Hotfix/cd emis kmtx by chengdang · Pull Request #221 · JCSDA/CRTMv3

chengdang · 2025-04-02T20:47:13Z

Description

Surface emissivity Jacobian bug fix PR

Issue(s) addressed

Resolves #217

Dependencies

None

Impact

Expect changes in value of surface emissivity Jacobians
Know impacts on qc tests

Checklist

I have performed a self-review of my own code
I have made corresponding changes to the documentation
I have run the unit tests before creating the PR

Minor modification to CRTM test: including amusa test over water covered surface

chengdang · 2025-04-02T21:40:37Z

Note that commit 964e34e reverted the changes to surface types.
All unit tests passed in comparison to the results generated by dev branch (except for surface emissivity Jacobians)

BenjaminTJohnson · 2025-04-03T02:10:11Z

When I compare to develop, I get

         91 - test_k_matrix_Simple_cris-fsr_n21 (Failed)
         92 - test_k_matrix_Simple_v.abi_g18 (Failed)
         93 - test_k_matrix_Simple_atms_npp (Failed)
         94 - test_k_matrix_Simple_cris399_npp (Failed)
         95 - test_k_matrix_Simple_v.abi_gr (Failed)
         96 - test_k_matrix_Simple_abi_g18 (Failed)
         97 - test_k_matrix_Simple_modis_aqua (Failed)
        104 - test_k_matrix_Zeeman_ssmis_f20 (Failed)
        105 - test_k_matrix_Zeeman_ssmis_f19 (Failed)
        106 - test_k_matrix_Zeeman_ssmis_f18 (Failed)
        107 - test_k_matrix_Zeeman_ssmis_f17 (Failed)
        108 - test_k_matrix_Zeeman_ssmis_f16 (Failed)
        109 - test_k_matrix_ChannelSubset_iasi_metop-b (Failed)
        110 - test_k_matrix_ClearSky_atms_n21 (Failed)
        111 - test_k_matrix_ClearSky_cris-fsr_n21 (Failed)
        113 - test_k_matrix_ClearSky_atms_npp (Failed)
        114 - test_k_matrix_ClearSky_cris399_npp (Failed)
        116 - test_k_matrix_ClearSky_abi_g18 (Failed)
        117 - test_k_matrix_ClearSky_modis_aqua (Failed)
        118 - test_k_matrix_ScatteringSwitch_atms_n21 (Failed)
        119 - test_k_matrix_ScatteringSwitch_cris-fsr_n21 (Failed)
        121 - test_k_matrix_ScatteringSwitch_atms_npp (Failed)
        122 - test_k_matrix_ScatteringSwitch_cris399_npp (Failed)
        124 - test_k_matrix_ScatteringSwitch_abi_g18 (Failed)
        125 - test_k_matrix_ScatteringSwitch_modis_aqua (Failed)
        134 - test_k_matrix_SOI_atms_n21 (Failed)
        135 - test_k_matrix_SOI_cris-fsr_n21 (Failed)
        137 - test_k_matrix_SOI_atms_npp (Failed)
        138 - test_k_matrix_SOI_cris399_npp (Failed)
        140 - test_k_matrix_SOI_abi_g18 (Failed)
        141 - test_k_matrix_SOI_modis_aqua (Failed)
        142 - test_k_matrix_SSU_ssu_n06 (Failed)
        143 - test_k_matrix_SSU_ssu_n14 (Failed)
        152 - test_adjoint_Simple_atms_n21 (Failed)
        155 - test_adjoint_Simple_atms_npp (Failed)
        160 - test_adjoint_ClearSky_atms_n21 (Failed)
        161 - test_adjoint_ClearSky_cris-fsr_n21 (Failed)
        163 - test_adjoint_ClearSky_atms_npp (Failed)
        164 - test_adjoint_ClearSky_cris399_npp (Failed)
        166 - test_adjoint_ClearSky_abi_g18 (Failed)
        167 - test_adjoint_ClearSky_modis_aqua (Failed)

[s4-submit] /data/users/bjohnson/CRTM/CRTMv3/build> ctest --rerun-failed --output-on-failure | grep "FAILURE"
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Zeeman(FAILURE) : RTSolution_K results are different!
 test_Zeeman(FAILURE) : RTSolution_K results are different!
 test_Zeeman(FAILURE) : RTSolution_K results are different!
 test_Zeeman(FAILURE) : RTSolution_K results are different!
 test_Zeeman(FAILURE) : RTSolution_K results are different!
 Example6_ChannelSubset(FAILURE) : RTSolution_K results are different!
 test_ClearSky(FAILURE) : RTSolution_K results are different!
 test_ClearSky(FAILURE) : RTSolution_K results are different!
 test_ClearSky(FAILURE) : RTSolution_K results are different!
 test_ClearSky(FAILURE) : RTSolution_K results are different!
 test_ClearSky(FAILURE) : RTSolution_K results are different!
 test_ClearSky(FAILURE) : RTSolution_K results are different!
 test_ScatteringSwitch(FAILURE) : RTSolution_K results are different!
 test_ScatteringSwitch(FAILURE) : RTSolution_K results are different!
 test_ScatteringSwitch(FAILURE) : RTSolution_K results are different!
 test_ScatteringSwitch(FAILURE) : RTSolution_K results are different!
 test_ScatteringSwitch(FAILURE) : RTSolution_K results are different!
 test_ScatteringSwitch(FAILURE) : RTSolution_K results are different!
 test_SOI(FAILURE) : RTSolution_K results are different!
 test_SOI(FAILURE) : RTSolution_K results are different!
 test_SOI(FAILURE) : RTSolution_K results are different!
 test_SOI(FAILURE) : RTSolution_K results are different!
 test_SOI(FAILURE) : RTSolution_K results are different!
 test_SOI(FAILURE) : RTSolution_K results are different!
 test_SSU(FAILURE) : RTSolution_K results are different!
 test_SSU(FAILURE) : RTSolution_K results are different!
 test_Simple(FAILURE) : RTSolution_AD results are different!
 test_Simple(FAILURE) : RTSolution_AD results are different!
 test_ClearSky(FAILURE) : RTSolution_AD results are different!
 test_ClearSky(FAILURE) : RTSolution_AD results are different!
 test_ClearSky(FAILURE) : RTSolution_AD results are different!
 test_ClearSky(FAILURE) : RTSolution_AD results are different!
 test_ClearSky(FAILURE) : RTSolution_AD results are different!
 test_ClearSky(FAILURE) : RTSolution_AD results are different!

chengdang · 2025-04-03T04:07:59Z

Good catch. Yes. I forgot to add that a lot k-matrix/ad tests should fail with the updated surface emissivity Jacobians.

Mostly for variable: RTSolution_K%Surface_Emissivity and RTSolution_AD%Surface_Emissivity

Here is a list of failed test from my end:

76% tests passed, 46 tests failed out of 191

Label Time Summary:
CRTM_Tests    = 174.35 sec*proc (191 tests)

Total Test time (real) = 177.72 sec

The following tests FAILED:
	 91 - test_k_matrix_Simple_atms_n21 (Failed)
	 92 - test_k_matrix_Simple_cris-fsr_n21 (Failed)
	 93 - test_k_matrix_Simple_v.abi_g18 (Failed)
	 94 - test_k_matrix_Simple_atms_npp (Failed)
	 95 - test_k_matrix_Simple_cris399_npp (Failed)
	 96 - test_k_matrix_Simple_v.abi_gr (Failed)
	 97 - test_k_matrix_Simple_abi_g18 (Failed)
	 98 - test_k_matrix_Simple_modis_aqua (Failed)
	106 - test_k_matrix_Zeeman_ssmis_f20 (Failed)
	107 - test_k_matrix_Zeeman_ssmis_f19 (Failed)
	108 - test_k_matrix_Zeeman_ssmis_f18 (Failed)
	109 - test_k_matrix_Zeeman_ssmis_f17 (Failed)
	110 - test_k_matrix_Zeeman_ssmis_f16 (Failed)
	111 - test_k_matrix_ChannelSubset_iasi_metop-b (Failed)
	112 - test_k_matrix_ClearSky_atms_n21 (Failed)
	113 - test_k_matrix_ClearSky_cris-fsr_n21 (Failed)
	115 - test_k_matrix_ClearSky_atms_npp (Failed)
	116 - test_k_matrix_ClearSky_cris399_npp (Failed)
	118 - test_k_matrix_ClearSky_abi_g18 (Failed)
	119 - test_k_matrix_ClearSky_modis_aqua (Failed)
	120 - test_k_matrix_ScatteringSwitch_atms_n21 (Failed)
	121 - test_k_matrix_ScatteringSwitch_cris-fsr_n21 (Failed)
	123 - test_k_matrix_ScatteringSwitch_atms_npp (Failed)
	124 - test_k_matrix_ScatteringSwitch_cris399_npp (Failed)
	126 - test_k_matrix_ScatteringSwitch_abi_g18 (Failed)
	127 - test_k_matrix_ScatteringSwitch_modis_aqua (Failed)
	136 - test_k_matrix_SOI_atms_n21 (Failed)
	137 - test_k_matrix_SOI_cris-fsr_n21 (Failed)
	138 - test_k_matrix_SOI_v.abi_g18 (Failed)
	139 - test_k_matrix_SOI_atms_npp (Failed)
	140 - test_k_matrix_SOI_cris399_npp (Failed)
	141 - test_k_matrix_SOI_v.abi_gr (Failed)
	142 - test_k_matrix_SOI_abi_g18 (Failed)
	143 - test_k_matrix_SOI_modis_aqua (Failed)
	144 - test_k_matrix_SSU_ssu_n06 (Failed)
	145 - test_k_matrix_SSU_ssu_n14 (Failed)
	154 - test_adjoint_Simple_atms_n21 (Failed)
	156 - test_adjoint_Simple_v.abi_g18 (Failed)
	157 - test_adjoint_Simple_atms_npp (Failed)
	159 - test_adjoint_Simple_v.abi_gr (Failed)
	163 - test_adjoint_ClearSky_atms_n21 (Failed)
	164 - test_adjoint_ClearSky_cris-fsr_n21 (Failed)
	166 - test_adjoint_ClearSky_atms_npp (Failed)
	167 - test_adjoint_ClearSky_cris399_npp (Failed)
	169 - test_adjoint_ClearSky_abi_g18 (Failed)
	170 - test_adjoint_ClearSky_modis_aqua (Failed)

chengdang · 2025-04-03T04:21:01Z

Could you add which tests are reporting these failures?

 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!
 test_Simple(FAILURE) : Atmosphere_K Jacobians are different!

All my tests failed due to RTSolution_K results are different!

BenjaminTJohnson · 2025-04-03T04:22:08Z

         90 - test_k_matrix_Simple_atms_n21 (Failed)
         91 - test_k_matrix_Simple_cris-fsr_n21 (Failed)
         92 - test_k_matrix_Simple_v.abi_g18 (Failed)
         93 - test_k_matrix_Simple_atms_npp (Failed)
         94 - test_k_matrix_Simple_cris399_npp (Failed)
         95 - test_k_matrix_Simple_v.abi_gr (Failed)
         96 - test_k_matrix_Simple_abi_g18 (Failed)
         97 - test_k_matrix_Simple_modis_aqua (Failed)
        104 - test_k_matrix_Zeeman_ssmis_f20 (Failed)
        105 - test_k_matrix_Zeeman_ssmis_f19 (Failed)
        106 - test_k_matrix_Zeeman_ssmis_f18 (Failed)
        107 - test_k_matrix_Zeeman_ssmis_f17 (Failed)
        108 - test_k_matrix_Zeeman_ssmis_f16 (Failed)
        109 - test_k_matrix_ChannelSubset_iasi_metop-b (Failed)
        110 - test_k_matrix_ClearSky_atms_n21 (Failed)
        111 - test_k_matrix_ClearSky_cris-fsr_n21 (Failed)
        113 - test_k_matrix_ClearSky_atms_npp (Failed)
        114 - test_k_matrix_ClearSky_cris399_npp (Failed)
        116 - test_k_matrix_ClearSky_abi_g18 (Failed)
        117 - test_k_matrix_ClearSky_modis_aqua (Failed)
        118 - test_k_matrix_ScatteringSwitch_atms_n21 (Failed)
        119 - test_k_matrix_ScatteringSwitch_cris-fsr_n21 (Failed)
        121 - test_k_matrix_ScatteringSwitch_atms_npp (Failed)
        122 - test_k_matrix_ScatteringSwitch_cris399_npp (Failed)
        124 - test_k_matrix_ScatteringSwitch_abi_g18 (Failed)
        125 - test_k_matrix_ScatteringSwitch_modis_aqua (Failed)
        134 - test_k_matrix_SOI_atms_n21 (Failed)
        135 - test_k_matrix_SOI_cris-fsr_n21 (Failed)
        137 - test_k_matrix_SOI_atms_npp (Failed)
        138 - test_k_matrix_SOI_cris399_npp (Failed)
        140 - test_k_matrix_SOI_abi_g18 (Failed)
        141 - test_k_matrix_SOI_modis_aqua (Failed)
        142 - test_k_matrix_SSU_ssu_n06 (Failed)
        143 - test_k_matrix_SSU_ssu_n14 (Failed)
        152 - test_adjoint_Simple_atms_n21 (Failed)
        155 - test_adjoint_Simple_atms_npp (Failed)
        160 - test_adjoint_ClearSky_atms_n21 (Failed)
        161 - test_adjoint_ClearSky_cris-fsr_n21 (Failed)
        163 - test_adjoint_ClearSky_atms_npp (Failed)
        164 - test_adjoint_ClearSky_cris399_npp (Failed)
        166 - test_adjoint_ClearSky_abi_g18 (Failed)
        167 - test_adjoint_ClearSky_modis_aqua (Failed)

BenjaminTJohnson · 2025-04-03T14:03:09Z

I also wonder why some of the _AD tests pass, some fail.

chengdang · 2025-04-03T15:07:45Z

I also wonder why some of the _AD tests pass, some fail.

It is also due to surface emissivity bug fix in function Assign_Common_Output_AD. This is called by AD calculations well.

Most AD tests from my end failed, tests # 154-170. It seems our total number of tests are different (likely due to the added amsua_n19 from this branch, which brings a total of 191 tests from this branch). But some did pass, the same applies to K_matrix tests. Let me dig more

I don't see any Atmosphere_K Jacobians failures though

chengdang · 2025-04-03T18:56:02Z

This is a list of k_matrix/AD tests that passed due to different reasons:

Tests that passed for certain sensors.

a). The following tests passed due to Azimuth_Fourier_Loop setup discussed here #222 (comment)

  Test #114: test_k_matrix_ClearSky_v.abi_g18 
  Test #117: test_k_matrix_ClearSky_v.abi_gr
  Test #122: test_k_matrix_ScatteringSwitch_v.abi_g18 
  Test #125: test_k_matrix_ScatteringSwitch_v.abi_gr
  Test #165: test_adjoint_ClearSky_v.abi_g18 
  Test #168: test_adjoint_ClearSky_v.abi_gr

b). The following tests passed because RTSolution_AD%Surface_Emissivity = SfcOptics_AD%Emissivity(SfcOptics_AD%Index_Sat_Ang,1) = 0, which is calculated by function CRTM_ADA_AD (emissivity_AD(1:nZ) = s_rad_up_AD(1:nZ) * RTV%Planck_Surface), in which the computed s_rad_up_AD(1:nZ) = 0.

  Test #155: test_adjoint_Simple_cris-fsr_n21 
  Test #158: test_adjoint_Simple_cris399_npp
  Test #160: test_adjoint_Simple_abi_g18
  Test #161: test_adjoint_Simple_modis_aqua

c). The following tests are newly introduced by this branch, therefore passed

 Test #99: test_k_matrix_Simple_amsua_n19
 Test #162: test_adjoint_Simple_amsua_n19

By design, K_matrix AOD tests do not check RTSolution_K.

  Test #100: test_k_matrix_AOD_cris-fsr_n21
  Test #101: test_k_matrix_AOD_v.abi_g18
  Test #102: test_k_matrix_AOD_cris399_npp
  Test #103: test_k_matrix_AOD_v.abi_gr
  Test #104: test_k_matrix_AOD_abi_g18
  Test #105: test_k_matrix_AOD_airs_aqua

With user emissivity, we likely don't call the updated codes (CONFIRMED).

  Test #128: test_k_matrix_User_Emissivity_atms_n21
  Test #129: test_k_matrix_User_Emissivity_cris-fsr_n21
  Test #130: test_k_matrix_User_Emissivity_v.abi_g18
  Test #131: test_k_matrix_User_Emissivity_atms_npp
  Test #132: test_k_matrix_User_Emissivity_cris399_npp
  Test #133: test_k_matrix_User_Emissivity_v.abi_gr
  Test #134: test_k_matrix_User_Emissivity_abi_g18
  Test #135: test_k_matrix_User_Emissivity_modis_aqua

By design, vertical coordinates tests do not check RTSoultion_K

  Test #146: test_k_matrix_VerticalCoordinates_atms_n21
  Test #147: test_k_matrix_VerticalCoordinates_cris-fsr_n21
  Test #148: test_k_matrix_VerticalCoordinates_v.abi_g18
  Test #149: test_k_matrix_VerticalCoordinates_atms_npp
  Test #150: test_k_matrix_VerticalCoordinates_cris399_npp
  Test #151: test_k_matrix_VerticalCoordinates_v.abi_gr
  Test #152: test_k_matrix_VerticalCoordinates_abi_g18
  Test #153: test_k_matrix_VerticalCoordinates_modis_aqua

chengdang · 2025-04-07T21:19:28Z

@BenjaminTJohnson Please see the updated notes on why certain tests failed/passed.
Category 1.a requires an additional bug fix, as proposed here #222 (comment)

liujake · 2025-05-06T20:40:31Z

Is there a timeline for fixing ctests and merging this PR into the develop?

chengdang added 3 commits March 25, 2025 11:30

Cloudy sky bug fix

734fc83

test for v2.4 implementation

137f7f1

Bug fix for surface emissivity jacobian

b6df808

Minor modification to CRTM test: including amusa test over water covered surface

chengdang added the needs review Asking others to review - often used for pull requests label Apr 2, 2025

chengdang requested a review from BenjaminTJohnson April 2, 2025 20:47

chengdang self-assigned this Apr 2, 2025

chengdang added 2 commits April 2, 2025 14:52

Merge branch 'develop' into hotfix/cd_emis_kmtx

251a7b9

Revert changes to Load_Sfc_Data.inc

964e34e

chengdang mentioned this pull request Apr 4, 2025

[bug] K_matrix calculation over Azimuth_Fourier_Loop #222

Closed

Fix code format for CRTM_K_Matrix_Module.f90

677afc8

Merge branch 'develop' into hotfix/cd_emis_kmtx

f2f6d52

BenjaminTJohnson approved these changes Jun 24, 2025

View reviewed changes

BenjaminTJohnson merged commit 8abded4 into develop Jun 24, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Hotfix/cd emis kmtx#221

Hotfix/cd emis kmtx#221
BenjaminTJohnson merged 7 commits into
developfrom
hotfix/cd_emis_kmtx

chengdang commented Apr 2, 2025 •

edited

Loading

Uh oh!

chengdang commented Apr 2, 2025 •

edited

Loading

Uh oh!

BenjaminTJohnson commented Apr 3, 2025

Uh oh!

chengdang commented Apr 3, 2025 •

edited

Loading

Uh oh!

chengdang commented Apr 3, 2025 •

edited

Loading

Uh oh!

BenjaminTJohnson commented Apr 3, 2025

Uh oh!

BenjaminTJohnson commented Apr 3, 2025

Uh oh!

chengdang commented Apr 3, 2025 •

edited

Loading

Uh oh!

chengdang commented Apr 3, 2025 •

edited

Loading

Uh oh!

chengdang commented Apr 7, 2025

Uh oh!

liujake commented May 6, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

chengdang commented Apr 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Issue(s) addressed

Dependencies

Impact

Checklist

Uh oh!

chengdang commented Apr 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

BenjaminTJohnson commented Apr 3, 2025

Uh oh!

chengdang commented Apr 3, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chengdang commented Apr 3, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

BenjaminTJohnson commented Apr 3, 2025

Uh oh!

BenjaminTJohnson commented Apr 3, 2025

Uh oh!

chengdang commented Apr 3, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chengdang commented Apr 3, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chengdang commented Apr 7, 2025

Uh oh!

liujake commented May 6, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

chengdang commented Apr 2, 2025 •

edited

Loading

chengdang commented Apr 2, 2025 •

edited

Loading

chengdang commented Apr 3, 2025 •

edited

Loading

chengdang commented Apr 3, 2025 •

edited

Loading

chengdang commented Apr 3, 2025 •

edited

Loading

chengdang commented Apr 3, 2025 •

edited

Loading