added model stop for ghg_input with nonsupported radiation options #1931
Conversation
Corrected a mis-spelled instance of "Climate"
|
@dudhia @weiwangncar |
|
Good question but also dangerous to change defaults. Need to modify the
tests.
…On Mon, Oct 30, 2023 at 1:31 PM Kelly Werner ***@***.***> wrote:
@dudhia <https://github.com/dudhia> @weiwangncar
<https://github.com/weiwangncar>
Some of the tests failed here because if we have the model stop if anyone
is using ghg_input and a physics scheme that doesn't work with it (e.g.,
goddard), it will fail. ghg_input is turned on, by default, in the
registry. Is that what we want?
—
Reply to this email directly, view it on GitHub
<#1931 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AEIZ77ABAI3OVHKBPDY4Y6TYB76CPAVCNFSM6AAAAAA6TNNKA6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBVHEYDMMZXGA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
@dudhia |
|
The problem is they won't notice and think they used it.
…On Mon, Oct 30, 2023 at 2:19 PM Kelly Werner ***@***.***> wrote:
@dudhia <https://github.com/dudhia>
So maybe something like "if using anything other than CAM, RRTM, RRTMG,
and RRTMG_fast, set ghg_input = 0" instead of stopping the model?
—
Reply to this email directly, view it on GitHub
<#1931 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AEIZ77D3DOBYEOOA7DWUOPDYCADW5AVCNFSM6AAAAAA6TNNKA6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBVHE3TCNRRGU>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
@kkeene44 Is it possible to add ghg_input = 0 to the namelist.input file for the failed runs in regression tests? |
|
@weiwangncar |
|
@weiwangncar |
|
Maybe you can add a blank space in a README?
…On Wed, Nov 1, 2023 at 2:56 PM Kelly Werner ***@***.***> wrote:
@weiwangncar <https://github.com/weiwangncar>
I've made the changes to the namelists in the regression tests. Is there a
way to manually resubmit the Jenkins tests without having to modify a file?
—
Reply to this email directly, view it on GitHub
<#1931 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AEIZ77EOY46KDAZGH5IRAPLYCKZP3AVCNFSM6AAAAAA6TNNKA6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBZGY3TENRXGI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
@kkeene44 Unfortunately you cannot start a new test without changing the code. README files do not count. |
|
Closing and opening a different PR #1936 so that reg tests will run. |
3 participants
TYPE: bug fix
KEYWORDS: ghg_input, radiation, module_check_a_mundo
SOURCE: internal
DESCRIPTION OF CHANGES:
Problem:
If a user had ghg_input turned on, but wasn't using one of the supported radiation physics options (i.e., not CAM, RRTM, RRTMG, or RRTMG_fast), the model would give an "ERROR" message, but wouldn't stop, nor would it make any changes to the settings. As long as everything else was correct, it would continue and write out the SUCCESS message at the end of the log files.
Solution:
Added a "count_fatal_error = count_fatal_error + 1" to the check for this in module_check_a_mundo.F.
LIST OF MODIFIED FILES:
M share/module_check_a_mundo.F
TESTS CONDUCTED:
Now, if non-supported radiation options are used with ghg_input, the model stops with the following message:
--- ERROR: ghg_input available only for these radiation schemes: CAM, RRTM, RRTMG, RRTMG_fast
And the LW and SW schemes must be reasonably paired together:
OK = CAM LW with CAM SW
OK = RRTM, RRTMG LW or SW, RRTMG_fast LW or SW may be mixed
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: LINE: 2794
NOTE: 1 namelist settings are wrong. Please check and reset these options
-------------------------------------------
Are the Jenkins tests all passing?
RELEASE NOTE: A bug fix was added to ensure the model stops when using ghg_input and a non-supported radiation physics option (i.e., anything other than CAM, RRTM, RRTMG, or RRTMG_fast).