Just before seeing this on HN, I randomly came across this article on "dark lightning" -- a burst of gamma rays that accompanies the RF burst that precedes visible lightning.
I believe this one also shows you an approximation of the thunder, so if you are close enough to a strike you can actually tell when you will hear the thunder. We were playing with it over the weekend and it was surprisingly accurate!
The Blitzortung paper mentions that sensor isn't accurate enough to use to determine the distance of a strike, so I don't think a crowdsourced phone-based network could work particularly well:
I also suspect that a phone's RF frontend would be awful for lightning detection as it is tuned for 900+Mhz signals. You can read more about lightning's RF properties here:
but it's clear that above 10Mhz you probably won't learn much of value about the lightning strike.
Elsewhere in this thread, an HN reader found that Nokia were working on lightning detection for phones, but it seems to use the FM radio receiver that was often built into SoCs, which could definitely tune itself to a much more useful frequency range.
Looks like Nokia had some plans (or at least patented something related) in 2007 [1]. According to ComputerWorld "The invention involves the use of radio frequency receivers in mobile phones, such as FM radio and GSM (Global System for Mobile Communications), to detect radio waves emitted by lightning."
Does lightning have any effect on the sensors that detect the geomagnetic field? Maybe even some small anomalies in the readings could be useful if many devices were working together.
It's huge. Normally, the curvature of the earth would make it impossible to detect anything at such distances. But the ionosphere's lower levels act like a waveguide [1], (and this tells you which frequencies you should listen on to take advantage of this).
The craziest phenomenon of this kind is whistlers [2]. Lightning strikes can be heard on radio at a point on the earth exactly symmetrical to the source, using the equator as plane of symmetry. In fact, the signal bounces around back and forth between the two points, following a line on the magnetosphere. The ones closer to the poles have longer paths, and their frequencies can get spread out over 3-4 seconds as they bounce around.
It seems to me that these strikes are happening in clusters at the same time 1000s of miles away. Is this related to radiation from the sun providing the tipping point?
Used this to find webcams of places with thunderstorms rolling in, and found this here in Switzerland [0], quite fun to watch when you skip through the history. Day started out lovely but now they are getting hammered.
I don't know how complete it claims to be (in terms of what fraction of lightning strikes in notionally covered areas it catches), but right now there's a storm very close to me (Cambridge, UK) and it's displaying a good fraction, but not all, of the strikes I see and hear.
(This is getting a bit pedantic but this is HN so we like to be accurate. And I learned the word on HN so might as well pass it on.)
Several of the other answers have mentioned "triangulation" but that's not quite right--that would imply that each listening station measures an angle towards the lightning source. But they don't--instead, they collect non-directional data consisting of arrival time stamps.
I believe the word for what they're doing is "trilateration".
Edit: I originally said "multilateration" but it turns out that's not quite right, either.
I would guess they have devices that detect light and sound. When the device detects a flash, it will start a timer and wait for the sound, which ends the timer.
Then based on the speed of sound, they can estimate the distance of the strike.
With many of these devices all over the place recording the data, using math, they can pinpoint the location of where the strike was.
So do the lines between stations and a strike mean that detector heard the strike? That's incredible if so... there's a station in California picking up strikes over Missouri.
From their FAQ, they pick up RF pulses, tag it with a GPS timestamp, and correlate on a central server.
"A lightning is a pulse not comparable to a broadcast signal on a single frequency. Such a pulse can be described best as a interference of several frequencies. Therefore a ferrite rod for lightning detection has to be a wide-band antenna."
The funny part is that they happen mostly in the midwest and that area is sometimes called the 'bible belt' (for its strong evangelical christian population) and lightning strikes are a metaphorical means of punishment by God (as in "May lightning strike me down if I'm lying ...")
That irony aside, I am wondering about that South America number, it seems like at equidistant latitudes there are similar lightning environments on both sides of the equator and yet a nearly 16x reduction in number of strikes? What is up with that?
Where I live in West Central Florida we get about 50 strikes per square mile per year, 1.5 million per year for the state. Lightning is common here during the rainy season.
cool, I'm looking forward for this evening's thunderstorm to confirm if it's actually/acurately working :) Average delay appears to be between 3s and 10s here in Europe, but closer to 3s.
They show the detectors that picked up the strikes. I don't know if a single detector can determine distance, or if the strike location is triangulated using angles from multiple detectors.
[edit] Oh, the detectors are quite simple. They use GPS for position and time and report time-of-arrival of the elctromagnetic pulses. The server then uses that information to figure out where the strikes occurred.
[+] [-] larcher|11 years ago|reply
http://www.sciencedaily.com/releases/2013/04/130424210319.ht...
[+] [-] moreentropy|11 years ago|reply
http://www.lightningmaps.org/realtime
[+] [-] netcraft|11 years ago|reply
[+] [-] hcarvalhoalves|11 years ago|reply
[1] http://www.blitzortung.org/Webpages/index.php?lang=en&page=3
[+] [-] joezydeco|11 years ago|reply
http://www.maxwell.com/products/microelectronics/docs/hsn100...
[+] [-] bri3d|11 years ago|reply
http://www.digikey.com/product-highlights/us/en/ams-as3935-f...
The Blitzortung paper mentions that sensor isn't accurate enough to use to determine the distance of a strike, so I don't think a crowdsourced phone-based network could work particularly well:
http://www.blitzortung.org/Documents/TOA_Blitzortung_RED.pdf
I also suspect that a phone's RF frontend would be awful for lightning detection as it is tuned for 900+Mhz signals. You can read more about lightning's RF properties here:
http://assets.cambridge.org/052158/3276/sample/0521583276WS....
but it's clear that above 10Mhz you probably won't learn much of value about the lightning strike.
Elsewhere in this thread, an HN reader found that Nokia were working on lightning detection for phones, but it seems to use the FM radio receiver that was often built into SoCs, which could definitely tune itself to a much more useful frequency range.
[+] [-] jpalomaki|11 years ago|reply
Does lightning have any effect on the sensors that detect the geomagnetic field? Maybe even some small anomalies in the readings could be useful if many devices were working together.
[1] http://www.computerworld.com/s/article/9021121/Nokia_to_add_...
[+] [-] j2kun|11 years ago|reply
[+] [-] arethuza|11 years ago|reply
http://en.wikipedia.org/wiki/DEFCON_%28video_game%29
[+] [-] neumann|11 years ago|reply
[+] [-] darsham|11 years ago|reply
The craziest phenomenon of this kind is whistlers [2]. Lightning strikes can be heard on radio at a point on the earth exactly symmetrical to the source, using the equator as plane of symmetry. In fact, the signal bounces around back and forth between the two points, following a line on the magnetosphere. The ones closer to the poles have longer paths, and their frequencies can get spread out over 3-4 seconds as they bounce around.
[1]: http://en.wikipedia.org/wiki/Earth%E2%80%93ionosphere_wavegu...
[2]: http://en.wikipedia.org/wiki/Whistler_%28radio%29
[+] [-] spektom|11 years ago|reply
[+] [-] rondon2|11 years ago|reply
[+] [-] RijilV|11 years ago|reply
[+] [-] TazeTSchnitzel|11 years ago|reply
[+] [-] phreeza|11 years ago|reply
[0] http://palacegstaad.roundshot.ch/
[+] [-] quarterto|11 years ago|reply
[+] [-] gjm11|11 years ago|reply
[+] [-] cipherzero|11 years ago|reply
[+] [-] IvyMike|11 years ago|reply
Several of the other answers have mentioned "triangulation" but that's not quite right--that would imply that each listening station measures an angle towards the lightning source. But they don't--instead, they collect non-directional data consisting of arrival time stamps.
I believe the word for what they're doing is "trilateration".
Edit: I originally said "multilateration" but it turns out that's not quite right, either.
https://en.wikipedia.org/wiki/Triangulation
https://en.wikipedia.org/wiki/Trilateration
https://en.wikipedia.org/wiki/Multilateration
The More You Know.
[+] [-] 4mnt|11 years ago|reply
http://www.blitzortung.org/Webpages/index.php?lang=en&page=3
[+] [-] pndmnm|11 years ago|reply
They use VLF radio detection to triangulate strikes.
[+] [-] cipherzero|11 years ago|reply
[+] [-] wyred|11 years ago|reply
Then based on the speed of sound, they can estimate the distance of the strike.
With many of these devices all over the place recording the data, using math, they can pinpoint the location of where the strike was.
[+] [-] exabrial|11 years ago|reply
[+] [-] chris_va|11 years ago|reply
"A lightning is a pulse not comparable to a broadcast signal on a single frequency. Such a pulse can be described best as a interference of several frequencies. Therefore a ferrite rod for lightning detection has to be a wide-band antenna."
[+] [-] thekevan|11 years ago|reply
Oceania: 28 strikes
Asia: 0 strikes
South America: 913 strikes
North America: 15,826 strikes!!
[+] [-] hcarvalhoalves|11 years ago|reply
[+] [-] ChuckMcM|11 years ago|reply
That irony aside, I am wondering about that South America number, it seems like at equidistant latitudes there are similar lightning environments on both sides of the equator and yet a nearly 16x reduction in number of strikes? What is up with that?
[+] [-] CapitalistCartr|11 years ago|reply
[+] [-] greggman|11 years ago|reply
[+] [-] CmonDev|11 years ago|reply
[+] [-] omnibrain|11 years ago|reply
[+] [-] toomuchtodo|11 years ago|reply
Ship Traffic (AIS): https://www.marinetraffic.com/en/
[+] [-] wazari972|11 years ago|reply
[+] [-] aarkling|11 years ago|reply
[+] [-] Gracana|11 years ago|reply
[edit] Oh, the detectors are quite simple. They use GPS for position and time and report time-of-arrival of the elctromagnetic pulses. The server then uses that information to figure out where the strikes occurred.
http://www.blitzortung.org/Webpages/index.php?lang=en&page=3
[+] [-] nraynaud|11 years ago|reply