CO2 Coalition's Most Recent Political Tract from Happer and Lindzen

A few years ago, I responded to a political "comment" from the CO2 Coalition in which Happer, Lindzen (and potentially ghost writers) made a series of self-contradictory claims about climate sensitivity. Recently I became aware of another political tract in which they simplified their argument tremendously. They've also hidden almost all their math, making it more difficult to check up on what they say. However, there is plenty we can say about this that shows that they are still actively changing their opinions, and this always in the direction of distancing themselves from the best scientific evidence. Much of this tract depends on the following graph:

The graph above is supposed to agree with another chart in the tract, and it mostly does, except that ΔT14 = 0.054°C above but is wrongly rounded up to 0.06°C. In the description of these calculations, this tract asks, "How much warming did CO2 cause when CO2 increased from 280 ppm in 1750 to the 420 ppm in the atmosphere today?" Their answer to this question let me make sure I can replicate their graph and chart and make sure they're consistently using the same sensitivity value they claim they're assuming. Here's their answer to this question:

Table 1 shows a temperature increase of only 0.32°C (from 5.12 to 5.44 °C)... Human emissions are increasing CO2 concentration so that in 75 years the CO2 level in the atmosphere will be about 600 ppm. Table 1 shows that the warming resulting from CO2 concentration increasing from 420 ppm to 600 ppm would only increase global temperatures 0.28°C, increasing from 5.44°C at 420 ppm to 5.72°C at 600 ppm."

So with this information I was able to replicate CO2 Coalition's chart, with only minor discrepancies likely resulting from additional rounding errors on the part of CO2 Coalition. And I was able to confirm that assuming an ECS of 0.538°C for 2xCO2 would replicate the above chart almost exactly. There is no math in the political tract that justifies their ECS value. It's just assumed. But this assumed value also directly contradicts previous claims by both Happer and Lindzen. Here's a sampling of their changing opinions beginning in 2020, all associated to some extent with the CO2 Coalition, and always changing in the direction of lowering the ECS value they want to promote.

2020

Happer: "Doubling CO2 concentrations with water vapor feedback increases the surface temperature warming to ∆θo = 2.3 K from ∆θo = 1.4 K, or by a factor of 1.6."[1]

Lindzen: "Each doubling of CO2 is expected to provide about 3.7 Watts per square meter (Pierrehumbert, 2011)." And, "However, the contribution of such feedbacks to the total feedback system is generally less than 0.3, which will only bring the total response to less than about 1.5°C, assuming that the Iris Effect only cancels (rather than exceeds) any other long-wave feedbacks. This is at the bottom of the current IPCC model-based range of 1.5°C to 4.5°C warming for a doubling of atmospheric CO2."[2]

2022

Happer & Lindzen #1: They introduced a formula ΔT = (5.35/3.2)*ln(C/Co) to calculate sensitivity. This formula assumes that 2xCO2 causes an RF of 3.7 W/m^2, and if you plug in the value of ln (2) for 2xCO2 you get a sensitivity value of 1.16°C.

Happer & Lindzen #2: In the exact same paper, they write, "to increase the flux by 1.1% and bring solar heating back into balance with radiative cooling, a temperature increase of dT/T, of ¼*1.1% = 0.28%. is needed. Since the absolute temperature of the Earth is approximately T = 300 K, (60°F), the required temperature increment is dT = 0.0028 x 300 K = 0.84 K = 0.84°C." Here both Happer and Lindzen assume that both the radiative forcing for 2xCO2 (3.0 W/m^2) and ECS (0.84°C) are smaller than they assumed earlier in the tract.

Happer & Lindzen #3: The value above (#2) suffers from a calculation error, since 60°F is actually 288.7 K, not 300 K. Correcting this mistake would lower their value to 0.81°C. All three of these results come from the same paper, and I document this with quotes and graphs from the paper here. They insist that ECS can't be higher than 1°C - "basic physics shows that doubling CO2 would result in a temperature increase of less than 1°C." This directly contradicts claim #1 in this tract and both of their opinions from 2020.

2024

Happer: In talks on YouTube (which I document here), Happer gives a series of calculations that arrive at an ECS value of 0.71°C, which he attempts to justify by saying that positive feedbacks are essentially impossible. He says, "it's almost impossible to justify this idea of a positive feedback, but you need it, otherwise CO2 is too wimpy to be worried about... feedbacks are negative... positive feedbacks almost never occur." This opinion of course directly contradicts his statement I quoted above in 2020, and it's obviously dishonest. Happer knows that the Planck Response is a large negative feedback that is larger than the sum of the positive feedbacks, and nothing in the physics here prevents ECS from being in the range estimated in the scientific literature (near 3°C). Here's a summary of their opinions published for the CO2 Coaltion:

Now to be fair, Lindzen and Choi 2009[3] did calculate a sensitivity value similar to the assumption in this tract. So perhaps this paper formed the inspiration for the value assumed in this political tract. But it's also clear that both Happer and Lindzen are willing to accept conflicting values for ΔF2xCO2 - both 3.0 W/m^2 and 3.7 W/m^2. And we can summarize their conflicting ECS opinions as follows:

1. Happer: ECS is 2.3°C and 1.16°C but must be less than 1°C and is also 0.84°C, 0.71°C or 0.54°C.
2. Lindzen: ECS is up to 1.5°C and 1.16°C but must be less than 1°C and is also 0.84°C or 0.54°C.

The IPCC correctly publishes a central estimate for ECS of 3°C with a likely range of 2.5 to 4°C. But this is not comparable to Happer and Lindzen. They are not claiming a range of values for ECS with a central estimate in the middle of their likely range. They are in 2026 claiming both that their central estimate for ECS is 0.54°C and that it must be less than 1°C (2022) and that it's either and 2.3°C or up to 1.5°C (2020). Both have progressively lowered their opinions since 2020 while directly contradicting themselves.

And Happer & Lindzen's current opinion is deeply problematic when compared to empirical data. Currently the GMST is ~1.38°C warmer than the 1850-1900 mean, and EEI is currently 1.4 W/m^2. Human forcings total 3.1 W/m^2. If we assume that Happer & Lindzen are right that ECS = 0.54°C, then sensitivity λ = 0.54/3.7 = 0.15°C/W/m^2. This lets us solve for ΔT in the energy balance equation and find out the theoretical expectation of global warming caused by human forcings if this is true:

ΔT = λ(ΔF - EEI)

ΔT = 0.15*(3.1-1.4) = 0.26°C

This means that of the 1.38°C warming, at least 1.1°C must be driven by natural forcings. Now let's rearrange the terms to find out the natural forcings required to make up the difference.

ΔF = ΔT/λ + EEI

ΔT = 1.1/0.15+1.4 = 8.7 W/m^2

So Happer and Lindzen need to identify 8.7 W/m^2 in natural forcings to explain current warming. Climate scientists have calculated the total natural forcings to be <0.1 W/m^2 (give or take a larger number), meaning that natural forcings have been negligible. Neither Happer nor Lindzen will ever try to find these natural forcings, nor will they point out this gaping whole in their assumed ECS value. This is not a problem if ECS is near 3°C, as estimated by the IPCC, making sensitivity λ = 0.81°C/W/m^2.

ΔT = λ(ΔF - EEI)

ΔT = 0.81*(3.1-1.4) = 1.38°C

There's a reason why the IPCC and the scientific literature have converged on an ECS near 3°C. It fits the empirical data.

The graphic above is what Happer and Lindzen use to support another major portion of this tract, including the following claims:

The Final Rule is based on the Human Caused Global Warming Theory, which predicts there would have been dangerously warm temperatures over the roughly 500 million years when CO2 was higher than today.

The chart contradicts the Theory with data. Although the data are based on various proxies, with the attendant uncertainties, they are good enough to demolish the Theory by observations. It also confirms our conclusions from physics that CO2 had virtually no warming power above 280 ppm.

The dishonesty here is patently obvious:

1. The claim that there were "dangerously warm temperatures over the roughly 500 million years when CO2 was higher than today" is supported by numerous papers, most notably Judd et al 2024.[4] Happer and Lindzen completely ignore this paper as well as the entirety of the paleoclimate literature.
2. The CO2 portion of the graph is mis-plotted from Berner's 2001 climate model. The model covered only the Phanerozoic. However, as it appears here (from Nasif Nahle's version from 2007), the CO2 portion was stretched to align more closely Scotese's temperature schematic (which covers the last 2 billion years). I document the incompetence of this graph here.  Berner's model is now over 25 years out of date. It's been updated numerous times and is now superseded by proxy evidence for the last 485 million years.
3. The Temperature portion of this graph is a rough schematic by Christopher Scotese that shows up on his website (here). This also is outdated, and Scotese has updated this multiple times since this graph was made in 2002 or so, even in the peer-reviewed literature. Happer and Lindzen appear to be completely unaware of Scotese's more recent work or for that matter any work on paleoclimate for the Phanerozoic..
4. Neither the CO2 nor the temperature portions of Nahle's chart are "based on various proxies." Berner's data is the result of his climate model, and Scotese's portion is a rough schematic based on a a general sense of when hothouse and icehouse conditions occurred.
5. Neither the CO2 nor the temperature portions plot any "attendant uncertainties." Worse, the uncertainties in Berner's climate model for CO2 were extremely large, but Nahle's chart ignored them. Scotese never published any uncertainties for the schematic he put on his website almost 25 years ago.
6. Nothing in climate science can be "demolished" by a mis-plotted graph based on an outdated climate model from 2001 and an outdated, rough schematic for temperature. Due to the incompetence of Nahle's chart, it cannot reliably identify any "inverse relationship between temperatures and CO2 levels."
7. The data in Judd et al 2024 supports that 2xCO2 correlates with ~8°C warming on geologic time scales. The best data we have directly contradicts their opinion that "CO2 had virtually no warming power above 280 ppm." Below I show a properly plotted accounting of CO2 and temperature that is based on proxy data (temperature is a data analysis) and shows the uncertainties. There is an apparent disconnect during the Mesozoic, but this is quite possibly due more to data poverty than anything else.

During the Cenozoic, when data poverty issues are the smallest, the correlation between temperature and ln(rCO2) is very strong, and the relationship is still ~8°C for 2xCO2, even when CO2 concentrations are high. There is no evidence for Happer's saturation myth in the paleoclimate data. 

These data from Judd's paper also contradict a couple other claims in this political tract:

1. CO2 was likely never anywhere near "about 7,000 ppm," but when CO2 was above 2000 ppm, temperatures were hot, even when the Sun was significantly weaker.
2. Temperatures were not "the highest they have ever been about 55 million years ago" during the PETM. But CO2 did increase dramatically at the onset of the PETM, and the geologic evidence supports that this warming was driven by a disruption of the carbon cycle.

Happer and Lindzen have not figured out how to make political tracts for the CO2 Coalition that qualify as being scientifically literate and honest. I know that they are capable of doing better than this. They certainly could do a proper analysis that estimates values for ΔF2xCO2 and ECS with properly calculated uncertainties.  The obviously can do better than just regurgitating nonsense from mis-plotted graphs have been wrong for almost a quarter century. I suspect there are a couple good reasons why they don't actually do good work here: 1) if they tried to do honest work, they would not be able to claim what they're claiming, and 2) they don't have to do honest work to for people to believe what they say. Many people will believe them simply because they have big reputations and their opinions match what they want to believe.

 

References:

[1] W. A. van Wijngaarden and W. Happer. Dependence of Earth’s Thermal Radiation on Five Most Abundant Greenhouse Gases. Unpublished manuscript. 2020. https://arxiv.org/pdf/2006.03098.pdf

[2] Richard Lindzen. On Climate Sensitivity. Unpublished Manuscript from CO2 Coalition. https://co2coalition.org/wp-content/uploads/2021/08/On-Climate-Sensitivity.pdf

[3] Lindzen, R. S., and Y.-S. Choi (2009), On the determination of climate feedbacks from ERBE data, Geophys. Res. Lett., 36, L16705, doi:10.1029/2009GL039628.

[4] Emily J. Judd et al., A 485-million-year history of Earth’s surface temperature. Science 385,eadk3705 (2024).DOI:10.1126/science.adk3705