LOL....

Well with a fast computer it's always easier to propose the "brute force computer power" approach first. In this way, one doesn't have to strain one's brain too much....

Another more efficient approach might use vector math operations (cross and dot) to identify the branch points for each "earthquake christmas tree" ringmap structure. One would just check earthquake pairs and then drop a red icon where the ringmap disks / outer edge (i.e. branch tips) are within a certain reach of each other in 3d space. The accumulated red icons would define the morely severely overlapped regions while skipping the more consuming previously mentioned. Call it the "quicker and dirtier" approach...(after it's been worked out).

The other approach would be to only examine the overlap regions for high population density areas. This would drop the computer load down bigtime (but might require shadow arrays to dogtag locations near population areas...no biggie with lots of RAM).

I can understand why there is an urge to look at historical seismicity...guess time will tell if it pans out huh ?

Clutter and Confusion...well the brain has an immense capacity to process the information one way or the other...sometimes there is order in chaos but sifting throught it might be easier if one knew exactly which data to look through. Looks like michael has some ideas too.

.02 worth