Physics

Public earthquake early warning systems have the potential to reduce individual risk by warning people of an incoming tremor but their development has been hampered by costly infrastructure. Furthermore, users' understanding of such a service and their reactions to warnings remains poorly studied. The smartphone app of the Earthquake Network initiative turns users' smartphones into motion detectors and provides the first example of purely smartphone-based earthquake early warnings, without the need for dedicated seismic station infrastructure and operating in multiple countries. We demonstrate here that early warnings have been emitted in multiple countries even for damaging shaking levels and so this offers an alternative in the many regions unlikely to be covered by conventional early warning systems in the foreseeable future. We also show that although warnings are understood and appreciated by users, notably to get psychologically prepared, only a fraction take protective actions such as "drop, cover and hold".

Classically, anisotropic surface wave tomography is treated as an optimisation problem where it proceeds through a linearised two-step approach. It involves the construction of 2D group or phase velocity maps for each considered period, followed by the inversion of local dispersion curves inferred from these maps for 1D depth-functions of the elastic parameters. Here, we cast the second step into a fully Bayesian probability framework. Solutions to the inverse problem are thus an ensemble of model parameters (\textit{i.e.} 1D elastic structures) distributed according to a posterior probability density function and their corresponding uncertainty limits. The method is applied to azimuthally-varying synthetic surface wave dispersion curves generated by a 3D-deforming upper mantle. We show that such a procedure captures essential features of the upper mantle structure. The robustness of these features however strongly depend on the wavelength of the wavefield considered and the choice of the model parameterisation. Additional information should therefore be incorporated to regularise the problem such as the imposition of petrological constraints to match the geodynamic predictions.

In recent years, a variety of Marchenko methods for the attenuation of internal multiples has been developed. These methods have been extensively tested on 2D synthetic data and applied to 2D field data, but only little is known about their behaviour on 3D synthetic data and 3D field data. Particularly, it is not known whether Marchenko methods are sufficiently robust for sparse acquisition geometries that are found in practice. Therefore, we start by performing a series of synthetic tests to identify the key acquisition parameters and limitations that affect the result of 3D Marchenko internal multiple prediction and subtraction using an adaptive double-focusing method. Based on these tests, we define an interpolation strategy and use it for the field data application.

Although many 3D One-Way Wave-equation Migration (OWEM) methods exist for VTI media, most of them struggle either with the stability, the anisotropic noise or the computational cost. In this paper we present a new method based on a mixed space- and wavenumber-propagator that overcome these issues very effectively as demonstrated by the examples. The pioneering methods of phase-shift (PS) and Stolt migration in the frequency-wavenumber domain designed for laterally homogeneous media have been followed by several extensions for laterally inhomogeneous media. Referred many times to as phase-screen or generalized phase-screen methods, such extensions include as main examples of the Split-step Fourier (SSF) and the phase-shift plus interpolation (PSPI). To further refine such phase-screen techniques, we introduce a higher-order extension to SSF valid for a 3D VTI medium with large lateral contrasts in vertical velocity and anisotropy parameters. The method is denoted Fourier Mixed-Domain (FMD) prestack depth migration and can be regarded as a stable explicit algorithm. The FMD technique was tested using the 3D SEG/EAGE salt model and the 2D anisotropic Hess model with good results. Finally, FMD was applied with success to a 3D field data set from the Barents Sea including anisotropy.

The stability of dry-snow avalanches is strongly dependent on the interaction between the snow slab above a weak-layer and, as presented in this work, the skier induced load. This induced load causes an additional stress field on the slab which eventually triggers an avalanche. I present the results of 3D finite element simulations in an elastoplastic domain. The plastic deformation of the weak-layer follows the Mohr-Coulomb-Cap model which provides a more realistic model as a pure elastic approach. I investigate how the stress field on top of the weak-layer changes if one is skiing down-slope or parallel to the slope. A layered snow slab changes the stress on top of the weak-layer and to investigate these changes I simulated two different representative layered slabs. One containing only soft layers to investigate how the weak layer is affected by the ski induced stress and the other hard-soft-hard layer to examine bridge effects caused by the hard layers. A hard layer in the snow slab forms a sort of bridge which spreads the induced stress over a larger lateral distance, at the same time decreasing the stress to the layers below the bridge. Furthermore, I show a possible connection between the plastic deformation and the critical crack length.

We create virtual sources and receivers in a 3D subsurface using the previously derived single-sided homogeneous Green's function representation. We employ Green's functions and focusing functions that are obtained using reflection data at the Earth's surface, a macro velocity model and the Marchenko method. The homogeneous Green's function is a Green's function superposed with its time-reversal. Unlike the classical homogeneous Green's function representation, our approach requires no receivers on an enclosing boundary, however, it does require the source signal to be symmetric in time. We demonstrate that in 3D, the single-sided representation is an improvement over the classical representation by applying the representations to numerical data. We retrieve responses to virtual point sources with an isotropic and with a double-couple radiation pattern and compare the results to a directly modeled reference result. We also demonstrate the application of the single-sided representation for retrieving the response to a virtual rupture that consists of a superposition of double-couple point sources. This is achieved by obtaining the homogeneous Green's function for each source separately, before they are transformed to the causal Green's function, time-shifted and superposed. The single-sided representation is also used to monitor the complete wavefield that is caused by a numerically modeled rupture. However, the source signal of an actual rupture is not symmetric in time and the single-sided represenation can therefore only be used to obtain the causal Green's function. This approach leaves artifacts in the final result, however, these artifacts are limited in space and time.

This review gives an overview of the development of machine learning in geoscience. A thorough analysis of the co-developments of machine learning applications throughout the last 70 years relates the recent enthusiasm for machine learning to developments in geoscience. I explore the shift of kriging towards a mainstream machine learning method and the historic application of neural networks in geoscience, following the general trend of machine learning enthusiasm through the decades. Furthermore, this chapter explores the shift from mathematical fundamentals and knowledge in software development towards skills in model validation, applied statistics, and integrated subject matter expertise. The review is interspersed with code examples to complement the theoretical foundations and illustrate model validation and machine learning explainability for science. The scope of this review includes various shallow machine learning methods, e.g. Decision Trees, Random Forests, Support-Vector Machines, and Gaussian Processes, as well as, deep neural networks, including feed-forward neural networks, convolutional neural networks, recurrent neural networks and generative adversarial networks. Regarding geoscience, the review has a bias towards geophysics but aims to strike a balance with geochemistry, geostatistics, and geology, however excludes remote sensing, as this would exceed the scope. In general, I aim to provide context for the recent enthusiasm surrounding deep learning with respect to research, hardware, and software developments that enable successful application of shallow and deep machine learning in all disciplines of Earth science.

We present radiometric 81 Kr dating results for ice samples collected at the outlets of the Guliya ice cap in the western Kunlun Mountains of the Tibetan Plateau. This first application of 81 Kr dating on mid-latitude glacier ice was made possible by recent advances in Atom Trap Trace Analysis, particularly a reduction in the required sample size down to 1 μ L STP of krypton. Eight ice blocks were sampled from the bottom of the glacier at three different sites along the southern edges. The 81 Kr data yield upper age limits in the range of 15-74 ka (90% confidence level). This is an order of magnitude lower than the ages exceeding 500 ka which the previous 36 Cl data suggest for the bottom of the Guliya ice core. It is also significantly lower than the widely used chronology up to 110 ka established for the upper part of the core based on δ 18 O in the ice.

The global volume of digital data is expected to reach 175 zettabytes by 2025. The volume, variety, and velocity of water-related data are increasing due to large-scale sensor networks and increased attention to topics such as disaster response, water resources management, and climate change. Combined with the growing availability of computational resources and popularity of deep learning, these data are transformed into actionable and practical knowledge, revolutionizing the water industry. In this article, a systematic review of literature is conducted to identify existing research which incorporates deep learning methods in the water sector, with regard to monitoring, management, governance and communication of water resources. The study provides a comprehensive review of state-of-the-art deep learning approaches used in the water industry for generation, prediction, enhancement, and classification tasks, and serves as a guide for how to utilize available deep learning methods for future water resources challenges. Key issues and challenges in the application of these techniques in the water domain are discussed, including the ethics of these technologies for decision-making in water resources management and governance. Finally, we provide recommendations and future directions for the application of deep learning models in hydrology and water resources.

Reflection seismic imaging usually suffers from a loss of resolution and contrast because of the fluctuations of the wave velocities in the Earth's crust. In the literature, phase distortion issues are generally circumvented by means of a background wave velocity model. However, it requires a prior tomography of the wave velocity distribution in the medium, which is often not possible, especially in depth. In this paper, a matrix approach of seismic imaging is developed to retrieve a three-dimensional image of the subsoil, despite a rough knowledge of the background wave velocity. To do so, passive noise cross-correlations between geophones of a seismic array are investigated under a matrix formalism. More precisely, the detrimental effect of wave velocity fluctuations on imaging is overcome by introducing a novel mathematical object: The distortion matrix. This operator essentially connects any virtual source inside the medium with the distortion that a wavefront, emitted from that point, experiences due to heterogeneities. A time reversal analysis of the distortion matrix enables the estimation of the transmission matrix that links each real geophone at the surface and each virtual geophone in depth. Phase distortions can then be compensated for any point of the underground. Applied to seismic data recorded along the Clark branch of the San Jacinto fault zone, the present method is shown to provide an image of the fault until a depth of 4 km with a transverse resolution of 80 m. Strikingly, this resolution is almost one eighth below the diffraction limit imposed by the geophone array aperture. The heterogeneities of the subsoil play the role of a scattering lens and of a transverse wave guide which increase drastically the array aperture.