Application of a convolutional neural network for seismic phase picking of mining-induced seismicity.

Monitoring mining-induced seismicity (MIS) can help engineers understand the rock mass response to resource extraction. With a thorough understanding of ongoing geomechanical processes, engineers can operate mines, especially those mines with the propensity for rock-bursting, more safely and efficiently. Unfortunately, processing MIS data usually requires significant effort from human analysts, which can result in substantial costs and time commitments. The problem is exacerbated for operations that produce copious amounts of MIS, such as mines with high-stress and/or extraction ratios. Recently, deep learning methods have shown the ability to significantly improve the quality of automated arrival-time picking on earthquake data recorded by regional seismic networks. However, relatively little has been published on applying these techniques to MIS. In this study, we compare the performance of a convolutional neural network (CNN) originally trained to pick arrival times on the Southern California Seismic Network (SCSN) to that of human analysts on coal-mine-related MIS. We perform comparisons on several coal-related MIS data sets recorded at various network scales, sampling rates and mines. We find that the Southern-California-trained CNN does not perform well on any of our data sets without retraining. However, applying the concept of transfer learning, we retrain the SCSN model with relatively little MIS data after which the CNN performs nearly as well as a human analyst. When retrained with data from a single analyst, the analyst-CNN pick time residual variance is lower than the variance observed between human analysts. We also compare the retrained CNN to a simpler, optimized picking algorithm, which falls short of the CNN's performance. We conclude that CNNs can achieve a significant improvement in automated phase picking although some data set-specific training will usually be required. Moreover, initializing training with weights found from other, even very different, data sets can greatly reduce the amount of training data required to achieve a given performance threshold.

Seismoacoustic Monitoring of a Longwall Face Using Distributed Acoustic Sensing.

Violent, dynamic failures of rockmasses in underground mines pose significant hazards to workers and operations. Over the past several decades, hardrock mines have widely adopted seismic monitoring to help address such risks. However, coal mines, particularly those employing the longwall mining method, have struggled to implement similar monitoring strategies. This is because typical longwall mines are much larger and mine more rapidly than hardrock mines. Moreover, regulations place significant restrictions on the subsurface use of electronics in coal mines due to potentially explosive atmospheres. We present a new monitoring concept that uses distributed acoustic sensing (DAS) to turn an entire longwall face into a seismoacoustic array. After exploring the acoustic response of our sensors in the laboratory, we deployed the array at an active underground longwall mine for several days. We examine 33 events recorded by both the in-mine DAS array and a surface seismic network. We observed that the array records both seismic vibrations traveling through rock and mining equipment as well as sound waves propagating in the workings. We show that waveform moveouts are clearly visible, and that the standard deviation of the audio recordings is a straightforward yet promising metric that could help quantify burst damage. Although improvements are needed before mines can routinely use this monitoring strategy, DAS-based seismoacoustic arrays may assist in understanding coal-burst mechanisms and managing associated risks in underground longwall mines as well as enable better understanding of damage associated with dynamic failures in other underground environments.

ObsPlus: A Pandas-centric ObsPy expansion pack.

Over the past decade, ObsPy, a python framework for seismology (Krischer et al., 2015), has become an integral part of many seismology research workflows. ObsPy provides parsers for most seismological data formats, clients for accessing data-centers, common signal processing routines, and event, station, and waveform data models. ObsPlus significantly expands ObsPy's functionality by providing simple data management abstractions and conversions between ObsPy classes and the ubiquitous pandas DataFrame (McKinney, 2010).