Presentation Information
[S09-11]Slow Slip Event Detection and Associated Seismic Activation in the Hikurangi Subduction Zone, New Zealand
*Natsumi Takeda1, Tomoaki Nishikawa2 (1. Graduate School of Science, Kyoto University , 2. Disaster Prevention Research Institute, Kyoto University)
Slow slip events (SSEs) are recognized as potential triggers of seismic activity, with such interactions observed across numerous subduction zones worldwide. However, standard statistical models of earthquake occurrence, such as the Epidemic-Type Aftershock Sequence (ETAS) model, do not account for this triggering effect. Nishikawa and Nishimura (2023) addressed this limitation by introducing an additional term into the ETAS framework, incorporating SSE data from the Hikurangi subduction zone in New Zealand, and demonstrated that the modified model more accurately represents regional seismicity. The Hikurangi margin is characterized by both frequent SSEs and regular earthquakes, and a variety of interactions between them have been reported (Nishikawa et al., 2021). Nonetheless, the model remains limited by the small number of SSEs included, despite the occurrence of many more in this region, indicating a clear opportunity for further refinement through expanded data integration.
This study aims to provide a more comprehensive understanding of earthquake activity in the Hikurangi subduction zone by constructing a detailed SSE catalog and assessing the relationship between individual SSEs and elevated seismicity. By identifying patterns in which SSEs are more likely to trigger earthquakes, this research seeks to clarify the relationship between aseismic slip and seismic events, with the long-term goal of contributing to improved earthquake forecasting in subduction zones.
We created a comprehensive SSE catalog from 2005 to 2024 using GNSS time series data from 41 stations along the east coast of New Zealand’s North Island, with one reference station on the island’s western side. The data were provided by the Nevada Geodetic Laboratory (Blewitt et al., 2018). An automated SSE detection method, developed based on Nishimura et al. (2013), was applied to the detrended GNSS time series (with the reference station subtracted), using an Akaike’s information criterion difference (AIC) threshold of < -8.5 to identify potential events. For each detected SSE, we performed a slip inversion using rectangular fault Green’s functions to estimate fault geometry, location, and slip. The inversion was conducted using the Powell method implemented in scipy. The event duration was determined by fitting a double-hinge line to the stacked displacement time series, and the moment magnitude was subsequently calculated. We then extracted earthquakes of Mw > 2.5 occurring within ±90 days and 100 km of each SSE’s fault center, using data from GeoNet (January 2004 to May 2025). To assess whether these events represented significant seismicity increases, we applied a simplified version of the detection method by Nishikawa and Ide (2017), based on a temporal ETAS model originally developed for earthquake swarm detection.
Through our auto-detection, we were able to detect 53 SSEs in 17 years, with the magnitude ranging from Mw 5.7 to 7.0, and 29 events had a notable seismicity increase within 60 days from the center date of the SSEs. 24 events out of these 29 events displayed spatial clustering as well, on or close to the estimated faults. In cases where multiple swarms were associated with a single SSE, most occurred within the same spatial region, although one SSE exhibited distinct swarms in separate areas.
This study successfully detected numerous SSEs across the Hikurangi subduction zone and associated seismicity. Our findings highlight the potential for SSE-based forecasting of earthquake likelihood, particularly if future models can distinguish which types of SSEs are more seismically active. This study can contribute to the broader goal of developing time-dependent seismic hazard models in subduction zones where slow slip is frequent and influential.
This study aims to provide a more comprehensive understanding of earthquake activity in the Hikurangi subduction zone by constructing a detailed SSE catalog and assessing the relationship between individual SSEs and elevated seismicity. By identifying patterns in which SSEs are more likely to trigger earthquakes, this research seeks to clarify the relationship between aseismic slip and seismic events, with the long-term goal of contributing to improved earthquake forecasting in subduction zones.
We created a comprehensive SSE catalog from 2005 to 2024 using GNSS time series data from 41 stations along the east coast of New Zealand’s North Island, with one reference station on the island’s western side. The data were provided by the Nevada Geodetic Laboratory (Blewitt et al., 2018). An automated SSE detection method, developed based on Nishimura et al. (2013), was applied to the detrended GNSS time series (with the reference station subtracted), using an Akaike’s information criterion difference (AIC) threshold of < -8.5 to identify potential events. For each detected SSE, we performed a slip inversion using rectangular fault Green’s functions to estimate fault geometry, location, and slip. The inversion was conducted using the Powell method implemented in scipy. The event duration was determined by fitting a double-hinge line to the stacked displacement time series, and the moment magnitude was subsequently calculated. We then extracted earthquakes of Mw > 2.5 occurring within ±90 days and 100 km of each SSE’s fault center, using data from GeoNet (January 2004 to May 2025). To assess whether these events represented significant seismicity increases, we applied a simplified version of the detection method by Nishikawa and Ide (2017), based on a temporal ETAS model originally developed for earthquake swarm detection.
Through our auto-detection, we were able to detect 53 SSEs in 17 years, with the magnitude ranging from Mw 5.7 to 7.0, and 29 events had a notable seismicity increase within 60 days from the center date of the SSEs. 24 events out of these 29 events displayed spatial clustering as well, on or close to the estimated faults. In cases where multiple swarms were associated with a single SSE, most occurred within the same spatial region, although one SSE exhibited distinct swarms in separate areas.
This study successfully detected numerous SSEs across the Hikurangi subduction zone and associated seismicity. Our findings highlight the potential for SSE-based forecasting of earthquake likelihood, particularly if future models can distinguish which types of SSEs are more seismically active. This study can contribute to the broader goal of developing time-dependent seismic hazard models in subduction zones where slow slip is frequent and influential.