Page 1 of 1
clustering error
Posted: November 18th, 2015, 12:49 am
by gkrkpm
hi;
I have got trajectories simulated for one source area at 500 m AGL. I have used 69 set of backward trajectories for cluster analysis. I have attached the screenshot for that error
Re: clustering error
Posted: November 18th, 2015, 9:02 am
by barbara.stunder
Open that tdump file in Notepad and confirm there is only one trajectory in the file. Sometimes trajectories starting at more than one level are in a single tdump file. The cluster program cannot input that.
Re: clustering error
Posted: November 18th, 2015, 10:06 am
by gkrkpm
for above trajectory input data vis archive model run.Is this method is correct or not?
total run time-48 hrs
start new trajectory every 6 hrs
max number of trajectories 30
I should no specify start new trajectory within total run time i.e for 48 hrs trajectory a new trajectory is started every 6 hrs .
Regards;
Karthikraj
Re: clustering error
Posted: November 23rd, 2015, 9:28 am
by barbara.stunder
For input to the cluster program there can only be one trajectory in the trajectory endpoints file. With the option "start new trajectory every 6 hours" that writes every trajectory to the one endpoints file, so that cannot be used in the cluster program.
One automated method to get trajectories for input to the cluster program is with the PC/Mac version, under / Trajectory / Special Runs / Daily. See the HELP there and the HYSPLIT Tutorial at
http://www.ready.noaa.gov/documents/Tut ... index.html, particularly Section 5.1 Frequency Analysis.
Links to the PC/Mac versions and archived meteorology are at
http://www.ready.noaa.gov/HYSPLIT.php