GIS scripting with backward trajectories

Post user applications of the HYSPLIT trajectory or dispersion model.
Post Reply
mab
Posts: 1
Joined: March 3rd, 2016, 9:43 am
Registered HYSPLIT User: Yes

GIS scripting with backward trajectories

Post by mab »

Hi Hysplit users,

I'm new to Hysplit, so a fascinating world just opened up for me. I am looking in to the effect of wild fires on N deposition. I want to check whether I can explain the seasonality in my N deposition (field measurements) by combining Terra and Aqua fire detections and backward trajectories starting from my measuring spot. For this, I have succeeded in batch generating daily trajectories (backwards, starting at 500 m agl, one week back) for a whole year. Additionally, I have a shape file with daily fire detections in my study area.

So far so good. However, I'm having difficulties now. I'm trying to combine both datasets in such way that I want to count the fire incidents in a 20 km buffer along the trajectory, keeping in mind a one-day buffer. This means that for a parcel arriving at my site on the 7th of april, I want to see where this trajectory passed subsequently on the 1st of April, and whether there was a fire within a 20 km radius on the 31st of March, or 1st of April. Then go one time step further to the 2nd of April and so on..

I can not do this manually, since there are 4 trajectories a day, times 365 days. Additionally there are over a million fires..

I suspect some of you are working with more or less the same concept, so I was wondering if anyone can help me out. And what implementation I should use (Python?? but I'm myself more of an R man...never used Python before...)

Thanks in advance!
Marijn
alicec
Posts: 411
Joined: February 8th, 2016, 12:56 pm
Registered HYSPLIT User: Yes

Re: GIS scripting with backward trajectories

Post by alicec »

Hi Marijn,
Glad to hear that you are enjoying HYSPLIT.
The tdump (trajectory output) files are in ascii format and so it is fairly easy to read them
into any program to do your analysis with. Personally, I really like to use python but you should
be able to use whatever language you like.

In case you decide you want to try python:
I was doing something somewhat similar - looking for computational particles which fell within a certain range of a lidar track. I found the python shapely package useful for this.

Melissa Cross has made some of her python code for HYSPLIT trajectories available on github.
https://www.youtube.com/watch?v=2mzhTC4Kp-Y
https://github.com/mscross/pysplit

Let us know how it goes.

Alice
Post Reply

Return to “User Applications”