Update HAFS RTs to use the new multi output grid capability by BinLiu-NOAA · Pull Request #19 · DusanJovic-NOAA/ufs-weather-model

BinLiu-NOAA · 2022-02-08T13:30:40Z

Update/unify existing HAFS RTs to use the new capability of the write grid component to write out both parent and nested grids.

…tput_grids branch.

… grid component to write out both parent and nested grids.

…@DusanJovic-NOAA

…@DusanJovic-NOAA's review suggestions.

…ared.IN.

junwang-noaa · 2022-02-08T15:58:50Z

+dx:                      @[DX]              # grid cell size in x/y direction (meters)
+dy:                      @[DY]              # needed for lambert_conformal
+
+<output_grid_02>


For the hafs test with single domain output, will this section still show up? Can we remove this section using hafs run script if these output grids are not needed?

If you think about fv3_conf/hafs_fv3_run.IN, this script doesn't now how many nests are configured. And by the time this script is executed module_configure is already created in the rundir.

@junwang-noaa and @DusanJovic-NOAA, this is one of the reasons I suggested using the column configuration method. We do not need to add these extra <output_grid_xx> sections. And if default values are the concern, we can just use empty as default values, for example:

output_grid: 'global_latlon', 'regional_latlon', 'rotated_latlon', 'rotated_latlon', 'rotated_latlon' imo: 360, , , , jmo: 181, , , , cen_lon: , -175.00, -86.00, -86.00, -86.00 cen_lat: , 0.00, 23.00, 23.00, 23.00 lon1: , -220.00, -45.00, -45.00, -35.00 lat1: , -45.00, -45.00, -45.00, -30.00 lon2: , -130.00, 45.00, 45.00, 35.00 lat2: , 45.00, 45.00, 45.00, 30.00 dlon: , 0.25, 0.25, 0.10, 0.05 dlat: , 0.25, 0.25, 0.10, 0.05

I personally think this will give us a much cleaner model_configure template/file. Thanks!

@junwang-noaa and @DusanJovic-NOAA, this is one of the reasons I suggested using the column configuration method. We do not need to add these extra <output_grid_xx> sections. And if default values are the concern, we can just use empty as default values, for example:

output_grid: 'global_latlon', 'regional_latlon', 'rotated_latlon', 'rotated_latlon', 'rotated_latlon' imo: 360, , , , jmo: 181, , , , cen_lon: , -175.00, -86.00, -86.00, -86.00 cen_lat: , 0.00, 23.00, 23.00, 23.00 lon1: , -220.00, -45.00, -45.00, -35.00 lat1: , -45.00, -45.00, -45.00, -30.00 lon2: , -130.00, 45.00, 45.00, 35.00 lat2: , 45.00, 45.00, 45.00, 30.00 dlon: , 0.25, 0.25, 0.10, 0.05 dlat: , 0.25, 0.25, 0.10, 0.05

I personally think this will give us a much cleaner model_configure template/file. Thanks!

Are you sure this table, with 'empty values', will properly parse using ESMF config parser? Shouldn't all values at a given row be the same type i.e. float?

I am not very sure. I recall ESMF_config can treat every line as string, then one can parse it into variables (string, integer, float, etc.). But, @theurich probably know this better.

You can certainly read in every line as a string. Of course, parsing this string, and tokenizing it then becomes your an application level responsibility.
If on the other hand you have as many elements in each line as there are columns, then you can use ESMF_Config API to read in each value. But that is where you would need dummies.

junwang-noaa · 2022-02-08T16:46:01Z

I am thinking of adding a section (function) to handle grids in the rt_fv3.sh. E.g. if configuration variable output_grid_xx is not set, then remove that section in the model_configure.

On Tue, Feb 8, 2022 at 11:32 AM Dusan Jovic ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In tests/parm/model_configure_hafs_shared.IN <#19 (comment)> : > +cen_lon: @[CEN_LON] # central longitude/latitude (degrees) +cen_lat: @[CEN_LAT] # needed for rotated_latlon, lambert_conformal +lon1: @[LON1] # longitude/latitude of lower-left corner +lat1: @[LAT1] # needed for regional_latlon, rotated_latlon, lambert_conformal +lon2: @[LON2] # longitude/latitude of upper-right corner +lat2: @[LAT2] # needed for regional_latlon, rotated_latlon +dlon: @[DLON] # grid spacing in longitude/latitude direction (degrees) +dlat: @[DLAT] # needed for regional_latlon, rotated_latlon +stdlat1: @[STDLAT1] # latitude of first/second standard parallel (degrees) +stdlat2: @[STDLAT2] # needed for lambert_conformal +nx: @[NX] # number of grid cells along x/y-axis +ny: @[NY] # needed for lambert_conformal +dx: @[DX] # grid cell size in x/y direction (meters) +dy: @[DY] # needed for lambert_conformal + +<output_grid_02> If you think about fv3_conf/hafs_fv3_run.IN, this script doesn't now how many nests are configured. And by the time this script is executed module_configure is already created in the rundir. — Reply to this email directly, view it on GitHub <#19 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AI7D6TM34MYBAIMJEMPRD5TU2FARZANCNFSM5N2PBCYA> . 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 commented.Message ID: ***@***.***>

How would you do that? And is it worth complicating the script to remove parts of the configure file that are not going to be read in anyway. In my opinion the much cleaner solution is to not have them in the model_configure in the first place. That means instead of having a template with 6 sections, have a script that will append just needed sections. But that will complicate scripting in a different way, which I'm also not sure is worth doing.

…model_configure.

BinLiu-NOAA · 2022-02-09T12:48:43Z

@junwang-noaa and @DusanJovic-NOAA, I added a few lines in run_test.sh to remove those unneeded the <output_grid_xx> sections from model_configure. For example,

if [[ "Q${INPUT_NEST03_NML:-}" != Q ]] ; then
    INPES_NEST=$INPES_NEST03; JNPES_NEST=$JNPES_NEST03
    NPX_NEST=$NPX_NEST03; NPY_NEST=$NPY_NEST03
    K_SPLIT_NEST=$K_SPLIT_NEST03; N_SPLIT_NEST=$N_SPLIT_NEST03
    atparse < ${PATHRT}/parm/${INPUT_NEST03_NML} > input_nest03.nml
else
    lnbeg=$(grep -i -n "<output_grid_03>" model_configure | cut -d: -f1)
    lnend=$(grep -i -n "</output_grid_03>" model_configure | cut -d: -f1)
    [[ ! -z "$lnbeg" ]] && [[ ! -z "$lnend" ]] && sed -i -e "${lnbeg},${lnend}d" model_configure
fi

It's not an elegant solution, but it at least works fine. Thanks!

DusanJovic-NOAA · 2022-02-09T13:35:07Z

@junwang-noaa and @DusanJovic-NOAA, I added a few lines in run_test.sh to remove those unneeded the <output_grid_xx> sections from model_configure. For example,

if [[ "Q${INPUT_NEST03_NML:-}" != Q ]] ; then
    INPES_NEST=$INPES_NEST03; JNPES_NEST=$JNPES_NEST03
    NPX_NEST=$NPX_NEST03; NPY_NEST=$NPY_NEST03
    K_SPLIT_NEST=$K_SPLIT_NEST03; N_SPLIT_NEST=$N_SPLIT_NEST03
    atparse < ${PATHRT}/parm/${INPUT_NEST03_NML} > input_nest03.nml
else
    lnbeg=$(grep -i -n "<output_grid_03>" model_configure | cut -d: -f1)
    lnend=$(grep -i -n "</output_grid_03>" model_configure | cut -d: -f1)
    [[ ! -z "$lnbeg" ]] && [[ ! -z "$lnend" ]] && sed -i -e "${lnbeg},${lnend}d" model_configure
fi

It's not an elegant solution, but it at least works fine. Thanks!

Thanks. Unnecessary complication that's not really necessary. But since it's been already added, let's leave it. This entire section is very repetitive, difficult to maintain, and error prone.

DusanJovic-NOAA · 2022-02-09T13:49:53Z

+    lnbeg=$(grep -i -n "<output_grid_04>" model_configure | cut -d: -f1)
+    lnend=$(grep -i -n "</output_grid_04>" model_configure | cut -d: -f1)
+    [[ ! -z "$lnbeg" ]] && [[ ! -z "$lnend" ]] && sed -i -e "${lnbeg},${lnend}d" model_configure


Can these three lines be replaced by something like:
sed -i -e "/<output_grid_04>/,/<\/output_grid_04>/d" model_configure

@DusanJovic-NOAA, I think your solution is simpler and better. Since the PR is already merged, please commit your change directly into your branch. Thanks a lot!

Done. f232add

BinLiu-NOAA added 2 commits February 8, 2022 00:41

Update submodule FV3 to point to the latest commit of the multiple_ou…

90c6887

…tput_grids branch.

Update/unify existing HAFS RTs to use the new capability of the write…

767644f

… grid component to write out both parent and nested grids.

DusanJovic-NOAA reviewed Feb 8, 2022

View reviewed changes

Comment thread tests/default_vars.sh Outdated

DusanJovic-NOAA reviewed Feb 8, 2022

View reviewed changes

Comment thread tests/default_vars.sh Outdated

DusanJovic-NOAA reviewed Feb 8, 2022

View reviewed changes

Comment thread tests/tests/hafs_regional_telescopic_2nests_atm

DusanJovic-NOAA reviewed Feb 8, 2022

View reviewed changes

Comment thread tests/tests/hafs_global_1nest_atm

BinLiu-NOAA added 4 commits February 8, 2022 15:02

Some further adjustments for HAFS RT output grids definitions based on …

0584cc0

…@DusanJovic-NOAA's review suggestions.

Use NX/Y, DX/Y instead of NX/Y_LC, DX/Y_LC in model_configure_hafs_sh…

6998937

…ared.IN.

Fix a comment in model_configure_hafs_shared.IN.

ce0c1e3

Remove duplicated comments in model_configure_hafs_shared.IN.

5579633

junwang-noaa reviewed Feb 8, 2022

View reviewed changes

DusanJovic-NOAA reviewed Feb 8, 2022

View reviewed changes

Comment thread tests/tests/hafs_regional_telescopic_2nests_atm Outdated

BinLiu-NOAA added 2 commits February 8, 2022 20:58

Update LAT1_3 in hafs_regional_telescopic_2nests_atm.

e949037

Add the capability to delete unneeded <output_grid_xx> sections from …

75430a1

…model_configure.

DusanJovic-NOAA merged commit f0af03e into DusanJovic-NOAA:multiple_output_grids Feb 9, 2022

DusanJovic-NOAA reviewed Feb 9, 2022

View reviewed changes

Conversation

BinLiu-NOAA commented Feb 8, 2022

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

BinLiu-NOAA Feb 8, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

junwang-noaa commented Feb 8, 2022 via email • edited by DusanJovic-NOAA Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

BinLiu-NOAA commented Feb 9, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

DusanJovic-NOAA commented Feb 9, 2022

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

BinLiu-NOAA Feb 8, 2022 •

edited

Loading

junwang-noaa commented Feb 8, 2022 via email •

edited by DusanJovic-NOAA

Loading

BinLiu-NOAA commented Feb 9, 2022 •

edited

Loading