Reservoir characterization using perforation shots: anisotropy, attenuation, and uncertainty analysis

Anisotropic parameters, when considered in a microseismic processing, are typically inverted using perforation shot data and/or simultaneously inverted with microseismic event locations. Inverting using microseismic data alone usually leads to an underconstrained inverse problem that is highly dependent on prior/initial information. We carefully processed the waveforms from perforation shots, and picked P-, SH- and SV-wave arrival times to mitigate this issue. Because the perforation shot locations are known, the inversion is better constrained by reducing the number of model parameters while increasing the number of observations. We applied both maximum a posteriori estimation and randomized maximum likelihood methods for anisotropic parameter estimation, uncertainty quantification and trade-off analysis. Results verified the stability of the inversion and revealed the uncertainty and trade-offs among model parameter. In addition, attenuation is generally not considered in microseismic modelling and processing. Our study found that hydraulic stimulation may lead strong increases in seismic attenuation to reservoirs. The attenuation can dramatically change waveform characteristics and cause velocity dispersion. Thus, sonic logs, which are acquired at frequencies much higher than seismic data frequency should not be used directly for data processing in hydraulic stimulated zones.