Several efforts were directed towards characterizing external field contamination in low- and high-latitude satellite magnetic field (Magsat) data in order to produce higher accuracy lithospheric anomaly fields. Ravat and Hinze [1993] derived an empirical ionospheric correction for the equatorial region by computing dip-latitude averages of suites of profiles grouped in longitude, time, and altitude. Alsdorf et al. [1994] employed a Fourier wavenumber correlation filter to adjacent south polar passes to extract the spatially correlated signal associated with the lithospheric field from the temporally varying signal associated with the ionospheric and magnetospheric currents. Their analysis was carried out independently on dawn and dusk passes in four altitude bins which were then adjusted to a common altitude of 430 km and combined in a final correlation filtering step. Arkani-Hamed et al. [1994] applied a degree-correlation filter in the spherical harmonic domain to retain covariant harmonics in selected quiet passes of Pogo, Magsat-dawn, and Magsat-dusk data; the equatorial ionospheric correction of Ravat and Hinze [1993] was first removed from the Magsat data. Their three data sets were combined using two criteria for the correlation level, one stringent and one relaxed, which trade off signal-to-noise against signal power. These new maps all reveal the anomaly field with greater accuracy and smaller uncertainties; as such they constitute a valuable resource for future geologic and geodynamic investigations, as well as a means of leveling individual aeromagnetic surveys for producing continent-scale maps. These new techniques provide a baseline scenario for analyzing future satellite magnetic field data and for merging them with existing data sets to further improve accuracy and resolution.