Re: [WRF-Chem] Error reading wrffirechemi files

[Gabriele Pfister]

Hello Vikas

I cannot see anything obviously wrong in your namelist as it relates to the fire emissions and not sure why it looks for a different filename. However, I am also not familiar with chem_opt=301 and we only can help with the by us developed suite of MOZART chemistry options. 

You mention that you experience these issues when running real.exe. However, you would not need to run real,exe with chemistry. I assume you must have run real.exe for meteorology only  to get wrfinput_d01 as needed to create your input emissions. So why are you running real.exe again? And do you experience the same issue when running wrf.exe? 

One other thing I noticed is that you have set phot_opt=0. Do you intentionally run without photolysis? 

Gabi

On Sun, Sep 8, 2024 at 1:43 AM vikas _ <vikas....@gmail.com> wrote:

Hi all, 

I am trying to run wrf-chem with chem_opt= 301, I have created anthro, biogenic, and fire emissions using anthro_emiss, bio_emiss and fire_emiss utilities. 

While running real.exe, i am getting the error:

(base) [wcrg23@login03 em_real]$ tail rsl.error.0000
d01 2022-11-01_00:00:00  DEBUG wrf_timetoa():  returning with str = [2022-11-01_00:00:00]
 mediation_integrate: med_read_wrf_chem_fireemissions: Open file wrffirechemi_d01
d01 2022-11-01_00:00:00 module_io.F: in wrf_open_for_read
d01 2022-11-01_00:00:00  NetCDF error: No such file or directory
d01 2022-11-01_00:00:00  NetCDF error in wrf_io.F90, line        1077
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE:  mediation_integrate.b  LINE:    2885
 med_read_wrf_chem_fireemissions: error opening wrffirechemi_d01
-------------------------------------------
Abort(1) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0

Seems like real program is looking for wrffirechemi_d01 instead of  wrffirechemi_d<domain>_<date> even though 
auxinput7_inname                     = 'wrffirechemi_d<domain>_<date>',

is set in the namelist.

I have tried running the real by linking the first hour wrffirechemi_d01_2022-11-01_00:00:00 to  wrffirechemi_d01, which strangely works even with 

auxinput7_interval_m                 = 60,

frames_per_auxinput7                 = 1,

I could not figure out whether the problem is related to the chemistry settings in the namelist.input file or the emissions input files.

Please guide if you have expertise on this.

Thank you

--

================================
Gabriele Pfister

Atmospheric Chemistry Observations & Modeling (ACOM) Lab

NSF National Center for Atmospheric Research (NCAR)

Email: pfi...@ucar.edu

Work Phone: +1 303 497 2915

Web: https://staff.ucar.edu/users/pfister

[vikas _]

Hi Gabi,

Thank you so much for your feedback and observations!

I'm currently following the instructions from WRF-Chem Exercise 6 along with a YouTube video on WRF-Chem/mozbc:

• WRF-Chem Exercise 6
• YouTube Video

Regarding real.exe, you're correct that I initially ran it just for meteorology to generate the wrfinput_d01 file. The subsequent runs were aimed at incorporating chemistry data and updating it for certain gaseous species and particulate matter using the mozbc utility.

Earlier, I used chem_opt=112 for the same emission dataset, where the second run of real.exe was successful, and the WRF output seemed fine.

During troubleshooting, I turned off some variables one at a time to pinpoint the issue and accidentally left phot_opt=0. I’ll be using phot_opt=1 for the final simulations, as photolysis is definitely necessary.

Interestingly, the WRF output seems to run fine without updating wrfinput and boundary files for chemical inputs, so I’m starting to think the issue might lie in the namelist settings.


I would really appreciate your thoughts on this.

Thanks again for your help!

Best regards,
Vikas

[Gabriele Pfister]

Hi Vikas
I was not aware of the youtube video but you will not need to run real.exe with chemistry for mozbc. You only need to set def_missing_var = .true. in your mozbc namelist input and the tool will add any missing variables to wrfinput and wrfbdy. See the README that comes with mozbc for more information.

Gabi