Chapter 5

Plate tectonics: a paradigm driven by data

“In the Earth sciences you’ve got to be led by the observations.”

“In the Earth sciences you’ve got to be led by the observations. The whole notion of plate tectonics is so bizarre: that you can have these things [continents] that cover something like a third of the Earth’s surface and the elastic part which actually governs their behaviour is only 30 kilometres thick. It’s completely bizarre and complete nonsense, apparently. But it works.” Looking to the future, McKenzie sees seismology becoming a much more powerful tool for Earth sciences, driven by increasing use of supercomputers. And he looks off Earth to the challenges raises by the discovery of thousands of planets around other stars. Above all, however, he stresses the need to collect data and work through what it means – an approach that he has followed throughout his career. “It’s no good thinking that you can actually do these problems by pure thought; you’ve got to be led by the hand by carefully working out the consequences of the observations,” notes McKenzie. “You’ve got to observe the thing, think it through and decide whether you believe it, not on the basis of whether it fits your preconceptions, but on whether it’s right or not. That’s absolutely crucial.”

“I also get a bit bored with doing things that I’ve been doing for a long time,” admits McKenzie. “I quite like learning new subjects and the one that I’ve been learning recently is anisotropic wave propagation in solids. That’s really an extremely complicated subject but it’s an extremely well understood subject by the applied mathematicians, more than the seismologists, The seismologists have not really kept up with the strange things that the applied mathematicians have been doing in that field. So I’ve had rather fun there.”

Seismologists can detect when there is an alignment of features – fractures, mineral grains or layering – in the rocks through which seismic waves travel. This is known as anisotropyanisotropydirectional dependence of physical properties. ; better processing and modelling mean that this may shed light on recent plate movement. “The plate motions themselves orient the crystals of olivineolivinean olive-green, grey-ish-green, or brown mineral occurring widely in basalt, peridotite and other basic igneous rocks. in the upper mantle, so you can see the whole movement of the plates through the upper mantle, in the part that’s deforming, from the anisotropy of surface waves, which is really something wonderful.”

“There’s been all kinds of ideas about how the Earth was formed from a dust cloud but we haven’t got any observations yet. It’s all speculative and it’s all model-dependent,” notes McKenzie. “Now we are actually getting observations from other planetary systems, they don’t look like anything that people expected. That to me is no surprise, it wasn’t based on sound observations in the first place so why should you believe it? If you go back further, the whole business of the synthesis of elements in the stars, you’d never have got that right without the measurements of the capture cross-sections. You’ve got to be helped by making observations and decide whether you believe the observations or not on the basis of the observations themselves, not on the basis of theory.”

Is there a key factor in earth science research that has underpinned your many insights?