Megan Segmentation fault problem

[Ediclê Fernandes]

Dear all,

I'm working with 3 Lambert domains (see attached namelist.wps), wrfinput_d01 file attached and mega.data in the link below.

The Megan preprocessor compiled normally as suggested by README.bio_emiss file.

https://www.acom.ucar.edu/wrf-chem/temp/wrfchem-20230622040631308109/megan.data.tar.gz

But when I run MEGAN I have the error message

  Finished megan2 bio emiss dataset shr200121_30sec.nc
 map_megan2_emissions: Before write_bioemiss
 map_megan2_emissions: After write_bioemiss
 
map_megan2_emissions:   3 cached grid(s)

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
#0  0x1531e4c23860 in ???
#1  0x1531e4c22a05 in ???
#2  0x1531e4854def in ???
#3  0x1531e48af94e in ???
#4  0x420712 in map_megan2_emissions
at /home/eduarte/WRFCHEM_02/MEGAN/bio_emiss.f90:242
#5  0x40391c in main
at /home/eduarte/WRFCHEM_02/MEGAN/bio_emiss.f90:9
--------------------------------------------------------------------------
Primary job  terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun noticed that process rank 0 with PID 0 on node pyro exited on signal 11 (Segmentation fault).

I already looked at the gmail wrf-chem-bio_emiss group but I could not find a solution for this problem.

I correctly installed megan prep in another supercomputer, but it gives me the same problem. I even downloaded smoller megan.data to see it solved, but it doesn't.

Please, do you know any solution for this problem? I couldn't share this essuie in gmail wrf-chem-bio_emiss because I do not have permission yet.

Thank very much!

[Ediclê Fernandes]

Is there any problem with dataset shr200121_30sec.nc? Because always break when reaches here.

[Gabriele Pfister]

Hello Ediclê

I tested your wrfinput_d01 and megan_bioemiss.inp on our computer and it worked just fine. Do you run into problems with the single domain or only when you run all three domains? If only the latter, can you please also share the wrfinput_d02 and wrfinput_d03. 

Gabi

--

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

Atmospheric Chemistry Observations & Modeling (ACOM) Lab

National Center for Atmospheric Research (NCAR)

Email: pfi...@ucar.edu

Temporary Home Office Phone: +1 303 241 1329

Work Phone: +1 303 497 2915

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

[Ediclê Duarte]

Thank  you for answering, Gabi.

I tried to simulate only with wrfinput_d01. Didn't work.

I also tried to simulate it for only 2 months. Didn't work.

Attached are wrfinput_d02 and d03.

I tried to run megan on 2 supercomputers and on my personal computer.

The model compiled normally, but keeps giving me the same segmentation fault problem.

The compile is gfortran for all computers.

Thank you for your feedback and help, Gabi.

 wrfinput_d02

 wrfinput_d03

Dr. Ediclê Duarte (Postdoc)

Universidade de Évora - Instituto de Ciências da Terra

University of Évora - Institute of Earth Sciences

Rua Romão Ramalho, 59
7000-671 Évora
Portugal

http://www.ict.uevora.pt/g1/

https://orcid.org/0000-0002-2785-6648

[Gabriele Pfister]

Hi Ediclê

the tool also works fine when I am running all three of your domains. Unfortunately I cannot help you with solving your issue with running the tool, but I attach the wrfbiochemi* files for your setup so that you can continue your runs. 

I hope this works, 

Gabi

[Ediclê Duarte]

Thank you so much, Gabi!

Unfortunately, I couldn't generate these files yet. When I figure it out, I'll share it here for those who may have the same problem.

You helped a lot!

I'll try these wrfbiochemi files for now. 

Many thanks!