JOURNAL OF GEOMAGNETISM AND GEOELECTRICITY VOL. 21, N0. 1, Schumann Resonances and Worldwide Thunderstorm Activity

Transcription

1 JOURNAL OF GEOMAGNETISM AND GEOELECTRICITY VOL. 21, N0. 1, 1969 Schumann Resonances and Worldwide Thunderstorm Activity Diurnal Variations of the Resonant Power of Natural Noises in the Earth-Ionosphere Cavity Toshio OGAWA, Yoshikazu TANAKA, and Michihiro YASUHARA Geophysical Institute, Kyoto University, Kyoto (Received October 16; revised November 21, 1968) Abstract Electric vertical and horizontal NS and EW components of natural ELF electromagnetic noises were observed through a year from February, 1967, to January, 1968, Mean diurnal variations of the Schumann resonance power were obtained for the first three modes from noise spectra. The variations show completely different patterns for each component, and change considerably month by month. The diurnal amplitude also varies by 1 to 7 db from the mean daily values. The results are compared with the diurnal variations derived from the simple theory considering worldwide thunderstorm activity as noise sources. There are fairly good agreements between the experimental and theoretical results, especially for EW component. For vertical and NS components thunderstorm effect from Africa seems relatively smaller then expected. The time of this effect is in local night. This shows that local night conditions of the lower ionosphere as the upper boundary of the earth-ionosphere cavity, make the received intensity of vertical and NS components decrease. 1. Introduction Diurnal variations of natural noise power of the first few modes of the Schumann resonances were investigated by Balser and Wagner (1962) for the electric vertical component. Gendrin and Stefant (1964) also investigated the diurnal variations for the magnetic component. They dealt with only one directional component of either electric or magnetic. Etcheto, Gendrin, and Karczewski (1966) first used data of both magnetic NS and EW components at Chambon-la-Foret (location: 4801'N, 226'E) and Port-aux-Francais (location: 'N, 70 10'E), and obtained different diurnal variations for respective components and for respective stations. They explained this result qualitatively by geometrical factors relating the positions of the signal sources of thunderstorm origin, and of the observational points. In the present investigation we observed electric three components through a year, and obtained diurnal variations of the Schumann resonance power. We discuss differences between respective components in respect to the worldwide thunderstorm activity as noise sources.

2 During the period from February, 2. Observations and analyses 1967, to January, 1968, the electric vertical component of natural ELF electromagnetic noises was measured in Kyoto (location: 3501'N, 13547'E) by using a ball antenna (Ogawa, Tanaka, Miura, and Yasuhara, 1966). Simultaneous measurements of the electric horizontal NS and EW components were done at Aso (location: 3253'N, 131OVE) by using two orthogonal ground antennas of 300 meters in length. Recordings were made by FM magnetic tape recorders with the tape speed of 7.6 cm/s. Record tape was played back at a speed 20 times higher than the original recording speed and signals were re-recorded by the audio-frequency tape recorder. When the tape thus obtained is played back again at the twice higher speed, the frequency ranges of theoriginal 17h (UT) MAY 16, 1967 Fig. 1 An example of the sonagram showing the Schumann resonance frequency time display. Signals were observed by the use of a ball antenna in Kyoto during a period from 1659 to 1705UT, May 16, Hz noise is from the nearby commertial power line. The wavy pattern of the 60 Hz noise was produced during the data processing. Fig. 2 An example of the sonagram showing the Schumann resonance power spectra which were obtained by integrating four and a half minutes signals shown in Fig. 1 as an example. The amplitude reads relative power as 1db/mm on the original sonagram paper. 60 Hz noise is used as a frequency reference.

3 signals are converted from 4-40Hz to 160-1, 600Hz for the vertical component, and from 3-30Hz to 120-1, 200Hz for the horizontal components. A sound spectrograph can then be used to get power spectrum from these noises. Analyses were made for the first four and a half minutes of each hour during a week each month. An example of the sonagram showing the frequency time display is given in Fig. 1. The signals shown in Fig. l were observed with the ball antenna in Kyoto during the time from 1659 to 1705UT, May 16, The 60Hz noise came from the nearby commertial power line. Wavy pattern of the 60Hz noise was produced during the data processing. The signals were integrated to make power spectrum which is shown in Fig. 2. There are eight such spectra in the sonagram of Fig. 2 in which first six modes of the Schumann resonances can be seen. The amplitude reads relative power as 1db/mm on the original sonagram paper. The peak at 60Hz is used as a frequency reference. The peak amplitude of the spectrum of each hour is read and normalized to the daily mean value of that day for each mode, and for each component. during a week are then obtained for each mode, each component, are given in Fig. 3 for the first three modes. were so disturbed by some artificial or electrostatic Mean diurnal variations and each month, which The April and May data of NS component noises that no curve could be deduced for the first mode. It is most interesting to see in Fig. 3 that each component shows a Fig. 3 Diurnal variations of the Schumann resonance peak power. Solid, broken, and dotted curves represent the first, second, and third modes, respectively.

4 T.OGAWA, Y. TANAKA and M. YASUHARA completely different variation pattern from the other components, and changes its diurnal amplitude month by month, by 1 to 7db from the mean daily values. Fig. 4 Diurnal variation of the relative thunderstorm activity derived from the world thunderstorm day map of each month in Handbook of Geophysics (1960). Fig. 5 Calculated diurnal variations of the Schumann resonance power. Solid and broken curves represent the first and second modes, respectively.

5 3. Theoretical results Let us make simple calculations of the expected power variations considering worldwide thunderstorm activity as signal sources. Letting 6 be the angular separation between the signal source and the observation site, peak intensity of each mode of the Schumann resonances may be proportional to APn(cos8) for the electric vertical component, where Pn is the n th Legendre polynomial, while the electric horizontal component may be proportional to Be/e8Pn(cos8), where A and B are factors including noise intensity at the source (Schumann, 1952). Fig. 4 shows mean diurnal variations of relative worldwide thunderstorm activity of each month derived from the data in Handbook of Geophysics (1960); using the world thunderstorm day maps the thunderstorm area by every 15 in longitude was calculated, being multiplied by the numbers of thunderstorm days, and the values thus obtained were distributed into five hours centered at 16 h local time which may be the most active time of thunderstorms in a day. The activity varies considerably month by month. There are three main thunderstorm activity centers in the continents of Asia, Africa, and America. The angular separations of these centers from the observation sites were estimated from the world thunderstrom day maps of each month. Also was estimated the angular separation of the activity center at 0 h UT, which is at 120W in longitude. Using relative thunderstorm activity shown by Fig. 4 as noise intensity at the source and the angular separations of those activity centers mean diurnal variations of the Schumann resonance power can then be calculated for each mode of each component. The results of the first two modes are given in Fig Discussions It is interesting to compare the experimental results in Fig. 3 with the theoretical results in Fig. 5. There seems fairly good agreement between both. The beautiful agreement between the experiment and the calculation for the EW component, is rather surprising although there is much difference for June. It should be noted that the experimental curves were obtained from only a week data each month, while the theoretical curves were obtained from thunderstorm distributions by month of different year. For the vertical and NS components, on the other hand, thunderstorm effect from Africa is relatively smaller than expected. The time of this effect is in local night in Japan. This is also shown in the continuous recording of integrated field intensity of the vertical component at the bandwidth of 4 to 40Hz. It shows diurnal variation in which intensity is high in the daytime and low in the nighttime. The field intensity begins to increase in the very early morning before ground sunrise, and to decrease at about ground sunset. This shows that local night conditions of the lower ionosphere as the upper boundary of the earth-ionosphere cavity, make the received intensity decrease. In the present investigation the electromagnetic radiation from lightning discharge was considered as only signal source for the Schumann resonances in the cavity between the conducting ionosphere and the earth. Some of the differences of the observed power varia-

6 Acknowledgment is made for the partial financial support given to this investigation by a grant from the Japan Society for the Promotion of Science. References Balser, M., and C.A. Wagner, Diurnal power variations of the earth-ionosphere cavity modes and their relationship to worldwide thunderstorm activity, J. Geophys. Research, 67, , Etcheto, J., R. Gendrin, and J.-F. Karczewski, Enregistrement simultane des resonances de la cavite terre-ionosphere en deux stations distantes de 12,000 km, Ann. Geophys., 22, , Gendrin, R., and R. Stefant, Magnetic records between c/s, in Propagation of Radio Waves at Frequencies below 300kc/s, edited by W.T. Blackband, pp , Pergamon Press, Ogawa, T., Y. Tanaka, T. Miura, and M. Yasuhara, Observations of natural ELF and VLF electromagnetic noises by using ball antennas, J. Geomag. Geoelectr., 18, , Schumann, W.O., Uber die strahlungslosen Eigenschwingungen einer leitenden Kugel, die von einer Luftschicht and einer Ionospharenhi lle umgeben ist, z. Naturforschg., 7a, , U.S. Air Force Geophysics Directorate, Handbook of Geophysics, Macmillan Company, New York, 1960.