REPORT NO.

UCB/EERC-89/03

MARCH 1989

PB91~229369

EARTHQUAKE ENGINEERING RESEARCH CENTER

IMPLICATIONS OF SITE EFFECTS IN THEMEXICO CITY EARTHQUAKE OF SEPT. 19, 1985FOR EARTHQUAKE-RESISTANT DESIGN CRITERIAIN THE SAN FRANCISCO BAY AREAOF CALIFORNIA

by

H. BOLTON SEED

JOSEPH I. SUN

A report on research sponsored bythe National Science Foundation

REPRODUCED liWU.S. DEPARTMENT OF COMMERCE

NATfONAL TECHNICALINFORMATION SERVICESPRINGFIELD, VA 22161

COllEGE OF ENGINEERING

UNIVERSITY OF CALIFORNIA AT BERKElEY

50272-10l

REPORT DOCUMENTATION 11• REPORT NO. .PAGE NSF/ENG-89022 I~

PB91-:229369 -

4. Title and Subtitle

Implications of Site Effects in the Mexico City Earthquake ofSept. 19, 1985 for Earthqauke-Resistant Design Criteria in the~Rn ~ • Or'ro "R"nT AT,,'" rlf r",l ifornia7. Author(s)

H.B. Seed and J.I. Sun9. Performing Organization Name and Address

Earthquake Engineering Research CenterUniversity of California, Berkeley1301 S 46th St.Richmond, CA 94804

12. Sponsoring Organization Name and Address

National Science Foundation1800 G. St. NWWashington, DC 20550

15. Supplementary Notes

!5. Report Date

March 1989

8. Performlna Organization Rept. No.UCBjEERC-89/03

10. Project/Task/Work Unit No.

11. Contract(e) or Grant(G) No.

(C)

(G) ECE86-11066

13. TYJ)e of Report & Period Covered

14.

16.. Abstract (Limit: 200 words)

One of the most dramatic aspects of the earthquake effects in the Mexico City earthquake ofSeptember 19, 1985 was the enormous differences in intensities of shaking and associatedbuilding damage in different parts of the city. This report examines the factors which arelikely to have influenced the response and degree of damage to structures in the heavy dam-age area of Mexico City in the earthquake of 1985, attempts to relate these factors to the,intensity of damage which occurred, uses the results of the studies to examine the possibleextent of damage to structures constructed on sites underlain by clay in other seismic re-gions, such as the San Francisco Bay area, which like Mexico City, is located near the edgeof a deep deposit of clay soil, examines the implications of structural performance in Mex-ico City for buildings in San Francisco in the light of the seismicity of the region, andexamines the effects of possible modifications in building codes which might seem desirablein the light of the Mexico City disaster in 1985.

17. Document Analysis a. Descriptors

b. Identifiers/Open·Ended Terms

c. COSATI Field/Group

18. Availability Statemen~

Release Unlimited19. Security Class (This Report)

unclassified20. Security Class (This Page)

unclassified22. Price

(See ANSI-Z39.1B) See Instructions on Revers. OPTIONAL FORM 272 (4-77)(Formerly NTIS-35)Department of Commerce

EARTHQUAKE ENGINEERING RESEARCH CENTER

IMPLICATIONS OF SITE EFFECTS IN THE MEXICO CITY EARTHQUAKEOF SEPT. 19, 1985 FOR EARTHQUAKE-RESISTANT DESIGN CRITERIA

IN THE "SAN FRANCISCO BAY AREA OF CALIFORNIA

by

H. Bolton Seed

Joseph I. Sun

Report No. UCB!EERC-89!03

A Report on Research Sponsored bythe National Science Foundation

March 1989

College of Engineering

Department of Civil Engineering

University of California

Berkeley, California

, ,

For sale by the National Technical Informa-tion Service, U.S. Department of Commerce,Springfield, Virginia 22161

See back of report for up to date listing ofEERC reports.

DISCLAIMERAny opinions, findings, and conclusions orrecommendations expressed in this publica-tion are those of the authors and do not nec-essarily reflect the views of the NationalScience Foundation or the Earthquake Engi-neering Research Center, University of Cali-fornia at Berkeley.

ACKNOWLEDGEMENTS

The research study described in this report was made possible by Grant

No. ECE 8611066 from the National Science Foundation. This support is

gratefully acknowledged, as is the encouragement of Dr. Clifford J. Astill,

Program Director.

The assistance provided by many colleagues and engineers, including

V. Bertero, R. Golesorkhi, R. Borcherdt, M. Romo, J. Lysmer and R. Pyke, is

also deeply appreciated.

ii

TABLE OF CONTENTS

PageNo.

I. INTRODUCTION

II. RELATIONSHIP BETWEEN DAMAGE INTENSITY, GROUND MOTIONS ANDDESIGN LATERAL FORCE COEFFICIENTS FOR BUILDINGS IN THEHEAVY DAMAGE ZONE OF MEXICO CITY

Damage Intensity in the Heavy Damage Area of Mexico City

Evaluation of Potential for Building Damage Due toEarthquake Shaking

Seismic Code Provisions for Mexico City

Ground Response in the Heavy Damage Area of Mexico City

Damage Potential Index for Heavy Damage Zone in Mexico City

III. EVALUATION OF POTENTIAL FOR BUILDING DAMAGE DUE TOEARTHQUAKE SHAKING IN THE SAN FRANCISCO BAY AREA

1

4

4

8

14

14

15

22

Seismic Environment of the San Francisco Bay Area 22

Soil Conditions in the San Francisco Bay Area 25

Dynamic Soil Properties of San Francisco Bay Mud 28

Shear Wave Velocity Profiles for Young San Francisco Bay Mud 32

Site Conditions for Three Bayshore Sites 32

Southern Pacific Building site 34

Embarcadero Site 37

Ravenswood Site 41

Characteristics of Rock Outcrop Motions in San Francisco 41

Ground Response Analyses for San Francisco Bayshore Sites 43

Ground Response Analyses for Magnitude 7-1/4 Earthquake

Near San Francisco Bay Area

(1) Southern Pacific Building Site

(2) Embarcadero Center Site

(3) Ravenswood Site

Summary of Ground Response Analyses for Magnitude 7-1/4Earthquake

51

51

54

54

59

iii

PageNo.

III. Contd.

Ground Response Analyses for Magnitude 8+ EarthquakeNear San Francisco Bay Area 64

(1) Southern Pacific Building Sit 64

(2) Embarcadero Center Site 67

(3) Ravenswood Site 67

Summary of Ground Response Analyses for Magnitude 8+ Earthquake 72

Ground Motions in Central Parts of San Francisco 72

IV. EARTHQUAKE-RESISTANT DESIGN PROVISIONS FOR CALIFORNIA 78

Current Seismic Design Provisions for Structures in California 80

V. EVALUATION OF DAMAGE POTENTIAL INDEX VALUES FOR STRUCTURESIN SAN FRANCISCO

Computation of Damage Potential Index Values for

San Francisco Bay Area

Lateral Force Requirements Recommended by SEAOC (1988)

Calculated Damage Potential Index Values for San FranciscoStiff Soil and San Francisco Bayshore Sites Based on1988 SEAOC Recommendations

REFERENCES

86

86

97

105

121

I.

IMPLICATIONS OF SITE EFFECTS IN THE MEXICO CITY EARTHQUAKEOF SEPT. 19, 1985 FOR EARTHQUAKE-RESISTANT DESIGN CRITERIA

IN THE SAN FRANCISCO BAY AREA OF CALIFORNIA

by

H. Bolton Seed1 and Joseph I. Sun2

I. INTRODUCTION

One of the most dramatic aspects of the earthquake effects in the

Mexico City earthquake of September 19, 1985 was the enormous differences in

intensities of shaking and associated building damage in different parts of

the city. In the south-west part of the city ground motions were moderate

and building damage was minor. However in the north-west part of the city,

catastrophic damages occurred and a record of the earthquake motions near

the southern end of this heavy damage area showed a very high intensity of

shaking. Similar patterns of building damage intensities have been observed

in previous earthquakes and the differences attributed to the differences in

soil conditions in different parts of the city. In the 1985 Mexico

earthquake these differences seem to be somewhat more accentuated than in

other earthquakes in the past 40 years, and the availability of recordings

of ground motions in different parts of Mexico City makes it possible to

explore, in greater detail than heretofore, the relationships between soil

conditions, intensities of shaking, and the associated extent of structural

damage.

Analyses of ground response for five sites in Mexico City in relation

to the soil conditions at the recording stations (Seed et al., 1987) have

1Cahill Professor of Civil Engineering, University of California, Berkeley,California.

2Graduate Research Assistant, Department of Civil Engineering, University ofCalifornia, Berkeley, California.

2

shown that if allowance is made for possible small deviations from the best

average deterministic ground response analysis parameters, and ground

response is considered on a probabilistic basis, ground response analyses

can provide very useful data for assessing the influence of local soil

conditions on the characteristics of the ground motions likely to develop at

sites in the old lake-bed area of Mexico City where motions varied widely

depending on the depth and stiffness of the clay deposits. Furthermore,

because of the generally good results obtained in using ground response

analyses to predict ground motions for the five sites at which motions and

soil characteristics are known in Mexico City, the same procedures can be

expected to provide a good basis for predicting motions at sites where

motions were not recorded in the September 19, 1985 earthquake. Thus it has

been possible to make analyses for a number of different soil depths

existing in the heavy-damage area of Mexico City and to develop a

representative spectrum for the average ground motions occurring in this

area in the 1985 earthquake (Seed et al., 1987).

Within the heavy damage area itself, the intensity of structural

damage was found to be different for structures of different heights,

presumably reflecting the influence of the soil conditions, the intensity

and frequency characteristics of the ground motions, the characteristics of

the structures and the criteria controlling the design of the structures.

It is the purpose of this report to examine the factors which are

. likely to have influenced the response and degree of damage to structures in

the heavy damage area of Mexico City in the earthquake of 1985, to attempt

to relate these factors to the intensity of damage which occurred, to use

the results of the studies to examine the possible extent of damage to

structures constructed on sites underlain by clay in other seismic regions,

3

such as the San Francisco Bay area, which like Mexico City, is located near

the edge of a deep deposit of clay soil, to examine the implications of

structural performance in Mexico City for buildings in San Francisco in the

light of the seismicity of the region, and to examine the effects of

possible modifications in building codes which might seem desirable in the

light of the Mexico City disaster in 1985.

4

II. RELATIONSHIP BETWEEN DAMAGE INTENSITY, GROUND MOTIONSAND DESIGN LATERAL FORCE COEFFICIENTS FOR BUILDINGSIN THE HEAVY DAMAGE ZONE OF MEXICO CITY

Damage Intensity in the Heavy Damage Area of Mexico City

The location of the heavy damage area in Mexico City in the 1985

earthquake is shown in Fig. 1. Following the earthquake, a detailed survey

was made of the intensity of damage to different classes of structures in

different parts of the city (Borja-Navarrete, et al., 1986). The damage

statistics for the heavy damage zone of the city are summarized in Table 1,

which shows, for buildings with different story height ranges, the datllage

intensity, defined as the ratio of the number of structures in any given

category which suffered major damage divided by the total number of

structures in that category existing in the heavy damage zone.

It is readily apparent that it was the mid-height buildings, with

about 6 to 20 stories, which suffered the highest damage intensities. This

trend is also clearly evidenced by the plot of these data shown in Fig. 2.

Since most seismic design procedures for buildings are based on

structural period rather than building height, it is useful to examine the

natural periods of the structures in Mexico City in relation to the number

of stories of the buildings. Emphasis will be placed on large-deformation

periods since it is these periods which are most indicative of building

behavior during major earthquakes (Bertero et al., 1988).

For North American practice, the fundamental period (in seconds) for

typical bUildings is typically about N/lO, where N is the number of stories.

However, in Mexico City, buildings are somewhat less stiff as compared to

United States practice, the foundation soils are much more compressible, in-

fill walls tend to crack early in an earthquake and buildings become less

5

Reproduced frombest available copy.

o

____Heavy damage zone

• Locotions 01 strong-moliona cce\erographS

\.~

cotas. en m

FIG. I PLAN OF MEXICO C11~ SHOWING DEP1H OF SOIL AND LOCA11DN DFINS1RUMENl SI~110NS

Ii

TABLE 1 DAMAGE STATISTICS FOR THE HEAVY DAMAGE AREA OFMEXICO CITY IN THE SEPT. 19, 1985 EARTHQUAKE(modified after Borja-Navarrete et al., 1986)

,I INumber of Number of Bldgs. with Total Number Damage

Stories Serious Damage of Buildings Intensity

1 - 2 :::: 297 :::: 15,000 :::: 2 %

3 - 5 :::: 154 :::: 5,400 :::: 3 %

6 - 8 :::: 117 :::: 650 :::: 18 %

9 - 12 :::: 62 :::: 215 :::: 29 %

> 12 :::: 21 :::: 92 :::: 23 %

6

60

'"""I---------------------------------, 24

Da

mo

ge

Inte

nsi

tyfo

rH

ea

vyD

am

og

eZ

on

ein

Me

xico

City

,1

98

5(M

=8.

101

ob

ou

l2

40

mileS?

".--

-...

,'".....

...,

..,'

.....

"'"

"'"

,..

,..

""..

,..

,..

,..

,..

.,..

, , , " ",",

""

" ..' .......--

-,,,,.

--a

II

II

II

I

a4

81

21

62

0N

um

be

ro

fS

tori

es

50

I-

..... e (I) u L ~4

0l-

I >- ~ Vl

30

l-e Q.l ..... e Q.l (J

l

02

0l-

E 0 0

10

I- FIG

.2

DAMA

GEIN

TENS

ITY

FOR

HEAV

YDA

MAGE

ZONE

INM

EXIC

OCI

TYDU

RING

THE

SEPT

.19

,19

85EA

RTHQ

UAKE

'-J

stiff, while periods also lengthen if the structures go beyond the elastic

range. Thus the "effective" building periods for structures ·in Mexico City

can be expected to be significantly longer than those normally estimated for

U.S. structures. Quantitative estimates of these effects are shown in

Table 2, and it seems reasonable to expect that the effective building

periods for many structures in the heavy damage area of Mexico/City is more

likely to have been of the order of N/6 (seconds) where N again represents

8

the number of stories. Taking this into account, the data in Fig. 2 are

replotted in Fig. 3 to show the damage intensity as a function of effective

building periods. It is evident that structures suffering the highest

damage intensities were those with fundamental periods in the range of 1.5

to 2.5 seconds.

Evaluation of Potential for Building Damage Due to Earthquake Shaking

In previous studies of earthquake damage intensity in relation to

ground motions it has been suggested that a useful index of the

vulnerability of a structure to damage caused by earthquake shaking can be

evaluated in terms of a simple ratio establishing a Damage Potential Index

(Seed et al., 1970): This index, which incorporates the idea of capacity

(the forces that a bUilding is designed to withstand) and demand (the forces

induced on the building by the earthquake shaking), was found to be

extremely useful in examining damage in Caracas, Venezuela (Seed et al.,

1970) and it will therefore be used, with minor modifications in the present

study.

In the approach proposed by Seed et al., 1970 for ductile buildings,

of the type which suffered major damage in Mexico City, the Damage Potential

Index (DPI) is evaluated as follows:

TABLE 2 ESTIMATION OF BUILDING PERIODS IN HEAVY DAMAGE AREA OFMEXICO CITY WITH RESPECT TO NUMBER OF STORIES IN THESEPT. 19, 1985 EARTHQUAKE

9

For U.S. structures, T = N I 10, where N = no. of stories

Conditions in Mexico City Effect on Building Period

1. Soil more compressible Increase by about 50%

2. Infill walls crack easily Increase by about 30%

3. Building go beyond elastic range Increase by about 30%

Thus for conditions in the heavy damage area of Mexico City:

Effective building period = N/5 to N/6 seconds

60I

I

50

I-

24

Da

ma

ge

Inte

nsi

tyfo

rH

ea

vyD

am

ag

eZ

on

ein

Me

xico

City

,1

98

5(M

=B

.lat

abau

t2

40

mileS?

,---

.....,'"

'....

,..

,'.

....,

...,

..,,

....,

,,

..,

..,

~

""

,..

,..

,..

./..

,/ , "

"I

"",

,",,"

....,.."-

".

..,,"

oI

I(

II

II

o4

81

216

20

Nu

mb

er

of

sto

rie

sI

II

•I

II

!I

o1

23

4

Bu

ildin

gP

eri

od

-se

con

ds

10

I-

-+-' c 0) U L- 0) 0.

40

I-

>.

-+-' Ul c

30

I-0

)-+

-' c 0) C1l

o2

0l-

E o o

FIG

.3

DAMA

GEIN

TENS

ITY

FOR

HEAV

YDA

MAGE

ZONE

INM

EXIC

OCI

TYDU

RING

THE

SEPT

.19

,19

85EA

RTHQ

UAKE

INRE

LATI

ONTO

ESTI

MAT

EDBU

ILDI

NGPE

RIOD

INM

EXIC

OCI

TY

I-'

o

Force induced on building by earthquake 0: w • Sag

11

Damaging Potential of ground motion 0: (Induced Force) x (Duration of force)

0: w . Sa x T x No. of cyclesg

where the duration of .the force is considered to be proportional to the

period of the building and the number of load cycles induced by the

earthquake. The number of cycles will be determined primarily by the

duration of shaking during the earthquake and this in turn will depend on

the Magnitude of the earthquake (Bolt, 1973). The damaging potential of the

ground motion, as expressed above, can be considered an approximate

expression of the Demand imposed on a structure by the earthquake shaking.

The design resistance of a structure (Capacity) is usually determined

by the building code requirements and for most codes, including the Mexico

City code, it is expressed as follows:

Design Lateral Force = k • W

where k is the design lateral force coefficient. Generally speaking, the

higher the design lateral force coefficient, the greater is the capacity of

a structure to withstand the effects of earthquake shaking.

The capacity will also depend, however, on the load combinations and

the allowable stresses prescribed by the Building Code. In anyone city

these will be the same for all structures, but in comparing structures in

different cities, the relative values of these factors, which also affect

design resistance, will have to be taken into account. On a comparative

basis they can be expressed by a factor termed the building resistance

factor, Rf , which expresses the relative design resistances as they are

affected by allowable stresses and load combinations, all other factors

12

being equal. Thus the capacity of a building to withstand earthquake damage

can be expressed by:

Design Resistance ex: k • W • Rf

and the vulnerability of a structure to earthquake damage can be expressed

in an approximate way by the ratio of Demand/Capacity, leading to the

development of a Damage Potential Index as follows:

Damage Potential Index ~ Induced Force x Duration of ForceDesign Resistance

W • Sa/g x T x No. of cycles

k • W • Rf

Sv• Duration Weighting Factor

k Rf

where the Duration Weighting Factor reflects the influence of the duration

of shaking and is a function of the earthquake magnitude. Since the intent

of this index is only to compare the relative vulnerabilities of different

structures, the Duration Weighting Factor can be assigned relative values,

based on judgment, which are determined by the Magnitude of the earthquake

involved. Suggested values of the Duration Weighting Factor (DWF) are

listed in Table 3. It is recognized that other engineers may make different

estimates of the effects of duration of shaking on potential damage, but the

values shown in Table 3 will be used in the present study as a first

approximation, and the results obtained with this approach will be compared

with the actual damage statistics for the heavy damage area of Mexico City

in the 1985 earthquake.

TABLE 3 SUGGESTED VALUES OF DURATION WEIGHTINGFACTOR (DWF) FOR STRONG GROUND MOTIONSWITH DIFFERENT DURATIONS

Duration DWF

15 sec 0.6

40 sec 1.0

65 sec 1.35

120 sec 2.0

13

14

In addition. since the term Rf is intended to express relative values

of design resistance in different cities or areas. it is most convenient to

assign this factor a value of unity for Mexico City; values for other cities

and areas would then be somewhat higher or lower than unity depending on

their individual Code requirements. Thus, for example, based on the Code

requirements for California (SEAOC. 1980), the value of Rf applicable for

reinforced concrete structures in California might be estimated to be about

1.3 (Bertero, 1988). For other areas appropriate values could be assessed

by knowledgeable structural engineers familiar with Code requirements in

Mexico City and local design codes.

Seismic Code Provisions for Mexico City

The first seismic design provisions adopted for use- in Mexico City

were developed in 1942. and they have been under constant revis ion since

that time (1957, 1966 and 1976). The 1976 code microzoned the Federal

District of Mexico into three parts: (1) the hilly and hard soil or rocky

zone; (2) the transition zone and (3) the lake bed zone. in recognition of

past experience which indicated that different intensities of shaking

developed in the city depending on the subsoil conditions. Table 4 shows

the required lateral force coefficients (the ratio of design lateral force

to total building weight) for buildings located in the lake-bed zone. For

all multi-story buildings having periods longer than 1 second, the required

design lateral force is equivalent to 6% of the building's weight. Lower

design requirements were used for the hilly zone and the transition zone.

Ground Response in the Heavy Damage Area of Mexico City

In a previous report (Seed et a1.. 1987), ground response analyses

were performed to study the ground motions developed in the lakebed areas of

15

Mexico City and for the heavy damage area of Mexico City during the 1985

earthquake. Fig. 4 shows, in terms of acceleration response spectra, the

recorded motions at the SCT recording station, located within the heavy

damage zone, together with the spectra for computed motions likely to have

been developed for clay depths ranging from 25 to 45 meters. Based on these

results a representative average response spectrum was determined which

could be considered to represent the general characteristics of the earth-

quake motions in the heavy damage zone, as shown in Fig. 4. These spectral

accelerations can readily be converted to spectral velocities to evaluate

Damage Potential Index values for bUildings in this zone.

Damage Potential Index for Heavy Damage Zone in Mexico City

The Damage Potential Index (DPI) has been defined previously as (see

page 12) as:

DPI = Sv . Duration Weighting Factork-Rf

and DPI thus has the units of velocity.

For Mexico City, Rf has been assigned a value of 1 in this study.

Thus with the aid of spectral velocities determined from the results shown

in Fig. 4 and lateral force coefficients determined from Table 4, values of

the Damage Potential Index for buildings with different periods can readily

be determined for buildings in the heavy damage zone of Mexico City as shown

in Table 5. The results shown in Table 5 are plotted in Fig. 5 to show the

computed Damage Potential Index values for buildings with a wide range of

periods. It can be seen that bUildings which have large-deformation natural

periods in the neighborhood of 2 seconds exhibit the highest damage poten-

tials, which corresponds well with the damage observed in Mexico City. The

1.0

0.80.9

5.0

4.5

4.0

Re

pre

sen

tati

veA

vera

ge

for

He

avy

Da

ma

ge

Are

a

Dam

ping

Rat

io=

5",

Co

mp

ute

dfo

rd

iffe

ren

td

ep

ths

toh

ard

laye

r

Sp

ect

rafo

rre

cord

ed

mo

tio

ns

at

seT

site

, '~""'"

~~~~

........ .

... 3=~.-.::,:::

I I II , I I I I I , I ,

I , , , I \ , , I"

2.0

2.5

3.0

3.5

Per

iod

.se

co

nd

s

. ",I I, ,I I, ,,

,I

I,

I,

II

I,

I,

,,

I,

I,

•I

,,

,,

,'

,,

.'-

_.'

\,

',

'I

',

I I ,

1.5

1.0

0.5

0.40.5

0.3

0.1

0.0

0.0

0.2

O'l

0.7

~ c o o L. +-'

U Q)

0-

if)

+-'

0.6

o L. Q} Q} u u «

FIG

.4

COMP

UTED

SPEC

TRA

FOR

HEAV

YDA

MAGE

AREA

FOR

SOIL

DEPT

HSRA

NGIN

GFR

OM25

TO45

MET

ERS

AND

EVAL

UATI

ONOF

REPR

ESEN

TATI

VEAV

ERAG

ESP

ECTR

UMFO

RHE

AVY

DAMA

GEAR

EAI-

'0

'\

TABLE 4 DESIGN LATERAL FORCE COEFFICIENTS FORMEXICO CITY LAKEBED ZONE (1976-1985)

Building Period Design Lateral Force(seconds) Coefficient

0.0 0.030

0.5 0.045

1.0 0.060

1.5 0.060

2.0 0,060

2.5 0.060

3.0 0.060

3.5 0.060

t

17

TABLE 5 CALCULATED DAMAGE POTENTIAL INDEX VALUES FOR HEAVY DAMAGEAREA OF MEXICO CITY IN THE 1985 EARTHQUAKE

I

MEXICO CITY - 1985

For Representative Spectrum in Heavy Damage Part of the City

Buildirg Spectral Spectral Lateral Force DamagePeriod Accelera- Velocity Coefficient Potential

tion (K = 0.8) Ind.ex

T sa SV k (Svjk.R ).IMF(sec) (g) (fps) (fps)

0.0 0.15 0.00 0.030 0

0.5 0.23 0.59 0.045 26

1.0 0.31 1.61 0.060 54

1.5 0.50 3.84 0.060 128

2.0 0.65 6.66 0.060 222

2.5 0.60 7.69 0.060 256

3.0 0.30 4.61 0.060 154

Notes:

1. The Duration Weighting Factor CDWF) used is 2.0 for theduration of strong ground motions in the 1985 earthquake,which lasted over 2 minutes.

2. The building resistance factor CRf ) is assigned a value of 1.0for typical reinforced concrete structures in Mexico City.

18

50

0

45

0

40

0u.. 3 0 ~

35

0""

- > Cf) ..........

30

0I x Q)

25

0.-

"'(J c - 0

20

0:.;

:; c Q) 4J 0 0...

15

0Q

)

Ol

0 E1

00

0 (:)

50 0

0

Da

ma

ge

Po

ten

tia

lIn

de

x(S

v/k

DW

Fin

fps)

for

He

avy

Da

ma

ge

Zo

ne

of

Me

xico

City

,1

98

5(M

=8

.1a

ta

bo

ut

24

0m

iles)

20

general trend of the relationship shown in Fig. 5 is also in good accord

with the relationship based on observed damage intensity previously shown in

Fig. 3. The relationships between observed damage intensity and computed

Damage Potential Index in the heavy damage area of Mexico City can be more

easily compared in Fig. 6, which superimposes the results presented in

Fig. 3 and Fig. 5. Again it can be seen that buildings which had natural

periods close to about 2 seconds suffered the most severe damage in

Mexico City in the September, 1985 earthquake and also that a calculated

Damage Potential Index of about 200 fps corresponds roughly to an observed

damage intensity of about 30% for the damage developed in Mexico City in

this earthquake.

60

~i-------------------1

'4

00

en C.

..... I l.L.3

00~ ----..:x '-. > (f) .......

.

x Q) "02

00

c 0 :;J c Q) ..... 0 0-

10

0Q

)O

l0 E 0 0

02

4

•

10

•A

ctu

al

valu

es

of

da

ma

ge

inte

nsit

y'

50I-

aC

om

pu

ted

valu

es

of

da

ma

ge

po

ten

tia

lin

de

x,S

v/k

oI

II

II

!I

o4

812

162

0N

um

be

ro

fS

tori

es

I.

__,

1_

~---.1...-

I~L.

l-

1

o

o4

J C Q)

U L. ~

40

£ ~3

0v .....

. c v 8'2

0E o o

o1

23

Bu

ildin

gP

eri

od

--s

eco

nd

s4

FIG

.6

ACTU

ALDA

MAGE

INTE

NSIT

YAN

DCA

LCUL

ATED

DAMA

GEPO

TENT

IAL

INDE

XFO

RHE

AVY

DAMA

GEZO

NEIN

MEX

ICO

CITY

N I-'

22

III. EVALUATION OF POTENTIAL FOR BUILDING DAMAGE DUE TOEARTHQUAKE SHAKING IN THE SAN FRANCISCO BAY AREA

Seismic Environment of the San Francisco Bay Area

In view of the good relationship observed between Damage Potential

Index values and actual damage intensities for the heavy-damage zone of

Mexico City, it is of interest to examine the significance of these results

to other areas of North America where structures are constructed on deep

layers of clay. Areas of maj or interest in this respect would certainly

include San Francisco, California, Salt Lake City, Utah, and Anchorage,

Alaska. The San Francisco Bay area was selected for special study in this

investigation. The Bay area has experienced 12 damaging earthquakes during

the past 150 years (Goldman, 1969), including the majo! San Francisco

earthquake of 1906 (Magnitude ~ 8.2), and can be expected to be subjected to

similar events in the future. In its latest evaluation the U.S. Geologic

Survey predicts a 50% probability of a Magnitude 7 earthquake in the

San Francisco Bay area in the next 30 years (Fig. 7). However a repetition

of the 1906 Magnitude 8 earthquake, although assigned a relatively low

probability of about 10% in the next 30 years, is also an important

consideration in the next 100 years.

Most of the major earthquakes that have occurred in the San Francisco

Bay area have been closely related to the active faults in the area, shown

in Fig. 8. These include the San Andreas fault which transects the

San Francisco and Marin Peninsulas, the Hayward fault which runs along the

base of the Berkeley Hills in the East Bay, and the Calaveras fault located

south of the Hayward fault. These faults and their branches are closely

related and together comprise an important part of the major fault system

which governs the seismicity of the San Francisco Bay area.

Gorda BasinMendocino, M = 7

\

\. Olema, M=8\

EXPLANATION

CUMULATIVE 30-YEAR PROBABILITY

••• ~ 60 percent

••• 2: 40 percent

--- ~ 20 percent

••• ~ '0 percent

--- ~ 5 percent

All other faults < 5 percent

23

San FranciscoPeninsula, M=7

100 200 KILOMETERSI

San Juan Bautista, M =6.5,,

"'I. Creeping, M= 7(7),'I.,,,,

Carrizo, M = 8

\

USGS hazard watchMammoth-Mono Lakes

\\\\\

FIG. 7 30-YEAR CUMULATIVE PROBABILITIES OF OCCURRENCE OF EARTHQUAKEALONG SELECTED FAULT SEGMENTS OF THE SAN ANDREAS FAULT SYSTEM(after Lindh, 1983)

.-

D

oA

~ }----~o ';I 40 ~ ~ ~ ........ ,...s.a..."'...-c'KO""C~....aTIOIIMClOlYILoPlIIIt .. tc:o..M.SSoIC*

o

o 0

D

o

.0

o

o

D

o

\o

24

FIG. 8 ACTIVE FAULT SYSTEM NEAR SAN FRANCISCO FAULT SYSTEM(after Bolt, 1978)

25

Soil Conditions in the San Francisco Bay Area

San Francisco Bay is located in a northwest-trending valley. The bay

is bounded mostly by marshlands, alluvial plains and beyond, by the Coast

Ranges. The major geological units of the San Francisco Bay area, shown in

Fig. 9 can be categorized broadly as bedrock, alluvium, and Bay mud, as

follows (Borcherdt, et. aI, 1975):

(1) Bed rock in the area is composed mainly of sandstone, siltstone,

chert and greenstone of the Franciscan formation.

(2) Alluvium. This unit contains late Quaternary floodplain deposits

of silt and clay, inter-layered with alluvial fan and stream-bed

deposits of sand and gravel, derived from weathering and erosion

of the uplands surrounding the San Francisco Bay. Some older

alluvial deposits (early Quaternary) may be more consolidated

and/or partially cemented.

(3) Bay Mud. These sediments are Holocene age, soft, water-

saturated, organic-rich silts and clays,occasionally interlayered

with sand deposits. They generally are derived from the

suspended materials brought into San Francisco Bay by the rivers

draining the Central Valley of California, as well as streams

from the southern Bay area. Table 6 shows some of the physical

and engineering properties of San Francisco Bay mud. It is the

behavior of this soft clay under seismic loading that concerns

many seismologists and engineers (after Goldman, 1969; Borcherdt,

1970; Borcherdt and Gibbs, 1976; Wilson, Warrick and Bennett,

1978).

As clearly illustrated in Mexico City, local geological conditions can

substantially change the characteristics of seismic waves and the intensity

EXPLANATION

".d'

•Bay mild~

A1ll1Villm

r-:l~

Iledrock

•IlecarcIiD« Ioeation

10....

II

IOu.atl(11tl.S

26

FIG. 9 GEOLOGICAL CONDITIONS NEAR SAN FRANCISCO BAY(after Borcherdt et al., 1975)

TABL

E6

PHYS

ICAL

AND

ENGI

NEER

ING

PROP

ERTI

ESOF

SAN

FRAN

CISC

OBA

YMU

D(m

od

ifie

daft

er

Bo

nap

arte

and

Mit

ch

ell

,19

79an

dER

TEC,

1981

)

-N

atu

ral

Hea

nT

hic

kn

ess

v.t

ee

Liq

uid

Pla

etl

cU

nit

e11

a.et

erS

hea

rC

OII

-o

fIl

udco

nte

nt

lI.1

tlI

.lt

wei

gh

tD

50In

Str

eng

thp

reu

lon

Sulf

tL

acH

lan

Ref

eren

ce(f

eet)

(per

cen

t)(p

erce

nt)

(per

cen

t)Ib

/ft3

.teto

n.

Ib/f

t2In

dex

C cS

ull

lp

sf/

H

Pro

po

sed

Tra

lkan

dR

ols

ton

0-4

594

-101

1.4

-10

.5sout~ern

cro

ssin

g(1

9511

Pro

po

sed

Tra

skan

dR

ols

ton

0-6

535

-80

41-6

8-

U-I

13

3.1

-13

.520

0-40

0,a

rall

el

cro

ssin

g11

951

)

Oak

land

Est

uar

yT

rask

and

Ro

lsto

n0

-45

U-I

OI

3.6

-5.1

(195

11

Ad

jace

nt

tob

rid

ge

See

dan

dR

eese

2035

-50

41

.52

3.5

112

-10

0-50

00

.36

-0.4

5O

akla

nd

1401

e(1

951)

Jala

laC

reek

Baa

lnL

ea(1

953)

128

52-6

2-

-10

2-10

1-

100-

1000

Can

dle

stic

kC

ove

Prl

vlt

eC

omm

unl-

>65

50-1

565

-119

21-3

598

-101

2-1

015

0-80

00

.5-1

.0cati

on

Ham

ilto

nA

irF

orc

eD

unc.

n(1

"5

)60

9088

U93

-96

-40

0-12

001

.10

.12

10B

I ..

L.s

hT

er.

lnal,

Po

rtD

uncs

nan

d10

055

5030

100

-20

0-10

0-

0.2

110

of

San

Fra

nci

sco

Buc

hlgn

anl

(191

31

Isla

IsC

raek

52C

loug

han

d8

1ta

r50

55-9

515

-100

30-4

510

5-

400-

1000

--0

.40

11(1

911)

Jala

ldC

reek

52

Clo

U9h

and

5It

ar

120

15

-95

15-1

0030

-45

"-

400-

500

--0

.11

4(1

911)

Fu

ryB

uil

din

gP

riv

ate

CO

Mm

unl-

--

--

98-

---

0.2

410

cati

on

(196

5)

E ..b

arca

der

oC

ente

rP

riv

ate

CO

Mm

unl-

--

--

110

-"'

00

-15

00

--0

.21

13cati

on

(197

1)

Ha.

.llt

on

Den

by(U

?8)

58B

6-90

85-8

831

-40

94-

400-

1300

.8-1

10

.34

10

.7

lIen

loP

ark

!lo

har.

at

11

,1

6-9

1"

41-1

.1(1

9731

So

uth

5.r

.B

ayII

rula

nan

dan

.96

-103

9345

90

.5et

al.

(197

11

Su

isu

nH

arsh

Th

isat

ud)'

7558

-120

90-9

852

-62

BO

-95

N -....J

28

of earthquake shaking. This was also made abundantly evident in the 1906

San Francisco earthquake. Fig. 10 shows the geological conditions in San

Francisco and Fig. 11 shows the intensity of shaking for San Francisco

during the 1906 earthquake (Borcherdt, 1975). Except for the south-western

part of the city where the San Andreas fault passes directly through the

city, most of the parts of the city showing high values of "apparent shaking

intensity" were those underlain by thick layers of Bay mud. This was

especially true for the downtown area of San Francisco. The need to re-

evaluate seismic damage potentials for the San Francisco Bay area in the

light of the Mexico City earthquake experience seems therefore especially

appropriate.

Dynamic Soil Properties of San Francisco Bay Mud

It is well-recognized that the appropriate forms of the modulus

reduction and damping ratio relationships with shear strain play a key role

in performing successful ground response analyses. A number of studies have

been performed to evaluate these dynamic properties for San Francisco Bay

mud. Bay mud samples tested in these investigations were taken from:

Suisun marsh in the north Bay (ERTEC, 1981), Hamilton Air Force Base in the

west-central part of the Bay (Isenhower, 1979; Isenhower and Stokoe, 1981)

and from Ravenswood, Dumbarton West and Agnew, three sites located in the

south Bay (Stokoe and Lodde, 1978; Lodde, 1982).

The dynamic properties of young Bay mud, summarized from these

investigations, are shown in Fig. 12. The upper part of the figure shows

how the shear modulus reduces with increasing shear strain while the lower

part of the figure shows the increase in damping ratio with shear strain.

The rate of reduction in modulus with increasing strain for young Bay mud is

A4

37°48'

37°42'

REDUCTION OF EARTHQUAKE HAZARDS. SAN FRANCISCO BAY REGION

122'30'

EXPLANATION

ilil Bay mud (in places covered by artificial fill as of 1906)

29

122°22'

~Ye~

Buena I.

I::}:::}}j Alluvium (>3Om (100 tt) thick)

lU] Alluvium «30m (100 ft) thick)r.-:-:;l Bedrock~

oIo

2 3 MILESII

3 KILOMETRES

FIG. 10 GENERALIZED GEOLOGICAL MAP OF SAN FRANCISCO(compiled by K. R. Lajoie from data ofSchlocker et al., 1958)

37°48'

37°42'

122°30'

vg$Buena I.

"Il ~

"~~ ...

-a".... ;:;'>j ~.

e....~ ta

"""

30

_ Very violent

• Violent

EXPLANATION

FX"::::l Very strong

o Strong L=::] Weak

3 KILOMETRES

o 2 3 MILES1-1----,----'--,-,---r,...L'----',o 2

FIG. 11 DISTRIBUTION OF APPARENT SHAKING INTENSITY INSAN FRANCISCO IN THE 1906 EARTHQUAKE(compiled by Borcherdt, 1975 after data fromWood, 1908)

31

Average GIGo foryoung Boy mud

\

\\\

\\'\

\\

\'\

\\

\\

\\\

\'\",

Range of test results for~oung Bl2Y mud~lsenhower,Lodde & ERTEC)

1.0

0.8

0.2

~ 0~6E

C>

"-C> 0,4

0.010 -4 10 ..., 10-2 10-1

Shear Strain, percent

10

10

Relationship usedin this study

10 -2 10-1

Shear Strain, percent

Typical dam~ing ratio for cloys(Seed & ldnss 1970)~._._._'

"-'''-'-'.-'-~-",--. ,-./ " ~.

.",oil" "", ",0/

./ " /0/0"" ,,,' /0

Test results for young Bay mud./ /

(Isenhower, Lodde) A'''-'''-''' /.,/."- /.",

.~.~ ",,0.,.""" ."...~ ...,.....- _.~ .."""_.-0-.___ ~.-----_.-- _.-'

60

50

....c:Q)

40uI-Q)Q.

0 30:;:;c~

01c: 20'0.Ec

CI10

010-4

FIG. 12 VARIATION OF SHEAR MODULUS AND DAMPING RATIO WITHSHEAR STRAIN FOR YOUNG SAN FRANCISCO BAY MUD

32

significantly less than that for typical sands, but it is in generally good

accord with values determined for other clays (see Sun et al., 1988). The

damping characteristics fall well within the range proposed by Seed and

Idriss (1970) for typical clays. Lodde (1982) has also tested some older

Bay sediments near Dumbarton West site. The average modulus reduction

relationship for these older Bay sediments, which consisted mainly of

gravelly sands and silts having a void ratio of about 0.63, was more like

that for sands than for Bay mud.

Shear Wave Velocity Profiles for Young San Francisco Bay Mud

Values of shear wave velocity for young San Francisco Bay mud have

been measured in studies by Warrick, 1974; Gibbs et al., 1975; Gibbs et al.,

1976; Gibbs et al., 1977; Wilson et al., 1978; Pyke, 1987; -and Liu et al.,

1988. The results provided by these investigations are summarized in

Fig. 13. The figure indicates that shear wave velocities for young Bay mud

are essentially constant for the top 30 feet, with a value of 250 fps, but

the velocity then gradually increases to about 500 fps at a depth of

60 feet.

Site Conditions for Three Bayshore Sites

Three Bayshore sites underlain by soft Bay mud were chosen for the

purpose of studying the ground motions that are likely to develop on such

sites in the event of earthquakes with Magnitudes 7! and 8+ occurring on the

-San Andreas fault. The soft Bay mud at these three sites appears to have

the same general characteristics as for other San Francisco Bayshore sites~

Two of these sites, the Southern Pacific Building site and the Embarcadero

Center Four site are in the city of San Francisco, and the Ravenswood site

is in the south Bay area.

33

-

-

2000

Shear Wave Velocity, fpso 500 1000 1500

Or-.--..--nr.....-.--r--,-r--r-.,.--r-.....---r---r--;r-r-.,.--.---r-,I JJ I I I I I I I I I I I I I I! i IiiiI. i Iij..-I)i !' Ii \,

't-'"I i'

.;...40 f-

20 -

\\

\\\\\\\\\\

.......-

..c::.......a.(l)

o

60 f-

80 f-

100 f-

i \i \I \I \

\

I \'- ._.-1 -'-'-'1

iii

-

-

-

-

-

120 ,...

140o

I I I I r

500

Son Mateo (Bridgeway Park)RavenswoodRedwood PointSon Mateo (Audubon School)Coyote HillEmbarcadero Center

I 1 1 1 1 I 1 I I

1000 1500I

-

-

2000

FIG. 13 MEASURED SHEAR WAVE VELOCITY PROFILE FOR YOUNGBAY MUD FOR SIX SAN FRANCISCO BAY SHORE SITES

34

Southern Pacific Building Site

The Southern Pacific BUilding Site has been the subject of several

studies in the past several decades (Idriss and Seed, 1968; Singh et al.,

1981). The soil conditions at the Southern Pacific Building site are shown

in Fig. 14. A sandy fill extends to a depth of about 20 ft and is underlain

by a 35 ft layer of soft clay followed by a 50 ft layer of medium stiff

clay. Below this is a 120 ft layer of stiff clay interbedded with a 10 ft

layer of dense sand at a depth of 125 ft. The stiff clay is underlain by a

layer of very dense sand and gravel which extends to bed rock at a depth of

about 285 ft.

Initial estimates of the shear wave velocity for each soil layer in

the soil profile were based on the shear strength and the stiffness of the

materials, as reported by Idriss and Seed (1968) and Rinne and Stobbe

(1979), together with the representative data for Bay mud shown in Fig. 13.

The shear wave velocity profile was then further calibrated by computing its

response to the 1957 Daly City, San Francisco earthquake as e~plained below.

An earthquake of Magnitude 5.3 located along the San Andreas Fault was

recorded in the basement of the II-story high Southern Pacific Building on

March 22, 1957. The earthquake was simultaneously recorded on rock in

Golden Gate Park, located roughly 7 miles from the Southern Pacific Building

site as shown in Fig. 15. The Golden Gate Park record was used as a rock

outcrop motion in dynamic response analyses of the Southern Pacific Building

site, but the acceleration values were scaled down by a factor of 0.65 to

account for the different distances of the two sites from the source of

energy release, as recommended by Idriss and Seed (1968) and Singh et al.

Sh

ea

rW

ave

Ve

loci

ty,

fps

I

II

I

I-\

\/

Re

pre

sen

tati

vea

vera

ge

•\

for

you

ng

Bay

mu

d-

\-

\\

I-\

\

I-\

.

I- - I- ,- l- I- I- ----

----

----

----

----

----

-,

22

5'--------------

28

5' I""

/~"X'~

~~"M7

~~w"

TZn:

ISd~

/R....~~'&....IA

'"'&~rn..........~".,I

.,.,.1

\.'\Y.

""M'/J

0t.rJ'

r"'M~,

.,.~lr

'VJ'I'

IJ

Ro

ck

Sa

nd

fill

Vs

=8

00

fps

20

'

So

ftcl

ay

Vs

=3

25

fps

55

'

Me

diu

mcl

ay

Vs

=6

80

fps

10

5'

Sti

ffcl

ay

Vs

=8

00

fps

12

5'

San

dV

s=

11

00

fps

13

5'

Vs

=1

90

0fp

s

Vs

=1

10

0fp

sS

tiff

cla

y

Sa

nd

an

dg

rove

l

20

00

16

00

12

00

80

04

00

oo

25

50

75

10

0

12

5..... '+- .r:

15

0..... 0

-(l

)

01

75

20

0

22

5

25

0

27

5

JOO

FIG

.14

SOIL

PROF

ILE

AND

ESTI

MAT

EDSH

EAR

WAVE

VELO

CITI

ESAT

SOUT

HERN

PACI

FIC

BUIL

DING

SITE

W \.It

'u:-~

V-\"'g;\~

-

(1981) .

0.06 g.

37

The peak rock acceleration used in the analyses was thus reduced to

Values of shear wave velocity for the different layers were first

selected based on those used in previous studies, and on the available data

for soft Bay mud, and they were then modified slightly to give a velocity

profile for which the computed response spectrum, for motions at the

basement level, waS in best agreement with the spectrum for the recorded

motion. The shear wave velocity profile determined in this way is shown in

Fig. 14 and the results of the ground response analysis in Fig. 16. Fig. 17

shows a comparison of the response spectra for the computed and recorded

motions. It can be seen from the results presented in Fig. 14 that the

shear wave velocity profile for soft Bay mud used for this site agrees well

with the average shear wave velocity data for soft Bay Mud discussed

previously (see Fig. 13) and that the computed motions are in excellent

accord with those recorded in the 1957 earthquake.

Embarcadero Center Site

The Embarcadero Center Four site is located east of Drum Street, south

of Clay Street and north of Sacramento Street near the waterfront of

downtown San Francisco. The subsoil conditions, shown in Fig. 18, consist

of 5 layers: about 20 ft of fill overlying roughly 90 ft of Bay mud,

followed by 20 ft of sand, 30 ft of silty clay and 50 ft of silty sand. The

bedrock, mainly shale and sandstone, is located at a depth of 210 ft below

the ground surface. Down hole seismic surveys were performed (Harding-

Lawson, 1977) to measure the shear wave velocity profile for the site. A

representative shear wave velocity profile interpreted from these data, is

plotted in Fig. 18.

Com

pute

dm

otio

nat

bas

emen

tle

vel

2J

45

6T

Ime

-se

co

nd

s7

B9

10

Stif

fcl

ayV

s=

11

00

fps

Gol

den

Gat

eP

ark

reco

rd(1

957)

109

87

45

6li

me

-se

con

ds

J2

0.1

01 I c: 0

0.0

:p e Q) 11'0 0 '""

-0.1

0

San

dan

dgr

avel

Vs

=1

90

0fp

s

22

5'--------------

285'

7J17'2

l:l\:-

y'!VTl

7~~Q:7

Z7}m~/

"1'A'\

~¥n'i.

A7'\Yr

rnw~*;

nZ'77\

.\Y.~n

'/~'1I

:~"U,

,'T.~,

"~1TI1

Roc

k

FIG

.16

GROU

NDRE

SPON

SEAN

ALYS

ISOF

SOUT

HERN

PACI

FIC

BUIL

DING

SITE

INSA

NFR

ANCI

SCO

INTH

E19

57EA

RTHQ

UAKE

w 00

0.3

0.3

01

So

uth

om

Pac

ific

Bui

ldin

gO

l

rS

ou

tho

mP

acif

icB

uild

ing

Roc

ordo

dm

oUon

ct

bo

sem

on

t10

1/01

Ico

mp

ule

dm

olio

n01

bo

sem

on

t10

1101

c(D

ampi

ngR

ollo

-51

1)C

(Dam

ping

Rol

lo-

511)

:!l0.

2~

0.2

00

San

dfil

lV

s•

80

0fp

s..

..II

II

20'

'ilIi

00

~~

So

ftcl

ayV

.-

32

5fp

s_

0.\

_0

.\0

~.. ...

55'

00

IIII

a.a.

III

III

Med

ium

clay

Vs

-6

80

fps

0,5

\.0

1.5

2,0

0.0

0.5

1.0

1.5

2.0

Per

iod

-se

cond

sP

erio

d-

aeco

nda

105'

Sti

ffcl

ayv.

-8

00

fps

125'

San

dV

s...

11

00

fps

135'

Sti

ffcl

ayV

s...

11

00

fps

0.3

0.3

01

Cal

don

Cot

oP

orll

reco

rd(\

1157

)0

1

~C

oldo

nG

ato

Par

llre

cord

(\9

57

)Ic

alod

100.

0659

IIc

olo

d10

0.06

5g

225'

c(D

ampi

ngR

atio

-51

1)c

1(D

ampi

ngR

atio

-51

1)~

0.2

~0.

2~

~

19

00

fps

IIII

)

San

dan

dgr

avel

Va-

IiIi

00

0~

«28

5'~

_0

.\_

0.\

0~

1fA\."fk..YI/\*~AtlNI/.,\(/A,Y~

.. ........

00

Roc

kII

II

JI"a. III

0.0

0.0

0.0

1.0

1.5

2.0

0,0

0.5

1.0

\.5

2.0

-se

cond

aP

.rlo

d-

a.co

nd

a

FIG

.17

ANAL

YSIS

OFGR

OUND

RESP

ONSE

ATSO

UTHE

RNPA

CIFI

CBU

ILDI

NGSI

TEIN

SAN

FRAN

CISC

OEA

RTHQ

UAKE

,19

57W \0

Sh

ea

rW

ave

Ve

loci

ty,

fps

Vs

=8

00

fps

Vs

=2

80

to7

80

fps

Vs

=1

20

0fp

s

Vs

=1

40

0fp

s

Vs

=5

00

fps

San

dfil

l

Silt

ysa

nd

San

d

Gre

ycl

ayey

slit

Silt

ycl

ay

0' 20'---------------

160'-----------------

110'---------------

130'--------------

2000

1600

1200

800

400

oo

iI

II

II

IIii

it

7525 50 125

100

.... ..... ~15

0:5 0

.(j

) o17

5

200

225

210''71\Wfk.,Vi~AW.7,('\{~~YIl

Sha

lea

nd

san

dst

on

e

250

275

30

0!

II

II

II

I!

!I

FIG

.18

SOIL

PROF

ILE

AND

SHEA

RWA

VEVE

LOCI

TIES

ATEM

BARC

ADER

OCE

NTER

SITE

~ o

41

Ravenswood Site

This site was chosen as typical of large areas near the edge of the

southern end of San Francisco Bay. The generalized geological section at

the site consists of a surface layer of young Bay mud covering the older Bay

sediments which extend to bedrock at a depth of about 600 ft, as shown in

Fig. 19.

The younger Bay mud, about 33 ft thick at this site, is a soft, low

Some sand and silt layers are interspersed within theshear-strength clay.

Bay mud where stream channels crossed the mud flats. The older bay

sediments, beneath the younger Bay mud, are of late Pliocene to Holocene age

and were formed by alluvial processes. The older Bay sediments vary greatly

in composition but in general, they are much stiffer than the younger Bay

mud. The depth to bedrock, mainly sandstone and greywacke, is about 600 ft.

The shear wave velocity profile adopted for this site was based on a

downhole seismic survey (Warrick, 1974) together with some minor

modifications to the profile recommended by Joyner et al., (1976).

Characteristics of Rock Outcrop Motions in San Francisco

Two levels of ground motion were used to evaluate the possible effects

of earthquakes that may be generated on the San Andreas Fault. The lower

level was a Magnitude 7! earthquake with an estimated duration of shaking of

about 30 to 40 seconds and the higher level earthquake was a Magnitude 8+

earthquake with shaking lasting for about 70 to 80 seconds. The latter is

compatible with the duration of shaking reported for the 1906 earthquake.

Since no reliable near-field rock records are available for Magnitude 7! and

8+ earthquakes at the present time, artificial records were used to provide

rock outcrop motions for use in the analyses. The sites were considered to

Sh

ea

rW

ave

Ve

locit

y,

fps

04

00

80

01

20

01

60

02

00

0

On

II

II

II

II

,0' 3

6'

So

ftcl

oy

Vs

=2

62

fps

Allu

viu

m(o

lde

rV

s=

85

0fp

s1

00

f-

L1

Boy

sed

ime

nts

)

13

1'

Allu

viu

m(o

lde

rV

s=

11

00

fps

20

0I

L1

Ba

yse

dim

en

ts)

23

0'

41

0'---------------

30

0I-

..... '+- ..c ..... 0.

(l)

40

0r

L..

-a

50

0I-

-

Allu

viu

m(o

lde

rB

ayse

dim

en

ts)

Allu

viu

m(o

lde

rB

ayse

dim

en

ts)

Vs

=1

25

0fp

s

Vs

=1

40

0fp

s

60

0~

~_

70

0I

II

I!

II

It

II

60

4'/~&lJWlA.v~iMlNa..~

Bro

wn

san

dst

on

ea

nd

gra

ywa

cke

FIG

.19

SOIL

PROF

ILE

AND

SHEA

RWA

VEVE

LOCI

TIES

ATRA

VENS

WOO

DUS

GSSI

TE.p

oN

43

be located about 6 miles from the major fault systems, and an appropriate

attenuation relationship for rock motions in Western U. S. earthquakes was

adopted, as shown in Fig. 20 (Seed and Idriss, 1983), to estimate the peak

ground accelerations on rock outcrops. It has been shown (Bolt, 1973) that

the duration of strong ground motions generally depends on the earthquake

magnitude. Table 7, which lists the duration of strong shaking for

earthquakes with different magnitudes, was also used as a guideline in

generating the rock records considered representative for these two

earthquake magnitudes.

Table 8 lists the main characteristics of the rock motions used in the

analyses and Figs. 21 and 22 show the acceleration time histories for

Magnitudes Hand 8+ earthquakes respectively. In addition to assigning

appropriate values of acceleration amplitudes and durations, it has been

shown that different magnitudes of western U. S. earthquakes also produce

typical frequency characteristics (McGuire, 1974; Joyner and Boare, 1982,

1988; Sadigh, 1983; Sadigh et al., 1986; Idriss, 1985). Fig. 23 and Fig. 24

compare the response spectra for the two selected rock motions with the

normalized magnitude-dependent spectra proposed by Sadigh et aL, (1986).

It can be seen that the acceleration response spectra for the two motions

used are in good agreement with the spectra based on empirical experience.

Thus it was considered that the synthetic motions employed in this study

have the appropriate characteristics of natural earthquakes (of comparable

magnitudes and distances to faults), both in amplitude and frequency

characteristics.

Ground Response Analyses for San Francisco Bayshore Sites

Ground response analyses were performed for the three sites described

previously to study the site responses for the two selected levels of ground

10

050

ZONE

51

020

HO

RIZO

NTA

LD

ISTA

NCE

FROM

OFEN

ERGY

REL

EASE

.m

i

2

CLO

SEST

a~I

II

I!

!I

!I

I:=

;?J=

-i?

,?3

0.8I

I

01

~-

BASE

DON

AVAI

LABL

EDA

TA----

-z

--

----

ESTI

MA

TIO

N0 - I- ~ 0::: W -J W W W ~ -J

0.4

~ I- z:

0 N - 0::: 0 ::I:

0.2

~ ~ W 0..

FIG

.20

ATTE

NUAT

ION

OFGR

OUND

ACCE

LERA

TION

INRO

CKW

ITH

DIST

ANCE

FOR

EART

HQUA

KEW

ITH

VARI

OUS

MAG

NITU

DES

(aft

erSe

edan

dId

riss

,19

83)

~ ~

TABL

E7

TYPI

CAL

DURA

TION

SFO

REA

RTHQ

UAKE

SW

ITH

DIFF

EREN

TM

AGNI

TUDE

S(f

rom

Ger

ean

dSh

ah,

1984

)

Ave

rage

Ave

rage

Rad

ius

ofR

egio

nN

um

ber

ofN

umbe

ro

fD

urat

ion

Sub

ject

edto

Earth~akes

Ear

thqu

akes

ofS

tro

nl

Str

ong

Gro

und

Ric

hter

per

ear

Eer

100

Yea

rsG

roun

dS

hain

gS

haki

ngM

agni

tude

for

the

Wor

ldor

Cal

ifor

nia

(sec

onds

)(k

ilom

eter

s)

8.0

to8.

91

130

to90

80to

160

7.0

to7.

915

1220

to50

50to

120

6.0

to6.

914

080

10to

3020

to80

5.0

to5.

990

040

02

to15

5to

30

4.0

to4.

98,

000

2,00

0ot

o5

oto

15

.po

l./l

TABLE 8 CHARACTERISTICS OF ROCK OUTCROP MOTIONS USED INTHIS STUDY

Magnitude Distance to Peak Ground UurationFault Acceleration

7.1 6 miles 0.45 g 32 seconds4

8+ 6 miles 0.55 g 75 seconds

46

0.6

0.4

0'

0.2

c o :.p0.

0E Q) Qi u u

-0.2

«

-0.4

-0.6

o5

1015

202

5

Tim

e-

seco

nd

s30

354

0

FIG

.21

SYNT

HETI

CRE

CORD

FOR

MAG

NITU

DE7-

1/4

EART

HQUA

KEM

OTIO

NON

ROCK

ATA

DIST

ANCE

OF6

MIL

ES.l:

'- .....

0.6

0.4

01

0.2

c:: o :p0

.0o L- ~ Q

) u u-0

.2«

-0.4

-0.6

o10

203

0 Tim

e4

0.

50

seco

nd

s60

7080

FIG

.22

SYNT

HETI

CRE

CORD

FOR

MAG

NITU

DE8+

EART

HQUA

KEM

OTIO

NON

ROCK

ATA

DIST

ANCE

OF6

MIL

ES(m

odif

ied

afte

rSe

edan

dId

riss

,19

69)

.p-

00

5.0

4.5

4.0

for

M=

7-1

/4e

art

hq

ua

ke6

mile

s

Da

mp

ing

Ra

tio=

5~

1.5

2.0

2.5

3.0

3.5

Pe

rio

d-

seco

nd

s

1.0r

Ma

gn

itu

de

de

pe

nd

en

tsp

ectr

um

for

M=

7-1

/4

ea

rth

qu

ake

(mo

dif

ied

fro

mS

ad

igh

.1

98

5)

2.4

2.2

2.0

1.8

0'1

1.6

c:1.

40 :;:; 0 L.

1.2

Q.)

Q) u

1.0

0 «0.

80 L

.4

J0.

60 Q.

) a. (f)0.

4

0.2

0.0

0.0

0.5

FIG

.23

SPEC

TRAL

SHAP

ESFO

RM

AGNI

TUDE

7-1/

4EA

RTHQ

UAKE

MOT

ION

ONRO

CK

+='

1.0

Da

mp

ing

Ra

tio=

5~

Syn

the

tic

$-1

mo

tio

nfo

rM

=8

+e

art

hq

ua

kea

ta

dis

tan

ceo

f6

mir

es

Ma

gn

itu

de

de

pe

nd

en

tsp

ectr

um

for

M=

8+

ea

rth

qu

ake

(mo

dif

ied

fro

mS

ad

igh

.1

98

5)

5.0

4.5

4.0

1.5

2.0

2.5

3.0

3.5

Pe

rio

d~

seco

nd

s

1.0

2.4-

2.2 2.0

1.8

0'1

1.6

c1

.40 :;:; a L

1.2

Q.l

Q.l U

1.0

u

shaking.

51

Computed surface motions were expressed in terms of the

corresponding acceleration response spectra. Uncertainties in the measured

shear wave velocity profiles, which played an important role in the ground

response studies of the Mexico City sites (Seed, et al., 1987) were also

taken into account in the analyses.

Ground Response Analyses for Magnitude 71 EarthquakeNear San Francisco Bay Area

(1) Southern Pacific Building Site

Fig. 25 shows the soil profile for the Southern Pacific Building

site together with the response spectra for both the rock outcrop

motion and the computed surface motion for a hypothetical Magnitude 71

earthquake. The ground response analysis was made using the computer

program SHAKE-86 (after Schnabel et al., 1971). For a peak

acceleration of 0.45 g at a rock outcrop, the peak ground acceleration

was computed to be about 0.26 g. The maximum strain induced in the

soil profile was about 0.62% at a depth of 52 feet, while the site

period lengthened from 1.08 seconds to 1.66 seconds due to the non-

linear behavior of the soils.

As a result of the study for the Mexico City earthquake

(Seed et al., 1987), which indicated that a 10 per cent uncertainty in

the in-situ measured shear wave velocities can have significant effects

on the computed ground surface motions, the effects of a similar

variation in shear wave velocity on the computed surface motions were

also evaluated for this site. The results of this study are shown in

Fig. 26. It may be seen that the effect of a 10 per cent variation in

shear wave velocities does not produce as significant an effect for the

Southern Pacific Building site (shown in Fig. 26) as for the

Mexico City sites.

55'--------------

Acc

eler

ollo

nre

spon

sesp

ectr

umlo

rco

mpu

ted

grou

ndsu

rfoc

em

ollo

n.

IDom

plng

Rol

lo•

!loI)

1.8

2.0

:U3.

03.

lI4.

04.1

15.

0

Pen

od-

seco

nds

2.0

1.8

1.8

'" I1.

4c

~,g

l.2

e .!!1.

0

J0.8

Ii 1:l0.

8

" e5t0.

4

0.2

0.0

0.0

0.8

1.0

Vs...

325

fps

Vs...

680

fps

Sn

nd

fill

Sof

tcl

ay

Med

ium

clay

105"

------------

Stif

fcl

ayVs

..8

00

fps

12

5'_

....

....--~--__._.._____::""Z":lI:"_.:.-

135'

Sand

Va..

1100

fps

2115

'

22

5'-------------

San

dan

dgr

avel

Vs'"

19

00

fps

Acc

eler

oUon

rup

on

.e.p

ectr

um

for

rock

outc

rop

mot

ion

(Dam

ping

Rol

lo•

!loI)

1.0

1.5

2.0

2.8

3.0

3.lI

4.0

4.5

5.0

Pen

od-

seco

nds

2.0

1.8

1.S

'" 11.4

c ~l.2

~1.

0

~o.

s~

i0.'

x. VI0.

4

0.2

0.0

0•0

0.5

Vs...

1100

fps

Stif

fcl

ay

T1~'(lfl::.W~IN'<IM.1I7\'WM.JlW/...«M.'qJ..«.~Y11

Roc

k

FIG

.25

RESU

LTS

OFGR

OUND

RESP

ONSE

ANAL

YSES

FOR

SOUT

HERN

PACI

FIC

BUIL

DING

SITE

-M

AGNI

TUDE

7-1/

4EA

RTHQ

UAKE

ATA

DIST

ANCE

OFAB

OUT

6M

ILES

VI

N

5.0

4.5

4.0

Da

mp

ing

Ra

tio=

5~

1.5

2.0

2.5

3.0

3.5

Pe

rio

d-

seco

nd

s

1.0

2.0

1.8

1.6

0' , 1

.4

c ~1.

20 '- Q) Q;

1.0

0 0

54

(2) Embarcadero Center Site

Figure 27 shows the response spectrum for the computed surface

motions at the Embarcadero Center site together with that for the rock

outcrop motions used in the analyses. The computed peak ground

acceleration was 0.3 g, the maximum induced strain was about 0.537. at a

depth of 25 feet and the site period lengthened from 1.01 seconds to

1. 36 seconds due to the non-linear behavior of soils resulting from

seismic straining. A 10 per cent deviation from the measured shear

wave velocities showed only a small effect on the response spectrum for

the computed surface motions as shown in Fig. 28.

(3) Ravenswood Site

The results of the ground response analyses for the Ravenswood

site in the south Bay are presented in Fig. 29 in the same format as

before. The computed peak ground acceleration was 0.30 g, the maximum

strain along the profile was about 0.48% at a depth of 30 feet and the

site period was 2.54 seconds as computed from the strain-compatible

soil properties. The initial small-strain site period was 1.81 seconds

before seismic straining. The effect of a 10 per cent variation in

measured shear wave velocities on the computed surface motions is

relatively small, as shown in Fig. 30. It is interesting to observe,

however, that the effects of this small variation in the shear wave

velocities used in the soil profiles has a more noticeable influence on

spectral accelerations in the lower period range, say below one second,

than for the higher period range for all three sites included in this

study (see Fig. 4-20, Fig. 4-22 and Fig. 4-24).

20'--------

_

210'

I1t~

TJ\\

YIM.

.@I@

IXWt

\f/I

\W!l

\W/.

..\'

{fA~

/~\VJ

J.N

Sha

lea

nd

sand

ston

e

110'

San

dVa

-1

20

0fp

a13

0'

Silt

ycl

ayV

s=

80

0fp

s16

0'

Silt

ysa

ndVs

..1

40

0fp

sA

ccII

.rot

ion

reap

onll

Iplc

tru

mlo

rto

ckQ

utcr

opm

olio

n

(Dam

ping

Rat

la-

!loll

Acc

eler

alio

nre

spon

sesp

ectr

um

for

com

pule

dgr

ound

.urf

oce

mol

ion.

(Dom

plnu

Rat

io-~)

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Peri

od-

seco

nd.

1.0

l.5

2.0

2.5

3.0

'.5

4.0

4.5

5.0

Peri

od-

seco

nds

2.0

U U

'" I1.

4c

....~

1.2

f! ~1.

0u .'i

0.8

'0 1i0.

8

! en0.

4

0.2

0.0 0

.00.

5

2.0

U 1.1

...'" I

1.4

c i1.2

U1.

0u .'i

0.8

e 'U0

.'v n. en

0.4

0.2

0.0

0.0

0.5

Vs

'"2

80

to7

80

fps

Va-

50

0fp

aS

and

till

Gre

ycl

ayey

alit

O'

FIG

.27

RESU

LTS

OFGR

OUND

RESP

ONSE

ANAL

YSES

FOR

EMBA

RCAD

ERO

CENT

ERSI

TE-

MAG

NITU

DE7-

1/4

EART

HQUA

KEAT

ADI

STAN

CEOF

ABOU

T6

MIL

ESV

tV

t

5.0

4.5

4.0

Da

mp

ing

Rat

io=

5~

1.5

2.0

2.5

'3.

03.

5

Per

iod

-se

con

ds

1.0

2.0

1.8

1.6

01 I

1.4

c 0 : Q)

1.0

0 0

2.0 ut

Acc

eler

atio

nre

spon

sesp

ectr

um

for

com

pute

dgr

ound

surf

ace

mot

ions

1.8

(llo

mpl

n9R

allo

-:10

1)

'" I1.

4

0'c

So

ftcl

oyV

s""

26

2fp

s..

:31.

23

6'

l! -il1.

0

Allu

vium

(Old

erV

s""

85

0fp

su

Boy

sed

imen

ts)

.:J.0.

8(;

131'

-e0.

8

Allu

vium

(Old

erV

s""

u1

10

0fp

sQ

.

Boy

sed

imen

ts)

V)

0.4

230'

0.2

0.0

0.0

O.S

1.0

1.S

2.0

2.S

J.OJ.

S4.

04-

SS.

O

Allu

vium

(Old

erV

s=

12

50

fps

Peri

od-

seco

nds

Boy

sed

imen

ts)

2.0

410'

1.1~

Ace

.lero

tlon

roop

on..

opoc

lrum

lor

rock

outc

rop

mot

ion

Allu

vium

{Old

er1

.'V

s""

14

00

fps

'"(O

amp'

n9R

allo

-!!o

r)

Boy

sed

imen

ts)

11

.4

c :31.

2G

604'

..l:

11I\

~/k.

Yf)M

$N.7

A.WJ

A\VI

I\W/

Ji..

'l.7

I\W7

I\w7

I\\V

I1J

Bro

wn

san

dst

on

ean

dgr

ayw

acke

10

.8

X- V)0.

4

0.2

0.0

0.0

0.5

1.0

1.5

2.0

:La

3.0

3.1

4.0

4.5

s.o

Peri

od-

seco

nd.

FIG

.29

RESU

LTS

OFGR

OUND

RESP

ONSE

ANAL

YSES

FOR

RAVE

NSW

OOD

SITE

-M

AGNI

TUDE

7-1/

4EA

RTHQ

UAKE

ATA

DIST

ANCE

OFAB

OUT

6M

ILES

Vt

'-I

2.0

ID

ampi

ngR

atio

=5~

1.8

1.6

01 I

1.4

c: 0 +-1.

20 I..

.(1

)

"i>1.

00 0 «

0.8

0 I...

...... g

0.6

a. (f)0.

4

0.2I

-~------------~-

.0.

00.

00.

51.

01.

52.

02.

53.

03.

54.

04.

55.

0

Pe

rio

d-

seco

nd

s

FIG

.30

INFL

UENC

EOF

±10

PERC

ENT

VARI

ATIO

NIN

SHEA

RWA

VEVE

LOCI

TYON

RESP

ONSE

SPEC

TRUM

FOR

COM

PUTE

DGR

OUND

SURF

ACE

MOT

ION

ATRA

VENS

WOO

DSI

TE-

EART

HQUA

KEM

AGNI

TUDE

=7-

1/4

\Jl

00

59

Summary of Ground Response Analyses for Magnitude 7; Earthquake

The acceleration response spectra for the computed surface motions at

the three sites are summarized in Fig. 31. The peaks of the response

spectra for all three sites reached values of about 1.0 g. A representative

average spectrum is also shown in Fig. 31.

In the Applied Technology Conference study (ATC-3, 1978) three sets of

site conditions were established and recommended for use in seismic building

codes. The site conditions, which were later adopted by the Seismology

Committee of the Structural Engineers Association of California (SEAOC) in

their "Tentative Lateral Force Requirements" blue book, ranged from rock-

like material and shallow stiff sites (Sl), deep stiff and dense soil

conditions (S2), to clay and loose sand site conditions (S3). Following a

study of the damage in Mexico City during the Mexico earthquake of 1985, an

additional site condition S4 was added to these categories,as shown in Table

9, to represent the anticipated response on deeper deposits of soft clays.

Soft clays are described as clays with shear wave velocities generally in

the range of 200 to 500 fps and a tentative spectrum for S4 soils was

indicated. It is interesting to note that this response spectrum for a soil

profile containing more than 40 feet of clay (the S4 site condition) is in

excellent general agreement with the representative spectrum for Bayshore

sites computed in this study for a Magnitude 7; earthquake, as indicated in

Fig. 32.

The representative spectrum for bay-shore sites underlain by clay for

a Magnitude 7; earthquake near the San Francisco Bay area shows signifi-

cantly higher spectral acceleration values at periods less than about 2

seconds than the representative spectrum for the heavy damage area of

Mexico City in the 1985 Mexico earthquake, as shown in Fig. 33.

Rep

rese

nta

tiv

esp

ectr

um

(Dam

ping

Rat

io=

5~)

--

So

uth

ern

Pac

ific

Bui

ldin

gsi

te--

----

-E

mb

arca

der

osi

te-'

-'-'

-R

aven

swoo

dsi

te

2.0

1.8

1.6

0'

1.4

c 01.

2:;::

; 0 l.- ID1.

0I

AID U U

TABLE 9 SITE COEFFICIENTS RECOMMENDED BY SEAOC (1988)

61

Type

S1

Description

A soil profile with either:

(a) A rock-like material characterizedby a shear-wave velocity greater than2,500 feet per second or by other suit-able means of classification, or

(b) stiff or dense soil condition wherethe soil depth is less than 200 feet.

A soil profile with dense or stiff soilconditions, where the soil thicknessexceeds 200 feet, or a profile consist-ing of a thin layer of soft clay up to20 feet thick overlying rocklike material.

A soil profile 40 feet or more in depthcontaining more than 20 feet of 80ft tomedium stiff clay but not more than40 feet of soft clay.

A soil profile containing more than40 feet of soft clay. Alternativelysoils falling into this category mayhave a design spectrum determined byspecial geotechnical study reflectingthe soil specific conditions.

8 Factor

1.0

1.2

1.5

2.0

The site factor shall be established from properly substantiated geotechnicaldata. In locations where the 80il properties are not known in sufficientdetail to determine the soil profile type, soil profile 83 shall be usedunless the building official determines that 84 type soil may exist at thesite in which case 84 shall be used.

54

spec

trum

(198

8)Dam

ping

Rat

io=

5lI

5.0

4.5

4.0

Pro

pose

dre

pres

enta

tive

spec

trum

for

M=

7-1

/4ea

rthq

uake

1.5

2.0

2.5

3.0

3.5

Pe

rio

d-

seco

nd

s

1.0

..£SEA

OC"

.--

, ......

0.5

2.0

1.8

1.6

0\

1.4

c: .Q1.

2..... 0 L.. Q)

1.0

Q) u u ~

0.8

0 L.. (J0.

6Q

) a.. ({)0.

4

0.2

0.0

0.0 FI

G.

32RE

PRES

ENTA

TIVE

SPEC

TRA

FOR

BAYS

HORE

SITE

SIN

M=

7-1/

4EA

RTHQ

UAKE

OCCU

RRIN

GAT

ADI

STAN

CEOF

ABOU

T6

MIL

ES

a-

N

5.0

4.5

4.0

Da

mp

ing

Ra

tio=~

Re

pre

sen

tati

vesp

ect

rum

for

he

avy

da

ma

qe

are

ao

fM

exi

coC

ity(1

98

5)

Re

pre

sen

tati

vesp

ectr

um

for

Ba

ysh

ore

site

son

cla

yfo

rM

=7

-1/4

ea

rth

qu

ake

1.5

2.0

2.5

3.0

3.5

Pe

rio

d-

se

co

nd

s

2.0

1.8

1.6

0"1

1.4

c 01.

2:.;

:i 0 L- eu1.

0eu u u «

0.8

0 L- t)0.

6Q.

) a.

(/)0.

4

0.2

0.0

0.0

0.5

1.0

FIG

.33

COM

PARI

SON

OFHE

AVY

DAMA

GEAR

EAOF

MEX

ICO

CITY

(198

5)AN

DM

AGNI

TUDE

7-1/

4EA

RTHQ

UAKE

ACTI

NGNE

ARSA

NFR

ANCI

SCO

BAY

AREA

'"w

64

Ground Response Analyses for Magnitude 8+ EarthquakeNear San Francisco Bay Area

Analyses were also made to compute the response of the same three

sites to the rock motions corresponding to a Magnitude 8+ earthquake on the

San Andreas fault located at a distance of about 6 miles, and producing a

peak acceleration in rock of about 0.55 g. The rock motion record used in

these analyses was that shown in Fig. 22. Analyses were made us ing the

SHAKE-86 computer program.

(1) Southern Pacific Building Site

Fig. 34 shows the response spectra for the computed surface

motions and the rock outcrop motions used in the analysis for the

magnitude 8+ earthquake. The computed peak ground acceleration was

about 0.34 g, the maximum shear strain developed in the soil profile

was about 1% at a depth of 52 feet and the strain-compatible site

period was 1.88 seconds as compared to the low strain-level site period

of 1.08 seconds. The surface acceleration response spectrum for the

Magnitude 8+ earthquake (Fig. 34) is significantly higher than that for

the Magnitude 7! earthquake (Fig. 25). This is especially evident at

longer periods. The higher acceleration level, the longer duration of

the shaking, the more abundant long period motions in the 8+ rock

outcrop motion used in the analyses, and the larger strain-softening

effects of the soil, which reduce the ability of the soil to transmit

high frequency motions effectively, are some of the factors that

contribute to this higher response for the magnitude 8+ earthquake.

The effects of small (±10%) variations in the shear wave velocities of

the soils on the computed surface motions for the Southern Pacific

Building site (Fig. 35), although more evident than for the Magnitude

55'-------------

Acc

eler

atio

nre

.po

ns.

spec

trU

mfo

rco

mpu

ted

grou

ndsu

rfac

em

olln

ne

(1Ia

mIII

ntilo

ilo-

lie)

1.1

:t.O

:l.5

J.O

U4.

04

"1.

0P

erio

d-

.eco

nd

.

2.0

1.8

1.1

... 11.4

~f1.2 J: ~Q.I 10.4 0.2 0.0 0.

00.

1U