Inter-episodes earthquake migration in the Bohai-Zhangjiakou Fault Zone, North China: Insights from numerical modeling

In this paper, we focus on why intraplate seismic initiation and migration occurs, which has widely been considered to be caused by static stress triggering caused by earthquakes, as well as post-seismic slips. To illustrate the mechanism underlying large earthquakes, in particular the migration caused by two key episodes that occurred after 1500 in the Bohai-Zhangjiakou Fault Zone (BZFZ) of North China, we developed a high-resolution three-dimensional viscoelastic finite element model that includes the active faults with vertical segmentation, their periodical locking, and the lithosphere heterogeneity. We used the birth and death of element groups to simulate stress intensity changes during the two episodes (named Episode I and II), with our results showing that the Tangshan earthquake was primarily triggered by the Sanhe-Pinggu M8.0 earthquake in 1679, whereas the Zhangbei M6.2 earthquake in 1998 was not triggered by earthquakes in Episode I. According to our work, the calculated stress changes in the different segments of the fault zone correspond to the magnitude of the triggered earthquakes. Further, the largest stress decrease was near the Sanhe-Pinggu fault and occurred the largest earthquake in Episode I, whereas the largest stress increase was near the Tangshan fault and occurred during the largest earthquake in Episode II. Given the above, we propose a model for seismic migration to describe the dynamic mechanisms of earthquake migration within the BZFZ and North China, in which the factors affecting both the seismic migration path and intensity primarily include the distance between the triggered active fault and the original fault, the coupling of the active faults, the location and scale of the low-velocity anomaly, its distance from the active fault, and the location and scale of the crustal thinning.

Long Short-Term Memory Networks for Pattern Recognition of Synthetical Complete Earthquake Catalog

Exploring the spatiotemporal distribution of earthquake activity, especially earthquake migration of fault systems, can greatly to understand the basic mechanics of earthquakes and the assessment of earthquake risk. By establishing a three-dimensional strike-slip fault model, to derive the stress response and fault slip along the fault under regional stress conditions. Our study helps to create a long-term, complete earthquake catalog. We modelled Long-Short Term Memory (LSTM) networks for pattern recognition of the synthetical earthquake catalog. The performance of the models was compared using the mean-square error (MSE). Our results showed clearly the application of LSTM showed a meaningful result of 0.08% in the MSE values. Our best model can predict the time and magnitude of the earthquakes with a magnitude greater than Mw = 6.5 with a similar clustering period. These results showed conclusively that applying LSTM in a spatiotemporal series prediction provides a potential application in the study of earthquake mechanics and forecasting of major earthquake events.

The Growth of Earthquake Clusters

Migration of hypocenters is a common attribute of induced injection seismicity and of earthquake swarms, which distinguishes them from aftershock sequences. Spreading of the triggering front is often examined by fitting the time dependence of hypocenter distances from the origin by the pore pressure diffusion model. The earthquake migration patterns however often exhibit not only spreading envelopes but also fast-growing streaks embedded in the overall migration trends. We review the observed migration patterns and show that in the case of earthquake-driven migration, where the new ruptures are triggered at the edge of previous ruptures, it is more suitable to examine the cluster growth as a function of the event index instead of time. We propose a model that relates the speed of seismicity spreading to the average rupture area and the effective magnitude of the hypocenter cluster. Application of the model to selected linearly growing clusters of the 2008 West Bohemia swarm gives an almost linear increase of the measured total rupture area with the event index, which fits the proposed model. This is confirmed by a self-similar scaling of the average rupture area with the effective magnitude for stress drops ranging from 0.1 to 1 MPa. The relatively small stress drop level indicates the presence of voids along the fault plane and a possible role of aseismic deformation.

What drives the growth of earthquake clusters?

<div> <p><span>Migration of hypocenters is a common attribute of induced injection seismicity and of earthquake swarms, which distinguishes them from aftershock sequences. Spreading of the triggering front is often examined by fitting the time dependence of hypocenter distances from the origin by the pore pressure diffusion model. The earthquake migration patterns however often exhibit not only spreading envelopes, but also fast-growing streaks embedded in the overall migration trends. We review the observed migration patterns and show that in the case of self-driven seismicity, where the new ruptures are triggered at the edge of previous ruptures, it is more suitable to examine the cluster growth as a function of the event index instead of time, which often discloses a continuous linear growth during time periods which appeared strongly discontinuous in the coordinate-time plots. </span></p> </div><div> <p><span>We propose a model that relates the speed of seismicity spreading to the average rupture area and the effective magnitude of the hypocenter cluster. Application of the model to selected linearly growing clusters of the 2008 West Bohemia swarm gives almost linear increase of the measured total rupture area with the event index, which fits the proposed model. This is confirmed by a self-similar scaling of the average rupture area with the effective magnitude for stress drops ranging from 0.1 to 1 MPa. The relatively small stress drop level indicates the presence of voids along the fault plane and a possible role of aseismic deformation. Further application of the model to seismic swarms from different areas confirms its validity and potential for distinguishing fluid-triggered seismicity.</span></p> </div>

Underlying mechanism of precursory activity from analysis of upward earthquake migration

In this paper we analyse the upward earthquake hypocentral migration in the ten known subduction zones and discuss a possible mechanism of such migration. The total time of the migration appears to range from 2.5 to 10 years. It leads to the estimation of the average velocity Vz~ 60−300 km yr−1. It probably corresponds to the movement of the forcing agent like stress or deformation wave from depths of the upper mantle (600–700 km) to the level of the lithosphere with subsequent initiation of fluid migration inside the crust to trigger shallow earthquakes. Averaged over all zones upward migration travel time is about 5 years (< Vz > ≈120 km yr−1) that coincides approximately with the period of characteristic temperature variation (El Nino) and crustal seismic periodicity in the Pacific region. These findings are helpful for the study of the seismic precursors and analysis of earthquake triggering.