+- Blitzortung.org Forum (https://forum.blitzortung.org)
+-- Forum: Public Forums (https://forum.blitzortung.org/forumdisplay.php?fid=29)
+--- Forum: General Discussion (https://forum.blitzortung.org/forumdisplay.php?fid=31)
+--- Thread: Lightning Location Logic Question (/showthread.php?tid=4021)
Lightning Location Logic Question - DrFX - 2025-03-11
- When lightning strikes, and many detectors pick it up, is there a max # of stations that the server will use data from to calculate the location?
- I've heard that a station can detect lightning close by, and then I heard the system will not use lightning data for a strike near by, and instead use data from stations hundreds of miles away?
2.a If you look at the attached station list, it shows stations sorted by closet to furthest for North America, most of them only show lightning hundreds of miles away? Is it possible for the server/station to detect lightning striking near the station?
I'm hoping someone can provide some insight.
Thank you in advanced.
RE: Lightning Location Logic Question - cutty - 2025-03-11
(2025-03-11, 04:03)DrFX Wrote:
- When lightning strikes, and many detectors pick it up, is there a max # of stations that the server will use data from to calculate the location?
- I've heard that a station can detect lightning close by, and then I heard the system will not use lightning data for a strike near by, and instead use data from stations hundreds of miles away?
2.a If you look at the attached station list, it shows stations sorted by closet to furthest for North America, most of them only show lightning hundreds of miles away? Is it possible for the server/station to detect lightning striking near the station?
I'm hoping someone can provide some insight.
Thank you in advanced.
1. Generally minimum of 8 - up to maximum 17, unless the algorithm has changed. I think NASA only uses max 15 decimal points for 'pi' in computations, so 17 data channels should be quite sufficient. And if ALL stations were perfectly the same, we should be able to nail that ground point within meters.... sigh... 'evil' deviation...
For any specific 'impulse', if a station actually sends 'signal data', that increments the station's "Total" count. From that 'signal data', ONE single channel may pass initial analysis, and be classified as "valid" (considered the 'best' channel). If that channel's data matches a minimum of (4), but typically 8+ OTHER stations' "valid" channels, the impulse typically is recognized as a strike ('stroke'), and the station is a 'detector' ("involved") for that stroke... . Up to 17 'detectors' ('locators' or 'primary detectors') are then used for location computation: "used". (Remaining 'involved detectors' then are sometimes referred to as 'secondary detectors'). I'm not about to try and explain 'regional assignments' and totals, since they vary and may change frequently, and are assigned by administrators. Original regions are Europe (1), Oceania (2) North America (3). The 'World' has now many 'regions', Europe now subdivided into Europe 1, 2, and 3, for e.g.
2. Nearby cells (e.g.<30-50km generally contain so much intense activity that a 'normally' configured station should go into interference mode. The activity is so frequent, jumbled, and intense, that it is virtually useless for our purposes.
2a. I personally have been able to disable all but one channel, minimize gains on that channel to bare minimum with high level threshold, and pulled strokes at ±6-12km (3-7mi) which classed as 'detected' a few times. Don't think any were 'used', however. Interesting, but of little value. The network/system is NOT designed to function in that manner. Very best H field data originates with 'ground wave' energy... with best configured stations is typically 50-120km range! Any H channel signal beyond about 80-100 km is likely a reflected impulse of some type, with mixed characteristics, not a 'ground wave' component. Those very long distance signals you note are 'sky waves', and manifest as a 'sombrero' signal, not a 'stroke' charge/discharge 'sharpie'.... . (E channel capable systems (Red and BLUE MAX) set up and configured appropriately, can produce better 'near' stroke data, assuming all H channels are disabled. In fact, 'near stroke data' was part of the rationale for E capability).