Severe analytical problems in dendroclimatology, part three

This is just a short comment.

I’ve put up two posts now discussing certain analytical problems in dendroclimatology. But I don’t want people to get the wrong impression or get carried away on this topic, given how much animosity and generally strange and counter-productive discussions that exist on the internet surrounding this field. There are many people in this field who are very well aware of a number of existing problems, and of steps needed to be taken to lessen/correct them. For example, I linked to this white paper by Keith Briffa and Ed Cook in part two (but that was a long post and the link is way down there so people might well have missed it). Please read that–it’s short and easy to read.

That paper is the only one linked to on the front page of the NOAA Paleo Tree Ring page, which means they have chosen to place it in clear view of those retrieving tree ring data. There are also a number of papers in the tree ring literature discussing various analytical issues in great detail (particularly with respect to calibration). Furthermore you can pick almost any field in science and find some kind of serious analytical problem exhibited by some group of researchers somewhere at some time (including ones that are worse than those in dendroclimatology). So it’s wise to keep these issues in perspective. The point of discussing problems is so that they can be better understood and addressed; to go beyond that is to enter into the counter-productive. Lastly, I do my best to be conscientious and get things right, but I’m far from infallible–if I say something wrong or questionable, raise it and let’s discuss it.

8 thoughts on “Severe analytical problems in dendroclimatology, part three

  1. “Lastly, I do my best to be conscientious and get things right, but I’m far from infallible–if I say something wrong or questionable, raise it and let’s discuss it.”

    What a refreshing and unusual statement in this arena! I wish my humble statistical and scientific background would allow me to engage you in a conversation at a sufficiently high level to be of general interest.

    • Thank you for the compliment, I do appreciate it. In my view, most people, including yourself, are eminently capable of understanding the issues involved, and it’s our job to communicate the science in the way that people can understand it. The only way to succeed at that is by give and take discussions I think.

    • Per your first sentence there, it’s really quite sad that the debate on the science has become so dogmatic and poisoned, due largely to a few particular individuals with substantial ego problems. We have to forge something new because the existing state of affairs is simply intolerable and broken.

  2. RE: . . . it’s really quite sad that the debate on the science . . .

    Actually, I think what’s really quite sad is there has been no real debate on the science. I have no doubt that’s because of, as you say, a few particular individuals with substantial ego problems. Has there ever been any other field of study where, “Now, we know everything” was soon achieved? (Religions exempted.) In fact, has there ever been any other field of study where, “Now we know everything” was ever achieved? (Religions exempted.)

  3. Jim: “it’s really quite sad that the debate on the science has become so dogmatic and poisoned, due largely to a few particular individuals with substantial ego problems.”

    Just to say, Jim, I think that completely misapprehends the problem. We like to focus on individuals since it’s easy for us to do, but the main problem here is a widespread fear-based denial of reality and resistance to change.

    Of course you know this, so really this comment is just to point out how easy it is to be distracted.

    • Hi Steve. I wasn’t referring just to scientists as causing problems with their egos, in case that wasn’t clear. I personally don’t like to focus on individuals, which is why I’ve aimed my critiques in this series at the issues only, not individuals. However, because it’s scientists who do science, criticisms of poor science will eventually point to somebody or other, like it or not. Becoming obsessive about certain individuals however, is a clear sign that something other than scientific criticism is at play. We all need to stick to the science issues, period.

      It’s not correct to just blame one side or the other in this thing, nor is it correct to lump people into one or the other of two opposing camps. It’s far more nuanced than that, regardless of what some people would like to have everyone believe. I will support the science 100% when I think it is right, and I will criticize it 100% when I think it’s wrong. Anything else is less than honest, and I will not do it.

  4. Interesting that the Briffa/Cook white paper lays out some of the serious problems with tree ring analysis, calls for improved methods in collection of the data to deal with (at least some) of this problem, but does not go further in giving explict warnings on the effect of previous use in reconstructions.

    Jim, is this related to the more general issue of avoidance/non-engagement you’ve elsewhere described (especially wrt Loehle2009)?

    • I really don’t know kch. I was glad to see that they recognized the problems inherent in typical field sampling, because I’ve rarely seen that raised anywhere else. And in their fourth paragraph under “Uncertainty in the Tree-Ring Data Themselves” they state directly that there are very serious problems with estimating long term trends (but go no further with that issue). And at the top of the second page they discuss two problems with RCS. But they do not recognize the problems I’ve been discussing, referring only to a somewhat similar but much less serious problem that Briffa and Melvin refer to as “end effects”. They were on the right track but didn’t quite get there.

      Overall, I think you have to give them kudos for recognizing and stating directly, what some serious problems are. But it’s just a set of notes from a small conference in Trieste Italy a few years ago. It’s another thing altogether to raise and clearly describe these problems in a peer-reviewed paper where someone or other might want to shoot you down.

Have at it