Complete Zhang McFarlane conversion to CCPP #186

cacraigucar · 2024-12-24T20:19:20Z

closes #66

…mpilation

…eanup3

Pull ZM 0.5 timestep removal into main

Remove 0.5*timestop logic from all to zm Register species from MICM

Replace original, locally-developed license with Apache 2.0 license.

Bring in development ESCOMP#108 (TUV-x) ESCOMP#112 (tropopause_find) to main

Bring in new directory structure from development

…stics

…mcleanup4

…allow as well

@jimmielin

Tag name (The PR title should also include the tag name): `atmos_phys0_07_001` Originator(s): @jimmielin List all `development` PR URLs included in this PR and a short description of each: * Update extraterrestrial flux in TUV-x prior to calculating rate constants ESCOMP#152 by @boulderdaze * Simplify deallocation of multiple objects associated with the TUV-x ESCOMP#156 by @boulderdaze * Fill in errmsg, errflg in check_energy schemes ESCOMP#160 by @jimmielin * Validates the MUSICA meta data against the CCPP standard names ESCOMP#162 by @boulderdaze * Add constituent tendency updater ESCOMP#111 by @peverwhee * Add cloud optical calculations for use in TUV-x ESCOMP#167 by @mattldawson * Add initialize_constituents scheme ESCOMP#149 by @peverwhee * Add diagnostics to TJ2016 test schemes ESCOMP#170 by @peverwhee * update "radians" to "rad" ESCOMP#173 by @peverwhee * Solar zenith angle and Earth-Sun distance ESCOMP#171 by @mattldawson * Update standard names for tropopause_find ESCOMP#140 by @jimmielin * Update surface albedo units ESCOMP#181 by @mattldawson * don't set water species property for species that air_composition handles ESCOMP#185 by @peverwhee List all test failures: N/A

jimmielin

Thanks @cacraigucar, and congratulations on completing ZM 🎉!!!

I have minor, hopefully just formatting related changes. As mentioned in the chat I will avoid proposing large changes (e.g., moving the ZM namelist scheme) in this PR so it can be merged and I can propose the changes in another future PR.

I have tested building and running of Hack shallow convection with cloud_fraction_fice and zm_conv_evap and verified they are within roundoff (and bit-for-bit for history output), as well as (modified for shallow) the interstitial schemes for conv fluxes to/from general/deep.

schemes/zhang_mcfarlane/set_general_conv_fluxes_to_deep.F90

schemes/zhang_mcfarlane/zm_convr_namelist.xml

schemes/sima_diagnostics/zm_diagnostics.F90

schemes/sima_diagnostics/zm_evap_tendency_diagnostics.F90

schemes/sima_diagnostics/zm_tendency_diagnostics.F90

jimmielin · 2025-01-09T19:03:25Z

schemes/utilities/to_be_ccppized_temporary.F90

@@ -0,0 +1,29 @@
+module to_be_ccppized_temporary


Perhaps a call to this has to be added to the CAM7 SDF as well which will use ZM?

schemes/zhang_mcfarlane/zm_conv_convtran.F90

jimmielin · 2025-01-09T19:05:28Z

suites/suite_zhang_mcfarlane.xml

+   <group name="physics_before_coupler">
+       <scheme>initialize_constituents</scheme>
+       <scheme>to_be_ccppized_temporary</scheme>
+       <scheme>zm_convr</scheme>


I'm not sure if this is within the scope of the PR, but maybe the calls to ZM need to be added to the corresponding locations in the CAM7 SDF as well.

@nusbaume and/or @peverwhee - Should ZM be added to the CAM7 SDF?

cacraigucar

I believe that I have addressed all of your comment, though I might have missed some.

cacraigucar · 2025-01-10T21:07:14Z

suites/suite_zhang_mcfarlane.xml

+   <group name="physics_before_coupler">
+       <scheme>initialize_constituents</scheme>
+       <scheme>to_be_ccppized_temporary</scheme>
+       <scheme>zm_convr</scheme>


@nusbaume and/or @peverwhee - Should ZM be added to the CAM7 SDF?

schemes/zhang_mcfarlane/zm_conv_convtran.F90

schemes/sima_diagnostics/zm_tendency_diagnostics.F90

schemes/sima_diagnostics/zm_momtran_tendency_diagnostics.F90

schemes/sima_diagnostics/zm_evap_tendency_diagnostics.F90

schemes/sima_diagnostics/zm_diagnostics.F90

schemes/zhang_mcfarlane/set_general_conv_fluxes_to_deep.F90

Upgrading to atmos_phys0_07_001 (ESCOMP#187)

jimmielin

Thanks @cacraigucar for addressing my comments!

There are a few comments I left open to get @nusbaume and @peverwhee's input, about the history variables and inclusion of ZM in the CAM7 SDF. I've marked the other comments as resolved.

peverwhee

hooray! zm! a few (hopefully) small things!

note: i didn't review the code in to_be_ccppized since I assumed it was copied over from CAM...

also, reminder that you'll want to regenerate NamesNotInDictionary.txt after all the reviews are done.

schemes/sima_diagnostics/zm_diagnostics.F90

schemes/sima_diagnostics/zm_tendency_diagnostics.F90

schemes/zhang_mcfarlane/zm_conv_convtran.F90

schemes/zhang_mcfarlane/zm_conv_evap.meta

peverwhee · 2025-01-13T20:49:13Z

schemes/zhang_mcfarlane/zm_conv_evap.F90


          work2 = max(fsnow_conv(i,k), work1)
          if (snowmlt(i).gt.0._kind_phys) work2 = 0._kind_phys
-          ntsnprd(i,k) = prdprec(i,k)*work2 - evpsnow(i) - snowmlt(i)
-          tend_s_snwprd  (i,k) = prdprec(i,k)*work2*latice
+          ntsnprd(i,k) = prdprec_gen(i,k)*work2 - evpsnow(i) - snowmlt(i)
+          tend_s_snwprd  (i,k) = prdprec_gen(i,k)*work2*latice
          tend_s_snwevmlt(i,k) = - ( evpsnow(i) + snowmlt(i) )*latice
-      else
-          ntsnprd(i,k) = prdsnow(i,k) - min(flxsnow(i,k)*gravit/pdel(i,k), evpsnow(i)+snowmlt(i))
-          tend_s_snwprd  (i,k) = prdsnow(i,k)*latice
-          tend_s_snwevmlt(i,k) = -min(flxsnow(i,k)*gravit/pdel(i,k), evpsnow(i)+snowmlt(i) )*latice
      end if


same with this block - if we're always using old_snow, can replace lines 207 - 219 with:

if (flxprec(i,k).gt.0._kind_phys) then work1 = min(max(0._kind_phys,flxsnow(i,k)/flxprec(i,k)),1._kind_phys) else work1 = 0._kind_phys endif work2 = max(fsnow_conv(i,k), work1) if (snowmlt(i).gt.0._kind_phys) work2 = 0._kind_phys ntsnprd(i,k) = prdprec_gen(i,k)*work2 - evpsnow(i) - snowmlt(i) tend_s_snwprd (i,k) = prdprec_gen(i,k)*work2*latice tend_s_snwevmlt(i,k) = - ( evpsnow(i) + snowmlt(i) )*latice

I asked Rich Neale about this, but he has not responded. Since this could fit under "code cleanup", I am inclined to leave it. Also, ZM may be having some science changes coming in, so we could bring that up when that PR comes along.

ok! I guess I'd like to be sure we haven't lost functionality in CAM with this PR? since there used to be a way to make old_snow=.false. (by passing in prdsnow)

schemes/zhang_mcfarlane/zm_conv_momtran.F90

schemes/zhang_mcfarlane/zm_convr.F90

schemes/zhang_mcfarlane/zm_conv_evap.meta

schemes/utilities/to_be_ccppized_temporary.meta

cacraigucar

I believe I have addressed all of @peverwhee's comments, but I may have missed one

schemes/utilities/to_be_ccppized_temporary.meta

schemes/zhang_mcfarlane/zm_convr.F90

schemes/zhang_mcfarlane/zm_conv_momtran.F90

schemes/zhang_mcfarlane/zm_conv_evap.F90

schemes/sima_diagnostics/zm_diagnostics.F90

schemes/zhang_mcfarlane/zm_conv_evap.meta

cacraigucar · 2025-01-14T20:41:04Z

schemes/zhang_mcfarlane/zm_conv_evap.F90


          work2 = max(fsnow_conv(i,k), work1)
          if (snowmlt(i).gt.0._kind_phys) work2 = 0._kind_phys
-          ntsnprd(i,k) = prdprec(i,k)*work2 - evpsnow(i) - snowmlt(i)
-          tend_s_snwprd  (i,k) = prdprec(i,k)*work2*latice
+          ntsnprd(i,k) = prdprec_gen(i,k)*work2 - evpsnow(i) - snowmlt(i)
+          tend_s_snwprd  (i,k) = prdprec_gen(i,k)*work2*latice
          tend_s_snwevmlt(i,k) = - ( evpsnow(i) + snowmlt(i) )*latice
-      else
-          ntsnprd(i,k) = prdsnow(i,k) - min(flxsnow(i,k)*gravit/pdel(i,k), evpsnow(i)+snowmlt(i))
-          tend_s_snwprd  (i,k) = prdsnow(i,k)*latice
-          tend_s_snwevmlt(i,k) = -min(flxsnow(i,k)*gravit/pdel(i,k), evpsnow(i)+snowmlt(i) )*latice
      end if


I asked Rich Neale about this, but he has not responded. Since this could fit under "code cleanup", I am inclined to leave it. Also, ZM may be having some science changes coming in, so we could bring that up when that PR comes along.

cacraigucar · 2025-01-14T20:41:42Z

schemes/zhang_mcfarlane/zm_conv_evap.F90

@@ -192,7 +176,7 @@ subroutine zm_conv_evap_run(ncol, pver, pverp, &
          evplimit   = min(evplimit, flxprec(i,k) * gravit / pdel(i,k))

 ! Total evaporation cannot exceed input precipitation
-          evplimit   = min(evplimit, (prec(i) - evpvint(i)) * gravit / pdel(i,k))
+          evplimit   = min(evplimit, (prec_gen(i) - evpvint(i)) * gravit / pdel(i,k))

          evpprec(i) = min(evplimit, evpprec(i))
          if( .not.old_snow ) then


See response on other old_snow comment

peverwhee · 2025-01-16T17:17:24Z

schemes/zhang_mcfarlane/zm_conv_convtran.meta

  intent = in
+  advected = true


Oops, should've clarified. advected = true is only for individual constituents (not the whole array). I imagine this "works" since you've tested it, but I'd remove this to avoid future confusion

agh i'm annoyed that this comment ended up here rather than in my review.

peverwhee · 2025-01-16T17:21:25Z

schemes/zhang_mcfarlane/zm_conv_evap.F90


          work2 = max(fsnow_conv(i,k), work1)
          if (snowmlt(i).gt.0._kind_phys) work2 = 0._kind_phys
-          ntsnprd(i,k) = prdprec(i,k)*work2 - evpsnow(i) - snowmlt(i)
-          tend_s_snwprd  (i,k) = prdprec(i,k)*work2*latice
+          ntsnprd(i,k) = prdprec_gen(i,k)*work2 - evpsnow(i) - snowmlt(i)
+          tend_s_snwprd  (i,k) = prdprec_gen(i,k)*work2*latice
          tend_s_snwevmlt(i,k) = - ( evpsnow(i) + snowmlt(i) )*latice
-      else
-          ntsnprd(i,k) = prdsnow(i,k) - min(flxsnow(i,k)*gravit/pdel(i,k), evpsnow(i)+snowmlt(i))
-          tend_s_snwprd  (i,k) = prdsnow(i,k)*latice
-          tend_s_snwevmlt(i,k) = -min(flxsnow(i,k)*gravit/pdel(i,k), evpsnow(i)+snowmlt(i) )*latice
      end if


ok! I guess I'd like to be sure we haven't lost functionality in CAM with this PR? since there used to be a way to make old_snow=.false. (by passing in prdsnow)

peverwhee

a couple lingering things

peverwhee · 2025-01-16T17:24:04Z

schemes/zhang_mcfarlane/zm_convr.meta

  intent = in
+  advected = true


looks like this one has "advected = true" twice

peverwhee · 2025-01-16T17:25:31Z

schemes/sima_diagnostics/zm_diagnostics.meta

+#[ dif ]
+#  standard_name = detrainment_of_cloud_ice_due_to_deep_convection
+#  units = kg kg-1 s-1
+#  type = real | kind = kind_phys
+#  dimensions = (horizontal_loop_extent,vertical_layer_dimension)
+#  intent = in


this can be removed

…mcleanup4

cacraigucar and others added 30 commits May 17, 2024 11:52

Metadata completed and partial namelist

c6f2699

Remove unused variables

c6ddcb4

Cleanup3 updates and namelist xml file added

2e96710

put back in dif inside zm_convr_run

3c6447a

Remove dif from calling list

2abb58d

Bring limcnv calculation into init

2d7b589

Update ZM metadata, add namelist and preliminary mods for CAM-SIMA co…

5b89cb5

…mpilation

Remove prdsnow completely

2208d45

Add missing declaration for standard_name

70c0450

hook up constituents

2252688

Updates to get ZM to compile in CAM-SIMA

c8076de

Merge tag 'atmos_phys0_03_000' into atmos_phys_zmcleanup3

102cd40

Fixes for cleanup3

755a07b

additional changes to get ZM running in CAM-SIMA

e8dc85f

Merge remote-tracking branch 'court/diagnostics' into atmos_phys_zmcl…

6cf8164

…eanup3

Preliminary zm_diagnostics

c790696

preliminary diagnostics for ZM

00c566d

Merge pull request ESCOMP#110 from ESCOMP/development

d9d0e5d

Pull ZM 0.5 timestep removal into main

Merge tag 'atmos_phys_0_04_001' into atmos_phys_zmcleanup3

721142e

Remove 0.5*timestop logic from all to zm Register species from MICM

Remove zm_org logic per Rich Neale's advice

34e609b

Changes to get ZM to run in CAM-SIMA

aba6bee

Changes to fix answers - now roundoff

094d677

Change Liceense to Apache 2.0

fd27d14

Replace original, locally-developed license with Apache 2.0 license.

Remove old license file.

592fe83

cleanup and fix a couple of bugs

bff9925

add some ZM diagnostics

752c550

Merge pull request ESCOMP#121 from ESCOMP/development

93a1dbf

Bring in development ESCOMP#108 (TUV-x) ESCOMP#112 (tropopause_find) to main

Merge pull request ESCOMP#130 from ESCOMP/development

f8ce60b

Bring in new directory structure from development

Next phase of updates to ZM, including introducing zm_tendency_diagno…

d17b4ed

…stics

preliminary diagnostics

307d8bc

cacraigucar added 8 commits December 9, 2024 13:31

Merge remote-tracking branch 'upstream/development' into atmos_phys_z…

5b34c32

…mcleanup4

Further cleanup

f0f9ac6

Finalize ZM diagnostics

3d44c2b

restore diagnostic message

3349c8f

Fix error message

73d6a7d

Merge remote-tracking branch 'upstream/development' into atmos_phys_z…

d119665

…mcleanup4

Make zm_conv_evap variables generic so it can be called by convect_sh…

2380fa6

…allow as well

Update NamesNotInDictionary ane preliminary ChangeLog

5766e67

cacraigucar requested review from jimmielin, peverwhee and nusbaume December 24, 2024 20:19

cacraigucar self-assigned this Dec 24, 2024

cacraigucar changed the base branch from main to development January 2, 2025 23:46

jimmielin requested changes Jan 9, 2025

View reviewed changes

Address reviewer comments

52e951e

cacraigucar commented Jan 13, 2025

View reviewed changes

cacraigucar requested a review from jimmielin January 13, 2025 17:49

cacraigucar added 2 commits January 13, 2025 10:55

Merge tag 'atmos_phys0_07_001' into atmos_phys_zmcleanup4

a6e3d34

Upgrading to atmos_phys0_07_001 (ESCOMP#187)

Address one more review comment

5789a10

jimmielin approved these changes Jan 13, 2025

View reviewed changes

peverwhee requested changes Jan 13, 2025

View reviewed changes

cacraigucar added 2 commits January 14, 2025 15:42

Address second round of reviewer comments

a7f33e5

one more change from second reviewer comments

9d03025

cacraigucar commented Jan 14, 2025

View reviewed changes

cacraigucar requested a review from peverwhee January 14, 2025 23:26

Update CAM7 SDF

4eaff0d

peverwhee reviewed Jan 16, 2025

View reviewed changes

peverwhee requested changes Jan 16, 2025

View reviewed changes

Merge remote-tracking branch 'upstream/development' into atmos_phys_z…

582a284

…mcleanup4

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Complete Zhang McFarlane conversion to CCPP #186

Complete Zhang McFarlane conversion to CCPP #186

cacraigucar commented Dec 24, 2024

jimmielin left a comment

jimmielin Jan 9, 2025

jimmielin Jan 9, 2025

cacraigucar Jan 10, 2025

cacraigucar left a comment

cacraigucar Jan 10, 2025

jimmielin left a comment

peverwhee left a comment

peverwhee Jan 13, 2025

cacraigucar Jan 14, 2025

peverwhee Jan 16, 2025 •

edited

Loading

cacraigucar left a comment

cacraigucar Jan 14, 2025

cacraigucar Jan 14, 2025

peverwhee Jan 16, 2025

peverwhee Jan 16, 2025

peverwhee Jan 16, 2025 •

edited

Loading

peverwhee left a comment

peverwhee Jan 16, 2025

peverwhee Jan 16, 2025

Complete Zhang McFarlane conversion to CCPP #186

Are you sure you want to change the base?

Complete Zhang McFarlane conversion to CCPP #186

Conversation

cacraigucar commented Dec 24, 2024

jimmielin left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

cacraigucar left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jimmielin left a comment

Choose a reason for hiding this comment

peverwhee left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

peverwhee Jan 16, 2025 • edited Loading

Choose a reason for hiding this comment

cacraigucar left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

peverwhee Jan 16, 2025 • edited Loading

Choose a reason for hiding this comment

peverwhee left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

peverwhee Jan 16, 2025 •

edited

Loading

peverwhee Jan 16, 2025 •

edited

Loading