T31A-1255 0800h
Thin and Thick Skinned Foreland Deformation in the Central Andes: A Numerical Simulation Study
The two main segments of the Central Andean plateau, Altiplano and Puna, demonstrate since the Late Miocene different styles of tectonic shortening. Initially pure shear shortening in the Altiplano plateau switched at 13-9 Ma into the simple shear mode accompanied by formation of one of the world largest thin skinned foreland belt. Further to the south, in the Puna, the pure shear shortening continued until much more recently, gradually transforming into mixed pure and simple shear mode with thick skinned deformation in the foreland (the Santa Barbara System). Through numerical simulation of thermo-mechanical processes we show that different shortening modes - pure and simple shear accompanied by thin or thick skinned tectonics - might be controlled by (i) strength of the foreland uppermost crust and (ii) temperature of the foreland lithosphere. As a numerical tool we use a 2-D parallel thermo-mechanical finite element code LAPEX-2D. The code combines explicit lagrangian finite element FLAC algorithm with particle-in-cell technique. Particles track not only material properties but also full strain and stress tensors minimizing numerical diffusion. We employ Maxwell visco-elastic rheology with temperature- and stress-dependent viscosity, simulating ductile flow, as well as Mohr-Coulomb elasto-plastic rheology, simulating brittle deformation. Both rheological models may experience strain softening. Previous geodynamic models indicated the importance of the lateral temperature variations in the lithosphere on the style of tectonic shortening. However, they failed to reproduce migration of the deformation from the Altiplano plateau into its foreland before the major uplift of the plateau. We show that deformation may easily migrate from the plateau into the foreland by rapidly propagating thin skinned thrust belt as a consequence of dramatic mechanical weakening of the Palaeozoic sediments overlying the cold lithosphere of the Altiplano foreland. The processes in the foreland sediments, which triggered their mechanical failure, remain unclear. We speculate that they might be related to the onset of the hydrocarbon maturation in the foreland sediments after the initial stage of the plateau uplift. Lack of the thick sediments further south, in the Puna foreland, precluded formation of the thin skinned thrusting. If the Puna foreland would have been as cold as the foreland at the latitude of the Altiplano, the Puna plateau would be continuously deformed in a pure shear mode with almost no deformation in the foreland; which is not the case. We show that the existing thick skinned deformation pattern of the Puna foreland might be explained by warmer foreland at the latitude of Puna than that at the latitude of Altiplano, which is in accord with the existing seismic Q-factor observations.
T31A-1256 0800h
Mega-rings Surrounding Timber Mountain Nested Calderas, Geophysical Anomalies: Rethinking Structure and Volcanism Near Yucca Mountain (YM), Nevada
Observed regional mega-rings define a zone $\sim$80-100 km in diameter centered on Timber Mountain (TM). The mega-rings encompass known smaller rhyolitic nested Miocene calderas ($\sim$11-15 my, $<$ 10 km circular to elliptical small "rings") and later stage basaltic features ($<$ 11 my, small flows, cones, dikes) in the Southwest Nevada Volcanic Field. Miocene rhyolitic calderas cluster within the central area and on the outer margin of the interpreted larger mega-ring complex. The mega-ring interpretation is consistent with observations of regional physiography, tomographic images, seismicity patterns, and structural relationships. Mega-rings consist of arcuate faulted blocks with deformation (some remain active structures) patterns showing a genetic relationship to the TM volcanic system; they appear to be spatially associated and temporally correlated with Miocene volcanism and two geophysically identified crustal/upper mantle features. A 50+ km diameter pipe-like high velocity anomaly extends from crustal depth to over 200 km beneath TM (evidence for 400km depth to NE). The pipe is located between two $\sim$100 km sub-parallel N/S linear trends of small-magnitude earthquake activity, one extending through the central NV Test Site, and a second located near Beatty, NV. Neither the kinematics nor relational mechanism of 100km seismically active N/S linear zones, pipe, and mega-rings are understood. Interpreted mega-rings are: 1) Similar in size to larger terrestrial volcanic complexes (e.g., Yellowstone, Indonesia's Toba system); 2) Located in the region of structural transition from the Mohave block to the south, N/S Basin and Range features to the north, Walker Lane to the NW, and the Las Vegas Valley shear zone to the SE; 3) Associated with the two seismically active zones (similar to other caldera fault-bounded sags), the mantle high velocity feature, and possibly a regional bouguer gravity anomaly; 4) Nearly coincident with area hydrologic basins and sub-basins; 5) Similar to features described from terrestrial and planetary caldera-collapse studies, and as modeled in laboratory scaled investigations (ice melt, balloon/sand). Post Mid-Miocene basalts commonly occur within or adjacent to the older rhyolitic caldera moats; other basaltic material occurs marginal to both the outer rings of the interpreted mega-ring system and high velocity pipe. The YM repository may be situated in an isolated structural setting within the mega-ring system; basaltic materials are absent in the block for over 11my for geologic reasons. The mega-ring model may better explain YM area structures (Highway 95 fault), tectonism, and volcanism. Coincident physiographic, geologic, and geophysical features associated with the mega-rings feature, and temporal characteristics of regional seismicity and volcanism suggest the need to critically re-assess regional scale and YM tectonic, seismotectonic, and volcanic models.
T31A-1257 INVITED 0800h
Timescales and topographic expression of lithospheric extension in the western Great Basin, NV, USA
One of the goals of the Plate Boundary Observatory is to determine how continental lithosphere responds to changes in driving forces. Because many geodynamic processes occur on timescales much longer than geodetic recording time intervals, longer term deformation measurements are required. On what timescale, however, should these longer term (geological) measurements be made to allow a meaningful integration with geodetic time series? Traditional geological and tectonic studies appear to indicate that continental fault systems are active continuously for millions of years, whereas more precise paleoseismological measurements often document irregular fault displacement. In order to study the evolution of individual fault systems, measurements are needed on an intermediate timescale: long enough to average over many seismic cycles, but short enough to provide the incremental strain history. For continental fault systems such as those of the Basin and Range Province (c. 10 nstrain at present, e.g., Bennett et al., TECTONICS 2003, Friedrich et al., JGR 2003), the expected time interval is on the order of a few hundred thousand years and the expected signal size should range from several to a few hundred meters. In climatically sensitive regions, such as the Great Basin, such surface deformation features (fault trace in alluvial sediments, triangular facets, etc.) may, on one hand, be preserved extremelly well over several hundred thousand years; On the other hand, however, such regions are also sensitive to weather extremes and medium-term climatic variations (tens of ka) as exhibited during wet periods. In the Great Basin, both cases are represented. For example, (1) on the 10 ka timescale, many internally drained basins filled to large lakes (Bonneville and Lahontan) which left thick sedimentary sections covering most pre-existing fault traces; and (2) on the 500 ka timescale, growth or reactivation of faults affected drainage, erosion and deposition patterns. We document such tectonogeomorphic features along the W-flank of the Shoshone Range and Cortez Mountains and surrounding region. We speculate that these features are the expression of active crustal or lithospheric thinning.
T31A-1258 0800h
Speculations on the Evolution of the East California Shear Zone and Associated Structures by Fault Propagation: Comparison with the Mechanics of Anatolia and the Aegean
Studies of the North Anatolian Fault System (NAFS) in Turkey and Greece have shown that it has evolved by slow growth (propagation) over millions of years. The system initiated about 13 Myrs ago in eastern Turkey as a result of the collision of Arabia with Asia. After crossing north Anatolia it reached the Marmara Sea in western Turkey 6 Myrs ago before extending into the Aegean. Prior to the arrival of the NAFS the Aegean had experienced slow extension for more than 15 Myrs. The compressional deformation field associated with the NAFS on its south-eastern side suppressed Aegean extension to the southeast of the fault. To the northwest increased extension rates resulted in the rapid opening of the North Aegean Trough and the Gulf of Corinth over the last 1Ma. A similar sequence of events can be proposed for the western USA. By about 5 Ma the southern San Andreas had jumped inland allowing the Gulf of California to open, creating the big bend and uplifting the Transverse Ranges. At about this time the Garlock fault propagated to the east reaching the present Owens, Panamint and Death (OPD) Valleys about 3 Ma. Prior to the arrival of the Garlock fault the Basin and Range had been subject to extension for about 20Ma. The processes of propagation suppressed extension south of the Garlock fault east of the present Mojave, and enhanced activity in the OPD valleys. This extension was associated with the onset of volcanic activity and as the extension propagated to the north volcanism appeared in the Mammoth lakes region about 1-2 Ma. Except for the extensive volcanism, the evolution is very similar to that observed in the Aegean. However the OPD faults and their associated mantle structures once formed became activated in shear associated with the American-Pacific Plate boundary at about 1 Ma. A new episode of propagation ensued with shear deformation extending south into the Mojave block and north to rejuvenate or enhance extension in the Carson Sink Dixie Valley region.
T31A-1259 0800h
The World Stress Map Project - Stress Orientations Near Plate Boundaries From Focal Mechanisms
The World Stress Map (WSM) is a global compilation of contemporary tectonic stress in the Earth's crust using a wide range of geophysical and geological stress indicators. In general there is no systematic deviation of SH azimuth (maximum horizontal stress) between the different stress indicators. However, near the San Andreas fault the SH azimuth from focal mechanisms give conflicting information to those derived from borehole breakouts. This had led to the hypothesis that the San Andreas fault is a `weak' fault and casts doubt on the accuracy of focal mechanisms as stress indicators. Assuming that all plate boundaries are potentially weak affects the data quality of stress data from focal mechanisms nearby. We proceed a statistical analysis of the data from the WSM release 2004 to develop criteria which detect focal mechanisms which are probably inaccurate for deducing the SH azimuth. We conclude that the SH azimuth from focal mechanisms is at a greater risk of being inaccurate if the earthquake has a) prevailing strike-slip component, b) occurred within a 100 km corridor around a presumably weak transform plate boundary, and c) if the azimuth from one of the two possible horizontal slip vectors of the focal mechanism is within 20° of the relative plate motion azimuth. This applies for 605 out of approximately 9000 focal mechanisms used in the WSM database. We mark these data sets as `possible plate boundary events', or PBE herein, indicating that these events are probably controlled by the fault geometry and not by the stress field. Case studies at transform plate boundaries reveal that the mean SH azimuth changes in the 100 km corridor by deselecting the PBE. Along the El Pilar fault in Northern Venezuela the change of mean SH azimuth is 33°. The impact on the San Andreas fault is very small (2.4°) due to the exceptionally high amount of stress data available from other sources than focal mechanisms. Examples from the southern Pacific ocean and the central Atlantic ocean show a change in mean SH azimuth of 22.0° and 33.3° respectively.
T31A-1260 0800h
A global coupled model of the lithosphere and mantle dynamics
Understanding the dynamics of global lithospheric motion is one of the most important problems in geodynamics today. Mantle convection is commonly accepted as the driving force for plate motion but, while the kinematics of plate movement is well known from space geodetic and paleomagnetic observations, we lack a rigorous description of the coupled mantle convection-plate motion system. Here we present first results from a coupled mantle convection-global lithosphere motion model following a similar effort by Lithgow-Bertelloni and Guynn. Our plate motion code is SHELLS, a thinsheet FEM code developed by Bird which computes global plate motion and explicitly accounts for faults. The global mantle convection code is TERRA, a high-resolution 3-D FEM code developed and parallelized by Bunge and Baumgardner. We perform simple modeling experiments in which the shear tractions applied to the bottom of the lithosphere arise directly from the mantle circulation model. Our mantle circulation model includes a history of subduction and accounts, among others, for variations in mantle viscosity and strong bottom heating from the core. We find that our results are sensitive to the amount of core heating, an inference that has received renewed attention lately, and that models with stronger core heating overall are in better agreement with observations of intraplate stresses derived from the World Stress Map.
T31A-1261 0800h
Coupling models of crustal deformation and mantle convection: An application of GeoFramework
Crustal and mantle deformation are two closely coupled dynamical systems, usually solved in isolation. To numerically solve this problem, it is desirable to have both the crust and mantle as active components of the dynamics. However, materials composing the crust and mantle respond to loading differently and two different constitutive relations are necessary to describe the rheology of this system. Deformations also occur over a wide range of length scales: from a few 100 m for fault zones to well over 10$^{4}$ km for the largest scales involved in mantle convection. As a result, the numerical cost for a single model to resolve all of these spatial features while also incorporating distinct material types is prohibitive. Coupling two distinct modeling codes within a computational framework is a natural avenue to tackle the multi-material and multi-scale dynamics associated with the crust-mantle system. Using GeoFramework (http://geoframework.org), an extension of the Pyre, Python-based modeling framework, the SNAC and CitcomS codes are dynamically coupled. CitcomS has been used in variety of studies of mantle convection; this finite element package has been entirely reengineered within the Pyre environment. SNAC is based on the FLAC algorithm and is well-suited to modeling crustal deformation because it can deal with linear elastic, Maxwell viscoelastic, or elastoplastic rheology with a Mohr-Coulomb criterion. Using the Pyre-coupled SNAC and CitcomS codes, we run 3D numerical experiments of the extension of lithosphere in the presence of a rising mantle plume within a regional spherical geometry. The full thickness of the crust is simulated with SNAC and mantle convection with CitcomS. In the far field, deformation is partly driven by prescribed velocities described by two diverging plates around a single Euler pole. This specific setting for the problem is intended to help understand the evolution of the Red Sea and Afar triple junction. We will show the dynamic outcome, including surface uplift, for different crustal rheologies, and mantle plumes during the formation of fault-bound grabens. Integrating observational dataset to constrain the models will be the next step.
http://geoframework.org
T31A-1262 0800h
Modeling Supercontinent Cycles in 3D Spherical Convection Simulations With Multiple Continents
In 1966 J. Tuzo Wilson suggested that the Atlantic Ocean basin had closed and then reopened, a process now commonly termed the Wilson Cycle. Since then, numerous paleomagnetic studies have shown that Wilson's original idea may be extended to describe a global cycle, punctuated by the periodic formation of supercontinents such as Pangea, Rodinia, and Nuna, separated by time scales of several hundred million years (Myr). It is generally accepted that these motions are coupled to large scale mantle convection. Early two dimensional (2D) mantle convection models demonstrated the dynamic feasibility of such supercontinent cycles. However, fully 3D spherical mantle convection models incorporating multiple continents have yet to be explored. Here we present the first high resolution, 3D spherical mantle convection models with multiple continents. A global grid spacing of $\sim$50 km permits us to model vigorous convection at Rayleigh number 10$^7$. We study models incorporating between two and ten continents in predominantly internally heated flow with radially stratified viscosity. From these models we find that continents aggregate and disperse cyclically, with periods of several 100 Myr. Supercontinents remain intact for roughly 200 Myr. Moreover, the presence of multiple continents promotes the development of mantle thermal heterogeneity on the longest length scales (spherical harmonic degrees 1 and 2) in some models. These results agree well with geologic observations and place dynamic constraints on global mantle flow models.
T31A-1263 0800h
Thrust Stacking and the Creation and Preservation of Cratonic Lithosphere
Cratons are areas of continental crust, and often the corresponding thick lithosphere, that exhibit long term stability from deformation. One suggested mechanism for the formation of cratonic lithosphere invokes the thrust stacking of proto-cratonic lithospheric material. This leads to the possible conundrum of how cratons can originate from deformative processes and then once formed, resist further deformation over long geologic time scales. To test the physical viability of formation of cratons via thrust stacking, as well as providing a mechanism for stabilization, we conducted numerical simulations and scaling analysis of simple analogues that incorporate a chemical layer of variable rheology within the upper thermal boundary layer of a convecting layer. We found that formation of cratonic lithosphere via thrust stacking is most viable for proto-cratonic lithosphere possessing low effective friction coefficient values. Once formed, preservation depends on the total thickness of the newly formed cratonic lithosphere, as well as the friction coefficient. Higher friction coefficient values and/or greater cratonic lithosphere thicknesses are more conducive to long-term stability as they provide higher integrated yield stresses within cratons. The high yield stress can offset convective stresses and thus, stabilize cratons. Thin cratonic lithosphere or cratonic lithosphere with low friction coefficient values may not provide adequate stability against the increasing convecting stresses, thus providing a potential explanation as to why some cratons are not long lived.
T31A-1264 0800h
Marginal Stability of Thick Continental Lithosphere
Xenoliths record a relationship between lithospheric mantle density, age, and thickness such that Archean-aged lithosphere is significantly thicker and chemically more buoyant than Proterozoic-aged lithosphere, and Proterozoic lithosphere in turn is thicker and more buoyant that Phanerozoic-aged lithosphere. The alteration of ancient lithosphere within plate interiors, independent of spatial proximity to continental margins, suggests that lithosphere may not be perfectly stable. These observations pose the question: what controls the thickness and stability of cratonic lithosphere? We argue that subcontinental mantle lithosphere, not just on its margins, but also within its interior, has persisted in a thermal and mechanical state near the threshold of stability, such that the local Rayleigh number for the lithosphere is close to a critical value, $Ra_c$. Recognizing that both chemical and temperature differences affect stability, we carried out laboratory and numerical experiments to examine how $Ra_c$ depends on chemical differences in density. We find that when a chemically buoyant layer overlies a hotter but otherwise denser layer, analogous to continental mantle lithosphere over asthenosphere, convective stability depends strongly on both the critical Rayleigh number and the buoyancy number, $B$, of the lithosphere-like layer. Sufficient cooling at low buoyancy number results in an oscillatory convective instability whereby the colder, more viscous, but chemically lighter layer is drawn into zones of downwelling flow adjacent to laterally extensive zones of upwelling. The critical Rayleigh number for instability increases with the buoyancy number from as little as $\approx 30$ for $B=0$ to $\approx 1000$ for $B\approx 0.5$. Applied to continental lithosphere in a thermal and mechanical state near the instability threshold, this relationship implies that the lithospheric thickness \emph{must} decrease as the mean density of the lithospheric mantle increases, consistent with the geological record. The experiments emphasize that stability must be assessed not only as a function of buoyancy, but also as a function of Rayleigh number. Insofar as continental lithosphere is in a state close to the threshold of instability, more depleted, less dense lithosphere corresponding to large $B$, requires a greater local Rayleigh number to become unstable. The dependence of $Ra_c$ on $B$ near the threshold of instability can therefore account for the existence of multiple stable states and the relationship between thickness and intrinsic density.
T31A-1265 0800h
Particular Mantle Dynamics Induced by Continental Roots
Continental roots represent extensive regions of seismically fast, cold and chemically buoyant material compared to the surrounding mantle. These thick masses may affect circulation in the mantle and impart thermal conditions at their base which determine to some extent the style of convection. A 2-D Cartesian viscous flow model in which the mantle and lithosphere are described as compressible, Newtonian fluids is used. The continental root is sufficiently viscous such that it does not deform. A range of realistic Rayleigh Numbers are considered. The mantle is internally heated and heat production by the decay of radioactive elements U,K,Th in the lithosphere is incorporated into the model. The amount of continental heat production determines the thickness of the continental thermal boundary layer, and in turn the amplitude of lateral temperature anomalies between oceans and continents. Vertical temperature gradients beneath continents are significantly super-adiabatic with a thick thermal basal boundary layer. The thickness of this basal thermal boundary layer and the amplitude of lateral temperature anomalies determines to what extent continents affect deep mantle circulation below and around the root. We characterize families of solutions in which, depending on the heat production in the continental lithosphere, continents either alter mantle circulation below them or are displaced over time without imposing significant disruption on the mantle dynamic system. Differentiating between the two types of solutions may help to understand the evolution of mantle temperatures and dynamics over geologic time.
T31A-1266 0800h
Elastic thickness and mechanical anisotropy of the lithosphere: Implications for the depth scale of lithospheric deformations.
We have determined the two dimensional coherence between Bouguer gravity and topography in the Canadian Shield to detect anisotropy in the flexural response of the lithosphere. We interpret the increase in the wavelength averaged coherence as indicating the direction where the lithosphere is weakest. Throughout the Canadian Shield with the exception of Hudson Bay, the flexural response of the lithosphere is strongly anisotropic. In general, this anisotropy is correlated with the geology and the weak axis is perpendicular to the main tectonic discontinuities: the Grenville Front and the Appalachian orogen in southeastern Canada, the east-west tectonic fabric of the south Superior Province, the New-Quebec and the Torngat Orogens in northern Quebec and Labrador, the Trans-Hudson Orogen in central Canada. There is also a strong correlation between the mechanical anisotropy and the seismic and electrical conductivity anisotropies where they have been observed. The weak (flexural) axis is oriented perpendicular to the seismic fast axis and the high electrical conductivity direction. In the areas of the Shield where the seismic and electrical measurements are absent, the isostatic response anisotropy is generally observed perpendicular to the boundaries between the main tectonic provinces. Shear-wave splitting results and the impedance tensor in magneto-telluric soundings mostly depend on the fossil strain recorded by the upper mantle while the flexural response is more sensitive to the mechanical properties of the crust and very shallow mantle. Thus in the regions where mechanical, electrical, and seismic anisotropies are correlated, these observations suggest that the same strain field was recorded in the crust and upper mantle during the last tectonic event. They are also consistent with the absence of major subsequent tectonic reworking. On the other hand, anisotropy is absent beneath the Hudson Bay basin, possibly because it was obliterated by a thermal perturbation preceeding the Basin subsidence.
T31A-1267 0800h
Who Needs Isostasy? Non-Isostatic Support for Major Mountain Belts, an Example From the Northern Iran-South Caspian System
What supports mountains and creates deep basins on earth? We present a new mechanism for the coeval development of an isostatically unsupported 2 to 4 km high mountain belt and flanking sedimentary basins. Within the Arabia-Eurasia continent-continent collision zone, contrasts in strength and geometry between northern Iranian continental lithosphere and south Caspian oceanic lithosphere focuses collision-related deformation at the Caspian northern-Iran interface. This process is responsible for Late Miocene to Recent rapid south Caspian subsidence, rapid uplift of northern Iran (Alborz Mountains) and subsidence within the Turkish Iranian Plateau (central Iran). A north-south oriented, 600 km long by 130 km deep, cross-sectional finite element model for the south Caspian and northern Iran system is centered on the continent-oceanic lithospheric interface at 10 million years before present when there was ~10km of sediment overlying the south Caspian oceanic lithosphere and low topography along the south Caspian margin. The model is displaced 90 km from the south with its northern margin fixed, causing the oceanic lithosphere to warp down and the continental lithosphere to warp up adjacent to the interface. Deformation decreases to the north and south with a lower amplitude depression forming south of the up-warped continental lithosphere and a lower amplitude up-warp forming north of the depressed oceanic lithosphere. This model fits geological and geophysical observations from northern Iran and indicates that high mountains can be supported flexurally by horizontal compression without calling on other mechanisms like Pratt isostasy and Airy isostasy. Similarly, the Caspian basin to the north and the central Iranian basin to the south are flexurally depressed explaining the great thickness of the south Caspian basin ($>$20 km) and the development of a broad basin within the Turkish-Iranian Plateau.
T31A-1268 0800h
Convergence and Extension Driven by Gravitational Instability in Continental Orogens
The mechanisms that drive the evolution of convergent continental orogens have been the research focus in many previous quantitative analyses of orogeny. One aspect that is still under much debate is the interplay of convergence and extension often observed in localized orogenic systems such as the Alboran Sea Basin and the Carpathian-Pannonian system. Extension often follows convergence but the spatio-temporal evolution is complex and in some cases extension is coeval with episodes of convergence in the surrounding orogenic belt. In several cases the extensional collapse of mountain belts is associated with dramatic thinning of the mantle part of the lithosphere, more so than the crust. Previous studies have shown that gravitational instabilities may play a fundamental role in the tectonics of mountain ranges. In general, the lithosphere is colder and thereby denser than the underlying asthenosphere. Under some circumstances this may cause the lithosphere to sink into the underlying asthenosphere. In this study we investigate how such gravitational instabilities may affect the evolution of continental orogens. The Alboran Sea and the Carpathian-Pannonian systems are both characterized by a roughly circular region of extension surrounded by arc-shaped mountain belts. Therefore we use axisymmetric finite element models to quantify the mechanisms that control the evolution of these systems. We investigate how parameters, such as density and viscosity (Newtonian or non-Newtonian) affect the evolution. In particular, we show how a crust initially thickened by localized convergence may promote lithospheric gravitational instabilities that cause the collapse of high topography and focused, depth-variable lithospheric thinning observed in the central basins of these regions.
T31A-1269 0800h
Lithospheric heterogeneities and crustal deformation
Crustal deformation depends on the disequilibrium between boundary stresses, either at the base of the deforming lithosphere or at its lateral boundaries, and buoyancy stresses arising form lateral density variations within the lithosphere itself. Based on the thin viscous sheet approximation, we propose a modelling technique, which account for both crustal and lithospheric thickness and density variations. The transport of the crustal moment is given by the continuity equation while the transport of the lithosphere, considered as a thermal boundary layer, is described from the heat equation. Combinations of theoretical and real Earth examples, in both extensional and compressional settings, highlight the fact that lithospheric instabilities and small-scale convection play a prominent role during crustal deformation. In extensional settings, particular attention is paid to the evolution of rifts, using additional insights from analogue modelling. We recognize a three stage evolution, i) boundary stresses dominate, ii) crustal buoyancy stresses dominate, iii) mantle buoyancy stresses counterbalance crustal buoyancy stresses and prevent the collapse of passive margins. In compressional settings, we consider the role of mantle lithosphere heterogeneities in the growth of high plateaus. In the Central Andes, the fast velocity lithospheric keel beneath the Eastern Cordillera may likely explain the kinematics of the Altiplano growth.
T31A-1270 0800h
Reevaluating plate driving forces from 3-D models of subduction
Subducting lithospheric slabs mechanically attached to tectonic plates provide the main driving force for surface plate motion. Numerical models historically simulate slab dynamics as a 2-D process and further simplify the problem into either a density driven model (no heat transfer) or a corner-flow problem (thermal convection) [Christensen, 2001; Enns et al., (in revision); van Keken, 2003]. Recent 3-D global models of density driven flow incorporating a history of plate motion (Conrad and Lithgow-Bertelloni, 2002) have succussfully ruled out slab "suction" (basal shear traction induced by downward flow of the slabs) as a major driving force, but exact partitioning of the remaining forces acting on the slab remain unconstrained. A survey of trenches around the world reveals that over half of the slabs presently subducted in the upper mantle have a discontinuous edge (either a slab tip on a young slab, or the side edge of a slab with finite width) around which mantle can flow: prime examples being slabs in the Mediterranean and Carribean. However, even slabs with a wide lateral extent (and where a 2-D approximation may seem appropriate), show signs of having 3-D complexity. For example, on the surface Tonga appears relatively symmetric, but when the history of subduction is considered, the slab has a twisted, 3-D structure due to significant eastward retreat of just the northern part of an originally N-S oriented trench edge. Similarly the widest slabs, South American and Kamchatka, show seismic anisotropy attributed to trench parallel mantle flow (Russo and Silver, 1994; Peyton, et al., 2001, respectively), while the Aleutian trench has oblique subduction varying in magnitude from west to east, and medium width Central American slab likely has a slab window allowing 3-D flow (Johnston and Thorkelson, 1997). Recent laboratory experiments of subduction have demonstrated the full complexity of flow occuring in 3-D geometry (Kincaid and Griffiths, 2003; Schellart, 2004), owing to the analog slab having a lateral extent smaller than the width of the box. These experiments clearly show subduction of a finite-width slab will generate a flow of material from behind the slab around both the side edges and under the nose of the slab into the mantle wedge. This rollback induced flow establishes a positive feedback with backward hinge migration on the surface, and has significant consequences for the composition and dynamics of the mantle wedge. Here we present results of 3-D numerical experiments aimed to quantify the partitioning between different forces acting on such a slab. These experiments include a high viscosity slab (relative to background mantle), a high viscosity lower mantle and a computational domain large enough so that the flow induced by subduction of a finite-width slab is not constrained by the side or bottom boundaries. We provide a self-consistent force balance and integrate the forces acting over the different portions of the slab, thereby partitioning such forces into specific components. We quantify the force due to rollback-induced flow, and signify its importance as a driving force relative to the other forces present: a net slab pull force, a force responsible for bending the slab at the subduction hinge, and a resistive force due to shear traction on the upper, lower, and nose (if present) surfaces of the subducted slab.
T31A-1271 0800h
Continental mountain belts and subduction zone backarcs
A critical problem of continental tectonics is the existence of long-lived active mobile belts compared with the long-term stability of cratons and platforms. At many continental margin plate boundaries, there are broad mobile mountain belts with a long history of distributed deformation, such as the Cordillera along the western margin of North and South America. Mobile belts cover nearly one quarter of the continental area on Earth, and are key regions of lithosphere deformation and mountain building. However, the origin and longivity of mobile belts are not well-understood. Mobile belts are mobile and deform readily because they are sufficiently weak to be deformed by plate boundary forces, whereas cratons and stable platforms are too strong. We conclude that they are weak because they are hot, and they are hot because they are in present or recent subduction zone backarcs. Nearly all backarcs are very hot, not just those with extensional or rift zones. For present-day continental backarcs (e.g., Cascadia), surface heat flow is commonly ~75 mW/m$^{2}$ over backarc widths of as much as 1000 km, with Moho temperatures of 800-900\deg C, and lithosphere thicknesses of 50-60 km. In contrast, cratons exhibit surface heat flow of 40-50 mW/m$^{2}$, Moho temperatures of 400-500\deg C and a 200-300 km thick lithosphere. The difference in thermal regime results in backarc lithosphere being more than a factor of ten weaker than cratons. Backarcs may be hot because of vigourous thermal convection in the shallow asthenosphere due to the viscosity reduction by water released from the subducting plate. The backarc regions appear to remain hot for approximately 300 my after subduction has terminated. This model provides a solution to another long-standing question in continental tectonics: the origin of the heat for orogeny. Hot and weak former backarcs are the locus of most deformation during continent or terrane collision orogeny, i.e., the vice or inherited weakness model. The orogenic heat required for weakening the crust, as indicated by the observed widespread orogenic granitic plutonism, high temperature-moderate pressure regional metamorphism, and ductile deformation at mid-crustal depths, comes from pre-existing high temperatures in the former backarc, and not from the orogenic deformation process itself.
T31A-1272 0800h
SE Carpathins - unique continental collision?
The Carpathians are a key zone for understanding the post-collisional fate of subducted lithosphere which can only be observed and monitored on a very few spots on the Earth. The slab detachment (?) and the following sinking of the detached lithosphere slab into the deeper mantle are strictly restricted to a few million years. The detachment process itself is rarely observable and not really understood. It is not clear whether the detachment takes place in form of brittle failure suddenly removing the slab or whether visco-elastic processes play a relevant role, so that the detachment is not a real "break", but more likely a long-winded elongation and thinning of parts of the subducted lithosphere. The Vrancea region in the SE-Carpathians is one of the regions where to study this geodynamic process. Seismicity beneath Vrancea is characterized by strong intermediate events repeated 10 years for Mw>6.5 and 50 years for Mw>7.4. A very complex approach is clearly necessary in order to shade light on the processes associated with this very active area. Seismic investigations of the Carpathian orogen area form one component of a few multidisciplinary initiatives. The main objectives are: (1) the architecture of the Tertiary/Quaternary; (2) the presence and geometry of structural detachments; (3) the relationship between crustal structures, deeper (mantle) structure and seismicity; and (4) integration with complementary studies in the Carpathian region, the evaluation and validation of competing geodynamic models. A complex tomography experiment from 1999 shows a good resolution down to 350 km depth. The high velocity body extends between 70 km and more than 350 km, deeper than the deepest event (220 km). The 350 km are key for ocean closure reconstruction as it represents the entire lithosphere subducted in the area of SE-Carpathians. The subduction of a foreland slab is associated to surface deformation as indicated by the crustal seismicity and by leveling comparison data. Repeated GPS measurements gives information on the precise location of deformations which has important implications for determination/confirmation of the main risked areas and therefore on the future developments. In particular, we try to answer if the subducted slab has detached from the foreland plate or not. We also have information on the origin of the convergence (eastern or western plate) and about the (de)coupling of the Carpathians with the foreland plate.
T31A-1273 0800h
Lithospheric structure of the West-Pannonian Basin, based on CELEBRATION 2000 and ALP2002 3D seismic data and mantle xenolith lithology: an integrated approach
The nature of the lower crust and upper mantle can only be revealed using geophysical methods or studying rocks which derive from this deep part of the lithosphere. In our work we tried to combine these methods in anticipation of more realistic view of the deep lithosphere in the Pannonian Basin. The seismic survey was designed to obtain not only in-line recordings along five profiles, but also fan recordings of the off-line shots. The station spacing was changing between 1,5 and 5 km, the average charge of the seismic shots were 500 kg TNT. The use of 409 single channel (Texan, PRS) recorders and 20 seismic sources provided a sufficient 3D ray coverage over an 250km * 240km area, and allowed for construction of a 3D model of the crustal structure. The seismic sections show clear first arrivals up to a distance of 180-200 km. For tomographic modeling 5700 picks of first arrivals were used. The P-wave velocity model was defined at equidistant nodes of the 3D rectangular grid. The distance between nodes was 0.5 km. The initial model used was established by apriori velocity data of the pre-Tertiary sediments. For the tomographic inversion of the areal data we used FAST package developed by C. Zelt (1998). The crustal velocities in the West-Pannonian Basin are relatively low, around 6.1-6.3 km/s in the upper crust, with thin lower crust characterized by velocities around 6.5 km/s, suggesting crustal extension. The Moho depth varies 30-32 km in the Transdanubian Central Range area, 27 km in the Mid-Hungarian Zone, with upper mantle velocities 7.8-8.0 km/s. Peridotite xenoliths hosted in young alkaline basalts from the West-Pannonian Basin (Little Hungarian Plain and Bakony-Balaton Highland) have been also extensively studied regarding their deformation microstructures and geochemical characteristics. Our results suggest that the formation of the Pannonian Basin was associated with deformation, induced by considerable (45-50 km) lithospheric thinning, and geochemical modification of the lithospheric mantle. Xenoliths from the shallow mantle (30-40 km) of the western Pannonian Basin display depleted major element and enriched incompatible trace element compositions indicative of partial melting and subsequent refertilization. Deformation microstructures designate multiple stage complex deformation, characterized by unusual olivine fabrics. Conversely, xenoliths from the deep lithosphere (>40 km) show fertile major element and depleted incompatible trace element contents similar to that of the asthenosphere and exhibit evidence for simple, single stage deformation related to asthenospheric flow. We suggest that the mantle portion represented by these xenoliths was only attached to the bottom of the lithosphere after cooling and the cessation of deformation. Our complex study demonstrates that considerable (40-50 km) lithosphere thinning took place during the formation of the Pannonian Basin, which manifested both in the lower crust and upper mantle associated with geochemical modification of the lower lithosphere. This integrated approach is a powerful tool in reconstructing the evolution of the Pannonian Basin.
T31A-1274 0800h
Kinematic Deformation of the Interior Western U.S. Extensional Regime with Mantle Flow
GPS campaign and continuous data were recorded and compiled for over 1600 stations in order to study the Yellowstone-Snake River Plain volcanic system within the larger tectonic framework of the western U.S. The Yellowstone-Snake River Plain volcanic system lies at the eastern and northern margins of the Basin-Range Province and is associated with the Yellowstone hotspot. Campaign GPS surveys from 1995-2003 have revealed up to 4 mm/yr SW extension across the Yellowstone caldera with respect to stable North America, while the SRP is extending at 2 mm/yr SW and the eastern Basin-Range at 2 mm/yr E-W. The high rates at the caldera are part of the local, time-varying deformation observed there and imply compression between the caldera and SRP. GPS data for the YSRP are combined with GPS data for the Wasatch front, Basin-Range, and the continental margin to determine the velocity and strain field of the western U.S. The resulting deformation field shows the SW motion for the YSRP rotating into E-W motion in the Basin-Range and westward motion rotating to the northwest and northeast at rates of 10-20 mm/yr in California and the Pacific Northwest, respectively. The deformation field was additionally constrained by fault slip data from USGS fault database, which provided slip rates and directions. Once the strain field has been determined from the velocity data, it remains to relate the strain to the distribution of stress. The stress field is consequently calculated from geoid and topography data. The resulting models describe the stress and strain fields of the lithosphere for the YSRP and the Basin-Range. These results are then compared with SKS split directions, which are indicators of flow in the upper mantle. Splitting directions recorded for the Basin-Range, Yellowstone caldera, and Snake River Plain are used. The comparison of the kinematic field with shear wave split alignments will enable us to estimate consistency of motion between the lithosphere and upper mantle, and to observe effects of the Yellowstone Hotspot in the upper mantle.
T31A-1275 0800h
Mantle Deformation and Seismic Anisotropy due to Oblique Collision, South Island, New Zealand
An important question in geodynamics is how does continental mantle lithosphere shorten in transpressional zones? We address this by measuring properties of the upper mantle beneath central South Island, New Zealand. The obliquely-convergent, Australian-Pacific plate boundary passes through the South Island and effectively links two subduction zones of opposite polarities by the Alpine Fault. Crustal deformation is now well documented across this continental transform, with oblique thrusts occurring at the Alpine Fault and distributed deformation to the east of the fault. About 90 km of shortening has occurred across central South Island in the past 7 myr. Deformation in the mantle, however, remains controversial and surface observations are equally explained by a variety of models. We use aftershocks of the 2003 Fiordland EQ to determine Pn beneath the root of the Southern Alps. High Pn speeds of 8.6 $\pm$ 0.1 km/s, as well as thickening of the crustal root from 45 $\pm$5 km in central (Mt Cook) to 49 $\pm$ 6 km in southern (Queenstown-Wanaka) South Island are the main results of our experiment. Comparison with parallel and crossing lines both on- and off-shore suggest similar Pn speeds on the Pacific and the Australian plate boundaries, but high values of Pn anisotropy of 11.5 $\pm$ 2 %\ on the Australian side than the 7.7 $\pm$ 2.7 %\on the Pacific side. We interpret the anisotropy as being due to finite strain in the mantle lid. Two further Pn-anisotropy measurements off-shore of 0 $\pm$ 2.5 %\ and a 6.5 $\pm$ 3 %\, define a E-W boundary to uppermost mantle deformation east of South Island. Furthermore, gravity modelling of the thick low-density crustal root shows, that the Southern Alps are not sufficiently high to compensate for it, requiring a region of positive density contrast in the mantle, and probable widening towards the South. We interpret this region as cold, thickened lithospheric mantle. Concomitant crustal root thickening, widening of Pn anisotropy and the mantle positive density contrast, suggest material accumulation, e.g. extrusion towards the southeast of South Island, and favour the thesis of thickening of the entire lithosphere.
T31A-1276 0800h
Gravitational Potential Energy of the Tibetan Plateau and the Role of Mantle Circulation in Driving the Indian Plate
We present a study of the vertically integrated deviatoric stress field for the Indian plate and the Tibetan plateau associated with gravitational potential energy (GPE) differences. In previous studies, the driving forces for the Indian plate have been attributed solely to the mid-oceanic ridges that surround the entire southern boundary of the plate. Stress magnitudes at the Tibetan plateau are presumed to provide a lower bound to the ridge-push force magnitude that is transmitted and stored as excess GPE at the Tibetan plateau. However, vertically integrated stress magnitude estimates of $\sim6-7\times10^{12}$ N/m in Tibet (Molnar and Lyon-Caen, 1988) far exceed those of $\sim3\times10^{12}$ N/m (Richardson, 1992) associated with GPE at mid-oceanic ridges. This apparent discrepancy calls for an additional force that is required to drive the Indian plate. We use the Crust 2.0 dataset to infer gravitational potential energy differences in the lithosphere. We then apply the thin sheet approach, in which Stokes equations of steady motion, $\frac{\partial{\sigma}_{ij}}{\partial{x}_{ij}} + \rho g \widehat{z}_i =0$, are integrated vertically and then solved to infer a global solution of vertically integrated deviatoric stresses associated only with gravitational potential energy differences. The results around Tibet and the Indian ocean dramatically illustrate the inadequacy of ridge-push forces driving the Indian plate into Tibet. For example, we show that deviatoric stresses associated with GPE differences between the elevated ridges, the deeper Indian ocean, and the elevated Tibetan plateau are insufficient to explain the onset of folding and reverse faulting that is now occurring in the Indian Ocean within the Indo-Australian plate boundary zone. In addition, our global deviatoric stress field solution indicates that both the ridge-push forces ($\sim1.5\times10^{12}$ N/m) and the forces associated with GPE differences around the Tibetan plateau ($\sim3.5\times10^{12}$ N/m) have previously been overestimated by a factor of 2 or more. These overestimates have resulted from either incorrectly simplified 2-D calculations or from defining total stress as $\sigma_{ij} = \tau_{ij} + \sigma_{zz}\delta_{ij}$, in which $\tau_{zz}= 0$, as opposed to the correct 3-D definition $\sigma_{ij} = \tau_{ij} + 1/3 \sigma_{kk}\delta_{ij}$, in which $\tau_{zz} \ne 0$. Our results of a global deviatoric stress field solution associated with GPE differences alone can be used to calibrate the magnitudes of shear tractions that have to be applied to the base of the lithosphere to give the expected styles of stresses in tectonically active regions. Such tractions where they exist are expected to be associated with buoyancy driven circulation of the sub-lithospheric mantle. For Tibet in particular, N-S deviatoric compressional stresses needed to cancel the large N-S deviatoric tension ($\sim3-3.5\times 10^{12}$ N/m) associated with Tibetan plateau GPE can be explained by the coupling of lithospheric dynamics with buoyancy driven mantle flow, most likely associated with the long history of subduction of the Indo-Australian plate, both below Tibet and elsewhere.
T31A-1277 0800h
Lithospheric structure of the Eastern Syntaxis of Tibet using receiver functions
In July of 2003, as part of a larger Continental Dynamics study, a temporary PASSCAL seismic network (Eastern Syntaxis Seismic Experiment) was installed throughout southeastern Tibet, consisting of 50 broad-band and 20 short-period seismometers. The aim of this multi-disciplinary study is to better understand the interaction between surface processes and tectonics. This dense network is used to investigate the ways in which changes in lithospheric rheology control coupling between crustal deformation and mantle flow. The seismic array extends eastward from Lhasa in central Tibet and straddles the eastern syntaxis, centered at the Gyala Peri - Namche Barwa Massif. The massif exposes mid-lower crustal rocks exhumed from depth, providing an opportunity to evaluate models of crustal flow. Initial receiver function results show a Moho depth that varies from 60-80 km depth. West of the Namche Barwa massif, Moho depth ranges from 70-80 km and upper crustal layering is observed. In the vicinity of the Namche Barwa massif the Moho shallows to approximately 60 km. Layering in the lower crust is observed and preliminary analysis of the tangential component of the receiver functions for anisotropy suggests at least one anisotropic layer in the lower crust. Additional analysis will constrain the orientation and magnitude of this anisotropy. East of the Namche Barwa Massif the Moho deepens to 70+ km and layering is present throughout the crust with possible anisotropic layers in the upper crust. The thinning of the crust in the vicinity of the Namche Barwa massif is suggestive of coupling between the upper and lower crust in relation to surface processes. The exact mechanism by which this coupling occurs and how this is related to mantle flow has yet to be determined and will be further investigated.
T31A-1278 0800h
Crustal Deformation and Mantle Flow: The Eastern Syntaxis Seismic Experiment
The Himalaya and high Tibetan Plateau are one of the most remarkable topographic features on Earth, and are widely taken to be the classic example of continent-continent collision. Across the northeastern margin of the Indian plate in southeastern Tibet, the Himalayan orogen terminates abruptly. Collisional processes responsible for the elevation of Tibet and the tectonics of the main Himalayan range are replaced by the strike-slip tectonics of the eastern Himalayan syntaxis. Steep lateral velocity gradients mark the eastern margin of the Indian plate, and incoming Indian lithosphere is partitioned as deeper Indian lithosphere continues north beneath Tibet, and shallower lithosphere decelerates and, together with overthrust Asian lithosphere, enters the clockwise deformation regime of the eastern syntaxis. As part of a larger multidiscipinary study we deployed a temporary seismic array of 50 broadband and 20 short period stations across the transition from the Tibetan plateau to the eastern indentor corner to examine how changes in lithospheric rheology are linked to changes in topography and lithospheric mechanics. We are particularly interested in how deformation in the mantle is coupled to deformation at the surface. The 50 station broadband array extends east from Lhasa, through the eastern Himalayan syntaxis, to the eastern edge of the Tibetan plateau. Station spacing averages 50 km. A denser array of 20 short period stations was deployed in the core of the syntaxis around the Gyala Peri - Namche Barwa Massiff. This massiff is the site of high relief, high topography, and rapid exhumation exposing mid to lower crustal rocks at the surface. Our array was in place from July 2003 through October 2004 recording local, regional, and teleseismic events. Here we present preliminary results from local and regional earthquake location and focal-mechanism solutions. These data are compared with recent GPS results from a co-located array and with shear-wave splitting analysis of SKS phases. The objective is to locate active faults and to understand their contributions in the general deformation of the plateau and to compare deformation at depth with surface observations.
T31A-1279 0800h
Insight into the lithospheric structure and deformation in Eastern Tibet from splitting and traveltime variations of core phases.
The evaluation of the degree of crust/mantle mechanical coupling is essential to better understand the mechanisms accountable for the formation and uplift of the Tibetan plateau. To that end, a dense IRIS PASSCAL seismic array composed of 48 broadband (BB) and 19 short-period stations was deployed in southeastern Tibet from July 2003 to October 2004. The Eastern Syntaxis Seismic Experiment was designed to explore the structural and physical properties of the southeastern Tibetan plateau in order to enhance our understanding of the deformation processes associated with the Indian-Eurasian continental collision. We present preliminary results inferred from the analyses of both shear-wave splitting and multichannel cross-correlation relative arrival time. Both techniques have been performed using teleseismic SKS phases recorded exclusively by broadband stations. Initial SKS analysis reveals the presence of a complicated anisotropic pattern within the Lhasa terrane and the eastern edge of the Qiangtang block. The delay times integrated along the core-receiver path range from null to a maximum of 1.3s near the edge of the Bangong suture.The measured fast polarization directions show spatial variability with a tendency to align close to the direction of the surficial structures. Although the crust is thick beneath the high plateau (60-80 km), the range of delay times implies that the splitting has a significant mantle component. The Fresnel zone approach indicates that the major part of the anisotropy is confined within the lithosphere as opposed to the sublithospheric mantle. One of the more remarkable features observed in our measurements is the south-eastward clockwise rotation of the fast axis of polarization that occurs along and east of the Bangong suture, where the Lhasa and Qiangtang terranes rotate around the eastern Himalayan syntaxis. This rotational anisotropic pattern is remarkably coherent with nearby preliminary GPS observations. The tendency of the fast polarization to align along surficial deformation, suggests coupling between the crust and the mantle in contradiction with the suggested presence of flow in the lower crust. Relative SKS arrival times inferred at all the BB stations using two Pacific events from the Tonga and Samoa Isl. show consistency. The Lhasa terrane appears to be divided into two distinct regions with negative delays (fast) up to 1s in the south and positive delays (slow) up to 1s in the north. These observations are consistent in the west part of the array with Moho thickening to the north, as revealed by a preliminary receiver function analysis (see poster of Zurek et al.). This coherence is not discerned in the eastern part. Although changes in crustal thickness contribute to the arrival times, the amount of delay time variations suggests a stronger dependence to mantle structure with the observed transition perhaps representing the northern edge of the underthrusting Indian lithosphere.
T31A-1280 0800h
Deep Crustal Seismic Investigation Across the Silkeborg Gravity High, Central Denmark
We recently completed a seismic refraction study across northwestern Denmark (the Jutland Peninsula) as part of our efforts to understand the origin and evolution of the Danish Basin. We were testing the hypothesis that the origin of the Mesozoic sedimentary basins in Denmark is related to extensional regimes and intensive magmatic activity during the Carboniferous and early Permian. Potential field data and existing seismic studies indicate that a large batholith is present within the lower crust below the Silkeborg Gravity High (Central Jutland). The goal of project ESTRID (Explosion Seismic Transect across a Rift In Denmark) is to define the shape and size of this batholith. Our main profile (more 140 km long) was laid out along the strike direction of the Silkeborg anomaly. Around 240 Texan instruments were deployed on this profile and ca. 150 more instruments where deployed in arcuate fan geometries, whose centres of curvature were located at the two extreme shotpoints on each end of the profile. These two shots were detonated in water, in the North Sea (shot 1) and in the Arhus Bay (shot 6). Additionally, 4 shots were fired at regular distances along the main profile. Preliminary analysis with seismic tomography locates the top of the crystalline crust at approximately 9 to 11 km depth, below a thick package of sedimentary rocks of Mesozoic and Palaeozoic age. This result is also confirmed by 1D modelling of the data. The tomographic study also defines the extension and depth to the top (ca. 12 km) of the supposed batholith and the depth to the Moho (ca. 30 km). The PmP (Moho reflection) signal is clearly identified along most of the profile, but not in a ca. 30 km long window at the centre of the gravity high. The Pn (diving waves in the upper mantle) is recognizable at the end of the profile. The reflection from the Moho shows a ringing which is interpreted as indicating a layered structure in the lower crust.
T31A-1281 0800h
Evolution of Composition, Thermal Structure, and Resistance to Deformation of Continental Lithosphere Through Time: Why Archean Diamonds Are Probably Proterozoic in Age.
Ductile deformation of lithosphere is temperature-dependent and is also controlled by composition. Understanding evolution of continental lithospheric strength through time, and its related resistance to deformation, thus depends on understanding on the thermal history of continental lithosphere and changes in composition through time. Composition and temperature also control lithospheric density on which gravitational body forces are dependent. Studies of mantle xenoliths of different ages and geotherms in lithospheres from which they originate indicate decreasing stability in mantle lithosphere to Rayleigh-Taylor convection from Archean through Proterozoic to Phanerozoic continental lithospheres. In extrapolating lithospheric conditions back in time, probably the most important parameter to consider is heat flow. Both secular cooling and the exponential decrease in energy release from radioactive decay require that heat loss from the early Earth was significantly greater than from the modern Earth. Although most modern heat loss is through the ocean floor, not all additional heat loss from the early Earth can be accommodated by increased oceanic heat loss, and some increase in continental geotherms associated with increased asthenosphere temperatures is predicted. In addition, unless there has been significant change in the incompatible element concentration and/or distribution within the surviving fragments of Archean lithosphere, the ancient stabilized geotherms in these columns of lithosphere must have been higher than today because of the exponential decay of their intrinsic radioactive heat production. These factors require average continental geotherms in the Archean to be significantly higher than modern geotherms, although the maximum geotherm was probably the same, buffered by the lithospheric solidus. Although gem-quality diamonds are dated by their inclusions to be Archean in age, the continental geotherms probably did not cool into the diamond stability field until Proterozoic times.
T31A-1282 0800h
The Long-Term Strength of Europe and its Implications for Plate Forming Processes
There is considerable debate on the methods to measure the effective elastic thickness, $T_e$, and its significance for the rheology of continental lithosphere. While some authors obtain both low and high $T_e$ from the Bouguer coherence and suggest that high values indicate a strong mantle, others only obtain estimates of $< 25$ km using the free-air admittance and suggest that the mantle is weak everywhere and the strength resides in the crust. We have used topography, free-air and Bouguer gravity anomaly data to estimate $T_e$ in Europe using both spectral techniques. We show that when both techniques are consistently formulated, the $T_e$ structure is equivalent: the old Archean and early Proterozoic terranes have large $T_e$ values ($> 60$ km), indicating that the mantle is strong, whereas young Phanerozoic terranes have low $T_e$ values and hence a comparatively weaker mantle. This strength contrast cannot be explained by a single thermal boundary layer cooling model. Rather, we suggest that it is related to differences in the plate forming processes through Earth's history. During Archean and early Proterozoic a higher mantle temperature and/or volatile content led to a greater degree of melting producing thick, highly depleted lithosphere which strengthened with time as it cooled. However, during the Phanerozoic, partial melting of a cooler and dryer sub-lithospheric mantle led to the formation of plates which are thinner and intrinsically weaker, due to thermal and compositional effects. This fundamental strength contrast has probably contributed to the concentration of deformation in Phanerozoic terranes. These terranes act as stress buffers for the strong Archean and early Proterozoic plates which remain undeformed.
T31A-1283 0800h
Numerical simulation of the thermo-mechanical processes resulting in the major pull-apart basins at the Dead Sea Transform
The continental transform boundary between the Arabian and African plates marked by the Dead Sea Transform (DST) accommodated ca 105 km of relative transform displacement during the last 15-20 Myr. The strike-slip deformation resulted in a string of the pull-apart basins along the DST, with up to 10 km thick sedimentary cover. The reconstruction of the pre-DST basement topography indicates that there was a crystalline basement high close to the DST trace between the Red Sea and the Dead Sea, possibly associated with either regionally thicker crust or thinner lithosphere. Moreover, recent seismological data indicate significant variations of the crustal thickness along and across the DST. We employ a fully coupled thermo-mechanical modelling numerical technique to study 3D deformation at the DST with focus on the investigation of the possible effects of inherited heterogeneity of the lithospheric structure on the origin of the two main pull-apart basins in the Southern part of the DST- Gulf of Aqaba and Dead Sea. Our finite-element, explicit, parallel computing code handles realistic temperature-, strain- and stress-dependant visco-elasto-plastic rheology of the lithosphere and is able to model spontaneous self generation of the faults in the upper crust and zones of the strain localization in the lower crust and upper mantle. The modelling shows that both Gulf of Aqaba and Dead Sea basins may have resulted from interaction between deformations imposed by the plate scale kinematics and inherited variations of the thermal structure of the lithosphere (lithospheric thickness) and variations of the crustal thickness. The models replicate major features of the both basins including their shape, depth, number and spacing of the major faults as well as general features of the strain partitioning between the faults.
T31A-1284 0800h
Comparative analysis of topographic, structural and microseismicity images of the two active transform structures in northern Iceland
The Tj\"ornes Fracture Zone is a dextral transform zone connecting the off shore Kolbeinsey ridge to the on land rift zone in northern Iceland. The present day seismic image of this zone reveals two distinct NNW-SSE lineaments : the Husavik-Flatey Fault and the Grimsey lineament. several scarps and fault zones marks the trace of the Husavik-Flatey Fault whereas no topographic evidence correspond to the seismic clusters recorded along the Grimsey lineament. Both from a topographical point of view and microseismicity distribution, the Grimsey Lineament is similar to the South Iceland Seismic Zone. We analysed time and space distribution of the 1994-2003 seismicity. The maximum magnitude were 4,7 on both Grimsey lineament and Husavik-Flatey Fault. In addition to a permanent low level activity, we observed two typical different types of crisis : both correspond to a huge of events, the first one, not associated with significiant energy relaxation, contrary to the other one. During the 9 years studied, only one high energy crisis occured on the Husavik-Flatey Fault, whereas four events of that type happened on the Grimsey Lineament. Total energy relaxed is the same on both active structures. Seismic activity is restricted to the Western part of the Husavik-Flatey Fault whereas the Grimsey Lineament is active to all along his trace. For the Husavik-Flatey Fault, magnitude distribution agree with the Gutenberg-Richter law for all magnitude larger than 0,9. The b-value is 0,91. Magnitude distribution on the Grimsey Lineament doesn't correspond to a similar single law. We observed a Gutenberg-Richter law only for magnitude ranging 1 to 2,5, larger events indicating clearly a default of large earthquakes. Similar breaked distribution was observed on the South Iceland Seismic Zone before the large events of June 2000. This study suggest very clear difference in the functioning cycle of the two active transform zones in northern Iceland. This differences have to be taking into account for the seismic risk evaluation.
T31A-1285 0800h
The Lithosphere of The East African Rift System: Insights From Three-Dimensional Density Modelling
We use the gravity data that cover the large part of the Afro-Arabian rift system, the eastern branch (Ethiopia-Afar and northern Kenya), in order to produce a regional density model. In an earlier work the new and old gravity data were compiled, evaluated and homogenised using a consistent data reduction procedures. Three basic constraints widely spaced over a 1500 km rift length have been generated between 1969 and 2003 by an international consortium with information from isostatic models, global tomography, geological, geochemical evidences, and petrological and experimental results. These are integrated and applied to the model to constrain upper and lower crustal structures underneath the Rift and Plateau areas. New crustal thickness estimations (Dugda et al., 2004 in press) and inferences from recent velocity models along the axis of the Main Ethiopian Rift (Keller et al., 2004) are added to the density model. Thirty parallel planes cutting across the entire plateau region and Rift system (Afar-Ethiopia-Kenya) are interactively modelled using a starting geometry that invoke asthenospheric upwelling. Densities for the upper crust are calculated using Nafe Drake method, averaged from earlier interpretation and measured ones from the Geological Survey of Ethiopia database (e.g. Geothermal project, GSE petrophysical laboratory, pers. communication). Densities for lower crust are estimated using the approach by Sobolov and Babyko (1994). We used also lower crustal densities calculated by (Simyu and Keller, 1997) for the northern part of Kenya rift. The preliminary model offers a possibility to quantify depth, thickness and volumes of different geological interfaces and bodies. As for example, the estimation of the volume of volcanic constructs on the western plateau of Ethiopia is relatively larger than the eastern plateau. The load map derived from the model indicated maximum crustal loads at the crust/mantle interface (ca. 40km) on the eastern and western flanks of the Main Ethiopian rift. A three dimensional image of the lower interface below the rift shows shallow depths and a large area affected by the asthensopheric upwelling in the northern part of the Main Ethiopian Rift (MER) and Afar rather than in the southern section of the MER and northern part of the Kenya Rift. With more constraints and careful modelling further work is on progress to improve and refine the model and forward important implications (e.g., structure and volume of melt or the low velocity layer along the entire rift, improved crustal thickness, load maps etc.).
T31A-1286 0800h
A Reassessment of the Thermal Plate Model for the Ocean Basins
The relationship between the average depth and age of the ocean floor can be approximated using a model of a cooling plate with constant basal temperature in isostatic equilibrium. In previous studies it has been assumed that the thermal conductivity ($k$), thermal expansivity ($\alpha$) and specific heat capacity $(c_p)$ of the lithosphere does not change with temperature $(T)$ and that the temperature under the ridge axis does not change with depth. We use a plate model incorporating laboratory measurements of $k(T)$, $\alpha(T)$ and $c_p(T)$ and an initial temperature structure corresponding to isentropic upwelling with melting, together with information about bathymetry, sediment thickness, sediment density and post-ridge volcanism in the North Pacific and North West Atlantic, to estimate the thickness of the thermal lithosphere under the oceans. We find that a value of approximately 87 km is consistent with both the average variation of unloaded depth with age and the average heat flux beneath the old ocean floor. We then examine the relationship between residual depth anomalies and variations in the intermediate wavelength residual gravity field.
T31A-1287 0800h
Testing the Utilization of Aeromagnetic Data for the Determination of Curie-Isotherm Depth
We examine the feasibility of using depths to Curie-temperature isotherm in regions with sparse heat-flow measurements, such as intraplate regions of the United States, to detect areas with elevated lower-crustal temperatures. Depth to Curie point is often considered to be a general indicator of temperature at depth. In this study, California was chosen as a test region because of its large number of well-documented surface heat-flow measurements and generally high quality aeromagnetic data. We estimate depths to Curie-temperature isotherm by spectral analysis of aeromagnetic data following the methods of Smith et al. (1974), Shuey et al. (1977), Boler (1978), Connard et al. (1983), and Blakely (1988). However, two modifications are applied to these previous methods as a result of the complex geology of California: We increase the dimensions of sub-regions in a stepwise manner until a peak is observed at the low-wavenumber end of the power spectrum, and we manually fit the observed power spectrum with a theoretical expression that directly yields the depth to the top and bottom of the magnetic layer. In general, we find good agreement between our calculated Curie depths for California and measured heat flow, and we obtain reasonable Curie depths using our modified spectral analysis approach. The Coast Ranges of California are characterized by high heat flow, and our calculated Curie depths are relatively shallow for this province, whereas the Great Valley and Sierra Nevada, characterized by low heat flow, have deeper Curie depths. On the other hand, we also find that the spatial resolution of our method is insufficient to distinguish areas of elevated lower-crustal temperatures in California with lateral dimensions smaller than about 200 km in the absence of heat-flow measurements.
T31A-1288 0800h
Correlation between continent area and elevation
This presentation is motivated by the following questions: (1) What factors determine the mean elevation and thickness of an individual continent? (2) How to explain the positive correlation between the mean height and area of individual continent? (3) Given total continental crust volume, what determines the mean thickness (and hence total area) of all continents? For example, Mean thickness of all continents is about 41 km. Mean land elevation is 874 m, and mean elevation of all continents (including land areas and continental shelves and slopes to 1000 meters below sea level) is about 800 m. Could mean continental thickness have doubled and continental area have halved in the geologic past? I present a first-order model to address these issues assuming that continental mean height is the steady state height controlled by uplift and erosion. The model predicts that it takes longer time to erode a larger continent. Hence mean continental height at steady state increases as continental area increases. This prediction is consistent with the general trend between present-day continental elevation and area (except for Antarctica), and can fit the trend well. This is the first time the relation between continental area and mean elevation is quantitatively explained. The model is further applied to investigate variations of mean thickness of continental crust over the last 600 Myr over which the continental crust mass and seawater volume are assumed to be constant. Because a change in the number of continents leads to change in the area of continents, it is predicted that the mean continental thickness increases as the number of continents decreases. Nevertheless, the thickness variation is small, amounts to about 10% from one continent to six continents. Change in the number of continents leads to a sea level fluctuation of about 0.3 km, with the lowest sea level coinciding with times of supercontinents. This prediction is consistent with prominent lows in sea level curves at the times of Pangea and Rodinia. It is concluded that the number of continents played a major role in Phanerozoic sea level changes.