CrIS viirs bug fix

[wx20jjung]

DUE DATE for merger of this PR into develop is 4/8/2024 (six weeks after PR creation).

Description
The new cloud and aerosol detection software (CADS) requires the corresponding imager CRTM coefficient files. There are satellite identifier mis-matches for CrIS-VIIRS instruments on NOAA-20 and NOAA-21. The VIIRS files are named J1 and J2 for NOAA-20 and -21 respectively. Eventually the CRTM group will develop the correct named files. When this happens read_cris should transition to the new file names without incident. The current code will fail.

The code changes in read_cris will transition between j2 and n21 without incident. This was tested on S4 and Hera. The same logic was put in for J1 and n20 but can't be tested. This PR will fix #706.

Fixes #706

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)

Ctests were conducted on Hera and Orion.

Hera stats:
Start 1: [=[global_4denvar]=]
Start 2: [=[rtma]=]
Start 3: [=[rrfs_3denvar_glbens]=]
Start 4: [=[netcdf_fv3_regional]=]
Start 5: [=[hafs_4denvar_glbens]=]
Start 6: [=[hafs_3denvar_hybens]=]
Start 7: [=[global_enkf]=]
1/7 Test #3: [=[rrfs_3denvar_glbens]=] ........ Passed 1210.74 sec
2/7 Test #4: [=[netcdf_fv3_regional]=] ........ Passed 1510.34 sec
3/7 Test #7: [=[global_enkf]=] ................ Passed 1689.43 sec
4/7 Test #6: [=[hafs_3denvar_hybens]=] ........ Passed 1826.83 sec
5/7 Test #2: [=[rtma]=] ....................... Passed 1996.78 sec
6/7 Test #5: [=[hafs_4denvar_glbens]=] ........ Passed 2187.12 sec
7/7 Test #1: [=[global_4denvar]=] ............. Passed 2213.32 sec

100% tests passed, 0 tests failed out of 7

Total Test time (real) = 2213.33 sec

Orion stats:
Start 1: [=[global_4denvar]=]
Start 2: [=[rtma]=]
Start 3: [=[rrfs_3denvar_glbens]=]
Start 4: [=[netcdf_fv3_regional]=]
Start 5: [=[hafs_4denvar_glbens]=]
Start 6: [=[hafs_3denvar_hybens]=]
Start 7: [=[global_enkf]=]
1/7 Test #4: [=[netcdf_fv3_regional]=] ........ Passed 666.36 sec
2/7 Test #7: [=[global_enkf]=] ................ Passed 683.18 sec
3/7 Test #3: [=[rrfs_3denvar_glbens]=] ........ Passed 730.58 sec
4/7 Test #2: [=[rtma]=] ....................... Passed 1088.72 sec
5/7 Test #6: [=[hafs_3denvar_hybens]=] ........ Passed 1399.00 sec
6/7 Test #1: [=[global_4denvar]=] ............. Passed 1623.90 sec
7/7 Test #5: [=[hafs_4denvar_glbens]=] ........***Failed 1639.01 sec

86% tests passed, 1 tests failed out of 7

Total Test time (real) = 1639.03 sec

The failure on orion is a time threshold test:
The runtime for hafs_4denvar_glbens_loproc_updat is 389.841067 seconds. This has exceeded maximum allowable threshold time of 389.025272 seconds,
resulting in Failure time-thresh of the regression test.

[RussTreadon-NOAA]

Thank you @ADCollard for your review and approval.

@wx20jjung ,thank you for running ctests on Hera and Orion. I'll run the tests on WCOSS2. We need one additional peer reviewer. Let me know who you want and we can and him/her to the reviewer list.

[wx20jjung]

I asked Dave Huber. He will do the 2nd review.
Jim

[RussTreadon-NOAA]

I asked Dave Huber. He will do the 2nd review. Jim

Thanks. I added @DavidHuber-NOAA as a reviewer.

[DavidHuber-NOAA]

Looks good, thanks for the catch @wx20jjung!

[RussTreadon-NOAA]

WCOSS2 ctests
wx20jjung:CrIS-VIIRS_bug at 76f22b1 was installed on Dogwood along with develop at 7c4a571. ctests were run with the following results

Test project /lfs/h2/emc/da/noscrub/russ.treadon/git/gsi/pr708/build
    Start 1: global_4denvar
    Start 2: rtma
    Start 3: rrfs_3denvar_glbens
    Start 4: netcdf_fv3_regional
    Start 5: hafs_4denvar_glbens
    Start 6: hafs_3denvar_hybens
    Start 7: global_enkf
1/7 Test #4: netcdf_fv3_regional ..............   Passed  483.07 sec
2/7 Test #3: rrfs_3denvar_glbens ..............   Passed  485.66 sec
3/7 Test #7: global_enkf ......................   Passed  611.07 sec
4/7 Test #2: rtma .............................   Passed  969.13 sec
5/7 Test #6: hafs_3denvar_hybens ..............   Passed  1093.05 sec
6/7 Test #5: hafs_4denvar_glbens ..............   Passed  1342.07 sec
7/7 Test #1: global_4denvar ...................   Passed  1442.62 sec

100% tests passed, 0 tests failed out of 7

Total Test time (real) = 1442.68 sec

[RussTreadon-NOAA]

@wx20jjung : Quick comment: All tests pass on Dogwood, Hera, and Orion with (this PR) or without (develop) the bug fix. Since our ctests yield Passed this suggests that our current suite of ctests do not exercise the code in question. We should update our ctests to exercise the corrected code. Doing so is not a requirement for this PR but it's something we don't want to forget. ctests should not be static. They need periodic updates, too.

[RussTreadon-NOAA]

Approve.

[RussTreadon-NOAA]

@wx20jjung : Quick comment: All tests pass on Dogwood, Hera, and Orion with (this PR) or without (develop) the bug fix. Since our ctests yield Passed this suggests that our current suite of ctests do not exercise the code in question. We should update our ctests to exercise the corrected code. Doing so is not a requirement for this PR but it's something we don't want to forget. ctests should not be static. They need periodic updates, too.

OK. I think I see what's going on. Looking at the issue this PR future-proofs the code. It's not fixing a current bug. It's looking ahead to when the current code will fail as is. Is this correct?

[RussTreadon-NOAA]

@wx20jjung : Quick comment: All tests pass on Dogwood, Hera, and Orion with (this PR) or without (develop) the bug fix. Since our ctests yield Passed this suggests that our current suite of ctests do not exercise the code in question. We should update our ctests to exercise the corrected code. Doing so is not a requirement for this PR but it's something we don't want to forget. ctests should not be static. They need periodic updates, too.

OK. I think I see what's going on. Looking at the issue this PR future-proofs the code. It's not fixing a current bug. It's looking ahead to when the current code will fail as is. Is this correct?

@wx20jjung , given your confirmation that my understanding is correct we can schedule this PR for merger into develop.

[wx20jjung]

Russ,

You are correct. This change circumvents a potential problem when the "consistent" crtm file names are available. When I tried this on S4, the gsi crashed, not just read_cris. Similar results on hera. I don't want this possibility to exist in operations.

[RussTreadon-NOAA]

Thanks, @wx20jjung . I'll work with the Handling Review team to schedule merger of this PR into develop. I appreciate your being proactive, especially with regards to operations.