Full Waveform Matching Based Microseismic Focal Mechanism Inversion And Its Application

Unconventional gas resources include tight sand gas, coalbed methane, and shale gas. They are playing an increasingly important role in supplying low carbon fuel for a growing global energy demand. Horizontal well drilling and hydraulic fracturing are the two key technologies in developing these low permeability reservoirs. Microearthquakes occur during fracking because of the stress perturbations and fracturing fluid leakage resulting from the hydraulic fracturing. Microseismic monitoring provides a way to image the overall geometry of the fractures and assess the stimulated reservoir volume. The main goal of microseismic monitoring is accurately locating induced microseismic events and determining their focal mechanisms to better delineate fracture distribution and fracturing mechanisms. Accurate microseismic maps and reliable source mechanism estimates not only reveal important information about the fracturing process, but also allow fracture characterization away from the wellbore, providing a prior information and critical constraints for building fractured reservoir models.At present, moment tensor inversion generally relies on the P wave first motion polarity and/or the S/P amplitude ratio. Due to microearthquake’s small magnitude, the released energy is limited and thus, only a few stations can record the waveforms. Conventional P wave first motion polarity and S/P amplitude ratio method cannot get the focal mechanism of small events accurately. Thus we use the full waveform matching based moment tensor inversion method in this study, through the best match of synthetic seismogram and the actual earthquake records to determine the focal mechanism solution. The objective function of full waveform matching method contains not only the waveform amplitude and phase information, but also the P wave initial motion and S/P amplitude ratio information. The method constrains the inversion by fully using the waveform information and reduces the uncertainty of inversion result by only using part of the waveform data.First of all, in order to verify the robustness of waveform matching method, we use different velocity models and seismic source parameters for the synthetic data test. Two types of model test results show that the waveform matching method is stable and reliable, which can be used in the focal mechanism determination of microearthquakes. Then, the waveform matching method is applied to the Cove Fort Sulphurdale Geothermal site, which is located in central-western Utah. Focal mechanism inversion result shows that the waveforms of synthetic data and real data match very well, and the first arrival P wave polarities and S/P amplitude ratios are close. Fault plane solutions have strikes trending in approximately S-N direction, which is consistent with the maximum horizontal principle stress direction indicated by the stress map of America. Previous researches show that the direction of extension stress in this region is nearly east-west, so the principal compressive stress direction should be close to the north and south direction, consistent with the results of stress analysis based on the focal mechanism and our current understanding of the tectonics in this region. In order to understand the structure characteristics of the fractures in Cove Fort Sulphurdale geothermal area more clearly, we also studied the anisotropy in the regions. Shear wave splitting analysis results show that the anisotropy mainly exists in the shallow area. The fast shear wave polarization directions at each station points to the S-N direction, indicating that if anisotropy is mainly caused by the fault zone structure in this region, the strikes of the main faults/fractures should be in the S-N direction.Finally, this study mainly discusses the focal mechanism inversion based on the assumption of the double couple point source, which can simplify the problem and reduce the solution uncertainty when the medium contains anisotropy and other inhomogeneity in the crust. However, the fault plane solution (FPS) resulting from the double couple source based inversion is not a complete solution. It does not include the volumetric changes and compensated linear vector dipole (CLVD) components. By knowing the non-double couple component, especially the volumetric change, is very important to understand the fracture failure process and the evaluation of hydraulic fracturing efficiency. To this end, in order to ensure the accuracy of focal mechanism inversion results, we expand the fault plane solution inversion to full waveform matching based full moment tensor inversion. Considering the precision and accuracy, the searching space for moment tensor parameters’is huge, which greatly slows down the grid search process. To overcome this problem, we add the differential evolution algorithm in the inversion. Synthetic data test results show that compared with the fault plane solution inversion, the DE improved moment tensor inversion can determine focal mechanism information accurately. Besides, the computational efficiency has raised dozens of times than the grid search method.