9:30 AM - 9:45 AM
[SCG54-03] Temporal evolution of seafloor crustal deformation obtained by the 10-year-long GNSS-A observations after the 2011 Tohoku-oki earthquake
Keywords:GNSS-A, Seafloor geodetic observation, Postseismic deformation
On March 11, 2011, the Tohoku-oki earthquake occurred in the western side of the Japan Trench. Seafloor geodetic observation using the GNSS-A technique detected the trench-ward seafloor movements of several tens of meters (Sato et al., 2011 Science; Kido et al., 2011 GRL). Differential bathymetric measurements revealed that the coseismic rupture had reached the trench axis in the off-Miyagi region (Fujiwara et al., 2011 Science), where the seafloor geodetic network in 2011 had no resolution. Additionally, tsunami data suggested that the shallow coseismic rupture had spread in wider area from the off-Iwate to off-Fukushima regions than the geodetically derived distributions (Satake et al., 2013 BSSA).
The stress induced by the coseismic slip causes transitional rock deformation following the mainshock. In the time scale of months to years, afterslip and viscoelastic relaxation are dominant in the postseismic crustal deformation (e.g., Wang et al., 2012 Nature). Whereas the afterslip consistently causes trench-ward surface motion, the viscoelastic relaxation drives land-ward and trench-ward movements above the source region and its downdip side, respectively. The viscoelastic relaxation additionally causes significant subsidence on and around the downdip edge of the main rupture (e.g., Luo and Wang, 2021 Nat. Geo.).
Because both the afterslip and viscoelastic relaxation similarly cause the trench-ward motion in the onshore region, it is difficult to evaluate the contributions of these processes. In contrast, the seafloor geodetic observation is capable of extracting the dominant source rather than the terrestrial observations. The landward and downward movements were detected at the sites above the main rupture using the three-year-long GNSS-A seafloor observations, which provided the definitive evidence for the dominance of viscoelastic relaxation (e.g., Watanabe et al., 2014 GRL, Sun et al., 2014 Nature). The spatial distribution of GNSS-A observation was improved by the Tohoku University in mid-2012 (Kido et al., 2015 GENAH), which lead to detecting more detailed deformation pattern in and around the whole source region (Tomita et al., 2015 GRL; Tomita et al., 2017 Sci. Adv.; Honsho et al., 2019 JGR). Many researchers had proposed the postseismic models (e.g., Sun et al., 2014 Nature; Sun and Wang, 2014 JGR; Iinuma et al., 2016 Nat. Commun.; Freed et al., 2017 EPSL), and consensus has reached that the viscoelastic relaxation was dominant in the main rupture area and that the afterslip less likely occurred in the downdip side of main rupture in the off-Miyagi region. On the other hand, there remain difficulties in decomposing the postseismic deformation sources in the northern and southern regions, because the geodetic data including the early postseismic stage are insufficient. Moreover, the lower resolution of the shallow coseismic slip in these areas increased uncertainty in the viscoelastic relaxation modeling. Although many postseismic models used the geodetically-constrained coseismic model (e.g., Iinuma et al., 2012 JGR) as an input, tsunami data indicated wider shallow slip in the north-south extensions which will significantly affect the viscoelastic deformation in the marginal areas.
Based on the decade of GNSS-A data after the event, we investigated the temporal evolution of postseismic crustal deformation in this study especially in the northern and southern regions. Our results indicated (1) that the afterslip occurred on the rupture edges at depths near the epicenter, (2) that the northern afterslip had almost decayed in the first three years, and (3) that the shallow coseismic slip had taken place in the off-Fukushima region. We will present the decadal GNSS-A results and discuss the postseismic behaviors in the offshore region.
The stress induced by the coseismic slip causes transitional rock deformation following the mainshock. In the time scale of months to years, afterslip and viscoelastic relaxation are dominant in the postseismic crustal deformation (e.g., Wang et al., 2012 Nature). Whereas the afterslip consistently causes trench-ward surface motion, the viscoelastic relaxation drives land-ward and trench-ward movements above the source region and its downdip side, respectively. The viscoelastic relaxation additionally causes significant subsidence on and around the downdip edge of the main rupture (e.g., Luo and Wang, 2021 Nat. Geo.).
Because both the afterslip and viscoelastic relaxation similarly cause the trench-ward motion in the onshore region, it is difficult to evaluate the contributions of these processes. In contrast, the seafloor geodetic observation is capable of extracting the dominant source rather than the terrestrial observations. The landward and downward movements were detected at the sites above the main rupture using the three-year-long GNSS-A seafloor observations, which provided the definitive evidence for the dominance of viscoelastic relaxation (e.g., Watanabe et al., 2014 GRL, Sun et al., 2014 Nature). The spatial distribution of GNSS-A observation was improved by the Tohoku University in mid-2012 (Kido et al., 2015 GENAH), which lead to detecting more detailed deformation pattern in and around the whole source region (Tomita et al., 2015 GRL; Tomita et al., 2017 Sci. Adv.; Honsho et al., 2019 JGR). Many researchers had proposed the postseismic models (e.g., Sun et al., 2014 Nature; Sun and Wang, 2014 JGR; Iinuma et al., 2016 Nat. Commun.; Freed et al., 2017 EPSL), and consensus has reached that the viscoelastic relaxation was dominant in the main rupture area and that the afterslip less likely occurred in the downdip side of main rupture in the off-Miyagi region. On the other hand, there remain difficulties in decomposing the postseismic deformation sources in the northern and southern regions, because the geodetic data including the early postseismic stage are insufficient. Moreover, the lower resolution of the shallow coseismic slip in these areas increased uncertainty in the viscoelastic relaxation modeling. Although many postseismic models used the geodetically-constrained coseismic model (e.g., Iinuma et al., 2012 JGR) as an input, tsunami data indicated wider shallow slip in the north-south extensions which will significantly affect the viscoelastic deformation in the marginal areas.
Based on the decade of GNSS-A data after the event, we investigated the temporal evolution of postseismic crustal deformation in this study especially in the northern and southern regions. Our results indicated (1) that the afterslip occurred on the rupture edges at depths near the epicenter, (2) that the northern afterslip had almost decayed in the first three years, and (3) that the shallow coseismic slip had taken place in the off-Fukushima region. We will present the decadal GNSS-A results and discuss the postseismic behaviors in the offshore region.