Generation of synthetic seismograms with layer reduction

Abstract

SUMMARY The surface motion in a plane multilayered half-space has been considered. Layer matrices of SH and P-SV motions are written as sums of matrices. Using these it is seen that when shear wave velocity of a layer, say the lth, is greater than the phase velocity, and layer thickness is large compared to wavelength, the surface motion can be written with the lth layer as the terminating half-space. Thus in computation of surface motion, the depth of the terminating half-space in a given structure depends on phase velocity and frequency. Further, if the source is below the terminating half-space, it has no contribution to surface motion at that phase velocity and frequency for which the terminating half-space has been obtained. This layer reduction formulation is applied in the generation of synthetic seismograms. As this generation requires computation over large ranges of phase velocity and frequency, the present method is important in avoiding excess computation time for unnecessary layers and in avoiding overflow due to the above mentioned Ith layer. The method is most useful in low phase velocity and high frequency ranges.