Premium
Validity of the long‐wave approximation in periodically layered media
Author(s) -
Arntsen Borge
Publication year - 2007
Publication title -
geophysical prospecting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.735
H-Index - 79
eISSN - 1365-2478
pISSN - 0016-8025
DOI - 10.1111/j.1365-2478.2006.00576.x
Subject(s) - wavelength , stack (abstract data type) , phase velocity , reflection coefficient , anisotropy , optics , reflection (computer programming) , approximation error , layering , phase (matter) , slab , geology , mineralogy , mathematical analysis , physics , geophysics , mathematics , botany , quantum mechanics , computer science , biology , programming language
In seismic modelling, a stack of thin layers is often replaced by an effective equivalent anisotropic homogeneous slab. For waves with finite wavelength, this is an approximation, and the error thus introduced can be quantified by considering the relative error in the phase velocity between the layer stack and the effective medium. For periodic layering, the relative phase‐velocity error can be expressed in closed form as a function of wavelength, reflection coefficients and layer thicknesses. By comparing the relative phase‐velocity error with laboratory measurements and numerical simulations, we find that the difference in seismic response between a periodic layer stack and an equivalent effective medium depends not only on wavelength, but it also depends significantly on reflection coefficients and the ratio between layer thicknesses. For a 1% relative error in the phase velocity, and if all layers have the same thickness measured in vertical traveltime, we find that the wavelength must be larger than approximately three times the layer period for a reflection coefficient of 0.1, but this increases to 13 times the layer period for a reflection coefficient of 0.9, which is highly unrealistic in a geological setting.