add to jcb-base.yaml.j2

[nicholasesposito]

Description

Updates jcb-base.yaml.j2

Companion PRs

Should be added before running the workflow/gdasapp and before the jcb-gdas PR#228 and likely at the same time as UFO#3951

Issues

Fixes issue #2035

[CoryMartin-NOAA]

Looks good, thanks. Presumably you want your jcb-gdas changes updated as part of this PR too?

[RussTreadon-NOAA]

Changes look good. Just a question if additional changes are needed.

[RussTreadon-NOAA]

The parm/atm/jcb-base.yaml.j2 which I successfully tested on Hercules includes additional changes.

@@ -101,16 +101,20 @@ atmosphere_obsdataout_suffix: "_{{ current_cycle | to_YMDH
 }}.nc"
 atmosphere_obsbiasin_path: "{{DATA}}/obs/"
 atmosphere_obsbiasin_prefix: "{{GPREFIX}}"
 atmosphere_obsbiasin_suffix: ".satbias.nc"
+atmosphere_obsbiasin_acft_suffix: ".acftbias.nc"
 atmosphere_obstlapsein_prefix: "{{GPREFIX}}"
 atmosphere_obstlapsein_suffix: ".tlapse.txt"
 atmosphere_obsbiascovin_prefix: "{{GPREFIX}}"
 atmosphere_obsbiascovin_suffix: ".satbias_cov.nc"
+atmosphere_obsbiascovin_acft_suffix: ".acftbias_cov.nc"

 atmosphere_obsbiasout_path: "{{DATA}}/bc/"
 atmosphere_obsbiasout_prefix: "{{APREFIX}}"
 atmosphere_obsbiasout_suffix: ".satbias.nc"
+atmosphere_obsbiasout_acft_suffix: ".acftbias.nc"
 atmosphere_obsbiascovout_prefix: "{{APREFIX}}"
 atmosphere_obsbiascovout_suffix: ".satbias_cov.nc"
+atmosphere_obsbiascovout_acft_suffix: ".acftbias_cov.nc"

 bias_files_dict:
     radiance_atms_n20: rad_varbc_params.tar
@@ -122,6 +126,7 @@ bias_files_dict:
     radiance_ssmis_f18: rad_varbc_params.tar
     radiance_cris-fsr_n20: rad_varbc_params.tar
     radiance_cris-fsr_npp: rad_varbc_params.tar
+    aircraft_temperature: acft_varbc_params.tar

 # Local Ensemble DA (LETKF)
 # -------------------------

The additional changes are

1. define atmosphere_obsbiasout_acft_suffix for the output aircraft bias correction file
2. define input and output suffixes for the aircraft bias correction covariance files. If JEDI can read and write aircraft bias correction coefficients and covariances from the same file, these additions are not necessary.
3. add aircraft_temperature to point at the tarball containing aircraft bias correction files. For my tests the tarball contains two files: atmosphere_obsbiasin_acft and atmosphere_obsbiascovin_acft. If we only need a single files for both quantities, we don't need a tarball. However, in this case we will need to modify the python scripting which stages the bias correction data. It assumes it's working with tarballs.

[nicholasesposito]

1. This makes sense.
2. There are no acft covariance files but the code still requires there be a covariance file, so couldn't we just re-use the atmosphere_obsbiasin_acft and atmosphere_obsbiasout_acft ?
3. Because we are merging all three aircraft yamls into one, we do not need aircraft_temperature anymore. They could just be called atmosphere. However if the point is to mention that it is specifically temperature correction, then sure we can keep "aircraft_temperature"