updates for integrated setup and coupled run of LADAS #81
Conversation
saraqzhang
added
enhancement
|
This PR is being prevented from merging because you have added one of our blocking labels: Contingent - DNA, Needs Lead Approval, Contingent -- Do Not Approve. You'll need to remove it before this PR can be merged. |
There was a problem hiding this comment.
@saraqzhang : I added a few inline comments that probably require further discussion at our next tag-up
modified: GEOSldas_App/ldas_setup modified: GEOSmetforce_GridComp/LDAS_Forcing.F90
modified: GEOSldas_App/ldas_setup
There was a problem hiding this comment.
@saraqzhang, I added a few inline comments/suggestions/questions. Please take a look, thanks
| VERSION: 1 | ||
| EXPID: GEOSldas_expid | ||
|
|
||
| COLLECTIONS: |
There was a problem hiding this comment.
I think we need to add the usual diagnostic tavg[*]_2d_lnd_Nx, and inst[*]_2d_lndfcstana_Nx collections that we will want to output from the LDAS whenever it's running in the coupled land-atm DA system.
The time step could be hourly (as in FP and reanalysis) or perhaps 3-hourly (as in SMAP L4).
Note that the aforementioned collection names are from the MERRA-2 era and reflect what is in FP now. For M21C, there's slightly different naming convention. I assume the newer convention will be adopted in FP when it transitions to the forthcoming "v12" (181-level) GCM, so maybe it's best to use the newer collection naming convention here.
There was a problem hiding this comment.
those diag collections can be added to the template. Because they are lat/lon outputs, more modification is needed in setup, we will need to insert the grid name ( from setup info) in the specification of "regrid_name", probably a corresponding "grid_label" as well.
|
|
||
| EXPID: GEOSldas_expid | ||
|
|
||
| COLLECTIONS: |
There was a problem hiding this comment.
See comment above -- add "lnd" and "lndfcstana" collections (ens avg)
| elseif (tmp_tag(ii)(1:3)=='bkg') then | ||
|
|
||
| use_bkg = .true. | ||
|
|
There was a problem hiding this comment.
I have to think more about this, but can't we simply replace "use_Predictor" with "use_bkg"? The purpose is the same... use lfo output from the Predictor segment of the GCM.
There would probably be issues with backward compatibility if we eliminated "use_Predictor" and the Nx+- syntax, but we may not need backward compatibility in this case.
This requires more thought
There was a problem hiding this comment.
I feel that replacing use_Predictor seems to be more disruptive to the current code.
| @@ -6646,6 +6704,7 @@ program ut_parse_G5DAS_met_tag | |||
| met_tag_in_vec = (/ & | |||
| 'gcmexpname' , & | |||
| 'gcmexpname__Nx+-' , & | |||
| 'gcmexpname__bkg' , & | |||
| 'gcmexpname__precCORRPREC', & | |||
| 'gcmexpname__precCORRPREC__Nx+-' , & | |||
| 'gcmexpname__Nx+-__precCORRPREC' , & | |||
There was a problem hiding this comment.
we may need the option to run bkg with precip corr, as in: 'gcmexpname__bkg__precCORRPREC' and/or 'gcmexpname__precCORRPREC__Nx+-' (not sure whether both options for ordering the substrings are needed (or what the difference may be)
requires more thought
There was a problem hiding this comment.
for LADAS with precip correction in agcm ( corrector and predictor) ?
|
|
||
| if (use_bkg) then | ||
|
|
||
| do j=1,N_GEOSgcm_vars | ||
|
|
||
| GEOSgcm_defs(j,3) = G5LADAS_defs(j,3) | ||
| GEOSgcm_defs(j,4) = G5LADAS_defs(j,4) | ||
|
|
||
| end do | ||
|
|
||
| end if |
There was a problem hiding this comment.
This should be handled in the if-block in lines 3783 etc
updates are implemented to integrate ldas_setup in the coupled LADAS setup procedure. adas fvsetup calls ldas_setup with optional command line inputs, so that coupled ldas experiments are setup according to adas experiment specifications such as exp path, exp id, begin date and gid.
Related PR: GEOS-ESM/GEOSadas#327