Last week, Judith Curry had a guest post about causality and climate. I was initially a little confused, because I thought it was about a paper published in Proceedings of the Royal Society A, that I’d discussed in an earlier post. However, even though this paper was mentioned in the post, the post is actually about a more recent paper that uses the same method and draws the same conclusion.
The Proceedings of the Royal Society A paper applied a stochastic approach to causality to a number of case studies, including looking at [a]tmospheric temperature and carbon dioxide concentration. The newer paper focusses only on the relationship between atmospheric temperature and concentration of carbon dioxide. The key results are:
changes in CO₂ concentration cannot be a cause of temperature changes.
and
All evidence resulting from the analyses of the longest available modern time series of atmospheric concentration of [CO₂] at Mauna Loa, Hawaii, along with that of globally averaged T, suggests a unidirectional, potentially causal link with T as the cause and [CO₂] as the effect.
In other words, the claim is that the increase in atmospheric CO2 is driven by increasing temperatures, and that this increase in atmospheric CO2 is not the cause of the increase in temperature.
As most of the regulars here will know, this is clearly nonsense. It’s virtually certain that the recent increase in atmospheric CO2 is driven by anthropogenic emissions. There are multiple lines of evidence that support this conclusion. It’s also unequivocal that human influences have warmed the atmosphere, ocean and land.
When the Proceedings of the Royal Society A paper came out last year, I emailed the editor to point out that they’d published a paper with a rather nonsensical result. I didn’t get a response. However, I was cc’d into a response to someone else who had also complained. This response was, unfortunately, rather dismissive and somewhat insulting.
The response said that the criticism had been discussed with the board member and subject editor who handled the paper. According to them, the criticism misinterpreted what was in the paper and was not well-founded. Apparently, the result was scientifically intriguing and would be of interest to many of their readers.
The latter may be true. However, a paper suggesting that the recent rise in atmospheric CO2 is not due to anthropogenic emissions and that this rise in atmospheric CO2 is not the cause of global warming is not scientifically intriguing, it’s simply wrong. The board member and subject editor is either being disingenuous, or understands this so poorly, they probably should not be a subject editor for the journal.
It’s one thing for a paper like this to slip through. Peer review isn’t perfect and clearly papers that aren’t very good do get published. It’s another thing, however, to double-down when people point out the problem. Today, it is virtually certain that the increase in atmospheric CO2 is due to anthropogenic emissions, and that this increase is driving global warming. A paper using a purely statistical causal method is not going to overthrow these extremely well-established results.
It’s one thing to see a paper making these claims appear in an MDPI journal, but you’d hope that the Royal Society would have higher standards for their journals. In my view, rather than regarding the results in the paper as being scientifically intriguing, the Royal Society should really be embarassed that this paper was published in the first place.