Mysteries Before Earthquakes

Page 1

201

Mysteries Before Earthquakes

The Behavior of Electric Appliances

Second hands of clocks were reported to have rotated rapidly before the Kobe,

Izmit and Taiwan-921 earthquakes and also to have moved backwards. Obviously

relativity theory is not a factor! Some have suggested that the Alice in Wonderland

Syndrome—a vision disturbance disorder—might be responsible, but EM waves

may account for the phenomenon. Above: An illustration of the Mad Tea Party by

John Tenniel in Alice’s Adventure in Wonderland, by Lewis Carroll.

Page 2

9. Mysteries Before Earthquakes: The Behavior of Electric Appliances

202

9.1 Introduction

The electronic age has added a new category of earthquake precursor reports to

the ancient legends.

They are accounts of odd behavior and noises from domestic electric appliances:

TVs, radios, clocks, refrigerators, mobile phones, fluorescent lamps, car navigators,

and possibly computers. The reports came in independently from Kobe, Izmit, and

Taiwan before the large earthquakes of 1995, and 1999 and are described in more

detail in Chapter 2.

Interestingly, exposure of some of these appliances to EM pulses in a laboratory

reproduced the behaviors described in the reports: clocks stopped, or their hands

rotated rapidly; fluorescent lamps lit up, radios went dead, or produced static;

color shifts and speckle noise appeared on TV screens, or sets fluctuated between

channels; refrigerator compressors switched on and off producing odd sounds;

cellular phones illuminated and rang but no record of any call was left; the needle

on a magnetic compass fluctuated.

Reports of malfunctioning devices that we did not attempt to replicate in the

laboratory setting could just as easily be ascribed to EM wave interference: random

self-operation of power windows in cars, spontaneous switching on of air condition-

ers and intercoms at midnight (five or six hours before the Kobe quake); the apparent

sudden switching on of a tape-recorder in Izmit so that a unscheduled call to prayer

resonated from a local mosque at 2 a.m.—one hour before the earthquake.

Several reports from slightly earlier than the electronic era are also investi-

gated in this chapter: How was it that nails hanging for weeks from a permanent

magnet dropped off two hours before the Ansei-Edo (Tokyo) Earthquake (M6.9)

in 1855? Why did iron chains in a military factory begin to swing against each

other two hours before the Eastern Nankai (M8) Earthquake in 1944? Why did arc

discharges occur between iron bars lying on the ground a day or two before the

Tangshan Earthquake in 1976? Why could a man no longer read the phosphorescent

numbers on his watch 2-3 days before the Great Kanto Earthquake, but see them

again one day later?

9.2 Magnetic earthquake precursors?

9.2.1 Nails falling from a magnet, not a magnetic anomaly!

In 1855 the Ansei Chronicle carried a report of 15 cm iron nails that dropped from

a big natural magnet two hours before the Ansei-Edo Earthquake. They had been

hanging end-to-end from the magnet in a spectacle store for some time as part of

an advertisement. The magnet regained its iron-attachment properties after the

earthquake.

Page 3

203

9.2 Magnetic Earthquake Precursors?

Attempts have been made to link the anomaly to a change in the earth’s magnetic

field—but as discussed earlier, the variation in the earth’s magnetic field before

earthquakes is minuscule; the earth’s magnetic field is 0.03 - 0.05 mT, and the

change, a few nano-Tesla (nT) or about one ten thousandth of the field (Rikitake,

1986). The surface intensity of a ferrite magnet is tens of mT and that of the most

intense permanent magnet made of Neomax (Nd-B-Fe alloy) is about 800 mT.

The magnetic field of the natural magnet mentioned in the Ansei Chronicle may

have been several mT at its surface. The only way to explain the phenomenon as a

magnetic one was to argue that a large magnetic field variation had occurred before

the earthquake. This was not credible, so the story came to be regarded by most

geophysicists as a misleading anecdote, though there were two similar European

reports (Rikitake, 2000).

Attempts to link the interrupted magnetic contact to the arrival of seismic P-

waves does not stand up either, as they dropped off two hours before the quake.

Had they fallen 20 seconds beforehand, the effect could plausibly have been argued

to be a P-wave effect, since the epicenter was about 160 km away from the store.

However, the story may be explicable as an electric effect.

Electric discharge experiment:

Several nails were attached to a magnet hung from a pole made of plastic LEGO

blocks ( See Figure 9.1). The floor under the LEGO construction was covered with

aluminum foil, which was connected to the high voltage sphere of a Van de Graaff

generator. When the high-voltage Van de Graaff sphere was electrically charged,

the nails repelled each other. When the voltage was turned on and off (to simulate

EM pulses of ULF waves) the iron nails began to swing increasingly, and finally

dropped off.

Explanation:

Electric induction by EM pulses:

The iron magnet and nails acted as an electrode when the aluminum foil was charged

up. Electrostatic induction generated a charge on the nails and an attractive force

was formed between the ground and the nails. The nails, charged at their tips with the

same electrical charge, repelled each other. When the charge was removed from the

floor, the nails returned to their original position. When it was applied and removed

several times, the nails began to swing, became unstable and dropped off.

So the nails that fell from the magnet at the spectacle store two hours before

the Ansei Earthquake may have been responding to electric charge appearing on

the ground, and not to changes in the Earth’s magnetic field or to land tilting. This

transforms the tale from a magnetic anomaly to an electrical one.

Page 4

9. Mysteries Before Earthquakes: The Behavior of Electric Appliances

204

Attempts have been made to attribute the story in the Ansei Chronicle to elec-

trostatic charges produced on the nails by charged aerosol, as was done for a report

from a watch repairman in Italy, who claimed a tiny watch mechanism jumped

into the air just before an earthquake (Tributsch, 1982). Because the hypothesis

of charged aerosol could also explain these results, such an effect was eliminated

by placing the magnet and nails in a plastic box to shield them from any ion wind.

The iron nails still swung and dropped during nearby electric discharges from the

Van de Graaff generator, meaning that electric induction was sufficient explanation

for the behavior of the nails.

9.2.2 Disturbed orientation of a magnetic compass

Statements:

Small compass magnets have been said to have fluctuated violently before the

Eastern Nankai and other earthquakes. The anomaly has been rather dubiously

attributed to cosmic radiation, explained thus:

The earth’s magnetic field changed before the earthquake so that cosmic

radiation penetrated to the non-polar area causing earthquake light similar

Figure 9.1 The induction of opposite charge at the nail tips made them drop off a

magnet—as they did two hours before the Tokyo Earthquake (M6.9) in 1855. The op-

posite charge was induced by the appearance of negative charge on ground aluminum foil

connected to the high voltage sphere of a Van de Graaff generator. (a) a diagram of

the experiment (b) a photograph of the same experiment [See also front color plates and

Figure 1.6 (a)].

Magnet

Nails

Van de Graaff

aluminum foil

wire

Nails

aluminum foil

Magnet

(a)

(b)

Page 5

205

9.2 Magnetic Earthquake Precursors?

to the auroral light caused by ionization by charged particles from solar

flares…

It could be added that odd hypotheses like these appear because no serious attempt is

being made to bring science to bear on unusual events before earthquakes. Although

the author might also be accused of producing odd hypotheses, his efforts are at

least a serious attempt at an objective scientific analysis of precursor reports.

Experiment:

A compass was placed near an antenna and a pulsed field applied. Naturally enough

the compass needle moved with the interaction between the magnetic needle and

the electric field. When systematically varying low frequency EM sine waves were

applied, the needle spun round.

Explanations:

As discussed, changes in the earth’s magnetic field before earthquakes are negligible:

not enough to draw increased cosmic radiation into the earth’s atmosphere, and

certainly not in amounts intense enough to cause unusual animal behavior before

earthquakes. If that were the case animals would have skin burns, as in overexposure

to X-rays or intense UV light.

A change of a few nano-Tesla (nT) in the Earth’s magnetic field before big

earthquakes is about one ten-thousandth of the Earth’s magnetic field. Given the

frictional forces impeding rotation of the compass needle, such a small change would

not be sufficient to move the needle. The fluctuation is much more explicable as

an effect of charges induced by an electric field.

9.2.3 Car navigator: Fluctuating direction arrow

Statements:

Drivers claimed that their car navigators did not function—or malfunctioned—a

day before the Kobe Earthquake. In one case the navigational arrow swung 180

degrees from the true in a car in the city of Kobe. Generally navigators did not

function normally before and after the earthquake, but returned to normal a month

or so later.

Experiment and Explanation:

Modern car navigators are set up to receive satellite signals coded as 0 and 1 to

cut out background noise. Electric discharges were made using a Van de Graaff

generator near such a modern navigator. Line noises caused by EM pulses were

observed on the screen, but there was no fluctuation in the arrow.

When the author spoke to a woman who provided such a report she said that the

navigator in the vehicle did not use the GPS system but an older system based on a

Page 6

9. Mysteries Before Earthquakes: The Behavior of Electric Appliances

206

magnetic compass. Hence, EM waves at low frequency may have been responsible

for the change in the arrow’s direction on the screen of the navigator, due to an

induced electric field.

9.2.4 Mischief by an invisible man?

Swinging iron chains:

Two hours before the Eastern Nankai Earthquake in 1944, iron chains hanging

from the ceiling in a military factory began to swing, knocking a nearby electric

furnace and making clanging sounds. [This was not coincidental with the vigorous

movement of the bubble in the surveying level (Chapter 8).] The phenomenon

was reproduced in the laboratory by inducing charge on small iron chains hang-

ing near the high voltage sphere of a Van der Graaff generator. The chains swung

towards the sphere.

So it is possible that an intense electric field, presumably electric pulses, may

have appeared before the earthquake inducing charges on metal objects so they

were attracted to each other, causing swinging in the iron chains.

Arc discharges between iron bars:

A day or two before the 1976 Tangshan Earthquake (M8.2), China, arc discharges

were observed between iron bars lying on the ground—as if an invisible man was

using an arc-welding machine (Dai, 1996). (A small arc discharge caused by static

electricity in a human body may be seen sometimes when one unlocks the car or

touches a doorknob.)

An intense electric field is formed between iron bars (electric conductors) if

charges appear on the ground. Opposite charges may appear at the ends of the bars

producing sparks by atmospheric breakdown at 3 MV/m.

A charge appearing on the ground intense enough to swing iron nails and iron

chains and move magnetic compass needles would also explain arc discharges

between the iron bars. The charge density of less than one millionth of a Coulomb

per square meter on the ground would be sufficient to cause discharges at the sharp

edge of iron bars. (This is less one thousandth of the charge necessary to run a

small piece of consumer electronic equipment for one second.)

9.3 Unusual behavior of electric home appliances

9.3.1 Faint glow of fluorescent lamp

Statements:

Fluorescent lamps glowed faintly before the Kobe Earthquake (Wadatsumi, 1995).

They also lit up spontaneously before the Tangshan Earthquake (Dai, 1996). The

Page 7

9.3 Unusual Behavior of Electric Home Appliances

207

author has also seen a faint glow from a fluorescent lamp in a room during a

thunderstorm.

Experiment:

A fluorescent lamp held close to the charged sphere of a Van de Graaff generator

in the dark, lit up because of the intense electric field. A plastic sheet charged by

frictional electricity also produced a glow from a fluorescent lamp (surprising pupils

in a lecture on earthquake precursor phenomena). Air-gap discharges between the

high voltage sphere and an electrically grounded rod generated pulsed EM waves

that also lit up the lamp in a darkened

room (Figure 9.2). A fluorescent lamp

exposed to EM waves at 1 MHz from

a Tesla coil lit up as it also does in a

microwave oven. (A Tesla Coil is a high

voltage transformer that generates very

high voltages at high frequency.)

Explanation:

In a suitable environment, EM waves

from thunderstorms (produced by at-

mospheric lightning) light a fluorescent

lamp. The faint light of a fluorescent

lamp before earthquakes could similarly

be due to an electric field of pulsed EM

waves.

9.3.2 Radio interference

Incidents:

Truck drivers reported radio interference on a highway near the epicenter in Kobe

about 5 a.m., 45 minutes before the Kobe Earthquake. Some were unable to get any

reception at all. Before the Tangshan Earthquake factory workers manufacturing

radios were unable to tune in any broadcast signals. The phenomenon seemed lim-

ited to that area, and reception returned to normal after the quake (Dai, 1996).

Experiments:

A transistor radio placed on the high voltage sphere of a Van de Graaff generator

went dead at a high voltage but worked normally when the sphere was grounded.

Air-gap discharges between the sphere and a grounded rod caused interference on

another radio nearby (Ikeya and Matsumoto, 1998). FM broadcasts were not affected

during these experiments. (A radio receiver is a good sensor of electromagnetic

Figure 9.2 A fluorescent lamp in a labora-

tory lit up when it was exposed to EM pulses,

as fluorescent lamps have been reported to

do during thunderstorm lightning and before

earthquakes.

Page 8

9.3 Unusual Behavior of Electric Home Appliances

208

waves; a small radio placed near a computer will pick up EM interference from

the computer.)

Explanation:

The air-gap discharges produced EM interference in the radio wave frequency

range. Radio receivers tuned into AM bands are easily affected by EM noise; their

longer, medium waves are more subject to interference than VHF FM waves. (Those

seeking to use radio broadcasts for earthquake forecasting should obviously tune

in to AM (rather than FM) broadcasts.)

Reported interference in FM broadcasts has been attributed to scattering of

broadcast VHF waves according to a reasonable inference that the ionosphere

descends locally from its usual height in response to the appearance of charges

on the ground i.e. to an electric field of EM waves in the ULF range (Kushida and

Kushida, 1998).

The reflection of FM waves might, rather, be caused by a change in the at-

mospheric conductivity around the epicenter due to charged aerosol. Whether the

ionosphere is actually affected or not has yet to be established. For more on the

apparent lithosphere-atmosphere-ionosphere (LAI) coupling, see Chapter 11.

9.3.3 “Barber-pole” color and speckle interference on TV screens

(a) Reproduction of barber-pole noise

Video films recording precursor noise: A video recording made 40 km away from

the Kobe Earthquake epicenter of an NHK program broadcast at 11.18 p.m. (about

six and half hours before the Kobe Earthquake), showed ghosting, “barber-pole”

color and white speckle interference lasting for about 30 seconds. The broadcast

and received images are shown respectively in Figure 9.3 (a) and (b). At this time

hamsters were reported to be agitated and biting each other and mice were much

more active than usual (Matsumoto et al., 1998) and Figure 5.5 (a).

A person living in Kyoto, about 80 km away from the Kobe epicenter, supplied

another video film that recorded the same section of the program. In this case no

“barber-pole” noise was recorded but flashing lines appeared across the screen at

about 11:18 p.m. No seismic waves were detected at the time, so low frequency

EM pulses (not detectable by seismographs) generated by local fractures in stressed

rock close to the observation site might have been the cause.

However, the times did not exactly correspond, so a time difference needs to

be accounted for. A speculative explanation might be that radiowave interference

(generated by pre-seismic EM waves) was emitted either heterogeneously (in a

focussed beam)—maybe because of some inexplicable polarization from a ferro-

electric orientation of electric dipoles (Ikeya et al., 1998)—or the EM waves moved

Page 9

9.3 Unusual Behavior of Electric Home Appliances

209

through the crust at an extraordinarily slow speed (Section 11.3.8, Table 11.2).

In any event, only the low frequency ULF component of EM pulses can be

propagated from underground to the surface without attenuation (due to the large

skin depth of these waves) and induce charges on the ground around fault planes.

Appearance of this intense charge on the ground would cause electric discharges

everywhere, generating EM noise in the high frequency (TV broadcast) range.

They might also create disturbance in the ionosphere, creating high frequency EM

waves by some unknown non-linear mechanism.

Analysis of videotapes:

Waveforms and signal spectra in both video recordings of the film were analyzed

using a digital storage oscilloscope (DSO). An image on a TV screen is composed

of 525 horizontal lines that are scanned from left to right in 0.063 ms by the NTSC

system used in the USA and Japan. The luminosity signal (EY) and sub carrier color

difference signals (EI and EQ) at a frequency of 3.58 MHz are related to components

of the red, blue and green signals (ER, EB and EG).

Figure 9.3 Screen distortion during a TV program broadcast by NHK at 11.18

p.m., January 16, 1995, about six and half hours before the Kobe Earthquake. (a)

The image without distortion. (b) An image distorted by speckle noise and “barber-

pole” interference. (c) Luminosity image EY isolated from (b) and showing speckle

interference. (d) Image of color difference signal EI, isolated from (b) and affected

by “barber-pole” noise (Matsumoto et al, 1998). For a color view of “barber pole”

noise, see the front color plates. (Videotape, Mr I. Kawakata.)

Page 10

9.3 Unusual Behavior of Electric Home Appliances

210

Figure 9.3 (c) and (d) respectively show the luminosity (brightness) and color

difference signals isolated from the distorted image (b) recorded before the earth-

quake. Only speckle noises appear in (c); there is no effect on luminosity. Barber-

pole noise appeared in the image showing color difference signals (d).

An attempt was made to reproduce barber-pole noise in a regular TV pro-

gram broadcast by exposing the TV antenna to EM sine waves from a microwave

synthesizer. A color shift appeared at 215 ą 5 MHz (within the TV broadcast

frequency range).

It seems reasonable to conclude that EM pulses composed of EM waves at a

wide range of frequencies before quakes can create interference on TV screens.

(b) Reproduction of speckle noise and channel fluctuation

There was a report of a faulty channel selection function in a TV remote for

several days at the epicenter area before the Kobe Earthquake and we wondered

if replication of channel-selection malfunction might be possible in a laboratory

using EM waves. Speckle noise also seemed a likely candidate for an EM wave

experiment given that the length of speckle interference in the video (based on

the horizontal-scanning rate of 63 microseconds) was estimated to be of the order

of microseconds, the same as pulsewidths observed during thunder, ten minutes

before a local earthquake, and also during the granite compression experiment

described in Chapter 5.

Experiments:

A TV antenna was exposed to EM waves produced by a microwave synthesizer

and by air-gap discharge from a Van de Graaff generator. The discharges created

speckle noise. A portable TV with a liquid crystal screen and a standard TV set

were placed, respectively, on the sphere and also one meter from the sphere, during

the air-gap discharge. Channel setting anomalies and noise were observed on the

portable TV (Figure 9.4) on the sphere and in the standard set at a meter distant,

during an arc discharge from the Van de Graaff generator.

Explanation:

Digital noise created by the EM waves scrambled the digital channel selection code

in the TV and interfered with the broadcast frequency.

It is interesting to note that a small earthquake of M2.1 occurred at 11:49 p.m. on

January 16 at the epicenter, apparently one of the four Kobe foreshocks. However,

an episode of fluctuating channels at the epicenter area (above) occurred 30 minutes

beforehand, meaning it can’t be linked to the foreshock itself. In addition no such

interference had ever been reported in connection with frequent small earthquakes

Page 11

9.3 Unusual Behavior of Electric Home Appliances

211

in the area. So the incident is puzzling unless it was in response to presismic EM

waves generated locally by increasing fault stress.

There was one report of distorted TV images a few seconds before the Kobe

Earthquake affecting not just one but at least two channels, followed by earth

sounds and then the tremor. This apparently indicates coseismic generation of

EM waves producing interference, and then earth sounds, just before the arrival

of S-waves (the quake).

Japan has just implemented a hi-tech digital broadcasting TV system that gives

a high-resolution color picture, but the new technology will still remain susceptible

to color changes and other interference from EM noise generated by earthquakes

or lightning. Even cable TV will be susceptible.

9.3.4 Strange sounds from refrigerators, spontaneous on-and-off switching

of appliances

Statements:

Reports of mysterious buzzes and alarm sounds over intercoms, and strange sounds

from refrigerators and air conditioners are among precursor reports collected after

Figure 9.4 The effects of EM interference in an image on two color TV sets. (a) Normal

image (b) Loss of image on the high voltage sphere of the Van de Graaff generator. (c)

Horizontal line noise and “snow-like” speckle noise produced by arc discharges (d)

Noise and an unstable channel setting (Channel 2).

Page 12

9. Mysteries Before Earthquakes: The Behavior of Electric Appliances

212

the Kobe Earthquake (Wadatsumi, 1995). Kobe citizens were also surprised and

puzzled by spontaneous switching of radios, TVs and air conditioners around 2 - 3

a.m. before the earthquake at 5:45 a.m.

Experiments:

Some of the electric appliances reported to have malfunctioned were exposed to EM

pulses generated by arc discharges using a Van de Graaff generator or by 100

MHz EM waves from a Tesla coil.

A TV connected to a video recorder, which could be operated by remote control,

switched on upon exposure to EM pulses.

A refrigerator was charged using the Van de Graaff generator. Electric discharges

occurred in the circuitry affecting internal devices and the temperature sensors,

creating malfunctions that switched the compressor off and then on again, creating

sounds and vibrations in the refrigerator.

Explanation:

Modern electrical equipment employs a simple switching system using digital

signals. Mains power is still connected to this quick switching apparatus even

when the switch is off, meaning preseismic EM pulses could cause spontaneous

switching of electric appliances before the Kobe Earthquake. (Malfunctioning

caused by EM waves will not occur if the appliance is not connected to the mains

socket. Only a fluorescent lamp operates differently.)

Buzzing sounds from intercoms and cellular phones may also be caused by EM

interference. In fact, spontaneous switching and malfunctioning of nearby electric

appliances in our laboratory surprised us sometimes during our experiments using a

Van de Graaff generator and a Tesla coil e.g. a laboratory oven at room temperature

began to warn us that it was overheating and a TV 2 m away switched on.

Intermittent short circuits within the refrigerators would have turned compres-

sors off and on causing the sounds and vibrations reported before earthquakes and

reducing cooling efficiency. This could explain reports of spoiled yogurt; certainly

the direct effect of EM pulses on lactate cells is doubtful since the refrigerator

would electrically shield the yogurt.

The reported malfunctioning of radios, TVs, air-conditioners and intercoms at

midnight before the earthquake is consistent with preseismic discharges and effects

of intense pulsed electric fields such as earthquake lightning. The unscheduled

broadcast of the Koranic prayer from a local mosque one hour before the Izmit

Earthquake in Turkey may have been due to spontaneous switching on of an elec-

tronic device. It should be noted that, at about the same time, budgerigars were

very agitated, and frightened children woke people up according to reports (Sec-

tion 2.3.3). These are consistent with the appearance of intense EM pulses in the

epicenter area before the Izmit Earthquake, just as before the Kobe Earthquake.

Page 13

9.3.5 Activated Mobile Phones with no Call Records

213

9.3.5 Activated mobile phones with no call records

The big surge in popularity of mobile phones occurred only three to four years ago

so there are not many reports of malfunctioning mobile phones before the Kobe

Earthquake in 1995. However, there

were reports from both Kobe and Izmit

of mobile phones whose call indicators

had activated but had received no calls.

In other words, the display lit up and

there may or may not have been a ring-

ing tone, but there was no record of any

caller (Figure 9.5).

Experiment: EM noise created by elec-

tric discharge

Ten mobile phones manufactured by dif-

ferent companies were placed on the Van

de Graaff generator and the grounded

sphere was placed close to the high volt-

age sphere. Electric discharges generated EM pulses. In two of the ten phones the

display lit up and the phones rang. (The others may have been better shielded.)

Again, lay reports of mobile phones receiving calls but leaving no record of a caller

may be explained as an effect of preseismic EM pulses.

9.3.6 Assorted unusual reports

(a) Burglar-proof alarm

A car security alarm, which used an electrostatic field, sounded at 9.30 p.m., Janu-

ary 16, about 8 hours before the Kobe Earthquake—about 1hr 45 mins before the

TV picture interference captured on video. The incident may be another case of

spontaneous switching on of an electronic device.

(b) Cat disappears from locked car

A cat mysteriously escaped from a locked car before a large aftershock in Turkey

(Ulusoy and Ikeya, 2002) while its owners were asleep inside the car. Woken by

the tremor they found the cat outside the locked car. It is possible that the automatic

power window, activated by intermittent EM pulses, opened briefly, letting the cat

out, and closed again.

There was a statement that personal computers were more troublesome than usual

just before the Taiwan-921 earthquake. If pulsed noise affects only one bit in the

Figure 9.5 Mobile phones exposed to

electric discharges of EM waves lit up and

rang but left no record of any caller.

Page 14

9. Mysteries Before Earthquakes: The Behavior of Electric Appliances

214

digital code, a calculation result can be completely skewed, but usually there is

sufficient redundancy built into system software to absorb these fluctuations. The

author didn’t notice anything unusual about the performance of his computer before

the Kobe earthquake, but then again it crashed so frequently—with a high voltage

line 30 m away and a bus station closer still—it would have been hard to know!

(Data on the frequency of crashes would have to be kept if computers are to be

used in earthquake forecasting.)

(d) Odd sounds from a power transformer

Two electric power transformers made strange “hoh-hoh-hoh” sounds from 13 to 6

hours before an earthquake (M5) in Canton, China in 1970 (Rikitake, 1979). Such

sounds are produced from a transformer by coil vibration when there are minor

voltage and amp instabilities in the power line, the instability having been induced

in the power lines from outside.

Dr T. Higuchi (General Research Institute of the Kansai Electric Power Com-

pany) has detected noise before earthquakes by measuring current in the neutral line

of three-phase power lines. Power lines are effectively huge antennae, or circuits,

capable of measuring earth potential differences, and are therefore possible detec-

tors of earth current and/or EM waves before earthquakes.

9.4 Mysterious clocks and the “Alice in Wonderland” Syndrome

9.4.1 Did time stop or speed up?

There was an anecdotal report that clocks in a local stock exchange did not show

the correct time before the Kobe earthquake. Quartz clocks in Beijing, at a distance

of 160 km from the epicenter were reported to have stopped eight hours before the

Tangshan earthquake (M8.2) in 1976. The same clocks also stopped before large

aftershocks (Dai, 1996).

A newspaper reported that a medical doctor in Osaka had a radio-clock which

adjusted its time every hour in response to a 40 kHz radiowave signal from a Gov-

ernment telecommunications laboratory, close to Tokyo. His clock lost two seconds

before the Kobe earthquake, but ran to time again after the earthquake.

A citizen of Ashiya City, 25 km away from the Kobe epicenter, telephoned the

author and said that the hands of her quartz wall-clock stopped moving and then

moved slightly backward for a few seconds a day before the Kobe earthquake.

They resumed normal speed and function after the earthquake. She wondered

if she had been “seeing things” because nobody believed her, and she sought a

scientific explanation. A girl student who also claims she saw the hands of a clock

move rapidly before the Kobe earthquake requested confidentiality in case people

thought she was a little strange.

Page 15

215

9.4. Mysterious Clocks and the “Alice in Wonderland” Syndrome

The “Alice in Wonderland” Syndrome?

According to the theory of relativity, time may pass faster or slower. One cosmologi-

cal theory says time may even reverse if the universe starts to contract. Obviously

erratic clock malfunction does not fall into the category of a real time anomaly, but

to someone who hears reports of hands spinning forwards, backwards and jerking

to and fro, it can sound like a case of ‘Alice in Wonderland’ syndrome (Burstein,

1998; Mizuno et al., 1998), in which symptoms of nausea and vision disturbance

accompany a recurrent throbbing headache on one side of the head. Those with

the syndrome have a disordered perception of time and space. The disorder can

also be induced by drugs and gained its name from the book Alice in Wonderland

by Lewis Carroll (1953).

In Alice in Wonderland (Chapter 7), the Hatter, who has become mad through

exposure to toxic mercury used in tanning leathers to make hats, says to Alice at

the Mad Teaparty,

Now if you only kept on good terms with him (Time), he’d do almost any-

thing you liked with the clock. For instance, suppose it were nine o’clock

in the morning, just time to begin lessons: you’d only have to whisper a

hint to him (Time), and round goes the clock in a twinkling! Half-past one,

time for dinner!

There have been suggestions that people who claim to have seen clock hands

jerk backwards and forwards or move at increased speed before earthquakes had

some sort of psychological condition at the time. Did they?

Experiments:

A quartz clock was placed on the high voltage-sphere of a Van de Graaff generator

(See Figure 9.6). No change in the

movement of the hands was observed

as the sphere was charged. However,

the hands stopped moving when an

air-gap discharge was made to a

grounded rod close to the high volt-

age sphere. The alarm rang and then

the second hand rotated at a speed

eight times faster than normal when

arc discharge occurred between the

plastic case and the high voltage

sphere.

Figure 9.6 The second hand of a quartz clock

rotates at eight times normal speed on the high

voltage metal sphere of the Van de Graaff gen-

erator.

Page 16

9. Mysteries Before Earthquakes: The Behavior of Electric Appliances

216

An exposure to EM waves from an antenna at an emitting power of 10 W in

a frequency range from 120 to 130 MHz stopped the clock. At the same time the

automatic focusing motor of a video camera taking a picture of the clock at a dis-

tance of 2 m malfunctioned, blurring the image. The intensity of the EM waves was

about 2 W/m2 (30 V/m) assuming uniform radiation, a little higher than that causing

unusual animal behavior in budgerigar and mice, as described in Chapter 4.

Explanation: Short circuits in a digital integrated circuit

The hands of a quartz clock are driven by a stepping motor, itself driven by electric

pulses fed from a digital integrated circuit (IC) composed of flip-flop circuits in a

state of either “0” or “1”. The clock’s plastic case was charged positively by col-

lecting charges from the air and the clock’s insulation was broken down by air-gap

discharges. This discharge state generated electronic noise, which disturbed the

digital circuit (essentially creating a short circuit by interfering with the digital code,

and producing errors, thus affecting the generation of pulses, motor and hands).

Sometimes, discharges of accumulated charge in a printed circuit will also

result in short-circuiting. Such short-circuiting would cause spontaneous ringing

of a buzzer.

The eightfold increase in the speed of the clock in the present experiment

suggests that three flip-flop circuits in the digital IC were short-circuited giving

eight times the number of electric pulses to the pulse motor, resulting in the faster

movement of the second hand (the loss of each flip-flop doubling the speed of the

second hand ).

Thus, the statements about rapidly moving clock hands before the Kobe earth-

quake appear to have nothing to do with the “Alice in Wonderland” Syndrome

(Ikeya et al., 1998a), but were probably real.

It should also be repeated that preseismic EM waves are not 120-130 MHz or

any particular frequency but are pulses composed of many different frequencies,

so having a much higher probability of affecting circuits than if they were one

specific frequency.

So, the reports of aberrant clocks before the Kobe, Izmit and Taiwan earthquakes

could have been real natural phenomena.

9.4.2 Dead phosphors on a wristwatch

A man passing through a tunnel near Izu (close to Tokyo) in a train looked at his

wristwatch. He was unable to read the phosphorescent markers and concluded

the phosphors had failed. This happened on August 29, 1923 and the Great Kanto

earthquake occurred two days later on September 1. On September 2 the man found

that the phosphorescence had reappeared (Kamei, 1976).

Page 17

9.5 Earthquake Forecasting Using Electric Appliances?

217

Experiment:

When a light emitting phosphor was placed under an intense electric field gener-

ated by the Van de Graaff generator, in a dark room, the phosphorescence faded

and then disappeared. It took hours for it to be restored.

Explanation:

In the 1920s phosphors in wristwatches were excited by alpha rays emitted by

radium. Alpha rays have a strongly ionizing effect, creating electrons and holes

(Figure 7.4 ). The slow recombination of electrons and holes in the phosphor was

responsible for the phosphorescence. An electric field removes ionized electrons

so reducing the efficiency of the electron-hole recombination in the phosphor.

Hence, phosphorescence may be degraded under an intense electric field before

an earthquake. When the field disappeared the distribution of the electrons would

return to normal in time.

9.5 Earthquake forecasting using electric appliances?

By exposing radios, TVs and air conditioners to EM noise at about 10 W/m2 or 60

V/m in many laboratory experiments we have reproduced the spontaneous switch-

ing of these devices and appliances that have been reported before earthquakes.

Essentially, the phenomenon of malfunctioning modern electric home appliances

before earthquakes (Figure 9.7) may be explained largely in terms of short circuits

Like animals we

also behave in

unusual ways

before earthquakes!

Figure 9.7 Electric home appliances—present and future—that can be expected to malfunc-

tion on exposure to EM waves.

Page 18

9. Mysteries Before Earthquakes: The Behavior of Electric Appliances

218

from digital data errors produced by electronic interference from intense EM pulses,

though some exaggeration or invention of reports cannot be excluded.

However it is risky to use electronic devices in earthquake forecasting. There

are too many other factors producing interference to make them reliable indicators

of an imminent large quake e.g. a sporadic E layer in the ionosphere formed by

solar flares (and maybe also by the apparent bending of the ionosphere towards

intense ground charges in epicenter areas), globally disturbs the transmission of

EM waves creating interference in radio and TV reception.

Radio and TV transmitters and receivers are not always maintained well, nor

is the power supply universally stable, and modern environments are noisy with

sudden EM discharges, so malfunctioning electric appliances are a fact of life for

many people. PC crashes are too frequent and commonplace to be used in predic-

tion. Consumer demand for better-performing appliances might eventually make

them more reliable indicators of imminent earthquakes, but electrical shielding of

circuitry will no doubt also improve.

In any event, if greater efforts are not made scientifically to predict earthquakes,

lay people may be able to get enough warning from sudden malfunctions in a group

of ordinarily well-performing electronic appliances to make sure life-saving and

property-protecting precautions are in place in case of a quake.

9.6 Care needed to avoid interference in data measurements

Many university geophysicists have tended to do their field work with cheap record-

ers and electrometers that are not made for regular outdoor use i.e. the manufac-

turer’s specifications stipulate use in air-conditioned rooms only. This introduces a

risk of electronic noise and errors. Perhaps the best-known example of interference

was the illusory discovery of “cold fusion.” In the late 1980s, scientists thought they

had found a way to produce energy by nuclear fusion at very low cost by putting

a current through heavy water at room temperature using special palladium and

platinum electrodes. But they were apparently misled by electronic interference

created partly by moisture precipitation on the circuit from the bubbling humidi-

fying action of the heavy-water cells. When they gained grants because of their

“discovery” and moved the operation into dehumidified rooms their cold fusion

discovery burst like the bubbles. They were unable to reproduce their result.

These cases are warnings to researchers, especially non-professionals, work-

ing in short term forecasting that the greatest care needs to be taken in measuring

seismo-electromagnetic signals (SEMS) because their equipment could be subject

to electronic interference—not least from EM waves themselves—and misleading

associations made.

Page 19

9.7 Summary

219

Necessary caution is one side of the story in EM/earthquake research. The other

side is an unwillingness or inability to see past an entrenched point of view. When

the laboratory response of a quartz clock to EM waves was described at a confer-

ence of the International Geophysical Union (IGU) at Birmingham, England in

1999, a leading geophysicist simply said “Rubbish! If a quartz clock in my more

than three hundred seismographs in California were to have malfunctioned, the

earthquake focus could not have been determined. We have not had an event like

that in the last twenty years!” The man dismissed the evidence outright because he

leapt to the defence of the clocks in his system (presumably well-shielded expensive

quartz clocks used in precise scientific instruments) and failed to see that cheap

quartz clocks in plastic casing fell into a different category altogether. In so doing

he failed to hear a legitimate case for EM effects on electronic objects.

In another, but much smaller, symposium a leading geophysicist said he had

considered certain seismograph clock data to be electronic error and omitted the

data from his results because no other local stations gave similar data. The author

upset him by suggesting the omitted data might have been a meaningful EM pulse

signal before the Northridge Earthquake and that the settings of their instruments

(viz. their low sampling frequency and the time window of their A-D converter)

meant the probability of two instruments picking up the same pulse would have

been very small. A skeptical referee also argued that electronic error might have

been the cause of the high activity levels reported in mice before the Kobe Earth-

quake (Chapter 5).

Strange data should not necessarily be blamed on electronic error; they could

be caused by preseismic EM interference.

9.7 Summary

Malfunctions in a range of modern electronic appliances, reported before four

large earthquakes in 1995 and 1999, have been reproduced in laboratory settings

by exposing them to EM waves, giving support to the emission of EM waves both

before and at the time of earthquakes.

Though EM interference can be produced from many sources, lay people should

be alert if they notice in a number of domestic electric appliances, over a short

period of time, odd functions of the kinds that have been described in this chapter.

An earthquake may follow, it may not, but it would be a good time to make sure that

precautions have been taken to save life and property in earthquake-prone areas.