Effective use of the wealth of information provided by Distributed Acoustic Sensing (DAS) for mass movement monitoring remains a challenge. We propose a semi-supervised neural network tailored to screen DAS data related to a series of rock collapses leading to a major failure of approximately 1.2 million m3 on 15 June 2023 in Brienz, Eastern Switzerland. Besides DAS, the dataset from 16 May to 30 June 2023 includes Doppler radar data for ground-truth labeling. The proposed algorithm is capable of distinguishing between rock-slope failures and background noise, including road and train traffic, with a detection precision of over 90%. It identifies hundreds of precursory failures and shows sustained detection hours before and during the major collapse. Event size and signal-to-noise ratio (SNR) are the key performance dependencies. As a critical part of our algorithm operates unsupervised, we suggest that it is suitable for general monitoring of natural hazards.

We report a new model of radially anisotropic crustal and upper mantle structure of the western United States (WUS324) obtained from full waveform inversion of earthquake data. We ran three multiscale inversion stages beyond model WUS256 (Rodgers et al., 2022) allowing them to approach convergence to fit a larger data set to a shorter minimum period of 16 s. WUS324 is based on 324 total iterations from its starting model, significantly more (16 times) than previous studies. Waveform misfit reductions are 66-70% for both the inversion data and an independent validation data set providing confidence in the predictive power of the model. WUS324 provides much better fits and reveals shear wavespeed, vS, structure of this large region with more detail than previous waveform tomography models. We show representative images demonstrating the resolution of diverse seismic structure across this highly heterogeneous region including oceanic lithosphere, subducting slabs and continental magmatism.

We present a seismic model of the African Plate, made with the technique of full-waveform inversion. The purpose of our model is to become a foundation for future use and research, such as quantitative geodynamic interpretations, earthquake-induced ground motion predictions, and earthquake source inversion. Starting from the first-generation Collaborative Seismic Earth Model (CSEM), we invert seismograms filtered to a minimum period of 35 s and compute gradients of the misfit function with respect to the model parameters using the adjoint state method. In contrast to the conventional FWI approach, we use dynamically changing data subsets (mini-batches) of the complete dataset to compute approximate gradients at each iteration. This approach has three significant advantages: (1) it reduces computational costs for model updates and the inversion, (2) it enables the use of larger datasets without increasing iteration costs, and (3) it makes it trivial to assimilate new data since we can add it to the complete dataset without changing the misfit function, thereby enabling “evolutionary FWI". We perform 130 mini-batch iterations and invert data from 397 unique earthquakes and 184,356 unique source-receiver pairs at the cost of approximately 10 full-data iterations. We clearly image tectonic features such as the Afar triple junction. Particularly interesting are the low-velocity zones below the Hoggar, Aïr, and Tibesti Mountains, pronounced more than in earlier works. Finally, we introduce a new strategy to assess model uncertainty. We deliberately perturb the final model, perform additional mini-batch iterations, and compare the result with the original final model. This test uses actual seismic data instead of artificially generated synthetic data and requires no assumptions about the linearity of the inverse problem.

Utilizing existing telecommunication cables for Distributed Acoustic Sensing (DAS) experiments has eased the collection of seismological data in previously difficult-to-access areas such as the ocean bottom. To assess the potential of submarine DAS for monitoring seismic activity, we conducted an experiment from mid-October to mid-December 2021 using a 45 km long dark fiber extending from the Greek island of Santorini along the ocean bottom to the neighboring island of Ios. This region is of great geophysical and public interest because of its historical and recent seismic and volcanic activity, especially along the Kolumbo volcanic chain. Besides recording anthropogenic noise and around 1000 seismic events, we observe the primary and secondary microseisms in the submarine section, the latter inducing Scholte waves in a sediment layer where the cable is well-coupled. By using the spectral element wave propagation solver Salvus, we compute synthetic strains for earthquakes with varying degrees of model complexity. Despite including topography, a water layer, and a heterogeneous velocity model, we are unable to reproduce the lack of coherence in our observed earthquake waveforms. Backpropagation simulations for four observed earthquakes indicate that clear convergence of the wavefield, and thus the ability to constrain a source region, is only possible when all model complexities are considered. We conclude that, despite the promising emergence of DAS, monitoring capabilities are limited by often unfavorable cable geometries, cable coupling, and the complexity of the medium. Interrogating multiple cables simultaneously or jointly analyzing DAS and seismometer data could help improve future monitoring experiments.

We use the full waveform inversion method to study the crustal-mantle seismic structure beneath Central Asia. By combining earthquake waveforms and ambient noise cross-correlations, we construct a 3D model of Vp and Vs down to a depth of 220 km. This model reveals a complex Indian-Asian plate configuration and interaction, resulting from the plate subduction, indentation, and break-off. Beneath the Hindu Kush, the marginal Indian slab with its lower crust is successfully imaged, the latter of which hosts vigorous intermediate-depth seismicity. The subducted marginal Indian slab can be traced further east to the Kohistan Arc, which is a previously undetected structure. We first imaged a flat cratonic Indian plate beneath the Pamir. The indentation of the cratonic Indian plate forces the Asian plate to delaminate, indicated by the south-eastwards dipping high-velocity anomalies, atop which a south-dipping low-velocity zone is observed with higher resolution than previous studies, which we interpret as the delaminated Asian lower crust. In addition, a sharp velocity transition at lithospheric depth is newly discovered and coincides with the Talas-Ferghana fault, delineating the boundary of the Ferghana basin with the Central Tian Shan. Low-velocity anomalies mainly focus beneath the south and northern part of the Central Tian Shan with deep Moho, indicating the lithosphere is possibly delaminated and the deformation of the Central Tian Shan is probably concentrated at the north and south margins by the Tarim basin and Kazakh Shield, respectively. In contrast, West Tian Shan displays a simpler lithospheric structure with a single deep Moho.

Geological interpretations, earthquake source inversions and ground motion modelling, among other applications, require models that jointly resolve crustal and mantle structure. With the second generation of the Collaborative Seismic Earth Model (CSEM2), we present a global multi-resolution tomographic Earth model that serves this purpose. The model evolves through successive regional- and global-scale refinements. While the first generation aggregated regional models, with this study, we ensure consistency between all individual submodels, resulting in a model that accurately explains wave propagation across scales. Recent regional tomographic models were incorporated, comprising continental-scale inversions for Asia and Africa, as well as regional inversions for the Western US, Central Andes, Iran, and Southeast Asia. Across all regional refinements, over 793,000 unique source-receiver pairs contributed. Moreover, the long-wavelength Earth model (LOWE) introduces large-scale structures outside of pre-existing local refinements. A global full-waveform inversion over a total of 194 iterations with a minimum period of 50 s on a large data set of 2,423 earthquakes and over 6 million source-receiver pairs ensures that regional updates in the crust and uppermost mantle correctly translate into updates of deeper, global-scale structure. To test the performance of CSEM2, we evaluate waveform fits between observed and synthetic seismograms at 50 s for an independent data set on the global scale, and on the regional scale for lower periods. We show that we can accurately simulate waveforms within and across the regional refinements, maintaining the original resolution of the submodels embedded in the global framework.

Avalanches and other hazardous mass movements pose a danger to the population and critical infrastructure in alpine areas. Hence, understanding and continuously monitoring mass movements is crucial to mitigate their risk. We propose to use Distributed Acoustic Sensing (DAS) to measure strain rate along a fiber-optic cable to characterize ground deformation induced by avalanches. We recorded 12 snow avalanches of various dimensions at the Vallée de la Sionne test site in Switzerland, utilizing existing fiber-optic infrastructure and a DAS interrogation unit during the winter 2020/2021. By training a Bayesian Gaussian Mixture Model, we automatically characterize and classify avalanche-induced ground deformations using physical properties extracted from the frequency-wavenumber and frequency-velocity domain of the DAS recordings. The resulting model can estimate the probability of avalanches in the DAS data and is able to differentiate between the avalanche-generated seismic near-field, the seismo-acoustic far-field and the mass movement propagating on top of the fiber. By analyzing the mass-movement propagation signals, we are able to identify group velocity packages within an avalanche that propagate faster than the phase velocity of the avalanche front, indicating complex internal structures. Importantly, we show that the seismo-acoustic far-field can be detected before the avalanche reaches the fiber-optic array, highlighting DAS as a potential research and early warning tool for hazardous mass movements.

We demonstrate the logistic feasibility and scientific potential of Distributed Acoustic Sensing (DAS) in alpine volcano-glacial environments that are subject to a broad range of natural hazards. Our work considers the Mount Meager massif, an active volcanic complex in British Columbia, estimated to have the largest geothermal potential in Canada, and home of Canada’s largest recorded landslide in 2010. From September to October 2019, we acquired continuous strain data, using a 3 km long fiber-optic cable, deployed on a ridge of Mount Meager and on the uppermost part of a glacier above 2000 m altitude. The data analysis detected a broad range of unexpectedly intense, low-magnitude, local seismicity. The most prominent events include long-lasting, intermediate-frequency (0.01 - 1 Hz) tremor, and high-frequency (5 - 45 Hz) earthquakes that form distinct spatial clusters and often repeat with nearly identical waveforms. We conservatively estimate that the number of detectable high-frequency events varied between several tens and nearly 400 per day. We also develop a beamforming algorithm that uses the signal-to-noise ratio (SNR) of individual channels, and implicitly takes the direction-dependent sensitivity of DAS into account. Both the tremor and the high-frequency earthquakes are most likely related to fluid movement within Mount Meager’s geothermal reservoir. Our work illustrates that DAS carries the potential to reveal previously undiscovered seismicity in challenging environments, where comparably dense arrays of conventional seismometers are difficult to install. We hope that the logistics and deployment details provided here may serve as a starting point for future DAS experiments.

We develop a Hamiltonian Monte Carlo (HMC) sampler which solves a multi-parameter elastic full-waveform inversion (FWI) in a probabilistic setting for the first time. This gives novel access to the full posterior distribution for this type of highly non-linear inverse problem. Typically, FWI has focused on using gradient descent methods with proper regularization to iteratively update models to a minimum misfit value. Non-uniqueness and uncertainties are mostly in this approach. Bayesian inversions offer an alternative by assigning a probability to each model in model space given some data and prior constraints. The drawback is the need to evaluate a very large number of models. Random walks from Markov chains counter this effect by only exploring regions of model space where probability is significant. HMC method additionally incorporates gradient information, i.e. local structure, typically available for numerical waveform tomography experiments. So far, HMC has only been implented for acoustic FWI. We implement HMC for multiple 2D elastic FWI set-ups. Using parallelized wave propagation code, wavefields and kernels are computed on an regular numerical grid and projected onto basis functions. These gradients are subsequently used to explore the posterior space of different target models using HMC. The free parameters in these experiments are P and S velocity, and density. Although simulating Hamiltonian dynamics in the resulting phase space is approximated numerically, the results of the Markov chain are nevertheless very insightful. No prior tuning of kernels, data or model space is required, under the constraint that the sampler is properly tuned. After a burn in phase during which the mass matrix is iteratively optimized, the Markov chain is run on multiple nodes. After approximately 100,000 samples (combined from all nodes) the Markov chain mixes well. The resulting samples give access to the full posterior distribution, including the mean and maximum-likelihood models, conditional probabilities, inter-parameter correlations and marginal distributions.

Seismologists working with fiber-optic sensing, commonly referred to as Distributed Acoustic Sensing (DAS), have yet to find an established way of automatically detecting signals of interest within its recordings. We propose a new research perspective within the field by examining the output of a DAS array as an image and processing the image to find signals of interest. In this manuscript, we show an example of such a method, where we automatically detect seismic events of interest within two different DAS datasets, finding, respectively 99 % and 96 % of the local earthquakes previously identified within the data by manual analysis. The method is based on simple image processing and computer vision techniques, which clean the image, and, ideally, leave nothing but the signal of interest. These simple image processing steps yield promising results, indicating that computer vision and image processing might have an immediate impact in geophysical applications of fiber-optic sensing.

A new seismic velocity model for the south Central Andes is derived from full waveform inversion, covering the Pampean flat and adjacent Payenia steep subduction segments. Strong focused crustal low-velocity anomalies indicate partial melts in the Payenia segment along the volcanic arc, whereas weaker low-velocity anomalies covering a wide zone in Pampean possibly indicates remnant melts in the past. Thinning and tearing of the flat Nazca slab below the Pampean is inferred by gaps in the high-velocity slab along the inland projection of the Juan-Fernandez-Ridge. A high-velocity anomaly in the upper mantle below the flat slab is interpreted as a relic Nazca slab segment, which indicates an earlier slab break-off during the flattening process, triggered by the buoyancy of the Juan-Fernandez-Ridge. In Payenia, large-scale low-velocity anomalies atop and below the re-steepened Nazca slab are associated with the re-opening of the mantle wedge and sub-slab asthenospheric flow, respectively.

We present a new seismic tomography model for the crust and upper-mantle beneath the Central Andes based on multi-scale full seismic waveform inversion, proceeding from long periods (40–80~s) over several steps down to 12–60~s. The spatial resolution and trade-offs among inversion parameters are estimated through the multi-parameter point-spread functions. P and S wave velocity structures with a spatial resolution of 30–40 km for the upper mantle and 20 km for the crust could be resolved in the central study region. In our study, the subducting Nazca slab is clearly imaged in the upper mantle, with dip-angle variations from the north to the south. Bands of low velocities in the crust and mantle wedge indicate intense crustal partial melting and hydration of the mantle wedge beneath the frontal volcanic arc, respectively and they are linked to the vigorous dehydration from the subducting Nazca plate and intermediate depth seismicity within the slab. These low velocity bands are interrupted at 19.8º–21°S, both in the crust and uppermost mantle, hinting at the lower extent of crustal partial melting and hydration of the mantle wedge. The variation of lithospheic high velocity anomalies below the backarc from North to South allows insight into the evolutionary foundering stages of the Central Andean margin. A high velocity layer beneath the southern Altiplano suggests underthrusting of the leading edge of the Brazilian Shield. In contrast, a steeply westward dipping high velocity block and low velocity lithospheric uppermost mantle beneath the southern Puna plateau hints at the ongoing lithospheric delamination.