Add Radiance (clear) and Brightness Temperature (clear) to RTS in K_matrix modules

[chengdang]

Hello @BenjaminTJohnson, I made some adjustment to the k_matrix modules that the RTS computed with it will also output Radiance/BT for clear-sky. With my test cases, the code changes in draft PR #135 works.

The major changes are:

1. adding the following code block:

             !** output Tb_clear in the case of n_clouds = 0  (note this is NOT aerosol cleared)
             IF (Atm%n_Clouds == 0 .OR. CloudCover%Total_Cloud_Cover < MIN_COVERAGE_THRESHOLD) THEN
                RTSolution(ln,m)%Tb_clear = RTSolution(ln,m)%Brightness_Temperature
                RTSolution(ln,m)%R_clear  = RTSolution(ln,m)%Radiance
             END IF

2. Comment out the IF statement:

           ! IF ( SpcCoeff_IsInfraredSensor( SC(SensorIndex) ) .OR. &
           !     SpcCoeff_IsMicrowaveSensor( SC(SensorIndex) ) ) THEN

With this IF statement, I wonder if all the visible jacobians with fractional cloud coverage are wrong.

[chengdang]

In PR #135, I also modified the unit test test_simple for k_matrix a bit to include the comparisons of RTS computed with Forward and K_matrix modules:

  ! Compare the RT structures
  ! --------------------------
  IF ( ALL(CRTM_RTSolution_Compare(RTSolution, RTSolution_forward)) ) THEN
    Message = 'RTSolution results are the same! (Forward vs. K_Matrix)'
    CALL Display_Message( PROGRAM_NAME, Message, INFORMATION )
  ELSE
    Message = 'RTSolution results are different!'
    CALL Display_Message( PROGRAM_NAME, Message, INFORMATION )
    STOP 1
  END IF

So far all the following ctests built upon this passed:

Test project /Users/dangch/Documents/CRTM/CRTM_dev/crtm_code_review/CRTMv3/build
    Start 73: test_k_matrix_Simple_atms_n21
1/8 Test #73: test_k_matrix_Simple_atms_n21 .......   Passed    0.11 sec
    Start 74: test_k_matrix_Simple_cris-fsr_n21
2/8 Test #74: test_k_matrix_Simple_cris-fsr_n21 ...   Passed    5.39 sec
    Start 75: test_k_matrix_Simple_v.abi_g18
3/8 Test #75: test_k_matrix_Simple_v.abi_g18 ......   Passed    0.09 sec
    Start 76: test_k_matrix_Simple_atms_npp
4/8 Test #76: test_k_matrix_Simple_atms_npp .......   Passed    0.10 sec
    Start 77: test_k_matrix_Simple_cris399_npp
5/8 Test #77: test_k_matrix_Simple_cris399_npp ....   Passed    1.06 sec
    Start 78: test_k_matrix_Simple_v.abi_gr
6/8 Test #78: test_k_matrix_Simple_v.abi_gr .......   Passed    0.09 sec
    Start 79: test_k_matrix_Simple_abi_g18
7/8 Test #79: test_k_matrix_Simple_abi_g18 ........   Passed    0.12 sec
    Start 80: test_k_matrix_Simple_modis_aqua
8/8 Test #80: test_k_matrix_Simple_modis_aqua .....   Passed    0.13 sec

These tests passed because RTsolution%R_clear and RTsolution%TB_clear are assigned in K_matrix moduls now, and because the TOA reflectance are conveniently all zeros. Otherwise, the TOA reflectance would be different because there are no cloud-fraction adjustment to reflectance, in CRTM TL, AD, and K_matrix modules.

[BenjaminTJohnson]

@chengdang is there anything else that needs to be done on this issue? as far as I can tell, it's completed.

[chengdang]

Yes it is completed. I'll close it for now. Thank you!