Changes required for oops#3147

[shlyaeva]

https://github.com/JCSDA-internal/oops/pull/3147 cleans up some legacy ensemble classes in oops, and incidentally also improves the runtime for LETKF/GETKF. It is unclear at this point when it will be merged, as someone needs to figure out why pyiri-jedi is failing with these changes.

When the oops PR is merged, marine analysis recentering will start outputting ensemble increments indexed 1..n (instead 0..n-1) and this would need to be changed accordingly:

GDASApp/parm/marine/marine_ecen_config.yaml.j2

Lines 74 to 75 in f2600d5

	# ocean increments are saved as 0..n-1
	- ['{{ DATA }}/ocn.recenter.ens.{{ (imem-1) | string }}.{{ WINDOW_END | to_isotime }}.PT0S.nc', '{{ COM_OCEAN_ANALYSIS_TMPL | replace_tmpl(tmpl_dict) }}/{{ APREFIX_ENS }}mom6_increment.i006.nc']

[shlyaeva]

https://github.com/JCSDA-internal/oops/pull/3147 was merged, and we will need this change in the gdasapp when we update hashes this week. @RussTreadon-NOAA FYI. I will try to test the expected change tomorrow pm.

[RussTreadon-NOAA]

@shlyaeva : Thank you for the heads up. Tonight's stable-nightly run should pull the updated oops hash into feature/stable-nightly.

[RussTreadon-NOAA]

@shlyaeva : The 20260327 run of stable nightly updated sorc/oops to hash 2dfeb62. As you explained,

316 - test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_marineanlecen_202103250000 (Failed)

The following change was made to the working copy of parm/marine/marine_ecen_config.yaml.j2

+++ b/parm/marine/marine_ecen_config.yaml.j2
@@ -71,8 +71,8 @@ data_out:
                           '${HH}': current_cycle | strftime('%H'),
                           '${MEMDIR}': 'mem%03d' | format(imem) }) %}
 
-    # ocean increments are saved as 0..n-1
-    - ['{{ DATA }}/ocn.recenter.ens.{{ (imem-1) | string }}.{{ WINDOW_END | to_isotime }}.PT0S.nc', '{{ COM_OCEAN_ANALYSIS_TMPL | replace_tmpl(tmpl_dict) }}/{{ APREFIX_E
NS }}mom6_increment.i006.nc']
+    # ocean increments are saved as 1..n
+    - ['{{ DATA }}/ocn.recenter.ens.{{ (imem) | string }}.{{ WINDOW_END | to_isotime }}.PT0S.nc', '{{ COM_OCEAN_ANALYSIS_TMPL | replace_tmpl(tmpl_dict) }}/{{ APREFIX_ENS
 }}mom6_increment.i006.nc']
 
     # ice restarts are saved as 1..n
     - ['{{ DATA }}/ens/cice_model.res.{{ imem | string }}.nc', '{{ COM_ICE_ANALYSIS_TMPL | replace_tmpl(tmpl_dict) }}/{{ cice_rst_date }}.analysis.cice_model.res.nc']

The test_gdasapp_C48mx500_hybAOWCDA suite of tests was rerun. All tests Passed

Test project /scratch3/NCEPDEV/da/Russ.Treadon/CI/hera/GDASApp/stable/20260327/global-workflow/sorc/gdas.cd/build
      Start 305: test_gdasapp_C48mx500_hybAOWCDA
 1/14 Test #305: test_gdasapp_C48mx500_hybAOWCDA ........................................   Passed    6.62 sec
      Start 306: test_gdasapp_C48mx500_hybAOWCDA_gdas_stage_ic_202103241800
      Start 308: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_stage_ic_202103241800
 2/14 Test #306: test_gdasapp_C48mx500_hybAOWCDA_gdas_stage_ic_202103241800 .............   Passed   17.81 sec
      Start 307: test_gdasapp_C48mx500_hybAOWCDA_gdas_fcst_202103241800
 3/14 Test #308: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_stage_ic_202103241800 .........   Passed   31.47 sec
      Start 309: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_fcst_202103241800
 4/14 Test #307: test_gdasapp_C48mx500_hybAOWCDA_gdas_fcst_202103241800 .................   Passed  303.11 sec
      Start 310: test_gdasapp_C48mx500_hybAOWCDA_gdas_prepoceanobs_202103250000
 5/14 Test #309: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_fcst_202103241800 .............   Passed  319.74 sec
      Start 311: test_gdasapp_C48mx500_hybAOWCDA_gdas_marinebmatinit_202103250000
 6/14 Test #310: test_gdasapp_C48mx500_hybAOWCDA_gdas_prepoceanobs_202103250000 .........   Passed   56.76 sec
 7/14 Test #311: test_gdasapp_C48mx500_hybAOWCDA_gdas_marinebmatinit_202103250000 .......   Passed   56.59 sec
      Start 316: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_marineanlecen_202103250000
      Start 312: test_gdasapp_C48mx500_hybAOWCDA_gdas_marinebmat_202103250000
 8/14 Test #316: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_marineanlecen_202103250000 ....   Passed   54.33 sec
 9/14 Test #312: test_gdasapp_C48mx500_hybAOWCDA_gdas_marinebmat_202103250000 ...........   Passed   67.45 sec
      Start 314: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlinit_202103250000
      Start 313: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_marineanlletkf_202103250000
10/14 Test #314: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlinit_202103250000 ........   Passed   55.10 sec
      Start 315: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlvar_202103250000
11/14 Test #313: test_gdasapp_C48mx500_hybAOWCDA_enkfgdas_marineanlletkf_202103250000 ...   Passed   81.84 sec
12/14 Test #315: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlvar_202103250000 .........   Passed   57.38 sec
      Start 317: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlchkpt_202103250000
13/14 Test #317: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlchkpt_202103250000 .......   Passed   60.87 sec
      Start 318: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlfinal_202103250000
14/14 Test #318: test_gdasapp_C48mx500_hybAOWCDA_gdas_marineanlfinal_202103250000 .......   Passed   74.02 sec

100% tests passed, 0 tests failed out of 14

Label Time Summary:
gdasapp    = 1243.10 sec*proc (14 tests)
script     = 1243.10 sec*proc (14 tests)

Total Test time (real) = 729.63 sec

The modified parm/marine/marine_ecen_config.yaml.j2 along with the updated JEDI hashes were committed to feature/stable-nightly at b2be058.

Branch feature/stable-nightly will be merged into GDASApp develop the morning of 20260330.

[shlyaeva]

@RussTreadon-NOAA thank you so much, you are, as usual, way ahead of me!

[RussTreadon-NOAA]

@shlyaeva No, you are ahead of me. You alerted me way in advance of a pending failure and provided the fix. Thank you for making my Friday morning much smoother than it could have been. Great teamwork!

[shlyaeva]

Yes, great teamwork! Thank you @RussTreadon-NOAA!