Display of final results and intermediate stage fields

59 views
Skip to first unread message

nininini

unread,
Nov 19, 2024, 8:04:38 AM11/19/24
to european-calpuff-user-discussion-group
Hello everyone, 

I have recently been delving into the world of CALPUFF. Like any new area, there are things that I don't understand completely well, however, I had some doubts regarding the display of the fields of the files generated throughout the process.

Do you somehow take a look at the generated files to see if everything is in order, and by this I mean graphs in figures?

If you use MMIF, do you take a look at what is generated in this part before using it to run CALPUFF?

Or if you use CALWRF, do you somehow see what is generated here before using CALMET? And do you see what is generated in CALMET before using it in CALPUFF?

Or on the other hand, when executing these programs, if there is no error message, it is simply assumed that everything is correct and there is no need to see the generated fields, even out of curiosity?

Well that is a concern... another has to do with knowing if you usually use CALPOST or do you simply stick with the CALPUFF outputs?

Finally, what software do you use to generate figures, either in the intermediate fields (if any) or for the final CALPUFF/CALPOST files?

Do you use any software such as MATLAB or PYTHON to analyze the generated fields?

I'm using Linux and focusing on using WRF outputs to use in CALPUFF, so my concerns are on that side...


Kind regards

Roberto Zambianchi

unread,
Nov 19, 2024, 8:22:25 AM11/19/24
to european-calpuff-user-discussion-group

I describe my experience: over 7 years as a CALPUFF user without a graphical user interface (GUI).

  • Checking error messages in the LST file.
  • Verifying every output file (the depth of the check depends on the case: complex orography requires more scrutiny than flat terrain);
  • Check the rosewind on different place of the domain.
  • (You could create a routine of your check)
And than I run CALPUFF. 
For generating figures and statistics, I use QGIS + Python: you can easily create scripts to calculate statistics and produce contour maps or other outputs as needed. For tabular results, you can use R or Python, or create a standard template with Excel.

Regards

Roberto

Jana Krajcovicova

unread,
Nov 19, 2024, 10:55:13 AM11/19/24
to european-calpuff-us...@googlegroups.com
Hi nininini, 
Your concerns are very legitimate :-). 
Unfortunately, the binary output files from CALMET and CALPUFF are not very user friendly. It is necessary to use  PRTMET and CALPOST in order to get ASCII outputs, which are also very awkward and take up a lot of disk space. 
I also run the models on Linux (HPC), and I wrote a python script for running whole year of CALPUFF hourly output to a single NETCDF file (similarly for CALMET ground data and all levels). NETCDF files can be easily analyzed by plotting or performing statistics in python. I suggest you doing the same, if you plan to spend a lot of time with those models. 

Best regards, 

Jana  

--
You received this message because you are subscribed to the Google Groups "european-calpuff-user-discussion-group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to european-calpuff-user-dis...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/european-calpuff-user-discussion-group/b0a7dfe4-68f1-48b5-b8cd-8f41583f7bbdn%40googlegroups.com.

nininini

unread,
Nov 21, 2024, 5:24:09 PM11/21/24
to european-calpuff-us...@googlegroups.com
Dear Roberto and Jana, thank you for your comments.

It's good to hear from your experience that it seems similar to the way I think I will adopt when working.

Jana, regarding the use of PRTMET, did you have errors when compiling it? I am encountering non-trivial problems that go beyond selecting and commenting the lines according to my operating system in the calutils.for file.


Greetings and thank you very much

Jana Krajcovicova

unread,
Nov 28, 2024, 9:55:51 AM11/28/24
to european-calpuff-us...@googlegroups.com
Hi, 
I haven't compiled PRTMET personally, it has been done by one of my colleagues a rather long time ago. I don't remember mentioning her any problems with the compilation. 

Good luck,



Reply all
Reply to author
Forward
0 new messages