IEEE Access (Jan 2024)

Stratum P- and S-Wave <italic>Q</italic> Factors Estimation via Improved Spectral Envelope Matching Method

  • Chao Wang

DOI
https://doi.org/10.1109/ACCESS.2024.3520157
Journal volume & issue
Vol. 12
pp. 198079 – 198095

Abstract

Read online

Seismic wave attenuation in viscoelastic subsurface environments is quantified by the quality factor Q. Accurate estimation of the S-wave Q factor is crucial for improving resolution, imaging quality, interpretation accuracy, and predicting reservoir fluids and permeability in multicomponent seismic data. Although widely used, the spectral matching method has significant limitations in stability and accuracy when applied to noisy field data, stemming primarily from challenges in noise attenuation and transmission loss compensation. To address these limitations, we propose the spectral envelope matching (SENVM) method, which, unlike traditional spectral matching, focuses on spectral envelope differences, significantly reducing spectral noise interference, thereby improving stability and accuracy in matching outcomes. Based on Ganley’s theory, we propose an improved spectral envelope matching (ISENVM) method that builds on SENVM and incorporates transmission loss effects during estimation, thereby contributing to more accurate Q estimates. This method first extracts a seismic wavelet, followed by the centroid frequency shift (CFS) method to estimate formation Q values and construct an initial attenuation model. Then, the optimal formation Q values are determined using ISENVM. Applied to both noise-free and noisy synthetic vertical seismic profiling (VSP) down-going P- and S-wave data, both SENVM and ISENVM demonstrate superior anti-noise capabilities, providing more stable and accurate Q value estimates for P- and S-waves in thin layers compared to CFS. The successful application of ISENVM to field zero-offset three-component VSP data validates its effectiveness, contributing to a continuous model for the P- and S-wave Q factors. Finally, empirical formulas derived from regression analyses of P- and S-wave Q factors and velocities facilitate the estimation of S-wave Q factors and velocities for the Tuofutai block in the absence of reliable S-wave data.

Keywords