T43A-01
Deformation Patterns and Subduction Behavior of Continental Lithosphere Entering a Trench
We perform 2-D numerical simulations of continental lithosphere entering a subduction zone, to better understand deformation patterns resulting from subduction of a continental margin. The model consists of a subduction zone in which an attached slab drives subduction of a passive continental margin beneath an oceanic plate. A particle-based 2-D visco-elasto-plastic thermo-mechanical finite element code is employed to study the dynamics of the system. A novel feature of the code is that the resolution of the model can be significantly increased in selected parts of the domain, which allows for self-consistent modelling of mantle-lithosphere interaction. In the present study we employ this feature to study how lithospheric-scale deformation around and within the subduction zone is influenced by surface processes such as erosion, and by flow in the upper mantle. Using systematic 2-D numerical simulations, we explore the parameters that are dominant in controlling near- surface structures, both with regards to changes in topography and trench location, and subsurface features such as Moho undulations. The main parameters that have been varied are: the lithospheric density structure; the lithospheric age and temperature structure; the strength of the lower crust; the presence of a weak zone at the plate interface; the amounts of erosion; the upper boundary condition (free surface versus free slip); rheology (non-Newtonian versus Newtonian, viscous, visco-elasto-plastic); and finally the effect of an imposed slab breakoff. In all cases we track surface uplift, subduction evolution and rock exhumation history. We find that the strength of the overriding plate influences surface uplift and the shape of subsurface deformation, and that the density and thermal structure of the subducting plate affects trench motion. Denser slab roll back, and younger, lighter slabs advance, while neither slab rheology nor the presence of erosion greatly affect trench location. For all cases, we observe some degree of symmetric subduction, but the presence of a weak serpentinite zone between plates helps to decouple subduction.
T43A-02
Instantaneous mantle flow induced by subduction of a freely sinking slab
We conduct three-dimensional (3-D) subduction experiments by a finite element approach to study flow around slabs which are prescribed based on a transient subduction stage from a laboratory model. Instantaneous velocity fields are examined for a slab that sinks freely into the mantle, focusing on the toroidal vs. poloidal components as a function of boundary conditions (BCs), plate width, and viscosity contrast between slab and mantle. Results show that the toroidal flow is important for the circulation geometries in the vertical plane. In particular, the material resumed at surface in the back-arc wedge by the return flow cell below the slab tip is minimal with respect to 2-D models, in agreement with laboratory models. Furthermore, we find that circulation is characterized by an upward flow component close to slab sides that could be important for local tectonic structures at slab edges. We show that BCs affect the magnitude and pattern of velocities. In particular, in proximity of the slab the flow field is similar for no- and free-slip BCs, while strong variations exist elsewhere. Moreover, the characteristic spatial length-scale is given by the box height. By modeling different viscosity contrasts between slab and mantle (η'), we find that significant return flow around edges can only be obtained for stiff slabs and that the strength of the toroidal/poloidal Ratio increases with η', nearly independent of slab width. For η' ≥ 103, the toroidal is ~60-70% of the poloidal component, while we estimate about 40-50% for lower viscosity contrasts. In our models, the toroidal term peaks for slab/mantle viscosity ratios η'max~ 102. This trend is found not only for transient but also for steady-state, rollback subduction. Estimates for effective viscosity contrasts in nature are comparable to, or somewhat higher than ηmax.
T43A-03
INFLUENCE OF FOREARC STRUCTURE ON THE EXTENT OF GREAT SUBDUCTION ZONE EARTHQUAKES
Structural features associated with forearc basins appear to strongly influence the rupture processes of large subduction zone earthquakes. Recent studies demonstrated that a significant percentage of the global seismic moment release on subduction zone thrust faults is concentrated beneath the gravity lows resulting from forearc basins. To better determine the nature of this correlation and examine its effect on rupture directivity and termination, we estimated the rupture areas of a set of Mw 7.5-8.7 earthquakes that occurred in circum-Pacific subduction zones. We compare synthetic and observed seismograms by measuring frequency- dependent amplitude and arrival time differences of the first orbit Rayleigh waves. At low frequencies, the amplitude anomalies primarily result from the spatial and temporal extent of the rupture. We then invert the amplitude and arrival time measurements to estimate the second moments of the slip distribution which describe the rupture length, width, duration and propagation velocity of each earthquake. Comparing the rupture areas to the trench-parallel gravity anomaly (TPGA, Song and Simons 2003) above each rupture, we find that in 12 of the 14 events considered in this study the TPGA increases between the centroid and the limits of the rupture. Thus, local increases in TPGA appear to be related to the physical conditions along the plate interface that favor rupture termination. Owing to the inherently long time scales required for forearc basin formation, the correlation between the TPGA field and rupture termination regions indicates that long-lived material heterogeneity rather than short time-scale stress heterogeneities are responsible for arresting most great subduction zone ruptures.
T43A-04
The Mw 8.8, 1906 Colombia-Ecuador Subduction Earthquake: Seismic Structure and Thermal Regime of the Plate Boundary.
The North Ecuador-SW Colombia active margin underwent four great subduction earthquakes in 1906 (Mw8.8), 1942(Mw7.8), 1958 (Mw7.7) and 1979(Mw8.2). The resulting seismotectonic segmentation of the margin may be a consequence of variable mass transfer, remarkable down-going plate basement relief, and margin transverse and along strike crustal faults. The margin basement consists of accreted oceanic terranes, and is underthrust by the northern flank of the19-km thick Carnegie Ridge, and 7-km-thick and structurally complex, Miocene Malpelo- Yaquina-Buenaventura spreading-transform system of the northern Nazca plate. The oceanic crust is covered by turbidite that strongly vary in thickness along the trench. South of 1°30'N, the oceanic crust shows N35- 75° and N120° faults and lineaments, and a N-trending chain of small seamounts, which delineate en-échelon, small pounded trench basins containing up to 500 m of turbidites. The steep adjacent inner trench slope is devoid of an accretionary wedge, but shows evidences for slope instabilities and mass wasting deposits. A 1-km thick subduction channel that is continuous with the hemipelagic cover of the oceanic crust is imaged landward, over a 30-km distance from the trench. This region shows evidence for thinning of the margin basement by seamount subduction, and correlates with the hypocenter of the 1942 subduction earthquake. North of 1°30'N, an up-to 4-km-thick deep-sea turbidite system has developed in the Colombia trench. The turbidite system is fed by the major Esmeraldas and Patia-Mira canyons and rivers, and dammed to the South by the Galera seamounts. This thick trenchfill appears to be dominantly subducting beneath an erosional margin basement between 1°30'N and 2°30'N, thus forming a km-thick subduction channel that is modulated by the subduction of a buried, large horst and graben structure. This region of thick sediment and rough oceanic crust subduction coincides with the hypocenters of the1958 and1979 subduction earthquakes, which occurred beneath the subsiding fore-arc basin. North of latitude 2°30'N, although the trench fill is thinner (~2 km), the Colombian accretionary wedge has developed and reaches 30 km in width at 3°30'N. The zones where each of the 1942, 1958 and 1979 earthquake rupture zones meet with the adjacent rupture zone have been shown to correlate with along strike changes in the fore-arc tectonic regime from uplift (1942) to subsidence (1958), and oblique compression (1979), and with transverse crustal faults that segment the margin. In addition, a fault-bounded outer basement high and a major splay fault that branches upward from the plate interface may control the seaward limit of the 1958 earthquake rupture zone. Thermal modeling, matching bottom simulating reflector (BSR)-derived heat flow, and heat flow measurements, suggests that the 1958 and 1979 earthquakes nucleated within a central to shallow portion of the seismogenic zone, at temperatures estimated to be ~160°C and that the updip limit of the seismogenic zone is controlled by low temperature (60-70°C) processes, with the exception of the 1958 event, where the updip limit is more likely related to prominent structural features fronting the margin.
T43A-05
Flexural Modeling of the Andean System Using Finite Element Method
The general equation of flexure of the lithosphere in cartesian coordinates is solved using a numerical Finite Element Method (FEM) with triangular elements in non-structured meshes. This alternative way to model bending of thin elastic plates lying over an inviscid fluid allows taking into account lateral variation of rigidity, plate discontinuities and full 3-D representation of loads. The numerical solution was initially compared with the analytical solution of bending of an elastic plate loaded by an uniformally distributed load. The method was applied to model flexure of a plate due to curved orogenic belts and the results were compared with solutions obtained if a 2-D approximation of plates and loads was considered. The proposed numerical method was applied to study flexural deformation of the western edge of the South American lithospheric plate due to the loads of the Andean mountains, using Te =75 km for both continuous and broken plates. The predicted forebulges agree with the observed distribution of positive gravity anomalies paralleling the negative gravity anomalies associated with the high topography of the Andes. Maximum amplitudes of forebulges correlate with Purus Arch in Solimões basin (W Brazil) and the Chaco Pampeana plain (Argentina), and between these two regions, a saddle point occurs over the Pantanal wetland (SW Brazil).
T43A-06
Carpathian Collision vs. Andean Subduction in Potential Fields
The arcuate Carpathian mountain range extends over a distance of almost 1500 km in Central and Eastern Europe. The Carpathians, together with the Pannonian Basin System, belong to the youngest tectonic units in this region. Their complicated structure is the result of several processes, such as convergence, subduction and plate boundary retreat, slab roll-back, transpressive-transtensive collision and asthenospheric upwelling during the neo-Alpine evolution. The Carpathians are typically divided into several distinctive segments: the Western, Eastern and Southern Carpathians (clockwise from left). The Andes, extending along the western continental margin of South America, are the longest system (~8000 km) of high mountain ranges on Earth. They are the result of the convergence of the oceanic Nazca and Antarctic Plates towards the continental South American Plate. The Andes are characterized by extreme along-strike variations, creating segments variable in width, altitude, climate and age. In this work we focus on the Western Carpathians and the Southern Andes at latitudes of 36 degrees - 42 degrees S. Despite their different evolution, a comparison of these mountain ranges shows some similarities in terms of the geological structures and their gravity anomalies. However, a significant difference of the forearc structure is revealed (Carpathian forearc gravity low versus forearc gravity high of the Andean subduction zone). The structural image of the Andean subduction zone was derived based on the 3-D density modelling, performed as a combined interpretation of the gravity anomalies, seismic and geological data. The model of the Carpathians was obtained using the 2-D integrated geophysical modelling and a preliminary 3-D gravity modelling, based on the results of the CELEBRATION 2000 seismic experiment and other geophysical data available.
T43A-07
The Close Connection Between the Subduction Processes, the Morphology of the Continental Margin and the Marine Basins on the accretionary segment of the Chilean Margin
The differences observed in the morphology of the accretionary segment of the Chilean margin and the architecture of the associated marine basins, are indicative of variations in the degree of erosion-accretion and uplift of the prism at different latitudes of the margin. The geometry of the slope basins as wedge-shaped half graben basins is controlled by differential subsidence and uplift of the upper levels of the slope. The dynamics of the deposition and deformation of the sequences suggest alternating tectonic events of extension, compression (or transpression) and relative stability. The main geodynamic elements controlling the tectonic evolution of the Chilean margin can be grouped into factors associated with the subducting oceanic Nazca Plate and those related to mass transfer processes. They influence the geometry and deformation of slope basins. Accretionary and non-accretionary or erosional episodes are linked to temporal variations in sediment thickness of the trench, which is mainly associated with climatic fluctuations. Glacial material flux contributed significantly to the trench infill, and has also likely increased accretion rates. In some cases an accelerated sediment supply to the trench may cause unsustainable rates. This can induce subsidence in the trench and sediment subduction to the base of the prism, which drives more basal underplating and strong uplift localized near surface extension in the entire convergent system. The examination of the seismic lines acquired during a Chilean expedition in the RV Vidal Gormaz suggests differential subsidence controlled by displacement on flanking faults as well as pre-kinematic, syn-kinematic and post-kinematic deposition. The thickness of basins also will influence the amount of sediment which is either distributed from the shelf to the trench, or trapped on the slope basins. The analyzed slope basins, between 33°30' - 36°50' S, exhibit variations along the margin. The deepest basin is observed near the Juan Fernandez Ridge (JFR) intersection with the margin, on seismic line VG02-18 (~1 km). While thinner sequences are observed to the south, along VG02-10 (~0,27 km and ~0,45 km) and VGO2-5 (~0 4 km and ~0,5 km). In proximity to the JFR, the high subsidence rate on a slope basin of the seismic line VG02-18, is influenced by the tectonic erosion at the base of the accretionary wedge linked with the progressive positions of subducting seamounts. The alternation between syn-tectonic and pre-post tectonic deposition documents cyclic tectonic events which can be associated with the accretionary and the non-accretionary, or erosive episodes for the development of the south Chile margin. The most probable mass transfer mode in the South Chilean accretive system is partly frontal accretion and partly underthrusting of sediment entering the trench with potential underplating. Frontal and basal accretion leads to characteristic growth patterns of the forearc area. In this study, episodes of frontal accretion can be associated with pre-post tectonic sequences along the slope, which evidences episodes of stability without deformation of sequences and parallel continuous deposition of sediment. The basal accretion of large volumes of underplated sediment beneath the buttress undergoes deformation and tilting and can be associated with the syn-tectonic units.