Lightning Stroke/TLE Analysis with VLF Receiver (Post with Missing Images)

[Jonathan]

Sometimes lightning strokes produce what are called transient luminous events, or TLEs. They are commonly known as sprites and jets. Not much is known about them but they are continuously studied. Columns and regions of atmosphere are ionized so much by the electric potentials caused by the lightning stroke that the air glows red. It’s present for a split second when the stroke occurs, so you need a quick eye and a camera to capture an image. Many enthusiasts who capture these beautiful events often use timed video and photography and use lightning stroke data to identify the specific lightning stroke, its channel current, and polarity. The "channel" is the conductive channel of ionized air where lightning current either flows upwards or downwards. VLF receivers detect the radio emissions from lightning strokes, called sferics. The sferic signal characteristics are fed into models that calculate stroke polarity and channel current. 

An enthusiast and photographer, Paul, captured a double TLE, showing both a sprite (the dendritic structure) and an ELVE (the upper dim region of red glowing air above the sprite). ELVEs are often, but not always, indicative of what is called a "continuing current", or a residual current flowing through the channel with ELF frequency components. 

Using a VLF receiver connected to a soundcard and vlfrx-tools software, it is possible to look at the sferic's impulse in a time domain plot. A continuing current will often show up as an "ELF tail" right after the initial impulse. This tail has ELF frequency components and is indicative of that "little wavy line" after the sferic's impulse. The plot below shows the sferic from the stroke that created the TLE above. The ELF tail is hard to see because it has some high frequency components on it, but it is there. Running the following signal processing chain in vlfrx-tools software produced the plot below:

vtread -T2022-12-14_02:09:20,+30s /data/vlf_96k | vtfilter -a th=5 | vtresample -r32000 | vtcat -T2022-12-14_02:09:38.2,+0.1 | vtplot -t "+266kA Stroke Nice Sprite/ELVE Combo"

The farther the VLF receiver is from the lightning stroke, the longer the ELF tail is. If this stroke occurred in Europe, it would be much longer, but it was captured in the US, closer to the VLF receiver. Here is another example of an ELVE:

This is the time domain plot with the ELF tail easier to see because there are much less high frequency components:

In recording these millisecond events, it is essential to use precision timing, which is why I use a GPS receiver to enable accurate and precision timestamping. I used the signal processing chain above to pull the spectrum data from the data store, filter out mains hum, resample to 32k to remove a lot of high frequency components, then feed the specific spectrum chunk into the plotting program.

With vlfrx-tools software and a network of VLF receivers, you can do lightning location as well. Here is a lightning map from a network of VLF receivers in India:

The red dots indicate the location of a stroke and the circles indicate VLF receiver locations.

Eventually, I would like to have a network of VLF receivers collecting sferic data for lightning location.

Jonathan

KC3EEY

[Kerry Case]

Very interesting. Is the light strokes at the area of the Luminous or is it miles away?

 

Kerry

N1URT

 

Sent from Mail for Windows

 

From: Jonathan
Sent: Friday, January 27, 2023 9:35 AM
To: Unknown; TangerineSDR Listserv
Subject: [HamSCI] Lightning Stroke/TLE Analysis with VLF Receiver (Post with Missing Images)

 

Sometimes lightning strokes produce what are called transient luminous events, or TLEs. They are commonly known as sprites and jets. Not much is known about them but they are continuously studied. Columns and regions of atmosphere are ionized so much by the electric potentials caused by the lightning stroke that the air glows red. It’s present for a split second when the stroke occurs, so you need a quick eye and a camera to capture an image. Many enthusiasts who capture these beautiful events often use timed video and photography and use lightning stroke data to identify the specific lightning stroke, its channel current, and polarity. The "channel" is the conductive channel of ionized air where lightning current either flows upwards or downwards. VLF receivers detect the radio emissions from lightning strokes, called sferics. The sferic signal characteristics are fed into models that calculate stroke polarity and channel current. 

 

An enthusiast and photographer, Paul, captured a double TLE, showing both a sprite (the dendritic structure) and an ELVE (the upper dim region of red glowing air above the sprite). ELVEs are often, but not always, indicative of what is called a "continuing current", or a residual current flowing through the channel with ELF frequency components. 

Using a VLF receiver connected to a soundcard and vlfrx-tools software, it is possible to look at the sferic's impulse in a time domain plot. A continuing current will often show up as an "ELF tail" right after the initial impulse. This tail has ELF frequency components and is indicative of that "little wavy line" after the sferic's impulse. The plot below shows the sferic from the stroke that created the TLE above. The ELF tail is hard to see because it has some high frequency components on it, but it is there. Running the following signal processing chain in vlfrx-tools software produced the plot below:

 

vtread -T2022-12-14_02:09:20,+30s /data/vlf_96k | vtfilter -a th=5 | vtresample -r32000 | vtcat -T2022-12-14_02:09:38.2,+0.1 | vtplot -t "+266kA Stroke Nice Sprite/ELVE Combo"

 

The farther the VLF receiver is from the lightning stroke, the longer the ELF tail is. If this stroke occurred in Europe, it would be much longer, but it was captured in the US, closer to the VLF receiver. Here is another example of an ELVE:

 

This is the time domain plot with the ELF tail easier to see because there are much less high frequency components:

In recording these millisecond events, it is essential to use precision timing, which is why I use a GPS receiver to enable accurate and precision timestamping. I used the signal processing chain above to pull the spectrum data from the data store, filter out mains hum, resample to 32k to remove a lot of high frequency components, then feed the specific spectrum chunk into the plotting program.

 

With vlfrx-tools software and a network of VLF receivers, you can do lightning location as well. Here is a lightning map from a network of VLF receivers in India:

 

The red dots indicate the location of a stroke and the circles indicate VLF receiver locations.

 

Eventually, I would like to have a network of VLF receivers collecting sferic data for lightning location.

 

Jonathan

KC3EEY

--
Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
---
You received this message because you are subscribed to the Google Groups "HamSCI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB1ObgqciLJcBmJKJuq2VpxdJ7n-C%2BQeQbCx-pSMOXa4qA%40mail.gmail.com.

 

[Gerald Creager]

Jonathon, 

Your network of receivers is similar in concept to the network I presented several years ago at the TAPR DCC. Let's talk. Your ELF tail and EVLE work suggest some things to me. A lot of work has been done in other spectral ranges than VLF, and there are data to capture at higher frequencies. Think "Citizen Weather Observer Network" but for lightning. I've also got some ideas on central processing for geolocation for GC and CC/IC impulses.

73

Gerry N5JXS

Maj Gerry Creager
National Health Services Advisory Team
Civil Air Patrol, U.S. Air Force Auxiliary
(M) 979.229.5301
GoCivilAirPatrol.com
Volunteers serving America's communities, saving lives, and shaping futures.

--

[Jonathan]

Kerry,

ELVES and sprites occur in the mesosphere above thunderstorms.

Gerry,

I'm still looking to create a network of VLF receivers and one of the functions would be lightning location. Other than impulse duration, what are the characteristics of a CG and intracloud stroke at VLF and how would distance from the receiver affect that? The vlfrx-tools software I use is very versatile in terms of signal processing, so what were you looking to do?

Jonathan

KC3EEY

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAG3zga4uy-pmvv_K2BhxxDF9OAijw1VRdehqi0YaLsVAW4XueA%40mail.gmail.com.

[Kerry Case]

Roger that understand now That is cool stuff

K

 

Sent from Mail for Windows

 

Using a VLF receiver connected to a soundcard and vlfrx-tools software, it is possible to look at the sferic's impulse in a time domain plot. A continuing current will often show up as an "ELF tail" right after the initial impulse. This tail has ELF frequency components and is indicative of that "little wavy line" after the sferic's impulse. The plot below shows the sferic from the stroke that created the TLE above. The ELF tail is hard to see because it has some high frequency components on it, but it is there. Running the following signal processing chain in vlfrx-tools software produced the plot below:

 

vtread -T2022-12-14_02:09:20,+30s /data/vlf_96k | vtfilter -a th=5 | vtresample -r32000 | vtcat -T2022-12-14_02:09:38.2,+0.1 | vtplot -t "+266kA Stroke Nice Sprite/ELVE Combo"

 

The farther the VLF receiver is from the lightning stroke, the longer the ELF tail is. If this stroke occurred in Europe, it would be much longer, but it was captured in the US, closer to the VLF receiver. Here is another example of an ELVE:

 

This is the time domain plot with the ELF tail easier to see because there are much less high frequency components:

In recording these millisecond events, it is essential to use precision timing, which is why I use a GPS receiver to enable accurate and precision timestamping. I used the signal processing chain above to pull the spectrum data from the data store, filter out mains hum, resample to 32k to remove a lot of high frequency components, then feed the specific spectrum chunk into the plotting program.

 

With vlfrx-tools software and a network of VLF receivers, you can do lightning location as well. Here is a lightning map from a network of VLF receivers in India:

 

The red dots indicate the location of a stroke and the circles indicate VLF receiver locations.

 

Eventually, I would like to have a network of VLF receivers collecting sferic data for lightning location.

 

Jonathan

KC3EEY

--
Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
---
You received this message because you are subscribed to the Google Groups "HamSCI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB1ObgqciLJcBmJKJuq2VpxdJ7n-C%2BQeQbCx-pSMOXa4qA%40mail.gmail.com.

--
Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
---
You received this message because you are subscribed to the Google Groups "HamSCI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAG3zga4uy-pmvv_K2BhxxDF9OAijw1VRdehqi0YaLsVAW4XueA%40mail.gmail.com.

--
Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
---
You received this message because you are subscribed to the Google Groups "HamSCI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB2HPk9jxq7Gc-9J9CFPjS1mFLeVQEsc-jSFtFJFi5jKXg%40mail.gmail.com.

 

[Stephen Kangas]

And for photographic proof, refer to the Twilight Zone movie of 1983 where John Lithgow sees one, or at least some distant relative.

 

Stephen W9SK

 

 

From: ham...@googlegroups.com <ham...@googlegroups.com> On Behalf Of Jonathan
Sent: Friday, January 27, 2023 9:42 AM
To: ham...@googlegroups.com; TangerineSDR Listserv <tanger...@lists.tapr.org>
Subject: Re: [HamSCI] Lightning Stroke/TLE Analysis with VLF Receiver (Post with Missing Images)

 

Kerry,

 

ELVES and sprites occur in the mesosphere above thunderstorms.

 

Gerry,

 

I'm still looking to create a network of VLF receivers and one of the functions would be lightning location. Other than impulse duration, what are the characteristics of a CG and intracloud stroke at VLF and how would distance from the receiver affect that? The vlfrx-tools software I use is very versatile in terms of signal processing, so what were you looking to do?

 

Jonathan

KC3EEY

On Fri, Jan 27, 2023 at 10:35 AM Gerald Creager <gcre...@cap.gov> wrote:

Jonathon, 

 

Your network of receivers is similar in concept to the network I presented several years ago at the TAPR DCC. Let's talk. Your ELF tail and EVLE work suggest some things to me. A lot of work has been done in other spectral ranges than VLF, and there are data to capture at higher frequencies. Think "Citizen Weather Observer Network" but for lightning. I've also got some ideas on central processing for geolocation for GC and CC/IC impulses.

 

73

Gerry N5JXS

 

Maj Gerry Creager

National Health Services Advisory Team

Civil Air Patrol, U.S. Air Force Auxiliary

(M) 979.229.5301

GoCivilAirPatrol.com

Volunteers serving America's communities, saving lives, and shaping futures.

 

 

On Fri, Jan 27, 2023 at 7:35 AM Jonathan <emum...@gmail.com> wrote:

Sometimes lightning strokes produce what are called transient luminous events, or TLEs. They are commonly known as sprites and jets. Not much is known about them but they are continuously studied. Columns and regions of atmosphere are ionized so much by the electric potentials caused by the lightning stroke that the air glows red. It’s present for a split second when the stroke occurs, so you need a quick eye and a camera to capture an image. Many enthusiasts who capture these beautiful events often use timed video and photography and use lightning stroke data to identify the specific lightning stroke, its channel current, and polarity. The "channel" is the conductive channel of ionized air where lightning current either flows upwards or downwards. VLF receivers detect the radio emissions from lightning strokes, called sferics. The sferic signal characteristics are fed into models that calculate stroke polarity and channel current. 

 

An enthusiast and photographer, Paul, captured a double TLE, showing both a sprite (the dendritic structure) and an ELVE (the upper dim region of red glowing air above the sprite). ELVEs are often, but not always, indicative of what is called a "continuing current", or a residual current flowing through the channel with ELF frequency components. 

Using a VLF receiver connected to a soundcard and vlfrx-tools software, it is possible to look at the sferic's impulse in a time domain plot. A continuing current will often show up as an "ELF tail" right after the initial impulse. This tail has ELF frequency components and is indicative of that "little wavy line" after the sferic's impulse. The plot below shows the sferic from the stroke that created the TLE above. The ELF tail is hard to see because it has some high frequency components on it, but it is there. Running the following signal processing chain in vlfrx-tools software produced the plot below:

 

vtread -T2022-12-14_02:09:20,+30s /data/vlf_96k | vtfilter -a th=5 | vtresample -r32000 | vtcat -T2022-12-14_02:09:38.2,+0.1 | vtplot -t "+266kA Stroke Nice Sprite/ELVE Combo"

 

 

The farther the VLF receiver is from the lightning stroke, the longer the ELF tail is. If this stroke occurred in Europe, it would be much longer, but it was captured in the US, closer to the VLF receiver. Here is another example of an ELVE:

 

This is the time domain plot with the ELF tail easier to see because there are much less high frequency components:

In recording these millisecond events, it is essential to use precision timing, which is why I use a GPS receiver to enable accurate and precision timestamping. I used the signal processing chain above to pull the spectrum data from the data store, filter out mains hum, resample to 32k to remove a lot of high frequency components, then feed the specific spectrum chunk into the plotting program.

 

With vlfrx-tools software and a network of VLF receivers, you can do lightning location as well. Here is a lightning map from a network of VLF receivers in India:

 

The red dots indicate the location of a stroke and the circles indicate VLF receiver locations.

 

Eventually, I would like to have a network of VLF receivers collecting sferic data for lightning location.

 

Jonathan

KC3EEY

--
Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
---
You received this message because you are subscribed to the Google Groups "HamSCI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB1ObgqciLJcBmJKJuq2VpxdJ7n-C%2BQeQbCx-pSMOXa4qA%40mail.gmail.com.

--
Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
---
You received this message because you are subscribed to the Google Groups "HamSCI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAG3zga4uy-pmvv_K2BhxxDF9OAijw1VRdehqi0YaLsVAW4XueA%40mail.gmail.com.

--
Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
---
You received this message because you are subscribed to the Google Groups "HamSCI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB2HPk9jxq7Gc-9J9CFPjS1mFLeVQEsc-jSFtFJFi5jKXg%40mail.gmail.com.

[kunde...@tbsys.com]

Hello All:

 

I am a heretofore silent observer of the hamsci group and Tangerine conversations. I am also a ham radio operator (K8ZTV) and a Meteorologist.

 

I have a Meteorologist friend who has been researching and observing Sprites for years.  He is a past President of the American Meteorological Society. His website is:

 

https://www.weathervideohd.tv/wvhd.php?mod=search&sev=68123&sp=1

 

I hope that you find his website interest and informative.

 

Ken

 

Kenneth H. Underwood, Ph.D., CCM (#466)

Santa Clarita, CA 91350

kunde...@tbsys.com

www.tbsys.com

661-965-9621 (cell)

661-309-6225 (office)

661-309-6229 (fax)

 

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/0f9001d93279%24c66c05a0%24534410e0%24%40kangas.com.

[Gerry Creager]

Ken

I suspect we need to talk. I was wondering when another meteorologist might join in! (Now that I'm retired, it's time to start looking at the CCM...). I was at OU and NSSL until SEP 2021, and exposure to the likes of Don McGorman, as well as a long history helping with the CWOP program, caused me to turn toward lightning. Sprites are not on my list... yet... but I think we need to be looking at them and GRB will be something we can make serious progress on in the future, and the developing Space Weather Station shows promise for all this.

73

Gerry N5JXS

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/00bd01d93296%2477e52750%2467af75f0%24%40tbsys.com.

--

Gerry Creager N5JXS

It's kind of fun to do the impossible. -- Walt Disney

[Robert McGwier]

Where in Maryland would you locate one?

Bob 

N4HY

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB2HPk9jxq7Gc-9J9CFPjS1mFLeVQEsc-jSFtFJFi5jKXg%40mail.gmail.com.

--

Dr. Robert W McGwier, Ph.D.
Adjunct Faculty, Virginia Tech

Affiliated Faculty, University of Scranton
ARDC Member of Board
N4HY: ARRL, TAPR, AMSAT, EARC, CSVHFS
Sky: AAVSO, Sky360, explorescu.org, Skyscrapers

[Jonathan]

Gerry,

No worries. Why do you prefer to use TDOA instead of TOGA?

Also, how does one characterize the stroke in VLF? Will looking at the analytic magnitude indicate anything?

In vlfrx-tools software, lightning location works in the following manner: the utility vttoga will identify impulsive signals and measure their TOGA (whether they are sferics, rain static, or local interference). It can also measure other parameters, such as ground wave and sky wave identification, analytic magnitude, frequency spectra, and unwrapped absolute phase. Still, vttoga just sees the impulse. To determine whether it came from a stroke or some other interference source, vttoga data from multiple receivers (including lat/long coordinates of the VLF receivers) is fed into the utility vtspot which computes a solution for the stroke using trilateration. Rain static and local interference won't yield a solution, but sferics will if it's within a location between the VLF receivers. vtspot is not perfect, but it generates pretty accurate solutions when fed with good data. You can read about how to use vttoga and vtspot for lightning location here. The hope is to build a dense network of VLF receivers for an overdetermined solution, but this will be volunteer based, so it'll take time to build.

Using VLF receivers with vlfrx-tools, do you feel it would be a good start? With the signal processing utilities available in vlfrx-tools, I think quite a lot can be done.

Jonathan

KC3EEY

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CA%2BK5gzdTiNv36FArMNNYr5ruGmzV_yS%3Dk%2B_p5U-nCEo3ShiUNQ%40mail.gmail.com.

[Gerald Creager]

Characterization of the impulse will require acquiring several million strikes, probably 3-6 months if we have at least 10 stations automatically logging the potential signatures. Characteristics of CG leaders and strikes have been defined in several texts, so that should be pretty straightforward. What I think we need to do is the initial data acquisition, then apply a filter to remove extraneous noise. This will leave CG leaders and strikes with a composite time-domain and frequency domain signal. Recording some n signatures  for a rapid table lookup allows a quick assignment of relative uniqueness and subsequent transmission for central processing. 

Also, while VLF is likely to simplest and narrowest spectrum for detection and, to some extent, characterization of lightning impulse, sferrics are not the sole spectral output, or more to the point, there's a lot of signal associated with a given impulse and it's present, at diminishing amplitude, well into the HF spectrum. Going back to surveillance data collection, characterizing which pieces of spectrum tend to possess more signal than others, and I don't expect the dispersion to be uniform, would allow a better characterization of each impulse. 

As to TDOA vs TOGA, I've worked with TDOA before, and it's similar to the work I did LONG ago with GPS systems.TGOA might work but I'm concerned that we'd be looking for similar phase offsets from stations that are not necessarily geospatially local to one another. Allowing each receiving station to timestamp each impulse, attach a signature to it, and send it for processing means it's likely to be a little easier to process with TDOA. 

Having said all that, I'm open to experimenting with TOGA. 

I'm going to have to come up to speed on VLFRX-tools as I've not used it before. It's always possible my problem set has been solved and my work is done!

73

Gerry N5JXS

Maj Gerry Creager
National Health Services Advisory Team
Civil Air Patrol, U.S. Air Force Auxiliary
(M) 979.229.5301
GoCivilAirPatrol.com
Volunteers serving America's communities, saving lives, and shaping futures.

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB2L6ERNK7vaa9B219csQme%3DWENnhzkg1hPH9ZoYzq4sGQ%40mail.gmail.com.

[Jonathan]

Gerry,

What frequency bands will be required for characterization then? If you're talking about HF and higher, you need a much denser network of receivers that are within short range to the lightning and some sophisticated autonomy to collect the waveforms and then perform trilateration. If the trilateration produces a good solution, it's most likely a stroke, but you need a lot of receivers and a lot of data will be produced. Once characterization is complete, will these HF and higher stations still be required or will VLF be adequate?

TOGA is a proven method for lightning location and it is utilized by many lightning location networks. vttoga of vlfrx-tools is particularly useful in that it can produce a lot of useful information from a sferic. It utilizes a scoring system to determine if an impulse is "sferic-like", however, the additional useful information can be used to further identify the sferic, possibly CG and intracloud. 

Here is an example of the sferic captured at my receiver in Pennsylvania originating from the -333kA stroke and ELVES event in May of last year in Kansas. Below is the sferic's time domain plot:

Running that narrow slice of spectrum into vttoga produces an output of two TOGAs, one being the real one and one a false detection. Using vttoga's options to produce extended measurements and a plotting script produces the following:

If you look at the image, you can see the plot title includes the sferic's TOGA and the various scoring based on how "sferic-like" it is. (see vttoga's documentation for more information) The top plot is a time domain plot showing the groundwave and successive skywaves. The second plot is the analytic magnitude which removes all phase information from the sferic's signal. The third plot is the instantaneous frequency computed from the analytic magnitude. The fourth plot is the frequency spectrum of the sferic, showing the peak at less than 5 kHz. Lastly, the fifth plot is the unwrapped absolute phase as a function of frequency.

Using vttoga's extended data option, it can be possible to characterize sferics and even differentiate from local interference. See the post here for more information. As you can see from Paul's remarks, sferics are difficult to identify and characterize.

Would this sort of analysis be useful?

Jonthan

KC3EEY

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAG3zga7hg%2B69NRrvYa8A5mOpKVWZs95sE5%3Dks9H%2Bo547swfkSA%40mail.gmail.com.

[Jonathan]

Hi Gerry,

One last thing I forgot to mention, the output of vtspot was used to generate the following plot, so with a network of simple receivers, vlfrx-tools and gnuplot is pretty powerful:

Jonathan

KC3EEY

[Gerry Creager]

Morning, Jonathan,

On Mon, Jan 30, 2023 at 5:38 AM Jonathan <emum...@gmail.com> wrote:

Gerry,

What frequency bands will be required for characterization then? If you're talking about HF and higher, you need a much denser network of receivers that are within short range to the lightning and some sophisticated autonomy to collect the waveforms and then perform trilateration. If the trilateration produces a good solution, it's most likely a stroke, but you need a lot of receivers and a lot of data will be produced. Once characterization is complete, will these HF and higher stations still be required or will VLF be adequate?

I anticipate we're looking at 100kHz-10 Mhz, but likely slices of spectrum in that region. I'm hoping to be able to characterize that more now that I've more time, but I didn't see sufficient local convective activity last season to do much good. That's going to be a dedicated SDR and server job, to capture sufficient data to be sure. And, the spectral space might expand to 15 MHz. I did look at this in the past, and saw signal that looked interesting up to 10-15 MHz. I did not do analysis at that time to determine if the signal was merely reproduced from lower frequencies.

I disagree that the receiver density will need to increase dramatically. Initiation occurs at altitude, as the charge formation leading to leader creation is usually high in the cumulus or cumulonimbus column in the midst of ice nucleation. Range will be pretty good at HF if your source initiates at 35,000 ft. 

TOGA is a proven method for lightning location and it is utilized by many lightning location networks. vttoga of vlfrx-tools is particularly useful in that it can produce a lot of useful information from a sferic. It utilizes a scoring system to determine if an impulse is "sferic-like", however, the additional useful information can be used to further identify the sferic, possibly CG and intracloud. 

TOGA is very popular for localized lightning detection networks. I will investigate it; I am not discounting it, but I need to evaluate it more. 

 

Here is an example of the sferic captured at my receiver in Pennsylvania originating from the -333kA stroke and ELVES event in May of last year in Kansas. Below is the sferic's time domain plot:

Running that narrow slice of spectrum into vttoga produces an output of two TOGAs, one being the real one and one a false detection. Using vttoga's options to produce extended measurements and a plotting script produces the following:

If you look at the image, you can see the plot title includes the sferic's TOGA and the various scoring based on how "sferic-like" it is. (see vttoga's documentation for more information) The top plot is a time domain plot showing the groundwave and successive skywaves. The second plot is the analytic magnitude which removes all phase information from the sferic's signal. The third plot is the instantaneous frequency computed from the analytic magnitude. The fourth plot is the frequency spectrum of the sferic, showing the peak at less than 5 kHz. Lastly, the fifth plot is the unwrapped absolute phase as a function of frequency.

Using vttoga's extended data option, it can be possible to characterize sferics and even differentiate from local interference. See the post here for more information. As you can see from Paul's remarks, sferics are difficult to identify and characterize.

Would this sort of analysis be useful?

This sort of characterization could well solve my signature issue. The question arises: How long does it take to analyze? I've other questions about TOGA (I've effectively forgotten everything I'd read about it 10 years ago thru lack of use) to resolve. I recall thinking, when I first looked at it, that it might be an easier solution than trilateration (which, with overdetermination can also produce an initiation point in 3D for the discharge), but for some reason abandoned it. I'll look for my notes as I unpack the boxes from the office. I've got an old-fashioned dead tree notebook I was keeping for documentation. 

73

Gerry N5JXS 

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB20TdA0yXKs9J%3DHi97UWjMUTZ9hhBVEpYghnya8AgkQ%2BQ%40mail.gmail.com.

[Jonathan]

Gerry,

In that case, would a VLF system be of any value to you then, at least for characterization? In the higher frequencies, you'd only be able to detect fairly local lightning, which is why I mentioned that you'd need a dense receiver network to cover a large geographical area in HF and VHF, at least for an active lightning location and detection system, providing both the location and the type of stroke. The network would not have to be as dense in the VLF band. Do you have some spectrograms of HF or VHF sferics?

vttoga produces that data almost instantly. Again, not every sferic will be an actual sferic, but you can modify the ascore options to trigger on almost all real sferics. 

Jonathan

KC3EEY

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAMyi%2BGP%3DRa0Z2RiVNaY7XXvtsaEODov0Pr3H-SODEr_F%3D%2BBnbg%40mail.gmail.com.

[Gerald Creager]

Yes, vir systems are a starting point, for sure. 

My experience suggests the detection range isn’t as severely degraded as you might expect. That said, if this effort works out, I expect we’ll be able to engage a pretty large group of users, which could augment the HamSci ranks, too.

I’m a bit leery of solely using vlf without waveform characterization especially at much longer ranges. vttoga may be the answer. I’ve downloaded vlftools and have started looking at it and reading documentation.

There’s a lot of information in lightning impulse signals useful for research. I’d like to see as much of that captured and made available as open data as possible.

73

Gerry N5JXS

To view this discussion on the web visit https://groups.google.com/d/msgid/hamsci/CAOY0kB2JY%3D-tycynd0pScPQP-a0x8BrPfS3vq6G_so3LDLtUdw%40mail.gmail.com.

--