The mechanical model for the October 15,
1979, earthquake event considers only the
rupture along the Imperial Fault (Archuleta,
1984). The lower four panels of Fig. 1.5 are graphs
of different physical quantities calculated using
the model and plotted on a vertical planar section
that approximates the more complex geometry of
the actual fault as suggested by the mapped trace
in the first panel. The model fault is about 12 km
in depth (ordinate) and 35 km in length (abscissa).
The physical quantities (rupture time, slip duration, strike slip, and dip slip) are represented by
contours of equal magnitude on these graphs.
Together these panels provide a remarkable visualization of the model slip event from the Earth’s
surface to the bottom of the rupture.
Figure 1.5b illustrates the position of the
leading edge of the model rupture to the north of
the point of rupture initiation, the hypocenter, at
times measured in seconds after initiation. What
happened to the south of the hypocenter is
ignored on these panels. At a given time, say 4 s,
that portion of the fault between the hypocenter
and the 4-s contour has slipped, while elsewhere
on the fault no slip has occurred. Clearly, slip on
the model fault does not initiate everywhere
simultaneously. Rather, the model rupture initiated at a point, at the hypocentral depth of about
8 km. Then, the rupture front advanced rapidly to
the north and less rapidly upward toward the
surface. The rupture took a total time of about 12 s
to spread the 35 km to the north end of the model
fault. Thus, the average rupture velocity was about
3 km s
Ϫ1 toward the north, approximately the
speed of seismic shear waves.
Figure 1.5c shows the total time that originally
adjacent particles on the two surfaces of the
model fault were in relative motion. For example,
along the contour labeled “1.6 s” the two surfaces
slipped for a total time of less than 2 s. You might
find this surprising given the fact that the total
duration of faulting was about 12 s. Clearly all
parts of the model fault were not slipping at the
same time. This is illustrated in the previous panel
by the pattern of dots next to the 8-s contour.
These dots cover the relatively small portion of
the fault that has already slipped and is still in the
process of slipping at the moment that the
rupture front lies along the 8-s contour. Between
these dots and the hypocenter the model fault has
slipped and stopped, whereas to the north and
above the 8-s contour the fault has not yet slipped.
At any particular location on the model fault the
slipping occurred over a period of time ranging
from a fraction of a second to almost 2 s as the
rupture front passed, and then slipping stopped.
Figures 1.5d and e show, respectively, two components of slip between the model fault surfaces
after the rupture has completed its propagation
from the hypocenter to the northern termination.
Strike slip varies from 1.4 m near the bottom
center of the fault to a few decimeters or less at
the surface. The strike slip is zero along the southern portion of the fault at the surface and this is
consistent with the observations shown on the
map in the first panel. Note that the surface measurements of slip, amounting to about 20 cm,
under-represent the slip at depth by a factor of
eight or more. The model fault slipped much
more at depth than at the surface. Dip slip is concentrated near the surface at the northern end of
the model fault with magnitudes approaching
a few decimeters. This is consistent with the
geological observations (see Fig. 1.1) that indicate
the rocks on the northeastern side of this part of
the Imperial Fault went down relative to those
on the southwestern side. The amount of dip slip
at the surface (up to about 20 cm) also is consistent with the field observations. The relative
motion on the Brawley Fault was also nearly pure
dip slip with the northwestern side down. In fact,
the region between the Imperial and Brawley
Faults is a topographic depression occupied by a
(usually) dry lake-bed. This suggests that the relative motion experienced during the 1979 earthquake is typical of the recent geological history of
this fault system.
This mechanical model gives us a picture
of active faulting that is reasonably consistent
with the available surficial and seismic data from
the 1979 event. It informs our intuition about
the physical process of faulting and provides a
glimpse into possible behavior along the Imperial
Fault at depth. Building models such as this one
and using these models to understand the process
of faulting is an exciting area of research in which
structural geologists can participate (Segall and
Pollard, 1980; Aydin and Schultz, 1990; Cowie and
8
MOTIVATIONS AND OPPORTUNITIES
1979, earthquake event considers only the
rupture along the Imperial Fault (Archuleta,
1984). The lower four panels of Fig. 1.5 are graphs
of different physical quantities calculated using
the model and plotted on a vertical planar section
that approximates the more complex geometry of
the actual fault as suggested by the mapped trace
in the first panel. The model fault is about 12 km
in depth (ordinate) and 35 km in length (abscissa).
The physical quantities (rupture time, slip duration, strike slip, and dip slip) are represented by
contours of equal magnitude on these graphs.
Together these panels provide a remarkable visualization of the model slip event from the Earth’s
surface to the bottom of the rupture.
Figure 1.5b illustrates the position of the
leading edge of the model rupture to the north of
the point of rupture initiation, the hypocenter, at
times measured in seconds after initiation. What
happened to the south of the hypocenter is
ignored on these panels. At a given time, say 4 s,
that portion of the fault between the hypocenter
and the 4-s contour has slipped, while elsewhere
on the fault no slip has occurred. Clearly, slip on
the model fault does not initiate everywhere
simultaneously. Rather, the model rupture initiated at a point, at the hypocentral depth of about
8 km. Then, the rupture front advanced rapidly to
the north and less rapidly upward toward the
surface. The rupture took a total time of about 12 s
to spread the 35 km to the north end of the model
fault. Thus, the average rupture velocity was about
3 km s
Ϫ1 toward the north, approximately the
speed of seismic shear waves.
Figure 1.5c shows the total time that originally
adjacent particles on the two surfaces of the
model fault were in relative motion. For example,
along the contour labeled “1.6 s” the two surfaces
slipped for a total time of less than 2 s. You might
find this surprising given the fact that the total
duration of faulting was about 12 s. Clearly all
parts of the model fault were not slipping at the
same time. This is illustrated in the previous panel
by the pattern of dots next to the 8-s contour.
These dots cover the relatively small portion of
the fault that has already slipped and is still in the
process of slipping at the moment that the
rupture front lies along the 8-s contour. Between
these dots and the hypocenter the model fault has
slipped and stopped, whereas to the north and
above the 8-s contour the fault has not yet slipped.
At any particular location on the model fault the
slipping occurred over a period of time ranging
from a fraction of a second to almost 2 s as the
rupture front passed, and then slipping stopped.
Figures 1.5d and e show, respectively, two components of slip between the model fault surfaces
after the rupture has completed its propagation
from the hypocenter to the northern termination.
Strike slip varies from 1.4 m near the bottom
center of the fault to a few decimeters or less at
the surface. The strike slip is zero along the southern portion of the fault at the surface and this is
consistent with the observations shown on the
map in the first panel. Note that the surface measurements of slip, amounting to about 20 cm,
under-represent the slip at depth by a factor of
eight or more. The model fault slipped much
more at depth than at the surface. Dip slip is concentrated near the surface at the northern end of
the model fault with magnitudes approaching
a few decimeters. This is consistent with the
geological observations (see Fig. 1.1) that indicate
the rocks on the northeastern side of this part of
the Imperial Fault went down relative to those
on the southwestern side. The amount of dip slip
at the surface (up to about 20 cm) also is consistent with the field observations. The relative
motion on the Brawley Fault was also nearly pure
dip slip with the northwestern side down. In fact,
the region between the Imperial and Brawley
Faults is a topographic depression occupied by a
(usually) dry lake-bed. This suggests that the relative motion experienced during the 1979 earthquake is typical of the recent geological history of
this fault system.
This mechanical model gives us a picture
of active faulting that is reasonably consistent
with the available surficial and seismic data from
the 1979 event. It informs our intuition about
the physical process of faulting and provides a
glimpse into possible behavior along the Imperial
Fault at depth. Building models such as this one
and using these models to understand the process
of faulting is an exciting area of research in which
structural geologists can participate (Segall and
Pollard, 1980; Aydin and Schultz, 1990; Cowie and
8
MOTIVATIONS AND OPPORTUNITIES