Testing Global Climate Models

From The Encyclopedia of Earth
Jump to: navigation, search
search entire Network
Temperature verification (main)

Testing Global Climate Models

December 18, 2010, 12:00 am
May 7, 2012, 6:55 pm
Content Cover Image

Observed global temperature changes versus simulations via the HadGEM1 GCM, assuming only natural climate forcing factors, anthropogenic forcing factors, or the sum of the two. The gray band indicates the uncertainty of the calculations for all forcing fa

Camel-icon.png

If we are given values for the past, present, and projected future of greenhouse gases, we can derive the total of all climatic forcing factors, and run global climate models (GCMs) to calculate the past, present, and likely future of Earth’s climate. Verifying a GCM involves simulating the climate for some time past and comparing the model’s output with what actually occurred. Discrepancies between a GCM and the climate record indicate problems with the model’s assumptions, algorithms, initial conditions, coordination among various submodels, or even the climate record. Tweaking a GCM to correct for such problems is a fine art because climate involves interactions among many variables that do not exhibit linear behavior.

Linear behavior dictates that changes in output be proportional to changes in input. In nonlinear behavior, small changes in conditions may have large consequences, or conversely, large changes may have no perceptible consequences. An example of nonlinear behavior is that lowering the temperature of an air mass has only slight effects on its transmission of electromagnetic radiation until water vapor in the air mass reaches saturation, a cloud forms, and, suddenly, albedo skyrockets.

The Intergovernmental Panel on Climate Change (IPCC), in its Fourth Assessment Report released in 2007, presents the results from 16 or more GCMs that simulated Earth’s climate under a range of forcing factors, both natural (e.g., orbital variations) and anthropogenic (e.g., human activities that emit additional greenhouse gases). The models were run three times: once with only natural climate forcing, the second time with only anthropogenic forcing, and, finally, with both natural and anthropogenic forcing. The best match for reconstructing global temperatures over the past 125 years occurred when the models included both types of forcing. In particular, anthropogenic forcing appears responsible for most of the 0.5°C rise in global temperatures experienced from 1970 until 2000.

This is an excerpt from the book Global Climate Change: Convergence of Disciplines by Dr. Arnold J. Bloom and taken from UCVerse of the University of California.

©2010 Sinauer Associates and UC Regents

Glossary

Citation

Bloom, A. (2012). Testing Global Climate Models. Retrieved from http://editors.eol.org/eoearth/wiki/Testing_Global_Climate_Models

0 Comments