Geophysicists Discover How Rocks Produce Magnetic Pulses — The Physics arXiv Blog — Medium
Geophysicists Discover How Rocks Produce Magnetic Pulses
…and how they might use these pulses to predict earthquakes
One of the most intriguing and puzzling of the many phenomena associated with earthquakes are magnetic pulses. For some years, geophysicists have been measuring these pulses in the days and weeks before certain tremors.
For example, during the weeks before the Alum Rock earthquake near San Jose, California, geophysicists recorded a series of unusual low-frequency magnetic pulses with amplitudes of up to 30 nanoTesla. (By comparison, the Earth’s magnetic field has an intensity of about 40,000 nanoTesla.) These pulses increased in number until the day of the earthquake on 30 October 2007.
That raises an interesting question. What causes these magnetic pulses?
Today, we get an answer thanks to the work of John Scoville at San Jose State University and a couple of pals. These guys suggest that certain kinds of rocks behave like semiconductors when placed under huge pressures and temperatures. It is the way these rocks conduct current that then causes them to emit magnetic pulses in the run up to a quake.
Scoville and co begin by explaining the chemistry that leads to igneous rocks behaving like semiconductors. They point out that when magma crystallises in the presence of water, the resulting silicates contain peroxy bonds consisting of OH groups.
Under huge pressures and temperatures, these bonds can break to form electron-hole pairs. The electrons become trapped near broken peroxy bonds but the holes are free to travel through the crystal structure. The natural diffusion of these holes leads to a separation of charge, creating regions of the rock that are positively and negatively charged.
The boundary between these regions behaves like the p-n junction of a diode, say Scoville and co. This allows current to flow in one direction but not the other. At least not until the potential difference reaches a certain value when the boundary breaks down allowing a sudden increase in current.
It is this sudden increase that generates a magnetic field. The sheer scale of this process over a volume of hundreds of cubic metres ensures that these magnetic pulses have an extremely low frequency. And since low-frequency fields can travel through the Earth’s crust, they can be detected on the surface.
Having described this effect, Scoville and co go on to create a model of the processes involved and then calculate the shape of the pulses that ought to be produced. It turns out that the predicted pulse shapes bear a remarkable similarity to ones that geophysicists have observed.
In particular, they show that the magnetic pulses measured in the lead up to a series of earthquakes near Lima in Peru have great similarities to the pulse shapes that Scoville and co calculated. “This suggests that pre-earthquake ultra-low frequency activity may be the result of geophysical semiconductor processes,” they say.
That’s interesting work that provides a realistic explanation for a phenomenon that has puzzled geophysicists for many decades. Ultra-low-frequency magnetic pulses have been observed in the run-up to earthquakes since the 1960s. What’s more, scientists have measured low-frequency electric currents associated with earthquakes for several centuries.
The great promise of this new model is that it points to a straightforward way of identifying regions of the crust that are at imminent risk. The idea is to use a number of ground stations to listen for magnetic pulses and then triangulate the source. Indeed, exactly this technique has already been tested in Peru with some success.
Any significant increase in the production of magnetic pulses suggests an increase in pressure and the sudden breakdown currents that Scoville and co have modelled. This in turn points to the strong possibility of an imminent quake.
“By triangulating the source of these magnetic pulses, the increased buildup of stress around future earthquake epicenters may be identiﬁed weeks in advance of seismicity,” they say.
Of course, there is no shortage of ideas for predicting earthquakes. Geophysicists can certainly give accurate probabilities of earthquakes over a timescale of decades to hundreds of years. That is useful for long-term strategies such as determining building standards and so on.
They can also predict earthquakes on a timescale of seconds. That’s useful for shutting down high-speed railways in Japan, for example.
What’s needed is a way of predicting earthquakes on a timescale of hours, days and weeks. That’s never been possible with any reliability. The prospect raised by this new mechanism is that magnetic pulses could provide warnings over that kind of timescale.
That’s an exciting vision. However, there is significant work ahead before that kind of prediction can even be contemplated. For a start, geophysicists will want to characterise the magnetic environment that exists in the crust when it is not stressed as well as when it is. That should tell them exactly how unusual these kinds of pulses are and whether they uniquely predict earthquakes.
Another question that will need answering is whether vulnerable rocks emit these kinds of pulses when an earthquake is not imminent. One or two false positives would dramatically reduce the confidence that any population might have in such a predictive technique.
Scoville and co have hit on an interesting and important course of further research. Nevertheless it is too early to say whether useful predictions using this method will ever be possible.
Ref: arxiv.org/abs/1405.4482 : Pre-earthquake magnetic pulses
Suggest a link for further reading
The Physics arXiv Blog
An alternative view of the best new ideas in science. About: http://tinyurl.com/p6ypk56