No abstract

The magnitude, colatitude, longitude, and distance from the center for radial dipoles were adjusted by a least‐squares procedure so as to reproduce the surface magnetic field synthesized from the Fineh‐Leaton (F.L.) 1955 Spherical Harmonic Coefficients and for the 1965 International Geomagnetic Reference Field (IGRF). The rms residual was reduced to 28 γ using 21 dipoles for the 1955 field (6th degree and order) and to 25 γ using 35 dipoles for the IGRF (8th degree and order). These residuals include all components of the surface field. The dipole parameters were then varied using a least‐squares method to approximate the secular change field. These solutions match the Z component of secular change field synthesized from Leaton's 1955 coefficients to a rms residual of 2.15 γ per year and the IGRF secular change field to a rms residual of 1.26 γ per year. In both cases the dipole sources were found to be approximately 0.2 earth radius from the center of the earth.

The first degree external spherical harmonic coefficients are obtained for each year of a 12-year period centered at 1958.5, using annual mean values of X, Y, and Z components from 54 magnetic observatories. Values of the coefficient g0e of the zonal first degree harmonic clearly show a solar cycle variation. The peak-to-peak amplitude of the variation is approximately 29 nT, which is consistent with values obtained earlier by first filtering data to retain only those variations having periods near the solar cycle and then subjecting the filtered data to a spherical harmonic analysis. The variation in g0e is found to correlate extremely well with the annual mean Ds, index and with the annual number of days having Ap>60. Based on statistics of the mean square successive difference, an explanation is presented why the obtained solar cycle variation in g0e, which is very much smaller in magnitude than the standard deviations calculated by the conventional method, is statistically meaningful. The determined absolute (not relative) values of gie are in agreement, within several nanotesla, with the expectation from a theoretical model of solar wind compression of the magnetosphere and an analysis of the Dst index.

The problem of collecting data for making geomagnetic charts including secular change is reviewed. Satellite data gives excellent coverage, but the satellite total field intensity data is not sufficient to properly define the field components. In constructing charts, the specific time and space filtering that is used should be specified so the user can make needed corrections in the use of the charts. An attempt should be made to include time variations having periods of one year or greater. In the space domain spherical harmonic coefficients up to degree 12 or 13 should probably be included. Adequate charts in the future will likely require the use of some external coefficients.

Techniques for resolving the components of a multipath field are investigated in this paper. Both linear and circular probing schemes are discussed. A method for analyzing the li?early P':lhed data is given, which is considerably simpler than the proce~ure followed by Watter:on _m an e3;rlier paper. The method is also extended to the circular case. It 1s found that the resolutiOn m the cucular case is poorer than for the linear scheme.