A statistical elaboration of multiple observations from a geophysical monitoring network.

The conditional probability of an exceptional event such as an earthquake can be defined through the variations of one of the geophysical observables that usually precede it. Starting from the frequencies of seismic events above a certain magnitude and the variation of an observable that exceeds its threshold, the probability gain of an earthquake is linked to the correlation between the two events. Some examples are discussed where observations made both from satellite, using MEPED onboard NOAA, and on the Earth's surface by ULF magnetic detectors produce probability gains around 4. The same approach is extended to multiple observations both from the ground and from satellites, considering not completely independent observables, where the probability gains are calculated for both distinguishable and indistinguishable observables. An example is also discussed in this case, showing how the overall probability gain increases and depends on the comparison volume of the observables. Finally, the probability gain is estimated by using a network of geophysical monitoring instruments.