The last 40 to 50 years have seen the theory of plate tectonics supported by more and more empirical evidence from sea-floor magnetism, seismicity, bathymetry and a growing number of other features that relate to Earth’s dynamism. Yet the original concepts of rigid plates and their dislocation from one another and the underlying mantle have been undermined to a degree by the wealth of data now available. Increasing resolution of seismic tomography is revealing what is happening in the depths of the mantle on which growing confidence can be placed. Matching these increasingly revealing sources of data has been the computing power to try to blend them all with rheological theory and thereby model the way the world works. The latest of these modelling ventures does seem to move plate theory onto a significantly higher plane (Stadler, G. et al. 2010. The dynamics of plate tectonics and mantle flow: from local to global scales. Science, v. 329, p. 1033-1038). The keys to this step are: increasingly sophisticated software that encompasses the contributory factors, akin to models used by mechanical and hydraulic engineers; faster computing that allows a decrease in the size of the 3-D cells used in assessing all the interactions as realistically as possible, and a great deal of graphic creativity so that we can visualise the results. At its centre is varying rock strength, the principal ‘engineering’ input derived from seismic tomography, blended with the gravitational and thermal forces that drive Earth’s ‘engine’.
Stadler et al.’s development divides up the planet into a 3-D mesh whose resolution varies according to the likely complexity of motions within and upon the Earth. For instance there is not much call for detail for what lies below abyssal plains of the ocean floor, so available computing power can be focused on the more intricate parts of the tectonic set-up, especially subduction zones that are both the most spectacular features of the Earth’s behaviour and the source of the main force that drives its surface parts – slab pull. Already the approach is producing more questions than answers. For instance, building in the data that show something of convection in the deep mantle makes the model’s output for the more shallow-seated and better known processes deviate more than expected from what is observed – less comprehensive and more coarse approaches previously seemed to be match deep and shallow processes quite well. This is a difficult topic to express merely in words, but fortunately the paper has been made freely available at http://users.ices.utexas.edu/~carsten/papers/StadlerGurnisBursteddeEtAl10.pdf
See also: Becker, T. 2010. Fine-scale modelling of global plate tectonics. Science, v. 329, p. 1020-1021.