Structure and secular change of the geomagnetic field

This report is a review of some of the recent developments in the knowledge of the structure and secular variation of the main geomagnetic field. It is hoped that this discussion will help to complement and update the excellent summary given by Zmuda [1973]. Many past papers have imbedded in them such phrases as ‘lack of adequate data, ‘uncertainty of high‐order coefficients’ and other qualifiers that lead the reader to infer that the numerical models (and maps) have large errors or that the conclusions of the studies must be considered tentative. The recent comprehensive global mapping of the field by near‐earth satellites, in scalar magnitude only, has partially quelled such statements and shown that many of the prior results were approximately correct. Confusion has often arisen since attention was focused on the fact that many of the individual higher‐order coefficients of the spherical harmonic expansions did not agree when computed by different analyses [see, e.g., Zmuda, 1971, p. 148]. It has now become evident that it is more meaningful to compare the spatial structure represented by the sum of the functions than to place undue emphasis on the correspondence of each individual term. This is true because aliasing is dependent on the degree of expansion and data distribution, so that two sets of coefficients can differ slightly in phase and amplitude and yet be fairly consistent with the data in most areas.