Debaters Behaving Badly, Part 6 - Misrepresenting Scientific Sources

This is Part 6 in a series on Debaters Behaving Badly. In this post, I'm considering the rampant misuse of scientific papers in public discourse, whether intentionally or unintentionally.

Monckton vs Lambert 2010

In Sidney on February 12, 2010, Christopher Monckton participated in a debate with Tim Lambert about climate change. The debate is on YouTube in 15 parts, and when referencing it, I'll link to the part of the debate that I'm referring to. In this debate, Monckton argued that satellite measurements have uncovered a global brightening from 1983 to 2001 that demonstrates natural forcings dominate climate influences, and sensitivity to CO2 must therefore be extremely small. That claim of Monckton came from some calculations made from a paper by Rachel Pinker published in 2005[1]. In the debate, he took a value for what he calls "cloud forcings" at 3.05 W/m^2 from 1983-2001 and added to it both CO2 and "other" external forcings, then came up with a total forcing of 3.85 W/m^2, and since there was 0.45°C warming during that time, he performed a simple calculation based on the following formula:

ΔT = λ*ΔF

0.45 = λ*3.85
λ = 0.12°C/W/m^2

For comparison, he cites the IPCC's value for λ to be 0.88 C/W/m^2. From there he calculated the equilibrium sensitivity for doubling CO2 given this value of λ and a 2xCO2 forcing of 3.71 W/m^2. His "ball park" back of the envelope calculation was ECS = 0.12*3.71 = 0.43°C. Since he says the IPCC estimates ECS to be 3.26°C, this means the IPCC's figure for ECS is exaggerated by 7.5x. Monckton then says he had some mathematicians look at his calculations, and one of them said the IPCC's estimate is actually 8x exaggerated. He then argues that if you take away this exaggeration, you take away the problem. And of course, if it's true that ECS is only 0.43°C, then there is no problem.

Astute readers, however, will already note problems here, the most obvious of which is that ECS is an "equilibrium" change in temperature, and the 0.45°C value for ΔT is not an equilibrium change in temperature. Monckton's calculation does not include the current energy imbalance, which is essential for any estimate of ECS, so the calculation is wrong, even if we were to accept this 3.05 W/m^2 from Pinker as being relevant to the calculations. But Monckton also misused Pinker's paper, and Lambert was on his game.  This was a point that Monckton had made before, and anticipating that he would do so again, Lambert came to the debate prepared. 

Monckton used this to get his 3.05 W/m^2 Value

The first thing he did was to play audio of Monckton praising Pinker for being a good scientist, complete with Monckton implying that he knew Pinker's motives while not realizing that Rachel Pinker is a woman, not a man. He plays Monckton saying this:

Now this paper was written in 2005 by a satellite nerd named Pinker. He is wholly unconcerned with with the global warming debate. It's as though he lives on another planet. He lives for verifying weather satellites and doing their job. That's what Pinker is all about.

Lambert had written Pinker and asked for a response to Monckton's use of her paper. Then he played audio of a recorded transcript of Rachel Pinker's response to Monckton's use of her paper.

The CO2 "radiative forcing" value that Mr. Christopher Monckton is quoting refers to the impact on the Earth's radiative balance as described above. The numbers that we quote in our paper represent the change in surface shortwave radiation due to changes in the atmosphere (clouds, water vapor, aerosols). These two numbers cannot be compared at their face value. To the best of my understanding this is the source of the misunderstanding. Our work was properly interpreted in the latest IPCC report (2007).

Through Pinker's response, Lambert showed that Monckton was comparing apples to oranges, and it was inappropriate for him to simply add 3.05 W/m^2 to radiative forcings used by the IPCC. Here's what Monckton seems to have done, and it appears to be based entirely on the abstract of Pinker's paper and one of her graphs (shown above). Pinker writes, "We observed an overall increase in S [solar radiation at the surface] from 1983 to 2001 at a rate of 0.16 watts per square meter (0.10%) per year." So over a 19 year period, at a rate of 0.16 W/m^2/year, that's 0.16 = 3.05 W/m^2 increase in solar radiation. That increase can be caused by many things, including changes in aerosols or changes in cloud cover, and even water vapor. And there is no indication that it's all natural, as he claims. It will be affected by decreases in anthropogenic aerosols and cloud feedbacks from warming due to anthropogenic CO2. What's clear from Pinker is that this value is NOT an external forcing that should be included in a back of the envelope calculation for ECS. In the paper, Pinker even writes, "the potential climatic effects of a sustained decrease or increase in surface solar radiation require extensive investigation of additional factors, such as the long-wave radiative effects of any associated cloud variation." Cloud feedbacks are complex, though in general terms high clouds and low clouds work in opposite directions - high clouds reflect light back to space, so a decrease in high clouds would let more light reach the surface. Low clouds contribute to the greenhouse effect. The net cloud feedback is more likely positive than negative, but this is still a feedback in response to warming. Pinker of course correctly identified Monckton's error.  Monckton was exposed for not understanding the papers he read, even to the point of not realizing that Rachel Pinker is a woman.

Congressional Testimony

Perhaps we can chalk this up to an isolated incident where Monckton made an honest mistake and simply lacked the competence to understand what Pinker was writing. However, undaunted by this correction to his misreading of Pinker's paper, on May 6, 2010, just a couple months later, Monckton offered testimony before Congress, in which Monckton essentially repeated his assertions about Pinker's paper.

What, then, caused the third period of warming? Most of that third and most recent period of rapid warming fell within the satellite era, and the satellites confirmed measurements from ground stations showing a considerable, and naturally-occurring, global brightening from 1983-2001 (Pinker et al., 2005). 

Allowing for the fact that Dr. Pinker’s result depended in part on the datasets of outgoing radiative flux from the ERBE satellite that had not been corrected at that time for orbital decay, it is possible to infer a net increase in surface radiative flux amounting to 0.106 W m–2 year–1 over the period, compared with the 0.16 W m–2 year–1 found by Dr. Pinker. 

Elementary radiative-transfer calculations demonstrate that a natural surface global brightening amounting to ~1.9 W m–2 over the 18-year period of study would be expected – using the IPCC’s own methodology – to have caused a transient warming of 1 K (1.8 F°). To put this naturally-occurring global brightening into perspective, the IPCC’s estimated total of all the anthropogenic influences on climate combined in the 256 years 1750-2005 is only 1.6 W m–2.

Monckton appears to have simply "updated" Pinker's value of 0.16 W/m^2/yr to 0.106 W/m^2/yr and shortened the the time from 1983-2001 to 18 years (before it was 19 years) and did essentially the same calculation: 0.108*18 = 1.9 W/m^2. Then he notes that this number (which Pinker has already explained is not comparable to external forcings used to estimate ECS) exceeds the 1.6 W/m^2 totaling "all the anthropogenic influences on climate combined in the 256 years 1750-2005." This is essentially the same mistake that Monckton made before with slightly different numbers. This again prompted a response from Rachel Pinker:

This statement in effect equates temperature change with surface solar radiation change which... is only one input into a complex climate process. Also, it is not necessarily the case that global brightening is naturally-occurring; it can be caused by anthropogenic aerosols or changes in the atmospheric moisture content as well as clouds, possibly affected by increasing CO2 levels.

At this point, the benefit of the doubt, at least for me, is gone. Monckton is presenting himself as having sufficient expertise to examine the scientific literature and assess whether AGW is a problem. He was given conclusive evidence that he failed to do so. But rather than correcting his mistake, he redressed it and offered it again as public testimony in an effort to affect policy. We cannot have any confidence that Monckton is even attempting to present the scientific literature honestly.

To this point, Monckton has also wisely advised us against believing him or taking him at his word. On October 14th, 2009, he gave a public speech in Minnesota in which he declared,

One point I do want to make is that you must not believe a word I say. I am NOT Al Gore. I am NOT going to tell you what the truth about the climate is. I am simply going to tell you a series of facts from the science and the data and the peer-reviewed literature, and I am going to allow you to draw the conclusion for yourselves that there is no problem with the climate. I am NOT here to proselytize or preach; I do not expect you to believe me, because science is not a belief system. Science is a rigorous process of inquiry.

Indeed. Monckton's point here appears to be that we shouldn't have to believe him because the evidence in the literature is clear, and the evidence doesn't care about whether or not we believe it. I couldn't agree more. A major component of skepticism is that we shouldn't just believe what we're told, but we should check up on claims to see if they are accurate. Doing so confirms that Monckton is flatly contradicted by the literature he uses in support of his claims.

Not Just Monckton

This is becoming an increasing problem in public discourse about climate change. Many contrarians want to show that reputable scientific evidence is on their side, so they scour through the peer-reviewed literature looking for papers that they can weaponize for their ideological cause. This frequently reduces to quote mining papers looking for statements that can be taken out of their context to give the impression that the paper supports conclusions that the paper simply doesn't endorse. A blog by the name of No Trick Zone specializes in this trick and frequently makes posts listing dozens to even hundreds of papers that are reported to support whatever claim NTZ wants to support. The blogposts frequently contain quotes lifted from the paper or screenshots of papers with highlighted text that can be construed as supporting NTZ's claim. If you click on the links and read the papers, though, you are almost certainly going to find a few contrarian papers published in predatory or semi-predatory journals and a lot of papers that simply do not conclude what NTZ wants you to believe. These papers can then be cited by contrarians as "proof" that there is no climate emergency, that the medieval warm  period was warmer than today or any number of claims.

Lucas Zeppetello describes this happening to one of his papers.[2] I don't know if this paper was featured at NTZ, but at any rate, somebody found it, promoted it on Twitter, and then it became a popular "weapon" against climate science, but it was based entirely on a misreading of the paper. He writes, 

My article’s attention score, an indicator of interest in a paper, was in the top 5% of all research tracked by Altmetric, largely because of a plethora of Twitter posts. Initially curious, I quickly became mortified as I scrolled through the page and found that something was wrong. Accounts with thousands of followers were using my article to substantiate their arguments that climate change is a hoax perpetrated by the scientific establishment.

This was Zeppetello's first academic paper published in the peer-reviewed literature. While his response may seem to be a bit extreme (mortified?), still I can understand the reaction. He wrote a paper that made a positive contribution to climate science only to have it misused as a weapon against the science he's contributing to, and using it to call that science a "hoax." Zeppetello continues:

In our abstract, my coauthors and I wrote that “surface downward longwave radiation is tightly coupled to surface temperature; therefore, it cannot be considered an independent component of the surface energy budget.” The community that promoted our paper on Twitter mistakenly took that to mean that global temperature change cannot be caused by changes in the longwave portion of the planetary energy budget, thus furthering their argument that greenhouse gas concentrations do not contribute to climate warming.

Is this a logical interpretation? Absolutely not. A strong feedback between surface temperature and downwelling longwave radiation does not mean that carbon dioxide emissions have no impact on the longwave component of the planetary energy budget. In fact, just the opposite is true: The strong feedback means that greenhouse emissions can have a substantial effect on longwave radiation. Yet I found myself scrolling through pages of posts jeering at climate scientists and dismissing science as politically motivated propaganda.

But this kind of behavior from many contrarians comes with the territory nowadays. I suspect a good portion of it comes from people reading papers that they do not sufficiently understand, but this does not explain all of it. As Monckton aptly demonstrates, some people apparently don't care if what they say about the peer-reviewed literature is true. All that matters is if whether they can weaponize the paper to make their point.

Conclusion

The root cause among those making honest mistakes, I think, is viewing the literature exclusively from from an apologetic point of view. If you have an ideological objection to climate science, and you lack the scientific literacy to understand the literature, it's easy to fall into the trap of assuming your conclusion and then using quotes from papers as "proof" of your conclusion. If you don't have the literacy to understand what you quoted or its context, you can easily misread papers to support your desired conclusion. And then you can just force the paper to support the apologetic interest you had before reading the paper anyway. 

I had an interesting example of this recently in Facebook conversation. To protect the person involved (who doesn't know I'm writing this), I won't name the person or link to my source. But discussions like this are rampant, and it won't be hard to find other similar examples. This discussion had to do with a recent paper[3] about the effect of AMOC on carbon sinks and offered new constraints on projected warming from climate models. This person describes this constraint as an "emerging phenomenon" (EP) and argued that this "constraint" explains why models predict twice as much warming as observations (they don't, but that's another post). His reading of the paper depended on the confusion of restraining uncertainty with reducing warming:

The effect of this EP is actually very large, and is estimated to be between about 30% to as much as 53% cooler than previous estimates.

Except that's not at all what the paper said. Here's a quote from the paper illustrating the mistake:

After incorporating the emergent constraint based on the observed SSS (Methods), the NH surface temperature warming is reduced from 5.9 ± 1.4 °C to 5.5 ± 1.0 °C (Fig. 4c) and the cumulative NA carbon uptake is increased from 50.1 ± 10.7 PgC to 54.7 ± 5.1 PgC under SSP5–8.5 (Fig. 4d). The uncertainties in future NH warming and cumulative carbon uptake are reduced by 30% and 53%, respectively.

The person I was debating didn't know what the word "constraint" means. The quote has do with constraints that reduce uncertainties in model projections. The mean shifted down a little (about 6-7%), but that's not the point. The point was that uncertainties were reduced for future NH warming (by 30%) and for cumulative carbon uptake (by 53%). The effect of this reduction of uncertainty is that warming of <3.5°C and >6.5°C are less likely. But warming within the range of 4°C and 6°C are therefore more likely.

I suspect this began as an honest mistake (though he never changed his tune on this point when I pointed it out). But it appears to have been motivated by an apologetic interest to use this paper to show that climate science, and in particular climate models, has been exaggerating projected warming with respect to observations. It was likely the apologetic interest (combined with ignorance) that lead to the misreading of the paper. I think we can mitigate this problem by putting apologetic interests aside when reading a paper. If you instead are concerned to understand the paper you read as accurately as possible and learn the physics that undergirds it, then understanding can become your primary concern. If a paper is published that is based on sound scientific evidence and still challenges AGW in some way, great. Let's discuss it. But understanding needs to come first before using papers as "proof" that everyone who understands the science better than you is wrong.

References:

[1] R. T. Pinker et al. ,Do Satellites Detect Trends in Surface Solar Radiation?.Science308,850-854(2005).DOI:10.1126/science.1103159

[2] Zeppetello, L. V. (2020), Don’t @ me: What happened when climate skeptics misused my work, Eos, 101, https://doi.org/10.1029/2020EO140098. Published on 17 February 2020.

[3] Park, IH., Yeh, SW., Cai, W. et al. Present-day North Atlantic salinity constrains future warming of the Northern Hemisphere. Nat. Clim. Chang. 13, 816–822 (2023). https://doi.org/10.1038/s41558-023-01728-y