Radio Observation of Solar Eclipse 2015

On the 20th of March 2015 a solar eclipse could be observed from Europe. This was a welcome opportunity to observe the propagation of vlf and lf radio waves during this event. A short and comprehensive introduction to this theme is given by P. Lassudrie-Duchesne and R. Fleury.

Radio observation was done at Algermissen, northern Germany (52.2527 N / 9.9786 E). Receiving system was a magnetic loop antenna (Wellbrook ALA 1530LF) and a software defined radio (RF Space SDRIQ) connected to a notebook running Spectrum Lab. Within Spectrum Lab a watch list was setup to monitor the following frequencies:

Transmitter Country Latitude Longitude Path length
Eclipse (UT)*
20.3 Is. Tavolara Italy 40.9229 N 9.732028 E 1259 09:26
20.9 Sainte-Assise France 48.5446 N 2.579429 E 684 09:29
22.1 Skelton Great Britain 54.7318 N 2.883033 W 884 09:33
24 Cutler USA 44.6449 N 67.28164 W 5480 -
57.4 Grindavik Iceland 63.8502 N 22.46552 W 2260 09:36
60 Anthorn Great Britain 54.9117 N 3.27848 W 919 09:33
62.6 Kerlouan France 48.6379 N 4.35062 W 1088 09:22
81 Inskip Great Britain 53.8301 N 2.834266 W 873 09:32

*at the transmitter site (calculated with NASA online solar eclipse explorer)

Figure 1 shows the path of total eclipse as well as of penumbra together with the transmitter sites and the receiving location. Maximum eclipse at Algermissen occurred at 09:43 h UT with a magnitude of 0.81. The penumbral shadow contacts Earth's surface first (P1) at 07:41 h and last (P4) at 11:50 h UT. As can be seen only the signal from Grindavik crossed the path of totality. All other signal paths crossed the penumbra at varying magnitudes.

Fig. 1: Path of solar eclipse March 20, 2015. The transmitter sites as well as the receiving site at Algermissen, northern Germany are plotted into the map. The points P1 and P4 mark the coordinates where the penumbral shadow first contacts and last contacts Earth's surface. Eclipse map courtesy of Fred Espenak - NASA / Goddard Space Flight Center. For more information on solar and lunar eclipses, see Fred Espenak's Eclipse Web Site:

Map of the path of the solar eclipse 20th March 2015 and observed radio

Continuous measurement of the amplitude of the transmitter signals was performed between 07:00 to 12:30 UT at the day of solar eclipse and within the same time span on the following days from 23.03. to 27.03.2015. The mean of the latter 5 days serves as reference against the solar eclipse day (=quiet day curve). Space weather conditions at these days were (taken from

Date Measured
solar flux
2.8 GHz
A index
Planetary K index
(3-hour intervals)
Number of flares
2015-03-27 137.8 109 7 32222111 C: 5
2015-03-26 136.1 103 7 21202223 C: 4
2015-03-25 137.8 115 11 22233232 C: 10
2015-03-24 133.0 127 8 11123322 C: 2
2015-03-23 128.1 119 16 44234323 C: 3
2015-03-20 112.7 27 21 43433234 C: 1 (01:40 UT)

During the eclipse no strong solar flares occurred whereas the geomagnetic conditions were rather unsettled. How much this influences the rf-signals in general is beyond my experience. Geomagnetic conditions during the 5 control days are quiet. These days show a number of small C-class flares which have no impact on the 5-day mean. A sliding 5-min average is applied to all measurements. Time resolution is 1 second. The rf-amplitudes are at minimum 10 dB above noise floor when signal is running through depressions.

Figure 2 shows the difference between the rf-signals at solar eclipse day and the quiet day curve. For each of the 8 signal paths distinct variations in the received rf-amplitude can clearly be identified around the time of the maximum eclipse. Whereas the amplitude of the transmitter-signals of eclipse day and quiet day curve converge at midday (with exception of Sainte-Assise and Cutler) the situation is different in the morning. Prior to the eclipse time an offset of up to 7 dB from the quiet day curve can be seen (with exception of Skelton and Anthorn). I wonder whether this could be an effect of the unsettled geomagnetic conditions in the morning of the eclipse day.

Fig. 2: Plots (= eclipse signatures) of the difference between the rf-signals of eclipse day and quiet day curve against time for each of the 8 signal paths and frequencies.

Plots of the difference between signals at eclipse day and
			prerecorded days from 8 radio stations

Only the transmitter signals from Grindavik run through the zone of totality. Nevertheless, all signal sequences show deviations from the quiet day curves; so even the penumbra has a remarkable effect on the propagation of vlf and lf. Each eclipse signature is unique and a result of the interference pattern due to multiple path propagation. It depends mainly on frequency, path length and conditions of ionospheric D-layer (its three dimensional electron density distribution) along this path. The forming of the D-layer is determined fundamentally by solar irradiation.

Fig. 3: Plot of local maxima and minima between 07:41 h (P1) and 11:50 h (P4) derived from Figure 2.

Plot of the simplified signal-courses from 8 radio stations

Modelling of eclipse signatures has been done elsewhere. So I will concentrate on parameterization of it. For characterization of the various shapes I selected the local maxima and minima (turn points) between the times of P1 and P4. Plotting these against time together with the signal amplitudes at the times of P1 and P4 results in figure 3. Some common types of reactions of the rf-signals to the solar eclipse become apparent. M-, W-, V- and A-shaped signal sequences can be described.

However, does a relationship exist between the distribution of turn points and solar eclipse maximum? To answer this question the distribution of the turn points along the time axis must be examined.

Fig. 4: Histogram of the frequency distribution of local maxima and minima during solar eclipse (data taken from fig. 3). Class interval is 28 min.

Plot of the frequency distribution of local maxima and minima from the signal curves of the 8 radio stations

Figure 4 shows a clear concentration of turn points between 09:24 h and 09:52 h UT. This corresponds well with the path of solar eclipse (see figure 1): within this time span a maximum shading of the signal paths between transmitters and receiving site takes place. Also from figure 4 an imbalance of turn points before and after the maximum can be detected. Again the question arises whether this is an effect of the unsettled geomagnetic conditions in the morning. Also the asymmetric progress of the eclipse in relation to the signal paths may be an explanation.