2-degree WACCM CMIP6DECK needs to match a no-ciso user-mod

[ekluzek]

Brief summary of bug

A compset with WACCM and CMIP6DECK for 2-degree should match a WACCM CMIP6 user-mod without Carbon isotopes turned on.

General bug information

CTSM version you are using: release-clm5.0.22
Does this bug cause significantly incorrect results in the model's science? No

Configurations affected: WACCM at 2-degree with CMIP6DECK

Details of bug

The matching for CLM_USER_MODS, currently for non-WACCM cases with CMIP6DECK on will match either regular or the "nociso" user-mod case at 2-degree. This same logic needs to be added to the CMIP6WACCMDECK compsets so that nociso is used for 2-degree and the regular user-mod for anything else.

Important details of your setup / configuration so we can reproduce the bug

Here's an example test case that fails: SMS_D_Ln9.f19_g17.BWmaCO2x4cmip6.cheyenne_intel.allactive-default

Important output or errors that show the problem

ERROR: Cannot interpolate from a run without c13/c14 to a run with c13/c14

[ekluzek]

@fvitt this is the fix I think is needed...

diff --git a/cime_config/config_component.xml b/cime_config/config_component.xml
index 7025283a..4a5ababe 100644
--- a/cime_config/config_component.xml
+++ b/cime_config/config_component.xml
@@ -246,6 +246,7 @@
       <value                  compset="_CLM50%[^_]*CMIP6DECK[%_]"     >$COMP_ROOT_DIR_LND/cime_config/usermods_dirs/cmip6_deck</value>
       <value grid="l%1.9x2.5" compset="_CLM50%[^_]*CMIP6DECK[%_]"     >$COMP_ROOT_DIR_LND/cime_config/usermods_dirs/cmip6_nociso_deck</value>
       <value                  compset="_CLM50%[^_]*CMIP6WACCMDECK[%_]">$COMP_ROOT_DIR_LND/cime_config/usermods_dirs/cmip6_waccm_deck</value>
+      <value grid="l%1.9x2.5" compset="_CLM50%[^_]*CMIP6WACCMDECK[%_]">$COMP_ROOT_DIR_LND/cime_config/usermods_dirs/cmip6_waccm_nociso_deck</value>
     </values>
     <group>run_component_clm</group>
     <file>env_case.xml</file>
diff --git a/cime_config/usermods_dirs/cmip6_waccm_nociso_deck/README b/cime_config/usermods_dirs/cmip6_waccm_nociso_deck/README
new file mode 100644
index 00000000..fda15948
--- /dev/null
+++ b/cime_config/usermods_dirs/cmip6_waccm_nociso_deck/README
@@ -0,0 +1,17 @@
+WACCM DECK runs should be configured similarly to CAM DECK runs, but
+without virtual elevation classes over Antarctica. At 2-degree we
+also decided to NOT turn on Carbon isotopes.
+
+(2018-10-22) In discussions between Bill Sacks and Bill Lipscomb: We
+have decided NOT to turn on virtual Antarctica columns for any of the
+WACCM simulations (DECK or ScenarioMIP): these weren't turned on in the
+WACCM PI-Control or historical runs (either by accident or
+intentionally: I'm not sure which, and can't find emails describing this
+decision). Since this setting needs to be compatible with the REFCASE,
+it's simpler if we keep it consistent for all WACCM runs, and there
+seems to be only limited value in turning on virtual Antarctica columns
+in the remaining WACCM runs (1% CO2, 4xCO2 and ScenarioMIP) when they
+weren't on in the WACCM PI-Control or Historical runs. And Mike Mills
+says that there won't be any more sharing of initial conditions between
+CAM and WACCM CMIP6 runs, so it's fine if the two sets remain different
+in this respect.
diff --git a/cime_config/usermods_dirs/cmip6_waccm_nociso_deck/include_user_mods b/cime_config/usermods_dirs/cmip6_waccm_nociso_deck/include_user_mods
new file mode 100644
index 00000000..64c3f6f6
--- /dev/null
+++ b/cime_config/usermods_dirs/cmip6_waccm_nociso_deck/include_user_mods
@@ -0,0 +1,2 @@
+../_includes/cmip6_glaciers_cplhist
+../output_crop