5. How earthquakes
work
Can we predict them?
Effects visible today from M 7.3 1959
Hebgen Lake, Montana earthquake
SAN FRANCISCO
EARTHQUAKE
April 18, 1906
3000 deaths
28,000 buildings
destroyed
(most by fire)
$10B damage
“The whole street was
undulating as if the waves
of the ocean were coming
toward me.”
“I saw the whole city
enveloped in a pile of dust
caused by falling buildings.”
“Inside of twelve hours half
the heart of the city was
gone”
Motion along ~ 500 km
of previously
unrecognized San
Andreas Fault
~ 4 m of ground motion
West side moved north
DD
6.2
ELASTIC REBOUND
Over many years, rocks on opposite sides of the fault move,
but friction on fault "locks" it and prevents slip
Eventually strain accumulated overcomes friction,
and fault slips in earthquake
DD 6.3
Took 60 years to figure out why this happens!
EARTH’S OUTER SHELL - PLATES
Plates move
at few cm/yr
(speed
fingernails
grow)
San
Andreas
fault:
boundary
between
Pacific &
North
American
plates
DD 9.7
1906 Commission insights for policy
San Francisco’s city government and boosters stressed that
most damage came from fire, because if businesses thought
another earthquake would happen soon, they wouldn’t invest.
Elastic rebound showed that strain on the fault built up slowly
over many years. Because the fault had just slipped in the
earthquake, it wouldn’t slip again for a long time. Thus
investors could buy bonds knowing that no similar earthquake
would happen on a time scale that mattered to them.
Buildings needed to be built carefully, but there was no need to
rush because it would be a long time before another large
earthquake. Because buildings on weak soil or landfill suffered
much greater damage than ones on solid rock, they
recommended “studying carefully the site of proposed costly
public buildings.”
1906 Commission insights for recurrence
Elastic rebound gave a method to tell whether and when
another big earthquake would happen. The twelve feet that the
fault had moved in a few seconds had accumulated over many
years. Thus measuring the slow motion across the fault could
tell how long it would be until another earthquake that big could
happen.
The commission recommended that geodetic measurements
be made to measure the motion across the fault over time as
strain built up before a future earthquake. They suggested that
"we should build a line of piers, say a kilometer apart" across
the fault so "careful determination from time to time of the
directions of the lines joining successive piers ... would reveal
any strains which might be developing."
Strain accumulated in rubber band applies force
(shear stress) to block
Block slips when shear stress overcomes friction
due to weight of block (normal stress)
GPS: GLOBAL
POSITIONING SYSTEM
Satellites transmit radio
signals
Receivers on ground
record signals and find
their position
from the time the
signals arrive
Find mm/yr motions
from changes in
position over time
DD 12.1
GPS receiver finds
position with radio
signals from different
satellites
Conceptually like
locating earthquakes
GPS positions 2-3
times more precise in
the horizontal than
vertical, because
signals arrive only from
above.
Davidson et
al, 2002
Activity 5.1: GPS clock
To locate something to a meter, assess how
accurate the clock must be
What’s the speed of light?
How long does it take to travel a meter ?
“Any sufficiently advanced technology is
indistinguishable from magic”
Arthur C. Clarke
GPS uses very precise atomic clocks.
GPS CLOCK
Synchronizing satellite clocks within nanoseconds
(billionths of a second) lets a receiver find its
position on earth within a few meters.
Atomic clocks use the fact that atomic transitions
have characteristic frequencies
(orange glow from sodium in table salt sprinkled
on a flame).
Like all clocks, they make the same event happen
over and over. For example, the pendulum in a
grandfather clock swings back and forth at the
same rate, and swings of the pendulum are
counted to keep time.
In a cesium clock, transitions of the cesium atom
as it moves back and forth between two energy
levels are counted to keep time.
Since 1967, the
International System
of Units (SI) defines
the second as the
duration of
9 192 631 770 cycles
of the radiation
produced by transition
between two energy
levels of the ground
state of the cesium133 atom
Satellites transmit code - timing signals - on two microwave
carrier frequencies synchronized to very precise on-board
atomic clocks.
Carrier has much higher frequency than code
METER PRECISION GPS
GPS receiver compares code signal
arriving from satellite to one it
generates and finds time difference
Time difference x speed of light
gives the distance
called pseudo range
$450 Handheld GPS unit
with MP3 player
Precision of several meters
MILLIMETER PRECISION GPS
Higher precision obtained using phase of microwave carriers
Carrier wavelengths are 19 and 24 cm
Phase measurements resolve positions
to a fraction of these wavelengths
Measuring phase to 1% gives 1-2 mm
precision
Geodetic quality receivers cost
about $10,000
SURVEY (EPISODIC) GPS
GPS antennas are set up over
monuments for short periods, and
the sites are reoccupied later.
Less expensive but less precise.
GPS = Great Places to Sleep
CONTINUOUS (PERMANENT) GPS
Continuously recording GPS
receivers permanently installed.
Give daily positions & can observe
transient effects
Precision of
velocity
estimates
depends on
precision of
site position &
length of time
GPS VELOCITY ESTIMATES
Velocity from a
weighted least
squares line fit
to positions
Precision
increases over
time
Horizontal
precision is
better
Sella et al.,
2002
Activity 5.2: GPS across the San Andreas
GPS site motions show deformation accumulating that
will be released in future earthquakes
Like a deformed
fence
Z.-K. Shen
How fast is the motion?
Activity 5.3: Longer term slip rate
and earthquake recurrence on the
San Andreas Fault
Wallace Creek is offset
by 130 m
This offset developed
over 3700 years
What’s the average fault
slip rate?
If this happens in large
earthquakes with about
4 m slip, how often on
average should they
occur?
San
Andreas
Fault
DD 9.8
Activity 5.4: Time between earthquakes from
paleoseismic record
DD 9.8
What’s the mean and standard deviation of the time between
large earthquakes? If the last was in 1857, when would you
expect the next one?
Activity 5.5: Why no earthquake yet?
What might the problems be?
What does this suggest about predicting
earthquakes?
“We are predicting
another massive
earthquake
certainly within the
next 30 years and
most likely in the
next decade or so.”
W. Pecora, U.S.
Geological Survey
Director, 1969
Traditional skepticism
“Only fools and charlatans predict earthquakes”
Charles Richter (1900-1985)
1970’s optimism
Scientists will “be able to predict earthquakes
in five years.”
Louis Pakiser , U.S. Geological Survey, 1971
“We have the technology to develop a
reliable prediction system already in hand.”
Alan Cranston, U.S. senator, 1973
“The age of earthquake prediction is upon
us”
U.S. Geological Survey, 1975
PARKFIELD, CALIFORNIA SEGMENT OF SAN ANDREAS
M 5-6 earthquakes about every 22 years: 1857, 1881,
1901, 1922, 1934, and 1966
In 1985, expected next in 1988; U.S. Geological Survey
predicted 95% confidence by 1993
Occurred in 2004 (16 years late)
Discounting
misfit of
1934 quake
predicted
higher
confidence
Science, 10/8//04
"Parkfield is geophysics' Waterloo.
If the earthquake comes without
warnings of any kind, earthquakes
are unpredictable and science is
defeated. " (The Economist)
No precursors in seismicity
(foreshocks), strainmeters,
magnetometers, GPS, creepmeter
$30 million spent on “Porkfield”
project
WHY CAN’T WE PREDICT EARTHQUAKES?
So far, no clear evidence for consistent changes in physical properties
(precursors) before earthquakes.
Maybe lots of tiny earthquakes happen frequently, but only a few grow
by random process to large earthquakes
In chaos theory, small perturbations can have unpredictable large
effects - flap of a butterfly's wings in Brazil might set off a tornado in
Texas
If there’s nothing special about the tiny earthquakes that happen to
grow into large ones, the time between large earthquakes is highly
variable and nothing observable should occur before them.
If so, earthquake prediction is either impossible or nearly so.
small perturbations grow in simple function
2x**2-1
1.5
1
2x**2-1
0.5
Series1
0
0
5
10
15
20
25
30
Series2
-0.5
-1
-1.5
time
Starting values
Series 1: 0.750
Series 2: 0.749
Deterministic contributor - stress changes
A big earthquake can change stress on
nearby faults, and make it easier or
harder for them to move in a big
earthquake. That's called "unclamping"
or "clamping" a fault.
Think back to the soap bar and yoga
mat. Increasing the tension in the
rubber band or reducing the weight of
the soap bar unclamps the soap and
makes slip easier. Conversely,
reducing tension in the rubber band or
putting weight on the soap clamps it,
making slip harder.
Stress transfer can cause earthquakes to migrate
along a plate boundary. The best example is in
Turkey, where since 1939 large earthquakes have
occurred successively further to the west along
the North Anatolian fault.
1906 San Francisco earthquake may have reduced failure stress on other faults in
the area, causing a "stress shadow" and increasing the expected time until the next
earthquake on these faults.
This is consistent with the observation that in 75 years before the 1906 earthquake,
the area had 14 earthquakes with Mw > 6, whereas only three occurred since
Stress
change
SAF
USGS
Status today
Meaningful prediction involves specifying the location, time,
& size of an earthquake before it occurs
Long-term forecast: some success
- Use earthquake history to predict next one
- Use rate of motion accumulating across fault and amount
of slip in past earthquakes
Short-term prediction:
-Find precursors - changes in earth before earthquakes
consistently resolvable from normal variability
Despite some claims, no reliable method yet…
Most claim of success to date have
been postdictions
Texas sharp shooter
Shoot at barn and then draw target around bullet holes