Using dynamical downscaling to close the gap between global change scenarios and local permafrost dynamics

TitleUsing dynamical downscaling to close the gap between global change scenarios and local permafrost dynamics
Publication TypeJournal Article
Year of Publication2007
AuthorsStendel, M, Romanovsky, VE, Christensen, JH, Sazonova, T
JournalGlobal and Planetary Change
Keywordsactive layer, climate change, climate model, ground temperature, Permafrost, snow cover

Even though we can estimate the zonation of present-day permafrost from deep-soil temperatures obtained from global coupled atmosphere-ocean general circulation models (GCMs) by accounting for heat conduction in the frozen soil, it is impossible to explicitly resolve soil properties, vegetation cover and ice contents in great details. On the local scale, descriptions of the heterogeneous soil structure in the Arctic exist only for limited areas. Semi-empirical approaches, e.g. based on the Stefan [Stefan, J., 1891. Uber die Theorie der Eisbildung, insbesondere uber Eisbildung im Polarmeere. Ann. Phys. 42, 269-286] formula, give a more realistic depiction of permafrost temperatures and active layer thicknesses while at the same time avoiding problems inevitably associated with the explicit treatment of soil freezing and thawing. The coarse resolution of contemporary GCMs models that prevents a realistic description of soil characteristics, vegetation, and topography within a model grid box is the major limitation for use in permafrost modelling. We propose to narrow the gap between typical GCMs on one hand and local permafrost models on the other by introducing as an intermediate step a high resolution regional climate model (RCM) to downscale surface climate characteristics to a scale comparable to that of a detailed permafrost model. Forcing the permafrost model with RCM output results in a more realistic depiction of present-day mean annual ground temperature and active layer depth, in particular in mountainous regions. By using global climate change scenarios as driving fields, one can obtain permafrost dynamics in high temporal resolution on the order of years. For the 21st century under the IPCC SRES scenarios A2 and B2, we find an increase of mean annual ground temperature by up to 6 K and of active layer depth by up to 2 m within the East Siberian transect. According to these simulations, a significant part of the transect will suffer from permafrost degradation by the end of the century.