Postseismic deformation following the 2016 Kumamoto earthquake detected by ALOS-2/PALSAR-2

Abstract

I have been conducting a study of postseismic deformation following the 2016 Kumamoto earthquake using ALOS-2/PALSAR-2 acquired till 2018. I apply ionospheric correction to interferograms of ALOS-2/PALSAR-2. L-band SAR gives us high coherence enough to reveal surface deformation even in vegetated or mountainous area for pairs of images acquired more than 2 years. Postseismic deformation following the Kumamoto earthquake exceeds 10 cm during 2 years at some spots in and around Kumamoto city and Aso caldera. Westward motion of ~ 6 cm/year was dominant on the southeast side of the Hinagu fault, while westward shift was detected on both sides of the Futagawa fault. The area of latter deformation seems to have correlation with distribution of pyroclastic flow deposits. Significant uplift was found around the eastern Futagawa fault and on the southwestern frank of Aso caldera, whose rate reaches 4 cm/year. There are sharp changes across several coseismic surface ruptures such as Futagawa, Hinagu, and Idenokuchi faults. Rapid subsidence between Futagawa and Idenokuchi faults also found. It is confirmed that local subsidence continued along the Suizenji fault, which newly appeared during the mainshock in Kumamoto City. Subsidence with westward shift of up to 4 cm/year was also found in Aso caldera. Time constant of postseismic decay ranges from 1 month to 600 days at selected points, but that postseismic deformation during the first epochs or two is dominant at point in the Kumamoto Plain. This result suggests multiple source of deformation. Westward motion around the Hinagu fault may be explained with right lateral afterslip on the shallow part of this fault. Subsidence along the Suizenji fault can be attributed to normal faulting on dipping westward. Deformation around the Hinagu and Idenokuchi faults cannot be explained with right lateral afterslip of Futagawa fault, which requires other sources. Deformation in northern part of Aso caldera might be the result of right lateral afterslip on a possible buried fault.

Nationwide urban ground deformation monitoring in Japan using Sentinel-1 LiCSAR products and LiCSBAS

“…15c). Similar downward velocities were also observed by ALOS-2 (Kobayashi 2018;Hashimoto 2020). Because most of this region is covered with rice paddy fields ( Figure S14a), this subsidence signal may include the bias related to fading signals (see the Section 4.1).…”

“…The time constant of the exponential decay seemed to be shorter than that of P4311, and the subsidence almost stopped around 2018. The consistent temporal behavior using only the Sentinel-1 descending data and infrequent ALOS-2 data was also reported in Bischoff et al (2020) and Hashimoto (2020), respectively.…”

“…The Mw 7.0 Kumamoto earthquake occurred on April 16, 2016, and associated large coseismic deformation (> 2 m) and small displacement lineaments were observed around the Futagawa and Hinagu faults (e.g., Himematsu and Furuya 2016;Fujiwara et al 2016;Fig S14b). While postseismic deformation was detected by ALOS-2 InSAR (Fujiwara et al 2016;Kobayashi 2018;Hashimoto 2020), the detailed temporal evolution was not revealed due to infrequent observations (2-4 times/year in general). This study shows the spatially and temporally detailed postseismic deformation revealed by the Sentinel-1 LiCSAR products and LiCSBAS, which is essential for understanding the complex postseismic deformation mechanism ( Figs.…”

“…The area ID with triple digits consists of a double digits prefectural ID (01-47, e.g., 01 Hokkaido) and a single digit sub-number. As clear deformation has been detected outside of the DID in Hirosaki (ID: 022), Kanto (ID: 131), and Kumamoto (ID: 431) by past studies (e.g., Morishita et al 2010;Nonaka et al 2020;Hashimoto 2020), their target areas were expanded to measure complete deformation signals. All the defined areas except Naha (ID: 471) had at least one ascending and descending data set each, and some had two data sets due to overlapping areas between adjacent tracks ( Fig.…”

“…Almost no measurement points were obtained in the southeast side of the faults because the region is covered with dense vegetation. Although L-band ALOS-2 can retain sufficient coherence in these areas, the temporal sampling frequency is much lower and the detailed temporal evolution of the postseismic deformation cannot be revealed (Hashimoto 2020). The next-generation Lband SAR missions with high observation frequency such as ALOS-4 (Motohka et al 2019) or NASA-ISRO SAR (NISAR; Jet Propulsion Laboratory 2020) will help clarify the complex postseismic deformation, even in such densely vegetated regions.…”