Computer Models Fail to Predict Climate

Computer Models Fail to Predict Climate
February 1, 2008

Computer models that form the basis for future global warming predictions have projected significantly more warming in recent years than has actually occurred, concludes a comprehensive new scientific study.

"A Comparison of Tropical Temperature Trends with Model Predictions," published in the December 2007 International Journal of Climatology, is the latest study to cast doubt on the efficacy of climate modeling. Climate scientists David H. Douglass, John Christy, and S. Fred Singer analyzed 22 climate models and found their predictions at odds with actual warming over the past 30 years.

No Human Fingerprint

Most of the models predicted significant middle- and upper-troposphere warming, yet actual warming was minimal.

Douglass and his colleagues write, "Model results and observed temperature trends are in disagreement in most of the tropical troposphere, being separated by more than twice the uncertainty of the model mean. In layers near 5 km, the modelled trend is 100 to 300% higher than observed, and, above 8 km, modelled and observed trends have opposite signs."

Christy, an Intergovernmental Panel on Climate Change (IPCC) contributor, noted in a December 6 press statement, "Satellite data and independent balloon data agree that atmospheric warming trends do not exceed those of the surface. Greenhouse models, on the other hand, demand that atmospheric trend values be 2-3 times greater. Satellite observations suggest that greenhouse models ignore negative feedbacks, produced by clouds and by water vapor, that diminish the warming effects of carbon dioxide."

Models Don't Reflect Causes

Many top climate scientists point out climate models are incapable of handling confounding factors such as cloud cover and water vapor (the dominant greenhouse gas), thus distorting climate predictions.

Additionally, they note, the models do not reflect the actual causes of warming. Richard Lindzen, a professor of meteorology at MIT, says the models used by the IPCC and other alarmists assign too much warming resulting from increases in atmospheric carbon dioxide, rendering the models' predictions inaccurate.

Singer writes, "Dire predictions of future warming are based almost entirely on computer climate models, yet these models do not accurately understand the role of water vapor. Plus, computer models cannot account for the observed cooling of much of the past century (1940-75), nor for the observed patterns of warming. For example, the Antarctic is cooling while models predict warming. And where the models call for the middle atmosphere to warm faster than the surface, the observations show the exact opposite."

Computer Programs Inadequate

Computers, no matter how big, cannot take account of all of the earth's complexities and processes, critics of the alarmist models also note. As a result, no current climate model can explain the causes of climate changes, accurately predict future climate, or form a sound basis for environmental policy.

"Mother Nature simply operates at a level of complexity that is, at this point, beyond the mastery of mere mortals (such as scientists) and the tools available to us," Christy was quoted as saying in The Wall Street Journal on November 1.

"Can the models accurately explain the climate from the recent past? It seems that the answer is no," summarized Douglass in the press statement.

"This new study adds another nail to the coffin of alarmist global warming theory," said Sterling Burnett, senior fellow at the National Center for Policy Analysis.

"Alarmist global warming theory is totally dependent on computer models predicting accelerating warming in the future," Burnett noted, "yet the models have predicted such warming in the past, and the predicted warming has failed to materialize. This hardly seems a reliable indicator of future warming."


Drew Thornley (dthornley@texaspolicy.com) is a policy analyst at the Texas Public Policy Foundation.