Runtime I/O filtering: unexpected internal variables appearing in wrfout_d0*

[Prolay Saha]

Hi,

I am trying to minimize my WRF output to include only rainfall or rainfall related variables using the run-time I/O option.


What I did:
I created the following I/O filtering files:

rain_only_d01.txt (same for d02):

---
-:h:0:LU_INDEX,ZNU,ZNW,ZS,DZS,VAR_SSO,U,...
+:h:7:XLAT,XLONG,XTIME,RAINC,RAINNC,RAINSH
---

And added in namelist.input:

---
&time_control
run_days = 2,
run_hours = 0,
run_minutes = 0,
start_year = 2022, 2022,
start_month = 05, 05,
start_day = 14, 14,
start_hour = 00, 00,
start_minute = 00, 00,
start_second = 00, 00,
end_year = 2022, 2022,


end_month = 05, 05,
end_day = 16, 16,
end_hour = 00, 00,
end_minute = 00, 00,
end_second = 00, 00,
interval_seconds = 21600,
input_from_file = .true., .true.,
history_interval = 60, 60,
frames_per_outfile = 24, 24,
restart = .false.,
!restart_interval = 1440,
restart_interval = 16000,
write_hist_at_0h_rst = .false.,
output_ready_flag = .false.,
io_form_history = 2,
io_form_restart = 2,
io_form_input = 2,
io_form_boundary = 2,
debug_level = 0,
auxinput4_inname = "wrflowinp_d<domain>",
auxinput4_interval = 360, 360,
io_form_auxinput4 = 2,
!!!!!!!!!!!!!! new suggestion for seperate rainfall out

iofields_filename = "rain_only_d01.txt","rain_only_d02.txt",
ignore_iofields_warning = .true.,
auxhist7_outname = "rain_d<domain>_<date>",
auxhist7_interval = 60, 60,
frames_per_auxhist7 = 24, 24,
io_form_auxhist7 = 2,
/
---

What I expected:
Only the selected variables (rainfall + coordinates) in the output (wrfout*).

What I actually get:
Even after filtering, several additional variables still appear in the "wrfout*" file, such as:

---
ITIMESTEP, THIS_IS_AN_IDEAL_RUN, BATHYMETRY_FLAG, O3_GFS_DU,
IVGTYP, ISLTYP, CH, CROPCAT, ISEEDARRAY_SPP_LSM, etc.
---

But rain_d0* files have all the variables that is mentioned via rain_only_d0*.txt files.

I got successful run when testing for 2 day simulations. (attaching my namelist.input and rain_only_d01.txt)

My query is:
Since i am removing lots of fields, is it save or correct way?
If i am doing wrong, then please suggest me the best way to handle default WRF output fields.

Thank you in advance!

 

[kwerner]

Hi, I'd like to try to test the issue you're having. To do that, I need the following:

1. Please attach your full namelist.input file
2. Given that you are getting the variable THIS_IS_AN_IDEAL_RUN, will you also let me know which ideal case you're running?
3. The standard error/out from your wrf run. If you run in parallel and get rsl* files, please package those into a single *.tar or zipped file and attach that. Otherwise, just attach the log file you create (e.g., ./wrf.exe >& wrf.txt - in this case, attach the wrf.txt file)
4. The version of WRF

Thanks!