Hello Stefan,

Most exciting to see your latest venture, near the 2kHz-4kHz 100km-1000km 
propagation wilderness (soon after reaching the attenuation summit* at ~ 17,000 
km / 17kHz)
* (signal strength increasing at distances greater than 17,000km/17kHz) 

Thunderstorms reduce (by up to 5%) the effective height of the ionosphere above 
them (increasing reflection and affecting the electron density gradient, the 
electron gyrofrequency and electron gyrofrequency gradient). Due to 
thunderstorm-induced changes in the four essential parameters (height, 
gyrofrequency and their gradients), a thunderstorm anywhere near an ELF-VLF 
transmitter-receiver path can significantly affect amplitude and phase of the 
received signal, and more substantially if the thunderstorm is strong and close 
to the transmitter or receiver. If many waveguide modes are involved, the 
phase/amplitude effects of the thunderstorm-induced changes to the ionosphere 
above the storm are greater. Shorter paths generally mean more waveguide modes 
(but very short paths mean groundwave stronger than all skywaves, so some 
intermediate distance range is worst case), and being near modal-cutoffs (~ 
1.6kHz-4 kHz) means even more variation, so received signal phase and amplitude 
in the region 2kHz-4kHz 100km-1000km are particularly sensitive to 
thunderstorms, time-of-day variations in the ionosphere, nighttime changes in 
the ionosphere, and small changes in transmitter-receiver separation distance. 
This makes 2kHz-4kHz 100km-1000km somewhat of a rough untracked propagation 
wilderness that will be described by experiments before models can accurately 
account for all of the variables, which makes your reports from the wilderness 
quite exciting.

The attached plot shows smooth-sailing at 1.57kHz, but the attached plot is for 
one particular scenario of season, time of day, surface type, solar weather 
condition, and tropospheric weather condition. It's certainly possible that 
some combination of these parameters, including the widespread strong 
thunderstorms that you reported, could elicit large amplitude and phase 
variations near 1.57 kHz. One could imagine the peaks and nulls in the attached 
plot moving left-right and many other ways as the four parameters vary due to 
thunderstorms (and as distance, surface type, time-of-day, etc. vary depending 
on the experimental configuration).
http://www.ittc.ku.edu/~callen/energy_harvesting/Cummer2000TAPpp1420-1429.pdf 
shows ~ 1.6kHz amplitude somewhat higher during daytime, or 20 dB lower during 
daytime, depending on model and conditions (including 3dB day/night difference 
at 200km, 20dB day/night difference at 50km, for one configuration in Figures 8 
vs 9 in the above paper). Widespread strong thunderstorms could affect 1.57 kHz 
propagation in many ways: given that thunderstorms reduce the height of the  
ionosphere above them as does daytime, and increase ULF/VLF reflectance as does 
daytime, one could suspect that thunderstorms might move a 1.6kHz null to a 
slightly higher frequency (as does daytime under particular conditions in 
Figures 6 vs 7 in the above paper), which would increase 1.57kHz signal 
strength without generally increasing lightning noise; realistically there are 
many other factors but it's safe to say that a substantial SNR peak during 
thunderstorms at 1.57kHz is not unlikely at short and intermediate distances. 

It's quite interesting to see your thunderstorm SNR data in the fascinating 
intermediate-distance, multi-mode waveguide-cutoff region; best of luck in the 
ongoing experiment at 1.57 kHz.

73,

Jim AA5BW
 

                   

-----Original Message-----
From: [email protected] 
[mailto:[email protected]] On Behalf Of DK7FC
Sent: Thursday, May 3, 2018 5:31 AM
To: [email protected]
Subject: Re: ULF: A new band is opened up! 191 km | DL0AO results

Paul, ULF,

There is an interesting unexpected observation: In the night of the 30th, there 
were strong thunderstorms all over Germany. The noise was very high of course. 
But when analysing the carrier for that single day (from the DL0AO data file), 
i'm getting a peak SNR of 10.98 dB for a 10 hour carrier starting 01:20 UTC. A 
6 hour carrier starting at the same time produces a similar SNR. That peak is 
the strongest in 5000 bins. 
SNR plot attached. I easily decoded a '*' message with good Eb/N0 and no false 
decodes.

I had serious doubts but then Markus got similar results with his totally 
different way of data analysis.

Can this be real? Could it be propagation? Could it be that the lightnings have 
an effect on the ionosphere to improve reflection significantly so that this 
S/N can be reached? You know that 191 km wave is at about lambda/2 at the 
height of the ionosphere relative to ground (at night).

We need more such data. The carrier is now running for more than 1 week without 
an interruption and it will continue further... Oh and we need more 
thunderstorms on the path! :-)

It would be most interesting to see if you're getting similar results from the 
data of DL4YHF, which is a different and longer path.

Another thought: I choose 1.57 kHz because this is below that 'cut-off' 
frequency for distant sferic propagation (the distant sferic S/N drops by say 
20 dB at 1.6 kHz at night), so the night QRN is much lower there. 
BUT, maybe it would be even better to transmit directly on that cut-off 
frequency, i.e. between 1.6...1.65 kHz?! No problem to move to that frequency 
but let's collect more data on 1570.01 Hz for the next weeks.
Looks like an ideal summer experiment!

73, Stefan


Am 01.05.2018 15:29, schrieb Paul Nicholson:
>
> So far, nothing at Todmorden, by stacking all the daytime signals, or 
> by chaining them for super narrow bandwidth.
>
> Some problems using Bielefeld vlf6, the stream often steps backwards 
> one sample at a time and I'm not sure where that leaves the phase.
>
> --
> Paul Nicholson
> --
>

E field, 1kHz-100kHz, 100km-1000km.jpg
Description: JPEG image