Daily wind cycling error in MOM6 ocean-only idealized model #631
Comments
|
For reasons I will not defend, the netcdf data needs certain attributes (not standard). I'm not sure where the documentation is (somewhere on the FMS pages) but an example of a file that does work is in |
|
Sorry, just reread and saw that you had already sorted out the attributes, and found a fix. The 365 days is a good gotcha - another oddity we didn't know about! |
|
Hi! I found something interesting in the MOM6 source code: (directory: MOM6/config_src/drivers/solo_driver/MOM_surface_forcing.F90) This is within subroutine wind_forcing_from_file(sfc_state, forces, day, G, US, CS): ... if (time_lev_daily < 31) then ; time_lev_monthly = 0 time_lev_daily = time_lev_daily+1 select case (CS%wind_nlev) I marked the vital component boldly, you could see there is no case of other time levels except daily cycling and monthly cycling. |
Refactored the solo_driver version of MOM_surface_forcing to modify how the time levels to read from forcing files, with a series of 11 new runtime parameters (WIND_FILE_DAYS_PER_RECORD, etc.) giving the user to specify how many days each time record spans (for positive values) or the number of time records in the file per day (for negative values). If these parameters are set to 31, the month returned by the FMS `get_date()` function sets the time level. If they are set to 0 (the default) the previous behavior is obtained, in which the time level is determined from the number of records in the file. To always take the first record in the file, set these parameters to large values (greater than 1000000 will always work, but in practice smaller values do the same thing). By default all answers are bitwise identical, but there are up to 10 new runtime parameters in some MOM_parameter_doc files for cases that have `WIND_CONFIG = "file"` or `BUOY_CONFIG = "file"`. The list of new runtime parameters is `WIND_FILE_DAYS_PER_RECORD`, `LONGWAVE_FILE_DAYS_PER_RECORD`, `SHORTWAVE_FILE_DAYS_PER_RECORD`, `EVAPORATION_FILE_DAYS_PER_RECORD`, `LATENTHEAT_FILE_DAYS_PER_RECORD`, `SENSIBLEHEAT_FILE_DAYS_PER_RECORD`, `RAIN_FILE_DAYS_PER_RECORD`, `SHORTWAVE_FILE_DAYS_PER_RECORD`, `RUNOFF_FILE_DAYS_PER_RECORD`, `SSTRESTORE_FILE_DAYS_PER_RECORD` and `SALINITYRESTORE_FILE_DAYS_PER_RECORD`. The problem identified in github.com/NOAA-GFDL/issues/631 is addressed by this commit, and that issue can be closed once this commit is merged onto the main branch of MOM6.
Refactored the solo_driver version of MOM_surface_forcing to modify how the time levels to read from forcing files, with a series of 11 new runtime parameters (WIND_FILE_DAYS_PER_RECORD, etc.) giving the user to specify how many days each time record spans (for positive values) or the number of time records in the file per day (for negative values). If these parameters are set to 31, the month returned by the FMS `get_date()` function sets the time level. If they are set to 0 (the default) the previous behavior is obtained, in which the time level is determined from the number of records in the file. To always take the first record in the file, set these parameters to large values (greater than 1000000 will always work, but in practice smaller values do the same thing). By default all answers are bitwise identical, but there are up to 10 new runtime parameters in some MOM_parameter_doc files for cases that have `WIND_CONFIG = "file"` or `BUOY_CONFIG = "file"`. The list of new runtime parameters is `WIND_FILE_DAYS_PER_RECORD`, `LONGWAVE_FILE_DAYS_PER_RECORD`, `SHORTWAVE_FILE_DAYS_PER_RECORD`, `EVAPORATION_FILE_DAYS_PER_RECORD`, `LATENTHEAT_FILE_DAYS_PER_RECORD`, `SENSIBLEHEAT_FILE_DAYS_PER_RECORD`, `RAIN_FILE_DAYS_PER_RECORD`, `SHORTWAVE_FILE_DAYS_PER_RECORD`, `RUNOFF_FILE_DAYS_PER_RECORD`, `SSTRESTORE_FILE_DAYS_PER_RECORD` and `SALINITYRESTORE_FILE_DAYS_PER_RECORD`. The problem identified in github.com//issues/631 is addressed by this commit, and that issue can be closed once this commit is merged onto the main branch of MOM6.
|
PR #793 provides the runtime parameters that are needed to deal with this issue, so it can now be closed. |
Hi! I want to set wind stress with 3 months of variability as input data in the MOM6 ocean-only idealized model, but I found it is constant without varying... could anyone help me with this issue? Thanks!
The tau_mag output (show as constant):
My input data showed below:
Also, my configuration setting:
Recent finding: The input data could only be read with 365 days, less or more days will make the MOM6 ocean-only model read wind stress as a constant defaultly. To derive wind stress variability with 3 months effect, I chose to save model output every 360 days, then it worked well.
The text was updated successfully, but these errors were encountered: