Meteors & VLF


This topic was first investigated at this observatory in the Spring of 2015 as a result of an internet literature search on radio reflections from meteors.

It became clear that little work had been done in the UK on how VLF impulses might be generated by a meteor. 

Some amateur measurements had already been made in 2009 by Jean-L. RAULT  F6AGR IMO (International Meteor Organization) Radio Commission. The work was a collaboration between Rault in France and Romero in Italy.

The initial VLF and VHF meteor observation campaign was performed with the help of Renato Romero. The live VLF data being received in Cumiana, Italy by Romero were retrieved from the Internet and compared to the VHF meteor pings detected near Paris, France. Rault comments that: “time synchronisation issues occurred, because the sample frequency used by Renato was never exactly the same as the one produced by my own computer, so the VLF / VHF correlation task wasn't possible.”

Rault also comments that: “Looking for correlations between meteors and ELF/VLF events is a very demanding and a very time consuming task. The detection of the interesting events can't be automated, because the ELF/VLF events signatures are not known in advance”.

He goes on to suggest: “The theories stating that some meteors can radiate low frequency electromagnetic energy seem to be supported by the present practical study which is based on hundreds of actual discrete observations of meteors and ELF/VLF events. However, more data are still needed to confirm such a conclusion.”

The stimulus for investigating VLF emissions from meteors arose from consideration of an even more unusual notion. This is the phenomenon known as "electrophonic sounds".

One of the earliest investigators was Keay [ COLIN S. L. KEAY Journal of Scientific Exploration, Vol. 7, No. 4, pp. 337-354, 1993 ]  who produced a paper entitled 'Progress in Explaining the Mysterious Sounds Produced by Very Large Meteor Fireballs' in 1993. His investigation was initiated by trying to understand reports that strange sounds had been heard simultaneously with the sighting of brilliant meteor fireballs, many tens of kilometres distant. 

It was proposed that the meteor produced a strong Electric field in a high power VLF transient which propagated to the ground at light speed and caused dried leaves and grasses to vibrate, producing a sound locally to the observer.

The term "electrophonic sounds" was widely used to describe them and to distinguish them from the normal sonic effects heard after the fireball has passed by. Keay presented the history of this perhaps neglected branch of meteor science in some detail, drawing attention to the theoretical difficulties which stood in the way of a scientifically plausible understanding until the early 1990s

Keay suggests that a large bolide sheds its kinetic energy at rates upwards of tens of gigawatts. Its luminous efficiency, a function of velocity and composition, is of the order of a few percent. Ionisation is of the same order, while the remaining energy is mainly liberated as heat. The extremely high energy density residing in the plasma trail should excite all EM oscillatory modes possible, including those at frequencies in the audio range (ELF/VLF radiation). The problem is to discover a realistic generation mechanism. One possibility appeared to be through excitation of a hybrid-mode magnetohydrodynamic wave within the plasma of the bolide trail.

Keay also introduces the notion of turbulence in the meteor plasma tail. Turbulent motions in the wake have characteristic frequencies upwards of around 500 Hz, as energy is transferred to smaller eddies. The turbulence excites vibrations of the geomagnetic field giving rise to the emission of electromagnetic radiation in the ELF/VLF region of the spectrum.

A major release of stored magnetic energy occurs when the conductivity falls, due to recombination or electron attachment as the plasma cools and the magnetic
Reynolds number falls to less than unity. The twisted and tangled "magnetic spaghetti" then relaxes, releasing its strain energy as vibrations of the geomagnetic field within the earth-ionosphere cavity. 

These field vibrations have wavelengths of the order of 100 km, corresponding to an electromagnetic wave frequency of 3 kHz.

Professional VLF Measurements (1999)

Price and Blum [ Colin Price & Moshe Blum, Department of Geophysics and Planetary Science, Tel Aviv University, RamatAviv 69978, ISRAEL ]   made measurements during the Leonid meteor storm on 18 November 1999.

Electromagnetic measurements were continuously recorded to try and detect the VLF radio waves produced by meteors. Since the best viewing location for the 1999 meteor shower was the Middle East, they were ideally located for this task. A permanent field site for observing ELF/VLF signals was located at the Desert Research Institute of Ben-Gurion University, at Sde Boker in the Negev Desert (30 N, 34 E).

The work suggests that the VLF pulse emitted from a meteor is distinct from that produced by lightning. See below and opposite.

These pulse shapes produce clearly different spectra.

Price and Blum produced convincing evidence to show that they can correlate the selected meteor VLF pulses with visually observed meteor counts for the Leonid shower of 1999. This result can be seen opposite.

The ELF/VLF method of counting the meteor flux produced a figure of 15,000 per hour as compared to 350 per hour using optical methods.

The only theoretical explanation of how these radio waves are produced has been presented by Keay. However these measurements challenge his theory in respect of observed VLF generation from sub-visible and small meteors rather than only from bright fireballs (bolides).

More information is available by opening the file { The_Generation_of_VLF_Emissions_by_Meteors.pdf } at the base of this page.

Setting up an Experiment to measure VLF Transients from Meteors

Various equipment was brought together with the aim of capturing both VLF and optical information during the Perseid meteor shower in August 2015.

The measurement system configuration is shown opposite.

There are a number of physical variables that were un-quantifiable at that point and would need to be carefully controlled in order that the relationship (if any) between meteors and VLF transients could be isolated.

The variables include:
1. Expected frequency range of transients – various papers suggest around 1 kHz or from 6 to 10 kHz.

2. Variation in transient rates with time of day 
(ie. ionospheric condition)

3. Meteorological conditions - specifically lightning

4. Meteor Radar transmitter location and meteor trail echo geometry to the receiver. The transmitter needs to be reasonably close to the receiver / observing location for echoes to be from meteors that are local to the area and which may be observed visually.

5. Optical visibility – clouds and phase of the moon + light pollution levels

Choice of Meteor Radar Transmitter
Most amateurs make use of the French GRAVES transmitter on 143.05 MHz when receiving meteor echoes. If we need to try and correlate meteor echoes with optical sightings however, we need echoes from nearby the receiving station. This would not be the case for Graves as it is located in central France and beams southward.
An alternative is to use the Belgian BRAMS transmitter at approximately 50 MHz as it is more local, but the transmitter power is only 150 Watts on 49.97 MHz and 50 Watts on 49.99 MHz. Therefore the number and strength of echoes is likely to be smaller than for the much more powerful Graves system.

As a preliminary experiment it was necessary to establish a detection threshold and echo rate for the two possible BRAMS transmitters to decide which one to use for the meteor shower measurements. This should be done in the period where only sporadic meteors are present – signals will be stronger and more frequent during the meteor shower.

Optical detection

Ideally we should like to be able to correlate the meteor radar echoes with visual sightings. This can be done by the traditional manual method, recording count rates and brightness on paper forms by observers.

A simple switched three level voltage source has been developed which has buttons for "faint, medium and bright" meteors. Each time one of these buttons is pressed the voltage reading is recorded on a laptop against time of day – replacing paper records. Later analysis can be performed on a spreadsheet to count / time each type of meteor.

It may also be possible to use low light TV cameras and low light webcams to make recordings of visible meteor trails. The aim will be to time-stamp the recordings made using VHS or DVD recorders to enable post measurement identification with radar echoes and any VLF transients.


Low light CCD camera

Low Light Camera Test

Measurement of VLF Transients during 2015 Perseid Meteor Shower

Information was gathered during the 12 hours through the nights of 5-6/8/15 to 17-18/8/15 covering the Perseid event. Although the data was gathered seamlessly every half second during the 12 hour periods, it was analysed in three four-hour ‘chunks’ each of ~ 28000 data samples. This choice was largely determined by the graph plotting limitations of Microsoft Excel, which restricts data files to <32000 points. 

Measurements made every half second included:
Type 1 Data - (.CSV file)
  •  IEPER Radar echo received signal level
  •  VLF signal level in two frequency bands 0.3 – 2kHz    (called the 1kHz Band)
  • 5 – 8 kHz (called the 6 kHz Band)
  •  RADAR Trigger (when the radar signal exceeded a preset threshold level)
  •  VLF Trigger (when the VLF signal exceeded a preset threshold level)

When the IEPER Beacon radar echo trigger was activated, the following data was automatically stored on PC hard drives:

Type 2 Data – Primary data
Recorded around days near the peak of the meteor shower:
  •  A 15 second long waterfall plot of Radar Echo signal and the 1kHz Band VLF signal on PC1.
  • A15 second long waterfall plot of the Radar Echo and the 6 kHz Band VLF signal on PC 2.
  •  A 4 second audio recording in .wav format of the echo and 1 kHz band VLF signals on PC1.
  • A 4 second audio recording in .wav format of the echo and 6 kHz band VLF signals on PC2.
If long echoes were detected the audio recordings continued through the period of the echo. In some cases these lasted for around 1 minute.

The Type 1 data was intended for statistical analysis, whereas Type 2 data would enable detailed scrutiny of single events.

The table opposite shows the 1kHz Band VLF & meteor data collected.

The optical Zenith Hourly Rate (ZHR) data from the IMO for the 2015 Perseid meteor shower is shown opposite and displays a strong peak on the night of 12th -13th of August. This plot was be used to compare meteor rate with both VLF transient activity and IEPER Radar Echo behavior from the 5th to the 18th of August.

The 1 kHz Band (0.3 - 2kHz) VLF transient hourly rate was plotted with the IMO optical meteor rate as shown below. There is a definite correlation, indicating some connection between the advent of the meteor shower and the increase in VLF transients.


The following conclusions were drawn from the analysis of these experiments :

VLF Transients
 The number of VLF transients increased during the 2015 Perseid meteor shower and reached a maximum at the same time as the observed Zenith Hourly Rate (ZHR) produced by the International Meteor Organisation (IMO).

 This finding supports the results of Price & Blum obtained during the 1999 Leonid meteor shower.

 The form of the number of VLF transients followed the 2015 Perseid ZHR profile.

 At the shower peak, the number of VLF transients in a 4 hour period was approximately six times the number before the shower began.

 The above conclusions apply to VLF transients detected in both the 1 kHz Band (0.3-2kHz) and the 6 kHz Band (5-8kHz. However the effect was more pronounced for the lower frequency band

 The number of VLF Transients showed a nocturnal variability for all 12 days where they were monitored during the meteor shower

 The same nocturnal variation was observed during a non shower period some weeks before the Perseid meteor shower and it is suggested that this is due to improved VLF propagation at night.

 By using the same 4 hour period data blocks to compare counts from day to day it is possible to minimise the effect of nocturnal variability.

 The number of VLF transients at any time also depends on the quantity and distribution of lightning in the UK and Europe. But this cannot be accurately quantified at present.

 The profile of lightning activity over the period of the Perseid meteor shower appears to be independent of the meteor shower and does not account for the increase in VLF transients which broadly follows the Perseid ZHR.

 It has not been possible to fully evaluate the contribution made by lightning to the VLF transient counts as a suitable lightning metric depends on the geographical coverage of the VLF receiver used in these experiments and which is presently undefined.

Generation Mechanisms
 The physical mechanism for the generation of VLF transients by meteors that has been proposed in the academic papers is concerned with the interaction of turbulent plasma in the meteor tail and the geomagnetic field. If this is so, it should be possible to relate transients and meteor echoes closely in time but there appears to be no proof of this.

 If this is the operative mechanism, the experiments reported here have failed to find clear evidence for it.

 It is possible that the measurements are not sophisticated enough to reduce the data to a subset of echoes and transients from the same location – as well as the same time.

 If the assumed mechanism is not correct and coincidences between echoes and transients are essentially random, then we may speculate about secondary mechanisms that could occur after the meteor event which then give rise to a VLF transient.

 Evidence of the relationship between increased lightning rates and cosmic ray showers and solar wind impacts, suggests the existence of some mechanism whereby energetic particles can trigger lightning or other discharge events resulting in VLF transients.

 Perhaps some related mechanism could account for the increase in VLF transients during meteor showers?

 A more thorough examination of these possibilities should be considered.

Finally - an example of the IEPER Beacon meteor echo and a VLF transient that ' 'appears' to coincide with it is shown opposite. It was not possible to prove a causal relationship between VLF transients and meteors on a one-to one basis.

Much more information on this measurement campaign can be found by opening the file  
at the base of this page.

The range of frequencies considered to be in VLF band are shown below as part of the electromagnetic spectrum.

Literature suggests that meteor generated VLF signal have frequencies between a few hundred Hz (ULF Band) up to perhaps 10 kHz in the VLF Band.

Early work suggested that VLF transients were only produced by meteor Fireballs.

A typical Fireball event

Some History.

There is no doubt about the electrophonic effects of a large bolide seen over England on the 19th of March, 1719. 

Edmund Halley (1719) reported some eye-witnesses "hearing it hiss as it went along, as if it had been very near at hand," but he dismissed such claims as "the effect of pure fantasy".

                             Edmund Halley

" This rejection is related to Halley's realisation, by careful triangulation from many observations, that "they abundantly evince the height thereof to have exceeded 60 English miles, which is far too distant for sound waves to arrive instantly". 

Halley was one of the first to show that meteors occur at a great height compared to most other atmospheric phenomena and that their velocity was "incredible", being "above 300 such miles in a minute."

Typical Large Loop antenna for VLF measurements

Hourly counts of optically observed meteors during the night of 17 -18 Nov 1999

An example of the VLF and meteor echo data gathered by the system is shown below.  It is possible to see the VLF transients or 'spikes' together with the meteor echo.

One of difficulties to be expected in the analysis is to distinguish between  VLF transients produced by meteors and those from lighting events. It is therefore necessary to have some measure of lightning rates during the observation. This was done by monitoring the output from Blitzortung .org website. An example of a typical lighting plot over Europe is given below.

More information about the experimental set-up can be obtained by opening the file
{ Identifying VLF transients from Meteors Part 1 (1).pdf   }  at the base of this page.

Some trial optical measurements should be made before the meteor shower to establish the right balance between system sensitivity and image integration time.

Bright stars can be used for this purpose if the sky is exposed for a second or so to mimic a meteor. 

A report on the measurements made of the sporadic meteor background - before the 2015 Perseids can be found by opening the file 
at the base of this page.

The 6 kHz Band (5 - 8kHz) VLF transient hourly rate was plotted with the IMO visual data as shown below. There is some correlation between the two data sets but it is not as marked as for the 1kHz Band.


IEPER Radar Echoes
 The number of radar echoes received from the 49.99MHz transmitter at IEPER in Belgium increased during the Perseid meteor shower.

 The form of the radar echo count followed closely the IMO ZHR profile.

 The radar echo count in a 4 hour period at the shower maximum was approximately 10 times that before the shower began.

 The number of strong, long echoes increased toward the shower maximum.

 The geographical area for which IEPER beacon echoes are generated is not known and this limits the ability to correlate echoes with transients.

Echo and Transient Coincidences (Type 1 Data)
 The simple statistical tests to uncover coincidences between meteor echoes and VLF transients produced a profile that followed the ZHR to a high degree.

 This is not surprising since it results from the combination of the VLF transient and meteor echo profiles.

 The coincidences appear to be random, as for all 12 days where measurement were made, the number of ‘Actual’ (at the same time) and ‘Delayed’ (VLF transients 5 seconds after the echo) were approximately the same.

 The % difference between ‘Actual’ and ‘Delayed’ coincidence counts tended to zero as the number of events increased through the Perseid shower maximum. This also suggests a degree of randomness.

 With this information it has not been possible to relate individual Echoes and transients and unfortunately sheds no light on the causal mechanism of VLF generation by meteors.

 The assumption that VLF transients and radar echoes occur close together (within the 0.5 second sample time) may be flawed. If so this will invalidate the ‘Actual’ vs ‘Delayed’ test and any conclusions about a random connection.

Echo and Transient Coincidences (Type 2 Data)
 Over 8,000 echo and transient waterfall plots were captured during the 12 day measurement period from 5/8/15 to 18/8/15.

 Over 8,000 audio recordings were made of both echo and VLF signals.

 Detailed examination of audio files for a number of strong, long IEPER beacon echoes and VLF transients that occurred at the same time, failed to show a clear or special coincidence of a transient with an echo.

 If a direct causal relationship exists between VLF transients and the meteors which are thought to generate them, this may be overshadowed by the reception of a large number of unrelated VLF signals from all over Europe produced by lightning and other sources.

 Only with a more sophisticated measurement system will it be possible to sift out a subset of echoes and transients that arrive not only at the same time – but also from the same location !

 It may be possible to examine the audio files in greater detail using bespoke software that might find evidence of echo and VLF transient coincidences.

 Collaboration would be welcomed.

David Morgan,
9 Feb 2016, 01:59
David Morgan,
9 Feb 2016, 02:03
David Morgan,
9 Feb 2016, 03:03
David Morgan,
9 Feb 2016, 01:29