GEOPHYSICAL RESEARCH LE•TTERS, VOL. 11, NO. 5, PAGES 511-514, MAY1984 OBSERVATIONS OF DOUBLE LAYER-LIKE AND SOLITON-LIKE STRUCTURES IN THE IONOSPHERE M. H. Boehm*, C. W. Carlson, J. McFadden*, F. S. Mozer* Space Sciences Laboratory, University of California, Berkeley, CA 94720 Abstract. Two types of large electric field signatures, individual pulses and pulse trains, were observed on a sounding rocket launched into the afternoon auroral zone on January 21, 1982. The typical electricfields in the indi- vidual pulses were 50 mV/m or larger, aligned mostly parallel to B, and the corresponding potentials were at least 100 mV (kT --0.3 eV). A lower limit of 15 km/sec can be set on the velocity of these structures, indicating that they were not ion acoustic double layers. The pulse trains, each consisting of on the order of 100 pulses,were observedin close association with intense plasma frequency waves. This correlation is consistent with the interpretation of these trains as Langmuir solitons. The pulse trains corre- late better with the intensity of the field-aligned currents than with the energetic electronflux. Introduction A sounding rocket was launched from Cape Parry, Canada,(--74 degrees invariant latitude)at 14:25MLT on January 21, 1982. the trajectory was northwest through a visible quiet auroral arc, with an apogee of 640 km. The instruments on this rocket included two axis DC and AC electric field measurements, a magnetometer, density and temperature probes, and an electron detector capable of measuring a full pitch angle distribution once per second. The AC electric field was telemetered as broadband analog waveforms with a bandwidth of 10 kHz for the component along B (the rocket was aligned with the magnetic field by an attitudecontrolsystem) and 20 kHz for the perpendicu- lar component. The AC density was similarly transmitted with a bandwidth of 10 kHz. Since various analytic, laboratory and computer simulations of double layers[Tor- ven and Anderson, 1979; Levine and Crawford, 1980; Lotko and Kennel, 1983; Hudson et al., 1983 and refer- ences therein] together with the recent observations of double layersby the S3-3 satellite [Temerinet al., 1982] suggested that such structures could also exist in the iono- sphere, these analog waveforms were searched for any large pulseswhich would indicate the presence of double layers or solitons. Observations of Isolated Pulses Several hundred isolated pulses, similar to that illus- trated in Figure 1, were found in the electric field data. The typically observed electric field was mostly parallel to B and usually upwards, with a magnitude of 50 mV/m, corresponding to a potential between the parallel spaced electrodes of--80 mV. The duration of the pulses was typically 50 - 100 tzs. Since they almost always saturated the automatic gain control amplifier through which all these Signals were fed, the actual amplitude of these sig- nals was determined from the size of the overshoot pro- duced by a 100 Hz high pass filter near the signal input. _ , *Also Physics Department. Copyright 1984 by the American Geophysical Union. Paper number 4L0461. 0148-0227/84/004L-0461503.00 The saturation also means that the observed ---100 tzs pulse width can only be interpreted as an upper limit, as it is unclear how long the AGC amplifier took to recover from saturation. Various checks were made to determine if these signals were real. The timing of the pulseswas checked againstall the periodicities in the rocket, i.e., high voltage sweep periods, spin period, etc., and no correlations were found. There are some coinciding signalswhich are not completely understood in the current monitors of the particle detec- tors, but it seems probable that these were caused by the large perturbations in the thermal distribution function which potential steps of a hundred millivolts or greater would probably carry with them. It is clear from the dark line in Figure 1, which is a laboratory oscilloscope trace of the output of the filter electronics in the payloadwhen it is excited by a 90 mV, 100tzs, input pulse, that the signal is coming from the initial differential amplifier, and not from some place in the middle of the electronics. Also, the per- pendicular signal occurred in both polarities, which would not be expected from an electronic problem. In addition, coinciding signals were seen in many casesin the two elec- tric field channelsand in the density, with the density sig- nal lasting much longer than the electric field signals (severalmilliseconds). All taken together, it seemsclear that the observed signals are coming from the plasma, and are not an electronic artifact. Two more isolated pulse events are shown in Figure 2. The top two panels are the two components of the electric field and the bottom panel is the wave density measure- ment. In these examples as well as typically, the measured perpendicular field in the pulses was a few percent of that in the parallel component. The density increase (third panel, Figure 2) starts at approximately the time of the electric field pulses (the relative timing betweenpanels is uncertain by 1 - 2 ms) and lasts for several milliseconds afterwards, whenever it occurs - which is mainly on the up leg of the flight below --400 km altitude. Observation of Pulse Trains In addition to the individual pulses, such as those shown in Figures 1 and 2, trains of electric field spikes - in which the individual pulses are not easy to discern because of the almost continuous saturation - were seen several times, mostly just outside a moderate arc which was crossednear apogee. These pulse trains are strongly correlated with intense Langmuir frequency waves which were measured with a swept frequency analyzer. The trail- ing end of one such pulse train is shown in Figure 3; the entire train usually lasted on the order of one to several hundred milliseconds. Near the beginning of Figure 3, the individual pulses are not clearly separated, but there are periods of long positive saturation in the parallel electric field (while there are no long periods of negative satura- tion), indicatingthe presence of large negative spikes. Near the end of the figure, the pulses separate into the characteristic negative spikes followed by a positive overshoot with the usual--2 ms decay time. The indivi- dual spikes here contain electric fields of several hundred millivolts per meter parallel to B, and only a few millivolts 511