11:00 〜 11:15
[SSS03-07] Investigation of a possible relationship between crustal heat flow and aftershock parameters of inland sequences occurred in Japan after the year 2000
キーワード：seismicity, heat flow, aftershocks
Introduction.
The correlation of seismicity parameters with other geophysical parameters is important for understanding the physics of earthquake occurrence and for earthquake hazard mitigation. Previous results obtained for Southern California indicate an inverse correlation between the aftershock productivity, Ko, of the Omori-Utsu law, and the α-value (a measure of the efficiency of a mainshock in generating aftershock activity relative to its magnitude) of the ETAS model with the crustal heat flow.
Data and method.
The aim of this study is to search for such possible correlations for Japan. We used the Japan Meteorological Agency (JMA) earthquake catalog to investigate the relationship between heat flow and aftershock parameters for 18 crustal, inland earthquake sequences, with mainshocks having magnitudes equal to 5.5 or greater that occurred in Japan since 2000 and were shallower than 20 km depth.We have also eliminated those events that were preceded by significant aftershock activity from nearby previous large events (i.e., visually confirming that the selected events are not aftershocks of a previous earthquake). The heat flow data has been measured in Hi-net borehole stations, belonging to the National Research Institute for Earth Science and Disaster Resilience (NIED)(Matsumoto, 2007). The aftershock datasets were created for each of the 18 earthquakes by imposing several space-time and magnitude conditions. The time-window for aftershock selection was set to 100 days and the spatial window was estimated based on the magnitude of the mainshock. The parameter Ko has been estimated for sequences of aftershocks where the minimum magnitude was chosen four magnitude units below that of the mainshock magnitude, while for the ETAS model the estimation of parameters (including the α-value) has been done for datasets with a minimum magnitude M = 2.0. For all estimations we checked that the minimum magnitude is equal to or above the magnitude of completeness of the data. A weighted average heat flow value was calculated for the area of each aftershock sequence.
Results and Discussion.
The α-values for the Japanese earthquakes estimated in this study had values in the range 1.0~2.0. On the other hand, Enescu et al (2009), reported α-values in Southern California in the range 1.5~3.5. In addition, in contrast to the previous results reported for Southern California, our findings indicate no significant correlations between the crustal heat flow and the aftershock parameters for the analyzed sequences in Japan. These findings may be related to differences in the tectonic conditions between Japan and Southern California. The α-values tend to be lower in Japan than in Southern California, which may indicate that the seismicity in Japan has more pronounced swarm-like characteristics. The productivity parameter, Ko, for several aftershock sequences occurred in northeast Japan, before the 2011 M9.0 Tohoku-oki earthquake, were relatively high, which might be related with their faulting style (reverse fault mainshocks), complexity of faulting, or a combination of these factors.
The correlation of seismicity parameters with other geophysical parameters is important for understanding the physics of earthquake occurrence and for earthquake hazard mitigation. Previous results obtained for Southern California indicate an inverse correlation between the aftershock productivity, Ko, of the Omori-Utsu law, and the α-value (a measure of the efficiency of a mainshock in generating aftershock activity relative to its magnitude) of the ETAS model with the crustal heat flow.
Data and method.
The aim of this study is to search for such possible correlations for Japan. We used the Japan Meteorological Agency (JMA) earthquake catalog to investigate the relationship between heat flow and aftershock parameters for 18 crustal, inland earthquake sequences, with mainshocks having magnitudes equal to 5.5 or greater that occurred in Japan since 2000 and were shallower than 20 km depth.We have also eliminated those events that were preceded by significant aftershock activity from nearby previous large events (i.e., visually confirming that the selected events are not aftershocks of a previous earthquake). The heat flow data has been measured in Hi-net borehole stations, belonging to the National Research Institute for Earth Science and Disaster Resilience (NIED)(Matsumoto, 2007). The aftershock datasets were created for each of the 18 earthquakes by imposing several space-time and magnitude conditions. The time-window for aftershock selection was set to 100 days and the spatial window was estimated based on the magnitude of the mainshock. The parameter Ko has been estimated for sequences of aftershocks where the minimum magnitude was chosen four magnitude units below that of the mainshock magnitude, while for the ETAS model the estimation of parameters (including the α-value) has been done for datasets with a minimum magnitude M = 2.0. For all estimations we checked that the minimum magnitude is equal to or above the magnitude of completeness of the data. A weighted average heat flow value was calculated for the area of each aftershock sequence.
Results and Discussion.
The α-values for the Japanese earthquakes estimated in this study had values in the range 1.0~2.0. On the other hand, Enescu et al (2009), reported α-values in Southern California in the range 1.5~3.5. In addition, in contrast to the previous results reported for Southern California, our findings indicate no significant correlations between the crustal heat flow and the aftershock parameters for the analyzed sequences in Japan. These findings may be related to differences in the tectonic conditions between Japan and Southern California. The α-values tend to be lower in Japan than in Southern California, which may indicate that the seismicity in Japan has more pronounced swarm-like characteristics. The productivity parameter, Ko, for several aftershock sequences occurred in northeast Japan, before the 2011 M9.0 Tohoku-oki earthquake, were relatively high, which might be related with their faulting style (reverse fault mainshocks), complexity of faulting, or a combination of these factors.