WACCM-X with specified dynamics?

[inos@bas_ac_uk]

Hi all,I was wondering if it's possible to run WACCM-X with specified dynamics. I am aware of WACCM compsets that do this, but I would like to study higher altitudes if possible, so WACCM-X would be better. Can this be done by adapting an existing compset?Thanks,Ingrid 

 

[joemci]

Hi Ingrid,We have recently put together a WACCM-X case using specified dynamics.  There is not an official compset so we set up a development compset based on F_2000_WACCMX.  We are in the process of documenting the steps we took to get this running and will include those here very soon.Joe

 

[inos@bas_ac_uk]

Ok - thanks!

 

[joemci]

Ingrid, what version of CESM are you going to use to run WACCM-X SD?
Joe

 

[inos@bas_ac_uk]

Hi Joe,I've recently set up CESM1.2.2 on our local cluster at the British Antarctic Survey, so I would in principle be looking to use that version, unless you advise me otherwise.
Ingrid

 

[joemci]

Hi Ingrid,Here is what to do:Go to download.hao.ucar.edu/pub/waccm and grab the files waccmx_sd.tar and f.e13b02.F_1955_2005_b_1950-2100_2deg_refc2_wa4_ma.cam.i.1975-01-01_103L_c150223.ncUntar waccmx_sd.tar into a directory and go into the scripts directory and create a case with the compset F_1975-2010_CCMI_REFC1SD_WACCMX_MAGo to the case directory and add the following line to the user_nl_cam filencdata="/f.e13b02.F_1955_2005_b_1950-2100_2deg_refc2_wa4_ma.cam.i.1975-01-01_103L_c150223.ncwhere is the path to this file on your system.Set up, build, and run and see what happens.
Joe

 

[inos@bas_ac_uk]

Thanks Joe! I'll try this.Ingrid 

 

[inos@bas_ac_uk]

Hi Joe,I finally started to set up my WACCM-X SD run. I followed your instructions, but I'm getting some errors - see the attached cesm log file. First, a number of times "NetCDF: Variable not found" is reported. It also complains a few times about an "Invalid dimension ID or name" and "Attribute not found". I'm not sure if these are a serious errors or if they can be ignored. Second, it can't find a file that it appears to need: "GETFIL: FAILED to get /work/n02/n02/inos/modeldata/inputdata/atm/cam/met/MERRA/1975/MERRA_19x2_19750102.nc". I've checked and it is indeed not there. Was I meant to download this separately from somewhere? The first one (MERRA_19x2_19750101.nc) is there. I also wonder why it needs data for 1975. I'm interested in an interval in 2013 (Oct 22 - Nov 11) and to allow for spinup I specified the start date as 2013-01-01 in my env_run.xml file. I understand that my run will require MERRA data for 2013 which clearly I don't have. Is this available from NCAR or do I need to get it myself?I also have a few more general questions:1) I'm not actually sure what version of WACCM-X I am using through the download you gave me. Can you provide more information on that? Is there a published paper I can refer to?2) Over what pressure level or height range am I nudging towards the MERRA data? And how could I modify that if I need to?3) I am also interested in running the same simulation without specified dynamics, to see what difference it makes. Will I accomplish that by setting up a case with the F_2000_WACCMX compset within the waccmx_sd directory I got from your tar file?I hope you can help with this.Regards,Ingrid

 

[jedwards]

The netcdf messages are all warnings and can be safely ignored.   I am uploading the MERRA data for 1975 and 1979 to the database, it should be available now and running thecheck_input_data script should retreve it for you.   

 

[inos@bas_ac_uk]

Thanks for your reply. However, when I run check_input_data nothing happens. Also, I'm not sure that I actually need the 1975-1979 MERRA data, as I'm interested in an interval in 2013... I think I may need to change something in user_nl_cam(?). I checked the cam.input_data_list file in Buildconf and it lists the following MERRA-related variables:met_data_file = /work/n02/n02/inos/modeldata/inputdata/atm/cam/met/MERRA/1975/MERRA_19x2_19750101.nc
met_data_path = /work/n02/n02/inos/modeldata/inputdata/atm/cam/met/MERRA
met_filenames_list = /work/n02/n02/inos/modeldata/inputdata/atm/cam/met/MERRA/CCMI_sd_filenames.txtThe CCMI_sd_filenames.txt file lists lots of files covering the period 1975 to 2012, but they are a) not being downloaded automatically, and b) the year 2013, which is the one I need, is not there (and c) the model doesn't actually try to find data for 2013 even though that's the year I'm trying to simulate). There are also a number of other outstanding questions from my earlier message. Since Joe McInerny supplied me with the WACCM-X SD code and instructions, he might be best placed to answer these (though I do appreciate your help).Cheers,Ingrid

 

[var@coastal_edu]

Hi Joe,Can you also please upload the MERRA data for 1976, 1977, and 1978 to the database? These years are also apparently not available on the Earth System Grid at NCAR.Thank you.

 

[mmills]

Dear Var,MERRA and MERRA2 are satellite era reanalysis products provided by NASA. MERRA goes back to the year 1979, and MERRA2 goes back to 1980. For the purpose of spinning up SD versions of CESM for several years prior to the start date of these products, we have copied files and adjusted the date variable to represent previous years. For example, fields from 1979 can be copied back to 1976, 1977, and 1978. Alternately, different years can be copied to different previous years, if one is concerned about matching the phase of the QBO or some other parameter. I have attached a shell script that can be used to change the dates on copied MERRA fields for this purpose.

 

[var@coastal_edu]

Dear Mike,Thank you for the clarification. I was confused by the mentioning of 1975 MERRA data in the earlier postings. Now, I understand. Also, thank you for the script.Best,Var