From:	mzhang@notes.cc.sunysb.edu
Sent:	19 June 2009 14:22
Subject: summary, next plan, and action items for the CFMIP-GCSS study
All,

It was great to meet many of you at the CFMIP-GCSS meeting last week
in Vancouver. This email is to remind you and report to you what we
have decided at the meeting: 


SUMMARY
--------------
1. The SCM models simulated a range of cloud distributions that
   share many features in some of the GCMs that have been examined. The
   SCM results also provided many hints of cloud feedbacks in some of
   the models.   

2. The LES results differ a lot among each other. To understand the
   reasons of the differences, we have decided to do more standardized
   LES simulations at s11. While trying to reconcile the differences,
   we recognize that our study is the first in GCSS in which the LES
   models are integrated to quasi-equilibrium states. We are therefore
   in new territory. But this is essential for CFMIP. 

3. We are still open for new submissions. A few modeling groups are
   very close to submitting their results or revised results (LMD,
   CCC, Univ. Wiscousin).  


NEXT PLAN
---------------
Future tasks of our case study are to 

(1) understand the LES results, 
(2) learn about what processes operate in the SCMs to account for the
    (simulated clouds in the control climate,  
(3) analyze the physical mechanism of cloud feedback in the SCMs, 
(4) conduct sensitivity tests of the cloud feedback in the SCMs based
    (on the understanding of physical mechanisms, 
(5) connect results from the case study with the behavior and feedback
    (of clouds in the GCMs. 

Our aspiration of this study is to gain enough insight of the possible
mechanisms of cloud feedbacks so that we can clearly interpret the
GCMs and to more effectively use observations. 

We are planning a small one-day-and-half invitation-only meeting at
Stony Brook (all participants of this case and possibly few others) in
middle November to early December this year. At the meeting, we will
discuss a joint overview paper, and share analysis results from each
participating model. We expect each group to present an in-depth
analysis of its model results for at least some of the tasks from (2)
to (5). The time constraint at last week's CFMIP-GCSS meeting had
prevented us from examining individual models in great details.  


ACTION ITEMS 
--------------------

1. Standardized LES simulation for those who can participate 

The specific LES configuration is as follow:

(1). Same radiation code. Peter Blossey and Marat Khairoutdinov will
    (make the CAM3 radiation code as a stanalone subroutine for all LES
     groups to use)
(2). 3-D simulations, horizontal resolution of 50 meters with 96X96
     grids, vertical resolution of 20 meters to an altitude of 4 km. 
(3). Upper boundary at 4 km will be prescribed as in the initial
     condition. The downward solar and infrared radiation at 4 km will be
     given (which will be based on calculation from the initial profiles). 
(4). Number density of cloud particles will be prescribed as 100 /cm3.
(5). Mean wind in the model domain will be relaxed toward the profile
     in the forcing data with a relaxation time scale of 10 minutes. 
(6). Drizzles are included.
(7). Surface fluxes are calculated internally by the model.

The standardized simulation will be done at s11 for both the control and perturbed climate.

Chris Bretherton, Marat Khairoutdinov, and Peter Blossey will be the
LES lead of the study. Using the same LES configuration is our first
step to finally obtain the benchmarks for the SCMs.  

Time Table:
-----------
July 15, radiation code ready (Blossey and Khairoutdinov)
July 15, addition of an initial water vapor profile, to the forcing
data file, for quicker spin up of the LES to equilibrium states, and
addition of the downward shortwave and longwave flux at 4 km (Zhang) 
Sept. 15, initial LES simulation by individual groups due
Oct. 15, final LES simulation due

2. SCM participants

Pending on further tests of the forcing setup and the experimental
results from LES, we may refine the case if this is found
necessary. You will be informed about this. Meanwhile, we expect you
to look into your results carefully and share your findings with the
group. You will be updated on the LES side of the study. 

3. Submitted Data

I will make plots of the submitted results from all models and post
them at an internal website. I will do this after this week's CCSM
workshop. You will be hearing from me soon. 


The content of this email will be posted on a website after I get back
to Stony Brook. Let me know if you have any commends or
questions. Mark Webb is working on a name to replace the long
"CFMIP-GCSS SCM/LES Case Study." Suggestions welcomed. 

Cheers,

Minghua