Abstract An earthquake with a reported magnitude of 4.4 ( M L ) was detected on 13 June 2015 in western central Alberta, Canada. This event the third felt this year near Fox Creek, shale gas exploration region. Our results from full moment tensor inversions regional broadband data show strong strike‐slip mechanism near‐vertical fault plane solutions. The decomposition solution is overwhelmingly double couple, while only modest (∼20%) contribution attributed to...

The 28 September 2004 Parkfield earthquake, arguably the best recorded earthquake ever, allows for detailed investigation of finite‐source models and their resolution. We have developed using GPS interferometric synthetic aperture radar geodetic data seismic strong motion waveform ( f ≤ 0.5 Hz) both independently combined. preferred model shows that rupture is predominantly unilateral to NW with a small component SE. Slip concentrated into two primary patches, one near hypocenter other...

More than 60 small earthquakes ( M L 0.7–3.0) were detected from December 2011 to March 2012 north of Cardston, Alberta, an area with little evidence for previous seismic activity. The timing these events closely correlates (>99.7% confidence) hydraulic fracturing completions the Devonian–Mississippian-age Exshaw Formation at a nearby horizontal well. Unanimous waveform multiplicity within swarm suggests that share similar origin and source mechanism. This observation is corroborated by...

On 2016 January 12, an intraplate earthquake with initial reported local magnitude (ML) of 4.8 shook the town Fox Creek, Alberta. While there were no damages, this was widely felt by residents and suspected to be induced nearby hydraulic-fracturing (HF) operations. In study, we determine source parameters using moment tensor inversions, then detect locate associated swarm a waveform cross-correlation based method. The broad-band seismic recordings from regional arrays suggest (M) 4.1 for...

Abstract Finite‐source inversions are performed using small earthquake waveforms as empirical Green's functions (eGf) to investigate the rupture process of repeating earthquakes along San Andreas Fault in Parkfield, California. The eGf waveform inversion method is applied a M w 2.1 Parkfield sequence three‐component velocity recorded by an array borehole seismometers. obtained models show circular slip distribution with ~20 m radius, 3.0–4.2 cm average main asperity, and peak displacement...

Abstract In volcanic regions, active earthquake swarms often occur in association with activity, and their rapid detection analysis are crucial for volcano disaster prevention. Currently, these processes ultimately left to human judgment require significant time money, making detailed real-time verification impossible. To overcome this issue, we attempted apply machine learning, which has been successfully applied various seismological fields date. For seismic phase pick, several models have...

Abstract Volcanic earthquakes provide essential information for evaluating volcanic activity. Because are often characterized by swarm-like features, conventional methods using manual picking require considerable time to construct seismic catalogs. In this study, a machine learning framework and trained model from earthquake catalog, we obtained detailed picture of during the past 12 years at Kirishima volcano, southwestern Japan. We detected ~ 6.2 times as many catalog high-resolution...

Microseismicity generated during stimulation of two horizontal wellbores in the Fayetteville shale was monitored simultaneously using a surface array, grid shallow wells and an extensive downhole array. Analysis is provided noise signal wavefields from reservoir depth to surface.

Abstract The rupture process of two M 4 repeating earthquake sequences in eastern Taiwan with contrasting recurrence behavior is investigated to demonstrate a link between slip heterogeneity and recurrence. 3.6–3.8 quasiperiodic earthquakes characterized by 3 years interval reveal overlapped concentrations. Inferred distribution for each event illustrates asperities peak 47.7 cm stress drop 151.1 MPa. Under the influence nearby 6.9 event, 4.3–4.8 separated only 6–87 min, however, an...

Southeast (SE) Asia is a tectonically complex region surrounded by many active source regions, thus an ideal test bed for developments in seismic tomography. Much recent development tomography has been based on 3-D sensitivity kernels the first-order Born approximation, but there are potential problems with this approach when applied to waveform data. In study, we develop radially anisotropic model of SE using long-period multimode waveforms. We use theoretical ‘cascade’ approach, starting...

A microseismic source mechanism is described by its moment tensor which used to define how the rock breaks. However estimation often biased various errors, such as event mis‐location, incorrectly accounting for path effects, background noise and receiver coverage. In this study, using constrained full waveform inversion, sensitivity of errors stated above examined. The impact potential on interpretation whether microseismicity corresponds fracture shearing or tensile opening illustrated.

Volcanic earthquakes provide essential information for evaluating volcanic activity. As are often characterized by swarm-like features, conventional methods using manual picking require much time in constructing seismic catalogs. In this study, a machine learning framework and trained model from earthquake catalog, we obtained detailed picture of during the past 12 years at Kirishima volcano, southwestern Japan. We could detect about 7.5 times larger than those catalog obtain high-resolution...

Precisely classifying earthquake types is crucial for elucidating the relationship between volcanic earthquakes and activity. However, traditional methods rely on subjective human judgment, which requires considerable time effort. To improve this, we developed a deep learning model using transformer encoder more objective efficient classification. Tested Mount Asama’s diverse seismic activity, our achieved high F1 scores (0.876 tectonic, 0.964 low-frequency earthquakes, 0.995 noise),...

Precisely classifying earthquake types is crucial for elucidating the relationship between volcanic earthquakes and activity. However, traditional methods rely on subjective human judgment, which requires considerable time effort. To improve this, we developed a deep learning model using transformer encoder more objective efficient classification. Tested Mount Asama’s diverse seismic activity, our achieved high F1 scores (0.876 tectonic, 0.964 low-frequency earthquakes, 0.995 noise),...

Understanding the source characteristics of hydraulic‐fracturing‐induced microearthquakes is expected to provide better understanding both fracturing process and influence pre‐existing structures on distribution events. However, details remain largely unknown. One controversial issue whether mechanism such events differs from natural tectonic events, that is, significant reductions effective stress and/or volumetric change occur because stimulation. Herein, address this question, we...

We introduce the 3-D attenuation tomographic approach that combines frequency-domain inversions scheme consisting of a set sub-inversions and site effect estimation based on residual spectra. apply it to data observed at Japanese Islands estimate three-dimensional P wave structure beneath islands. The results show Qp-1 distribution is highly variable over entire Islands, with spatial wavelength 2-3 degrees. This heterogeneous material property underground may be related tectonic state Islands.

Recognizing seismic waves immediately is very important for the realization of efficient disaster prevention. Generally these systems consist a network detectors that send real time data to central server. The server elaborates and attempts recognize first signs an earthquake. current problem with this approach it subject false alarms. A critical trade-off exists between sensitivity system error rate. To overcame problems, artificial neural based intelligent learning can be used. However,...

Abstract In volcanic regions, active earthquake swarms often occur associated with activity, and their rapid detection measurement are crucial for volcano disaster prevention. Currently, however, these processes ultimately left to human judgment require much time money, making detailed verification in real impossible. To overcome this issue, we attempted apply machine learning, which has been studied many seismological fields recent years. Several models have already trained using a large...

In this study, we extended double-couple constarained focal mechanism inversion code developed by Snoke (2003) to retrieve non-double couple component from hydraulic fracturing induced microearthquakes. We constrained the be shear slip, tensile displacement, or combination of these mechanisms, since it is reasonable model for stimulation microseismicity. addition, has smaller parameters stabilize than that full moment tensor case. Synthetic test performed under various condition and showed...