Quantitative gas saturation estimation by frequency‐dependent amplitude‐versus‐offset analysis

Abstract

Seismic amplitudes contain important information which can be related to fluid saturation.The amplitude-versus-offset (AVO) analysis of seismic data based on Gassmann's theory and approximation of the Zoeppritz equations, has played an central role in reservoir characterization. However, this standard technique faces a long standing problem, its inability to distinguish between partial gas and "fizz-water" with little gas saturation. In this paper, we studied seismic dispersion and attenuation in partially saturated poroelastic media by using frequency-dependent rock physics model, through which the frequency-dependent AVO (FAVO) response is calculated as a function of porosity and water saturation. We propose a crossplotting of two attributes derived from FAVO response to differentiate partial gas saturation and "fizz-water" saturation. One of the attributes is a measure of "low frequency"or Gassmann, reflectivity, while the other is a measure of the "frequency-dependence" of reflectivity. This is in contrast to standard AVO attributes, where there is typically no such separation. A pragmatic FAVO inversion for rock and fluid properties is also established based on Bayesian theorem. Synthetic study is performed to explore the potential of the method to estimate gas saturation and porosity variations. An advantage of our work is that the method is in principle predictive, opening the way to further testing and calibration with field data. We believe that such work should guide and augment more theoretical studies of FAVO analysis.

Estimating gas saturation in a thin layer by using frequency‐dependent amplitude versus offset modelling

“…That is to say, its reciprocal actually plays the role of fluid mobility, which is the ratio of permeability to fluid viscosity. Wu et al (2014) has discussed the practical implementation of these equations by relating the effective bulk and shear moduli with reference P-wave velocity, S-wave velocity and density. We can then calculate the frequencydependent reflectivity from Zoeppritz equations generalized by Schoenberg and Protazio (1992).…”

“…The corresponding viscoelastic case has been studied by Borcherdt (2009). Wu et al (2014) calculated frequency-dependent reflectivity for an elastic layer overlying a dispersive and attenuative lower layer. The elastic behaviour of the upper layer allowed them to construct solutions in which the slowness vector in the upper medium was real, and only the vertical components of slowness in the lower layer were complex -a substantial simplification.…”

“…We show how this forward seismic model can act as a basis for a Bayesian inversion scheme that estimates gas saturation and layer thickness with suitable prior information. Our results from a thin layer study show that with sufficient prior geological knowledge, we can handle interfering reflections and the quantitative gas saturation estimation method by Wu et al (2014) can be effectively extended to the thin-layer case.…”

“…Since then, authors have analyzed similar problems using models which do show significant gas-related dispersion at seismic frequencies, and have concluded that frequency-dependent reflectivity could be potentially important Odebeatu et al 2006;Ren et al 2009;Innanen 2011). Wu et al (2014) proposed a method for inverting pre-stack reflection data for gas saturation based on frequency-dependent reflectivity and rock physics theory. The method involves forward modelling of the reflection coefficient using frequency-dependent rock physics theories, applying spectral decomposition and balancing to seismic data to obtain the spectral amplitudes of reflection coefficients at varying angles of incidence, and making comparison between the observed data and model responses to perform a Bayesian inversion for water saturation.…”

“…Despite the progress from Wu et al (2014), a number of limitations underlie the technique: there has to be only one interface in the reservoir and dispersion only occurs in the lower half space as the upper layer is assumed to be elastic. A key weakness of the single-interface assumption is that the results from spectral decomposition can be corrupted by closely spaced interfering reflections from a thin layer in the circumstance where the reservoir consists of multiple layers.…”