2d microtremor array

Using Two-station Microtremor Array

Method to Estimate Shear-wave Velocity

Profiles in Seattle and Olympia,

Using Two-station Microtremor Array Method to Estimate Shear-wave Velocity Profiles in Seattle and

Olympia, Washington1

Washington

*Koichi Hayashi (Geometrics)

Recep Cakir (Washington State Department of Natural Resources)

Timothy J. Walsh (Washington State Department of Natural Resources)

Outline



• Introduction

• Spatial Auto-correlation

• Data acquisition and processing

• Dispersion curve analysis in terms of Spatial

Autocorrelation

• Conclusions

Introduction (1)



• Surface ground motion of earthquakes highly depends on

subsurface geological structure.

• For estimating the local site effect, S-wave velocity (Vs) down to a

depth of several 10m (e.g. AVS30) is most popular in all over the

world.

• Recent several sever earthquakes have revealed that much deeper

and two or three dimensional structures also play important role.

• Reflection and refraction methods have been applied to delineate

deeper Vs structure.

• These conventional methods are expensive and time consuming so

that the development of cheaper and simpler methods has been

desired.

• Active and passive surface wave methods have been increasingly

getting popular in last 10 years.

Introduction (2)



• The passive method or microtremor array measurements

(MAM) is particularly attractive because the method does not

require any artificial source and the depth of investigation can

be increased easily.

• Large scale microtremor measurements have been widely used

in last 10 years in Japan for estimating S-wave velocity

structure down to a depth of several kilometers.

• In these investigations, triangle arrays with size of several

kilometers are used for calculating a phase velocity in

frequency range from 0.2 to 1Hz.

• The investigation using the large scale microtremor

measurements revealed that abrupt change of deep bedrock

depth caused disaster concentration in the 1995 Earthquake in

Kobe, Japan.

Spatial Auto-correlation (2D)

ReceiverCentral receiver

Micro-tremors does not propagate with particular direction

Receiver

Receiver

Receiver Receiver

Receiver



Spatial auto correlation (2D)

Receiver

Infinite apparent velocity (wave-number is 0)Using Two-station Microtremor Array Method to Estimate Shear-wave Velocity Profiles in Seattle and


Center receiver

Direction of Propagation

Coherence

12 14 16

Frequency (Hz)

1

0.8

0.6

0.4

0.2

0

- 0.2

- 0.4

- 0.6

- 0.8

- 1

0 2 4 6 8 10


Center receiver


Receiver

High apparent velocity (small wave-number)

Coherence

12 14 16

Frequency (Hz)



1

0.8

0.6

0.4

0.2

0

- 0.2

- 0.4

- 0.6

- 0.8

- 1

0 2 4 6 8 10


ReceiverCenter receiver

Coherence


Frequency (Hz)

1

0.8

0.6

0.4

0.2

0

- 0.2

- 0.4

- 0.6

- 0.8

- 1

Low apparent velocity (large wave-number)



0 2 4 6 8 10 12 14 16


Receiver

Receiver

Receiver Receiver

Receiver

-1

- 0.8

- 0.6

- 0.4

- 0.2

0

0.2

0.4

0.6

0 2 4 6 8 10 12 14 16

Spatial Auto-correlation (2D) Bessel function

Averaging all direction coherence (cos function）results in Bessel functionUsing Two-station Microtremor Array Method to Estimate Shear-wave Velocity Profiles in Seattle and


0.8

1

Coherence


Frequency (Hz)

SPAC with Small Number of Sensors



• Most people use a spatial autocorrelation (SPAC) method for

calculating phase velocities from ambient noise and the method

requires at least 4 or 7 sensors placed on center and corners of

triangles.

• Margaryan et al. (2009) showed that the SPAC using only two

sensors yields almost identical phase velocities as one of 4 or 7

sensors on triangles.

• The SPAC using small number of sensors enables us to perform the

MAM much easily.

• Hayashi and Underwood (2012a) have shown that S-wave velocity

profiles to a depth of 2 to 3 km can be determined by the SPAC

using two sensors in the San Francisco Bay area

• We have performed the SPAC using two sensors at several sites in

Seattle and Olympia, Washington.


Averaging all direction coherence (cos function）results in Bessel function

Receiver

Receiver

Receiver Receiver

Receiver

- 0.6

- 0.8

-1

Frequency (Hz)

- 0.4

- 0.2

0

0 2 4 6 8 10 12 14 16

Spatial Auto-correlation (2D) Bessel function



0.8

0.6

0.4

0.2

1

Coherence




Averaging all direction coherence (cos function）results in Bessel function

- 0.4

- 0.6

- 0.8

-1

Frequency (Hz)

- 0.2

0.8

0.6

0.4

0.2

0

1

0 2 4 6 8 10 12 14 16

Coherence

Bessel function




Data Acquisition and Processing



• Microtremor array measurements have been performed at

three sites in Seattle and two sites in Olympia.

• Separation of two seismographs varies from 10 to 3102m.

• In each measurement, 10 to 60 minutes ambient noise was

recorded.

• Sampling time is 10msec.

• Two seismographs including three-component

accelerometer (McSEIS-MT Neo) made by OYO

Corporation were used for data acquisition.

• The seismographs include GPS clock and two

seismographs can be synchronized in any distance.

S'te of Investigation

14Using Two-station :tv:ricrotremor .ArrayMethod toEstimate Shear-wave Velocity Profiles in Seattle and

Olympia, Washington

Examp e ofarray configuration

15Using Two-station :tv:ricrotremor .ArrayMethod toEstimate Shear-wave Velocity Profiles in Seattle and

Olympia, Washington

Seismograph (McSEIS-MT Neo)

Sensor : Servo-Accelerometer

Resolution : 1μG

Sensitivity : 2.0V/G

Range : +/-4G

Frequency Response : 0.1 – 200Hz

Dynamic range : 120dB(Measured)

A/D Converter : 32bits

Sample Time : 2, 4, 10, 20, 50msec

Dimension :

220mm(W) × 2 45mm(D) × 2 50mm

(H)Weight : Approx. 7.5kg

(Main unit 5kg, internal battery 2.5kg)



Example of Data

Small array in

Downtown Olympia

Centennial Park (051)

Large array in

Downtown Seattle

Denny park (048)



Example of Spatial

Auto-correlationCoherences as a function of frequency

Denny park (048) Centennial Park (051)

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5

Frequency (Hz)

6 7 8 9 10

Co

he

ren

ce

8m

16m

27m

54m

75m

128m

200m

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Co

her

en

ce

Frequency (Hz)

100m

150m

200m

250m

300m

400m

500m

702m

1116m

1968m

2547m

3102m



Example of Spatial

Auto-correlation

Coherences as a function of distance


-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

0 500 1000 1500 2000

Distance (m)

2500 3000 3500

Co

he

ren

ce

0.20757Hz

1362m/sec

0.30525Hz

10755m/sec

0.40293Hz

955m/sec

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

0 20 40 60 80 100 120 140 160 180 200

Co

he

ren

ce

Distance(m)

1.50183Hz

735m/sec

2.00244Hz

302m/sec

3.50427Hz

245m/sec



Example of SpatialAuto-correlation

Error between observed coherences and theoretical Bessel functions


0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Fre

quency

(Hz)

0 200 400 600 800

Phase velocity (m/s)

1000 1200 1400 1600 1800 2000

Source= 0.0m

Dispersion curve : Coherence_all_appended.coh

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

10.0

Fre

quency

(Hz)

0 200

Phase velocity (m/s)

400 600 800 1000

Source= 0.0m

Dispersion curve : Coherence_all.cohUsing Two-station Microtremor Array Method to Estimate Shear-wave Velocity Profiles in Seattle and


Comparison of Observed

Dispersion Curves

0

200

400

600

800

1000

1200

1400

1600

0.1 1 10 100

Ph

ase

velo

city

(m/s

ec)

Frequency (Hz)

48 : Seattle (Denny park) 49 : Seattle (Alki)50: Seattle (Hiwatha park)51 : Olympia (Centennial Park)52: Olympia (Regional Athletic Park) Wave length = 200mWave length = 500m Wave length = 1km Wave length = 2km Wave length = 5km Wave length = 10km



200m

500m

1km

2km

5km

10km

Comparison of S-wave Velocity

Models Obtained by Inversion

0

50

100

150

200

250

300

350

400

450

500

0 100 200 300 400 500 600 700 800 900 1000 1100

Dep

th(m

)

S-wave velocity (m/sec)

48 : Seattle (Denny park)

49 : Seattle (Alki)

50 : Seattle (Hiwatha park)

51 : Olympia (Centennial Park)

52 : Olympia (Regional Athletic Park)



Shallow region

0

500

1000

1500

2000

2500

3000

3500

2000 2500

Dep

th(m

)

Deep region0 500



S-wave velocity (m/sec)

1000 1500

48 : Seattle (Denny park)

49 : Seattle (Alki)

50 : Seattle (Hiwatha park)

51 : Olympia (Centennial Park)

52 : Olympia (Regional Athletic Park)

Comparison of S-wave Velocity

Models Obtained by Inversion

Conclusions



• 2ST-MAM were performed at several sites in Seattle and Olympia,

Washington in order to estimate deep Vs structures of the area and

evaluate the applicability of the method to such investigations.

• Our investigation results imply that the 2ST-MAM can detect accurate

phase velocities down to a frequency of 0.2Hz and a maximum

penetration depth as deep as 2 to 3km.

• At the downtown Seattle, the bedrock with an Vs higher than 1500m/s

was determined at a depth of greater than 2500m.

• At the central Olympia, a low velocity layer with Vs less than 400m/s

was determined to a depth of 90m.

• These preliminary results have shown that using the 2ST-MAM

method is applicable to deep and shallow depth-to-bedrock

investigations, and provides fast, cost-effective and accurate Vs

estimates.

2d microtremor array

Education

shearwave velocity profiles

surfacewave method

swave velocity vs

seattle andolympia

phase velocity

passive surface wave

washington4the passive

depth of investigation