We identified resonance due to poor geophone-borehole cou- pling in downhole microseismic data and proposed to use spik- ing deconvolution and relative spectrum analysis to remove its effect. The resonance may hinder the arrival time picking and contaminate microseismic waveform spectrum. We designed a spiking deconvolution filter to recover the impulse response of the earth. Also, we proposed to use relative spectrum anal- ysis to study microseismic source parameters. The applica- tion of deconvolution on Marcellus shale dataset improved the identifiability of the multiple arrivals. Additionally, the rela- tive spectrum analysis is more effective in revealing the actual spectrum characteristics of microseismic waveforms.

We applied Bayesian inference for simultaneous inversion of multiple microseismic data for event locations and velocity models. The traditional method of using a predetermined ve- locity model for event location is subject to large uncertainty if prior information of the velocity model is poor. Our study shows that microseismic data can improve the velocity model, which is usually a major source of location uncertainty. Also, the developed method can quantify the uncertainty of the mi- croseismic location estimation. Its successful application on both synthetic examples and Newberry enhanced geothermal system (EGS) demonstrates its robustness over the traditional least-square traveltime inversion. Comparison with location result of the traditional method shows that we can effectively improve the accuracy of microseismic event location thanks to the improved velocity model.

In this paper, we show that the location of microseismic events can be significantly improved by incorporating information on head wave arrival time. The traditional method of using direct arrival times and P-wave polarizations leads to increased error due to the large uncertainty in polarization. We integrated head wave arrival time to P- and S-wave arrival time to achieve better resolution in microseismic event location. To this end, we developed a Bayesian inference framework for joint event location and velocity model calibration. The developed method was applied for both microseismic event as well as perforation shot location in a project in Marcellus shale. Comparison with location results provided by contractor shows that the developed method can effectively improve the accuracy of microseismic event location. Based on the improvement, we propose a new acquisition geometry and strategy to reduce microseismic monitoring cost and improve event location accuracy.