Ground Monitoring of Microseismic Based on Low Signal-to-Noise Ratio

At present, the principle, data acquisition, data processing, and/or interpretation of many microseismic monitoring methods around the world are far from the requirements of microseismic monitoring characteristics, and impossible to analyze the microseismicity. The main technical reason for the situation is still the lack of understanding the characteristics of microseismic and corresponding monitoring for it, so that the monitoring R&D and application are not based on strict seismology, geology, rock mechanics, a large number of reliable experiments and mathematical statistics. We first summarize the characteristics of microseismic and monitoring for it. Based on this, as well as the basic requirements of seismometry, various monitoring methods are discussed, including their applicable conditions, limitations and development prospects. The summary and discussion show that the development and application of microseismic monitoring, even avoiding strong noise sources as much as possible during data acquisition, and effectively denoising during processing, have to face the reality of low signal-to-noise ratio (S/N): in most cases, whether the microseismic signal is implied in the background noise recording, the number of microseismics, and the initial motion form of any microseismic arrival are not known. We then report that in the past 2-3 years, our Vector Scanning (VS) for microseismic ground monitoring has been greatly improved, including: an in-depth understanding of the available principles, the refinement of the conditions necessary for the success of the application with a high probability, and the quantitative integration of automated data processing and interpretation; Among them, the most important is an in-depth understanding of the existing principles: VS uses the focal mechanism (i.e., the relationship between the strain and the stress fields) to implement large-scale migration and stacking, carry out various possible combinations of positive and negative initial movements for all seismic stations, and select the spatiotemporal distribution with high probability of the greater microseismic released energy (i.e., the correlation coefficient recorded of stations, also the minimum S/N). A large number of cases are available for mathematical statistics, which provide a basis for analyzing the details of microseismicity. Finally, we describe the specific morphology of the stimulated rock volume in stimulation, the equivalent microseismic focal mechanism, and the effect of production measures such as in-situ pump shutdown. The necessary conditions, monitoring output patterns and analyses described in the paper also provide a basis for the test of the microseismic methods.