Individual And Joint Effects Of Early


Seismological Laboratory, California Institute of Technology, Pasademãng cầu, CA, USA

Correspondence to: X. Mao,


Bạn đang xem: Individual and joint effects of early

Seismological Laboratory, California Institute of Technology, Pasadena, CA, USA

Search for more papers by this author

Xem thêm: Như Thế Nào Là Nhà Có Mặt Tiền Là Gì ? Bật Mí Những Mẫu Nhà 2 Mặt Tiền Đẹp


Xem thêm: Truy Lùng Danh Tính 9 Loại Hoa Trang Trí Phòng Khách Đẹp Nao Lòng

1 Introduction

Subduction initiation is a vital phase of the plate tectonic cycle since it fundamentally alters the global force balance on tectonic plates. Numerical studies have advanced our understanding of under what circumstances & with what physical processes a new subduction zone can develop (e.g., Gurnis et al., 2004; Nikolaeva et al., 2010; Regenauer-Lieb et al., 2001; Thielmann & Kaus, 2012; Toth & Gurnis, 1998), but uncertainty still exists and key parameters related to lớn subduction initiation remain poorly quantified mainly due khổng lồ the lack of good constraints on numerical models. Subduction initiation can be either induced or spontaneous: induced subduction initiation begins with svào compression và uplift (Gurnis et al., 2004), whereas spontaneous initiation begins with rifting & subsidence (Stern, 2004). Therefore, topographic changes that result from subduction initiation can be used to lớn distinguish different initiation modes and can potentially be used to quantify parameters that control the initiation process. While the importance of topographic change in subduction initiation has been noticed previously (Gurnis et al., 2004), applying topographic changes as a constraint on subduction initiation process at a specific subduction zone has not been addressed, as most of the early record needed to constrain the dynamics is overprinted by later deformation và volcanism for mature subduction zones, lượt thích the well-known examples of the Eocene initiation of Izu-Bonin-Marianas và Tonga-Kermadec subduction zones (Sutherlvà et al., 2006). On the other h&, some possible incipient subduction zones are so young (Gorringe Bank, the Owen Ridge, the Hjort Trench, & Mussau Trench) that the slab may not have sầu yet started to lớn bkết thúc into the mantle (Gurnis et al., 2004).

Luckily, one subduction zone overcomes these limitations: the Puysegur Incipient Subduction Zone (PISZ). The Puysegur Trench and Ridge khung the northern over of the Macquarie Ridge Complex (MRC) defining the Australian-Pacific plate margin south of New Zealvà (Figure 1a). Since about 20Ma, highly oblique convergence beneath the Puysegur Ridge results in a maximum total convergence of 150–200km at Puysegur as suggested by a Benioff zone with seismicity down to 150km depth (Figure 1c) (Sutherland et al., 2006). Subduction-related igneous rocks, especially adakite, which is formed by the partial melting of young oceanic crust under eclogitic facies conditions, are sparsely distributed on the overriding plate at Solander Islvà (Reay và Parkinson, 1997). This confirms that the slab enters the mantle. The morphology of the Puysegur Ridge (Figure 1b) shows a characteristic change from uplift in the southern part, where the total convergence is less, and subsidence in the northern part, the Snares Zone (or Snares Trough), where the total convergence is largest và is roughly consistent with geodynamic models of induced subduction initiation (Gurnis et al., 2004). Discrete flat-topped segments, which are interpreted as the results of subaerial exposure & erosion, are also evident at both northern & southern parts of the Puysegur Ridge (Figure 1d). The southernmost segment is cđại bại khổng lồ sea level (−120m), while a peak subsidence of ∼1800m is found in the Snares Zone in the north, suggesting that the southern part has only experienced uplift, while there was uplift followed by subsidence in the Snares Zone (Collot et al., 1995; Gurnis et al., 2004; Lebrun et al., 1998). The width of the Puysegur ridge also widens northward from less than 50km at 49.5°S to lớn ∼80km at 47.5°S. A confined strike-slip fault zone is found near the peak of the ridge in the south, while a splayed fault zone structure is suggested in the trough of the Snares Zone (Figure 1d). Together, they show that the overriding plate cthất bại to lớn the trench is under compression in the south & potentially in extension in the north (Collot et al., 1995; Lamarbít & Lebrun, 2000). The corresponding trench depth in the south is about 1km shallower than in the north (Collot et al., 1995). These spatial variations in structure along the PISZ are thought khổng lồ represent different time periods in the evolution, and a space-for-time substitution can be made lớn compare the time evolution from 2-D models with the spatial variation along the PISZ.


The two-fault system at PISZ, with a thrust fault at the trench và a vertical fault inboard of the thrust fault, is recognized through the dual rupture mode for large earthquakes & interpretations of multibeam bathymetric, sonar imagery, seismic reflection, và geopotential data (Collot et al., 1995; Ruff et al., 1989). However, how this two-fault structure formed is still open khổng lồ debate. Ruff et al. (1989) propose that the thrust interface formed through the propagation và connection of disconnected small thrust faults behind the vertical fault, while this hypothesis preceded the mapping of these two faults. Collot et al. (1995) propose that it developed through progressive sầu adjustments of two adjacent vertical weak zones under compression, which requires the development of the ridge và differential uplift of the crustal bloông chồng at one weak zone, and the rotation of the other. Here we propose that the two-fault system formed through developing a new thrust fault near the preexisting vertical fault during subduction initiation. Previous studies suggest that there is a transition in the force balances from being forced externally lớn a state of self-sustaining subduction under its own negative buoyancy for induced subduction initiation (Gurnis et al., 2004; Leng & Gurnis, 2011). However, it is not clear whether PISZ is self-sustaining, or if the along strike variation in the uplift and subsidence of the Puysegur Ridge represents this transition. Here we use 2-D geodynamic models, which have sầu a true miễn phí surface to lớn traông chồng topographic changes, and mã sản phẩm cài đặt và boundary conditions tailored for PISZ, to lớn thử nghiệm our hypothesis for the formation of the two-fault system at PISZ, and to explore the factors that control the transition in the force balance, while focusing on the evolution of topography and state of căng thẳng.

2 Method

Although a number of studies have tracked the topography in subduction zone models (e.g., Billen và Gurnis, 2001; Kaus et al., 2008; Gerya & Meiliông chồng, 2011; Zhong & Gurnis, 1994), predicting reliable topographic evolution in subduction zone remains a challenge. By reliable topography, we mean that not only is the predicted topography consistent with that observed but the Mã Sản Phẩm should also be able lớn include most of the important geophysical, petrological, & geochemical processes that affect the force balance, and the model thiết đặt, boundary conditions, & the evolution of material properties need to be self-consistent và properly constrained. During subduction initiation, the driving forces must overcome the resisting forces from the friction on the sliding interface, the bending of lithosphere và the buoyancy of oceanic crust (before the basalt-to-eclogite phase change) (e.g., McKenzie, 1977; Toth và Gurnis, 1998). With the subduction of the downgoing slab, phase changes in the crust lead khổng lồ an increasing crustal mật độ trùng lặp từ khóa, & this part of the resisting force evolves to lớn drive sầu subduction. Fluids released from the subducting crust contribute lớn the decoupling between subducting and overriding plates, which may reduce the resisting force. Also influencing the force balance is the development of topography và surface process, which generate loads that affect lithospheric và mantle dynamics (Kaus et al., 2008, 2010).

A true không lấy phí surface is tracked in pTatin3 chiều (May et al., 2014, 2015), based on the Arbitrary Lagrangian Eulerian (ALE) finite element method, và is used lớn follow the dynamic mantle-surface interactions & the topographic evolution. Initial topography is calculated from isostasy (Figure 2), và topography is updated between time steps with surface velođô thị under the constraint that the vertical topographic change is smaller than 20m lớn avoid topographic oscillation (Kaus et al., 2010). A simplified surface process Mã Sản Phẩm, based on linear topographic diffusion, is implemented (e.g., Avouac, 1996). A 5km thichồng altered basaltic crust is placed on the top of dry pyrolite, & sediments are generated at the surface with our surface process. Density và không tính phí water nội dung for different phase assemblages are gained by referring to lớn precalculated 4-D (temperature, pressure, roông xã type, & total water content) phase maps using Perplex (Connolly, 2005). Darcy"s law is used to migrate miễn phí water, & a linear water weakening is applied khổng lồ the mantle material (Hirth & Kohlstedt, 1996). The Drucker-Prager yield criteria with a maximum yield bức xúc are employed for material plastithành phố, & the accumulated plastic strain is recorded on tracers và used khổng lồ reduce the material friction coefficient & cohesion. Elastiđô thị alters the bít tất tay within the slab, & we include a new visco-elastic formulation called the Elastic Viscous Stress Splitting (EVSS) method (e.g., Keunings, 2000) in our Mã Sản Phẩm. Energy change with shear heating is treated as heat source terms & stored on tracers. Thermal and rheological parameters are given in Table S1 in the supporting information. More details on the numerical implementations of surface process, phase change, và EVSS are found in the supporting information.