07/05/2012

TOPICSofLECTURESinSOILDYNAMICS(SD)
1. OverviewofSD.TheElastic1DoFSystem
2. TheInelasticSystems:Sliding,Overturning.
RetainingWallsandSlopes
i i
ll
d l
3. Review:SoilBehavior,WavePropagation,
Liquefaction ExamplesfromRecentEQs
4. [Link]
5. StiffnessandDampingofFoundations
6. SoilFoundationStructureInteraction
7. [Link]

SoilAmplification:AnalysisofCaseHistories

1)MEXICOCITY:1985MichoacanEarthquake
2)Hokkaido(2003), TreasureIsland(1989),
KobePortIsland(1995), etc.

07/05/2012

Even the most refined theories

(before
b f
they
h
can be
b established
bli h d)
must be VALIDATED by
COMPARISONS against the
REALITY that these theories

describe

The SIMPLER and more DIRECT is

the method of analysis,

the more convincing is the
interpretation of Reality
Reality .

07/05/2012

saturation

MS

MW

The Seismic Problem

Fault
Rupture:the
SOURCE

P
Propagation
i
throughSoil
FOCUS

07/05/2012

Effects of Soil on Ground Motion

S
(g)

S
(g)

S
(g)

C
t

DEPTH
0.1 km

Soil
Amplification

10 km

Wave
Propagation
Fault Rupture

SH Wave Propagation Through the SOIL

B
A

H
C

07/05/2012

B
A

H
C

B
A

H
C

07/05/2012

B
A

H
C

B
A

H
C

07/05/2012

B
A

H
C

B
A

H
C

07/05/2012

B
A

H
C

B
A

H
C

07/05/2012

B
A

H
C

Wave Propagation through the SOIL

B
VS = G/

Rock Outcrop

C
Base Rock

07/05/2012

A =

2 /

uA
uC

5%

10 %

f1

(2 / ) /3

f2

f3

f ( Hz)

f 1 = VS / 4H.2

10

Slide 19
.2

gazetas; 06/12/2009

07/05/2012

SOILAMPLIFICATION
HarmonicExcitation
2 /

a
Asurf
(T)
arock

x
H

, V

( I1, I2 )
1
T1
5

Period T

4H

T1
3

T1 = V
s

Propagation of seismic waves

Surface
0.43 gsurface

0.0

50
0

40
0

30
0

20
0

-0.5
0

10

15

20

t :s

0.5

z:m

BASE

a:g

10
0

VS : m / s
0

a:g

0.5

0.29 g outcropping bedrock

12

16

0.0

20
-0.5
0

10

15

20

24

t :s

11

07/05/2012

50
0

40
0

30
0

20
0

outcropping
Base

bedrock

10
0

1.5

SA : g

VS : m / s

Surface
surface

0.5

0
0

0.5

1.5

z:m

T:s

Dynamic
amplification

16

12

2
20

1
0
0.0

0.5

1.0

1.5

24

T:s

RESPONSEACCELERATIONSPECTRA

=?
1,5

Sa ( g )
0,5

0
0

0,2

0,4

0,6

0,8

T ( sec )

12

07/05/2012

TheMEXICOCITY1985Disaster:
InstrumentalObservation,Analysis

300 km
MEXICO

Gulf of
Mexico
Pacific
Ocean

13

07/05/2012

MEXICO 1985: MS = 8.1

Mexico
City

380 km
surprisingly

enormous damage,

small
damage

BUT only in a small

part of downtown
Mexico City

14

07/05/2012

G l i zones
Geologic
of
Mexico City

15

07/05/2012

Peak Accelerations , Velocities

SCT (0.19 g, 60 cm/s )

CDAO

(0.10 g,

UNAM
(0.04 g, 10 cm/s)

40 cm/s)

Subduction Zone arthquakes

1

SCT
0.20 Zone
0.1

0.06 Hilly
0.04 Zone

[g]
0.01

MS = 8
Empirical
E
i i l
Attenuation
Relations

0.01
0.005

Mexico
City

Statistically
y
expected

0.001
Distanse [km]

16

07/05/2012

Peak Accelerations , Velocities

SCT (0.19 g, 60 cm/s )

CDAO
UNAM
(0.04 g, 10 cm/s)

(0.10 g,

40 cm/s)

Epicentral
Region

UNAM
VIV
CDAO
(H60m)

SCT
(H40m)

17

07/05/2012

Comparison of Response Spectra

Epicentral
Region

Spectral
Acceleration :

ZoneC (SCT)

Sa /g

ZoneA
Period

T : sec

Mexico City Recorded Spectra

Sa
g

T : sec

18

07/05/2012

Mexico City CLAY

Natural Water Content

Mexico
Clay
Mexico Clay

Depth
(m)

wn 200%
600%

IP 200 +

HighlyCompressible:CC 6, Vs,max 40m/s

1 3
2

W%

undisturbed
max Su 80kPa

disturbed

l
log

Sensitivity :

Su undisturbed
Su disturbed

10

19

07/05/2012

Mexico City CLAY

G
1

G ()
max

Monotonic

20

07/05/2012

Summary of Experimental Results

( Dobry + Vucetic, Seed et al)

21

07/05/2012

Agreement with Simplest Theory

UNAM

SCT ??

H 40m

UNAM

Homogeneous Soil Layer,

Vertical Shear Waves

Soil Layer Properties: Seed 1985

UNAM

SCT

H 40
40 m

Fundamental soil period :

Ts =

Amplification : (Resonance) A1

4H
Vs

4 40
80

2
0.05

= 2 sec
= 12.7

22

07/05/2012

Comparison of Response Spectra

Spectral
Acceleration :

Epicentral
Region
A

LakeZone
(SCT)

SA /g

A1

uA
uC

0.08 g
Hills

Period

T : sec

Agreement with Simplest Theory

UNAM

SCT

H 40m

at the Natural Period of 2 sec

Sa, SCT

A1 xSa, UNAM 12.7 x 0.08 g 1.0g

23

07/05/2012

1.0 g
SCT

SA
g
CDAO

T : sec

2
UNAM

SCT

CDAO

DetailedAnalysis(Waves in Layered Soil)

G ()
()

24

07/05/2012

RESPONSEACCELERATIONSPECTRA

=?
1,5

Sa ( g )
0,5

0
0

0,2

0,4

0,6

0,8

T ( sec )

SIMPLE
Analysis

Detailed
Analyses

Average of Recorded
2 Components

DESIGN

25

07/05/2012

DOUBLE RESONANCE

But, to be successful,
a theory must
explain NOT ONE but
ALL aspects of reality.

26

07/05/2012

The Records

CDAO

Sa

??

??

T : sec

Soil Period :

Ts =

4H
Vs

Amplification at Resonance:

A1

4 60
70

3.4 sec
CDAO

Average
12.7
Records
Analysis

0.05
0 05

UNAM
SoWHYdontwehaveasharppeak
similartothatofSCTatT =3.4sec ??

27

07/05/2012

Comparison of Response Spectra

Epicentral
Region
A

Spectral
Acceleration
l
i
:

LakeZone
(SCT)

SA /g

0.08 g
0.03 g!

uA
A1
uC

Period

T : sec

Indeed, although the two Amplifications

at Resonance are the same:
A1

12 7
12.7

the values of Sa, rock that they amplify

are very different: 0.03g versus 0.08g
at T = 3.4 sec
Sa, CDAO

A1 xSa, UNAM 12.7 x 0.03 g 0.40g

28

07/05/2012

CDAO
Average
Records

Analysis

UNAM

AnotherRecordatStationCAF

CAF

UNAM

29

07/05/2012

The question marks are for the students

to explain (i.e., to analyse)

a double (remaining) question:

why damping 0.05 ?
how significant is the nature
off soil
il ?

30

07/05/2012

actual

31

07/05/2012

Some other recent Case

Histories
and Concepts

PORTISLAND:Kobe1995

accelerographs

32

07/05/2012

KOBE1995:RECORDS

uA
uC
I

uA
uB

Amax

rockVR
soilVS

Amax

1
I

-1

( /2)

33

07/05/2012

KINEMATIC DISTRESS of an ACTUAL PILE

in Hokkaido Japan during an Earthquake
Adapted from : Y. Miyamoto & K. Koyamada (2007)

Tokachi-oki Earthquake 2003 , M = 7.9

Konan Junior High School ( R = 240 km ):

footings on piles
accelerograms: at 0 m and 153 m depth !!

TOKACHI-oki 2003 Earthquake: M = 7.9 ....

2003 /09/26

34

07/05/2012

KONAN High School

AccelerographArray

50 m

HighSchool

SOIL LAYERS and PHC (d=0.40 m) PILE

0.35 g

60 90

190

Vs [ m/s ]

0m

Peat

6m

Clay

Shear wave
velocity

20 m

Sandy
Silt

30 m

0.05 g

400 m/s

153 m
Sandstone

35

07/05/2012

ACCELEROGRAMS
0.35 g

Recorded at
Ground Surface

cm / s2

Recorded at -153 m
(Input Motion)

cm / s2

0.05 g
50

sec
100

150

Response Spectra ( 5 % damping)

0m

0.35 g

153 m

0.05 g

[DrawtheanalogywiththeMexicoCity1985
spectraofthemotionsrecordedat
SCT(soilsurface)versusUNAM(rockoutcrop)!!]

36

07/05/2012

Vs (m/s)

Peat

60

6
Clay

20

90

m
190

SiltySand

30
Gravel

40

320

LomaPrieta
1989

Treasure Island (E-W)

SA : g

ACC. : cm/s 2

Yerba Buena Island (E

(E-W)
W)
Treasure
Island

Yerba Buena
Island
Time : s

Period : s
Infrastructure Group Lecture, May 25, 2011

37

07/05/2012

PGA:ROCKversus SOIL
0.6

PGA

on Soil

(g)

1989 Loma Prieta

0.4
0.2

1985MexicoCity

0
0

0.2
PGA on

0.4

0.6

Rock (g)

Infrastructure Group Lecture, May 25, 2011

2 D Valley
2-D
V ll
(B
(Basin)
i ) Eff
Effects
t

38

07/05/2012

1-D
B

2-D
A
C

2 D Aggravation
gg
Factor :

AF

= UA / UB

AF = f (valley geometry, location, frequency)

Findings of last 20 years show that :

Inalluvialbasins:1D soilamplificationoften
underpredicts motionatgroundsurface!

2D,3Dwaveeffects areevident:
in records (mostlyofweakmotions
mostly of weak motions),
),
inrecords
linearanalyses ,
distribution+extentofobserveddamage

39

07/05/2012

Two main reasons for aggravation

(a) Multiple Wave Reflections at the Valley edges

AF > 1

15o
75o

60o

60o

300

60o

SanchezSesma (1985)

(b)

Generation of Surface Waves

Surface Waves

AF > 1

Surface Waves
SH waves

SV waves

Aggravationatthecentre:
constructiveinterferenceofLove orRayleigh waves

40

07/05/2012

Scope

ThisStudyexplorestheSensitivity ofValley
Effectsand2DAggravationphenomenato:

excitation frequency

soil nonlinearity

Ohba Valley in Futsizawa, Japan

Cross section
of Ohba Valley

L = 520 m
80 m

360 m

80 m

H= 24 m

Simplified geometry

41

07/05/2012

Numerical model / soil properties

80 m

360 m

24 m
m/s

36 m

80 m

VS = 60
VR= 400 m/s

Absorbing boundary

1-D
B

2-D
A
C

2 D Aggravation
gg
Factor :

AF

= UA / UB

AF = f (valley geometry, location, frequency)

42

07/05/2012

Linear response of the valley to

Ricker wavelets

Excitation Pulses
Time histories

Fourier Spectra

1.0
3.0

Ricker 3

0.5
0

2.0
1.0

a [ m / s2 ]

- 0.5
0

1.0

Ricker 0.5

0.5

10

10

10

20.0
15.0
10.0

5.0
- 0.5
05
0

1.0
8.0

Ricker 1

0.5

6.0
4.0

2.0

- 0.5
0

t [sec]

f [sec-1]

43

07/05/2012

Input: Ricker 3

largest Aggravation: 2
2.5

AF

2
1.5
1
0.5
0

-300

-200

-100

100

200

300

Input: Ricker 0.5

largest Aggravation: > 1.7

AF

1
0
-300

-100

100

300

44

07/05/2012

Input: Ricker 1
largest Aggravation: 1.7
2
1,6

AF

1,2
0,8
30 m

0,4
0
-300

-200

-100

100

200

300

Horizontal Waveforms - Ricker 1

X (m)

Ca

R2 R1
70 m/s

200

100

200 m/s

X(m
)

-100

110 m/s
-200

t (s)

45

07/05/2012

NonLinear response of the

valley to
Ricker wavelets

Soil Elasticity

Excitation
Ricker 3:
Ricker 0.5:

PGA =

0.30 g ,

PI = 50

PGA =

0.30 g ,

PI = 50

Elastic vs Non-linear Ricker 3

Aggravation

AF

Elastic

2.0

Nonlinear
1.5
1.0
0.5

- 200

- 100

100

200

X[m]

Ricker 3

46

07/05/2012

Elastic
Aggravation

vs

Non-linear Ricker 0.5

AF
2.0
16
1.6
1.2
0.8

- 200

- 100

0.4

200

X[m]

Ricker 0.5

Elastic
X[m]

vs Non-linear Ricker 0.5

waveforms

Elastic

t[s]

Non-linear

t[s]

47

07/05/2012

Conclusions

Valley effects are significant IN GENERAL

However :

1. They are extremely sensitive to

frequency content of input motion

2. They reduce drastically with increasing

soil non-linearity

HE END

48

07/05/2012

Two main reasons for aggravation

(a) Multiple Wave Reflections at the Valley edges

AF > 1

15o
75o

60o

60o

300

60o

SanchezSesma (1985)

(b)

Generation of Surface Waves

Surface Waves

AF > 1

Surface Waves
SH waves

SV waves
Aggravation at the centre:
constructive interference of Love or Rayleigh waves

49

07/05/2012

G / GO

PI

PI

Horizontal Waveforms - Ricker 3

Converted Excel Data

X (m)

Ca

R2

Ca

R1

R1

R2

200 m/s

60 m/s
85 m/s

t (s)

50

07/05/2012

Horizontal Waveforms - Ricker 0.5

X (m)

R1

R1

200

230 m/s

X
(m
)

100

-100

-200

2
R1

t (s)

t (s)

Basic Conclusions

Regardlessofexcitationfrequency:absolutely
LARGEST
LARGESTaggravationinPGA:
ti i PGA
AF

Spatialdistributionofaggravation: SENSITIVEto
FREQUENCYofexcitation:
( ) Low frequency input : significant aggravation
(a)

at center (Rayleigh waves interfere with SV)

(b) High frequency input : significant aggravation

near the edges (wave focusing effects)

51