A distraction of Himalayan proportions

It was one of the most startling predictions in climate science. By 2035 the great glaciers of the Himalayas were supposed to have largely disappeared, threatening the water supplies of tens of millions of people who rely on the ice to feed the great rivers of Asia, from the Indus and the Ganges in the west to the Brahmaputra and the Yangtze in the east.

But the prediction, made by the Nobel Prize-winning body charged with overseeing global climate science, also managed to astonish the scientists who actually knew about Himalayan glaciers. For them, the 2035 timeframe meant that the great slabs of ice sitting on top of these mountains, some of which are hundreds of metres thick, must be melting about 25 times faster than expected – an extraordinary claim.

In science, extraordinary claims demand extraordinary evidence, but in this case the "evidence" turned out to be to be non-existent, which is why the Intergovernmental Panel on Climate Change (IPCC) had to admit this week that, on this occasion, its "well-established standards" of assessing the value of scientific information had failed in relation to the 2035 prediction. The Himalayan glaciers are still melting but not so fast that they are likely to have disappeared by 2035, the panel said.

The error, made by the authors of the IPCC fourth assessment report published in 2007, was to rely on so-called "grey literature" rather than peer-reviewed scientific journals where evidence is double-checked by other experts before publication. But this was only half the story; it is clear the authors made mistakes of their own that compounded the problem.

It began with a news story published in an Indian publication, Down to Earth, in April 1999. It quoted an Indian scientist, Syed Iqbal Hasnain, then vice chancellor of Jawaharlal Nehru University in New Delhi, who said the Himalayan glaciers were receding faster than glaciers in any other part of the world, and if the rate continued they would be gone by 2035.

The story also quoted a respected glaciologist called Vladimir Kotlyakov of the Russian Academy of Scientists, who appeared to back up Hasnain's claim with the quote: "The glacier will be decaying at rapid, catastrophic rates. Its total area will shrink from the present 500,000 to 100,000 square km by the year 2035." (In fact, Kotlyakov was seriously misquoted, but more of this later.)

These claims may not have gone much further had the story not been read by Fred Pearce, a highly experienced and respected environment journalist. Pearce contacted Hasnain to verify his position and, satisfied that the scientist was not misquoted, wrote a version of the story for New Scientist. Hasnain has since told Pearce the claim was "speculative.

Again, the 2035 timeframe may not have emerged from the magazine's archives had it not been recycled in a 2005 report by the environmental body WWF, formerly the World Wildlife Fund. It claimed that the 2035 figure came from the semi-official International Commission for Snow and Ice, written by Hasnain, but it turned out that this commission had never included that timeframe in its reports.

It subsequently turned out that the WWF report, and the 2035 claim, had been based on Pearce's story in New Scientist, and, indirectly, the Down to Earth article of the same year. This was grey literature at its secondhand greyest – a non-peer reviewed report based on another non-peer-reviewed report.

Matters became even greyer when the IPCC began writing up its section on Himalayan glaciers for its fourth assessment report published in 2007. The authors were part of IPCC Working Group II, which is involved in assessing the "impacts" of climate change. Interestingly, it is Working Group I, which is responsible for the science of climate change, which actually has the expertise on glaciers and yet its own report included no mention of the 2035 timeframe.

However, the Working Group II report was quite unequivocal in its statement: "Glaciers in the Himalaya [sic] are receding faster than in any other part of the world and, if the present rate continues, the likelihood of them disappearing by the year 2035 and perhaps sooner is very high if the Earth keeps warming at the current rate. Its total area will likely shrink from the present 500,000 to 100,000 square km by the year 2035."

It included a reference to the 2005 WWF report to support the statement, but only after expert reviewers had questioned the claim during the drafting of the report. Nevertheless, the phrasing is almost identical to Down to Earth's, including the reference to Kotlyakov's glacial shrinkage.

In fact, anyone with a knowledge of the total surface area of Himalayan glaciers would have known that Kotlyakov was not referring to the shrinking of Himalayan glaciers, but to the shrinking of all mountain glaciers, outside the polar regions. More importantly his report, seriously misquoted by Down to Earth, said this shrinkage will occur by 2350 – not 2035.

To add insult to injury, the table of Himalayan glaciers used by the IPCC to back up its statement contains a serious mathematical error, first identified by Professor John Nielsen-Gammon of Texas A&M University. One of the Himalayan glaciers is listed as retreating by 2,840 metres between 1845 and 1966, which is 23 metres per year. Yet the IPCC's report lists it as retreating by 135 metres per year. Whoever did the calculation within the IPCC had divided the total glacier retreat by 21 years, not 121.

It would be easy to condemn the IPCC for these lapses but it must be remembered that the organisation is essentially composed of working scientists and it was the science community that identified and exposed the errors. The 2035 claim was not mentioned in the IPCC's "summary for policymakers" so was not presented as one of its central arguments.

However, such an astounding claim was bound to receive wider currency and doubts about it were initially rejected in the higher echelons of the IPCC. Yet it was through the dogged investigation of ordinary climate scientists that the truth emerged. Professor Graham Cogley of Trent University in Ontario traced the 2035 timeframe back to the 1999 news story in New Scientist and had alerted Pearce.

Professor Cogley said that once the IPCC had been presented with the detailed critique of the 2035 claim, it had moved swiftly to admit the error and make the correction. However, the head of the IPCC, Rajendra Pachauri, when he was confronted over the 2035 claim last year, denounced those questioning his body's research as "voodoo scientists".

Professor Cogley said: "In Working Group I the guidelines are pretty clear about not relying on grey literature unless absolutely necessary. But in Working Group II that rule is more relaxed. As far as the public perception goes, a good question to ask is how many minds will change [ due to this debacle]? The climate sceptics will continue being sceptics. The reality in the Himalayas is bad enough without exaggerating it. The glaciers are losing mass and we are fairly sure they are losing mass faster now than a few years ago. We also know the Himalayan glacier water could become a non-renewable resource," he said.

Professor Jeffrey Kargel, an expert on Himalayan glaciers at the University of Arizona, who was also involved in exposing the error, said that overall the IPCC's fourth report on climate change impacts was otherwise "very solid and very accurate".

"There is indeed a consensus that the glaciers in the Himalayas, as in other parts of the world, are retreating relatively quickly," Professor Kargel said. While most glaciers in the chain are in retreat, others appear stable and a few seem to be advancing because of heavier snowfall – which is not inconsistent with warmer temperatures. The mountains in the east, which accumulate snow in the wetter summer period of the monsoon, are particularly vulnerable to small changes in average temperature, which can quickly turn annual snowfall into annual rain.

The 2035 timeframe may be discredited, but that doesn't mean that there is not a problem. Increasingly, this century, the gradual loss of the Himalayan glaciers are likely to impinge on the reliability of the water supply of people living in the the lower alpine valleys of the mountains.

Professor Kargel said: "When sceptics talk of 'glaciergate', it hurts. That word suggests an elaborate conspiracy when there isn't. This is a self-correcting system, that's what happened, that's what science is."