PROPERTIES OF THE MAGNETOSPHERIC HOT PLASMA DISTRIBUTION DEDUCED FROMWHISTLER-MODE WAVE INJECTION AT 2400 HZ - GROUND-BASED DETECTION OF AZIMUTHAL STRUCTURE IN MAGNETOSPHERIC HOT PLASMAS

Siple station VLF wave injection experiments aimed at finding the prop erties of the magnetospheric hot plasma were conducted for a 9-hour pe riod between 1705 and 0210 UT on January 23-24, 1988. A special freque ncy versus time format, lasting 1 min and transmitted every 5 min, con sisted of a sequence of pulses, frequency ramps, and parabolas, all in a 1-kHz range centered on 2400 Hx. The transmitted signals, after pro pagating along a geomagnetic field-aligned duct, were recorded at Lake Mistissini, Canada. At various times during the 9-hour interval the S iple signals showed features characteristic of wave-particle interacti ons, including wave growth, sidebands, and triggered emissions. Our ob servations, primarily at 2400 Hz, show that (1) there were no correlat ions between the initial levels, the growth rates, and the saturation levels of constant-frequency pulses; (2) in general, the values of gro wth rate and saturation level of two pulses injected within 30 s were nearly the same; (3) the initial level, growth rate, and saturation le vel showed temporal variations over 5-15 min and 1-2 hour timescales; (4) the leading edges of constant-frequency signals underwent spatial amplification; and (5) under conditions of saturation the received sig nal bandwidth (similar to 20 Hz) remained constant over a 1-hour perio d, although the saturation level and growth rate varied during the sam e period. On the assumption that gyroresonant interactions were respon sible for the observed wave growth and saturation, the timescales over which those phenomena varied provide constraints on the possible ener getic electron population within the duct. In the reference frame of t he duct (L similar to 5.1, N-e similar to 280 cm(-3)) the particle flu xes showed no variation over a 30-s timescale but varied over 5-15 min and 1-2 hour timescales. The 5-15 min timescale variations indicate l ongitudinal structures ranging from similar to 0.20 or similar to 100 km (in the equatorial plane) for electrons with energy E = 0.6 keV and pitch angle alpha = 40 degrees, to similar to 5 degrees or similar to 2800 km for electrons with energy E = 11 keV and pitch angle alpha = 80 degrees. The hour-long time variations indicate longitudinal struct ures ranging from similar to 2 degrees or similar to 1100 km (in the e quatorial plane) for electrons with energy E = 0.6 keV and pitch angle alpha = 40 degrees, to similar to 45 degrees or similar to 25,000 km for electrons with energy E = 11 keV and pitch angle alpha = 80 degree s. We conclude that ground-based active and passive wave experiments h ave substantial potential for investigating properties of the hot plas ma of the magnetosphere.