The Weekend Australian
By: Jeffrey Ball, Keith Johnson
Source: The Wall Street Journal
A new strategy to restore credibility will focus on the basics
Over the next few days, the world's leading authority on global warming plans to roll out a strategy to tackle a tough problem: restoring its own bruised reputation.
A months-long crisis at the Intergovernmental Panel on Climate Change has upended the world's perception of global warming, after hacked emails and other disclosures revealed deep divisions among scientists working with the UN-sponsored group.
That has raised questions about the panel's objectivity in assessing one of today's most hotly debated scientific fields.
The problem stems from the IPCC's thorny mission: to take sophisticated and sometimes inconclusive science, and boil it down to usable advice for politicians. To meet that goal, scientists working with the IPCC say they sometimes faced institutional bias toward oversimplification.
Richard Alley, a geoscientist who helped write the IPCC's latest report, issued in 2007, described a trip that northern summer to Greenland's ice sheet with senators who urged him to be as specific as possible about the potential for sea-level rise. The point many of them made, he said, was: Give more explicit advice -- because, if the sea rises, "the levee has to be built some height''.
The tension within the IPCC stretches back a decade or more, according to interviews with scientists and a review of hundreds of IPCC documents and emails.
It has complicated the panel's work on matters ranging from the study of tree rings to the proper use of massively complex climate computer models.
To restore its credibility, the IPCC will focus on enforcing rules already on the books, IPCC chairman Rajendra Pachauri and other officials said in interviews.
Scientific claims must be checked with several experts before being published. IPCC reports must reflect disagreements. And those who write reports must refrain from advocating specific environmental actions.
The reforms aim to "ensure that even the slightest possibility of someone not adhering to procedures is eliminated completely'', Mr Pachauri said. "We just have to act like monitors at every stage.''
The IPCC shared a Nobel Peace Prize with former US vice-president Al Gore in 2007. The group won for a report that declared climate change was "unequivocal'' and "very likely'' caused by emissions of carbon dioxide and other greenhouse gases due to human activity.
Formed in 1988, the group doesn't conduct or fund scientific research, but filters the work of researchers worldwide to assess what is, and isn't, known about climate change. About 30 paid staffers help thousands of scientists who volunteer to assemble voluminous "assessment reports'' every five or six years.
The goal is to be "policy-relevant'' but "never policy-prescriptive,'' the IPCC says.
The organisation's troubles raise questions about its quality control in summarising science. But many scientists say the crisis does not undermine independent research demonstrating man's influence on the climate.
"There is a very broad and deep consensus that I buy into that we're producing too much CO2 and it's going to cause problems eventually,'' said John Marburger, science adviser to former US president George W. Bush. Many details remain uncertain, he said, but "I think it's unequivocal that there is a human component''.
The IPCC's annual budget, about $US7 million ($7.9m) this year, is funded mainly by contributions from the US and other industrialised nations.
IPCC leaders including Mr Pachauri say the IPCC is rigorous and transparent. The group issued a statement last month expressing "regret'' for what IPCC leaders describe as a particularly embarrassing mistake in the 2007 report: an erroneous projection that Himalayan glaciers could melt by 2035. "The organisation has an impeccable record of having performed,'' Mr Pachauri said, and its work "always includes the quantifications of uncertainties''.
As climate change gained public attention in recent decades, some IPCC-affiliated scientists privately expressed concerns conclusions were at risk of getting oversimplified.
Keith Briffa, a climate scientist at East Anglia, expressed this worry in emails to colleagues in 1999, as work intensified on the IPCC's third major report, published in 2001.
Mr Briffa's particular concern: tree rings.
Scientists use tree rings and other proxies to assess temperatures thousands of years ago, before thermometers existed. Wider rings indicate greater growth, generally suggesting warmer temperatures, or higher precipitation, or both. Mr Briffa pioneered the technique.
"I know there is pressure to present a nice tidy story as regards `apparent unprecedented warming in a thousand years or more','' he wrote to other researchers in the email, among those hacked at East Anglia. "In reality the situation is not quite so simple.''
He failed to identify the source of the pressure. A university spokesman said Mr Briffa would not comment.
The problem: using Mr Briffa's tree-ring techniques, researchers in the 1990s built charts suggesting that temperatures in the late 20th century were at the highest levels in a millennium. The charts were dubbed "hockey sticks'' because they showed temperatures relatively flat for centuries, then angling much higher recently.
But Mr Briffa fretted about a potential problem. Thermometers show temperatures have risen since the 1960s, but tree-ring data do not move in tandem, and sometimes show the opposite.
In the same 1999 email, Mr Briffa said tree-ring data overall did show "unusually warm'' conditions in recent decades. But, he added, "I believe that the recent warmth was probably matched about 1000 years ago.''
In other words, maybe the chart should not resemble a hockey stick.
The data were the subject of heated back-and-forth in the months before the IPCC's 2001 report. John Christy, one of the section's lead authors, said at the time that he tried in vain to make sure the report reflected the uncertainty.
Mr Christy said some of the pressure to downplay the uncertainty came from Michael Mann, a fellow lead author of that chapter, a scientist at Pennsylvania State University, and developer of the original hockey-stick chart.
The "very prominent'' use of the hockey-stick chart "overrules what tentativeness some of us actually intended,'' Mr Christy wrote to the National Research Council in the US a month after the report was published. Mr Christy, a climate scientist at the University of Alabama, provided that email.
"I was suspicious of the hockey stick,'' Mr Christy said. Had Mr Briffa's concerns been more widely known at the time he said, "the story coming out of the (report) may have been different in tone and confidence''.
Mr Mann said: "I was not pushing `hard' for anything of the sort.'' The chapter's authors, he said, "engaged in a robust, good faith discussion of what the level of certainty was.'' Mr Mann also noted that his original 1998 hockey-stick paper stressed the uncertainties involved in reconstructing past temperatures.
Chris Folland, a leader of that chapter and a research fellow at the Hadley Centre, a prominent British climate lab, said last week he did not recall Mr Briffa's concerns about the data. He added that research since 2001 had bolstered the notion that the late 20th century was warmer than prior centuries.
Complicating matters, a simplified version of the hockey-stick chart appeared prominently in the 2001 report's "summary for policy makers''-- a 34-page distillation of the full report. Thomas Stocker, a climate scientist at the University of Berne and member of the team that wrote the summary, said the team wrestled with how to make it "faithful to the full report and yet still comprehensible'' to policy makers.
The hockey-stick chart is "the textbook example'' of "how difficult the job really is'' to summarise the full report, said Mr Stocker, one of the top scientists overseeing the IPCC's next report.
In retrospect, he said, the simplified version should have had more detail. It could suggest a clearer conclusion.
"That was part of the problem--that we simplified it,'' he said. "It's not suppressing information, but it's making it harder for the rapid reader to have the full picture.''
Another crucial issue: The accuracy of highly complex computer models that underpin climate science. Run on supercomputers, these models try to predict how greenhouse-gas emissions might affect temperatures, and how temperatures might affect everything from glaciers to hurricanes.
In September 2000, Filippo Giorgi, a researcher at the International Centre for Theoretical Physics in Trieste, Italy, wrote a worried email to colleagues. He said he felt pressure to cite climate simulations that were yet to be published in a scientific journal. He worried it showed a relaxation of standards.
The IPCC's rules "have been softened to the point that in this way the IPCC is not any more an assessment of published science (which is its proclaimed goal),'' he wrote in the email. Mr Giorgi added: "At this point there are very little rules and almost anything goes. I think this will set a dangerous precedent.''
In an interview, Mr Giorgi said the pressure he felt came from the panel of experts overseeing the portion of the report he worked on. That panel was co-chaired by John Houghton, a scientist who previously had chaired the IPCC as a whole.
Mr Houghton defended his panel's oversight. "Nobody was arguing for 'anything goes','' he said. "My concern as chairman was to make the most useful presentation of results'' by using the new models. "Nobody was arguing for making choices that selected anything more dramatic or with a particular message,'' he said. "Everybody wanted to present the results in the most helpful as well as honest way.''
Mr Giorgi said in the interview that including the data ultimately did no harm, because the IPCC report included a disclaimer noting it had not appeared in a scientific journal. Eventually, he added, the work was published in a journal.
Some researchers continued to feel pressure to boil down complex science as work began on the IPCC's fourth major report, published in 2007. Things that are "very difficult to quantify must be quantified to keep the policy makers happy,'' said Mr Alley, the geoscientist, who teaches at Penn State University. "It's a very frustrating thing.''
Mr Alley walked that tightrope in helping write the chapter covering his speciality: the degree to which massive Greenland and Antarctic ice sheets might melt, raising sea levels. The problem, he said: "Ice-sheet models are not very good.''
Many conversations with policy makers world-wide, including Mr Gore, the senators on the Greenland trip and Christian Gaudin, a French senator, left the clear impression that "we scientists had better get better numbers'', said Mr Alley, adding he understood their need for detail.
So the scientists put numbers into the 2007 study, along with a big caveat -- what Mr Alley calls a "punt''. The study took into account things like glacier melt in most of the world, but it noted that it excluded what is happening in the Greenland and Antarctic ice sheets, which "we can't predict'', Mr Alley said.
Inevitably, Mr Alley said, some people have quoted the numbers without mentioning that important caveat.
In a statement, a spokeswoman for Mr Gore said the former vice-president understood the uncertainties, and that Mr Gore pointed out in public statements "that there was essentially an asterisk'' on the 2007 report's sea-level projections. "As he understands the situation from the ice-science community, the uncertainty in sea level applies in both directions'', meaning sea-level rise could be greater or smaller than projected, the statement said.
Senator Gaudin said he recalled having lunch with Mr Alley at which they discussed the interplay between scientists and politicians on the "big questions that interest society'', notably climate change. Scientific reports, including the IPCC's, "need to have more precision'', Senator Gaudin said. It is "difficult for politicians to make a decision'' otherwise.
Mr Marburger said he frequently heard policy makers express frustration at the lack of certainty in many areas of science, including climate change. "Why can't we get better numbers? Everybody asks that,'' he said. "But science rarely gives you the right answer. Science tells you what the situation is, but it doesn't tell you what to do.''
As the IPCC gears up for its next big assessment report, Mr Pachauri and other IPCC leaders say the organisation needs to be more watchful than ever.
"There's really no other scientific endeavour of the size and scale and scope of what the IPCC has done,'' Mr Pachauri said. The next step, he added, is that "we really need to build in some set of checks and balances''.