Search

During the past few years, distributed acoustic sensing (DAS) has become an invaluable tool for recording high-fidelity seismic wavefields with great spatiotemporal resolutions. However, the considerable amount of data generated during DAS experiments limits their distribution with the broader scientific community. Such a bottleneck inherently slows down the pursuit of new scientific discoveries in geosciences. Here, we introduce PubDAS—the first large-scale open-source repository where several DAS datasets from multiple experiments are publicly shared. PubDAS currently hosts eight datasets covering a variety of geological settings (e.g., urban centers, underground mines, and seafloor), spanning from several days to several years, offering both continuous and triggered active source recordings, and totaling up to ∼90 TB of data. This article describes these datasets, their metadata, and how to access and download them. Some of these datasets have only been shallowly explored, leaving the door open for new discoveries in Earth sciences and beyond.

Continuous seismic monitoring for quantifying CO2 plume migration and detection of any potential leakages in the subsurface is essential for the security of long-term anthropogenic carbon dioxide geologic storage. Traditional time-lapse full-waveform inversion (TLFWI) methods aim to map the CO2 distribution by estimating seismic velocity changes, but recent studies find that CO2-induced attenuation is an important complement to seismic velocity for tracking the CO2 plumes and even quantifying the CO2 saturation. We have developed a novel data-assimilated TLFWI method to construct high-resolution time-lapse velocity and attenuation changes from dense time-lapse monitoring data. This method consists of two theoretical developments: visco-acoustic full-waveform inversion (QFWI) and multiparameter hierarchical matrix-powered extended Kalman filter (mHiEKF). The method is capable of (1) posing temporal constraints to retrieve time-lapse information from dense monitoring data by using mHiEKF, (2) accurately recovering high-spatial-resolution velocity and attenuation perturbations using first-order equation system-based QFWI, and (3) providing the model uncertainty by estimating their model standard deviation. With numerical examples, we first find the effectiveness of the new QFWI on estimating accurate velocity and attenuation models simultaneously. Then, a CO2 leakage case and a realistic Frio-II CO2 monitoring case are presented to find the advantages and applicability of our data-assimilated QFWI method for estimating time-lapse changes using dense time-lapse monitoring surveys. By assimilating time-lapse seismic monitoring data over time, our data-assimilated QFWI method can improve the resolution of velocity and attenuation changes and decrease their model uncertainties.

The Distributed Acoustic Sensing (DAS) method re-purposes fiber optic cables into a very-dense array of strain/strain-rate sensors, capable of detecting different types of seismic events. However, DAS data are characterized by lower SNRs compared with standard seismic sensors, mainly because of a) strong directivity effects, 2) ground coupling inhomogeneities, and 3) site effects. Hence, beyond the array geometry, specific noise sources may reduce the potential of DAS for seismic monitoring. Previous research has already shown successful case-studies for event detection/location. Nevertheless, a coherent test on the performances of various arrays of different sizes and geometries is still lacking.In this study, an extensive DAS database is organized for such a goal, including 15 DAS arrays that recorded at least one seismic event (located at a range of distances from the arrays). P wave arrival times are exploited to estimate the epicentral parameters with a Markov Chain Monte Carlo method. Then, to analyze the effects of cable geometry and potential sources of noise/ambiguity on the location uncertainties, a series of synthetic tests are performed, where synthetic traveltimes are modified as follows: a) adding noise with equal variance to all the DAS channels (SYNTH-01), b) adding noise characterized by an increasing variance with the distance from the event (SYNTH-02), c) simulating the mis-pick between P and S phases (SYNTH-03) and d) adding noise with a variance influenced by cable coupling inhomogeneities (SYNTH-04). Results show that the epicentral locations with automatic P wave arrival times have different degrees of uncertainty, given the geometrical relation between the event and the DAS arrays. This behavior is confirmed by the SYNTH-01 test, indicating that specific geometries provide a lower constraint on event location. Moreover, SYNTH-04 shows that simulating cable coupling inhomogeneities primarily reproduces the observed location uncertainties. Finally, some cases are not explained by any of the synthetic tests, stressing the possible presence of more complex noise sources contaminating the signals.

We have formulated the Fréchet kernel computation using the adjoint-state method based on a fractional viscoacoustic wave equation. We first numerically prove that the 1/2- and the 3/2-order fractional Laplacian operators are self-adjoint. Using this property, we find that the adjoint wave propagator preserves the dispersion and compensates the amplitude, whereas the time-reversed adjoint wave propagator behaves identically to the forward propagator with the same dispersion and dissipation characters. Without introducing rheological mechanisms, this formulation adopts an explicit [Formula: see text] parameterization, which avoids the implicit [Formula: see text] in the conventional viscoacoustic/viscoelastic full-waveform inversion ([Formula: see text]-FWI). In addition, because of the decoupling of operators in the wave equation, the viscoacoustic Fréchet kernel is separated into three distinct contributions with clear physical meanings: lossless propagation, dispersion, and dissipation. We find that the lossless propagation kernel dominates the velocity kernel, whereas the dissipation kernel dominates the attenuation kernel over the dispersion kernel. After validating the Fréchet kernels using the finite-difference method, we conduct a numerical example to demonstrate the capability of the kernels to characterize the velocity and attenuation anomalies. The kernels of different misfit measurements are presented to investigate their different sensitivities. Our results suggest that, rather than the traveltime, the amplitude and the waveform kernels are more suitable to capture attenuation anomalies. These kernels lay the foundation for the multiparameter inversion with the fractional formulation, and the decoupled nature of them promotes our understanding of the significance of different physical processes in [Formula: see text]-FWI.

First arrival time picking is one of the critical processing steps of acoustic emission (AE)/microseismic (MS) monitoring for studying rock fracture processes. Because of massive monitoring data, the automatic arrival time picking technique is particularly desired. Inspired by recent successful applications of machine learning (ML) in earthquake phase identification, we propose a deep learning (DL)-based P-wave first arrival time picking method named AE Network (AEnet) for laboratory AE monitoring data. Our approach consists of two steps: classification and picking. The convolutional neural network (CNN) is used to classify each sample point of acoustic waveforms into either noise or signal. Different from prior DL-based phase picking studies using raw waveforms, we combine the waveform and high-order statistics as the input to enrich the input data features and accelerate the CNN model learning process. Our approach is examined using the laboratory AE monitoring data and the performance of each component of AEnet is also analyzed. The results show that the CNN model can classify the sample points accurately for the picking procedure. With this classification result, we pick the first arrival time of each trace using the curve fitting method and an unsupervised clustering algorithm. To evaluate the performance of AEnet, we apply Akaike Information Criterion-Short Term Averaging/Long Term Averaging Method (AIC-STA/LTA), one of the most popular and traditional picking methods, on the same waveforms and use the manual picks as the reference. Error analysis results show that AEnet outperforms AIC-STA/LTA.

Prior studies have shown an ambiguous relationship between gas-hydrate saturation and seismic attenuation in different regions, but the effect of gas-hydrate morphology on seismic attenuation of hydrate-bearing sediments was often overlooked. We have combined seismic data with rock-physics modeling to elucidate how gas-hydrate saturation and morphology may control seismic attenuation. To extract P-wave attenuation, we process the vertical seismic profile data within a frequency range of 30–150 Hz and the sonic logging data within 10–15 kHz from three wells in the south Hydrate Ridge, offshore Oregon (USA), collected during Ocean Drilling Program Leg 204 in 2000. We calculate the P-wave attenuation using spectral matching and centroid frequency shift methods, and we use Archie’s relationship to derive gas-hydrate saturation from the resistivity data above the bottom-simulating reflection at the same wells. To interpret the observed seismic attenuation in terms of the effects of the gas-hydrate saturation and the morphology, we use the hydrate-bearing effective sediment rock-physics model. By comparing the observed and model-predicted attenuation values, we infer that (1) seismic attenuation appears to not be dominated by any single factor — instead, its variation is likely governed by the gas-hydrate saturation and the morphology; (2) the relationship between seismic attenuation and gas-hydrate saturation varies with different hydrate morphologies; (3) the squirt flow, occurring at different compliances of adjacent pores driven by pressure gradients, may be responsible for the significantly large or small attenuation over a broad frequency range.

Quantifying the response of human activity to different COVID-19 measures may serve as a potential way to evaluate the effectiveness of the measures and optimize them. Recent studies reported that seismic noise reduction caused by limited human activity due to the COVID-19 lockdown had been observed with seismometers. However, it is difficult for the current seismic infrastructure in urban cities to characterize spatiotemporal seismic noise during the post-COVID-19 lockdown, because of their sparse distribution. Here, we show key connections between progressive COVID-19 measures and spatiotemporal seismic noise changes recorded by a distributed acoustic sensing (DAS) array deployed in State College, Pennsylvania. We first show a spatiotemporal seismic noise reduction (up to 90%) corresponding to reduced human activity in different city blocks during the stay-at-home period. We also show partial noise recovery corresponding to increased road traffic and industrial machinery in phase yellow and phase green of the lockdown. Nonrecovery seismic noise in the 0.01–10 Hz band suggests the low level of pedestrian movement during phase yellow and phase green. According to a linear correlation between Google mobility change and seismic noise change, we emphasize that DAS recordings using city-wide fiber optics could provide a way for quantifying the impact of COVID-19 measures on human activity in different blocks.

Thunderstorms are a common atmospheric phenomenon that cause abundant acoustic disturbances, which can interact with the ground surface, creating a link between atmospheric and solid Earth processes. This article reports seismological observations of four thunderstorms through the spring and summer of 2019, as recorded by the distributed acoustic sensing fiber-optic array (4.9 km) on the Penn State campus in State College, Pennsylvania. With a dense sensor array in the local region, we are able to construct the seismic full waveform response of the thunderstorm events (hereafter referred to as thunderquakes) and track the wave propagation across the array. We use a time-domain grid search to obtain the back azimuth and slowness of the waves, and a modified Geiger’s method to pinpoint source locations of the thunderquakes. Correlated with the time of the recorded signal, this data allows reconstruction of thunderstorm movement as well as offering measurements of the seismic velocity.

SUMMARY Seismic attenuation (quantified by the quality factor Q) has a significant impact on the seismic waveforms, especially in the fluid-saturated rocks. This dissipative process can be phenomenologically represented by viscoelastic models. Previous seismological studies show that the Q value of Earth media exhibits a nearly frequency-independent behaviour (often referred to as constant-Q in literature) in the seismic frequency range. Such attenuation can be described by the mathematical Kjartansson constant-Q model, which lacks of a physical representation in the viscoelastic sense. Inspired by the fractal nature of the pore fluid distribution in patchy-saturated rocks, here we propose two fractal mechanical network (FMN) models, that is, a fractal tree model and a quasi-fractal ladder model, to phenomenologically represent the frequency-independent Q behaviour. As with the classic viscoelastic models, the FMN models are composed of mechanical elements (spring and dashpots) arranged in different hierarchical patterns. A particular parametrization of each model can produce the same complex modulus as in the Kjartansson model, which leads to the constant-Q. Applying the theory to several typical rock samples, we find that the seismic attenuation signature of these rocks can be accurately represented by either one of the FMN models. Besides, we demonstrate that the ladder model in particular exhibits the realistic multiscale fractal structure of the saturated rocks. Therefore, the FMN models as a proxy could provide a new way to estimate the microscopic rock structure property from macroscopic seismic attenuation observation.

Seismic data for studying the near surface have historically been extremely sparse in cities, limiting our ability to understand small-scale processes, locate small-scale geohazards, and develop earthquake hazard microzonation at the scale of buildings. In recent years, distributed acoustic sensing (DAS) technology has enabled the use of existing underground telecommunications fibers as dense seismic arrays, requiring little manual labor or energy to maintain. At the Fiber-Optic foR Environmental SEnsEing array under Pennsylvania State University, we detected weak slow-moving signals in pedestrian-only areas of campus. These signals were clear in the 1 to 5 Hz range. We verified that they were caused by footsteps. As part of a broader scheme to remove and obscure these footsteps in the data, we developed a convolutional neural network to detect them automatically. We created a data set of more than 4000 windows of data labeled with or without footsteps for this development process. We describe improvements to the data input and architecture, leading to approximately 84% accuracy on the test data. Performance of the network was better for individual walkers and worse when there were multiple walkers. We believe the privacy concerns of individual walkers are likely to be highest priority. Community buy-in will be required for these technologies to be deployed at a larger scale. Hence, we should continue to proactively develop the tools to ensure city residents are comfortable with all geophysical data that may be acquired.

SUMMARY Rapid development of time-lapse seismic monitoring instrumentations has made it possible to collect dense time-lapse data for tomographically retrieving time-lapse (even continuous) images of subsurface changes. While traditional time-lapse full waveform inversion (TLFWI) algorithms are designed for sparse time-lapse surveys, they lack of effective temporal constraint on time-lapse data, and, more importantly, lack of the uncertainty estimation of the TLFWI results that is critical for further interpretation. Here, we propose a new data assimilation TLFWI method, using hierarchical matrix powered extended Kalman filter (HiEKF) to quantify the image uncertainty. Compared to existing Kalman filter algorithms, HiEKF allows to store and update a data-sparse representation of the cross-covariance matrices and propagate model errors without expensive operations involving covariance matrices. Hence, HiEKF is computationally efficient and applicable to 3-D TLFWI problems. Then, we reformulate TLFWI in the framework of HiEKF (termed hereafter as TLFWI-HiEKF) to predict time-lapse images of subsurface spatiotemporal velocity changes and simultaneously quantify the uncertainty of the inverted velocity changes over time. We demonstrate the validity and applicability of TLFWI–HiEKF with two realistic CO2 monitoring models derived from Frio-II and Cranfield CO2 injection sites, respectively. In both 2-D and 3-D examples, the inverted high-resolution time-lapse velocity results clearly reveal a continuous velocity reduction due to the injection of CO2. Moreover, the accuracy of the model is increasing over time by assimilating more time-lapse data while the standard deviation is decreasing over lapsed time. We expect TLFWI-HiEKF to be equipped with real-time seismic monitoring systems for continuously imaging the distribution of subsurface gas and fluids in the future large-scale CO2 sequestration experiments and reservoir management.

The spatial derivatives in decoupled fractional Laplacian (DFL) viscoacoustic and viscoelastic wave equations are the mixed-domain Laplacian operators. Using the approximation of the mixed-domain operators, the spatial derivatives can be calculated by using the Fourier pseudospectral (PS) method with barely spatial numerical dispersions, whereas the time derivative is often computed with the finite-difference (FD) method in second-order accuracy (referred to as the FD-PS scheme). The time-stepping errors caused by the FD discretization inevitably introduce the accumulative temporal dispersion during the wavefield extrapolation, especially for a long-time simulation. To eliminate the time-stepping errors, here, we adopted the [Formula: see text]-space concept in the numerical discretization of the DFL viscoacoustic wave equation. Different from existing [Formula: see text]-space methods, our [Formula: see text]-space method for DFL viscoacoustic wave equation contains two correction terms, which were designed to compensate for the time-stepping errors in the dispersion-dominated operator and loss-dominated operator, respectively. Using theoretical analyses and numerical experiments, we determine that our [Formula: see text]-space approach is superior to the traditional FD-PS scheme mainly in three aspects. First, our approach can effectively compensate for the time-stepping errors. Second, the stability condition is more relaxed, which makes the selection of sampling intervals more flexible. Finally, the [Formula: see text]-space approach allows us to conduct high-accuracy wavefield extrapolation with larger time steps. These features make our scheme suitable for seismic modeling and imaging problems.

This is a Cuda based code to simulate 3D viscoelastic wave propagation in isotropic or VTI attenuation media

To efficiently simulate wave propagation in a vertical transversely isotropic (VTI) attenuative medium, we have developed a viscoelastic VTI wave equation based on fractional Laplacian operators under the assumption of weak attenuation ([Formula: see text]), where the frequency-independent [Formula: see text] model is used to mathematically represent seismic attenuation. These operators that are nonlocal in space can be efficiently computed using the Fourier pseudospectral method. We evaluated the accuracy of numerical solutions in a homogeneous transversely isotropic medium by comparing with theoretical predictions and numerical solutions by an existing viscoelastic-anisotropic wave equation based on fractional time derivatives. To accurately handle heterogeneous [Formula: see text], we select several [Formula: see text] values to compute their corresponding fractional Laplacians in the wavenumber domain and interpolate other fractional Laplacians in space. We determined its feasibility by modeling wave propagation in a 2D heterogeneous attenuative VTI medium. We concluded that the new wave equation is able to improve the efficiency of wave simulation in viscoelastic-VTI media by several orders and still maintain accuracy.

The generation of microseismic events is often associated with induced fractures/faults during the extraction/injection of fluids. A full characterization of the spatiotemporal distribution of microseismic events provides constraints on fluid migration paths in the formations. We have developed a high-resolution source imaging method — a hybrid multiplicative time-reversal imaging (HyM-TRI) algorithm, for automatically tracking the spatiotemporal distribution of microseismic events. HyM-TRI back propagates the data traces from groups of receivers (in space and time) as receiver wavefields, multiplies receiver wavefields between all groups, and applies a causal integration over time to obtain a source evolution image. Using synthetic and field-data examples, we revealed the capability of the HyM-TRI technique to image the spatiotemporal sequence of asynchronous microseismic events, which poses a challenge to standard TRI methods. Moreover, the HyM-TRI technique is robust enough to produce a high-resolution image of the source in the presence of noise. The aperture of the 2D receiver array (azimuth coverage in 3D) with respect to the microseismic source area plays an important role on the horizontal and vertical resolution of the source image. The HyM-TRI results of the field data with 3D azimuthal coverage further verify our argument by producing a superior resolution of the source than TRI.

Quantifying the dynamics of sequestered CO2 plumes is critical for safe long-term storage, providing guidance on plume extent, and detecting stratigraphic seal failure. However, existing seismic monitoring methods based on wave reflection or transmission probe a limited rock volume and their sensitivity decreases as CO2 saturation increases, decreasing their utility in quantitative plume mass estimation. Here we show that seismic scattering coda waves, acquired during continuous borehole monitoring, are able to illuminate details of the CO2 plume during a 74-h CO2 injection experiment at the Frio-II well Dayton, TX. Our study reveals a continuous velocity reduction during the dynamic injection of CO2, a result that augments and dramatically improves upon prior analyses based on P-wave arrival times. We show that velocity reduction is nonlinearly correlated with the injected cumulative CO2 mass and attribute this correlation to the fact that coda waves repeatedly sample the heterogeneous distribution of cumulative CO2 in the reservoir zone. Lastly, because our approach does not depend on P-wave arrival times or require well-constrained wave reflections it can be used with many source-receiver geometries including those external to the reservoir, which reduces the risk introduced by in-reservoir monitoring wells. Our results provide an approach for quantitative CO2 monitoring and plume evolution that increases safety and long-term planning for CO2 injection and storage.

Recent world-wide quieting of seismic noise caused by COVID-19 lockdown measures has been observed by seismometers. However, current seismic network that has a few seismometers or none in a city-sc...

The FORESEE Distributed Acoustic Sensing (DAS) Array records roughly 1/3 terabyte of data per day along 5 kilometers of dark fiber optic telecommunications cable underneath the Pennsylvania State University campus. The campus sits in the Allegheny Mountain region of the US, and our aim is to understand urban hydrology and detection of geohazards (particularly karst features). We have verified a number of features of these data similar to prior urban seismic studies, both in ambient noise and in distant earthquake records, which builds further evidence that dark fiber can be a useful tool for seismology in cities.These data also contain a number of new signals not observed on previous dark fiber arrays. We see a stronger response to air waves than prior experiments. For instance, musical bass lines are clearly observed in the 30-100 Hz range during a concert, and we can see the spatial decay of higher versus lower frequencies throughout the array. This is the first dark fiber array in the eastern US, where thunderstorms occur with some frequency, and we have observed clear recordings of ground motion due to thunder. Source inversion of the waveforms throughout the array leads to locations that show reasonable agreement compared to the National Lightning Detection Network. These thunderquake signals could be an important source of broadband energy for seismic imaging in an area with little earthquake seismicity.We have performed ambient noise interferometry throughout the array with a variety of pre-processing workflows, but some subsets of the array are strongly affected by nearby sources. With the wide variety of natural and manmade signals in these data, we are working towards further efficient automation to detect repeatable signals that could be used for targeted interferometry, and methods to automate filtering of non-ideal noise sources. As one example of filtering a specific noise, we were surprised the array is able detect the paths of individuals walking along a sidewalk by the fiber. While this array records data on a public college campus, a likely future area of research may include urban areas with a mix of commercial and residential purposes, so we desire tools to remove individual signals as they are recorded. Thus, we have developed a neural network to detect and remove footsteps from data before those data are shared with researchers. To encourage others working on urban seismic acquisition to remove similar signals, we are generalizing these methods for footstep removal to different scales.

Ambient noise seismology has greatly reduced the cost of acquiring data for seismic monitoring and imaging by reducing the need for active sources. For applications requiring time-lapse imaging or continuous monitoring, we desire sensor arrays that require little effort, money, and power to maintain over long periods of time. Distributed Acoustic Sensing repurposes a standard fiber optic cable as a series of single-component strain rate sensors with spacing at the scale of meters over distances of kilometers. With a single location providing the power source and recording all data, along with the ability to use existing underground fiber optic networks, a small team is now able to easily establish a monitoring network and acquire massive amounts of strain rate data continuously.This talk will explore two conceptual changes when using DAS data for ambient noise interferometry: greatly increased data volumes, and the difference between velocity and distributed strain-rate data. These two challenges will be illustrated in the context of experiments with applications in near-surface Vs imaging with applications in earthquake hazard analysis, permafrost thaw monitoring, and urban geohazard and hydrology monitoring.On the issue of data volumes: Orders of magnitude more sensors and high sample rates (often in the kilohertz range) quickly result in data quantities that exceed the limits of computational infrastructure and algorithms available to many seismologists, potentially at the petabyte/year scale for modern acquisition instruments. New algorithms focused on reduced data movement are improving our ability to analyze more data with existing resources. This talk will include a brief overview of some recent algorithmic improvements for both ambient noise interferometry for imaging, and interferometry-based event detection.On the issue of changing from velocity to distributed strain rate data: Because strain rate is a tensor quantity and velocities are a vector quantity, the sensitivity of DAS to seismic sources at different orientations is quite different from typical seismometers. This difference can be clear both in polarity and amplitude of the signal, and is particularly significant in shear and Love wave recordings. We will describe simple models to describe expected changes in how seismometers and DAS record the same noises, and the corresponding changes expected in noise correlation functions. These sensitivity differences are more pronounced in ambient noise correlation functions than they are in raw signal recordings, effectively emphasizing a different distribution of ambient noise sources. Modeling these sensitivities helps determine which sensor orientations are reliable for use in ambient noise interferometry imaging.

The calculation of the gradient in full-waveform inversion (FWI) usually involves crosscorrelating the forward-propagated source wavefield and the back-propagated data residual wavefield at each time step. In the real earth, propagating waves are typically attenuated due to the viscoelasticity, which results in an attenuated gradient for FWI. Replacing the attenuated true gradient with a [Formula: see text]-compensated gradient can accelerate the convergence rate of the inversion process. We have used a phase-dispersion and an amplitude-loss decoupled constant-[Formula: see text] wave equation to formulate a viscoacoustic FWI. We used this wave equation to generate a [Formula: see text]-compensated gradient, which recovers amplitudes while preserving the correct kinematics. We construct an exact adjoint operator in a discretized form using the low-rank wave extrapolation technique, and we implement the gradient compensation by reversing the sign of the amplitude-loss term in the forward and adjoint operators. This leads to a [Formula: see text]-dependent gradient preconditioning method. Using numerical tests with synthetic data, we demonstrate that the proposed viscoacoustic FWI using a constant-[Formula: see text] wave equation is capable of producing high-quality velocity models, and our [Formula: see text]-compensated gradient accelerates its convergence rate.