Call it the mystery of the missing thermometers.
Two months after "climategate" cast doubt on some of the science behind global warming, new questions are being raised about the reliability of a key temperature database, used by the United Nations and climate change scientists as proof of recent planetary warming.
Two American researchers allege that U.S. government scientists have skewed global temperature trends by ignoring readings from thousands of local weather stations around the world, particularly those in colder altitudes and more northerly latitudes, such as Canada.
In the 1970s, nearly 600 Canadian weather stations fed surface temperature readings into a global database assembled by the U.S. National Oceanic and Atmospheric Administration (NOAA). Today, NOAA only collects data from 35 stations across Canada.
Worse, only one station -- at Eureka on Ellesmere Island -- is now used by NOAA as a temperature gauge for all Canadian territory above the Arctic Circle.
The Canadian government, meanwhile, operates 1,400 surface weather stations across the country, and more than 100 above the Arctic Circle, according to Environment Canada.
Yet as American researchers Joseph D'Aleo, a meteorologist, and E. Michael Smith, a computer programmer, point out in a study published on the website of the Science and Public Policy Institute, NOAA uses "just one thermometer [for measuring] everything north of latitude 65 degrees."
Mr. D'Aleo and Mr. Smith say NOAA and another U.S. agency, the NASA Goddard Institute for Space Studies (GISS) have not only reduced the total number of Canadian weather stations in the database, but have "cherry picked" the ones that remain by choosing sites in relatively warmer places, including more southerly locations, or sites closer to airports, cities or the sea -- which has a warming effect on winter weather.
Over the past two decades, they say, "the percentage of [Canadian] stations in the lower elevations tripled and those at higher elevations, above 300 feet, were reduced in half."
Using the agency's own figures, Smith shows that in 1991, almost a quarter of NOAA's Canadian temperature data came from stations in the high Arctic. The same region contributes only 3% of the Canadian data today.
Mr. D'Aleo and Mr. Smith say NOAA and GISS also ignore data from numerous weather stations in other parts of the world, including Russia, the U.S. and China.
They say NOAA collects no temperature data at all from Bolivia -- a high-altitude, landlocked country -- but instead "interpolates" or assigns temperature values for that country based on data from "nearby" temperature stations located at lower elevations in Peru, or in the Amazon basin.
The result, they say, is a warmer-than-truthful global temperature record.
"NOAA . . . systematically eliminated 75% of the world's stations with a clear bias towards removing higher latitude, high altitude and rural locations, all of which had a tendency to be cooler," the authors say. "The thermometers in a sense, marched towards the tropics, the sea, and to airport tarmacs."
The NOAA database forms the basis of the influential climate modelling work, and the dire, periodic warnings on climate change, issued by James Hanson, the director of the GISS in New York.
Neither agency responded to a request for comment Wednesday from Canwest News Service. However Hanson did issue a public statement on the matter earlier this week.
"NASA has not been involved in any manipulation of climate data used in the annual GISS global temperature analysis," he said. "The agency is confident of the quality of this data and stands by previous scientifically-based conclusions regarding global temperatures."
In addition to the allegations against NOAA and GISS, climate scientists are also dealing with the embarrassment this week of the false glacier-melt warning contained in the 2007 report of the UN Panel on Climate Change. That report said Himalayan glaciers are likely to disappear within three decades if current rates of melting continue.
This week, however, the panel admitted there is no scientific evidence to support such a claim.
The revelations come only two months after the "climategate" scandal, in which the leak or theft of thousands of e-mails -- private discussions between scientists in the U.S. and Britain -- showed that a group of influential climatologists tried for years to manipulate global warming data, rig the scientific peer-review process and keep their methods secret from other, contrary-minded researchers.