HYSPLIT Forum: hysplitbbs.arl.noaa.gov

I'm trying to cluster trajectories for a couple of EPA CASTNET ozone sites over a three year period (about 1050 back trajectories for each site). I'm using the 64 bit version of HYSPLIT (June 2015/svn 761). The length of each trajectory is 72 hours. For one monitoring site I was able to get all the trajectories processed/clustered. On the second monitoring site, I hit the "Run cluster analysis" button on the userform. After a short period I received the following error message:
The cluster analysis continued to run until 180 passes were completed and then stopped. I tried this process on another computer and got the same error message? Any ideas on what I might be doing wrong? Thanks in advance for your help!!

Erik Gribbin

First off, I have to apologize for posting the same problem that a different user posted a few months ago and the same problem that I apparently had a couple of years ago (time flies and I'm getting older I guess).

That being said, I followed Ms. Stunder's advice ran cluster.exe from the command line. I did not get a more enlightening error message, but I did notice something else -- the clustering process works consistently! Is it possible that there is something in the GUI that is in inhibiting the clustering process for larger numbers of trajectories?

Erik

Are you saying the clustering worked when you ran it from the command line, but not in the GUI?

Correct! Clustering with about 1050 trajectories is working consistently from the command line, but not from the GUI.

Erik Gribbin

Well then it must be something in the GUI, but I don't know offhand what it would be. For now run cluster.exe from the command line, and then you can go back to the GUI for the post-processing.

Will do!

Thanks Barbara!

Erik Gribbin