Synchronization trends and rhythms of multifractal parameters of the field of low-frequency microseisms

The field of low-frequency microseisms is investigated with the use of data from 83 stations of the F-net broadband network in Japan over the period from the beginning of 1997 through June 2008. Vertical components with a sampling step of 1 s are used for analysis, as well as signals with a sampling step of 1 min obtained from the initial data by averaging and thinning. Long-period regularities of changes in the singularity spectrum support width Delta alpha and the generalized Hurst exponent alpha* for the field of low-frequency microseisms were revealed by estimating multifractal singularity spectra in consecutive time windows 30 min long for 1-s data and 24 hour long for 1-min data. The average value of the parameter alpha* for 1-s data significantly decreased before the Hokkaido earthquake of September 25, 2003 (M = 8.3), and was not restored subsequently to its previous level. Prior to September 2003, 1-min data on alpha* variations experienced strong annual changes, which completely ceased afterwards. Both these effects are interpreted as an increase in the degree of synchronization of microseismic noise on Japan's islands after the September 25, 2003, earthquake. This hypothesis is also supported by estimates of the measures of correlation and spectral coherence between variations in the average values of Delta alpha and alpha* calculated for 1-min data inside five spatial clusters of stations from consecutive time fragments two months long. Based on the well-known statement of the theory of catastrophes that synchronization is one of the flags of an approaching catastrophe, it was suggested that the Hokkaido event could be a foreshock of an even stronger earthquake nucleating in the region of Japan's islands.