3:30 PM - 3:45 PM
[SSS05-07] New stable slip on the plate in a wide area after 2015 detected by postseismic deformation model of 2011 Tohoku-Oki Earthquake
Keywords:the 2011 off the Pacific coast of Tohoku Earthquake, crustal deformation, postseismic deformation, stable slip, GNSS time series, predicting model
The time series of the postseismic deformation of the 2011 off the Pacific coast of Tohoku earthquake (Tohoku-Oki earthquake) is fitted by a function to predict the future changes over a wide area. In this presentation, we validate the model of the postseismic deformation using data up to December 2020, and find that a new stable slip on the plate has occurred since 2015 in the residuals between the observed data and the model.
Method
It has been shown by Tobita (2016) that a mixed model of two logarithmic and one exponential functions, represented by the following equation, can be used to predict with high accuracy the spatial distribution of postseismic deformation:
D(t) = a ln(1+ t/b) + c + d ln(1+ t/e) - f exp(– t/g) + Vt
where D(t) is each component of the postseismic deformation, t is days after the earthquake, ln is the natural logarithm, b, e, and g are the relaxation time constants of logarithmic or exponential functions common to all stations, and V is the steady-state velocity at each station. The time constants were determined for stations, Miyako, Yamoto, Minase, and Choshi, which show representative variations, as used in Tobita (2016); for the other stations, the determined time constants were given in common, and the amplitude coefficients were determined for each station and component by the least-squares method.
Results
Fig. 1 shows the observed values of the east-west component at two GEONET stations, the predicted values based on the 3.9-year fitting (until February 2015), and their residuals. The predicted values agreed with the observed values to within 1 cm until February 2015, but then showed a different trend after earthquakes in Sanriku-Oki (M6.9) in February 2015 and Miyagi-Oki (M6.8) in May 2015. This is not only shown in the residuals, but also the time constant g of the exponent in the above prediction model becomes computationally saturated if we include 2015 and beyond in the fitting period, indicating that another event occurred in 2015 different from the postseismic deformation of the Tohoku-Oki earthquake.
Obtaining a slip model from residuals
The systematic changes in residual from the prediction model after 2015 occurred simultaneously over a wide area from Hokkaido to the Chubu region, and each continued at an almost constant rate. In order to investigate the cause of this phenomenon, a slip velocity model on the plate surface was calculated as shown in Fig. 2, which shows a few centimeters of slip per year.
The location of this slip can be roughly divided into two areas: Sanriku-Miyagi-Oki and Urakawa-Oki. The Sanriku-Miyagi-Oki area is almost coincident with the afterslip region that began immediately after the Tohoku-Oki earthquake. In contrast, no significant slip was observed in the Urakawa offshore area before 2015, and it was newly added after 2015.
The location and timing of the systematic residuals coincide with those of the Sanriku-Oki earthquake (M6.9) in February 2015 and the Miyagi-Oki earthquake (M6.8) in May 2015. This suggests that the onset of the slip and the occurrence of these medium-sized earthquakes are somehow related.
Conclusion
The prediction model by Tobita (2016) is able to predict the postseismic deformation of the Tohoku-Oki earthquake with an accuracy of a few centimeters even after 9 years. The residuals obtained from the prediction model are very effective in detecting the occurrence of other events that are not the postseismic deformation (e.g. Ozawa et al. 2016), and are expected to be used in various fields.
References
Ozawa S, Tobita M and Yarai H (2016) Earth Planets Space. https://doi.org/10.1186/s40623-016-0430-4.
Tobita M (2016) Earth Planets Space. https://doi.org/10.1186/s40623-016-0422-4.