Many of our readers will no doubt be aware of the long-standing dispute between Steve McIntyre and members of the climate science community whose data McIntyre is keen to get hold of.
For those of you less familiar with the story, here’s some background. McIntyre, who runs the Climate Audit blog, is best known for questioning the validity of the statistical analyses used to create the ‘hockey stick’ graph. The ‘hockey stick’ is the graph that illustrates the past 1000 years of climate based on palaeo proxy data and was published by Penn state climatologist Michael Mann and co-authors in Nature back in 1998.
More recently, McIntyre has turned his attention to criticizing the quality of global temperature data held by institutes such as NASA’s Goddard Institute of Space Studies. Several organizations worldwide collect and report global average temperature data for each month. Of these, a temperature data set held jointly by the Climatic Research Unit (CRU) at the University of East Anglia and the UK Met Office’s Hadley Centre in Exeter, known as HadCRU , extends back the farthest, beginning in 1850.
Since 2002, McIntyre has repeatedly asked Phil Jones, director of CRU, for access to the HadCRU data. Although the data are made available in a processed gridded format that shows the global temperature trend, the raw station data are currently restricted to academics. While Jones has made data available to some academics, he has refused to supply McIntyre with the data. Between 24 July and 29 July of this year, CRU received 58 freedom of information act requests from McIntyre and people affiliated with Climate Audit. In the past month, the UK Met Office, which receives a cleaned-up version of the raw data from CRU, has received ten requests of its own.
I’ve reported the full story for this week’s Nature, but here’s a breakdown, plus some details that didn’t make the cut.
Why does McIntyre want the data?
Given McIntyre’s track record in critiquing data that comprise a significant part of the evidence for global warming, one might wonder whether he is in fact interested in having a go at reproducing the global temperature record. But McIntyre insists hat he’s not interested in challenging the science of climate change, or in nit-picking; rather he is simply asking that the “data be made available”.
Why won’t Jones give McIntyre the data?
Jones says that he tried to help when he first received data requests from McIntyre back in 2002, but says that he soon became inundated with requests that he could not fulfill, or that he did not have the time to respond to. He says that, in some cases, he simply couldn’t hand over entire data sets because of long-standing confidentiality agreements with other nations that restrict their use.
Although Jones agrees that the data should be made publicly available, he says that “it needs to be done in a systematic way”. He is now working to make the data publicly available online and will post a statement on the CRU website tomorrow to that effect, with any existing confidentiality agreements. “We’re trying to make them all available. We’re consulting with all the meteorological services – about 150 members of WMO – and will ask them if they are happy to release the data”, says Jones. But getting the all-clear from other nations could take several months and there may be objections. “Some countries don’t even have their own data available as they haven’t digitized it. We have done a lot of that ourselves”, he says.
Are there likely to be ‘holes’ in the data’?
Everyone agrees that raw station data are imperfect; that’s why they are cleaned up before being handed over to the UK Met Office. Jones says that existing issues include stations being relocated without being renamed, but he emphasises that these minor errors do not affect the global temperature trend, because there are thousands of individual stations collecting data worldwide at any one time.
McIntyre says that he does not expect to find any major errors in the data. But he also believes that too few resources are put into quality checking climate data, and that independent professional statistical services should be employed to check the data. Any thoughts on who might offer such services?
What was the CRU ‘data purge’ about?
A couple of weeks ago, it became clear that McIntyre had in fact retrieved some of the HadCRU data from a server on the CRU website. On realizing this, CRU immediately removed the data from their website, leading to speculation about a CRU ‘data lockdown’ over on Climate Audit. It transpires, however, that these data were on an anonymous ftp server intended for Met Office Hadley Centre project partners only, and were not for public use.
Given that McIntyre’s wish for access to the data will take time to be granted, this dispute will likely continue for some time. He’s especially aggrieved by the fact that hurricane expert Peter Webster at Georgia Tech University was recently provided with data that had been refused to him. McIntyre’s point here is that he should be treated as a legitimate academic given his background and publication record.
But Webster points out that he was allowed access because of the nature of his request, which was very specific and will result in a joint publication with Phil Jones. “Reasonable requests should be fulfilled because making data available advances science”, says Webster, “but it has to be an authentic request because otherwise you’d be swamped".
Once the data become publicly available, Jones wants McIntyre to produce a global temperature record. “Science advances that way. He might then realize how robust the global temperature record is”, says Jones. Asked if he would take on the challenge, McIntyre said that it’s not a priority for him, but added “if someone wanted to hire me, I’d do it”.