Changes in seismicity pattern due to the 2016 Kumamoto earthquake sequence and implications for improving the foreshock traffic-light system

TL;DR

This study analyzes seismicity changes before and after the 2016 Kumamoto earthquakes, highlighting stress variations and aseismic slips, and discusses how these insights can enhance foreshock prediction systems.

Contribution

It reviews and extends previous seismological laws to better understand stress changes and preslip phenomena, proposing improvements for foreshock traffic-light systems.

Findings

High stress zones were identified near future earthquake epicenters before the sequence.

Aseismic preslip reduced stress locally prior to the M7.3 earthquake.

Stress decreased along the fault zone after the sequence.

Abstract

Crustal deformation due to the 2016 earthquake sequence in Kumamoto, Japan, that culminated in a preceding earthquake of magnitude M6.5 and a subsequent M7.3 earthquake 28 hours later, caused stress perturbation on and around the causative Futagawa-Hinagu fault zone. Monitoring changes in seismicity pattern along this zone plays a role in understanding the process before and after major earthquakes. For this purpose, stress-dependent laws in statistical seismology can be used: the Gutenberg-Richter frequency-size law and the Omori-Utsu aftershock-decay law. We review the results obtained by using these laws in previous studies to show a zone of high stress near the eventual epicenters of the M6.5 and M7.3 earthquakes before the start of the Kumamoto sequence, and after it, showing a decreasing trend in stress along the Futagawa-Hinagu fault zone. Detailed analysis suggests aseismic slips along the causative fault zone. The aseismic preslip locally reduced stress just prior to the M7.3 earthquake near its epicenter. Recently, a system was proposed by Gulia and Wiemer (2019) that utilizes the Gutenberg-Richter frequency-size law to judge, immediately after a large earthquake, whether it was the mainshock or a foreshock to a future event. Based on the reviewed results and our new results, further research that takes into account the spatial variation of frequency-size distribution, allowing the exploration of the possibility of a local preslip of a future nearby earthquake, is needed to improve this system.