17:15 〜 18:45
[U16-P04] 能登地域の歪み速度場に対する2011年東北沖地震の影響
キーワード：2024年能登半島地震、群発地震、流体の上昇、歪み速度場、2011年東北沖地震
The Mw 7.5 earthquake occurred in the Noto Peninsula on January 1, 2024. The earthquake swarms have been observed since November 2020 with the Mw 6.2 earthquake on May 5, 2023. The low seismic velocity (Nakajima, 2022) and the low electrical resistivity (Yoshimura et al., 2022, SGEPSS) imply the existence of fluid in the lower crust. Fluid can increase pore fluid pressure weakening fault strength, which is considered to have triggered the earthquake swarms. However, what caused the fluid migration has hardly been discussed. Therefore, we estimate strain-rate fields around the Noto Peninsula from GNSS data before and after the 2011 Tohoku-oki earthquake. We also discuss the cause of fluid migration before the earthquake swarms.
We estimate strain-rate fields using the method of Okazaki et al. (2021) based on Yabuki & Matsu'ura (1992), which used the framework of Bayesian inversion and took the smoothness of a velocity field as prior information (L2-regularization). We express a velocity field by the superposition of cubic B-spline functions and minimize the objective function composed of 2 terms: residuals between observed and estimated velocities and the L2-norm of second derivatives of the velocity field. The weight between the 2 terms is specified by a hyperparameter, which is determined by the ABIC minimization.
We use the daily coordinates of the F5 solution provided by the Geospatial Information Authority of Japan (GSI) as GNSS data. We select 2 analysis periods before the 2011 Tohoku-oki earthquake (2003-2006 and 2007-2010), and 2 analysis periods after the 2011 Tohoku-oki earthquake (2012-2015 and 2016-2019). In obtaining the average velocity during each period at each observation station, we remove annual and semiannual sinusoidal variations and coordinate offsets associated with large earthquakes and equipment maintenance.
The obtained strain-rate fields are significantly different between before and after the 2011 Tohoku-oki earthquake. First, the eastward extension is estimated around the Noto Peninsula only after the 2011 Tohoku-oki earthquake. Second, the dilatation rates decrease with time, although dilation still proceeds in Toyama Bay even in the period from 2016 to 2019. The temporal change of the strain-rate fields implies that the extension around the Noto Peninsula after the 2011 Tohoku-oki earthquake made upward fluid migration from depth easier.
For further discussion, we estimate the value of hydraulic diffusivity, assuming that the fluid migrated upward from the depth of 30 km (around the Moho) to the depth of 15 km (the maximum depth of earthquake swarms) in 4-8 years including the periods of 2012-2015 and 2016-2019. Using the equation introduced by Shapiro et al. (1997), the diffusivity of the fluid is estimated to be about 0.07-0.14 m2/s, which is equivalent to or smaller than the average value of the subsurface diffusivity, 0.1-10 m2/s (Talwani et al. 2007). This result supports the assumption that the fluid migrated from the Moho to the seismogenic depth over 4-8 years after the 2011 Tohoku-oki earthquake. This suggests that the 2011 Tohoku-oki earthquake may have caused upward fluid migration, which triggered the earthquake swarms around the Noto Peninsula after 2020.
We estimate strain-rate fields using the method of Okazaki et al. (2021) based on Yabuki & Matsu'ura (1992), which used the framework of Bayesian inversion and took the smoothness of a velocity field as prior information (L2-regularization). We express a velocity field by the superposition of cubic B-spline functions and minimize the objective function composed of 2 terms: residuals between observed and estimated velocities and the L2-norm of second derivatives of the velocity field. The weight between the 2 terms is specified by a hyperparameter, which is determined by the ABIC minimization.
We use the daily coordinates of the F5 solution provided by the Geospatial Information Authority of Japan (GSI) as GNSS data. We select 2 analysis periods before the 2011 Tohoku-oki earthquake (2003-2006 and 2007-2010), and 2 analysis periods after the 2011 Tohoku-oki earthquake (2012-2015 and 2016-2019). In obtaining the average velocity during each period at each observation station, we remove annual and semiannual sinusoidal variations and coordinate offsets associated with large earthquakes and equipment maintenance.
The obtained strain-rate fields are significantly different between before and after the 2011 Tohoku-oki earthquake. First, the eastward extension is estimated around the Noto Peninsula only after the 2011 Tohoku-oki earthquake. Second, the dilatation rates decrease with time, although dilation still proceeds in Toyama Bay even in the period from 2016 to 2019. The temporal change of the strain-rate fields implies that the extension around the Noto Peninsula after the 2011 Tohoku-oki earthquake made upward fluid migration from depth easier.
For further discussion, we estimate the value of hydraulic diffusivity, assuming that the fluid migrated upward from the depth of 30 km (around the Moho) to the depth of 15 km (the maximum depth of earthquake swarms) in 4-8 years including the periods of 2012-2015 and 2016-2019. Using the equation introduced by Shapiro et al. (1997), the diffusivity of the fluid is estimated to be about 0.07-0.14 m2/s, which is equivalent to or smaller than the average value of the subsurface diffusivity, 0.1-10 m2/s (Talwani et al. 2007). This result supports the assumption that the fluid migrated from the Moho to the seismogenic depth over 4-8 years after the 2011 Tohoku-oki earthquake. This suggests that the 2011 Tohoku-oki earthquake may have caused upward fluid migration, which triggered the earthquake swarms around the Noto Peninsula after 2020.