I have 2 kinds of events I want to examine correlation between, lightning strikes and earthquake events. The lightning strikes are defined by lat lon points and a timestamp, whereas I have represented the earthquakes with buffers of variable radius(corresponding to earthquake magnitude) around an epicentre. I am interested in the time just before a quake happens, and whether seismic processes effect lightning activity. A graphic of my data is included below, where colour gradients represent times.

enter image description here

I feel like the Space Time Pattern Mining toolbox would do what I need, but I only have access to ArcGIS for desktop 10.2.2.

I am able to code stuff in python if needs be but Im not sure where to start.

1 Answer 1


Could you write a script that takes each epicentre and creates a buffer. This buffer selects lightening strike points within that radius, then select a subset of this that is within an appropriate time envelope. You could then run this with different parameters ( increasing buffer size around the epicenter) and also time envelope definition (I.e. before the quake, after the quake and an overlap of before and after). You said you can script in Python so assume you have the knowledge to do this?

You would also need to generate some random dummy points to see if any patterns you see are only associated with quake events. Similarly you would run the earthquake epicentres at random times to ensure lightening activity is due to that quake event and not just the type of landscape.

As a gis exercise it's pretty simple; you just need to run the analysis varying over space at the same time as an earth quake and over varying time on the same space an earth quake has occurred.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.