Comparative Study of the Refraction Microtremor (ReMi) Method: Using Seismic noise and standard P-wave refraction equipment for

deriving 1-D S-wave profiles

6th SEG-J International Symposium, Tokyo, Japan

Satish Pullammanappallil and Bill Honjas

John N. Louie

J. Andrew Siemens

Hidetoshi Miura

University of Nevada Seismological Laboratory, Reno, Nevada, USA

Siemens & Associates, Bend, Oregon, USA

Terra Corporation, Tokyo, Japan

Optim LLC, Reno, Nevada, USA

2003 Optim LLC

Refraction Microtremor TechniqueMicrotremor Technique

• Based on two fundamental ideas– Standard refraction equipment deployed similar to a

shallow P-wave refraction survey to record ambient noise (microtremor)

• 4.5 to 14 Hz (or higher) vertical geophones depending on application. Depth of penetration is inversely proportional to natural frequency of geophones used.

– Slowness-frequency (p-f) transformation of the recorded microtremor

• Separate Rayleigh waves from other seismic arrivals and allow identifying true phase velocity against apparent velocities

SeisOptSeisOpt®® ReMiReMi™™

2003 Optim LLC

• Disadvantages of commonly used methods– Drilling and logging S-wave velocities (such as OYO logger)

• Expensive & time consuming• Permitting required• Physical restrictions

– Surface methods SASW & MASW• Expensive & time consuming• Specialized recording equipment required• Artificial seismic source required

• Advantages of using SeisOpt ReMi– Data acquisition and analysis takes about three hours– No physical restrictions beyond required 100 m to 200 m line deployment space,

minimal permitting• Data can be acquired along roads, in buildings & at active construction sites

– No specialized recording equipment required• Standard refraction seismograph & refraction P-wave geophones

– No artificial seismic source • Uses ambient noise: “Quiet” site not required

– Can be used offshore as effectively as on-shore

Why SeisOpt® ReMi™

2003 Optim LLC

SeisOpt® ReMi™ Method•Acquire 15-20 seconds microtremor data along a linear array. •Array length depends on depth of investigation – recommended minimum length 100m. Crooked line geometry can be handled.

A(p====p0++++ldp, ττττ====kdt) = ΣΣΣΣ A(x====jdx, t====idt ==== τ+τ+τ+τ+px)

Step 1: p-τ transformation or the Slant Stack operation (Thorson and Claerbout, 1985) of the vertical particle velocity

Step 2: Fourier transformation: p-τ to p-f domain (McMechan and Yedlin, 1981)

F A(p,f ==== mdf) ==== ΣΣΣΣ A(p,ττττ====kdt)ei2π π π π m df kdt

Surface Waves Data Courtesy of Terracon, Las Vegas, NV, USA

2003 Optim LLC

SeisOpt® ReMi™ Method

Step 3: Velocity Spectral Analysis (Louie, 2001): Power spectrum

SA(|p|,f ) = [SA(p,f)]p>=0 + [SA(-p,f)]p<0 : Stotal(|p|,f ) = ΣΣΣΣ SAn(p,f)]

•Lower limit of the apparent phase velocities can be recognized as the true phase velocities (Louie, 2001)

Dispersion Picks

SA(p,f ) = FA*(p,f) FA(p,f)

2003 Optim LLC

SeisOpt® ReMi™ Method

Step 4: Interactive Forward Velocity Modeling

• Avoids ambiguities associated with inversion, while providing the user with ability to include constraints while modeling

2003 Optim LLCCase Study: Oregon State University Geotechnical

Engineering Field Research Test Site

•36 10-Hz geophones with 10 ft spacing deployed along a linear array

p-f Image

2003 Optim LLC

-100

-80

-60

-40

-20

0300 800 1300 1800 2300

Shear-Wave Velocity, ft/s

Dept

h, ft

Vs, ft/s

Case Study: Oregon State University Geotechnical Engineering Field Research Test Site

•36 10-Hz geophones with 10 ft spacing deployed along a linear array

2003 Optim LLCCase Study: Oregon State University Geotechnical

Engineering Field Research Test Site

•Shear-wave profile from ReMi overlain on refraction velocity model from SeisOpt @2D

2003 Optim LLC

-100-90-80-70-60-50-40-30-20-10

0300 500 700 900 1100 1300 1500 1700 1900 2100 2300

Shear-Wave Velocity, ft/s

Dept

h, ft

Seismic Cone Penetration Test (SCPT)Circa 1997

Case Study: Oregon State University Geotechnical Engineering Field Research Test Site

•Comparison of SeisOpt ReMi with SCPT at OSU

2003 Optim LLC

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0300 800 1300 1800 2300

Shear-Wave Velocity, ft/s

Dep

th, f

t

ReMi Analysis September

Spectral Analysis of Surface Waves (SASW) Geocon Northwest Circa 1997

Profile from SCPT measurement

Case Study: Oregon State University Geotechnical Engineering Field Research Test Site

•Comparison of SASW with SCPT at OSU

2003 Optim LLC

-160

-140

-120

-100

-80

-60

-40

-20

0400 500 600 700 800 900 1000 1100 1200 1300 1400

Shear-Wave Velocity, ft/s

Dep

th, f

t

Cross-Hole-99-01

Pre-Grout ReMi, V30=673 ft/s

Post-Grout ReMi, V30=827 ft/s

Grout TreatmentZone

Case Study: Wickiup Dam, Deschutes National Forest, Oregon

•Comparison of Cross-hole and SeisOpt ReMi at Wickiup

2003 Optim LLC

100

300

500

700

900

1100

1300

1500

1700

1900

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Period, s

Rayl

eigh

Wav

e Pha

se V

eloc

ity, m

/s

Modeled DispersionPicked Dispersion

Case Study: Loire Estuary, western France

Purpose: Soil classification for harbor extensionData collected by SISMOCEAN SAS, France (www.sismocean.com)

•24, standard hydrophones, 5m spacing deployed along a linear array•Source: 40 cu/in air gun. Hydrophone spread and air gun towed along seabed at 2 knots

2003 Optim LLC

The resulting ReMi S-wave velocity profile revealed boundaries that correlated well with logged cores from drill holes, and also provided

the same soil classification standard as the drill holes.

-40

-35

-30

-25

-20

-15

-10

-5

0100 300 500 700 900 1100 1300 1500 1700 1900 2100

Shear-Wave Velocity, m/s

Dept

h, m

V30 = 328 m/s (NEHRP D)

Layer 1: Gravel with fine layer of mud

Layer 2: Fine sand and clay

Layer 3: Fine sand and clay

Layer 4: Bedrock (Gneiss)

Layer boundaries from drill hole data

Case Study: Loire Estuary, western France

Contact Jerome Adamy (jerome.adamy@sismocean.com) for data acquisition and project details

2003 Optim LLC

SeisOpt® ReMi™ Market Applicability

• ReMi Vs profiles can be used for:– Earthquake site response– Liquefaction analysis– Mapping the subsurface and estimating the strength

of subsurface material– Complementing seismic refraction analysis in areas

characterized by near-surface velocity reversals– Finding buried cultural features, such as dumps and

fill material in submerged structures– Determining soil classification for offshore projects

2003 Optim LLC

Conclusion: SeisOpt® ReMi™

• Compares well with previously used 1-D shear wave measurement techniques: Economic, accurate and reliable– Correlates with SCPT measurements

• Detects velocity reversals – Matches average velocities obtained using OYO logger– Greater depth of investigation compared to borehole and surface methods– Trends similar to velocity measurements from cross-hole– Data acquisition and analysis takes about 3 to 4 hours

• Determine shear strength of subsurface material– Concrete, water saturation material, buried fill material, etc.

• Save money in performing seismic site characterization studies– Minimizes number of boreholes required– No permitting required– Can be carried out in urban settings

• Uses ambient noise as seismic energy source– Uses standard seismic refraction recording equipment

• Offshore application– Determine seismic soil classification standards for offshore projects