Comparing and integrating artificial intelligence and similarity search
detection techniques for seismic sequences in Southern Italy
Abstract
Understanding mechanical processes occurring on faults requires detailed
information on the microseismicity that can be today enhanced by
advanced techniques for earthquake detection. This problem is more
challenging when seismicity rate is low and most of the earthquakes
occur at depth. In this study we compare three detection techniques, the
autocorrelation FAST, the machine learning EQTransformer and the
template matching EQCorrScan, for catalog improvement associated with
seismic sequences in the normal fault system of Southern Apennines
(Italy) using data from the Irpinia near fault observatory. We found
that the integration of the machine learning and template matching
detectors, the former providing templates for the cross-correlation,
largely outperforms with respect to the techniques based on
autocorrelation and machine learning alone, featuring an enrichment of
the automatic and manual catalogs of factors 21 and 7 respectively. The
output catalogs can be polluted by many false positives; so, we applied
refined selection based on the cumulative distribution of the similarity
level to clean up the detection lists and analyze final subsets
dominated by real events. The magnitude of completeness decreases by
more than one unit as compared to the reference value for the network.
We report b-values smaller than the average, likely corresponding to
larger differential stresses than for the background seismicity of the
area. For all the analyzed sequences, we found that main events are
anticipated by foreshocks, indicating a possible preparation process for
mainshocks at sub-kilometric scales.