Geology Reports⌕ Search
Home/Archive/USGS

USGS · 70230358

Identification of low-frequency earthquakes on the San Andreas fault with deep learning

A. M. Thomas·A. Inbal·J. Searcy·David R. Shelly·R. Bürgmann
View original source ↗Export metadata ↓

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/10.1002/2017jb014047 as 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/10.1002/2017jb014047 catalog. 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.

Explore related subjects

Geophysical Research LettersArticle

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

A. M. Thomas, A. Inbal, J. Searcy, David R. Shelly, R. Bürgmann. 2021-07-09. Identification of low-frequency earthquakes on the San Andreas fault with deep learning. https://doi.org/10.1029/2021gl093157

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Emergent headwater types for watershed monitoring, research, and planning

Headwaters—that is headwater streams and the watersheds they drain—strongly influence water quality, ecosystem services, and hydrologic connectivity, yet remain poorly mapped, monitored, and understood at large spatial scales. Using available spatial data to classify headwater “types” could therefore enhance the utility of sparse monitoring data to advance research and watershed planning—and help update literature-based perceptual models of headwaters as steep, forested, and minimally disturbed watersheds. Here, we analyze headwaters across the contiguous United States to quantify how their attributes (climate, topography, land cover, soils) vary regionally and differ from their downstream watersheds. Our analysis (a) identified seven dominant headwater types with distinct combinations of biophysical and climatic attributes and (b) revealed systematic contrasts with downstream watersheds, with strongest differences in attributes representing potential flowpath depth. Our findings offer a transferable framework for classifying, modeling, and managing these critical yet understudied components of the hydrologic network.

conterminous United States↗
USGS2026

Corundum discovered by SuperCam and the Perseverance rover at Jezero crater, Mars

Mars is primarily composed of mafic mineral assemblages and their alteration products, but small, scattered rocks strewn across the landscape offer clues to greater petrological diversity. While traversing the Jezero crater rim, the Perseverance rover encountered several plagioclase-rich light-toned float rocks. SuperCam identified the distinctive signature of corundum (α-Al 2 O 3 ) in these rocks using time-resolved luminescence spectroscopy. Two strong peaks (692.7 and 694.1 nm) with millisecond lifetimes, and additional supporting lines, are consistent with Cr 3+ substitution for Al 3+ in corundum. Corundum forms in Al-rich, Si-depleted environments through magmatic or metamorphic processes. Given the rocks' small size, association with plagioclase, and location on the crater rim, we interpret the most plausible formation scenario to be impact induced metamorphism at the interface of a felsic and a mafic/ultramafic member with the likely action of fluids at some stage, although other possibilities are not excluded.

Geophysical Research Letters↗
USGS2026

Subsurface vertical connectivity shapes solute transport to montane streams: Insights from watershed-scale geophysics-informed modeling

Headwater catchments provide essential water and nutrients to downstream ecosystems. The timing and evolution of their transport is shaped in part by the “invisible” subsurface structure but have been largely unexplored at the watershed scale. This raises a fundamental question: how and to what extent does subsurface structure influence streamflow and solute transport in montane watersheds? Here, we test controls of three-dimensionally resolved subsurface structure on solute transport, which reflects source waters and water residence time. We integrate the geophysically mapped architecture of Coal Creek, a headwater catchment in the Colorado River Basin, into numerical flow-transport models. Our results reveal greater vertical connectivity elongates flow paths and enhances deep groundwater contributions to streams, therefore substantially influencing the timing and magnitude of solute transport. In contrast, watershed-integrated outlet discharge can be predicted without high-resolution subsurface information. These findings demonstrate incorporating subsurface architecture is critical for predicting ecosystem health and contaminant transport.

Colorado↗