Seismic waves from earthquakes and other sources are used to infer the structure and properties of Earth’s interior. The availability of large-scale seismic datasets and the suitability of deep-learning techniques for seismic data processing have pushed deep learning to the forefront of fundamental, long-standing research investigations in seismology. However, some aspects of applying deep learning to seismology are likely to prove instructive for the geosciences, and perhaps other research areas more broadly. Deep learning is a powerful approach, but there are subtleties and nuances in its application. We present a systematic overview of trends, challenges, and opportunities in applications of deep-learning methods in seismology. Description Large-scale learning The large amount and availability of datasets in seismology creates a great opportunity to apply machine learning and artificial intelligence to data processing. Mousavi and Beroza provide a comprehensive review of the deep-learning techniques being applied to seismic datasets, covering approaches, limitations, and opportunities. The trends in data processing and analysis can be instructive for geoscience and other research areas more broadly. —BG The ways in which deep learning can help process and analyze large seismological datasets are reviewed. BACKGROUND Seismology is the study of seismic waves to understand their origin—most obviously, sudden fault slip in earthquakes, but also explosions, volcanic eruptions, glaciers, landslides, ocean waves, vehicular traffic, aircraft, trains, wind, air guns, and thunderstorms, for example. Seismology uses those same waves to infer the structure and properties of planetary interiors. Because sources can generate waves at any time, seismic ground motion is recorded continuously, at typical sampling rates of 100 points per second, for three components of motion, and on arrays that can include thousands of sensors. Although seismology is clearly a data-rich science, it often is a data-driven science as well, with new phenomena and unexpected behavior discovered with regularity. And for at least some tasks, the careful and painstaking work of seismic analysts over decades and around the world has also made seismology a data label–rich science. This facet makes it fertile ground for deep learning, which has entered almost every subfield of seismology and outperforms classical approaches, often dramatically, for many seismological tasks. ADVANCES Seismic wave identification and onset-time, first-break determination for seismic P and S waves within continuous seismic data are foundational to seismology and are particularly well suited to deep learning because of the availability of massive, labeled datasets. It has received particularly close attention, and that has led, for example, to the development of deep learning–based earthquake catalogs that can feature more than an order of magnitude more events than are present in conventional catalogs. Deep learning has shown the ability to outperform classical approaches for other important seismological tasks as well, including the discrimination of earthquakes from explosions and other sources, separation of seismic signals from background noise, seismic image processing and interpretation, and Earth model inversion. OUTLOOK The development of increasingly cost-effective sensors and emerging ground-motion sensing technologies, such as fiber optic cable and accelerometers in smart devices, portend a continuing acceleration of seismological data volumes, so that deep learning is likely to become essential to seismology’s future. Deep learning’s nonlinear mapping ability, sequential data modeling, automatic feature extraction, dimensionality reduction, and reparameterization are all advantageous for processing high-dimensional seismic data, particularly because those data are noisy and, from the point of view of mathematical inference, incomplete. Deep learning for scientific discovery and direct extraction of insight into seismological processes is clearly just getting started. Aspects of seismology pose interesting additional challenges for deep learning. Many of the most important problems in earthquake seismology—such as earthquake forecasting, ground motion prediction, and rapid earthquake alerting—concern large and damaging earthquakes that are (fortunately) rare. That rarity poses a fundamental challenge for the data-hungry methods of deep learning: How can we train reliable models, and how do we validate them well enough to rely on them when data are scarce and opportunities to test models are infrequent? Further, how can we operationalize deep-learning techniques in such a situation, when the mechanisms by which they make predictions from data may not be easily explained, and the consequences of incorrect models are high? Incorporating domain knowledge through physics-based and explainable deep learning and setting up standard benchmarking and evaluation protocols will help ensure progress, as is the nascent emergence of a seismological data science ecosystem. More generally, a combination of data science literacy for geoscientists as well as recruiting data science expertise will help to ensure that deep-learning seismology reaches its full potential. Deep-learning processing of seismic data and incorporation of domain knowledge can lead to new capabilities and new insights across seismology. A. MASTIN/SCIENCE, TOP RIGHT: CARA HARWOOD/UNIVERSITY OF CALIFORNIA-DAVIS/CC BY-NC-SA 3.0 MIDDLE RIGHT: JOHAN SWANEPOEL/SCIENCE SOURCE

Deep-learning seismology

Over the last three decades, fiber optic sensors (FOS) have gained a lot of attention for their wide range of monitoring applications across many industries, including aerospace, defense, security, civil engineering, and energy. FOS technologies hold great promise to form the backbone for next‐generation intelligent sensing platforms that offer long‐distance, high‐accuracy, distributed measurement capabilities and multiparametric monitoring with resilience to harsh environmental conditions. The major limitations posed by FOS are 1) cross‐sensitivity, 2) enormous volume and large data generation, 3) low data processing speed, 4) degradation of signal‐to‐noise ratio over the fiber length, and 5) overall cost of sensor and interrogator systems. These challenges can be overcome by building advanced data analytics engines enabled by recent breakthroughs in machine learning (ML) and artificial intelligence (AI). This article presents a comprehensive review of recent studies that integrate ML and AI algorithms with FOS technologies. This review also highlights several FOS technology development directions that promise a significant impact on widespread use for several industrial applications, with an emphasis on energy systems monitoring. A perspective on future directions for further research development is also provided.

Recent Advances in Machine Learning for Fiber Optic Sensor Applications

Cite this article as Lellouch, A., N. J. Lindsey, W. L. Ellsworth, and B. L. Biondi (2020). Comparison between Distributed Acoustic Sensing and Geophones: Downhole Microseismic Monitoring of the FORGE Geothermal Experiment, Seismol. Res. Lett. XX, 1–13, doi: 10.1785/0220200149. We compare the performance of a downhole distributed acoustic sensing (DAS) fiberoptic array with that of conventional geophones. The downhole collocated arrays are part of the Frontier Observatory for Research in Geothermal Energy (FORGE) geothermal experiment, in which stimulation of the rock volume in an enhanced geothermal system (EGS) causes microseismic events. The DAS acquisition system yields data sampled at every 1 m at 2000 samples per second for the entire length of the well, spanning to a depth of 985 m from the surface. Whereas single DAS channels are substantially noisier than geophones at the same location, their large number and spatial coherency allow for the application of effective array processing techniques. We follow a complete workflow for the fiber-optic array: velocity model building, event detection, event location, and magnitude estimation. Estimated velocity models agree well with sonic logging in a nearby well and map a granitic contact accurately. Detection performance is somewhat worse than geophones and yields magnitude completeness of −1:4 compared to −1:7 for geophones. Using a single vertical fiber array, we cannot retrieve the azimuth of the events relative to the well. However, we can very accurately estimate their depth and horizontal distance from the array. Magnitude estimation with DAS approaches geophone results to within a standard deviation ofM 0:115 and negligible mean difference. The DAS processing results outperform a regional and local surface array, consolidated with a shallow borehole sensor. Although downhole geophones in the FORGE experimental layout performed better, DAS holds several critical practical benefits that were not demonstrated. Thanks to its heat resistance, it can be deployed much closer to the reservoir; fibers can be deployed along cased active wells, eliminating the need for a dedicated monitoring well; the permanently installed fiber can be used for years or decades. Therefore, we argue that DAS holds vast potential for long-term monitoring of EGS projects.

Comparison between Distributed Acoustic Sensing and Geophones: Downhole Microseismic Monitoring of the FORGE Geothermal Experiment

This study presents the first demonstration of the transferability of a convolutional neural network (CNN) trained to detect microseismic events in one fiber-optic distributed acoustic sens...

/pdf/application-of-machine-learning-to-microseismic-event-32ugvz8rlg.pdf

Application of machine learning to microseismic event detection in distributed acoustic sensing data

Icequakes, microseismic earthquakes at glaciers, offer critical insights into the dynamics of ice sheets. For the first time in the Antarctic, we explore the use of fibre optic cables as Distribute...

Distributed Acoustic Sensing (DAS) for Natural Microseismicity Studies: A Case Study From Antarctica

We have developed a case study demonstrating the use of an “L”-shaped downhole fiber-optic array to monitor microseismicity. We use a relatively simple method to detect events from continuo...

/pdf/microseismic-monitoring-using-a-fibre-optic-distributed-l0gvi95kzt.pdf

Microseismic monitoring using a fibre-optic Distributed Acoustic Sensor (DAS) array

Fiber-optic distributed acoustic sensing (DAS) cables are now used to monitor microseismicity during hydraulic-fracture stimulations of unconventional gas reservoirs. Unlike geophone arrays...

/pdf/characteristics-of-microseismic-data-recorded-by-distributed-5fhat0ldc0.pdf

Characteristics of microseismic data recorded by distributed acoustic sensing systems in anisotropic media

Summary Fibre-optic Distributed Acoustic Sensing (DAS) cables are now used to monitor microseismicity during hydraulic fracture stimulations of unconventional gas reservoirs. Unlike geophone arrays, DAS systems are sensitive to uniaxial strain along the fibre direction and thus provide a single-component recording, which makes identifying the directionality of incoming waves difficult to infer. Using synthetic examples, we show some fundamental characteristics of microseismic recordings on DAS systems for purposes of hydraulic fracture monitoring in a horizontal well in anisotropic (VTI) shales. We demonstrate that SH arrivals dominate the recorded signals since their polarization is aligned along the horizontal cable at near offset. The amplitude of the SH phase along the cable exhibits a characteristic pattern with bimodal peaks, the width of which relates to the distance of the event from the cable. Furthermore, we find that shear-wave splitting recorded on DAS systems can be used to infer the inclination of the incoming waves, overcoming a current limitation of event locations which have constrained events to lie in a horizontal plane. Low amplitude qSV arrivals suggest an event depth similar to that of the DAS cable. Conversely, steep arrivals produce higher amplitude qSV waves, with shear-wave splitting increasing with offset along the cable.

/pdf/modelling-of-fibre-optic-das-response-to-microseismic-3djrxcss2u.pdf

A. Clarke

Papers

Application of machine learning to microseismic event detection in distributed acoustic sensing data

Microseismic monitoring using a fibre-optic Distributed Acoustic Sensor (DAS) array

Characteristics of microseismic data recorded by distributed acoustic sensing systems in anisotropic media

Modelling of Fibre-Optic DAS Response to Microseismic Arrivals in Anisotropic Media