fix Exner bug in CLUBB interface and change CLUBB namelist

bld/namelist_files/namelist_defaults_cam.xml

@@ -2214,7 +2214,7 @@
 <clubb_l_intr_sfc_flux_smooth phys="cam7"         > .true.  </clubb_l_intr_sfc_flux_smooth>
 <clubb_l_lmm_stepping                             > .false. </clubb_l_lmm_stepping>
 <clubb_l_lscale_plume_centered                    > .false. </clubb_l_lscale_plume_centered>
-<clubb_l_min_wp2_from_corr_wx                     > .true.  </clubb_l_min_wp2_from_corr_wx>
+<clubb_l_min_wp2_from_corr_wx                     > .false.  </clubb_l_min_wp2_from_corr_wx>
 <clubb_l_min_xp2_from_corr_wx                     > .true.  </clubb_l_min_xp2_from_corr_wx>
 <clubb_l_mono_flux_lim_rtm                        > .true.  </clubb_l_mono_flux_lim_rtm>
 <clubb_l_mono_flux_lim_spikefix                   > .true.  </clubb_l_mono_flux_lim_spikefix>

doc/ChangeLog

@@ -1,3 +1,16 @@
+One-line Summary: Fix Exner bug in CLUBB interface and change CLUBB namelist
+
+Purpose of changes (include the issue number and title text for each relevant GitHub issue):
+
+The computation of the Exner function in the CLUBB interface code currently passes an incorrect version to the PBL utilities. The PBL utilities expect the "Stull" definition of the Exner function rather than the traditional "atmospheric" Exner function.
+(Github issue 1222)
+
+The CLUBB group has recommended a namelist change to address this issue.
+(Github issue 1208)
+
+
+Expect baseline failures for all applications using CLUBB.
+
 ===============================================================
 
 Tag name: cam6_4_056

src/physics/cam/clubb_intr.F90

@@ -4700,8 +4700,8 @@ subroutine clubb_tend_cam( state,   ptend_all,   pbuf,     hdtime, &
     ! --------------------------------------------------------------------------------- !
     do i=1,ncol
       do k=1,pver
-         !use local exner since state%exner is not a proper exner
-         th(i,k) = state1%t(i,k)*inv_exner_clubb(i,k)
+         !subroutine pblind expects "Stull" definition of Exner
+         th(i,k) = state1%t(i,k)*state1%exner(i,k)
          !thv should have condensate loading to be consistent with earlier def's in this module
          thv(i,k) = th(i,k)*(1.0_r8+zvir*state1%q(i,k,ixq) - state1%q(i,k,ixcldliq))
       enddo