A short drive from the windswept North Sea coast of England, the Climatic Research Unit occupies a squat, weather-beaten grey concrete building on the campus of the University of East Anglia.
This scientific bunker holds the world's largest trove of climate-change data, gleaned from Siberian tree-ring counts, Greenland ice-layer measurements and centuries-old thermometer readings.
Now the pirating of thousands of e-mail messages from within its walls has revealed a dangerous bunker mentality among the scientists who guarded those records and a data-fudging scandal that has created a crisis of confidence in global-warming science that is threatening to destroy the political consensus around next week's carbon-policy summit in Copenhagen.
Said one scientist working at the institute: "It wouldn't be an exaggeration to say that this has set the climate-change debate back 20 years."
The crisis intensified yesterday as the head of the UN's Intergovernmental Panel on Climate Change, the main scientific and political authority on global warming, announced an investigation into the university's practices and the reliability of the findings that have underpinned the UN's climate-change conclusions. The university has launched its own inquiry and on Wednesday ordered the CRU's embattled head, Phil Jones, to step down until it is complete.
On a political level, coming on the eve of the Copenhagen summit, the controversy has been catastrophic: In the last few days, it has prompted opposition politicians in the United States, Britain and Australia to argue that human-caused global warming is a myth.
Saudi Arabian officials now say that they will argue in Copenhagen that carbon-emission controls are pointless because the CRU scandal has nullified any evidence of human-caused atmospheric temperature increase.
The reports the CRU produced from its now-controversial data were the main source of the UN's key global-warming document, the IPCC's report of 2007, which concluded that "warming of the climate system is unequivocal" and that "most" of the global temperature increase since the mid-20th century has been caused by human activity - a conclusion, still supported by the majority of atmospheric scientists, that most governments adopted as the basis of their carbon-emissions policy.
That consensus has been shaken by hundreds of pages of messages, apparently stolen from the lab's servers, which have been interpreted as suggesting that the scientists at the CRU manipulated data to make it deliver a more dramatic message about the human contribution to global warming, destroyed data files that did not support their hypothesis, and tried to prevent critics within the scientific community from having access to their raw information and methods.
Unusually, even sympathetic scientists and some activists have concluded that the credibility of climate science has been seriously harmed.
"We should not underestimate the damage caused by what has happened, either for the science or for the politics of climate change, and potentially it could have some very far-reaching consequences," said Mike Hulme, a climate scientist at East Anglia whose e-mails were among those included in the pirated files and who has been critical of the secrecy and lack of impartiality in his colleagues' work.
Independent scientists are quick to point out that the actions described in the e-mails do not describe anything like a fabrication of global-warming evidence, and that two other major sets of historical data drawn from the same sources, both held by NASA institutions in the United States, also show a historical warming trend.
That has not stopped right-wing politicians in Western countries from using the scandal to dramatic effect: Yesterday, a group of Hollywood conservatives launched a campaign to revoke the Academy Award given to Al Gore, the former vice-president and a carbon-cap advocate, for his climate documentary An Inconvenient Truth.
But perhaps more important than the ammunition the CRU affair has given to conspiracy theorists is what it has revealed about the awkward role scientists have come to play in the heated world of climate policy.
"I think there is a serious problem with the way scientists are used, and the way they position themselves, in climate-policy debates," Prof. Hulme said. "Wherever you look around climate change, people are bringing their ideologies, beliefs and values to bear on the science."
The CRU files, apparently hacked or leaked from the institute's server, began appearing on websites on Nov. 17, and reached the attention of climate-skeptic groups and the media two days later.
The most contentious e-mail was written by Prof. Jones, the director of the CRU, who wrote to colleagues in 1999, as they studied measurements of Siberian tree rings, which scientists have long realized do not reflect local temperature changes after 1961: "I've just completed Mike's trick of adding in the real temps to each series for the last 20 years (i.e. from 1981 onwards) and from 1961 to hide the decline."