The First Generation Coupled Global Climate Model

Canadian Centre for Climate Modelling and Analysis

The first version of the Canadian Centre for Climate Modelling and Analysis (CCCma) Coupled Global Climate Model, CGCM1, and its control climate are described by Flato et al. (2000). The details of the model and discussion of primary results may also be found in Climate Change Digest published by Environment Canada. The atmospheric component of the model is essentially The Second Generation Atmospheric General Circulation Model described by McFarlane et al. (1992). It is a spectral model with triangular truncation at wave number 32 (yielding a surface grid resolution of roughly 3.7°x3.7°) and 10 vertical levels. The ocean component is based on the GFDL MOM1.1 code and has a resolution of roughly 1.8°x1.8° and 29 vertical levels. The model uses heat and water flux adjustments obtained from uncoupled ocean and atmosphere model runs (of 10 years and 4000 years duration respectively), followed by an `adaption' procedure in which the flux adjustment fields are modified by a 14 year integration of the coupled model. A multi-century control simulation with the coupled model has been performed using the present-day CO2 concentration to evaluate the stability of the coupled model's climate, and to compare the modelled climate and its variability to that observed.

An ensemble of four transient climate change simulations has been performed and is described in Boer et al. (2000). Three of these simulations use an effective greenhouse gas forcing change corresponding to that observed from 1850 to 1990, and a forcing change corresponding to an increase of CO2 at a rate of 1% per year (compounded) thereafter until year 2100 (the Intergovernmental Panel on Climate Change "IS92a"forcing scenario). The direct forcing effect of sulphate aerosols is also included by increasing the surface albedo (as in Reader and Boer, 1998) based on loadings from the sulphur cycle model of Langner and Rodhe (1991). The fourth simulation considers the effect of greenhouse gas forcing only. The change in climate predicted by a model clearly depends directly on this specification of greenhouse gas (and aerosol) forcing, and of course these are not well known. The prescription described above is similar to the IPCC "business as usual" scenario, and using a standard scenario allows the results of this model to be compared to those of other modelling groups around the world. Some initial results from these simulations are presented below.

The ability of a climate model to reproduce the present-day mean climate and its historical variation adds confidence to projections of future climate change. A comparison of modelled and observed global, annual average surface air temperature anomaly from 1900 to 1990 is shown in Figure 1. The observations, from Jones (1994), do not completely cover the globe, particularly in the earlier portion of the record and so to make the comparison as consistent as possible, only those model values for which there was a corresponding observed value were used to construct the averages. Only one of the ensemble of transient runs is shown in this figure, but the overall agreement in temperature trend, and the magnitude of stochastic interannual variability, is the same for the others as well. In both the model and observations, the increase in global mean temperature over this century is roughly 0.6°C.

Figure 1: Modelled and observed global, annual average surface air temperature anomaly from 1900 to 1990. The observations are from Jones (1994)

Figure 1: Modelled and observed global, annual average surface air temperature anomaly from 1900 to 1990. The observations are from Jones (1994).

An expanded time series of modelled global, annual mean air temperature spanning the period 1900 to 2100 is shown in Figure 2. In this figure the results from three model simulations are included. The lowermost curve is the result of a control simulation with fixed greenhouse gas forcing. The small drift in this control run illustrates that the flux adjustment scheme is relatively successful. The middle curve is the transient calculation with increasing greenhouse gas and aerosol forcing, the first 100 years of which was shown in Figure 1. The uppermost curve is the result of an experiment in which the green house gas forcing evolves as in the previous case, but the aerosol amount is fixed at its 1900 level.

Figure 2: Modelled global, annual average surface air temperature from 1900 to 2100. Lower curve is from a control simulation with constant greenhouse gas and aerosol forcing. Middle curve is from a simulation with increasing greenhouse gas and aerosol forcing. Upper curve is from a simulation with greenhouse gas forcing only. See text for further details.

Figure 2: Modelled global, annual average surface air temperature from 1900 to 2100. Lower curve is from a control simulation with constant greenhouse gas and aerosol forcing. Middle curve is from a simulation with increasing greenhouse gas and aerosol forcing. Upper curve is from a simulation with greenhouse gas forcing only. See text for further details.

Between years 1980 and 2050 the prescribed CO2 concentration doubles, and over this time the greenhouse gas only run (the upper curve) exhibits an increase in temperature of 2.7°C. The increase over the same period in the greenhouse gas plus aerosol run is 1.9°C; the difference of 0.8°C is the cooling effect of the aerosols. One can contrast these results with the equilibrium calculation of Boer et al. (1992)who used the same atmospheric model without the aerosol effect. They obtained a global average warming of 3.5°C upon doubling CO2 concentration. The difference between the transient warming at the time of CO2 doubling, 2.7°C, and the equilibrium value of 3.5°C illustrates the ocean's role in sequestering heat. A more detailed study of the aerosol effect in equilibrium climate change calculations is provided by Reader and Boer (1998).

Figure 3: Difference in annual mean temperature between years 1971-1990 and years 2041-2060 of a transient model simulation with increasing greenhouse gas and aerosol forcing. The colour scale bar gives temperature ranges in celsius.

Figure 3: Difference in annual mean temperature between years 1971-1990 and years 2041-2060 of a transient model simulation with increasing greenhouse gas and aerosol forcing. The colour scale bar gives temperature ranges in °C.

Finally, the spatial pattern of warming obtained from one realization of the transient model with greenhouse gases and aerosols is illustrated in Figure 3 by the difference in temperature between 20 year periods centered on years 1980 and 2050 (again to illustrate the warming at the time of CO2 doubling). The overall pattern closely resembles that of the earlier equilibrium calculations (Boer et al., 1992) in that the warming is largest in the polar regions, and larger over land than over ocean. A notable difference is the hemispheric asymmetry in warming whereby the warming is larger in the northern hemisphere than in the southern hemisphere, in contrast to the equilibrium calculation where the warming is roughly equal in the northern and southern polar regions. This asymmetry in the fully-coupled model results from the more vigorous vertical mixing between the surface and deep waters of the Southern Ocean and is a general feature of fully-coupled simulations. Indeed, changes in vertical exchange at certain locations in both the Southern Ocean and the Labrador Sea actually lead to a slight local cooling of the surface temperatures, despite global average warming.

Global climate models are continuously being improved. Horizontal and vertical resolution are being increased as computing power advances, with many modelling groups taking advantage of the ability to split intensive calculations between multiple, parallel processors on a single computer (`multi-tasking'). Improvements in parameterizations of physical processes in all component models are being made. Some examples are more sophisticated land surface schemes such as CLASS (Verseghy et al., 1993), improved representation of sub-grid-scale mixing in the oceans, improved treatments of clouds, convec tion, radiation, and the indirect effect of aerosols in the atmosphere (e.g. Zhang and McFarlane, 1995), and the inclusion of sea ice dynamics (e.g. Flato and Hibler, 1992). Some of these improvements have been incorporated in the CCCma's second generation global climate model, The Second Generation Coupled Global Climate Model. In addition, the `technology' of coupling various model components is receiving considerable attention with new methods of `spinning up' the ocean model to equilibrium, reducing the `shock' the model components experience upon coupling, and minimizing or eliminating the requirement for flux adjustment.

Acknowledgements

The development of CGCM1 was a team effort involving G.J. Boer and G.M. Flato along with W.G. Lee, N.A. McFarlane, D. Ramsden, M.C. Reader, and A.J. Weaver.

References:

Boer, G.J., Flato, G.M., Reader, M.C., and Ramsden, D., 2000a: A transient climate change simulation with historical and projected greenhouse gas and aerosol forcing: experimental design and comparison with the instrumental record for the 20th century. Climate Dynamics, 16, 405-425.

Boer, G.J., Flato, G.M, and Ramsden, D., 2000b: A transient climate change simulation with historical and projected greenhouse gas and aerosol forcing: projected climate for the 21st century. Climate Dynamics, 16, 427-450.

Boer, G.J., N.A. McFarlane, and M. Lazare, 1992: Greenhouse Gas-induced Climate Change Simulated with the CCC Second-Generation General Circulation Model. J. Climate, 5, 1045-1077.

Flato, G.M., Boer, G.J., Lee, W.G., McFarlane, N.A., Ramsden, D., Reader, M.C., and Weaver, A.J., 2000: The Canadian Centre for Climate Modelling and Analysis Global Coupled Model and its Climate. Climate Dynamics, 16, 451-467.

Flato, G.M. and Hibler, W.D. III, 1992: Modelling Pack Ice as a Cavitating Fluid. J. Phys. Oceanogr., 22, 626-651.

Jones, P.D. (1994): Hemispheric Surface Air Temperature Variations: A Reanalysis and an Update to 1993. J. Climate, 7, 1794-1802.

McFarlane, N.A., G.J. Boer, J.-P. Blanchet, and M. Lazare, 1992: The Canadian Climate Centre Second-Generation General Circulation Model and Its Equilibrium Climate. J. Climate, 5, 1013-1044.

Reader, M.C., and Boer, G.J., 1998: The modification of greenhouse gas warming by the direct effect of sulphate aerosols. Clim. Dyn.,14, 593-607.

Verseghy, D.L., N.A. McFarlane, and M. Lazare, 1993: A Canadian Land Surface Scheme for GCMs:II. Vegetation model and coupled runs. Int. J. Climatol., 13, 347-370.

Zhang, G.J. and N. A. McFarlane, 1995: Sensitivity of climate simulations to the parameterization of cumulus convection in the CCC GCM. Atmos.-Ocean, 3, 407-446.