Faulting processes have created large damage zones with complex structures in the field; however, estimating the width and geometry of such fault structures in the subsurface is challenging due to a lack of data. Seismic attributes (e.g., coherence and variance) from seismic surveys have been used for the characterization of faults, but most cases do not detail the effectiveness of this approach. By using forward modeling and the associated seismic attributes of variance, four fault models of idealized damage zones are characterized and the frequency effect is evaluated on the width estimation of fault damage zones in the subsurface. The main results indicate that (1) the general geometric pattern of damage zones could be identified by using simulated amplitude and seismic variance with main frequencies of 10, 25, and 40 Hz; (2) the estimated widths of damage zones at a low frequency of 10 Hz are larger (up to twofold) than those at frequencies of 25 and 40 Hz; for large damage zones (>400 m), the width is best estimated by a frequency of 25 Hz; and (3) scattering noise and diffraction around the fault are found in data at a high frequency of 40 Hz, which results in width overestimation of the damage zones by approximately 17%. The internal structures are difficult to distinguish as scattering noise and chaotic reflections dominate seismic signals. More factors that may influence the accuracy of damage zone width estimation via seismic attributes, include the bedding thickness, fracture density, and velocity. An in-depth understanding of this approach is useful in the application of seismic variance to characterize fault damage zones that may significantly control the fluid migration in the subsurface.

You do not currently have access to this article.