The centennial evolution of geomagnetic activity and its driving mechanisms

This cumulative thesis is concerned with the evolution of geomagnetic activity since the beginning of the 20th century, that is, the time-dependent response of the geomagnetic field to solar forcing. The focus lies on the description of the magnetospheric response field at ground level, which is particularly sensitive to the ring current system, and an interpretation of its variability in terms of the solar wind driving. Thereby, this work contributes to a comprehensive understanding of long-term solar-terrestrial interactions. The common basis of the presented publications is formed by a reanalysis of vector magnetic field measurements from geomagnetic observatories located at low and middle geomagnetic latitudes. In the first two studies, new ring current targeting geomagnetic activity indices are derived, the Annual and Hourly Magnetospheric Currents indices (A/HMC). Compared to existing indices (e.g., the Dst index), they do not only extend the covered period by at least three solar cycles but also constitute a qualitative improvement concerning the absolute index level and the ~11-year solar cycle variability. The analysis of A/HMC shows that (a) the annual geomagnetic activity experiences an interval-dependent trend with an overall linear decline during 1900–2010 of ~5 % (b) the average trend-free activity level amounts to ~28 nT (c) the solar cycle related variability shows amplitudes of ~15–45 nT (d) the activity level for geomagnetically quiet conditions (Kp<2) lies slightly below 20 nT. The plausibility of the last three points is ensured by comparison to independent estimations either based on magnetic field measurements from LEO satellite missions (since the 1990s) or the modeling of geomagnetic activity from solar wind input (since the 1960s). An independent validation of the long-term trend is problematic mainly because the sensitivity of the locally measured geomagnetic activity depends on geomagnetic latitude. Consequently, A/HMC is neither directly comparable to global geomagnetic activity indices (e.g., the aa index) nor to the partly reconstructed open solar magnetic flux, which requires a homogeneous response of the ground-based measurements to the interplanetary magnetic field and the solar wind speed. The last study combines a consistent, HMC-based identification of geomagnetic storms from 1930–2015 with an analysis of the corresponding spatial (magnetic local time-dependent) disturbance patterns. Amongst others, the disturbances at dawn and dusk, particularly their evolution during the storm recovery phases, are shown to be indicative of the solar wind driving structure (Interplanetary Coronal Mass Ejections vs. Stream or Co-rotating Interaction Regions), which enables a backward-prediction of the storm driver classes. The results indicate that ICME-driven geomagnetic storms have decreased since 1930 which is consistent with the concurrent decrease of HMC. Out of the collection of compiled follow-up studies the inclusion of measurements from high-latitude geomagnetic observatories into the third study’s framework seems most promising at this point.