Climategate software codes should have been available, say MPs

The Professor at the centre of the leaked climate emails scandal has been criticised, alongside other scientists, by MPs who found he had been too secretive over complex research algorithms.

Share

The Professor at the centre of the leaked climate emails scandal has been criticised, alongside other scientists, by MPs who found he had been too secretive over complex research algorithms.

Last November, hackers had accessed a server at Professor Phil Jones’ Climatic Research Unit at the University of East Anglia, revealing emails that were reported to have demonstrated data had been withheld or manipulated to support arguments that climate change was a result of human activity.

But Jones was today largely exonerated by the Commons Science and Technology Committee over his refusal to share scientific data and methodologies in academic research journals. The committee recognised that it was not yet common industry practice to share software codes in publications, but insisted this needed to change.

The codes are typically hugely complex and hundreds of thousands of lines long, the committee acknowledged in its report. While these could not practically be made available in print, it was easy enough to do so on the internet, the committee said. It acknowledged that Jones had recognised scientific practice in this area could be changed for the better.

In another issue on secrecy, Jones had failed to respond to Freedom of Information requests, and the committee criticised him for this. But it laid more blame with the university, which it said should have supported scientists in responding to the requests and should have tackled any perceived culture of secrecy. The committee said it did not consider the FoI matter to be over, and urged the Information Commissioner to take action.

The committee did not assess the validity of Jones’ research, instead leaving that judgement to the Independent Climate Change Email Review and the Scientific Appraisal Panel, both currently conducting their reviews.

But it did highlight potential issues with the software code used to produce Jones’ CRUTEM3 chart, a widely-used dataset on climate change. While some of the code used was similar to that run by the Met Office and available since December, other code was tailored specifically for the CRU research.

John Graham-Cumming, a professional programmer interviewed by the committee for the report, strongly criticised the CRU code, saying he had found “easily-identified bugs, no visible test mechanism, [code that] was not apparently under version control and was poorly documented”. He added: “It would not be surprising to find that other code written at the same organisation [the CRU] was of similar quality”.

“What this inquiry revealed was that climate scientists need to take steps to make available all the data that support their work and full methodological workings, including their computer codes,” said MP and committee chairman Phil Willis. If the data and computer codes had been available, he said, “many of the problems” at Jones’ Climatic Research Unit “could have been avoided”.

Find your next job with computerworld UK jobs