Paper that claimed the Sun caused global warming gets retracted
A paper published last June was catnip for those who are desperate to explain climate change with anything but human-caused greenhouse gas emissions. It was also apparently wrong enough to be retracted this week by the journal that published it, even though its authors objected.
The paper’s headline conclusion was that it described a newly discovered cycle in the motion of the Sun, one that put us 300 years into what would be a thousand-year warming period for the Earth. Nevermind that we’ve been directly measuring the incoming radiation from the Sun and there has been no increase to explain the observed global warming—or that there is no evidence of a 2,000 year temperature cycle in the paleoclimate record.
Those obvious issues didn’t stop some people from taking this study as proof that past warming was natural, and only mild and unavoidable warming lies in our future.
What goes up must come down… in a cycle?
The lead author of the paper was Valentina Zharkova, a mathematician and astrophysicist at Northumbria University who has a bit of a track record. If you’ve ever read one of the dozens (hundreds?) of UK tabloid stories declaring that we’re about to start an impending “mini ice age” driven by a declining solar cycle, it was probably supported by a quote from Professor Zharkova. A mini ice age can be difficult to fit into a 1,000 year warming trend, of course, but that didn’t stop Zharkova from publishing her new claim.
Immediately after her paper was published in Scientific Reports—an open access journal among the sprawling family published under the same roof as Nature—criticism of the work started coming in. In fact, much of it is documented in a long comment thread on PubPeer, a site designed to host a sort of post-publication, public peer review. (That truly incredible thread is recommended reading, dear reader.)
The objections started with Ken Rice, a University of Edinburgh astrophysicist and climate blogger. He challenged the paper’s central claim—that the distance between the Earth and the Sun would change as a result of the cycle they were describing. And that’s where things got really wild.
Zharkova engaged in a spirited back-and-forth with Rice that generated more heat than light. Both agreed that the Sun is known to wobble around the precise gravitational balance point of the Solar System, pulled slightly off its mark by the attraction of the larger planets like Jupiter and Saturn. But the study seemed to ignore the fact that the Earth’s orbit also shifts in response to those giant planets, causing it to maintain a constant distance from the Sun. The paper instead assumed that Earth’s orbit was unaffected so that any motion of the Sun would alter its distance from the Earth. If that’s not true, then there has been no change in the strength of sunlight reaching the Earth, and there is no mechanism for their centuries-long warming trend.
As several people tried in vain to point out that this constant Earth-Sun relationship is well-known, Zharkova posted, “Oh dear, You suggest that the Earth does follow in its orbit this solar inertial motion? And its orbit is not stable? You have to have a very vivid imagination assuming that the Earth moves like a drunken men…[sic]”
At one point, after Rice provided a simple orbital simulation calculating the gravitational interactions in the Solar System, Zharkova replied, “Your simulations are extremely biased by the idea you believe in.”
(Zharkova also demonstrates an affinity for an array of arguments against the clear evidence for human-caused climate change, sharing (unprompted) claims that humans are not responsible for increasing atmospheric CO2, challenging the accuracy of global temperature data, and failing to grasp the important difference between local temperature data and global records.)
On Wednesday, Scientific Reports—for which Zharkova is listed as an editor, by the way—formally retracted the paper. The retraction note states that “concerns were raised regarding the interpretation of how the Earth-Sun distance changes over time and that some of the assumptions on which analyses presented in the Article are based are incorrect.” One of the paper’s four authors apparently agreed to retract the paper, while the other three (Zharkova among them) objected.
When contacted by Retraction Watch for their post on this, Zharkova told them, “We consider this retraction by the Editor of Scientific Reports as a shameful step to cover up the truthful facts about the solar and Earth orbital motion reported by the retracted paper, in our replies to the reviewer comments, and in the further papers.”
Correlation is not cause for celebration
In a blog post noting the retraction, NASA climate scientist Gavin Schmidt reflected on the ever-growing catalog of studies claiming to find solar cycles in Earth’s climate. “[T]here has been a long history of people assuming that they ‘know’ that solar cycles have an effect and then just looking ever more deeply for the mechanism,” he writes.
The problem is that if you try enough data sets—of local rather than global temperatures, for example—you can eventually find the cycle correlation you want. Extrapolating that correlation into the future often makes for splashy headlines at outlets that don’t know how to cover science, have a fondness for hype, or both.
Schmidt points out that the predictions never seem to pan out. Math, it turns out, won’t tell you much about the behavior of the climate system without some physics behind it.
That rub is on full display in the PubPeer discussion of this paper. In response to challenges that the study’s Solar System model contradicts physics, Valentina Zharkova repeatedly seems to argue that their correlation is too good to be wrong. Obviously, it doesn’t work that way. Correlations generate testable hypotheses, and some of those hypotheses will undoubtedly be wrong. It could be that the correlation is explained by something else—including the possibility that the correlation is a meaningless quirk of your data set or statistical method.
And in this case, a poorly tested correlational hypothesis was certainly in no position to overturn the positively mountainous pile of physics-based evidence that clearly shows humans have caused modern global warming. No matter how badly some people liked the sound of it.
My Lesson Planning,Sawagi Noise
via Ars Technica https://arstechnica.com
March 7, 2020 at 05:03AM