Earth Cubed

Distributed Climate Science and Computing

A Comparison of GCMs and Some History (Two Links)

I found two links which seem to be a good place to start with understanding the differences between the current GCM, and the history of GCM.

The GCM-Reality Intercomparison
Project for SPARC (GRIPS):
Scientific Issues and Initial Results
S. Pawson,a,b,c K. Kodera,d K. Hamilton,e T. G. Shepherd,f S. R. Beagley,g B. A. Boville,h
J. D. Farrara,i T. D. A. Fairlie,j A. Kitoh,d W. A. Lahoz,k U. Langematz,c E. Manzini,l
D. H. Rind,m A. A. Scaife,n K. Shibata,e P. Simon,o R. Swinbank,n L. Takacs,p
R. J. Wilson,e J. A. Al-Saadi,q M. Amodei,o M. Chiba,r L. Coy,p J. de Grandpré,g
R. S. Eckman,q M. Fiorino,s,t W. L. Grose,q H. Koide,d J. N. Koshyk,f D. Li,k,n
J. Lerner,m J. D. Mahlman,e N. A. McFarlane,u C. R. Mechoso,i A. Molod,p
A. O’Neill,k R. B. Pierce,q W. J. Randel,h R. B. Rood,b and F. Wuh
To investigate the effects of the middle atmosphere on climate, the World Climate Research Programme is supporting the project “Stratospheric Processes and their Role in Climate” (SPARC). A  entral theme of SPARC, to examine model simulations of the coupled troposphere–middle atmosphere system, is being  erformed through the initiative called GRIPS (GCM-Reality Intercomparison Project for SPARC). In this paper, an overview of the objectives of GRIPS is given. Initial activities include an assessment of the performance of middle atmosphere climate models, and preliminary results from this evaluation are presented here. It is shown that although all 13 models evaluated represent most major features of the mean atmospheric state, there are deficiencies in the magnitude and location of the features, which cannot easily be traced to the formulation (resolution or the parameterizations included) of the models. Most models show a cold bias in all locations, apart from the tropical tropopause region where they can be either too warm or too cold. The strengths and locations of the major jets are often misrepresented in the models. Looking at three-dimensional fields reveals, for some models, more severe deficiencies in the magnitude and positioning of the dominant structures (such as the Aleutian high in the stratosphere), although undersampling might explain some of these differences from observations. All the models have shortcomings in their simulations of the present-day climate, which might limit the accuracy of predictions of the climate response to ozone change and other anomalous forcing.

1965-75: Spread of AGCMs

By 1965, three groups in the United States had established ongoing efforts in general circulation modeling:

  • Geophysical Fluid Dynamics Laboratory
  • UCLA Dept. of Meteorology
  • National Center for Atmospheric Research

At this point, AGCMs and modeling techniques began to spread by a variety of means. Commonly, new modeling groups began with some version of another group’s model. Some new groups were started by post-docs or graduate students from one of the three original AGCM groups. Others built new models from scratch. An AGCM family tree offers a visual map of these relationships, with links to some of the models and modeling groups.

I plan to have more to say on this later.


August 14, 2009 Posted by | GCM (General Circulation Models) | 1 Comment