English  Sprachen Icon  |  Gebärdensprache  |  Leichte Sprache  |  Kontakt


Numerical Plate Tectonics. Computer simulations of an ever-moving earth

von Giampiero Iaffaldano (13.01.2009)

The uppermost 100 Km of the Earth, so-called lithosphere (from ancient Greek ‘lithos’ = rocky + sphere), is in reality not an intact shell around the Earth’s deep interior. As a matter of fact it is broken into several pieces or units, which we refer to as tectonic plates, like for example the Africa or the South America plates. Tectonic plates move continuously in different directions, at velocities on the order of few centimeters per year (as a reference, fingernails in human beings grow about 5 centimeters long every year). Relative motions across plate boundaries can be divergent (plates move apart from each other along mid-ocean ridges), convergent (one plate undergoes the other at subduction zones) or conservative (plates simply slide past each other along transform faults). Differential motion along plate boundaries generates deformation of the Earth surface and is the origin of strong, localized seismic, volcanic activity, and associated risk for the population.

Comparison of observed plate motions 10 million years ago and at present day[Bildunterschrift / Subline]: Figure 1. Comparison of observed plate motions 10 million years ago and at present day. Plate motions exhibit significant variations through time. Note, changes in the convergent motion between Nazca (NZ) and South America (SA) plates. The principle of inertia tells us that a change in plate motion must necessarily be related to a change in one or more forces acting upon plates; therefore such variations serve as probe for the dynamics of lithosphere.

Motions of tectonic plates are traditionally estimated by sampling the record of the Earth magnetic field imprinted on rocks arising from the deep interior along mid-oceanic ridges. This allows geoscientists to reconstruct the history of plate motions over periods of million years. However, only with the advent over the last decade or so of highly precise geodetic techniques like the Global Positioning System (GPS) we are capable to measure present-day plate motions at unprecedented precision and compare them to their motion in the past. The ability to consider past as well as present plate motions is an incredible source of information. The principle of inertia in fact tells us that a change in plate motion (see figure 1) must necessarily be related to a change in one or more forces acting upon plates; therefore plate-motion changes serve as a probe to assess forces in the lithosphere. Unfortunately, geologic events are by nature neither repeatable nor controllable. Hence, we must build sophisticated computer models in order to test hypotheses and to understand the dynamics of how plates move.

Numerical models of global plate motions dynamics linked with processes occurring in the deeper interior[Bildunterschrift / Subline]: Figure 2. Numerical models of global plate motions dynamics linked with processes occurring in the deeper interior. Numerical grid is visible on top of tectonic plates (outlined in solid black). Color scale shows temperature distribution in the Earth interior, beneath South America. Cold dense material (blue) sinks at the subduction zone under South America, whereas hot buoyant material (red) is visible beneath the mid-Atlantic ridge.

Within the Elitemodul THESIS (Complex processes in the Earth: THeory, Experiments, SImulationS) we have developed a novel numerical approach that allows simulations of global lithosphere dynamics linked with processes occurring in the deeper interior (figure 2). We take advantage of highly-efficient clustered computers to simulate the motion of tectonic plates and, by reproducing observed plate-motion changes over time, to learn what is the history of magnitude and distribution of forces acting upon tectonic plates. A fascinating example for how plate movements change over time is represented by the history of convergent motion between Nazca and South America plates during the past 10 million years. A slowdown in the relative convergence of some 30% is documented for these two plates. With the aid of our models we have succeeded in reproducing this change in a computer simulation. What is important is that our numerical models relate the velocity reduction entirely to the growth of the Andean belt, especially the high Altiplano in the vicinity of the plate boundary. The large gravitational load of the Andes, developed over the past 10 million years, has strongly enhanced the contact pressure within the top few kilometers of the Nazca/South America plate boundary, making it far more difficult for the Nazca plate today to subduct under South America as opposed to 10 million years ago (see figure 3). The large pressure increase along the plate boundary thus is the chief tectonic force responsible for the observed velocity reduction. To our best knowledge, this is the first time that plate motion changes are reproduced faithfully in a computer model. Such advance in computational geodynamics allows us to recognize large mountain belts as key players in the dynamics of the lithosphere.

[Bildunterschrift / Subline]: Figure 3. Resisting pressure (red color scale) developed at the interface between Nazca and South America plates (outline in black, coastline in white) following the growth of the Andes (relief in green color scale). Pressure is responsible for slowing down by some 30% the convergent motion between South America and Nazca plates over the past 10 million years (visible in figure 1). This demonstrates that large topographic feature such as the Andean belt generate enough force to control tectonic plate motions.

  • 2003
  • MS, magna cum laude, Physics (Advisor: Prof. Michele Caputo), Dissertation: Experimental and theoretical memory diffusion of water in sand, Universita’ La Sapienza Rome, Italy
  • 2007
  • PhD, magna cum laude, Geophysics (Advisor: Prof. Hans-Peter Bunge), Dissertation: Balancing the budget of plate tectonics along the Nazca/South America plate margin, Ludwig-Maximilians Universitaet Munich, Germany
  • Since 2008
  • Daly Postdoctoral Fellow at Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA. U.S.A

  • Awarded Grants
  • 2005-2006
  • BaCaTec Program funded by the Bavarian government. Project: Global coupled models of lithosphere and mantle dynamics.
  • 2007-2008
  • Vigoni Program funded by DAAD. Project: Joint modeling of subduction zone dynamics using analog and computer models of the lithosphere.
  • Publications
  • G. Iaffaldano and H-P. Bunge (2008). Relating rapid plate motion variations to plate boundary forces in global coupled models of the mantle/lithosphere system: effects of topography and friction, in review.
  • G. Iaffaldano and H-P. Bunge (2008). Strong plate coupling along the Nazca/South America convergent margin, Geology, v. 36 (6), p. 443 – 446.
  • O. Heidbach, G. Iaffaldano and H-P. Bunge (2008). Topography growth drives stress rotations in the central Andes – observations and models, Geophysical Research Letters, v. 35, L08301 (6 pages).
  • G. Iaffaldano, H-P. Bunge and M. Buecker (2007). Mountain belt growth inferred from histories of past plate convergence: A new tectonic inverse problem, Earth and Planetary Science Letters, v. 260, p. 516 – 523.
  • G. Iaffaldano, H-P. Bunge and T. H. Dixon (2006). Feedback between mountain belt growth and plate convergence, Geology, v. 34 (10), p. 893 – 896.
  • G. Iaffaldano, M. Caputo and S. Martino (2006). Experimental and Theoretical Memory Diffusion of Water in Sand, Hydrology and Earth System Sciences, v. 10, p. 93 – 100.