2019 AAPG Annual Convention and Exhibition:

Datapages, Inc.Print this page

Seismic Azimuthal Anisotropy Analysis Applied to Natural Fracture Intensity and Azimuth Prediction: Barnett Shale Example


Studying seismic responses of velocity and amplitude on wide/full azimuth seismic data is now common for unconventional reservoir characterization. In general, azimuth variation with velocity (VVAz) and azimuth variation with amplitude (AVAz) are two major tools for azimuthal anisotropy analysis. To accelerate azimuthal anisotropy analysis, a new attribute-based AVAz workflow was employed and validated using a Barnett Shale wide azimuth data. The seismic data is migrated into eight azimuths. Prestack structure-oriented filtering is applied to the seismic amplitude data for noise reduction. Spectral peak magnitude, envelope, and P-wave impedance inversion were computed on each azimuthally-limited seismic volume. Then, 2D-flattened computed attributes and seismic amplitude along picked horizons were extracted and input into the AVAz workflow. The results indicate intensity, orientation, and confidence of azimuthal anisotropy effects on seismic velocity and amplitude which can be referred to smaller scale vertical natural fractures.

All the analysis results consistently reveal four zones of high anisotropy intensity with the corresponding azimuth throughout the entire Barnett formation that can be tied to either the regional structures or older local stress field. To further correlate the outcome with regional structural features and natural fractures, a regional production map was projected on the anisotropy map and confirms the anisotropy interpretation result that regional anisotropy is mainly caused by the vertical sealed fractures and the sealed fracture network is inhibiting production in the study area according to the literature and field observations. The result indicates that horizon-based azimuthal anisotropy analysis effectively avoids the seismic migration error and can be utilized on the fracture and regional stress field prediction.