Require help running WRF-Chem

[Akhtar Memon]

I am running a 37 days simulation but the simulation stops without giving any proper error. I have tried a lot to set the error but still the simulation stops. I am attaching a namelist and an error file.  It might be due to less computational resources that is why I am mentioning that I have  a system with 24 CPU cores and 64 GB RAM. If the error is due to computational constraints, let me know how much CPU cores and RAM is required to run the simulation smoothly.

Note: I am running WRF-Chem with EDGAR, MEGAN, FINN and mozbc for anthropogenic emissions, biogenic emissions, fire emissions and IC/BC respectively.

--

Regards

Akhtar Ali Memon

Cell # : +923322788727

[Gabriele Pfister]

Hello Akhtar

a few suggestions: 

- set debug_level to 12 or higher, this way you will get more information on the potential error in the rsl files

- when you run for such a long time you might want to use at least grid nudging (fdda), otherwise your meteorology will drift from the real situation

- test your setup without chemistry and make sure it runs

- reduce frames_per_outfile = 1000 else your wrfout files will be much too large. My suggestions is to use no more than frames_per_outfile = 24

- are you indeed only interested in output once a day (history_interval         = 1440)? 

I cannot give you advice on the computing needs and I am also not familiar with chem_opt=34. As mentioned previously, we only can help with the by us supported suite of MOZART schemes and the related processors. chem_opt=34 is also a non-KPP implementation and these are very old and have not been revisited in a long time. 

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