Thanks for a nudge from Fred, I've taken a new look at some
lightning data being collected by Blitzortung.org. You can see
their live lightning map at:
Note: You can do this in a new APRSIS32 instance (just copy the .EXE to a new empty directory) that can run n a receive-only mode (let the passcode stay at -1).
For instance, here's a current snapshot of a storm system off the central east coast of the US. I think "Hamilton" might be in Bermuda? Wherever it is, there's an island down there.
I'm soliciting suggestions for improvements on this. One I'm already exploring is to change the gridsquare box to a "convex hull" (google it) shape surrounding just the strikes in the square. For some storms this will give a better indication of the collection of the strikes.
I'm also likely going to drop off the individual strike objects in favor of just the gridsquare storm outlines. There's just way too many strikes to sustain this for a really long time. Strikes are names based on the hours/minutes/seconds that they occurred with an added character to account for quick strikes, but this will quickly make lots of unique IDs.
Both strikes and storms are killed (15 minutes for the former, whenever strike count goes to zero for the latter), but sometimes a few may be left orphans if I restart my generator (which is based on APRSISMO, BTW). Strike object data is persistently recorded so they should be killed after a restart, but the storm objects are not yet persisted, so if the strikes go to zero while I'm not running, you'll be left with a gridsquare object that just needs to age out.
I'm anxious to see how the gridsquare objects pan out count-wise with an eye towards possibly moving them to the APRS-IS once the whole thing is stable. Don't count on that, however, as firenet is more the host for such volumes of data.
Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32