ARW Trajectory Diagnostics #84
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
KEYWORDS: trajectory SOURCE: internal DESCRIPTION OF CHANGES: The original "trajectory" code in routine module_em.F only computed the trajectory spatial coordinates per trajectory, per time step. The new code, in share/module_trajectory.F, takes trajectory coordinate values and linearly interpolates in x,y, and z to get the value of up 100 variables per trajectory per time step. The code will detect and handle trajectory coordinate values that are between the grid parcels. The user may have up to 1000 trajectories per domain and as mentioned up to 100 variables per trajectory. LIST OF MODIFIED FILES: M Registry/Registry.EM_COMMON M Registry/Registry.NMM M chem/chem_driver.F M chem/depend.chem M dyn_em/module_em.F M dyn_em/start_em.F M external/RSL_LITE/module_dm.F M main/depend.common M share/Makefile M share/module_check_a_mundo.F A share/module_trajectory.F M share/solve_interface.F TESTS CONDUCTED: WTF reg test, version 3.04, has been successfully completed. The code has been tested in ACOM by Mary Barth, Megan Bella, and Gabrielle Pfister.
davegill
reviewed
Dec 27, 2016
| @@ -438,6 +438,7 @@ module_sf_clm.o: module_cam_shr_kind_mod.o \ | |||
| module_cam_shr_const_mod.o \ | |||
| module_cam_support.o \ | |||
| module_sf_urban.o \ | |||
| module_ra_gfdleta.o \ | |||
There was a problem hiding this comment.
Why is there a new dependency on module_ra_gfdleta.o?
davegill
reviewed
Dec 27, 2016
| @@ -606,6 +606,13 @@ SUBROUTINE check_nml_consistency | |||
| model_config_rec%num_traj = 0 | |||
| END IF | |||
|
|
|||
| #elif( NMM_CORE == 1 ) | |||
There was a problem hiding this comment.
@jamiebresch
Do we need a similar ifdef for 3dvar in check_a_mundo?
There was a problem hiding this comment.
Can a "#endif" added before line 597 to conclude the EM_CORE test?
Then, start another ifdef for the trajectory thing?
if NMM_CORE == 1 or DA_CORE==1, no trajectory.
There was a problem hiding this comment.
@jamiebresch
Has this ifdef issue in check_a_mundo been satisfactorially addressed? It seems like you wanted something like this:
#endif
#if ((EM_CORE == 1) && (DA_CORE != 1))
!----------------------------------------------------------------------------
! If trajectory option is turned off, make sure the number of trajectories is
! zero.
!----------------------------------------------------------------------------
IF ( ( model_config_rec%traj_opt .EQ. 0 ) .AND. &
( model_config_rec%num_traj .NE. 0 ) ) THEN
WRITE (wrf_err_message, FMT='(A,A)') '--- WARNING: traj_opt is zero, but ', &
'num_traj is not zero; setting num_traj to zero.'
CALL wrf_message ( wrf_err_message )
model_config_rec%num_traj = 0
END IF
#else
!----------------------------------------------------------------------------
! If NMM core or DA, and trajectories are on, then halt.
!----------------------------------------------------------------------------
IF ( model_config_rec%traj_opt /= 0 ) THEN
call wrf_error_fatal( 'Trajectories not supported in NMM or DA' )
END IF
#endif
There was a problem hiding this comment.
When I tested it some days ago, 3DVAR and 4DVAR both ran fine. DA doesn't care about the traj_opt setting.
I just tested 4DVAR with stacy's latest TRJ branch but with an WRFPLUS that doesn't yet have the trajectory mods in start_em.F. It still runs fine with traj_opt=1.
The current TRJ branch looks OK.
|
Stacy,
|
|
Dave,
Would it be useful if I sent you an early draft documentation regarding the
trajectory code?
Stacy
…On Tue, Dec 27, 2016 at 4:34 PM, Dave Gill ***@***.***> wrote:
Stacy,
1. There are no new variables in the Registry.EM_COMMON. If the output
is in a netcdf file, where are these variables defined?
2. Give an example of how you turn the option on/off, control which
variables, starting and stopping time, initial locations, periodically
follow same starting (i,j,k), etc.
3. Please add a sample of what the netcdf -h returns for the new file.
4. What happens to a trajectory during DFI (when the model runs
backwards)?
5. How about restarts? Is there any bit-for-bit results that you guys
have manufactured, where a long-running long used trajectories?
6. Are you able to get bit-wise identical answers using different core
counts with MPI? Are you able to use OpenMP with multiple cores and get the
same bit-identical answers as with a single processor?
7. Since this is an additional output file and is not "gridded", how
do you use this new file to work with trajectories?
8. This generic info on trajectories could be put in a top-level
README.trajectory, but let's wait for the rest of the WRF Dev to weigh-in
on that idea (which can be added later, also).
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAki8vxsAA-VajfMna6dsWtMamOg2Fks5rMaCAgaJpZM4LWjrO>
.
|
|
Jamie,
Hate to say it but I don't know much about 3dvar. I only concentrated on
the EM and NMM cores
so that the code would pass the reg tests. All our testing has been with
em_real.
Stacy
…On Tue, Dec 27, 2016 at 4:19 PM, Dave Gill ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In share/module_check_a_mundo.F
<#84 (review)>:
> @@ -606,6 +606,13 @@ SUBROUTINE check_nml_consistency
model_config_rec%num_traj = 0
END IF
+#elif( NMM_CORE == 1 )
@jamiebresch <https://github.com/jamiebresch>
Do we need a similar ifdef for 3dvar in check_a_mundo?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#84 (review)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkrfTv5VPI20LXtnyY33Hd5mMtL-Yks5rMZz4gaJpZM4LWjrO>
.
|
|
Dave,
I added this dependency for module_sf_clm.o because phys/module_sf_clm.F
use associates
module_ra_gfdleta at line # 3919. Without this dependency you are playing
roulette with multi
thread compilations.
Stacy
…On Tue, Dec 27, 2016 at 4:17 PM, Dave Gill ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In main/depend.common
<#84 (review)>:
> @@ -438,6 +438,7 @@ module_sf_clm.o: module_cam_shr_kind_mod.o \
module_cam_shr_const_mod.o \
module_cam_support.o \
module_sf_urban.o \
+ module_ra_gfdleta.o \
Why is there a new dependency on module_ra_gfdleta.o?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#84 (review)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkgd5NXf5jhCN4_bFgwp6ti9ImUArks5rMZykgaJpZM4LWjrO>
.
|
|
|
Stacy,
Great job tracking down busted dependencies, but no good deed goes
unpunished.
We will want that single line pulled out and put in as a separate PR.
Dave
On Tue, Dec 27, 2016 at 4:45 PM, stacywalters <notifications@github.com>
wrote:
… Dave,
I added this dependency for module_sf_clm.o because phys/module_sf_clm.F
use associates
module_ra_gfdleta at line # 3919. Without this dependency you are playing
roulette with multi
thread compilations.
Stacy
On Tue, Dec 27, 2016 at 4:17 PM, Dave Gill ***@***.***>
wrote:
> ***@***.**** commented on this pull request.
> ------------------------------
>
> In main/depend.common
> <#84 (review)>:
>
> > @@ -438,6 +438,7 @@ module_sf_clm.o: module_cam_shr_kind_mod.o \
> module_cam_shr_const_mod.o \
> module_cam_support.o \
> module_sf_urban.o \
> + module_ra_gfdleta.o \
>
> Why is there a new dependency on module_ra_gfdleta.o?
>
> —
> You are receiving this because you authored the thread.
> Reply to this email directly, view it on GitHub
> <#84 (review)>,
or mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/AIHAkgd5NXf5jhCN4_
bFgwp6ti9ImUArks5rMZykgaJpZM4LWjrO>
> .
>
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AMFFeg_2fEVWP6vem5Pz6c9q2Incdmgtks5rMaMugaJpZM4LWjrO>
.
|
|
Dave,
Does that mean I need to go through the process to generate a pull request
again?
Stacy
…On Tue, Dec 27, 2016 at 4:57 PM, Dave Gill ***@***.***> wrote:
Stacy,
Great job tracking down busted dependencies, but no good deed goes
unpunished.
We will want that single line pulled out and put in as a separate PR.
Dave
On Tue, Dec 27, 2016 at 4:45 PM, stacywalters ***@***.***>
wrote:
> Dave,
>
> I added this dependency for module_sf_clm.o because phys/module_sf_clm.F
> use associates
> module_ra_gfdleta at line # 3919. Without this dependency you are playing
> roulette with multi
> thread compilations.
>
> Stacy
>
> On Tue, Dec 27, 2016 at 4:17 PM, Dave Gill ***@***.***>
> wrote:
>
> > ***@***.**** commented on this pull request.
> > ------------------------------
> >
> > In main/depend.common
> > <#84 (review)>:
> >
> > > @@ -438,6 +438,7 @@ module_sf_clm.o: module_cam_shr_kind_mod.o \
> > module_cam_shr_const_mod.o \
> > module_cam_support.o \
> > module_sf_urban.o \
> > + module_ra_gfdleta.o \
> >
> > Why is there a new dependency on module_ra_gfdleta.o?
> >
> > —
> > You are receiving this because you authored the thread.
> > Reply to this email directly, view it on GitHub
> > <#84 (review)>,
> or mute
> > the thread
> > <https://github.com/notifications/unsubscribe-auth/AIHAkgd5NXf5jhCN4_
> bFgwp6ti9ImUArks5rMZykgaJpZM4LWjrO>
>
> > .
> >
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#84 (comment)>, or
mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/AMFFeg_
2fEVWP6vem5Pz6c9q2Incdmgtks5rMaMugaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkqSki6UDBBzcb2YYjdAzSBV7dSf_ks5rMaXxgaJpZM4LWjrO>
.
|
|
Jamie,
The variable:
no_trajectory
is defined in frame/module_state_description.F (line # 180 ).
Stacy
…On Tue, Dec 27, 2016 at 4:47 PM, Jamie Bresch ***@***.***> wrote:
@stacywalters <https://github.com/stacywalters> @davegill
<https://github.com/davegill>
1. Is the variable no_trajectory defined?
2. There might be various places that if DA_CORE!=1 needs to be added.
I'll need some time to figure it out.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAksS83ooUZlsRmynjfWLEIpFS56vlks5rMaOcgaJpZM4LWjrO>
.
|
|
Stacy,
Yes and no.
No: On your local sand box clone, in the TRJ branch, revert the mod to the
file (depend.common). Then, push the code back up to github.
Yes: From your master branch, cut over an new branch (DEPENDENCY_FIX). Put
that one-liner in there, then: add, commit, push. You'll have two PRs, one
that is very close to the first, and one that is for the bug-fix only.
Dave
On Tue, Dec 27, 2016 at 5:03 PM, stacywalters <notifications@github.com>
wrote:
… Dave,
Does that mean I need to go through the process to generate a pull request
again?
Stacy
On Tue, Dec 27, 2016 at 4:57 PM, Dave Gill ***@***.***>
wrote:
> Stacy,
>
> Great job tracking down busted dependencies, but no good deed goes
> unpunished.
>
> We will want that single line pulled out and put in as a separate PR.
>
> Dave
>
> On Tue, Dec 27, 2016 at 4:45 PM, stacywalters ***@***.***>
> wrote:
>
> > Dave,
> >
> > I added this dependency for module_sf_clm.o because
phys/module_sf_clm.F
> > use associates
> > module_ra_gfdleta at line # 3919. Without this dependency you are
playing
> > roulette with multi
> > thread compilations.
> >
> > Stacy
> >
> > On Tue, Dec 27, 2016 at 4:17 PM, Dave Gill ***@***.***>
> > wrote:
> >
> > > ***@***.**** commented on this pull request.
> > > ------------------------------
> > >
> > > In main/depend.common
> > > <#84 (review)
>:
> > >
> > > > @@ -438,6 +438,7 @@ module_sf_clm.o: module_cam_shr_kind_mod.o \
> > > module_cam_shr_const_mod.o \
> > > module_cam_support.o \
> > > module_sf_urban.o \
> > > + module_ra_gfdleta.o \
> > >
> > > Why is there a new dependency on module_ra_gfdleta.o?
> > >
> > > —
> > > You are receiving this because you authored the thread.
> > > Reply to this email directly, view it on GitHub
> > > <#84 (review)
>,
> > or mute
> > > the thread
> > > <https://github.com/notifications/unsubscribe-
auth/AIHAkgd5NXf5jhCN4_
> > bFgwp6ti9ImUArks5rMZykgaJpZM4LWjrO>
> >
> > > .
> > >
> >
> > —
> > You are receiving this because you commented.
> > Reply to this email directly, view it on GitHub
> > <#84 (comment)>, or
> mute
> > the thread
> > <https://github.com/notifications/unsubscribe-auth/AMFFeg_
> 2fEVWP6vem5Pz6c9q2Incdmgtks5rMaMugaJpZM4LWjrO>
> > .
> >
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#84 (comment)>, or
mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/
AIHAkqSki6UDBBzcb2YYjdAzSBV7dSf_ks5rMaXxgaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AMFFev2Ypno4UF6EdceKx_lwbLiyCjmlks5rMadKgaJpZM4LWjrO>
.
|
|
@stacywalters Oh, right. I was looking at WRFDA's version of frame/module_state_description.F which is different from ARW. |
|
Dave,
Quit a response.
(1) Indeed there are no "new" variables in Registry.EM_COMMON; just a
redefinition of num_traj.
(2) Hope the word document I sent you helps here.
(3) can easily do
(4) I must admit my ignorance of the base WRF is showing. I had no idea
that one could run
WRF in "reverse time". But if the trajectory subroutine in the file
dyn_em/module_em.F
can handle reverse time then the trajectory code should work. But
I'd feel far more confident
if you have a simple DFI case around to test.
(5) restarts work and should generate bit for bit results assuming the code
without trajectories enabled
generates bit for bit results.
(6) haven't tried differing MPI counts on the same code. This should be
hard to verify. We never run
our chemistry simulations with OpenMP turned on. I can run a pure
WRF simulation and test this
out.
I have a question here; are you talking about "pure" OMP or MPI + OMP
or both?
(7) Not quite sure what you are driving at here.
Stacy
…On Tue, Dec 27, 2016 at 4:34 PM, Dave Gill ***@***.***> wrote:
Stacy,
1. There are no new variables in the Registry.EM_COMMON. If the output
is in a netcdf file, where are these variables defined?
2. Give an example of how you turn the option on/off, control which
variables, starting and stopping time, initial locations, periodically
follow same starting (i,j,k), etc.
3. Please add a sample of what the netcdf -h returns for the new file.
4. What happens to a trajectory during DFI (when the model runs
backwards)?
5. How about restarts? Is there any bit-for-bit results that you guys
have manufactured, where a long-running long used trajectories?
6. Are you able to get bit-wise identical answers using different core
counts with MPI? Are you able to use OpenMP with multiple cores and get the
same bit-identical answers as with a single processor?
7. Since this is an additional output file and is not "gridded", how
do you use this new file to work with trajectories?
8. This generic info on trajectories could be put in a top-level
README.trajectory, but let's wait for the rest of the WRF Dev to weigh-in
on that idea (which can be added later, also).
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAki8vxsAA-VajfMna6dsWtMamOg2Fks5rMaCAgaJpZM4LWjrO>
.
|
|
@stacywalters From my limited tests, 3DVAR/4DVAR/WRFPLUS all ran fine with your TRJ branch code as it is. |
|
Jamie,
Many thanks for the testing.
Stacy
…On Tue, Dec 27, 2016 at 6:17 PM, Jamie Bresch ***@***.***> wrote:
@stacywalters <https://github.com/stacywalters> From my limited tests,
3DVAR/4DVAR/WRFPLUS all ran fine with your TRJ branch code as it is.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkg-RYYEagD5P-GWRpw3snTNY4pF6ks5rMbicgaJpZM4LWjrO>
.
|
|
Dave,
Sorry to be so cautious on this but can you verify the following steps:
No: On your local sand box clone, in the TRJ branch, revert the mod to the
file (depend.common). Then, push the code back up to github.
(1) git reset HEAD main/depend.common
Yes: From your master branch, cut over an new branch (DEPENDENCY_FIX). Put
that one-liner in there, then: add, commit, push. You'll have two PRs, one
that is very close to the first, and one that is for the bug-fix only.
(1) git checkout -b DEPENDENCY_FIX
Thanks, Stacy.
…On Tue, Dec 27, 2016 at 5:15 PM, Dave Gill ***@***.***> wrote:
Stacy,
Yes and no.
No: On your local sand box clone, in the TRJ branch, revert the mod to the
file (depend.common). Then, push the code back up to github.
Yes: From your master branch, cut over an new branch (DEPENDENCY_FIX). Put
that one-liner in there, then: add, commit, push. You'll have two PRs, one
that is very close to the first, and one that is for the bug-fix only.
Dave
On Tue, Dec 27, 2016 at 5:03 PM, stacywalters ***@***.***>
wrote:
> Dave,
>
> Does that mean I need to go through the process to generate a pull
request
> again?
>
> Stacy
>
>
> On Tue, Dec 27, 2016 at 4:57 PM, Dave Gill ***@***.***>
> wrote:
>
> > Stacy,
> >
> > Great job tracking down busted dependencies, but no good deed goes
> > unpunished.
> >
> > We will want that single line pulled out and put in as a separate PR.
> >
> > Dave
> >
> > On Tue, Dec 27, 2016 at 4:45 PM, stacywalters <
***@***.***>
> > wrote:
> >
> > > Dave,
> > >
> > > I added this dependency for module_sf_clm.o because
> phys/module_sf_clm.F
> > > use associates
> > > module_ra_gfdleta at line # 3919. Without this dependency you are
> playing
> > > roulette with multi
> > > thread compilations.
> > >
> > > Stacy
> > >
> > > On Tue, Dec 27, 2016 at 4:17 PM, Dave Gill ***@***.***
>
> > > wrote:
> > >
> > > > ***@***.**** commented on this pull request.
> > > > ------------------------------
> > > >
> > > > In main/depend.common
> > > > <#84 (review)-
14510503
> >:
> > > >
> > > > > @@ -438,6 +438,7 @@ module_sf_clm.o: module_cam_shr_kind_mod.o \
> > > > module_cam_shr_const_mod.o \
> > > > module_cam_support.o \
> > > > module_sf_urban.o \
> > > > + module_ra_gfdleta.o \
> > > >
> > > > Why is there a new dependency on module_ra_gfdleta.o?
> > > >
> > > > —
> > > > You are receiving this because you authored the thread.
> > > > Reply to this email directly, view it on GitHub
> > > > <#84 (review)-
14510503
> >,
> > > or mute
> > > > the thread
> > > > <https://github.com/notifications/unsubscribe-
> auth/AIHAkgd5NXf5jhCN4_
> > > bFgwp6ti9ImUArks5rMZykgaJpZM4LWjrO>
> > >
> > > > .
> > > >
> > >
> > > —
> > > You are receiving this because you commented.
> > > Reply to this email directly, view it on GitHub
> > > <#84 (comment)>,
or
> > mute
> > > the thread
> > > <https://github.com/notifications/unsubscribe-auth/AMFFeg_
> > 2fEVWP6vem5Pz6c9q2Incdmgtks5rMaMugaJpZM4LWjrO>
> > > .
> > >
> >
> > —
> > You are receiving this because you were mentioned.
> > Reply to this email directly, view it on GitHub
> > <#84 (comment)>, or
> mute
> > the thread
> > <https://github.com/notifications/unsubscribe-auth/
> AIHAkqSki6UDBBzcb2YYjdAzSBV7dSf_ks5rMaXxgaJpZM4LWjrO>
> > .
> >
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#84 (comment)>, or
mute
> the thread
> <https://github.com/notifications/unsubscribe-
auth/AMFFev2Ypno4UF6EdceKx_lwbLiyCjmlks5rMadKgaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkkQ9mZbc7dbZXxgiesPWBaN-UpQgks5rMaoqgaJpZM4LWjrO>
.
|
|
Stacy,
Yep, those are correct.
Dave
On Wed, Dec 28, 2016 at 9:44 AM, stacywalters <notifications@github.com>
wrote:
… Dave,
Sorry to be so cautious on this but can you verify the following steps:
No: On your local sand box clone, in the TRJ branch, revert the mod to the
file (depend.common). Then, push the code back up to github.
(1) git reset HEAD main/depend.common
Yes: From your master branch, cut over an new branch (DEPENDENCY_FIX). Put
that one-liner in there, then: add, commit, push. You'll have two PRs, one
that is very close to the first, and one that is for the bug-fix only.
(1) git checkout -b DEPENDENCY_FIX
Thanks, Stacy.
On Tue, Dec 27, 2016 at 5:15 PM, Dave Gill ***@***.***>
wrote:
> Stacy,
> Yes and no.
>
> No: On your local sand box clone, in the TRJ branch, revert the mod to
the
> file (depend.common). Then, push the code back up to github.
>
> Yes: From your master branch, cut over an new branch (DEPENDENCY_FIX).
Put
> that one-liner in there, then: add, commit, push. You'll have two PRs,
one
> that is very close to the first, and one that is for the bug-fix only.
>
> Dave
>
> On Tue, Dec 27, 2016 at 5:03 PM, stacywalters ***@***.***>
> wrote:
>
> > Dave,
> >
> > Does that mean I need to go through the process to generate a pull
> request
> > again?
> >
> > Stacy
> >
> >
> > On Tue, Dec 27, 2016 at 4:57 PM, Dave Gill ***@***.***>
> > wrote:
> >
> > > Stacy,
> > >
> > > Great job tracking down busted dependencies, but no good deed goes
> > > unpunished.
> > >
> > > We will want that single line pulled out and put in as a separate PR.
> > >
> > > Dave
> > >
> > > On Tue, Dec 27, 2016 at 4:45 PM, stacywalters <
> ***@***.***>
> > > wrote:
> > >
> > > > Dave,
> > > >
> > > > I added this dependency for module_sf_clm.o because
> > phys/module_sf_clm.F
> > > > use associates
> > > > module_ra_gfdleta at line # 3919. Without this dependency you are
> > playing
> > > > roulette with multi
> > > > thread compilations.
> > > >
> > > > Stacy
> > > >
> > > > On Tue, Dec 27, 2016 at 4:17 PM, Dave Gill <
***@***.***
> >
> > > > wrote:
> > > >
> > > > > ***@***.**** commented on this pull request.
> > > > > ------------------------------
> > > > >
> > > > > In main/depend.common
> > > > > <#84 (review)-
> 14510503
> > >:
> > > > >
> > > > > > @@ -438,6 +438,7 @@ module_sf_clm.o: module_cam_shr_kind_mod.o
\
> > > > > module_cam_shr_const_mod.o \
> > > > > module_cam_support.o \
> > > > > module_sf_urban.o \
> > > > > + module_ra_gfdleta.o \
> > > > >
> > > > > Why is there a new dependency on module_ra_gfdleta.o?
> > > > >
> > > > > —
> > > > > You are receiving this because you authored the thread.
> > > > > Reply to this email directly, view it on GitHub
> > > > > <#84 (review)-
> 14510503
> > >,
> > > > or mute
> > > > > the thread
> > > > > <https://github.com/notifications/unsubscribe-
> > auth/AIHAkgd5NXf5jhCN4_
> > > > bFgwp6ti9ImUArks5rMZykgaJpZM4LWjrO>
> > > >
> > > > > .
> > > > >
> > > >
> > > > —
> > > > You are receiving this because you commented.
> > > > Reply to this email directly, view it on GitHub
> > > > <#84 (comment)>,
> or
> > > mute
> > > > the thread
> > > > <https://github.com/notifications/unsubscribe-auth/AMFFeg_
> > > 2fEVWP6vem5Pz6c9q2Incdmgtks5rMaMugaJpZM4LWjrO>
> > > > .
> > > >
> > >
> > > —
> > > You are receiving this because you were mentioned.
> > > Reply to this email directly, view it on GitHub
> > > <#84 (comment)>,
or
> > mute
> > > the thread
> > > <https://github.com/notifications/unsubscribe-auth/
> > AIHAkqSki6UDBBzcb2YYjdAzSBV7dSf_ks5rMaXxgaJpZM4LWjrO>
> > > .
> > >
> >
> > —
> > You are receiving this because you were mentioned.
> > Reply to this email directly, view it on GitHub
> > <#84 (comment)>, or
> mute
> > the thread
> > <https://github.com/notifications/unsubscribe-
> auth/AMFFev2Ypno4UF6EdceKx_lwbLiyCjmlks5rMadKgaJpZM4LWjrO>
> > .
> >
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#84 (comment)>, or
mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/
AIHAkkQ9mZbc7dbZXxgiesPWBaN-UpQgks5rMaoqgaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AMFFehpFHXwyT3731yLLt3PzeFbkmtwhks5rMpIHgaJpZM4LWjrO>
.
|
|
In addition to a document, I think we could have a short README for this as well as something in examples.namelist. This can all be done later. |
mkavulich
approved these changes
Dec 30, 2016
mkavulich
reviewed
Jan 6, 2017
|
|
||
| i_beg = max( 1,its-1 ) | ||
| i_end = min( ite+2,ide-1 ) | ||
| do j=max(1,jts-1),min(jte+2,jde-1) |
There was a problem hiding this comment.
@davegill Is this the right set of "mu"s to be using here?
davegill
reviewed
Jan 9, 2017
| i_end = min( ite+2,ide-1 ) | ||
| do j=max(1,jts-1),min(jte+2,jde-1) | ||
| do k=kms,kme | ||
| u(its:i_end,k,j)=ru_m(its:i_end,k,j)/muu(its:i_end,j)*msfu(its:i_end,j) |
There was a problem hiding this comment.
Stacy,
Change these three lines
-
Make the "mu" variables upper case.
muu -> MUU
muv -> MUV
mut -> MUT -
Split the loop into two pieces
DO k = kms,kme-1
the lines with u and v
DO k = kms,kme
the line with w is here
|
Stacy, muu -> muus Your line numbers may vary, but in dyn_em/solve_em.F, in the call to trajectory: Original codeNew code |
KEYWORDS: trajectory SOURCE: internal DESCRIPTION OF CHANGES: The original "trajectory" code in routine module_em.F only computed the trajectory spatial coordinates per trajectory, per time step. The new code, in share/module_trajectory.F, takes trajectory coordinate values and linearly interpolates in x,y, and z to get the value of up 100 variables per trajectory per time step. The code will detect and handle trajectory coordinate values that are between the grid parcels. The user may have up to 1000 trajectories per domain and as mentioned up to 100 variables per trajectory. Inputs are in the namelist files wrfinput_traj_d<domain>. Outputs are in the netcdf files wrfout_traj_d<domain>. LIST OF MODIFIED FILES: M Registry/Registry.EM_COMMON M Registry/Registry.NMM M chem/chem_driver.F M chem/depend.chem M dyn_em/module_em.F M dyn_em/start_em.F M external/RSL_LITE/module_dm.F M main/depend.common M share/Makefile M share/module_check_a_mundo.F A share/module_trajectory.F M share/solve_interface.F TESTS CONDUCTED: WTF reg test, version 3.05, has been successfully completed. The code has been tested in ACOM by Mary Barth, Megan Bella, and Gabrielle Pfister.
KEYWORDS: trajectory SOURCE: internal DESCRIPTION OF CHANGES: The original "trajectory" code in routine module_em.F only computed the trajectory spatial coordinates per trajectory, per time step. The new code, in share/module_trajectory.F, takes trajectory coordinate values and linearly interpolates in x,y, and z to get the value of up 100 variables per trajectory per time step. The code will detect and handle trajectory coordinate values that are between the grid parcels. The user may have up to 1000 trajectories per domain and as mentioned up to 100 variables per trajectory. Inputs are in the namelist files wrfinput_traj_d<domain>. Outputs are in the netcdf files wrfout_traj_d<domain>. LIST OF MODIFIED FILES: M Registry/Registry.EM_COMMON M Registry/Registry.NMM M chem/chem_driver.F M chem/depend.chem M dyn_em/module_em.F M dyn_em/start_em.F M external/RSL_LITE/module_dm.F M main/depend.common M share/Makefile M share/module_check_a_mundo.F A share/module_trajectory.F M share/solve_interface.F TESTS CONDUCTED: WTF reg test, version 3.05, has been successfully completed. The code has been tested in ACOM by Mary Barth, Megan Bella, and Gabrielle Pfister.
|
Dave,
I just tried to compile a clone I got today and the variables:
muus and muvs
are not in the grid object. muts is in the grid object. This is a compile
without any trajectory code.
Stacy
…On Mon, Jan 9, 2017 at 6:52 AM, Dave Gill ***@***.***> wrote:
Stacy,
Also, Wei pointed out that the "mu" variables from the solver should be
fixed.
muu -> muus
muv -> muvs
mut -> muts
Your line numbers may vary, but in dyn_em/solve_em.F, in the call to
trajectory:
Original code
3469 !$OMP PARALLEL DO &
3470 !$OMP PRIVATE ( ij )
3471 DO ij = 1 , grid%num_tiles
3472
3473 call trajectory (grid,config_flags, &
3474 grid%dt,grid%itimestep,grid%ru_m, grid%rv_m, grid%ww_m,&
3475 grid%mut,grid%muu,grid%muv, &
New code
3469 !$OMP PARALLEL DO &
3470 !$OMP PRIVATE ( ij )
3471 DO ij = 1 , grid%num_tiles
3472
3473 call trajectory (grid,config_flags, &
3474 grid%dt,grid%itimestep,grid%ru_m, grid%rv_m, grid%ww_m,&
3475 grid%muts,grid%muus,grid%muvs, &
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkq_DRzANCvbYHF9t0yD38ionDR1Lks5rQjupgaJpZM4LWjrO>
.
|
|
Stacy, |
|
Dave,
To be clear the muts variable is part of grid. Right?
And what about all the commotion regarding rasm_diag?
Did Jamie actually add registry.rasm_diag to Registry/Registry.EM_CHEM?
I gather that no matter what I should start basically start over by doing a
update_fork?
Stacy
…On Wed, Jan 11, 2017 at 9:53 AM, Dave Gill ***@***.***> wrote:
Stacy,
Looks like muus and muvs are "i1", not "state". So, get rid of the "grid%"
part. The variables are just muus and muvs, not part of the grid DDT.
Dave
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAknmH9SiaXKPtczJq_YDMt9bzlk3vks5rRQkJgaJpZM4LWjrO>
.
|
KEYWORDS: trajectory SOURCE: internal DESCRIPTION OF CHANGES: The original "trajectory" code in routine module_em.F only computed the trajectory spatial coordinates per trajectory, per time step. The new code, in share/module_trajectory.F, takes trajectory coordinate values and linearly interpolates in x,y, and z to get the value of up 100 variables per trajectory per time step. The code will detect and handle trajectory coordinate values that are between the grid parcels. The user may have up to 1000 trajectories per domain and as mentioned up to 100 variables per trajectory. Inputs are in the namelist files wrfinput_traj_d<domain>. Outputs are in the netcdf files wrfout_traj_d<domain>. LIST OF MODIFIED FILES: M Registry/Registry.EM_COMMON M Registry/Registry.NMM M chem/chem_driver.F M chem/depend.chem M dyn_em/module_em.F M dyn_em/start_em.F M external/RSL_LITE/module_dm.F M main/depend.common M share/Makefile M share/module_check_a_mundo.F A share/module_trajectory.F M share/solve_interface.F TESTS CONDUCTED: WTF reg test, version 3.05, has been successfully completed. The code has been tested in ACOM by Mary Barth, Megan Bella, and Gabrielle Pfister.
Bring local TRJ branch up to latest master branch.
|
@davegill @stacywalters @mkavulich We will probably need to run a WRFDA regression test on the TRJ branch. |
|
Dave,
This code has been in my TRJ branch for some time. When I last ran WTF on
just the trajectory additions
there were no reported errors.
Should I wait to do a "squash and merge" until the WRFDA WTF test is done?
I gather that Jamie is going
to do the WTF test. Is that correct?
Please advise.
Stacy
…On Thu, Jan 12, 2017 at 11:30 AM, Dave Gill ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In share/module_check_a_mundo.F <#84>
:
> @@ -606,6 +606,13 @@ SUBROUTINE check_nml_consistency
model_config_rec%num_traj = 0
END IF
+#elif( NMM_CORE == 1 )
@jamiebresch <https://github.com/jamiebresch>
Has this ifdef issue in check_a_mundo been satisfactorially addressed? It
seems like you wanted something like this:
#endif
#if ((EM_CORE == 1) && (DA_CORE != 1))
!----------------------------------------------------------------------------
! If trajectory option is turned off, make sure the number of trajectories is
! zero.
!----------------------------------------------------------------------------
IF ( ( model_config_rec%traj_opt .EQ. 0 ) .AND. &
( model_config_rec%num_traj .NE. 0 ) ) THEN
WRITE (wrf_err_message, FMT='(A,A)') '--- WARNING: traj_opt is zero, but ', &
'num_traj is not zero; setting num_traj to zero.'
CALL wrf_message ( wrf_err_message )
model_config_rec%num_traj = 0
END IF
#else
!----------------------------------------------------------------------------
! If NMM core or DA, and trajectories are on, then halt.
!----------------------------------------------------------------------------
IF ( model_config_rec%traj_opt /= 0 ) THEN
call wrf_error_fatal( 'Trajectories not supported in NMM or DA' )
END IF
#endif
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkrJc0XaJQkewT_n3Ah4TYkhuvDH6ks5rRnFRgaJpZM4LWjrO>
.
|
|
Stacy,
Jamie and I are okay with you merging it now. It doesn't break any of our
existing capabilities.
Michael Kavulich, Jr.
Associate Scientist, National Center for Atmospheric Research (NCAR)
Mesoscale & Microscale Meteorology Laboratory (MMM)
Data Assimilation Section
*kavulich@ucar.edu <kavulich@ucar.edu>*
On Thu, Jan 12, 2017 at 11:51 AM, stacywalters <notifications@github.com>
wrote:
… Dave,
This code has been in my TRJ branch for some time. When I last ran WTF on
just the trajectory additions
there were no reported errors.
Should I wait to do a "squash and merge" until the WRFDA WTF test is done?
I gather that Jamie is going
to do the WTF test. Is that correct?
Please advise.
Stacy
On Thu, Jan 12, 2017 at 11:30 AM, Dave Gill ***@***.***>
wrote:
> ***@***.**** commented on this pull request.
> ------------------------------
>
> In share/module_check_a_mundo.F <https://github.com/wrf-model/
WRF/pull/84>
> :
>
> > @@ -606,6 +606,13 @@ SUBROUTINE check_nml_consistency
> model_config_rec%num_traj = 0
> END IF
>
> +#elif( NMM_CORE == 1 )
>
> @jamiebresch <https://github.com/jamiebresch>
> Has this ifdef issue in check_a_mundo been satisfactorially addressed? It
> seems like you wanted something like this:
>
> #endif
> #if ((EM_CORE == 1) && (DA_CORE != 1))
> !-----------------------------------------------------------
-----------------
> ! If trajectory option is turned off, make sure the number of
trajectories is
> ! zero.
> !-----------------------------------------------------------
-----------------
> IF ( ( model_config_rec%traj_opt .EQ. 0 ) .AND. &
> ( model_config_rec%num_traj .NE. 0 ) ) THEN
> WRITE (wrf_err_message, FMT='(A,A)') '--- WARNING: traj_opt is zero, but
', &
> 'num_traj is not zero; setting num_traj to zero.'
> CALL wrf_message ( wrf_err_message )
> model_config_rec%num_traj = 0
> END IF
>
> #else
> !-----------------------------------------------------------
-----------------
> ! If NMM core or DA, and trajectories are on, then halt.
> !-----------------------------------------------------------
-----------------
> IF ( model_config_rec%traj_opt /= 0 ) THEN
> call wrf_error_fatal( 'Trajectories not supported in NMM or DA' )
> END IF
> #endif
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#84>, or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AIHAkrJc0XaJQkewT_
n3Ah4TYkhuvDH6ks5rRnFRgaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AMHfAhb3xRU6KjdM_hrIr5O_-jvvz0oqks5rRnY2gaJpZM4LWjrO>
.
|
|
Michael,
Since I'm a git "neophyte" I'd like to confirm that you mean go ahead with
the "squash and merge" process via github.
Stacy
On Thu, Jan 12, 2017 at 12:35 PM, Michael Kavulich <notifications@github.com
… wrote:
Stacy,
Jamie and I are okay with you merging it now. It doesn't break any of our
existing capabilities.
Michael Kavulich, Jr.
Associate Scientist, National Center for Atmospheric Research (NCAR)
Mesoscale & Microscale Meteorology Laboratory (MMM)
Data Assimilation Section
***@***.*** ***@***.***>*
On Thu, Jan 12, 2017 at 11:51 AM, stacywalters ***@***.***>
wrote:
> Dave,
>
> This code has been in my TRJ branch for some time. When I last ran WTF on
> just the trajectory additions
> there were no reported errors.
>
> Should I wait to do a "squash and merge" until the WRFDA WTF test is
done?
> I gather that Jamie is going
> to do the WTF test. Is that correct?
>
> Please advise.
>
> Stacy
>
> On Thu, Jan 12, 2017 at 11:30 AM, Dave Gill ***@***.***>
> wrote:
>
> > ***@***.**** commented on this pull request.
> > ------------------------------
> >
> > In share/module_check_a_mundo.F <https://github.com/wrf-model/
> WRF/pull/84>
> > :
> >
> > > @@ -606,6 +606,13 @@ SUBROUTINE check_nml_consistency
> > model_config_rec%num_traj = 0
> > END IF
> >
> > +#elif( NMM_CORE == 1 )
> >
> > @jamiebresch <https://github.com/jamiebresch>
> > Has this ifdef issue in check_a_mundo been satisfactorially addressed?
It
> > seems like you wanted something like this:
> >
> > #endif
> > #if ((EM_CORE == 1) && (DA_CORE != 1))
> > !-----------------------------------------------------------
> -----------------
> > ! If trajectory option is turned off, make sure the number of
> trajectories is
> > ! zero.
> > !-----------------------------------------------------------
> -----------------
> > IF ( ( model_config_rec%traj_opt .EQ. 0 ) .AND. &
> > ( model_config_rec%num_traj .NE. 0 ) ) THEN
> > WRITE (wrf_err_message, FMT='(A,A)') '--- WARNING: traj_opt is zero,
but
> ', &
> > 'num_traj is not zero; setting num_traj to zero.'
> > CALL wrf_message ( wrf_err_message )
> > model_config_rec%num_traj = 0
> > END IF
> >
> > #else
> > !-----------------------------------------------------------
> -----------------
> > ! If NMM core or DA, and trajectories are on, then halt.
> > !-----------------------------------------------------------
> -----------------
> > IF ( model_config_rec%traj_opt /= 0 ) THEN
> > call wrf_error_fatal( 'Trajectories not supported in NMM or DA' )
> > END IF
> > #endif
> >
> > —
> > You are receiving this because you were mentioned.
> > Reply to this email directly, view it on GitHub
> > <#84>, or mute the thread
> > <https://github.com/notifications/unsubscribe-auth/AIHAkrJc0XaJQkewT_
> n3Ah4TYkhuvDH6ks5rRnFRgaJpZM4LWjrO>
>
> > .
> >
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#84 (comment)>, or
mute
> the thread
> <https://github.com/notifications/unsubscribe-
auth/AMHfAhb3xRU6KjdM_hrIr5O_-jvvz0oqks5rRnY2gaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AIHAkgprbjnzjwn43U0j8oWrGaJagZmjks5rRoBvgaJpZM4LWjrO>
.
|
|
Stacy,
Yes, let's wait for official confirmation from Jamie or Mike that the code
ran through the WRF DA regression test OK.
Dave
On Thu, Jan 12, 2017 at 12:52 PM, stacywalters <notifications@github.com>
wrote:
… Michael,
Since I'm a git "neophyte" I'd like to confirm that you mean go ahead with
the "squash and merge" process via github.
Stacy
On Thu, Jan 12, 2017 at 12:35 PM, Michael Kavulich <
***@***.***
> wrote:
> Stacy,
>
> Jamie and I are okay with you merging it now. It doesn't break any of our
> existing capabilities.
>
>
>
> Michael Kavulich, Jr.
> Associate Scientist, National Center for Atmospheric Research (NCAR)
> Mesoscale & Microscale Meteorology Laboratory (MMM)
> Data Assimilation Section
> ***@***.*** ***@***.***>*
>
> On Thu, Jan 12, 2017 at 11:51 AM, stacywalters ***@***.***
>
> wrote:
>
> > Dave,
> >
> > This code has been in my TRJ branch for some time. When I last ran WTF
on
> > just the trajectory additions
> > there were no reported errors.
> >
> > Should I wait to do a "squash and merge" until the WRFDA WTF test is
> done?
> > I gather that Jamie is going
> > to do the WTF test. Is that correct?
> >
> > Please advise.
> >
> > Stacy
> >
> > On Thu, Jan 12, 2017 at 11:30 AM, Dave Gill ***@***.***>
> > wrote:
> >
> > > ***@***.**** commented on this pull request.
> > > ------------------------------
> > >
> > > In share/module_check_a_mundo.F <https://github.com/wrf-model/
> > WRF/pull/84>
> > > :
> > >
> > > > @@ -606,6 +606,13 @@ SUBROUTINE check_nml_consistency
> > > model_config_rec%num_traj = 0
> > > END IF
> > >
> > > +#elif( NMM_CORE == 1 )
> > >
> > > @jamiebresch <https://github.com/jamiebresch>
> > > Has this ifdef issue in check_a_mundo been satisfactorially
addressed?
> It
> > > seems like you wanted something like this:
> > >
> > > #endif
> > > #if ((EM_CORE == 1) && (DA_CORE != 1))
> > > !-----------------------------------------------------------
> > -----------------
> > > ! If trajectory option is turned off, make sure the number of
> > trajectories is
> > > ! zero.
> > > !-----------------------------------------------------------
> > -----------------
> > > IF ( ( model_config_rec%traj_opt .EQ. 0 ) .AND. &
> > > ( model_config_rec%num_traj .NE. 0 ) ) THEN
> > > WRITE (wrf_err_message, FMT='(A,A)') '--- WARNING: traj_opt is zero,
> but
> > ', &
> > > 'num_traj is not zero; setting num_traj to zero.'
> > > CALL wrf_message ( wrf_err_message )
> > > model_config_rec%num_traj = 0
> > > END IF
> > >
> > > #else
> > > !-----------------------------------------------------------
> > -----------------
> > > ! If NMM core or DA, and trajectories are on, then halt.
> > > !-----------------------------------------------------------
> > -----------------
> > > IF ( model_config_rec%traj_opt /= 0 ) THEN
> > > call wrf_error_fatal( 'Trajectories not supported in NMM or DA' )
> > > END IF
> > > #endif
> > >
> > > —
> > > You are receiving this because you were mentioned.
> > > Reply to this email directly, view it on GitHub
> > > <#84>, or mute the thread
> > > <https://github.com/notifications/unsubscribe-
auth/AIHAkrJc0XaJQkewT_
> > n3Ah4TYkhuvDH6ks5rRnFRgaJpZM4LWjrO>
> >
> > > .
> > >
> >
> > —
> > You are receiving this because you were mentioned.
> > Reply to this email directly, view it on GitHub
> > <#84 (comment)>, or
> mute
> > the thread
> > <https://github.com/notifications/unsubscribe-
> auth/AMHfAhb3xRU6KjdM_hrIr5O_-jvvz0oqks5rRnY2gaJpZM4LWjrO>
> > .
> >
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#84 (comment)>, or
mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/
AIHAkgprbjnzjwn43U0j8oWrGaJagZmjks5rRoBvgaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AMFFegxvtqc0cBhDcjIlXFMBpva8pAJ2ks5rRoRjgaJpZM4LWjrO>
.
|
|
Yes, the TRJ branch works for all the DA tests Mike and I did. Jamie |
|
Yes, Squash and merge as described in the email I sent out last Friday.
Michael Kavulich, Jr.
Associate Scientist, National Center for Atmospheric Research (NCAR)
Mesoscale & Microscale Meteorology Laboratory (MMM)
Data Assimilation Section
*kavulich@ucar.edu <kavulich@ucar.edu>*
On Thu, Jan 12, 2017 at 12:52 PM, stacywalters <notifications@github.com>
wrote:
… Michael,
Since I'm a git "neophyte" I'd like to confirm that you mean go ahead with
the "squash and merge" process via github.
Stacy
On Thu, Jan 12, 2017 at 12:35 PM, Michael Kavulich <
***@***.***
> wrote:
> Stacy,
>
> Jamie and I are okay with you merging it now. It doesn't break any of our
> existing capabilities.
>
>
>
> Michael Kavulich, Jr.
> Associate Scientist, National Center for Atmospheric Research (NCAR)
> Mesoscale & Microscale Meteorology Laboratory (MMM)
> Data Assimilation Section
> ***@***.*** ***@***.***>*
>
> On Thu, Jan 12, 2017 at 11:51 AM, stacywalters ***@***.***
>
> wrote:
>
> > Dave,
> >
> > This code has been in my TRJ branch for some time. When I last ran WTF
on
> > just the trajectory additions
> > there were no reported errors.
> >
> > Should I wait to do a "squash and merge" until the WRFDA WTF test is
> done?
> > I gather that Jamie is going
> > to do the WTF test. Is that correct?
> >
> > Please advise.
> >
> > Stacy
> >
> > On Thu, Jan 12, 2017 at 11:30 AM, Dave Gill ***@***.***>
> > wrote:
> >
> > > ***@***.**** commented on this pull request.
> > > ------------------------------
> > >
> > > In share/module_check_a_mundo.F <https://github.com/wrf-model/
> > WRF/pull/84>
> > > :
> > >
> > > > @@ -606,6 +606,13 @@ SUBROUTINE check_nml_consistency
> > > model_config_rec%num_traj = 0
> > > END IF
> > >
> > > +#elif( NMM_CORE == 1 )
> > >
> > > @jamiebresch <https://github.com/jamiebresch>
> > > Has this ifdef issue in check_a_mundo been satisfactorially
addressed?
> It
> > > seems like you wanted something like this:
> > >
> > > #endif
> > > #if ((EM_CORE == 1) && (DA_CORE != 1))
> > > !-----------------------------------------------------------
> > -----------------
> > > ! If trajectory option is turned off, make sure the number of
> > trajectories is
> > > ! zero.
> > > !-----------------------------------------------------------
> > -----------------
> > > IF ( ( model_config_rec%traj_opt .EQ. 0 ) .AND. &
> > > ( model_config_rec%num_traj .NE. 0 ) ) THEN
> > > WRITE (wrf_err_message, FMT='(A,A)') '--- WARNING: traj_opt is zero,
> but
> > ', &
> > > 'num_traj is not zero; setting num_traj to zero.'
> > > CALL wrf_message ( wrf_err_message )
> > > model_config_rec%num_traj = 0
> > > END IF
> > >
> > > #else
> > > !-----------------------------------------------------------
> > -----------------
> > > ! If NMM core or DA, and trajectories are on, then halt.
> > > !-----------------------------------------------------------
> > -----------------
> > > IF ( model_config_rec%traj_opt /= 0 ) THEN
> > > call wrf_error_fatal( 'Trajectories not supported in NMM or DA' )
> > > END IF
> > > #endif
> > >
> > > —
> > > You are receiving this because you were mentioned.
> > > Reply to this email directly, view it on GitHub
> > > <#84>, or mute the thread
> > > <https://github.com/notifications/unsubscribe-
auth/AIHAkrJc0XaJQkewT_
> > n3Ah4TYkhuvDH6ks5rRnFRgaJpZM4LWjrO>
> >
> > > .
> > >
> >
> > —
> > You are receiving this because you were mentioned.
> > Reply to this email directly, view it on GitHub
> > <#84 (comment)>, or
> mute
> > the thread
> > <https://github.com/notifications/unsubscribe-
> auth/AMHfAhb3xRU6KjdM_hrIr5O_-jvvz0oqks5rRnY2gaJpZM4LWjrO>
> > .
> >
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#84 (comment)>, or
mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/
AIHAkgprbjnzjwn43U0j8oWrGaJagZmjks5rRoBvgaJpZM4LWjrO>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#84 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AMHfAmp74dFZ8iKLE1MJvmL1Q4TJ6ueAks5rRoRjgaJpZM4LWjrO>
.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
TYPE: enhancement
KEYWORDS: trajectory
SOURCE: internal
DESCRIPTION OF CHANGES:
The original "trajectory" code in routine module_em.F only computed the trajectory
spatial coordinates per trajectory, per time step. The new code, in share/module_trajectory.F,
takes trajectory coordinate values and linearly interpolates in x,y, and z to get the
value of up 100 variables per trajectory per time step. The code will detect and handle
trajectory coordinate values that are between the grid parcels. The user may have
up to 1000 trajectories per domain and as mentioned up to 100 variables per trajectory.
Output goes to netcdf files wrfout_traj_.
LIST OF MODIFIED FILES:
M Registry/Registry.EM_COMMON
M Registry/Registry.NMM
M chem/chem_driver.F
M chem/depend.chem
M dyn_em/module_em.F
M dyn_em/solve_em.F
M dyn_em/start_em.F
M external/RSL_LITE/module_dm.F
M main/depend.common
M share/Makefile
M share/module_check_a_mundo.F
A share/module_trajectory.F
M share/solve_interface.F
TESTS CONDUCTED:
WTF reg test, version 3.04, has been successfully completed. The code has been tested in ACOM by
Mary Barth, Megan Bella, and Gabrielle Pfister.