MOZBC trying to a read file that does not exist

[Erica Burrows]

Hello,

I am a student at the University of Alabama in Huntsville attempting to run mozbc to fix an issue with my gocart files (originally I was using the NU-WRF model to create my GOCART files but they do not show any dust emissions for the 2 day period which is not correct).

My steps before running mozbc were of course to make the met_em files from WPS, then I ran real without chemistry. I then ran MEGAN to produce my necessary emissions files. All of those files are located in my WRF directory specified in the input for mozbc.

Both with and without the addition of "def_missing_var = .true." within my MOZCART.inp file, I get the error: 

 failed to open /nas/rhome/eburrows/Models/WRF3.9/MOZ/Data/h9999.nc

 No such file or directory

My file is named h0000.nc (originally I had h00001.nc). I understand mozbc is trying to find a file for a previous time stamp, however my entire time period is located within that file (I downloaded the data from https://www.acom.ucar.edu/wrf-chem/mozart.shtml and renamed the file).

Is there any way to prevent mozbc from doing that? Or did I download the wrong data? I ultimately want to run MOZCART.

Attached are my input and output files.

Thankyou for your help.

Erica

[Gabriele Pfister]

Hi Erica, 

you should keep the filename as h0001.nc, if you use h0000.nc then mozbc will not know how to increment the file.
Have you tried to run:
ncdump -v date,datesec h0001.nc
to check that your MOZART global fields indeed cover the time period of your wrfinput and wrfbdy files?

Gabriele

--

================================
Gabriele Pfister, Deputy Director

Atmospheric Chemistry Observations & Modeling

National Center for Atmospheric Research

Email: pfi...@ucar.edu

Phone: +1 303 497 2915

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

[Erica Burrows]

Gabriele,

Yes the issue was that it did not contain the first time stamp. Thank you!

However now I am having another issue. I am on the same step and I am trying to run MOZCART. I am not quite sure what the issue is this time. The last line says it cannot find the variable within my h0001.nc file, do I need to download something else? Or should I turn off that variable (which does not sound right)?

I have attached the new input and output files. 

[Gabriele Pfister]

Hi Erica

the problem here is simply that in the provided mapping we include most species that are in the chemical mechanisms, while in the global MOZART output we only archive longer-lived species since the short-lived compounds will have negligible effect for boundary conditions. You can take species like O_VMR_inst out of your mozbc.inp as well as all the others that mozbc does not find in your h0001.nc and that are short-lived (e.g. OH, HO2, ...). 

Gabriele

[Erica Burrows]

Thank you. That worked. 

Erica