OPINION
LOOKING
BACK
FROM THE
YEAR
3,000
Many
people
have
asked
me
how
I can
justify
living
with the
earthquake
threat
in
the
Los
Angeles
area.
My
answer
is
usually
that
we
have
some
reasonably
strict
building
codes
and
that
the
threat
of
earthquakes
to
life
safety
is minimized
if
our
buildings
survive
our
coming
quakes.
The
current
building
code
calls
for
buildings
to
sustain
at
most
repairable
damage
from
the
strongest shaking
that
is anticipated
with
a
10%
probability
in 50
years.
If
the
building
is
a
critical
structure,
such
as
a
hospital,
then
the
requirement
is
in-
creased
to
10%
in
100
years. Furthermore,
buildings
of
both
classes
should
not
collapse
for
the
strongest
ground
shaking
that
can
be
anticipated
at
the
location
of
the
building.
If
the
building
code
works
as
it's
supposed
to,
then
we
endure
far
greater
risks
from
other
factors
than
from
earth-
quakes.
Although
it is comforting
to
know
that
our
buildings
are
designed
so
that
they
will
not
collapse
for the
strongest
shaking
that
can
be
anticipated
at
a
site,
it is
also
disturbing
that
we
have
seen
important deficiencies
in
our
building
practices
following
each
strong
earthquake
located
close
to
our
urban
regions.
For
example,
the
1971
San
Fernando
earthquake
(Mw6.7)
re-
vealed
fundamental
deficiencies
in
the
ductility
of
reinforced
concrete
frame
buildings.
Concrete
frame
buildings
Although
it is comforting
to
know that
our
buildings
are
designed
so
that
they
will not
collapse
for
the
strongest
shaking
that
can
be anticipated
at a
site,
it is
also
disturbing
that
we
have
seen
important
deficiencies
in
our
building
practices
following each
strong
earthquake located
close
to our
urban regions.
that
seemed
perfectly
adequate
to
designers
25
years
ago
are
now
viewed
as
potential
collapse
hazards,
and
there
are
several
committees
struggling
with
the
question
of
how
to
identify
and
strengthen
these
buildings.
The
recent
Northridge
earthquake
(Mw6.7)
provides
another
disturbing
example;
cracks
were
observed
in
many
of
the
welded
connections
in
steel
frame
buildings.
Prior
to
the
earthquake,
it was
generally
believed
that
the
steel
beams
would
yield
over
many
cycles
before
any
of
these
connections
failed.
Unfortunately,
it
is
not
yet
clear
what
should be
done
to
repair
damaged
buildings
or
to
prevent
such
behavior
in
future
earthquakes
in
the
thousands
of
steel
frame
buildings
in
California.
Other
examples
are
the
generally
poor
perfor-
mance of
parking
structures
(even
new
ones)
in
the
Northridge
earthquake,
and
the
collapse
of
freeway
bridges
in
both the
Northridge
earthquake
and
the
1989
Loma
Prieta
earthquake
(Mw6.9).
It
is
not
only
the
engineering
community
that
has
re-
ceived
uncomfortable
surprises.
The
Northridge
earthquake
also
revealed deficiencies
in
our
understanding
of
earth-
The
opinions
expressed
in
this
article
are
those
of
the
author
and
do
not
necessarily
reflect
the
views
of
the
Seismological
Society
of
America.
quake
phenomena.
For
example,
how
well
can
we
recog-
nize
the
size
of earthquakes
that
may
occur
at
a given
site?
I
don't
believe
that
anyone
was
planning
on
an
earthquake
on
a south-dipping
buried
thrust
fault
beneath
the San
Femando
Valley.
About
15
years
ago,
I recall
working
on
the
seismic
design
criteria
for
a liquefied
natural
gas
facility
proposed
at
the
west
end
of
the
Santa
Barbara
Channel.
Although
there
was
abundant
evidence
of
late
Quaternary
folding
of
sedi-
mentary
rocks
in
the
area,
no
clear
evidence
could
be
iden-
tified
for
major
faults
that
cut
Ho-
locene
deposits.
We
argued
that
the
folding
process
was
likely
to
be
be-
nign,
and
we
concluded
that
a major
earthquake
was
unlikely
directly
be-
neath
the
site.
Since
that
time,
we
have
seen important
earthquakes
on
blind
faults
beneath
analogous
areas
of
folded
rocks,
such
as
Coalinga
and/or
Northridge.
Worse
yet,
consider
the
1952
Kern
County earthquake
(Mw7.5).
There
was
very
little
surface
faulting
considering
the
large
magni-
tude
of
this
earthquake.
How
can
we
exdude
the
possibility
of
similar
earth-
quakes
in
other
regions
such
as
the
Los
Angeles
or
Ventura
basins?
It
was
also
a mere
dozen
years
ago
that
I reviewed
a
safety
analysis
report
to
the
Nuclear
Regulatory
Commission
for
a
power
plant
in
coastal
Washington.
The
applicant
argued
strongly
that
the
Cascadia
subduction
zone
is
com-
pletely
aseismic.
While
there
is still
much
to
be
learned
about
the
earthquake
potential
of
the
Cascadia
subduction
zone,
I
think
few
would
still
argue
so
confidently
that
there
is
no
potential
for
large
earthquakes.
Even
if we
can
anticipate
future earthquakes,
how
well
do
we
know
the
ground
motions
that
will
result?
In
the
case
of
the
Northridge
earthquake,
ground
motions
were
gener-
ally
higher
than
acknowledged
by
building
codes,
especially
for
sites
located
above
the
northern
part
of
the
rupture
zone
where
longer period
(_>1
sec)
ground
motions
were
espe-
cially
large.
Fortunately,
taller
buildings,
which
were
vulner-
able to
this
type
of
motion,
were
not
located
in
this
region
of
intense
long-period
shaking.
Well,
maybe
it's
unfortunate
that
tall
buildings
were
not
located
in
this
region.
If they
had
been,
we
might
have
learned
some
important
lessons.
Since
the
earthquake
was
at
4:30
AM,
the
lessons
would have
been
Seismological
Research
Letters
Volume
66,
Number
2, March-April
1995
3
expensive,
but
few
lives
would
have
been
endangered.
Even
with
the
important
new
records
from
Northridge,
we
have
relatively
few
near-source
recordings
of
large
earthquakes.
We
have
all
read
intriguing
tales
of
extremely
strong
shaking
in
close
to
large
earthquakes
(objects
thrown
significant
distances).
Are
we
really
so
perceptive
that
we
understand
how
to
interpret
these
anecdotal
reports?
Do
we
really
know
how
strong
the
maximum
shaking
can
be
at
a
given
site,
even
if
we
could
anticipate
the
earthquake?
Can
you
see
where
this
discussion
is
leading?
Our
build-
ing
codes
are
supposed
to
insure
safety
for
the
worst
shaking
that
can
be
anticipated
at
a
site,
but
in
mere
decades,
we
are
changing
our
understanding
of
the
types
of
earthquakes
that
can
occur
at
a
particular
site,
the
nature
of
earthquake
ground
shaking
that
would
result,
and
of
the
corresponding
building
response
to
that
ground
motion.
How
can
we
claim
to
build
for
events
that
might
happen
in
the
next
1,000
years,
when
our
ideas
don't
hold
up
for
20
years?
("I'm
sorry,
but
the
warranty
on
that
research
expired
last
year.")
Fortunately,
writers
of
our
codes
have
used
some
common
sense
wisdom
to
circumvent
our
very
incomplete
under-
standing
of
the
physics
of
both
earth-
quakes
and
buildings.
That
is,
the
code
is
based
largely
on
past
experience.
Design
practices
that
produced
unacceptable
be-
havior
in
past
earthquakes
were
modi-
fied.
If
one
has
enough
experience
with
buildings
and
earthquakes,
then
if
one
finds
a
design
that
works,
is
it
critical
that
we
understand
why
it
works?
I
suspect
that
it
is
this
experience
factor
that
has
allowed
engineers
to
feel
"comfortable"
with
their
designs.
But
how
good
is
our
experience?
In
a
sense,
the
Mw6.7
Northridge
earthquake
is
the
largest
one
to
occur
beneath
a
modern
urban
area
of
the
US,
and
in
that
case
taller
buildings
were
not
in
the
region
of
the
strongest
long-period
shaking.
If
we
could
look
back
from
the
year
3,000,
would
we
be
surprised
to
hear
that
several
of
our
cities
had
been
visited
by
much
larger
earthquakes?
How
about
several
Mw>7.5
earth-
quakes
in
Los
Angeles,
San
Francisco,
and
Oakland?
Or
perhaps
a
Mw>_9.0
on
the
Cascadia
subduction
zone?
Have
we
learned
the
lessons
that
these
earthquakes
will
teach
us?
I
suspect
that
the
answer
is
no.
I
am
particularly
skeptical
of
our
current
understanding
of
moment-resisting-frame
buildings.
These
are
the
flexible
buildings
that
are
usually
built
with
a
rectangular
skeleton
of
either
steel
or
reinforced
concrete
columns
and
beams.
These
buildings
appear
more
susceptible
to
large
ground
displacements
than
stiffer
buildings
are.
Since
moment-re-
sisting-frame
buildings
have
not
yet
been
in
the
near-source
area
of
a
very
large
earthquake,
I
don't
think
we
can
claim
that
they
have
ever
been
field
tested.
How
large
can
the
ground
motions
be
close
to
a
Mw_>7.5
earthquake?
3
meters
in
several
seconds
seems
perfectly
reasonable.
How
about
6
meters?
Do
we
understand
what
will
happen
if
the
base
of
a
tall
structure
is
displaced
a
large
distance
in
a
short
time?
When
these
large
earthquakes
occur
beneath
our
cities,
we
may
pay
a
very
high
price
for
codes
that
are
based
on
lessons
from
smaller
earthquakes.
Frankly,
I
am
surprised
at
the
boldness
of
those
who
have
designed
tall
buildings
with
such
a
partial
understand-
ing
of
earthquakes
and
building
responses.
If
a
future
large
urban
earthquake
renders
a
large
number
of
buildings
use-
less,
or
worse,
if
a
significant
number
of
them
collapse,
then
it
is
almost
certain
that
society
will
demand
dramatic
changes
in
design
criteria.
Is
it
only
through
the
experience
of
future
large
earth-
quakes
that
we
can
learn
about
our
shortcomings?
Of
course
not;
that
is
what
basic
research
is
all
about.
More
complete
modeling
and
testing
could
tell
us
quite
a
bit
about
the
I
wonder
what
scientists
and
engineers
in
the
year
3,000
will
think
about
the
understanding
that
we
are
employing
to
design
large
buildings
and
bridges
that
will
withstand
the
inter-
vening
earthquakes.
capacity
of
buildings
to
survive
different
types
of
shaking.
We
should
use
this
re-
search
to
decide
which
building
designs
most
practically
respond
well
to
a
wide
variety
of
ground
motions,
including
ground
motions
with
large
displacements
or
long
durations.
Of
course,
support
for
this
research
has
been
a
big
problem,
and
it
will
continue
to
be
unless
we
more
clearly
state
the
fundamental
shortcom-
ings
of
current
understanding.
Perhaps
earthquake
professionals
should
declare
a
moratorium
on
the
approval
of
types
of
design
until
their
capacity
is
well
under-
stood.
For
example,
we
might
allow
only
those
buildings
for
which
we
have
the
greatest
confidence
of
their
capacity,
such
as
concrete
shear-wall
construction
with
a
height
limit
of
5
stories
(and
don't
skimp
on
the
concrete
and
rebar).
I
suspect
while
such
a
move
would
be
painful,
it
would
also
make
it
clear
that
far
more
research
is
required
to
confidently
erect
tall
buildings.
If
society
wants
more
daring
designs,
then
it
should
invest
far
more
in
research
about
the
conse-
quences
of
those
designs.
Wouldn't
it
make
sense
to
plow
some
reasonable
percentage
of
the
cost
of
a
building
(per-
haps
1%)
back
into
research
about
the
overall
performance
of
buildings?
In
1995,
I
look
back
on
arguments
made
in
the
1970's
that
now
seem
pretty
poorly
founded.
I
wonder
what
scien-
tists
and
engineers
in
the
year
3,000
will
think
about
the
understanding
that
we
are
employing
to
design
large
build-
ings
and
bridges
that
will
withstand
the
intervening
earth-
quakes.
I
think
that
the
next
time
someone
asks
how
I
can
justify
living
with
the
earthquake
threat
in
the
Los
Angeles
area,
I
will
simply
answer
by
telling
them
that
I
work
in
an
office
converted
from
a
wood-frame
house.
Besides,
I
ride
a
motorcycle
to
work.
[;~
Thomas
H.
Heaton
President
Seismological
Society
of
America
4
Seismological
Research
Letters
Volume
66,
Number
2
March-April1995