VLF: Mysterious Echo on NML LaMoure?

[Markus Vester <markusvester@aol.com>]

The signal from NML (LaMoure, North Dakota, 25.2 kHz) is showing the familiar ripple pattern of dual-path propagation: http://www.df6nm.darc.de/vlf/vlfgrabber.htm

Similar effects have been regularly observed on NWC (Australia, 19.8 kHz) and VTX1 (India, 16.3 kHz), due to superposition of short and long path components. The pitch of the ripples in the frequency domain is the inverse of the time delay, and thus proportional to the path difference.

The strange thing is that the pitch on NML is about 45 Hz, much wider than the others. This corresponds to a path difference of only 6.7 Mm. The direct great-circle range from LaMoure to Nuernberg is 7.4 Mm, thus the second component should have traveled around 14.1 Mm. Thus in this case, the long great-circle path (32.6 Mm) clearly has to be ruled out.

We also find that the pattern moves up slightly after our sunrise, indicating decreasing delay difference. This would imply that the delayed component seems to accelerate more during the progress of daylight than the direct component - quite the opposite of what is observed in the short-long path scenario.

So what is the physical mechanism of the echo? I really have no clue! A reflector on the North Pole?

I'm pretty sure that it is not a receiver artifact. As the strongest effect occurred around our sunrise (3 UT), I was tempted to speculate about "grey-line" propagation, picked up and guided by the terminator, which is currently not very far north of the Atlantic great circle path. But quite clearly the extra path length would be way too small, and we'd still have to assume some magic to explain most of the 22 ms delay.

Comparing ripple patterns simultaneously received at different locations could provide some insight. Even with relatively small receiver spacing (a few km between sites, e.g. the two receivers in Heidelberg), we might already observe slightly different phasing and thus shifted maxima. This could let us deduce the arrival angle of the late component. At larger RX spacing, the variation of time difference could be large enough to result in visibly different frequency pitch.

All the best,
Markus (DF6NM)

NML_multipath.jpg
Description: JPEG image

LaMoure_Nuernberg_3UT.jpg
Description: JPEG image