Magnetometer array studies: A review of the interpretation of observed fields

Abstract

A magnetometer array study consists of a number of variometers recording simultaneously across a two-dimensional area. Data recorded by an array should satisfy certain criteria, showing that transient magnetic events have been thoroughly monitored on the plane of observation. Interpretation of such data in terms of solid-earth geophysics involves two separate classes of problem. In the first the local geoelectric structure changes over a shorter lateral distance than does the inducing field. The observed response of the local structure is modelled numerically, taking the inducing field to be uniform. The parameters modelled are relationships between regional and anomalous horizontal and vertical field components. Operation of many instruments in an array should give better estimates of these parameters than operation of instruments singly, or in small groups. Numerical models constructed to fit observed data are non-unique, though the wide frequency range of geomagnetic events may perhaps be used to greater advantage in distinguishing between conductors in the upper crust, lower crust, and upper mantle. In constructing models a further complication arises in distinguishing between conductors simply concentrating current induced elsewhere, and conductors in which the induction is itself taking place. The second class of problem comprises determination of conductivity as a function of depth, utilizing non-uniformity in inducing fields. Here a large array of instruments should enable estimation of field gradients more accurately than is possible with fewer instruments. Interpretations published to date have perhaps not exploited this aspect of array information as fully as might prove possible. Some miscellaneous comments are: (1) daily-variation data in particular could resolve some depth ambiguity problems; (2) there is confusion in the literature between two possible definitions of phase; (3) ideas on optimum station spacing are still evolving; (4) it is not easy to decide an optimum ratio for time spent collecting data to time spent interpreting data; (5) array data have relevance to the study of electric currents in the ionosphere.