No abstract

No abstract

Long-period ground motions in plain and basin areas on land can cause large-scale, severe damage to structures and buildings and have been widely investigated for disaster prevention and mitigation. However, such motions in ocean-bottom areas are poorly studied because of their relative insignificance in uninhabited areas and the lack of ocean-bottom strong-motion data. Here, we report on evidence for the development of long-period (10–20 s) motions using deep ocean-bottom data. The waveforms and spectrograms demonstrate prolonged and amplified motions that are inconsistent with attenuation patterns of ground motions on land. Simulated waveforms reproducing observed ocean-bottom data demonstrate substantial contributions of thick low-velocity sediment layers to development of these motions. This development, which could affect magnitude estimates and finite fault slip modelling because of its critical period ranges on their estimations, may be common in the source areas of subduction earthquakes where thick, low-velocity sediment layers are present.

The seismoacoustic scattering due to an irregular fluid-solid interface (i.e. the ocean bottom) must be considered when modelling seismic wave propagation in oceanic regions. In this paper, we present an accurate method to treat this problem. The method is an extension of the reflection/transmission matrix approach that was developed to model scattering due to an irregular interface between two elastic solids. This approach uses the discrete wavenumber representation of the wavefield and the Rayleigh ansatz; the wavefield is decomposed into a set of plane waves, and we assume that the scattered wavefield due to an incident plane wave can be expanded in terms of only the waves moving away from the interface. The plane waves at both sides of the interface are then connected to each other based on this hypothesis and proper boundary conditions. The elements of the reflection/transmission matrices describe these connection coefficients.As the fluid-solid interface, the direct application of the above method is difficult because the tangential displacement at the interface is not necessarily continuous. To avoid this problem, we use pressure as the variable inside the fluid layer. The reflection/transmission matrices between the pressure wave (i.e. the acoustic wave) and the elastic waves are then developed by using the standard boundary conditions at the fluid-solid interface (i.e. normal stress continuity, zero tangential stress, and normal displacement continuity) and the Rayleigh ansatz.To demonstrate the validity and feasibility of our method, we compare the results of our method with those of other methods. First, we examine the reflection of plane elastic waves by a periodic sinusoidal interface for a single frequency. The interface profile is characterized by its periodicity or wavelength (D) and peak to trough height (h). The behaviour of the reflection calculated by our method shows almost complete agreement with that calculated by a boundary element method (BEM) up to a relatively steep interface whose height to wavelength ratio (h/D) is 0.5.Second, we calculate the synthetic time domain waveforms for a plane vertically incident P wave into basin-like fluid layers. We compare the waveforms calculated by the present method with those calculated by the finite difference method (FDM). In general, the waveforms calculated by the two methods match well. A quantitative study shows that the degree of agreement indicated by the RMS residuals between the waveforms of the two methods becomes better when we use a smaller grid interval in the FDM calculation. For the models treated here, the cases of 30 grid-points per wavelength results in RMS residuals less than about 10-15 per cent. This grid interval is smaller than that usually applied (more than 5-10 grid-points per wavelength). Since, in general, the accuracy of the FDM calculation becomes better as the grid interval decreases, these results indicate the validity of the present method.

We first incorporate the large effect of near-source heterogeneity on teleseismic body waveforms in the inversion of the slip distribution of the 2006 Java tsunami earthquake. We incorporate the effect by computing the response of an assumed "2.5D" model structure of the Java trench by a 2.5D finite-difference method. Based on a simulation of inversion, we suggest that intense smearing is possible when we apply 1D Green's functions in the analysis, and that it may obscure the slip pattern. In the inversion of real data, we confirm macroscopic features, such as a long duration (∼165 s) and a slow rupture velocity (∼1.25 km/s). The region of the initial rupture is found to be isolated from the eastern broad region in which we further identify a heterogeneous slip distribution. Most of these regions are likely to be at the sedimentary plate interface where the accreted sediment and the subducting plate are in contact. In particular, the nearly "isolated" feature of a shallow slip region suggests a possible faulting in the shallowest part of the sedimentary plate interface without being strongly enforced by the rupture propagated from the deeper part of the fault. Such heterogeneity suggests a highly variable frictional behavior at the sedimentary plate interface.