ELASTIC REBOUND THEORY
A. INTRODUCTION
Earthquakes occur along breaks within the earth's crust known as
faults. Most
earthquakes occur along
pre-existing faults, but a new
fault can be created during an
earthquake. Note that large faults
(e.g., the San Andreas Fault) do not move all at once. Instead, only
certain segments move during any
given earthquake.
Two terms are used to describe the point of origin for earthquakes:
1. FOCUS
This is the actual location where fault movement begins. Almost
every earthquake
has its focus located below the earth's surface.
2. EPICENTER
This is
the point on the land surface directly above the focus and
is the
location normally reported in the news or shown on maps.
Note that an
epicenter need not be located on a fault line.
B. OBSERVATIONS ABOUT EARTHQUAKES
1. WHERE EARTHQUAKES OCCUR
Recent
worldwide seismicity
Worldwide seismicity
(1975 - 1991)
Recent U.S.
seismicity
Large
historic earthquakes (U.S.)
Most earthquakes (esp.
large
ones) occur at plate boundaries,
but there are many exceptions
(e.g., Missouri
and Hawaii).
2. MAGNITUDE AND FREQUENCY
In
general, the larger the magnitude of an earthquake, the larger
is its
recurrence
interval (i.e., time between earthquakes) and the
greater the amount of fault movement
during the earthquake.
Very large earthquakes
can cause sizeable offset (e.g., the
1906
San
Francisco earthquake, which had a magnitude of more than
8.0 and has a
recurrence interval of over 300 years, caused more
than
20 feet of displacement along the San Andreas Fault).
At the other end of the spectrum
is the slow
continual movement
along a
fault (called fault
creep), which produces earthquakes so
small
they can only be detected with seismometers.
3. PRECURSORY EVENTS
Many earthquakes are preceded by events that suggest stress is
building up along the fault where the earthquake
a. Fault Zone "Bulges" (Dilatancy)
b. Earthquake Lights
(EQLs) natural
gas
c. Foreshocks
(Parkfield)
d. Animal Behavior
(Haicheng, China)
1. HYPOTHESIS
Based on the observations listed above, geologists hypothesize
that if rocks
are under stress, they deform elastically
(analogous
to a
rubber band). Strain builds up until
either: the rocks break
(creating a new fault), or movement
occurs on an existing fault.
As stored strain is released during an earthquake, the
deformed
rocks
"rebound" to their undeformed shapes. The magnitude of
the
earthquake reflects how much strain was released. Residual
strain
not released during the initial earthquake is often released
in
smaller
aftershocks.
Does this hypothesis allow geologists to predict earthquakes?
a. Predicting Where Earthquakes Will Occur
Segments
of active faults that have not moved recently are
known as
seismic
gaps. These are locations where stress
is building up and where earthquakes can be expected in
the future.
In fact, the 1989 Loma Prieta earthquake occurred in along a
portion of the San Andreas Fault identified as a seismic gap.
b. Predicting When Earthquakes Will Occur
In cases where there is sufficient historical information (e.g.,
Parkfield,
CA), recurrence intervals can be determined and
used to calculate the
probability of an earthquake for some
specified time period extending into the future.
3. DIFFICULTIES
Testing the elastic rebound model is difficult because:
a. Recurrence
intervals are known only for a few well-studied
fault zones (e.g., the San Andreas Fault). By
contrast, little
is known about recurrence intervals or strain rates
for areas
where large intraplate
earthquakes have occurred.
b. The rates of strain build up could change through time, and
this would change earthquake recurrence intervals.
c. Strain is sometimes distributed along
several parallel faults,
so it is difficult to know how much strain has built up on any
particular fault segment.
Thus, the state of the art is to forecast
earthquake probabilities,
and
earthquake prediction has achieved only limited success.
