VLF radiation in the
atmosphere
Israel Silber
What is VLF?
Electro-Magnetic (EM) waves in the
frequency range of 3-30KHz, generated by
various natural and man-made sources.
So, what are these sources?
Natural VLF Sources
•
•
•
•
•
Volcanic eruptions.
Dust storms.
Tornadoes.
Aurora.
Lightning
Discharges.
Man-Made VLF Sources
• VLF transmitters.
• ‘Wireless’ Antennas:
– HF heaters.
– Cubic non-linearity
mechanism.
• Nuclear explosions.
Earth’s Ionosphere
•
•
•
The charged part of the atmosphere (partially ionized gas).
Solar UV and X-ray radiation dominates the ionization processes during the day.
During the night, non-solar sources (e.g., cosmic rays, meteoric ionization, etc.)
maintain the smaller free electrons and ions concentrations.
Earth’s Ionosphere
• The ionosphere is divided into
four main regions: D, E, F1, and
F2 regions. Each of the regions
is characterized by its ion’s
composition and concentration,
as well as its own physics.
• The D and F1 regions vanish at
night, and the E region becomes
much weaker, because the
composition of these regions, is
mainly of molecular ions.
Hargreaves, 1995
The Earth-Ionosphere Waveguide
• Due to the high concentration of electrons and ions in
it, the ionosphere may be treated as a conductor. As a
result, a waveguide is created between the conductive
Earth and the ionosphere.
• Propagating Electro-Magnetic (EM) waves within the
waveguide, are reflected at an altitude that satisfies:
1
f N  f cos i0
Where f is the wave’s frequency, fN is the plasma
frequency, and i0 is the angle of the radio wave from
zenith.
The Earth-Ionosphere Waveguide
• Propagating Electro-Magnetic (EM) waves within the
waveguide, are reflected at an altitude that satisfies:
1
f N  f cos i0
Where f is the wave’s frequency, fN is the plasma
frequency, and i0 is the angle of the radio wave from
zenith.
1
• Because for electrons f N   80.5 N  2 , where N is the
electron number density, a higher incident angle
demands a lower N for reflection, thus leading to the
reflection of the wave at a lower altitude.
The Earth-Ionosphere Waveguide
NASA, 1998
VLF Reflection Height
• VLF signals are usually reflected by the D region, at an
altitude of ~65 km during daytime, and ~85 km during
nighttime. The received signals inherently contain
information of the reflection height’s region and its
variability.
• Therefore, VLF measurements allow studying the
ionosphere, mainly in the D region. This is a great
advantage, as these heights are too high to study them by
using balloons, and too low for in-situ measurements of
satellites. Moreover, measurements with the use of
rockets, are very transient and spatially limited.
The Regents of
the University of
California, 2003
VLF Signals Attenuation
• VLF radio signals propagate within the waveguide over long
distances (tens of megameters) with low attenuation (~2-3 dB
per Mm).
Barr et al., 2000
VLF Cut-off Frequency
• Every waveguide has a cut-off frequency determined by
its dimensions, as described by:
 2
f c  nc 2h
where, n is the waveguide mode, c is the speed of light
in vacuum, and h the distance between the waveguide’s
boundaries.
• In the case of a VLF signal’s propagation in the Earthionosphere waveguide, the boundaries are the D region
reflection height and the Earth’s surface.
• For the average VLF reflection height of ~75 km, the
received cut-off frequency is fc≈2 kHz.
Group Velocity Near the Cut-off
Frequency
• An EM pulse propagating within a waveguide will have a group
velocity determined by the equation (in c.g.s units)
 3
v g  c 1   fc f 
2
• Thus, the closer the frequencies of the pulse’s components to the
cut-off frequency, the slower these component will propagate.
• This will result in the pulse’s dispersion and the change of its
structure as a function of time and distance from the source.
• The group velocity represents the propagation speed of the pulse’s
energy. Thus, the energy of a component with a frequency f=fc won’t
propagate within the waveguide, as the group velocity is 0 at fc.
• Moreover, signals with frequencies close to fc will be strongly
attenuated (strong attenuation constant β – see Jackson, 1962,
section 8.5).
Data Acquisition
• VLF radio signals are received in loop
antennas, following the Maxwell-Faraday
equation:
 4

 E   B
t
where, E and B are the electric and magnetic
fields, respectively.
Tjostheim, 1997
Data Acquisition
• In order to intensify the received signal, the
loop’s wire is usually coiled. This amplification
of the signal is described by Faraday’s law of
induction (received by using the Stokes’s
theorem over equation (4)):
 5
 B
 N
t
where,  is the magnitude of electromotive
force (measured in volts), N is the number of
turns of the wire, and  B is the magnetic flux.
Data Acquisition
• Probably, all magnetic VLF receiving systems operate
two orthogonal loop antennas, aligned to the northsouth and east-west directions of propagation, in
order to derive the signal’s source azimuth.
• According Poynting’s vector equation
 6
S
1
EB
0
where, S is the Poynting vector and  0 is the vacuum
permeability, the direction of the loop’s alignment is
also the direction from which the energy has arrived
to that component.
Data Acquisition
Line
Receiver
Computer
Data Acquisition
Mount Hermon station
Sde-Boker station
Data Acquisition Modes
• There are two modes for VLF data acquisition:
– Broadband.
– Narrowband.
• Broadband data acquisition records full range VLF
amplitude data time series, which can be later
transformed into the frequency domain, by using
windowed FFT, in order to get the VLF dynamic spectrum.
Broadband data is usually recorded in synoptic mode.
• Narrowband data acquisition records phase and
amplitude in a specific frequency, which corresponds to
the frequency of a VLF transmitter, located somewhere
around the globe.
Broadband Measurements
• Broadband data give the option of studying lightning strokes
and their waveforms.
• Among other uses of the broadband data, The VLF reflection
height between a signal’s source and a receiver can be
deduced by analyzing the received waveform (e.g., Lay and
Shao, 2011).
• In addition, a signal’s dynamic spectrum informs us of its
propagation path and distance from the receiver, e.g., a
“tail”/”hook” near the cut-off frequency tells us that the signal
has propagated long enough for it to be strongly disperesed
(remember equation (3)?).
• Thus, VLF signals may be classified by their dynamic spectrum,
e.g., sferics, tweeks, whistlers, etc.
Sferics
kHz
• As large part of the VLF range is comparable
with the human hearing range (20-20,000Hz),
VLF signals “can be heard” if fed directly into a
speaker.
NASA, 2011
Sferics
•
•
Sferics, short for "atmospherics", are impulsive signals emitted by lightning.
They are caused by lightning strokes within a couple of thousand kilometers away from
the receiver.
The spectrograms of sferics are characterized by vertical lines on the frequency-time
graph indicating the simultaneous arrival of all of the audio frequencies.
The sound of sferics consists of sharp crackling noises like twigs snapping or sizzling
noises like bacon frying.
kHz
•
•
NASA, 2011
Tweeks
•
•
Tweeks result when sferics are ducted in the earth-ionosphere waveguide distances
much greater than a couple of thousand kilometers (up to 20,000 kilometers).
The spectrogram of a tweek shows a vertical line at the higher frequencies with a
curved section (“hook”), appearing at ~2 kHz, as a result of the signal’s dispersion.
Tweeks sound much different than sferics. Instead of the sharp crackling sound, tweeks
have a quick musical sound somewhat like the ricochet sound bullets make (at least in
the movies).
kHz
•
NASA, 2011
Whistlers
•
•
•
Under certain conditions, the VLF signal penetrates the ionosphere and returns by
traveling along a magnetic field line. During this long path, dispersion is much greater
than with tweeks.
While tweeks might disperse a few hundred kHz over a few thousandths of a second,
whistlers show a dispersion of a second or more over several thousand kHz.
On the spectrogram, whistlers appear as long sweeping arcs showing the sequential
arrival of the frequencies.
The sound of a whistler is a musical descending tone that lasts for a second or more.
kHz
•
NASA, 2011
Whistler Echo Train
kHz
• Echo trains result when the radio wave bounces back and forth between
magnetic conjugate points. Each time the signal bounces off the ionosphere,
some of the energy leaks down in the lower atmosphere and is heard as a
whistler.
• All of the whistlers in the train are the result of a single lightning stroke.
• Successive "hops" of the whistler are seen with increasing dispersion time as
the distance traveled grows with each bounce.
NASA, 2011
Chorus
kHz
• Occasionally, especially in the quiet times of the morning, a
chorus can be heard.
• Chorus seems to be the result of many brief, short-path whistlerlike emissions occurring at almost the same time.
• Chorus sounds like many birds calling in turn.
NASA, 2011
Someone Will Always Push It To the
Limit…
McGreevy, 2012
Broadband Time Series and Dynamic
Spectrum From Sde-Boker
Narrowband Measurements
• VLF transmitters broadcast at
a
constant
frequency,
amplitude, and phase.
• A transmitter’s signals reach
the
receiver’s
antenna
through the Great Circle Path
(GCP)
between
the
transmitter and receiver,
according to the Fermat’s
Principle.
[dB]
Narrowband Amplitude Time Series
From Mount Hermon
Lightning Induced VLF Narrowband
Perturbations
• High altitude atmospheric phenomena (e.g., Lightning Induced
Electron Precipitation (LEP), sprites, elves, meteor showers, etc.),
create modifications in Earth-ionosphere waveguide (mainly in the Dregion), by ionizing, heating, or adding ions and electrons to the
phenomenon’s region.
• Because VLF transmitters generally broadcast at a constant amplitude
and phase, modifications of the received signals can be used to
“probe” the D-region (and hence, the waveguide’s medium),
providing an important measurement technique of localized
disturbances.
• These disturbances in the received signal can be classified (by their
duration, onset time, etc.), into several typical categories, each with
its own probable source.
Trimpis’
Inan et al., 2010
Trimpis’
Inan et al., 2010
Trimpis’
• Source:
Whistlers and LEP
Fry, 2010
Early/Fast Events
Armstrong, (as Cited in Inan et al., 2010)
Early/Fast Events
• Source:
Sprites?
Elves?
Inan et al., 2010
Trimpis’ Vs. Early/Fast Events
Early/Slow Events
Inan et al., 2010
Early/Slow Events
• Source:
IC lightning Induced Elves?
Inan et al., 2010
“Long recovery” Events
Inan et al., 2010
“Long recovery” Events
• Source:
Gigantic Jets?
Elves? [Haldoupis
et al., 2013]
NASA Images
Comparison Between
Perturbations
Trimpi’s
Early/Fast
Early/Slow
Long Recovery
Probable source
Whistlers (LEP)
Sprites/Elves
IC induced Elves
Gigantic Jets
Location in
respect of
lightning
usually poleward
to lightning
above lightning
above lightning
above lightning
Area
Wide region
Small region
Small region
Small region
Delay from
lightning
1-2 seconds
<100 ms
<100 ms
<100 ms
Common amp.
change
negative
positive
positive
positive
Typical onset
time
1-2 seconds
>50 ms
1-2 seconds
>50 ms
Conclusions
• VLF measurements, both in broadband and
narrowband data acquisition modes, are a low cost
powerful tool for monitoring and studying the D region,
as well as lightning discharges, its accompanied
Transient Luminous Events (TLEs), and also other
phenomena (Meteor showers, solar flares, etc.).
•VLF Rulzzz!!!!!
Links Between Mesopause
Temperatures and Ground-Based
VLF Narrowband Radio Signals
Israel Silber1, Colin Price1, Craig J. Rodger2, and
Christos Haldoupis3
1Department
of Geophysical, Atmospheric and Planetary Sciences Tel-Aviv
University, Israel
2Department of Physics, University of Otago, New Zealand
3Physics Department, University of Crete, Greece
[email protected]
[email protected]
Atmospheric Region of Interest
Mesosphere-Lower
Thermosphere (MLT)
region (~60-80km up
to ~100km).
This region consists:
o The Mesopause.
o The ionospheric D-layer
and part of the E-layer.
Kamide and Chian, 2007
Initial Motivation – The Long Term
Changes
• Troposphere warming.
• Cooling of the stratosphere
and mesosphere.
• Downward displacement of
ionospheric layers.
Lastovicka et al., 2006
SABER and VLF Data (MH-NSC is Shown)
NSC,
Sicily
45.9KHz
2,070Km from MH
SABER and VLF Data (MH-NSC)
VLF data was taken as the
average of 12-13UT (during the
most stable period of the day).
Observation Systems
• VLF antennas.
• The SABER instrument
(on-board the TIMED
satellite).
Sde-Boker, Israel
GATS Inc., 2011
VLF NB Amplitudes and
Temperatures Correlations
DN-NWC 2006-7
SABER and VLF Correlation’s
Summary
-Periods with low amount of data points were compared after a 15-days
running average instead of 20.
Days of running
Max negative R
MAX Negative
Significant (Paverage
correlation coefficient correlation Height [Km]
val<0.05)
SB-NRK2007 23/03/2007-31/08/2007
15
0.66423-26
X
SB-NRK2008 08/04/2008-02/08/2008
15
0.97774-77
V
SB-NRK2009 10/05/2009-10/12/2009
20
0.82889-92
V
SB-DHO2007 23/03/2007-31/08/2007
15
0.81065-68
V
SB-DHO2008 08/04/2008-02/08/2008
15
0.93877-80
V
SB-DHO2009 10/05/2009-10/12/2009
20
0.89086-89
V
SB-DHO2010 12/04/2010-02/08/2010
15
0.92626-29
V
SB-NSC2007 23/03/2007-31/08/2007
15
0.83480-83
V
SB-NSC2008 08/04/2008-02/08/2008
15
0.99944-47
V
SB-NSC2009 10/05/2009-10/12/2009
20
0.91486-89
V
SB-NSC2010 12/04/2010-02/08/2010
15
0.75835-38
X
MH-NSC2009-11 01/04/2009-31/04/2011
20
0.50986-89
V
DN-NWC2006-7 0106/2006-31/05/2007
20
0.88880-83
V
CR-DHO2008-9 01/08/2009-31/12/2009
20
0.84695-98
V
Date set
Data set period
High correlation between VLF mid-day amplitude and mesopause
temperatures in many data sets.
Applying PCA on the Measurements
• VLF mid-day amplitude.
• SABER 80-90Km average temperature.
• Total Solar Irradiance (TSI) , multiplied with the solar
zenith angle for mid-day at the middle of the
Transmitter-Receiver GCP.
-All data sets were standardized for a correct PCA
analysis.
-Data gaps were filled with harmonic analysis.
PCA of Mesopause Temperatures,
VLF, and TSI
Attributed to the solar
annual cycle
74.5% of
variance
Attributed to short time
scale forcing on VLF and
mesopause
temperatures (not
connected to the TSI).
17.9% of
variance
Attributed to solar
insolation changes with a
positive feedback,
contrary to the negative
feedback in PC1.
7.4% of
variance
PCA Summary
Average:
PC1 - ~60%
PC2 - ~28%
PC3 - ~11%
Mesopause Temperature Estimation
Using the PCA
Conclusions
• The relation between VLF NB amplitudes and mesopause
temperatures which was found in this study, being either
direct or indirect, has not been identified previously.
However, a connection between mesopause temperatures
and radio waves absorption has been predicted before [e.g.,
Taubenheim, 1983].
• The variability of the UMLT temperatures and VLF
amplitudes can be explained by global seasonal solar
irradiance changes (~72% of the variability), while the
remaining variability has its origins from other sources
(~28%).
• High resolution mesopause temperature estimates might
be achieved in the future by combining VLF NB observations
and calculated solar irradiance variability (as a function of
hour, day, and location, i.e., latitude).
Thank You!
References
Barr, R., D. Llanwyn Jones, and C. J. Rodger (2000), ELF and VLF radio waves, Jr. Atmos. Sol.
Terr. Phys., 62, 1689–1718.
Fry,
GATS
A. B. (2010), On The Phenomena Of Lightning, The Astronomist,
http://theastronomist.fieldofscience.com/2010/03/on-phenomena-oflightning.html.
Inc.,
Atmospheric
Science
inc.com/projects_saber.htm.
(2011),
SABER,
http://www.gats-
Hargreaves, J. K. (1995), The Solar-terrestrial Environment: An Introduction to Geospace –
The Science of the Terrestrial Upper Atmosphere, Ionosphere, and Magnetosphere,
pp. 6-9, 13-29, 208 – 232, Cambridge Univ. Press, New York.
NaitAmor, S., M. A. AlAbdoadaim, M. B. Cohen, B. R. T. Cotts, S. Soula, O. hanrion, T.
Neubert, and T. Abdelatif (2010), VLF observations of ionospheric disturbances in
association with TLEs from the EuroSprite-2007 campaign, J. Geophys. Res., 115,
A00E47, doi:10.1029/2009JA015026.
NASA (1998), Radio Jove,
http://ds9.ssl.berkeley.edu/lws_gems/6/secef_9.htm
http://radiojove.gsfc.nasa.gov/education/educ/radio/tran-rec/exerc/iono.htm.
NASA Images,
http://www.nasaimages.org/luna/servlet/detail/NVA2~4~4~6449~106975:GiganticJets-Over-Oklahoma.
Hunsucker, R. D., and J. K. Hargreaves (2003), The high-latitude ionosphere and its effects on
radio propagation, pp. 140-169, Cambridge Univ. Press, New York.
Reuveni, Y., and C. Price (2009), A new approach for monitoring the 27-day solar rotation
using VLF radio signals on the Earth’s surface, J. Geophys. Res., 114, A10306,
doi:10.1029/2009JA014364.
Haldoupis, C., M. B. Cohen, E. Arnone, B. Cotts, and S. Dietrich (2013), Step-like and long
recovery Early VLF perturbations caused by EM pulses radiated by powerful +CG
llightning discharges, J. Geophys. Res., Submitted.
Rodger, C.J. (1999), Red sprites, upward lightning, and VLF perturbations, Reviews of
Geophysics, 37, 317-336.
Inan, U. S., S. A. Cummer, and R. A. Marshall (2010), A survey of ELF and VLF research on
lightning-ionosphere interactions and causative discharges, J. Geophys. Res., 115,
A00E36, doi:10.1029/2009JA014775.
Rodger, C. J., and R. J. McCormick (2006), Remote sensing of the upper atmosphere by VLF,
Sprites, Elves and Intense Lightning Discharges, Nat. Sci. Ser., vol. 225, edited by M.
Fullekrug, E. A. Mareev, and M. J. Rycroft, 167–190, Springer, Dordrecht,
Netherlands.
Kamide, Y, and A. Chian (2007), Handbook of the Solar-Terrestrial Environment, pp. 7-8, 189190, 197-199, 217, 223 Springer-Verlag Berlin Heidelberg.
Kelley, M. C. (2009), The Earth's ionosphere plasma physics and electrodynamics, pp. 4 - 16,
Elsevier, London.
Jackson, J. D. (1962), Classical Electrodynamics, pp. 170-173, 189-190, 208-212, 235-264,
John Wiley and Sons, New-York.
Lastovicka, J., R.A. Akmaev, G. Beig, J. Bremer, and J.T. Emmert (2006), Global change in the
upper atmosphere, Science, 314, 1253–1254.
Lay, E. H., and X.-M. Shao (2011), High temporal and spatial-resolution detection of D-layer
fluctuations by using time-domain lightning waveforms, J. Geophys. Res., 116,
A01317, doi:10.1029/2010JA016018.
Marshall, R. A., and U. S. Inan (2010), Two-dimensional frequency domain modeling of
lightning EMP-induced perturbations to VLF transmitter signals, J. Geophys. Res.,
115, A00E29, doi:10.1029/2009JA014761.
NASA (2011), The Inspire Project,
http://theinspireproject.org/default.asp?contentID=17.
Russell III, J. M., M. G. Mlynczak, L. L. Gordley, J. Tansock, and R. Esplin (1999), Overview of
the SABER experiment and preliminary calibration results, Proc. SPIE Int. Soc. Opt.
Eng., 3756, 277– 288, doi: 10.1117/12.366382.
Silber, I., C. Price, C. J. Rodger, and C. Haldoupis (2013), Links between mesopause
temperatures and ground-based VLF narrowband radio signals, J. Geophys. Res., In
Press. Taubenheim, J. (1983), Meteorological control of the D region, Space. Sci. Rev.,
34, 397-411, doi: 10.1007/BF00168831.
The
Regents of the University of California (2003),
http://ds9.ssl.berkeley.edu/lws_gems/6/secef_9.htm.
Living
with
a
star,
The World Data Center for Remote Sensing of the Atmosphere, NDMC Mission,
http://wdc.dlr.de/ndmc/index.php.
Tjostheim, T.A. (1997), Department of Electronic Engineering, University Of Surrey,
McGreevy, S. P. (2012) Natural VLF Radio Phenomena, http://www.auroralchorus.com/.
McRae, W. M., and N. R. Thomson (2000), VLF phase and amplitude: daytime ionospheric
parameters, Jr. Atmos. Sol. Terr. Phys., 62, 609-618, doi: 10.1016/S13646826(00)00027-4.
http://info.ee.surrey.ac.uk/Teaching/Courses/EFT/dynamics/html/faradayslaw.html.