Bad Data Hurts Model Performance

An article appearing in the April 19 issue of Science looks at how accurately climate models are able to reproduce current and past climate. What the authors conclude is that the models do a decent job of simulating the observed data, but that the data itself may not be that good.

“We can now test how well climate models simulate century-scale variations in the observed climate record,” say the authors, Thomas Smith, Thomas Karl and Richard Reynolds, at the National Climatic Data Center. “There have been numerous intercomparisons of various climate model simulations of 20^th-century climate, based on the best available estimates of the climate forcing.”

It is assumed that these simulations closely reproduce observed climate variability, but, “This assumption mustbe viewed with caution,” say the authors. “Observational errors, sampling errors, and time-dependent biases degrade the climate record.” Although researchers have attempted to remove these errors, there are still problems.

The authors compare the errors in the computer models with the errors in the observed temperature record of worldwide sea surface temperatures (SSTs) which stretch back to the 19^th century. To illustrate model errors the authors run three separate computer simulations with identical forcings, but starting at different initial conditions. By doing this they were able to “estimate the magnitude of the uncertainty introduced by a chaotic climate system.”

For the observed temperature record the authors calculated the uncertainty by comparing the range of SSTs derived from different observational errors, sampling errors and time-dependent bias adjustments. “The errors in analyzed SSTs,” say the authors, “are comparable to the uncertainty estimate associated with climate chaos over much of the 20^th century.”

They conclude, “Todays models are thus within the observed uncertainty of the observations, at least with respect to the global SST record. This does not imply that the model simulations are perfect; rather, it indicates that more attention must be given to improving the records of past climate and ensuring that future climate records have little or no time-dependent biases.

“It is unsettling that the uncertainty related to treatment of the data is increasing in recent decades in the most-sampled oceans. This points to the importance of developing a global observing system that not only has good spatial coverage, but more importantly, strictly adheres to guidelines and principles articulated by the U.S. National Academy of Sciences.”

• In an April 21 editorial in the New York Times, former Vice President Al Gore called Dr. Rajendra Pachauri, the new chairman of the United Nations Intergovernmental Panel on Climate Change, a “lets drag our feet candidate who is known for his virulent anti-American statements.”

Dr. Pachauri responded in a letter to the editor in the May 1 New York Times, stating, “Mr. Gores derogatory statements about me reflect a deep disappointment at my election as chairman of the [IPCC], with 76 votes for me against 49 for his protg, Dr. Robert T. Watson.

“In a 1999 speech, Mr. Gore, referring to my commitment, vision and dedication, said; Pachy is the one person in the world who could bring us all here. He is known all over the community of concerned men and women as someone with the intellect and the heart.

“In Earth in the Balance, Mr. Gore acknowledged me among the other scientists who have been helpful in giving me advice during the writing of this book.

“Would the real Al Gore stand up? Does what he says today hold no value tomorrow?”

Announcements

• The Cooler Heads Coalition together with the Science and Environmental Policy Project and Frontiers of Freedom will hold a briefing on “Whatever Happened to Global Warming?” on Monday, May 13, from Noon to 2 PM in Room B-339 of the Rayburn House Office Building. Speakers will include John Daly, who runs the highly regarded Still Waiting for Greenhouse web site in Australia, and climate scientists from Europe and North America. Lunch will be served. Attendance is free of charge, but reservations are required. Please contact Ericka Joyner of CEI at (202) 331-1010, ext. 267, or e-mail her at ejoyner@cei.org. Include your name, affiliation, and phone number.