5:15 PM - 6:45 PM
[SSS11-P12] Interrelation between fault parameters in the active fault database of Japan by AIST and along-strike slip rate distribution
Keywords:Active fault, Active Fault Database inJapan, Slip rate destribution, Fault displacement
In the event of crustal earthquakes with magnitudes around 7 or greater, surface faulting leads to substantial damage to buildings due to fault displacement, particularly above the surface rupture (e.g., Hisada et al., 2020). Accurate prediction of fault displacements and distributions is crucial for disaster preparedness. Understanding seismic displacement distribution along anticipated faults is essential in addition to identifying fault positions through methods like aerial photointerpretation. Recent notable studies have highlighted the heterogeneous distribution of co-seismic displacements and long-term cumulative displacements along active faults, often forming triangular-shaped profiles (e.g., Manigetthi et al., 2005).
While some studies in Japan have explored scaling laws for earthquake faults, comprehensive analyses of various parameters associated with active faults are scarce, as are studies examining the shape of these slip profiles of active faults nationwide. However, the Active Fault Database of Japan (AFDB) of the National Institute of Advanced Industrial Science and Technology (AIST) compiles information on active faults distributed throughout Japan in one accessible platform.
This study analyzed and compiled data from the AFDB using text mining and other methods. The initial focus was on representative values, which are activity parameters calculated for each active segment, revealing no strong correlations among slip rate, slip per event, and fault length. Nonetheless, a trend reflecting the scaling relationship shown by Awata (1999) around the regression line between fault length and slip per event was observed. Of these, the distribution based on actual measurements after removing data presumed to be calculated by empirical formulas showed a regression line with a more gradual slope compared to before removal.
Subsequently, the shape of the slip rates profile of active segments throughout Japan was examined using the compiled data. Two distribution types were created: (1) reflecting points with a large slip rate and (2) reflecting the nationwide average distribution profile using normalized slip rates.
(1) In the case of vertical displacement, the slip rate distribution showed a triangular shape with a maximum slip rate of 0.3 [m/kyr] at 41% of the total segment length from the edge of the active segment. Therefore, the points with the largest slip rate tended to be distributed near the center of the active segment. In the case of horizontal displacement, the slip rate reached a maximum of 1.3 [m/kyr] at 11% of the total segment length from the end of the active segment, and a bimodal shape with a near maximum value of 1.2 [m/kyr] at 44%, and a linear decrease from the center to the other edge. Therefore, the points with the largest slip rate were distributed near the edge and the center of the active segment.
(2) In the case of vertical displacement, the normalized slip rate distribution showed that the normalized slip rate reached a maximum of 0.3 at 30% of the total segment length from the edge and maintained a large value towards the center before linearly decreasing. This suggests that the national average shape of the slip rates profile is biased in the case of vertical displacement. In the case of horizontal displacement, a triangular profile with a maximum normalized slip rate of 0.34 was observed at 12% of the total segment length from the edge of the active segment. Therefore, in the case of horizontal displacement, many of the active segments had their maximum values near the edge, resulting in a triangular profile with a large bias as revealed by Manigetthi et al. (2005).
However, it is crucial to note that the data used in this study heavily rely on the number of surveys conducted, rendering the analysis insufficient for segments lacking numerical information or with few surveys. Therefore, future investigations are necessary to increase data accuracy for such faults.
While some studies in Japan have explored scaling laws for earthquake faults, comprehensive analyses of various parameters associated with active faults are scarce, as are studies examining the shape of these slip profiles of active faults nationwide. However, the Active Fault Database of Japan (AFDB) of the National Institute of Advanced Industrial Science and Technology (AIST) compiles information on active faults distributed throughout Japan in one accessible platform.
This study analyzed and compiled data from the AFDB using text mining and other methods. The initial focus was on representative values, which are activity parameters calculated for each active segment, revealing no strong correlations among slip rate, slip per event, and fault length. Nonetheless, a trend reflecting the scaling relationship shown by Awata (1999) around the regression line between fault length and slip per event was observed. Of these, the distribution based on actual measurements after removing data presumed to be calculated by empirical formulas showed a regression line with a more gradual slope compared to before removal.
Subsequently, the shape of the slip rates profile of active segments throughout Japan was examined using the compiled data. Two distribution types were created: (1) reflecting points with a large slip rate and (2) reflecting the nationwide average distribution profile using normalized slip rates.
(1) In the case of vertical displacement, the slip rate distribution showed a triangular shape with a maximum slip rate of 0.3 [m/kyr] at 41% of the total segment length from the edge of the active segment. Therefore, the points with the largest slip rate tended to be distributed near the center of the active segment. In the case of horizontal displacement, the slip rate reached a maximum of 1.3 [m/kyr] at 11% of the total segment length from the end of the active segment, and a bimodal shape with a near maximum value of 1.2 [m/kyr] at 44%, and a linear decrease from the center to the other edge. Therefore, the points with the largest slip rate were distributed near the edge and the center of the active segment.
(2) In the case of vertical displacement, the normalized slip rate distribution showed that the normalized slip rate reached a maximum of 0.3 at 30% of the total segment length from the edge and maintained a large value towards the center before linearly decreasing. This suggests that the national average shape of the slip rates profile is biased in the case of vertical displacement. In the case of horizontal displacement, a triangular profile with a maximum normalized slip rate of 0.34 was observed at 12% of the total segment length from the edge of the active segment. Therefore, in the case of horizontal displacement, many of the active segments had their maximum values near the edge, resulting in a triangular profile with a large bias as revealed by Manigetthi et al. (2005).
However, it is crucial to note that the data used in this study heavily rely on the number of surveys conducted, rendering the analysis insufficient for segments lacking numerical information or with few surveys. Therefore, future investigations are necessary to increase data accuracy for such faults.