Thursday, June 01, 2006


Well, following a polite enquiry to GRL, the Editor has agreed to consider a revised Comment. This seems very reasonable - as I said before, the original decision didn't actually seem to be justified by the reviews. Of course, we still have the fun of trying to satisfy 3 rather diverse reviewers within the 2 page (and 2 week) limit while trying to keep some of our own opinion in there too :-) So the final outcome is far from a foregone conclusion. Dave is engaged in some interesting email discussions with us which should make the final result more useful.

In Other News, today I also received an email from Nature saying that they are considering my "appeal" on the Hegerl et al thing. This came as somewhat of a surprise - I had sent them a rather intemperate and sarcastic email immediately following our rejection several weeks ago, and not heard anything (nor really expected to) since. Perhaps they were just clearing out the spam trap and found I'd been filtered for using naughty words (no, I didn't really). Of course I'm sure they won't take the comment, but who knows, they might give me another amusing excuse when they reconfirm the rejection :-)


Anonymous said...

I'm still trying to follow what sounds more and more like a petty vendetta against Dr. Frame (you coincidentally bash everything he's done, up to and including the recent Hegerl paper?).

Basically, "Group A" believes we shouldn't rule out high sensitivities. "Group B" believes we can due to expert priors, and because they conveniently sweep high sensitivities under the rug. Both groups are using climate models with pretty high uncertainties in general. So you basically make up an "expert prior" because you have decided a priori that you just can't stand high sensitivities based on what you think has happened in the past. As if AGW will just repeat some past interglacial event I guess.

So it sounds like you are fighting to be "king of the hill" on a dung heap. Guys like you keep the skeptics of the world in business! Doesn't the Japanese gov't want you to do any science at all? Do you even bother to use the Earth Simulator anymore? ;-)

Anonymous said...

Hey, James, how sure are you that the Nature editors don't read your blog? Be nice... :)

Actually I wish the extreme sensitivity stuff would be kept more under wraps. There should be more emphasis on the effects of 2x sensitivity at 3C, especially since increasing knowledge about abrupt effects seems likely to amount to a steady stream of bad news. As I've mentioned before, the implications of the near-term loss of the Tibetan glaciers seems like a perfectly good case study as to why continued global warming is a very bad idea, and yet there seems to be very little information on that. This is of course no criticism of climate scientists, as their job is more or less done as soon as the trend has been described, but they could play a role in pushing for the necessary follow-up analyses to be done. Maybe the AR4 will have something on this, though.

James Annan said...


I think you illustrate neatly why this debate needs to be held in the usual manner in the literature. I can see how you might have got your mistaken impressions from reading what has been published, and this is precisely the problem.

For the record, I don't think that Dave Frame can be held particularly responsible for the failings of Hegerl et al. But we certainly reserve the right to comment critically on papers that we find ourselves disagreeing with. For what its worth, the primary author in that case agreed that we had a point worth making, and Frame et al made some similar noises in their Reply. Why are you taking it so personally anyway? We are allowed to disagree, it's how science happens!


Gavin Schmidt has a nice quote expressing a similar POV on abrupt climate change in USA Today.

Anonymous said...


Just a note to say that I definitely support James's right to comment critically on work he disagrees with. As he says, that's how science works. The basic idea of criticism is pretty fundamental to science. I'm optimistic James, Julia and I will be able to come up with a fairly productive exchange of views, and hopefully GRL will think it useful.

Steve: The really messy stuff will come when we try to break it down into regional analyses and hard variables like precipitation. I wouldn't be surprised to see a whole new round of methodological arguments over that step. [I just hope I won't be playing too much of a role in it!]

Anonymous said...

"As I've mentioned before, the implications of the near-term loss of the Tibetan glaciers seems like a perfectly good case study as to why continued global warming is a very bad idea"

There are a number of areas where glacial "regulation" of the outflow from mountains has been important for the growth of large communities. The Chinese and Indian examples are probably the most well known.

One argument may be reched if one were to tot-up all the regions where a reliance on glacial fed hydrology exists (as much to regulate flow as one expects the supplying precipitation may be unaffected), estimate when the glaciers in each region will be affected enough cause problems*, and add up the total population that will thus be affected in each decade from now.

This could be done for warming that's already in the system, and maybe for a selected emission scenario and/or a 3C climate sensitivity?

*the most difficult part of the exercise, I'd guess.

James Annan said...


I think what Dave was getting at - and certainly I think it will be the next challenge - is the difficulty of not just saying "we think this sort of thing is likely to happen" but actually putting meaningful 5-95% confidence intervals on detailed regional outcomes. There are good reasons why we started with global temperature, and even that has proved a bit of a struggle!

Anonymous said...


My post was a response to Steve's as a way of strengthening the case made by his example (and possibly as the natural next question posed by it).

The point you and Dave have made is what I was assuming in my little asterisked aside. However, as some people have been making a stab at regional temperature predictions, I did wonder if these could be used as an initial first-pass.

I realise that confidence intervals are a way away,...and there seems to me to bit of a catch22 situation, in that without having meaningful predictions of glacial melt timescales, we don't know how long we have to come up with meaninful predictions before it's a bit late to act on them. And that point isn't meant as flippantly as it sounds. :-)