An effective polarity correction method for microseismic migration-based location

Microseismic methods are important tools for monitoring the status and consequences of hydraulic fracturing. Because microseismic data recorded at the surface have a low signal-to-noise ratio, migration-based algorithms are widely used to determine the locations of microseismic events. However, there may be polarity changes in waveforms at different receivers due to the source mechanisms, which will cause the stacking images to not reach a maximum at the event location. One way for polarity correction is to perform the source mechanism and the source location inversions simultaneously, which, however, is computationally expensive and not good for real-time monitoring. We have developed an effective polarity correction method in the data domain for migration-based location methods called the polarity correction migration-based (PCM) method. This method uses an amplitude trend least-squares fitting procedure to determine the polarities along the receiver line with low additional computational cost. Then, the fitted waveform polarities are used to convert the signs of the amplitude values to stack them consistently. Due to curve fitting, this method is more suitable for microseismic data acquired with regular arrays than with scattered arrays. Numerical experiments of synthetic and real data sets demonstrate that the proposed PCM method can improve accuracy in the detection and location of microseismic events.

Minimum semblance weighted stacking with polarity correction for surface microseismic data processing

Diffraction-stack-based algorithms are the most popular microseismic location methods for surface microseismic data. They can accommodate microseismic data with low signal-to-noise ratio by stacking a large number of traces. However, changes in waveform polarity across the receiver line due to source mechanisms may prevent stacking methods from locating the true source. Imaging functions based on simple stacks have low resolution, producing large uncertainty in the final location result. To solve these issues, we introduce a minimum semblance weighted stacking method with polarity correction, which uses an amplitude trend least-squares fitting algorithm to correct the polarity across the receiver line. We adapt the semblance weighted stacking for better coherency measure to improve the imaging resolution. Moreover, the minimum semblance is used to further improve the resolution of location results. Application to both synthetic and real data sets demonstrates good performance of our proposed location method.