Investigation of the Ionosphere by the Method of Lunar Radar Ranging

Until now, the properties of the ionosphere have been studied chiefly by reflection of radio waves of various frequency (see, for example, 1,7). However, the method of radio-wave reflection has a number of inherent limitations. Thus, for example, it does not make it possible to study regions of the ionosphere situated above the level of maximum ionization. Its natural supplement should be the method of “transillumination” of the ionosphere. In principle, radio waves reaching us from interplanetary space may be used for this purpose. Such, for example, are the radio emission of the Sun and of the Galaxy. However, the radio emission of the Sun depends strongly on solar activity; moreover, when it is considerable (during periods of intensified solar activity), it is especially variable, while at other times it is rather weak. In principle it is possible also to use the radio emission of individual cosmic objects, but great difficulties arise along this route because of the low intensity of such emission. For the purpose of studying the ionosphere, it is more suitable to “transilluminate” it with radio pulses by means of lunar radar ranging. Already in the first lunar radar experiments a number of anomalies were observed in the amplitude and direction of the received signal 2,3,4. One possible cause of these anomalies could be librations of the Moon—oscillations of the Moon with respect to the line connecting it with the Earth. Another cause is the peculiarities of the propagation of radio waves in the upper layers of the ionosphere. In this connection, work was undertaken on the systematic study of the ionosphere by the method of lunar radar ranging 5, 6. A broadcasting station in Shepparton (Australia, state of Victoria) was used as the transmitter, with a power of 70 kw on a wavelength of 13.9 m and 50 kw on a wavelength of 16.8 m.

Fig. 1. Oscillogram of the direct and reflected signal.

The antenna used for transmission was the stationary antenna of the radio station. The angular coordinates of the maximum radiation of the antenna were: azimuth—63°, elevation above the horizon—9°.

The receiver was located 600 km from the transmitter, and in the later experiments also 180 km away. The pass band of the receiver intermediate-frequency circuit was 70 cps, but when operating with short pulses of 1 msec it was widened to 100 cps. From the receiver the signals were fed to a cathode-ray oscilloscope having a screen with afterglow. Figs. 1 and 2 show examples of the sweeps obtained. The experiments were carried out on days when the time of passage of the Moon through the direction of emission of the antenna’s radio waves coincided with a break in the transmissions of the station’s broadcasting program. During the course of a year (November 1947—November 1948) there were only 30 such days. Of this number, transmissions reflected by the Moon im-

pulses were obtained only in 24 cases; on the remaining 6 days, despite normal operation of the transmitter and receiver, no reflected pulses were received. A similar picture—an inexplicable absence in some cases of reflected signals—was also observed in \(^{3}\).

The authors studied: 1) temporal variations in the intensity of the reflected signal, 2) the direction of the radio beam, 3) its intensity.

1. Three types of temporal variations were observed:
   a) Rapid variations of the reflected signal over a time of the order of a second. To investigate these variations, single pulses of duration 2.2 sec were used. Comparison with theory shows that variations of this type can be explained by librations of the Moon. Comparing the speed and amplitude of these variations with the librations of the Moon, the authors also draw a conclusion about the nature of the reflection of radio waves by the lunar surface. It turns out that the surface of the Moon reflects radio waves not according to the law of specular reflection, but diffusely. This thereby confirmed the correctness of the idea of N. D. Papaleksi \(^{8}\), expressed by him in 1946, that for radio waves “the surface of the Moon should undoubtedly be regarded more correctly as rough (in the optical sense).” As was established and calculated by N. D. Papaleksi and L. I. Mandelstam, the form of the reflected pulse depends on the law of reflection of radio waves by the lunar surface (Lambert’s law or the Lommel–Seeliger law). The authors of the work under review give theoretical curves for the shape of the reflected pulse in both cases, without mentioning, however, Papaleksi’s work. In order to establish which of these two laws of diffuse reflection applies, it is necessary to send to the Moon pulses shorter than 0.116 sec (the travel time of radio waves “along the Moon,” and back).

Fig. 2. Three successive echoes obtained from pulses of 2.2 sec.

The authors tried to solve this problem with the aid of short-duration pulses of 1 μsec. However, working with such pulses required widening the passband of the receiver. As a result of considerable ionospheric and “cosmic” noise, the signal-to-noise ratio of the receiver was thereby reduced so much that no reliable data on the form of the reflected pulses could be obtained in this case. It was only established that the duration of the reflected pulse exceeds the duration of the transmitter pulse. The very fact of the spreading of the pulse indicates the diffuse character of the reflection, but does not yet make it possible to decide unambiguously the question of its law.

b) Slower oscillations of the intensity of the reflected signal were also observed, with a period on the order of several minutes. It may be assumed, for the time being only qualitatively, that they are caused by changes in the conditions in the ionosphere (winds in the ionosphere).

c) A third type of variation of the reflected signal is a change in its intensity from one day to another. The variations of this last type are very considerable. No explanation has yet been given for them.

1. Direction of the radio beam. Since the refractive index in the ionosphere is variable (it is determined by the variable concentration of electrons), the path of the radio beam is curved (Fig. 3). Therefore the angle \(\delta\) of the radio beam with the horizon must be greater than the angle \(\alpha\)—the altitude of the Moon. Analysis of the experimental curves confirms this.

Having assumed some law for the change of ionization with height (in the \(F\) layer such a law is taken to be parabolic), one can calculate the minimum height of the Moon above the horizon at which reflected signals first begin to appear. However, theory disagrees with experiment; the angle at which the Moon is first detected by means of radio echoes exceeds the theoretical one—the difference reaches \(10^\circ\). In order to determine whether this and other anomalies are connected with features of the behavior of the \(F\) layer, the stations neighboring the one that was sending pulses to the Moon simultaneously determined, by the pulse-reflection method, the height and critical frequency of the \(F\) layer. It turned out that the magnitude of the discrepancy between the theoretical angle of Moonrise for radio waves and the experimental data depends substantially on the ratio \(f/f_{0F_2}\) of the wave frequency to the critical frequency of the \(F\) layer at the given time.

Fig. 3. Trajectory of the radio beam in the ionosphere.

1. Intensity of the radio echo. In addition to the absorption of radio waves in the ionosphere, there is another very important cause of attenuation of the radio beam. When a plane-wave front penetrates into the ionosphere, it is curved,\(^9\) the energy flux density decreases, and the signal intensity falls. Taking a parabolic law for the variation of \(N\) with height and introducing the corresponding correction for the curvature of the Earth, the authors calculate the expected attenuation of the signal.

We emphasize that this calculation, like the calculation of the angle \(\delta\), is carried out in the approximation of geometrical optics.

However, experiment does not confirm the theory. The signal amplitudes measured experimentally are many times smaller than the theoretical ones. Moreover, the magnitude of the discrepancy with theory is highly variable and changes strongly both within the limits of a single experiment and from one experiment to another. The signal amplitude was found, furthermore, to depend on the ratio \(f/f_{0F_2}\). This circumstance indicates that the attenuation of the signal most likely occurs in the \(F\) layer. The observed discrepancies between theory and experiment may have two explanations: first, that the \(F\) layer is inhomogeneous not only in the vertical but also in the horizontal direction; second—that at large angles of incidence

tions with which the authors worked, the approximation of geometrical optics, on the basis of which all the calculations were carried out, is invalid.

The authors draw two main conclusions: 1) their experiments disagree with the results of the usual theory of radio-wave propagation in the ionosphere, which applies the approximation of geometrical optics and regards the ionosphere as a system of ionized layers homogeneous in the horizontal direction; 2) for a wave of 16.8 m, and still more so for shorter waves, the surface of the Moon, in calculating the reflection, must be considered “rough” in the optical sense.

M. Ginzburg

CITED LITERATURE

1. V. L. Ginzburg, Theory of Radio-Wave Propagation in the Ionosphere, Gostekhizdat, 1949.
2. J. H. De Witt, Jr. and E. K. Stodola, PIRE 37, 229 (1949).
3. Z. Bay, Hungarica Acta Physica 1, 1 (1946).
4. V. S. Vavilov, UFN 39, 353 (1949).
5. F. J. Kerr, C. A. Shain and C. S. Higgins, Nature 163, 310 (1949).
6. F. J. Kerr and C. A. Shain, PIRE 39, 230 (1951).
7. Ya. L. Alpert, Propagation of Radio Waves in the Ionosphere, Gostekhizdat, 1947.
8. N. D. Papaleksi, UFN 29, 250 (1946); Elektrichestvo 51 (1946).
9. Bremmer, Terrestrial Radio Waves, Amsterdam, 1949, p. 271.