problem ./real.exe for WRF+SLUCM with NARR as input

[lslrsgis]

Dear WRF Community,

I am writing for a question on WRF_SLUCM. Version 4.0 WPS/WRF is used, while NARR is treated as initial/lateral boundary. After ./real.exe:

lsl@lsl-OptiPlex-7060:~/WRF/WRFV4.0_UCM/test/em_real_TEST_NEWYORK_NARR$ ./real.exe
Ntasks in X 1 , ntasks in Y 1
REAL_EM V4.0 PREPROCESSOR
*************************************
Parent domain
ids,ide,jds,jde 1 293 1 292
ims,ime,jms,jme -4 298 -4 297
ips,ipe,jps,jpe 1 293 1 292
*************************************
DYNAMICS OPTION: Eulerian Mass Coordinate
alloc_space_field: domain 1 , 2209130348 bytes allocated
Time period # 1 to process = 2006-07-12_00:00:00.
Time period # 2 to process = 2006-07-12_06:00:00.
Time period # 3 to process = 2006-07-12_12:00:00.
Time period # 4 to process = 2006-07-12_18:00:00.
Time period # 5 to process = 2006-07-13_00:00:00.
Total analysis times to input = 5.

-----------------------------------------------------------------------------

Domain 1: Current date being processed: 2006-07-12_00:00:00.0000, which is loop # 1 out of 5
configflags%julyr, %julday, %gmt: 2006 193 0.00000000
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 401
error opening met_em.d01.2006-07-12_00:00:00.nc for input; bad date in namelist or file not in directory
-------------------------------------------
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 401
error opening met_em.d01.2006-07-12_00:00:00.nc for input; bad date in namelist or file not in directory
-------------------------------------------

However, these files generated by WPS V4.0 do exist in the folder.
(base) lsl@lsl-OptiPlex-7060:~/WRF/WRFV4.0_UCM/test/em_real_TEST_NEWYORK_NARR$ ls -l met_em.d0*.2006-07-12*
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d01.2006-07-12_00:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d01.2006-07-12_00:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d01.2006-07-12_06:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d01.2006-07-12_06:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d01.2006-07-12_12:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d01.2006-07-12_12:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d01.2006-07-12_18:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d01.2006-07-12_18:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d02.2006-07-12_00:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d02.2006-07-12_00:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d02.2006-07-12_06:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d02.2006-07-12_06:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d02.2006-07-12_12:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d02.2006-07-12_12:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d02.2006-07-12_18:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d02.2006-07-12_18:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d03.2006-07-12_00:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d03.2006-07-12_00:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d03.2006-07-12_06:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d03.2006-07-12_06:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d03.2006-07-12_12:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d03.2006-07-12_12:00:00.nc
lrwxrwxrwx 1 lsl lsl 64 3月 31 22:33 met_em.d03.2006-07-12_18:00:00.nc -> /home/lsl/WRF/WPS_NARR_NEWYORK/met_em.d03.2006-07-12_18:00:00.nc

Anyone can tell me why this happens? Thanks.

LSL

P.S. namelist.input is attached. In this file, urban option is not open yet.

 

[kwerner]

Hi,
The packaged files you sent don't seem to be the correct ones for the problem you are mentioning. The namelist you sent is for a 2015/2016 date range, and the rsl.* files are for a wrf.exe run with a segmentation fault.

1) I assume if the error is related to a problem with the met_em* files, this is likely a real.exe error, and not a wrf.exe error. Is that correct?
2) Can you please send the correct files?
Thanks.

 

[lslrsgis]

Yes, I have uploaded the wrong package.

The namelist.wps and namelist.input (with log file for ./real.exe) are updated now. Thanks.

 

[kwerner]

Hi, Thanks for sending the correct files. Can you go into your /home/lsl/WRF/WPS_NARR_NEWYORK/ directory and issue

ls -ls met_em* >& ls.txt

and send me that ls.txt file? Can you also attach your file met_em.d01.2006-07-12_00:00:00 so that I can run a test? Thanks!

 

[lslrsgis]

The list_met_em.txt is attached.

met_em.d01.2006-07-12_00_00_00.nc can be found throught the link:
link：https://pan.baidu.com/s/1NTzaGEf5_rwPS7HrF_NZVw
code for extraction：shku

Thanks.

LSL

 

[kwerner]

Hi,
Unfortunately when I click the link to get the met_em file, all the wording is in a language I am unable to read, so I don't know what I'm clicking on. Is there a way that you can put it somewhere with English wording? If not, and if it's too large to attach here, there are instructions on the home page of this forum for uploading larger files. Thanks.

 

[lslrsgis]

Thanks for the quick response. I have uploaded the met_em.d01.2006-07-12_00_00_00.nc file to https://nextcloud.mmm.ucar.edu/nextcloud/index.php/s/DkPamzUKYhQkWEI#.

 

[kwerner]

Hi,
Thanks for sending that. I was able to get it. I finally figured out the problem, and apologize for not recognizing this to begin with. The met_em* files you are using are in the name format:
met_em.d01.2006-07-12_00_00_00.nc
and the real.exe program is looking for the format:
met_em.d01.2006-07-12_00:00:00.nc
(with colons instead of underscores between the hours/minutes/seconds). You'll need to change the format of the file names and then you should be able to get past this problem.