Search for: All records

Abstract Low‐frequency earthquakes are a seismic manifestation of slow fault slip. Their emergent onsets, low amplitudes, and unique frequency characteristics make these events difficult to detect in continuous seismic data. Here, we train a convolutional neural network to detect low‐frequency earthquakes near Parkfield, CA using the catalog of Shelly (2017),https://doi.org/10.1002/2017jb014047as training data. We explore how varying model size and targets influence the performance of the resulting network. Our preferred network has a peak accuracy of 85% and can reliably pick low‐frequency earthquake (LFE) S‐wave arrival times on single station records. We demonstrate the abilities of the network using data from permanent and temporary stations near Parkfield, and show that it detects new LFEs that are not part of the Shelly (2017),https://doi.org/10.1002/2017jb014047catalog. Overall, machine‐learning approaches show great promise for identifying additional low‐frequency earthquake sources. The technique is fast, generalizable, and does not require sources to repeat. 

Low-frequency earthquakes (LFEs) are small-magnitude earthquakes that are depleted in high-frequency content relative to traditional earthquakes of the same magnitude. These events occur in conjunction with slow slip events (SSEs) and can be used to infer the space and time evolution of SSEs. However, because LFEs have weak signals, and the methods used to identify them are computationally expensive, LFEs are not routinely cataloged in most places. Here, we develop a deep-learning model that learns from an existing LFE catalog to detect LFEs in 14 years of continuous waveform data in southern Vancouver Island. The result shows significant increases in detection rates at individual stations. We associate the detections and locate them using a grid search approach in a 3D regional velocity model, resulting in over 1 million LFEs during the performing period. Our resulting catalog is consistent with a widely used tremor catalog during periods of large-magnitude SSEs. However, there are time periods where it registers far more LFEs than the tremor catalog. We highlight a 16-day period in May 2010, when our model detects nearly 3,000 LFEs, whereas the tremor catalog contains only one tremor detection in the same region. This suggests the possibility of hidden small-magnitude SSEs that are undetected by current approaches. Our approach improves the temporal and spatial resolution of the LFE activities and provides new opportunities to understand deep subduction zone processes in this region. 

Deep long-period earthquakes (DLPs) are an enigmatic type of volcanic seismicity that sometimes precedes eruptions but mostly occurs at quiescent volcanoes. These earthquakes are depleted in high-frequency content and typically occur near the base of the crust. We observed a near-periodic, long-lived sequence of more than one million DLPs in the past 19 years beneath the dormant postshield Mauna Kea volcano in Hawaiʻi. We argue that this DLP sequence was caused by repeated pressurization of volatiles exsolved through crystallization of cooling magma stalled beneath the crust. This “second boiling” of magma is a well-known process but has not previously been linked to DLP activity. Our observations suggest that, rather than portending eruptions, global DLP activity may more commonly be indicative of stagnant, cooling magma. 

Abstract We deployed a network of 68 three-component geophones on the slow-moving Two Towers earthflow in northern California. We compute horizontal-to-vertical spectral ratios (HVSRs) from the ambient seismic field. The HVSRs have two prominent peaks, one near 1.23 Hz and another between 4 and 8 Hz at most stations. The 1.23 Hz resonance is a property of the background noise field and may be due to a velocity contrast at a few hundred meters depth. We interpret the higher frequency peaks as being related to slide deposits and invert the spectral ratios for shallow velocity structure using in situ thickness measurements as a priori constraints on the inversion. The thickness of the shallowest, low-velocity layer is systematically larger than landslide thicknesses inferred from inclinometer data acquired since 2013. Given constraints from field observations and boreholes, the inversion may reflect the thickness of deposits of an older slide that is larger in spatial extent and depth than the currently active slide. Because the HVSR peaks measured at Two Towers are caused by shallow slide deposits and represent frequencies that will experience amplification during earthquakes, the depth of the actively sliding mass may be less relevant for assessing potential slide volume and associated hazard than the thicknesses determined by our inversions. More generally, our results underscore the utility of combining both geotechnical measurements and subsurface imaging for landslide characterization and hazard assessment.