Invalid time coordinate info on files in /cluster/shared/noresm/inputdata/atm/waccm/lb/cmip7_ghg_version20260607/

[mvertens]

@DirkOlivie -
It seems that the time coordinate in the CO2 bottom level file that is being used in n1850 compsets - /cluster/shared/noresm/inputdata/atm/waccm/lb/cmip7_ghg_version20260607/ghg_cmip7_CMIP_CR-CMIP-1-0-0_1850_monthly_gnz_0p5degLat_version20260207.nc
is all special values.

ncdump -v time -t /cluster/shared/noresm/inputdata/atm/waccm/lb/cmip7_ghg_version20260607/ghg_cmip7_CMIP_CR-CMIP-1-0-0_1850_monthly_gnz_0p5degLat_version20260207.nc
 time = _, _, _, _, _, _, _, _, _, _, _, _ ;

This seems to be the case with all the files in
/cluster/shared/noresm/inputdata/atm/waccm/lb/cmip7_ghg_version20260607/

This seems problematic to me. I am trying to modify the 1850 file to be used for NorESMhub/NorESM#825 (comment)

[mvertens]

@DirkOlivie @gold2718 - I am seeing that the date has valid values - so I think maybe I am not understanding how CAM handles the time interpolation if only the date variable on the file has valid values.

[DirkOlivie]

@mvertens I thought that CAM was using the date variable for doing the time-management of the GHG files, and that variable is of the form YYYYMMDD. I better should have put the time variable also to values that make sense with a corresponding "since 1750-01-01". I will modify the files such that they have also the time variable well defined.

[mvertens]

@DirkOlivie - I think you are right regards CAM using the date variable. However, I was trying to extend these files to be used for NorESMhub/NorESM#825 (comment) and the time coordinate was problematic. It would be good to add a calendar attribute to the time coordinate as well. This is not critical at this point since I am creating a new file with longitude coordinates that are needed for CDEPS to ingest this data.