There have been a number of posts on Watts Up With That discussing climate sensitivity. Climate sensitivity is basically an estimate (calculation?) of how much the global surface temperature would increase if the CO2 levels doubled. The first post was a guest post by Nic Lewis called An objective Bayesian estimate of climate sensitivity in which he discussed his own paper that had just been accepted for publication in the Journal of Climate. I believe that this is a reputable journal and so this is highly commendable.
The post discussed previous work that suggested that if CO2 levels doubled, there was a 95% chance that global surface temperatures would rise by between 2.1 and 8.9 K, with a mode (most likely) of 2.9 K. The mode is quite close to the lower-end of the range and so suggests that this is quite a skewed distribution and that the actual value is closer to 2.1 K than to 8.9 K. Nic Lewis’s work redid this and found a mode of 2.4 K and a 95% range of 2.0 to 3.6 K. It’s a narrower range, but seems fine. I don’t know enough to really comment. He goes on to suggest that a more realistic range (based on various assumptions about the Bayesian priors) is from 1.0 to 3.0 K, with a mode of 1.6 K. Again, I don’t really know enough to comment further and I have no reason to suspect that this isn’t someone who is quite capable of carrying out this analysis. If anything, I was quite impressed by this post as it discussed what appeared to be a perfectly valid bit of scientific work that is to be published in a peer-reviewed journal. Additionally the comments were also quite balanced.
The latest post is one by Anthony Watts called Another paper finds lower climate sensitivity. This new work appears to suggest that there is a 90% chance that the climate sensitivity is between 1.2 and 5.1 K and that the mode is between 1.5 and 2.0 K. Again, seems like a reasonable bit of work. Anthony Watts’ post, however, finishes with
All this blows the laughable Skeptical Science claim Climate Sensitivity Single Study Syndrome, Nic Lewis Edition out of the water. Dana should quit while he’s ahead, because his arguments aren’t convincing.
I found this an odd comment to make, given that all these studies seem consistent at some level. What Anthony Watts is referring to is a post on the Skeptical Science site, suggesting that the response to Nic Lewis’s paper by those on WUWT is a form of single-study syndrome.I read the post on Skeptical Science and I thought it fairly balanced. It had some criticisms of Nic Lewis’s work, but nothing that isn’t part of normal scientific discourse. If anything, he was quite complementary about what Nic Lewis had done. The main point of the Skeptical Science piece was that a single piece of work doesn’t immediately prove or disprove something, and with this I completely agree. This is one of the things that does really bug me when I discuss climate change with those that are skeptical. They often seem to misunderstand, or misuse, the scientific method.
It seems that there is a perception that if you can find an error in a piece of work, then it makes it completely wrong. It also seems that many think that if they can point at one scientific study that supports their view, it makes them right. This is not the scientific method. The scientific method is a complicated process in which some area of science is investigated and, over time, there is a general convergence towards what is regarded as the most likely scenario. It doesn’t make it correct and there is always the chance that something will come along to completely change peoples views about a particular science area. However, in the same way that a single study doesn’t prove that a certain idea is correct, a single study is almost never going to be enough to prove that something is fundamentally wrong. Although it is excellent that many people are investigating climate sensitivity, just because a couple of papers suggest that it is lower than previously thought, doesn’t make it so.