Recalibration of the CHAMP satellite magnetic field measurements

The eight years of continuous CHAMP operation allows us to investigate the long-term behaviour of the vector magnetometer on board the satellite. For the processing of the FGM vector magnetometer data primarily nine instrument parameters are needed (three offsets, three scale factors and three misalignment angles). These are determined in pre-flight calibration and routinely in the in-flight scalar calibration. A review of the temporal evolution shows that the scale factors exhibit a clear change with time. This variation of the sensitivity can be well described by a logarithmic function. The offsets and the misalignment angles show no long-term trend. The Z component offset performs medium-term variations within the range of 1 nT. These have been corrected on a daily basis. When processing the FGM data over the whole period with the derived parameters, the difference between the FGM and OVM readings stays below 0.5 nT (corresponding to about 0.2 nT standard deviation). For a further improvement the error model has to be extended beyond the scalar calibration. Dynamic spacecraft magnetic fields varying over an orbit are falsely interpreted by the scalar calibration and cause apparent variations of the instrument parameters. In the case of CHAMP, the magnetic field generated by the power system is responsible for most of the remaining errors. We have applied an extended error model that can account for the time-varying disturbance fields. When considering them properly the standard deviation of scalar errors less than 0.1 nT can be achieved