Added bkg_lfo collections for LDAS processing #707
Conversation
gmao-jstassi
added
0 diff trivial
There was a problem hiding this comment.
@gmao-jstassi which tag of the DAS have you started from? @saraqzhang and @gmao-rreichle is this a variable typically distributed among GMAO products? I ask to know about the consequences of commenting this out ...
@rtodling: The "lfo_inst" and "lfo_tavg" Collections are routinely output in FP, M-2, GEOS-IT, etc as diagnostics from the Corrector segment. Their purpose is to drive offline land simulations. What we need here (for the coupled land-atm analysis) is the output of the same variables but from the Predictor segment (i.e., the bkg...lfo... Collections). What exactly do you want to comment out, or what do you see that is commented out but think it shouldn't be? |
|
@rtodling, I started from your GEOSadas-5.41.1 tag and updated the HISTORY.rc.tmpl file with a version given to me by Sara. There shouldn't be any other changes. |
… from these collections: * tavg1_2d_lfo_Nx- * tavg1_2d_lfo_Nx+- * bkg.tavg1_2d_lfo
|
The HISTORY.rc.tmpl file was updated with version from Sara Zhang with SWLAND removed from these collections:
|
There was a problem hiding this comment.
@saraqzhang, @rtodling, @gmao-jstassi: I added a few inline comments below. Please take a look.
| steps: | ||
| - name: Checkout repo | ||
| uses: actions/checkout@v3.3.0 | ||
| uses: actions/checkout@v4 |
There was a problem hiding this comment.
The changes in push-to-develop.yml and validate_yaml_files.yml aren't related to the intent of the PR. It doesn't matter to me if they're included, just pointing this out to avoid inadvertent changes and making sure nothing is out of sync.
| 'bkg.eta' | ||
| 'bkg.sfc' | ||
| 'cbkg.eta' | ||
| 'bkg.inst1_2d_lfo' |
There was a problem hiding this comment.
What do we want these Collection names to be?
bkg.inst1_2d_lfo includes most but not all of the name of the standard (diagnostic) collection.
Using bkg.inst1_2d_lfo_Nx would be more consistent, where the "Nx" part expresses that the file is at the "native" resolution.
Since this is a "bkg" Collection, however, it may be better to keep only the essence of the standard collection name, as in bkg.lfo_inst1, or maybe even bkg.lfo_inst if we don't want to hardwire the 1-hour time step into the name.
My suggestion would be to use bkg.lfo_inst and bkg.lfo_tavg, which is probably most consistent with the other "bkg" Collection names.
(Note that GEOS-ESM/GEOSldas_GridComp#81 will need matching edits in LDAS_Forcing.F90.)
There was a problem hiding this comment.
" Nx " should have not been in the lfo name since it conventionally indicates a lat/lon grid collection. We have had it in lfo Nx+- collection for the convenience ( existing name format in ldas) even though they output in CS resolution. We certainly can modify the names as Rolf suggested here, then there are several places needed to match in the adas PR #327 and in ldasGC PR#81.
| tavg1_2d_lfo_Nx+-.end_time: >>>IOEBKGT<<< , | ||
| tavg1_2d_lfo_Nx.fields: 'SLRSF' , 'SOLAR' , 'SWGDN' , | ||
| 'SWLAND' , 'SURFACE' , | ||
| tavg1_2d_lfo_Nx+-.fields: 'SLRSF' , 'SOLAR' , 'SWGDN' , |
There was a problem hiding this comment.
There's no need to keep the "lfo_Nx+-" Collections in the template. They are superseded by the "bkg.lfo" Collections. Both the "Nx+-" inst and tavg Collections should be removed.
| bkg.tavg1_2d_lfo.end_time: >>>IOEBKGT<<< , | ||
| bkg.tavg1_2d_lfo.fields: 'SLRSF' , 'SOLAR' , 'SWGDN' , | ||
| 'LWS' , 'IRRAD' , 'LWGAB' , | ||
| 'PCU' , 'SURFACE' , 'PRECCU' , |
There was a problem hiding this comment.
FYI, in the updated file specs of M21C, we now rename as follows:
PCU --> PRECRAINCU (instead of PRECCU)
PLS --> PRECRAINLS (instead of PRECLS)
We could make this change now in the bkg.lfo Collections, or we could stick with the older names (PRECCU, PRECLS) until the FP file specs are updated to the M21C conventions.
There was a problem hiding this comment.
Let's not worry about this now.
If this is merged into |
5 participants
Added collections for LDAS processing