GSL UPP developments up to January 2023

[SamuelTrahanNOAA]

1. Increases precision of VIL_ON_ENTIRE_ATMOS from <scale>4.0</scale> to <scale>7.0</scale>
2. Adds an experimental diagnostic, developed for the HRRR to correct for the HRRR's low bias in cloud cover. For an FAA deliverable due Feb 2023.
3. Able to use either cldfra or cldfra_bl when reading cloud fraction from INITPPOST_NETCDF. (Will use whichever is available; uses cldfra if both are present.)

The GSL in-flight icing developments were merged in pull requests #536 and #621

Fixes #610

[SamuelTrahanNOAA]

These changes were from @EricJames-NOAA and @hu5970

[EricJames-NOAA]

@SamuelTrahanNOAA thanks!

Regarding change (3), I don't understand why EMC reads in a different field for cloud fraction from FV3. GSL has always had cldfra_bl output. We may need to ask other developers about this.

[WenMeng-NOAA]

@SamuelTrahanNOAA thanks!

Regarding change (3), I don't understand why EMC reads in a different field for cloud fraction from FV3. GSL has always had cldfra_bl output. We may need to ask other developers about this.

@yangfanglin and @ericaligo-NOAA Could you confirm if EMC side has switched cloud fraction model output from cldfra to cldfra_bl for RRFS?

[ericaligo-NOAA]

I added the universal 3D cloud fraction,cldfra, in 2020 I believe in order for it to be used in all microphysics applications including the GFDL microphysics. There was no other instantaneous cloud fraction in the FV3 output at that time.Eric -------- Original message --------From: WenMeng-NOAA ***@***.***> Date: 1/26/23 9:02 PM (GMT-05:00) To: NOAA-EMC/UPP ***@***.***> Cc: ericaligo-NOAA ***@***.***>, Mention ***@***.***> Subject: Re: [NOAA-EMC/UPP] GSL UPP developments up to January 2023 (PR #620) @SamuelTrahanNOAA thanks! Regarding change (3), I don't understand why EMC reads in a different field for cloud fraction from FV3. GSL has always had cldfra_bl output. We may need to ask other developers about this. @yangfanglin and @ericaligo-NOAA Could you confirm if EMC side has switched cloud fraction model output from cldfra to cldfra_bl for RRFS? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>

[SamuelTrahanNOAA]

I've merged the bug fix from #621 that is needed to get the inline post working again, after my libIFI changes broke it. (I don't deallocate some arrays after the next iteration of the post.) If this PR will take a while to review, then we should merge #621 first since it'll be a shoo-in.

EDIT: The bug fix in #621 has been merged.

[SamuelTrahanNOAA]

The bug fix in #621 has been merged to develop, and I've synced this branch with develop again.

[WenMeng-NOAA]

@EricJames-NOAA From the UPP RT tests on Hera, there are following changed results in rap/hrrr:

[Wen.Meng@hfe07 ~/stmp2]$ cat rap_2020072316/WRFPRS.GrbF16.diff
697:190713718:HGT:cloud base:rpn_corr=0.895998:rpn_rms=3364.31
698:191867295:PRES:cloud base:rpn_corr=0.91432:rpn_rms=15296.7
708:195473891:CDCON:convective cloud layer:rpn_corr=-nan:rpn_rms=undefined
[Wen.Meng@hfe07 ~/stmp2]$ cat rap_2020072316/WRFNAT.GrbF16.diff
1070:282595626:HGT:cloud base:rpn_corr=0.895998:rpn_rms=3364.31
1071:283749203:PRES:cloud base:rpn_corr=0.91432:rpn_rms=15296.7
[Wen.Meng@hfe07 ~/stmp2]$ cat hrrr_2020060118/WRFTWO.GrbF04.diff
121:85620587:HGT:cloud base:rpn_corr=0.932971:rpn_rms=2896.63
122:88068643:PRES:cloud base:rpn_corr=0.931465:rpn_rms=12453.8

I would assume these changes come from updates in CLDRAD.f?

Now I saw valid cloud fraction in GSL RRFS test at /scratch1/NCEPDEV/stmp2/Wen.Meng/post_rrfs_2022102713.
Would you please verify my testing? Thanks!

[WenMeng-NOAA]

@SamuelTrahanNOAA Would you add change logs in two source code files CLDRAD.f and INITPOST_NETCDF.f?

[EricJames-NOAA]

@EricJames-NOAA From the UPP RT tests on Hera, there are following changed results in rap/hrrr:

[Wen.Meng@hfe07 ~/stmp2]$ cat rap_2020072316/WRFPRS.GrbF16.diff
697:190713718:HGT:cloud base:rpn_corr=0.895998:rpn_rms=3364.31
698:191867295:PRES:cloud base:rpn_corr=0.91432:rpn_rms=15296.7
708:195473891:CDCON:convective cloud layer:rpn_corr=-nan:rpn_rms=undefined
[Wen.Meng@hfe07 ~/stmp2]$ cat rap_2020072316/WRFNAT.GrbF16.diff
1070:282595626:HGT:cloud base:rpn_corr=0.895998:rpn_rms=3364.31
1071:283749203:PRES:cloud base:rpn_corr=0.91432:rpn_rms=15296.7
[Wen.Meng@hfe07 ~/stmp2]$ cat hrrr_2020060118/WRFTWO.GrbF04.diff
121:85620587:HGT:cloud base:rpn_corr=0.932971:rpn_rms=2896.63
122:88068643:PRES:cloud base:rpn_corr=0.931465:rpn_rms=12453.8

I would assume these changes come from updates in CLDRAD.f?

Now I saw valid cloud fraction in GSL RRFS test at /scratch1/NCEPDEV/stmp2/Wen.Meng/post_rrfs_2022102713. Would you please verify my testing? Thanks!

@WenMeng-NOAA yes, we changed the formulation of the exp2 ceiling diagnostic, which would affect these fields. We are seeking to improve the diagnostic for RRFS for the FAA. Your GSL RRFS test looks good to me! Thanks!

[WenMeng-NOAA]

The UPP RT tests were completed on WCOSS2, Hera and Orion. The updates of baselines for rap and hrrr are needed with this PR.

[WenMeng-NOAA]

@SamuelTrahanNOAA Would you add change logs in two source code files CLDRAD.f and INITPOST_NETCDF.f so I can proceed to merging?

[SamuelTrahanNOAA]

Would you add change logs in two source code files CLDRAD.f and INITPOST_NETCDF.f so I can proceed to merging?

It is done.