Automatic Microseismic Events Detection Using Multiscale Morphological Characteristic Function

Abstract

Surface-acquired microseismic events are commonly unpredictable and appear as weak signals with low signal-to-noise ratios (SNRs), which present a significant challenge in microseismic data analysis and events detection. We have developed the multiscale morphology method for automatically detecting microseismic events. The technique is based on the fact that the effective signal and noise have different features in the multiscale morphological section. According to these characteristics, the effective signal can be identified by waveform consistency analysis. The microseismic data are first decomposed into different scales by multiscale morphology decomposition. Then the multiscale morphology characteristic function is calculated by waveform cross correlation approach and the potential microseismic events are evaluated using the peak value of the multiscale morphology characteristic function. Applications of the proposed method on the synthesized data and field data have been explored in this article. Various examples demonstrate that the proposed method has the advantage of suppressing the effect of Gaussian noises adaptively. Furthermore, it is excellent in automatically detecting the low SNR signal so that more microseismic events can be identified. In addition, the method could be applied to other fields, such as in the identification of earthquake signals.