Added code to perform hail size forecasting diagnostic

[ywangwof]

Description

FV3-HAILCAST hail size forecasting diagnostic (Adams-Selin and Ziegler 2016 MWR).

Authors: John Henderson and Rebecca Adams-Selin AER, Conrad Ziegler NSSL.

How Has This Been Tested?

It was tested on Jet. In order to turn on the HAILCAST package, this namelist block should be added to input.nml.

&fv_diagnostic_nml
do_hailcast = .true.
/

and this entry should be added to diag_table

"gfs_dyn", "hailcast_dhail_max", "HAILCAST_DHAIL", "fv3_history2d", "all", .false., "none", 2

Checklist:

Please check all whether they apply or not

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published in downstream modules

[lharris4]

This is a very nice addition to FV3 and I am glad to see that has been contributed back to the community. I have a couple of recommendations:

1. I would prefer that the diagnostics code and variables all be contained within the one HAILCAST module. In particular the code being put into fv_arrays.F90 and fv_nggps_diag.F90 should all be put within the one module_diag_hailcast.F90, with only an initialization routine and an execution routine being exposed to the rest of the code base. The files fv_arrays.F90 and fv_nggps_diag.F90 are already too large and unwieldy.

2. Is this implementation of HAILCAST this designed for any particular microphysical scheme? If so this should be explicitly stated and checked for at initialization. Alternately, if the scheme is relatively generic, but needs the microphysics to make available certain fields, this should also be checked for at initialization, with either a warning or a fatal error thrown if the appropriate fields are not found.

Thanks for your contribution to FV3.

[bensonr]

@ywangwof - @lharris4 requested changes to this PR. If you are unable or unwilling to address these by the end of January 2022, we will be closing this PR.

[ywangwof]

Hello @bensonr, we are working on the request. Unfortunately, we are occupied by some time-sensitive projects recently. Could we request for an extension to the end of February for this task? Thanks.

[bensonr]

@ywangwof - we can leave this open.

[lharris4]

Hi, Yunheng. No problem, whenever you are ready. Lucas

…

On Tue, Jan 18, 2022 at 4:45 PM Rusty Benson ***@***.***> wrote: @ywangwof <https://github.com/ywangwof> - we can leave this open. — Reply to this email directly, view it on GitHub <#164 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AMUQRVCM6DCPG6HHM4SC5C3UWXNOPANCNFSM5H57G5YA> . 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.Message ID: ***@***.***>

[lharris4]

This is much better, exactly what I was asking for. As long as this is checked so it compiles and works correctly I would approve this merge. Thanks for working on this.

[bensonr]

I agree with @lharris4, this is a much cleaner implementation. Please see the review comments and address.

[ywangwof]

Thanks for all your comments. We are still having some issues. After I have fully tested the module, I will ask for your review again.

[ywangwof]

Hello @lharris4, @bensonr, please review this PR again and I finally had it tested with Thompson, NSSL 2-moment and GFDL microphysical schemes with the latest develop branch of the "ufs-weather-model". Thanks.

[junwang-noaa]

remove extra spaces?

[ywangwof]

That is true. I would suggest that this practice should be established as a code convention. Anyway, almost all editors can be set to prune extra spaces at the end of lines automatically before saving.

Furthermore, we swap the orders of the IF statement and the DO loop statements for code efficiency. Thanks.

[lharris4]

This is much improved but I still have a question about the hailcast-specific code in the fv_nggps_diag.

[ywangwof]

Thanks for your comments and suggestions. Here are the summary about my last commit based on your comments.

1. I moved the namelist and reading from module_diag_hailcast to fv_nggps_diag_init in fv_nggps_diags_mod;
2. Listed the variables and subroutines imported from module_diag_hailcast explicitly in fv_nggps_diag.F90;
3. Removed unnecessary comments as noted by @junwang-noaa;
4. Replaced all GOTO statements with EXIT statements;
5. Introduced new variables to replaced some exponent expressions as possible.

Please review them again at your convenience. Thanks again.

[laurenchilutti]

@ywangwof is there a fv3atm and ufs-weather-model PR in place to bring these changes into the UFS weather model?

[ywangwof]

Thanks. @laurenchilutti. I just issued two PRs for them.
fv3atm, NOAA-EMC/fv3atm/pull/518.
ufs-weather-model, ufs-community/ufs-weather-model/pull/1164

[junwang-noaa]

@laurenchilutti @bensonr FYI. we are running RT in UFS (PR#1156)[https://github.com/ufs-community/ufs-weather-model/pull/1156] with this PR. We will let you know when all the RT test pass, please commit the changes (probably on 4/29). Thanks

[binli2337]

@laurenchilutti @bensonr All tests passed in UFS (PR#1156). Please commit the changes from PR#164. Thanks.