Seismic instruments: article

System for automatic recognition of types of sources of regional seismic events
V.E. Asming1
S.V. Asming1
A.V. Fedorov1
Z.A. Yevtyugina1
Ye.N. Chigerev2
E.O. Kremenetskaya1
1 Kola Branch of Geophysical Survey of the Russian Academy of Sciences, Apatity, Russia 2 Dorodnicyn Computing Center, Federal Research Center “Computer Science and Control” of the Russian Academy of Sciences, Moscow, Russia
Journal: Seismic instruments
Tome: 58
Number: 2
Year: 2022
Pages: 39-56
UDK: 550.34.064, 550.34.03
DOI: 10.21455/si2022.2-2
Full text
Keywords: seismic event, earthquake, explosion, discrimination, Bayesian belief network
Аnnotation: An approach to the recognition of the types of seismic event sources is proposed, based on the combination of several heterogeneous parameters of the event records and information about the territory where the seismic event occurred. The following recording parameters are used: the ratio of the amplitudes of body waves, the ratio of parts of the spectra at high and low frequencies, the magnitude, and the parameter of spectral constancy. Territory information includes data on the presence of water bodies, glaciers, mines and simplified information on natural seismic activity. Their joint use is done using a special type of Bayesian belief network. The decisions made by the network are probabilistic in nature, the probability is understood in the Bayesian sense, i.e., as the degree of confidence in the truth of the judgment, which consists in attributing an event to one of the types (mine explosion, other explosion on land, underwater explosion, earthquake, icequake). The approach is implemented in the form of a software system, which is included in the program for the interactive analysis of seismic event records LOS.