date: Fri, 29 Nov 1996 17:17:05 +0100 from: stefan@nis.pik-potsdam.de (Stefan Rahmstorf) subject: Copenhagen to: Millennia@dkrz.de Dear Millennia friends, let me inform you that the 'model debate' has come to an end through a meeting we had in Hamburg last week (see below). Thus we can go ahead with our original model design, plus we will get help from DKRZ with coupling an ice model to our Millennia model. This is clearly the best outcome we could have wished for the success of our project. On the question of what to do in Millennia 2 we have not had enough discussion yet, I think, and I hope we will be prepared enough in Copenhagen to come to a good work programme. The contributions we had so far in our mailbox could perhaps be summarised as follows. What we may want to do in Millennia 2: 1. Analysis of "one-way" coupled run (performed as part of Millennia 1), i.e. ocean model forced with atmos model output but no feedback to atmosphere. Scientific focus: - air-sea coupling (compare to fully coupled run) - decadal ocean variability - ocean response to coupling: drift - check sensitivity to ocean time steps (and other parameters?) 2. Initial fully coupled runs to finalise/tune coupled model 3. Preparation/testing of data storage/retrieval/analysis machinery 4. Part way through the run: analysis of short term variability, i.e. ENSO in the model 5. Continue analysis of old ECHAM/LSG run 6. Freshwater sensitivity experiments Focus: high latitude convection patterns, stability of thc A comment on our schedule for Copenhagen: is it sensible to have the status reports on the second day and discuss the future on the first? Maybe it is, given Hans' absence on thursday afternoon (crazy jet-setter...). Maybe on the first day everyone can give a very short summary (one overhead slide) of what they have done, what they will do in the second year of Millennia 1, and what they would like to do in Millennia 2, without going into any scientific details. Then discussion of where we should be heading. On day 2 then time to discuss some scientific results. What do you think? Look forward to the meeting, Stefan -------------------------------------------------------------- Summary of Meeting at MPI Hamburg, 23.12.1996 Scientists present: Lennart Bengtson Klaus Hasselmann Uwe Mikolajewicz (joined later) Antonio Navarra Stefan Rahmstorf Hans von Storch This meeting was called to coordinate the plans for long coupled GCM experiments to be conducted at DKRZ. One such experiment is the European Union's MILLENNIA project with participation of 10 European laboratories, which aims to run a coupled model for several thousand years; this project has been going for almost one year. Other more or less long experiments are being planned. Klaus Hasselmann strongly advocated that in view of the high expenses both in CPU time and model development effort, there should not be too many different coupled models. While at DKRZ only the ECHAM atmosphere model is being used, there are four ocean models used in various coupled experiments: LSG, OPYC, HOPE and MOM. The meeting participants agreed that it is valuable to have conceptually different ocean models such as LSG and OPYC, but that HOPE and MOM are conceptually so similar that it would not be advisable to pursue competing strands of coupled ECHAM/HOPE and ECHAM/MOM runs at DKRZ. As both MOM and HOPE have advantages and drawbacks, the meeting felt that one should aim to combine the best aspects of both models in one "Euro-model" for coupled experiments. This could either be done by adding HOPE features into the basic MOM code structure or vice versa, whichever is more practical. The participants saw the major strength of MOM in its user-friendly and flexible modular "community model" structure with numerous compile-time options (different stream function solvers, various tracer mixing parameterisations including the GM scheme, different tracer advection schemes and so on). As an advantage of HOPE the well-developed ice model was mentioned. To decide on the technical aspects of the future model, a task force is to be formed which includes interested scientists with a working knowledge of HOPE and/or MOM; Stefan Rahmstorf was charged with organising this. The programming work involved would be supported by staff from DKRZ's "Modellbetreuung" department. The leadership of this long-term model development project should not be at MPI or DKRZ to avoid a "Hamburg dominance" in climate modeling, but rather at one of the ocean modeling centres (AWI, Kiel). As a more immediate step to support the MILLENNIA project, DKRZ staff will help to couple a good sea ice model to the global MOM model which is used for MILLENNIA. -- Dr Stefan Rahmstorf Potsdam Institute for Climate Impact Research (PIK) PO Box 60 12 03 14412 Potsdam Germany -- E-mail: rahmstorf@pik-potsdam.de Tel: +49 331 2781 160 (home: +49 331 715429) Fax: +49 331 2781 204 (home: +49 331 715429)