New Hampshire experts debate how global warming is measured

Union Leader, July 25, 2007

 HYPERLINK "http://www.unionleader.com/article.aspx?headline=New+Hampshire+experts+debate+how+global+warming+is+measured&articleId=be437b68-68a6-42f8-bf68-5c0ae76f5c26" http://www.unionleader.com/article.aspx?headline=New+Hampshire+experts+debate+how+global+warming+is+measured&articleId=be437b68-68a6-42f8-bf68-5c0ae76f5c26

By John Distaso

Have Northeast winters grown dramatically warmer in the last 30 years? A researcher for the Union of Concerned Scientists says yes, emphatically. UNH researcher Dr. Cameron Wake's findings of a four-degree increase from 1970 to 2000 have been cited in a congressional global warming hearing and contributed to UCS reports last year and this month that predicted huge changes ahead.

But a retired Boston meteorologist says the finding is the product of biased data selection. It is something the scientists group has been accused of before.

"I can get anything I want out of the data, too, depending on what years I start and finish. That's not rocket science," said Dr. Fred Ward, longtime Channel 7 weatherman, now retired and living in New Hampshire,

Wake said last week his figures are accurate, the product of intense research of temperatures recorded by 75 government-sanctioned observation stations throughout the region. He said it was Ward's research, using data from only 11 National Weather Service stations in New England, that was "shoddy.''

Ward's figures show winter temperatures have risen only two-tenths of a degree since the early 1970s. He compared the average temperature reported by the stations from 1971 to 1975 to the average temperature reported by the stations from 2001 to 2005.

The New Hampshire Union Leader checked both men's findings with data available at the National Climatic Data Center. The North Carolina-based agency is part of the U.S. Department of Commerce and is the collection center and custodian of all United States weather records. NCDC is the central source of historical weather information, some of which it publishes as serial or periodic publications.

For the period 1970 to 2000, the numbers appear to be far closer to the researcher Wake's than to weatherman Ward. But Ward says data and date selections are key.

Wake said he used 1970 as a benchmark date to show the recent change in climate because when scientists "use the global temperature curve, we see a distinct change since 1970."

But Ward points out that the NCDC Web site shows that the winter temperature in 1970 was "a big dip" to an average of 21 degrees, which provided a low starting point. Ward says that one need only shift the beginning and end years of a time period to show different results.

The NCDC data shows, for instance, that while the average winter temperature in the Northeast increased by 3.81 degrees from 1970 to 2000, it increased by 3.0 degrees from 1975 to 2005. The winter of 1975 happened to be unusually warm, averaging 27 degrees, according to the NCDC, while in 1976, the winter averaged 24 degrees and in 2006, it averaged nearly 28 degrees. As a result, the 30-year increase from 1976 to 2006 was 4.02 degrees, or 1.34 degrees per decade.

Over the shorter time period of 1983 to 2003, the winter temperature increase was only 1.2 degrees, or .61 degrees per decade.

"He picked the time period to give him the maximum change," Ward said.

But Wake said that is simply not the case because extremes were thrown out in his data to achieve an accurate result from a "broad regional average." He said Ward "has done a bad analysis, I wouldn't let my graduate students do that type of research."

Research questioned

Charges and counter-charges of biased and poor research, and sometimes biased reporting of such research, are often part of the scientific and political debate over global warming.

The Union of Concerned Scientists was recently chided for poor research and has changed its surveying methods as a result.

A survey it released earlier this year claimed to show the Bush administration engaged in "wide-ranging political interference in research related to global warming.''

The study cited a survey of more than 1,600 federal climate scientists as evidence, saying 43 percent of respondents, "perceived or personally experienced changes or edits during review of their work that changed the meaning of the scientific findings.''

But Wall Street Journal opinion editor James Taranto and a self-described non-partisan statistical group affiliated with George Mason University criticized the survey as flawed, both for the low response rate (fewer than 300, or 17 percent, of those queried) and for lumping in those who "perceived'' changes with scientists who said their work had actually been changed. Only 15 percent of respondents were in the latter category.

The Statistical Assessment Service (STATS) is also affiliated with the Center for Media and Public Affairs. It said the UCS survey may have a "selection bias,'' in that "the scientists who took the time and effort to fill out and return the questionnaire might be precisely those most upset about perceived interference.''

Francesca Grifo, director of UCS's scientific integrity program, told the New Hampshire Union Leader her group took the criticism by STATS "very seriously" because UCS is not interested in presenting data that can be easily criticized by political opponents.

She said the survey response rate was low because UCS wanted to "maintain the anonymity of the scientists" and "had no way of tracking who had, and who had not, responded, so we could not go back and send reminders to get the response rate up." She said UCS has now contracted with Iowa State University's Center for Statistics and Methodology to improve future survey response rates.

Grifo also said, "We never intended to present these numbers as a representative sample" of the seven agencies surveyed. "The media did that.

"But what is valid," Grifo said, "is that all of those numbers should have been zero," with no scientists feeling pressured to reach certain conclusions or that their work had been changed.