Detecting and locating microseismic events with stacking velocity analysis for surface monitoring

We have analyzed microseismic monitoring data sets obtained from the surface and downhole-monitoring arrays recorded during the first experiment of hydraulic fracturing in Poland. Using the downhole-monitoring network, we were able to record and locate 844 microseismic events, including 10 perforation shots from six stages of the stimulation. We detected 2 perforation shots and no microseismic events using the surface array, which was operational only during the first two stages of the stimulation. To explain the poor detectability of the surface array, we analyzed the spectral content of the events from the downhole-monitoring array. We found that the detectability of the perforation shots on the surface array was consistent with the low-frequency part of the signal on the downhole recordings. Our observation is in agreement with the fact that microseismic events with low-frequency signal weaker than the two detected perforation shots were not detected by the surface-monitoring array. Using the low-frequency part of the spectra of the events recorded by the downhole array, we predicted the surface-array detection threshold. We found that some events from the later stages could have been detected if only the surface array had been operational during that time.

Download Full-text

Joint microseismic moment tensor inversion and location using P- and S-waves diffraction stacking

Geophysics ◽

10.1190/geo2021-0104.1 ◽

2021 ◽

pp. 1-62

Author(s):

Xu Jincheng ◽

Wei Zhang ◽

Xing Liang ◽

Jiaojun Rong ◽

Junlun Li

Keyword(s):

Shale Gas ◽

Moment Tensor ◽

Moment Tensor Inversion ◽

P Wave ◽

Origin Time ◽

S Waves ◽

Surface Monitoring ◽

Microseismic Events ◽

Location Methods ◽

Fracture Growth

The microseismic location methods based on diffraction stacking which does not require arrival picking can yield accurate and reliable source location for data with a low signal-to-noise ratio. However, due to the complex radiation pattern from a rupturing source, variation in the waveform polarities brings challenges to the diffraction-stacking based methods. The current implementations of joint source mechanism inversion and location methods which only use P-wave amplitudes have limitations in noise resistance and location accuracy. To mitigate those issues, we develop a new method for joint microseismic moment tensor inversion and event location using diffraction stacking with P- and S-waves amplitudes, both of which are used to invert for the moment tensor of a microseismic event, and then the inverted moment tensor is used to correct the waveform polarity changes before stacking. In addition, to expedite the large amount of calculations required for moment tensor inversion at each potential source position and origin time, we develop an optimized grid search scheme and implement the algorithm with GPUs. The proposed location method does not require manual picking of the first arrivals, and can automatically detect and locate microseismic events from continuous data. We first validated the method with two synthetic examples, and then applied it to a surface monitoring dataset for hydraulic fracturing at a shale gas well pad in the southern Sichuan Basin, China, where billions of cubic meters of shale gas are being produced annually. The locations of the microseismic events are nicely correlated with the fracturing stages and the determined source mechanisms are also consistent with the expected fracture growth. The proposed method is feasible for microseismic surface monitoring with dense nodal arrays and can provide important information for fracture growth and regional stress characterization.

Download Full-text