Journal of Natural Disaster Science

HYPOCENTER ESTIMATION SYSTEM FOR EARTHQUAKE COUNTERMEASURES

Shigeru NODA
Associate Professor, Department of Social Systems Engineering, Tottori University, Tottori
and
Hiroyuki KANO
Assistant Manager, Center for Disaster Management and Supply Control, Tokyo Gas Co., Ltd., Tokyo

(Received 7 April 1993 and revised 27 December 1993)

Abstract

A real-time strong-motion data processing system has been developed to detect and locate local earthquakes in and near the Kanto area. This strong-motion monitoring network system composed of three subsystems is operated by Tokyo Gas Co., Ltd. The first subsystem estimates the arrival times of P and S waves by the AIC method for AR model and the Kalman filter algorithm. The second subsystem determines the hypocenter of the event detected by the first subsystem. The earthquake hypocenter with a non-negative depth is located by an iterative method, after which the travel-times, the derivatives of the travel-times with respect to hypocenter location, and the angle of incidence at the hypocenter are computed from an earth model consisting of fiat-lying constant velocity layers. The third subsystem calculates the magnitude of the earthquake. The reliability of the location of the hypocenter and the time of origin based on real arrival time data for a four-station seismic array as calculated by this automatic system is demonstrated.

Key words

strong-motion monitoring system, hypocenter determination, onset time detection, adaptive Kalman filter