Abstract
We estimate static displacement and long-period pulse from strong motion records in the near fault region of the Mw7.0 2016 Kumamoto earthquake and compare them with previous prediction equations. In addition, we examine the relation between the observed data and the other parameters unused for the previous equations to aim for the improvement or development of prediction equations in the feature.
We derive velocity and displacement time history from acceleration time history of K-NET, KiK-net, JMA-95 type and local government strong motion records. Then we estimated the static displacement Dp, the period of long-period (>2 s) velocity pulse Tp and the PGV for the maximum Dp direction (Fling-P).
Tp of Fling-P component at Nishihara village with the distance is near 0 km is 2.6 s and the average of Tp at eight stations with the distance less than 15 km is 3.1 s. The average of Tp at 16 stations with the distance less than 30 km is 4.0 s. One cosine-shape pulse at the S wave potion is observed in the extremely near fault region. On the other hand the fling step is started between P wave and S wave portions at stations with the distance longer than 15 km and so the Tp becomes longer. Tp predicted from equations by Kamai et al. (2014) and Burks and Baker (2016) are 5.5 s and 3.5 s, respectively. These two equations were developed using data with the distance less than 30 km or more and modeled by only Mw. Kamai et al. used synthetic data for scenario earthquakes. Burks and Baker used both synthetic data and strong motion records. However no records of crustal earthquakes in Japan were used. Tp equations could be improved by using the distance as one of predictors and Tp obtained from the Kumamoto earthquake as data. Tp of vertical component is almost the same to Tp of Fling-P component.
PGV of Fling-P and vertical components agree with previous equations on the average. However PGV of Fling-P component at Nishihara village is 277 cm/s, which is larger than the average plus the standard deviation of the equation by Si and Midorikawa (1999). The PGV ratios of Fling-P component to the orthogonal component are 2 or 2.5 times at three stations with the distance less than 1 km and become smaller to unity as the distance is longer. PGV of vertical component at Nishihara village is 152 cm/s. This is larger than the average plus the standard deviation of the equation by Satoh (2008).
Dp of Fling-P component observed at Nishihara village is 154 cm. Predicted Dp by four previous equations in which the average slip is predicted using Mw are 80 to 90 cm at the distance of 0 km. On the other hand the predicted Dp in which the average slip is predicted using the fault area and Mw is consistent to Dp observed in the distance from 0 to 30 km. The Dp predicted by the best equation are 141 cm at a distance of 0 km and 56 cm at a distance of 10 km in the hanging-wall side. When the ratio of the distance along the strike direction from the fault edge is smaller, the observed Dp becomes smaller. This feature is consistent to a previous rupture shape model developed for slip distribution from point measurements of rupture displacement of many faults. This parameter would be a good predictor to improve the Dp equation. Dp of vertical component at Nishihara village is 179 cm. However, there were no equations of Dp and Tp for vertical component considering normal-slip.
