The Real Climate Science Debate?

An interesting piece from last year at Daily Climate suggests that scientists underplay climate impacts.

The late William Freudenberg in a study claiming this skew suggested that members of the press

reporters, he warned, “need to learn that, if they wish to discuss ‘both sides’ of the climate issue, the scientifically legitimate ‘other side’ is that, if anything, global climate disruption is likely to be significantly worse than has been suggested in scientific consensus estimates to date.”

Is this a fair assessment? I’m not sure. It’s my claim that the purpose of IPCC is to establish the center of gravity of the conversation, not to close it down. I think what is fair is that equal time be given to fringes in some sense equidistant from the IPCC on the complacency/alarm spectrum.

But Naomi Oreskes is quoted at the link with a similar conclusion to Freudenberg’s, which also argues that the argument we are having is backwards:

Oreskes has spent a career studying climate science. She finds ample evidence that climate scientists are indeed biased – just not in the way portrayed by politicians such as Texas Gov. Rick Perry, who claimed scientists paint a bleak picture to secure more research funding.

In reality, Oreskes said, scientists skew their results away from worst-case, doomsday scenarios. “Many people in the scientific community have felt that it’s important to be conservative – that it protects your credibility,” she said. “There’s a low-end bias. It has led scientists to understate, rather than overstate, the impacts.”


  1. [Devil's Advocate] "It has led scientists to understate, rather than overstate, the impacts." Not a brilliant conclusion for anyone trying to argue scientists do a good job of neutrally presenting their evidence. If the fact that they're capable of finessing their findings is established, does the direction it's finessed in matter?

    Where does this leave any arguments about the self-correcting process of ongoing peer review and rigorous self-policing?

  2. The process remains somewhat self-correcting.

    First of all, the most important bias is toward the center of the spectrum - a herd mentality. It is safer at the center of the herd even if the herd is wrong. The bias toward understatement is a constant burden on those of us actually alarmed; I think there is also a bias toward overstatement that attracts some atypical scientist types, arguably myself included. But behind it all is respect for the scientific method.

    The way to get yourself ignored in science is to behave with indifference to truth. Also disqualifying is a lack of set of skills that is fruitful in testing hypotheses and converging on truth, i.e., scientific incompetence. Many people prominent in the public debate and wearing a scientist hat have relatively weak scientific skills and little impact within any productive group in science.

    This filtering still works, eventually. But the mechanisms whereby the pre-formal consensus is established within climate science are only distantly related to the pretensions of the journals.

  3. danolner,

    I think you are missing the point. What Oreskes is saying is that by deliberately presenting their evidence neutrally they are downplaying the dangers. To put it crudely, by trying to be honest they are in fact lying! I think I have said something similar here.

    There is an interesting paper in Nature Climate Change this month doi:10.1038/nclimate1547 which is related to this. It argues "Communicators are convinced of the importance of emphasizing the scientific evidence about climate change risks. But research shows that science-literate individuals are not necessarily the most concerned about global warming."

    While I agree whole heartedly with those sentiments, it is the conclusion of the paper which I feel relates to this discussion. The abstract states that "public divisions over climate change stem not from the public’s incomprehension of science but from a distinctive conflict of interest: between the personal interest individuals have in forming beliefs in line with those held by others with whom they share close ties and the collective one they all share in making use of the best available science to promote common welfare."

    Rather than "conflicts of interest" I would say their views stem from their political convictions. Liberals will accept global warming as a threat, conservatives dismiss it as a scare story. The problem is that scientists are just people and they have political opinions too. Most are relatively well paid and tend to be conservative. Thus they naturally think that AGW is not a serious threat. They are not really lying when they play down the dangers. They really believe that we are safe. Until we can get the conservative scientists to accept the dangers then we have no hope of convincing the conservative general public.

    Hoping what I have written makes sense,

    Cheers, Alastair.

  4. "What Oreskes is saying is that by deliberately presenting their evidence neutrally they are downplaying the dangers."

    That's not what the article says, though. In the quote above it's quite specific: it's bias, just a conservative bias. That isn't the same as being neutral. Unless you're referring to something Oreskes has written somewhere else?

    And there seems to be a contradict between claiming scientists are "presenting their evidence neutrally" while also saying "the problem is that scientists are just people and they have political opinions too. Most are relatively well paid and tend to be conservative. Thus they naturally think that AGW is not a serious threat. They are not really lying when they play down the dangers." If it's the case that political outlook is shaping interpretation of scientific fact, that isn't neutral, is it?

    Unless I'm nitpicking, or we need to separate out the 'core science' being done from the interpretation being placed on it?

  5. That Daily Climate article was a bit dodgy. What it called 'observed annihilation rates' were actually extrapolations of the prevalence of future extinction risk (the risk of being at risk) based on studies that included observed changes to the population, range or habitat of various species. The 'predicted' future prevalence differed in that it was derived from scenario-driven studies - projections, not predictions. So the 'observed' rates were observation-based projections and the 'predicted' rates were projection-based projections. (Or something. I haven't eaten much today, so I'm a bit ga-ga.) Study and SI here:

    The article's opening sentence is dodgy, too.

    'The warnings were dire: 188 predictions showing that climate-induced changes to the environment would put 7 percent of all plant and animal species on the globe - one out of every 14 critters - at risk of extinction.'

    No. They weren't. They didn't. I can't be arsed to explain why. (Time to eat.)

    The study also looks dodgy. That's probably because I'm ga-ga, so ... Oh, what the hell. The authors' mathematical jiggery-pokery produced very few species facing extinction because of climate change before 2100. The first one on the list is a nemotode that's endemic to Antarctica and whose numbers in a particular valley system declined by 65% between 1993 and 2005. I think the authors pretty much straightlined observed population declines to arrive at their extinction risks but in this case even a more sophisticated projection method would probably indicate doom for the poor nematode by 2100.

    The trouble is, that drastic population decline was wholly due to a regional cooling. Climate change, to be sure, and perhaps at least partly anthropogenic, but how long will the cooling persist? If it ends soon, that's already reduced the study's 'observed' extinction-risk rate by a fifth - and I've only looked at one paper, the first on the list. Not promising.

  6. The process is indeed self-correcting. Data are pouring in, and in almost every case it is clear scientists indeed are biased towards understatement.

    On the whole, people are unable to cope with the dread facing the truth creates.

    (My last attempt, which involved some homework, was lost so I'm using this route. Don't know why, but with two sick parents I don't have a lot of time.)

  7. Being conservative in a scientific sense is a quite different thing from the political sense, although both may be present in the same individual. Being reticent about conclusions to be drawn from the evidence is the former at work, not the latter.

    Re the IPCC, I agree that its tendency to underemphasize the worst case stuff isn't good, but much worse is the fact that the science is advancing quickly and the WG1 reports are at least a year out of date before the ink is dry, let alone 5+ years later.

  8. Pingback: Another Week of GW News, September 30, 2012 – A Few Things Ill Considered

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.