Tuesday, September 20, 2016

Scientists published climate research under fake names. Then they were caught.

By Ben Guarino

September 19

The scientists briefly known as Den Volokin and Lark ReLlez thought they had found something big.

It was not data wrung from a clever experiment or a lucky field observation. Instead, the pair had constructed a model, a mathematical argument, for calculating the average surface temperature of a rocky planet. Using just two factors — electromagnetic radiation beamed by the sun into the atmosphere and the atmospheric pressure at a planet’s surface — the scientists could predict a planet’s temperature. The physical principle, they said, was similar to the way that high-pressure air ignites fuel in a diesel engine.

If proved to be the case on Earth, the model would have dramatic implications: Our planet is warming, but the solar radiation and our atmosphere would be to blame, not us.

[I don't know what to make of this report/study. On the one hand, it seems to suggest non-anthropogenic climate change. But there is a critque at the end of this article by a NASA researcher. So not definitive.]

“That there is a simple analytic tool [for planetary temperature], based on simple parameters — that’s a valid and interesting point,” astrobiologist David Grinspoon told The Washington Post. Grinspoon, a researcher at the Tucson-based nonprofit Planetary Science Institute, was not involved in the research.

There was a problem, though. The paper, Grinspoon said, “seems to have an agenda.”

There was another problem, too. The names Den Volokin and Lark ReLlez are fake.

The pair thought their mathematics, if accurate, could topple conventional models. As the paper’s authors argued via a 50-page manuscript, the temperature model worked for known temperatures of rocky heavenly bodies such as Mars, the moon and Venus.

And, crucially, it worked for Earth.

But before they could shift anyone’s paradigms, the authors would have to publish their study. Therein lay an obstacle: In 2011, the pair’s previous work sparked sometimes intense — and sometimes misunderstood, they said — discussion in the climate-skeptic blogosphere. To would-be publishers, that association was deadly.

As Volokin wrote to The Washington Post in a statement prepared as a response to this article, “journal editors and reviewers would reject our manuscripts outright after Googling our names and reading the online discussion.”

So the scientists submitted their model under fake names. “We wanted pseudonyms that could relatively easily be linked to our true identifies if needed in the future,” Volokin said, though the names also had to fool a peer reviewer. The pseudonyms succeeded on both counts.

The pair had used the same fake names once before, in fact, for the journal SpringerPlus in 2014. This second pseudonymous temperature study passed review, too. On Aug. 18, 2015, the journal Advances in Space Research published Volokin and ReLlez’s study online.

Within months after publication, the pseudonyms fell apart.

The journal withdrew the paper, as the website Retraction Watch reported earlier in September. In place of the study, a statement now reads: “This article has been withdrawn upon common agreement between the authors and the editors and not related to the scientific merit of the study. The Publisher apologizes for any inconvenience this may cause.”

Pascal Willis, the editor in chief of Advances in Space Research, declined to elaborate to The Washington Post. “I think that the explanation on the journal Web page is sufficient,” he said.

The scientists came clean to Willis in September 2015. “We have been forced by unfortunate circumstances to use pseudonyms,” Volokin wrote to the editor, in an email dated Sept. 27 that the scientist provided to The Post.

“Den Volokin and Lark ReLlez are nothing more than our actual names spelled backwards.”

Behind the pseudonyms were Ned Nikolov, a physical scientist, and Karl Zeller, a retired meteorologist. Both worked together at the U.S. Forest Service. The withdrawn study received no federal funding or support, Nikolov told The Post, or any external support at all. The pair worked independently and in off hours.

Nikolov, a.k.a. Den Volokin, does not consider himself to be a climate skeptic. He does not deny the climate is changing.

For years, he said, he aligned with the mainstream scientific consensus: Global warming is real and humans are to blame. At least 97 percent of working climate scientists — as well as the American Association for the Advancement of Science, the Intergovernmental Panel on Climate Change and other leading government and research organizations — agree that the planet is undergoing human-caused climate change.

Later in his career, Nikolov had a change of heart. In 2009, hackers released thousands of climate researchers’ emails in a scandal ultimately named Climategate. Deniers and skeptics pointed to certain emails and cried foul. But formal investigations found neither scientific wrongdoing nor a global-warming conspiracy.

Nikolov’s convictions, though, were shaken. He decided to investigate climate-change data the way he knew how, as a physicist. Out of that work came the model he published with Zeller, his former boss at the U.S. Forest Service. (Together, the pair had published studies in 2003 and 2005 under their real names.)

Along with that model, too, came the corollary conclusion — the fault was our star’s radiation and our atmosphere, not the carbon emissions that the vast majority of researchers link to climate change.

With that conclusion came another sort of discovery: the alienation and frustration of working from the scientific fringe.

“There is no doubt that trying to publish research results, which do not conform to accepted theories or mainstream beliefs, poses a challenge in today’s world of academic political correctness,” Nikolov said. “This is not just our experience, and it is not just happening in climate science.”

In defense of pseudonymity, Nikolov cited a paper by the neuroscientist and writer known only as NeuroSkeptic that details the long history of such research. In a famous 20th-century case, William Sealy Gosset, a chemist employed by the Guinness Brewery and barred from publishing research outside his job, penned papers under the name Student. (The Student’s t-test, a pseudonymous Gosset creation, is an important statistical method still used today.)

What’s more, Nikolov and Zeller argued, was that the model stood on its own. “We think that the scientific merit of a research article should ultimately be judged by its content rather than authors’ identity,” Nikolov wrote to Willis. “After all, Den Volokin and Lark ReLlez are as good names as Ned Nikolov and Karl Zeller as far as the quality of science is concerned.” The paper is currently under review at another publication, Nikolov said.

But many disagree that scientists should have carte blanche to use pseudonyms.

“I don’t think that anonymity is a sort of guarantor of objectivity” Julian Reiss, a Durham University philosopher who writes about scientific objectivity, said. Author names and affiliations can act as signals, he said, to other readers for the validity of an argument.

According to Agriculture Department guidelines, papers written by authors who identify as employees should be subject to an internal “peer-review process where applicable.” Nikolov and Zeller, as Volokin and ReLlez, listed their affiliation as Tso Consulting, a company owned by Zeller’s daughter.

Nikolov argued that the pseudonyms are a way to disarm bias trapped in the publishing process. That the peer-review process is flawed is a critique that has existed as long as the system itself. It is still around. In 2015, for instance, a group of female biologists had their paper rejected because they lacked a male co-author.

To that end, journals such as Nature now allow researchers to submit manuscripts through an optional double-blind peer review. (In traditional, single-blind peer review, scientists do not know who critiques their papers. In double-blind review, neither the study authors nor the reviewers know each other’s identities.) But even in double-blind review, real names are ultimately attached to the published paper.

“It’s not appropriate for authors to give false names in a submission without the express permission of the editors,” Virginia Barbour, chair of the Committee on Publishing Ethics, said by email. Only in a few exceptions should pseudonymous or anonymous work be allowed, she said. For instance, health workers may be in danger if they publish science from within a repressive regime. “However, in any such circumstance the editors have to have assurance about the identity of the authors.”

The pseudonyms further unraveled with a tweet from NASA researcher Gavin A. Schmidt, which alerted Retraction Watch to the withdrawal. “Top tip for climate contrarians: When you submit nonsense papers to journals,” Schmidt wrote, “spell your name backwards so no one knows who you are.”

The withdrawn study “is just a curve-fitting exercise of five data points using four free parameters and as many functional forms as they could think of,” Schmidt, an expert in atmospheric climate modeling, said in an email. Like the previous pseudonymous research, “it too has nothing fundamental to add.”

He added, “The authors’ insistence that they are ‘contradicting mainstream theory’ is just delusional self-aggrandizement.”

Grinspoon, too, said the model does not invalidate decades of research into Earth’s atmospheric science. “I don’t think they’ve made this case,” he said. “Certainly not enough to rearrange everything we know about climate.”

There were other red flags embedded within the study. Nikolov and Zeller recalculated Mars’ pressure and temperature data, in lieu of using the “known data for Mars that people had been carefully studying for decades,” Grinspoon said. “If they hadn’t, their model would not have worked quite as well.”

Rather than aiming to be a universal paradigm-buster, Grinspoon said the study is better served as a handy mathematical approximation. “It’s a kind of clever, back-of-the-envelope way to calculate planet temperatures,” Grinspoon said. Should scientists find themselves with limited exoplanet data, something like the Volokin and ReLlez model could be a simple way to approximate distant temperatures.

As such, Nikolov and Zeller have positioned themselves within a burgeoning scientific discipline. There is a relatively new effort underway for scientists to apply Earth climate models to other planets. For instance, Grinspoon recently took a climate model traditionally applied to our planet and ran simulations on Venus. The results were a compelling case that 1 billion or 2 billion years ago, Venus may have been habitable.

Rather than pseudonyms or even double-blind reviews, Grinspoon called for a more open pursuit of climate science. He said he would love to see Nikolov and Zeller — presumably, under their own names — welcomed at a conference of planetary scientists. “People would want to talk to them,” he said. But the “sneaky” approach “doesn’t promote that kind of collegiate exchange of ideas.”

No comments: