11:00 AM - 1:00 PM
[SSS07-P24] What controls the expansion rate of the aftershock area? An example of magnitude-7 class earthquakes in Japan
Keywords:Seismicity, Aftershock area, Differential stress
The area of aftershocks following a major earthquake tends to expand in proportion to the logarithm of the time elapsed since the mainshock (e.g., Peng and Zhao, 2009; Ross et al., 2017). However, it is not well understood what factors control the expansion rate of the aftershock area. In this study, we compare the expansion rates of aftershock areas for multiple mainshocks of similar magnitude in Japan.
Using the earthquake catalog data provided by Japan Meteorological Agency, we define earthquakes with a magnitude of 7.0-7.4 since 2000 as mainshocks and earthquakes with a magnitude 3 or greater that occurred near the hypocenter within 100 days of the mainshocks as aftershocks. For evaluating the expansion of the aftershock area, we use a 3-D distance from the hypocenter of the mainshocks. To quantify the front of the expansion of aftershock areas, we focus on the slope on a graph where the horizontal axis is the logarithmic time and the vertical axis is the 3-D distance from the mainshock. To automatically estimate this slope from the data, we develop an algorithm that consists of the following steps: partitioning the time window of aftershock occurrence, removing outliers, determining the farthest point in each time window, and regressing the farthest point against the logarithmic time.
We find a strong negative correlation between the slope (representing the expansion rate of the aftershock area) and the b-value of the Gutenberg-Richter law, on the other hand, no correlation with the depth and magnitude of the mainshocks. There is also no spatial trend of the slope. Together with the existing empirical relation (Scholz, 2015), we obtain a new equation to infer differential stress from the expansion rate of the aftershock area. The equation is mechanically reasonable in that the aftershock area expands faster with higher differential stress.
Using the earthquake catalog data provided by Japan Meteorological Agency, we define earthquakes with a magnitude of 7.0-7.4 since 2000 as mainshocks and earthquakes with a magnitude 3 or greater that occurred near the hypocenter within 100 days of the mainshocks as aftershocks. For evaluating the expansion of the aftershock area, we use a 3-D distance from the hypocenter of the mainshocks. To quantify the front of the expansion of aftershock areas, we focus on the slope on a graph where the horizontal axis is the logarithmic time and the vertical axis is the 3-D distance from the mainshock. To automatically estimate this slope from the data, we develop an algorithm that consists of the following steps: partitioning the time window of aftershock occurrence, removing outliers, determining the farthest point in each time window, and regressing the farthest point against the logarithmic time.
We find a strong negative correlation between the slope (representing the expansion rate of the aftershock area) and the b-value of the Gutenberg-Richter law, on the other hand, no correlation with the depth and magnitude of the mainshocks. There is also no spatial trend of the slope. Together with the existing empirical relation (Scholz, 2015), we obtain a new equation to infer differential stress from the expansion rate of the aftershock area. The equation is mechanically reasonable in that the aftershock area expands faster with higher differential stress.