Stanford Research Communication Program
  Home   Researchers Professionals  About
Archive by Major Area


Social Science

Natural Science

Archive by Year

Fall 1999 - Spring 2000

Fall 2000 - Summer 2001

Fall 2001 - Spring 2002

Fall 2002 - Summer 2003




Electromagnetic Precursors to Earthquakes: Do They Exist?

Darcy Karakelian
Geophysics Department
Stanford University
June 2001

Earthquakes are powerful phenomena that kill an average of 10,000 people a year and are consequently the topic of widespread research. Previous studies have suggested that major earthquakes may be preceded by anomalous electromagnetic field signals. My research focuses on exploring the relationship, if one exists, between electromagnetic field activity and earthquakes; a relationship that may provide insight into the processes that cause earthquakes as well as a possible warning for these earth-shaking phenomena.

If you live in a place like California, you are very familiar with the consequences of living near a major plate-boundary such as the San Andreas Fault. The plates are continuously subject to forces from within the earth that move them toward or along other plates. When the plates stick, and the forces at the plate-boundaries build up until the plates break loose and begin moving extremely fast, an earthquake results.

By studying the seismic waves that earthquakes emit, seismologists have learned much about their location, size, and type as well as the geology and structure of the faults surrounding the earthquake zones. Yet a lot of questions regarding the actual origin of earthquakes remain unanswered: When and where will an earthquake happen? What sorts of processes occur in the fault zone just before and immediately after an earthquake? When exactly does an earthquake begin and finish?

Within the past few decades, researchers have begun speculating as to whether electromagnetic (EM) waves, waves comprised of both a magnetic and an electric field, are produced during the earthquake "preparation process." If this is true, these EM waves may provide the means for gaining a better understanding of the earthquake process and ultimately predicting when and where earthquakes are likely to occur. I am interested in understanding if an EM wave is produced primarily during this preparation phase, but also whether it is produced during any other stage of the earthquake cycle (i.e. during or after the earthquake).

Throughout history, many scientific discoveries have been stumbled upon by pure accident. One such case occurred when Tony Fraser-Smith, Professor of Electrical Engineering at Stanford University, discovered anomalous magnetic field activity prior to the magnitude (M) 7.1 Loma Prieta earthquake that shook central California in October 1989, while he was monitoring magnetic field changes associated with the electric train system in the San Francisco Bay Area. His magnetic field sensor, which was fortuitously located only 7 km away from the epicenter, recorded a large increase in magnetic field activity two weeks before the earthquake, followed by an even larger increase just three hours prior to the main shock. This anomalously high magnetic field activity continued for over three months after the main shock. Although the Loma Prieta signals are by far the most spectacular, observations of strange magnetic and electric field signals prior to major (M > 6.0) earthquakes have been reported throughout the world for several decades. However, most of these reported earthquake precursors were recorded serendipitously by measurement systems established for other purposes, and therefore lack the associated measurements needed to exclude other potential sources of EM activity such as the sun, transmission/power lines and electric trains.

In order to test the credibility of these claims, some long-term observatories have been established along major faults, in particular the San Andreas, to monitor EM fields. However, to test whether an earthquake has associated EM signals, permanent observatories may require researchers to wait for decades for the occurrence of even one earthquake that is sufficiently large enough and close enough to the observatory. I have, therefore, helped construct a transportable recording system for rapid relocation to the epicentral region of a major earthquake immediately following the main shock. Although we will miss any potential EM activity produced prior to the main shock, placing recorders in the aftershock region will allow us the opportunity to record both continuing EM activity due to the main shock, as well as precursory EM signals associated with aftershocks, should any or all of these phenomena occur.

Our transportable system, which we bury directly in the ground, records three components (east-west, north-south, and vertical) of the magnetic field and two components (east-west and north-south) of the electric field. Because our EM sensors are also sensitive to motion, a seismometer collocated with our sensors allows us to rule out EM signals that are generated by motion of the sensors as opposed to tectonic sources. Most of the EM signals we measure are global, atmospheric variations generated by the sun. In order to isolate EM signals generated locally within the Earth, we can subtract out these solar signals from our data using signals measured at a remote EM station. Our systems are powered by 12 Volt batteries in the field, and portable data loggers store the data until we download it onto laptop computers.

After installing and testing a prototype of this transportable system near Stanford University, I was awoken early one morning in October 1999, by a phone call from one of my professors informing me that there had just been a large earthquake in the Mojave Desert in Southern California. This was our big chance! I immediately raced down to the desert with all of my equipment and some strong field hands to install two of our systems. This earthquake, named the Hector Mine earthquake, was as large as the Loma Prieta (M 7.1) and was located in a relatively quiet (no electric trains) area of the desert. However, as with most fieldwork, we weren't really prepared for the adventure we were about to encounter. The Hector Mine earthquake ruptured the Lavic Lake fault, which was entirely contained within the 29 Palms Marine Corps Combat Center, a patch of desert in which all U.S. government bombs are tested. So amidst dodging bombshells and trying to avoid Marine exercises, we were barely able to install two of our systems within two weeks after the main shock, though it did provide us with enough time to catch quite a few large aftershocks.

I am now in the process of searching for strange EM signals in the data that may be related to the main shock or one of the aftershocks that we captured. If we find anomalous signals, then not only do we strengthen the argument that EM signals are related to earthquakes, but we may learn something about how faults reorganize or "heal" after an earthquake. This healing process may lead to the production of EM fields under certain circumstances. If we do not find anomalous signals in our data, however, then we will have learned that not all earthquakes, and certainly not all aftershock earthquakes, can generate measurable EM signals.