Bulletin o f the Seismological Society o f America. Vo l . 5 1 , N o . 2 , p p . 1 7 5 - 1 8 9 .

Ap r i l , 1961
I NT E GRAT E D VELOCI TY AND DI S P L ACE ME NT OF
STRONG EARTHQUAKE GROUND MOTI ON
BY G. V. BERG AND G. W. HOTJSNEI~
ABSTRACT
The earthquake ground accelerations recorded at E1 Centro, 30 December 1934; E1 Centro, 18
May 1940; Olympia, Washington, 13 April 1949; and Taft, California, 21 July 1952, have been
integrated with a digital computer to obtain at each location the three components of ground
velocity and ground displacement. Maximum horizontal displacements of 10 and 20 inches are
indicated, and maximum vertical displacements of 5 inches are indicated.
St r ong gr ound mot i ons are usual l y r ecor ded by accel eromet ers since t he gr ound
accel erat i on is t he qua nt i t y most per t i nent t o engi neeri ng studies; however, for
some engi neeri ng purposes a knowl edge of gr ound di spl acement s dur i ng st rong
ear t hquakes is i mpor t ant . I n addi t i on t o accel eromet ers (period 0.06 see) t he Uni t ed
St at es Coast and Geodet i c Sur vey mai nt ai ns a number of Car der di spl acement
met er s (period 2.5 sec) and a few di spl acement met er s havi ng a nat ur al peri od of
10 seconds. The nat ur al peri od of t he Car der di spl acement met er is much t oo shor t
t o give an accur at e measur e of t he gr ound di spl acement duri ng st rong gr ound mo-
t i on, 1 and even t he 10-second peri od i nst r ument pr obabl y does not give accur at e
measur ement s of st rong gr ound mot i on. The difficulties i nvol ved in installing and
mai nt ai ni ng a net wor k of di spl acement met er s havi ng peri ods sufficiently long t o
r ecor d accur at el y st rong gr ound mot i ons are t r ul y formi dabl e. The l at e Ha r r y
Wood, of t he Seismological Labor at or y at t he Cal i forni a I ns t i t ut e of Technol ogy,
designed and had const r uct ed a pai r of hori zont al pendul ums t ha t pr esumabl y
woul d have accompl i shed sat i sf act or y measur ement s but t hei r size and wei ght
pr ecl uded t hei r i nst al l at i on in t he basement s of or di nar y commerci al buildings
where t he aecel eromet ers are installed.
Al t hough gr ound velocities and di spl acement s are not of as much i nt erest t o en-
gineers as gr ound accel erat i ons t her e is still i nt erest in knowi ng t he magni t udes
i nvol ved and, t herefore, in t he past t her e have been at t empt s at i nt egr at i ng t he
accel erat i ons t o obt ai n t hese quant i t i es. ~,3 When done graphi cal l y or numer i cal l y
t he sat i sf act or y i nt egr at i on of t he accel erograms is ve r y difficult. The chief difficulty
arises f r om t he f act t ha t t he t r ue axis (zero accel erat i on) of t he accel erogram is not
known and hence it mus t be appr oxi mat el y l ocat ed by t ri al and error. Recent l y
when t he t hr ee component s of gr ound accel erat i on for t he four st rongest r ecor ded
gr ound mot i ons were put on punch cards, it was a r el at i vel y simple ma t t e r t o do
t he i nt egr at i ons for t he velocities and di spl acement s.
The accel erograms were put on punch cards as follows. Fi r st t he t r ace was ap-
pr oxi mat ed as f ai t hf ul l y as could r easonabl y be accompl i shed wi t h a series of st r ai ght
Manuscript received for publication August 27, 1960, and, as revised, October 13, 1960.
1 Housner, G. W. and D. E. Hudson, "The Port Hueneme Earthquake of March 18, 1957,"
Bull. Seism. Soc. Am., 48:163-168 (1958).
2 Neumann, F., United States Earthquakes, 1940, (Serial No. 647), U. S. Department of Com-
merce, 1942.
3 I-Iousner, G. W., "Ground Displacement Computed From Strong-Motion Accelerograms,"
Bull. Seism. Soc. Am., 37:299-305 (1947).
175
176 BULLETIN OF THE SEISMOLOGICAL SOCIETY OF AMERICA
line segments. A temporary straight base line was drawn on the accelerograms and
the coordinates of the intersection points of successive line segments, measured
from the base line using some convenient scale of measurement, were recorded on
punch cards.
It was assumed that the recorded initial ground velocity and ground displacement
were zero. It was further assumed that in the true time-acceleration coordinates
the equation of the temporary base line would take the form Co -~ clt ~- c2t 2, where
the constants Co, cl, and c2 would have the values that would make the mean
square computed ground velocity a minimum. The punch cards were processed by
computer to introduce appropriate scale factors for time and acceleration, to find
the equation of the base line meeting the above conditions, and to adjust the ac-
celeration readings accordingly. The computer produced a deck of scaled and cor-
rected punch cards as its output.
The integrations were performed in the following manner. Let h be an interval
of time such that the ground acceleration is a linear function of time in the interval
(t, t -t- h), let ao, v0, and x0 be the values of ground acceleration, velocity, and
displacement at time t, and let a~ be the ground acceleration at time t -t- h. The
ground velocity and displacement at time t -t- h are then
vl = v0 + h ( a0 -4- al)
h 2
xl = xo A- hvo -t- ~ (2a0 A- al).
By repeated use of these formulas one can advance step by step through the solu-
tion. No truncation error occurs, and roundoff error is negligible.
All computation was performed on the IBM 704 computer at The University
of Michigan Computing Center.
The question of accuracy of the computed velocities and displacements is asso-
ciated with the accuracy of the accelerometer as well as the accuracy of the dig-
itized data in representing the smooth accelerogram. Comparative measurements
made with a USCGS accelerometer and an instrument of much more extended
frequency range indicate that the USCGS instrument gives a quite accurate repre-
sentation of earthquake ground acceleration. 4 There would appear to be no reason
why the integrated velocities and displacements should be grossly in error, but it
is not possible to make an assessment of the accuracy of the results. The three com-
ponents of acceleration were integrated to obtain the velocities and displacements
shown in the accompanying diagrams for the following records:
E1 Centro, California, 18 May 1940 (figs. 1, 2, and 3)
E1 Centro, California, 30 December 1934 (figs. 4, 5, and 6)
Olympia, Washington, 13 April 1949 (figs. 7, 8, and 9)
Taft, California, 21 July 1952 (figs. 10, 11, and 12)
Pasadena, California, 21 July 1952 (fig. 13)
Specific comments on the integrated curves are as follows:
4 Hudson, D. E. , J. L. Alford, and G. W. Housner, "Measured Response of a St ruct ure to an
Expl osi ve-Generat ed Ground Shock, " Bull. Seism. Soc. Am., 44:513-527 (1954).
STRONG EARTHQUAKE GROUND MOTI ON 177
E1 Centro, 18 May 1940. This shock has been given a Richter Magnitude of 7.0
(formerly 6.7). The instrument was somewhat less than 10 miles southwesterly of
the trace of the fault which showed relative surface displacement over a length of
approximately forty or fifty miles, with a maximum relative displacement of ap-
2 0
U
ID
.N
j . .
>
O
15
I0
5
5
10
15 0
WEST
I
TIME, Seconds
Z
W
:E
bJ
_J
(1.
u)
r~
Z
1 5
I 0
5
WEST
|
o F
0
EAST
t
TIME, Seconds
FI G. 1. E1 Ce n t r o , 1940, E - W c o mp o n e n t .
3O
178 B UL L E T I N OF T HE S E I S MOL OGI C AL S OCI E T Y OF AME R I C A
proximately twelve feet. The center of the slipped length of fault was estimated t o
be approximately 25 t o 30 miles southeasterly of the instrument. The integrated
displacements indicate permanent displacements of approximately 9 inches to t he
we s t and 7 inches t o the north. This is consistent in direction with the observed
15'
I O
o , 5 .
I 0
20.
5,
t / )
4 )
0
..Re O.
Z
bJ
_.1
a.
u)
c~
z
t o ,
( ~
(.9
NORTH
3 0
TIME, Seconds
SO~TH
I
' 5 0 ~ , b ; 5 ~
T I M E , Seconds
FIG. 2. E1 Cent r o, 1940, N- S component .
50
STRONG EARTHQUAKE GROUND MOTION 179
f aul t slipping. The r esul t ant per manent di spl acement of 11 t o 12 inches is not un-
reasonabl e in vi ew of t he 6 f eet experi enced at t he cent er of t he surface t race. A
per manent downwar d di spl acement of appr oxi mat el y 2 inches is also i ndi cat ed.
15
, 0
-
> 5
a
z
0 I0
15
- DOWN -
" ~ _ ~ ~"~_ _J 11 ~ V V "" " ~ - V V " - " Y " ~ v , , ,' ~
VV, l l V- -ij,w
- U P -
T I ME, S e c o n d s
N)
G
t . -
u 5'
c:
I1.
m_
r~
Z
D 5
0
n*
(D
I0
- DOWN-
- UP -
15 2 0 25 3 0
T I M E , S e c o n d s
:FzG. 3. E1 Centro, 1940, vertical component.
180 BULLETI N OF THE SEI SMOLOGI CAL SOCI ] ~TY OF AMERI CA
El Centro, 30 December 1934. This shock of Magni t ude 6.5 had a center approxi-
mat el y 35 miles south of the instrument. The slipping along the fault did not extend
to the surface. The east-west component of the integrated motion does not indicate
-

W
>
Q
Z
0
n,"
(.0
15
I O
O ~
5
I0
15 0
SOUTH
I
I
N O R T H
T I ME , Seconds
IS
( n
4 )
e -
~o
z
w
N 5
J
Q.
Q
Q
z 0
0
S
0
S O ~ T H
I
N O R T H
TI ME, Seconds
FIG. 4. E1 Centro, 1934, N-S component.
STRONG EARTHQUAKE GROUND MOTION 181
a ny pe r ma ne nt di spl acement . The nor t h- s out h component woul d ~ppear t o indi-
cat e a pe r ma ne nt di spl acement of 3 or 4 inches, but t he st ep-l i ke appear ance of t he
vel oci t y cur ve i ndi cat es t ha t per haps t he i ns t r ument was not behavi ng pr oper l y.
15
i :
I 0 15 2O
TI NE, Seconds
Z5
I 0 -
m
o
Z
w
. J
o.
(n
,m
z
:::) 5,
0
n~
(.9
A
EAST
T I M E , Seconds
FIO. 5. E1 Centro, 1934, E-W component.
182 BULLETIN OF THE SEISMOLOGICAL SOCIETY OF AMERICA
In the early days of the accelerometer certain difficulties were encountered wi t h
the suspension system and shifting baselines.
15
~ D
o l
D
J
( . I
c 5
0
E ]
Ld
> 5
Z
: 3
0 I0
n~
(.9
15
- DOWN-
- U P -
: - : : : ;
5 I 0 15 2 0 25
TI ME, Seconds
t / /
Q . )

~ 5
W 0
Q
Z 5
0
O~
I0
- DOWN-
- U P -
TI ME, Seconds
FzG. 6. E1 Cent ro, 1934, vert i cal component .
STRONG EARTHQUAKE GROUND MOTION 183
Olympia, 13 April 1949. This shock of Magni t ude 7.1 centered approximately
45 miles north-northeast of the instrument, and the hypocenter was located approxi-
15
.)
c
I - -
9
w
>
z
0
I0
5
0
5 ,
I 0.
, ~ ~ o
TIME, Seconds
30
5
~' O.
(J
lad
IaJ
.3
<I
. - I
CL
( D
Q I 0' "
Z
::~
0
1 5
0
A
N- 8Oe- E
I
i
S-8Oo-W
, b , ~ 2 ' o
TIME, Seconds
FzG. 7. Ol ympi a, 1949, N- 80- E c ompone nt .
N-8OO-E
S- 8Oe- W
30
184 B U L L E T I N OF THE SEISMOLOGICAL SOCIETY OF AMERICA
mately 45 miles beneath the surface of the ground. The north-south component of
computed ground displacement indicates no permanent displacement but the other
15'
4 )
"~ ,o.
e -
0 o
q
> 5
Q
z
~ I0
0
15
0
$
I
S- l Oe- E
t
30
TIME, Seconds
W
5
N- I O- W
0
I 0-
I
S- I Oe- E
t
TI ME, Seconds
30
FIG. 8. Olympia, 1949, N-10-W component .
STRONG EARTHQUAKE GROUND MOTI ON 1 8 5
horizontal component indicates perhaps 5 inches of permanent displacement. The
vertical component also indicates some upward displacement.
15-
I 0,
( n
0
z
( . 9
15
- D O W N -
- k i P -
TIME, Seconds
5 " '
~)
o - D O W N -
0
c
I--
z
w
r ~
Z
:D
0
o~
I0
O
- UP -
TIME, Seconds
FI G. 9. Ol y mp i a , 1949, v e r t i c a l c o mp o n e n t .
186 B UL L E T I N OF T HE S E I S MOL OGI C AL S OCI E T Y OF AME R I C A
Taft, 21 J ul y 1952. This shock of Magnitude 7.7 centered approximately 40 miles
east-southeast of t he instrument. The White Wolf fault, on which t he shock orig-
inated, is perpendicular to t he San Andreas fault and t he slipping was predominantly
15
Q )
m
W
~ 5
0
Z
:~ I 0
0
I
S-69e-E
'50 ~ ,~ ,~ ~
TIME, Seconds
K)
ttJ
.A
G.
Z
::)
O
fig
(.9
N-69e-W
5 '
0 ' '
5
' % s ~ ,~ ~ ~
T I M E , S e c o n d s
FIG. 10. Taf t , 1952, N-69-W component .
STRONG EARTHQUAKE GROUND MOTION 187
vertical rat her t han t he horizontal slipping usually associated with destructive
California earthquakes. No permanent horizontal ground displacement is indicated
but a small permanent downward displacement is indicated. The nat ure of t he
15
(5
41)
U )
G )
c 5
>:
o -
o ~
Ld
O
S-21o-W
I
!
N- 21* - E
t
TIME, Seconds
30
5 S-21-W
4 ) I
i o
N 5
!
I 0 N' 21"- E
t
o ~ i b , ~ z ' o
TIME, Seconds
FIG. 11. Ta f t , 1952, N- 21- E component .
~ 3o
188 BULLETI N OF THE SEI SMOLOGI CAL SOCI ETY OF AMERI CA
f aul t slip is refl ect ed in t he t r ansi ent ver t i cal gr ound di spl acement whi ch is rela-
t i vel y l arger t ha n for t he ot her t hr ee ear t hquakes. A per manent downwar d dis-
pl acement woul d be consi st ent wi t h t he nat ur e of t he faul t i ng.
15'
U
I0~
~ 5
F
0
.~ I 0
15 0
- DOWN-
, . A ~ ' q ~ / I .,A ,., , . ~, _ , t , ~ J ~ ^ . ~ . A ~ ~^A ^
- UP -
I , , ! , , ~
TI ME, Seconds
I0
I n
l i d
r -
u 5
_c
z
hi
W
~o
. J
m_
a
Q
o
I 0
- U P -
T I M E , S e c o n d s
FzG. 12. Taft, 1952, vertical component.
STRONG EARTHQUAKE GROUND MOTION 189
SOUTH
!
w
q)
.1=
o
Z
bJ
:E
bJ
0
<[ I
. J
Q.
U)
C~
a
z
0
n*
(.9
;R ATED
3'
12 14 16 18 2 0 2 2 2 4 2 6 2 8 3 0 3 2 3 4 3 6
T I ME , Seconds
FIG. 13. Pasadena, 1952, N-S component.
Pasadena, 21 July 1952. The accelerometer on t he campus of the California In-
st i t ut e of Technology was approximately 70 miles from t he center of t he shock and
it recorded a maximum acceleration of 5 per cent of gravity. A 10-second period
displacement meter also recorded the motion at t he same location. Figure 13 shows
a comparison bet ween t he north-south recorded and integrated displacements
covering only the first t went y-four seconds of the ground motion. The 16-second
period component t hat appears in the integrated displacement is t oo long t o be
recorded by t he 10-second displacement meter. Aside from this, however, there is
reasonable agreement between t he recorded and integrated displacements.
G . V . B . G . W . I I .
DEPARTMENT OF CIVIL ENGINEERING DIVISION OF t~NGINEERING,
UNIVI~RSITY OF MICHIGAN CALIFORNIA INSTITUTE OF TEcHNoLoGY~
ANN ARBOR, MICHIGAN PASADENA, CALIFORNIA.