MODIS snow cover fraction assimilation #512
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
1) Updates regarding MODIS obs. reader
- Added the description of MODIS obs. reader at the subroutine
(Table of longitude band is moved up)
- Changed the hdf-related function to read (vdata vs. SDS dataset)
: Previous hdf-related functions are cleaned up
- Updated and checkced the QC procedure by comparing against Matlab result
2) Updates on subroutine cat_enkf_increment
- Removing the cat_diagS at ths subroutine
:Cat_diagS is also removed when the subroutine is called (e.g., ensupd_enkf_update.F90)
- Computed the snow diagnostics to apply condition when to add/remove snow or no further action needed
: This commit has been made to update the error status located @ subroutine get_ind_obs()
: Basically, not significant change made on the code compared to the previous commit. All the print statements for debugging purposes are commented out
: Minor change made on the MODIS SCA reader(in clsm_ensupd_read_obs.F90) to correctly read the MODIS observation
: Things need to be done
1) Check the compressed vector ('Observation') lat, long, and tilenum to see whether it correctly calculates or not
2) Check the tile_grid properties
** Brief updates compared to commits made on 11/08/2021] - No more errors related to the subroutine get_ind_obs() - Checked the details of tilenum, Observation vector, and Obs_pred vector (meeting slide at 11/19/2021) - Another floating error occured when calculating the snow heat content (located within the subroutine clsm_ensupd_upd_routine.F90) code line where showed error: TSN = min(met_force(n)%Tair-MAPL_TICE,0.0) - I've tried to calculate the different between met_force%Tair -MAPL_TICE before entering the loop, and i defined it as deduction (array) and call nth deduction value in the if loop. Still, it revealed a problem. - For the review purpose, It would be great help if you can look close into the subroutine clsm_ensupd_upd_routine.F90 (from line 4076 ot 4261)
- fixed typos - fixed indentation - added/edited comments - removed redundant initialization of cat_progn_incr - removed redundant cat_progn_incr%wesn=0 etc in "do nothing" case of snow analysis - undid partial (!) deletion of an old, commented-out subroutine
…_fraction_assimilation land_assim: yes; pertubation:0; no mwRTM file
…ensup_read_obs.F90)
|
@jpark50 I just fixed your read_obs according to my understanding of your program ( I may be wrong though, please double check). But I believe the more serious error is the dimension of met_force that you passed into the subroutine cat_enkf_increments. The number N_catd and nTiles_ana are different if I understand correctly. You need to have a forcing that matches and re-arranges with nTiles_ana. |
…ame for clarity (clsm_ensupd_read_obs.F90)
…n_assimilation (manually resolved conflict in clsm_ensupd_enkf_update.F90)
|
All GEOSldas nightly 0-diff tests passed after most recent commit. |
|
@gmao-rreichle, the newest iteration with Aqua and Terra runs cleanly. |
biljanaorescanin
approved these changes
Mar 1, 2024
mathomp4
added
0-diff trivial
biljanaorescanin
deleted the
feature/rreichle/MODIS_snow_cover_fraction_assimilation
branch
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Jul 10, 2023 update
Clean up and modifications to MODIS SCF assimilation:
* The apha, beta, and max_incr_swe terms are still hard coded within upd_routines. Rolf, Qing and I discussed where these should ultimately go, ideally in a location that can be changed at run time. Rolf will look into this.
* Hard coded snow layer target thickness partially remains (Line 3519) The max top layer thickness can be imported from catch_constants.F90; however, the remaining values, which should be the distribution of thickness in layers 2 through N_snow in terms of fraction. These could be calculated within upd_routines as 1.0/real(N_snow-1, kind=4) following Stieglitz or could alternatively be exported along with the top layer thickness. In my mind, this decision depends on what relayer2 really needs. Does it simply need reasonable values (0.08 0.5 0.5) or should it be using the forecast values?
* When there is snow in the forecast and the analysis, the code currently calculates the swe and heat content of each layer based on the swe ratio; however, there could be instances where the forecast snow layer is at 0 deg C and has liquid water within it. So, it may be more correct to calculate the heat content as a function of the forecast snow temperature and frozen fraction, which would look something like what I include in a comment on line 4568. I'd be interested in anyone's thoughts.
tagging @gmao-rreichle, @saraqzhang, @gmao-qliu, @amfox37
%%%%%%%
This PR continues where #434 left off.
Starting from Jongmin's branch (
feature/jpark50/MODIS_snow_cover_fraction_assimilation), I removed three bad and redundant assignments in the MODIS SCA obs reader.I also created an associated GEOSgcm_GridComp branch and edited the components.yaml file accordingly. The GEOSgcm_GridComp branch and its connection to the GEOSldas branch is not meant to be final and will need to be revisited.
These changes were implemented in Jongmin's sandbox but hadn't made it into his branch.
For reference, I copied Jongmin's sandbox and experiments to here:
/discover/nobackup/rreichle/Jongmin_SCA_DA/from_Jongmin/
cc: @lcandre2, @weiyuan-jiang, @gmao-qliu