see also Chapter 4, Section 4.2) and changes in the vegetation associated with the thickening of the active layer. The first contributes directly to the global radiative forcing, while the second alters parameters of the radiation balance and surface hydrology.
Other reports in this collection