[S-CG53] Reducing risks from earthquakes, tsunamis & volcanoes: new applications of realtime geophysical data

[SCG53-P03] Constructing envelope functions of seismic intensity for the evaluation of EEW in Japan

*Hong Peng¹, Stephen Wu^1,2, Masumi Yamada^3,4 (1.The Institute of Statistical Mathematics, 2.The Graduate University for Advanced Studies, 3.Kyoto University, 4.Disaster Prevention Research Institute)

Keywords:Earthquake early warning, Envelope function, Performance metrics

The Integrated particle filter (IPF) and propagation of local undamped motion (PLUM) are two Earthquake Early Warning (EEW) systems employed in Japan. Despite their proven efficacy, false alarms have been recorded (e.g., the Torishima-Kinkai earthquake in July, 2020, which has a seismic intensity (SI) of 5 upper, but no seismic station recorded an SI of 1 or higher). In this study, we pick 500,000 events from JMA seismic stations, Kyoshin Network (K-NET) and Kiban Kyoshin Network (KiK-NET) in the past 20 years as the database, and introduce two empirical envelope functions and corresponding correction methods for simulating the SI curves of them, with a specific focus on evaluating the final alarm selection of EEW system by Japan Meteorological Agency (JMA) in real-time. Our findings suggest a substantial potential for using envelope functions as reliable performance metrics when comparing the performance of different EEW systems in real-time.