Something in the introduction for the Cook Lecture of next Friday stood out for me. This is how the Consensus Project is presented (my emphasis):
In 2013, John Cook lead the Consensus Project, a crowd-sourced effort to complete the most comprehensive analysis of climate research ever made. They found that among relevant climate papers, 97% endorsed the consensus that humans were causing global warming.
Relevant can of course mean many things, it all depends on what they want to prove with the survey. According to their paper (my emphasis):
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).
What they actually did was searching for papers in which the terms “global warming” or “global climate change” were mentioned in the title or the text of the abstract. But it is not because the term “global warming” or “global climate change” is in the title/abstract of a paper that it is relevant to the cause of global warming.
It should be no surprise that they got a bunch of papers that don’t deal at all with the physics of global warming or with the causes. Most of the papers are on Impacts and Mitigation. Those papers already assume global warming is real, that is their starting point. Mitigation also assumes the anthropogenic nature. None of the two say anything about whether humans are the cause, let alone how much.
There is even a category with non climate related papers. These papers obviously don’t deal with attribution.
The category Methods has the potential to come the closest to attribution, but as expected aren’t about cause. Of course they assume global warming and assume it is anthropogenic. It is also a bin for everything that can not be put in the other categories and it shows.
But, you could say, it is a starting point for many scientists in their research, doesn’t that indicate that they stand behind it? Well, yes, but it doesn’t say anything WHY they believe it. It could be that they understand the cause (not likely, otherwise they certainly would have published the proof), but could be any other reason. Looking at for example Lennart Bengtsson who had to resign from GWPF Advisory Board because of strong peer pressure. Would it be a surprise that in such a highly polarized debate skeptical scientists write neutral looking titles and abstracts and then be classified on the No position pile?
If we want an answer to the question how much the current global warming in the last, say, 100 years is caused by human activity, is a survey of papers the right instrument to do this? And which papers to select? The question about attribution is a very complex one that keeps scientists busy already some time and their opinions vary a lot. This raises the question: is a proof possible with our current understanding and with the limited data we gathered until now?
Even if we think a survey would be the best idea to clarify the physics of global warming, would the outcome matter? It is not because the majority thinks that something is true, that it therefor is true. And what is a consensus on anthropogenic global warming worth if the papers it was based on weren’t even related to studying attribution?
Trust, yet verify 