The Arctic Response to CO₂-Induced Warming in a Coupled Atmosphere-Ocean General Circulation Model

Abstract

A major development in climate research over recent years has been the establishment, at a number of modelling centres, of global coupled general circulation models (GCMs) of the climate system for studies of climate and climate change. These models are based on the dynamical and physical equations of their component systems, atmosphere, oceans, land surface, and cryosphere. Their basic variables, such as atmospheric temperature, humidity and wind, are represented on a grid of points (or their equivalent in spectral space) with a typical spacing of order 300km or so. To maintain numerical stability on such grids, it is necessary to specify a high viscosity in the ocean dynamics. As a consequence, the ocean models used for coupled climate studies at present do not simulate the magnitudes of the ocean current systems very well. Nevertheless, they do enable the first order effects of the thermal inertia of the oceans to be represented, which is an important factor in determining rates of climate change. A major concern for climate modelling is the representation of the many important physical processes which take place on scales smaller than the model grid. For example the transfers of heat and momentum associated with boundary layer turbulence, clouds and their interaction with solar and terrestrial (long wave) radiation, precipitation processes and the drag associated with breaking gravity waves. The approach used is to “parametrise” the grid square averaged effects of these processes in terms of the large scale basic model variables.