Supershear Rupture Indicator in Near-fault Particle Motion
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.
Kai: We're looking at the paper: "Supershear Rupture Indicator in Near-fault Particle Motion".
Mira: Here's the gist of the paper. Earthquake rupture propagation speed is an essential source parameter that fundamentally controls hazard and risk.
Kai: This study develops and demonstrates the capability of near-fault seismic records in delineating rupture speeds of strike-slip earthquakes through inspecting displacement particle motion.
Mira: The researchers apply a new method on near-fault strong-motion data in global M7+ strike-slip earthquakes to identify diverse particle motion signatures associated with various rupture behaviors, including "sustaining subshear rupture, sustaining supershear rupture, supershear transition, oblique slip, initial rupture expanding process, and multiple rupture fronts".
Kai: They focus on the particle motion of displacement rather than velocity pulses because it seeks a "simple and robust approach to identify rupture speeds" aiming at applying in real time.
Title and authors: Kai: So we've established what this paper is about, but let's talk about what they are actually trying to prove with this "Supershear Rupture Indicator in Near-fault Particle Motion."
Mira: They are really focusing on how you can use these near-fault records to distinguish between different ways a rupture can propagate. It’s not just measuring speed; it’s about identifying the specific signatures that tell you if it’s subshear, supershear transition, or something else entirely.
Lev: Which is interesting because those signatures are what we have to try and model when we look at real hardware data later. We need to know what the actual physical manifestation of a fast rupture looks like in terms of particle motion before we can even think about the computational overhead.
Kai: Right, so they are developing a method that inspects these displacement motions—not just velocity pulses—to identify things like "sustaining subshear rupture" or "oblique slip" and how those translate to speed.
Mira: And the core idea they push is that the shape of this particle motion actually contains first-order information about the rupture speed, which could let us infer it very quickly.
Lev: That’s a big claim because if it's true, it means we don't need to run these full dynamic models just to get a preliminary idea of the propagation characteristics.
Kai: It’s about moving from complex waveform analysis to something that can be more robust for real-time applications, which is exactly what we want in hazard assessment.
Mira: And it moves the focus away from just looking at velocity pulses and toward analyzing the whole displacement signature across different components.
Lev: So they’re trying to make sense of those complex, multi-wiggle waveforms that we struggle with when we look at stations like CLC or Kahramanmaraş because they can't easily tell if a wiggle is radiation or just fault geometry.
The paper's summary: Kai: Now for the actual summary of "Supershear Rupture Indicator in Near-fault Particle Motion." They apply this new technique to strong-motion data from global M7 plus strike-slip earthquakes.
Mira: They are identifying distinct particle motion signatures that correspond to specific rupture behaviors, like "initial rupture expansion" or the transition into a supershear phase. For instance, they point out four stations that show typical supershear particle motions: NPW, PS10, NAR, and LUC.
Lev: So they are using these known examples to build a classification scheme for what different motion patterns mean in terms of rupture physics. That sounds like a solid step toward creating a feature set for an AI system later.
Kai: They also look at the transition itself, which they describe as the particle motion separating into an initial supershear pattern followed by a trailing Rayleigh pattern. That separation is key to their identification process.
Mira: And they’re testing these ideas with shape metrics, specifically the normalized particle motion area and the particle motion anisotropy, to see if they have predictive power over the rupture speed.
Lev: If those metrics work well, it suggests that we can filter out a lot of noise or irrelevant complexity from the raw data before we try to run simulations on it.
Kai: They found that motions under subshear speeds show high motion area and low anisotropy, while supershear ruptures show the opposite: high anisotropy and low motion area. That’s a really sharp contrast.
Mira: And they state that these metrics are effective for identifying supershear-like motions because all the stations they previously identified as supershear speeds feature a motion area below zero point two and an anisotropy over zero point nine.
The paper's improvements: Kai: Let's talk about what they suggest as improvements to this work, because the authors are already pointing toward future applications.
Mira: They argue that the shape of the particle motion contains first-order information regarding rupture speed, which allows for a potential rapid inference of that speed without needing a full simulation run.
Lev: That’s what we're aiming for when we look at real hardware; you want to reduce the computational load so you can run more complex error correction algorithms on the data itself.
Kai: They test two specific shape metrics, the normalized particle motion area and the particle motion anisotropy, and they show these are good for rapid identification of supershear-like motions.
Mira: The improvement is really in creating an efficient first-order screening tool, which essentially lets you quickly filter out data that isn't indicative of a fast rupture.
Lev: That screening capability would be invaluable for real-time hazard assessment, allowing us to flag potential supershear events before the standard processing pipelines get bogged down.
Kai: They also note some limitations, which is important for reality: the sensitivity of the motion trend to rupture speed actually decreases with the off-fault distance.
Mira: So they restrict their analysis to stations within three kilometers from the fault to ensure they have enough resolution for this speed determination.
Conclusion: Kai: To wrap up, what this paper really delivers is a strong argument for using near-fault data in a new way to resolve earthquake rupture propagation speeds.
Mira: They propose these shape metrics as an efficient first-order screening tool for identifying potential supershear particle motion based on the trajectory of the near-fault data.
Lev: From an engineering standpoint, this is powerful because it shifts the burden away from needing perfect knowledge of material properties or exact fault geometry during the initial speed estimation phase.
Kai: So, in simple terms, they’re showing that you can use what you measure on the ground—the particle motion—to rapidly screen for supershear activity.
Mira: It moves us toward a system where we can get a quick indicator of rupture speed just by looking at the data shape, which is a huge step for emergency planning.
Lev: It’s promising, but we have to remember their caveat: this method loses sensitivity when the station is too far away from the fault, so it's not a universal solution yet.
Kai: So, the paper "Supershear Rupture Indicator in Near-fault Particle Motion" gives us a new way to look at earthquake data, focusing on those displacement signatures.
Mira: It’s less about measuring velocity and more about analyzing the shape of the particle motion to infer speed, which is what we need for better hazard understanding.
Lev: We'll keep watching how this translates into actual algorithms that can handle the noise and complexity that comes with real-world seismic data.
Department of Earth and Environmental Sciences, The Chinese University of Hong Kong · Laboratoire de Géologie, École Normale Supérieure, CNRS-UMR 8538, PSL Research University · Shenzhen Research Institute, The Chinese University of Hong Kong · Institute of Environment, Energy and Sustainability, The Chinese University of Hong Kong · Institute of Space and Earth Information Science, The Chinese University of Hong Kong · Bureau de Recherches Géologiques et Minières (BRGM)
physics.geo-ph
Submitted: 2026-06-09
Updated: 2026-11-03
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 89/100
The gist: Earthquake rupture propagation speed is an essential source parameter that fundamentally controls hazard and risk.
Key concepts
- Particle Motion Analysis
- Researchers analyze the displacement of particles on a fault using acceleration data. This is done by rotating the motion into fault-parallel (FP) and fault-normal (FN) components based on local strike directions. This approach focuses on displacement rather than velocity pulses to identify rupture speeds simply and robustly.
- Supershear Signatures
- The study identifies specific particle motion patterns associated with different rupture behaviors, such as 'initial rupture expansion' or 'supershear transition'. These signatures are distinct from subshear motions and are crucial for characterizing the speed at which the earthquake ruptures along the fault.
- Shape Metrics
- The paper demonstrates that the geometric shape of particle motion contains first-order information about rupture speed. Two metrics tested—normalized particle motion area and anisotropy—show clear differences: subshear speeds yield high area and low anisotropy, while supershear ruptures show high anisotropy and low area.
Terminology
Summary
Earthquake rupture propagation speed is an essential source parameter that fundamentally controls hazard and risk. p. 0 Bhat2
, and Hideo Aochi2,6
1
Department of Earth and Environmental Sciences, The Chinese University of Hong
Kong, Hong Kong, China
2
Laboratoire de Géologie, École Normale Supérieure, CNRS-UMR 8538, PSL
Research University, Paris, France
3
Shenzhen Research Institute, The Chinese University of Hong Kong, Shenzhen,
Guangdong, China
4
Institute of Environment, Energy and Sustainability, The Chinese University of Hong
Kong, Hong Kong, China
5
Institute of Space and Earth Information Science, The Chinese University of Hong
Kong, Hong Kong, China
6
Bureau de Recherches Géologiques et Minières (BRGM), 45100 Orléans, France
* Corresponding author: hyang@cuhk. hk
Abstract
Earthquake rupture propagation speed is an essential source parameter that
fundamentally controls hazard and risk. hk
Abstract
Earthquake rupture propagation speed is an essential source parameter that
fundamentally controls hazard and risk. In this paper, we develop and demonstrate the
capability of near-fault seismic records in delineating rupture speeds of strike-slip
earthquakes through inspecting displacement particle motion.Where the paper says it
Where the paper says it
Methodology for Particle Motion Analysis
The researchers apply a new method on near-fault strong-motion data in global M7+ strike-slip earthquakes to identify diverse particle motion signatures associated with various rupture behaviors, including sustaining subshear rupture, sustaining supershear rupture, supershear transition, oblique slip, initial rupture expanding process, and multiple rupture fronts
. They focus on the particle motion of displacement rather than velocity pulses because it seeks a simple and robust approach to identify rupture speeds
aiming at applying in real time. The analysis involves integrating acceleration data to obtain velocity and displacement waveforms, which are then rotated into fault-parallel (FP) and fault-normal (FN) components according to local strikes and the positive FN direction is defined as the S-wave particle motion direction of the preceding rupture
.
Identification of Rupture Speed Signatures
The study identifies distinct features associated with different rupture behaviors, such as We identify four stations exhibiting typical supershear particle motions: NPW (2025
Mw7. 7 Myanmar), PS10 (2002 Mw7. p. 4 We identify four stations exhibiting typical supershear particle motions: NPW (2025
Mw7. 7 Myanmar), PS10 (2002 Mw7. 7 Myanmar), PS10 (2002 Mw7. 8 Denali), NAR (2023 Mw7. 8 Denali), NAR (2023 Mw7. 8 Kahramanmaraş),
and LUC (1992 Mw7. 8 Kahramanmaraş),
and LUC (1992 Mw7. 3 Landers) (Fig.initial rupture expansion, oblique slip, and supershear transition
. They observe four stations exhibiting typical supershear particle motions: NPW (2025 Mw7. p. 4Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
subshear FN-FP-FN motion pattern
. Furthermore, they note that during the transition, the particle motion separates into an initial supershear FP-dominant motion followed by a trailing Rayleigh FN-dominant motion
.
Shape Metrics for Rapid Inference
The paper demonstrates that This property enables
potential rapid inference of rupture speed. Therefore, we further test shape metrics to
quantify trajectory characteristics of subshear or supershear rupture speeds, including
the normalized particle motion area and the particle motion anisotropy (see details in
Materials and Methods) (Fig. This property enables
potential rapid inference of rupture speed. Therefore, we further test shape metrics to
quantify trajectory characteristics of subshear or supershear rupture speeds, including
the normalized particle motion area and the particle motion anisotropy (see details in
Materials and Methods) (Fig.the shape of particle motion contains first-order information regarding rupture speed
which enables potential rapid inference of rupture speed. They test two shape metrics: the normalized particle motion area and the particle motion anisotropy. p. 6Where the paper says it
high motion area and low anisotropy, whereas supershear ruptures show the opposite pattern with high anisotropy and low motion area
. These metrics are effective for identifying supershear-like particle motion, as all stations previously identified with supershear speeds feature motion area below 0.2 and anisotropy over 0. p. 6
.Where the paper says it
Influence of Fault Geometry and Complexity
The analysis shows that the sensitivity of the motion trend to rupture speed decreases with the off-fault distance, restricting analysis to stations within 3 km to ensure adequate resolution of local rupture speed. The influence of fault bend, dipping angle, and oblique slip on particle motion is generally minor on the overall trend. p. 6 To assess the robustness of the
particle motion patterns, we test the sensitivity of particle motion to the dipping angle
and rake angle by considering a dipping angle of 75° and oblique slip with rake angles
of 20° (lateral-reverse) and -20° (lateral-normal). As shown in Figure S15, the two
parameters can modulate the motion to some degree while the overall motion trend
remains. Those pulses are likely associated multiple slip
episodes on the local complex fracture network (Litchfield et al. Station
KMMH16 is located close to both the hypocenter and a bend, with oblique shallow slip
reported (Shirahama et al.Where the paper says it
Where the paper says it
Conclusion on Near-Fault Data Utility
The study highlights the application of near-fault data in resolving earthquake rupture propagation. The authors propose shape metrics for rapid rupture speed identification using the near-fault trajectory. p. 7 We also
identify the particle motion associated with other rupture behaviors such as initial
rupture expansion process and early supershear transition. Moreover, we propose shape
metrics for rapid rupture speed identification using the near-fault trajectory. We apply the new
method on near-fault strong-motion data in global M7+ strike-slip earthquakes and
identify diverse particle motion signatures associated with sustaining subshear rupture,
sustaining supershear rupture, supershear transition, oblique slip, initial rupture
expanding process, and multiple rupture fronts. This study highlights the superior
application of near-fault observations in rapid rupture speed determination. Having a series of stations along the rupture is effective to
characterize transient supershear (Deng et al. Nevertheless, the metrics provide
an efficient first-order screening tool for identifying potential supershear particle
motion.Where the paper says it
Where the paper says it
Where the paper says it
Table S1
Table S1 lists the information of near-fault data used in this study, including event Mw Faulting type Station Off-fault distance (km) strike/dip/rake (GCMT) Data source.
Table S2
Table S2 lists the local strikes used for waveform rotation, showing the strike degree for various stations.
Table S3
Table S3 presents a 1-D velocity model adopted in the dynamic rupture model, detailing depth and seismic velocities.
Figure 1
Figure 1 illustrates synthetic near-fault particle motions in a strike-slip earthquake, showing subshear and supershear patterns under different speeds.
Figure 2
Figure 2 shows near-fault particle motions in global M7+ strike-slip earthquakes, marking stations with identified supershear particle motions in red.
Figure 3
Figure 3 displays particle motions near the hypocenter and supershear transition, illustrating subshear and supershear rupture propagation in a dynamic model.
Figure 4
Figure 4 presents particle motion shape metrics, comparing synthetic motions with real data, where stations with inferred supershear rupture speeds are marked in red.
Table S1-S3
Table S1-S3 contains the information of near-fault data used in this study.
Table S2 details the local strikes used for waveform rotation.
Table S3 provides the 1-D velocity model adopted in the dynamic rupture model.
Figure S9
Figure S9 illustrates schematic diagram illustrating radiation from different positions on the fault and velocity pulses under different rupture speeds.
Figure S10-S12
Figures S10-S12 show synthetic particle motions at different off-fault distances for a total rupture length of 120 km, demonstrating the capability of resolving sustained rupture speeds.
Figure S15
Figure S15 plots the influence of dip and rake angle on particle motion, showing how these parameters modulate the motion under subshear and supershear conditions.
Figure 5
Figure 5 shows near-fault particle motions for KMMH16 station in the 2016 Mw 7.1 Kumamoto earthquake. p. 9 & Miyakawa, A. Characteristics of the surface ruptures associated with the
2016 Kumamoto earthquake sequence, central Kyushu, Japan. & Miyakawa, A. Characteristics of the surface ruptures associated with the
2016 Kumamoto earthquake sequence, central Kyushu, Japan.Where the paper says it
Where the paper says it
Figure S4
Figure S4 shows particle motion for KEKS station in the 2016 Mw 7.8 Kaikōura earthquake. p. 6 We observe at least three FP and FN pulses at the station KEKS
station during the 2016 Kaikōura Mw 7. 8 earthquake (Fig.Where the paper says it
Figure S5
Figure S5 shows particle motion for KMMH16 station in the 2016 Mw 7.1 Kumamoto earthquake.
Figure S6
Figure S6 shows particle motion for ROLC station in the 2010 Mw 7.2 Darfield earthquake. p. 4 We further identify stations displaying the subshear FN-FP-FN motion pattern (Fig. 2C), including most stations along the Amanos segment in the Kahramanmaraş
earthquake and the ROLC station in the Darfield earthquake.Where the paper says it
Figure S7
Figure S7 shows particle motion for PS10 station in the 2002 Mw 7.9 Denali earthquake. p. 1 For
instance, the 2002 Denali Mw 7. 9 earthquake had only a single record within 3 km to
the fault, at Pump station 10 (PS10) operated by Alyeska Pipeline Service Company
(Ellsworth et al.Where the paper says it
Figure S8
Figure S8 shows particle motion for LUC station in the 1992 Mw 7.3 Landers earthquake. p. 7 , & Fukuyama, E. Three‐dimensional nonplanar simulation of the
1992 Landers earthquake.Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Automated rupture speed inference from displacement particle motion: AI systems can be trained to analyze raw displacement particle motion data and automatically classify signatures as subshear, supershear transition, or pure supershear rupture based on learned patterns in FP and FN components.
-
Real-time hazard assessment for near-fault regions: An improved system can use the proposed shape metrics, such as normalized particle motion area and anisotropy, to rapidly screen incoming seismic data to identify potential supershear events before traditional methods can process them fully.
-
Fault geometry constraint integration: AI models can incorporate fault geometry parameters like dipping angle and rake angle into rupture simulation frameworks to better predict particle motion patterns at complex features like fault bends, improving the accuracy of speed determination in heterogeneous fault zones.
Abstract
Earthquake rupture propagation speed is an essential source parameter that fundamentally controls hazard and risk. In this paper, we develop and demonstrate the capability of a single near-fault seismic record in delineating earthquake rupture speed by inspecting displacement particle motion. We apply this method to near-fault strong-motion data in M7+ strike-slip earthquakes and identify diverse particle-motion signatures associated with subshear rupture, supershear rupture, supershear transition, initial rupture expansion, and potential multiple rupture fronts. This study highlights the value of near-fault observations for determining rupture speed.
Related papers
- Geometry-Aware Transformer for Five-Dimensional Seismic Noise Attenuation and First-Arrival Picking
- Generative deep learning improves reconstruction of global historical climate records
- Field-Deployable Hybrid Gravimetry: Projecting Absolute Accuracy Across a Remote 24km squared Survey via Daily Quantum Calibration
- Parameter-Efficient Transfer Learning for Microseismic Phase Picking Using a Neural Operator
- Scaling-law-informed neural point processes for earthquake sequence forecasting
- Prediction of Extreme Events in Multiscale Simulations of Geophysical Turbulence using Reinforcement Learning