The Japan Meteorological Agency (JMA) has been generating the Unified Seismic Catalog using data on automatically detected earthquakes. The JMA classifies automatically detected hypocenters into reliable results and unreliable results by visual inspections. It was extremely difficult to classify all of them through visual inspection reviews due to the enormous number of seismic events that occurred in the case of the 2016 Kumamoto Earthquake. Given this, we are proposing a more efficient

more »

... ication method by using cross-correlation to select reliable results from the large number of events that have not yet been reviewed. If an automatically processed event ("target event") is found to be close to a reviewed event ("template event") and their waveforms around the P and S phases are similar, the target event can be regarded as a hypocenter with high reliability without any need to perform a visual inspection review. Using 3,337 template events, we applied this method to 35,921 target events that were recorded between 21:00 on April 14 and 24:00 on April 28 around the seismically active area of the 2016 Kumamoto Earthquake. As a result, 20,970 events (58%) were classified as being closely related events. Since these events can be regarded as highly reliable and there is no longer any need to perform visual inspections, this method can contribute to greater work efficiency. Even for the events that could not be classified as being closely related events, we were able to extract other appropriate hypocenters by adding their waveforms to the templates after visually inspecting them. We also expect to be able to use this method to classify noise or blast events more efficiently by using the waveforms of noise or blast events as templates.