The most surprising find in the evidencesquared episode 2 podcast (see previous post) is a statement by John Cook about the endorsement levels of the Cook 2013 paper. This is how he reacts to his observation that some skeptics claim that they belong to the 97% because they believe that the world is warming and humans have some influence (my emphasis):
Yeah, and the way that the consensus position is often expressed by people who don’t accept the consensus: “Well, I am part of the 97%, because I think that humans are causing some of global warming, but I don’t think it is most and therefor that makes me part of the 97%”.
Right, we saw that with criticism from publishing of the Wall Street Journal, for example.
Yeah, we come to that in a moment, but the problem with that argument is, in our 2013 study, we explicitly ruled that out as an option. Whenever there was an expression that humans are causing less than half of global warming, we categorized that as rejection of the consensus. So the person who thought that humans are causing some, just not more than half, that is a rejection of the consensus.
I remembered from reading the Cook 2013 paper that there were only two levels of endorsement that quantified the consensus and the papers in those two levels just amounted to less than 1% of all papers in their dataset. How on Earth could he now claim that they considered it rejecting the consensus when a paper expressed that “humans are causing less than half of global warming” when almost all data had no clear quantification?
Let’s look into what the paper actually said. It starts promising for his claim in the introduction:
We examined a large sample of the scientific literature on global CC, published over a 21 year period, in order to determine the level of scientific consensus that human activity is very likely causing most of the current GW (anthropogenic global warming, or AGW).
Remember, they were looking at the title and abstract of the papers to classify their level of endorsement to the claim that “humans cause global warming”.
Looking further into the paper, the methodology gives a glimpse that it is not what it seems:
Explicit endorsements were divided into non-quantified (e.g., humans are contributing to global warming without quantifying the contribution) and quantified (e.g., humans are contributing more than 50% of global warming, consistent with the 2007 IPCC statement that most of the global warming since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations).
This means that level 1 (explicit endorsements with quantification) should contain everything between 50 and 100% endorsements of AGW. There were only few papers classified in level 1. Level 2 (explicit endorsement without quantification) is explained as “humans are contributing to global warming without quantifying the contribution”. That is rather trivial and goes beyond the only-more-than-50% claim.
The claim that humans contribute (non-quantifiable) to global warming is something most skeptics can agree with, I think. It doesn’t end there. Just look at the table of levels of endorsement:
|Level of endorsement||Description||Example|
|(1) Explicit endorsement with quantification||Explicitly states that humans are the primary cause of recent global warming||“The global warming during the 20th century is caused mainly by increasing greenhouse gas concentration especially since the late 1980s”|
|(2) Explicit endorsement without quantification||Explicitly states humans are causing global warming or refers to anthropogenic global warming/climate change as a known fact||“Emissions of a broad range of greenhouse gases of varying lifetimes contribute to global climate change”|
|(3) Implicit endorsement||Implies humans are causing global warming. E.g., research assumes greenhouse gas emissions cause warming without explicitly stating humans are the cause||“…carbon sequestration in soil is important for mitigating global climate change”|
|(4a) No position||Does not address or mention the cause of global warming|
|(4b) Uncertain||Expresses position that human’s role on recent global warming is uncertain/undefined||“While the extent of human-induced global warming is inconclusive..”|
|(5) Implicit rejection||Implies humans have had a minimal impact on global warming without saying so explicitly E.g., proposing a natural mechanism is the main cause of global warming||“…anywhere from a major portion to all of the warming of the 20th century could plausibly result from natural causes according to these results”|
|(6) Explicit rejection without quantification||Explicitly minimizes or rejects that humans are causing global warming||“…the global temperature record provides little support for the catastrophic view of the greenhouse effect”|
|(7) Explicit rejection with quantification||Explicitly states that humans are causing less than half of global warming||“The human contribution to the CO2 content in the atmosphere and the increase in temperature is negligible in comparison with other sources of carbon dioxide emission”|
That claim of Cook shed a new light on this table. Now I understood the classification of those levels. Until now, I assumed that these levels ware in ascending order, from very sure to not sure at all. Which didn’t make much sense back then. My view now is that these levels represent three big groups: endorsing the consensus (1, 2 and 3), neutral (4a and 4b) and rejecting the consensus (5, 6 and 7).The groups endorsing and rejecting the consensus are then further divided in three subgroups depending on how clear the endorsement/rejection is stated. That also explain why they took level 1, 2 and 3 to calculate the consensus position.
There is however one thing that spoils the party: the definition of those levels is overly broad as seen in the beginning of the post. According to these definitions and examples, a statement claiming that humans cause some global warming can be classified in at least three levels of endorsement depending on the wording:
- (2) Explicit endorsement without quantification (if global warming is stated as a known fact)
- (3) Implicit endorsement (if it is implied that humans are the cause)
- (6) Explicit rejection without quantification (if the catastrophic view is rejected).
At least three, because it also might be classified under (4b) Uncertain, depending on one’s definition of “undefined” or “inconclusive”. But this will open a new can of worms. What is then the difference between non-quantified and undefined or between rejecting and inconclusive?
Coming back to the claim that papers expressing less than half of the warming is human caused were classified as rejecting the consensus: how did they evaluated whether a paper expressed humans are causing more/less than half of global warming, when agreement is not specifically specified in level 2, 3, 5 and 6? Only categories 1 and 7 quantified the endorsement and these only had few papers in it.
There might however be a way to check that out. Cook made a searchable database with (most?/all?) the scientific papers in the Cook 2013 paper. Those papers were classified according to their category and level of endorsement. That gives me the opportunity to figure out how the titles/abstracts were classified: I know the definition of all levels, so if I read a title and abstract, then I can try to reverse engineer what their original definition was and answer the question which definition the authors of the Cook 2013 paper used. Is it the definition claimed in the podcast by Cook or the (much broader) definition that was defined in the paper with as lead author the same Cook?
The first level (explicit endorsement with quantification) should be straight forward because a quantification is provided in the title/abstract. So I will focus on level 2 and 3. According to their definition of level 2, it should be clearly stated that humans are the cause, so I would look at words like “human”, “anthropogenic, “most”, “primary” and so on. It would be a bit messier when looking at level 3 in which the human influence and the statement that they cause most of the warming is implied. That gives even more degrees of freedom.
My first surprise came when I searched for the words “human”, “anthropogenic”, “most” or “primary” in level 2. Almost a third of these papers did not have any of these words in their title nor in their abstract. How were these papers then classified as supporting that humans cause more than 50% of the warming?
If the paper for example referenced the IPCC or if it used (the outcome of) a global circulation models, then it was considered to endorse the consensus that more than half of the warming was caused by humans.
Another criterion seem to be stating concern about emissions. That seems odd. It is not because a researcher is stating concern about emissions that it is based on the assumption that more than half of the warming is caused by humans. It very well could be, but not necessarily.
Yet another criterion is when the paper was categorized under “Mitigation” (when the paper is for example about measures to limit emissions). I can understand that when someone proposes measures to limit emissions, that it is probably based on the assumption that humans cause a significant part of it (otherwise it would not be worth the effort), but that doesn’t guarantee that it is based on the assumption that humans cause more than 50% of the warming. That threshold could be (much) lower than 50%.
An example of a mitigation paper from level 2 (1994):
N2o Emissions At Solid-waste Disposal Sites In Osaka City
Nitrous oxide (N2O) is a trace gas contributing to stratospheric ozone depletion and global warming. Although a large quantity of information exists about N2O emissions from various ecosystems, this study was initiated to demonstrate the features of N2O emissions from sea-based waste disposal sites in Osaka City in relation to CH4 emissions. Average N2O emissions at an active landfill (S-Site) were several times higher than those at a closed landfill (N-Site). Average CH4 emissions were also much greater at the S-Site. Regarding the nature of N2O emissions, remarkable emissions often were observed with aerobic waste layers at the N-Site, suggesting almost inversely related N2O emissions with CH4 production at the N-Site. However, at the S-Site a few exceptionally high N2O emissions were noted in cases of high CH4 emissions.
Just looking at the title and the abstract (as was done according to the Methodology section of the paper), I can understand the human part of it (nitrous oxide emissions from waste disposal sites) and N2O is a trace gas contributing to global warming, but there is nothing that reveals the endorsement level. It could be higher than 50%, but as well lower than 50%. That is not clear from this information.
Another example, this time from level 2 in the category “Methods” (1993):
Rate Coefficients For Reactions Of Several Hydrofluorocarbons With Oh And O((1)d) And Their Atmospheric Lifetimes
The rate coefficients for the reaction of OH with CH3F, CHF3, C2H5F, C4H2F8, and C5H2F10 were measured at temperatures between 232 and 378 K using the pulsed laser photolysis-laser-induced fluorescence technique. The rate coefficients for the reaction of O(1D) with the above molecules and CH2F2 were measured at room temperature using time-resolved vacuum-UV atomic resonance fluorescence detection of O(3P). The atmospheric lifetimes needed for the evaluation of global warming potentials were calculated for all six molecules using a one-dimensional atmospheric model using the kinetic data obtained in the present study.
Also here, I can understand the human part of this. Those hydrofluorocarbons come from human emissions and they add to the greenhouse effect. But then, no clear endorsement level was given although it was classified as “explicitly” endorsing the consensus. This paper was probably rated as explicitly endorsing the consensus because the global warming potentials were calculated by using an atmospheric model?
Also from level 2, but then in the category “Impacts” (1993):
Agriculture In A Greenhouse World
While agriculture in some temperate regions may benefit from global climate change, tropical and subtropical regions may suffer. Even where potential production will improve, the required adjustments may disrupt ecosystems and land-use patterns. Agricultural zones will shift toward high latitudes, while heat stress and increased droughts will reduce productivity in lower latitudes. On the positive side, higher CO2 may enhance photosynthesis and water-use efficiency. Future hazards include sea-level rise, insect infestation, and greater evaporation losses. Some agricultural activities augment the greenhouse effect by releasing CO2, CH4, and N2O. Understanding the potential impacts of climate change is a prerequisite to developing societal responses.
This by no means shows that the paper endorses the human part of global warming. The paper only mentions “(global) climate change”, nowhere is said (or even implied) that humans are considered the cause of the warming. Based on the title and abstract, endorsement could go from 0% to 100%. The authors probably classified this as explicitly endorsing the consensus because agriculture emits various greenhouse gases that contribute to global warming? In that case, to classify this under level 2, the broadest range of the definition need to be used.
It is not my intention to quantify how many papers go beyond their definition. The very fact that I easily found several examples at the start of their dataset confirms that the definition used in the Cook 2013 paper is actually as broad as it is stated in the beginning of their paper and is certainly not the definition that Cook stated in the podcast.
To conclude, I do not believe for one moment that only papers expressing that “humans cause more than 50% of global warming” got classified as endorsing the consensus in the Cook 2013 paper and it is not that difficult to find out. One just has to look at how the authors interpreted the definition in order to come to their classification.