date: Fri, 06 Feb 2004 10:24:31 -0700 from: Tom Wigley subject: MAGICC stuff to: Sarah Raper , Sarah Raper Dear Sarah, Rachel continues to amaze me by her ignorance. Still, it *would* be nice to add some more models to the set that you have calibrated against MAGICC -- independent of Rachel's request. I presume that one could do this for any models that are in the CMIP data base -- ??. As you know, I will be going to the meeting in Oxford on April 15,16. I was planning to come over earlier, probably before Easter. There are various things I want to do/ people to see. Perhaps I could come up to Norwich around April 7,8? There are a few things that we could sort out much more easily if we were together. For instance, I still don't understand what your concerns are about the effective climate sensitivity as backed out of MAGICC -- but I am sure you can enlighten me quickly face to face. At the same time, we could calibrate a few more AOGCMs, since these issues must be closely related. If this were a possibility then do we need to get more numbers from CMIP? The other thing I am playing around with is estimating a pdf for sensitivity from observed temperature, using MAGICC. I have devised two different ways to do this that I think are more transparent than what others have done, and which circumvent what I see as defects in previous analyses. I should have got far enough with this by April for us to be able to look at it together. This leads on to the IPCC sensitivity meeting. The dates are July 26 -- 29. I have hard copy of the program. The participants (speakers) include you on the first day on "Climate sensitivity as a function of time". Neither I nor Ben have been invited. I am a bit disappointed by this, since I have done a lot of work on the subject over the years. However, I haven't published much on this recently. On the other hand, late July is a difficult time for me to travel. The fact that I am not going and that you apparently are is part of my reason for wanting to get together in April. For calculating lambdas, this is dim in my mind too. The LAMCALC subroutine was written in 1995. The issue of negative lambdas came up more recently and I put a flag in to warn users if a negative value occurred -- with a suggestion to change XKLO. It is interesting that Boer gets negative lambdas -- I will have to check this out. The subroutine calculates equilibrium temperatures over the four boxes, but I don't think I ever looked at these. Might be interesting. It is good that Jean has the TSU job -- can't think of anyone better for such a job. Also, good that Tim Lenton has a job at UEA. If he were there before Easter I would very much like to talk to him. Do you know when he is coming to UEA? Let me know about pre-Easter. All the best, Tom.