hide
Free keywords:
-
Abstract:
The magnetic field of the Earth varies in space and time. Geomagnetism as research area that aims to describe and understand the sources of these variations is supported by two pillars: first, regular high-precision measurements in the global network of magnetic observatories and repeat stations are necessary to register the field and its variations at all. Second, mathematical methods are required in order to extract magnetic field models from this large data set. Methods applied to data offer insights in the mechanisms generating the magnetic field. This thesis covers both subjects. In a first part, following the description of the state of the art in observatory instrumentation, I explain in detail two instruments that have the potential to streamline the classical procedures: The Geomagnetic AUtomated SyStem GAUSS paves the way to automated absolute measurements, up to now only possible manually. The newly developed DI3 technique improves and simplies the standard manual measurements signicantly and thus reduces the requirements placed on observers. The second part deals with the mathematical tools available for geomagnetic field modelling. I focus on harmonic splines that can be derived from the classical approach of spherical harmonics. These base functions are interpolatory and have a localised shape while satisfying Laplaces equation. Hence, they are applicable to fit data regionally or globally. The harmonic splines are used with a data set made of repeat station and observatory measurements from Southern Africa. This region is of special interest because the field intensity is very low and and both spatial and temporal field gradients exist. Subdivided into an analysis of ancient (years 1961-2001) and recent (2005-2009) data, two continuous regional field models SAMS and X-SAMS are derived. From the analysis of the field models, a better understanding of the field behaviour is gained. Finally, the harmonic splines are used in a case study on globally distributed secular variation data. Rotating the data set in a system of coordinates aligned with the dipole axis and modelling it with the harmonic splines reveals the external origin of observed fast variations.
