Global warming is really happening — really. There was no conspiracy or cover-up. Peer review did not fail and the scientists who have spent decades working out the best way to handle and process data turned out to know how to handle and process data after all. Thank you Berkeley Earth Surface Temperature (BEST) study.

Four papers released by the BEST team at the University of California, Berkeley, last week are of undoubted interest to the media, given that they support what is portrayed as the mainstream scientific position on climate change. They could also find traction in politics, especially in the United States, where they could be used to combat the assertions of Republicans, who have effectively tossed climate science away. But the headline scientific conclusion, that a century and a half of instrumental measurements confirm a warming trend, is, well, all a little 1990.

Of course, reproduction of existing results is a valid contribution, and the statistical methods developed by the BEST team could be useful additions to climate science. But valid contributions and useful additions alone do not generate worldwide headlines, so the massive publicity associated with the release of the papers (which were simultaneously submitted to the Journal of Geophysical Research) is a curious affair.

There was predictable grumbling at the media coverage from within the scientific community, which saw it as publicity in lieu of peer review. Reporters are more than happy to cover the story now, while it's sexy, but will they cover it later, when the results are confirmed, adjusted or corrected in accordance with a thorough vetting? The short answer is no, many of them will not. Barring an extraordinary reversal of message, the wave of press coverage is likely to be only a ripple when the papers are finally published. And this is what upsets the purists: the communication of science in this case comes before the scientific process has run its course.

Members of the Berkeley team revelled in their role as scientific renegades. Richard Muller, the physicist in charge, even told the BBC: "That is the way I practised science for decades; it was the way everyone practised it until some magazines — particularly Science and Nature — forbade it."

This is both wrong and unhelpful. It is wrong because for years Nature has explicitly endorsed the use of preprint servers and conferences as important avenues for scientific discussion ahead of submission to this journal, or other Nature titles. For example, on page 493 this week we publish a paper that discusses the dwarf planet Eris, based on results that the lead author presented (with Nature's knowledge and consent) at a conference several weeks ago. Journalists are, of course, welcome to report what they come across in such venues — as several did on Eris. What Nature discourages is authors specifically promoting their work to the media before a peer-reviewed paper is available for others in the field to read and evaluate.

Muller's statement is unhelpful because such inflammatory claims can only fuel the heated but misguided debate on climate-sceptic blogs and elsewhere about the way science works and how it treats those who insist on viewing themselves as outsiders.

To solicit input on results before publication is not a guerrilla action against a shadowy scientific elite. Witness the posting on a preprint server last month of the paper reporting neutrinos that apparently travel faster than light: the authors made it clear that they were seeking help from the wider community to explain the findings, and the media stories (if not the headlines) mostly reflected that. To pretend otherwise can only erode public trust in science, as it is practised by all.