Search for: All records

ABSTRACT Oceanic transform faults are a significant component of the global plate boundary system and are well known for generating fewer and smaller earthquakes than expected. Detailed studies at a handful of sites support the hypothesis that an abundance of creeping segments is responsible for most of the observed deficiency of earthquakes on those faults. We test this hypothesis on a global scale. We relocate Mw ≥5 earthquakes on 138 oceanic transform faults around the world and identify creeping segments on these faults. We demonstrate that creeping segments occur on almost all oceanic transform faults, which could explain their deficiency of earthquakes. We also find that most of the creeping segments are not associated with any large-scale geological structure such as a fault step-over, indicating that along-strike variation of fault zone properties may be the main reason for their existence. 

Earthquake clustering can be promoted by local, regional, and remote triggering. The interaction between faults by static and dynamic stress transfer has received much attention. However, the role of quasi‐static stress interaction mediated by viscoelastic flow is still poorly understood. Here, we investigate whether the tight synchronization of moment‐magnitude 6 earthquakes every about 6 years on distant asperities in the Gofar‐Discovery fault system of the East Pacific Rise may be caused by mechanical coupling within the lithosphere‐asthenosphere system. We build a three‐dimensional numerical model of seismic cycles in the framework of rate‐ and state‐dependent friction with a brittle layer overlaying a viscoelastic mantle with nonlinear rheology to simulate earthquake cycles on separate asperities. The brittle section of the West Gofar fault consists of two frictionally unstable 20 km‐long by 5 km‐wide asperities separated by a velocity‐strengthening barrier, consistent with seismic observations, allowing stress transfer by afterslip and viscoelastic relaxation. We find that viscoelastic stress transfer can promote the synchronization of earthquakes. Even if the asperities are separated by as far as 30 km, synchronization is still possible for a viscosity of the underlying mantle of 10¹⁷ Pa s, which can be attained by dislocation creep or transient creep during the postseismic period. Considering the similarities in tectonic and structural settings, viscoelastic stress transfer and earthquake synchronization may also occur at 15’20 (Mid‐Atlantic Ridge), George V (Southeast Indian Ridge), Menard and Heezen transform fault (Pacific‐Antarctic Ridge).

 

Deep learning models have achieved state-of-the-art performance in semantic image segmentation, but the results provided by fully automatic algorithms are not always guaranteed satisfactory to users. Interactive segmentation offers a solution by accepting user annotations on selective areas of the images to refine the segmentation results. However, most existing models only focus on correcting the current image’s misclassified pixels, with no knowledge carried over to other images. In this work, we formulate interactive image segmentation as a continual learning problem and propose a framework to effectively learn from user annotations, aiming to improve the segmentation on both the current image and unseen images in future tasks while avoiding deteriorated performance on previously-seen images. It employs a probabilistic mask to control the neural network’s kernel activation and extract the most suitable features for segmenting images in each task. We also design a task-aware architecture to automatically infer the optimal kernel activation for initial segmentation and subsequent refinement. Interactions with users are guided through multi-source uncertainty estimation so that users can focus on the most important areas to minimize the overall manual annotation effort. Extensive experiments are performed on both medical and natural image datasets to illustrate the proposed framework’s effectiveness on basic segmentation performance, forward knowledge transfer, and backward knowledge transfer. 

We propose to jointly analyze experts' eye movements and verbal narrations to discover important and interpretable knowledge patterns to better understand their decision-making processes. The discovered patterns can further enhance data-driven statistical models by fusing experts' domain knowledge to support complex human-machine collaborative decision-making. Our key contribution is a novel dynamic Bayesian nonparametric model that assigns latent knowledge patterns into key phases involved in complex decision-making. Each phase is characterized by a unique distribution of word topics discovered from verbal narrations and their dynamic interactions with eye movement patterns, indicating experts' special perceptual behavior within a given decision-making stage. A new split-merge-switch sampler is developed to efficiently explore the posterior state space with an improved mixing rate. Case studies on diagnostic error prediction and disease morphology categorization help demonstrate the effectiveness of the proposed model and discovered knowledge patterns. 

Synchronization behavior of large earthquakes (rupture of nearby faults close in time for many cycles) has been reported in many fault systems. The general idea is that the faults in the system have similar repeating intervals and are positively coupled through stress interaction. However, many details of such synchronization remain unknown. Here, we built a numerical model in the framework of rate‐and‐state friction to simulate earthquake cycles on the west Gofar fault, East Pacific Rise. Our model consists of two seismic patches separated by a barrier patch, which are constrained by seismic observations. We varied the parameters in the barrier to understand its role on earthquake synchronization. First, we found that when the barrier is relatively weak, synchronization can be achieved by afterslip or post‐seismic creep in the barrier patch. Second, static stress transfer can lead to synchronization, opposite to the suggestion by Scholz (2010,https://doi.org/10.1785/0120090309), which was based on results from a spring‐slider model using rate‐and‐state friction. Third, the width of the barrier is more important than its strength. When the barrier is narrow enough (no more than half the width of the seismic patch in our model), the system can achieve synchronization even with a very strong barrier. Fourth, for certain simulations, the interaction between the two seismic patches promotes partial rupture in the seismic patches and leads to complex behavior: the system switches from synchronized to unsynchronized over 10–20 cycles.

 

In this study, we explore the use of low rank and sparse constraints for the noninvasive estimation of epicardial and endocardial extracellular potentials from body-surface electrocardiographic data to locate the focus of premature ventricular contractions (PVCs). The proposed strategy formulates the dynamic spatiotemporal distribution of cardiac potentials by means of low rank and sparse decomposition, where the low rank term represents the smooth background and the anomalous potentials are extracted in the sparse matrix. Compared to the most previous potential-based approaches, the proposed low rank and sparse constraints are batch spatiotemporal constraints that capture the underlying relationship of dynamic potentials. The resulting optimization problem is solved using alternating direction method of multipliers . Three sets of simulation experiments with eight different ventricular pacing sites demonstrate that the proposed model outperforms the existing Tikhonov regularization (zero-order, second-order) and L1-norm based method at accurately reconstructing the potentials and locating the ventricular pacing sites. Experiments on a total of 39 cases of real PVC data also validate the ability of the proposed method to correctly locate ectopic pacing sites.