RODNEY HIDE - Raw data that isn't raw: Australia catches up with what we were told to ignore

Australians woke last week to a story I knew well. The Australian's Graham Lloyd reported that a forensic audit had found chunks of the Bureau of Meteorology's "raw" temperature data copied and pasted from one period into another.

The standout example is Sydney Observatory. The minimum temperatures shown for August 1867 are the same as those for August 1907. The paper checked the handwritten logbooks at the National Archives. The 1867 numbers online are simply wrong.

The auditors, Tom Berger and Geoffrey Sherrington, say they found hundreds of red flags. Mr Berger likens the odds to winning the lottery every week for a year.

The Bureau admits errors and will fix them, but won't commit to a full review. The auditors claim far more. Keep the two apart.

What struck me is that the story ran on the front of a national newspaper. Fifteen years ago in New Zealand, asking much milder questions got you labelled a denier.

In November 2009 Richard Treadgold of the Climate Conversation Group collated a short paper with the NZ Climate Science Coalition called "Are We Feeling Warmer Yet?". It took NIWA's graph of New Zealand temperatures, the seven-station series, which showed warming of 0.92C a century, and plotted NIWA's own unadjusted data for the same seven stations. The Coalition's calculation gave a trend of just 0.06C a century. Near enough flat.

The difference was the adjustments. That can be perfectly proper; sites change. But you have to say what you adjusted, by how much, and why.

That was all anyone asked. Show your working.

I asked in person. On the evening of 9 December 2009 Nick Smith called MPs to a briefing at Parliament from NIWA's chief climate scientist, Dr David Wratt. I brought Dr Vincent Gray as my adviser. Nick ordered him out. It was a private meeting of MPs, he said, no outsiders. The outsiders from NIWA stayed. No recording, no notes.

After 25 minutes on the IPCC, I asked about the seven-station graph. When I put it to them that the unadjusted trend was basically flat, they agreed. They explained Hokitika. I said that was what I wanted for every site. Dr Wratt, visibly rattled, told me I should write a paper and get it published in a reputable journal. I didn't need to write a paper. They needed to show their working. They said they would post it all on their website.

I walked out. I was so upset. It was a horrible experience. I loved science and thought scientists did too.

The next day I wrote to Dr Wratt asking when all the adjustments and their rationale would be made public. NIWA published Hokitika. The adjustments traced back to Dr Jim Salinger's 1981 PhD thesis. Science Minister Wayne Mapp told Parliament the computer records behind it were no longer available. As reported at the time, the calculations were lost when Victoria University changed computers in 1983.

I was ACT leader then and a minister. On 25 March 2010 I pointed out in the House that the series had been built at NIWA in the 1990s, not in 1981, and asked how we could conclude anything other than that NIWA had something to hide.

I make no claim about motives. But what scientist loses the records for the most important graph in his country's climate debate?

Salinger worked at the University of East Anglia's Climatic Research Unit from January 1980 to February 1982 and submitted his thesis while there. That unit was at the centre of Climategate, which broke in November 2009, the month before my NIWA meeting. In one leaked email from April 2003, Salinger told Phil Jones, Michael Mann and others he applauded their moves against the journal Climate Research over its editor, Chris de Freitas, adding: "I will ensure that the appropriate sources get to know!" The Bureau's Neville Nicholls was copied in. De Freitas later co-wrote the 2014 reanalysis. And CRU's global record draws on 98 homogenised Bureau series and 13 from NIWA, the New Zealand ones received "through personal communication". To my mind that is a narrow, closed shop, and it has had a devastating effect on public policy.

The graph mattered. As I told Waikato farmers in 2010, it arrived just as the government was weighing carbon taxes and an ETS: "The NIWA graph was most helpful." In the House I said it had been used "for everything from sworn expert evidence to advisory booklets for local government".

To its credit NIWA published a rebuilt series in December 2010. It came out at 0.91C a century, as before. It was peer-reviewed by none other than Australia's Bureau of Meteorology.

The sceptics weren't satisfied. A 2011 statistical audit argued NIWA hadn't applied its own published method properly, and that done correctly the trend was 0.34C. A 2014 peer-reviewed reanalysis by de Freitas, Dedekind and Brill found 0.28C. NIWA scientists replied in the same journal in 2018 saying that figure was "substantially too low". That is how science should work, in the open, and only because people refused to be shouted down.

Credit where it's due. Richard kept the record online, post after post. Barry Brill, a former National MP and under-secretary for energy and science, chaired the Coalition, acted for its trust in court and co-wrote the 2014 reanalysis. Behind them were retired scientists and engineers working unpaid on the numbers. They did the work our science agencies should have done.

The Coalition's trust went to court and lost. In 2012 Justice Venning found NIWA's method credible and internationally recognised, and ordered $89,238.90 in costs. But note what he said: whether there is global warming "is a scientific issue, not suitable for determination by a Court". It declined to referee the science.

David Evans wrote in The Australian that the "trust me" approach must end. Nullius in verba, as the Royal Society puts it: take nobody's word for it.

That is all we ever asked of NIWA. It is all Australians are now asking of their Bureau. The difference is that this time nobody is pretending the question is illegitimate.

I remember back then asking Dr Wratt what their testable hypothesis was. He said Climate Science didn’t work like that. It was thousands of papers and lots of testable hypotheses. It was simple confirmation that none of it is science.

Rodney Hide is a former Minister and ACT Party leader