General circulation models

While radiative-convective models are very useful tools for investigating how the atmospheric composition affects the radiation field, and in turn the role of the radiation field on atmospheric photochemistry, the fact that they have only one spatial dimensional (height) poses some fundamental limitations. The real earth is a three-dimensional distribution of atmospheric and surface properties that are linked by atmospheric dynamics and ocean circulation, whose accurate

Table 11.7 Climatic effects of changing global cloud cover, from a present value set at 0.50. Methane, carbon monoxide and hydrogen levels in ppmv. Water-vapour column Wh2o in g cm-2. Stratospheric peak water-vapour mixing ratio H2O St in ppmv. The planetary albedo is a.









































WH2 o








h2o St
















representation requires the inclusion of all three dimensions, in what has become known as a general circulation model (GCM). Because the physics of the atmosphere and of the ocean are quite different, we usually produce separate atmospheric and oceanic GCMs and link them together to form a coupled general circulation model (CGCM).

In order for 3D general circulation models to adequately treat the dynamics of the atmosphere, they have to have quite high resolution in space and time, which imposes an enormous computational burden. To compensate, they can include only a limited number of chemical species and reactions, while the radiation budget is treated through broadband parameterizations, that is they use only a limited number of spectral intervals to perform the radiative-transfer calculation. There is still scope, therefore, for radiative-convective models, since they allow a rapid sensitivity analysis of the effects of radiation transfer and chemistry on the mean global vertical temperature structure. The RC models can be linked with a photochemical model to examine in detail the spectral radiation effects on the vertical photochemistry and temperature structure, in a way that is beyond the scope of present-day GCMs. There is a need for both kinds of model, therefore, although we may expect that in time the two will merge. For the present, RC models lead when studying radiative or chemical processes, while GCMs are essential for most dynamical studies and, crucially, for long-term climate predictions.

Richardson (1922) reported one of the earliest attempts to simulate atmospheric motions mathematically knowing, however, that the complexity of modelling the global atmosphere required significant computational power. An early computer pioneer, von Neumann (see Aspray 1990, Charney et al. 1950), played a key role in the development of numerical weather prediction. The Royal Swedish Air Force Service was the first to begin real-time numerical weather forecasting in December 1954, using a model developed by the Institute of Meteorology at the University of Stockholm. By the 1960s, the computing power available was

Table 11.8 3D global and regional climate models used by the Hadley Centre, UK.





Atmosphere GCM

Coupled to land-surface and cryosphere,

used for atmospheric process studies



Ocean circulation and sea ice



Coupled atmospheric—oceanic circulation



Regional climate with boundaries from AOGCMs


Carbon cycle

Land—surface cycle from AOGCM,

marine cycle from OGCM


Atmospheric chemistry

Atmospheric composition changes

sufficient for a first attempt at modelling the global general circulation. By the mid- to late 1960s, the Geophysical Fluid Dynamics Laboratory at Princeton was established, and Smagorinsky, Manabe, and their collaborators (Manabe et al. 1965) had implemented a nine-level, hemispheric primitive-equation GCM and used it to perform carbon-dioxide doubling experiments (e.g. Manabe and Bryan, 1969). The mid-980s saw the first serious attempts at climate modelling that involved coupled atmospheric and oceanic circulation models.

As models and research groups have proliferated around the world, liaison between groups and comparisons of their products has become more important, and also more difficult. In 1990, the World Climate Research Programme established the Atmospheric Model Intercomparison Project, which lists details of the existing AGCMs and the centres worldwide that perform climate-simulation experiments with them, and establishes a common protocol for intercomparisons, to assist in this endeavour. The Intergovernmental Panel on Climate Change (IPCC) was established in 1988 by two United Nations organizations, the World Meteorological Organization (WMO) and the United Nations Environment Programme (UNEP) to assess the 'risk of human-induced climate change'. Its reports are widely cited in almost any debate related to climate change and generally regarded as authoritative, although obviously considerable disagreement remains, particularly about the evolution of such a complex system as the global climate.

11.4.1 Types of climate models

An example of the various models used by one leading climate-modelling institute is given in Table 11.8 that lists those used at the UK Meteorological Office Hadley Centre. The report by Roeckner et al. (2003) on the ECHAM5 model of the Max Planck Institute for Meteorology also gives full details of the complexity of GCM modelling and the relevant equations.

Due to the complexity of radiative transfer, numerous assumptions and simplifications are made within all GCMs. Generally, these include

1. The plane-parallel assumption, i.e. that columns of atmosphere and ocean, within a resolution element, have no curvature.

2. Gas constituents are homogeneously mixed, i.e. that there is no compositional variation (including cloud properties) or temperature contrast within each cell.

3. Gas constituents are in local thermodynamic equilibrium (LTE), which is generally true in the troposphere and lower stratosphere but not at higher levels.

4. The spectral resolution for both shortwave and longwave radiation transfer is broadband rather than detailed monochromatic.

5. For efficiency reasons the radiative-transfer computation in GCMs is performed less frequently than the dynamics and other parameterizations.

6. Shortwave scattering/absorption is usually parameterized by implementing either the delta-Eddington approximation (§3.5.6 and §6.7), or the two-stream approximation (§3.5.7), while longwave absorption/emission is parameterized by absorptance formulations (§4.5). Also, for infra-red and near-infra-red bands the correlated-fc approximation (§6.2.2) is used to convert the non-linear dependence of optical depth on absorber amount to a linear dependence.

Below, we consider some specific examples of the treatment adopted by some of the world's leading climate modelling groups. These illustrate the commonality and also the diversity of methods and assumptions within models.

11.4.2 Solar radiation transfer in GCMs HadAMS Scheme This is the atmospheric component of the HadCM3 coupled GCM that was used extensively by the IPCC in its evaluation of global climate predictions, as we discuss below. In HadAM3, the shortwave fluxes are calculated by a two-stream method, where the spectrum is divided into 6 bands with the wavelength limits: 0.20-0.32; 0.32-0.69 (2 bands); 0.69-1.19; 1.19-2.38; and 2.38-5.00 ¡m. Within each band, data are treated using the correlated-textitk approximation method using gaseous absorption data from the HITRAN and LOWTRAN databases.

Rayleigh scattering by gases is explicitly represented but scattering by clouds and aerosols are much more difficult to incorporate. The amount of background aerosol at any given time or location is unknown, so a calculation of a fixed amount of climatological background aerosol is incorporated as an interim approach. Cloud-radiative interactions are parameterized separately for water and ice as functions of water content and effective radius. Ice particles are treated as spheres with a fixed radius of 30 ¡m. Stratified clouds are treated as overlapping layers and convective clouds as vertical towers; mixed-phase clouds are represented by horizontally adjacent regions of ice and water cloud. NCAR CAM 3.0 Scheme This is a state-of-the-art model from the National Center for Atmospheric Research in the USA. In this, the shortwave scattering/absorption is parameterized by the delta-Eddington approximation applied in 19 discrete spectral intervals; 7 for O3, 1 for the visible, 7 for H2O, 3 for CO2, and 1 for the near-infra-red. The radiation scheme allows for gaseous absorption by O3, CO2, O2, and H2O. Molecular scattering and scattering/absorption by cloud droplets and aerosols are included. The model atmosphere consists of a discrete vertical set of horizontally homogeneous layers within which radiative heating rates are specified.

Each of these layers is considered to be a homogeneous combination of several radiatively active constituents. Five chemical species of aerosol are considered, including sea salt, soil dust, black and organic carbonaceous aerosols, sulphate, and volcanic sulphuric acid. Direct and semidirect effects of tropospheric aerosols on shortwave fluxes and heating rates are included. CAM 3.0 differentiates between the cloud drop effective radius for clouds diagnosed over maritime and continental regimes and over pristine surfaces (sea-ice, snow-covered land). The cloud parameterizations can treat random, maximum, or an arbitrary combination of overlap between clouds. ECHAM5 Scheme This is the 5th generation of the ECHAM general circulation model produced by the Max Planck Institute for Meteorology in Hamburg, Germany. Its scheme for calculating radiative transfer of solar energy within the atmosphere uses the Eddington approximation for the integration over the zenith and azimuth angles and the delta-Eddington approximation for the reflectivity of a layer. The scheme includes Rayleigh scattering, absorption by water vapor and ozone, both varying in space and time, and CO2, N2O, CO, CH4 and O2 as uniformly mixed gases. Aerosols and cloud particles are effective through absorption and scattering. Interactions between scattering processes and absorption are considered for water vapor and for the uniformly mixed gases, but not for ozone, because gaseous absorption rather than scattering is assumed to dominate in the stratosphere.

The computation of transmissivities and reflectivities across a vertical column is split into two separate calculations for the clear-sky and cloudy-sky parts. The scheme has 4 spectral bands; 1 for the UV-visible (0.20-0.69 ¡m) and 3 for the near-infra-red (0.69-1.19, 1.19-2.38, 2.38-4.0 ¡m). The near-infra-red range is resolved by 3 bands to account for the wavelength dependence of the optical properties of cloud particles or aerosols. The interaction of scattering processes and gaseous absorption is included for the near-infra-red bands, but neglected in the visible band. Single scattering properties of clouds use Mie calculations for idealized size distributions for both cloud droplets and ice crystals. For each of the 4 spectral intervals polynomial fits are derived in order to express the respective single scattering properties as functions of the effective radii, which are parameterized in terms of the liquid and ice water content.

Was this article helpful?

0 0
Guide to Alternative Fuels

Guide to Alternative Fuels

Your Alternative Fuel Solution for Saving Money, Reducing Oil Dependency, and Helping the Planet. Ethanol is an alternative to gasoline. The use of ethanol has been demonstrated to reduce greenhouse emissions slightly as compared to gasoline. Through this ebook, you are going to learn what you will need to know why choosing an alternative fuel may benefit you and your future.

Get My Free Ebook

Post a comment