ABSTRACT Through downhole monitoring, the focal mechanisms of microearthquakes can be quantitatively determined, thus providing valuable information for characterizing the fracturing process and the in situ stress status. The double-couple (DC) and moment tensor (MT) source models are commonly used to study microearthquakes. However, the DC model fails to include non-DC mechanisms, and MT inversion from single-well data is still challenging. One possible way to address this is using the shear–tensile general dislocation (GD) source model. We provide a detailed comparison of the DC, GD, and MT models, and introduce the differences in their modeling and inversion theories. These three models are described by four, five, and six parameters, and correspond to a single point, a straight line, and the entire space in the Hudson source-type plots, respectively. Both the DC and GD models yield nonlinear inversions, whereas the MT inversion is linear. Synthetic tests set up from a field single-well monitoring case are performed to study the resolvability of the DC, GD, and MT models in single-well focal mechanism inversions. The results indicate that the inversion error increases from DC→GD→MT for a single-well acquisition system, and the GD and DC inversions are both stable, whereas the MT inversion deviates from the inputs in cases with a perfectly vertical receiver array, 5% model velocity perturbations, 10 m horizontal source location errors, or 40% noise levels. We also find that the focal mechanism inversion mainly depends on the horizontal source–receiver azimuth coverage, and that the nonvertical well direction is helpful for constraining single-well inversions. According to our study, focal mechanism inversions based on the GD model can obtain reliable solutions from near-vertical single-well data, which will help improve non-DC earthquake studies.
Read full abstract