Friday night (July 24/25)
was stunning. Coming home from a beer after local
midnight, I noticed spectacular VLF spherics resonances
on the DK7FC remote garden grabber (screenshot http://df6nm.bplaced.net/VLF/spherics/dk7fc_VLF_150725_0704.jpg from
the morning). I decided to restart the dual-loop
directional spectrogram, the one which sometimes feeds
my VLF wideband grabber.
After a couple of
minutes, the first spectrum appeared, and I was
literally swept away: Fat resonance peaks at all
multiples of 1.62 kHz up to the tenth harmonic, and up
to 20 dB above the noise: http://df6nm.bplaced.net/VLF/spherics/vlfwide_150725_0700_cursor-2347.png.
This was really surprising, as my suburban loops are
usually very prone to local interference, and I had
never seen or heard something like that before. The
colour direction finder displayed the resonances in
grey, which means "no direction", just as expected for a
vertical-incidence circular polarized wave.
http://df6nm.bplaced.net/VLF/spherics/tweek_resonances_dk7fc_df6nm_Blitzortung_150725_0700.png is
a combined image showing two screenshots from Stefan in
Heidelberg (top) and myself in Nuernberg (bottom),
stacked above one another with identical timescales. The
resonances in Heidelberg were strongest between 22 and
23 UT, became significantly weaker at 23:20, and died
out about 0:10 UT. Resonances in Nuernberg came about an
hour later, strongest at 23:40, and fadeout between 0:30
and 1:10 UT. Interestingly, the resonance frequency goes
through a flat minimum at the time of maximum amplitude
and sharpness, which is different for the two locations..
The spectrograms can be
compared to a Blitzortung screenshot taken at 0:40 UT,
inserted at the bottom left of the image. Blitzortung sorts displayed flashes in 20
minute bins, indicated by colours from red (oldest,
22:40 - 23 UT), through shades of orange, to white (last
20 minutes, i.e. 0:20 to 0:40 UT). For clarity, I
repeated the Blitzortung colours in rectangles
corresponding to the respective parts of the spectrogram
timescale.
The position of the
stormfront at the time of fadeout allows us to estimate
the lateral range across which resonance amplification
could be used to enhance communication. We find that tweek resonances were
strong up to about 90 km away, and became very weak at
about 180 km (indicated by blue circles around
Heidelberg). Unfortunately this means that the distance
between DK7FC and DF6NM (180 km) is probably already too
large to benefit much from such a vertical resonance
mode.
But on the other hand, it
is known that tweeks have often been heard from great
distances, up to 3000 or 6000 km over seawater. So why
are the resonances restricted to a couple of 100 km? I think the answer for this
is dispersion - the same effect that creates the
characteristic chirp sound in distant tweeks. For a
purely vertical wave, the time interval between
successive hops is always 2h/c, about 0.6 ms with h = 90
km ionosphere height- This is the case if the source is
close to the receiver. However at increased ranges, the path is
initially at lower angle, and becomes steeper and
steeper for later hops. Thus the time interval between
successive pulses is initially smaller, and grows
towards later (more vertical) hops. Source distance can
be calculated from a tweek's knee in a spectrogram - far
away sources have less rectangular, more rounded knees. In the resonance spectrum, this
means that energy from for further away sources will be
more spread out, and the maximum shifted to higher
frequencies. This seems to be exactly what we have
observed in the slow spectrograms: the sharpest
resonances at the lowest frequencies occur when the
storm passes just overhead.
At times, the
resonances even appeared to split up. Look at the zoomed
section around the fourth harmonic at the bottom: During
the maximum in Nuernberg (23:45 UT), there are two
distinct peaks, one at 6430 Hz and another at 6518
Hz. In the correponding orange Blitzortung crosses, we
indeed find two separate sections of lightning activity,
a smaller and nearer one passing south of Nuernberg,
and another broader one passing north.
Let's take another look in
time domain. http://df6nm.bplaced.net/VLF/spherics/dualloops_tweeks_scope.png are
SpecLab scope shots from a couple of tweeks from
recordings. For these, the frequency range had been
restricted to 4 - 16 kHz to eliminate interference from
3 kHz railway noise and military MSK stations, and
SpecLab's autonotch filter was engaged to
suppress residual mains and railway harmonics. The two
traces correspond to the two loops, red is the
North-South loop (actually pointing 330° azimuth), and
blue is East-West (60°). Tweek #3 shows a direct
groundwave received only by the red channel, and a
subsequent train of echoes which are received by both,
with slightly increasing echo spacing. We find that
undulations in the blue trace preceed those on the red
one, showing that the circular polarized magnetic field
vector rotates counterclockwise, i.e. from east to
north.
All the best,
Markus (DF6NM)
PS: I started this thread
on the rsgb LF reflector, but it may be appropriate to
post to the VLF / Natural radio group guess I'll just
do both.
Sent: Saturday, July 18, 2015 6:50 AM
Subject: VLF: Tweek mode resonances
The screenshot http://df6nm.bplaced.net/VLF/spherics/dk7fc_VLF_150718_1326.jpg shows a number of narrow tweek-mode resonances at
multiples of 1.72 kHz. These are obviously spherics from
nearby lightnings, bouncing multiple times vertically
between the ionosphere and ground (much the same as
clapping your hands between two parallel brick walls). The resonances are rather sharp,
indicating a high Q-number (ie. around 100 bounces until
decay). They are
visible up to about 20 kHz, showing unusually small
damping of vertical incidence reflections at these
frequencies. There is a small variation of resonance
frequency over time, reflecting the variable height of
the ionospheric ceiling. The fundamental resonance at
1.7 kHz is probably not visible due to the frequency
response of the loop and receiver.
The tweek resonances were received on the loop
antenna in the garden but not on the E-field antenna of
the (somewhat whitened out) city grabber
http://df6nm.bplaced.net/VLF/spherics/dk7fc_wideband_150718_0330.jpg.
This corroberates the notion of near vertical incidence
and horizontal H-field polarisation. According to the
literature, tweek tails are usually circular polarized
as only one sense of rotation exhibits a high reflection
coefficient. They are predominately excited by
horizontal current components in intra-cloud lightnings.
Of course the resonances will also be there in quiet
nights without spherics, so they could probably be
employed to enhance fieldstrength (up to a factor of Q)
for medium-range VLF communication experiments using
magnetic transmit and receive antennas. When Stefan
still had his earth dipole, we already discussed a 2 kHz
tweek-mode experiment, which for various reasons hasn't
taken place yet. Now it looks like one could even employ
the fifth mode near 8.5 kHz...
All the best,
Markus (DF6NM)
