Earthquakes result from fast slip that occurs along a fault surface. Interestingly, numerous dense geodetic observations over the last two decades indicate that such dynamic slip may start by a gradual unlocking of the fault surface and related progressive slip acceleration. This first slow stage is of great interest, because it could define an early indicator of a devastating earthquake. However, not all slow slip turns into fast slip, and sometimes it may simply stop. In this study, we use a numerical model based on the discrete element method to simulate crustal strike‐slip faults of 50 km length that generate a wide variety of slip‐modes, from stable‐slip, to slow earthquakes, to fast earthquakes, all of which show similar characteristics to natural cases. The main goal of this work is to understand the conditions that allow slow events to turn into earthquakes, in contrast to those that cause slow earthquakes to stop. Our results suggest that fault surface geometry and related dilation/contraction patterns along strike play a key role. Slow earthquakes that initiate in large dilated regions bounded by neutral or low contracted domains, might turn into earthquakes. Slow events occurring in regions dominated by closely spaced, alternating, small magnitude dilational and contractional zones tend not to accelerate and may simply stop as isolated slow earthquakes.
We use the discrete element method to create numerical analogs to subduction megathrusts with natural roughness and heterogeneous fault friction. Boundary conditions simulate tectonic loading, inducing fault slip. Intermittently, slip develops into complex rupture events that include foreshocks, mainshocks, and aftershocks. We probe the kinematics and stress evolution of the fault zone to gain insight into the physical processes that govern these phenomena. Prolonged, localized differential stress drops precede dynamic failure, a phenomenon we attribute to the gradual unlocking of contacts as the fault dilates prior to rupture. Slip stability in our system appears to be governed primarily by geometrical phenomena, which allow both slow and fast slip to take place at the same areas along the fault. Similarities in slip behavior between simulated faults and real subduction zones affirm that modeled physical processes are also at work in nature.
23 July 2020 in Geological Society, London, Special Publications
A database of 134 apatite fission track (AFT), and apatite and zircon (U–Th)/He analyses has been assembled for eastern Mexico. Most of these samples have reset ages and track lengths reflecting rapid cooling. Time–temperature histories were modelled for 99 localities, and were converted to depth using a constant gradient of 30°C km −1 . Maps of these results reveal smooth temperature patterns in space and time, indicating that heating was due to regional burial rather than hydrothermal circulation. Cooling began by 90 Ma in the west and 50 Ma along the eastern edge of the Sierra Madre Oriental. These ages mimic the duration of the Mexican Orogeny, which verifies that most of these AFT ages have event significance. The elongate Mayrán Basin, a part of the Mexican foreland basin system, formed and grew across and above the eastern toe of the active Sierra Madre Oriental. This basin subsided between at least 70 and c. 40 Ma, and reached a minimum depth of 6 km. It was a both a catchment and routing system for sediment from US and Mexican sources. The shape of the basin suggests that early outflow was directed through the Burgos Basin into the Gulf of Mexico (GoM). By 50 Ma, some outflow potentially routed southwards through the Tampico Misantla Basin area. The Mayrán Basin subsided until 40 Ma, and then began to uplift and erode. This inversion mobilized the stored sediment and redeposited it into the GoM, filling the offshore Bravo Trough. Volcanism swept eastwards between 90 and 40 Ma, driven by northeastward-directed flat-slab subduction, which may also have driven the contraction. Local subsidence during contraction suggests there was dynamic pull-down created by the underriding flat slab. Subsidence ceased at c. 40 Ma, as volcanism swept back westward and asthenosphere replaced the flat slab. The crust rebounded, creating an ensuing period of massive erosion which peaked around 20 Ma. Southern Mexico was relatively quiet until rapid uplift began in Oaxaca in late Oligocene–early Miocene time. Uplift progressed eastwards to the Chiapas Massif in the late Miocene, commensurate with the eastward translation of the Chortis Block.
We propose a binary classification model rooted in state‐of‐the‐art deep learning techniques to predict whether or not complete‐interface rupture is imminent along a numerical megathrust fault. The models are trained on labeled 2D space‐time input features taken from the synthetic fault system. We contrast the performance of two neural networks trained on three types of data, to determine the relative predictive power of each. The neural networks are able to discriminate imminent complete rupture precursors from everything else, thus providing a relative size and time forecast. Vertical displacements along the fault demonstrate relatively good predictive power. The results confirm previous qualitative observations that precursory deformation scales with upcoming event size, consistent with the preslip model for earthquake nucleation. The methods we propose are adaptable and can be modified to use 3D data in the future.
The framework of the dataset and movies: The data, scripts, and movies provided here support the work of Yannick Caniven, Julia K Morgan, and David G Blank on the role of the along-fault dilatancy on fault slip behavior. This work concerns discrete element modeling using the code Ricebal developed at the Rice University, Houston, Texas, USA. The approach used here allows for simulating the behavior of a strike-slip fault at the seismic time scale, investigating the nucleation processes of slow-earthquakes and fast earthquakes. The dataset: The directory DATASET.zip provides all data and results described in Caniven et al., 2021. The data set provides files containing: - particle positions (b-*.out) - time-series of stress ratio and dilatancy (dy/dx) at the wall of the model (WALL-FRIC_frames_*.txt and Dilation-frames_*.txt) - time-space maps folders containing stacked slip-velocity and dilatancy along-fault profiles (MAP_Dilatancy_*_*.txt) All these files can be used to plot all visualizations presented from Figure 2 to Figure 8 in the related paper. For each figure directory, we provide an example of scripts (*.sh and *.py) that can be used to display the data using the GMT software (Global Mapping Tools) or Python. The provided scripts have been developed to be used with GMT 5.4.4 and Python 3.9.2 (using Spyder 4.2.3 in Anaconda 1.10). The use of other versions may require some adjustments. All scripts from Figure 2 to Figure 6 can be launched straightforwardly without any additional setting. To use the script in the Figure7-8 directory, please follow the instructions in the script header. The supporting movies: Movie S1 shows an example of the evolution of particle displacements at the global model scale, here during the occurrence of the fast Eq4 earthquake sequence (see Figure 7 in the paper). Note the slow nucleation of slip associated with the precursory dilation (see Figure 7b in the paper) before the onset of the dynamic stage and propagation of seismic waves. Movie S2 highlights the detailed evolution of particle displacements across the regions of the isolated slow-earthquake SEq1. The nucleation zone of this event exhibits several areas close to their failure points (apex) at different states that accommodate both fault dilation and contraction in response to the far-field imposed displacement (see Figure 8a in the paper). Movie S3 highlights the detailed evolution of particle displacements across the regions of the fast earthquake Eq3. Most asperities on each side of the main narrow nucleation zone approach failure in a synchronized state leading to general dilation and fast slip (see Figure 8b in the paper).