Induction studies with satellite data

The natural variations of the Earth's magnetic field of periods spanning fr om milliseconds to decades can be used to infer the conductivity-depth prof ile of the Earth's interior. Satellites provide a good spatial coverage of magnetic measurements, and forthcoming missions will probably allow for obs ervations lasting several years, which helps to reduce the statistical erro r of the estimated response functions. Two methods are used to study the electrical conductivity of the Earth's ma ntle in the period range from hours to months. In the first, known as the p otential method, a spherical harmonic analysis of the geomagnetic field is performed, and the Q-response, which is the transfer function between the i nternal (induced) and the external (inducing) expansion coefficients is det ermined for a specific frequency. In the second approach, known as the geom agnetic depth sounding method, the C-response, which is the transfer functi on between the magnetic vertical component and the horizontal derivative of the horizontal components, is determined. If one of these transfer functio ns is known for several frequencies, models of the electrical conductivity in the Earth's interior can be constructed. This paper reviews and discusses the possibilities for induction studies us ing high-precision magnetic measurements from low-altitude satellites. The different methods and various transfer functions are presented, with specia l emphasis on the differences in analysing data from ground stations and fr om satellites. The results of several induction studies with scalar satelli te data (from the POGO satellites) and with vector data (from the Magsat mi ssion) demonstrate the ability to probe the Earth's conductivity from space . However, compared to the results obtained with ground data the satellite results are much noisier, which presumably is due to the shorter time serie s of the satellite studies. The results of a new analysis of data from the Magsat satellite indicate hi gher resistivity in oceanic areas than in continental areas. However, since this holds for the whole range of periods between 2 and 20 days, this diff erence probably is not caused purely by differences in mantle conductivity (for which one would expect less difference for the longer periods). Furthe r studies with data from recently launched and future satellites are needed .