Scientists recognize the main mistakes in studying the speed of global warming

Scientists behind a major study that Earth's oceans are warming faster than previously thought, say their work contains accidental errors that made their conclusions more certain than they they really are.

Two weeks after the publication of this highly publicized study in the journal Nature, its authors have submitted corrections to the publication. The Scripps Institution of Oceanography, which hosts several of the researchers involved, also noted the problems encountered in the scientists' work and corrected a press release posted on its website. The latter had previously stated that the study explained how the Earth's oceans "absorb 60% more heat than we previously thought.

"Unfortunately, we have made mistakes here," said Ralph Keeling, a Scripps climatologist, co-author of the study. "I think the main lesson is that you work as fast as you can to correct mistakes when you find them."

According to Keeling, the central problem has emerged in the way researchers have treated the uncertainty of their measurements. As a result, the conclusions raise too much doubt to definitively support the conclusion of the document on the amount of heat absorbed by the oceans over time.

The central conclusion of the study – that the oceans retain more and more energy as more and more heat is retained in the Earth's climate system – is consistent with other studies that have led to similar conclusions. And that did not change much despite the mistakes. But Keeling said the authors' miscalculations meant that the margin of error in the results was much greater, meaning that researchers could weigh with less certainty than they thought.

"I take responsibility for what happened because it's my job to make sure that kind of detail is communicated," said Keeling.

The lead author of the study was Laure Resplandy from Princeton University. Other researchers were working for institutions in China, Paris, Germany and the US National Center for Atmospheric Research and Geophysical Fluid Dynamics Laboratory.

"Maintaining the accuracy of scientific data is of paramount importance to us, the publishers, and we recognize our responsibility to correct the errors in the documents we have published," Nature said in a statement to The Post. "The issues related to this document have been brought to the attention of Nature and we are looking at them carefully. We take all concerns related to the articles we have published very seriously and will post an update as soon as further information becomes available. "

The first study, published on October 31, has developed a new method for measuring the amount of heat absorbed by the oceans. The authors essentially measured the volume of gas, in particular oxygen and carbon dioxide, that had fled the oceans in recent decades and were heading towards the warming atmosphere. They found that warming "lies at the top of previous estimates" and suggested that the rate of global warming itself could be more accelerated.

The results, the authors wrote, may suggest that there is less time than expected to reduce greenhouse gas emissions. The study has attracted considerable media attention, including from The Post.

However, shortly after its publication, a UK-based independent researcher, Nicholas Lewis, published a long article on his blog stating that he had discovered a "major problem" in research.

"As far as I know, their method vastly underestimates uncertainty," said Lewis in an interview Tuesday, "as well as a significant increase of nearly 30% in the central estimate."

Lewis added that he tended "to read a lot of articles and that, having both a background in mathematics and physics, I tend to look at them carefully and see how they are doing." They have a meaning. And where they do not make sense – with this one, it's pretty obvious that it does not make sense – I look at them more deeply. "

Lewis has argued in previous studies and commentary that climatologists predict excessive warming because of their use of computer simulations, and that current data from the planet itself suggest that global warming would be less severe than expected.

It is unclear whether the authors agree with all of Lewis's critics, but Keeling said, "We agree that issues similar to those he identified have been identified.

Paul Durack, a researcher at Lawrence Livermore National Laboratory in California, said that quickly recognizing errors in the study "is the right approach in the interest of transparency."

But he added in an e-mail: "This study, although there are other questions, confirms the long-known result that the oceans are warming from record levels and that the rate of warming has increased. ", did he declare.

Gavin Schmidt, director of NASA's Goddard Institute for Space Studies, closely followed the growing debate around the study on Twitter and said that measures relating to heat absorption in the oceans were tainted data problems – and that new research in this area is difficult.

"Of course, you rely on your co-authors and proofreaders to solve most problems, but sometimes things go wrong," writes Schmidt in an email.

Schmidt and Keeling agreed that other studies also supported higher ocean heat content than that of the Intergovernmental Panel on Climate Change (IPCC) in a landmark 2013 report.

Overall, said Schmidt, the episode can be considered positive.

"The key is not whether mistakes are made, but how they are treated – and the answer from Laure and Ralph here is exemplary. Do not panic, but a careful re-examination of their work – despite a somewhat hostile environment, "he wrote.

"So, plus one for some post-publication reviews and one for the authors to re-examine the overall calculation constructively. We will all be wiser. "