Eos,
Vol.
66,
No.
44,
October
29,
1985
Filming
Seismograms
and
Related
Materials
at
the California
Institute
of
Technology
PAGES
737-739
Judith
R.
Goodstein
Paul Roberts
Institute
Archives,
Robert
A. Millikan
Kresge
Seismological
Laboratory,
California
Memorial
Library,
California
Institute
of
Institute
of
Technology, Pasadena
Technology, Pasadena
As
part
of
the
worldwide
effort
to
create
an
international
earthquake data
bank,
the
seismology
archive
of
the
California Institute
of
Technology (Caltech)
has
been
organized,
labeled, described,
and
microfilmed.
It
includes
a wide variety of
original
records, documents,
and
printed
materials
relating
to local
and
distant
earthquakes.
The
single largest
and
most
complex
component
of
the
task has
been
the
preparation
and
microfilming
of
Caltech's
vast collection of
original
seismograms.
The
original proposal
envisioned
a modest
project
in which
a se-
lected
number
of seismographic records
at Caltech could
be
made more
general-
ly
available to
the
scientific
community.
These
single-copy
records
are
stored
at
Kresge
Laboratory
and
comprise
thousands
of
individual
photographic
sheets,
each
30
x
92
cm.
In
the
end,
we
microfilmed
both
the
Pasadena
station
records
and
those
written
at
the
six
original stations
in
the
Caltech
network.
This
task
got
underway
in
June
1981
and
was
completed
in
January
1985.
In
the
course
of
the
project,
the
staff
sorted,
arranged,
inventoried,
copied,
and
refiled
more than
276,000 records written
between
January
10, 1923
and
December
31, 1962.
The
microfilm
edition
of
the earthquake
records
at the
Seismological
Laboratory
at
Pasadena
and
at
auxiliary stations
at
Mount
Wilson,
Riverside,
Santa
Barbara,
La
Jolla,
Tinemaha,
and
Haiwee
(the
latter
two
in
the
O\t.ens
Valley) consists of
461
reels
of film.
The
film
archive
is cataloged
and
available to
researchers
in
Cal-
tech's Millikan
Library
in
Pasadena,
at the
U.S. Geological
Survey
in
Menlo
Park,
Calif.
and at
the
World
Data
Center
(National
Oceanic
and
Atmospheric Admin-
istration)
in
Boulder,
Colo.
Origins of
the
Seismology
Archive
Seismology
at
the
California
Institute
of
Technology (Caltech) arose
from an
"ar-
ranged marriage"
between
two
different
tra-
ditions: a
European
interest
in global
earth-
quakes
and
an
American concern
for
local
earthquakes
[Goodstein,
18741.
German-born
and
Giittingen
University-trained,
Beno
Gu-
tenberg
brought to
Caltech
the European tra-
dition of
viewing seismology
as
a research
tool.
Rigorously
trained
in
physics
and
math-
ematics,
he
used
earthquake
records
to
inves-
tigate
the
physical
properties
and
structure
of
the
earth's interior. Earthquake instruments
installed
in
seismological
stations
around
the
world
provided the
data
for
his analysis.
For
Gutenberg,
the
globe
was
his scientific
labora-
tory.
However,
Gutenberg's American
col-
leagues,
Harry
Oscar
Wood
included,
took
a
much
more
pragmatic
view
of
the
world.
Few
in
number
and
concentrated
in
California,
the
American
seismological
community
saw
their
research
in
terms
of
finding a
solution
to
the
"California
problem,"
as they
called
re-
gional
earthquakes.
To
a seismologist like
Wood, a veteran
of
the
San Francisco
quake
and
the
expert on
the
seismic
history of
Cali-
fornia, Gutenberg's
global
problem
shrank
to
the
size
of
southern
Califhrnia.
In bringing these
two
men
together
in
Pas-
adena
in
1930, Robert
A.
Millikan,
the
head
of Caltech, set
in
motion a chain
of
events
1-
Cover.
Planet
Earth:
Twenty-five years
ago,
scientists
from around
the
world
joined
forces
in
a venture
of
unprecedent-
ed
scale
aimed at
achieving
a major
ad-
vance in
our
knowledge
of
the
earth.
The
International
Geophysical Year
(IGY)
took
place in 1957-1958
and
was
successful
be-
yond
all
hopes
of
the
participants.
"Planet
Earth"
is an
upcoming
major
new PBS
prime-time
series
and
television
course
that
will
explore the
multifaceted revolu-
tion
in
scientific
thought
that
came
in
the
wake
of
IGY.
Special previews of this se-
ries
will
be
featured
at
the 1983
AGU
Fall
Meeting
in
San Francisco,
Calif.,
Decem-
ber
8-13.
that
was
ultimately
to
lead
Charles
Richter to
develop
the
first
earthquake magnitude
scale.
Charles
Richter took
his
Ph.D.
in
theoreti-
cal physics
at Caltech.
He
became Wood's
as-
sistant
in
1927, 3
years
before Gutenberg's
arrival.
He
spent the
next
several years
mea-
suring and
filing
seismograms.
Much of
the
work
was
tedious
and
mechanical,
and
aside
from
Wood,
Richter's contacts
with seismolo-
gists
remained
limited.
A theoretical
physicist
who fell
into
seismology
more
or
less
acciden-
tally,
Richter
desperately
needed
a scientific
mentor. Gutenberg
fit
the
bill;
indeed,
if
Gu-
tenberg
had not come
to
Richter,
Richter
would
have
gone
to
Gutenberg.
As
things
turned
out,
they
spent
30 years
under
the
same
academic
roof.
American
interest in
seismology
got
its big-
gest boost
from the
1906
San Francisco
earth-
quake.
At
that time,
no
American university
boasted
a department
of seismology,
let
alone
any
professional
seismologists.
The
quake
triggered,
among other
things,
the
birth
of
the
Seismological Society
of
America,
an or-
ganization
dedicated
to
stimulating interest
in
geophysical
matters
in
general
and
earth-
quake problems
in
particular.
Meanwhile,
at
the
state level,
a commission
under
the
direc-
tion
of
Andrew
C.
Lawson,
a Berkeley geolo-
gist,
investigated
the
tremblor
itself.
The
Car-
negie
Institution
of
Washington, a
private
philanthropic
foundation,
supplied
the
neces-
sary
funds
when
the
Sacramento,
Calif., state
legislators,
under
pressure
from
the
business
community.
refused to
do
so.
Lawson
tapped Wood,
who
was
then
an
in-
structor
in
the
University
of
California,
Berkeley,
Geology
Department, to
study
in
detail
the
extent
and
nature
of
the
earth-
quake damage
within
the
city
itself.
Wood
went
into
the
exercise
a field
geologist
and
came
out
a seismologist.
It was
Wood who
brought
seismology
to
southern
California.
His
campaign
began
in
1916
with
the
publication
of two
papers.
Stressing
the
importance
of
taking a
regional
approach
to
the
study
of
Local
earthquakes,
he
suggested
that
the
plan
be
tested
on
a
modest
scale
in
southern
California.
Wood
singled
out
southern
California
for
two
reasons:
because
the
region
had
no
re-
cording
instruments,
and
because
he
expect-
ed
the next large
earthquake to
occur
there.
The
1857
earthquake
along the
San
Andreas
fault
had
been
the
last
great
shock
in
south-
ern
California.
His
research
program
also
stressed
the
need
for
a new
generation
of
instruments:
there
could
be
no
hope
of
measuring
short-
I
Fig.
1.
Schematic
drawing
of
the
Wood-Anderson
torsion
seismometer.
0096-394118516644-0737$1.00
Copyright
1985
by
the American
Geophysical
Union
Eos,
Vol.
66,
No.
44,
October
29, 1985
period
local
earthquakes
with
instruments
de-
signed
to
measure
long-period
distant
earth-
quakes.
Finally,
he
emphasized
the impor-
tance of
field
work,
in particular,
to locate
weak
shocks. Like
most
of his
contemporar-
ies, Wood
was
convinced
that
if geologists
could identify the
active
faults associated
with
weak
shocks, they could
then
"deduce.
.
.the
places
where
strong
shocks
are
to
originate,
considerably
in
advance of
their
advent,''
[Woodr,
19161.
Big
shocks, in
other
words, fol-
low
weak
shocks.
He
believed
that,
in time,
it
would be possible
to make, in
his
words,
a
"generalized prediction"
of
when
and
where
to expect
the
next
big
quake.
The
compelling reason
for
setting
up
branch
seismological
stations
was
therefore
to
detect
and
register
the
weak
shocks systemati-
cally.
Yet
the routine
registration of
hun-
dreds and hundreds
of
small
California
earthquakes did
not confirm
Wood's
hypoth-
esis.
Nevertheless,
the lure
of that
elusive
idea drives
seismological
research to this day.
Like
Lawson
before him,
Wood
ultimately
found a patron
in
the Carnegie Institution.
Southern
California's first
seismological
pro-
gram
began
operation
in Pasadena
in June
1921,
under
Wood's
direction.
For
the
next
6
years,
Wood
ran the
project
from
an
office
at
the
Mount
Wilson
Observatory, located
a
short
distance
from
the
Caltech
campus.
To
succeed,
the
project
needed
the
right
instrument
for
recording
nearby
earthquakes.
Fortune
favored
Wood
in
the
person
of
John
Anderson,
one
of
Mount
Wilson's
ablest as-
tronomers.
As
part
of
the
school's
defense ef-
fort
during
World
War
I,
Anderson had
worked
on submarine
detection
instruments
sensitive
enough
to record
short
vibrations.
The
apparatus made
use of
the
piezoelectric
properties
of crystals,
including
Rochelle salt
and
quartz,
to
produce
and
detect supersonic
waves.
Anderson's
war-honed
skills
matched
Wood's
peacetime needs.
The
Wood-Ander-
son collaboration began immediately
after
Wood
had
settled
into
his office.
To
do
what Wood
wanted it
to
do,
the
in-
strument
had
to
be
sensitive
enough to
re-
cord
shocks
having
a period
varying
from
0.5
to 2.0
s. Seismometers
designed
for
recording
distant earthquakes
typically
have
longer
pe-
riod response.
In the
case
of Berkeley's sta-
tion,
the
instruments
in
use
had periods of
15
and
6 s, respectively.
In
the
early
1920's,
in-
struments on the
Atlantic
seaboard
could
measure
the
time
and
place of California
shocks
with
greater
precision
than
could com-
parable
instruments
located in California.
In the autumn
of 1922,
after
several false
starts,
Wood
and
Anderson had
designed
a
reliable, compact,
portable instrument
that,
when
placed vertically,
consistently
recorded
the
east-west
and
north-south components
of
the
earth's motion
during
an earthquake.
In
practice,
the
Wood-Anderson torsion
seis-
mometer
was
an
ideal
instrument
for
record-
ing
the
earth's
horizontal movements
over a
short
distance
during an
earthquake;
it
proved
less
successful
for
recording
the
earth's up-and-down
motions. Shortly
before
Gutenberg arrived
to
take
up
his
duties
as
professor of geophysics
at
Caltech in 1930,
Hugo
Benioff,
Wood's
assistant, designed
and
built
a vertical seismometer
to
meet
Wood's
needs. Routine
recording
of
local
shocks
with
Benioffs instrument
began
in
1931,
by
which
time
Wood
was
predicting
the
new
vertical
component
seismometer
would
surpass
any
existing vertical
then
in
use
for the
registra-
tion of
distant earthquakes
as
well.
Both
the
Wood-Anderson
and
Benioff
instruments
have since become
standard
equipment
in
seismic
stations
around the
world.
All
seismographs consist of
a damped
me-
chanical oscillator of some type,
along
with
a
mechanical, optical,
or
electromagnetic
re-
cording apparatus.
In the
case of
the
Wood-
Anderson instrument (Figure
I),
a perma-
nent
magnet provides critical
damping
of
the
pendulum
motion.
The
pendulum
mass
is
mounted
on
a taut
wire suspension
and ro-
tates about
the
wire
against
the
restoring
force of torsion.
This
rotation
is optically
magnified
and
photographically
recorded
by
means of
a mirror
attached
to the pendulum
and the
recording
drum.
Benioffs vertical
component
seismometer
(Figure
2)
works
on
the
same principle as
the
telephone transmitter. Pendulum
movement
is converted
into
electric
current
by
means
of
Eos,
Vol.
66,
No.
44,
October
29,
1985
a variable
reluctance
transducer.
A galvanom-
seismic
cycle
of
the
local
shocks
and
the re-
Seismology
Records
in
eter,
activated
by
this
current,
records the
up-
currence
time of
large earthquakes requires
and-down
com~onent
of
ground
motion.
seismicity
data over a
long
~eriod
of
time.
Microform
The
first
~dod-AnderGn
instrumental
records
were written
in
December
1922;
the
first
extant
records
date
from
mid-January
1923. Ironically,
the Wood-Anderson
torsion
seismometer
did more than
its
creators
in-
tended.
Wood
had
wanted a
short-period
in-
strument to
register
local
earthquakes, but
when
the instrument
was
put
to
the
test
in
1923,
he
discovered
that
it also
registered
the
first
phases of
distant
earthquakes.
Wood
had
unwittingly
altered the
course
of
his own
pro-
gram.
By
the spring
of
1924
the experimental
torsion
seismometers
installed
in
the
base-
ment
of
the
observatory
office
and
the
phys-
ics
building
on
the
Caltech
campus
had re-
corded
dozens
of
earthquakes, near
and far.
including
the
initial
short-period
phases
of
the devastating
Japanese earthquake
of
Sep-
tember
1,
1923.
The
fact
that
Wood
had
re-
corded
this
event
on
an instrument
designed
to
register
local
earthquakes
was
not
lost
on
seismologists
elsewhere.
When
Gutenberg
heard the
news
in
Germany,
from
a colleague
who
had
attended
an
international
gathering
of geophysicists
in
Xladrid,
Spain,
he
held
up
the publication
of his book
on
the
fundamen-
tals
of seismology
long
enough
to
insert a dia-
gram
of
the
apparatus.
By
the
end
of
the
1920's,
13
cities
in
the
U.S.
and one
overseas
boasted
Wood-Anderson
instruments.
In
1925, Caltech
started a
geology
pro-
gram.
The
follo~ving
year,
Slillikan
forrnally
invited
the
Carnegie Institution
to
conduct
its
earthquake
research
in
the
institute's
new
Seismology
Laboratory,
which
was
located
in
the
foothills
of
the
San Rafael
hlountains, a
short
drive
from the campus.
In
January
1927, Wood left
his
temporary quarters
at
the
observatory
office
and
moved
into
the
new
building.
The
time
to
go
earthquake
hunting
in
earnest
had
begun.
By
1929,
six
outlying
stations,
all
within
a 300-mi
(480-km)
radius
of
the
central
station
in
Pasadena
were
in place
and
working.
Each
station
was
equipped
with
a pair
of
horizontal
compo-
nent
torsion
instruments and recording
drums,
as
well
as
with
radio-timing
equip-
ment.
Records
were
sent
\\.eekly
to
Pasadena
for photographic
processing,
registration,
and interpretation.
Historical
Seismogram
Filming
Project
Why
develop a
historical file
of seismo-
grams?
Aside
from
the
practical
consider-
ation that
there
is always
the
danger
of fire
destroying
the
present
single copies,
there
is
no
doubt
that
the
records
are
scientifically
important.
One
reason, certainly,
is that
the
art and
science of
interpreting
such records
has
improved
considerably
since
the end
of
the
last
century,
when
instrumental
data
first
came
into
existence.
There
is still
much
to
be
learned
about
the
physics
of
the
earth and
of
earthquakes
from
analyzing old
seismograms.
Consider
the
problem
of
earthquake
pre-
diction
and
the
assessment of
risk
due
to
large
earthquakes.
Progress
in
this
area
de-
pends,
in
part,
on
knowledge
of seismicity
rates
and
historical
changes
of seismicity in
a
certain region.
The
understanding
of
the
This
applies
to
distant
eaGkquakes
also.
Pasa-
dena's historical
file
is important
here
be-
cause
it is a continuous
long-term
record
for
both kinds of
events.
Since
the
seismic
cycle
itself
is a long-term
process,
and
since
no
two
earthquakes
are
ever
identical,
the need
to
consult
old
records remains
vital.
Over
the
last
several
decades, researchers
at Hokkaido
University (Japan),
the
Universi-
ty
of
Tokyo,
and
the
Japan
Meteorological
Agency,
as well
as
at
Stanford
University
(Stanford, Calif.), Caltech,
and
other
places,
have used
historical
seismograms
to
attack
a
wide
range of
geophysical
problen~s.
Such
studies
include
accuracy
of
earthquake
cata-
logs, seismic
gaps,
global
and
regional
seis-
micity,
tsunami
earthquakes,
some
previously
unstudied great
earthquakes
around
the
world,
and patterns
of
foreshocks
and
after-
shocks associated with
large
earthquakes.
In
the
absence of field work
or
a good
local
network,
seismologists
have
found other
ways
to
expand the data
base
for
regions
that
have
strong earthquakes. For example,
distant
earthquake records can
be
used to
supply the
fault
orientation,
rupture
mechanics,
and
re-
lated
information
through
the
use
of
a com-
puter
modeling
technique
known
as
"synthet-
ic seismology."
This
is one
way
in
which
re-
searchers
utilize
Caltech's
distant
earthquake
seismogranls.
Still
another
technique
involves
the
use
of
unpublished
primary accounts
in
archival
repositories, such
as
diaries
and
re-
ports,
to document
the
effects of
great earth-
quakes
in
the
past
and
prepare comprehen-
sive
catalogs.
Nuclear
test
detection
has
also
focused
at-
tention
on
seismology
archives.
The
problem
of
discriminating
earthquakes from
explo-
sions in
connection
with
the nuclear
test
ban
treaty
led
to
improvements
in
the
worldwide
network of
seismic
stations.
Locally,
many
us-
ers
of Caltech's historical
file seek
out
records
of
nuclear
tests,
both
ours
and
those
of
other
nations.
In
their day, Gutenberg
and
Richter
used
nuclear
explosion
records
for
scientific
reasons:
they
measured
their
size
and
the
travel
time of
the
seismic waves.
They are
still
used
for
this
purpose
today.
Caltech's historical file of
seismograms has
been
filmed
as
part
of
an
international
pro-
gram
to
preserve
records
written
prior to
the
creation
in
1963
of
the
World-Wide
Network
of
Standardized Seismographs
(WWSSS)
and
to make
them
more
accessible
to
the
scientific
community.
The
World
Data
Center
A for
Solid
Earth
Geophysics
in
Boulder,
Colo.,
ad-
ministers
the
program
in
the United States,
working
with
the
International
Association
of
Seismologists
and
Ph~sicists
of
the Earth's
In-
terior
(IASPEI)
and the
United Nations
Edu-
cational,
Scientific,
and
Cultural Organization
(UNESCO)
in
other
countries.
Financial
sup-
port for the filming
in
the
United States
has
been provided
by
the
U.S.
Geological
Survey,
the
National Oceanic
and
Atn~ospheric
Ad-
ministration
(NOAA), and
the
Nuclear Regu-
latory
Con~mission.
The
35,235
seismograms
recorded
at
Slount Hamilton,
Calif.,
bet\veen
191
1 and
1962
have also been microfilmed
as
part
of
this
project,
as have
seismograph
sta-
tion
records
from
College, Alaska;
Honolulu,
Hawaii;
and
the
University of
Tokyo.
The
long-term
goal
is to
film
several
million
pre-
WWNSS
seismograms.
Caltech's
seismology
archive includes
a
wide variety
of
original records,
documents,
and printed
materials
relating
to
local
and
distant
earthquakes.
In
1979,
the
Institute
Archives
prepared,
labeled, described,
and
filmed
a group
of
published and unpublished
items,
including the
Bulletin
of
the
CIT
Seismo-
logical
Laboratotg
(Pasadena
and
Auxiliary
Sta-
tions),
1931-1968,
various station
clock
cor-
rections,
Beno
Gutenberg's
annotated
copy
of
the
International
Seismologiral
Summaq,
1918-
1942,
and the
original Gutenberg-Richter
worksheets
for
Seismicitj
of
the
Earth
(Prince-
ton
University
Press,
Princeton,
N.
J.,
1954).
The
notepads, more than
100
in
all,
include
calculations
and data relating
to
the
magni-
tude
scales
used
by
the
two
men
ill
tl~eir
cata-
log.
(See
Goodstein
et
al.
[1980]
for a
more de-
tailed
description
of
this project.)
Since
then,
we
have
concentrated
on
film-
ing the original
documentation on earth-
quakes registered at
the
Seismological
Labo-
ratory
at Pasadena
and
at
auxiliary stations
at
Mount Wilson, Riverside,
Santa
Barbara,
La
Jolla,
Tinemaha,
and
Haiwee.
In
1981,
we
completed the
microfilm
publication of
the
phase
cards compiled at
the
laboratory
and
at
the
auxiliary
stations belonging
to the south-
ern
California
network
of
seismological
sta-
tions.
There
are
133
rolls
of film,
which
cover
data
from
April
1.5,
1927 to December
31,
1969.
We
also
prepared
a
nlicrofilm
index
of
the
collection.
In
addition
to
the phase cards, the contents
of
five
loose-leaf
binders
that
were located
in
the
laboratory's
measuring
room
were
also
filmed.
This
material
is contained
on
a sepa-
rate
roll
of
film
marked
"Richter Notebooks:
Local
Shocks;
Long Beach." Binder
B
is con-
cerned
exclusively with
the
Long
Beach
shock
of March
10, 1933,
and
contains
graphs and
tabulations of
readings
from
all
stations
re-
cording the
earthquake
and
its
aftershocks,
March
1933-June 1936.
The
other
binders
contain
material relating
to
instruments and
stations
in
the
1930's,
tabulation
of local
shocks
between
October
1926
and
December
1930,
contemporary
accounts
of local
and
dis-
tant
shocks between
1933
and
1935,
and
mis-
cellaneous tables,
news
reports, and
geologi-
cal
notes.
Some
General
Information
About
the Phase
Cards
The
microfilm
edition
of
the
phase
cards
mirrors
the arrangement
of
the
original
cards
in
the
shock
file. Every
card
was
filmed.
Re-
marks,
diagrams,
calculations,
and other
in-
formation noted
on the
reverse
of
the
card
were
also
filmed.
The
arrangement
of
the
shock file,
com-
posed of
guide and
file
cards,
was
established
by
C.
F.
Richter
in
1929,
and
while
changes
in
the
earthquake
measuring routine
have
oc-
curred
over
the
years,
the
phase
card
layout
remains largely
intact.
The
cards
themselves
are
filed
in
chronological
order
within
each
drawer.
Each shock
is represented
by
a pri-
mary
guide
card,
followed
by
a series
of
col-
or-coded
file
cards.
The
primary guide card
has
a center,
right,
or
left
tab.
Center tab
cards contain information
about
local
shocks;
right-hand
tab
cards about
teleseismic
shocks.