An investigation of the propagation of whistlers in magnetosperic ducts by means of ray-tracing, curve-fitting and direction-finding techniques

Strangeways, Hal John (1978) An investigation of the propagation of whistlers in magnetosperic ducts by means of ray-tracing, curve-fitting and direction-finding techniques. University of Southampton, Doctoral Thesis.

Abstract

It is shown that upgoing whistler-mode waves can become trapped through the sides of field-aligned electron density enhancements (ducts). This occurs when latitudinal gradients of refractive index result in upgoing whistler-mode waves becoming incident on a duct with wave normals which are field aligned, or nearly so, at an altitude near or above that of the minimum in the phase refractive index along field lines corresponding to the duct position. The whistler exit points for trapping by this mechanism are investigated for magnetospheric models which have both symmetric and asymmetric plasma distributions about the equatorial plane. The exit points are generally found to be at a higher or lower L-value than that of the duct position and are typically about 10 in latitudinal extent at 300 km altitude. For an asymmetric plasma distribution, it is found that, for whistlers that have propagated down the same duct, there can be a difference in size and geomagnetic latitude between whistler exit points in the northern and southern hemispheres.The frequency-time relationships for whistlers trapped both through the base and through the side of ducts are determined by ray-tracing. Residual patterns are obtained by curve-fitting to these frequency time relationships and are shown to be consistent with residual patterns obtained by. curve-fitting to whistlers.Trapping in ducts with different cross-sections and with gradual ends is also investigated. It is shown that for waves trapped in ducts with complex cross-sections, a furt:ier trapping effect can take place resulting in waves being constrained to follow ducted pathsin a fine structure peak in the density enhancement.It is concluded that trapping of VLF waves generally occurs both through the base and the side of ducts and that the proportion of wave energy trapped in a duct in these-two ways depends on, the altitude of the base of the duct,its cross-section,and the way.,.its density enhancement tapers off at its ends.It is shown that.for whistlers received prior to an eclipse of the sun in Eastern Canada on 10 July 1972, the mechanism of trapping through the side of ducts can explain a discrepancy between the L-value of propagation of the whistlers (found from their nose frequencies) and the L-value of their exit points (found using direction-finding goniometer receivers). It is also shown that the upper cut-off frequency of these whistlers supports the suggestion that they resulted from trapping through the side of ducts and might enable the enhancements of the ducts to be estimated.The polarisation and multipath errors for crossed-loops and crossedloops plus vertical monopole goniometer direction-finding systems are investigated in some detail. It is shown that for the crossedloops goniometer technique, the bearing errors due to wave polarisation and multipath propagation are generally of opposite sign, producing a combined bearing error that 'can be ' smaller than thatof either of these two effects alone. For the crossed-loops goniometer technique, the variation of the bearing error with ionospheric and ground parameters, with the distance from exit point to receiver and with latitude and direction (north to south or south to north) of propagation' are investigated for the two different approximations of neglecting phase dependent terms in the goniometer receiver response and of assuming a phase continuous VLF source. It is found that the combined bearing error due to wave polarisation and multipath propagation is, generally less than tea degrees in magnitude and fairly constant with varying wave. frequency when the, phase-dependent terms in the receiver response are ignored. For the phase continuous source approximation, however, the bearing error varies considerably with wave frequency but would generally be less than ten degrees in magnitude if the bearing `were averaged over a typical whietlgr bandwidth.

This record has no associated files available for download.

Contributors

Download statistics