Different heights and duration time of trajectories

Topics about the trajectory clustering program for HYSPLIT.
Post Reply
wuy
Posts: 5
Joined: April 8th, 2014, 12:50 pm
Registered HYSPLIT User: Yes

Different heights and duration time of trajectories

Post by wuy »

One question is about if or not I can make a cluster analysis on those backward trajectories starting from
the same site (same lat and longitude) but having different heights and duration time.

Though the HYSPLIT/cluster can be run through, such cluster analysis process make sense or not?

Thank you.

Wu
barbara.stunder
Posts: 451
Joined: November 9th, 2012, 4:23 pm
Registered HYSPLIT User: Yes

Re: Different heights and duration time of trajectories

Post by barbara.stunder »

The cluster process requires trajectories of the same duration. If one set of trajectories has duration d1 and another set d2, where d2 > d1, then in the 2nd set of trajectories only the endpoints out to time d1 are used.

I haven't run clusters with trajectories at different start heights. If you are looking at transport within the boundary layer, it would make sense to use different start heights, because no single level is representative. However, with different winds at different heights, the resulting clusters may to some extent correlate with the trajectory start heights.
wuy
Posts: 5
Joined: April 8th, 2014, 12:50 pm
Registered HYSPLIT User: Yes

Re: Different heights and duration time of trajectories

Post by wuy »

Dear Barbara,
Thank you.
For sure, the transport duration time of aloft aerosol plumes are different due to their height differences.

To get more reliable cluster results, should I set the same duration time and same starting height for the
HYSPLIT backward trajectory run?
I doubt that with the same starting height, the wind speed might be different on the different days (trajectories).

Wu
barbara.stunder
Posts: 451
Joined: November 9th, 2012, 4:23 pm
Registered HYSPLIT User: Yes

Re: Different heights and duration time of trajectories

Post by barbara.stunder »

The trajectories being clustered must have the same duration. Wind speed usually varies with height and time, and if so, the trajectory lengths will vary both by start height for a given start time, and by start time for a given start height. The problem is that no one height is representative of transport in the boundary layer.
Post Reply

Return to “Cluster Analysis”