William Benjamin Frank

Interactions between slow slip and earthquakes remain poorly understood. We examine such an interaction in the central Hikurangi subduction zone where several moderate (Mw 4–5+) earthquakes occurred during a deep, 2‐year M7 slow slip event that started in 2021. We apply three different geodetic inversion methods to 6 years (2018–2024) of continuous GNSS data from 97 stations across New Zealand's North Island to retrieve the time‐dependent evolution of fault slip at depth: (a) a fixed‐window (1, 3, 6 months) static offset approach; (b) an Independent Component Analysis‐based inversion Method (ICAIM); and (c) an elastic block modeling approach utilizing TDefnode. All three methods yield a consistent, high‐resolution slip history in time and space. The 2021–2024 long‐term, deep slow slip event (equivalent 7.1–7.2) evolved with five to six distinct sub‐events, three of which represent an increase in slow slip event moment release within weeks after 4+ normal mechanisms (intraslab) earthquakes. In contrast, plate interface events ( 5) during this period do not seem to impact the slow slip event evolution. Static Coulomb modeling shows that cumulative slow slip increased stress by 40–60 kPa at the hypocenters of two plate‐interface earthquakes, suggesting that slow slip may have contributed to bringing these faults closer to failure. Our results reveal an asymmetric interplay between slow slip and seismicity: intraslab earthquakes modulate ongoing slow slip, while slow slip loading may trigger interface events, demonstrating two‐way interactions at considerably lower magnitudes than typically documented at other subduction zones.

The deformation regime that accommodates tectonic motion at plate boundaries changes as pressures and temperatures increase with depth, transitioning from the shallow frictional sliding of seismic ruptures to deep, viscous flow. We use recurring swarms of low-frequency earthquakes in this transition zone to measure the recurrence intervals and durations of accompanying slow slip across four plate boundaries. We find these time scales systematically decrease with depth and linearly scale with one another. Assuming a transition zone governed by episodic faulting, the observed time scales produce an average slow slip rate of 7 ± 2 mm/d across all four plate boundaries that is independent of depth. Thermal models place these slow slip dynamics within a common, narrow range of temperatures. Our results suggest deep slow slip is the result of the episodic unjamming of the plate boundary by frictional heterogeneities whose relaxation is modulated by surrounding temperature-dependent viscous material.

Slow‑slip events and their interactions with earthquakes remain poorly understood. To investigate such interactions, we examine a deep, two-year M7 slow slip event which started in 2021 in the central Hikurangi subduction zone, which also coincided with a period of several moderate (Mw 4-5+) earthquakes in the region. We analyze six years (2018–2024) of continuous GNSS data from 97 stations across New Zealand’s North Island. Three different methods are assessed to retrieve time-dependent slip at depth: (1) fixed‑window static offsets (1, 3, 6 months); (2) Independent Component Analysis-based inversion Method (ICAIM); and (3)~a simplex minimization approach utilizing TDefnode. All three methods yield a consistent, high‑resolution slip history in time and space. The 2021-2024 long-term, deep slow slip event (equivalent Mw 7.1-7.2) evolves with five to six distinct sub-events, three of which represent an increase in slow-slip event moment release within weeks after Mw4+ intraslab earthquakes. In contrast, plate interface events (M_w>5) during this period do not seem to impact the slow slip event evolution. Static Coulomb modeling of stress changes due to long-term slow slip events prior to two interface earthquakes show positive stress changes of 20–60kPa at the earthquake hypocenters, suggesting that slow slip events may have played a role in triggering these interface events. Our results highlight that Hikurangi slow slip events and moderate earthquakes engage in two‑way communication at considerably lower magnitudes than documented elsewhere, underscoring the value of real‑time geodetic monitoring for seismic hazard assessment.

Earthquakes abruptly release tectonic stress that builds slowly over time through the coupled evolution of faults and the surrounding crust. Seismic wavespeeds track crustal deformation and stress changes, but typical monitoring methods are most sensitive to shallow depths. Using receiver functions, we tracked rupture-zone wavespeed and anisotropy changes throughout the crust during the 2019 Ridgecrest earthquake sequence. Shallow coseismic wavespeed reductions recovered within months, whereas a deeper postseismic wavespeed drop persisted without measurable recovery over several years. The deep, persistent wavespeed drop likely reflects accumulating damage driven by postseismic deformation, suggesting two possible scenarios: (i) a slow interseismic recovery where wavespeed and anisotropy track long-term stress evolution; or (ii) permanent deformation of an immature fault zone. Both scenarios affect the dynamics and energy budget of the seismic cycle.

Subduction megathrusts release stress not only seismically through earthquakes, but also through creep and transient slow deformation, called slow slip events (SSEs). Understanding the interplay between fast and slow slip is essential for illuminating the deformation processes on the subduction interface. The Chilean subduction margin, while one of the most seismically active regions worldwide, has few reports of SSEs. Furthermore, there are no comprehensive reports of tectonic tremors or low‐frequency earthquakes (LFEs), seismic signals typically accompanying SSEs, tracking deformation at small spatial and temporal scales. Here, we perform a systematic search for tectonic tremors and LFEs in the Atacama segment in Northern Chile, a region hosting both shallow and deep SSEs. Using dense seismic networks, we investigate 3.3 years between November 2020 and February 2024. Due to the network geometry, we focus on deep tremor and LFEs. We apply two complementary methods, envelope correlation for tremor search and deep learning detection for LFEs, to generate initial catalogs. To validate the potential detections, we use clustering, matched filtering, heuristics, and extensive manual inspection. While our initial search provides numerous candidates, after verification, we find no evidence for tectonic tremor or LFEs in the region. In contrast, our approaches successfully recover tremors and LFEs in two reference regions outside Chile with known tremor and LFE activity. Our observations show that tremors and LFEs in Northern Chile are either of lower moment rate than in other regions, have substantially longer recurrence rates, or are absent altogether, potentially due to the cold subduction.

We report the first catalog of low‐frequency earthquakes in the Hikurangi subduction zone, located beneath the Kaimanawa Range of the North Island at 50 km depth, downdip of regularly recurring (every 4–5 years) deep M7 slow slip events. To systematically detect low‐frequency earthquakes within the regional continuous seismic data, we utilized a matched‐filter approach with template waveforms derived from previous observations of tectonic tremor. We built our catalog of 36 low‐frequency earthquake sources, that produced almost 21,000 events over more than a decade, with two matched‐filter search iterations. In each iteration, the detections were gathered into families and their coherent waveforms processed and stacked to extract high‐quality waveforms, allowing us to pick seismic phase arrivals to locate the low‐frequency earthquakes. We highlight three characteristic features to validate that our detected events are indeed low‐frequency earthquakes: the eponymous deficit of high frequencies in their seismic waveforms, the episodic swarms of activity that define their activity through time, and their location at the plate boundary with a double‐couple source mechanism and geometry consistent with the subduction interface. Considering the observed low‐frequency earthquakes' relationship to neighboring slow slip, we observe the event swarms to occur much more frequently than the M7 slow slip events located just updip. Similar to other deep low‐frequency earthquakes in other subduction zones, we suggest that this characteristic clustering in time is driven by more frequent, smaller slow slip events that are not clearly observable at the surface.

The underlying mechanism of the ongoing seismic swarm in the Noto Peninsula, Japan, which generates earthquakes at 10 times the average regional rate, remains elusive. We capture the evolution of the subsurface stress state by monitoring changes in seismic wave velocities over an 11-year period. A sustained long-term increase in seismic velocity that is seasonally modulated drops before the earthquake swarm. We use a three-dimensional hydromechanical model to quantify environmentally driven variations in excess pore pressure, revealing its crucial role in governing the seasonal modulation with a stress sensitivity of 6 × 10 ⁻⁹ per pascal. The decrease in seismic velocity aligns with vertical surface uplift, suggesting potential fluid migration from a high–pore pressure zone at depth. Stress changes induced by abnormally intense snow falls contribute to initiating the swarm through subsequent perturbations to crustal pore pressure.

We introduce BPMF (backprojection and matched filtering)—a complete and fully automated workflow designed for earthquake detection and location, and distributed in a Python package. This workflow enables the creation of comprehensive earthquake catalogs with low magnitudes of completeness using no or little prior knowledge of the study region. BPMF uses the seismic wavefield backprojection method to construct an initial earthquake catalog that is then densified with matched filtering. BPMF integrates recent machine learning tools to complement physics-based techniques, and improve the detection and location of earthquakes. In particular, BPMF offers a flexible framework in which machine learning detectors and backprojection can be harmoniously combined, effectively transforming single-station detectors into multistation detectors. The modularity of BPMF grants users the ability to control the contribution of machine learning tools within the workflow. The computation-intensive tasks (backprojection and matched filtering) are executed with C and CUDA-C routines wrapped in Python code. This leveraging of low-level, fast programming languages and graphic processing unit acceleration enables BPMF to efficiently handle large datasets. Here, we first summarize the methodology and describe the application programming interface. We then illustrate BPMF’s capabilities to characterize microseismicity with a 10 yr long application in the Ridgecrest, California area. Finally, we discuss the workflow’s runtime scaling with numerical resources and its versatility across various tectonic environments and different problems.