The first satellite observations were carried out by Larkina et al. (1983). Most of them were case studies when the INTERKOSMOS-19 satellite would fly over an earthquake’s epicentre. These studies highlighted an increase in intensity for very low frequency (VLF) waves, from a few dozen minutes to a few hours prior to the event. They also showed that these intensified waves appeared on a wider area (longitude-wise) which can be attributed to ionospheric plasma’s longitudinal drift.
French scientists also conducted surveys on data gathered by GEOS satellites. A statistical analysis of GEOS-2 data was conducted for magnitude 5 earthquakes, based on a test comparing average intensity before and after a seismic event. A similar test was conducted on randomly selected data recorded when seismic activity was non-existent to compare both sets of results. This comparison led to the following conclusions:
- positive tests percentages are stable (44%) when compared to the control set when the longitudinal distance between the epicentre and the satellite is over 30°;
- these percentages rise up to 54% when that distance is under 20°, and the epicentre is located on a continent rather than in an ocean.
Case studies conducted on GEOS-2 and AUREOL-3 data showed simultaneous detection of intensity increases of the electromagnetic signal on GEOS-2 antennas and disturbances on ground ionospheric sounding data from the intermittent E layer’s electronic density. These readings coincide with earthquakes whose epicentre was located close to the Dkibouti ionospheric station (11.52°N, 42.83°E) and GEOS-2’s position (00.52°N; 37.64°E). These studies also showed intensity increases in filter outputs for signals received by AUREOL-3 (from 10 Hz to 15 kHz) when the satellite flew over seismically active areas.