Mechanisms and Models of Seismic Attenuation
MetadataShow full item record
Seismic attenuation is a subject of great interest for both industry and academia. In exploration seismology, wave attenuation must be well understood for interpreting seismic data and laboratory experiments with rocks, and improving the quality and resolution of reflection imaging of the subsurface. To achieve such understanding, mechanisms of seismic attenuation and the associated physical models need to be studied in detail. This dissertation focuses on analyzing several attenuation mechanisms and building first-principle mathematical models for them. The effects of seismic attenuation can be broadly subdivided into two groups: 1) caused by inelasticity of the material and 2) caused by small-scale elastic structures of the material or subsurface. From the first of these groups, I study solid viscosity and internal friction due to squirt flows and wave-induced fluid flows (WIFF) at different scales. This approach is based on a new rheological law called the General Linear Solid (GLS) and recently developed to describe macroscopic inelastic effects in multiphase solids. The GLS is a model composed by time/frequency independent parameters and based on Lagrangian continuum mechanics. By utilizing the GLS framework, I extend the well known-model called the Standard Linear Solid (SLS) to include internal inertial forces, which explains the primary wave and reveals additional highly diffusive wave modes. I also use the GLS to model P-waves with squirt flow dissipation by different configurations of the density, moduli, drag and solid viscosity matrices. Seismic wave attenuation may not only be caused by inelastic properties but also by elastic processes such as reflectivity and scattering. I examine two types of such effects of the elastic structure of the material. First, in a laboratory experiment with several rock types, there is a modest influence of sample size on the measured level of attenuation and modulus dispersion. Second, in a field experiment aimed at measuring Q from seismic reflectivity, the effect of elastic layering can be extremely strong and even completely equivalent to that of the Q. An important general observation from this study is that amplitude decays and phase delays measured from reflection seismic data can always be interpreted as either caused by inelasticity or by small-scale elastic structures. An important complementary goal of studying the mechanisms and effects of seismic attenuation consists in correcting for its effects in seismic records and increasing the resolution of seismic images. In this dissertation, I briefly consider attenuation-correction techniques and develop a novel method for such correction by using time-domain deconvolution. Synthetic and field data are used to illustrate and test the performance of this approach.
DegreeDoctor of Philosophy (Ph.D.)
SupervisorMorozov, Igor B.
CommitteePan, Yuanming; Shevyakov, Alexey F.; Merriam, Jim; Butler, Samuel
Copyright DateJune 2017
General Linear Solid