Running WRF using MANAL(JMA), error in ungrib

[HirotoUi]

Hello.

I am trying to run a model using MANAL(JMA) as input data for WRF.
When using Vtable.JMAGSM, the surface RH(relative humidity) and TT(temperature) are not recognized.

If level1 is set to 1.5, the following error occurs:

============
Start_date = 2022-07-07_00:00:00 , End_date = 2022-07-10_00:00:00
output format is WPS
Path to intermediate files is ./
At line 214 of file parse_table.f90
Fortran runtime error: Bad integer for item 1 in list input


Error termination. Backtrace:
#0 0x751d96c23960 in ???
#1 0x751d96c244d9 in ???
#2 0x751d96c2510f in ???
#3 0x751d96e6e5eb in ???
#4 0x751d96e715aa in ???
#5 0x751d96e722aa in ???
#6 0x61b819fc4691 in ???
#7 0x61b819f79ec0 in ???
#8 0x61b819f7b076 in ???
#9 0x751d96829d8f in __libc_start_call_main
at ../sysdeps/nptl/libc_start_call_main.h:58
#10 0x751d96829e3f in __libc_start_main_impl
at ../csu/libc-start.c:392
#11 0x61b819f74824 in ???
==============


How should I modify the Vtable to read 1.5m-level data correctly?


Additionally, I edited ungrib./src/rd_grib2.F as follows:
==============
! Height above ground (m)
if (gfld%ipdtmpl(12) .eq. 2. .or.
& gfld%ipdtmpl(12) .eq. 1.5 .or. ! temp fix for hrrr maxref
& gfld%ipdtmpl(12) .eq. 10. ) then
level=200100.
==============


Thanks.

 

[kwerner]

Hi,
I'd like to first apologize for the long delay in response. My notification for this got pushed way down in my emails.

Unfortunately WPS is not able to accept data at 1.5 meters. The requirement is for 2-meter data for TT and RH. You can still use the MANAL data, but if it doesn't contain those variables at 2m, you will need to supplement it with another data type. See Using Multiple Meteorological Data Sources for guidance on how to do so.

 

[HirotoUi]

Thank you for reply.

I have decided not to obtain data from MANAL (every 3 hours) at a height of 1.5m. Instead of, I changed to obtain the data from a different meteorological source (every 6 hours).

For the result, the values from the MANAL meteorological data are all zero at met_em*.

Could you give me some advice?

 

[kwerner]

Can you provide the namelist.wps file you're using to process metgrid, as well as the metgrid.log file? Can you also use the util/rd_intermediate.exe utility to show what is in the MANAL intermediate files (the files that are created during ungrib)? To do so, from the WPS directory, issue something like

./util/rd_intermediate.exe ONE-OF-THE-MANAL-INTERMEDIATE-FILES >& manal.txt

and then also attach the manal.txt file. Thanks!

 

[HirotoUi]

Thank you for your reply. The namelist.wps, metgrid.exe, and manal.txt files are attached.

Thank you.

 

[kwerner]

Hi,
Thanks for sending those. Can you also share the following files? These files will probably be too large to attach, so see the home page of this forum for instructions on sharing large files.

1. a couple of your met_em* output files, showing the "zero" values for the variables you're having trouble with
2. a few time periods of the intermediate files for all 3 data types
3. the Vtables you used to process each data type
4. the original gribbed data for each type

Just a few times is okay. I'd just like to try to test this on my end. Can you also let me know a couple specific variables to look at - that will just help me to identify this faster. Thanks!

 

[kwerner]

Hi,
Thank you so much for your patience over these last few weeks while we had other obligations to tend to. Since I've started working with you on this, a few others have mentioned this same issue. See this post, on which a solution is suggested, and let me know if that works for you.

 

[HirotoUi]

Thank you for your reply!

In my case, changing the versions of gcc and gfortran did not work for the issue. The values from the MANAL meteorological data are all zero at met_em*.

I tested by gcc and gfortran v.10, 11, and 12.

 

[kwerner]

To reduce confusion for anyone reading this in the future, @HirotoUi and I chatted a few times offline. I informed them that when I tested only the MANAL data, I got errors indicating the data did not cover the full domain:

WARNING: Field PMSL has missing values at level 201300 at (i,j)=(1,1)
ERROR: Missing values encountered in interpolated fields. Stopping.

This is an error that often occurs when only a subset of data is downloaded. Due to map projection differences, sometimes the corners and/or edges of the domain will not be covered by the subset of data downloaded.

@HirotoUi,
Can you try downloading a full global dataset of the MANAL data and trying that to see if it gets past that error? You can just run a quick test through WPS, using only those data for now.