Indian Journal of Radio & Space Physics
Vol. 35, December 2006, pp. 396-400
Detection of 2004 Leonid meteor shower by observing its effects on
VLF transmission
S S De,1 B K De,2 A Guha,1 & P K Mandal1
1
Centre for Advanced Study in Radio Physics & Electronics, University of Calcutta, Kolkata 700 009, India
2
Department of Physics, Tripura University, Tripura 799 130, India
E-mail: [email protected]
Received 3 March 2006; revised 29 May 2006; accepted 26 October 2006
Results of the detection of 2004 Leonid meteor shower over Kolkata (22°34´ N, 88°30´ E), India is presented in this
paper, by using a VLF amplifier tuned at one of the transmission frequencies of Indian Navy Traffic stations at
Vijayananarayanam (8º25´59.88´´ N, 77°48´ E) at 16.3 kHz. The shower was predicted to exhibit a peak activity on 19 Nov.
2004. In spite of low ZHR predicted, the peak activity had been observed earlier than the predicted times, which confirms
the nongravitational ‘A2 effect’ on meteoroid trails. The observation also suggests that electromagnetic detection of meteor
shower is better than the visual observation, as any time of the day and night its effect on VLF transmission can be recorded.
Keywords: Meteor shower, VLF radio wave, ionospheric disturbances.
PACS No: 47.20.Ft, 96.50.Kr, 94.20.Vu, 95.85.Bh
1 Introduction
Meteors produce audible sound along with optical
signature during their entry through the earth’s
atmosphere1,2. These sounds are generated by
shockwaves or other audio signals produced by the
meteorite itself. The explanation of these sounds was
not so clear because of the time delay between optical
signals and audio signals coming from a meteor due
to the difference in their propagation velocities.
Also, meteors entering the earth’s atmosphere
produce electromagnetic waves in the VLF range due
to interaction with the medium, which propagate and
reach the ground at the same instance as the optical
signals3. The low frequency waves generate
electrophonic phenomena inducing perceptible sound
vibrations at electrically conducting objects near the
observer.
The present work envisages the detection of Leonid
Meteor shower by recording its effects on a
transmitted VLF signal VTX1 at 16.3 kHz from one
of the Indian Navy stations.
The Leonids emanate from the trail of comet
Tempel-Tuttle swings past the Sun approximately
every 33 years, and during each close approach it
emits a dense stream of dust and small particles. With
time, these dust trails extend along the whole length
of the comet's orbit, but the trails remain very narrow
and concentrated in space taking hundreds of years to
spread out. Comet Tempel-Tuttle revolves around the
Sun in the opposite direction to the Earth, so when the
Earth encounters the trails of particles, they enter the
atmosphere at very high speeds, about 257,495 kmph.
Most of the dust grains are very small, like grains of
sand, and vaporize on entry within the first few
seconds at heights around 96.5 km. Every year
roughly between 14 and 20 Nov. 2004, the Earth
meets a stream of ancient debris, leading to the annual
meteor shower visible in the early morning hours,
high in the north-east sky after midnight during those
dates. The cometary dust particles move in very
similar orbits and the resulting meteor shower appears
to radiate from a point in the constellation Leo, hence
the term Leonids.
It was thought that the Leonid period was over
after 2002 but there was a fair activity4 in 2003. For
2004, another activity was predicted. Not a storm
level, but the Earth was predicted to pass close to two
streams: the 1333 and the 1733. The 1733 stream was
already encountered in 2002 and the 1333 stream was
thought to be responsible from the 1998 storm, but
not all models agree about that. Nevertheless, the
three current models5-7 all agreed to have some
Leonid activity on 2004.
It was predicted that any outburst from the 1333
trail will peak at 0642 hrs UT, on 19 Nov 2004. Rates
would not be high, ZHR = 10 at best. The second
DE et al.: 2004 LEONID METEOR SHOWER EFFECT ON VLF TRANSMISSION
1733 trail would arrive at 2149 hrs UT, when the rates
could go up as high as ZHR = 65. Even though
predicted rates were not as high as in past Leonid
storms, it was important to continue observing these
showers to learn how dust is distributed by the parent
comet 55P/Tempel-Tuttle.
Many workers reported the detection of Leonid
showers by recording VLF signals emitted by the
meteors during their passage through ionosphere. VLF
signals at 19 kHz produced by 2002 Leonid meteor
shower has been recorded from Kolkata8. Generation
of electrophonics was also confirmed to be the results
of VLF emission during a meteor shower9.
The authors are able to continuously record a VLF
signal at 16.3 kHz, which is transmitted by VTX1
Indian Navy station located at Vijayananarayanam
(8°25´59.88´´ N, 77°48´ E), at a distance of about
2000 km from Kolkata. Continuous recording at this
frequency enables us to monitor ionospheric
disturbances due to lightning activity, solar flare, etc.
Recently, the authors observed the effect of recent
Leonid meteor shower on the propagation of VLF
signals. Regarding the effects of meteor propagation,
choice of VLF frequency is immaterial. In spite of
that, the transmitted signal chosen was at a frequency
where the influence of atmospherics is relatively low
compared to the atmospherics at 12 kHz and lower
frequencies. Moreover, suitable signal strength of
16.3 kHz at the receiving station led to its choice.
Interestingly, not only the shower could be detected,
but also the non-gravitational ‘A2 effect’ on Leonid
meteor trails could be confirmed.
2 Experimental set-up
To receive the VLF signal a horizontal 8 SWG
straight copper wire of length 120 m was installed as
an antenna at a height of 20 m from ground. The
antenna is sensitive to the vertical electric field of the
electromagnetic signal. To record the signal, a
Gyrator-II VLF receiver10 was fabricated. A block
diagram of the recording system is given in Fig. 1.
The VLF receiver was tuned at 16.3 kHz with a
quality factor around 250. The overall gain of the
amplifier is around 40 dB. The rms value of the signal
was recorded using a Pentium IV computer sound
card at a sample rate of 10/s. The recorded data were
analysed later using Origin 5.0.
3 Observational results
Two types of perturbation at the receiving signal
strength of VLF signal can happen due to the passage
397
Fig. 1 ⎯ Block diagram of VLF measurement system
of meteor through the ionosphere. Figure 2(a) depicts
the condition of normal one-hop VLF propagation in
a very simple manner. In normal condition and in a
given time frame, “signal 1” can reach the receiver
but “signal 2” cannot. Figure 2(b) shows how the
integrated VLF signal strength can increase due to
passage of meteor. The ionized column produced by
the meteor path reflects “signal 2” and it can easily
reach the receiving station increasing the effective
signal strength when both “signal 1” and “signal 2”
are in phase. On the other hand, the effective signal
strength at the receiving station can decrease due to
reflection of “signal 1” from the ionized path
produced by the meteor. The situation is shown in
Fig. 2 (c).
The days from 14 to 20 Nov. 2004 had very clear
sky and no serious ‘thunder-bolt’ related events were
reported at Kolkata11. So apart from the solarterrestrial influences on the ionosphere, the period
was ideal for observing meteor shower. Moreover, the
predicted peak activity periods on 19 Nov. 2004 were
around 0642 hrs UT and 2149 hrs UT, respectively
and no solar flare events were reported around that
time by GOES 10 and GOES 12 satellites that
continuously monitor solar activity12. At the predicted
peak activity period, there were no local lightning or
flare generated perturbations in the ionosphere that
could alter the average signal received at Kolkata.
A total of three occurrences of meteor shower were
detected. Two of these meteor showers were on 19
Nov. 2004 and one was on 22 Nov. 2004. The
variation of 16.3 kHz signal strength at the same
durations of the recorded meteor showers one day
before and one day after has been shown in Fig. 3
[(a)-(f)]. It is clear from the plots that in normal
condition, there is no noticeable variation in the signal
strength received at Kolkata.
Figures 4 and 5 show the effect of Leonid meteor
shower on VLF signal on 19th November. In both the
398
INDIAN J RADIO & SPACE PHYS, DECEMBER 2006
Fig. 2 ⎯ Schematic diagram showing (a) simplified one-hop VLF
signal path under normal atmospheric condition and (b) how
signal strength at the receiving station can increase (c) due to
meteor shower
cases, the signal level increased six to seven times the
normal value. The authors believe that the extra
ionization produced by the supersonic meteoroids
during their passage through lower ionosphere was
the cause of this enhancement. The ionized column
produced by the meteors reflected some signals from
the VLF transmitter, which could otherwise miss the
receiving station in undisturbed conditions. The first
occurrence of Leonid meteor shower on 19 Nov. 2004
has been analyzed event-wise and the result is shown
in Table 1. On the second occurrence of the shower
on 19 Nov. 2004, the successive events were so dense
that individual events could not be resolved as the
Fig. 3 ⎯ Plot of 16.3 kHz VLF signal strength on (a) 18 Nov.
2004, from 0530 to 0730 hrs UT; (b) 18 Nov. 2004, from 1600 to
0100 hrs UT; (c) 20 Nov. 2004, from 0530 to 0730 hrs UT; (d) 20
Nov. 2004, from 1600 to 0100 hrs UT; (e) 21 Nov. 2004, from
0530 to 0730 hrs UT and (f) 23 Nov. 2004, from 0530 to 0730 hrs
UT
resolution of the measurement system was not
sufficient.
Figure 6 shows the Leonid meteor shower on 22nd
November, 2004. The shower was not as strong as on
19th November. The individual events have been
analyzed and the results are presented in Table 2. One
interesting observation was that, both the increments
and decrements of signal level have been found
during the entry of the meteor into the atmosphere.
There was another interesting feature of the
observation. The peak activity was noticed earlier
DE et al.: 2004 LEONID METEOR SHOWER EFFECT ON VLF TRANSMISSION
399
Table 1 ⎯ Leonid shower data form on 19 Nov. 2004
(First Occurrence)
Events
Event property
Average signal Peak signal Enhancement Duration
level v1, mV
level v2, 20 log (v2/v1), of the
mV
dB
event, s
Fig. 4 ⎯ Leonid meteor shower on 19 Nov. 2004 (1st occurrence)
Event 1
Event 2
Event 3
Event 4
Event 5
Event 6
Event 7
Event 8
Event 9
Event 10
Event 11
Event 12
Event 13
Event 14
Event 15
34.81
59.66
58.93
62.33
56.88
50.45
42.35
46.05
45.85
45.03
43.62
45.88
50.83
32.87
51.28
36.7
66.05
66.91
70.82
60.70
54.70
50.02
52.9
48.89
48.51
48.62
51.79
58.70
40.20
58.98
0.46
0.88
1.1
1.11
0.56
0.7
1.45
1.2
0.56
0.65
0.94
1.05
1.25
1.75
1.22
107
32
92
42
58
111
42
37
46
42
88
200
280
131
300
Total period of the Shower (from 06:04:43 UT to 06:49:51 UT) ≈
45 min
Predicted visual zenithal hourly rate ≈ 10
Observed zenithal hourly rate from the variation of E.M. signal ≈
20
Fig. 5 ⎯ Leonid meteor shower on 19 Nov. 2004 (2nd occurrence)
than the predicted times. The first peak activity on 19
Nov. 2004 was predicted at 0642 hrs UT, but the peak
activity started at 0605 hrs UT. This is 37 min earlier
than the predicted time. The second activity started at
1842 hrs UT, which is 3 h 07 min earlier than the
scheduled time at 2149 hrs UT. Based on the reports
of International Meteor Organization13, Vaubaillon
et al.14 estimated that the times of peak activities
would be earlier than the predicted values due to nongravitational ‘A2 effect’ on 1333 and 1733 meteor
trails. Using some model calculations they showed
that the peak activity from 1333 trail would be earlier
by as much as five hours, from the predicted time and
peak activity from 1733 trail would be earlier by as
much as two to three hours, from the scheduled time.
Although the authors noticed 1333 trail effect to be
only 37 min earlier than the predicted time, taking
into account the ‘A2 effect”, the activation of peak
activity for 1733 trail matched closely with their
Fig. 6 ⎯ Leonid meteor shower on 22 Nov. 2004
observation. Here they observed the peak activity to
start at 3 h 7 min earlier than the predicted time.
4 Possible explanation of the ionized path
During the entry of the Leonid into the earth’s
atmosphere, there will be strong fluctuations of
charge distribution in the medium, which enhances
the rate at which the energy gets randomized. As a
result, instability is produced. For this, the relative
electron-ion drift velocity may exceed the value for
the onset of Kelvin-Helmoltz instability. The
compressible ionospheric plasma driven by velocity
INDIAN J RADIO & SPACE PHYS, DECEMBER 2006
400
Table 2⎯ Leonid shower data on 22 Nov. 2004
Events
Event property
Average
signal level
v1, mV
Event 1
Event 2
Event 3
Event 4
Event 5
Event 6
Event 7
Event 8
Event 9
Event 10
32.70
28.27
20.11
29.04
24.55
30.58
34.43
32.05
34.30
30.8
Peak signal Enhancement Duration of
level
20 log (v2/v1), the event,
dB
(Sec.)
v2, mV
11.61
32.59
11.85
37.18
11.96
49.46
11.48
11.92
13.07
49.37
-8.99
1.24
-4.59
2.15
-6.25
4.18
-9.54
-8.59
-8.38
4.1
145
102
205
390
200
185
435
125
210
-
Total Period of the Shower (from 05:30:00 UT to 07:23:59 UT) ≈
114 min
Observed Zenithal Hourly Rate from the variation of E.M. signal
≈5
shear and earth’s magnetic field at the frontal path of
the meteor increases the growth rate of KelvinHelmoltz instability. Strong turbulence is developed, in
which non-linear perturbation at the particle trajectories
towards the rear zone of the meteor acts to stabilize the
turbulent flow leaving a strongly ionized trail.
The station signal at 16.3 kHz frequency gets
depleted at this ionized zone during its travel towards
the ionospheric layer for reflection and for this, the
effective reception of the reflected signals at the
ground will be weak. Such ionized trail can also
contribute to the process of reflection of station
signals at the ground receiver along with the reflected
wave from the ionosphere and consequently higher
signal strength will be pronounced.
5 Discussion
The detection of meteor shower by electromagnetic
signals is very effective, since the presence of a
shower can also be detected during daytime unless
there is severe thunderstorm activity. If the system is
properly designed, it is also useful in detecting very
weak or low ZHR meteor showers, which are not
detectable with visual observation even at night. It
was daytime at 1212 hrs LT, when the first shower
occurred on 19 Nov. 2004. Nevertheless, it was
successfully observed using electromagnetic waves
from standard sources. To confirm non-gravitational
‘A2 effect’, it is presumed that simultaneous
observations are necessary, which will be possible by
synchronizing time at different places.
Acknowledgements
The authors are thankful to Indian Space Research
Organization (ISRO), Bangalore, for funding this
research through S K Mitra Centre for Research in
Space Environment, University of Calcutta, Kolkata
700 009, India.
References
1 Keay C S L, Anomalous sounds from the entry of meteor
fireballs, Science (USA), 210 (1980) 11.
2 Keay C S L, Progress in explaining the mysterious sounds
produced by very large meteor fireballs of large meteoroids
in Earth's atmosphere, J Sci Explor (USA), 7 (1993) 337.
3 Beech M, Brown P & Jones J, VLF detection of fireballs,
Earth Moon & Planets (Netherlands), 68 (1995) 181.
4 Vaubaillon J, Lyytinen E, Nissinen M & Asher D, The 2003
Leonid shower from different approaches, WGN J Intern
Meteor Org (Belgium), 31 (2003) 131.
5 McNaught R & Asher D, Leonid dust trails and meteor
storms, WGN (Belgium), 27 (1999) 85.
6 Lyytinen E & Flandern T V, Predicting the Strength of
Leonid Outbursts, Earth Moon & Planets (Netherlands), 82
(2000) 149.
7 Vaubaillon J, Activity level prediction for 2002 Leonids,
WGN J Intern Meteor Org (Belgium), 30 (2002) 114.
8 Chakrabarti S K, Pal S, Acharya K, Mandal S, Chakrabarti S,
Khan R & Bose B, VLF observation during Leonid meteor
shower–2002 from Kolkata, India, Indian J Phys, 76B (6)
(2002) 693.
9 Colin P & Blum M, ELF/VLF Radiation Produced by the
1999 Leonid Meteors, Earth Moon & Planets (Netherlands),
82 (2000) 545.
10 www.aavso.org.
11 www.imd.ernet.in.
12 www.sel.noaa.gov.
13 www.imo.net.
14 Vaubaillon J, Lyytinen E, Nissinen M & Asher D, The
unexpected 2004 Leonid meteor shower, WGN J Intern
Meteor Org (Belgium), 32:5 (2004) 125.