Unusual Auroral Observations in the 144 MHz Band

“Off-Normal Auroral Scattering” in Amateur Radio

Volker Grassmann, DF5AI

November 10, 2002


Radio amateurs report unusual Aurora communication which cannot be interpreted in terms of the ideal bistatic radar case which generally applies in Auroral backscattering in ham radio. This article considers Oliver's (DL1EJA) email information, see below, to analyse his Aurora QSOs to Peter, SM2CEW, in north Sweden. There is strong evidence that their observations may be interpreted by Auroral scattering which violates the ideal scatter geometry in field-aligned radio propagation. This phenomenon is known from high-powered ionospheric radar systems but in amateur radio it is reported for the first time.

Email from Oliver, DL1EJA, November 5 and 7, 2002

Hi Volker,
I read your article about Aurora/FAI with great interest. Your calculations fit with my observations in almost every contact I made in the past 10 years, but there are some QSO's to the SM3/OH-Region and specially to SM2CEW which can't be explained with this algorithms. I had a conversation with Peter, SM2CEW, after the Aurora-QSO and he told me it was a really Aurora with a QTF of 250° at his place and about 5-10° at my side. The scatter point was somewhere around JP43. I discussed this also with other OM like SM5BSZ, SM4IVE and they came to the same conclusion that there is a way to QSO with beam headings between 240-270° and 90-120° (to the south).
Oliver, DL1EJA (JO31DS)
Figure 1. 144 MHz array, DL1EJA in western Germany (JO31DS).Figure 2. Estimated common volume illuminated by the antenna systems.
Figure 3. 144 MHz array, SM2CEW in northern Sweden (KP15CR).

Auroral backscattering from JO31 to KP15

Using the BeamFinder analysis software [2], the E-region scatter volumes (110 km) and the corresponding Aurora dx radius is calculated for the geographical position of DL1EJA, see figure 4. The analysis is similiar to [3] and [4], in which the dx radius is calculated for various European regions. Figure 4 indicates that Auroral backscattering between the JO31 square and northern Sweden is impossible, see the blue area which covers the south of Sweden but excludes northern Scandinavia.

In the KP15 square, terrestrial communication associated with Auroral backscattering is not possible at all. This blanking effect is discussed in detail in [3] and [4]. However, the KP15 square just misses the opportunity of Aurora dxing because Auroral backscattering is available approximately 200 kilometers further south, see figure 7 in [3] and [4].

Thus, Auroral backscattering between JO31 and KP15 is impossible, i.e. DL1EJA and SM2CEW cannot establish QSO via Aurora. But they did and they did more than once. Taking Oliver's and Peter's antenna headings into consideration, we may localise their common scatter volume in the JP square above Norway, see figure 2.


Figure 4. Aurora dx radius (blue) and the corresponding scatter volumes (green) calculated for the JO31 square (DL1EJA).

The ideal geometry of Auroral backscattering

The geometry of Auroral and FAI backscattering of radiowaves is discussed in [1], [3] and [4]. Figure 5 shows the scatter geometry in the so-called bistatic radar case. The wave vector k_tx is directed from the transmitter to the scatter volume, and the wave vector k_rx is directed from the scatter volume to the receiver location. Neglecting possible Doppler effects, the wave vectors have identical length but, of course, different directions. In Aurora and FAI backscattering, the difference vector k_tx - k_rx is directed perpendicular to the magnetic field line passing through the scatter volume.

The field strength of the scattered radio wave drops significantly if the difference wave vector deviates from the normal direction. Aurora communication in amateur radio therefore corresponds to the ideal bistatic radar case because even high-powered amateur radio transmitters cannot compensate the degradation of the reflected radio wave in off-normal scattering (see figure 6).


Figure 5. Ideal bistatic radar case in which the difference between the wave vectors (black) is directed perpendicular to the magnetic field line (red).

Figure 6. Bistatic radar case in which the angle between the difference vector and the field line deviates from 90 degree (off-normal backscattering)

Off-normal Auroral scattering (ONA)

In figure 7, the red curves denote DL1EJA's and SM2CEW's range of vision (E-layer, 110 km). We may conclude that any scatter communication between JO31 and KP15 originates from a scatter volume within the corresponding overlap area.

Fortunately, BeamFinder's graphical user interface enables the user to change the scatter angle in FAI analyses, see the appendix. Adjusting this parameter empirically, a scatter volume was found which may indeed support Aurora communication between DL1EJA and SM2CEW, see the green dot marks in figure 7. Comparing this result to figure 2, the geographical position of the scatter volume corresponds to the common volume which results from the antenna headings.

Having identified the scatter volume, the calculation was reversed to analyse the corresponding dx target areas. The blue crosses in figure 7 display two remarkable features, i.e. a geographical band which extends from south Norway to the Bottnic gulf providing dx opportunities to DL1EJA and, on the other hand, a circular target area from Wales to Germany, providing potential dx targets to SM2CEW.


Figure 8. DX radius in off-normal Auroral scattering for the JO31 (left) and KP15 (right) square.

Figure 7. Auroral backscattering between JO31 and KP15 assuming an angle of 77.5 deg between the difference vector and the magnetic field line. The green dot marks (JP01/JP11) indicate the scatter volume which is not available in the ideal geometry of field-aligned backscattering.

Total ONA dx radius

The BeamFinder software was modified to calculate the dx radius in off-normal scattering similiar to the images in [3], [4] and in figure 4.

The surprising results are displayed in figure 8. In comparison to the "ordinary" Auroral backscattering (see the upper panels in figure 8), ONA significantly improves the dx reach in Auroral scatter communication. For example, VHF radio amateurs in western Germany (JO31 square) can establish Auroral scatter communication to Sweden (SM3 and SM2), Finland and even to the south coast of Iceland (see the panel labeled 77.5 degree on the left).

In northern Sweden (SM2), ONA is the only available Auroral scatter mode because "ordinary" Aurora communication is impossible.


In general, Auroral backscattering in amateur radio correspond to the ideal backscatter geometry in field-aligned radio propagation. However, this model cannot interpret the existance of unusual Aurora QSOs. There is evidence that those unusual Aurora QSOs may be interpreted in terms of an off-normal Aurora scattering (ONA). *)

Taking ONA scattering into account, we may conclude:

  • ONA predicts the existance of scatter volumes in situations in which the ordinary model does not provide any solutions
  • the geographical position of the scatter volumes correspond to the cross section of the antenna beams, see figure 2 and 7
  • ONA predicts long-distance Auroral scattering on 144 MHz from western Germany (JO31) to SM3, SM2 and OH which is not available in the ordinary Aurora model, see figure 7 and 8
  • ONA predicts long-distance Auroral scattering on 144 MHz from northern Sweden (SM2) to western Germany, the Netherlands and the UK which is not available in the ordinary Aurora model, see figure 7 and 8
  • ONA explains the azimuth range of the dx stations worked from northern Scandinavia, see Oliver's email ("beam headings between 240-270 deg and 90-120 deg") and the blanking of southern azmuths in the lower right panel of figure 8

On the other hand, the findings are surprising from a geophysical point of view because off-normal scattering was believed excluded in amateur radio communication. We may therefore assume that ONA scattering is the domain of high-sophisticated amateur radio stations involving high-gain antennas and high transmitter power. We may also speculate, that off-normal scattering is not restricted to the polar ionosphere, i.e. the discovery of off-normal FAI scattering in mid-latitudes (ONF) is very likely.

At first, the author was reluctant to publish these early findings because not enough observations are available for verification purposes and because it is the first ever use of the BeamFinder software for off-normal scatter analyses, i.e. the author first wanted to await further observations from other radio amateurs and use a number of reports to test the calculation tools. However, such observations of unusual Aurora communications are rare in the amateur radio community, so it was decided to write this article, in the hope that it will encourage more radio amateurs to investigate this interesting phenomenon.

*) Please note that the term normal is used in its mathematical meaning (perpendicular), i.e. off-normal denotes the non-perpendicular direction of the difference wave vector in respect to the magnetic field line. However, normal also means typical, usual or expected. Thus, the term off-normal scattering denotes both, i.e the non-perpendicular and the unexpected nature of this propagation mode.



Rückstreuungen ultrakurzer Wellen an Feldlinien-orientierten Irregularitäten

Dubus, 3, p. 182-189, 1987


The BeamFinder analysis software

Grassmann, V., DF5AI, http://www.df5ai.net/BeamFinder/bfintro.html


Dx radius in Aurora and FAI radio propagation

Grassmann, V., DF5AI, http://www.df5ai.net/ArticlesDL/FAIRadius/FAIRadius.html


Dx radius in Aurora and FAI radio propagation

Grassmann, V., DF5AI, Dubus, 3, 23-32, 2002


ONA analyses using BeamFinder

To use the BeamFinder software in off-normal scattering analyses, select the Preferences... item in the Edit menu of BeamFinder. In the dialog box, click the Models section and drag the upper slider to the left position which is labeled Relaxed, see figure 9.

Note to users of BeamFinder version 1.x.x: ONA analyses are restricted to the Scatter curve... item in the RadioProp menu of BeamFinder. In version numbers higher than 1.x.x, the Backscatter locations... and the Dx range... item also supports off-normal scatter analyses.


Figure 9. Adjusting the Aurora/FAI model to off-normal backscattering in the Preferences dialog of BeamFinder.


From: http://www.df5ai.net

Copyright (C) of Volker Grassmann. All rights reserved. The material, or parts thereof, may not be reproduced in any form without prior written permission of the author.