WRF-Chem v4.5 crash: “Error trying to read metadata” with wrffirechemi (auxinput7

[Sakshi Kumari]

Dear all,

I am running WRF-Chem v4.5 with FINNv2.5.1 fire emissions (wrffirechemi as auxinput7) for 20 Sep–30 Nov 2024 on a single domain (dx = 27 km, max_dom = 1).

The model runs for several days but crashes (late Sep–early Oct) with:
“Error trying to read metadata” followed by MPI termination. Changing time_step only shifts the crash timing.

The wrffirechemi_d01_YYYY-MM-DD_HH:00:00 files exist, and checked files have matching internal Times. However, the failure consistently occurs when reading later timesteps.

Note: although max_dom = 1, several namelist variables (e.g., auxinput7_interval_m, frames_per_auxinput7, and chem options) still contain multiple entries.

Any suggestions on diagnosing whether this is due to file inconsistency or namelist configuration would be appreciated.

Best regards,
Sakshi Kumari

[Gabriele Pfister]

Hello Sakshi

what makes you think the wrffirechemi files are the issue? Does your model run fine without fire emissions? Did you use fire_emiss to generate the emissions? 

It would also be helpful if you share your namelist.input and an example wrffirechemi file. 

Gabi

--

================================
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

[Sakshi Kumari]

Dear Gabriele,

Thank you for your response.

The model stops with the following error during runtime:

FATAL CALLED FROM FILE: <stdin> LINE: 315
Possibly missing file for = auxinput7

The fire emissions were generated using fire_emis from FINNv2.5.1 data, and the wrffirechemi files are linked in the run directory.

I will share my namelist.input and an example wrffirechemi file output for further inspection.

Best regards,
Sakshi Kumari

[Gabriele Pfister]

Hello Sakshi

there seem to be some issues in your namelist. 

Just a few points I noticed:  I see that you are using chem_opt=303 (RADM2 with GOCART). Since we can only support the by us developed MOZART chemistry schemes, I cannot tell you whether this option has been fully connected to the online plumerise, and work with FINN fire emissions. biomass_burn_opt =2 is defined for the MOZCART chemical scheme and might not work for chem_opt=303. Also, when using FINN emissions with online plumerise you always have to set scale_fire_emiss  = .true., otherwise you would be double counting the fire emissions.  

phot_opt=1 is an outdated version, when running with any of our MOZART chemistry schemes we strongly recommend using phot_opt=4. 

I suggest carefully reviewing your namelist.input and make sure that you have settings appropriate for the chemical option of your choice. 

Good luck, 

Gabi