On long-term trends in European geomagnetic observatory biases

We investigated the European geomagnetic observatory biases over 42 years, considered as contributions of the crustal field, and generally assumed to be constant in time. To estimate these biases, we compared observatory annual means to predictions given by the continuous CM4 model, and to four other core field models for different epochs. Solar-cycle related external fields are clearly present in the residuals. Although well-known, no suitable model to minimise them exists. We found that an empirical approach, taking advantage of the homogeneity of the external influences in the European region, can minimise these influences. Their reduction is better than when the external field description included in the comprehensive CM4 model is used. At several locations clear long-term trends remain after subtraction of the core field and minimisation of external fields. We investigated whether they are due to an insufficient description of the core field secular variation by the CM4 model, or to changes in induced lithospheric fields.