Move NoahMP to its own submodule #551
Comments
|
Will the update of Noah-MP in the ccpp-physics repository be used for the
CCPP v5 release and the associated UFS SRW App v1.0 release (code freeze
January 22)? This is getting very close to the deadline to allow thorough
testing.
…On Mon, Jan 11, 2021 at 11:54 AM grantfirl ***@***.***> wrote:
@DomHeinzeller <https://github.com/DomHeinzeller> @ligiabernardet
<https://github.com/ligiabernardet> @barlage <https://github.com/barlage>
@ncarchen <https://github.com/ncarchen>
This is a placeholder issue for setting up the NoahMP LSM as its own
submodule within the ccpp-physics repository.
According to @barlage <https://github.com/barlage> via the AMS 2021
meeting, @ncarchen <https://github.com/ncarchen> and @barlage
<https://github.com/barlage> have already begun a GitHub repository to
contain the unified (between WRF and the NWM) NoahMP LSM:
https://github.com/NCAR/noahmp
This should be what ends up as the ccpp-physics submodule. Depending on
what the owners of the unified NoahMP repository want to do in terms of the
internal NoahMP code, it may be necessary to follow the route of RRTMGP and
have only the CCPP interface to NoahMP exist in the ccpp-physics
repository, *OR*, if all users/developers of the code within the unified
NoahMP repository will be using CCPP-compliant hosts, it would be possible
to avoid the extra software layer and make the code within the unified
NoahMP repository CCPP-compliant.
In the meantime, @barlage <https://github.com/barlage> and the land team
at EMC have plans to update the NoahMP version in the ccpp-physics
repository to match a snapshot of the code in the unified NoahMP
repository, since what exists in the ccpp-physics today "is equivalent to
WRF v3.6 so probably 5-6 years old likely" (from personal communication
with @barlage <https://github.com/barlage>).
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#551>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AE7WQAV2JUUQF4LSN2XGEGLSZNCNBANCNFSM4V57WSWQ>
.
|
|
@ligiabernardet I don't have any details on the timeline. Mike Barlage mentioned that there was probably not time to do this for the upcoming release. I put an issue here just so that we don't forget about it and can reference that the issue has been closed when completed. |
|
Sounds good. In talking with SRW App release code lead (Jeff Beck) he said
this is not needed for the SRW App v1.0 release.
So we are all on the page that this change will not be used for the CCPP v5
release.
…On Tue, Jan 12, 2021 at 9:15 AM grantfirl ***@***.***> wrote:
@ligiabernardet <https://github.com/ligiabernardet> I don't have any
details on the timeline. Mike Barlage mentioned that there was probably not
time to do this for the upcoming release. I put an issue here just so that
we don't forget about it and can reference that the issue has been closed
when completed.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#551 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AE7WQAQNZAOEY6YKYJPP3TDSZRYQHANCNFSM4V57WSWQ>
.
|
|
I am going to talk with Mike tomorrow. I think we should include those Noah MP bug fixes to CCPP v5 release. The code is under my fork and has been tested. My branch is several commits behind and I will synchronize with the master today. So they will be ready to be merged if the decision is made. |
|
It is always good to have bug fixes. However, note that we have limited
time for further testing of CCPP v5 and note that this will impact the code
distributed with the UFS Short-Range App (attn @jeff Beck - NOAA Affiliate
<[email protected]>). They both have code freezes this week. Can you
please tell us the nature of these bug fixes?
…On Mon, Jan 18, 2021 at 8:33 AM HelinWei-NOAA ***@***.***> wrote:
I am going to talk with Mike tomorrow. I think we should include those
Noah MP bug fixes to CCPP v5 release. The code is under my fork and has
been tested. My branch is several commits behind and I will synchronize
with the master today. So they will be ready to be merged if the decision
is made.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#551 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AE7WQAU5MRPQIKM33BVZO23S2RIFVANCNFSM4V57WSWQ>
.
|
|
Is the UFS Short-Range App configured to use NOAH MP as the default in CCPP
v5 release ? If not, I think the bug fix can go in at a later date with
other NOAH MP updates -- Fanglin
On Mon, Jan 18, 2021 at 6:16 PM ligiabernardet <[email protected]>
wrote:
… It is always good to have bug fixes. However, note that we have limited
time for further testing of CCPP v5 and note that this will impact the code
distributed with the UFS Short-Range App (attn @jeff Beck - NOAA Affiliate
***@***.***>). They both have code freezes this week. Can you
please tell us the nature of these bug fixes?
On Mon, Jan 18, 2021 at 8:33 AM HelinWei-NOAA ***@***.***>
wrote:
> I am going to talk with Mike tomorrow. I think we should include those
> Noah MP bug fixes to CCPP v5 release. The code is under my fork and has
> been tested. My branch is several commits behind and I will synchronize
> with the master today. So they will be ready to be merged if the decision
> is made.
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#551 (comment)
>,
> or unsubscribe
> <
https://github.com/notifications/unsubscribe-auth/AE7WQAU5MRPQIKM33BVZO23S2RIFVANCNFSM4V57WSWQ
>
> .
>
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#551 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AKY5N2OHARX6YRG6NITUJSDS2S6K5ANCNFSM4V57WSWQ>
.
--
*Fanglin Yang, Ph.D.*
*Chief, Model Physics Group*
*Modeling and Data Assimilation Branch*
*NOAA/NWS/NCEP Environmental Modeling Center*
*https://www.emc.ncep.noaa.gov/gmb/wx24fy/fyang/
<https://www.emc.ncep.noaa.gov/gmb/wx24fy/fyang/>*
|
|
The UFS SRW App will be distributed with two CCPP suites: GFSv15p2 (operational GFS) and RRFS_v1alpha (which uses NoahMP). Suite RRFS_v1alpha, with NoahMP, is very important for the UFS SRW App release. |
|
The bug fixes includes:
(1) Some updates to prevent the model from crashing. The Noah MP is running
in single precision currently and it has caused a significant number of
crashing cases from our tests. Almost all crashing cases were gone after we
used double precision for Noah MP like other physics.
(2) Noah MP uses its own albedo and emissivity, which are different from
those used in the surface radiation calculations outside Noah MP.
(3) The wrong downward longwave forcing to Noah MP. Noah MP expects the
downward longwave before emission, however the current driver passed the
one after emission. Inside Noah MP it is multiplied by surface emissivity
again. So generally Noah MP is getting less downward longwave.
For (1) and (3), they are straightforward and we only need to modify the
code under CCPP physics. For (2), we need to make some modifications in the
other parts of the model. I am waiting for them to remove IPD completely
then commit my change.
Helin
On Mon, Jan 18, 2021 at 6:16 PM ligiabernardet <[email protected]>
wrote:
… It is always good to have bug fixes. However, note that we have limited
time for further testing of CCPP v5 and note that this will impact the code
distributed with the UFS Short-Range App (attn @jeff Beck - NOAA Affiliate
***@***.***>). They both have code freezes this week. Can you
please tell us the nature of these bug fixes?
On Mon, Jan 18, 2021 at 8:33 AM HelinWei-NOAA ***@***.***>
wrote:
> I am going to talk with Mike tomorrow. I think we should include those
> Noah MP bug fixes to CCPP v5 release. The code is under my fork and has
> been tested. My branch is several commits behind and I will synchronize
> with the master today. So they will be ready to be merged if the decision
> is made.
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#551 (comment)
>,
> or unsubscribe
> <
https://github.com/notifications/unsubscribe-auth/AE7WQAU5MRPQIKM33BVZO23S2RIFVANCNFSM4V57WSWQ
>
> .
>
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#551 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ALPHKYEMV5UJWBHTHTES7QLS2S6K3ANCNFSM4V57WSWQ>
.
|
|
@HelinWei-NOAA Thanks for providing information about the nature of the fixes. I am looking forward to seeing a PR. |
|
@ligiabernardet, yes, we will definitely need to run a few tests with the RRFS_v1alpha SDF within the SRW App (all three pre-defined domains with varying ICs/LBCs) to ensure these changes don't cause any unexpected results. |
|
I feel like maybe I missed some discussion along the way since this took a turn to bug fixing Noah-MP rather than the issue of moving Noah-MP to a submodule. Should we have a new issue that can be referenced by Helin's bug-fix PR? |
|
Yes, pls start a new issue
…On Tue, Jan 19, 2021 at 7:34 PM Michael Barlage ***@***.***> wrote:
I feel like maybe I missed some discussion along the way since this took a
turn to bug fixing Noah-MP rather than the issue of moving Noah-MP to a
submodule. Should we have a new issue that can be referenced by the PR?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#551 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AE7WQAVHTD5ZZI4IUABWFIDS2Y6LTANCNFSM4V57WSWQ>
.
|
@DomHeinzeller @ligiabernardet @barlage @ncarchen
This is a placeholder issue for setting up the NoahMP LSM as its own submodule within the ccpp-physics repository.
According to @barlage via the AMS 2021 meeting, @ncarchen and @barlage have already begun a GitHub repository to contain the unified (between WRF and the NWM) NoahMP LSM: https://github.com/NCAR/noahmp
This should be what ends up as the ccpp-physics submodule. Depending on what the owners of the unified NoahMP repository want to do in terms of the internal NoahMP code, it may be necessary to follow the route of RRTMGP and have only the CCPP interface to NoahMP exist in the ccpp-physics repository, OR, if all users/developers of the code within the unified NoahMP repository will be using CCPP-compliant hosts, it would be possible to avoid the extra software layer and make the code within the unified NoahMP repository CCPP-compliant.
In the meantime, @barlage and the land team at EMC have plans to update the NoahMP version in the ccpp-physics repository to match a snapshot of the code in the unified NoahMP repository, since what exists in the ccpp-physics today "is equivalent to WRF v3.6 so probably 5-6 years old likely" (from personal communication with @barlage).
The text was updated successfully, but these errors were encountered: