Test satellite radiance bias correction for ATMS NPP in MPAS-JEDI

[xyzemc]

For applying the satellite radiance bias correction in JEDI, there are some pre-process steps:

1. convert the gsi satbias and satbias_pc files to ioda format with GSI satbias converter

   • copy RDASApp/build/bin/satbias2ioda.x to your_work_directory

   • create satbias_converter.yaml

      input coeff file: ./rrfs.spinup.2023042720_satbias
      input err file: ./rrfs.spinup.2023042720_satbias_pc
      default predictors: &default_preds
      - constant
      - zenith_angle
      - cloud_liquid_water
      - lapse_rate_order_2
      - lapse_rate
      - cosine_of_latitude_times_orbit_node
      - sine_of_latitude
      - emissivity
      - scan_angle_order_4
      - scan_angle_order_3
      - scan_angle_order_2
      - scan_angle
      output:
      - sensor: atms_npp
        output file: satbias_atms_npp.nc4
        predictors: *default_preds
      - sensor: atms_n20
        output file: satbias_atms_n20.nc4
        predictors: *default_preds

• Create the tlaps file manually by copying the 2-4 columns of the instrument atms_npp.tlapse.txt

atms_npp 1 0.242396E+00
atms_npp 2 0.161521E+00
atms_npp 3 0.122284E+01
atms_npp 4 0.184939E+01
atms_npp 5 0.276065E+01
atms_npp 6 0.343459E+01
atms_npp 7 0.347138E+01
atms_npp 8 0.249312E+01
atms_npp 9 0.124182E+01
atms_npp 10 -0.104760E+01
atms_npp 11 -0.268708E+01
atms_npp 12 -0.534154E+01
atms_npp 13 -0.828398E+01
atms_npp 14 -0.726963E+01
atms_npp 15 -0.251210E+01
atms_npp 16 0.433186E+00
atms_npp 17 0.153876E+01
atms_npp 18 0.291233E+01
atms_npp 19 0.381283E+01
atms_npp 20 0.461646E+01
atms_npp 21 0.554495E+01
atms_npp 22 0.619940E+01

2. To convert GSI bias correction coefficient files (satbias_in and satbias_pc), run satbias2ioda.x executable, passing it a yaml configuration file:

satbias2ioda.x satbias_converter.yaml The output is: satbias_atms_npp.nc4

3. Copy these three files to Data/obs

cp satbias_atms_npp.nc4 Data/obs/. cp satbias_atms_npp.nc4 Data/obs/satbias_atms_npp_cov.nc4 cp atms_npp.tlapse.txt Data/obs/.

4. Modify the YAML file to add the bias correction

obs bias: input file: Data/obs/atms_npp.satbias.nc4 output file: Data/bc/out_atms_npp.satbias.nc variational bc: predictors:

• name: constant
• name: lapse_rate order: 2 tlapse: &atms_npp_tlapse Data/obs/atms_npp.tlapse.txt
• name: lapse_rate tlapse: *atms_npp_tlapse
• name: emissivity
• name: scan_angle order: 4
• name: scan_angle order: 3
• name: scan_angle order: 2
• name: scan_angle covariance: minimal required obs number: 20 variance range: [1.0e-6, 10.0] step size: 1.0e-4 largest analysis variance: 10000.0 prior: input file: Data/obs/atms_npp.satbias_cov.nc4

  input file: /scratch1/NCEPDEV/fv3-cam/Xiaoyan.Zhang/mpas_jedi_conus/satbias_atms_Emily/gdas.t18z.atms_npp.satbias_cov.nc4

  input file: Data/obs/atms_npp.satbias_cov_2022052619.nc4

  inflation: ratio: 1.1 ratio for small dataset: 2.0 output file: Data/bc/out_atms_npp.satbias_cov.nc

[xyzemc]

Histogram plot for O-B with/without bias correction:

[guoqing-noaa]

@xyzemc Do you have special considerations for using satbias files from a spin-up cycle?

[xyzemc]

@xyzemc Do you have special considerations for using satbias files from a spin-up cycle?

Nothing special. Just follow what RRFS is doing now to use the RRFS satbias file for a spin-ip cycle, or start from a zero satbias file.