xrun.scr comparison

Post any questions or comments regarding the LINUX version of HYSPLIT. This includes the model execution, GUIs, results, or graphics. Be sure to mention the version of LINUX you are using.
Post Reply
Posts: 1
Joined: February 12th, 2015, 3:31 pm
Registered HYSPLIT User: Yes

xrun.scr comparison

Post by dustinthewind » February 12th, 2015, 3:57 pm

I was able to compile and run xrun.scr on my linux box (RHEL).

A few things that I noticed:

1) On page 15 of the canonical file, there is a plot that says 'Dispersion (UNIX ONLY) using 4 processors....The corresponding page in my file has this strange rose plot that talks about height. I wasn't using/running MPI on my linux system.

2) There is an extra plot of model particle cross sections in my file for the Volcano Soufriere Hills simulation.

3) For the dispersion calculations, I got different minimum values, and sometimes different maximum values. I assume that this is due to the difference in RNG used and/or RNG state. Usually I can tell that I am modeling 'the same thing', just there are some slight differences in the shapes of the fields. I'm assuming that this is normal, but wanted to check.

Posts: 647
Joined: November 7th, 2012, 3:14 pm
Registered HYSPLIT User: Yes

Re: xrun.scr comparison

Post by ariel.stein » February 13th, 2015, 12:06 pm

1) If you did not compiled the mpi version then you should get a message in the ERROR_LOG.txt file that says "Multi-processor executable (310) not available"
2)Not sure why you are getting an extra plot.
3)It is entirely normal to get slight differences when using different computers and OS. Different fortran compilers produce different random numbers used to calculate the dispersion of the particles.

Post Reply