Magnetospheric observations of whistler‐mode emissions were made by the OGO 1 satellite over the frequency range 0.3 to 100 kHz. Taken during the geomagnetically quiet period from September 1964 through May 1965, the data cover geocentric distances from 2 to 24 RE and geomagnetic latitudes in the range ±50°. Emissions are found within L ∼ 10 at all local times except for L > 5 in the midnight‐to‐dawn sector, with intensity peaking at L ∼ 4 and L ∼ 9 near 10 hours local mean time. Their spectra resemble those of chorus and mid‐latitude hiss observed on the ground. An important new result is the observation that the upper‐frequency limit of most of these emissions is proportional to the minimum electron gyrofrequency along the magnetic fieldline passing through the satellite. This is interpreted to mean that the source of these emissions lies close to the equatorial plane. Estimates of the average intensity of these emissions agree with the intensity required to explain the precipitation of electrons through pitch‐angle scattering by whistler‐mode waves. Emissions observed on the dayside beyond L = 8 are burst‐like, and their occurrence tends to decrease with increasing radial distance. The location of the maximum intensity in this region agrees with that found for 1‐kev electron fluxes, giving strong support to theories of generation based on electron cyclotron resonance. No emissions were observed beyond the estimated position of the shock boundary.

Two new types of low‐frequency noise, designated ‘broadband’ and ‘highpass’ have been detected in the distant magnetosphere by the VLF/LF experiment on the OGO 1 satellite. Broadband noise extends over the entire range of observations from 0.2 to 100 kHz and the intensity decreases with increasing frequency. It occurs in bursts having durations of a few minutes or less. It shows no connection with any of the expected plasma cutoff or resonance frequencies and is believed to be a nonpropagating disturbance generated in the vicinity of the satellite. Highpass noise extends from a characteristic low‐frequency cutoff to above 100 kHz and occurs in bursts lasting tens of minutes. This cutoff has only been observed above 20 kHz. Above the cutoff the intensity shows little variation with frequency. Both types of noise are observed predominantly at L greater than 5 in or near the night hemisphere. The occurrence of both types of noise is highly correlated with the auroral electrojet index. Several times noise bursts began within 2 min of the onset of micropulsations in the polar region even though the satellite was near apogee (24 earth radii). The peak rms magnetic intensity in a highpass noise burst has a maximum of 8×10−5 γ Hz1/2 and averages 10−5 γ Hz−1/2. In free space these magnetic intensities correspond, respectively, to 10−12 and 3×10−14 watt m−2 Hz−1. The peak levels of both types of noise are 3 or more orders of magnitude stronger than emissions from extraterrestrial sources observed with interplanetary probes.

A new whistler phenomenon has been identified through measurements at ground stations, on an Aerobee rocket between 100 and 200 km, and on the Alouette satellite at 1000 km. The new phenomenon is called the 'subprotonospheric' or 'SP' whistler, since most of its path appears to be restricted to the region below about 1000 km. The first example of an SP whistler was reported by Barrington and Belrose. In the present report a large number of observations are summarized, and the basic characteristics of the new phenomenon are described. Experimental results are presented which suggest that the whistler ray path is confined to the region between roughly 100-and 1000-km altitude, and that the whistler energy can echo back and forth between these levels. The SP phenomenon occurs mostly at night, typically within a few hours after sunset. SP events are often observed over a period of one or two hours in duration and, for a single Alouette pass, have been observed over a north-south range as great as 2000 km in extent. The evidence. suggests that the SP phenomenon occurs mostly near sunspot minimum and at dipole latitudes greater than 45 degrees.

Whistler-mode signals from low-frequency ground-based transmitters in the 60-100 kHz range are observed by the OGO-1 satellite at frequencies unusually close to the local electron gyrofrequency, [t•. The intensity of signals observed on outbound passes near 40 ø geomagnetic latitude remains high until the decrease in [t• causes the normalized frequency [/[t• to approach 0.9. Here the signal disappears abruptly into the background noise, a change of at least 20 dB. Ray-tracing calculations provide a simple explanation for the sudden loss of signal. Rays in this frequency range which are launched just inside the plasmapause reach an altitude where [/[• = 0.9 before abruptly bending inward toward the earth's equator. Rays launched outside the plasmapause are reflected in the form of a "Spitze" at the point where the plasma frequency decreases to the wave frequency. Thus the primary features of the observed LF propagation in the magnetosphere can be explained using classical cold-plasma expressions for the refractive index. The narrow spread of the distribution of observed upper-cutoff frequencies indicates that the cutoff is caused by the loss of ducting at 0.5 [•o as suggested by Smith [1961]. Satellite-borne receivers, which extend the observations to nonducted signals, show the presence of whistler-mode signals at higher frequencies relative to the gyrofrequency. Using the OGO-3 satellite, Angerami [1970] has observed whistlers at frequencies as high as 0.7 •Present address: ESL, Inc., 495 Java Dr., Sunnyvale, California 94086. Copyright (• 1977 by the American Geophysical Union. [a. These whistlers had apparently leaked from ducts as their frequency increased above 0.5 Signals from ground-based transmitters operating near 18 kHz have been studied by Heyborne [ 1966, pp. 111-202] using the OGO-1 satellite. Transmissions were generally observed from low altitudes out to the region where [/[• decreased to •-0.3, which on outbound passes occurred in the vicinity of 30 ø to 40 ø geomagnetic latitude. A slow fadeout rather than a sudden cutoff was commonly observed. After the fadeout there sometimes occurred magnetic equator, J. Geophys. Res., 75, 6115-6135. Angerami, J. J., and D. L. Carpenter (1966), Whistler studies of the plasmapause in the magnetosphere, 2, Electron density and total tube content near the knee in magnetospheric ionization, J. Geophys. Res., 71, 711-725. MAN-MADE WHISTLER-MODE SIGNALS 829 Angerami, J. J., and J. O. Thomas (1964), Studies of planetary atmospheres, 1, The distribution of electrons and ions in the earth's exosphere, J. Geophys. Res., 69, 4537-4560. i3udden, K. G. (1966), Radio Waves in the Ionosphere, p. 269, Cambridge University Press, London. Carpenter, D. L. (1967), Relations between the dawn minimum in the equatorial radius of the plasmapause and Dst, Kp, and local K at Byrd Station, J. Geophys. Res., 72, 2969-2971. Carpenter, D. L. (1968), Ducted whistler-mode propagation in the magnetosphere; A half-gyrofrequency upper intensity cutoff and some associated wave growth phenomena,...

Natural whistlers observed in the ionosphere, but not by nearby ground receivers, may exhibit a frequency versus time curve that differs appreciably from the typical curve of whistlers observed at ground level. For wave frequencies well below the minimum value of electron gyro‐frequency along the path, the frequency‐time curve of most ground‐observed whistlers can be approximated within a few per cent by the relation D=tƒ1/2=constant where t represents travel time at frequency ƒ. However, a number of recent Alouette whistler observations at 1000 km show relatively large departures from this relation. The purpose of this note is to describe a particular departure in which the quantity D(ƒ) = tƒ1/2 exhibits an anomalous increase with increasing frequency.

Type III solar noise bursts have been observed in the frequency range 25-100 kHz with the VLF detector on OGO 3. The bursts decrease in frequency from 100 kHz (the highest frequency of observation) to as low as 25 kHz in approximately 45 min. The intensity at 100 kHz increases for about 20 min then decays over a period of approximately 1 h. The variation of the intensity with time becomes less regular at lower frequencies. Observed maximum intensities near 80 kHz range from 3 • 10 -I8 to 2 • 10 -16 W m -~ Hz -1. Bursts are predominantly associated with west-limb flares. Their commencement at 100 kHz tends to follow type III bursts observed at 2-4 MHz by about 10 min. Observed drift rates and decay times correspond roughly to those extrapolated from higher frequency measurements. Type III and the so-called 'high-pass' noise bursts often occur simultaneously, presenting a problem in identification. The solar noise events can be distinguished by their relatively slow time variation, smooth spectrum, and low intensity.

Type III bursts have been observed in the 30-100kHz range which show localized enhancements in their spectra. The enhancements exhibited by a series of type III bursts extending over a period of several hours show a consistent decrease in frequency. The rate of this decrease is consistent with the movement of an interplanetary shock. The passage of such a shock by the Earth is suggested by the occurrence of sudden commencements at approximately the expected time.Examination of solar type III bursts at kilometric wavelengths shows that the intensity of emissions generally decreases with decreasing frequency. In contrast, occasional bursts are observed which display band-limited enhancements of their spectra. Two such bursts are discussed below.The observations reported here were made by the Stanford University vlf experiment on the OGO-3 spacecraft. This experiment measured the characteristics of radio waves in the 0.2-100 kHz range; only the data from a receiver sweeping the range 10-100 kHz each 2.3 min are shown here. Further details concerning this experiment are given by Dunckel et al. (1972) andDunckel (1974). The digital output from this receiver has been converted into the 'ampligram' shown in Figure 1. The height of an individual trace is proportional to the logarithm of the receiver output; the overlap produced by strong emissions causes the large white areas. The narrow spectral lines of invariant frequency, as at 1415 UT at 93 kHz, are due to spacecraft interference. Figure 1 shows a sequence of three type III bursts that exhibit large enhancements in their spectra near 70 kHz. These enhancements have power spectral densities about ten times greater than that at 100 kHz. The frequency of these enhancements decreases from 72 kHz in the first event at 12 UT to 64 kHz in the third event at 19 UT. The enhancements do not appear to be related to the plasma immediately surrounding the Earth, since the low-energy electron detector on board the OGO-3 spacecraft indicates that it was outside the bow shock during ~he period shown in Figure 1 except for 1203-1243UT and 1250-1311UT (personal communication, V. Vasyliunas). Neither do the enhancements appear to result from the effects of the local interplanetary medium, since Vela 3 measurements at 18-21 UT on this day indicate a plasma frequency of 28 kHz (Bame et al., 1970). The second event is associated with a flare at W51 commencing at 1414 UT and reaching maximum at 1418 UT. The third event is associated with a flare in the same sunspot region commencing at 1753 and reaching maximum at 1806 UT. No known flares are associated with the first