Review of questioned IPCC report says conclusions 'well-founded'

Dutch government finds minor inaccuracies in contested paper, but reasserts that 'climate change poses "substantial risks" to most parts of the world'
Juliette Jowit guardian.co.uk 5 Jul 10;

The first major independent review of criticisms of the global assessment of climate change led by the United Nations declared today that it found "no errors that would undermine the main conclusions" of the panel of international scientists that climate change will have serious consequences around the world.

However the Dutch panel of experts claims it found 12 errors - from a criticism of the number of people in Africa at risk of water shortages to mistakes in references or typing. It also suggested the summary version of the report had portrayed an over-dramatic picture by putting the emphasis on negative impacts of climate change, and it failed to explain some of the threats were not only driven by climate change.

Among several recommendations, it said the UN's Intergovernmental Panel on Climate Change, paid for by governments including the UK and Netherlands, should in future pay researchers to review the report in more detail.

The report was officially welcomed by the IPCC and scientists who worked on the last assessment report, published in 2007, however only a small number of the "errors" have been corrected. The remaining errors were not accepted by the scientists, said Professor Martin Parry, who was co-chair of the section of the report that was under scrutiny.

"The conclusions are not undermined by any errors, and we'd like that to be the message the world will take," said Parry. "[They found] a very small number of near-trivial errors in about 500 pages [and] probably 100,000 statements. I would say that's pretty good going."

The scientists also rejected the potentially more damaging complaint that the IPCC's Summary for Policy Makers report, which condenses eight chapters on regional impacts to a single page of 32 statements, ignored positive impacts such as the ability to grow new crops in some parts of the world, or opening of shorter Arctic sea routes.

The summary, vetted "line by line" by governments, highlights the biggest impacts on humans and the environment which need political attention, said Parry. Positive benefits tended to be local and relatively small, said Professor David Vaughan of the British Antarctic Survey, a lead coordinating author of the next IPCC assessment.

The agency was asked to examine eight chapters about the regions in the 500-page section on global impacts put together by Working Group II, which itself formed half of the full 1,000-page IPCC assessment.

One "minor inaccuracy" the Dutch panel said it found was an estimate of people in Africa who could be exposed to water stress, which they said should be narrowed from 75m-250m to 90m-220m. However Professor Nigel Arnell, the source of the data, said although the underlying models could have been added differently, to recalculate the total would be to "over-interpret" the data by suggesting a level of accuracy the IPCC does not claim.

As well as the 12 errors, the Dutch reviewers made 23 criticisms of the "quality" of statements. These ranged from failure to explain that forecast water stress and heat deaths also had other causes such as population growth, to pointing out a link to underlying research did not work. Arnell said the IPCC report "repeatedly stresses" its estimates of numbers are a comparison to what would happen if forecast climate change did not happen.

Despite rejecting many criticisms in the Dutch report, the IPCC has employed more reviewers for the fifth assessment, and should consider other changes, including paying scientists to make sure every line of the report is scrutinised before it is published, said Parry.

The Dutch government asked the environment agency to investigate the IPCC report after international controversy about two mistakes in the 2007 assessment: the date by which Himalaya glaciers were expected to melt, and a claim that 55% of the Netherlands is below the sea level. The agency report admits this mistake was based on information it provided and says the real figure was 29%.

Dutch agency admits mistake in UN climate report
Arthur Max, Associated Press Yahoo News 5 Jul 10;

THE HAGUE, Netherlands – A leading Dutch environmental agency, taking the blame for one of the glaring errors that undermined the credibility of a seminal U.N. report on climate change, said Monday it has discovered more small mistakes and urged the panel to be more careful.

But the review by the Netherlands Environmental Assessment Agency claimed that none of the errors effected the fundamental conclusion by U.N. panel of scientists: that global warming caused by humans already is happening and is threatening the lives and well-being of millions of people.

Mistakes discovered in the 3,000-page report by the Intergovernmental Panel on Climate Change last year fed into an atmosphere of skepticism over the reliability of climate scientists who have been warning for many years that human-induced emissions of greenhouse gases could have catastrophic consequences, including rising sea levels, drought and the extinction of nearly one-third of the Earth's species.

The errors put scientists on the defensive in the months before a major summit on climate change in Copenhagen in December, which met with only limited success on agreeing how to limit carbon emissions and contain the worst effects of global warming.

The underlying IPCC conclusions remain valid, said Maarten Hajer, the Dutch agency's director. The IPCC report is not a house of cards that collapses with one error, but is more like a puzzle with many pieces that need to fit together. "So the errors do not affect the whole construction," he said at a news conference.

But he said the boiled-down version of the full IPCC report, a synthesis meant as a guideline for policymakers, included conclusions drawn from "expert judgments" that were not always clearly sourced or transparent.

With some conclusions, "we can't say it's plainly wrong. We don't know," and can't tell from the supporting text, Hajer said. The IPCC should "be careful making generalizations."

The IPCC, in a statement from its Geneva headquarters, welcomed the agency's findings, which it said confirmed the IPCC's conclusion that "continued climate change will pose serious challenges to human well-being and sustainable development."

It said it will "pay close attention" to the agency's recommendations to tighten up review procedures.

The Dutch agency accepted responsibility for one mistake by the IPCC when it reported in 2005 that 55 percent of the Netherlands is below sea level, when only 26 percent is. The report should have said 55 percent is prone to flooding, including river flooding.

The mistake happened when a long report was compressed into a short one, and two figures were meshed into one. "Something was lost, and it wasn't spotted," said Hajer.

"The incorrect wording in the IPCC report does not affect the message of the conclusion," that the Netherlands is highly susceptible to sea level rise, the agency's report said. "The lesson to be learned for an assessment agency such as ours is that quality control is needed at the primary level."

The second previously reported error claimed the Himalayan glaciers would melt by 2035, which the Dutch agency partly traced to a report on the likely shrinking of glaciers by the year 2350.

The review, which lasted five months, also found several other errors in the IPCC report on regional impacts of climate change — one of four separate IPCC reports in 2007 — although it said they were inconsequential.

The original report said global warming will put 75 million to 250 million Africans at risk of severe water shortages in the next 10 years, but a recalculation showed that range should be 90 million to 220 million, the agency said.

Another error it found involved the effect of wind turbulence on anchovy fisheries on Africa's west coast.

The Dutch agency said it examined 32 conclusions in the summary for policy makers on the impact of climate change in eight regions.

"Our findings do not contradict the main conclusions of the IPCC," the report said. "There is ample observational evidence of natural systems being influenced by climate change ... (that) pose substantial risks to most parts of the world."

It said future IPCC reports should have a more robust review process and should look more closely at where information comes from. It also recommended more investment in monitoring global warming in developing countries.


Dutch review backs UN climate panel report
Environment correspondent, BBC News Flooding in Holland in 1953 The IPCC erred over data on flooding in the Netherlands, the inquiry says
Richard Black BBC News 5 Jul 10;

A Dutch inquiry into the UN's climate science panel has found "no errors that would undermine the main conclusions" on probable impacts of climate change.

However, it says the Intergovernmental Panel on Climate Change (IPCC) should be more transparent in its workings.

The Dutch parliament asked for the inquiry after two mistakes were identified in the IPCC's 2007 report.

The inquiry is the latest in a series that have largely backed "mainstream" climate science against detractors.

The Netherlands Environmental Assessment Agency (PBL) does not give the panel a completely clean bill of health, however.

Whereas the IPCC's landmark Fourth Assessment (AR4) from 2007 "conclusively shows" that impacts of human-induced climate change are already tangible in many places around the world and will become more serious as temperatures increase, PBL also says the foundation for some of the specific projections "could have been made more transparent".

The Netherlands inquiry adds that the IPCC's summaries tended to emphasise "worst-case scenarios".

However, this was disputed by scientists who had played a leading role in AR4.

"The net impacts of climate change are not beneficial," said David Vaughan, science leader at the British Antarctic Survey, who co-ordinated the AR4 chapter on polar impacts.

Martin Parry, visiting professor at the Grantham Institute for Climate Change Research at Imperial College London who co-chaired AR4 Working Group 2 on climate impacts, welcomed the PBL report.

"We welcome the conclusion of this report, which is esentially that our conclusions are safe, sound and reliable," he said.

"The IPCC is about to venture into the next assessment; so it's important that we learn from these issues, and it's important not to be defensive, and I think that's how the IPCC is approaching things now."

Flood issues

PBL's central remit was to look at how accurately the IPCC's chapters on regional projections of climate impacts reflected the science available, and whether those chapters were adequately distilled into the summary that was given to government representatives for signing off.

The panel has already admitted making a mistake over the projected date for disappearance of glaciers in the Himalayas.

It projected that glaciers could largely disappear by 2035, which the IPCC acknowledges (and the PBL inquiry confirms) is highly unlikely.

A furore erupted in the Netherlands over a different claim in AR4, namely that 55% of the country was vulnerable to flooding because it was situated below sea level.

The source of the claim was a report from PBL itself, which said that 55% of the Netherlands was prone to flooding.

But that report said only 26% of the country was at risk because it lies below sea level, with the remainder affected by river flooding.

PBL now accepts the blame for the mistake lies within its own doors.

"We acknowledge that this error was not the fault of the IPCC... the error was made by a contributing author from the PBL, and the (co-ordinating) lead authors (of AR4 chapters) are not to blame for relying on Dutch information provided by a Dutch agency," it said.

PBL also says it has uncovered another minor error in AR4's summary.

The IPCC said that by the year 2020, between 75 million and 250 million Africans would be at risk of "water stress" (ie not having enough water). PBL says that based on the science available, the figures should be 90-220 million - but that the IPCC projections fit within the "range of uncertainty" in the science.

However, Nigel Arnell, head of the Walker Centre at the University of Reading who led the water chapter in AR4, disputed the PBL assessment.

"The figures are based on a series of assumptions and calculations," he said.

"I think the way in which it was projected with a wide range encapsulated the huge uncertainties, and we think that (narrowing it to) 90-220 million is an over-interpretation of the information that the chapter authors had at the time."

Overall, there were 35 instances where PBL cited errors or made other comments. The vast majority were minor, such as typographical errors or the mis-labelling of a graph; Professor Parry and other IPCC authors accept 12, and have noted them as errata on the IPCC website.

PBL closes its investigation with a number of recommendations for the IPCC, including setting up a public website for the submission of information on possible errors, additional funding for staff to assist with quality control, and taking care with public statements that "could be perceived... as heightening the projected impacts of climate change".

Testing sequence

This is the latest in a series of reviews and inquiries that have endorsed the central conclusions of mainstream climate science, while finding issues of concern in how it is practiced.

Two reviews into issues arising from the emails hacked from the University of East Anglia's Climatic Research Unit (CRU) late last year have concluded there was no evidence of scientific malpractice or manipulation of data.

However, they asked for greater openness, compliance with Freedom of Information law and better collaboration with professional statisticians.

The third review into the CRU issue will be published on Wednesday.

Meanwhile, the Sunday Times was recently forced to apologise for claiming that IPCC projections on die-back of the Amazon rainforest were unsubstantiated.

The main international review of the IPCC - conducted by the InterAcademy Council, a network of national science academies such as the UK's Royal Society - is ongoing, with formal publication due in October.