The paper is deeply flawed
There are a number of problems with this paper and many have pointed them out.
Schwartz et al, sum it up.
- Erroneous history of ¹⁴CO₂ in air that is at odds with direct observations
- Neglect of the consequences of the large input of ¹⁴CO₂ into the
atmosphere from nuclear weapons tests in the 1950s and 1960s
- Failure to account for isotope exchanges between the atmosphere,
ocean, and land biosphere that occur independent of net change in
amount of atmospheric CO₂;
- Neglect of multiple independent lines of evidence that CO₂ emitted from fossil-fuel combustion is
the principal contributor to the increase of atmospheric CO₂ over
the industrial era.
They call for the paper to be retracted.
On the basis of the several arguments presented here, we conclude that the paper of Skrable et al. (2022) should be retracted in its entirety.
The authors responded to Schwartz et al and Andrews and Tan. While they acknowledge some errors, they fail to directly address many of the issues below, or claim they are not relevant.
Health Physics is a journal about radiation science, not atmospheric science.
Health Physics, first published in 1958, provides the latest research to a wide variety of radiation safety professionals including health physicists, nuclear chemists, medical physicists, and radiation safety officers with interests in nuclear and radiation science.
Many have pointed out this could lead to problems with peer-review. Andrews & Tans...
Health Physics ventured outside its area of expertise when it accepted the paper by Ken Skrable, George Chabot, and Clayton French on atmospheric ¹⁴C... [Health Physics editors] should have included a qualified reviewer from the radiocarbon or atmospheric science community. They clearly did not because any reviewer with previous knowledge of atmospheric ¹⁴C would have found the fatal errors cited below. An extensive peer-reviewed literature exists on studying the carbon cycle with ¹⁴C, but next to none of it is cited by the authors.
The editor-in-chief of Health Physics did respond justifying their acceptance of the paper, but does not address any of the specific concerns, nor the flaw.
No collaboration with atmospheric scientists or radiocarbon specialists
Musolino further points out the lack of reaching out to atmospheric scientists who might have provided better data and helped point out fundamental flaws.
[the authors of the paper] did not appear to attempt a direct engagement with the primary scientific community of atmospheric scientists to whom they posed a widely divergent and controversial opinion. The draft was presented for peer review to experts in health physics but not to scientists who are expert in CO₂ emissions and who study CO₂ and ¹⁴C in the atmosphere.
Their value for ¹⁴CO₂ at 1750 is too high.
Both Andrews and Tam and Schwartz et al point out that the model used in the paper is at odds with existing observations. In the graph below, the smooth brown curve is the model used in the paper, the big spike is multiple papers based on actual observations, and the light blue line is a model if there was no nuclear testing.
The first issue is the paper uses a value 20% too high for its 1750 baseline; in the graph above the paper's line starts at 16 rather than everyone else starting at 14. Andrews and Tans explain...
Skrable et al.’s “educated guess” of 16.33 dpm (gC)−1 for the specific activity of ¹⁴C in the 1750 atmosphere is not bad for a guess, but it is 20% too high. It corresponds to a ∆¹⁴C of 200‰ (parts per thousand, see below). Except during the era of atmospheric nuclear testing, such a value has not been seen for over 10,000 years (Cheng et al. 2018). A 20% specific activity error in ¹⁴C converts to a dating error of over 1,800 years. Carbon-14 (¹⁴C) dating is much better than that because the atmosphere’s specific activity during historical times is accurately known. It is meticulously calibrated from materials of known age such as tree rings. There was no need to guess its starting point in 1750 or its trajectory since. There was no need for the authors to limit themselves to only the recent data from Niwot Ridge.
Skrable et al replied they knew their value might be too high, and there are errors, but believe their methodology has value.
We also recognized the likelihood of the excessive value of the specific activity, S(0), of ¹⁴C in 1750, and noted our rationale for maintaining a high value. We realize that the values we have determined for the anthropogenic fossil component, CF(t), and non-fossil component, CNF(t), may include some errors, but we do believe there is value to our methodology.
They fail to account for the effects of nuclear bomb testing
There's a huge spike in observed ¹⁴CO₂ which the paper completely ignores. This spike is due to nuclear bomb testing. The paper claims they can ignore this effect citing [checks notes] Wikipedia. :facepalm:
A Wikipedia link for ¹⁴C describes the increase in the concentration of ¹⁴CO₂ in the atmosphere that resulted from high altitude nuclear bomb tests, circa 1955–1963. Based on the figure in the Wikipedia link, ¹⁴CO₂ from the atmospheric bomb tests during this period would be significant in 1955 to about 2005.
As can be seen in the graph above, when nuclear testing is omitted from modelling the result diverges significantly from observations. They misinterpreted the data, nuclear testing remains significant and cannot be omitted.
Schwartz et al explains...
The contribution of bomb ¹⁴CO₂ to atmospheric ¹⁴CO₂ was explicitly examined in a modeling study (Graven et al. 2020) in which the source of bomb ¹⁴CO₂ was omitted, shown also in Fig. 1; the signal of bomb ¹⁴CO₂ is given by the difference between the measurements and the model in the absence of the bomb ¹⁴CO₂ source. This difference shows that residual bomb ¹⁴CO₂ cannot be dismissed over the years 2004–2012 used by Skrable et al. in their analysis. Failure to include this residual bomb ¹⁴CO₂ in their apportionment of the increase of the increase of atmospheric CO₂ to fossil fuel and non-fossil fuel sources completely vitiates this apportionment.
Skrable et al replied...
Exchange among the reservoirs was inconsequential for our work, except for the possible residual effects of bomb ¹⁴CO₂, which we admittedly assumed were not significant. Information provided by Schwartz et al. and others indicate that there likely was current feedback of bomb ¹⁴CO₂ from reservoirs to the atmosphere, and the claim is being made that this invalidated our findings.
Failure to account for isotope exchanges between the atmosphere, ocean, and land biosphere
Schwartz et al list a host of issues with their assumptions about how carbon flows between various parts of the Earth.
The Skrable et al. (2022) framework assumes that carbon in the environment can be divided into two categories: (1) preindustrial carbon, which is taken as uniform in the ¹⁴C/¹²C ratio, and (2) fossil carbon, which is devoid of radiocarbon. They further assume that these categories maintain their identities as carbon is exchanged between the atmosphere, ocean, and land biosphere. In fact, the flows of different carbon isotopes are not connected as assumed by Skrable et al.... Thus, importantly here, ¹⁴C can be exchanged between the atmosphere and ocean with no net exchange of carbon as a whole. An important consequence of this independence is that impact of emissions of fossil-fuel CO2 on the ¹⁴C/¹²C ratio of atmospheric CO₂ is much greater than the impact of these emissions on atmospheric CO₂ as a whole... These critical aspects were not considered by Skrable et al., causing their calculations to underestimate the input of fossil-fuel CO₂.
Skrable et al responded...
While we do recognize the reality of exchange of CO2 among the atmosphere and reservoirs, it was not our intent to attempt to develop a model that considered these, nor do we believe this was necessary for our purposes.
Incorrectly attributes the ocean as a source of CO₂
The paper claims the world's oceans have been a net source of CO₂ since 1950...
This DCNF(t) value of
31.07 ppm in 1950 results from the annual redistribution of CO₂ among its reservoirs, primarily a net release of CO₂ from the oceans due to increases in temperatures from solar insolation in 1950 and afterwards.
Schwartz et al points out this is incorrect, the oceans absorb CO₂.
From the well quantified rates at which CO₂ is building up in the atmosphere and rates of CO₂ release from fossil-fuel burning, it is clear that around 50% of the emitted carbon remains in the atmosphere, with the balance absorbed by other reservoirs, of which the oceans and the land biosphere are the most important. The ocean and land biosphere are thus together acting as a major sink not a source of CO₂. The excess CO₂ in the ocean is now also well measured, and the uptake rate is consistent with a wide body of other evidence on rates of mixing and carbon chemical properties of seawater. This understanding of the rate at which excess carbon is being redistributed into the ocean and land is independently supported by measurements of trends in atmospheric O₂ and ¹³C/¹²C ratio in addition to radiocarbon
Neglect of multiple independent lines of evidence
Schwartz et al point out that radiocarbon is not the sole piece of information about the present increase of atmospheric CO₂.
Finally, the present understanding of the controls on atmospheric CO₂ buildup importantly rests on many convergent strands of evidence in addition to radiocarbon. From the well quantified rates at which CO₂ is building up in the atmosphere and rates of CO₂ release from fossil-fuel burning, it is clear that around 50% of the emitted carbon remains in the atmosphere...
The paper cannot draw such sweeping conclusions about climate change without also addressing other evidence to the contrary.