high resolution WACCM

[ljdiao@gmail_com]

I am installing FW5 compset with resolution ne30_g16. The cesm_setup got the error message as:"Macros script already created ...skipping
 Machine/Decomp/Pes configuration has already been done ...skipping
 Running preview_namelist script
 infile is /tng4/users/ldiao/CESM/cesm1_2_0/scripts/FW5_2/Buildconf/cplconf/cesm_namelist
CAM build-namelist - ERROR: No default value found for ncdata
user defined attributes:
key=ic_md  val=00010101
Died at /tng4/users/ldiao/CESM/cesm1_2_0/models/atm/cam/bld/build-namelist line 3379.
ERROR: cam.buildnml.csh failed
ERROR: /tng4/users/ldiao/CESM/cesm1_2_0/scripts/FW5_2/preview_namelists failed: 25344"

How can I solve this problem? Do I need to download new input files and which they are? thanks,

 

[jedwards]

I think that you would want to use ne30_ne30 as the resolution for this case.   WACCM support in the cam-se dycore is still in development, I'm not sure that we are ready for you to run this case yet.

 

[mmills]

This is a WACCM5 compset, and is included in the release for development purposes. It is not supported at any resolution, whether finite volume (FV) or spectral element (SE). We aim to have a supported compset in the next release, targeted for May/June 2014. The ability to run WACCM (4 or 5) with full chemistry under the spectral element dycore will depend on the development of a new advection scheme that is currently under development.

 

[santos]

I'm the main developer for WACCM on the SE dycore. It is still very experimental. The only WACCM compset that can be run at ne30 in CESM 1.2 is FWSC, which is not scientifically validated, and it is known to have a significant bug (see the post marked "Please read" at the top of the WACCM forums). We are working on a WACCM5 SC compset as well, but right now we are waiting for the dycore to have a more efficient advection scheme ready before adapting the compset with full chemistry.

 

[ljdiao@gmail_com]

Hi, Dr. santos:As you suggested, I created FWSC compset with resolution ne30_f19_g16, During the building, I got the error messages as:"svn: URL 'https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/ne30np4/map_ne30np4_TO_fv0.9x1.25_aave.120712.nc' non-existent in that revision
File was not found in svn repo: https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/ne30np4/map_ne30np4_TO_fv0.9x1.25_aave.120712.nc
svn: URL 'https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/fv0.9x1.25/map_fv0.9x1.25_TO_ne30np4_aave.120712.nc' non-existent in that revision
File was not found in svn repo: https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/fv0.9x1.25/map_fv0.9x1.25_TO_ne30np4_aave.120712.nc
svn: URL 'https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/fv0.9x1.25/map_fv0.9x1.25_TO_ne30np4_aave.120712.nc' non-existent in that revision
File was not found in svn repo: https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/fv0.9x1.25/map_fv0.9x1.25_TO_ne30np4_aave.120712.nc
svn: URL 'https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/ne30np4/map_ne30np4_TO_fv0.9x1.25_aave.120712.nc' non-existent in that revision
File was not found in svn repo: https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/cpl/gridmaps/ne30np4/map_ne30np4_TO_fv0.9x1.25_aave.120712.nc"I checked the input data repository and there are no mentioned files. How can I find the missing files then?thanks,  

 

[jedwards]

Not sure where you got the f19 - just use ne30_g16. Jim

 

[santos]

FWSC is a non-coupled compset (data ocean, not active ocean), so if you want to run this, you actually want to use just "ne30_ne30" as the resolution, with no f19 nor g16. You should also apply the fix for bug 1766 from the known problems page:http://www.cesm.ucar.edu/models/cesm1.2/tags/cesm1_2/knownproblems.htmlThe patch refers to file models/atm/cam/src/dynamics/se/gravity_waves_sources.F90(Also, I don't have a PhD, so I'm technically not Dr. Santos.)

 

[ljdiao@gmail_com]

Hi, Santos:I am thinking implementing the full waccm-mozart chemistry into FWSC by changing the atm_in file of FWSC. Is it doable? I have experiences of modifying the B_2000_WACCM_CN compset chemistry, but I never tried adding chemical species.My another question is when the waccm supported SE resolution will be availabe, in the mid of next year? It is not clear from your above replies. Thanks,

 

[santos]

It is not a good idea to try modifying a specified chemistry case to resemble the full chemistry. Specified chemistry cases are not simply a different chemistry package; the chemical constituents are actually removed from a large number of physics processes, and their contributions to those processes, if any, are prescribed.In any case, you could theoretically run WACCM with full chemistry and the SE dycore right now, if you generated a suitable initial conditions file. The major obstacles to using WACCM with the SE dycore mainly have to do with performance. The big two obstacles are:

1. Adjusting the dynamics settings so that the simulation is stable and also gives good performance. This is a delicate balance, and I'm not even entirely satisfied with performance of specified chemistry yet, though it is usable. This is difficult mainly due to interactions between the gravity wave paramerization and the SE dycore, which in practice seem to make the SE dycore much less stable than we would theoretically expect based on e.g. CFL considerations.
2. The SE dycore does not do advection efficiently enough that we are prepared to sink resources into doing runs with many constituents (i.e. more than the twenty in CAM5). We are hoping that the addition of the CSLAM scheme to CAM-SE addresses this issue, at which point we can move forward with interactive chemistry.

For next year's release (targeted at late May or early June 2014), I expect that we can have a WACCM5 specified chemistry compset at some level of support (it may or may not be scientifically vetted). CSLAM is also expected to be ready before that release, so there is a good chance that we will have some WACCM compset with full chemistry on the SE dycore. However, it will probably be somewhat experimental.