Chances and limits of single-station seismic event clustering by unsupervised pattern recognition

Sick, Benjamin; Guggenmos, Matthias; Joswig, Manfred

doi:10.1093/gji/ggv126

Cited by 36 publications

(17 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The methodology followed here consists of optimizing nanoseismic monitoring (Wust-Bloch and Joswig, 2006;Joswig, 2008;Sick et al, 2012Sick et al, , 2015 and validating its picking and location scheme with standard residual-based approaches. Nanoseismic monitoring was originally developed to characterize extremely weak seismicity (M L ≥ −3:0) at short slant distances (10-10 4 m; Wust-Bloch and Joswig, 2006).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Optimizing Event Detection and Location in Low‐Seismicity Zones: Case Study from Western Switzerland

Vouillamoz

Wust-Bloch

Abednego

et al. 2016

Bulletin of the Seismological Society of America

View full text Add to dashboard Cite

Obtaining robust event catalogs in regions of low seismicity can be timeconsuming, because quality events are less frequent and sensor coverage is generally sparse. Optimizing event detection and location in such regions is all the more crucial because these areas tend to host a higher density of sensitive infrastructures. The methodology proposed consists of reprocessing existing data recorded by a permanent network and boosting the final catalog resolution by temporarily deploying portable sparse mini-arrays in the target area. Sonogram analysis is applied on both existing and new datasets to detect waveforms barely emerging from the background noise. A visual interactive event analysis module is used to test for phase picking, event association, waveform cross correlation, and location ambiguities. It also estimates back azimuth and slowness when sparse array data are available. The method is applied to a low-seismicity region in the western Swiss Molasse basin where two sparse miniarrays were temporarily deployed. The detection of earthquakes is improved by a factor of 9 when reprocessing four yrs (2009-2013) of available data recorded by two accelerometers and one broadband station in a 2500 km 2 target area. Magnitude estimations are empirically calibrated over four magnitude units, down to −1:7 M L , lowering the existing catalog completeness by close to one magnitude unit. After validating picking and location accuracies with a standard residual-based scheme, 174 newly detected events are relocated, illuminating zones of previously undetected microseismic activity.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Being self-adaptive (Joswig, 1995), sonograms do not require prior information, and event detection can be carried out without an initial template, which is not the case for standard template matching techniques. Automated sonogram-based detection algorithms are presently being tested (Sick et al, 2015).…”

Section: Integrated Catalog (Fricat)mentioning

confidence: 99%

Optimizing Event Detection and Location in Low‐Seismicity Zones: Case Study from Western Switzerland

Vouillamoz

Wust-Bloch

Abednego

et al. 2016

Bulletin of the Seismological Society of America

View full text Add to dashboard Cite

show abstract

“…1) Classification of seismic events: Various studies have used machine learning for seismic events classification. Some of the methods used are: artificial neural networks [57], [58], [59], self-organizing maps [60], [61], hidden Markov models [62], [63] and support vector machines [64]. We use a data set collected from a seismic catalog by the Geophysical Institute of Israel.…”

Section: Classification In the Embedding Spacementioning

confidence: 99%

Multi-view diffusion maps

et al. 2020

View full text Add to dashboard Cite

In this paper, a reduced dimensionality representation is learned from multiple views of the processed data. These multiple views can be obtained, for example, when the same underlying process is observed using several different modalities, or measured with different instrumentation. The goal is to effectively utilize the availability of such multiple views for various purposes such as non-linear embedding, manifold learning, spectral clustering, anomaly detection and non-linear system identification. The proposed method, which is called multi-view, exploits the intrinsic relation within each view as well as the mutual relations between views. This is achieved by defining a cross-view model in which an implied random walk process is restrained to hop between objects in the different views. This multi-view method is robust to scaling and it is insensitive to small structural changes in the data. Within this framework, new diffusion distances are defined to analyze the spectra of the implied kernels. The applicability of the multi-view approach is demonstrated for clustering, classification and manifold learning using both artificial and real data. 2 The problem of learning from two views has been studied in the field of spectral clustering. Most of these studies have been focused on classification and clustering that are based on spectral characteristics of the data while using two or more sampled views. Some approaches, which address this problem, are Bilinear Model [9], Partial Least Squares [10] and Canonical Correlation Analysis [11]. These methods are powerful for learning the relation between different views but do not provide separate insights or combined into the low dimensional geometry or structure of each view. Recently, a few kernel based methods (e.g [12]) propose a model of co-regularizing kernels in both views in a way that resembles joint diagonalization. It is done by searching for an orthogonal transformation that maximizes the diagonal terms of the kernel matrices obtained from all views. A penalty term, which incorporates the disagreement between clusters from the views, was added. Their algorithm is based on alternating maximization procedure. A mixture of Markov chains is proposed in [13] to model multiple views in order to apply spectral clustering. It deals with two cases in graph theory: directed and undirected graph where the second case is related to our work. This approach converges the undirected graph problem to a Markov chains averaging where each is constructed separately within the views. A way to incorporate a given multiple metrics for the same data using a cross diffusion process is described in [14]. They define a new diffusion distance which is useful for classification, clustering or retrieval tasks. However, the proposed process is not symmetrical thus does not allow to compute an embedding. An iterative algorithm for spectral clustering is proposed in [15]. The idea is to iteratively modify each view using the representation of the other view. The problem of two manifolds, ...

show abstract

“…In the two studies, the authors comment that their algorithms produced a number of false detections, but the uncertainties associated to those false detections were much higher than those associated to real earthquakes. Unsupervised algorithms based upon regrouping event waveforms by similarity have also been explored [96,97,98].…”

Section: Earthquake Detectionmentioning

confidence: 99%

Machine learning and fault rupture: a review

Ren¹,

Hulbert²,

Johnson³

et al. 2020

Preprint

View full text Add to dashboard Cite

Geophysics has historically been a data-driven field, however in recent years the exponential increase of available data has lead to increased adoption of machine learning techniques and algorithm for analysis, detection and forecasting applications to faulting. This work reviews recent advances in the application of machine learning in the study of fault rupture ranging from the laboratory to Solid Earth.

show abstract

Chances and limits of single-station seismic event clustering by unsupervised pattern recognition

Cited by 36 publications

References 27 publications

Optimizing Event Detection and Location in Low‐Seismicity Zones: Case Study from Western Switzerland

Optimizing Event Detection and Location in Low‐Seismicity Zones: Case Study from Western Switzerland

Multi-view diffusion maps

Machine learning and fault rupture: a review

Contact Info

Product

Resources

About