AFTERSHOCK
ACCELEROGRAMS
RECORDED
ON
A
TEMPORARY
ARRAY
By
J.
G.
ANDERSON,
ScRIPPS
INSTITUTION
OF
OcEANOGRAPHY;
and
T.
H.
HEATON,
CALIFORNIA
INSTITUTE
OF
TECHNOLOGY
CONTENTS
Page
1\bstract
--------------------------------------------------
443
Instrumentation
------------------------------------------
443
Recovereddata
--------------------------------------------
443
1\cknowledgrnents
___
_____
_____
_____
___________
_ __________
445
References
cited
_
____
_
___
___
___
_____
_______
______
_____
___
_
445
ABSTRACT
We
recovered
52
timed
analog
accelerograms
from
25
aftershocks
of
the
1979
Imperial
Valley
earthquake,
between
3:33p.m.
P.d.t.
Octo-
ber
16
and
5:43
a.m.
October
31.
The
largest
aftershock
that
we
recorded
(ML
=4.9)
occurred
at
4:16p.m.
October
16.
This
aftershock
triggered
eight
accelerographs;
preliminary
estimates
of
epicentral
distance
range
from
7
to
35
km.
The
data
from
this
aftershock
may
be
useful
for
study
of
both
source
and
wave-propagation
phenomena
in
the
Imperial
Valley.
INSTRUMENTATION
In
the
evening
hours
after
the
October
15,
1979,
earthquake,
a
field
crew
was
organized.
Beginning
early
the
next
morning,
they
installed
10
accelerog-
raphs
along
the
Imperial
fault
from
north
of
Brawley,
Calif.,
to
the
United
States-Mexican
border,
with
the
object
of
recording
aftershocks.
The
accelerographs
(Kinemetrics
model
SMA-1)
record
three
components
of
acceleration
on
70-mm
film,
in
the
amplitude
range
of
less
than
0.01
to
more
than
1
g
and
in
the
frequency
band
0
to
about
25Hz.
These
instruments
were
slated
to
be
installed
in
the
Los
Angeles
area
as
part
of
a
network
being
established
there
by
the
University
of
Southern
California
and
the
California
Division
of
Mines
and
Geology.
Each
instrument
has
an
internal
clock
(Kinemetrics
TCG-1)
that
generates
a
time
code
to
be
recorded
on
the
film.
The
clocks
have
a
temperature-compensated
crys-
tal
oscillator
with
a
frequency
stability
of
±3x
10-
7
in
I
the
temperature
range
0°-50°C;
this
stability
implies
a
daily
drift
of
less
than
26
ms,
provided
the
crystal
fre-
quency
is
set
properly.
The
clocks
tended
to
drift
much
less
than
this
limit.
Clocks
were
corrected
in
the
labora-
tory
before
instrument
deployment,
at
the
time
of
in-
stallation
on
October
16,
at
most
stations
on
October
17,
and
at
the
time
the
instruments
were
recovered
on
November
1
or
2.
There
is no
reason
to
suspect
an
uncer-
tainty
of
greater
than
10
ms
in
the
clock
corrections
interpolated
to
the
occurrence
times
of
aftershocks.
Table
48lists
the
locations,
elevations,
and
orientations
of
the
accelerograph
installations
(see
fig.
296),
as
well
as
the
sensor
characteristics
for
each
accelerometer.
RECOVERED
DATA
One
or
more
of
the
accelerographs
were
triggered
by
at
least
25
aftershocks
with
preliminary
local
mag-
nitudes
estimated
at
from
less
than
3.0
to
4.
9.
Because
of
clock
problems,
the
times
of
5
of
the
52
total
triggerings
are
uncertain.
Triggering
times
and
S-wave
arrival
times
for
the
remaining
4 7
records
(table
49)
were
read
directly
from
the
film
accelerograms.
In
addition
to
these
data,
table
49
lists
the
stations
that
recorded
the
aftershocks
and
the
peak
accelerations
recorded
for
each
aftershock.
These
peak
accelerations,
obtained
from
the
maximum
trace
amplitudes,
have
not
been
corrected
for
the
response
of
the
accelerometers.
Figure
297
illustrates
the
spatial
relation
between
the
pre-
liminary
aftershock
locations
and
the
accelerographs
that
recorded
them.
About
75
percent
of
the
triggerings
occurred
on
the
three
northernmost
accelerographs
(stations
3689,
3692,
3698,
fig.
296);
the
two
south-
ernmost
stations
(3695,
3696)
did
not
record
any
after-
shocks.
This
pattern
is
consistent
with
the
distribution
of
the
aftershock
sequence,
in
which
most
of
the
later
aftershocks
occurred
north
of
Brawley,
Calif.
(Johnson
and
Hutton,
this
volume).
443
444
THE
IMPERIAL
VALLEY,
CALIFORNIA,
EARTHQUAKE
OF
OCTOBER
15,
1979
TABLE
48.-Temporary
accelerograph
sites
[Location
is
within
±0.003°.
Sensor
orientation
is direction
of
ground
motion
that
causes
positive
trace
motion
on
the
film;
upward
groun~
acceleration
causes
positive
trace
motion
on
the
vertical
component.
Accelerograph
characteristics
were
supplied
by
Kinemetrics;
damping
is
60
percent
of
critical
on
all
instruments
Location
Site
Station
Latitude
N.
Longitude
W.
Del
Rio
Country
Club
______________
3698
33.0095°
115.5207°
Brawley
Airport
------------------
3689
32.9914°
115.5167°
Sam
Etchegaray
Livestock
Co.
____________________
3692
32.9816°
115.4743°
Doyle
McDuffy
home
______________
3681
32.9133°
115.4920°
Memory
Gardens
Cemetery
________
3693
32.8919°
115.5676°
Sharon
Fox
home
__________________
3678
32.8437°
115.4996°
University
of
California
Field
Station
------------------------
3697
32.8044°
115.4467°
McGrew
Farm
____________________
3688
32.7769°
115.4448°
Jasper
Road
----------------------
3695
32.7090°
115.4244°
Tuttle
Ranch
______________________
3696
32.6940°
115.3745°
Station
3698
did
not
record
any
events
between
4:16
p.m.
P.d.t.
October
16
and
1:33
p.m.
the
next
day
be-
cause
the
film
jammed.
The
vertical
component
of
the
station
3692
instrument
was
knocked
badly
out
of
alinement
on
the
trip
to
the
Imperial
Valley
or
during
installation
and,
therefore,
did
not
record
any
data.
Stations
3688, 3693, 3695,
and
3696
were
not
on
battery
chargers,
and
the
batteries
were
drained
when
those
instruments
were
recovered
on
November
2.
The
largest
aftershock
recorded
was
an
M
L
=
4.
9
event
that
occurred
at
4:16p.m.
P.d.t.
October
16;
this
after-
shock
triggered
the
eight
northernmost
accelerographs.
The
preliminary
epicentral
location
is
7¥2
km
north-
northwest
of
station
3698,
at
a
depth
of
5 km.
This
event
is
characterized
by
strong
relatively
long
period
accel-
erations
on
the
horizontal
components
of
the
three
northernmost
stations,
and
by
high-frequency
shaking
on
the
vertical
components.
At
stations
3689
and
3698,
Sensor
orientation
Accelerograph
characteristics
Elevation
(azimuth)
Natural
(ft)
Axis
Sensitivity
fre(}l=}cy
Longitudinal
Transverse
(mm/g)
-145
081°
351°
Long.
18.9
25.8
Vert.
18.8
25.6
Trans.
18.3
25.7
-130
250"
160°
Long.
18.7
25.0
Vert.
18.6
24.8
Trans.
19.0
24.9
-135
338°
24SO
Long.
18.5
24.5
Vert.
18.5
25.5
Trans.
17.4
26.4
-132
334°
244°
Long.
19.7
24.2
Vert.
18.2
25.7
Trans.
17.5
26.3
-80
095°
005°
Long.
19.6
24.2
Vert.
18.0
25.6
Trans.
19.3
24.7
-100
357°
267°
Long.
18.1
26.1
Vert.
19.4
25.6
Trans.
18.3
25.8
-55
000"
270°
Long.
18.9
24.9
Vert.
19.4
25.2
Trans.
17.6
26.0
-25
35~
266°
Long.
19.5
24.3
Vert.
17.6
25.8
Trans.
17.0
27.2
15
180°
090°
Long.
18.1
26.3
Vert.
18.7
25.3
Trans.
19.3
25.2
25
100"
010°
Long.
17.8
26.2
Vert.
18.2
25.4
Trans.
17.1
26.9
near-field
effects
are
clearly
visible
on
the
horizontal
components
before
the
arrival
of
the
S
wave
(fig.
298).
These
records
may
be
useful
for
studies
of
wave
propa-
gation
and
attenuation
along
the
fault
zone.
Several
accelerograms
in
figure
298
have
been
digi-
tized
on
the
automatic
digitizer
at
the
University
of
Southern
California
and
integrated
according
to
the
accelerogram-processing
techniques
of
Trifunac
and
Lee
(1973).
Figure
299
shows
the
results
for
the
three
northernmost
stations.
This
processing
is
not
optimal
for
all
these
records
because
long-period
ringing
from
the
filter
may
be
present
on
some
processed
displace-
ment
traces,
and
nearfield
permanent
displacements
are
removed
by
the
baseline
correction.
The
peak
accel-
erations
on
these
digitized
and
processed
records
are
smaller
than
those
listed
in
table
49
for
all
stations,
and
for
the
vertical
component
at
station
3689
the
reduction
is
large-from
0.63
to
0.26g.
This
effect,
which
has
been