Error in real.exe

[sroopa]

Dear all,
I am encountering the foll error in a WRF simulation.
Domain 1: Current date being processed: 2019-08-27_01:00:00.0000, which is loop # 2 out of 73
configflags%julyr, %julday, %gmt: 2019 239 1.000000
metgrid input_wrf.F first_date_input = 2019-08-27_01:00:00
metgrid input_wrf.F first_date_nml = 2019-08-27_00:00:00
d01 2019-08-27_01:00:00 input_wrf.F: SIZE MISMATCH: namelist ide,jde,num_metgrid_levels= 100 100 32
d01 2019-08-27_01:00:00 input_wrf.F: SIZE MISMATCH: input file ide,jde,kde = 100 100 34
d01 2019-08-27_01:00:00 ---- ERROR: Mismatch between namelist and input file dimensions
NOTE: 1 namelist vs input data inconsistencies found.

But my met_em* file has exactly 32 vertical levels, which I have specified in the namelist.input file.

Any idea on how kde = 34 comes from since I have not set it.

Request any help go overcome this problem

The namelist.input, rsl files and header from met_em* are attached.

Thanking you,
Warm Regards
Roopa

 

[kwerner]

Hi,

I have moved this inquiry from your original post to a new topic since this is a new problem. We like to keep threads organized to help future users search for problems efficiently.

Can you also check the met_em.d01 and met_em.d02 files to make sure they also have only 32 levels?

If they do, will you please issue the following:

ls -ls >& log.ls.wrf

in the wrf running directory, and

ls -ls >& log.ls.wps

in the wps directory, and then please attach those log files? Will you also please attach the three met_em.d01(d02/d03).2019-08-27_00:00:00 files?

Thanks

 

[sroopa]

Dear Sir,
Thanks very much for our reply. I have checked the met_em* files for dom 1 and 2 and they have exactly 32 levels. I am attaching the log files you mentioned.

I would also like to mention that this problem is with simulation using GFS 0.25 degree data for initial and boundary conditions obtained from https://www.nco.ncep.noaa.gov/pmb/products/gfs/.

Just for my curiosity, I did the same run using exactly same namelist.input file but with FNL 0.25 degree data and it was completed without any problem. In the simulation with FNL data, num_metgrid_levels is 34.

Thanking you for your time and consideration,
Warm Regards
Roopa

 

[kwerner]

Thanks for sending those. Can you also attach the following files:

met_em.d01.2019-08-27_00:00:00.nc
met_em.d01.2019-08-27_00:00:00.nc
met_em.d02.2019-08-27_00:00:00.nc
met_em.d02.2019-08-27_00:00:00.nc
met_em.d03.2019-08-27_00:00:00.nc
met_em.d03.2019-08-27_01:00:00.nc

Thanks!

 

[sroopa]

Dear Sir,
Required files are attached.

Thanks
Warm Regards
Roopa

 

[kwerner]

Roopa,
Thank you for sending those. I noticed that the size of your met_em* files initial time was slightly smaller than that of all the other times. For e.g.,

 19396 -rw-r--r-- 1 roopa pusers  19857972 Sep  4 10:36 met_em.d01.2019-08-27_00:00:00.nc
 19856 -rw-r--r-- 1 roopa pusers  20330004 Sep  4 10:37 met_em.d01.2019-08-27_01:00:00.nc
 19856 -rw-r--r-- 1 roopa pusers  20330004 Sep  4 10:37 met_em.d01.2019-08-27_02:00:00.nc
 19856 -rw-r--r-- 1 roopa pusers  20330004 Sep  4 10:38 met_em.d01.2019-08-27_03:00:00.nc

So I wanted to check the levels in the initial time, and the next time to compare. It looks like your met_em.d0*.2019-08-27_00_00_00.nc files all have 32 levels, but the next time (for each domain) has 34 levels, which is causing the error you are seeing. I assume this is the way the data was when you downloaded it, though I'm not sure why. You may want to contact someone in the support group for the team that manages those data files and see if there is an explanation or a way around it. If you figure anything out, please let us know so that it may help future users with the same problem. Thanks!

 

[sroopa]

Dear Sir,
Thank you very much for your reply ans insight on the error. I have posted the query on GFS "contact us" page. As soon as I here from them, I will post the same on this forum too.

Thanks again,
Warm Regards
Roopa

 

[sroopa]

Dear Sir,
Basically the error occurred since the number of vertical levels in gfs*.anl and gfs*.f000 files are different. For running WRF, it is better to use gfs*.f000 file for initializing the model rather than gfs*.anl. More information is available here https://ncarrda.blogspot.com/2015/04/whats-difference-between-gfs-and-fnl.html.

Thanks again
Warm Regards
Roopa

 

[kwerner]

Roopa,
Thank you so much for the update!

 

[sroopa]

Dear Sir,
You are most welcome. I am indeed thankful to you for being supportive and patiently answering my queries.

Warm Regards
Roopa

 

[PhungQuangNghia3108]

Hi there, I've met the same trouble and I'm tryna fix it up.
I got an idea to check num_metgrid_levels:
Type ncdump -h met_em.d01.2020-07-01_00\:00\:00.nc | grep -i metgrid,
This form for this stuff is that: ncdump -h <your_met-em_file> | grep -i metgrid
And u will recieve something like this:
num_metgrid_levels = 34 ;
float PRES(Time, num_metgrid_levels, south_north, west_east) ;
float GHT(Time, num_metgrid_levels, south_north, west_east) ;
float RH(Time, num_metgrid_levels, south_north, west_east) ;
float VV(Time, num_metgrid_levels, south_north_stag, west_east) ;
float UU(Time, num_metgrid_levels, south_north, west_east_stag) ;
float TT(Time, num_metgrid_levels, south_north, west_east) ;
:TITLE = "OUTPUT FROM METGRID V3.4.1" ;
:NUM_METGRID_SOIL_LEVELS = 2 ;
:FLAG_METGRID = 1 ;
I mean that's what we need,
Thanks,

 

[kwerner]

@PhungQuangNghia3108,
Can you open up a new forum topic for your problem and make sure to attach your namelist.input file, along with your real.exe output/err file (e.g., rsl.error.*), along with the details of your run? Thanks!

 

[PhungQuangNghia3108]

yep, i'll open it up soon, i have been in big trouble <3

 

[kgassi]

Hi all,
I've met the same trouble and I'm trying to fix it up can any one help me!! pleas it's kind of emergency!!!
well in its always about real.exe the same problem " starting wrf task 0 of 1" and when i checked the rsl.error.0000 i found the flowing lines alloc_space_field: domain 1 , 197623856 bytes allocated
d01 2022-05-16_00:00:00 Yes, this special data is acceptable to use: OUTPUT FROM METGRID V4.0
d01 2022-05-16_00:00:00 Input data is acceptable to use: met_em.d01.2022-05-16_00:00:00.nc
metgrid input_wrf.F first_date_input = 2022-05-16_00:00:00
metgrid input_wrf.F first_date_nml = 2022-05-16_00:00:00
d01 2022-05-16_00:00:00 input_wrf.F: SIZE MISMATCH: namelist ide,jde,num_metgrid_levels= 74 61 32
d01 2022-05-16_00:00:00 input_wrf.F: SIZE MISMATCH: input file ide,jde,kde = 74 61 34
d01 2022-05-16_00:00:00 ---- ERROR: Mismatch between namelist and input file dimensions
NOTE: 1 namelist vs input data inconsistencies found.
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 1248
NOTE: Please check and reset these options
-------------------------------------------
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
[unset]: write_line error; fd=-1 buf=:cmd=abort exitcode=1
:
system msg for write_line failure : Bad file descriptor
The namelist.input;WPS, rsl files are attached.

Thanking you,

 

[kwerner]

@kgassi,
Can you please issue this series of commands:

foreach f (met_em*)
ncdump -h $f | grep BOTTOM-TOP_GRID_DIMENSION
end

and then please send me the print out for that. Please also attach your namelist.input file. Thanks!