adding updated RAS #585

SMoorthi-emc · 2021-03-04T00:50:05Z

This PR is to replace current RAS with an updated one that fixes a potential issue that can cause model crash.

Further changes:

In MG2 and MG3 the single if loop belowif (lamr(i,k) > qsmall .and. one/lamr(i,k) < Dcs) then is expanded to two nested ifs like

if (lamr(i,k) > qsmall) then
if (one/lamr(i,k) < Dcs) then

because on some computers the first if may lead to divide by zero leading to crash.

In radsw_main.F90, the code is slightly rewritten to avoid sqrt of a negative number and potential divide by zero.
These issues were encountered during debugging. After fixing these two issues runs haven't crashed.
reactivate 3d mass flux diagnostic tendencies in GFS_DCNV_generic_post (from @climbfuji).

Associated PRs:

#585
NOAA-EMC/fv3atm#252
ufs-community/ufs-weather-model#448

For regression testing, see ufs-community/ufs-weather-model#448

climbfuji

Looks good to me. This PR will be combined with an fv3atm PR from my side (for updating the submodule pointer for ccpp-physics and potential updates in GFS_typedefs.F90, if needed), and a ufs-weather-model PR that enables the fv3_rasmgshoc regression test.

climbfuji · 2021-03-04T02:19:08Z

physics/rascnv.F90

@@ -8,7 +8,7 @@ module rascnv
      implicit none
      public :: rascnv_init, rascnv_run, rascnv_finalize
      private
-      logical :: is_initialized = .False.
+      logical, save :: is_initialized = .False.


This should not be necessary, module variables are by default 'save' variables. But it doesn't hurt to have it. Same for all of the module variables below.

climbfuji · 2021-03-04T02:22:16Z

physics/rascnv.F90


              CNV_MFD(ipt,ll)   = CNV_MFD(ipt,ll)   + flx(ib)/dt

-!             CNV_DQLDT(ipt,ll) = CNV_DQLDT(ipt,ll)
-!    &                          + max(0.,(QLI(ib)+QII(ib)-qiid-qlid))/dt
+!!            CNV_DQLDT(ipt,ll) = CNV_DQLDT(ipt,ll)


Are these double exclamation marks !! intentional? They will probably be picked by doxygen when creating the scientific documentation (same a few lines below).

SMoorthi-emc · 2021-03-04T02:29:38Z

Dom, That double "!!" was intentional to the extent that it is different from the single "!". No particular significance. Moorthni

…

On Wed, Mar 3, 2021 at 9:25 PM Dom Heinzeller ***@***.***> wrote: ***@***.**** commented on this pull request. Looks good to me. This PR will be combined with an fv3atm PR from my side (for updating the submodule pointer for ccpp-physics and potential updates in GFS_typedefs.F90, if needed), and a ufs-weather-model PR that enables the fv3_rasmgshoc regression test. ------------------------------ In physics/rascnv.F90 <#585 (comment)>: > @@ -8,7 +8,7 @@ module rascnv implicit none public :: rascnv_init, rascnv_run, rascnv_finalize private - logical :: is_initialized = .False. + logical, save :: is_initialized = .False. This should not be necessary, module variables are by default 'save' variables. But it doesn't hurt to have it. Same for all of the module variables below. ------------------------------ In physics/rascnv.F90 <#585 (comment)>: > CNV_MFD(ipt,ll) = CNV_MFD(ipt,ll) + flx(ib)/dt -! CNV_DQLDT(ipt,ll) = CNV_DQLDT(ipt,ll) -! & + max(0.,(QLI(ib)+QII(ib)-qiid-qlid))/dt +!! CNV_DQLDT(ipt,ll) = CNV_DQLDT(ipt,ll) Are these double exclamation marks !! intentional? They will probably be picked by doxygen when creating the scientific documentation (same a few lines below). — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#585 (review)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALLVRYRZ4RH342JHAAZA24TTB3VQVANCNFSM4YSIKIKQ> .

-- Dr. Shrinivas Moorthi Research Meteorologist Modeling and Data Assimilation Branch Environmental Modeling Center / National Centers for Environmental Prediction 5830 University Research Court - (W/NP23), College Park MD 20740 USA Tel: (301)683-3718 e-mail: [email protected] Phone: (301) 683-3718 Fax: (301) 683-3718

climbfuji · 2021-03-04T02:30:59Z

Dom, That double "!!" was intentional to the extent that it is different from the single "!". No particular significance. Moorthni
…

Ok, thanks. @ligiabernardet as our doxygen expert, do you think these !! will be picked up mistakenly by doxygen and end up in the scientific documentation?

ligiabernardet · 2021-03-04T05:34:05Z

Dom, That double "!!" was intentional to the extent that it is different from the single "!". No particular significance. Moorthni
…

Ok, thanks. @ligiabernardet as our doxygen expert, do you think these !! will be picked up mistakenly by doxygen and end up in the scientific documentation?

@mzhangw Can you please check this?

mzhangw · 2021-03-05T19:59:21Z

It is OK with single line starting with "!!"

climbfuji

@SMoorthi-emc can you also say something about the MG microphysics changes and the radsw_main.F90 changes, please? The description of the PR says only two files are changed (RAS .F90 and .meta), plus what I added for the DCNV convective mass flux tendencies. Thank you!

climbfuji · 2021-03-15T22:51:16Z

@mjiacono there are some small changes in radsw_main.F90, do you want to take a look?

SMoorthi-emc · 2021-03-15T23:59:53Z

In MG2 and MG3 the single if loop below
"if (lamr(i,k) > qsmall .and. one/lamr(i,k) < Dcs) then" is expanded to two nested ifs like
"if (lamr(i,k) > qsmall) then
if (one/lamr(i,k) < Dcs) then".
this is done based on Dom's recommendation as on some computers the first if may lead to divide by zero leading to crash.
In the radsw_main.F90, the code is slightly rewritten to avoid sqrt of a negative number and potential divide by zero.
These issues were encountered during debugging. After fixing these two issues runs haven't crashed.

climbfuji · 2021-03-16T01:50:33Z

In MG2 and MG3 the single if loop below
"if (lamr(i,k) > qsmall .and. one/lamr(i,k) < Dcs) then" is expanded to two nested ifs like
"if (lamr(i,k) > qsmall) then
if (one/lamr(i,k) < Dcs) then".
this is done based on Dom's recommendation as on some computers the first if may lead to divide by zero leading to crash.
In the radsw_main.F90, the code is slightly rewritten to avoid sqrt of a negative number and potential divide by zero.
These issues were encountered during debugging. After fixing these two issues runs haven't crashed.

Great, thank you Moorthi. I'll copy & paste this into the description at the top.

mjiacono · 2021-03-16T15:02:58Z

Dom, The proposed changes to radsw_main.F90 look OK to me. Mike From: Dom Heinzeller ***@***.***> Reply-To: NCAR/ccpp-physics ***@***.***> Date: Monday, March 15, 2021 at 6:51 PM To: NCAR/ccpp-physics ***@***.***> Cc: "Iacono, Mike" ***@***.***>, Mention ***@***.***> Subject: Re: [NCAR/ccpp-physics] adding updated RAS (#585) Caution: This email originated outside of the organization @mjiacono<https://urldefense.com/v3/__https:/github.com/mjiacono__;!!CAFLEWIB!WrNJiyZ8s3SKcPOIqf4mYADNJ0j_sQXHbyyyPunR36-PU8zw6tQaOWR90j4$> there are some small changes in radsw_main.F90, do you want to take a look? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<https://urldefense.com/v3/__https:/github.com/NCAR/ccpp-physics/pull/585*issuecomment-799807013__;Iw!!CAFLEWIB!WrNJiyZ8s3SKcPOIqf4mYADNJ0j_sQXHbyyyPunR36-PU8zw6tQadTCVbWo$>, or unsubscribe<https://urldefense.com/v3/__https:/github.com/notifications/unsubscribe-auth/AE7W55D3WMIH35QAA6V4HTTTD2FPJANCNFSM4YSIKIKQ__;!!CAFLEWIB!WrNJiyZ8s3SKcPOIqf4mYADNJ0j_sQXHbyyyPunR36-PU8zw6tQaEaOv7Js$>.

…

________________________________ This email is intended solely for the recipient. It may contain privileged, proprietary or confidential information or material. If you are not the intended recipient, please delete this email and any attachments and notify the sender of the error.

climbfuji · 2021-03-16T17:13:45Z

Dom, The proposed changes to radsw_main.F90 look OK to me. Mike From: Dom Heinzeller @.> Reply-To: NCAR/ccpp-physics @.> Date: Monday, March 15, 2021 at 6:51 PM To: NCAR/ccpp-physics @.> Cc: "Iacono, Mike" @.>, Mention @.***> Subject: Re: [NCAR/ccpp-physics] adding updated RAS (#585) Caution: This email originated outside of the organization @mjiacono<https://urldefense.com/v3/https:/github.com/mjiacono;!!CAFLEWIB!WrNJiyZ8s3SKcPOIqf4mYADNJ0j_sQXHbyyyPunR36-PU8zw6tQaOWR90j4$> there are some small changes in radsw_main.F90, do you want to take a look? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<https://urldefense.com/v3/https:/github.com/NCAR/ccpp-physics/pull/585*issuecomment-799807013;Iw!!CAFLEWIB!WrNJiyZ8s3SKcPOIqf4mYADNJ0j_sQXHbyyyPunR36-PU8zw6tQadTCVbWo$>, or unsubscribe<https://urldefense.com/v3/https:/github.com/notifications/unsubscribe-auth/AE7W55D3WMIH35QAA6V4HTTTD2FPJANCNFSM4YSIKIKQ;!!CAFLEWIB!WrNJiyZ8s3SKcPOIqf4mYADNJ0j_sQXHbyyyPunR36-PU8zw6tQaEaOv7Js$>.
…

Thanks, Mike!

SMoorthi-emc · 2021-03-17T23:15:36Z

Thanks Dom.

…

On Wed, Mar 17, 2021 at 7:07 PM Dom Heinzeller ***@***.***> wrote: Merged #585 <#585> into master. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#585 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALLVRYXNMYGKVAFDGODWK5LTEEYZTANCNFSM4YSIKIKQ> .

-- Dr. Shrinivas Moorthi Research Meteorologist Modeling and Data Assimilation Branch Environmental Modeling Center / National Centers for Environmental Prediction 5830 University Research Court - (W/NP23), College Park MD 20740 USA Tel: (301)683-3718 e-mail: ***@***.*** Phone: (301) 683-3718 Fax: (301) 683-3718

adding updated RAS

67c46a8

SMoorthi-emc requested review from climbfuji, grantfirl, JulieSchramm and llpcarson as code owners March 4, 2021 00:50

climbfuji reviewed Mar 4, 2021

View reviewed changes

This was referenced Mar 4, 2021

Update to RAS in CCPP, update submodule pointer for ccpp-framework (CI tests) NOAA-EMC/fv3atm#252

Merged

Final step to remove IPD (remove _ccpp labels from regression tests), RAS updates and regression tests ufs-community/ufs-weather-model#448

Merged

updating MGx routines and SW main for some potential issues and fixes

8b947ba

SMoorthi-emc and others added 3 commits March 7, 2021 21:28

fixing radsw_main

a483369

updating RAS with some minor modifications

51a6327

Turn on mass flux diagnostics

8d3ed1d

climbfuji reviewed Mar 15, 2021

View reviewed changes

climbfuji requested a review from junwang-noaa March 15, 2021 22:57

junwang-noaa approved these changes Mar 16, 2021

View reviewed changes

climbfuji mentioned this pull request Mar 16, 2021

Removed unused/old 3D diagnostic tendencies from GFS_typedefs.F90 and CCPP-physics GFS_DCNV_generic ? NOAA-EMC/fv3atm#235

Closed

climbfuji approved these changes Mar 16, 2021

View reviewed changes

climbfuji merged commit 7242a6d into NCAR:master Mar 17, 2021

grantfirl mentioned this pull request Apr 15, 2021

Updates from v2 #626

Merged

HelinWei-NOAA pushed a commit to HelinWei-NOAA/ccpp-physics that referenced this pull request Feb 26, 2023

Save routehandles in fv3cap initialize phase (NCAR#585)

a01627f

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

adding updated RAS #585

adding updated RAS #585

SMoorthi-emc commented Mar 4, 2021 •

edited by climbfuji

Loading

climbfuji left a comment

climbfuji Mar 4, 2021

climbfuji Mar 4, 2021

SMoorthi-emc commented Mar 4, 2021 via email

climbfuji commented Mar 4, 2021

ligiabernardet commented Mar 4, 2021

mzhangw commented Mar 5, 2021

climbfuji left a comment

climbfuji commented Mar 15, 2021

SMoorthi-emc commented Mar 15, 2021

climbfuji commented Mar 16, 2021

mjiacono commented Mar 16, 2021 via email

climbfuji commented Mar 16, 2021

SMoorthi-emc commented Mar 17, 2021 via email

adding updated RAS #585

adding updated RAS #585

Conversation

SMoorthi-emc commented Mar 4, 2021 • edited by climbfuji Loading

climbfuji left a comment

Choose a reason for hiding this comment

climbfuji Mar 4, 2021

Choose a reason for hiding this comment

climbfuji Mar 4, 2021

Choose a reason for hiding this comment

SMoorthi-emc commented Mar 4, 2021 via email

climbfuji commented Mar 4, 2021

ligiabernardet commented Mar 4, 2021

mzhangw commented Mar 5, 2021

climbfuji left a comment

Choose a reason for hiding this comment

climbfuji commented Mar 15, 2021

SMoorthi-emc commented Mar 15, 2021

climbfuji commented Mar 16, 2021

mjiacono commented Mar 16, 2021 via email

climbfuji commented Mar 16, 2021

SMoorthi-emc commented Mar 17, 2021 via email

SMoorthi-emc commented Mar 4, 2021 •

edited by climbfuji

Loading