Thursday, July 29, 2010

Remembering Stephen Schneider, and voting No on Proposition 23

(This is a repost from the Brian for Water District campaign blog.)

Driving to a meeting of the Water District's Environmental Advisory Committee earlier [last] week, I heard the sad news about the unexpected death of the prominent Stanford climatologist, Stephen Schneider.

While a student at Stanford Law School, I participated in one seminar where he guest-lectured and heard him on other occasions during school and afterwards. I thought he gave the most convincing demonstration of who to trust in the climate debate by showing a survey of the mainstream climatologists and the small number of scientists that doubted climate change. Schneider showed that the mainstream scientists were reasonably confident of their predictions but also admitted a wide margin for error. The few skeptic climatologists admitted nothing, and were absolutely confident that they were right. He had given the best demonstration I could imagine of scientific honesty on one side and over-confident hubris on the other.

Schneider's death comes as California wrestles with Proposition 23's demand to suspend its premier climate change law, AB32, a theoretical suspension that would actually kill it if Proposition 23 passes. I know that Schneider actually had some criticisms of his own of AB32, but I can't imagine he would favor the misguided effort to kill the law and do nothing in return.

I'll be voting No on Proposition 23, an initiative that will harm efforts to fight climate change and efforts to protect our water and watersheds. We will have to learn from Schneider's legacy as the state and country move forward.

-Brian

For a eulogy about Steve Schneider, read RealClimate here.

7 comments:

  1. Do you remember which "skeptic climatologists" were involved? Or which survey he was referring to? My impression as to where the bulk of the "overconfident hubris" is has been the exact opposite of that Schneider presented, but I suspect selection bias on both sides. Or maybe it depends on the topic about which confidence is being expressed...

    ReplyDelete
  2. Lindzen was one of them. His 95% certainty range was tiny, less than half a degree for 2xCO2 (IIRC). Remember it's not just what Lindzen predicted but what chance Lindzen gave that he could be wrong.

    I think there were one or two other skeptics, but I can't remember who. I think it was a questionnaire that Schneider developed and sent to climatologists.

    ReplyDelete
  3. Interesting. My reverse impression mostly comes from following RealClimate versus ClimateAudit. The CA refrain at one point was that the true certainty range on things like MBH99 was floor-to-ceiling - the stats were too uncertain to say much with any confidence. In response, Gavin and Mann have often said "If McIntyre doesn't like these reconstruction he should do one of his own." Thereby *completely missing the point* - McIntyre rejected publishing his own reconstruction the exact same reason he rejected MBH - McIntyre had a lack of confidence in the statistical significance of conclusions based on that sort of reconstruction. Meanwhile Mann and company skip merrily along making what seem like overconfident claims regarding specific years or decades being "the warmest in 2000 years" or some such.

    In that debate, the ones claiming absolute confidence tend to be at RC; the ones reluctant to make or accept claims of high certainty tend to be at CA.

    ReplyDelete
  4. Anonymous6:47 AM

    Uhm, maybe there's a bit of a distinction between the level confidence in one's own work and attacking the work of others? It's not a symmetric situation.

    -Jeff S.

    ReplyDelete
  5. The other distinction is that Schneider's survey asked climatologists to reach an overall conclusion about climate change and state their certainty levels/range of error. I've no doubt that on the proxy record, the mainstream scientists will state they're much more certain that it's useful than CA folks would.

    I'd be interested to see what chance each side would give that their analysis of the proxy record, that it's statistically significant or statistically meaningless, is right. I'd bet the both sides would be extremely confident of their position.

    Also on the proxies, it might be similar to an argument I made about climate models a while back: if the skeptics think you can manipulate it to say whatever you want to say, then go ahead and demonstrate an analysis that doesn't show a hockey stick but is at least as defensible as the hockey stick analysis. The point would not be to show confidence in their own reconstruction, but to show that the whole thing doesn't work. If CA can't do it, then they've got a problem (and it clearly isn't from a lack of resources, CA has spent years on this issue).

    ReplyDelete
  6. Anonymous7:03 AM

    Your last point is similar to another point you made on this blog that I don't think gets made often enough, or really, ever. If it is possible to construct a plausible, defensible climate model wherein a doubling of atmospheric CO2 leads to minimal warming, it is reasonable to expect that the allied forces of the fossil fuel industry (the largest industry in the world) and the skeptic community could have produced one by now. As Sherlock Holmes might have it, it's the dog that doesn't bark that tells the story.

    -Jeff S.

    ReplyDelete
  7. Have you read Craig Loehle's papers? I'd say his reconstruction is at least as defensible as the MBH/Mann08/Mann09 approach. One problem with saying skeptics should "demonstrate an analysis that doesn't show a hockey stick" is that they *have*, but a bigger one is that nobody seems to agree on what it means to say a chart "shows a hockey stick". RC sees them everywhere, but even the IPCC spaghetti chart doesn't look very stickish these days. The difference of opinion mostly has to do with whether you're looking at the shaft part or the blade part. (And also, I suppose, with whether the reader has read _How to Lie With Statistics_ closely enough to realize that it is *not okay* to combine data that was smoothed in different ways into the same chart and thereby draw relative conclusions.)

    ReplyDelete

Note: Only a member of this blog may post a comment.