202411 Release of SHiELD_physics

GFS_layer/GFS_driver.F90

@@ -206,7 +206,7 @@ subroutine GFS_initialize (Model, Statein, Stateout, Sfcprop,    &
     si = (Init_parm%ak + Init_parm%bk * p_ref - Init_parm%ak(Model%levr+1)) &
              / (p_ref - Init_parm%ak(Model%levr+1))
     call rad_initialize (si, Model%levr, Model%ictm, Model%isol, &
-           Model%ico2, Model%iaer, Model%ialb,                   &
+           Model%ico2, Model%fco2_scaling , Model%iaer, Model%ialb, &
            Model%disable_radiation_quasi_sea_ice, Model%iems,    &
            Model%ntcw, Model%num_p2d,  Model%num_p3d,            &
            Model%npdf3d, Model%ntoz,                             &
@@ -266,9 +266,10 @@ subroutine GFS_initialize (Model, Statein, Stateout, Sfcprop,    &
 !-----------------------------------------------------------------------
 
     if (Model%do_cosp) then
-       do nb = 1, nblks
-          call cosp2_init (size(Grid(nb)%xlon,1), Model%levs)
-       enddo
+       !ONLY call with the first block, which will be equal to the blocksize.
+       ! this will (should?) allocate the right amount of data. Then calls to
+       ! COSP will have the right data set up.
+       call cosp2_init (size(Grid(1)%xlon,1), Model%levs)
     endif
 #endif
 
@@ -645,6 +646,9 @@ subroutine GFS_rad_time_vary (Model, Statein, Tbd, sec)
           !--- for testing purposes, replace numrdm with '100'
           Tbd(nb)%icsdsw(ix) = numrdm(i+Model%isc-1 + (j+Model%jsc-2)*Model%cnx)
           Tbd(nb)%icsdlw(ix) = numrdm(i+Model%isc-1 + (j+Model%jsc-2)*Model%cnx + Model%cnx*Model%cny)
+#if defined (USE_COSP)
+          !add random seeds for COSP here
+#endif
         enddo
       enddo
     endif  ! isubc_lw and isubc_sw

GFS_layer/GFS_radiation_driver.F90

@@ -552,7 +552,7 @@ subroutine radinit( si, NLAY, me )
      &          '  May 01 2007'
         print *, VTAGRAD                !print out version tag
         print *,' - Selected Control Flag settings: ICTMflg=',ictmflg,  &
-     &    ' ISOLar =',isolar, ' ICO2flg=',ico2flg,' IAERflg=',iaerflg,  &
+     &    ' ISOLar =',isolar, ' ICO2flg=',ico2flg,' CO2_scaling=',co2_scaling, ' IAERflg=',iaerflg,  &
      &    ' IALBflg=',ialbflg,' IEMSflg=',iemsflg,' ICLDflg=',icldflg,  &
      &    ' ICMPHYS=',icmphys,' IOZNflg=',ioznflg
         print *,' IVFLIP=',ivflip,' IOVRSW=',iovrsw,' IOVRLW=',iovrlw,  &
@@ -1387,7 +1387,9 @@ subroutine GFS_radiation_driver                       &
       ! air and moles of co2 per square meter following the method in
       ! radlw_main.f.  These can be used later to compute a global mean carbon
       ! dioxide volume mixing ratio diagnostic if requested.
-      call compute_column_integrated_moles_of_dry_air_and_co2(Statein, gasvmr, IM, LMK, NF_VGAS, Diag)
+      if (Model%ldiag3d) then
+        call compute_column_integrated_moles_of_dry_air_and_co2(Statein, gasvmr, IM, LMK, NF_VGAS, Diag)
+      endif
 
       if (Model%do_diagnostic_radiation_with_scaled_co2) then
          do n = 1, Model%n_diagnostic_radiation_calls
@@ -1627,7 +1629,7 @@ subroutine GFS_radiation_driver                       &
                          tracer1(:,1:lmk,Model%ntiw), &
                          tracer1(:,1:lmk,Model%ntsw), &
                          tracer1(:,1:lmk,Model%ntgl), qa, &
-                         Sfcprop%slmsk, Sfcprop%snowd, &
+                         Sfcprop%slmsk, &
                          tracer1(:,1:lmk,Model%ntclamt),&
                          im, lmk, lmp, clouds, cldsa, mtopa, mbota)    !  ---  outputs
           endif

README.md

@@ -1,6 +1,6 @@
 # SHiELD_physics
 
-The source contained herein reflects the 202305 release of the SHiELD_physics from GFDL.
+The source contained herein reflects the 202411 release of the SHiELD_physics from GFDL.
 
 SHiELD_physics contains the infrastructure and physical parameterizations used within the SHiELD atmosphere model.
 
@@ -12,9 +12,7 @@ More information is available on the [GFDL SHiELD page](https://www.gfdl.noaa.go
 
 ```IPD_layer``` was developed by GFDL personnel, with community input, under funding from NWS Office of Science and Technology Integration.
 
-```atmos_drivers``` is derivative code from the NOAA-GFDL [Atmospheric Drivers project](https://github.com/NOAA-GFDL/atmos_drivers).  This will be removed in a future release.
-
-```simple_coupler``` is derivative code from the NOAA-GFDL [FMS Coupler project](https://github.com/NOAA-GFDL/FMSCoupler).  This will be removed in a future release.
+```atmos_shared``` contains files needed to compile SHiELD with the full FMS coupler.
 
 # Disclaimer
 

RELEASE.md

@@ -1,3 +1,59 @@
+# RELEASE NOTES for FV3 202411: Summary
+FV3-202411-public --- November 2024
+Lucas Harris, GFDL [email protected]
+
+This version has been tested with:  
+FV3 Dynamical Core release FV3-202411-public from https://github.com/NOAA-GFDL/GFDL_atmos_cubed_sphere  
+FMS release 2024.03 from https://github.com/NOAA-GFDL/FMS  
+FMS Coupler release 2024.03.01 from https://github.com/NOAA-GFDL/FMScoupler  
+Atmos Drivers release FV3-202411-public from https://github.com/NOAA-GFDL/atmos_drivers  
+
+This release includes the following:
+- Support for running SHiELD in the full FMSCoupler (Joseph)
+- GFDL Microphysics (Linjiong)
+  - Replaced the hardcoded qa with aerosol data
+  - Replaced the hardcoded hydrostatic with the input hydrostatic variable
+  - Added two namelist options (fast_fr_mlt and fast_dep_sub) to control freezing/melting and deposition/sublimation in the fast microphysics.
+  - Included a missing term in the energy conservation formula (credit: Tristan Abbott).  May affect  prediction of processes depending strongly on microphysics. Compile the model with -DENG_CNV_OLD to revert this change.
+  - Added diagnostic cloud content and cloud effective radii of all cloud hydrometeors (qc*, re*).
+  - Added diagnostic microphysical process rate (mpp*).
+  - Removed unused Keihl et al. (1994) cloud water effective radius diagnosis.
+  - A new COSP output variable named clisccp was added.
+- Surface processes and PBL (Kun, Linjiong, Kai, Spencer)
+  - Brought EMC updates from dev/emc branch related to TKE-EDMF, deep and shallow convection schemes
+  - Added a constraint to the PBL bottom layer to prevent negative water vapor. 
+  - Revise the surface process to prevent excessive downward latent heat transfer 
+  - Exposed additional tunable parameters in TKE-EDMF PBL scheme
+  - Fix miscalculated albedo over the ocean when ialb = 2
+  - Slab ocean bug fixes and enhancements
+  - Upgraded land-ice related processes in Noah-MP based on version 4.5 of the official Noah-MP repository and some updates from NCAR/ccpp-physics, mostly from EMC.
+    - Computing the snow height within the snowwater_glacier and snowwater subroutines
+    - Limiting the compaction of snow in various places
+    - Calculating the equilibrium state of snow only if snow exists. The maximum water depth becomes a tunable parameter and its default value remains 5 m.
+    - Prescribed soil color capability
+    - Prescribed snow albedo capability
+- Convection schemes (Kun, Linjiong)
+  - Brought EMC updates from dev/emc branch
+  - Merged samfshalcnv_gfdl.f and samfshalcnv.f.  Use gfs_physics_nml::limit_shal_conv = .true. to recover samfshalcnv_gfdl. 
+  - Created the shallow convection option 5 based on option 3, to turn off the shallow convection scheme if the diagnosed cloud depth or cloud top exceeds certain critical values (set by cthk_shal and top_shal, respectively)
+  - Added a parameter dxcrtas to the deep convection; for grid-spacing less than this (default 8000 m) the AS's quasi-equilibrium assumption is considered invalid.
+- Radiation (Kai)
+  - Added a namelist option, gfs_physics_nml::fco2_scaling to uniformly scale CO2
+- Diagnostics (Spencer)
+  - Added pressure_level_extrapolate, blended-area-weighted, and simplest model_level_area_weighted coarse-graining strategies.
+  - Implemented area-weighted mean over the dominant surface type for the snoalb, shdmin, and shdmax surface fields.
+  - Fixed a bug that affected coarse-graining the soil type field and some other soil properties downstream.
+  - Additional CO2 and surface-type fraction diagnostics.
+- Software updates (Spencer, Lucas, Joseph)
+  - Updates for running SHiELD in the FV3net Python wrapper
+  - Bugfix in ozphys.F for ldiag3d = .true..
+  - Cleaned up obscure diagnostics in radiation_astronomy.f.
+  - Namelist option gfs_physics_nml::landseaprt to allow turning off land/sea partitioned global intervals in stdout.
+  - Added variable-by-variable defined missing_value to gfdl_diag_type.
+  - Add additional atmos driver files to run SHiELD and SHiEMOM with the full FMScoupler, mimicking the structure of AM4 atmos_phys.
+  - Several diagnostic bug fixes.
+
+
 # RELEASE NOTES for FV3 202305: Summary
 FV3-202305-public --- May 2023
 Lucas Harris, GFDL [email protected]

gsmphys/gcycle.F90

@@ -191,7 +191,8 @@ SUBROUTINE GCYCLE (nblks, Model, Grid, Sfcprop, Cldprop)
                      size(Model%input_nml_file),                  &
                      Model%input_nml_file,                        &
                      Model%ialb, Model%isot, Model%ivegsrc,       &
-                     trim(tile_num_ch), i_index, j_index)
+                     trim(tile_num_ch), i_index, j_index, &
+                     Model%sst_perturbation)
 #ifndef INTERNAL_FILE_NML
       close (Model%nlunit)
 #endif