Comparisons of Darfield and Christchurch Ground Motions with NGA-W1 GMPEs

Comparisons of Darfield and Christchurch Ground Motions with

NGA-W1 GMPEs

David M. BooreWorkshop on Update of Pacific Northwest Portion of the U.S. National Seismic Hazard

Maps (NSHMs)March 21-22, University of Washington, Seattle

(http://www.geonet.org.nz/canterbury-quakes/)

Stations for which Rjb<= 200 km

170o 171o 172o 173o 174o-45o

-44o40'

-44o20'

-44o

-43o40'

-43o20'

-43o

-42o40'

-42o20'

-42o

-41o40'

Longitude

La

titu

de

Christchurch (M 6.1)NEHRP class DNEHRP class Chypocenter

File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_st

atio

n_

ma

p_

r_le

_2

00

.dra

w;

Da

te:

20

12

-03

-17

;T

ime

:1

6:4

1:4

6

Most class C sites to the north

Class C sites at greater distance may have smaller depth of sediments than those near Christchurch

1 10 100 1000

200

300

400

500

1000

RJB (km)

VS

30

(m/s

)

DarfieldChristchurchNEHRP B

NEHRP C

NEHRP D

NEHRP E

File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_d

arf

ield

_vs

30

_vs

_rj

b.d

raw

;D

ate

:2

01

2-0

3-1

7;

Tim

e:

16

:40

:20

Note: Some sites classified as “E” by Bradley & Cuprinovski (2011)

• Many sites with Vs30 = 275 and 660 m/s because values were estimated, not measured

• Distributions of Vs30 similar for both events

• Most records with R<100 km are from class D sites

Comparison of Darfield and Christchurch Ground Motions

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=0

.2s)

(g)

NEHRP Class D (Darfield, M 7.0)NEHRP Class D (Christchurch, M 6.1)NEHRP Class C (Darfield, M 7.0)NEHRP Class C (Christchurch, M 6.1)

File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_d

arf

ield

_p

sa_

t0.2

_o

bs.

dra

w;

Da

te:

20

12

-03

-17

;T

ime

:1

6:1

7:3

0

Observed RotD50 reduced to GMRotI50 using factors from Boore (2010); maximum effect is 4% at T=5 s.

In this and subsequent comparisons, the Y-axis spans three orders of magnitude.

Concentrate on H components, as the 2008 NGA GMPEs were only for H.

• Motions from both events at close distances are comparable

• Apparently no or small site effect (but note difference in spatial locations for different class sites, so source effects could compensate for site effects)

• Nonlinear soil response could have reduced motions for the larger event

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=1

.0s)

(g)


File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_d

arf

ield

_p

sa_

t1.0

_o

bs.

dra

w;

Da

te:

20

12

-03

-17

;T

ime

:1

6:1

9:2

1

• Christchurch (M 6.1) class D motions comparable or even larger than Darfield (M 7.0) at close distances

• Darfield class C motions greater than Christchurch motions (R>50 km)

1 2 10 20 100 200

0.001

0.01

0.1

RJB (km)

5%

-da

mp

ed

PS

A(T

=5

.0s)

(g)


File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_d

arf

ield

_p

sa_

t5.0

_o

bs.

dra

w;

Da

te:

20

12

-03

-17

;T

ime

:1

6:2

3:1

3

• Christchurch (M 6.1) class D motions now smaller than Darfield (M 7.0) at all distances

• Apparent site effect small or not existent

Comparison of Observed Motions and Motions from NGA-

W1 GMPEs: Darfield

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=0

.2s)

(g)

NEHRP Class DNEHRP Class CAS08, Vs30=275 m/sBA08, Vs30=275 m/sCB08, Vs30=275 m/sCY08, Vs30=275 m/sAS08, Vs30=660 m/sBA08, Vs30=660 m/sCB08, Vs30=660 m/sCY08, Vs30=660 m/s

Darfield (M 7.0)

File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\d

arf

ield

_p

sa_

t0.2

_o

bs_

ng

a_

r_le

_2

00

.gm

pe

s_b

lue

_re

d.d

raw

;D

ate

:2

01

2-0

3-2

1;

Tim

e:

14

:23

:44

• Overall comparison good

• Apparent lack of site effect in data is consistent with GMPEs

• Reddish GMPEs for class D

• Bluish GMPEs for class C

• Don’t try to follow curves for individual GMPEs

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=1

.0s)

(g)


Darfield (M 7.0)

File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\d

arf

ield

_p

sa_

t1.0

_o

bs_

ng

a_

r_le

_2

00

.gm

pe

s_b

lue

_re

d.d

raw

;D

ate

:2

01

2-0

3-2

1;

Tim

e:

14

:25

:06

• Note separation in observed class D values; GMPEs predictions between the two groups

• Site effect in GMPEs, but not apparent in observations (but little overlap in distance range for class C and D)

1 2 10 20 100 200

0.001

0.01

0.1

RJB (km)

5%

-da

mp

ed

PS

A(T

=5

.0s)

(g)


Darfield (M 7.0)

File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\d

arf

ield

_p

sa_

t5.0

_o

bs_

ng

a_

r_le

_2

00

.gm

pe

s_b

lue

_re

d.d

raw

;D

ate

:2

01

2-0

3-2

1;

Tim

e:

14

:26

:32

• GMPEs underpredict observations (except class C at greater distances)


Comparison of Observed Motions and Motions from NGA-

W1 GMPEs: Christchurch

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=0

.2)

(g)


Christchurch (M 6.1)

File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

psa

_t0

.2_

ob

s_n

ga

_r_

le_

20

0.g

mp

es_

blu

e_

red

.dra

w;

Da

te:

20

12

-03

-21

;T

ime

:0

6:3

6:0

5

• Overall comparison good

• Apparent lack of site effect in data is consistent with GMPEs

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=1

.0)

(g)



File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

psa

_t1

.0_

ob

s_n

ga

_r_

le_

20

0.g

mp

es_

blu

e_

red

.dra

w;

Da

te:

20

12

-03

-21

;T

ime

:0

6:3

6:3

2

• GMPEs tend to underpredict class D motions at close distances


1 2 10 20 100 200

0.001

0.01

0.1

RJB (km)

5%

-da

mp

ed

PS

A(T

=5

.0)

(g)



File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

psa

_t5

.0_

ob

s_n

ga

_r_

le_

20

0.g

mp

es_

blu

e_

red

.dra

w;

Da

te:

20

12

-03

-21

;T

ime

:0

6:3

7:1

3

• GMPEs severely underpredict observations at shorter distances

• Agreement better for greater distances


Sensitivity of Predicted T=5 s PSA to sediment depth and to

magnitude

• Show AS08 and class D only

• Z1.0=1000 m from Bradley & Cuprinovski (2011)

• M 6.33 from Holden (2011)

• Use of these data in NGA-W2 without Z1.0 and larger M could result in biased results.

• Does Z1.0 vary spatially, with it being smaller for stations at greater distances? If so, this would help explain the discrepancy at greater distance. This could also be due to a difference in geometrical spreading due to lateral changes in crustal structure.

1 2 10 20 100 200

0.001

0.01

0.1

RJB (km)

5%-d

ampe

dP

SA

(T=

5.0)

(g)

NEHRP Class DAS08, Vs30=275 m/s, Z1.0=480 mAS08, Vs30=275 m/s, Z1.0=1000 mAS08, as above, M 6.2AS08, as above, M 6.33


File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_p

sa_

t5.0

_o

bs_

ng

a_

r_le

_2

00

.cla

ss_

d_

as_

on

ly.c

om

pa

re_

zse

d.a

dd

_m

6.2

_6

.33

.dra

w;D

ate

:20

12

-03

-21

;Tim

e:

06

:39

:47

Vertical Motion

(Fry et al., 2011)

(Bradley & Cuprinovski, 2011)

Effects Producing Spatial Variability in Ground Motions

•Source: Radiation Pattern & Directivity•Path: volcanic vs sediments•Basin Waves•Sediment Depth•Shallow Site Response

•Linear•Nonlinear

LPCCLPCC

HVSCHVSC

65 Dip

SHLC, PRPC,CMHSSHLC, PRPC,CMHS

Fault Normal and Fault Parallel Velocity Time Series

0 20 40 60 80

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

LPCC,Fault Normal,158

0 20 40 60 80

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

LPCC,Fault Parallel,68

0 20 40 60 80

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

HVSC,Fault Normal,158

0 20 40 60 80

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

HVSC,Fault Parallel,68

0 20 40 60 80

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

CMHS,Fault Normal,158

0 20 40 60 80

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

CMHS,Fault Parallel,68

0 10 20 30 40 50

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

PRPC,Fault Normal,158

0 10 20 30 40 50

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

PRPC,Fault Parallel,68

0 10 20 30 40

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

SHLC,Fault Normal,158

0 10 20 30 40

-100

050

100

Time (Sec)

Vel

ocity

(cm

/s)

SHLC,Fault Parallel,68 CMHSCMHSPRPCPRPCSHLCSHLC

• Drastic change in frequency content going from sites on rock (HVSC & LPCC) to sites on Quaternary sediments.

(from B. Chiou)

Evidence for Nonlinear Soil Response

10 15 20 25-1

-0.5

0

0.5

1

as7

7w

g

NNBS

10 15 20 25-1

-0.5

0

0.5a

s13

eg

10 15 20 25-1

-0.5

0

0.5

1

Time (s)

au

pg

File

:C

:\n

ew

_ze

ala

nd

_ch

rist

chu

rch

_2

011

\20

110

22

1_

23

51

42

_N

NB

S_

V2

A.x

ls.d

raw

;D

ate

:2

01

2-0

3-1

7;

Tim

e:

21

:09

:00

20 30 40 50-400

-200

0

200

400

Acc

el(

cm/s

/s)

2001 M 6.8 Nisqually (SDS)

NS

20 30 40 50-400

-200

0

200

400A

cce

l(cm

/s/s

) EW

20 30 40 50-200

-100

0

100

200

Time (s)

Acc

el(

cm/s

/s) UD F

ile:

C:\

nis

qu

ally

_2

00

1\s

ds_

un

filt.

dra

w;

Da

te:

20

12

-03

-18

;T

ime

:0

8:0

9:2

9

(see Frankel et al., 2002, for details)

(Fry et al., 2011)

Negative vertical accelerations are “clipped”. This may be due to a different nonlinear process than that producing the cusps shown in the previous figures.

Conclusions• M 7.0 Darfield and M 6.1 Christchurch motions similar

for close distances, short periods• M 7.0 Darfield motions higher than M 6.1 Christchurch

motions for longer periods (as expected from the difference in magnitudes)

• Site response not too obvious, but this may be because of the different spatial distributions of the site classes (most close sites are class D)

• Observed motions influenced by many effects, including• Lateral changes in geology• Local linear and nonlinear site response• Basin waves (?)

• GMPEs are in reasonable agreement with observations for close distances, short periods

• GMPEs underpredict longer period motions, using metadata in current NGA-W2 flatfile

END

From B. Chiou, Source: GNS Science



LPCCLPCC

HVSCHVSC

Source: George Walker

65 Dip

SHLC, PRPC,CMHSSHLC, PRPC,CMHS

Directivity Effect and Velocity Pulse

(from B. Chiou)



172.4 172.6 172.8 173

-43.

8-4

3.6

-43.

4

Longitude (°)

Latit

ude

(°)

0282 Christchurch, New Zealand

SHLC

PRPC

CMSHHVSC

LPCC

(from B. Chiou)

Fault Rupture

• Reverse faulting on a buried fault• Assumed fault plane

– Strike = 68 (from USGS CMT)– 65 dip, to the south– Top of rupture is at 2 km depth (assumed)– Bottom of rupture is at 12 km (assumed)– Rupture length ~ 15 km (length of the aftershock

zone).

(from B. Chiou)

20 30 40 50-300

-200

-100

0

100

200

300

Acc

ele

ratio

n(c

m/s

ec2

)

2001 Nisqually, Washington, earthquake (M 6.8)

20 30 40 50

-50

0

50

20 30 40 50-300

-200

-100

0

100

200

300

Time (sec)

Acc

ele

ratio

n(c

m/s

ec2

)

20 30 40 50

-50

0

50

Time (sec)

File

:C

:\m

etu

_0

3\r

ec

_p

roc

_s

tro

ng

_m

oti

on

\sd

s_

sd

n_

un

filt

_fi

lt.d

raw

;Da

te:2

00

3-0

9-1

7;

Tim

e:

11:0

4:4

1

SDS (NS): unfiltered SDS: filtered from 10--20 Hz

SDN: filtered from 10--20 HzSDN (NS): unfiltered

• SDS within 200 m of SDN• liquefaction at SDS, not at SDN• Note cusps at SDS and increased amplitude at high frequencies

(after Frankel et al., 2002)

1 2 10 20 100 200

0.001

0.01

0.1

RJB (km)

5%

-da

mp

ed

PS

A(T

=5

.0)

(g)

NEHRP Class DNEHRP Class CAS08, Vs30=275 m/s, Z1.0=480 mAS08, Vs30=275 m/s, Z1.0=1000 mCB08, Vs30=275 m/s, Z1.0=326 mCB08, Vs30=275 m/s, Z1.0=1000 mCY08, Vs30=275 m/s, Z2.5=2.5 kmCY08, Vs30=275 m/s, Z2.5=4.1 km


File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_p

sa_

t5.0

_o

bs_

ng

a_

r_le

_2

00

.gm

pe

s_b

lue

_re

d.c

om

pa

re_

zse

d.d

raw

;D

ate

:2

01

2-0

3-1

7;

Tim

e:

15

:39

:35

1 2 10 20 100 200

0.001

0.01

0.1

RJB (km)

5%

-da

mp

ed

PS

A(T

=5

.0)

(g)

NEHRP Class DNEHRP Class CAS08, Vs30=275 m/s, Z1.0=480 mAS08, Vs30=275 m/s, Z1.0=1000 mAS08, as above, M 6.2CB08, Vs30=275 m/s, Z1.0=326 mCB08, Vs30=275 m/s, Z1.0=1000 mCY08, Vs30=275 m/s, Z2.5=2.5 kmCY08, Vs30=275 m/s, Z2.5=4.1 km


File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

psa

_t5

.0_

ob

s_n

ga

_r_

le_

20

0.g

mp

es_

blu

e_

red

.co

mp

are

_zs

ed

.ad

d_

m6

.2.d

raw

;D

ate

:2

01

2-0

3-1

7;

Tim

e:

15

:40

:01


Stations for which Rjb<= 200 km

169o 170o 171o 172o 173o 174o-45o

-44o40'

-44o20'

-44o

-43o40'

-43o20'

-43o

-42o40'

-42o20'

-42o

-41o40'

Longitude

Latit

ude

Darfield (M 7.0)NEHRP class DNEHRP class Chypocenter

File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\da

rfie

ld_

sta

tion

_m

ap

_r_

le_

20

0.d

raw

;Da

te:2

01

2-0

3-1

7;T

ime

:1

6:4

1:1

5

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

PG

A(g

)


File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

da

rfie

ld_

pg

a_

ob

s.d

raw

;D

ate

:2

01

2-0

3-1

7;

Tim

e:

16

:15

:29

1 2 10 20 100 2000.001

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=2

.0s)

(g)


File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_d

arf

ield

_p

sa_

t2.0

_o

bs.

dra

w;

Da

te:

20

12

-03

-17

;T

ime

:1

6:2

1:2

6

1 2 10 20 100 200

1

10

100

RJB (km)

PG

V(c

m/s

)


File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\ch

rist

chu

rch

_d

arf

ield

_p

gv_

ob

s.d

raw

;D

ate

:2

01

2-0

3-1

7;

Tim

e:

16

:15

:02

Observed RotD50 reduced to GMRotI50 using factors from Boore (2010); maximum effect is 4% at T=5 s.

In this and subsequent comparisons, the Y-axis spans three orders of magnitude.

Concentrate oh H components, as the 2008 NGA GMPEs were only for H.

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

PG

A(g

)


Darfield (M 7.0)

File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\da

rfie

ld_

pg

a_

ob

s_n

ga

_r_

le_

20

0.g

mp

es_

blu

e_

red

.dra

w;

Da

te:

20

12

-03

-20

;T

ime

:2

1:1

7:3

9

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=2

.0s)

(g)


Darfield (M 7.0)

File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\d

arf

ield

_p

sa_

t2.0

_o

bs_

ng

a_

r_le

_2

00

.gm

pe

s_b

lue

_re

d.d

raw

;D

ate

:2

01

2-0

3-2

0;

Tim

e:

21

:16

:27

1 2 10 20 100 200

1

10

100

RJB (km)

PG

V(c

m/s

)


Darfield (M 7.0)

File

:C:\

pa

cnw

_w

ork

sho

p_

21

-22

ma

r12

\da

rfie

ld_

pg

v_o

bs_

ng

a_

r_le

_2

00

.gm

pe

s_b

lue

_re

d.d

raw

;D

ate

:2

01

2-0

3-2

0;

Tim

e:

21

:16

:02

1 2 10 20 100 200

0.01

0.1

1

RJB (km)

PG

A(g

)



File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

pg

a_

ob

s_n

ga

_r_

le_

20

0.g

mp

es_

blu

e_

red

.dra

w;

Da

te:

20

12

-03

-21

;T

ime

:0

6:3

5:3

9

1 2 10 20 100 2000.001

0.01

0.1

1

RJB (km)

5%

-da

mp

ed

PS

A(T

=2

.0)

(g)



File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

psa

_t2

.0_

ob

s_n

ga

_r_

le_

20

0.g

mp

es_

blu

e_

red

.dra

w;

Da

te:

20

12

-03

-21

;T

ime

:0

6:3

6:4

9

1 2 10 20 100 200

1

10

100

RJB (km)

PG

V(c

m/s

)



File

:C

:\p

acn

w_

wo

rksh

op

_2

1-2

2m

ar1

2\c

hri

stch

urc

h_

pg

v_o

bs_

ng

a_

r_le

_2

00

.gm

pe

s_b

lue

_re

d.d

raw

;D

ate

:2

01

2-0

3-2

1;

Tim

e:

06

:37

:36

(Fry et al., 2011)

Documents

Comparisons of Darfield and Christchurch Ground Motions with NGA-W1 GMPEs