[S12-1] An interdisciplinary approach towards earthquake prediction studies I

[S12-1-01] Combining probabilistic seismicity models with precursory information: application to long-delayed aftershocks

^○Peter Shebalin (Institute of Earthquake Prediction Theory and Mathematical Geophysics)

invited

Quantitative assessments of increasing probabilities of earthquakes at a presence of precursors or precursory patterns still remain a big challenge. A strong and rising interest, however, is manifested last years to the methods of combining probabilistic seismicity models with any kind of precursory phenomena. Shebalin et al. (2014) have proposed a method to combine earthquake forecast rate-based models with any information that could locally increase the forecasted earthquake rates. The method is based on estimation of differential probability gains calculated in error diagram that evaluates the performance of the input information with respect to the reference model. The differential probability gain, that depend on a measurable quantity of the precursory phenomena, is then used at each point in space and time as an amplifying factor of the initial seismicity model. An important advantage of this combining method is its capacity to produce high expected rates. Recursive application of the method solves also a problem of the interdependence of the precursory phenomena.
In this talk I present first results on combining simple aftershock rate model with a precursory pattern of large aftershocks. The idea came from the premonitory analysis of the early aftershock statistics (forecasting algorithm EAST). In the considered case the EAST method is extended (but actuary drastically simplified) to the analysis of seismicity within aftershock sequences, and actually secondary aftershocks and main shocks, which are themselves aftershocks of the major main shock, are considered. The research was supported by Russian Science Foundation (Project N 16-17-00093).