11:05 AM - 11:20 AM
[MIS34-07] Rupture process of the 2016 Kumamoto earthquake based on waveform inversion with empirical Green's functions
Keywords:2016 Kumamoto earthquake , rupture process, empirical Green's funcion, waveform inversion, strong ground motion
Rupture process of the 2016 Kumamoto earthquake was estimated based on waveform inversion with empirical Green's functions. Records from two small events, namely, the April 15 0:50 event (MJ=4.2) and the April 15 15:27 event (MJ=4.2) were used as empirical Green's functions. The conventional least-squares linear waveform inversion (Hartzell and Heaton, 1983) was adopted.
A fault plane with a dimension of 40 km times 20 km was assumed, whose strike and dip angles were set to be 46 and 96 degrees, respectively. The fault was divided into 20 times 10 fault elements. The rupture front is assumed to start at the JMA hypocentral time and to propagate radially at a constant velocity of 2.5 km/s. The moment rate function at each fault element after passage of the rupture front was assumed to be a convolution of the moment rate function of the small event and an impulse train. The impulse train is composed of 12 impulses at equal time intervals of 0.25 s. The height of each impulse was determined through the inversion. Conventional corrections for the geometrical spreading and time shifts (Irikura, 1983) were applied to the empirical Green's functions to represent arrivals from each fault element. The shear wave velocity
in the source region was assumed to be 3.55km/s. Absolute time information for both the main shock and small event recordings was used.
The result indicates that both slip and slip-velocity were small in the west of the hypocenter. Between the hypocenter and KMMH16 (Mashiki), a large slip and slip-velocity region existed but it was restricted at the deeper part of the fault, indicating that the large amplitude ground motion in Mashiki was not a result of a forward directivity effect. A region of significantly large slip and slip-velocity existed about 20 to 25 km east of the hypocenter.
A fault plane with a dimension of 40 km times 20 km was assumed, whose strike and dip angles were set to be 46 and 96 degrees, respectively. The fault was divided into 20 times 10 fault elements. The rupture front is assumed to start at the JMA hypocentral time and to propagate radially at a constant velocity of 2.5 km/s. The moment rate function at each fault element after passage of the rupture front was assumed to be a convolution of the moment rate function of the small event and an impulse train. The impulse train is composed of 12 impulses at equal time intervals of 0.25 s. The height of each impulse was determined through the inversion. Conventional corrections for the geometrical spreading and time shifts (Irikura, 1983) were applied to the empirical Green's functions to represent arrivals from each fault element. The shear wave velocity
in the source region was assumed to be 3.55km/s. Absolute time information for both the main shock and small event recordings was used.
The result indicates that both slip and slip-velocity were small in the west of the hypocenter. Between the hypocenter and KMMH16 (Mashiki), a large slip and slip-velocity region existed but it was restricted at the deeper part of the fault, indicating that the large amplitude ground motion in Mashiki was not a result of a forward directivity effect. A region of significantly large slip and slip-velocity existed about 20 to 25 km east of the hypocenter.