Climate Feedback

How a computer expert corrected the Met Office

Cross-posted from Daniel Cressey on The Great Beyond

Another day, another fault discovered in the climate change catechism? This time it’s not the usual suspects claiming a grand global conspiracy though. It’s a bit more interesting than that.

At the heart of this story is professional computer programmer John Graham-Cumming, who decided to write some software to number crunch temperature data released by the UK’s Met Office.

“I thought it would be a fun hobby project to use those records to reproduce the worrying charts that show the increase in global temperatures,” he wrote on his blog last week.

“Because I was working with unfamiliar data I put special functions into my program to ensure that I wasn’t making any mistakes. To my surprise these functions began reporting that there was something wrong with temperature data in Australia and New Zealand.”

Once he told the Met Office, he writes, they admitted he was right

The story has been picked up by the Times, which opines that “although the errors do not alter the bigger picture on climate change, they have been seized upon as a further sign that scientific institutions have not been sufficiently transparent”.

For their part, the Met Office says it is grateful to the computer expert for catching their mistake and actually this shows how open they are already being. “We have put an exhaustive amount of information out there to show people exactly what we do,” says Peter Stott, head of climate monitoring and attribution at the Met Office, in the Times.

Graham-Cumming is not a climate sceptic, he tells the Times, “but this does show why the raw data and not just the results should be available”.

Read Graham-Cumming’s story in his own words on his blog, and in an edited version from The Times. Embedded below is a video he made last year showing a program written to analyze and visualize the Met Office data.


  1. Report this comment

    MarkB said:

    “The corrections made have minimal impact on changes at the largest scales. Even in the two regions most affected by the changes the United States and Australasia – the impact is only discernible in the nineteenth century when there are few stations and the published uncertainties are largest. The new version falls within the stated 95% confidence limits of the old version much more than 95% of the time.”

    “I thought it would be a fun hobby project to use those records to reproduce the worrying charts that show the increase in global temperatures”

    Doesn’t look like any change in the “worrying charts”.

  2. Report this comment

    David B. Benson said:

    Still, good to have it fixed.

Comments are closed.