Detection of Natural Fractures from Observed Surface Seismic Data Based on a Linear-Slip Model

作者：Chen, Huaizhen^*; Zhang, Guangzhi

来源：Pure and Applied Geophysics, 2018, 175(8): 2769-2784.

DOI：10.1007/s00024-018-1848-3

• Fracture detection
• Azimuthal seismic data
• Bayesian inference
• Seismic inversion
• Linear-slip model

Natural fractures play an important role in migration of hydrocarbon fluids. Based on a rock physics effective model, the linear-slip model, which defines fracture parameters (fracture compliances) for quantitatively characterizing the effects of fractures on rock total compliance, we propose a method to detect natural fractures from observed seismic data via inversion for the fracture compliances. We first derive an approximate PP-wave reflection coefficient in terms of fracture compliances. Using the approximate reflection coefficient, we derive azimuthal elastic impedance as a function of fracture compliances. An inversion method to estimate fracture compliances from seismic data is presented based on a Bayesian framework and azimuthal elastic impedance, which is implemented in a two-step procedure: a least-squares inversion for azimuthal elastic impedance and an iterative inversion for fracture compliances. We apply the inversion method to synthetic and real data to verify its stability and reasonability. Synthetic tests confirm that the method can make a stable estimation of fracture compliances in the case of seismic data containing a moderate signal-to-noise ratio for Gaussian noise, and the test on real data reveals that reasonable fracture compliances are obtained using the proposed method.

• 出版日期2018-8
• 单位中国石油大学（北京）