Slow Slip Earthquakes Research Articles

Seismicity and velocity structure in the vicinity of repeating slow slip earthquakes, northern Hikurangi subduction zone, New Zealand

Using an 11-month (April 2012 to March 2013) deployment of Ocean Bottom Seismographs along the northern Hikurangi subduction zone, New Zealand, we located 2005 earthquakes (M 0.7-4.5) and obtained detailed 3-D Vp and Vs models for a region where a spectrum of subduction fault slip behavior is observed. Our analysis over a 120 × 110 km2 onshore-offshore area reveals high Vp/Vs (>1.8) in the subducting 10 km thick Hikurangi Plateau oceanic crust below an area of repeating slow slip events (SSEs), that are assumed to occur on the plate megathrust. Low Vp/Vs is mapped in a 15 km wide domain along the coastline, offshore Gisborne, marked by low Qs/Qp. Seismicity is absent from the core of high Vp/Vs anomalies (∼2.0) and within the band with lower Vp/Vs, but earthquakes are distributed vertically and are widespread in adjacent high Vp/Vs (∼1.85-1.95) regions along the trend of the coastline. Down-dip and along-strike variation of the Vp/Vs structure of the subducting Hikurangi Plateau could be the result of dehydration process caused by reactivated bending-related faults. Zones of vertical extensional fault–fracture in the subducting crust may act as conduits for the discharge of overpressured fluid and episodic hydrofracturing. During our deployment a large SSE (Mw6.8) occurred in February 2013. In response, seismicity became active in central Hawke Bay and at the down-dip edge of a previously mapped high-amplitude reflectivity zone. Microseismicity appeared synchronous with the slip propagation of the SSEs and may have been initiated by the slip. Our results hold implications for understanding of the physical environments conducive to slow slip generation and represent an important baseline for future seismicity work in the region.

Data mining for seismic slip

Seismic data during the time interval between larger earthquakes could contain information about fault displacements and potential for future failure, suggest analyses of data from laboratory and real-world slow-slip earthquakes using machine-learning techniques.

An image segmentation based algorithm for imaging of slow slip earthquakes

Laboratory experiments next to a variety of observations, especially in subduction zones, have explored the existence of a premonitory stable slow slip growth phase preceding large earthquakes. These phenomena play an important role in the earthquake cycle and thus precise imaging and monitoring of these events are of great significance. In the literature, ENIF (extended network inversion filter) has been proposed as a rigorous algorithm capable of isolating signal from different types of noise and thereby provides us with deep insight into spatio-temporal evolution of slow slip events. Despite its considerable advantages, ENIF still suffers from some limitations. ENIF applies Tikhonov method of regularization with a quadratic form of cost function. While anomalous slip regions have clear contrast with the background slip in reality, Tikhonov regularization tends to over smooth (globally smooth) the slipping portion on the estimated images. In order to avoid over smoothing phenomenon, we have incorporated into ENIF an image segmentation step which tries to preserve edges of slow-slip event. As a second limitation, due to the nonlinearity imposed by such constraint as non-negativity of slip rate, uncertainty propagation through model is not simple. As the core of ENIF, EKF (extended Kalman filter), performs uncertainty propagation by linearization of nonlinear model using Jacobian and Hessian matrices. As an alternative for EKF, we have also investigated the application of UKF (unscented Kalman filter) which uses UT (unscented transform) for uncertainty propagation. Finally, we tested our proposed algorithm using a low signal to noise ratio synthetic data set. The results show a significant improvement in the performance of ENIF when the segmentation step is incorporated into the algorithm.

Vertical tectonics at an active continental margin

Direct observations of vertical movements of the earth's surface are now possible with space-based GPS networks, and have applications to resources, hazards and tectonics. Here we present data on vertical movements of the Earth's surface in New Zealand, computed from the processing of GPS data collected between 2000 and 2015 by 189 permanent GPS stations. We map the geographical variation in vertical rates and show how these variations are explicable within a tectonic framework of subduction, volcanic activity and slow slip earthquakes. Subsidence of >3 mm/yr is observed along southeastern North Island and is interpreted to be due to the locked segment of the Hikurangi subduction zone. Uplift of 1–3 mm/yr further north along the margin of the eastern North Island is interpreted as being due to the plate interface being unlocked and underplating of sediment on the subduction thrust. The Volcanic Plateau of the central North Island is being uplifted at about 1 mm/yr, which can be explained by basaltic melts being injected in the active mantle-wedge at a rate of ∼6 mm/yr. Within the Central Volcanic Region there is a 250 km2 area that subsided between 2005 and 2012 at a rate of up to 14 mm/yr. Time series from the stations located within and near the zone of subsidence show a strong link between subsidence, adjacent uplift and local earthquake swarms.

High pore pressure in seafloor related to slow slip earthquakes

Lying within the seafloor off the southern coast of southwestern Japan, the Nankai Trough is the surface expression of the subduction of the Philippine tectonic plate beneath the Eurasian plate. In the past, this fault has been responsible for producing large, tsunami‐generating earthquakes—the most recent one, a magnitude 8.2 event in 1944. Similar events are expected in the future. One geologic feature that may predispose the region to such earthquakes is the presence of a low velocity zone previously identified that stretches shoreward from the Nankai Trough. The low velocity zone, a subsurface region through which primary seismic waves propagate with reduced velocity, may be related to increased seafloor uplift during an earthquake, a property that would tend to generate tsunamis.

Seismogenic zone temperatures and heat-flow anomalies in the To-nankai margin segment based on temperature data from IODP expedition 333 and thermal model

The location of the seismogenic zone across the To-nankai margin segment has been widely investigated using geodetic, tsunami, seismologic, and heat-flow data as well as thermal models. Seafloor heat-flow measurements show large scatter, raising questions about the thermal state of the incoming and overriding plates. In this study, new temperature and conductivity measurements recorded in the Philippine Sea Plate (PSP) and in the accretionary prism during IODP Expedition 333 are integrated with recent data on the margin structure and seismicity to improve constraints on the thermal structure of the incoming plate and the subduction zone as well as the thermal conditions within the seismogenic zone.IODP Expedition 333 measurements provide heat-flow values that are lower than the average of nearby seafloor measurements. Thermal modeling for the PSP suggests that hydrothermal warming is insignificant in the trench. Thus, an observed widespread positive thermal anomaly in the Shikoku Basin that peaks at the Kashinozaki knoll more likely results from vigorous thermal convection in the upper mantle in back-arc context.The modeled thermal structure of the margin from the trench to the mantle shows a significantly colder interplate contact than in previous studies. The seismogenic zone is likely to be 30–170km from the deformation front, corresponding to the 100–340°C temperature range. The 1944 co-seismic slip zone extends farther southward where temperatures are about 60°C. Slow-slip earthquakes are located in the downdip transition zone between 340 and 440°C across the intersection of the Moho and the megathrust fault. This improved thermal structure from the trench to the mantle results mainly from a better estimate of oceanic plate age, slab dip and sedimentation rate variations in the trench.This last parameter cools the interplate contact until a depth of ∼35km, shifting the thermally defined seismogenic zone landward by ∼20km and significantly diminishing its influence near the corner flow area.

Clay anomaly responsible for Mexican slow slip earthquakes

Over the past decade, seismologists have uncovered a previously unknown class of earthquakes. Known as slow slip earthquakes (SSEs), these events are marked by their long duration; periodic or repetitive nature; and substantial, though harmless, increases in the velocity of intersecting tectonic plates. In southern Mexico the subduction of the Cocos plate under the North American tectonic plate has been found to harbor SSEs that last for up to a year. Compared to traditional earthquakes, which occur because of a sudden lurch between two frictionally locked plates, SSEs have been traced to the deeper rock underlying this locked layer. The rock properties that lend this deeper layer to the occurrence of SSEs, however, are poorly known. Measuring how seismic waves propagate through the varying rock layers of the south Mexican fault, Song and Kim identify the properties of rock that make this particular system susceptible to slow slip earthquakes.

Seismic and geodetic investigation of the 1996-1998 earthquake swarm at Strandline Lake, Alaska

SUMMARY Microearthquake swarms occur frequently in volcanic environments, but do not always culminate in an eruption. Such non-eruptive swarms may be caused by stresses induced by magma intrusion, hydrothermal fluid circulation or possibly other tectonic processes such as slow-slip earthquakes. The Strandline Lake region of southcentral Alaska, located 30 km northeast of Mt Spurr volcano, experienced an intense earthquake swarm between 1996 August and 1998 September. A total of 2507 earthquakes were recorded by the Alaska Volcano Observatory's permanent seismic network during the swarm period, with a maximum magnitude of ML 3.6. The cumulative seismic moment of the swarm was 1.2 × 1015 N m, equivalent to that of a single Mw 4.0 earthquake. Because of the swarm's distance from the nearest Holocene volcanic vent, seismic monitoring was minimal and gas emission and GPS data do not exist for the swarm period. However, combined waveforms from a dense seismic network on Mt Spurr and from several regional seismic stations allow reanalysis of a representative set of swarm and background earthquakes. Swarm hypocentres calculated using a newly formulated 1-D velocity model and station corrections indicate a roughly circular swarm volume with dimensions of approximately 5 km, centred below 10 km below sea level (BSL). Composite fault-plane solutions for swarm earthquakes indicate oblique strike-slip faulting with a northeast-trending P-axis orientation. In contrast, a composite fault-plane solution for background earthquakes indicates a fundamentally different, though poorly constrained, local stress field orientation. Interferometric Synthetic Aperture Radar images spanning the swarm period unambiguously show no evidence of surface deformation, but do not rule out subcentimetre-scale deformation during the swarm. While a shallow ( 10 km BSL) magma intrusion representing an intrusive or protovolcanic segment of the Aleutian arc.

Modeling monthlong slow slip earthquakes

The slow crawl of the Earth's tectonic plates is periodically punctuated by the release of decades or centuries of accumulated stress. Standard earthquakes, lasting from fractions of a second to minutes in duration, are known to range from imperceptible to those that can flatten a city, but recent geodetic measurements have led to the discovery of a previously unrecognized style of fault slip—episodic slow earthquakes. These slow earthquakes recur roughly annually, involve a weeklong or monthlong speeding up of a plate's motion by a factor of 10—100, and are associated with low‐level seismic tremors. First detected about 10 years ago in Japan and western North America, they occur along the same faults that produce devastating magnitude 9 earthquakes but at greater depths. How and why slow earthquakes differ from regular earthquakes, and the extent to which they might foreshadow the shallower truly great earthquakes, are currently topics of wide interest. (Geochemistry, Geophysics, Geosystems, doi:10.1029/2010GC003386, 2011)

Coseismic source model of the 2003Mw6.8 Chengkung earthquake, Taiwan, determined from GPS measurements

A coseismic source model of the 2003Mw6.8 Chengkung, Taiwan, earthquake was well determined with 213 GPS stations, providing a unique opportunity to study the characteristics of coseismic displacements of a high‐angle buried reverse fault. Horizontal coseismic displacements show fault‐normal shortening across the fault trace. Displacements on the hanging wall reveal fault‐parallel and fault‐normal lengthening. The largest horizontal and vertical GPS displacements reached 153 and 302 mm, respectively, in the middle part of the network. Fault geometry and slip distribution were determined by inverting GPS data using a three‐dimensional (3‐D) layered‐elastic dislocation model. The slip is mainly concentrated within a 44 × 14 km slip patch centered at 15 km depth with peak amplitude of 126.6 cm. Results from 3‐D forward‐elastic model tests indicate that the dome‐shaped folding on the hanging wall is reproduced with fault dips greater than 40°. Compared with the rupture area and average slip from slow slip earthquakes and a compilation of finite source models of 18 earthquakes, the Chengkung earthquake generated a larger rupture area and a lower stress drop, suggesting lower than average friction. Hence the Chengkung earthquake seems to be a transitional example between regular and slow slip earthquakes. The coseismic source model of this event indicates that the Chihshang fault is divided into a creeping segment in the north and the locked segment in the south. An average recurrence interval of 50 years for a magnitude 6.8 earthquake was estimated for the southern fault segment.