update era5/ifs soil field processing

[jimbresch]

Modifications to rd_grib2.F to allow decoding of the era5/ifs soil fields. (referenced in #264 and #265). . Note that EC uses a local table for the grib codes and that g2print is not able to process these fields yet. Decoding was successful for a sample file.

[gabrielcassol]

@jimbresch @JoshLovesFun I downloaded the modified rd_grib2.F file, recompiled my WPS, and tested it with the operational ECMWF IFS 0.25º data using your Vtable.ERA5 and it didn't work. The soil temperature and moisture data still aren't being read. Maybe your contributions work for the ERA5 data, but they don't work for the operational IFS data.

[JoshLovesFun]

@jimbresch @JoshLovesFun I downloaded the modified rd_grib2.F file, recompiled my WPS, and tested it with the operational ECMWF IFS 0.25º data using your Vtable.ERA5 and it didn't work. The soil temperature and moisture data still aren't being read. Maybe your contributions work for the ERA5 data, but they don't work for the operational IFS data.

@gabrielcassol Thanks for your testing. I don't expect the Vtable.ERA5 to work for the operational ECMWF. It is designed to only work for the ERA5 data which I don't think had any changes to soil processing. Thus, a new Vtable specific to the latest operational ECMWF would need to be developed.

Josh

[dudhia]

I am asking someone at NCAR who might use IFS data.

…

On Mon, Jan 13, 2025 at 12:08 PM Joshua ***@***.***> wrote: @jimbresch <https://github.com/jimbresch> @JoshLovesFun <https://github.com/JoshLovesFun> I downloaded the modified rd_grib2.F <https://github.com/wrf-model/WPS/blob/d278f384b9301275170ce9be5b8b207face5b79b/ungrib/src/rd_grib2.F> file, recompiled my WPS, and tested it with the operational ECMWF IFS 0.25º <https://data.ecmwf.int/forecasts/> data using your Vtable.ERA5 <https://github.com/wrf-model/WPS/pull/265/files/d3bdfcf87c274bf1760c490fda8e8c5083aaa05b#diff-bc87956177fc012ed7b2cf0cc558737085bea988d0a913fe4917ec7036c4a9cb> and it didn't work. The soil temperature and moisture data still aren't being read. Maybe your contributions work for the ERA5 data, but they don't work for the operational IFS data. [image: image] <https://private-user-images.githubusercontent.com/43749970/402575049-6e7d7551-915f-4bab-86a6-0d034c2cee1f.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MzY3OTUyMDUsIm5iZiI6MTczNjc5NDkwNSwicGF0aCI6Ii80Mzc0OTk3MC80MDI1NzUwNDktNmU3ZDc1NTEtOTE1Zi00YmFiLTg2YTYtMGQwMzRjMmNlZTFmLnBuZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLSUFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNTAxMTMlMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjUwMTEzVDE5MDE0NVomWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPWYzZjk3ODgyZTcyNWE5YTljYmZkMjlkYTAzYjMwMmYxYzIwZDAwZWUxYmJmMjg0YWRmN2Q2N2NkYWNiNDdhNjImWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0In0.Qu1zvyKCxJJj5hPP-9Tt2E0gPDO691c5Y-2OmO9hTAk> @gabrielcassol <https://github.com/gabrielcassol> Thanks for your testing. I don't expect the Vtable.ERA5 <https://github.com/wrf-model/WPS/pull/265/files/d3bdfcf87c274bf1760c490fda8e8c5083aaa05b#diff-bc87956177fc012ed7b2cf0cc558737085bea988d0a913fe4917ec7036c4a9cb> to work for the operational ECMWF. It is designed to only work for the ERA5 data which I don't think had any changes to soil processing. Thus, a new Vtable specific to the latest operational ECMWF would need to be developed. Josh — Reply to this email directly, view it on GitHub <#266 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEIZ77GXCOD4PQ5HDPZO7O32KQFLXAVCNFSM6AAAAABUBHQLKSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKOBXHE3TGOBSGM> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[dudhia]

There is a Vtable.ECMWF in recent versions.

…

On Mon, Jan 13, 2025 at 2:05 PM Jimy Dudhia ***@***.***> wrote: I am asking someone at NCAR who might use IFS data. On Mon, Jan 13, 2025 at 12:08 PM Joshua ***@***.***> wrote: > @jimbresch <https://github.com/jimbresch> @JoshLovesFun > <https://github.com/JoshLovesFun> I downloaded the modified rd_grib2.F > <https://github.com/wrf-model/WPS/blob/d278f384b9301275170ce9be5b8b207face5b79b/ungrib/src/rd_grib2.F> > file, recompiled my WPS, and tested it with the operational ECMWF IFS > 0.25º <https://data.ecmwf.int/forecasts/> data using your Vtable.ERA5 > <https://github.com/wrf-model/WPS/pull/265/files/d3bdfcf87c274bf1760c490fda8e8c5083aaa05b#diff-bc87956177fc012ed7b2cf0cc558737085bea988d0a913fe4917ec7036c4a9cb> > and it didn't work. The soil temperature and moisture data still aren't > being read. Maybe your contributions work for the ERA5 data, but they don't > work for the operational IFS data. > > [image: image] > <https://private-user-images.githubusercontent.com/43749970/402575049-6e7d7551-915f-4bab-86a6-0d034c2cee1f.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MzY3OTUyMDUsIm5iZiI6MTczNjc5NDkwNSwicGF0aCI6Ii80Mzc0OTk3MC80MDI1NzUwNDktNmU3ZDc1NTEtOTE1Zi00YmFiLTg2YTYtMGQwMzRjMmNlZTFmLnBuZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLSUFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNTAxMTMlMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjUwMTEzVDE5MDE0NVomWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPWYzZjk3ODgyZTcyNWE5YTljYmZkMjlkYTAzYjMwMmYxYzIwZDAwZWUxYmJmMjg0YWRmN2Q2N2NkYWNiNDdhNjImWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0In0.Qu1zvyKCxJJj5hPP-9Tt2E0gPDO691c5Y-2OmO9hTAk> > > @gabrielcassol <https://github.com/gabrielcassol> Thanks for your > testing. I don't expect the Vtable.ERA5 > <https://github.com/wrf-model/WPS/pull/265/files/d3bdfcf87c274bf1760c490fda8e8c5083aaa05b#diff-bc87956177fc012ed7b2cf0cc558737085bea988d0a913fe4917ec7036c4a9cb> > to work for the operational ECMWF. It is designed to only work for the ERA5 > data which I don't think had any changes to soil processing. Thus, a new > Vtable specific to the latest operational ECMWF would need to be developed. > > Josh > > — > Reply to this email directly, view it on GitHub > <#266 (comment)>, or > unsubscribe > <https://github.com/notifications/unsubscribe-auth/AEIZ77GXCOD4PQ5HDPZO7O32KQFLXAVCNFSM6AAAAABUBHQLKSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKOBXHE3TGOBSGM> > . > You are receiving this because you are subscribed to this thread.Message > ID: ***@***.***> >

[gabrielcassol]

@dudhia I have already tested this Vtable.ECMWF with the new operational IFS data and it doesn't work either. I reported this here #264

[weiwangncar]

The fix @jimbresch provided is based on a particular dataset a user pointed to us. Depending on how your data is obtained, the data may have peculiarities. Can you point us to or supply us a data file that you have problem with?

[smileMchen]

I will download and process IFS data to make sure what is going on.
I will keep you updated ....

[weiwangncar]

@jimbresch Do you think this is ready for WPS v4.7?

[andrewsoong]

@jimbresch Do you think this is ready for WPS v4.7?

@weiwangncar This bugfix was not run correctly when real.exe stoped with (grid%tsk unreasonable).
Because the skin temp Vtable is wrong.

[andrewsoong]

Modify Vtable in line 20 to：
235 | 1 | 0 | | SKINTEMP | K | Skin Temperature | 0 | 0 | 17 | 1 |

WRF4.7 works.

[andrewsoong]

Correct line 20 to the follow：
235 | 1 | 0 | | SKINTEMP | K | Skin Temperature | 0 | 0 | 17 | 1 |

The WRF runs successfully.

[MHBalsmeier]

@jimbresch I also tested it and can confirm it works with the modification by @andrewsoong

I have two small suggestions though:

• The command module load is often not necessary when not running on derecho but on standard Linux systems like Ubuntu. To prevent "command not found" errors one could do something like:

if command -v module >/dev/null 2>&1
then
  module load ncarenv/23.09  intel/2024.2.1
  module load eccodes/2.36.0
fi

• The WRF option smooth_cg_topo does not work with this data since the soilhgt variable is not found. Maybe this can be fixed in the Vtable or soilhgt can be computed based on soilgeo by ungrib.