Evolving work budget of fault initiation, linkage and growth within accretionary systems

点击翻译按钮获取中文摘要

Will the next damaging earthquake occur on a fault that has had a long history of slip or will it happen on a newly formed fault that is not yet recognized by geologists? The growth of new faults presents a significant challenge for accurate assessments of earthquake hazards in seismically active regions of the world. Prediction of new fault growth would allow better understanding of the development of active fault systems. Through the accumulation of many earthquakes that each incrementally advance the fault system development, an initial distribution of 1 cm cracks can evolve into a 100-km long fault capable of damaging earthquakes. Because fault growth happens across a wide range of spatial scales and in the subsurface, geologists are unable to directly observe the growth of a new fault. This project uses a combination of physical laboratory experiments and computer models that simulate fault growth to explore this problem in the absence of direct observation. The computer models build on the idea that new faults grow to minimize work and achieve a lower energy state. The computer predictions, which will be compared with laboratory experiments, will simulate the development of new faults in front of an accretionary wedge, the Nankai trough south of Tokyo, Japan. This extremely well imaged subduction zone provides good constraints for testing the crustal fault growth models. The Nankai trough has produced several devastating earthquakes and insights learned from this subduction zone will guide our understanding of other subduction zones such as Cascadia margin off Seattle and the Aleutians in Alaska. Additional desired societal outcomes include full participation of women and persons with disabilities in STEM, development of a diverse competitive STEM workforce through graduate student and post-doctoral researcher training, international collaboration, and enhanced infrastructure for research by development and distribution of new computer modeling code.Accretionary prisms are ideal locations for the study of fault development because the material incorporated at the toe of the prism is relatively un-faulted. Field observations suggest that through-going, prism-scale faults develop via the interaction and linkage of smaller fractures within accretionary systems. This coalescence results in the localization of strain along discrete faults from initially diffuse strain along many small fractures. This project examines the evolution of crack coalescence to fault formation within both scaled physical experiments and the Nankai prism. Accretionary systems will be modeled with an innovative new numerical code, GROW, which predicts fracture propagation through the optimization of external work. Crack coalescence will be simulated within four distinct systems: (1) biaxial and triaxial systems for validation with similarly loaded laboratory tests run at the University of Oslo, Norway, that use tomography to track crack coalescence; (2) scaled physical experiments of accretion run at the University of Cergy-Pontoise, France, that provide known boundary conditions, material properties and detailed quantitative observations of the evolution of faulting and associated deformation; (3) accretion simulations of intermediate complexity and scale that provide a critical step between (2) and (4) and allow investigation of increasingly realistic parameters; and (4) simulation of the Kumano transect across the Nankai accretionary prism, which provides a well-constrained example of the development of in-sequence and out-of-sequence crustal thrusts. For the latter, the proposed megathrust development will be simulated over the past 2 Ma in several successive time intervals. All four suites of models will investigate fault development considering the premise that faults grow to optimize the total work on the system. Many of the models will initiate fault growth and coalescence from initial distributed seed points, that represent weak heterogeneities in the experiment and crust. For all suites of models, the evolving work budget within the systems reveals the competing influences of different deformation mechanisms. By investigating the relationship between strain localization, fracture interaction and various energy consuming processes within each system, the researchers will be able to compare the differences in these relationships between models of cm-scale, laboratory physical experiments and km-scale, natural orogens.