An advanced approach to finding magnetometer zero levels in the interplanetary magnetic field

For a magnetometer that measures weak interplanetary fields, the in-flight determination of zero levels is a crucial step of the overall calibration procedure. This task is more difficult when a time-varying magnetic field of the spacecraft interferes with the surrounding natural magnetic field or when the spacecraft spends only short periods of time in the interplanetary magnetic field. Thus it is important to examine the algorithms by which these zero levels are determined, and optimize them. We find that the method presented by Davis and Smith (1968 EOS Trans. AGU 49 257) has significant mathematical advantages over that published by Belcher (1973 J. Geophys. Res. 71 5509) as well as over the correlation technique published by Hedgecock (1975 Space Sci. Instrum. 1 83-90). We present an alternative derivation of the Davis-Smith method which illustrates that it is also a correlation technique. It also works with first differences as well as filtered data as input. In contrast to the postulate by Hedgecock (1975 Space Sci. Instrum. 1 83-90), we find that using first differences in general provides no advantage in determining the zero levels. Our new algorithm obtains zero levels by searching for pure rotations of the interplanetary magnetic field, with a set of sophisticated selection criteria. With our algorithm, we require shorter periods (of the order of a few hours, depending on solar wind conditions) of interplanetary data for accurate zero level determination than previously published algorithms.