Preliminary
Report
on
the
16
October
1999
M
7.1
Hector
Mine,
California,
Earthquake
Scientists
from
the
U.S.
Geological
Survey,
Southern
California
Earthquake
Center,
and
California
Division
of
Mines
and
Geology
INTRODUCTION
The
Mw
7.1
Hector
Mine,
California,
earthquake
occurred
at
9:46
GMT
on
16
October
1999.
The
event
caused
mini-
mal
damage
because
it was
located
in
a remote,
sparsely
pop-
ulated
part
of
the
Mojave
Desert,
approximately
47
miles
east-southeast
of
Barstow,
with
epicentral
coordinates
34.59°N
116.27°W
and
a hypocentral
depth
of
5
±
3 km.
Twelve
foreshocks,
M
1.9-3.8,
preceded
the
mainshock
dur-
ing
the
previous
twelve
hours.
All
of
these
events
were
located
close
to
the
hypocenter
of
the
mainshock.
The
Hector
Mine
earthquake
occurred
within
the
East-
ern
California
Shear
Zone
(ECSZ).
By
virtue
of
its
remote
location,
the
societal
impact
of
the
Hector
Mine
earthquake
was,
fortunately,
minimal
in
spite
of
the
event's
appreciable
size.
The
ECSZ
is
characterized
by
high
seismicity,
a high tec-
tonic
strain
rate,
and
a broad,
distributed
zone
of
north-
northwest-trending
faults
(ECSZ;
Figure
1;
Dokka
and
Travis,
1990;
Sauber
et al.,
1986;
Sauber
et al.,
1994;
Sieh
et
al.,
1993).
Data
regarding
the
slip
rates
offaults
within
the
ECSZ
suggest
that
on
the
order
of
15%
of
the
Pacific-North
American
plate
motion
occurs
along
this
zone
(Sauber
et al.,
1986;
Wesnousky,
1986).
Most
of
the
faults
in
the
ECSZ
have
low
slip
rates
and
long
repeat
times
for
major
earth-
quakes,
on
the
order
of
several
thousands
to
tens
of
thou-
sands
of
years.
The
occurrence
of
the
Hector
Mine
earthquake
within
seven
years
and
only
about
30
km
east
of
the
1992
Mw
7.3
Landers
earthquake
suggests
that
the
closely
spaced
surface
faults
in
the
ECSZ
are
mechanically
related.
The
Hector
Mine
event
involved
rupture
on
two
previ-
ously
mapped
fault
zones-the
Bullion
Fault
and
an
unnamed,
more
northerly-trending
fault
that
is
informally
referred
to
in
this
paper
as
the
Lavic
Lake
Fault
(Dibblee,
1966,
1967a,b).
Traces
of
the
Bullion
Fault
exhibit
evidence
of
Holocene
displacement
and
were
zoned
as
active
in
1988
under
California's
Alquist-Priolo
Earthquake
Fault
Zoning
Act
(Hart
and
Bryant,
1997).
The
pattern
of
rupture
along
more
than
one
named
fault
was
also
observed
from
the
1992
Landers
earthquake
(Hauksson
eta!.,
1993;
Sieh
eta!.,
1994).
Much
of
the
fault
zone
that
produced
the
Hector
Mine
earthquake
had
been
buried
by
relatively
young
stream
deposits,
and
the
fault
scarps
in
bedrock
have
a subdued
mor-
phology.
It
appears
that
these
faults
have
not
experienced
sig-
nificant
offset
for
perhaps
10,000
years
or
more
(Hart,
1987).
Planned
future
investigations
will
refine
the
age
of
the
last
event
on
these
faults.
The
portion
of
the
Lavic
Lake
Fault
that
ruptured
between
the
northern
end
of
the
Bullion
Mountains
and
Lavic
Lake
had
not
previously
been
mapped.
However,
our
field
investigations
have
identified
ancient,
subdued
fault
scarps
along
portions
of
the
1999
rupture
zone
in
this
area.
It
thus
appears
that
the
entire
segment
of
the
Lavic
Lake
Fault
that
was
involved
in
the
1999
event
had
ruptured
in
the
past.
As
is
typical
for
most
faults
within
the
Eastern
California
Shear
Zone,
the
rate
of
movement
along
the
Lavic
Lake
Fault
may
be
quite
slow
(
<
1 mm/yr)
and
should
produce
earth-
quakes
only
infrequently.
This
event
is
a reminder
that
faults
that
have
ruptured
in
late
Quaternary
time,
but
that
lack
evi-
dence
of
Holocene
displacement,
can
still
produce
earth-
quakes
in
this
low-slip-rate
tectonic
setting.
Additionally,
the
Hector
Mine
earthquake
is
notewor-
thy
for
a couple
of
other
reasons.
First,
it clearly
produced
triggered
seismicity
over
much
of
southern
California,
from
the
rupture
zone
toward
the
south-southwest
in
particular.
Second,
as
we
will
discuss,
the
event
may
provide
new
data
and
insight
into
recently
developed
paradigms
concerning
earthquake
interactions
and
the
role
of
static
stress
changes.
Questions
such
as
these
will,
of
course,
be
the
subject
of
extensive
detailed
analyses
in
years
to
come.
Fortunately,
the
Hector
Mine
sequence
will
provide
one
of
the
best
data
sets
obtained
to
date
for
a significant
earthquake
in
the
United
States.
Because
it occurred
when
major
upgrades
to
both
the
regional
seismic
network
(TriNet)
and
the
regional
geodetic
network
(SCIGN)
were
well
underway,
the
Earth
science
community
will
have
abundant
high-quality
data
with
which
to
explore
the
important
and
interesting
questions
that
have
been
raised.
In
this
paper,
we
present
and
discuss
the
basic
data
and
preliminary
results
from
the
Hector
Mine
earthquake.
GEOLOGIC
FIELD
OBSERVATIONS
On
Saturday,
16
November
1999,
from
3:00
to
6:00
P.M.
(local
time),
scientists
made
a
helicopter
reconnaissance
Seismological
Research
Letters
Volume
71,
Number
1
January/February
2000
11
35
00'
34
°
40
'
34
20
'
LUG
~
-
11
7"
00'
-
11
6°
00
'
.A.
Fi
gure
1.
Eastern
Mojave
region
inc
l
uding
regiona
l
faults
(light
pink
lines)
and
topography
(gray-scale).
Open
black
circles
indicate
seismicity
(mainshocks
and
ear
ly
aftershocks)
associated
wit
h
the
1992
J
oshua
T
ree
-
L
anders
-
Big
Bear
sequence
.
Filled
yellow
c
ir
cl
es
i
nd
icate
aftershocks
of
the
10
/16
/
99
Hector
Mine
earthquake
. R
ed
lines
show
surface
rupture
associa
t
ed
with
theM
7.3
L
anders
earthquake
and
with
t
heM
7.1 H
ector
Mine
earthquake
.
Ai
g
ht
an
d found
the
n
ort
hwest
a
nd
so
uthe
as tern
e
nd
s
of
th
e
s
urf
ace
ruptur
e,
as
wel
l as
the
a rea
of
maximum
slip.
Th
ey
made
a prelimin
a ry
m
ap
of
the
fa
ult
bre
aks,
upon
which
Ai
g
ht
lin
es
we
re
pl
a
nn
ed
for
ae
ri
al
photo
gra ph
y.
Militar
y
ope
ratio
ns res
um
ed a
ll
day
lon
g
on
S
und
ay,
17
O c
tob
e
r.
A se t
of
275
air
photo
s was
then
tak
en
at
1:
I 0 ,
000
sca
le
at
d aw
n
(for
l
ow
s
un
angl
e)
o n
Monda
y,
18
O ctober.
On
18
a
nd
22-
24
Octob
e r,
geo
logical
Field
crews we re ab le
to
make c
ompre
-
hensi
ve
meas
ur
em e
nt
s a
ll
al
ong
the
fault
zone
during
a
res
pit
e in
the militar
y trainin
g opera
ti o ns
at
t he
ba
se.
Tn
the wee
k followin
g the H ector
Min
e ea
rthquake
,
up
to
rhirt
een rea
ms
of
geolo
gists from
the
U.S.
G eolo
gi
ca
l
S
ur
vey,
S
outh
ern
Ca
lifo rni
a Earthquake
Ce
nt
e r, a
nd
Ca
liforni
a Divi
-
sio n
of
Min
es
a
nd
Geo
logy
co
ll
aborated
to
map
the ex te
nt
a
nd
co
mpl
exity
of
s
urf
ace
fau
lt
rupture
.
Fa
ultin
g
wit
h
observed
right
-l
ate
ral dis
pl
ace
me
nt
was
trac
ed
for
a
ppro
xi-
m
at
el
y
4 1
km
alo ng the
mapped
trace
s
of
rh e Lavic
Lak
e
and
Bullion
Fa
ult
s (
Fi
g
ur
e 2) .
Minor
fault di
s
pl
ac
ement
s were
a lso
obser
ve
d
on
spla
ys
ex te
ndin
g from
the
main
r
uptur
e zo
ne a
nd
alon
g s
ubp
a rall
el
fa
ult
tr
aces
located
within
a few
kilom
eters
.A.
Fi
gure
2.
Simplified
map
of
surface
fault
rupture
associated
with
the
H
ec
t
or
Mi
ne
earth
qu
ake
(ye
ll
ow).
The
r
upture
involved
the
sou
thernm
os
t
portion
of
the
Bullion
F
ault
,
previously
recognized
as
Hol
ocene-active
,
and
the
previously
unnamed
"
Lavic
Lake
"
fault.
Pri
or
to
the
earthquake
,
the
Lav
ic
Lak
e
Fault
had
been
mapped
in
t
he
Bullion
Mountains
only
and
did
not
dis
p
lay
evidence
of
H
olocene
activity
. H
oloce
n
e-active
traces
of
other
nearby
f
aults
in
the
vic
i
nity
ar
e
identi
f
ied
by
the
dashed
whit
e
lines
. N
atu
r
al
color
digital
orthoimagery
,
at
.5
meter
resolution
,
was
provided
by
Earth
-
Data
International
under
contract
to
the
U.S.
Marine
Co
rp
s
Air
Ground
Combat
Center
,
Geospatiallnformation
and
Services/
R
emote
Sensing
L
ab
.
of
the main
zo
ne. Includin
g thes
e additi
o nal fault
s, rhe
o
ve
ra
ll
leng
th
of
s
urf
ace
fa
ultin
g is
appro
xim
ately
45
km.
Th
e fau
lt
ruptur
e is ch aracte
ri
zed b
y lin
ear
sca
rp
s, r
ig
ht
-
lat
era
lly offse
t s
tr
e
am
channels
, mo le tra
c
ks
,
tr
e nches,
s
hut
-
ter rid
ges,
a
nd
left
-s
teppin
g en
ec
he
l
on
fractur
es
co
mm
on
ly
c
onn
ected
by
thru
st faults.
Co
mpl
ex
ity
of
s
ur
face
faultin
g
va
ri
es
from
a simpl
e lin
ea r scar
p
to
multipl
e
br
a nches w
ithin
a zo
ne seve
ral
hundr
ed
met
ers w
id
e. Overall
, there
was
less
d i
str
i
but
ed
fau
lti ng
than
was
obse
r
ve
d
from
the
1
992
La
nd
ers
ea
rthqu
ake,
bur
there
a re
seve
ral r
eg
ion
s where fau
lt
p
atterns
a re
quir
e c
omple
x. In
pa rticular,
dis
tribut
ed
rupture
occur
s betwee
n the
north
ern
e
nd
of
the Bullion
Mount
a in s
a
nd
Lavi
c Lake.
Also
, a
zone
of
surfa
ce
ruptur
e a
bout
7
00
meter
s wid
e delin
ea
tes
the
comp
lex
jun
ct
io n
between
the
Lav
ic Lak
e a
nd
Bulli
on
Fa
ult
s.
eve
ral kilom
eters
north
of
the
s
outh
e rn
e
nd
of
the
ruptur
e,
the
Bullion
Fau
lt bifurcat
es
12
Seismological
Re
sea
r
ch
Letters
Vo
l
ume
71,
Numb
er
1
January
/
Febru
ary
2000
700.0
600.0
500
.0
.
9-
4
00
.0
iii
-e
Ql
3
~
300
.0
.c
0>
a:
200
.0
1
00
.0
0.0
0 .
00
North
,,
I
~
1
50
E
.!:!.
a.
:::>
Ql
1
00
'tl
iii
~
:>
50
e
.!:!.
0
·50
a.
-100
:::>
-8
iii
i
·
150
w
0. 00
North
•
~
I
~
!
··
..
Horizontal
Component
of
Slip
EPICENTER
~
l
!~
-
-
·
~-------~
·
---.,--
In
c
omp
lete
rnea
su
r
e
.
mrtn'i?'"
~
.
.
5.00
1
0.00
15.00
20.00
25.00
30.00
Al ong
Faul
t Dlstanco
(km)
Vertical
Component
of
Slip
EPICENTER
~
.
..
•
!_t
\.:.-.
.
.
__.,
.
_...
_
....
.!__~'-7-'
'---
•.
.
·
..
BULLION
MOUNTAINS
~
...
,
..
---.
5 .
00
1
0.00
15 .
00
20
.
00
25
.
00
0
30
.
00
Along
Fault
Distance
(km)
H
0 0
if
-t¥\o-
35.00
35.00
--
.
40 .
00
40 .
00
45
.
00
South
45
.
00
South
.A.
Figure
3.
Pr
e
limin
a
ry
s
lip
di
s
tribution
for
th
e
He
c
tor
Mine
e
ar
t
hquak
e fa
ult
ruptur
e
tak
en f
ro
m
fi
e
ld
mea
su
reme
nt
s.
Op
en
c
ir
cl
es
r
ep
r
ese
nt
d
ata
fr
om
th
e Wo
od
Canyon
ar
ea.
wher
e
ruptur
e o
cc
ur
s o n
two
major
paralle
l s
tr
a
nd
s (s
lip
rep
rese
nt
s a s
um
of
th
e a
ve
ra
ge
valu
es
ac
r
oss
th
e
tw
o s
tr
an
ds
). T
he
open
tri
a
ngl
es
repre
se
nt
data
fr
om a
para
ll
el
we
s
tern
s
trand
s
outhw
es
t of
th
e
main
ruptur
e
zo
n
es
at t
he
so
uth
end o f t
he
sys
tem. E
rror
b
ars
re
fl
ect
t
he
total
ra
n
ge
of
uncertainty
in
th
e s
li
p
es
timat
e
as
reported
by
I
ndividu
al
inv
es
tigat
or
s.
into
two
distin
c t
br
a nch
es
a
bout
1
km
a part.
Bo th
of
the
se
pr
eviously
ma pped
traces
ruptured
in
thi
s event
(
Fi
gure
2 ).
Th
e fau
lt
rupture
map
is prelimin
ary
a
nd
will
be
refin
ed
as
mo re c
omplet
e o
bserv
a tion
s
of
fault
slip
are
mad
e.
Slip
wa
s mea
sured
a t over
300
site
s a
long
the
m ain
rup
-
tur
e.
Thes
e
me
a
surement
s a
re co
mbined
to
fo rm
a s
lip
distri-
buti
o n
pr
o
fi
le along
th e
ruptur
e (Figur
e 3).
Wh
e re
mu
ltipl
e
fa ult
s
trands
ar
e
pr
ese
nt,
tora
l slip
va lues
at
each
site a
re
o
bt
ained b
y s
ummin
g
di splace
m e
nt
vec
to rs.
Al
t ho ugh
pr
ese
nt
d ata
are
inco
mpl
ete,
th
e dist ributi
o n o f h
o
ri
zo
n
ta
l
slip
alo ng t
he
fa
ult a
pp
ea rs
to
be rem a rka
bly sy
mm
e trical,
with
ra
th
e r a
brupt
te
rmin
a
ti
o ns o
f slip
at the e
nd
s
of
the
ruptur
e.
Alth
o ugh
rupture
a
nd
surf
ace
c rackin
g
has
been
observed ove r a leng
th
of m
ore
th
a n
40
km
, signifi
c
ant
ho r-
izo
nt
al slip
(i.e.,
mor
e tha n a few
ce
ntim
e ter
s)
occurs
over
a
le ngth
o f less
t han
35
km
.
Th
e la rges
t h
o
ri
zo
nt
al displa
ce-
Se
ismolo
gi
c
al
R
ese
a
rc
h L e
tt
ers
Vol
um
e
71,
Numb
er 1 J
a
nuary
/Fe
bru
ar
y
20
00
13
ment
measured
within
the
first
week
,
525
±
85
em
of
right-
latera
l slip,
is
located
app
roximately
4
km
sout
h
of
the
epi-
ce
nter
in
the
Bullion
Mountains.
Preliminary
results
indi
cate
an
average
slip
across
the
entire
fault
zone
of
approximately
250-300
em.
The
largest
observed
vertical
displacement
was
nearly
200
em,
but
valu
es
o f
100
em
or
less
were
more
commonly
observed.
Overall
the
sense
of
this
displa
ce
ment
component
is
not
co
nsistent,
which
is
typica
l for
strike
-slip
ruptures.
Between
the
Bullion
Mountains
and
Lavic
Lake,
there
is
consistent
west-
side
-u p
disp
lace
ment
where
the
fault
curves
northward,
producing
a
sma
ll
relea
s
ing
bend.
East-side-up
displ
ace
ment
is
observed
just
to
the
south
of
the
epicenter
wher
e
the
fault
be nds we
st-
ward
(a
compress
ive
bend)
.
TriNet:
MODERN
REAL-TIME
SEISMOLOGY
Th
e
U.S.
Geolog
ical Sur
ve
y,
Ca
ltech,
and
the
Ca
lif
o rni
a
Division
o f
Min
es
a nd
Geology
have
bee n
running
a
nd
co
n-
tinuing
to develop
TriN
e t, a state-of-the-art
broadband
and
strong-motion
seismic
network
in
southern
Ca
lifornia.
Tri
-
Net
is
a
multipurpo
se
network
designed
to
record
and
an
a-
lyz
e on-scale
ground
motions,
distribute
that
informati
on
quickly
, improve
our
und
erstanding
of
earthquakes
and
their
effects,
contribute
to
improving
building
codes
and
struc
-
tural
de
sign
,
and
facilitate
eme
rgency
respon
se
in
coo
per
a-
tion
with
other
age
ncies
. A
s
of
October
1999
there
wer
e
120
USG
S-Ca
ltech
real
-time
sta
ti o ns
online
a
nd
nea rly
200
CD
MG
s
trong
-
motion
sta
tions
for
which
da
ta
is
accessed
vi
a a
dial-up
protocol.
E
ventu
ally
,
TriNet
will
include
over
600
stations
(see
Mori
et
al.,
199
7;
Hauksson
et
al
. ,
1999).
The
He
ctor
Mine
eart
hquake
provided
a
good
mid-
pro
je
ct
test
and
eva
luation
of
the
performanc
e
ofTriNet
. Tri-
Net
is
designed
so
that
within
the fir
st
minute
following
the
re
co
rded
sha king
from
an
event,
rh e eve
nt
epicenter
, mag
ni -
tude,
a
nd
ground
motion
parameters
a
re
availab
le from
the
USGS-Ca
l
tech
component
of
the
network.
Within
minutes,
data
from
most
of
the
important
near
-so
urce
CDMG
sta-
tions
a re
reported.
A m
o re co
mpl
e
te
CDMG
contribution
is
avai
lab
le
within
approximately
thirty
minute
s.
An
initial
magnitude
a
nd
loca
ti
o n
is
obtained
within
minute
s
of
a n
event
using
th
e
real
-t
ime
station
s.
For
the
Hector
Mine
earthquake,
a
loc
at
ion
a
nd
preliminary
ML
est
im
a te
of
6.6
were
obtained
approximately
90
seco
nds
after
the
origin
time
of
the
event.
A m
ag
nitud
e
of7.0
wa
s deriv
ed
from
estimated
radi ated
energy
(Kanamori
et al.,
199
3)
approximately
30
seco
nds
later.
Loca
tio n a
nd
magnitude
estimates
for
significant
events
a
re
broadcast
via
email,
the
W
o rld
Wide
Web
,
and
pager
message
s via
the
CUBE
sys
tem
(Mori
et
al.,
1997).
The
first
CUBE
page
following
He
c
tor
Mine
, including
the
M L
6.6
value,
was
received
a
ppro
x
imat
ely
3.5
minute
s
af
ter
the
event;
an
updated
page
with
the
es
timat
e
M7.0
wa
s received
approximately
thirty
seconds
later.
One
critical
product
of
TriNet
is
"S
hakeMa
p",
which
involve
s rapid
(within
3- 5
minut
es)
generation
of
maps
dis-
playing
ground-motion
par
a mete
rs
throughout
sou
thern
Ca
lifornia.
These
param
eters
includ
e
observed
gro
und
motion
value
s
as
well
as
intensity
va
lues
derived
from
newly
developed
relationships
between
reco
rded
ground-motion
parameters
a
nd
expected
s
haking
int
ensity
(Wa
ld
et
al.,
1999a)
. Estimation
of
shaking
over
the
entire
regional
ex
tent
of
southern
California
is
o btained
via
spatia
l
int
erpolation
of
the m eas
ured
gr
ound
motions,
whi
ch are
reco
rded
on
a fairly
spa
rse,
non
unif
o rmly
space
d
netw
o rk
of
station
s. A
unif
o rm
sampling
grid
is o
btain
ed u
sing
geo
l
og
ica
ll
y base
d fre
quen
cy
a
nd
a
mplitude-dependent
site
co
rrec
tio ns
(Wald
et
a
l.
,
1999b).
The
observations
a re
a
ugm
e
nted
with
predicted
ground
motions
in
a reas
without
se ismi
c sta
tion
s, which
for
this
earthquake
include
s the e
ntir
e nea r-fault
region.
Production
of
maps
is
automatic,
trigge
red
by
a ny
si
g-
nific
a
nt
ea rthquake
in
southern
Ca
lif
o rni
a (see
Wald
et
al.
,
t
999b
fo r more
details)
.
TriNet
has
provided
funds
to
includ
e ShakeMap
as
input
d a ta
for
loss-e
stim
at
io n
too
ls
used by
the
Ca
lif
ornia
Office
of
Emergency
Services
(OES)
.
Although
ShakeMap
data
files
were
used for
rapid
loss-est
i-
matio n calcul
at
ions
for
the
He
c to r
Min
e ea rthquake,
the
result
s were
n
ot
of
signifi
ca
nt
societa
l
imp
act
because
th
e
eve
nt
occ
urred
in
an
uninh
a
bit
ed r
eg
ion
a
nd
ca used
o nly
moderat
e s
haking
in
sparsel
y
popul
ated
parts
of
the
M
ojave
Desert.
The
first
ShakeMap
for
the
He
c
tor
Min
e earthquake
was
produced
within
four
minute
s
of
the eve
nt.
Initi
ally
,
ShakeMap
g
round
motio
ns in
the nea r-so
ur
ce
region
were
est
im
ated
usin
g
ground-motion
reg
r
es
sion
from
a
po
in t
so urce
at
the
epicenter.
The
m ap was
rev
is
ed l
ate
r,
as
inf
o r-
m
at
ion
a
bout
fa
ult
dimensi
o ns
became
ava
ilabl
e (
Fi
g
ur
e 4).
Fau
lt
dimen
sion
inf
o rmati
o n was
initi
a
ll
y d e riv
ed
from
the
aftershock
di stributi
o n
and,
later
,
the
observed
s
urf
ace
trace.
Results
from
both
lin
es
of
evidence
were
no ted
to
be
co
nsis-
tent
with
a prelimin
a ry
finit
e-fa
ult
model
d et
erm
ined
within
hours
of
th e event
using
wavefor
m d
ata
from
the
TriNet
array
(D.
Dreg
er,
http
://www.seismo.berke
ley.edu/seismo/
eqw/99.10.16
_ff.html
).
SOURCE
MODELING
RESULTS
The
mainshock
hyp
o"ce
nt
e r is locate
d
approximately
2
km
east
o f
the
north
end
of
the
Bulli
o n
Fa
ult.
The
first
motion
foca
l m ec ha nism i
s po
o
rl
y co
nstr
ained
because
t he
SCS
N
st
at
ion
di
st
ribution
is sparse
in
the e
pice
ntr
al r eg
io n.
Main-
shock
mom
e
nt
ten
sor
solution
a nd
ce
ntr
o id d
ept
h d
ete
rmi
-
nations
were
performed
using
d ata
from
60
bro
ad
ba
nd
TriNet
stations,
rangin
g from
70
to
400
km
in
ep
ice
ntr
al dis
-
tances.
We use
the
C
AP
sou
rce
inv
e
rsi
o n
technique
which
utili
zes
bo th
th
e
P nl
and
surface
wave
portions
of
bro
a
db
an
d
wav
efo
rm
s to
co
ns
tr
ain
so
ur
ce
o
ri
e
nt
ation
a
nd
depth
(Z
hu
and
Heimber
ge
r,
1
996)
. T he sta
nd
a rd
so
uthern
Ca
li fornia
velocity
model
(Hadley
a
nd
Ka n
amo
ri
, 19
7 9)
is used t
o ca l-
culat
e
the
Green
's functions.
T he
CAP
technique
allow
s tim
e
shift
s between
dat
a a
nd
sy
nth
e tic
s whi
le
comput
ing
the
waveform
misfit
s so
that
the
influ
ences
of
so
ur
ce
mi
sl
ocat
io n
14
Seismological
Resear
ch
Letters
Volume
71,
Number
1
January/February
2000
TriNet
Rapid
Instrumental
Intensity
Map
fo r Hector
Mine
Earthquake
OCT
16 199
9 02:46
:45
PDT
M7.1
N34
.5956
W11
6.268
10:9108645
(site c
orr
ected)
~
60km
11
r w
PERCEIVED
N
ot
Felt
W
eak
Ligh
t
Mode
rate
Strong
Very
Strong
SHAKING
Se
vere
Viol
ent
Ex
treme
Moderate/
He
avy
H
eavy
Ve
ry H
eavy
POTENTIAL
None
None
None
V
ery Light
Li
gh
t
Modera
te
DAMAGE
PEAK
ACC
.(%g )
<.17
. 17-1.4
1.4-3.9
3.9
-9.2 9.2
-
18
18-34
34-6
5
6
5-124
>
124
PEAK
VEL.(cm
/s
)
<0.1
0.1-1.1
1.1-3.4
3 .4-8 .1
8.1-16
16-31
31-60
60
-116
> 116
~~~~~;-
----r-----r-----+------r----
-r-------+--
--
-
INSTRUMENTAL
INTENSITY
11-111
IV
v
VI
VII
VI
•
Figure
4.
Tr
iN
e!
rapid
i
nstrumen
t
al
intens
i
ty
ShakeMa
p
of
the
1
0/
1
6/99
H
ecto
r
Mine
,
California,
earth
q
uake.
Shaking
scale
is
indica
t
ed
at
bottom
of
figure.
a nd
c rusta
l h
eteroge
neiti
es
a re m
ini
mi
zed . Fo r thi
s
prel
im i-
na ry source
in
ve
rsi
on,
a do
ub
le-co
upl
e
po
i
nt
so
ur
ce
of
9 s
dur
atio n
is
ass
um
ed . Fig
ur
e 5 s
hows
th
e wave
form
mi
sfit
erro rs fo r
diff
ere
nt
so
ur
ce
de
pth
s w
he re
th
e mi
sfit
err
or
is
m
eas
ured b
y
the L2
-n
o rm
of
th
e d iffere
nce
betwee
n t
he
v
el
oc
ity
reco
rds a
nd
synt
h eti
cs,
no rm
a
li
ze
d b
y the data
va
ri-
a nce. T he
opt
imal ce
nt
ro id d
e
pth
is de term
ined
to
be 13.5
km.
(M
isfit
within
1
O
o/o
of
the o
pt
ima
l
va
lue is o
bt
ained f
or
a d
e
pth
rang
e of rou
ghly 1
2-
15
km
).
T he
opt
imal
fa
ul
t plane s
olution
h
as
a s
tr
ike
of
N29°W
a nd
a dip
of
7r
to
the east , w
ith
p
ur
e rig
ht
-lateral
mo
t ion
a nd
m o me
nt
of
3.4
X
10
26
N m
(Mw
7.0).
T he
sca
lar
mome
nt
is s
lightl
y s
ma
ll
er t
ha n the
es
tim
ate de
ri
ved
fr
o m
1
00
80
......
60
~
Ill
40
~
20
0
0
5
10
15
20
Depih
(km)
•
Figure
5.
Mis
fi
t
as
a
function
of
centroid
depth.
For
each
depth
,
the
m
echa
n
is
m
and
Mw
val
ue
are
also
plo
tt
ed.
Seismologica
l
R
esearc
h L
etters
Vo
lu
me
71,
Number
1 J
anuary/February
2000
15