Chengping Chai

The 8 September 2017 Mw 8.2 Tehuantepec earthquake was the largest instrumentally recorded normal-faulting earthquake in Mexico. The mainshock occurred offshore within the Tehuantepec seismic gap, generating >30,000 aftershocks in the following year. We applied an open-source, machine learning (ML)–assisted workflow to construct a high-resolution aftershock catalog using data from temporary and permanent seismic networks in southern Mexico. The workflow integrates PhaseNet for phase detection; GaMMA for phase association; and VELEST, HypoInverse, and HypoDD for velocity modeling and relocation. We processed seven months of continuous waveform data from 29 broadband stations, including a temporary rapid-response deployment that improved station coverage of the offshore rupture zone. To evaluate performance, we compared our results against analyst-reviewed picks and event locations from the Servicio Sismológico Nacional catalog. The resulting catalog contains 11,374 relocated earthquakes and represents the most comprehensive published dataset for this sequence, incorporating the first full use of the temporary network. Relocated hypocenters show improved depth control and align well with the Slab2.0 subduction geometry, revealing clearer separation between offshore slab events and onshore crustal seismicity. This study demonstrates that combining ML-based detection with established methods provides a scalable and reproducible approach for constructing high-quality earthquake catalogs in tectonically complex environments and offers practical guidance for adapting similar workflows to other earthquake sequences.

Summary: Reliable automatic phase picking is important for many seismic applications. With the development of machine learning approaches, many algorithms are proposed, evaluated and applied to different areas. Many of these algorithms are single station based, while recent proposed methods start to combine surrounding stations into consideration in the problem of phase picking. Among these algorithms, the Phase Neural Operator (PhaseNO) shows promising results on regional datasets comparing to existing algorithms. But there are many use cases for the local seismic networks in our community, therefore in this paper we evaluate the performance of PhaseNO on 4 different local datasets and compare the results to PhaseNet and EQTransformer. We used both individual phase picking metrics as well as association metrics to illustrate the performance of PhaseNO. By manually reviewing the newly detected events, we find the PhaseNO model outperforms the single station-based approaches in the local-scale use cases due to its consideration of coherent signals from multiple stations. We also explored PhaseNO’s behaviors when only using one station, as well as gradually increasing the number of stations in the seismic network to better understand its behavior. Overall, using the off-the-shelf machine learning based phase pickers, PhaseNO demonstrated its good performance on local-scale seismic networks.

We use a nodal seismic network deployed at the Redmond Salt Mine in Utah as a proxy for industrial monitoring to evaluate the efficacy of seismic interferometry for detecting and locating persistent anthropogenic seismic sources. We observe cyclic noise patterns corresponding to mining operations, with distinct stationary and migrating persistent signals. We present an enhanced backprojection location technique to deconvolve source finiteness, better resolving concurrent discrete sources for increased monitoring resolution of anthropogenic activity over time. Our results show that migrating persistent seismic signals consistently focus near lateral blasting areas during evening hours, suggesting they originate from blast preparation activities. In addition, we identify a stationary source near a surface industrial ventilation fan, which likely generates the continuous ground-coupled airwave observed in our data. Our results further demonstrate the utility of seismic interferometry to monitor continuous, high frequency, anthropogenic seismic sources generated by subsurface industrial facilities.

Seismic sensors deployed near roadways effectively capture ground vibrations generated by passing vehicles. Although both traditional and machine-learning algorithms have been utilized for analyzing such signals, independent validation of detected vehicle events remains limited. We applied two unsupervised machine-learning algorithms, uniform manifold approximation and projection for dimension reduction, and hierarchical density-based spatial clustering of applications with noise, to continuous seismic data collected along a road on the main campus of Oak Ridge National Laboratory. The algorithms identified seven distinct cluster labels across the entire dataset. By comparing these cluster labels with precipitation records from a nearby weather station and image-derived labels from a local camera system, we identified one cluster associated with rainfall and another with vehicle activity. Our algorithms identified a greater number of vehicle-related labels compared to the camera-derived labels because seismic data are unaffected by poor lighting conditions. The arrival times of the newly detected vehicle signals corresponded well with the road’s speed limit, supporting our findings. Our algorithm outperformed the short-term average/long-term average method and k-means clustering. Our results suggest that seismic data, when analyzed with machine-learning algorithms, can complement existing vehicle monitoring systems, particularly under challenging environmental conditions.

Deep learning has emerged as a transformative tool for enhancing the extraction of reliable information from seismograms, addressing the increasing demand for precise and efficient seismic data analysis. We introduce an innovative encoder–decoder deep learning model, named WaveDenoiser, designed for noise reduction in the time domain, thereby eliminating the need for spectrogram computations that have been used for existing deep learning tools and significantly improving processing speed. Utilizing the benchmark dataset that is Stanford Earthquake Dataset, we developed three models of varying sizes: base, medium, and large. Notably, the large (referred to as WaveDenoiser) model demonstrated superior performance, achieving a median signal-to-noise ratio improvement of 8.8 dB on in-distribution unseen data (in the same geographic region) and 7.7 dB on out-distribution unseen data (in a new geographic region), outpacing both the base and medium models. Further evaluation of the WaveDenoiser model revealed a reduction in median arrival-time errors by 0.02 s for P waves and 0.01 s for S waves when processing waveforms prior to phase picking using PhaseNet on in-distribution unseen data. When tested on out-distribution unseen data, the model also effectively reduced the P-wave median arrival-time error by 0.02 and 0.01 s in median arrival-time error for S waves. Importantly, the application of WaveDenoiser resulted in a significant reduction of phase picking outliers by 1.1% to 3.6% for both P and S waves. In addition, we achieved over five times acceleration in processing speed compared with the seisBench implementation of DeepDenoiser. Our findings underscore the potential of WaveDenoiser as a powerful tool for improving seismic data analysis and processing efficiency.

The Ecuadorian portion of the South American subduction zone presents an interesting case study in the structure and complex evolution of an upper plate. There are outstanding questions about its tectonic history, composition, and magmatic processes. While previous studies have employed ambient noise tomography to image the Ecuadorian upper plate, surface wave inversions alone often lack sensitivity at relevant shallow depths. This limitation can be overcome with an independent, complementary data set, such as gravity. We have jointly inverted Rayleigh wave phase velocities and Bouguer gravity anomalies to provide a more detailed seismic velocity model of the Ecuadorian upper plate. Our joint inversion has yielded several key improvements from previous models. First, we observe much shallower slow velocities beneath major basins (the Manabí, Progreso, and Gulf of Guayaquil), better aligning with expected basin structure. Second, we identify a high‐velocity block beneath the entire forearc, corresponding to the Piñon Terrane, with velocities suggesting the presence of ultramafic material. Third, we highlight a new narrow swath of slow velocities beneath the Ecuadorian Andes, which closely follows the active volcanoes along the Eastern Cordillera. The extent of these slow velocities coincides with the termination of active arc volcanism and the predicted location of the subducted Carnegie Ridge. The predicted compositions for the mid to lower crust in the region preclude a purely compositional explanation for these velocities, suggesting that some level of partial melt is necessary.

Precisely measuring seismic arrival times is a labor-intensive task but is critical for both earthquake monitoring and subsurface imaging. Recently published deep learning models have demonstrated superior performance compared to traditional automatic approaches for picking arrival times. Although existing deep learning models have shown promising results, further advancements are necessary as their performance is not yet satisfactory especially when applied to new regions and station networks. Increasing model size has led to improved performance in other machine learning applications. We aimed to investigate whether enlarging deep learning models can increase performance on accepted benchmarks. We trained three models of varying sizes, small (1X), medium (4X), and large (16X), using globally distributed local and regional earthquake signals and background noise waveforms from a benchmark dataset, Stanford Earthquake Dataset. Our results indicate that the largest model (PickerXL) outperforms both the smaller models and Seisbench implementation of the PhaseNet model, which has the same number of parameters as our small model. The PickerXL model’s enhanced capacity to extract complex patterns from seismograms contributes to its superior arrival picking abilities compared to the smaller model.