cc: carl mears , "'Dian J. Seidel'" , Frank Wentz , Karl Taylor , Leopold Haimberger , Melissa Free , peter gleckler , Phil Jones , Steve Klein , Steven Sherwood , "Thorne, Peter" , Tom Wigley , Tom Wigley date: Fri, 7 Dec 2007 23:48:07 -0500 from: John Lanzante subject: Re: [Fwd: sorry to take your time up, but really do need a scrub to: santer1@llnl.gov, John Lanzante Dear Ben and All, Since I've been on travel for most of this week I haven't been able to focus on the various emails that have been flying to and fro. Now that I've returned, processing them collectively, things seem to be boiling down to several distinct issues (then again, maybe it's just exhaustion from not having gotten much sleep the last few days!). I think it's important to separate out two issues (perhaps this is something akin to optimal detection by rotating away from either scientific or political space). I think it's very important to separate as much as possible the scientific and political responses. Regarding the political aspect, there are few certainties in life, but the fact that no matter how rationally and exhaustively we respond as scientists, the contrarians are going to come back at us with half-truths, distortions, cherry-picking of facts, etc. is probably as certain as it gets. When Charlie Brown goes up to kick the football, Lucy is most surely going to pull it away at the last second. So I suggest that we carefully consider 2 separate activities: (1) the political response to the recent contrarian paper(s) (2) the scientific response to new developments in the field. I don't think it is very fruitful to get worked up into a frenzy in an attempt to deal with (1) in a largely scientific way. The fact is, prior to CCSP SAP1.1 Santer et al. (2005), and other contemporary studies we had good scientific reason to believe there was no real discrepancy (sfc vs. upper warming, models vs. obs). These related studies greatly strengthened the argument, and, subsequently a number of recently published and developing works strengthen the case even more. But little of this is relevant to (1). Steve Sherwood said: > I'm not sure I agree that it's worth wasting a lot of time doing more > work and writing responses in the scientific literature, motivated only > by this, unless you think there would be real scientific value there. > But a short review piece or commentary to Science might make sense, if > they are interested in it, once there are enough new studies to include. I agree with this. I think this would be much more fruitful and timely than if a lot of new scientific work is to be done, and since in fact it is not necessary to do any new scientific work to refute what is being said. Steve Sherwood said: > I suggest disseminating this tidbit to Revkin and any other reporters we > come across. I think it qualifies as scientific fraud. Maybe we could > get the Times to do a story on that. I'm tired of this crap. To some extent I see some merit in this approach. In addition, one could point out, matter of factly, that it is rather odd that John Christy, who signed off on the conclusions of CCSP SAP1.1 as a co-author is now the co-author of a paper that contradicts SAP1.1. At the same time I think any such actions would have to be done very, very, carefully, taking the high road, and sticking to the facts as conscientiously as possible, to avoid being dragged into a mud-slinging contest. As far as (2) is concerned I would suggest thinking very carefully about what new and emerging findings are of scientific relevance and charting a strategy to exploit them. The plan that Ben laid out looks pretty good, although I haven't had the time to really ponder it as much as I'd like. I think the biggest factor is that there are new datasets, and I believe these almost universally support the contention that models and obs are not in disagreement. Dian Seidel said: > Second, choosing a different set of models for one > variable (temperature) than for another (humidity) seems highly > problematic. If we are interested in projections of other variables, > e.g. storm tracks or cloud cover, for which D&A has not been done, which > group of models would we then deem to be most credible? I don't have a > good alternative to propose, but, in light of these considerations, > maybe one-model-one-vote doesn't appear so unreasonable after all. I think the notion of selecting "good" vs. "bad" models is a very knotty one. When this issue is brought up I'm reminded of an analogy presented by my former GFDL colleague and long-time friend Tony Broccoli. During the development phase of the new GFDL modeling system (fms) a lot of diagnoses were presented on various aspects of new model configurations. Tony likened the effort to create the "best possible model" to that of packing the car for vacation (he even had cute little graphic to illustrate it). In trying to pack the car you discover that there is something that won't fit. So you remove everything from the trunk and repack so as to fit the left-out item, only to discover that now something else won't fit. One can repeat the exercise ad infinitum without resolution of the problem. What it boils down to is that if you try to "fix" a particularly deficiency in a given model, there is a non-trivial chance that you are going to degrade something else. The relevant point is that if one considers various variables, metrics, standards, etc., some models are going to be good at some things, and poorer at others. While perhaps one could designate some subset of models as being poorer in a lot of areas, there probably never will be a single universally superior model or set of models. We should keep in mind that the climate system is complex, so that it is difficult, if not impossible to define a metric that captures the breath of physical processes relevant to even a narrow area of focus. So while there may be merit in identifying and either eliminating or down-weighting models based on some criteria developed with a specific application in mind, there is probably also merit in the "one model one vote" approach. Finally, let me apologize in advance for slow responses during the next few weeks as I expect to be taking time off in perhaps a haphazard fashion, having additional "use or lose" annual leave to burn as well as the time period from Christmas to the week after New Years. Best regards, ______John