Geology Reference
In-Depth Information
Spatial Cluster
Temporal Cluster
Trend
Random
0
8
16
24
32
40
48
earthquake
Distance
Distance
Distance
Distance
Regional Earthquake Scenarios
Fig. 6.1
Rupture scenarios for regional fault systems.
Contrasting spatial and temporal patterns for earthquakes on a regional system of faults. For spatial clusters, multiple
earthquakes occur on a single fault over an interval of time and then that fault turns off as the rupture process switches
to a different fault. For temporal clusters, multiple faults occur within a brief time window and then an interval of
quiescence follows. Spatial-temporal trends in earthquakes suggest that, as one event releases stress on a given fault or
section of a fault, it increases stress on nearby faults or segments, driving them to subsequent failure. In contrast to the
other three models, random ruptures preclude predictability for any given fault. Modified after Burbank
et al.
(2002).
the stratigraphy of beds that have been affected
by faulting and are now exposed in an artificial
trench across a fault - have been employed in
paleoseismological studies. The calculation of
recurrence intervals and of rates of displace-
ment depends both on the correct interpretation
of the geological record of past offsets and on
reliable dating of the timing of those offsets.
Approaches to dating and some of the pitfalls
and applications of various dating techniques
were described in Chapter 3. Generally, these
approaches are not discussed in more detail
here. Instead, in this chapter, we focus on some
of the many techniques that have been success-
fully used to reconstruct the record of seismicity
along faults that have been active during late
Quaternary times.
earthquake and is calculated as the average of
the stresses before and after the earthquake:
s
mean
=
(
s
start
−
s
finish
)/2
(6.2)
If it is assumed (as is often done) that the fault
is stress-free at the end of rupture (
s
finish
=
0),
then the mean stress is equal to one-half of the
stress at the start,
s
mean
½
s
start
.
Although comparison of earthquakes on the
basis of the energy released is now both possible
and very useful, traditionally the size of an
earthquake has been assessed on the basis of its
magnitude. The first quantitative measure of size
was the local magnitude (
M
L
) scale, which was
based on Richter's (1935) observation that, with
increasing distance from seismic sources in
southern California, the maximum amplitude of
ground motion decayed along a predictable
curve. When the data for distance versus the
logarithm (to base 10) of the amplitude of ground
shaking were compared for several earthquakes,
they followed parallel curves of decay with
increasing distance. By measuring the amplitude,
A
, of shaking in a given earthquake and comparing
it with the amplitude,
A
0
, of a “reference event,” a
local magnitude (
M
L
) can be defined:
M
L
=
log
A
−
log
A
0
=
Δ
s
=
½
Seismic moment and moment magnitudes
Any time an earthquake occurs, it releases
seismic energy,
E
s
. The energy released is pro-
portional to the area
A
of the rupture plane, the
average displacement
d
along it, and the stress
drop
Δ
s
across the fault during the earthquake:
=
Δ
s
dA
E
s
½
(6.1)
(6.3)
The stress drop actually refers to the mean
stress,
s
mean
, acting across the fault during the
For a reference earthquake with
M
L
=
0, the
amplitude of shaking at a distance,
D
, of 100 km
