GroningenNet: Deep Learning for Low-Magnitude Earthquake Detection on a Multi-Level Sensor Network

Automatic detection of low-magnitude earthquakes has become an increasingly important research topic in recent years due to a sharp increase in induced seismicity around the globe. The detection of low-magnitude seismic events is essential for microseismic monitoring of hydraulic fracturing, carbon capture and storage, and geothermal operations for hazard detection and mitigation. Moreover, the detection of micro-earthquakes is crucial to understanding the underlying mechanisms of larger earthquakes. Various algorithms, including deep learning methods, have been proposed over the years to detect such low-magnitude events. However, there is still a need for improving the robustness of these methods in discriminating between local sources of noise and weak seismic events. In this study, we propose a convolutional neural network (CNN) to detect seismic events from shallow borehole stations in Groningen, the Netherlands. We train a CNN model to detect low-magnitude earthquakes, harnessing the multi-level sensor configuration of the G-network in Groningen. Each G-network station consists of four geophones at depths of 50, 100, 150, and 200 m. Unlike prior deep learning approaches that use 3-component seismic records only at a single sensor level, we use records from the entire borehole as one training example. This allows us to train the CNN model using moveout patterns of the energy traveling across the borehole sensors to discriminate between events originating in the subsurface and local noise arriving from the surface. We compare the prediction accuracy of our trained CNN model to that of the STA/LTA and template matching algorithms on a two-month continuous record. We demonstrate that the CNN model shows significantly better performance than STA/LTA and template matching in detecting new events missing from the catalog and minimizing false detections. Moreover, we find that using the moveout feature allows us to effectively train our CNN model using only a fraction of the data that would be needed otherwise, saving plenty of manual labor in preparing training labels. The proposed approach can be easily applied to other microseismic monitoring networks with multi-level sensors.

The Long-Term Residual Effects of Low-Magnitude Mechanical Stimulation On Murine Femoral Mechanics

As an alternative to drug treatments, low-magnitude mechanical stimulation (LMMS) may improve skeletal health without potential side effects from drugs. LMMS has been shown to increase bone health short term in both animal and clinical studies. Long term changes to the mechanical properties of bone from LMMS are currently unknown, so the objective of this research is to investigate the long-term effects of whole body vibration therapy on the elastic and viscoelastic properties of bone. In this study 10-week old female BALB/cByJ mice were given LMMS (15 min/day, 5 days/week, 0.3 g, 90 Hz) for 8 weeks; SHAM did not receive LMMS. Two sets of groups remained on study for an additional 8 or 16 weeks post LMMS (N=17). MicroCT and histomorphology of these femurs were studied and results were published by Bodnyk et al. [1]. Femoral quasi-static bending stiffness trended 4.2% increase in stiffness after 8-weeks of LMMS and 1.3% increase 8-weeks post LMMS compared to SHAM. Damping, tan delta, and loss stiffness, significantly increased by 17.6%, 16.3%, and 16.6% respectively at 8 weeks LMMS compared to SHAM. Finite element models of applied LMMS signal showed decreased stress in the mid-diaphyseal region at both 8-week LMMS and 8-week post LMMS compared to SHAM. Residual mechanical changes in bone during and post-LMMS indicates that LMMS could be used to increase long-term mechanical integrity of bone.

Protective effects of low-magnitude high-frequency vibration on high glucose-induced osteoblast dysfunction and bone loss in diabetic rats

Objective Low-magnitude high-frequency vibration (LMHFV) has been reported to be capable of promoting osteoblast proliferation and differentiation. Reduced osteoblast activity and impaired bone formation were related to diabetic bone loss. We investigated the potential protective effects of LMHFV on high-glucose (HG)-induced osteoblasts in this study. In addition, the assessment of LMHFV treatment for bone loss attributed to diabetes was also performed in vivo. Method MC3T3-E1 cells induced by HG only or treated with LMHFV were treated in vitro. The experiments performed in this study included the detection of cell proliferation, migration and differentiation, as well as protein expression. Diabetic bone loss induced by streptozotocin (STZ) in rats was established. Combined with bone morphometric, microstructure, biomechanical properties and matrix composition tests, the potential of LMHFV in treating diabetes bone loss was explored. Results After the application of LMHFV, the inhibiting effects of HG on the proliferation, migration and differentiation of osteoblasts were alleviated. The GSK3β/β-catenin pathway was involved in the protective effect of LMHFV. Impaired microstructure and biomechanical properties attributed to diabetes were ameliorated by LMHFV treatment. The improvement of femur biomechanical properties might be associated with the alteration of the matrix composition by the LMHFV. Conclusion LMHFV exhibited a protective effect on osteoblasts against HG by regulating the proliferation, migration and differentiation of osteoblasts. The function of promoting bone formation and reinforcing bone strength made it possible for LMHFV to alleviate diabetic bone loss.

The Understudied Risks of Low-Magnitude Eruptions

Volcanologists have historically focused on the risks of large-scale eruptions, but new research highlights how small eruptions can combine with human-made vulnerabilities to cause catastrophic impacts.

Full-Wavefield, Full-Domain Deterministic Modeling of Shallow Low-Magnitude Events for Improving Regional Ground-Motion Predictions

ABSTRACT At the fundamental level, seismic risk analysis relies on good modeling tools for predicting the ground motion resulting from hypothetical earthquake events, which is traditionally approximated using many variations of ground-motion prediction equations (GMPEs). The main benefit of these equations lies in their low computational cost, allowing one to run Monte Carlo simulations in which event probabilities are dictated by regional catalogs comprising historical observations. These equations, however, rely on approximations that are only accurate in a statistical sense. In this study, we consider cases in which regional high-resolution 3D earth models are available from exploration reflection seismology. These high-fidelity velocity models allow us to perform deterministic elastic ground-motion simulations at local distances, given a prescribed synthetic earthquake event, and compare the results with those predicted by GMPEs. This full-wavefield full-domain modeling approach is significantly more costly and particularly challenging due to the slow shear-wave velocity at the near surface, which requires fine spatial and temporal discretizations. With the aid of powerful computational resources, we use an adaptive mesh generator and an efficient wave solver to model the 3D elastic and anelastic wave propagation from the hypocenter all the way to the ground surface. This approach can simultaneously account for 3D subsurface structures, near-surface site effects, topographic relief, and the radiation pattern of the source. In areas where observations are sparse, the modeling results can fill the gap between stations and provide a test bed that can be used for improving the development and accuracy of GMPEs. This approach is well suited for areas where shallow low-magnitude-induced seismic events can occur. Lastly, to demonstrate our approach, we consider an observed seismic event at the Groningen gas field and compare the recorded ground motions with both—those predicted by our approach and those predicted by GMPEs.

Increase of landslide activity after a low magnitude earthquake inferred by DInSAR interferometry

On August 16th, 2018, an Mw 5.1 earthquake struck the Molise region (central Italy), inducing 84 earthquake-triggered landslides that involved soil covers of clayey materials and flysch on gently-dip slopes predominantly. To quantify the spatio-temporal landslide activity in the months immediately after the earthquake, a Differential SAR Interferometry (DInSAR) analysis was carried out in a time span comprising two years before the earthquake and one followed, recognising both first-time and reactivated landslides. The results showed a clear increase in landslide activity following the low magnitude earthquake occurrence with respect to the one recorded in the same months of the previous years. Several coherent landslides (earth slides and earth flows) were observed following the seasonally recurrent rainfall events. Such an increase was observed for both reactivations and first-time landslides, showing a decrease of inactivity period as well as activity over wider periods. Furthermore, spatial density distribution of the landslides was investigated in the post-seismic time along transepts perpendicular and parallel to the direction of the tectonic element responsible for the seismic event, respectively. An asymmetrical distribution was deduced parallel to the fault strike with the higher number of landslides located inside the compressional sector according to a strike-slip faulting mechanism.

EFFECTIVENESS OF VIBRATION (CYCLIC LOADING) IN ACCELERATING BONE REMODELING AND ORTHODONTIC TOOTH MOOVEMENT: A SHORT REVIEW

This paper assesses the effectiveness of vibration in accelerating bone remodeling and orthodontic tooth movement. Databases of PubMed, Web of Science, and ScienceDirect were searched from January 2017 to March 2019 for randomized or quasi-randomized controlled trials that evaluated the effectiveness of vibration in accelerating bone remodeling and orthodontic tooth movement. The inclusion criteria were as follows: (i) studies that assessed the efficacy of vibration (cyclic loading) in bone remodeling and orthodontic tooth movement and (ii) those that employed groupings (experimental vs. control/placebo groups) on the basis of the use of vibration (cyclic loading). Eight clinical trials were included in this short review. Five studies met the eligibility criteria for bone remodeling and orthodontic tooth movement. Four studies found that low-magnitude high-frequency vibration could accelerate bone remodeling. However, contradictory results were obtained with regard to the acceleration of orthodontic tooth movement by vibration in human participants. Low-magnitude high-frequency vibration can accelerate bone remodeling and orthodontic tooth movement. However, this acceleration is dependent on the magnitude and frequency. Further research is necessary to determine the most feasible protocols for investigating the effects of magnitude and frequency of vibration on the acceleration of orthodontic tooth movement in human participants.