Add WAM molecular diffusion #120
Conversation
…tmos_cubed_sphere into udhmhj.04 fix conflicts.
…ed_sphere into udhmhj.04 merge without conflict
resolve conflicts. Conflicts: driver/fvGFS/atmosphere.F90 model/dyn_core.F90 model/fv_arrays.F90 model/fv_control.F90 model/fv_dynamics.F90 model/fv_mapz.F90 model/nh_utils.F90 model/sw_core.F90 model/tp_core.F90 tools/fv_diagnostics.F90
|
I have not merged the latest GFDL dev/emc to my branch. I will do it soon.
…On Wed, Oct 6, 2021 at 11:25 AM Rusty Benson ***@***.***> wrote:
@XiaqiongZhou-NOAA <https://github.com/XiaqiongZhou-NOAA> - I believe we
are almost ready for this one. It looks like there is a conflict with the
makefile, which I think has been deleted.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#120 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ALQE7USC622GE2CL4FIZM3TUFRS5HANCNFSM5BCO5KVA>
.
Triage notifications on the go with GitHub Mobile for iOS
<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675>
or Android
<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.
|
|
OK. Glad to see that it works. |
|
@XiaqiongZhou-NOAA - thanks for the update. Let us know when the PRs are in for fv3atm and ufs-weather-model |
|
@junwang-noaa - can you please review this. |
|
@bensonr The PRs in fv3atm and ufs are planned to merge Tuesday (10/12) |
|
@bensonr @junwang-noaa |
|
Hi, Rusty. The existing ranges are explicitly designed for regular
lower-atmosphere modeling, while in the thermosphere winds can be
considerably stronger and temperatures both much colder and warmer. So
these need modification for range_warn to work in the WAM.
Lucas
…On Thu, Oct 14, 2021 at 9:41 AM Rusty Benson ***@***.***> wrote:
***@***.**** commented on this pull request.
Can you explain why the changes for the ranges are needed in
fv_diagnostics? I ask because it will cause merge issues in the future with
what's in the master branch.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#120 (review)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMUQRVESTWK3EV2RPGIRISTUG3M2DANCNFSM5BCO5KVA>
.
Triage notifications on the go with GitHub Mobile for iOS
<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675>
or Android
<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.
|
|
@bensonr @lharris4 The error message on DELL is as the following: |
|
Lucas - I understand the ranges and the reason for them. It's the logic
changes in the range variable definitions and the functional statements I'm
asking about.
vrange = (/xxx, yyy/) is now vrange(1) = xxx ; vrange(2) = yyy
…On Thu, Oct 14, 2021 at 9:48 AM lharris4 ***@***.***> wrote:
Hi, Rusty. The existing ranges are explicitly designed for regular
lower-atmosphere modeling, while in the thermosphere winds can be
considerably stronger and temperatures both much colder and warmer. So
these need modification for range_warn to work in the WAM.
Lucas
|
|
@XiaqiongZhou-NOAA - can you check a compile with a different version of gnu (preferably newer) to see if this is a bug in only one version of something that will be persistent. If it doesn't happen on the hera/gnu compiles, how much longer do we need to support the wcoss p3 dells? |
|
@bensonr This is not a bug for gnu/9.2.0 on Hera. All RTs can be compiled OK there. |
|
@XiaqiongZhou-NOAA - I am not a fan of a fix only needed for a specific version of gnu that seems to indicate a bug. But that being said, I'll let @lharris4 make the call on whether he wants to keep this mismatch or have the change brought back into all of the GFDL versions. |
|
It's the intel compiler on dell, we do not have any tests compiled with gnu
on dell.
…On Thu, Oct 14, 2021 at 10:50 AM Rusty Benson ***@***.***> wrote:
@XiaqiongZhou-NOAA <https://github.com/XiaqiongZhou-NOAA> - I am not a
fan of a fix only needed for a specific version of gnu that seems to
indicate a bug. But that being said, I'll let @lharris4
<https://github.com/lharris4> make the call on whether he wants to keep
this mismatch or have the change brought back into all of the GFDL versions.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#120 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AI7D6TP7CJVEH4MGL43QOO3UG3UZHANCNFSM5BCO5KVA>
.
Triage notifications on the go with GitHub Mobile for iOS
<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675>
or Android
<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.
|
|
I would prefer sticking with the syntax used for the original definition,
which should be standard Fortran.
Lucas
…On Thu, Oct 14, 2021 at 11:17 AM Jun Wang ***@***.***> wrote:
It's the intel compiler on dell, we do not have any tests compiled with gnu
on dell.
On Thu, Oct 14, 2021 at 10:50 AM Rusty Benson ***@***.***>
wrote:
> @XiaqiongZhou-NOAA <https://github.com/XiaqiongZhou-NOAA> - I am not a
> fan of a fix only needed for a specific version of gnu that seems to
> indicate a bug. But that being said, I'll let @lharris4
> <https://github.com/lharris4> make the call on whether he wants to keep
> this mismatch or have the change brought back into all of the GFDL
versions.
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <
#120 (comment)
>,
> or unsubscribe
> <
https://github.com/notifications/unsubscribe-auth/AI7D6TP7CJVEH4MGL43QOO3UG3UZHANCNFSM5BCO5KVA
>
> .
> Triage notifications on the go with GitHub Mobile for iOS
> <
https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675
>
> or Android
> <
https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub
>.
>
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#120 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMUQRVH7WNVL3VEPYHBBJLTUG3YA5ANCNFSM5BCO5KVA>
.
Triage notifications on the go with GitHub Mobile for iOS
<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675>
or Android
<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.
|
|
@junwang-noaa |
|
If we need the work-around to get it to work on the Dells then it is OK,
just put in a comment saying that this is the reason for the differing form.
Lucas
…On Fri, Oct 15, 2021 at 9:17 AM XiaqiongZhou-NOAA ***@***.***> wrote:
@junwang-noaa <https://github.com/junwang-noaa>
Please advice how to solve this issue. With the original definition with
some vrange = (/xxx, yyy/), I got "bus error" when running compiling jobs
on DELL_P3.
I do not know if there are other methods to solve instead of using
vrange(1) = xxx ; vrange(2) = yyy.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#120 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMUQRVEPV5BHOCDT5YVOTT3UHASW7ANCNFSM5BCO5KVA>
.
Triage notifications on the go with GitHub Mobile for iOS
<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675>
or Android
<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.
|
|
I will revise the range definition back and keep the original form. Another method will be used to fix the compiling issue on DELL: compiling fv_diagnostics.F90 with -O0 instead of -O2. |
…n be fixed by using -O0 instead of -O2 to compile fv_diagnostics.F90
|
@bensonr @laurenchilutti The ufs-weather-model RT passed on all the platforms except orion, where the jobs stayed in the queue the whole day and never got submitted. We decided to start committing the code. Would you please merge this PR? Thanks |
Description
Molecular diffusion developed by Henry Juang is added as a compiling option to replace Rayleigh Damping to configure FV3WAM with the model top reaching several hundreds km.
Fixes # (issue)
How Has This Been Tested?
It was tested on Hera with Intel compiler and will tested on other machines following UFS regression test rules.
Checklist:
Please check all whether they apply or not
** Dependences:
NCAR/ccpp-physics#712
NOAA-EMC/fv3atm#354
ufs-community/ufs-weather-model#714