The Nature of Meteor Radio Echoes in Observations of the Resonance Effect at a Wavelength \(\lambda = 10\) m

É. K. Nemirova

(Presented by the scientific seminar of the Radio Engineering Faculty)

Introduction

In radar observations of meteor trails, under certain conditions the character of the reflected signal proves to depend on the polarization of the incident wave relative to the axis of the meteor trail. The sensitivity to polarization is explained by the occurrence of plasma resonance, which is caused by the component of the field \(E\) perpendicular to the trail axis. Plasma resonance is observed mainly in trails of the underdense type (linear electron density in the trail \(\alpha < 2.4 \cdot 10^{12}\ \frac{e}{\text{cm}}\)).

Consideration of the character of the reflected signals under resonance conditions shows that, for “perpendicular” polarization, the signal amplitude, phase, and the form of the diffraction pattern (the character of the change of the signal with time) change significantly in comparison with the case of “parallel” polarization. The degree of this change depends on the parameters of the meteor and of the medium (electron density \(\alpha\), velocity \(v\), diffusion coefficient \(D\)), and also on the wavelength of the radar \([1, 2, 3]\). As an example, diffraction patterns are given in Fig. 1, obtained by calculation \([3]\). These diffraction patterns are characterized by an increase in amplitude and a delay of the signal in

Fig. 1. Examples of envelopes of the amplitudes of reflected signals: 1—ideal diffraction pattern (in the absence of attenuation and resonance). 2—diffraction pattern in the absence of resonance, with attenuation taken into account; 3, 4, 5, 6—diffraction patterns under resonance for different lengths of the resonating section of the trail \(l_{\text{res}}\) \(\bigl(l_{\text{res}}(3) < l_{\text{res}}(4) < l_{\text{res}}(5) < l_{\text{res}}(6)\bigr)\).
\(U_{0}\)—a quantity proportional to time.

time (shift to the right), smoothing of the pulsations and a decrease in their periods. For meteors with low electron density, the depth of the pulsations may increase due to resonance.

Experimental investigation of the polarization effect, the results of which were published in the works of Clegg, Closs, Kaiser [4], and Billem and Brown [5], was carried out at wavelengths \(\lambda = 4\) and \(5\) m. Calculations show that, for these wavelengths, changes in amplitude and phase are mainly characteristic; the other effects are weakly expressed [3]. Therefore, the results obtained in these observations (the distribution of the ratio of amplitudes for two mutually perpendicular types of polarization) characterize the resonance effect incompletely. Additional information about the resonance effect can be obtained by conducting observations at a longer wavelength. Of particular interest here is the form of the diffraction pattern under resonance, since distortion of the diffraction pattern can cause errors in certain types of meteor observations [1, 2, 6].

2. Observational method and apparatus

This article presents some results of observations of the resonance effect at the radar station of the Tomsk Polytechnic Institute, intended for meteor studies.

The main parameters of the station were: wavelength \(\lambda = 10\) m, pulse power \(P_u = 60\) kW, repetition frequency \(F_u = 600 \dfrac{\text{pulses}}{\text{sec}}\), pulse duration \(\tau_u = 5\) μsec.\(^{1}\)

The observations were carried out on June 8 and 9, 1961, during the activity of the daytime meteor shower ζ–Perseids, with radiant coordinates: declination \(\delta = 24.4^\circ\), right ascension \(\alpha = 61.5^\circ\).

During the observations the station transmitter operated simultaneously into two directional antennas of identical dimensions, one of which had horizontal polarization and the other vertical polarization.

Fig. 2. Diffraction pattern in the absence of a polarization effect (photograph from the speed-indicator screen).

Both antennas, located in close proximity to one another, were oriented in the direction with azimuth \(160^\circ\) and angle

\(^{1}\) The station parameters are given for this experiment.

location 35 (astronomical azimuth). With this arrangement of the antennas, at the time corresponding to the position of the stream radiant (at 11–12 hours decree time), the antenna with horizontal polarization radiated a wave polarized “perpendicularly” with respect to the meteor trails of the stream, while the antenna with vertical polarization radiated a “parallel”-polarized wave.

The reflected signals received by both antennas were fed to the receiver input in alternation (the switching frequency was equal to half the pulse-repetition frequency). The sequence of received pulses was fed to the meteor-velocity indicator, whose sweep, with a duration of \(\sim 0.1\) sec, was triggered by the arrival of the reflected signal. The screen of the indicator was photographed on film together with the range indicator.

If the signals received by the two antennas are exactly the same, a diffraction pattern is observed, an example of which is shown in Fig. 2. When a polarization effect is present, two different diffraction patterns are obtained on one sweep (Fig. 3)¹.

Fig. 3. Example of an echo showing the polarization effect.

3. Some results of observations

As a result of the observations carried out, the presence of a polarization effect at \(\lambda = 10\) m and distortion of the diffraction pattern due to resonance were confirmed.

The following types of reflections were characteristic:

1. Reflections of an unstable type (showing a noticeable decrease in amplitude over the course of a sweep of duration 0.1 sec), exhibiting a polarization effect.

2. Reflections of an unstable type, not exhibiting a polarization effect.

3. Reflections of a stable type with a distinctly expressed polarization effect.

4. Reflections of a stable type, not exhibiting a polarization effect.

5. Anomalous reflections.

For reflections of the 1st type, the polarization effect is manifested in an increase of amplitude, smoothing of pulsations, and delay of the di-

...of the diffraction pattern, and sometimes a shortening of the pulsation periods. Examples of type-1 reflections are presented in Fig. 4.

In reflections of this type, two different forms of the time dependence of the polarization ratio (the ratio of amplitudes for the two kinds of polarization) are observed.

(a)  (c)

(b)  (d)

(v)  (e)

Fig. 4. Characteristic forms of type-1 reflections.

For some reflections the polarization ratio first increases and then decreases, tending toward 1 (Fig. 4 a, b, v). Such reflections were observed in experiments at \(\lambda = 4\) and \(5\ \text{m}\). As calculation [3] shows, they are characteristic of trails with short resonating sections \(l_{\text{res}}\) (the length of the resonating section is proportional to

\[ \frac{\lambda^{2} v}{D} \]
).

The typical change of \(\rho\) with time for these reflections is shown in Fig. 5.

However, reflections corresponding to a comparatively large \(l_{\text{res}}\) occur more often. This is explained by the fact that the sensitivity of the apparatus is comparatively low. When meteors are detected in the direction of the maximum lobe of the antenna radiation pattern, the minimum detectable electron density at \(R = 150\ \text{km}\) is

\[ \sigma_{\min} \simeq 3 \cdot 10^{11}\ \frac{e}{\text{cm}} . \]

According to calculation, such reflections should be characterized by the changes shown in Fig. 1 (curves 5 and 6). Typical examples of reflections of this type are shown in Fig. 4 (c, d, e).

The exact value of the polarization ratio for these reflections was not found, since over the course of \(0.1\ \text{sec}\) it did not decrease...

reached 1, which did not make it possible to introduce a correction factor taking into account the difference in the intensity of the signals received by the two antennas for reasons not dependent on the polarization effect.

However, it may be noted that, as follows from the calculation, the polarization ratio increases over a prolonged time; the decay of the resonant echo occurs considerably more slowly than that of the nonresonant echo.

Fig. 5. Change with time of the polarization ratio for unstable echoes (example).

In most cases a shift of the reflection maximum to the right is observed. The magnitudes of the measured shifts \(\Delta t\) lie within the range from 0 to \(0.01\) sec.

For reflection 4 (e) \(\Delta t/T_1 = 0.35\) (where \(T_1\) is the first pulsation period); for reflection 4 (g) \(\Delta t/T_1 = 0.3\). According to the calculation, \(\Delta t/T_1\) lies within the range from 0 to 0.8.

Reflection 4 (g) shows a decrease in the pulsation periods: the first by 14%, the second by 20%, which also falls within the limits given by the calculation and confirms the calculated data that the second period should change more than the first.

Fig. 6. Resonant echo of the stable type.

Reflections of the second type (unstable echoes that do not give a polarization effect) have a polarization ratio of \(0 \div 1\) and a similar form of diffraction patterns. Reflections of this kind may be obtained for trails located at approximately identical angles to the polarization planes of the antennas. There are comparatively few such reflections.

Most meteors of the stable type do not exhibit a polarization effect, which agrees with the Kaiser–Closs theory [7].

At the same time, reflections of a stable type are often encountered, revealing a polarization effect at the beginning of the reflection. Characteristic reflections of this type are shown in Fig. 6. The polarization ratio decreases with time to 1. The value of $\rho_{\max}$ for such echoes averages 1.5. It was not possible to trace a regular change in the diffraction pattern due to resonance because of distortions caused by other factors.

In addition to the types considered, anomalous echoes with a chaotic variation of $\rho$ in time were observed.

Conclusions

The observations of the polarization effect that have been carried out make it possible to arrive at the following conclusions:

1. At a wavelength $\lambda = 10\ m$, plasma resonance may exist.
2. Because of the resonance, distortion of the diffraction pattern occurs. The principal regularities of this distortion, obtained by calculation, are confirmed experimentally.
3. In a number of cases a polarization effect is observed in stable meteor trails, which is at variance with the theory of Kaiser and Closs, but is partly confirmed by Keitel’s data [8].

It should be noted that the data obtained require additional processing in order to obtain the corresponding statistical regularities.

In conclusion I express my gratitude to senior engineer L. D. Butakov for his assistance in carrying out the observations.

References

1. E. K. Nemirova, Polarization effect in the reflection of radio waves from meteor trails, Proceedings of the Siberian Physico-Technical Institute, issue 37, 1959.
2. E. K. Nemirova, On the role of resonance effects in measuring meteor velocities, Astronomical Journal, XXXVI, 3, 1959.
3. E. K. Nemirova, Resonance scattering of radio waves by paraboloidal meteor trails, Izvestiya TPI, vol. 100.
4. R. L. Closs, J. A. Clegg, T. R. Kaiser, An experimental study of radio reflections from meteor trails, Phil. Mag., 1953, v. 44, N 350.
5. E. R. Billam, J. C. Browne, Observation of polarisation effects in radio echoes from meteor trails, “Meteors”, ed. by T. R. Kaiser, 1955, p. 73.
6. E. K. Nemirova, Errors in measuring meteor velocities by the spaced-reception method, Izvestiya TPI, vol. 105, 1960.
7. T. R. Kaiser, R. L. Closs, Theory of radio reflections from meteor trails, Phil. Mag., ser. 7, 1952, v. 43, N 336.
8. G. H. Keitel, Certain mode solution of forward scattering by meteor trails, J. Geophys. Res., 1951, 56, p. 37.

1. The pulsations of the diffraction pattern are not always clearly expressed because of the influence of noise and other distorting factors. ↩