Coupling CTSM with WRF #1275
Coupling CTSM with WRF #1275
Conversation
…d ESMF library by renaming wrf esmf modules. This will take care of that.
…ESMF_Initalize inside WRF. Plean NOTE that you should use the configure.wrf file here for compiling for now. Later we can make changes so it would be permanant in WRF configuration.
Key off of the environment variable WRF_CTSM_MKFILE. If that is set, change / add settings needed for linking with CTSM.
This way, you don't need to have this variable set when you compile, only when you configure.
This was put in for testing, but we don't want it in the final version
Three fixes:
(1) Pass XLONG into the subroutine: previously, this was an
uninitialized local variable
(2) Fix reshape routine to know lower bounds of arrays
(3) Convert longitudes to 0..360: WRF has longitudes -180 to 180, but
the mesh file has them 0..360. Convert WRF's longitudes to 0..360
so that they match the longitudes on the mesh file.
Also, a bit of minor cleanup.
For now, move create_gindex there. Other things will be moved there next.
Now named lilac_2d_to_1d
Especially, remove statement that end times may not be needed, based on discussion in ESCOMP/CTSM#877
Rather than piggy-backing off of an existing option, we now have a dedicated option for CTSM. In addition, CTSM calls are protected by available. To determine where CTSMSCHEME may need to be referenced, I searched for clmscheme and noahmpscheme (case-insensitive) throughout the repository. I have added a number of TODO notes where I would like someone else to help determine if CTSMSCHEME should be referenced in various places. (This is not yet tested and likely has some problems.)
Since we don't read things off of the wrf input file, this incompatibility shouldn't matter, I think.
(Not yet tested.)
|
@negin513 The PR commit message lists 18 files, and github says 16 files are changed. |
|
@negin513 |
@davegill Thanks so much for reviewing this. :-) The PR first message listed README and phys/module_ra_cam.F as changed files but the changes in those two files are reverted. I updated the PR message to show the 16 changed files: M Registry/Registry.EM_COMMON |
|
@weiwangncar @dudhia |
|
@davegill Should we inform the DA group so that they can conduct a compilation test? |
|
@liujake |
This fixes the code to work with WRFPlus. Kudos to Jamie Bresch - AS ALWAYS! modified: wrftladj/solve_em_ad.F
|
reopen testing |
|
reopen |
|
reopen |
|
@negin513 @billsacks |
Thanks so much for your help with this. |
Coupling CTSM with WRF (wrf-model#1275)
TYPE: New Feature KEYWORDS: WRF-CTSM coupling, Community Terrestrial Systems Model (CTSM), Community Land Model (CLM), coupled land-atmosphere models SOURCE: Negin Sobhani <sup>1</sup> , Dave Lawrence <sup>1</sup> , Bill Sacks <sup>1</sup>, Sam Levis <sup>2</sup>, Mariana Vertenstein <sup>1</sup>, Dave Gill <sup>1</sup>, Mike Barlage <sup>1,3</sup>, and Fei Chen <sup>1</sup> <sup>1</sup> National Center for Atmospheric Research (NCAR) <sup>2</sup> SLevis Consulting <sup>3</sup> National Centers for Environmental Prediction (NCEP) DESCRIPTION OF CHANGES: This PR includes the capability to couple WRF with Community Terrestrial Systems Model (CTSM) via LILAC (Lightweight Infrastructure for Land-Atmosphere Coupling). CTSM can be used with WRF as one of WRF's land surface model schemes by setting `sf_surface_physics` in the WRF namelist.input file: ``` &physics sf_surface_physics = 6, 6, 6, / ``` For instructions on how to run WRF with CTSM please check [instructions on using CTSM with WRF.](https://escomp.github.io/ctsm-docs/versions/master/html/lilac/specific-atm-models/index.html) LIST OF MODIFIED FILES: M Registry/Registry.EM_COMMON M arch/Config.pl M arch/configure.defaults M arch/postamble M arch/preamble A doc/README.CTSM M dyn_em/module_first_rk_step_part1.F M dyn_em/module_initialize_real.F M dyn_em/solve_em.F M main/depend.common M phys/Makefile M phys/module_physics_init.F A phys/module_sf_ctsm.F M phys/module_surface_driver.F M share/input_wrf.F M share/module_check_a_mundo.F M wrftladj/solve_em_ad.F TESTS CONDUCTED: 1. The jenkins tests are all passing. 2. Initial assessment of the performance of WRF-CTSM5(NWP) compared to WRF-Noah and WRF-NoahMP has been completed over the CONUS domain. Overall, we are finding that WRF-CTSM5(NWP) is performing reasonably well, in some cases even showing improved performance improvement. For example, the following plot shows the bias for maximum daily temperature from NOAH, NOAH-MP, and CTSM for June 2013 over METAR observation sites:  RELEASE NOTE: The capability to couple CTSM with WRF is added and activated by setting namelist option `sf_surface_physics` to 6. This is the initial beta release of WRF-CTSM coupling capability. For instructions on how to run WRF with CTSM please check [instructions on using CTSM with WRF.](https://escomp.github.io/ctsm-docs/versions/master/html/lilac/specific-atm-models/index.html)
TYPE: New Feature
KEYWORDS: WRF-CTSM coupling, Community Terrestrial Systems Model (CTSM), Community Land Model (CLM),
coupled land-atmosphere models
SOURCE: Negin Sobhani 1 , Dave Lawrence 1 , Bill Sacks 1,
Sam Levis 2, Mariana Vertenstein 1, Dave Gill 1, Mike Barlage 1,3,
and Fei Chen 1
1 National Center for Atmospheric Research (NCAR)
2 SLevis Consulting
3 National Centers for Environmental Prediction (NCEP)
DESCRIPTION OF CHANGES:
This PR includes the capability to couple WRF with Community Terrestrial Systems Model (CTSM) via LILAC (Lightweight
Infrastructure for Land-Atmosphere Coupling). CTSM can be used with WRF as one of WRF's land surface model
schemes by setting
sf_surface_physicsin the WRF namelist.input file:For instructions on how to run WRF with CTSM please check instructions on using CTSM with WRF.
LIST OF MODIFIED FILES:
M Registry/Registry.EM_COMMON
M arch/Config.pl
M arch/configure.defaults
M arch/postamble
M arch/preamble
A doc/README.CTSM
M dyn_em/module_first_rk_step_part1.F
M dyn_em/module_initialize_real.F
M dyn_em/solve_em.F
M main/depend.common
M phys/Makefile
M phys/module_physics_init.F
A phys/module_sf_ctsm.F
M phys/module_surface_driver.F
M share/input_wrf.F
M share/module_check_a_mundo.F
M wrftladj/solve_em_ad.F
TESTS CONDUCTED:
completed over the CONUS domain. Overall, we are finding that WRF-CTSM5(NWP) is performing reasonably well, in
some cases even showing improved performance improvement. For example, the following plot shows the bias for
maximum daily temperature from NOAH, NOAH-MP, and CTSM for June 2013 over METAR observation sites:
RELEASE NOTE:
The capability to couple CTSM with WRF is added and activated by setting namelist option
sf_surface_physicsto 6. This is the initial beta release of WRF-CTSM coupling capability. For instructions on how to run WRF with CTSM please check instructions on using CTSM with WRF.