172495a0Nature1724376195309124954960028-0836195310.1038/172495a0ukNatureNatureNATUREnatureNature is a weekly international journal publishing the finest peer-reviewed research in all fields of science and technology on the basis of its originality, importance, interdisciplinary interest, timeliness, accessibility, elegance and surprising conclusions. Nature also provides rapid, authoritative, insightful and arresting news and interpretation of topical and coming trends affecting science, scientists and the wider public./nature/journal/v172/n4376issueJournal homeArchiveCurrent issueAdvance online publicationPrivacy policySubscribeNature Publishing GroupCurrent issue172495a0Audio-Frequency Spectrum of Atmospherics
AU  - CHAPMAN, F. W.
AU  - MATTHEWS, W. D.Wheatstone Physics Laboratory, King's College, London, W.C.2. June 12.THE observations on the low-frequency or /`slow tail/' component in the waveform of an atmospheric described recently by Hepburn and Pierce in Nature1 is of particular interest to us since we have been making similar observations for some years. The earlier observations made by Appleton and Chapman2 dealt with the evolution of atmospheric waveforms. It was shown that many atmospherics consist of an oscillatory portion, composed of a sequence of quasi half-cycles of increasing duration, followed by a /`slow tail/' of usually two rounded half-cycles. The oscillatory portion was found to move ahead of the slower disturbance as the distance of propagation increased. Further, the quasi half-period of the components of the oscillatory portion was found to obey a relation different from the /`slow tail/'.During the past few years we have supplemented the waveform observations by simultaneously recording the different frequency components in the spectrum of atmospherics from the response of a number of narrow-band tuned receivers3. The results given below were derived from an apparatus covering the frequency range 40 c./s.-16 kc./s. split by twenty-seven band-pass filters arranged logarithmically at three channels per octave. The amplitude frequency spectrum is presented directly on the screen of a cathode-ray tube and photographed with the corresponding waveform in order that there should be no doubt as to the type of atmospheric to which the spectrum relates. The records obtained show in a striking way the relative amplitudes of the spectrum components in the oscillatory portion and 'slow tail' of atmospherics as recorded at different distances from the source. Like the Cambridge workers, we are indebted to the Meteorological Office "Sferics" Organization for the location of sources of atmospherics.
In Fig. 1 is shown the relative amplitude of the frequency components G{A) for atmospherics of the type described above as received at four different distances from the source. The parts of the spectrum resulting from the 'slow tail' and the higher-frequency precursor are indicated. As is to be expected, 6r(ci>) decreases as the distance of propagation increases ; but an interesting feature of the curves is that the frequency of the greatest component in the 'slow tail' decreases as the distance increases, while for the higher-frequency oscillatory portion the opposite effect occurs. The marked drop in the curves at frequencies in the region of 2 kc./s. and the widening of the trough with distance are indicative of selective attenuation of these components on an increasing scale with distance.
Fig. 1. Relative amplitude of the frequency components of atmospherics resulting from 'return strokes' of lightning flashes observed by day at different distances from the source. (A) shows the effect of distance on the maximum component in the 'slow tail', and (B) the corresponding effect for the higher-frequency oscillatory portion
Fig. 2. Effect of distance on the frequency of maximum components in the 'slow tail' of atmospherics. Results as follows : Chapman and Matthews ;	, Hepburn and Pierce ;Watson-Watt. D, day ; JV, nignt
Fig. 3. Frequency of maximum components in the oscillatory portion of atmospherics compared with results deduced from waveform observations by Watson-Watt (indicated as W.-W.) at various distances from the source. I), day ; iV, night
An exact comparison of our results with those of Hepburn and Pierce1 and Watson-Watt, Herd and Lutkin4 for the 'slow tail' is not possible, since their data were obtained from a measurement of the quarter-period T/4, the time occupied by the first quarter-cycle of the 'slow tail' waveform. If we assume T does, in fact, relate to maximum frequency components, and taking 1/T as the frequency at which these occur, we may compare the variation of this quantity with distance with our data. Fig. 2 shows our spectrum results for atmospherics received under both day and night conditions of propagation, together with the corresponding curves derived as above from the data supplied by Hepburn and Pierce1. Watson-Watt, Herd and Lutkin4 did not specify day or night observations. All the results show the same general trend and indicate clearly that the frequency of the greatest component in the 'slow tail' for a given distance is higher at night than by day, whereas for the oscillatory portion, Fig. 3, not dealt with by Hepburn and Pierce, it is lower at night than by day.
This investigation is part of a long-term study on the spectrum of atmospherics and has been aided by grants from the Department of Scientific and Industrial Research. A detailed account of the present work is in course of preparation.Hepburn, , F., and Pierce, , E. T., Nature, 171, 837 (1953).ISIAppleton, , E. V., and Chapman, , F. W., Proc. Roy. Soc., A, 158, 1 (1937).Chapman, , F. W., and Edwards, , A. G., Proc. U.R.S.I., 8, Pt. 2, 351 (1950).Watson-Watt, , R. A., Herd, , J. F., and Lutkin, , F. E., Proc. Roy. Soc., A, 162, 267 (1937).