METEOR OBSERVATIONS USING RADARS*

At the beginning of the 1930s, in ionospheric studies and in studies of the propagation of radio waves, short-lived radio echoes were occasionally observed. They arose at an altitude of about 100 km, i.e., in the region of the \(E\) layer. These echoes were observed both by day and by night, and therefore could not be associated with the ultraviolet radiation of the Sun. Skellett and other investigators ascribed them to meteor ionization, but for a long time this supposition remained unproved.

In November 1940, Pierce\(^1\), using an ionospheric station, established that the reflections arose after the passage of bright meteors that were observed visually. Subsequently, the study of short-lived echoes began to be carried out with the aid of radar equipment. They are visible on the screen of a cathode-ray range oscilloscope in the form of “bursts” lasting from fractions of a second to several tens of seconds. Their amplitude is sometimes 10–20 times greater than the amplitude of “noise.”

In England, the study of short-lived echoes was carried out by Appleton, Ekersley, and other investigators. Since January 1944, at the Research-

... station in Slough began continuous observations of these echoes with the aim of studying their diurnal and annual variation.

Since October 1944, Hey and Stewart began systematic investigations of short-duration echoes by means of radars that had been used during the war for tracking V-2 rockets. These radars have a wavelength of 4–5 m and a peak power of 150 kW. With their aid, on average 10 echoes per hour are observed. The preliminary results of these investigations, which confirmed the meteoric nature of the echoes, were published in October 1946.^2

Fig. 1.

Night observations with a radio beam directed vertically upward showed that all meteors observed visually near the zenith were accompanied by radio reflection. However, several times more echoes were observed that did not coincide with visible meteors and, evidently, were caused by telescopic meteors. In systematic daily observations over long periods, the hourly numbers of echoes gave noticeable maxima on days of active meteor showers, which was confirmation of their meteoric nature.

Another, far more interesting test of the meteor hypothesis was provided by simultaneous observations of the δ Aquarid meteor shower at stations with inclined radio beams directed along different azimuths. For each station its own curve of hourly numbers was obtained (Fig. 1). At station I the curve has no clearly expressed maximum, while for stations II and III the moments of the maximum are of various magnitudes. This result is explained as follows: the ion clouds produced by meteors have the form of long columns. It is natural to suppose that the best conditions for obtaining an echo occur when the column is irradiated perpendicularly to its length. Assuming that this condition was fulfilled at the time when the maximum was observed at a given station, one could delineate on the celestial sphere the zone in which the meteor radiant could be located (Fig. 2). The zones found for two stations intersected, and it was to be expected that the radiant lay near the middle of the overlapping part \((R)\). The equatorial coordinates of point \(R\) are: \(\alpha = 345^\circ,\ \delta = 10^\circ\), whereas the meteor radiant has coordinates \(\alpha = 340^\circ,\ \delta = -17^\circ\). Directing the radio beam of station I during those days was unfavorable for observing meteors of the Aquarid shower, and therefore the maximum was absent. The few echoes observed for any direction of the beam are due to the meteor background.

From Current Literature

During the Draconid meteor shower on the night of October 9–10, 1946, observations in England were carried out on several radar installations. Appleton and Naismith1 observed at a frequency of 27 Mc/s. A special antenna radiated mainly vertically upward. The pulses had a duration of 15 microseconds and a frequency of 50 per sec. A strip of photographic paper moved uniformly in front of the screen of a cathode oscillograph, perpendicular to the range sweep, and on it the range and duration of the echo were recorded (Fig. 3). The observations made it possible to construct a curve of the number of meteors during the Draconid shower.

Fig. 2.

At the meeting of the Royal Astronomical Society on December 13, 1946, Dr. A. Lovell and Hey reported on the results of their observations2. A. Lovell conducted observations at a frequency of 72 Mc/s. The transmitter emitted 150 pulses per second, each of 8 microseconds’ duration, with a peak power of 150 kW. A special receiver gave an easily distinguishable signal at an incoming power of \(10^{-14}\) W. The radio beam could be turned in any direction. The echoes were observed visually on the screen of a cathode oscillograph. During the maximum of the meteor shower the screen was filled with sparks.

With the antenna directed at an angle of \(90^\circ\) to the radiant, continuous observation was carried out for 71 hours from 8 to 11/X. The entire period of activity of the stream was contained between 0 h 00 m and 6 h 00 m on 10/X. At the beginning of this interval the number of short-duration echoes was 0.03 per minute (the normal frequency for the given installation in the period when meteor streams are absent). The number of echoes rapidly increased and at 4 h 40 m reached 168 per minute (an increase by a factor of 5000). The peak of activity was extremely brief—after five minutes the number of echoes had fallen to 50 per minute.

The theory developed by Herlofson indicates that for a transmitting beam directed at right angles to the meteor trail, the strength of the received echo is given by the formula

\[ \varepsilon = 8{,}88 \cdot 10^{-28}\frac{a^2\lambda^2}{R^3}P_0G \text{ watts}, \]

where \(a\) is the number of electrons produced by the meteor over 1 cm of path, \(R\) is the echo range in cm, \(P_0\) is the peak power of the transmitter in watts, \(\lambda\) is the wavelength in cm, \(G\) is the directivity coefficient of the receiving and transmitting antennas. For the maximum, the period of the maximum was found; \(R\) was also measured, and from these \(a\) was calculated. At a given speed \(a\) is proportional to the size of the meteor.

At the beginning of the shower’s activity, 21 coincidences with visually observed meteors were noted. Of the echoes lasting more than 0.5 sec, 50% coincided with visible meteors. (The same result had been obtained earlier for the Perseids.)

Hey’s report was devoted to the determination of meteor velocities. Improvement of the apparatus, in particular improvement of photographic recording, made it possible to study the fine structure of the echo. The resolution was increased by sweeping the range band from 80 to 115 km over the full width of the cathode-ray tube.

Fig. 3.

On the tape, at first a weak trace is obtained, associated with the approaching meteor and caused by ionization in the immediate vicinity of the meteor itself. Then comes the main echo from the ionized column at the minimum distance. In the initial weak trace, a change of range with time is obtained corresponding to a body moving rectilinearly with constant velocity. The latter can be found by measuring 3 points on each trace. This was done for 22 meteors, and the mean velocity was found to be 22.9 km/sec with a dispersion of 1.3 km/sec. The extra-atmospheric velocity, calculated from astronomical data, is 23.7 km/sec. Some traces show deceleration toward the end of the path.

E. Appleton, who was present at the meeting, expressed his delight at Hey’s work and called it “epoch-making.” He went on to say:

“When the war in Europe ended, there were in our country many radar installations and staffs of operators for whom no work remained. I was asked whether I could suggest something to keep them occupied, and I proposed observing sporadic meteors. As a result we have an enormous amount of material from all over the country, with round-the-clock observations throughout the whole year. This is unique material and, moreover, such as will hardly ever be repeated, considering the number of human efforts expended.

We do not yet have the results of this work, but Naismith and I, in our investigations of the ionosphere, came to the conclusion that nighttime ionization is maintained by sporadic meteors.”

Appleton noted that radio observations showed a noticeable maximum of meteors at noon and a seasonal maximum in summer. Possibly this is connected with the fact that, in the presence of ionization produced by solar radiation, weaker meteors begin to give radio echoes.

[Figure: photographic panels numbered 1–20; wavelength scale marked 216, 219, 250 nm.]

Fig. 4.

Lovell and Hey noted that, although the observations were not analyzed from the point of view of atmospheric winds, drift velocities of up to 200 km/hour, noted by visual observers of meteors, undoubtedly exist. One echo with a duration of 95 sec. gave a drift velocity of 600 km/hour.

In the United States, observations of meteors by means of radars were carried out on a considerably smaller scale than in England. With the participation of the Princeton University Observatory, during the Draconid meteor shower observations were made on 21 military radar installations[^5]. Wavelengths from 3 m to 3 cm were used, but positive results were obtained only on 3-meter waves (SCR-270-type radar). The frequencies of microwave radars exceed the critical frequencies of the ionic clouds produced by meteors.

The observations were made on a range oscilloscope. In the course of 23 minutes, filming of the image on the screen was carried out at a frequency of 1 frame per second. Several dozen reflections lasted more than 1 sec. (up to 27 sec.) and therefore appeared on a number of frames (Fig. 4). According to a preliminary analysis, the effective area of trails of sufficient duration ranges from 50 to 4000 m².

The results presented above show how great are the observational possibilities that the use of radar equipment opens up to meteor researchers.

B. Yu. Levin

References

1. J. A. Pierce, Phys. Rev. 59, 625—626 (1941).
2. Y. S. Hey and G. S. Stewart, Nature 158, 481—482 (1946).
3. E. Appleton and R. Naismith, Nature 158, 936—937 (1946).
4. The Observatory 67, 3—8 (1947).
5. J. Q. Stewart, M. Ference, J. J. Slattery, H. A. Zahl, Sky and Telescope, VI, 3—5 (1947).