Implement O3v2 in Linoz #3047

tangq · 2019-07-07T23:51:39Z

Previous PR #3042 has some issues that I couldn't push new commits to the remote branch. The issue was fixed by deleting and recreating the remote branch.

Implement the newly designed Linoz O3v2

Add ozone sink in the boundary layer with a fix e-folding
decay to a typical ozone concentration in the boundary layer
Disable the tropospheric ozone rewrite to mozart climatology
The parameters controlling this new ozone sink are name list variables.
Turn off ozone dry deposition (Currently turned off in the
test runs via user_nl_cam, need to do it by changing compset settings
in future commit)

[Non-BFB] - Non Bit-For-Bit

Implement the newly designed Linoz O3v2 * Add ozone sink in the boundary layer with a fix e-folding decay to a typical ozone concentration in the boundary layer * Disable the tropospheric ozone rewrite to mozart climatology * Turn off ozone dry deposition (Currently turned off in the test runs via user_nl_cam, need to do it by changing compset settings in future commit) * Need to change hard-coded parameters to name list variables with future commit. [Non-BFB] - Non Bit-For-Bit

Change the hard-coded Linoz tuning parameters to name list variables * The tuning parameters are for the boundary layer ozone loss [BFB] - Bit-For-Bit [NML] - Namelist Changing

The new Linoz O3v2 shares the tracer index with real O3. This requires turning off the dry deposition for the redesigned Linoz O3v2's loss in the boundary layer to work. When the interactive tropospheric chemistry is implemented, the ozone dry deposition should be turned on again. [BFB] - Bit-For-Bit [NML] - Namelist Changing

…ngq/atm/O3v2 Modify commit message

tangq · 2019-07-12T07:08:08Z

After the initial implementation, the following up commits 1) change the hard-coded parameters to name list variables enabling more flexible tuning in the future, and 2) turn off the ozone dry deposition by compset settings for all CMIP6 compsets on the master.

rljacob · 2019-07-25T04:26:43Z

Where is the code review document for this feature?

tangq · 2019-07-31T12:39:53Z

@rljacob , the document is on this page: https://acme-climate.atlassian.net/wiki/spaces/ED/pages/976486422/W15+O3v2+Design+Document

wlin7 · 2019-08-26T17:23:04Z

components/cam/bld/namelist_files/namelist_definition.xml

@@ -5454,6 +5454,24 @@ if <varname>linoz_data_type</varname> is 'FIXED'.
 Default: 0 seconds
 </entry>

+<entry id="linoz_lbl" type="integer" category="linoz"


It will be helpful to list here the actual default values -- as specified in build-namelist for linoz_lbl, linoz_tau and linoz_sfc.

Good catch @wlin7 . I will make these changes when cori is back.

The default O3v2 parameter values are now added in the name list definitions.

wlin7 · 2019-08-26T18:46:51Z

The changes look good. Special attention paid to the scoping of the dummy arrays involved in the new subroutine "lin_strat_sfcsink" and all is good.

One more comment to the drydep_list nml. Ideally, the change to the drydep list should be realized via build-namelist (which would invoke set_dep_lists in ChemNamelist). That said, the current format is ok for the handful of compsets in use.

wlin7 · 2019-08-26T20:11:47Z

Hi @tangq , another note concerning the drydep_list as eventually specified in drv_flds_in. What is the consequence if "O3" remains in the nml file after the new O3v2 is in used? Can the model still run properly? It may happen to compsets not covered in this PR.

Change needs to be made to one other file at least: 2000_cam5_av1c-04p2.xml. This is the use case file for FC5AV1C-04P2. I know AV1C- compsets are deprecated and no need to maintain. But FC5AV1C-L is still used for many of the nightly tests. It is an alias to FC5AV1C-04P2. Until another compset is selected for the tests, the change to 2000_cam5_av1c-04p2.xml is needed.

tangq · 2019-08-26T20:14:27Z

It will give the wrong results for ozone. I will add the change to 2000_cam5_av1c-04p2.xml when cori's back.

This drydep change won't be needed till the interactive tropospheric chemistry is implemented later.

tangq · 2019-08-27T18:14:24Z

@wlin7 , I added the changes you suggested yesterday - name list default values and turning off ozone dry deposition in 2000_cam5_av1c-04p2.xml. It should be ready to go now.

wlin7 · 2019-08-26T17:32:22Z

components/cam/bld/namelist_files/use_cases/1850_cam5_CMIP6_bgc.xml

@@ -162,6 +162,9 @@
 <linoz_data_path            >atm/cam/chem/trop_mozart/ub</linoz_data_path>
 <linoz_data_type            >CYCLICAL</linoz_data_type>

+<!-- Turn off ozone dry deposition, as Linoz O3v2 and ozone are not separated for now. Need to turn on ozone dry deposition when interactive tropospheric chemistry is implemented -->
+<drydep_list            >'H2O2', 'H2SO4', 'SO2'</drydep_list>


Reminder: this change will also be needed for the several F-2010 compsets, either via this PR or a future one, depending on the order of merging them to master.

wlin7 · 2019-08-27T20:17:47Z

Thanks @tangq . All is good now. F2010SC5-CMIP6 is in the master now. I can add the change to drydep_list for it after merging this PR to master.

@golaz , can we have your approval to get this PR in? Thanks.

rljacob · 2019-08-27T20:38:43Z

@golaz approved the code review document https://acme-climate.atlassian.net/wiki/spaces/ED/pages/976551960/W15+O3v2

Implement O3v2 in Linoz The following elements are added: . Add ozone sink in the boundary layer with a fix e-folding decay to a typical ozone concentration in the boundary layer . Disable the tropospheric ozone rewrite to mozart climatology . The parameters controlling this new ozone sink are name list variables. . Turn off ozone dry deposition (Currently turned off in the test runs via user_nl_cam, need to do it by changing compset settings in future commit) [Non-BFB] - Non Bit-For-Bit [NML] drv_flds_in changed on the list for drydep_list.

wlin7 · 2019-08-28T20:30:21Z

Merged to next.

Implement O3v2 in Linoz The following elements are implemented in the newly designed Linoz O3v2: . Add ozone sink in the boundary layer with a fix e-folding decay to a typical ozone concentration in the boundary layer . Disable the tropospheric ozone rewrite to mozart climatology . The parameters controlling this new ozone sink are name list variables. . Turn off ozone dry deposition (Currently turned off in the test runs via user_nl_cam, need to do it by changing compset settings in future commit) [Non-BFB] - Non Bit-For-Bit [NML] drydep_list in drv_flds_in changed

wlin7 · 2019-08-29T15:10:54Z

Merged to master. Requests submitted to bless DIFFs: sandistoss3 (38), melvin(7), cori-knl(2). Two FAILs on anvil will need to be blessed after inputdata issue is resolved.

SMS_Ld2.ne30_oECv3_ICG.A_WCYCL1850S_CMIP6.anvil_intel.allactive-v1cmip6
SMS_Ln5.ne30_ne30.FC5AV1C-L.anvil_intel.cam-cosplite

wlin7 · 2019-09-05T13:39:13Z

Request submitted to bless the anvil DIFFs, which were FAILs previously due to machine issue. Note that PR #3057 could also cause DIFF for SMS_Ld2.ne30_oECv3_ICG.A_WCYCL1850S_CMIP6.anvil_intel.allactive-v1cmip6.

tangq added 2 commits July 6, 2019 15:57

Make Linoz tuning parameters name list variables

123f11f

Change the hard-coded Linoz tuning parameters to name list variables * The tuning parameters are for the boundary layer ozone loss [BFB] - Bit-For-Bit [NML] - Namelist Changing

tangq added Atmosphere non-BFB PR makes roundoff changes to answers. labels Jul 7, 2019

tangq added this to the v2.0alpha milestone Jul 7, 2019

tangq requested review from wlin7 and golaz July 7, 2019 23:51

rljacob assigned wlin7 Jul 8, 2019

tangq added 3 commits July 11, 2019 23:56

Merge branch 'tangq/atm/O3v2' of github.com:ACME-Climate/ACME into ta…

1448ecc

…ngq/atm/O3v2 Modify commit message

rljacob assigned singhbalwinder and unassigned wlin7 Aug 14, 2019

wlin7 reviewed Aug 26, 2019

View reviewed changes

tangq added 2 commits August 27, 2019 11:08

Update the namelist_definition to include default O3v2 parameter values

592f844

Turn off O3 drydep for 2000_cam5_av1c-04p2

031b28f

wlin7 approved these changes Aug 27, 2019

View reviewed changes

wlin7 merged commit 031b28f into master Aug 29, 2019

tangq deleted the tangq/atm/O3v2 branch December 8, 2019 05:07

rljacob mentioned this pull request Dec 21, 2021

Implement full interactive chemistry and Fast-J photolysis rates #4707

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Implement O3v2 in Linoz #3047

Implement O3v2 in Linoz #3047

tangq commented Jul 7, 2019

tangq commented Jul 12, 2019

rljacob commented Jul 25, 2019

tangq commented Jul 31, 2019

wlin7 Aug 26, 2019

tangq Aug 26, 2019

tangq Aug 27, 2019

wlin7 commented Aug 26, 2019

wlin7 commented Aug 26, 2019

tangq commented Aug 26, 2019

tangq commented Aug 27, 2019

wlin7 Aug 26, 2019

wlin7 commented Aug 27, 2019

rljacob commented Aug 27, 2019

wlin7 commented Aug 28, 2019

wlin7 commented Aug 29, 2019

wlin7 commented Sep 5, 2019

Implement O3v2 in Linoz #3047

Implement O3v2 in Linoz #3047

Conversation

tangq commented Jul 7, 2019

tangq commented Jul 12, 2019

rljacob commented Jul 25, 2019

tangq commented Jul 31, 2019

wlin7 Aug 26, 2019

Choose a reason for hiding this comment

tangq Aug 26, 2019

Choose a reason for hiding this comment

tangq Aug 27, 2019

Choose a reason for hiding this comment

wlin7 commented Aug 26, 2019

wlin7 commented Aug 26, 2019

tangq commented Aug 26, 2019

tangq commented Aug 27, 2019

wlin7 Aug 26, 2019

Choose a reason for hiding this comment

wlin7 commented Aug 27, 2019

rljacob commented Aug 27, 2019

wlin7 commented Aug 28, 2019

wlin7 commented Aug 29, 2019

wlin7 commented Sep 5, 2019