Phil Jones: facing up to his critics.

Phil Jones holds himself defensively, his arms crossed tightly in front of his chest as if shielding himself from attack. Little wonder: Jones has spent the past three months being vilified for his central role in what is now called 'climategate'.

Jones was director of the Climatic Research Unit (CRU) at the University of East Anglia in Norwich, UK, when, last November, more than 1,000 e-mails and documents were illegally obtained from the university and posted on the Internet. Their contents quickly embroiled him in a controversy that has shaken the climate community and threatened his career. Jones has stood down from his post while several independent investigations look into the affair, including one headed by Muir Russell, former vice-chancellor of the University of Glasgow, UK, which will assess allegations that the e-mails contain evidence of poor scientific practice at the CRU.

Speaking exclusively to Nature, Jones is reluctant to discuss how the past few months have affected him personally, and says he cannot comment on allegations that freedom of information requests for raw climate data were mishandled by the university. But he is eager to set the record straight on the science.

Central to the Russell investigation is the issue of whether he or his CRU colleagues ever published data that they knew were potentially flawed, in order to bolster the evidence for man-made global warming. The claim specifically relates to one of Jones's research papers1 on whether the urban heat island effect — in which cities tend to be warmer than the surrounding countryside — could be responsible for the apparent rise in temperature readings from thermometers in the late twentieth century. Jones's study concluded that this local effect was negligible, and that the dominant effect was global climate change.

In the paper1, the authors used data from weather stations around the world; those in China "were selected on the basis of station history: we chose those with few, if any, changes in instrumentation, location or observation times", they wrote.

I don't think we should be taking much notice of what's on blogs because they seem to be hijacking the peer-review process. Phil Jones

But in 2007, amateur climate-data analyst Doug Keenan alleged that this claim was false, citing evidence that many of the stations in eastern China had been moved throughout the period of study. Because the raw data had been obtained from a Chinese contact of one of Jones's co-authors, Wei-Chyung Wang of the University at Albany in New York, and details of their location had subsequently been lost, there was no way of verifying or refuting Keenan's claim.

Jones says that approaching Wang for the Chinese data seemed sensible at the time. "I thought it was the right way to get the data. I was specifically trying to get more rural station data that wasn't routinely available in real time from [meteorological] services," says Jones, who asserts that standards for data collection have changed considerably in the past twenty years. He now acknowledges that "the stations probably did move", and that the subsequent loss of the stations' locations was sloppy. "It's not acceptable," says Jones. "[It's] not best practice." CRU denies any involvement in losing these records.

Jones says that he did not know that the weather stations' locations were questionable when they were included in the paper, but as the study's lead author he acknowledges his responsibility for ensuring the quality of the data. So will he submit a correction to Nature? "I will give that some thought. It's worthy of consideration," he says.

"The science still holds up" though, he adds. A follow-up study2 verified the original conclusions for the Chinese data for the period 1954–1983, showing that the precise location of weather stations was unimportant. "They are trying to pick out minor things in the data and blow them out of all proportion," says Jones of his critics.

That Nature trick

One of the most politically charged allegations is that Jones, together with scientific collaborators, tried to systematically downplay the importance of the Medieval Warm Period (MWP), a brief phase of natural, pre-industrial warming that may have occurred around 1000 AD. But if the MWP was restricted to mild local warming, it would mean that present-day global warming is unprecedented for the past 1,000 years, as claimed by climatologist Michael Mann of Pennsylvania State University, University Park, in his famous 'hockey stick' global temperature reconstruction3. That claim, however, relies on controversial data from tree rings.

In one of the leaked e-mails, Jones wrote that he had "just completed Mike's Nature trick of adding in the real temps to each series for the last 20 years to hide the decline". Jones was referring to the fact that he — like Mann — used only direct temperature measurements to reconstruct temperatures over the past 20 years or so, rather than using proxy data.

Palaeoclimatologists are confident that the width of tree rings reliably represents real temperatures because they tally with data from thermometers and other instruments taken since the nineteenth century. After the 1960s, however, there is a divergence, with most tree-ring proxy temperatures seeming to be lower than those from instrumental records across the Northern Hemisphere4. The exact cause of this problem is unknown, and is still being investigated by scientists.

Some argue that if the tree-ring data are unreliable for the recent past, including them in older temperature reconstructions is highly questionable, and could understate historic warming — including the MWP — relative to the present day. "It potentially does," admits Jones, but says that analyses using other methods — proxy temperature markers from ice core samples, for example — still show much the same temperature change over the past 1,000 years, backing up Mann's hockey stick.

Jones says that he is unconvinced that the MWP was a global phenomenon, but categorically denies that he has tried to downplay it. "A lot of people have this view that there was a MWP and then a little ice age," he says. "It might not be the case."

"We need more reconstructions from different parts of the world to reproduce a better history of the past thousand years". Jones challenges his critics to help with those efforts. "Why don't they do their own reconstructions?" he asks. "The work that's been published has been through the peer-review process; if they want to criticize that they should write their own papers."

Redefining review

But some of his critics say that Jones has tried to influence that very process by censoring them. In one e-mail, dated July 2004, Jones wrote to Mann: "I can't see either of these papers being in the next IPCC report. Kevin [Trenberth] and I will keep them out somehow — even if we have to redefine what the peer-review literature is!" Jones stresses that both papers did appear in the final version of the Fourth Assessment Report from the United Nations' Intergovernmental Panel on Climate Change (IPCC). But he also defends the right of IPCC authors to exclude papers from the report, based on their expertise.

"The IPCC [report] is an assessment, it's not a review," he says, "so the authors have to know something about the subject to assess which are the important papers to bring in to the particular chapter." In doing so, authors naturally would exclude papers that are scientifically weak or irrelevant, argues Jones.

But he fears that the aftermath of the climategate affair is undermining the integrity of the scientific review process. "I don't think we should be taking much notice of what's on blogs because they seem to be hijacking the peer-review process," says Jones.

It is now essential for climate researchers to stand up for their science, he says. "[I'd] like to see the climate science community supporting the climate science more. Lots of them are trying but they're being drowned out."