10:15 AM - 10:30 AM
[SSS08-12] Fault Displacement of the 2016 Kumamoto Earthquake deduced by DInSAR analysis
Keywords:The 2016 Kumamoto earthquake, Surface rupture, Fault displacement, DInSAR
1. Introduction
Differential SAR Interferometry (DInSAR) was fully utilized for the extraction of the surface ruptures associated with the 2016 Kumamoto earthquake (Fujiwara et al., 2016). The remote sensing technique will be a very effective tool to make the fault displacement database, because it can search source area without overlooking in a wide range. In this study, we measure each displacements of the surface ruptures of the Kumamoto earthquake using DInSAR technique and elucidate characteristics of the distribution.
2. Three dimensional ground deformation observed by DInSAR
At first three dimensional ground deformation associated with the earthquake was analyzed by two pairs of ALOS-2 PALSAR-2 data acquired before and after the earthquake (2016/3/7-2016/4/18, 2016/3/29-2016/4/26). The quasi-EW and quasi-UD components were obtained by combining LOS displacements of each pair based on DInSAR. The NS component was obtained by combining displacements in the satellite flight direction of each pair based on MAI processing. As a result, it is observed that remarkable subsidence in the northern part of the Futagawa fault zone and slight subsidence in the region between the western extension of the Futagawa fault and the Hinagu fault. In the horizontal component, asymmetric movements are observed across the Futagawa-Hinagu fault zone. Its north side moves to E ~ ENE direction while its south side moves to SW ~ S direction. This trend is consistent with the displacements observed by the neighboring GNSS stations. In addition, it is very similar to the distribution obtained by Kobayashi et al. (2017) using the pixel offset of SAR images. Accordingly it is considered that the ground deformation associated with the earthquake was properly captured in this study.
3. Measurement of fault displacement
Using the obtained three component of ground deformation data, the displacements of the 228 faults extracted by Fujiwara et al. (2016) were measured as follows: Transverse lines were set at 200 m intervals on the each fault trace, and the amount of fault displacement was measured on each transverse line. When non-interferometric areas occupy a large number on the same profile, they are excluded from measurement targets. Consequently, displacements of faults along the Futagawa fault zone are hardly obtained. On the other hand, a lot of displacement data were obtained in the area away from the fault zone. The vertical components are dominant at many points. Total slip is as small as 10 to 20 cm in the city area of Kumamoto. Meanwhile, it becomes relative large as 80 cm in the northwest part of Aso Caldera.
4. Spatial relation between fault displacements and the active faults
The amount of displacement of the surface rupture tends to decrease exponentially according to the distance from the known active faults. This tendency has also been obtained by the authors' investigation results on the 2011 Fukushima-ken Hamadori earthquake. In addition, it is similar to the results of Takao et al. (2013) which sorted the displacements of the secondary fault according to the distance from the primary fault of the 16 domestic earthquakes. Therefore, it is concluded that the amount of displacement of the surface rupture generally decreases with distance from the active fault. However, when Futagawa-Hinagu fault zone is considered as the primary fault, the resulting values at the northwest side of Aso Caldera are probably too large for that distance. Although it is necessary to investigate the origin of these surface ruptures, we think it is important to comprehensively examine the spatial relationship from active faults where cumulative displacement is observed for rational fault displacement prediction.
Differential SAR Interferometry (DInSAR) was fully utilized for the extraction of the surface ruptures associated with the 2016 Kumamoto earthquake (Fujiwara et al., 2016). The remote sensing technique will be a very effective tool to make the fault displacement database, because it can search source area without overlooking in a wide range. In this study, we measure each displacements of the surface ruptures of the Kumamoto earthquake using DInSAR technique and elucidate characteristics of the distribution.
2. Three dimensional ground deformation observed by DInSAR
At first three dimensional ground deformation associated with the earthquake was analyzed by two pairs of ALOS-2 PALSAR-2 data acquired before and after the earthquake (2016/3/7-2016/4/18, 2016/3/29-2016/4/26). The quasi-EW and quasi-UD components were obtained by combining LOS displacements of each pair based on DInSAR. The NS component was obtained by combining displacements in the satellite flight direction of each pair based on MAI processing. As a result, it is observed that remarkable subsidence in the northern part of the Futagawa fault zone and slight subsidence in the region between the western extension of the Futagawa fault and the Hinagu fault. In the horizontal component, asymmetric movements are observed across the Futagawa-Hinagu fault zone. Its north side moves to E ~ ENE direction while its south side moves to SW ~ S direction. This trend is consistent with the displacements observed by the neighboring GNSS stations. In addition, it is very similar to the distribution obtained by Kobayashi et al. (2017) using the pixel offset of SAR images. Accordingly it is considered that the ground deformation associated with the earthquake was properly captured in this study.
3. Measurement of fault displacement
Using the obtained three component of ground deformation data, the displacements of the 228 faults extracted by Fujiwara et al. (2016) were measured as follows: Transverse lines were set at 200 m intervals on the each fault trace, and the amount of fault displacement was measured on each transverse line. When non-interferometric areas occupy a large number on the same profile, they are excluded from measurement targets. Consequently, displacements of faults along the Futagawa fault zone are hardly obtained. On the other hand, a lot of displacement data were obtained in the area away from the fault zone. The vertical components are dominant at many points. Total slip is as small as 10 to 20 cm in the city area of Kumamoto. Meanwhile, it becomes relative large as 80 cm in the northwest part of Aso Caldera.
4. Spatial relation between fault displacements and the active faults
The amount of displacement of the surface rupture tends to decrease exponentially according to the distance from the known active faults. This tendency has also been obtained by the authors' investigation results on the 2011 Fukushima-ken Hamadori earthquake. In addition, it is similar to the results of Takao et al. (2013) which sorted the displacements of the secondary fault according to the distance from the primary fault of the 16 domestic earthquakes. Therefore, it is concluded that the amount of displacement of the surface rupture generally decreases with distance from the active fault. However, when Futagawa-Hinagu fault zone is considered as the primary fault, the resulting values at the northwest side of Aso Caldera are probably too large for that distance. Although it is necessary to investigate the origin of these surface ruptures, we think it is important to comprehensively examine the spatial relationship from active faults where cumulative displacement is observed for rational fault displacement prediction.