SHAKEN AND SHOCKED

It sounds like a paranormal phenomenon: people reporting an unexplained glow, lasting up to several minutes, near the epicenter of a major earthquake that takes place at night.

Such reports have come for decades from far-flung, seismically active places--from California and Oregon, Turkey, Chile, Japan. The tantalizing anecdotes suggest that big quakes generate brief but intense electrical currents that can create what amounts to a "spark" along a ruptured fault. The phenomenon, associated mainly with quakes over magnitude 6.0, gained some short-lived scientific attention in the 1970s, as well as a name--"earthquake lightning." But most scientists have not taken it very seriously.

Eric Ferré does. If earthquake lightning exists, it may open possibilities for an earthquake early-warning network based not on seismic waves, which travel at about the speed of sound, but on electrical currents, which travel at the speed of light. Such a network, he says, could give near-real-time warning of a quake whose waves are still many minutes away from a big city.

Scientists in the United States, Europe, and Japan have scrutinized electrical data after the fact and have detected minute increases in electrical activity just before or during big quakes. They've suggested various alternative explanations for these findings, such as earthquake-triggered shifts in the water table or failure of power grids. So Ferré, an assistant professor of geology at SIUC, has gone to the rocks for answers.

The best evidence for earthquake lightning, he says, is locked up in dark veins that are often found cutting through rocks in quake-prone areas. The thin, sheet-like veins resemble a glassy black volcanic rock called tachylite--hence the name "pseudotachylites." Ferré calls them the "black boxes" of earthquakes because, he says, they record information crucial to understanding catastrophic seismic events.

The origin of pseudotachylites was a mystery until the early 1970s. "They thought that these had to do with magma from volcanoes," Ferré says, "but then they figured out that they're very often associated with faults and with breccia [crushed rock]. They're formed by frictional melting--the same type of process that is used industrially to weld two surfaces together without solder."

When two blocks of rock on opposite sides of a fault suddenly slip past the sticking point, the friction creates heat that melts the rock along the plane of the fault. The molten rock acts as a lubricant, allowing the two slabs of rock to slip as far as necessary to release built-up stress. Once the rock melts, incidentally, its ability to conduct electricity soars.

As the molten rock rapidly solidifies into a glassy material, iron bonds with oxygen to form magnetite crystals--the same material that records information on your credit card. Over millions of years, other minerals in the glass crystallize, making the veins darker.

Pseudotachylites can be formed in almost any kind of rock, by quakes of about magnitude 4.0 or higher. The presence or absence of these veins can reveal much about a region's earthquake history. But they may have more to say than scientists had imagined.

Pseudotachylites frequently are much more highly magnetized than the surrounding rock, meaning that they were exposed to a strong magnetic field when they formed. This fact suggested to Ferré that reports of earthquake lightning aren't so farfetched: electrical currents generate magnetic fields. He believes that, in their molten phase, pseudotachylites act as conduits--essentially, lightning rods--for electrical currents generated by earthquakes. They are, he says, "very conducting for a very short period of time." And the magnetite preserves the "memory" of this event.

In 2003 the National Science Foundation awarded Ferré a $110,000 grant to test his hypothesis, with an additional $80,000 going to his chief collaborator, John Geissman, a geologist with the University of New Mexico (UNM).

The first order of business was collecting pseudotachylite samples. Ferré and his students, Geissman, and scientists from Japan and France collected samples in the Italian Alps, near Palm Springs, Calif. ("just back of the PGA golf course," Ferré says), and from the Japanese island of Kyushu. They're three of the world's most earthquake-prone regions, and all are heavily populated areas.

Pseudotachylite veins are formed at a fault's point of rupture, often several kilometers below ground. They can extend all the way up to the surface--provided that solid rock extends to the surface, rather than the fault being covered by layers of sediment. To be sure of getting sufficient pseudotachylite samples, the team chose to work with old faults, where uplift and weathering over millions of years has raised an abundance of pseudotachylites up from the depths. In California, for example, they worked not on the San Andreas fault but on a parallel "fossil" fault active 60 million years ago.

Geologically, that's still recent. The continental plates were bumping and grinding in pretty much the same places then as now, so Ferré's results will have implications for these same regions today.

At SIUC, geology master's student Matthew Zechmeister and several undergraduate students prepared samples in various shapes and sizes for advanced measurements of their magnetic properties. Some of the measurements were done on campus, some at UNM, and some at the University of Minnesota's Institute for Rock Magnetism. Ferré recently won additional funding from the NSF for equipment to double the capabilities of SIUC's rock magnetics lab, so more analyses will be done here in the future and students will gain better training.

"We spend a lot of time in the lab trying to make these rocks talk to us," says Ferré. "Anything that happens in an earthquake is potentially recorded in or near the pseudotachylite. The problem is, how do you read this information?

"The information we retrieve is not only due to the seismic slip event. There are things that happen after the earthquake. The rocks have recrystallized after they were molten. They may have been exposed to groundwater, or to weathering at the surface. All of these changes potentially affect the magnetic properties."

The arduous part is separating out the clues. But what's clear, says Ferré, is that the pseudotachylite samples "were exposed to magnetic fields 40 times higher than the earth's magnetic field." Their magnetic properties are similar to those of rocks struck by lightning bolts, he says.

Could pseudotachylites have been formed by atmospheric lightning? No, says Ferré, rocks hit by such bolts don't magnetize in a vein formation. Furthermore, some of the team's samples were collected from quarries, where the rocks had never been exposed to the surface.

Is there any explanation other than a large electric current? No plausible one that Ferré knows of. "There's still room for other hypotheses," he says, "but the fact that [pseudotachylites occur] along the fault plane limits possible explanations."

More evidence is accumulating. Matthew Zechmeister, who graduated in spring 2005 and is now doing doctoral research on mining-induced quakes in South African gold mines, relayed an interesting report to Ferré in June. During a recent magnitude-3.0 quake in one of these very deep mines, an engineer close to the quake's focus described seeing a glow along the fault plane.

Other hints come from far above the earth's surface. In August 2004 the European Space Agency launched Demeter, a satellite designed to see if variations in the atmosphere's electromagnetic field were linked to seismic activity. "The first results, which were reported in December 2004, indicate that the big anomalies observed in the electromagnetic field coincide perfectly with the boundaries of the continental plates," says Ferré.

Pseudotachylites can reveal a lot about the mechanics of earthquakes. By analyzing the variation in magnetic flux along pseudotachylite veins, it's possible to pinpoint the earthquake focus--the point of rupture. "We can get a better understanding of how big these ruptures are," Ferré says.

Scientists, he says, have historically conceptualized quakes as single ruptures. But sensitive instrumentation has found that in California, for example, often two or three major ruptures happen simultaneously. "With magnetics, we might be able to help understand these processes in fossil faults," Ferré says--knowledge to improve earthquake modeling and prediction.

Much more needs to be done to understand earthquake lightning itself. "Timing is the most unknown issue," says Ferré. "For example, it's important to find out how long the current might last, which may tell us about the source."

There are two competing explanations for earthquake lightning: triboelectricity (basically static electricity), in which friction between two surfaces shears off electrons, and piezoelectricity, in which electrons escape their bonds as quartz crystals are skewed under pressure. Ferré and Navani Mathanasekaran, an SIUC master's student in electrical engineering, have done computer simulations of the electrical properties of fault rocks to explore these issues. But Ferré thinks that rock deformation experiments will be needed to get solid answers.

Japanese colleagues Aiming Lin and Toshi Shimamoto have done some work in that direction by making pseudotachylites in the lab. When two rock samples are rotated together at high speed under carefully controlled conditions, the frictional melting at the interface creates a pseudotachylite that bonds the samples when it cools. As part of the NSF project, the SIUC team analyzed some of these experimental pseudotachylites to help them understand the properties of naturally occurring ones.

Ferré is an advocate for the development of an earthquake early-warning system based on electrical currents. Currently, experimental warning systems are based on early-arriving seismic waves called P waves, which precede the more-damaging S waves of quakes. Both types of waves spread out from the fault rupture like ripples on a pond. Both are sluggish compared to electromagnetic waves.

Producing measurable electrical currents may take a quake of at least magnitude 6.0, Ferré thinks--right around the level where earthquakes become seriously damaging. So a warning system based on these currents could give a city valuable extra minutes to prepare for a major quake. Seismic-wave experiments in Mexico and Chile have given warnings of offshore quakes that have ranged from a few minutes up to half an hour; Ferré says electric-current monitoring could double or perhaps even quadruple that.

Because seismometers are fragile and can't be placed directly on faults, a grid of seismometers is required to monitor an entire region. But electrical receivers, which cost far less than seismometers and wouldn't be disabled by quake activity, need be placed only along major active faults.

"The electrical signal has to travel along the fault plane because it has the best electrical conductivity," Ferré says. "Any current, even if it's faint, can be intercepted with one station almost instantaneously."

So what's the hitch? "We haven't invested in electrical warning systems mainly because nobody has believed in it," Ferré says.

True, there are obstacles to be overcome. Big quakes disturb the power grid, which can create ground leaks of electrical current. To have a reliable detection system, says Ferré, "you have to know how to filter out those industrial signals."

A Greek team of engineers has developed an earthquake warning system that is attempting to do just that, he adds, "and they've been laughed at by most of the rest of the world. But we think that now the tide is changing."

Ferré suggests employing a dual-sensor setup, with one of each pair of sensors on the fault and the other some distance away. By measuring the differential between the two, you could separate seismic electrical currents from industrial ground leaks.

That sort of work will involve other specialists. Ferré sees his role as decoding the messages of pseudotachylites so that engineers can apply that knowledge.

"There's a lot of information encapsulated in these rocks," he says, "and it's important to translate it in terms useful for others."

--by Marilyn Davis, ed.

Comments: Perspectives Webmaster
Copyright © 2005, Board of Trustees, Southern Illinois University | Privacy Policy