New Methods for Analyzing the Nature of Seismic Regime Non-Stationarity

Three new methods in geophysics for analyzing the non-stationarity of data series are used to analyze earthquake catalogs of the Kuril-Kamchatka and Mid-Atlantic Ridge regions. The possibility of identifying the non-stationary component of the seismic regime and the nature of non-stationarity are discussed. The use of new methods confirmed a number of known (expected) patterns and indicated a number of non-trivial points. Among these, the following were identified: a tendency of increasing non-stationarity with increasing characteristic time, which may conform to the correspondence of seismic activity spectrum to the flicker noise; 2) a difference in the nature of the magnitude distribution, possibly corresponding to a decrease in b-values, for clustering main events; 3) detection of two tendencies in the seismicity regime over time — clustering at smaller relative distances and repulsion at larger ones; these tendencies may correspond to the epochs of seismic activity growth and subsequent attenuation, during the accumulation of tectonic stresses. The results indicate the prospects of using these new analysis methods in seismology, providing clarification of the non-stationary nature of the seismic process.

non-stationarity analysis, new methods, earthquake distribution by magnitude, time intervals between events, event flow, “nearest neighbor” graph