cam6_4_005: Limit vertical domain used by COSP #1010

brian-eaton · 2024-04-17T18:25:31Z

The COSP simulator was not working with "FMT" compsets. This compset has a model top of about 1 Pa which is above where the cloud parameterizations operate. The COSP interface routine was modified so that COSP operates on the same vertical domain as the cloud parameterizations which is set the the namelist variable trop_cloud_top_press (1 mb by default). In addition a lot of code cleanup was done, and a bug fix was made for the layer interface values of height and pressure passed from CAM to COSP.

resolves #967- COSP prevents running "FMT" compsets.

Removed old tools for topo file generation.

resolves #1005 - Remove old topo generation software from CAM

…cosp-vlim

src/physics/cam/cospsimulator_intr.F90

cacraigucar · 2024-04-29T22:23:49Z

src/physics/cam/cospsimulator_intr.F90

-         dtau_s(1:ncol,1:pver),dem_c(1:ncol,1:pver),                           &
-         dem_s(1:ncol,1:pver),dtau_s_snow(1:ncol,1:pver),                      &
-         dem_s_snow(1:ncol,1:pver),state%ps(1:ncol),cospstateIN,cospIN)
+    call subsample_and_optics( &


It might be good to have a comment that subsample_and_optics works on only the active columns. Arrays that are dimensioned by pcols need to be subsetted during the call to be :ncol. I'm thinking about the scientist who adds a new variable to this list. They might or might not understand this nuance. When all arrays were being subsetted, the pattern was obvious. Now it is not so obvious ( but is more memory efficient). There is the added nuance of arrays going from ktop:pver (the new nlay). I suggest a comment to help the person modifying this call would be useful.

I added a hopefully clarifying comment. Let me know if it could be improved.

That sounds great and hopefully will help folks in the future. "alread" needs the "y" added to it.

src/physics/cam/cospsimulator_intr.F90

src/utils/hycoef.F90

…cosp-vlim

cacraigucar

One minor typo, otherwise my review is complete

src/physics/cam/cospsimulator_intr.F90

cacraigucar · 2024-05-20T22:41:39Z

src/physics/cam/cospsimulator_intr.F90

-         dtau_s(1:ncol,1:pver),dem_c(1:ncol,1:pver),                           &
-         dem_s(1:ncol,1:pver),dtau_s_snow(1:ncol,1:pver),                      &
-         dem_s_snow(1:ncol,1:pver),state%ps(1:ncol),cospstateIN,cospIN)
+    call subsample_and_optics( &


That sounds great and hopefully will help folks in the future. "alread" needs the "y" added to it.

nusbaume

Great cleanup! I do have some questions and suggestions, but none of them are really required for this particular PR, so will go ahead and approve it now.

nusbaume · 2024-06-01T21:31:49Z

doc/ChangeLog

+the same vertical domain as the cloud parameterizations which is set the
+the namelist variable trop_cloud_top_press (1 mb by default).  Changing to


Typo:

Suggested change

the same vertical domain as the cloud parameterizations which is set the

the namelist variable trop_cloud_top_press (1 mb by default). Changing to

the same vertical domain as the cloud parameterizations which is set by

the namelist variable trop_cloud_top_press (1 mb by default). Changing to

nusbaume · 2024-06-02T18:56:58Z

src/physics/cam/ref_pres.F90

+      'hPa', trop_pref, positive='down')
+
+   allocate(trop_prefi(nlev+1), stat=istat)
+   call alloc_err(istat, sub, 'trop_prefi', nlev)


Should this be nlev+1 here?

Suggested change

call alloc_err(istat, sub, 'trop_prefi', nlev)

call alloc_err(istat, sub, 'trop_prefi', nlev+1)

Yes. Fixed.

nusbaume · 2024-06-10T14:45:08Z

src/physics/cam/cospsimulator_intr.F90


+    ! The COSP init method was run from cospsimulator_intr_register in order to add


I assume this comment is referring to the setcosp2values call? If so then I might add it to the text here:

Suggested change

! The COSP init method was run from cospsimulator_intr_register in order to add

! The COSP init method (setcosp2values) was run from cospsimulator_intr_register in order to add

nusbaume · 2024-06-10T15:03:32Z

src/physics/cam/cospsimulator_intr.F90

+            'PARASOL-like mono-directional reflectance ', flag_xyfill=.true., fill_value=R_UNDEF)
+       call addfld('CFAD_SR532_CAL', (/'cosp_sr','cosp_ht'/), 'A', 'fraction', &
+            'Calipso Scattering Ratio CFAD (532 nm)', flag_xyfill=.true., fill_value=R_UNDEF)
+       call addfld('MOL532_CAL', (/'trop_pref'/), 'A', 'm-1sr-1', &


I might put a space here:

Suggested change

call addfld('MOL532_CAL', (/'trop_pref'/), 'A', 'm-1sr-1', &

call addfld('MOL532_CAL', (/'trop_pref'/), 'A', 'm-1 sr-1', &

nusbaume · 2024-06-10T15:12:16Z

src/physics/cam/cospsimulator_intr.F90

+       call addfld('CLD_CAL_TMP', (/'cosp_ht'/), 'A', 'percent', &
+            'NOT SURE WHAT THIS IS Cloud Fraction', flag_xyfill=.true., fill_value=R_UNDEF)
+       call addfld('CLD_CAL_TMPLIQ', (/'cosp_ht'/), 'A', 'percent', &
+            'NOT SURE WHAT THIS IS Cloud Fraction', flag_xyfill=.true., fill_value=R_UNDEF)
+       call addfld('CLD_CAL_TMPICE', (/'cosp_ht'/), 'A', 'percent', &
+            'NOT SURE WHAT THIS IS Cloud Fraction', flag_xyfill=.true., fill_value=R_UNDEF)
+       call addfld('CLD_CAL_TMPUN', (/'cosp_ht'/), 'A', 'percent', &
+            'NOT SURE WHAT THIS IS Cloud Fraction', flag_xyfill=.true., fill_value=R_UNDEF)


@brianpm Any idea what these history fields are? We don't have to deal with these for this particular PR but it might be good to eventually either provide better long names or remove them depending on what these quantities are actually representing.

These are temperatures ("TMP").

CLD_CAL_TMP = Calipso Cloud Temperature
CLD_CAL_TMPLIQ = Calipso Liquid Cloud temperature
CLD_CAL_TMPICE = Calipso Ice Cloud temperature
CLD_CAL_TMPUN = Calipso Undefined-Phase Cloud Temperature

Updated field long names and units.

nusbaume · 2024-06-10T20:03:13Z

src/physics/cam/cospsimulator_intr.F90

-    reff_cosp(1:ncol,1:pver,1:nhydro) = 0._r8
-    ! note: reff_cosp dimensions should be same as cosp (reff_cosp has 9 hydrometeor dimension)
-    ! Reff(Npoints,Nlevels,N_HYDRO)
-
    use_precipitation_fluxes = .true.      !!! consistent with cam4 implementation.


I know you didn't write the code, but it looks like use_precipitation_fluxes is always true, so why have it as a variable at all?

use_precipitation_fluxes removed.

nusbaume · 2024-06-10T20:11:52Z

src/physics/cam/cospsimulator_intr.F90

    end do

+    grpl_ls_interp = 0._r8


My understanding it is that PUMAS microphysics is now capable of simulating graupel. Should there be an attempt to try and couple it to COSP here? This is probably yet another question for @brianpm (and doesn't have to be done for this particular PR).

I don't know. This is a good question, but I would need to look into whether the satellite products think they see graupel.

nusbaume · 2024-06-10T20:22:55Z

src/physics/cam/cospsimulator_intr.F90

+    !! if use_reff=.false. then all sizes use DEFAULT_LIDAR_REFF = 30.0e-6 meters
    use_reff = .true.


Is this variable actually being used? I can't seem to find its use anywhere.

use_reff removed.

nusbaume · 2024-06-10T20:39:18Z

src/physics/cam/cospsimulator_intr.F90

+    cospstateIN%lat             = state%lat(:ncol)*180._r8/pi
+    cospstateIN%lon             = state%lon(:ncol)*180._r8/pi


Could 180._r8/pi be saved as a parameter instead of being re-calculated every call? I do realize this could cause round-off changes and so can certainly be ignored if answers are currently bit-for-bit for this PR.

Done. Defined parameter rad2deg.

nusbaume · 2024-06-10T20:52:18Z

src/physics/cam/cospsimulator_intr.F90

       call outfld('RH_COSP',        cospstateIN%qv,               ncol,lchnk)
-       call outfld('Q_COSP',         q(1:ncol,1:pver),             ncol,lchnk)


Just double-checking that cospstateIN%qv, which was originally set to be specific humidity, has been converted to what I assume is supposed to be relative humidity (RH_COSP)?

I didn't see the conversion, but maybe it happens somewhere. I asked the same in my response about the names above.

COSP takes specific humidity as an input. I replaced RH_COSP (which was set to specific humidity) by Q_COSP in the history output.

Merge pull request ESCOMP#1010 from brian-eaton/cosp-vlim cam6_4_005: Limit vertical domain used by COSP ESCOMP commit: 08d0326

brian-eaton added 12 commits March 29, 2024 13:44

mods to limit COSP top

98d2461

mods to limit COSP top

3b235e4

add coordinate trop_pref

f3000de

cosp outputs use new vertical coords as appropriate

6bc197b

merge cam/physpkg.F90 changes to cam_dev/physpkg.F90

8582549

address compiler messages for cospsimulator_intr

dbf1564

fix interface height and pressure inputs

9c60563

Merge branch 'cam_development' of https://github.com/ESCOMP/CAM into …

04e7d9c

…cosp-vlim

add ChangeLog

66534c9

fix for restart with COSP

168c302

remove old topo file generation tools

e41b167

update to cam6_3_156

f43047c

brian-eaton self-assigned this Apr 17, 2024

brian-eaton added BFB bit for bit tag bug-fix This PR was created to fix a specific bug. labels Apr 17, 2024

brian-eaton added 2 commits April 18, 2024 10:46

update ChangeLog

f18cf23

update to cam6_3_157

715123b

cacraigucar self-requested a review April 22, 2024 22:48

cacraigucar requested changes May 6, 2024

View reviewed changes

brian-eaton added 3 commits May 7, 2024 19:03

add allocate stat checks; add some comments

37961cf

Merge branch 'cam_development' of https://github.com/ESCOMP/CAM into …

616d75f

…cosp-vlim

update ChangeLog

5550915

brian-eaton requested a review from cacraigucar May 13, 2024 16:34

brian-eaton mentioned this pull request May 15, 2024

Radiatively active cloud water missing from COSP #1027

Closed

cacraigucar approved these changes May 20, 2024

View reviewed changes

cacraigucar requested a review from nusbaume May 20, 2024 22:48

fix typo in comment

267a37f

nusbaume approved these changes Jun 10, 2024

View reviewed changes

cacraigucar changed the title ~~Limit vertical domain used by COSP~~ cam6_4_003: Limit vertical domain used by COSP Jun 17, 2024

cacraigucar changed the title ~~cam6_4_003: Limit vertical domain used by COSP~~ cam6_4_004: Limit vertical domain used by COSP Jun 17, 2024

cacraigucar changed the title ~~cam6_4_004: Limit vertical domain used by COSP~~ cam6_4_005: Limit vertical domain used by COSP Jun 24, 2024

brian-eaton added 4 commits July 1, 2024 12:00

address review comments

9ebd4be

update to cam6_4_004

7dd6cbd

fix misplaced USE_COSP ifdef

b210815

update ChangeLog

bffdd2e

brian-eaton merged commit 08d0326 into ESCOMP:cam_development Jul 1, 2024

gold2718 pushed a commit to gold2718/CAM that referenced this pull request Oct 16, 2024

ESCOMP tag: cam6_4_005

994168a

Merge pull request ESCOMP#1010 from brian-eaton/cosp-vlim cam6_4_005: Limit vertical domain used by COSP ESCOMP commit: 08d0326

gold2718 pushed a commit to gold2718/CAM that referenced this pull request Oct 16, 2024

ESCOMP tag: cam6_4_005

187799a

Merge pull request ESCOMP#1010 from brian-eaton/cosp-vlim cam6_4_005: Limit vertical domain used by COSP ESCOMP commit: 08d0326

gold2718 pushed a commit to gold2718/CAM that referenced this pull request Nov 8, 2024

ESCOMP tag: cam6_4_005

ff0b850

Merge pull request ESCOMP#1010 from brian-eaton/cosp-vlim cam6_4_005: Limit vertical domain used by COSP ESCOMP commit: 08d0326

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

cam6_4_005: Limit vertical domain used by COSP #1010

cam6_4_005: Limit vertical domain used by COSP #1010

brian-eaton commented Apr 17, 2024

cacraigucar Apr 29, 2024

brian-eaton May 7, 2024

cacraigucar May 20, 2024

cacraigucar left a comment

cacraigucar May 20, 2024

nusbaume left a comment

nusbaume Jun 1, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 2, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 10, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 10, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 10, 2024

brianpm Jun 12, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 10, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 10, 2024

brianpm Jun 12, 2024

nusbaume Jun 10, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 10, 2024

brian-eaton Jul 1, 2024

nusbaume Jun 10, 2024

brianpm Jun 12, 2024

brian-eaton Jul 1, 2024

		the same vertical domain as the cloud parameterizations which is set the
		the namelist variable trop_cloud_top_press (1 mb by default). Changing to

	call alloc_err(istat, sub, 'trop_prefi', nlev)
	call alloc_err(istat, sub, 'trop_prefi', nlev+1)


		! The COSP init method was run from cospsimulator_intr_register in order to add

	! The COSP init method was run from cospsimulator_intr_register in order to add
	! The COSP init method (setcosp2values) was run from cospsimulator_intr_register in order to add

	call addfld('MOL532_CAL', (/'trop_pref'/), 'A', 'm-1sr-1', &
	call addfld('MOL532_CAL', (/'trop_pref'/), 'A', 'm-1 sr-1', &

		!! if use_reff=.false. then all sizes use DEFAULT_LIDAR_REFF = 30.0e-6 meters
		use_reff = .true.

		cospstateIN%lat = state%lat(:ncol)*180._r8/pi
		cospstateIN%lon = state%lon(:ncol)*180._r8/pi

		call outfld('RH_COSP', cospstateIN%qv, ncol,lchnk)
		call outfld('Q_COSP', q(1:ncol,1:pver), ncol,lchnk)

cam6_4_005: Limit vertical domain used by COSP #1010

cam6_4_005: Limit vertical domain used by COSP #1010

Conversation

brian-eaton commented Apr 17, 2024

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

nusbaume 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

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

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

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

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment