Repository navigation
Change data streams to use xmlvariables from hardcoded years (no behavior changes, more flexibility) - #459
Change data streams to use xmlvariables from hardcoded years (no behavior changes, more flexibility)#459manishvenu wants to merge 2 commits into
Conversation
The JRA55 (IAF, 1p4_2018, 1p5_2023) and CORE2 IAF datm streams, the matching drof IAF and JRA streams, drof GLOFAS, dice ssmi_iaf and docn interannual hardcoded stream_year_align/first/last. A case could not choose its forcing years, and every year of data was expanded into the stream file list and the input data list. Point these streams at DATM_YR_*, new DROF_YR_* and DICE_YR_* variables, and SSTICE_YEAR_* (docn), the same way the land forcing modes already use DATM_YR_*. The compset defaults equal the old hardcoded values, so default stream output is unchanged. Setting the variables now aligns the data and narrows the file list. Repeat-year, normal-year, ERA5 and the aerosol, deposition, ozone and CO2 time series streams are unchanged.
The docn.rst entries called the variables SSTICE_YR_* and said they were only used in prescribed mode. Use the real SSTICE_YEAR_* names, add interannual mode, and render the taxmode quotes as plain quotes.
|
@alperaltuntas Are you able to add a reviewer to this? I don't think I am able to. |
|
@alperaltuntas - I'm assigning you as reviewer on the CESM side. @mvertens - I'm assigning you as a reviewer in case you want to review this for NorESM. Feel free to remove yourself if you don't want to review it. @manishvenu - What testing has been done on this? I'd like to see a run of aux_cdeps testing with baseline comparisons. (There should be baselines for aux_cdeps in every CESM prealpha baseline.) |
Thanks, @billsacks. I ran aux_cdeps with --compare alpha11e and it looks like everything passed in the baseline comparison. You can see test ouput here: /glade/derecho/scratch/manishrv/aux_cdeps_pr459/full |
Description:
The JRA, CORE2 IAF and GLOFAS data streams had their years hardcoded, so a case couldn't choose its forcing years, and every year of data went into the input data list. We made the data stream years settable through xml variables, the way the land forcing streams already use
DATM_YR_*. Setting them also narrows the file list to just those years.This lets us make sure we are using the correct year of JRA or other data product, which is useful for shorter cases where the year of the forcing might matter.
Changes:
DATM uses the existing
DATM_YR_*. DROF and DICE get newDROF_YR_*andDICE_YR_*variables, and we added similar functionality to docn: itsinterannualmode now readsSSTICE_YEAR_*. The defaults match the old hardcoded years, so nothing changes unless you set them.pr_mom.derecho, all b4b).