WASHINGTON - At a tumultuous time in UN-led climate negotiations, one of the world's most credible scientific groups agreed Wednesday to plug the recent cracks in the authoritative reports of the United Nations' Nobel Prize-winning global warming panel.

"We enter this process with no preconceived conclusions," said Robbert Dijkgraaf, a Dutch mathematical physicist who co-chairs the group, the InterAcademy Council of 15 nations' national academies of science.

UN Secretary-General Ban Ki-moon asserted "there were a very small number of errors" in the 3,000 pages of the beleaguered U.N. Intergovernmental Panel on Climate Change's last major synthesis of climate data in 2007.

But those errors, which include projections of retreats in Himalayan glaciers, have put public confidence in the panel's work at risk, and have been seized on by climate skeptics opposed to the UN-led efforts to conclude a legal international agreement on global warming this year.

The nonbinding Copenhagen accord brokered by President Barack Obama in the final hours of the December climate change summit in the Danish capital has the support of major polluters and ecomomies such as the U.S., China and India. But it fell well short of its original ambition of a legally binding treaty controlling the world's emissions of carbon dioxide and other gases blamed for global warming.

Dijkgraaf, told reporters that his Netherlands-based group, which agreed to the UN's request to review the panel's work, "will definitely not go over all the data, the vast amount of data in climate science," but will instead look at focus on how the panel does its job, in light of the unsettling errors that have surfaced recently.

The group will pick a panel of experts, no easy task since the bulk of climate experts already participate in the UN panel, and wrap up its independent review by the end of August, said Dijkgraaf, also president of the Royal Netherlands Academy of Arts and Sciences.

Among the questions are whether the climate panel should use non-peer reviewed literature, how governments review IPCC material, and even how the IPCC communicates with the public.

No errors surfaced in the earlier and most well-known of the reports, which said the physics of a warming atmosphere and rising seas is man-made and incontrovertible. But there have been several mistakes discovered in the second of the four climate research reports produced in 2007, mainly owing to the use of government reports or even advocacy group reports instead of peer-reviewed research.

For example, in the Asian chapter, five errors in a single entry on glaciers in Himalayas say those glaciers would disappear by 2035 -- hundreds of years earlier than other information suggests -- with no research backing it up. A sentence in the chapter on Europe says 55 per cent of the Netherlands is below sea level, when it's really about half that amount.

And a section in the Africa chapter that talks about northern African agriculture says climate change and normal variability could reduce crop yields. But it gets oversimplified in later summaries so that lower projected crop yields are blamed solely on climate change.

Ban said the mistakes in the IPCC reports, found in recent months, don't undercut the broad consensus on global warning.

"Nothing that has been alleged or revealed in the media recently alters the fundamental scientific consensus on climate change," the secretary-general said. "Nor does it diminish the unique importance of the IPCC's work."

He did not respond to a question about how the errors might be affecting the UN-led negotiations.

The errors have shaken the credibility of climate scientists and given skeptics of global warming ammunition.

Longtime climate skeptic John Christy of the University of Alabama said he wasn't familiar with the InterAcademy Council, but he cheered the outside review.

"I hope people like me have input, otherwise its just the usual members of the establishment defending to themselves what's been done," said Christy, a researcher.

The review will involve a mix of outside experts and climate scientists who have worked with the IPCC before but are "far enough removed to be truly independent," Dijkgraaf told The Associated Press in a telephone interview. The idea is to have expertise and insight into how IPCC works without including current leaders, he said.

"The full panel needs gravitas and I think scientific stature," Dijkgraaf said. The members of the panel haven't been chosen, but they likely will be 10 scientists.

The evaluation group will be chosen when the InterAcademy's board meets on March 22, Dijkgraaf said. The InterAcademy has done science reviews before for the United Nations.

The IPCC was formed in 1989 by the United Nations and the World Meteorological Organization to study global warming and its causes and effects.

Prominent mainstream climate scientist Kevin Trenberth at the National Center for Atmospheric Research said "climate science has become a political hot potato." He said the reviewers should not just look at the IPCC but the standards of its critics.

The IPCC, which is mostly a collection of volunteer scientists, shared the Nobel Peace Prize in 2007 with former Vice-President Al Gore.