Geomagnetic induction studies in Southern Scotland

Abstract

The Southern Uplands of Scotland is at present a region of great interest to both geologists and geophysicists. The complex tectonic history associated with the closing of the proto-Atlantic ocean has yet to be determined. Previous geophysical studies have shown that the Southern Uplands is atypical of normal continent. Geomagnetic investigations have indicated a zone of anomalously high electrical conductivity underlying the Southern Uplands region at a depth of 12-30 km. In order to determine further the exact structure and spatial variation of this conductivity zone, two geomagnetic studies have been carried out in the region. A Geomagnetic Deep Sounding Array of 19 Gough-Reitzel variometers recorded the naturally varying Earth's magnetic field during December 1973 and January 1974. This was followed by Magneto-Telluric observations on lines perpendicular and parallel to the strike of the supposed anomaly. In this thesis, reviews of various regional MT studies, the geophysical significance of conductivity measurements and the known geology and geophysics of the Southern Uplands region are given. The aims, and relevant theory, of induction studies are also presented. The observational procedures for both the GDS and MT techniques, and the type and form of the MT activity, are described. Techniques for analysing the MT data are developed in some detail. Methods are proposed, and examples given, for (a) estimating the gross structural strike direction, (b) averaging response function estimates, (c) estimating the confidence intervals of the response functions, (d) estimating new forms of coherence functions, which exhibit many desirable properties, and (e) a frequency-time analysis· for estimating the response.functions for sub-intervals of the data set. The full MT and GDS estimates from analysis of the single station data - rotated major and rotated minor impedance estimates, azimuthal angles, skew factors and real and imaginary induction vectors - are presented and qualitatively discussed. A one-dimensional interpretation of two-dimensional MT data is examined and shown to be valid for 'rotated major' impedance estimates from locations sufficiently distant from gross lateral inhomogeneity. Various methods for determining an 'optimum model' that best satisfies, in some manner, the observed MT responses are reviewed. A Monte-Carlo inversion procedure is developed and applied to the 'rotated major' data from six of the thirteen locations. It was considered, for various reasons, unjustifiable to interpret all the data. The GDS and MT results agree on the complexity of the conductivity variations in the Midland Valley, the Southern Uplands and Northern England. The simple 'Eskdalemuir anomaly' proposed by Edwards et al. (1971) cannot explain the observations. A conductive layer is required beneath the Midland Valley at a depth of no greater than 11 km. The conductive zone underlying the Southern Uplands is at a depth greater than 24 km. For the Northern England response, the top layer of highly conducting sediments 'screen' the possible effects of a 'lower crustal/ upper mantle' conductive layer. The geological and geophysical implications of the acceptable MT models are discussed. In this work, the quantitative information, offered by the MT technique, is shown to be necessary for a full interpretation of the conductivity distribution. Also, estimation of the phase response, as well as the amplitude response, of the impedance tensor elements is shown to resolve the surface structure of the acceptable models. Various suggestions are made about further investigation of the region.