Zum Inhalt Zur Navigation

Profil der Anden Logo TU Berlin Logo GFZ Potsdam Logo Uni Potsdam

Subproject F1: 3D patterns of mass transfer and deformation resulting from oblique convergence along the Chilean forearc

 

Research Field: Geophysics, Geodynamics, Numerical and Analogue Process-Simulation, (Neo)tectonics

 

 

Staff

Project-Leader(s)

Dr. Nina Kukowski

 

Co-Leader(s)

Prof. Dr. Georg Dresen

Prof. Dr. Hans-Jürgen Götze

Dr. Christoph Janssen

Prof. Dr. Onno Oncken

 

Members

Dr. Katrin Huhn

Dipl.-Geophys. Antje Kellner

Dr. Jo Lohrmann

Susann Wienecke

 

Working area of the subproject F1

 

f1_working_area

Fig. 1: Working area of project F1

 

Abstract

The transmission of forces as well as related deformation and mass transfer at convergent margins are thought to be the key elements for understanding forearc evolution and its role in subduction orogeny. At the Chilean margin, the fundamental mass transfer and deformation patterns observed at its accretive and erosive portions, respectively, are qualitatively understood based upon the results of geological and 2D seismic data as well as 2D simulation with analog materials. As convergence along the Andean margin is oblique, a quantitative understanding of the regional mass transfer and deformation patterns including their controlling parameters requires 3D simulation. An equivalent open problem is the aspect of force transmission in an oblique environment, in particular with the effect on subduction orogeny with or without the formation of non-collisional plateaus. Of special interest in this respect are various types of plate-margin parallel strike-slip faults (Atacama Faults, Precordilleran Fault, Liquiñe Ofqui Fault (LOF)) which segment or border the forearc against the interior of the orogen. We therefore propose an integrated multidisciplinary study on the composition, structure, and dynamics of the Andean forearc.

 

We plan to address this aim with a quantitative investigation of the 3D stress patterns, deformation, strike-slip fault evolution and mass transfer modes with several simulation methods. These include sandbox analog experiments, static stress modeling, dynamic finite element stress modeling, and fault zone modeling with distinct elements. We will focus on two model areas, an erosive part of the forearc between 20°S and 25°S and an accretive portion of the margin between 36°S and 41°S. We plan to systematically evaluate the parameters which we identify as the main factors controlling forearc and orogen evolution. These will then be evaluated with a more global perspective in terms of their role in controlling forearc evolution and stress transfer at convergent margins.

 

 

Objectives, methods, work plan, and schedule

The aim of subproject F1 is to model controlling factors for the evolution of the upper plate forearc and its boundary towards the arc at different scales by simulating 3D deformation and mass transfer modes in two key-areas (20°S to 25°S and 36°S to 41°S) of the Andean forearc which represent the erosive part (northern model area) as well as accretive part (southern model area) of the Chilean convergent margin. This will lead to a better and quantitative understanding of the parameters controlling the relations between both plate kinematic and dynamic boundary conditions and transmission of forces towards the hinterland. Therefore, we propose to evaluate "internal" and "external" parameters possibly controlling orogenic evolution. In order to achieve this goal, we will apply multi-disciplinary, integrated modeling approaches including transient numerical model calculations (Finite Element Methods FEM, Distinct Element Methods DEM), analogue sandbox experiments, and "static" modeling (Fast Fourier Transformation FFT). The focus of the proposed work will be on the following main goals and specific questions related to each of them:

 

Thus, within subproject F1 we will investigate how much the forearc, its variability and evolution influences the development of the orogen and how strongly the forearc is coupled to the main orogen. As the margin parallel strike-slip faults have a complex history with motion along them occurring episodically, special emphasis is put upon their role in coupling the individual domains of the Andean orogen. To achieve this, we need a more thorough understanding of the properties of the fault zones and the history of their activity.

 

The other interface we propose to investigate is the plate interface and the variability of its properties along strike. Parameters that clearly differ between the northern and southern working areas are sediment input, heterogeneities within the incoming sediment, climate and probably also position, friction, and roughness of the plate interface. Therefore, we identify these parameters as the most challenging candidates for a quantitative analysis of the role in controlling forearc and also orogen evolution. Due to the obliquity of convergence, achieving a complete understanding of mass transfer modes and deformation patterns requires a 3D approach.

 

Our model areas basically comprise about 5° "long" slices of the entire forearc from the trench to the plate margin parallel strike-slip faults with the plate interface as the lower boundary. We will extend these areas eastward beyond the plate margin parallel strike-slip faults to investigate their role not affected by boundary conditions and to generate a certain overlap with the target transect of subproject G2 (Handy et al.) for better comparison and calibration of mutual results.

 

As "internal" parameters we mainly identify physical parameters, namely Young's modulus, internal friction, rigidity, and their spatial and temporal variation of the upper plate as well as the properties of the plate interface, i.e. its friction and roughness as well as its position relative to the top of the basement of the lower plate. As "external" parameters we identify boundary conditions like convergence rate and obliquity (Somoza, 1998), the age and thickness of the downgoing oceanic plate, and the thickness, composition, and properties of sediments entering the trench as well as their variation along strike and with time. We will focus on these parameters as they so far appear to be the most promising candidates as first order controlling parameters. Knowledge of the actual spatial distribution of these parameters will partly come from other subprojects (F3 Shapiro et al.: seismic properties; F2 Klotzet al.: actual direction and magnitudes of horizontal displacements as derived from GPS measurements). To evaluate these parameters we will need to undertake intensive parameter sensitivity studies. Within subproject F1, for the present state of the Andean forearc, we are able to examine the following parameters with "static" modeling:

 

We will start with an input from 3D-density models of the lithosphere obtained by project F4 (density and rigidity distribution in 3D); stress modeling will include the (1) fore- to back arc and (2) volcanic arc.

 

Stress responds to high topography and crustal density distributions which will be modeled by density models under the assumption that mechanical properties (e.g. Young's modulus, Poisson's ratio) are known from other SFB-projects. We will compare our results with those of the seismic project (F3) and with similar calculations e.g. in the area of the Cascade Subduction Zone (R. Blakely, USGS, pers. comm.). We expect that the local deviatoric stress differences caused by topography and 3D density distribution are as large as the tectonic stress resulting from plate boundary and body forces.

 

While analogue modeling is restricted to very specific rheologies, i.e. Coulomb rheology for sandbox experiments, which non-the-less have been very successfully applied to investigate mass transfer at convergent margins, numerical methods enable the investigation of more complex rheologies, e.g. visco-elastic, depth dependent Coulomb rheology with strain hardening/softening and the influence of temperature. Here, we aim to apply different rheological concepts including multi-layer rheological approaches to test the role of the parameters expected to be the most likely as main factors controlling the orogen including critical assessment of other yet unidentified parameters. Here again, evaluation of our results will include comparison with similar work applied to other convergent margins (Peru, Makran, Cascadia) to finally extract the overall fundamental parameters controlling mass transfer and deformation in forearcs and their role for orogenic evolution.

 

While previous work on fault evolution was mainly descriptive, we will use in this project existing and new field data to develop numerical modeling in order to quantify fault zone evolution processes. Here, we are interested in whether the intensity of deformation increases progressively towards the fault. We propose to quantify the influence of obliquity, degree and gradient of coupling, shear resistance, and the amount of separation in those portions of the forearc, which are bordered by a plate margin parallel strike-slip faults towards the interior of the mountain belt. To complete existing results regarding fault zone dynamics and deformation mainly obtained for the West Fissure, supplementary field work is necessary for the seismically active part of the LOF.

 

For the present-day situation, a comparison of the results of our simulations with observations is straightforward and will be achieved through co-operation with other subprojects (mainly F2, F3, F4, G1, and G2). This also will enable evaluation of the results of the proposed transient simulation. We will determine if and how the evolution of the forearc has affected plateau formation in the northern working area and investigate the evolution of the non-plateau orogen in the southern model area.

 

Proposed methods

In order to address the specific goals of subproject F1 and to obtain a quantitative understanding of the influence of the proposed factors probably controlling forearc evolution, we will apply several simulation tools including

 

We will apply all of these methods as each of them is most appropriate and efficient for the assessment of the sub-goals of F1. One important aspect of the "element" methods we will apply is that they can be used on a large variety of scales including very heterogeneous spatial resolution of the model area.

3D forward modeling and property inversion (density and susceptibility) will be done with our own software: the 3D modeling program IGMAS (interactive gravity and magnetic application system), FFT and non-linear inversion respectively. An important step is the calculation of model loads of the density models by IGMAS in different depths for the stress calculations.

The total stress field will be separated into hydrostatic and deviatoric components, of which the first is of no relevance here. To ease the calculations of the deviatoric components we assume that the density inhomogeneties in each layer are equivalent to additional loads acting at the layer boundaries. Therefore, all lateral (horizontal) parameter changes are linked with boundary conditions and the equation of equilibrium can be expressed by a standard biharmonic differential equation for a displacement vector after the hydrostatic term was removed.

To determine regional stresses we assume that the underground to be modelled can be divided into several horizontal plates (with constant elastic parameters) of an elastic half space; this leads to linearization and therefore simplification of the numerical problem. The elastic plates are placed on top of a low-viscous fluid. From the conditions of equilibrium of such system it is possible to calculate e.g. its shear- and vertical stress distribution. We developed our own algorithm (in cooperation with M. Kaban) which operates in the wave number domain by FFT. These techniques are fast and robust and enable testing of a large number of different data sets.

3D sandbox experiments will be used to simulate material transfer in case of oblique (and also perpendicular for reference) convergence using granular materials with different mechanical properties (sand, mortar, micro glass beads, micro mica flags) to simulate different rocks. The 3D approach also allows us to take into account along strike variation of parameters like thickness of sedimentary input, friction and roughness of the basal or a possible mid-level decollement. We have already constructed and successfully tested a prototype 3D sandbox to perform such experiments. This prototype will be further developed and adjusted to the specific needs of the planned experiments during the project. Each experiment will be documented with video and with photos at specific convergence intervals, i.e. every few cm, and then will be interpreted using tools familiar to structural geologists, i.e. mapping of fault zones developing during the experiment, characterization of domains of different modes of deformation, etc. To quantify particle paths and finite strains, we will track markers incorporated in the incoming "sediment pile" as well as the "backstop" against which accretion will take place, or which will be affected by erosion, respectively. For this, optical and CAD applications have been developed by our research group including the construction of strain ellipses. While the preparation and performance of a 2D experiment normally is done within some days, thorough processing and interpretation needs about two weeks. This time need will be more than doubled for 3D experiments. Therefore, it will be necessary to carefully decide on which specific experiments will be performed. This will be done with the help of the results of the "static" and "dynamic" numerical simulations. Here, our main focus will be on the northern model areas and therefore erosive mass transfer with some work related to the southern model area, as this still is the least understood pattern of mass transfer.

3D transient finite element simulations on various scales and with different spatial resolution will be performed for both model areas. Finite element methods (FEM) among the numerical methods are most appropriate to explicitly compute directions and magnitudes of the stress tensor. The FEM enables geometrically complex models. Furthermore, wide ranges of material parameters, in particular material gradients and their influence on stress and strain distribution, can be investigated and allows the application of complex rheological approaches. After starting with an overall rheology comparable to the one of the sandbox experiments (Coulomb), we then will incorporate more complex approaches subsequently. Here, we also will adopt the results of thermo-mechanical modeling (old project C2A and proposed project G2) and Curie temperature distribution estimation to extract the influence of temperature on forearc rheology. For these simulations, we will use several commercial codes available at PB 3.1 of GFZ as well as academic software and routines developed in our group.

The distinct element method (DEM) is based on investigations of the mechanical behavior of granular media like sand or glass beads, offering advantages in comparison continuum techniques (FEM) with regard to the possibility to simulate mass transfer. Examination of interparticle stress (which however does not allow the estimation of magnitudes of the stress tensor) and strain as a function of material parameter distribution can be used to investigate the influence and sensitivity of the 'key' parameters for the mechanical behavior of forearc area, similar to FEM studies. Furthermore, effects of boundary conditions, loading, particle size, and particle distribution can be tested against the results of analogue sandbox simulations. This method enables to apply large deformations using a Coulomb rheology. Within the frame of sub-project F1, DEM 1) will serve as a second numerical simulation technique to compare results from FEM simulations with analogue sandbox models and 2) will simulate fault zone processes in more detail. So far, DEM has only been applied in two dimensions in geo-sciences. When applying DEM, we will also restrict our research to two dimensions within sub-project F1. For our investigations, commercial codes as well as an academic source code, which will be further specified for our purposes, is available to PB 3.3 and 3.1 of GFZ.

To achieve the overall aim of the project, all components of the work proposed will be closely linked to one another as is shown in Fig.1. Results obtained on different scales then will be integrated for the final evaluation of parameters

F1_1

Fig. 2: Relation among sub-components of project F1 (fz = fault zone).

 

 

Proposed work

Data compilation and construction of model geometry
This part of the project will be done continuously to define the geometric and kinematic boundary conditions for the modeling studies and to re-format available data to enter the simulations. Relevant geologic data for the respective sections/data cubes at 20°S to 25°s and 36°S to 41°S are:

 

Compilation of existent and collection of still missing data will be performed in two stages: a low resolution, first-order synthesis will be provided for the modeling studies after several months. By the end of the second year, this data base will be continuously supplemented, integrating results from ongoing SFB-projects and by data collected by the Hamburg group in southern Chile, with the aim of providing a more detailed basis for the calibration of simulation results. Definition of the geometry of the forearc wedge and its first-order structures will be performed by assembling the available reflection seismic and wide angle data as well as seismological data in 2D-sections and 3D data cubes, adding the digital topography of the US Defense Agency (800 m grid) and the first order features of the geological maps. This data set will be the basis for the construction of 3D-models (employing GOCAD and 3D-Move) of the analyzed portions of the forearc. These models and the above data will define the modeling space and the constraints for numerical and analogue modeling.

 

Density inversion and static stress modeling

During the first year, data compilation and software adjustment for model input as well as installation of interfaces between computer programs to be used in the course of the project will provide the base for further work. We will also perform modeling of stress components for the regional density models (from F4), including congress presentation of the first results. In the second year, a publication of new results with a preliminary interpretation will be prepared. The main task will be continuing modeling of stress components along the northern and southern traverses, and also modeling in the area of the volcanic arc. Refinement of local density models in the area of the volcanic arcs and completion of regional models will be done in the third year. This will allow comparisons with the Cascadia subduction zone and the Peninsula of Kamtschatka. Results will be presented at IUGS to be held at Florence (Italy) and AGU (San Francisco). The new data sets together with interpretations will be prepared for publication.

 

3D sandbox modeling

In order to set the base for the more specific analogue work, we will need to perform a series of 3D experiments to evaluate parameters relevant for 3D transfer such as the degree of obliquity, along-strike variation of sediment input, etc. This will be done intensively during the first year and completed including the preparation of a publication during the second year. The analysis of 3D oblique erosive mass transfer will be a main focus of our work applying analogue modeling beginning in the second year and being the main target during the third year.

 

This will provide the first 3D analogue work on erosive systems ever done. As the Andean margin is a most prominent margin comprising large erosive parts, we expect very promising results.

 

Regional numerical modeling applying FEM and DEM

3D model simulation will be performed for both model areas. In the first year, after constructing the geometrical models and compiling information on physical parameters, we will focus on the actual state of deformation and displacements which will enable evaluation of presently active modes of force transmission. As we are interested in the evolution of the stress and force transmission with time, we then will extend our model calculations to the time interval between about 25Ma and the present. This as well as extensive parameter sensitivity analyses will be done in the second year. Here we will strongly interact with the subprojects quantifying physical parameters. In the third year, we will integrate these models, compare the results for both model areas and evaluate them by comparison with results from other regions. This will be the base for the final extraction of the main parameters controlling forearc evolution.

 

Research related to the plate margin parallel strike-slip faults

Most of the existing field observations are related to portions of plate-margin parallel faults which presently are not active. Supplementary field work in the northern part of the LOF is therefore necessary at the beginning of the project in early 2002 to observe present deformation. The main purpose of the field work is to examine the spatial extension of the process-zone, the fault zone kinematics, and to determine the amount of offset along the seismically active part of the LOF. This will enable comparison with the results obtained for an aseismic fault zone (Precordilleran Fault) which will be the base for a synoptic model of the evolution and propagation of plate margin parallel strike-slip faults. This model will then serve as input for numerical DEM model calculations, which will provide a parameter study on the distribution of stress and deformation in the vicinity of the fault zones, and which will be integrated in the regional simulation of deformation and stress accumulation in case of oblique subduction.

 

Integrated interpretation

While working with the specific themes and methods proposed for F1 (see also Fig.1) we will continuously integrate the results of the simulations, e.g. compare the results of FEM-modeling and sandbox experiments. A next step towards overall integration will be the incorporation of the results of fault zone modeling into the regional continuum approach simulations which will help to estimate the contribution of the fault zone to force transmission. This integration will begin during the planning of the specific simulations and will be done continuously during the project. In the third year, the internal integration will be extended to work on the integrative interpretation of simulation results coming from F1 and G2 to better estimate the overall significance of parameters controlling subduction orogeny. Related will be the integration of simulation and data interpretation, especially in cooperation with F3 and F4.

 

 

Collaboration with external research groups

Names of important partners in the host countries

The closest partners in Chile are Prof. Dr. G. Chong and Prof. Dr. G. Gonzalez (Antofagasta), both of whom have already been partners in earlier periods during the field studies at 21°S and during operation of the ANCORP-Project. In southern Chile cooperation has been agreed with Prof. Dr. G. Alfaro, Prof. A. Quinzio, and Prof. O. Figueroa (all Concepción), all of which are already involved in the joint GFZ-funded project studying the south Chilean margin. Additional cooperation is planned with Dr. G. Yañez (Santiago) in the fields of numerical and analogue modeling. Excellent working relationships and letters of agreement exist with the national geological survey (SERNAGEOMIN; Dr. Mpodozis), with the national copper industry (CODELCO; Dr. Behn) and with the national hydrocarbon company (ENAP; Dr. Patricio Bravo), all of which have made available a substantial body of existing data and evidence. In Argentina, the closest partner for subproject F1 will be Prof. Dr. V. Ramos, U de Buenos Aires.

 

Collaboration with national and international research groups external of SFB

The geological constraints for the southern working area will be analyzed by the Hamburg group (Prof. Dr. C.-D. Reuther) and by members of section 3.1 at the GFZ - headed by Oncken - who are analyzing the mass transfer and the related kinematics of the south-Chilean forearc and the Liquiñe-Ofqui fault employing structural geologic, thermo-barometric, and geo-chronologic techniques (funded by GFZ). Modeling of the southern area will be organized in close contact with these groups. Offshore data for this area will be obtained from the jointly planned marine survey SPOC with the German research vessel RV Sonne scheduled for fall 2001 and organized by the BGR. Interpretation of these data as well as of earlier data in the north (project CINCA'95) is agreed upon with the PI of both marine surveys, Dr. C. Reichert. Concerning aero-gravity, ongoing collaboration with Dr. G. Strykowski, Geodetical Survey of Denmark, is planned to be continued. We also plan to continue existing collaboration with Drs. S. Lallemand and J. Malavieille, both of Montpellier, regarding analog modeling and collaboration with Prof. Dr. J. Morgan, Houston, regarding numerical modeling, who for several years has been among the very few applying DEM to convergent systems.

 

 

Publications

 

Posters

The SFB 267 poster index is available on the poster page

 

Literature

Reviewed publications

Lindsay, J., de Silva, S., Trumbull, R., Emmermann, R., Wemmer, K. (2001): The La Pacana Caldera, N. Chile: a re-evaluation of one of the world's largest resurgent calderas. - J. Volcanology and Geoth. Res, 106: 145-173.. - []

Yuan, X., Sobolev, S.V. & R. Kind (2002): Moho topography in the central Andes and its geodynamic implications. - Earth and Planet. Sci. Letters, 199: 389-402.

Yuan, X., Sobolev, S.V., Kind, R., Oncken, O., Bock, G., Asch, G., Schurr, B., Greaber, F., Rudloff, A., Hanka, W., Wylegalla, K., Tibi, R., Haberland, Ch., Rietbrok, A., Giese, P., Wigger, P., Röwer, P., Zandt, G., Beck, S., Wallace, T., Pardo, M., Comte (2000): Subduction and collision process in the Central Andes constrained by converted seismic phases. - Nature, 408: 958-961.

Zandt, G., M. Leidig, J. Chmielowski, D. Beaumont and X. Yuan (2003): Seismic detection and Characterization of the Altiplano-Puna Volcanic Complex, Central Andes. - PAGEOPH, 160: 789-807; . - []


© 2006 Freie Universität Berlin | Feedback |
Stand: 31.10.2006