Acquisition of Active Multichannel Analysis of Surface Waves (MAS

Scholars' Mine
Doctoral Dissertations Student Theses and Dissertations
Spring 2017
Acquisition of active multichannel analysis of surface waves

(MASW) data in karst terrain
Ghassan Salem Alsulaimani
Follow this and additional works at: https://scholarsmine.mst.edu/doctoral_dissertations
Part of the Geological Engineering Commons, and the Geophysics and Seismology Commons
Department: Geosciences and Geological and Petroleum Engineering
Recommended Citation
Alsulaimani, Ghassan Salem, "Acquisition of active multichannel analysis of surface waves (MASW) data
in karst terrain" (2017). Doctoral Dissertations. 2735.
https://scholarsmine.mst.edu/doctoral_dissertations/2735
This thesis is brought to you by Scholars' Mine, a service of the Missouri S&T Library and Learning Resources. This
work is protected by U. S. Copyright Law. Unauthorized use including reproduction for redistribution requires the
permission of the copyright holder. For more information, please contact scholarsmine@mst.edu.
ACQUISITION OF ACTIVE MULTICHANNEL ANALYSIS OF SURFACE WAVES
(MASW) DATA IN KARST TERRAIN
by
GHASSAN SALEM ALSULAIMANI
A DISSERTATION
Presented to the Faculty of the Graduate School of the
MISSOURI UNIVERSITY OF SCIENCE AND TECHNOLOGY
In Partial Fulfilment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
in
GEOLOGICAL ENGINEERING
2017
Approved by
Neil L. Anderson, Advisor

J. David Rogers
Kelly Liu
Stephen S. Gao
Jeffery D. Cawlfield
© 2017
Ghassan Salem Alsulaimani
All Rights Reserved

iii
ABSTRACT
This study was designed to verify the effects and data reproducibility when the
length of receiver array, receiver spacing, source offset and array orientation parameters
are changed for data acquired using multichannel analysis of surface waves (MASW), at
intended target depth of 30ft (9m), and to compare the results with electrical resistivity
tomography (ERT) data obtained for the same study site.
The MASW data acquired for 34 sites, along four profiles for each site using
variable source offsets of 10ft (3m) and 30ft (9.1m), and variable receiver spacings of
2.5ft (0.76m) and 5.0ft (0.76m), concurrently. Out of the 272 profiles studied, 136
profiles were oriented east-west, and 136 profiles were oriented north-south. The MASW
data was used in conjunction with ERT data to ensure the accuracy of the ERT data.
The comparative analysis indicated the profile configuration measurements have
significant influence on the quality of the data and that the best inversion analysis is
obtained when the dispersion curves are created using the north-south oriented arrays.
The MASW survey study concluded that the most consistent and beneficial karst
terrain dispersion images were those obtained from the predicted optimal acquisition,
using receiver spacing (dx) = 2.5ft, source offset (X1) =10ft and depth of investigation of
about 30ft.
iv
ACKNOWLEDGEMENTS
First of all, I dedicate everything to God Almighty, who made all things possible.
I sincerely extend my gratefulness to my advisor, Dr. Neil L. Anderson, for the
guidance he provided as well as the understanding and patience. Through his mentorship,
I have availed with a well-grounded experience with the consistency of my research
objectives. He further provided me with encouragement not only to broaden my
experimentalist approach, but also to harness my independent thought. The
overwhelming contributions of my committee members, including Dr. J. David Rogers,
Dr. Kelly H. Liu, Dr. Stephen S. Gao and Dr. Jeffery D. Cawlfield, are also greatly
thanked for their generous support.
I would like to thank the Minister of Energy and Industry and Mineral Resources,
Saudi Arabia, Mr. Khalid Al-Falih. I thank the Saudi Geological Survey President, Dr.
Zohair Nawab, for his support as well.
I would like to extend my gratitude and eternal thanks to my father Salem, my
mother Amnah, my sisters, and my brothers for instilling in me the ability to pursue my
ambitions with their love and encouragement. I am also exceedingly grateful to my
father-in-law, Khalid and mother-in-law, Hayat for their support and prayer.
This work is dedicated to my loving wife, Kholood. Without your love and
sacrifice, I would not be achieving this milestone and encouragement throughout the
whole period of my study, and also to my beloved kids, Salem and Juwan.
Finally, special thanks go to the efforts of Mr. Rafat Ghandoura and Dr. Adel
Kerry. Without them, I would not have been able to complete the project successfully.
v
TABLE OF CONTENTS
Page
ABSTRACT ....................................................................................................................... iii
ACKNOWLEDGEMENTS ............................................................................................... iv
LIST OF ILLUSTRATIONS ............................................................................................. ix
LIST OF TABLES ........................................................................................................... xvi
SECTION
1. INTRODUCTION .............................................................................................. 1
1.1. OVERVIEW .........................................................................................1
1.2. RESEARCH OBJECTIVES .................................................................1
1.3. STRUCTURE OF THE DISSERTATION ...........................................2
1.4. LITERATURE REVIEW .....................................................................3
1.5. SIGNIFICANCE OF THE RESEARCH ..............................................5
2. GEOLOGY OF SOUTHWEST MISSOURI ...................................................... 7
2.1. INTRODUCTION ................................................................................7
2.2. OVERVIEW OF GEOLOGICAL AND STRATIGRAPHIC

SUCCESSION OF MISSISSIPPIAN SYSTEM IN SOUTHWEST
MISSOURI ...........................................................................................8
2.2.1. Lower Mississippian of Lower Osagean............................. 11
2.2.2. Lower Mississippian of Uperer Osagean . .......................... 11
2.2.3. Overview of Structure and Faulting Based on the

Work of Coots (2007)......................................................... 13
2.3. FORMATION OF KARST FEATURES ...........................................14
2.3.1. Groundwater Recharge. ...................................................... 16
2.3.2. Sinkhole or Sink.................................................................. 16

vi
2.3.3. Soluble Bedrock. ................................................................. 16
2.3.4. Natural Bridge or Tunnel. ................................................... 17
2.3.5. Losing Stream. .................................................................... 17
2.3.6. Cave. ................................................................................... 17
2.3.7. Spring. ................................................................................. 18
2.4. OVERVIEW OF SINKHOLE FORMATION PROCESS .................18
2.4.1. Collapse Sinkholes. ............................................................. 18
2.4.2. Solution-Sinkholes. ............................................................. 19
2.4.3. Cover-Subsidence Sinkhole. ............................................... 19
2.5. KARST TOPOGRAPHY IN MISSOURI ..........................................21
3. GEOPHYSICAL INVESTIGATIONS ............................................................. 23
3.1. MULTICHANNEL ANALYSIS OF SURFACE WAVE (MASW)..23
3.1.1. Overview. ............................................................................ 23
3.1.2. Basic Concept.. ................................................................... 24
3.1.3. Seismic Waves. ................................................................... 24
3.1.4. Body Waves. ....................................................................... 24
3.1.5. Surface Waves. ................................................................... 25
3.1.6. Rayleigh Waves in Homogeneous Elastic Half-Space ....... 27
3.1.7. Rayleigh Waves in Vertically Heterogeneous Elastic

Half-Space. ......................................................................... 32
3.2. MULTICHANNEL ANALYSIS OF SURFACE WAVES

DATA ACQUISITION...…...……………………………………….35
3.2.1. Geophone Spread Length. ................................................... 38
3.2.2. Source Offset. ..................................................................... 39

vii
3.2.3. Receiver Spacing ................................................................ 42
3.2.4. Orientation .......................................................................... 43
3.2.5. Topographical Conditions................................................... 44
3.2.6. Resolution of MASW Data. ................................................ 45

DATA PROCESSING ........................................................................46
3.3.1. Dispersion Analysis. ........................................................... 47

DATA INTERPRETATION ..............................................................50
3.5. ELECTRICAL RESISTIVITY TOMOGRAPHY (ERT) ..................51
51
3.5.1. Overview. ............................................................................ 57
3.5.2. Ohm’s Law and Resistivity................................................. 59

52
3.5.3. Relationship between Geology and Resistivity. ................. 57
3.5.4. Electrical Resistivity Array Configuration ......................... 59
3.5.5. Wenner Array...................................................................... 60
3.5.6. Schlumberger Array ............................................................ 62
3.5.7. Dipole-Dipole Array. .......................................................... 62
3.6. ELECTRICAL RESISTIVITY TOMOGRAPHY

DATA ACQUISITION… ...................................................................63

DATA PROCESSING ........................................................................66

DATA INTERPRETATION ..............................................................68
4. RESEARCH EXPERIMENTS ......................................................................... 70
4.1. STUDY SITE ......................................................................................70

viii

(MASW)……. ....................................................................................72
4.2.1. Data Acquisition. ................................................................ 72
4.2.2. Data Processing and Interpretation.. ................................... 74
4.3. ELECTRICAL RESISTIVITY TOMOGRAPHY (ERT) ..................81
4.3.1. Data Acquisition. .................................................................81
4.3.2. Data Processing and Interpretation. .....................................81

82
5. COMPARATIVE ANALYSES ........................................................................ 89
5.1. QUALITATIVE COMPARISON ......................................................89
5.1.1. Comparison of Overtone Images and Dispersion Curves. .. 89
5.1.2. MASW Traverse No. 1 Oriented West-East ....................... 89
5.1.2. MASW Traverse No. 1 Oriented North to South. .............. 97
5.1.3. MASW Traverse No. 2 Oriented West-East. .................... 106
5.1.4. MASW Traverse No. 2 Oriented North-South. ................ 114
5.2. DISPERSION OF RESOLUTION CURVE .....................................123
5.3. COMPARISON OF 1-D SHEAR-WAVE VELOCITY

PROFILES….. ..................................................................................132
5.4. QUANTITATIVE ANALYSES .......................................................136
6. CONCLUSIONS............................................................................................. 142
7. RECOMMENDATIONS ................................................................................ 143
BIBLIOGRAPHY ........................................................................................................... 144
VITA…………. .............................................................................................................. 156

ix
LIST OF ILLUSTRATIONS
Page
Figure 2.1. Surface Geology of Missouri............................................................................ 7
Figure 2.2. Regional Distribution of the Mississippian System in Missouri. ..................... 8
Figure 2.3. Geological Map of Southwest Missouri. .......................................................... 9
Figure 2.4. Stratigraphic Column for southwestern Mississippian System in

Southwest Missouri. ...................................................................................... 12
Figure 2.5. Faults and Lineament in Southwest Missouri (Coots, 2007). ........................ 13
Figure 2.6. Figure Showing Karst Topography Features.. ................................................ 16
Figure 2.7. Buildings collapse into a sinkhole at Disney World in, Florida.. ................... 20
Figure 2.8. Buildings collapse into a sinkhole at the Summer Bay Resort in
Clermont, Florida. ......................................................................................... 20
Figure 3.1. Elastic deformations and ground particle motions associated with the
passage of body waves. (a) P-wave. (b) S-wave (Bolt, 1982)....................... 25
Figure 3.2. (a) Rayleigh wave motion. (b) Love wave motion. ........................................ 26
Figure 3.3. Distribution of compressional, shear and Rayleigh waves generated

by a point source in a homogeneous half-space, isotropic, elastic half-
space. ............................................................................................................. 28
Figure 3.4. Displacement amplitude of Rayleigh waves versus dimensionless

depth .............................................................................................................. 30
Figure 3.5. Rayleigh wave components with different Rayleigh wavelengths

propagating through a layered medium. Wave components with
different frequencies reflect soil properties at diverse depths. ...................... 35
Figure 3.6. The instrumentation used in the MASW tomography survey. ....................... 38
Figure 3.7. Refraction arrivals showing the shingling degree pattern often
associated with layers. ................................................................................... 40
Figure 3.8. Topographical conditions are found to have an effect on the quality of
the recorded multichannel surface wave data. Receivers should be
placed on relatively flat terrain for optimum results (A & B). Surface
reliefs greater than 0.1d and the recorded data (C & D). .............................. 44
x
Figure 3.9. (A) Acquisition seismic time series data; (B) Dispersion curves
extraction frequency and phase velocity; and (C) 1-D shear-wave
velocity profiles Vs. 1-D depth curve. ........................................................ 48
Figure 3.10. Equipotential and current lines for a pair of current electrodes, A and
B, in a homogeneous half-space. ................................................................... 55
Figure 3.11. The resistivity of rocks, soils, and minerals. ................................................ 59
Figure 3.12. These are some commonly used electrode arrays and their geometric
factors. Note that for the multiple gradient arrays, the total array
length is‘(s + 2) an’, the distance between the center of the potential
Dipole pair is P1-P2 and the center of the current pair C1-C2 is given
by ‘ma.’ 'British Geological Survey (c) NERC 2013. K = Geometric
Factor. ............................................................................................................ 61
Figure 3.13. SuperSting R8/IP resistivity (left). Switch Box (AGI) connecting
passive cables (middle). ERT field setup (right). .......................................... 65
Figure 3.14. An example of a field data set with a few bad data points. The
apparent resistivity data in (a) pseudosection form and in (b) profile
form.. ............................................................................................................. 67
Figure 4.1. Southwest Missouri map where the study was conducted. ............................ 71
Figure 4.2. Map showing the location of the study area Zone A and Zone B. ................. 71
Figure 4.3. Active MASW data acquisition. ..................................................................... 73
Figure 4.4. Schematic diagram of MASW data acquisition setup. ................................... 74
Figure 4.5. The multichannel seismic records with overlay of preliminary data
quality evaluation provided by the processing software (SurfSeis4). ........... 75
Figure 4.6. The multichannel seismic records with overlay of preliminary data
Figure 4.7 The multichannel seismic records with overlay of preliminary data
Figure 4.8. (A) The multichannel seismic record collected with the source offset
of 10ft. (3.04m) and receiver spacing of 2.5ft. (0.76m); (B)
Corresponding overtone image with the fundamental and higher
modes identified.. .......................................................................................... 75
78
xi
Figure 4.9 (A) The multichannel seismic record collected with offset of 10ft.
(3.04 m) and receiver spacing of 2.5ft. (0.76 m) with applied muting.
(B) Corresponding overtone image with the fundamental and higher
modes identified.. .......................................................................................... 79
76
Figure 4.10. The results of the borehole BH1. In the same graph, the 1-D shear
wave velocity profile deduced from the MASW survey along the
Traverse No. 5 is shown. The MASW method estimated with
accuracy the shear wave velocity in the layers and assisted in the
layer identification of all the neighboring traverses... ................................... 80
Figure 4.11 ERT field set up... .......................................................................................... 82
Figure 4.12. Typical example of 2-D uninterpreted ERT model. Traverse No. P1
and three overlapping ERT traverses acquired at the study site along
2,440ft............................................................................................................ 83
Figure 4.13 Typical example of 2-D uninterpreted ERT model. Traverse No. P2
and three overlapping ERT traverses acquired at the study site along
2,440ft............................................................................................................ 83
Figure 4.14. 2-D Interpreted versions of resistivity Traverse No. P1. Depth to top
of bedrock is around 16 ft. and corresponds to the 125 ohm-m contour
interval... ........................................................................................................ 84
Figure 4.15 2-D interpreted versions of resistivity Traverse No. P2. Depth to top
of bedrock varies from 10 to 19 ft. and corresponds to the 125 ohm-m
contour interval.............................................................................................. 85
Figure 4.16. Interpreted versions of resistivity Traverse No. P1. The top of rock
correlates reasonably well with the 125 ohm-m contour interval. The
borehole location has been superposed in a red dashed line. The
200ft. mark on the resistivity profile corresponds with MASW profile
No.1; the 800ft. mark corresponds with profile No. 2 location; the1,
400ft. mark corresponds with profile No.3; the1, 800ft. mark
corresponds with profile No. 4; and the 2,200ft. mark corresponds
with profile No.5.. ......................................................................................... 86
Figure 4.17 (A) ERT Traverse No. P1 tied with MASW profile No. 1. at the 200ft.
(61m) mark; (B) Corresponding 1-D shear-wave velocity profile.
MASW depth to top of weathered rock (“acoustic” top of rock) is
identified at 15ft. (4.6m) depth. Red color line on Figure 8.8 (B)
indicates interpreted depth to top of rock... ................................................... 87
xii
Figure 4.18. (A) ERT Traverse No. P1 tied with MASW profile No. 2.at the 800ft.
(244m) mark; (B) Corresponding1-D shear-wave velocity profile.
MASW depth to top of weathered rock (“acoustic” top of rock) is
identified on 7ft. (2.1m) depth. Red color line on Figure 8.9 (B)
indicates interpreted depth to top of rock.... .................................................. 87
Figure 4.19 (A) ERT Traverse No. P1 tied with MASW profile No. 5 and
borehole (BH1), at the 2,200ft. (670m) mark; (B) Corresponding 1-D
shear-wave velocity profile. MASW depth to top of weathered rock
(“acoustic” top of rock) is identified at 8.5ft. (2.6m) depth. Red color
line on Figure 8.10 (B) indicates interpreted depth to top of rock.... ............ 88
Figure 5.1. MASW survey data collected along Traverse No.1 oriented west-east
array. (A) The 24 channel record (shot gather); (B) The dispersion
image (overtone) with superposed dispersion curve (phase velocity
versus frequency) where the fundamental mode is quite clear. (C) The
1-D shear wave velocity profile, deduced from the inversion
technique. 1) Risk of higher mode; 2) The signal-to-noise ratio (S/N);
3) R.M.S. error; 4) Iterations. ........................................................................ 91
xiii
Figure 5.5. MASW survey data collected along Traverse No. 1 oriented north to
south array. (A) The 24 channel record (shot gather); (B) The
dispersion image (overtone) with superposed dispersion curve (phase
velocity versus frequency) where the fundamental mode is quite clear.
(C) The 1-D shear wave velocity profile, deduced from the inversion
3) R.M.S. error; 4) Iterations. ...................................................................... 101
Figure 5.7 MASW survey data collected along Traverse No. 1 oriented north to
Figure 5.9. MASW survey data collected along Traverse No. 2 oriented west-east
versus frequency); (C) The 1-D shear wave velocity profile, deduced
from the inversion technique. 1) Risk of higher mode; 2) The signal-
to-noise ratio (S/N); 3) R.M.S. error; 4) Iterations. ..................................... 108
Figure 5.10. MASW survey data collected along Traverse No. 2 oriented west-
east array. (A) The 24 channel record (shot gather); (B) The
velocity versus frequency); (C) The 1-D shear wave velocity profile,
deduced from the inversion technique. 1) Risk of higher mode; 2) The
signal-to-noise ratio (S/N); 3) R.M.S. error; 4) Iterations. .......................... 110
xiv
Figure 5.13. MASW survey data collected along Traverse No.2 oriented north to
Figure 5.17. The multichannel seismic record; (B) Corresponding overtone image
in which the fundamental mode dispersion is identified; (C) The
multichannel seismic record collected with doubled receiver spacing;
(D) Corresponding overtone image in which the fundamental mode
dispersion is identified with noticeable discrepancies.. .............................. 130
126
xv
Figure 5.18. (A) The multichannel seismic record; (B) Corresponding overtone
image in which the fundamental mode dispersion is identified; (C)
The multichannel seismic record collected with doubled receiver
spacing; (D) Corresponding overtone image in which the fundamental
mode dispersion is identified with visible distortion..……………………135 128
Figure 5.19. Dispersion images (overtones) deduced from the surface wave
records collected using various MASW array configurations (north to
south orientation).. .......................................................................................130
135
Figure 5.20. Dispersion images (overtones) deduced from the surface wave
records collected using various MASW array configurations (west to
east orientation).. .........................................................................................131
120
Figure 5.21. Comparison of 1-D shear wave velocity profiles, deduced using the
inversion technique from the seismic data sets collected with various
configurations of active MASW oriented west to east array…..…………122 135
inversion technique from the seismic data sets collected with various
configurations of active MASW oriented north to south array... ................135
130
Figure 5.23. Overtone images obtained from MASW data of (A) good quality, (B)
fair quality, and (C) poor quality... ..............................................................137
130
Figure 5.24. Histogram showing how often MASW data of good, fair, poor, and
severe quality were acquired using a specific array configuration
(west to east direction)... .............................................................................140
135
Figure 5.25. Histogram showing how often MASW data of good, fair, poor, and
severe quality were acquired using a specific array configuration
(north to south)... .........................................................................................141
135
xvi
LIST OF TABLES
Page
Table 2.1. Geologic and Stratigraphic Units in southwest Missouri. ............................... 11
Table 3.1. Shear wave velocity (Vs) of some earth materials. ............................................. 51
Table 3.2. Resistivity of common earth materials. ........................................................... 68
Table 4.1. Survey parameters used during MASW surveys in this study......................... 73
Table 5.1. Results from all the dispersion curves and generated 1-D shear wave
velocity profiles for MASW Traverse No. 1. X1 - source offset; dx -
geophone spacing; D - receiver spread length; 𝜆𝑚𝑎𝑥 maximum
resolvable Rayleigh wave wavelength; 𝜆𝑚𝑖𝑛 minimum resolvable
Rayleigh wave wavelength; Z max penetration depth; S/N - signal-to-
noise ratio; RMS -root-mean-square error... ................................................. 90
Table 5.2. Results from all the dispersion curves and generated 1-D shear wave
velocity profiles for MASW Traverse No. 2. X1 - source offset; dx -
geophone spacing; D - receiver spread length; 𝜆𝑚𝑎𝑥 maximum
resolvable Rayleigh wave wavelength; 𝜆𝑚𝑖𝑛 minimum resolvable
Rayleigh wave wavelength; Z max penetration depth; S/N - signal-to-
noise ratio; RMS -root-mean-square error... ............................................... 106
Table 5.3. Average Vs at ten 1-D shear-wave velocity profile No.1. ............................. 135
133
Table 5.4. Encoded categories for three criterias. ........................................................... 136
Table 5.5. Data quality categories................................................................................... 138
Table 5.6. Quantitative analysis performed for 136 MASW profiles acquired in
west to east direction... ................................................................................ 139
Table 5.7. Quantitative analysis performed for 136 MASW profiles acquired in
north to south direction................................................................................ 139
1. INTRODUCTION
1.1. OVERVIEW
Subsurface imaging is very critical to study subsurface in complex karst terrain.
Application of non-destructive geophysical methods is one of the effective ways to
evaluate the subsurface conditions. A number of geophysical methods are currently used
to imaging subsurface in karst terrain, including Electrical Resistivity Tomography
(ERT), Self-potential (SP), Induced Polarization (IP), Ground Penetrating Radar (GPR),
Seismic Refraction, Seismic Reflection, and Multichannel Analysis of Surface Waves
(MASW).
The active Multichannel Analysis of Surface Waves (MASW) technique is
typically used to generate 1-D shear-wave velocity profiles that extend to depths of
approximately 100ft. (30.5m). The use of optimal acquisition parameters is vital to
assuring the generation of the most accurate 1-D shear-wave velocity profiles. The
identification of optimal acquisition parameters can be a challenging task, especially in
complex karst terrain, because the acoustic properties of the subsurface can vary due to
the complex and irregular nature of the bedrock surface, the presence of solution-widened
joints, etc. In this research, four configurations are tested in order to develop optimum
acquisition parameters to achieve better data resolution and the minimum investigation
depth of 30ft. (9m).
1.2. RESEARCH OBJECTIVES
The primary objective of the study is to develop optimum MASW array
configuration that can be used to image the subsurface in complex karst terrain. There is
2
an expectation that the use of optimum MASW array configuration will allow to image to
a depth of 30ft., while providing a good data quality. The following are means to
accomplish this objective:
 Acquire active MASW data along a 16 ERT traverses and using different array
configurations,
 Evaluate quality of MASW data on the basis of overtone images,
 Process and interpret MASW and ERT data using bore hole data,
 Compare the MASW and ERT data interpretations in terms of estimating top of
bedrock,
 Perform a statistical data analysis in order to develop an optimum MASW array
configuration.
1.3. STRUCTURE OF THE DISSERTATION
Sections one contains an introduction to the dissertation. The main objective and
significance of the research are discussed in this chapter. A critical review and analysis of
the relevant literature is also provided.
Section two discusses geology of southwest Missouri, overview of geological and
stratigraphic succession of Mississippian system, formation of karst features, overview of
sinkhole formation process and karst in topography in Missouri and geophysical
investigation in karst terrain.
Section three provides overviews of two geophysical methods – MASW and ERT.
Basic concepts, data acquisition, data processing and data interpretation are discussed for
each of the method.

3
Section four presents the experiments conducted in this research. The description
of the study site is also provided in this chapter.
Section five contains comparative analysis of the data. Discussions with focus on
the optimum MASW array configuration are presented in this chapter.
Section six provides the conclusions of the research.
Section seven summarizes recommendations for further studies.
The references that have been used during the study are listed at the end of the
dissertation.
1.4. LITERATURE REVIEW
Geophysical exploration methods, such as Electrical Resistivity Tomography
(ERT) and Multichannel Analysis of Surface Waves (MASW), have been used to study a
wide range of targets within the ground from discovering the deep structure of the
subsurface at thousands of meters to near surface structures and their properties at depths
of a few tens of meters (Elkrry et al., 2015). These methods have been employed in
various applications in engineering geology (Miller et al., 1999a, 1999b; Torgashov and
Varnavina, 2016; Park et al., 1998, 1999b, 2000; Xia et al., 1998, 1999, 2000a, 2000b).
The MASW and ERT methods have different applications in geotechnical
investigations. Each technique has its own advantages and drawbacks. To increase the
accuracy of geophysical investigation, the combination of methods can be effectively
used to constrain and verify the interpretations, especially for geophysical data collected
in complex karst terrain environment. Although karst terrain is often manifested at the
4
subsurface, it is difficult to determine that without more detailed subsurface
investigations.
The ERT method is used to estimate lateral and vertical variations in ground
resistivity values and can be used to map geologic variations. It has been used for
multiple geotechnical projects, including the imaging of the spatial distribution of
moisture in pavement sections (Buettner et al., 1999), river terrace sand and gravel
deposit reserve characterization and estimation using 3-D ERT arrays for bedrock surface
detection (Chambers et al., 2013), imaging of the subsurface in karst terrain (Torgashov,
2012), and characterization of soils using ERT and geotechnical investigations (Sudha,
2009).
The MASW method is used to evaluate the shear-wave velocity distribution and
arrangement of overburden soil deposits and the bedrock as well. A considerable amount
of research has been published regarding the use of both geophysical methods in complex
terrain environment (Doolittle et al., 1998; Kidanu et al., 2016; Miller et al., 2005;
Thierry et al., 2005; He et al., 2006; Schrott and Sass, 2008). Since the method was
introduced into the geophysical community in late 90s, it has been extensively used for
various applications and researched for its data acquisition and processing strategies
(Park et al., 1999). As noted by Park (2013), shear-wave velocity is a proper indicator of
material stiffness, but for a specific rock types, such as carbonate rocks, shear-wave
velocity is affected by its porosity and pore structure, as noted by Baechle et al., 2009.
Shear-wave velocity can be used to differentiate various types of unconsolidated soils and
bedrock (Odum et al., 2007), as typically bedrock exhibits higher velocities than soil,
fractured rock, or karst terrain. Recent studies have shown that the instrumental
5
configurations used for field data acquisition and processing can affect the dispersion
results obtained from surface wave seismic surveys. Attempts to obtain accurate or more
efficient dispersion curves were established by selection of particular acquisition
parameters and careful processing. The optimum parameters of MASW data acquisition,
including the optimum distance between the source and receiver spread (source offset)
and geophone array length, were discussed by Park, Miller, & Miura (2002), Park and
Carnevale (2010), and by Anderson (2012). (Xu et al., 2006) demonstrated the possibility
to develop optimum field parameters for particular MASW survey if preliminary
information on the phase velocity range and interface depth is available. A successful
attempt to develop a formula for quantitative evaluation of a layered homogeneous elastic
model estimation using specific offset was performed by (Xu et al., 2006). Attempted to
develop acquisition and processing parameters for MASW using 3-D ERT as control. In
environmental studies, the MASW method was employed to generate 2-D shear-wave
velocity fields calculated from inversion of Rayleigh-wave phase velocities, define top-
of-rock and subsurface geological structures from 6 to 100 ft. (1.8 to 30 m) (Xia et al.,
1998; Miller et al., 1999), and to determine a collapse feature in an extremely noisy
environment (Xia et al., 2004a).
1.5. SIGNIFICANCE OF THE RESEARCH
The proposed study is not only significant for imaging in karst terrain in
southwest Missouri, but it also has a broader impact on the social and economic lives of
people at all areas. Some of the significant aspects of this research include:
6
 Developing optimum MASW data acquisition and processing parameters to
enhance overall data quality and increase investigation depth.
 Verifying the efficiency of using integrated geophysical tools (MASW, ERT, and
bore hole) to map complex subsurface systems in karts terrain.
 Improving reliability of geophysical data interpretation in complex karst terrain.
 Detecting probabilities for catastrophic collapse, and therefore, minimizing
pollution and contamination of aquifers.

7
2. GEOLOGY OF SOUTHWEST MISSOURI
2.1. INTRODUCTION
Much of the Missouri state is underlain by carbonate rock (Fig. 2.1.), and most of
it is exposed (Vandike, 1997). Karst-related features are formed in dolomite and
limestone bedrocks in Missouri. The presence of clay and mud and shale with low
permeability curbs downward movement of surface water and slows the advancement of
solution cavities in the underlying carbonate rock. Hence, fewer sinkholes and caves are
found in Northern Missouri compared to Southern and Central Missouri.
Figure 2.1. Surface Geology of Missouri (MSDIS, 2003).

http://oewri.missouristate.edu/assets/OEWRI/KShade-Thesis.pdf.
8
2.2. OVERVIEW OF GEOLOGICAL AND STRATIGRAPHIC SUCCESSION

OF MISSISSIPPIAN SYSTEM IN SOUTHWEST MISSOURI
The stratigraphic succession of rock units in the Mississippian System is not
uniform throughout Missouri. Because of that, Missouri is divided into six zones:
northwestern, east-central, southeastern, central, southwestern, and northwestern. Figure
2.2 highlights that southwest Missouri is located in the southwestern region to facilitate a
description of Mississippian Age rocks.
Figure 2.2. Regional Distribution of the Mississippian System in Missouri (Vineyard,

1992).
The geologic map and units in southwest Missouri are displayed in Figure 2.3.
Each geologic rock unit has characteristics that need be addressed on a site-by-site basis.
Most of the southwest Missouri is composed of Burlington-Keokuk Limestone (Mbk) a

9
material that is particularly prone to dissolve, thereby increasing the probability of
catastrophic collapses, and through which surface water reaches the ground water supply.
Figure 2.3. Geological Map of Southwest Missouri (ESRI data source: Missouri
Geological Survey GEOSTRAT system, Sept 2015).
The Mississippian System is divided into two series: the Kinderhook and Osage.
The Mississippian formations of the southwest Missouri are presented in Table 2.1. The
total thickness of the Mississippian System in this part is about 640ft (195m), (Vineyard,
1992).
The Mississippian System starts with the Compton Formation, which has a
thickness of less than 30ft (9m). The Compton Formation is overlain by the Northview
Formation with an average thickness up to 80ft (24m). It is represented by brown
siltstone and blue or bluish-green shale.

10
The Mississippian System continues with the deposition of the Pierson, Reeds
Spring, and Elsey Formations. The total thickness of these three formations is about 260ft
(79m). Cherty limestone is a dominant component in all three formations.
The deposition of the Mississippian System is finalized by the Burlington-Keokuk
Formation. Further, the youngest exposed rock unit is in southwest Missouri. The
thickness of this formation is approximately between 150 to 270ft (45 to 83m) and varies
in thickness from place to place due to erosion. The majority of springs and caves in
southwest Missouri are found in Burlington-Keokuk Limestone.
According to conventional geotechnical procedures, the collections of boreholes
were drilled in southwest Missouri. The bedrock in the study location is intensely
dissolved Burlington-Keokuk Limestone and can be characterized by the presence of
pinnacles and cutters (Fellows, 1970), meaning that the depth in the bedrock varies
significantly. Furthermore, even though the subsurface information obtained at a location
is highly accurate, the interpolation between boreholes can sometimes be incorrect due to
significant lateral variability in karst terrains.
In the study site, the shallow subsurface is mainly represented by rocks of the
Osagean Series in the Mississippian System. The Mississippian System is represented
Table 2.1, by the Pierson, Reeds-Spring, Elsey, and Burlington-Keokuk Formations.
These four formations have similar lithologic characteristics, and they are sometimes
difficult to differentiate.
11
Table 2.1. Geologic and Stratigraphic Units in southwest Missouri (Vandike, 1993).
The following sections discuss the late two collective units of Mississippian
deposition system as shown in Figure 2.4.
2.2.1. Lower Mississippian of Lower Osagean. Lower Mississippian is Lower
Osagean (Mlo) Elsey, Reeds Spring, and Pierson Formations. The Elsey Formation is
light-gray, crystalline to micritic limestone with chert fragments and some crinoids. The
Reeds Spring Formation is gray to brown; finely crystalline limestone with chert
fragments. The Pierson Formation is brown to brown-gray and comprised of Magnesian
limestone with chert nodules. The fine-grained matrix contains some fossil fragments
(Muchaidze, 2008).
2.2.2. Lower Mississippian of Upper Osagean. Lower Mississippian is Upper
Osagean (Muo) or (Mbk) Burlington, and Keokuk Formation, in southwest Missouri. The
Burlington-Keokuk limestone is the most important formation in this research because
the bedrock in the study site is from this formation. The Burlington Keokuk Formation is
coarsely crystalline, light gray limestone with some fossiliferous limestone and chert
12
nodules. The limestone is comprised almost entirely of crinoid fragments. The
Burlington-Keokuk Formation weathers to a red to reddish brown residual soil that
contains a variable amount of chert fragments. It is almost entirely comprised of pure
calcite. Thus, this formation is susceptible to weathering through the dissolution process.
Uneven dissolution of this formation has resulted in a highly irregular bedrock-
overburden interface (Fellows, 1970).
Figure 2.4. Stratigraphic Column for southwestern Mississippian System in southwest

Missouri (Fellows, 1970).
13
2.2.3. Overview of Structure and Faulting Based on the Work of Coots
(2007). The geologic structures in southwest Missouri are highly jointed with orientations
that are trending N 20º W, and N 60º E and are approximately orthogonal to each other.
Coots (2007) discussed three main fault types that are common in the area, namely, the
Kinser Bridge fault, the Danforth Graben Fault, and the Pearson Creek Fault (Figure 2.5),
all of which have a northwest – southeast trend with normal deracination. The Kinser
Bridge Fault trends west-northwest with an average deracination of about 50ft (15m); the
Danfort Graben Fault has a vertical deracination of about 70ft (21m), and trends
northwestward; and the Pearson Creek Fault has a strike direction of N 55º W and a
normal deracination of between 10 and 20 ft (3 and 6m).
Figure 2.5. Faults and Lineament in Southwest Missouri (Coots, 2007).

14
2.3. FORMATION OF KARST FEATURES
According to the Missouri Department of Natural Resources dictionary, karst is “a
type of topography that is formed on limestone, gypsum, and other rocks by dissolution,
and is characterized by caves, sinkholes, and underground drainage”. Karst terrain is
characterized by the presence of springs, caves, sinkholes, and a unique hydrogeology
(USGS, 2012). Karst is a term derived from the German form of the Slavic word “Kras”
or “Krs,” meaning a bleak waterless place (Monroe, 1970).
Features inherent in karst terrain are caves, sinkholes, springs, dry valleys, and
loosening or disappearing streams and springs. The complex terrain results from the
chemical weathering of carbonate and other soluble rocks, and the formation of karst
features are controlled in part by pre-existing fractures within the bedrock (Ford &
Williams, 2007). Sinkhole formation is also controlled by overburden thickness,
fluctuation of the water table, soil type, and the presence of recharge or discharge zones
(Denizman, 2003).
Karst is formed when rain falls and carbon dioxide in the atmosphere is dissolved
making the rainwater acidic and consequently dissolving soluble rocks. It is worth noting
that the acidic rainwater passes through dead plant materials/debris in the soil and even
becomes more acidic as it percolates through cracks, consequently dissolving the bedrock
(limestone, dolostone, marble, gypsum and salt). Dissolution continues as the water
moves sidelong along bedding planes and joints and fractures in the rock, forming
conduits in the rock.
The carbonate rocks, limestone, and dolostone are predominantly composed of
calcite mineral (CaCO3) and dolomite mineral (CaMg (CO3)2), and both mainly calcite,
15
are susceptible to dissolution when slightly acidic water acts on them. Meteoric water
absorbs carbon dioxide (CO2) from the atmosphere and thus becomes slightly acidic.
After meteoric water saturates the ground, it passes through soil that may increase its
CO2 concentration. At the point where water enters carbonate rock, it reacts with soluble
minerals. Dissolved matter will be washed away, and as a result, features such as
dissolution-widened fractures form. As erosion continues underground, caves get hollow
enough and the roofs thin and eventually collapse to form a sink.
Karst terrain is characterized by caves, sinkholes, underground streams, and other
features formed by the slow dissolution, rather than the mechanical eroding, of bedrock.
People have discovered the difficulties (e.g., sinkhole collapse, sinkhole flooding, and
easily polluted groundwater) of living on those terrains as the population has grown and
expanded into those areas as shown in Figure 2.6, and therefore either serve as recharge
or discharge areas depending on certain geological and hydrogeological factors.

16
Figure 2.6. Figure Showing Karst Topography Features.

(www.springfieldmo.gov/documentcenter/view/11091).
2.3.1. Groundwater Recharge. Groundwater recharge (also called deep drainage
or deep percolation) is a hydrologic process where water moves downward from surface
water to groundwater. Recharge is the primary method through which water enters an
aquifer. This process usually occurs in the vadose zone below plant roots and is often
expressed as a flux to the water table surface. Recharge occurs both naturally through the
water cycle or through anthropogenic processes, such as rainwater, melting snow, or
other surface water.
2.3.2. Sinkhole or Sink. A sinkhole or sink is a collapsed part of bedrock above a
void. They may be a sheer vertical opening into a cave or a shallow depression of many
acres. They can appear suddenly, creating havoc for Ozarks landowners or highway
managers.
2.3.3. Soluble Bedrock. Soluble bedrock is characterized by underground
drainage systems with sinkholes and caves. It has also been documented for weathering-
17
resistant rocks. Subterranean drainage may limit surface water with few to no rivers or
lakes. However, in regions where the dissolved bedrock is covered (perhaps by debris) or
confined by one or more superimposed insoluble rock strata, distinctive karst surface
developments might be totally missing.
In Missouri, karst forms in limestone (calcium carbonate) and dolomite
(magnesium calcium carbonate).
2.3.4. Natural Bridge or Tunnel. Natural bridges may be the remains of a cave
that collapsed, with only a portion of the ceiling still standing. But while the entrance to
caves can be mysterious or even foreboding, natural bridges are more inviting with the
light at the end of the tunnel beckoning a visitor explore. Natural bridges are usually
shorter than a tunnel and air-filled rather than filled partially with water.
2.3.5. Losing Stream. A losing stream is a stream or river that loses water as it
flows downstream. The water infiltrates the ground, recharging the local groundwater
because the water table is below the bottom of the stream channel. Losing streams are
also common in regions of karst topography where the stream water may be completely
captured by an underground cavern system and become a subterranean river.
2.3.6. Cave. A cave is a hollow place in the subsurface, or an air-filled
underground void. A specifically, a cave is natural underground space large enough for a
human to enter and often extends deep underground. The formation and development of a
cave is known as speleogenesis. Caves are formed by various geologic processes and can
be variable sizes. This may involve a combination of chemical processes, erosion from
water, tectonic forces, microorganisms, pressure, and atmospheric influences.

18
2.3.7. Spring. A spring can be developed where surface water has infiltrated the
Earth's surface (recharge area) due to karst topography, thereby becoming part of the
area’s groundwater. The underground water then penetrates through a network of cracks
and fissure openings, ranging from intergranular spaces to large caves. The water
emerges from below the surface in the form of a karst spring. A confined aquifer forces
the spring to appear on the surface land.
2.4. OVERVIEW OF SINKHOLE FORMATION PROCESS
Sinkholes, also called dolines, form in the same manner that caves are formed
(dissolution of soluble rocks by surface water and groundwater to create a void or an
opening in the soil or rock). Sinkholes are the most diagnostic surface expression of karst
landscapes. They are important near-surface indicators of active karst features such as
dissolution-enlarged fractures, caves, and conduits. Sinkholes can vary from a few feet to
hundreds of acres wide and from less than 1 to more than 100 feet deep. Some are
rounded in shape, whereas others have vertical walls. Although most sinkholes drain
rapidly, some have natural plugs and may hold water for many years. There are three
types of sinkholes: Collapse Sinkholes, solution-sinkholes, and cover-subsidence
sinkholes.
2.4.1. Collapse Sinkholes. Collapse Sinkholes are karst-related features that are
not bound by a defined drainage area. Collapse sinkholes form as a void in the soil layer
migrates toward the surface. A sudden collapse occurs when the roof of the cavity can no
longer support its own weight and caves-in, creating an abrupt and dramatic sinkhole. A
collapse sinkhole will commonly develop in the floor of an existing depression sinkhole.
19
Sinkhole collapses are very common in the Ozarks. On average, the southwest Missouri
responds to over 30 reported collapses every year. A sinkhole of this type can be
dangerous if someone were to fall in and should therefore be repaired or fenced off as
soon as possible.
2.4.2. Solution-Sinkholes. Solution-sinkholes occur in areas where carbonate
rocks like limestones and dolostones are exposed at land surface or are covered by thin
layers of soil and permeable sand. Dissolution is most active at the limestone surface and
along joints, fractures or other openings in the rock that permit water to move easily into
the subsurface. When rain falls, surface water percolates through joints in the carbonate
rock and the dissolved rock is carried away from the surface, forming a small depression.
Further drainage of the carbonate surface accelerates the dissolution process and enlarges
the depression. As more debris is carried into the developing sinkhole, water outflow may
be plugged to form a pond and may hold water for years. Solution sinkholes tend to have
gently sloping sides and seldom pose a hazard by collapsing.
2.4.3. Cover-Subsidence Sinkhole. Cover-Subsidence Sinkhole is one of the
most dramatic forms of land subsidence is the collapse of the ground surface into natural
underground caverns. The material can no longer support its own weight, and a sudden
collapse forms this type of sinkhole. Heavy rainfall, drought, or mechanical loading can
trigger this type of sinkhole. The collapse is generally abrupt and can be catastrophic. A
sinkhole formed in this manner is often fairly circular with steeply sloping sides. The
sinkhole that developed in Exeter, Missouri is pictured in Figure 2.7 and Figure 2.8.
20
Figure 2.7. Buildings collapse into a sinkhole at Disney World in, Florida.
(http://www.csmonitor.com/Science/2013/0813/Sinkhole-Disney-World-visitors-
walking-on-holey-ground).
Figure 2.8. Buildings collapse into a sinkhole at the Summer Bay Resort in Clermont,
Florida. (http://www.amusingplanet.com/2013/09/disastrous-sinkholes).
21
In Missouri, this type of sinkhole is more common than the other types based on
the mechanism of its formation and nature of overburden materials. In areas where
limestone is buried beneath a sufficient thickness of unconsolidated material, few
sinkholes generally occur. If the overburden is dense plastic clay, its low permeability
may impede downward movement of surface water and slows the development of
solution cavities in the underlying limestone.
2.5. KARST TOPOGRAPHY IN MISSOURI
Missouri is one of seven states with karst terrain. Approximately 59% of Missouri
is covered by carbonate rock, most of which is exposed (Vandike, 1997). Major karst
areas in Missouri occur in the Mississippian rocks.
Three of the four largest metropolitan areas in Missouri (St. Louis, Springfield,
and Columbia) are located almost entirely on karst terrain. Most of the karst features in
Missouri are advanced in the southwest Missouri and Salem plateaus. However, there are
also karst features north of the Missouri River. Karst terrain is particularly susceptible to
catastrophic collapse. Furthermore, contaminants can also become concentrated in karst
depressions, so knowledge of the bedrock topography can be useful in remediation work
(Vandike, 1997).
The Ozark Plateau Aquifer is a large system comprised of many smaller aquifers
spread over a large geographic region across Oklahoma, Missouri, Kansas, and Arkansas.
This system consists of Ozark aquifers, St. Francois aquifers, and the Springfield Plateau.
It has been the most significant water source for southwest Missouri, northeastern
Oklahoma, southeast Kansas, and northern Arkansas (Macfarlane et al., 2005).

22
The Ozark area is experiencing a significant population growth in southern
Missouri regions; the population grew increased by 11% in 1990 and 27% in 2000,
(Missouri Department of Conservation, 2005). According to the Missouri Department of
Natural Resources, there are over 5,500 caves, more than 9,500 sinkholes, more than
2,800 springs, and loosing streams recorded in Missouri (Vandike, 1997).

23
3. GEOPHYSICAL INVESTIGATIONS
Several techniques have been used to characterize karst terrain. More detailed
investigations of the shallow subsurface often rely on boreholes control to characterize
subsurface conditions, but due to the spatial variability of karst terrain, and information
from individual boreholes may be insufficient for a complete site evaluation. Geophysical
exploration methods have been used in several fields to study a wide range of targets
within the subsurface, such as discovering the shallow structure of subsurface and
properties at depths up to 100ft. (30m).
Geophysical data can provide suitable continuous coverage between boreholes
and significantly decrease ambiguity (subsurface conditions). Numerous geophysical
techniques including multichannel analysis of surface waves (MASW) and electrical
resistivity tomography (ERT) can be employed to characterize the subsurface in complex
terrain like karst regions (Doolittle et al., 1998; Miller et al., 2005; Thierry et al., 2005;
Schrott & Sass, 2008). A combination of methods is often used to constrain the
interpretation because each technique has inherent advantages and drawbacks.
3.1. MULTICHANNEL ANALYSIS OF SURFACE WAVES (MASW)
3.1.1. Overview. Multichannel Analysis of Surface Waves (MASW) is a non-
destructive seismic survey method based on the measurement of seismic surface waves
generated by seismic source (active MASW) or ambient noise from cultural activities
(passive MASW) for evaluation of the elastic condition (stiffness) of the ground for
geotechnical purposes. Nowardays MASW is a commonly used geopgysical technique
because of its effectiveness for providing shear-wave velocities measurements and

24
simplicity of use (Park, 2013). The MASW method has become one of the main surface
wave methods to determine shear-wave velocitiy variations for geophysical and civil
engineering applications with observed difference of approximately 15% error between
the results obtained by MASW and borehole control measurements (Xia et al., 2002; Xia,
2014). Typical MASW survey can be broken down into three steps: data acquisition, data
processing, data interpretation. Final shear-waves profile can be presented in 1D, 2D and
3D formats (Park and Taylor, 2010).
3.1.2. Basic Concept. MASW method utilizes frequency-dependent properties of
particular type of seismic surface waves (fundamental-mode Rayleigh waves)
horizontally travelling along the earth surface directly from the impact point to the
receiver spread. The method provides the shear wave velocity information in 1-D (single
vertical shear wave velocity profile), 2-D (shear-wave velocity cross-section) or 3-D
(interpolation between densely distributed 1-D profiles) formats.
3.1.3. Seismic Waves. In general, seismic waves are parcels of strain energy that
propagate outwards from a seismic source. There are two main categories of seismic
waves that propagate within the ground and along its surface: body waves and surface
waves (Aki & Richards, 2002, Evrett, 2013).
3.1.4. Body Waves. There are two types of body waves: compression waves (the
longitudinal, primary or dilatational) and S-waves (the transverse, secondary, or shear-
wave).
Body waves can propagate through the internal energy of an elastic solid and may
be one of two types. In compressional/dilatational primary (or P-) waves, the particles of
the medium move in the direction of wave travel, involving alternating expansion and
25
contraction of the medium. The particle motion of compressional waves is parallel to the
motion of the wave propagation, causing the dilatation and compression of elementary
volume particles (Aki & Richards, 1980), as shown in Figure 3.1(a)).
Figure 3.1. Elastic deformations and ground particle motions associated with the passage
of body waves. (a) P-wave. (b) S-wave (Bolt, 1982).
3.1.5. Surface Waves. Surface waves are results of interfering P-waves and/or S-
waves travelling primarely along the free surfaces or along the boundary of dissimilar
materials (Kearey, Brooks, & Hill, 2002). They induce particles motion which is
perpendicular to the direction of wave propagation and has both a vertical and a
horizontal component. There are two types of surface waves that are most interesting for
engineering purposes based on their modes of propagation, dispersion velocities and the
depth range of the associated particle motion: Rayleigh waves and Love waves. Rayleigh
waves travel along a boundary with particle motion in vertical direction along the wave
patch and always exist in the presence of a free surface. Love waves induce particles
26
motion which is horizontal and transverse to the direction of wave propagation and only
generated in conditions where a soft layer overlying a stiffer layer (Parasnis, 1997). Since
their particle motion is always horizontal, Love waves are rarely recorded in seismic
surveys where only vertical source and receivers are used (Park et al., 1997), as shown in
Figure 3.2.
They represent the strongest portion of the signal received during a seismic
survey. For the preliminary location, an average value of Vp/Vs = 1.73 of the ground
crust is used. However, Vp/Vs can be determined with a fair degree of accuracy by the
Wadati-plot method (Wadati, 1933).
Vp = 1.7 Vs (3.1)
The speed of wave propagation is NOT the speed at which particles move in
solids (~ 0.01 m/s).
Figure 3.2. (a) Rayleigh wave motion. (b) Love wave motion
(http://thinkgeogeek.blogspot.com).
27
More specifically, in this technique, phase velocities are calculated for each
component frequency of the field-recorded Rayleigh waves. The resulting dispersion
curve (phase velocity vs. frequency) is inverted using a least squares approach, and a
vertical shear-wave velocity profile is generated (Miller et al., 2000; Nazarian et al.,
1983; Stokoe et al., 1994; Park et al., 1999a, 1999b, 2000; Xia et al., 1999).
Rayleigh waves result from the interaction of P-waves and vertical (SV) waves
with the surface of the ground (Aki & Richards, 1980). The particle motion of Rayleigh
waves moves perpendicular to the surface but travels along the wave path (Kearey,
Brooks & Hill, 2002). The wave motion is retrograde (counterclockwise) closest to the
surface, but becomes prograde (clockwise) at greater depths. More than two-thirds of the
seismic energy that is generated is imparted (Loke, 2000).
Particle motion associated with Love waves is parallel to the free surface and
perpendicular to the direction of the waves, whereas Rayleigh waves move perpendicular
to the surface but travel along the wave path (Kearey, Brooks and Hill, 2002). Love
waves are a form of a polarized shear wave and are observed in a multilayer media when
the shear wave velocity of the top layer is less than that of the lower layer (Parasnis,
1997). They are the fastest surface wave, move along the ground from side-to-side and
are confined to the surface of the crust. Since their particle motion is always horizontal,
(Love) waves are seldom recorded in seismic surveying where only vertical sources and
receivers are used (Park et al., 1997).
3.1.6. Rayleigh Waves in Homogeneous Elastic Half-Space. Rayleigh waves
generated by a vertical point source on a free surface propagate along cylindrical
wavefronts away from the impact point (Richart, Hall, & Woods, 1970). Body waves
28
propagate radially, from the exterior of the source, both into the medium and along the
surface, along a hemispherical wavefront, as shown in Figure 3.3. As the waves transfer
outward and encounter an increasingly larger volume of material, their amplitude
decreases.
Figure 3.3. Distribution of compressional, shear and Rayleigh waves generated by a point
source in a homogeneous half-space, isotropic, elastic half-space (Richart et al., 1970).
This is called geometrical spreading (or geometrical damping) (Aki & Richards,
1980; Richart et al., 1970). In a homogeneous elastic half-space, the amplitude of
Rayleigh waves decreases as r-0.5, where r is the distance from the impact point. For
comparison, the amplitude of body waves decreases as r-2 along the surface and r-1 into
the medium. As two-thirds of the total seismic energy is imparted into Rayleigh waves
and Rayleigh waves decay more slowly with distance than body waves, the wave field
becomes dominated by Rayleigh wave motion at nearest distances from the seismic
source (Richart et al., 1970). Around 67% of the seismic energy is imparted into Rayleigh
waves, while about 26% is imparted into shear waves and 7% into compressional waves
(Woods, 1968).
29
The amplitude of Rayleigh waves decays exponentially with depth (Richart et al.,
1970). The simplest method is attribution of a factored shear wave phase velocity
(usually 0.9 times the Rayleigh wave velocity (Joh 1996, Foti 2000 & Okada 2003) to a
depth equivalent to a fraction of the Rayleigh wavelength, λ. Fractional depth factors
range from λ/4 to λ/2 (Jones1958, Ballard & McLean 1975; Abbis, 1981).
Gazetas (1982) recommended that λ/4 be used where the stiffness increases
significantly with depth and that λ/2 be used for more homogeneous stiffness profiles.
However, a factor of λ/3 is most commonly used (Bullen 1963 & Richart et al., 1970)
because a significant proportion of the particle motion in the ground associated with
Rayleigh wave propagation is approximately this depth. The horizontal and vertical
Rayleigh wave displacement amplitudes as a function of dimensionless depth are shown
in Figure 3.4 for several values of Poisson’s ratio (v) = 0.25 (Richart et al., 1970).
Rayleigh waves have unique properties that allow them to be transformed into
subsurface shear-wave velocity profiles (Surf-Seis, 2006). Rayleigh waves are dispersive
in nature (different frequencies travel with different phase velocities). The highest usable
Rayleigh wave frequency that has been recorded (for geotechnical purposes) involves
particle motion within the shallowest depth range (approx. 1 Rayleigh wavelength;
typically upper few ft) and travels with a velocity that is mostly a function of the average
shear-wave velocity within that depth range.
Intermediate frequencies for Rayleigh waves involve particle motions to
intermediate depths (to approx. 1 Rayleigh wavelength), and they travel with velocities
that are a function of the average shear-wave velocity over those intermediate depth
ranges. Also, the lowest recorded usable frequency involves particle motions to the
30
greatest depth (1 Rayleigh wavelength) and travels with a velocity that is a function of
the shear-wave velocity over that depth range.
Figure 3.4. Displacement amplitude of Rayleigh waves versus dimensionless

depth(Richart et al., 1970).
Equations describing Rayleigh waves propagate along the free surface of the earth
with particle motions that decay exponentially with depth. The lower component
frequencies of Rayleigh waves involve particle motions at greater depths. In a
homogeneous (non-dispersive) medium, Rayleigh wave phase velocities are constant.
Rayleigh wave phase velocities are a function of both the shear-wave and the
compression wave velocities of the subsurface. The interrelationships between Rayleigh

31
wave velocities (VR), shear-wave velocities (β) and compression wave velocities (α) in a
uniform medium are expressed in Equation 3.2 (Anderson, 2010):
VR6 - 8β2VR4 + (24 - 16β2 /α2) β4VR2 + 16(β2/α2 – 1) β6 = 0 (3.2)
where
VR is the Rayleigh wave velocity within the uniform medium,
β is the shear-wave velocity within the uniform medium (also denoted as Vs), and
α is the compressional wave velocity within the uniform medium (Also denoted as
Vp).
Although the Raleigh wave phase velocity is a function of both compressional (α)
and shear-wave (β) velocities, it is much more sensitive to variations in β than in α in
Equation 3.3.
VR< β < α (3.3)
Lower frequencies involve particle motion at greater depths, causing VR to also
vary with frequency.
Equation 3.4 might initially suggest that it would be difficult to extract the shear-
wave velocity because the equation contains two unknowns (shear and compression wave
velocities). Fortunately, this is not the case because Rayleigh wave phase velocities are
influenced much less by changes in compression wave velocity than by changes in shear
wave velocity. In a uniform medium, Rayleigh wave velocity (VR) and shear-wave
velocity (β) are related by Equation 3.4 (Anderson, 2010):
β = VR/C (3.4)
32
The propagation velocity of individual frequency components is referred to as
phase velocity C. The variable C is a constant that changes slightly depending on
Poisson’s ratio of the material through which the seismic waves travel. Even in extreme
variations of Poisson’s ratio, C only ranges from 0.874 to 0.955 (Anderson, 2010). It is
suggested that if a value for C is assumed and the frequencies (with their respective
surface wave velocities) are recorded, then a shear-wave velocity profile can be
developed through analysis, and a velocity image of the subsurface can be generated
(Anderson, 2010).
Rayleigh wave velocities, as noted in Equation 3.2, are a function of both the
shear-wave velocity and the compressional wave velocity of the subsurface. A plot of the
frequency versus the phase velocity, known as a dispersion curve, visualizes these
relations. The shape of the dispersion curve is referred to as the dispersion characteristic
of the Rayleigh wave (Everett, 2013).
3.1.7. Rayleigh Waves in Vertically Heterogeneous Elastic Half-Space. Shear-
wave and compressional wave velocities vary with depth. Hence, different component
frequencies of Rayleigh waves (involving particle motion over different depth ranges)
exhibit different phase velocities (Bullen, 1963). The phase velocity of each component’s
frequency is a function of the variable body wave velocities over the vertical depth range
associated with that particular Rayleigh wavelength. More specifically, in the subsurface,
the Rayleigh wave phase velocity equation takes the following form:
VR (fj, CRj, β, α, ρ, h) = 0 (j = 1, 2, …., m) (3.5)
where
fj is the frequency in Hz,

33
VRj is the Rayleigh-wave phase velocity at frequency fj,
β = (β1, β2,….., βn) T is the shear-wave velocity vector,
βi is the shear-wave velocity of the i th layer,
α = (α1, α2, ….., αn) T is the compressional P-wave velocity vector,
αi is the P-wave velocity of the i th layer,
ρ = (ρ1, ρ2,…., ρn) T is the density vector,
ρi is the density of the i th layer,
h = (h1, h2,…., hn-1) T is the thickness vector,
hi is the thickness of the i th layer and,
n is the number of layers within the earth model,
The velocity of Rayleigh waves is comparable to the velocity of shear waves. In a
rock formation with a Poisson ratio of approximately 0.25, the velocity of the Rayleigh
wave is approximately 92 % of the velocity of the shear wave. In materials with rates
from 0.4 to 0.5, the percentage increases to 94 to 95.5 %, respectively (Steeples, 1998).
As a general rule, the velocities of Rayleigh waves are assumed to be approximately 90 to
92 percent of the respective shear-wave (Ivanov, Park, & Xia, 2009; Parasnis, 1997). The
shear-wave velocity can be estimated within a ten percent margin of error using these
assumptions (United States Corps of Engineers, 1995).
The shear-wave velocity of a material is very important when predicting the
impact that an earthquake’s seismic waves will have on a material as the waves pass
through it (Wood, 2009). Knowing the shear-wave velocity of material, one can
determine the shear modulus using the relationship between the shear-wave and soil or
rock as related by Equation 3.3:

34
μ=ρVs2 (3.6)
where
μ = shear modulus,
p = mass density, and
Vs = shear-wave velocity.
Rayleigh wave velocities, as noted in Equation 3.5, are a function of both the
shear-wave velocity and the compressional wave velocity of the subsurface. A plot of
frequency versus phase velocity, known as a dispersion curve, visualizes these relations.
The shape of the dispersion curve is referred to as the dispersion characteristic of the
Rayleigh wave (Evrett, 2013). In MASW, phase velocities are calculated for each
component frequency of the field-recorded Rayleigh waves. Then the resulting dispersion
curve (phase velocity vs. frequency) is inverted using a least squares approach, and a
vertical shear-wave velocity profile is generated (Miller et al., 2000; Nazarian et al.,
1983; Stokoe et al., 1994; Park et al., 1999a, 1999b, 2000; Xia et al., 1999). Typical
dispersion curve has multimodal character (multiple phase velocities existing for a certain
frequency). The mode with the lowest phase velocity (at each frequency) is defined as
fundamental mode and exists at all frequencies. Other higher modes, called the first
mode, second mode, etc. They have higher phase velocities and are only exist above a
cut-off frequency that depends on the mode (Evrett, 2013).

35
3.2. MULTICHANNEL ANALYSIS OF SURFACE WAVES DATA

ACQUISITION
Rayleigh wave velocities are, in general, shown to increase with depth (e.g. waves
with longer Rayleigh wavelengths and lower frequencies) and propagate faster than those
with shorter wavelengths. The relation between frequency (f) and Rayleigh wavelength is
called a phase velocity (c(f)). These unique characteristics result in a different Rayleigh
wavelength (λ(f) given as (Kramer, 1996) , as shown in Figure 3.5.
Figure 3.5. Rayleigh wave components with different Rayleigh wavelengths propagating
through a layered medium. Wave components with different frequencies reflect soil
properties at diverse depths.
The material properties of the topmost layer have an impact on the phase velocity
of wave component (1), whereas the phase velocities of wave components (2) and (3)
also depend on the properties of the deeper layers.
λ (f) = c(f)/f (3.7)

36
The variable C is a constant that changes slightly depending on Poisson’s ratio of
the material through which the seismic waves travel. Even in extreme variations of
Poisson’s ratio, C only ranges from 0.874 to 0.955 (Anderson, 2010). It is suggested that
if a value for C is assumed and the frequencies (with their respective surface wave
velocities) are recorded, then a shear-wave velocity profile can be developed through
analysis, and a velocity image of the subsurface can be generated (Anderson, 2010).
These unique characteristics result in a different Rayleigh wavelength λ.
Depending on the nature of the seismic source, MASW method can be
categorized as active or passive. The active MASW adopts the conventional seismic
refraction mode of surveying by using an active seismic source, such as a sledgehammer,
accelerated weight drop or vibroseis deployed in traverse along a linear array of receivers
(Park et al., 1999). The passive MASW method utilizes surface waves generated by
natural sources or cultural activities, highway traffic and construction equipment (Park et
al., 2007). Depending on the receiver configuration there are two ways to conduct a
passive MASW survey - the passive remote MASW (Park et al., 2004; Park et al., 2005)
employs a two-dimensional receiver array and the passive roadside MASW (Park and
Miller, 2008) uses linear one-dimensional receiver array. Data acquisition is more
tolerant in parameter selection than other seismic methods because of the easily achieved
high level of signal-to-noise ratio in seismic field records.
For the acquisition of active MASW data, low frequency (for example, 4.5 Hz),
vertically polarized geophones are lined on the test site surface at the appropriate equal
intervals. The number of geophones used is typically 24 or more with a constant inter-
geophone spacing that is optimized for site-specific geologic conditions (Donohue,

37
Forristal, & Donohue, 2013; Lin, Chang, & Chang, 2004). Each geophone is connected to
a separate recording channel (Park et al., 1997) and the whole array is connected to
engineering seismograph and field laptop equipped with software necessary for data
recording. A surface wave is generated with an impact load at one end of the lineup and
the geophones record the resulting wave motion as a function of time. The seismic source
may be a 20-lb sledgehammer, a mechanical impact device, a shotgun or explosives,
depending upon the depth of the investigation and the site-specific conditions. The arrival
of surface wave is detected along a set of geophones and recorded on a seismograph, with
the output of each geophone being displayed as a single trace.
Proper setup of data acquisition parameters is essential for the success of the
surface wave survey. The important field parameters are total length of the receiver
spread, source offset (the distance between the source and the nearest geophone), and
receiver spacing.
As the geophones only record vertical motion, it is important that they are placed
vertically in the subsurface. Each geophone is connected through a separate recording
channel to a data acquisition card and a computer equipped with the necessary data
acquisition software (Figure 3.6.).

38
Figure 3.6. The instrumentation used in the MASW tomography survey (Park et al.,
2002).
As illustrated in Figure 3.6, geophones are profiled up on the surface of the test
site. A wave is generated, and the wave propagation is recorded. The maximum depth of
the investigation (Zmax) varies with the site, the natural frequency (fe) of the geophones
and the type of seismic source that is used. The maximum investigation depth is
determined by the longest Rayleigh wavelength that is obtained during data acquisition
(λmax). The following is a commonly adopted empirical criterion (Park & Carnevale,
2010):
Zmax ≈ 0.5λmax (3.8)
3.2.1. Geophone Spread Length. The overall geophone spread length (D)
defines the longest Rayleigh waves wavelength that can be analyzed. It is also related to
the maximum investigation depth (Zmax), since the length of the receiver array is related
to the longest Rayleigh wavelength that is obtained during data acquisition. A common
39
criterion is that the longest Rayleigh wavelength (λmax) that can be analyzed is
approximately equal to the length of the receiver array (Park & Carnevale, 2010):
λmax ≈ D. (3.9)
Attempts to analyze longer Rayleigh wavelengths than those indicated by
Equation 3.9 can lead to less accurate results. Recent studies have shown that although
there is fluctuating inaccuracy, it will remain within 5% for the interval D ≤ λmax ≤2D
(Park & Carnevale, 2010), because of the uncertainties (noise) always included in the
measurement. A very long receiver spread should be avoided. Surface waves generated
by the most commonly used seismic sources (e.g. reasonably heavy sledgehammers) will
have attenuated noise level at the end of an excessively long receiver spread, making the
signal from the furthermost receivers too noisy to be usable (Park et al., 1999; Xia et al.,
2009). The maximum wavelength that can be accurately analyzed is approximately equal
to the half of the length of the overall geophones spread.
3.2.2. Source Offset. The appropriate choice of source offset eliminates the near-
field effect (the risk that geophones will pick up surface waves that were not fully
developed. Long source offset could generate energy for long-wavelength surface waves.
It can also result in lack of short-wavelength wave components due to the attenuation.
For long offsets, where the distance traveled along the refractor exceeds several Rayleigh
wavelengths, the head wave amplitude decreases by a factor close to the inverse square of
the offset. At smaller offsets, which is likely to be the case for many LVLs (low-velocity
layer or weathering surveys), the amplitude decrease is slightly less. In addition to the
variation of amplitude with the offset, significant attenuation occurs if the refractor is thin
(Press et al., 1954). This effect is normally associated with shingling, a phenomenon
40
characterized by a shift of energy to successively later cycles as the offset increases,
producing an echelon pattern, as shown in Figure 3.7.
Figure 3.7. Refraction arrivals showing the shingling degree pattern often associated with
layers.
The source offset needs to change in proportion to the intended maximum
investigation depth, with a conservative calculation being that the source offset is equal to
half of the maximum investigation depth (Park & Miller, 2005). The importance of
choice of optimum offset is that it controls the degree of contamination by a range of
near‐field effects. Surface wave cannot fully develop before it strikes the first geophone
in case of source positioned too close to the receiver spread. The risk of non-planar
surface waves being picked up by the geophones can be minimized by carefully choosing
the source offset (Park & Carnevale, 2010). The minimum source offset required to avoid
near-field effects depends on the longest Rayleigh wavelength that is being analyzed
41
(λmax). In most cases, plane-wave propagation of surface waves first occurs when the
source offset is greater than half of the longest Rayleigh wavelength (Park et al., 1999).
The source offset also needs to be sufficiently sized to extend to the primary depth
range of interest. This concept conforms to the common practice with Equation 3.10 and
3.11:
X1 ≥ 0.5D (3.10)
when λmax ≈ D (3.11)
where D is the length of the receiver array. However, studies have shown that this
criterion can be relaxed significantly for MASW surveys (Park et al., 1999, 2002; Park &
Shawver, 2009).
Studies have shown that this criterion can be relaxed significantly for MASW
surveys (Park et al., 1999, 2002; Park & Shawver, 2009). A long source offset, X1 ≥ D,
can potentially enhance energy for long- Rayleigh wavelength wave components, thus
increasing λmax for a receiver array of a given length (Park & Carnevale, 2010).
However, such a long source offset can result in a lack of short- Rayleigh wavelength
components, due to excessive attenuation (Park & Shawver, 2009). A long source offset,
X1≥ D, can potentially enhance energy for long- Rayleigh wavelength wave components,
thus increasing λmax for a receiver array of a given length (Park & Carnevale, 2010).
However, such a long source offset can result in a lack of short- Rayleigh wavelength
components, due to excessive attenuation (Park & Shawver, 2009). The suggested
minimum (X1, min) and maximum (X1, max) source offsets for use in practice are as
follows (Park, 2015):
X1, min = 0.2D and X1, max = D. (3.12)

42
The optimum offset (A), based on a layered ground model due to the longer
Rayleigh wavelength components of Rayleigh waves requires longer time or larger
offsets to develop into plane waves (Zhang et al., 2004).
𝜆𝑚𝑎𝑥 𝐶𝑅 𝑚𝑖𝑛
𝐴= (3.13)
4∆𝐶𝑅
where 𝜆𝑚𝑎𝑥 ,𝐶𝑅 𝑚𝑖𝑛 , and ∆𝐶𝑅 are the longest Rayleigh wavelength, the minimum
phase velocity of Rayleigh waves, and the difference between the maximum and
minimum phase velocities, respectively.
For example, the first shot was acquired with the nearest source offset of 15ft, the
longest Rayleigh wavelength of λmax = 500 ft/sec/25 Hz = 20ft, the minimum phase
velocity of CRmin =180ft/sec and the difference between the maximum and minimum
phase velocities of ΔCR = 500ft/s - 180ft/sec = 320ft/sec. Therefore, the suggested
optimum offset A would be 2.8ft.
3.2.3. Receiver Spacing. The receiver spacing (dx) is defines the shortest
wavelength that can be analyzed and the shallowest resolvable depth of MASW survey.
The minimum investigation depth λmin relates to the minimum distance between
geophones (O’Neill 2003). This assumes that the maximum depth of penetration is
approximately half of the maximum recorded Rayleigh wavelength (Park et al., 2009).
The receiver spacing is related to the shortest Rayleigh wavelength (λmin) that can be
included in a dispersion curve. In general, the receiver spacing should not be greater than
half of the shortest Rayleigh wavelength in order to avoid spatial aliasing (Xia et al.,
2009):
dx ≈ λmin ≈ Zmin
43
dx ≤ 0.5 λmin. (3.14)
Moreover, the receiver spacing acts as a guideline for determining the minimum
thickness (hmin) of the shallowest layer of the layered earth model used in the inversion
analysis (Park et al., 1999; Xia et al., 2009). In other words,
h1 ≥ hmin ≈ dx (3.15)
where h1 is the thickness of the topmost layer of the earth model.
The receiver spacing (dx) also determines the computational artifacts caused by
the aliasing impact. It sometimes generates some curved streaks in the dispersion image,
caused at the point of dispersion, where the Rayleigh wavelength becomes less than one-
half of (dx).
3.2.4. Orientation. The orientation of the profiles should best depict the
structures of the subsurface, which is as well important for the best possible
interpretation. While interpreting the seismic data, well log data and result from previous
geological surveys and studies of the region are often used in relation to the seismic data
(Sheriff & Geldart, 1995). Furthermore, the effect spacing needs to be small in all
orientation to avoid irregular surface. The subsurface coverage should be uniform with a
consistency between the contribution of different source offsets and azimuths (Bacon et
al., 2003).
Typically, there was a reasonable match between the two surveys from opposite
orientations. If a survey in one orientation produced a poor quality and a difficult-to-
extract dispersion curve, the survey in the opposite orientation generally produced a
dispersion curve of similar quality, which is largely presumed to be a result of subsurface

44
lateral heterogeneities. Processing the data surveyed in both orientations helped limit the
influence of disturbances.
To phrase it differently, both receiver spacing and source array methods can add
additional orientation effects. Subsequently, comprehensively considering the parameters
of the source array, the receiver spacing and the geometry in an acquisition system will
lay a foundation for a high quality and highly efficient seismic survey project.
3.2.5. Topographical Conditions. Topographical conditions are known to have
effect on the quality of the recorded surface wave data and therefore the quality of the
resulting dispersion curves. They can also affect the quality of the acquired surface wave
data and therefore affect the quality of the generated dispersion images (Zeng, Xia,
Miller, Tsoflias, & Wang, 2012). For optimum results, the receivers should be placed on
relatively flat terrain, suitable for Active MASW, as shown in Figure 3.8 (A) and 3.8 (B).
Figure 3.8. Topographical conditions are found to have an effect on the quality of the
recorded multichannel surface wave data. Receivers should be placed on relatively flat
terrain for optimum results (A & B). Surface reliefs greater than 0.1d and the recorded
data (C & D). Adopted from Park (2015).
Surface reliefs within the receiver spread greater than 0.1 d can have a significant
effect on the generation of surface waves (Park, 2015), as shown in Figure 3.8 (C), and
45
should therefore be avoided. The slope of the surface along the receiver spread can also
affect the quality of the surface wave records, as shown in Figure 3.8 (D).
However, topography can also interfere with surface wave propagation. Results
of numerical investigations presented by Zeng et al. (2012) indicated that the slope of the
topography along the survey profile (θ) should preferably be less than 10º. A steeper
topography (i.e., a slope angle θ > 10º) can lead to significant errors (greater than 4%) in
estimation of the Rayleigh wave dispersion characteristics (Zeng et al., 2012).
In theory, maximum investigation depth is defined by the longest surface wave
wavelength generated during the data acquisition. On practice, the maximum
investigation depth also depends on the survey site and type (strength) of seismic source.
A heavier source provides increased investigation depth. For example – the use of 20 lb.
sledgehammer typically results in 30 – 100ft. (10 – 30m) depth of investigation. Recent
studies reported that use of non-metallic plate placed in the impact point can generate
stronger energy at the lower frequency part of surface waves than a conventional metallic
plate (Cui, 2013). Also, the maximum investigation depth is determined by the longest
Rayleigh wavelength that is obtained during data acquisition.
Recording frequency of 1000 Hz (sampling interval of 1ms) is most commonly
used in active MASW surveys. The total recording time of 1 second is usually used for
the impulsive seismic source MASW surveys.
3.2.6. Resolution of MASW Data. Another important thing to consider is the
desired resolution of MASW data. Resolution is mostly a function of the receiver
spacing, which is directly related to the shortest wavelength and therefore determines the
shallowest resolvable depth of investigation. The total length of the receiver spread is
46
directly related to the longest wavelength that can be analyzed, which in turn defines the
maximum depth of investigation. The source offset distance controls the possible degree
of contamination by the near-field effects (surface waves are formed through interference
of body waves generated from reflections and refractions, so they require to propagate a
certain minimum distance from the source to fully develop, which may result in both
underestimation or overestimation of Rayleigh wave phase velocity due to the body
waves contamination near to the source). Vertical stacking with multiple impacts can
suppress ambient noise significantly, producing a high signal to noise ratio. This is good
in surveys in urban areas. Three to five vertical stacks are recommended during survey.
3.3. MULTICHANNEL ANALYSIS OF SURFACE WAVES DATA PROCESSING
Various component frequencies of Rayleigh waves involve particle motion at
various depths. For example - the lower frequencies involve particle motion at greater
depths. So the different component frequencies of Rayleigh waves exhibit different phase
velocities. The whole MASW technique is based on the relationship between the
Rayleigh wave phase velocities and the depth-range of associated particle motion.
The quality of the acquired surface wave records can be evaluated in terms of the
resolution of the phase velocity spectrum, i.e., the sharpness of the amplitude peaks
observed at each frequency, the extractable frequency range and the continuity of the
fundamental mode high-amplitude band.
The quality of the MASW data can also be affected by natural geologic conditions
that may not produce well-defined dispersion curves and cannot be used to calculate
reliable shear-wave velocity models.

47
After data acquisition, field records are analyzed using SurfSeis software
developed by Kansas Geological Survey (KGS). In order to convert seismic record to an
estimate of shear wave velocity, three crucial steps must be performed: generation of
dispersion image (overtone - a graphic representation of intensity in phase velocity and
frequency space) and extraction of the fundamental-mode dispersion curve from it. The
software calculates phase velocities for each component frequency of recorded MASW
data to generate the resultant dispersion image to extract the dispersion curve. Then, the
curve is inverted using a least–squares approach to generate a vertical shear-wave
velocity profile. Basically, software performs wavefield transformation from offset-to-
time domain into phase velocity-to-frequency domain (multichannel record, time-space
domain, to dispersion image, frequency-phase velocity domain). Then MASW data are
being inverted to generate 1-D (depth) shear wave velocity profile. Figure 3.9 displays
the raw seismic record, dispersion curve extracted from the overtone image, and the
resulting shear wave velocity profile.
3.3.1. Dispersion Analysis. The intent of dispersion analysis is to estimate one or
more dispersion curves that are in turn passed into the next step of inversion process. The
following are the most influential factors affecting the dispersion analysis:
1. Frequency range for depth of investigation,
2. Approximate phase velocity range,
3. Easy characterization of higher modes, and
4. Identification and reduction of noise events.
The influence of these factors on the analysis is highly dependent on the data
quality (signal-to-noise ratio, S/N). A “good” quality data set suggests that surface-wave
48
is the most prominent seismic event (with highest S/N), whereas a “bad” quality data set
is usually contaminated by noise. In the dispersion curve analysis, the fundamental mode
surface waves are the signal, and everything else is noise. Noise includes all higher-mode
surface waves as well as all body wave events.
Figure 3.9. (A) Acquisition seismic time series data; (B) Dispersion curves extraction
frequency and phase velocity; and (C) 1-D shear-wave velocity profiles Vs. 1-D depth
curve (http://www.kgs.ku.edu/software/surfseis/gifs/masw).
Software also performs evaluation of an approximate phase velocity range for the
surface waves. It generally ranges from as low as 650ft/sec (or 200m/sec) to as high as
2,500ft/sec (760m/sec) depending on the material type. This information is used by the
program to initiate the analysis by searching within this range for a phase velocity
49
corresponding to a certain surface wave frequency with the greatest coherence throughout
the entire range of offset and the highest signal-to-noise ratio which is indicative of a
high confidence in the acquired phase velocity (Figure 3.9). The SurfSeis software
calculates phase velocities within the specified frequency range. This calculation can be
run multiple times using different values and varieties of parameters, examining the
output curves until an optimum solution is identified. In general, the curve with the
highest signal-to-noise ratio (S/N) represents the best option. The quality of a dispersion
curve is judged according to two criteria: the signal-to-noise ratio (S/N) and the general
dispersion curve direction.
The next processing step is an extraction of a fundamental-mode dispersion curve
from the dispersion image. This extracted curve is called a "measured" dispersion curve
that is an input data to the next data analysis step (inversion). The quality of the acquired
surface wave records can be evaluated in terms of the resolution of the phase velocity
spectrum, i.e., the sharpness of the amplitude peaks observed at each frequency, the
extractable frequency range and the continuity of the fundamental mode high-amplitude
band. The quality of a dispersion curve is judged according to two criteria: the signal-to-
noise ratio (S/N) and the general dispersion curve direction. The quality of the ‘match’
between the two curves is evaluated on the root‐mean‐square error (Xia et al., 1999). The
inversion of a dispersion curve using SurfSeis Software is particularly straightforward, as
it is a fully automated process that removes any human error incurred during the
calculations.
50
3.4. MULTICHANNEL ANALYSIS OF SURFACE WAVES DATA

INTERPRETATION
Data interpretation of MASW involves the analysis of the variations of the shear
wave velocities with depth with a goal to transform the output velocity model into a
geologic model of the subsurface.
Interpretation of the subsidence features in karst terrain environment is difficult
because of the inherent complexity of the subsurface and the resolution limits of MASW
technique. Shear wave velocity models can clearly show the low-velocity zones, which
can be interpreted as sinkholes, voids, fractured bedrock, or naturally in-filled collapsed
features.
In Missouri for example, the shear wave velocity value used to determine the
depth to top of bedrock is mostly 1000ft/sec. but can vary. Again, highly stiff earth
materials have relatively high shear wave velocities compared to fractured or weak earth
materials (Table 3.1).
Reliability of output shear-wave velocity data decreases as lateral and vertical
heterogeneity of soil/rock increases. Therefore, site conditions should be considered; the
immoderate dip of the subsurface layer along the survey line (more than about 10
percent), disadvantageous topography, or known high lateral variability in soil or rock
properties may be reasons to reject field data as inconvenient for interpretation in terms
of simple vertical variation of subsurface.

51
Table 3.1. Shear wave velocity (Vs) of some earth materials (National earthquake hazards
reduction program).
3.5. ELECTRICAL RESISTIVITY TOMOGRAPHY (ERT)
3.5.1. Overview. Electrical Resistivity Tomography (ERT) is a nondestructive
geophysical technique based on the electrical resistivity method. ERT has been used for
decades as an effective environmental and geotechnical tool. In particular, ERT
methodology is widely used for determining the maximum depth of rock, acquiring
information on the elevation of soil, the top of rock, etc. This method is especially
preferred for vision characterization in karst terrain (Zhou, 1999).
The purpose of an electrical resistivity survey is to determine the subsurface
resistivity distribution in karst characteristics, such as caves, that may or may not be
easily recognizable on the subsurface. Areas where the top of rock is limestone or
gypsum, like it is in Missouri have a high probability of karst development. Karst areas
commonly lack surface water and have numerous streambeds that are dry except during
52
periods of high runoff. When an electrical resistivity tomography survey is conducted in
karst terrain, current flow is generally assumed to be electrolytic rather than electronic.
Surface electrical resistivity surveying is based on the principle that the partition
of electrical potential on the subsurface around a current-carrying electrode depends on
the soil’s materials and rock to demonstrate the variations in their electrical resistivity
because of the variations in their mineral content, permeability, fluid saturation, porosity,
etc. After that, areas of the subsurface undergoing dissolution can be differentiated from
the top of rock by measuring the resistivity of the subsurface in good resolution.
3.5.2. Ohm’s Law and Resistivity. The fundamental principle behind the
combination and interpretation of electrical resistivity measurements was created in the
electrical physical theory of Ohm’s law. Ohm’s law, Equation 3.16, states that the
product of the electrical current, I, and the resistance of the wire, R, through which the
current passes is equivalent to the potential difference, V, across the conductor:
V = IR (3.16)
According to Gibson and George (2003), this relationship is best represented by
envisioning a current passing through a thin wire. The expounded application of Ohm’s
law has made this relationship a capstone concept in the study of electrical theory. Units
for electrical potential, current and resistance are volts, amperes and ohms, respectively.
As indicated, the conductor element can be tangibly described as a wire element.
The resistance of the wire is related to both the geometric shape and material
attributes of the wire. The geometry of the wire is typically cylindrical, possessing a
length and cross-sectional area, and is made of a conductive material. The total resistance
53
of the wire element, R, is the product of the material resistivity, ρ, and the ratio of the
wire length and cross-sectional area:
R = ρ (L/A). (3.17)
Considering the physical relationship between the geometry of the conductor and
the material property, Equation 3.16 can be manipulated to from Equation 3.17 to
determine the material resistivity of the conductor element.
This form states that the units for resistivity depend on the volume of the space
through which the current travels. Typical units for resistivity, ρ, include ohm-meters and
ohm-centimeters (Gibson & George, 2003).
In a similar situation, the measurement of the potential difference can be related to
the dissipation of the electrical current within an infinite, homogenous half-space. In this
instance, the application of an electrical current travels in radial fashion out from the
point of origin. During the current application, the resistance at any location away from
the source point within the homogeneous mass can be found by determining the radius
from the point of origin and the surface area of the respective hemispherical equipotential
surface. Relating this model to the original wire example, Equation 3.17 can be rewritten
using the radius, r, as the distance for which the current travels and the surface area of the
resulting equipotential surface, 2πr2.
R = ρ(r/2πr2) = (ρ/2πr) (3.18)
Equation 5.4 describes the system resistance at any point away from the source
point within the homogeneous mass. Using the resistance term from the mentioned
homogeneous earth model, Equation 3.19 relates the resistance of the earthen model to
Ohm’s law:
54
V = IR = I (ρ/2πr) (3.19)
where
U = potential, in V,
ρ = resistivity of the medium, and
r = distance from the electrode.
Likewise, the potential difference between any two points within the
homogeneous mass would be the difference between the two equipotential surfaces, as
expressed in Equation 5.5 (Gibson & George, 2003):
𝜌𝐼 𝜌𝐼 𝜌𝐼 1 1
𝑈 = 2𝜋 𝑟𝐴 − 2𝜋𝑟 𝐵 = [𝑟 𝐴 − 𝑟 𝐵 ] (3.20)
2𝜋
where
rA and rB = distances from the point to electrodes A and B
Equation 3.21 relates the applied current, I, and measured potential difference, V,
to a constant value that accounts for spatial considerations, or the way in which the
reading was acquired. This model, a concept of equipotential surfaces, and means of
measuring potential differences between various surfaces, is fundamental in the
interpretation of collected field data (Gibson & George, 2003).
ρ = (2πU/I) [1/ [(1/ra)-(1/rb)] (3.21)
Figure 3.10 explains the electric field around the two electrodes regarding
equipotential and current lines. The equipotential represents the imagery projectiles or
bowls surrounding the current electrodes where the electrical potential is equal on each
one. The current lines represent a sampling of the infinite paths followed by the current,
which are defined by the condition that they must be everywhere and normal to the
equipotential surfaces.
55
Figure 3.10. Equipotential and current lines for a pair of current electrodes, A and B, in a
homogeneous half-space.
In a homogenous condition, the measured resistivity will be equivalent to the real
value of resistivity at a given location within the media. However, the occurrence of a
homogenous condition is rare in practice, if not nonexistent. A collected reading is
considered an apparent resistivity measurement in order to account for the inherent
heterogeneity of the subsurface. Visual resistivity is the resistivity of a theoretical,
homogeneous half-space that complements the measured current and potential difference
for a particular measurement scheme (United States Corps of Engineers, 2001). Mostly,
the obvious resistivity value is an average reading of the energized soil mass engaged
during the measurement. Numerically, the obvious resistivity can be expressed by
Equation 3.22:
56
𝜌𝐼 1 1 1 1
𝑉 = 𝑈𝑀 − 𝑈𝑁 = [ 𝐴𝑀 − 𝐵𝑀 + 𝐵𝑁 − 𝐴𝑁] (3.22)
2𝜋
where
UM and UN = potentials at M and N, and
AM = distance between electrodes A and M, etc.
These distances are always the actual distances between the respective electrodes,
whether or not they lie on a line. The quantity of the brackets, S is a function only of the
various electrode spacing. The quantity is denoted by 1/K, which allows the equation to
be rewritten as:
𝜌𝐼 1
𝑉= (3.23)
2𝜋 𝐾
where
K = the array geometric factor.
Equation 3.24 can be solved for ρ to obtain

𝑉
𝜌 = 2𝜋𝐾 . (3.24)
1
The resistivity of the medium can be established from the measured values of V, I
and K, the geometric factor. The variable K is a function only of the geometry of the
electrode arrangement.
When these measurements are made over real heterogeneous earth, as
distinguished from the fictional homogeneous half-space, the symbol ρ is replaced by ρa
for apparent resistivity. The resistivity surveying problem is reduced to its essence. The
use of apparent resistivity values from field monitoring at various locations, and with
various electrode configurations, is designed to estimate the actual resistivity of several
earth materials present at a site and to locate their confines spatially below the surface of
the site.
57
An electrode array with stationary spacing is used to investigate the lateral
changes in apparent resistivity, reflecting the lateral geologic variability or localized
anomalous features. The size of the electrode array is varied as we investigate how
changes in resistivity affect depth.
The resistivity is clearly affected by a material at increasing greater depths (hence
the larger volume) as the electrode spacing is increased. Because of this impact, a plot of
apparent resistivity against electrode spacing can be used to signal vertical variations in
resistivity.
The geometric coefficient, K, varies with array types. The spacing and layout of
the current and potential electrodes impact the induced equipotential fields created within
the earthen mass. The geometric factor for a general four probe system can be derived
from Equation 3.21(Gibson & George, 2003).
3.5.3. Relationship between Geology and Resistivity. Electric current flows in
subsurface materials at shallow depths through two main methods. They are electronic
conduction and electrolytic conduction. In electronic conduction, the current flow is via
free electrons, such as in metals. In electrolytic conduction, the current flow is via the
motion of ions in groundwater. In engineering and environmental surveys, electrolytic
conduction is the more common mechanism. However, electronic conduction is
important when conductive minerals are present, such as metal sulfides and graphite in
mineral surveys.
The resistivity of common rocks, chemicals, and soil materials (Keller &
Frischknecht, 1966: Daniels & Alberty, 1966: Telford et al., 1990) is shown in Figure
3.11. Igneous and metamorphic rocks typically have elevated resistivity values. The
58
resistivity of these rocks is greatly dependent on the degree of fracturing and the
percentage of the fractures that are filled with groundwater. Thus, a given rock type can
have a large range of resistivity, from about 1000 to 10 million Ω⋅m, depending on
whether it is wet or dry. This characteristic is useful in the detection of fracture zones and
other weathering features, such as in engineering and groundwater surveys.
Sedimentary rocks, which are usually more porous and have higher water content,
normally have lower resistivity values compared to igneous and metamorphic rocks. The
resistivity values range from 10 to about 10000 Ω⋅m, with most values below 1000 Ω⋅m.
The resistivity values are dependent on the porosity of the rocks and the salinity of the
contained water.
Unconsolidated sediments generally have even lower resistivity values than
sedimentary rocks, with values ranging from about 10 to less than 1000 Ω⋅m. The
resistivity value is dependent on the porosity (assuming all of the pores are saturated) as
well as the clay content. Clay soil normally has a lower resistivity value than sandy soil.
However, one should note the overlap in the resistivity values of the different classes of
rocks and soils should be noted. This is because the resistivity of a particular rock or soil
sample depends on a number of factors, including the porosity, the grade of water
saturation, and the concentration of dissolved salts. The resistivity of groundwater varies
from 10 to 100 Ω⋅m, depending on the concentration of the dissolved salts (Loke el al.,
2011).
59
Figure 3.11. The resistivity of rocks, soils, and minerals (Keller & Frischknecht, 1966;
Daniels & Alberty, 1966; Telford et al., 1990).
3.5.4. Electrical Resistivity Array Configuration. In theory, soil resistivity
could be measured by using a single current source and receiver element. In practice, this
is not feasible due to the contact resistance between the earth and the electrode pair. To
overcome this phenomenon, four electrodes are used for measurement: two electrodes
provide current to the subsurface and two electrodes measure the potential difference
between the earth materials (Milson, 1996). Current electrodes are identified as C1 and
C2 (or A and B), and potential electrodes are identified as P1 and P2 (or M and N) (Loke,
2000), as shown in Figure 3.12.
The types of electrode arrays that are most commonly used are Wenner (Figure
3.12a, Schlumberger (Figure 3.12b), pole-dipole (Figure 3.12c) and Dipole-Dipole

60
(Figure 3.12d). Other electrode configurations are either used experimentally, on non-
geotechnical problems, or are no longer popular. Some of these include the Lee, half-
Schlumberger, Polar Dipole, and gradient arrays.
In any case, the geometric factor for any four-electrode system can be found from
Equation 3.22 and can be advanced for more complicated systems by using the rule
illustrated in Equation 3.20. It can also be seen from Equation 3.25 that the current and
potential electrodes can be interchanged without affecting the results; this property is
called reciprocity (Milson, 1996). For the purpose of this research, the discussion will
concentrate on the Dipole-Dipole array, which was used during these studies.
Unlike to the Wenner and Schlumberger arrays, the configuration of the Dipole-
Dipole array does not place the potential electrode pair inside the current electrode pair.
Current and potential electrode pairs have common interior spacing and are separated by
a distance ten times the internal spacing of the electrode pair. The Dipole-Dipole array is
ordinarily used for performing tomography survey due to the array’s ability to resolve
lateral variations. In comparison to the Wenner and Schlumberger arrays, the Dipole-
Dipole array has a weak signal and is more susceptible to the effects of ambient or
cultural, noise.
If the division between both pairs of electrodes is the same, a, and the division
between the centers of the Dipoles is restricted to a (n+1), the resistivity is given by
𝑉
ρa = πaπ(n+1) (n+2) 1 . (3.25)
3.5.5. Wenner Array. The Wenner array is described by the equal spacing
between all four electrodes. The two current electrodes, C1 and C2, are placed on the
outside of the array during which the potential electrodes, P1 and P2, reside inside of the
61
array. The potential measurements are possessed at the mid-span of the potential
electrodes at a depth of approximately 0.5 to 1.0 times the electrode spacing (also known
as the spacing). Different depth measurements are made by varying the period spacing of
the array, as shown in Figure 3.12a.
Figure 3.12. These are some commonly used electrode arrays and their geometric factors.
Note that for the multiple gradient arrays, the total array length is‘(s + 2) an’, the distance
between the center of the potential Dipole pair is P1-P2 and the center of the current pair
C1-C2 is given by ‘ma.’ 'British Geological Survey (c) NERC 2013. K = Geometric
Factor.
In an exemplified homogeneous earth model, the sensitivity of the Wenner array
provides a pattern with strong horizontal layering immediately below the potential
electrode pair. Because the focus of vertical electrical sounding (VES) is to differentiate
between horizontal layering beneath a common point, the Wenner array is a practical
62
array for this application. The high signal of the Wenner array also makes the array
suitable for use in noisier environments (Loke, 2000).
3.5.6. Schlumberger Array. The Schlumberger array is arranged with two
current electrodes on the outside of the array, set apart by a distance of at least five times
the space between the two interior potential electrodes. The potential difference
measurement is believed to show incorrect at the mid-span of the potential internal
electrodes, at a depth of approximately one-half of the length between the current
external electrodes. Similar to the Wenner array, the Schlumberger array provides a high
signal directly below the potential electrode pair, as shown in Figure 3.12b.
The Schlumberger array is preferred for VES applications due to the strong
horizontal resolution and ease of setup in the field. As compared to the Wenner array
where all four electrodes must be repositioned after each test,
The Schlumberger array is preferred for VES applications due to its strong
horizontal resolution and ease of setup in the field. As compared to the Wenner array,
where all four electrodes must be repositioned after each test, the Schlumberger array
only requires that the two exterior current electrodes be moved to acquire a new
measurement. The potential interior electrodes are moved only as the current electrodes
are spaced beyond the practical limits of the survey. That movement occurs when the
ratio between the potential electrode spacing and the space between the exterior current
electrode and positional electrodes in the mid-span is greater than 0.4 (United States
Corps of Engineers, 1995).
3.5.7. Dipole-Dipole Array. The Dipole-Dipole array is logistically the most
appropriate in the field, particularly for large spacing and for 2-D imaging. The
63
convention for the Dipole-Dipole array shown in Figure 3.12d is that the current and
voltage electrode spacing are the same, a, and the spacing against them is an integer
multiple of a. The Dipole-Dipole array is usually used for performing tomography
surveying due to the array’s ability to resolve lateral variations. For a comparison against
the Wenner and Schlumberger arrays, the Dipole-Dipole array has a weaker signal and is
very susceptible to the effects of ambient, or cultural, noise. The apparent resistivity
reading recorded using the Dipole-Dipole array represents a condition present at the mid-
span of the array length that occurs at a depth amounting to one-half the product of the
Dipole electrode spacing, a, and one plus the separation factor (n+1).
3.6. ELECTRICAL RESISTIVITY TOMOGRAPHY DATA ACQUISITION
ERT data are typically acquired with the use of resistivity meter. Resistivity meter
is connected to the electrode cables, and the electrode cables are attached to metal stakes
pounded to the ground using rubber-band. In some instances, it is essential to water the
ground in proximity of the metal electrode to ensure a good contact between the metal
stake and ground. Newer resistivity meters are equipped with multiple channels, which
allows multiple electrodes to be engaged and measurements to be taken through each
channel. For instance, the SuperSting R8 resistivity meter, produced by Advanced
Geosciences, Inc., is equipped with eight channels. Subsequently, the system engages
nine electrodes to collect eight different potential difference measurements for each
current injection (Advanced Geosciences, Inc., 2006).
Planning and preparation are a significant part of ERT survey. Desired depth of
investigation, acquisition time, data resolution must be considered prior to each ERT
64
survey. The sequencing information considers the array style and information pertaining
to the electrode locations (or electrode address) during each measuring sequence. There
are no theoretical limits to the depth of penetration. Therefore, as the electrode spacing
increases, the signal strength decreases. At a certain electrode separation distance, the
signal strength is too low to provide reliable measurements of the potential difference.
Practical limits should be instilled that consider the signal strength of the particular array
type and equipment ability. Advanced Geosciences, Inc., (2008) suggested that when
considering depth of penetration for tomography applications, practitioners can generally
assume that the depth of penetration is approximately 15 to 25 % of the length of the
electrode spread. The survey resolution is also related to electrode spacing.
Coskun (2012) recommends the commonly used Dipole-Dipole array because it
offers better lateral and vertical resolutions than the other arrays. It is also worth noting
that the data acquisition time for a Dipole-Dipole array using 72 electrodes is typically
45minutes, and around 2 hours and 3 hours for 84 electrodes, and 168 electrodes,
respectively after survey set-up. Loke (1999) shares a similar view as Coskun, adding
that the Dipole-Dipole array is more suitable for investigating karst terrain because of its
ability to detect sharp changes in bedrock topographies.
Current practices propose that the electrode spacing should not be greater than
twice the size of the object or feature to be imaged. The design of the survey (i.e., survey
run length, electrode spacing, and array type) directly impacts the depth of penetration
and resolution (Advanced Geosciences Inc., 2008). It is not always possible or practical
to image a survey line or area in one deployment of electrodes. However, this can vary
depending on the preference between the resolution and imaged depth. Typically,
65
increasing the electrode spacing will increase the imaged depth and reduce the resolution
of the section and vice versa. However, it is often recommended that if the desire is to
increase the imaged depth, more electrodes should be added to the section instead of
increasing the electrode spacing. To continue a survey after completion of the initial data
collection, electrodes may be configured to collect additional data along a common
survey line or area, using roll-along survey techniques (Loke 2000), as shown in Figure
3.13.
Figure 3.13. SuperSting R8/IP resistivity (left). Switch Box (AGI) connecting passive
cables (middle). ERT field setup (right).
66
3.7. ELECTRICAL RESISTIVITY TOMOGRAPHY DATA PROCESSING
After data acquisition, the raw data are typically transferred to the laptop for
processing. During data processing inversion software transforms the apparent resistivity
values measured from the field to true resistivity by applying backward and forward
modeling in a process called inversion. There are a variety of processing software
available these days, including RES2DINV (Geotomo), Earth Imager (AGI), ZondRes2D
(Zond), etc. In this study, Res2DInv is used for ERT data processing. The RES2DINV
program uses the cell-based method in which the model parameters are the resistivity
values of the model cells, and the data provides the measured apparent resistivity values
(Loke, 2011). The mathematical link between the model parameters and the model
response for the 2-D and 3-D resistivity models is provided by either the finite-difference
(Dey & Morrison, 1979) or finite-element methods (Silvester & Ferrari, 1990).
The ERT data processing involves two main steps. The first step is to the
inspection of the resistivity data sets for the presence of any points that have anomalously
high or low apparent resistivity values, called “bad data points”, and subsequently
removing them when necessary (Figure 3.14b). Bad data points can be a result of several
factors, such as the failure or malfunction of equipment during the survey and very poor
electrode subsurface contact due to dry soil or shorting across the cable caused by very
wet subsurface conditions.
The second step is to run an inversion of the data. The inversion involves some
iterative calculations and the generation of a 2-D resistivity image of the subsurface to
represent the actual resistivity of the sections. The final result of the inversion is a 2-D
resistivity image of the subsurface showing the distribution of resistivity across a profile.
67
Additionally, pseudo section can be displayed and shows distribution of apparent
resistivity values (Figure 3.14a).
The root-mean squared (RMS) error is used to measure the difference between
calculated and measured apparent resistivity values. In practice, the lower the error, the
better the data quality is. However, the model with the low RMS error can sometimes
show unrealistic variations in the model resistivity values, and might not be best model
from geological perspective. A careful approach is to choose the model at the iteration
after which the RMS error does not change significantly. According to Loke (1999), an
RMS error of 5% is recommended for a geologic model of good quality.
Figure 3.14. An example of a field data set with a few bad data points. The apparent
resistivity data in (a) pseudosection form and in (b) profile form. (Loke, 1999).
68
3.8. ELECTRICAL RESISTIVITY TOMOGRAPHY DATA INTERPRETATION
The interpretation is based on the inverse model generated form the inversion
software, RES2DINV. The chart showing colors with various ranges of resistivity is used
to estimate the resistivity of the imaged subsurface earth materials.
Factors such as porosity, conductivity, saturation, salinity, clay content, lithology
and temperature can affect the ability of different materials to conduct electrical current.
Accordingly, materials of the same mineral content may exhibit different resistivity
values. For example, both the top-of-rock limestone and air-filled voids typically are
characterized by high resistivity values ranging from 50 ohm-m to 107 ohm-m., which
can be seen in Table 3.2. When an air-filled void is entirely embedded in limestone, it
usually cannot be easily detected on the resistivity data because of low resistivity
contrast.
Furthermore, dry soil usually has a much higher resistivity than saturated soil. The
same situation appears with weathered and un-weathered rock. Weathered rock is usually
more porous and fractured, and it becomes more saturated with groundwater; as a result,
weathered rock has a lower resistivity than un-weathered/the top of rock.
Table 3.2. Resistivity of common earth materials (Robinson, 1988).

69
According to previous studies (Torgashov, 2012) conducted in southwest
Missouri, typical resistivity values for the subsurface materials are characterized as
follows:
 Moist clays in southwest Missouri are normally characterized by low resistivity
values, usually less than 100 ohm-m, and may vary due to different degrees of
saturation, porosity, and layer thicknesses.
 Moist soils and intensively fractured rocks intermixed with clay typically have
resistivity values between 100 and 400 ohm-m. Such variation is explained by
different porosity, saturation, clay content and layer thicknesses.
 Relatively top-of-limestone with minimal clay content is characterized by higher
resistivity values, typically more than 400 ohm-m. Resistivity values of the top of
limestone may vary due to varying layer thickness, moisture content, porosity,
saturation, and impurities.
 Air-filled cavities usually show very high resistivity values, usually more than
10,000 ohm-m, but again these are variable depending on the conductivity of the
surrounding strata and depth/size/shape of a void.
 Zones of electrical resistivity contrast are where relatively the top of rock is
surrounded by moist, loose materials (such as clay) or where air-filled voids are
embedded in relatively the top of limestone. These zones can be successfully
detected by electrical resistivity tools.

70
4. RESEARCH EXPERIMENTS
4.1. STUDY SITE
The study site is located immediately within the southwestern limits of southwest
Missouri of the Ozarks physiographic region which is characterized by undulating to
rolling plains. Elevations of the region range between ~900 to ~1,500 ft. (asl). Southwest
Missouri lies on the western side of the Ozark Uplift where the rock layers dip gently
towards the west with minor faulting and folding, regionally. It is underlain by thick
Mississippian-age limestone with high porosity and susceptible to dissolution. Uneven
dissolution of this formation has resulted in a highly irregular bedrock-overburden
interface. Therefore, karst density in the area is among Missouri’s highest (Figures 4.1
and 4.2). The details of the studies are described in the next two sections.
This research emphasizes the need to apply efficient means for investigating
hazards to urban development such as an unstable soil foundation for structures, flood
hazards, groundwater contamination, and public safety in regards to potential catastrophic
collapse.
71
Figure 4.1. Southwest Missouri map where the study was conducted.
(http://www.nationsonline.org/oneworld/map/USA/missouri_map.htm).
Figure 4.2. Map showing the location of the study area Zone A and Zone B (Courtesy of
Google Earth).
72
4.2. MULTICHANNEL ANALYSIS OF SURFACE WAVES (MASW)
4.2.1. Data Acquisition. MASW data were acquired at 34 locations parallel and
perpendicular to the ERT traverses oriented west to east. Total number of 272 MASW
profiles in both west to east and north to south directions were acquired.
Data acquisition starts by the recording software launch and arming of the trigger
geophone. Once the MASW array initialized, operator gives the command to release the
seismic source (20 lb. sledgehammer) which notifies the seismograph to start data
recording. From three to five seismic records were collected at each traverse for a
purpose of staking to minimize the environment noise.
The main goal of the study was to develop appropriate acquisition parameters for
MASW surveys conducted in karst terrain topography in order to improve the data
resolution and maximize the depth of investigation. Figure 4.3 shows employed MASW
equipment: Seistronix RAS-24 seismograph and 24 vertically polarized 4.5 Hz
geophones. Source offsets of 10 ft. (3 m) and 30 ft. (9 m) were used for data acquisition.
Receiver spacings of 2.5 ft. (0.76 m) and 5 ft. (1.5 m) were used in attempt to increase the
vertical resolution of MASW data and to minimize lateral smearing. These parameters
were chosen to achieve the investigation depth of approximately 30 ft. (9m).
For each MASW traverse, eight measurements per site were obtained (four
records with seismic arrays oriented in west-east direction and four records with seismic
arrays oriented in north-south direction (parallel and perpendicular to the direction of
corresponding ERT traverses). Table 4.1 and Figure 4.4 shows parameters used to collect
MASW data in the study.

73
Figure 4.3. Active MASW data acquisition.
Table 4.1. Survey parameters used during MASW surveys in this study.
74
Figure 4.4. Schematic diagram of MASW data acquisition setup.
4.2.2. Data Processing and Interpretation. All seismic records acquired in this
research were processed using SurfSeis4 software, developed by the Kansas Geological
Survey.
Figures 4.5 to 4.7 display examples of the multichannel seismic records with
overlay of preliminary data quality evaluation provided by the processing software. This
step of data processing is to check overall data quality. This step is required to get a
primary idea about the overall data quality, surface wave velocity range, dominant
frequency, and the relative probability of higher modes contamination. “Excellent” and
“good” quality data sets suggest that the surface waves are the most prominent seismic
events (with highest signal-to-noise ratio) in the particular seismic record. “Fair” and
“poor” quality data sets are typically contaminated by noise which can be a result of the
irregularity of the bedrock surface, presence of voids, proximity of the array to the
75
construction site, receivers malfunction, etc. These seismic records can be not suitable for
processing due to a lack of any recognizable surface wave signal.
Figure 4.5. The multichannel seismic records with overlay of preliminary data quality
evaluation provided by the processing software (SurfSeis4).
Figure 4.6. The multichannel seismic records with overlay of preliminary data quality
76
Figure 4.7 The multichannel seismic records with overlay of preliminary data quality
Following the visual evaluation, initial processing was initiated and included
staking multiple shots as well as applying muting. The purpose of muting application is
to remove specific seismic energy which arrives in specific time frame as a result of its
velocity of propagation characteristics (Miller, Ivanov, et al., 2001). After proper muting,
the quality and observation range of the desired mode (i.e., fundamental mode) improve
significantly. The downside of muting is that it introduces a high velocity gradient feature
at the low frequency end of the dispersion curve, which is not present before muting. This
artifact can be easily handled by using the dispersion curve from the non-muted data that
covers the same frequency range, so the artificially high velocity values can be ignored
and are thus considered harmless (Miller, Ivanov, et al., 2001).
To show the effect of muting, the muting was applied to the seismic record
collected with the source offset of 10ft. (3.04 m) and the receiver spacing of 2.5ft. (0.76)
as it shown in Figures 4.8 and 4.9 to compare their dispersion curves before and after
application of muting. It can be seen from Figure 4.8A that seismic energy associated
77
with surface waves dominates in the seismic record. Figure 4.8B displays the dispersion
image with up-warping artifact appeared at high frequencies range that can be caused by
contamination from higher modes. The fundamental mode dominates a wide frequency
range between 38 and 61 Hz (with corresponding velocities between 1,060ft/sec and
1,735ft/sec) on the overtone image. However, the resulting 1-D shear wave velocity
profile can be distorted if the extracted dispersion curve will be improperly picked from
the ambiguous overtone image.
Figure 4.9B displays the dispersion image generated from the muted seismic
record. The presence of higher mode surface waves in the original seismic record is
almost completely eliminated, and the seismic energy associated with higher modes
disappeared. The fundamental mode can be observed within a frequency range between
36 and 63 Hz (with corresponding velocities between 1,040ft/sec to 1,720ft/sec).
Their attempt to observe seismic energy associated with fundamental mode in the
range between 36 and 67 Hz by boosting the amplitude of the phase-velocity – frequency
spectrum did not bring any results. Thus, the seismic energy associated with higher
modes is significantly higher, and corresponding dispersion images inhibit the ambiguous
character of dispersion curve extraction of the fundamental mode trend below 40 Hz.
78
Figure 4.8. (A) The multichannel seismic record collected with the source offset of 10ft.
(3.04 m) and receiver spacing of 2.5ft. (0.76m); (B) Corresponding overtone image with
the fundamental and higher modes identified.
79
Figure 4.9. (A) The multichannel seismic record collected with offset of 10ft. (3.04 m)
and receiver spacing of 2.5ft. (0.76 m) with applied muting. (B) Corresponding overtone
image with the fundamental and higher modes identified.
80
Following the generation of overtone images, fundamental mode dispersion
curves were estimated. The curves then were inverted to obtain 1-D shear wave velocity
profiles (Figure 4.10).
Figure 4.10. The results of the borehole BH1. In the same graph, the 1-D shear wave
velocity profile deduced from the MASW survey along the Traverse No. 5 is shown. The
MASW method estimated with accuracy the shear wave velocity in the layers and
assisted in the layer identification of all the neighboring traverses.
81
Based on the author’s previous experience, shear wave velocity in excess of 1,500
ft/sec corresponds to top of weathered rock. The value was used for this study for the
purpose of mapping top of bedrock. To verify the interpretations, the MASW data were
compared with bore holes in terms of depth to bedrock. As seen in Figure 4.10, the depth
of interpreted top of bedrock on the MASW profile of collected by the 10 x 2.5ft. array
configuration corresponded well to the top of rock obtained by drilling.
4.3. ELECTRICAL RESISTIVITY TOMOGRAPHY (ERT)
4.3.1. Data Acquisition. ERT data were acquired along a total of sixteen
traverses oriented in W-E direction. The traverse orientations were selected to image the
dominant north-south trending joint sets in the study area. The ERT traverses were
spaced at 400ft. intervals.
The ERT data were acquired using an AGI SuperSting R8/IP, a multichannel
electrode system powered by two 12-Volts batteries (Figure 4.11). During the field set-
up, electrode cables were attached to metal stakes pounded to the ground using rubber-
band. Switch box was also used to connect the electrodes (passive electrodes) to the
resistivity meter. In an effort to ensure good quality ERT data are acquired, the field
crews routinely performed contact resistance tests. The contact resistance test was
performed prior data acquisition to ensure that all of the metal stakes are connected
properly to the cables and to the ground.
One hundred sixty eight electrodes were spaced at 5ft. intervals covering a length
of 835ft. and with expectation to image up to the depth of 100ft. As the length of the ERT
traverses exceeded the array length of 835ft., 50% roll-along overlap option was used. A
82
dipole-dipole electrode configuration was utilized in this study. This array gives a better
lateral resolution than the other types of arrays (Schlumberger and Wenner arrays) and is
more suitable for investigating karst terrain.
Figure 4.11. ERT field set up.
4.3.2. Data Processing and Interpretation. ERT data were processed using
Res2DInv software. As a first step of data processing, “bad” data points were removed
manually. The bad data points were either anomalously high or low values. Then,
“forward modeling” subroutine was used to specify the subsurface resistivity in order to
calculate the apparent resistivity that would be measured by a survey over such a
structure. As a result of data processing, 2D cross-sectional images were generated, as
shown in Figures 4.12 and 4.13.

83
Figure 4.12. Typical example of 2-D uninterpreted ERT model. Traverse No. P1 and
three overlapping ERT traverses acquired at the study site along 2,440ft.
Figure 4.13. Typical example of 2-D uninterpreted ERT model. Traverse No. P2 and
three overlapping ERT traverses acquired at the study site along 2,440ft.
84
Regarding the interpretation of the ERT data, , it was determined that moist soils
are characterized by resistivity values of less than 125 ohm-m; dry soils by resistivity
values greater than 125 ohm-m; moist weathered and/or fractured rock by resistivity
values less than 600 ohm-m; moist fractured rock with moist piped clay/soil-fill by
resistivity values less than 125 ohm-m; and drier, possibly less weathered rock by
resistivity values greater than 600 ohm-m (Anderson et al., 2006; Muchaidze, 2009; Myat
et al., 2008; Robison & Anderson, 2008). Figures 4.14 and 4.15 show the interpreted top
on two representatives ERT traverses (Traverse No. P1 and Traverse No. P2).
Figure 4.14. 2-D Interpreted versions of resistivity Traverse No. P1. Depth to top of
bedrock is around 16 ft. and corresponds to the 125 ohm-m contour interval.
85
Figure 4.15. 2-D interpreted versions of resistivity Traverse No. P2. Depth to top of
bedrock varies from 10 to 19 ft. and corresponds to the 125 ohm-m contour interval.
Boring control was used to verify the interpretations. Bore hole (BH1) was drilled
along ERT Traverse No. P1. The borehole was drilled to a depth of 98ft. (30.6m) below
subsurface. The top of rock was encountered at a depth of about 15ft. (4.6m), which was
consistent with the ERT interpretations (Figure 4.16).
The ERT data interpretations were also compared with MASW data
interpretations in terms of estimated depth to the top of bedrock (Figures. 4.17, 4.18, and
4.19). The comparison revealed a good agreement between the two.

86
Figure 4.16. Interpreted versions of resistivity Traverse No. P1. The top of rock
correlates reasonably well with the 125 ohm-m contour interval. The borehole location
has been superposed in a red dashed line. The 200ft. mark on the resistivity profile
corresponds with MASW profile No.1; the 800ft. mark corresponds with profile No. 2
location; the1, 400ft. mark corresponds with profile No.3; the1, 800ft. mark corresponds
with profile No. 4; and the 2,200ft. mark corresponds with profile No.5.
87
Figure 4.17. (A) ERT Traverse No. P1 tied with MASW profile No. 1. at the 200ft. (61m)
mark; (B) Corresponding 1-D shear-wave velocity profile. MASW depth to top of
weathered rock (“acoustic” top of rock) is identified at 15ft. (4.6m) depth. Red color line
on Figure 8.8 (B) indicates interpreted depth to top of rock.
Figure 4.18. (A) ERT Traverse No. P1 tied with MASW profile No. 2.at the 800ft.
(244m) mark; (B) Corresponding1-D shear-wave velocity profile. MASW depth to top of
weathered rock (“acoustic” top of rock) is identified on 7ft. (2.1m) depth. Red color line
on Figure 8.9 (B) indicates interpreted depth to top of rock.
88
Figure 4.19. (A) ERT Traverse No. P1 tied with MASW profile No. 5 and borehole
(BH1), at the 2,200ft. (670m) mark; (B) Corresponding 1-D shear-wave velocity profile.
MASW depth to top of weathered rock (“acoustic” top of rock) is identified at 8.5ft.
(2.6m) depth. Red color line on Figure 8.10 (B) indicates interpreted depth to top of rock.
89
5. COMPARATIVE ANALYSES
A comparative analysis is followed to analyze the performance of each MASW
configuration. Two types of comparative analyses are performed in this research:
qualitative and quantitative. The qualitative comparisons are solely based on visual
assessments of overtone images. The quantitative comparisons are based on the
assessment of the quantities calculated for each MASW configuration.
5.1. QUALITATIVE COMPARISON
The goal of qualitative analysis approach is twofold: 1) to compare overtone
images and dispersion curves; 2) to compare 1-D shear wave velocity profiles. The
comparisons are performed using different array configurations.
5.1.1. Comparison of Overtone Images and Dispersion Curves. Two MASW
profiles (Traverse No.1 and Traverse No. 2) collected by different array configurations
(geophone spacing, source offset, and the array orientation) were analyzed in attempt to
evaluate the effect of different parameters on the process of dispersion curve extraction
followed by modelling of 1-D shear wave velocity profiles. More specifically, the
MASW data results were analyzed for frequency and phase velocity ranges, interpreted
depth to top of rock, maximum depth of the investigation, signal-to-noise ratio, difference
between the theoretical and the experimental dispersion curves (root-mean-square error),
and the maximum number of iterations for inversion procedure.
5.1.2. MASW Traverse No. 1 Oriented West-East. Results from dispersion
curves and shear wave velocity profiles for all the tested parameters are presented in
Table 5.1.
90
Table 5.1. Results from all the dispersion curves and generated 1-D shear wave velocity
profiles for MASW Traverse No. 1. X1 - source offset; dx - geophone spacing; D -
receiver spread length; 𝜆𝑚𝑎𝑥 maximum resolvable Rayleigh wave wavelength; 𝜆𝑚𝑖𝑛
minimum resolvable Rayleigh wave wavelength; Z max penetration depth; S/N - signal-
to-noise ratio; RMS -root-mean-square error.
Figure 5.1A displays the MASW data collected along Traverse No.1 oriented
west-east with source offset of 10ft. and geophone spacing of 2.5ft. Phase velocities were
manually picked using 11 equally spaced points from the fundamental-mode dispersion
curve in Figure 5.1B (represented by white dots) and used for an inversion process to
generate a 1-D shear wave velocity model (Figure 5.1C).
The dispersion analysis shows that the data possesses dominant frequencies at
approximately 𝑓𝑚𝑖𝑛 = 31 Hz to 𝑓𝑚𝑎𝑥 = 65 Hz. The phase velocity varied between 𝑉𝑝ℎ𝑚𝑎𝑥
= 1,541ft/sec and 𝑉𝑝ℎ𝑚𝑖𝑛 = 895ft/sec, corresponding to the Rayleigh waves wavelength of
𝜆𝑚𝑖𝑛 = 14ft. and 𝜆𝑚𝑎𝑥 = 55ft. The picked frequency and phase velocity ranges fall within
the determined fundamental mode. With these values and the traverse length, the
minimum and maximum penetration depths is 𝑍𝑚𝑎𝑥 = 18ft.
These parameters yielded the inverted Vs model that was obtained with the high
signal-to-noise ratio (S/N) of 98 % and an RMS error of 6.5 % with nine iterations. The
91
fundamental mode is clearly observable in the absence of a higher mode signs. The
dispersion curve is continuous and distinguishable at frequencies higher than 30 Hz. The
theoretical dispersion curve matches the experimental curve at higher frequencies. These
results can be due to the employed combination of the receiver spread length and the
source offset.
Figure 5.1. MASW survey data collected along Traverse No.1 oriented west-east array.
(A) The 24 channel record (shot gather); (B) The dispersion image (overtone) with
superposed dispersion curve (phase velocity versus frequency) where the fundamental
mode is quite clear. (C) The 1-D shear wave velocity profile, deduced from the inversion
technique. 1) Risk of higher mode; 2) The signal-to-noise ratio (S/N); 3) R.M.S. error; 4)
Iterations.
92
west-east with source offset of 10ft. and geophone spacing of 5ft. Phase velocities were
= 1,696 ft/sec and 𝑉𝑝ℎ𝑚𝑖𝑛 = 917ft/sec, corresponding to the Rayleigh waves wavelength
of 𝜆𝑚𝑖𝑛 = 17ft. and 𝜆𝑚𝑎𝑥 = 56ft. The picked frequency and phase velocity ranges fall
within the determined fundamental mode. With these values and the traverse length, the
signal-to-noise ratio (S/N) of 88 % and an RMS error of 9.2 % with eight iterations. The
fundamental mode lacks separation from the higher mode (this overlap might be due to
the similarities in phase velocities in that frequency ranges). The dispersion curve is non-
coherent at frequencies higher than 30 Hz. The theoretical dispersion curve fairly
matches the experimental curve at higher frequencies. Low quality survey results can be
due to the employed combination of the receiver spread length and the source offset.
93
Figure 5.2. MASW survey data collected along Traverse No. 1 oriented west-east array.
Iterations.
Figures 5.3 and 5.4 display the MASW data collected along Traverse No. 1
oriented west-east on the basis of the comparison between the MASW survey output for
the employed source offset of 30 ft. (9.14m) with a receiver spacing of 2.5ft. (0.76m) and
the source offset of 30ft. (9.14m) with a receiver spacing of 5ft. (1.52m), respectively.
Figures 5.3 and 5.4 show how the various combinations of source offset and geophone
94
spacing can affect the resulting overtone images and extracted dispersion curves for the
fundamental and higher modes.
= 1,533 ft/sec and 𝑉𝑝ℎ𝑚𝑖𝑛 = 859 ft/sec, corresponding to the Rayleigh waves wavelength
signal-to-noise ratio (S/N) of 85 % and an RMS error of 13.2 % with six iterations. The
fundamental mode is hardly distinguishable and very limited in the presence of a higher
mode. The dispersion curve is continuous and distinguishable at frequencies higher than
30 Hz. The theoretical dispersion curve fairly matches the experimental curve at higher
frequencies. These results can be due to the employed combination of the receiver spread
length and the source offset.

95
Iterations.
generate a 1-D shear wave velocity model (Figure 5.4C). The dispersion analysis shows
that the data possesses dominant frequencies at approximately 𝑓𝑚𝑖𝑛 = 38 Hz to 𝑓𝑚𝑎𝑥 = 53

96
Hz. The phase velocity varied between 𝑉𝑝ℎ𝑚𝑎𝑥 = 1,744ft/sec and 𝑉𝑝ℎ𝑚𝑖𝑛 = 995ft/sec,
corresponding to the Rayleigh waves wavelength of 𝜆𝑚𝑖𝑛 = 19ft. and 𝜆𝑚𝑎𝑥 = 46ft. The
picked frequency and phase velocity ranges fall within the determined fundamental
mode. With these values and the traverse length, the minimum and maximum penetration
depths is 𝑍𝑚𝑎𝑥 = 19ft. These parameters yielded the inverted Vs model that was obtained
with the high signal-to-noise ratio (S/N) of 92 % and an RMS error of 12.6 % with six
iterations. The fundamental mode lacks separation from the higher mode (this overlap
might be due to the similarities in phase velocities in that frequency ranges). The
dispersion curve is non-coherent at frequencies higher than 30 Hz. The theoretical
dispersion curve fairly matches the experimental curve at higher frequencies. Low quality
survey results can be due to the employed combination of the receiver spread length and
the source offset.

97
Iterations.
5.1.2. MASW Traverse No. 1 Oriented North to South. Figure 5.5A displays
the MASW data collected along Traverse No. 1 oriented west-east with source offset of
10ft. and geophone spacing of 2.5ft. Phase velocities were manually picked using 11
equally spaced points from the fundamental-mode dispersion curve in Figure 5.5B
(represented by white dots) and used for an inversion process to generate a 1-D shear
wave velocity model (Figure 5.5C).

98
= 1,740ft/sec and 𝑉𝑝ℎ𝑚𝑖𝑛 = 850ft/sec, corresponding to the Rayleigh waves wavelength
minimum and maximum penetration depths is 𝑍𝑚𝑎𝑥 = 21.5ft.
signal-to-noise ratio (S/N) of 98 % and an RMS error of 4.8 % with nine iterations. The
fundamental mode is clearly observable in the absence of a higher mode signs. The
dispersion curve is continuous and distinguishable at frequencies higher than 30 Hz. The
theoretical dispersion curve matches the experimental curve at higher frequencies. These
source offset.
99
Figure 5.5. MASW survey data collected along Traverse No. 1 oriented north to south
array. (A) The 24 channel record (shot gather); (B) The dispersion image (overtone) with
Iterations.
Figure 5.6A displays the MASW data collected along Traverse No. 1 oriented

100
signal-to-noise ratio (S/N) of 70 % and an RMS error of 4 % with twelve iterations, the
lowest of the eight trials. The fundamental mode is observable and can be distinguished
from the higher mode. The dispersion curve is continuous at frequencies higher than 30
Hz. The theoretical dispersion curve fairly matches the experimental curve at higher
frequencies. These results can be due to the employed combination of the receiver spread
length and the source offset.

101
Iterations.
oriented north-south on the basis of the comparison between the MASW survey output
for the employed source offset of 30ft. (9.14m) with a receiver spacing of 2.5ft. (0.76m)
and the source offset of 30ft. (9.14m) with a receiver spacing of 5ft. (1.52m),
respectively. Figures 5.7 and 5.8 show how the various combinations of source offset and
102
geophone spacing can affect the resulting overtone images and extracted dispersion
curves for the fundamental and higher modes.
north-south with source offset of 30ft. and geophone spacing of 2.5ft. Phase velocities
were manually picked using 9 equally spaced points from the fundamental-mode
dispersion curve in Figure 5.7B (represented by white dots) and used for an inversion
process to generate a 1-D shear wave velocity model (Figure 5.8C).
signal-to-noise ratio (S/N) of 88 % and an RMS error of 19 % with six iterations. The
fundamental mode is hardly distinguishable and limited in the presence of a higher mode.
The dispersion curve is continuous and distinguishable at frequencies higher than 30 Hz.
The theoretical dispersion curve fairly matches the experimental curve at higher
frequencies. Low quality survey results can be due to the employed combination of the
receiver spread length and the source offset.

103
Figure 5.7 MASW survey data collected along Traverse No. 1 oriented north to south
Iterations.
north-south with source offset of 30ft. and geophone spacing of 5ft. Phase velocities were

104
signal-to-noise ratio (S/N) of 76 % and an RMS error of 12 % with ten iterations. The
fundamental mode is hardly distinguishable and limited in the presence of a higher mode.
The dispersion curve is continuous and distinguishable at frequencies higher than 30 Hz.
The theoretical dispersion curve fairly matches the experimental curve at higher
frequencies. Low quality survey results can be due to the employed combination of the
receiver spread length and the source offset.

105
Iterations.
Eight configurations of MASW survey setup were tested. An average dispersion
curve was obtained by combining all 8 records using the frequency range from 27 to 54
Hz and the phase velocity range from 805ft/sec to 1,696ft/sec. This process was repeated
for all of the receiver array types, with the dominant frequency of the surface waves
varied from 30 to 53 Hz for the various array types. An excellent signal-to-noise ratio
was obtained for all of the records until a satisfactory match was reached between the
106
experimental and the theoretical dispersion curves with a relative error of less than 10%
to 15%.
The following conditions were common for all three inversions: 1) the 1-D shear-
wave velocity profile geometry was defined by ten model blocks (layers) with fixed
thicknesses increasing by 5% with depth, plus a model half-space; 2) the depth to the
average model half-space was determined as 18.6ft., corresponding to approximately
one-third of the maximum resolvable wavelength; 3) a maximum of 8 iterations is
possible.
5.1.3. MASW Traverse No. 2 Oriented West-East. Results from dispersion
curves and shear wave velocity profiles for all the tested parameters are presented in
Table 5.2.
Table 5.2. Results from all the dispersion curves and generated 1- D shear wave velocity
profiles for MASW Traverse No. 2. X1 - source offset; dx - geophone spacing; D -
receiver spread length; 𝜆𝑚𝑎𝑥 maximum resolvable Rayleigh wave wavelength; 𝜆𝑚𝑖𝑛
minimum resolvable Rayleigh wave wavelength; Z max penetration depth; S/N - signal-
to-noise ratio; RMS -root-mean-square error.
107
manually picked from the fundamental-mode dispersion curve in Figure 5.9B
of 𝜆𝑚𝑖𝑛 = 13.7ft. and 𝜆𝑚𝑎𝑥 = 40ft. The picked frequency and phase velocity ranges fall
signal-to-noise ratio (S/N) of 95 % and an RMS error of 5.2 % with eight iterations. The
fundamental mode is clearly observable in the absence of higher modes. The dispersion
curve is continuous and distinguishable at frequencies higher than 30 Hz. The theoretical
dispersion curve matches the experimental curve at higher frequencies. These results can
be due to the employed combination of the receiver spread length and the source offset.
108
superposed dispersion curve (phase velocity versus frequency); (C) The 1-D shear wave
velocity profile, deduced from the inversion technique. 1) Risk of higher mode; 2) The
signal-to-noise ratio (S/N); 3) R.M.S. error; 4) Iterations.
109
signal-to-noise ratio (S/N) of 85 % and an RMS error of 9 % with eight iterations. The
fundamental mode is observable in the presence of a higher mode. The dispersion curve
is continuous and distinguishable at frequencies higher than 30 Hz. The theoretical
dispersion curve matches the experimental curve at higher frequencies. Low quality
the source offset.

110
oriented west-east on the basis of the comparison between the MASW survey output for
the employed source offset of 30ft. (9.14m) with a receiver spacing of 2.5ft. (0.76m) and
the source offset of 30ft. (9.14m) with a receiver spacing of 5ft. (1.52m), respectively.
Figures 5.11 and 5.12 show how the various combinations of source offset and geophone
spacing can affect the resulting overtone images and extracted dispersion curves for the
fundamental and higher modes.

111
minimum and maximum penetration depths are 𝑍𝑚𝑎𝑥 = 14ft.
signal-to-noise ratio (S/N) of 98 % and an RMS error of 9.5 % with ten iterations. The
the similarities in phase velocities in that frequency ranges). The dispersion curve is
moderately distinguishable at frequencies higher than 30 Hz. The theoretical dispersion
curve fairly matches the experimental curve at higher frequencies. Low quality survey
source offset.
112
113
minimum and maximum penetration depths are 𝑍𝑚𝑎𝑥 = 14ft.
fundamental mode is visible and can be separated from the higher mode. The dispersion
curve is moderately distinguishable at frequencies higher than 30 Hz. The theoretical
dispersion curve fairly matches the experimental curve at higher frequencies. Low quality
the source offset.

114
5.1.4. MASW Traverse No. 2 Oriented North-South. Figure 5.13A displays the
MASW data collected along Traverse No.2 oriented north-south with source offset of
10ft. and geophone spacing of 2.5ft. Phase velocities were manually picked using 10
equally spaced points from the fundamental-mode dispersion curve in Figure 5.13B

115
fundamental mode is clearly observable in the absence of higher mods. The dispersion
116
Figure 5.13. MASW survey data collected along Traverse No.2 oriented north to south
117
signal-to-noise ratio (S/N) of 95 % and an RMS error of 10 % with ten iterations, the
lowest of the eight trials. The fundamental mode is clearly observable in the absence of
higher modes. The dispersion curve is continuous and distinguishable at frequencies
higher than 30 Hz. The theoretical dispersion curve matches the experimental curve at
higher frequencies. These results can be due to the employed combination of the receiver
spread length and the source offset.

118
Figures 5.15 and 5.16 display the MASW data collected along Traverse No.2
oriented north-south on the basis of the comparison between the MASW survey output
for the employed source offset of 30ft. (9.14m) with a receiver spacing of 2.5ft. (0.76 m)
and the source offset of 30ft. (9.14m) with a receiver spacing of 5ft. (1.52m),
respectively. Figures 5.15 and 5.16 show how the various combinations of source offset
119
and geophone spacing can affect the resulting overtone images and extracted dispersion
curves for the fundamental and higher modes.
north-south with source offset of 30ft. and geophone spacing of 2.5ft. Phase velocities
were manually picked using 10 equally spaced points from the fundamental-mode
dispersion curve in Figure 5.15B (represented by white dots) and used for an inversion
process to generate a 1-D shear wave velocity model (Figure 5.15C).
fundamental mode is clearly observable in the absence of higher modes. The dispersion
120
121
the similarities in phase velocities in that frequency ranges). The dispersion curve is non-
coherent at frequencies higher than 30 Hz. The theoretical dispersion curve fairly
matches the experimental curve at higher frequencies. Low quality survey results can be
due to the employed combination of the receiver spread length and the source offset.
122
Eight configurations of MASW survey setup were tested. An average dispersion
curve was obtained by combining all 8 records using the frequency range from 29 to 62
Hz and the phase velocity range from 834 ft/s to 1,541 ft/s. This process was repeated for
all of the receiver array types, with the dominant frequency of the surface waves varied
from 30 to 53 Hz for the various array types. An excellent signal-to-noise ratio was
obtained for all of the records until a satisfactory match was reached between the
123
experimental and the theoretical dispersion curves with a relative error of less than 10%
to 15%.
The following conditions were common for all three inversions: 1) the 1-D shear-
wave velocity profile geometry was defined by ten model blocks (layers) with fixed
thicknesses increasing by 5% with depth, plus a model half-space; 2) the depth to the
average model half-space was determined as 15ft., corresponding to approximately one-
third of the maximum resolvable wavelength; 3) a maximum of 9.7 iterations is possible.
When a satisfactory match is reached between the experimental and the numerical
dispersion curves (usually with a relative error of less than 10% to 15%), then the Vs
profile has been found and the researcher can proceed with determining the seismic
subsurface classification.
Generally, velocity values of less than 1,200ft/sec (366m/sec) are interpreted as
sand and unconsolidated material, while velocity values greater than 1,200ft/s (366m/s)
are interpreted as soft, weathered limestone.
5.2. DISPERSION OF RESOLUTION CURVE
Resolution of dispersion curve is another important factor in determining quality
of MASW data. The resolution along the velocity axis defines the ability to discriminate
a particular phase velocity from other velocities for a given frequency. The resolution
along the frequency axis determines the ability to discriminate a particular frequency
from other frequencies for a given velocity. The dispersion curve width is related to the
relationship between the frequency and the velocity of the surface waves.
124
To determine the resolution of the dispersion image in the frequency-velocity
domain (f-v), the length of the receiver spread needs to be increased (Park et al., 1998).
Forbriger (2003a) provided an analytical result to assess the resolution of the dispersion
image based on the Equation (5.1).
Δd = 1 / f D (5.1)
where Δd is the half-width between the neighboring minimum of dispersion
energy in the (f-v) domain, f is the frequency, and D is the length of the receiver array.
However, then understand that the resolution of the dispersion image could vary with the
algorithms that were used to generate the dispersion image in the frequency-velocity
domain. A current comparison of several different algorithms can be found in (Dal Moro
et al., 2003).
Three to five impacts generated by a 20 lb. sledgehammer were vertically stacked
at each MASW location using SurfSeis software developed by KGS. A record length of
2,048 milliseconds at a 1-millisecond sampling rate interval was employed.
The influence of these factors on the analysis is highly dependent on the data
quality such as high or low the signal-to-noise ratio, the confidence of dispersion curve
extraction, the R.M.S. error and the maximum number of iterations. A “good” quality
data set suggests the surface-wave energy is the most prominent seismic event in the
seismic record (i.e., the highest S/N), whereas a “bad” quality data sets are usually
contaminated by noise.
The quality of the acquired surface wave records can be evaluated in terms of the
resolution of the phase velocity spectrum, i.e., the sharpness of the amplitude peaks
125
observed at each frequency, the extractable frequency range and the continuity of the
fundamental mode high-amplitude band.
The quality of a dispersion curve is judged according to two criteria: the signal-to-
noise ratio and the general dispersion curve direction. A high S/N indicates a high
confidence in the obtained phase velocity. An S/N higher than 0.5 or more than 55% is
considered acceptable.
The quality of the MASW data can also be affected by natural geologic conditions
that may not produce well-defined dispersion curves and cannot be used to calculate
reliable shear-wave velocity models, only on the basis of the comparison between the
nearest source offset and variable receiver spacings. After that, the comparison between
the farther source offset and variable receiver spacings.
Two configurations of MASW geophone array results were generated with the
data Figure 5.17. The first nearest source offset was (X1) = 10ft. with a receiver spacing
of (dx) = 2.5ft., and the second nearest source offset was (X1) = 10ft. with a receiver
spacing of (dx) = 5ft.
The output of two different MASW array configurations are shown in Figure
5.17. Figure 5.17A and 5.17B show seismic field records collected using geophone
spacing of 2.5ft. (0.76m) with corresponding dispersive image; Figures 5.17 C and 5.17D
show seismic field records collected using geophone spacing of 5ft. (1.52m) with
corresponding dispersive image. For the estimation of dispersion curve resolution, a
double-ended arrow at frequencies of 30 Hz and 50 Hz is superposed over the dispersion
image. The double-ended arrow at 30 Hz is twice as long as that at 50 Hz. This is
determined by Equation (5.1) that indicates the resolution of the dispersion image is one-
126
half at 30 Hz compared with that at 50 Hz. For Figures 5.17C and 5.17D, the resolution
reduction of one-half (Δd is doubled) as shown in Figure 5.17D was expected. The
double-ended arrows in Figures 5.17A and 5.17B at 30 and 50 Hz are longer than 60%,
not 100% like the arrows in Figure 5.17C and 5.17D at 30 and 50 Hz, respectively.
Figure 5.17. The multichannel seismic record; (B) Corresponding overtone image in
which the fundamental mode dispersion is identified; (C) The multichannel seismic
record collected with doubled receiver spacing; (D) Corresponding overtone image in
which the fundamental mode dispersion is identified with noticeable discrepancies.
The third and fourth MASW array configurations are shown in Figure 5.18. The
third farthest source offset was 30 ft. (9.14m) with a receiver spacing of 2.5 ft. (0.76m),
and the fourth farthest source offset was 30 ft. (9.14m) with a receiver spacing of 5ft.
(1.52m).
The output of the third and fourth MASW array configurations are shown in
Figure 5.18. Figure 5.18A and 5.18B show seismic field records collected using
geophone spacing of 2.5ft. (0.76m) with corresponding dispersive image; Figures 5.18C
127
and 5.18D show seismic field records collected using geophone spacing of 5ft. (1.52m)
with corresponding dispersive image. For the estimation of dispersion curve resolution, a
double-ended arrow at frequencies of 30 Hz and 50 Hz is superposed over the dispersion
image. The double-ended arrow at 30 Hz is twice as long as that at 50 Hz. Both double-
ended arrows in Figure 5.18C and Figure 5.18D at 30 and 50 Hz were the same length
lengths, which also supports Equation (5.1) in that the resolution of the dispersion image
is determined by the length of the receiver array and the receiver spacing. Also, the
resolution (indicated by a double-ended arrow around the frequency of 30 Hz) of the
dispersive energy from the data with the farther source offset of 30ft. (9.14m) is better
than that of the data generated from seismic record collected with the nearest source
offset of 10ft. (3.04m).
As demonstrated by previous examples, Equation (5.1) might work well as the
frequency changes when the length of the receiver array is fixed. Furthermore, it is
usually necessary to double the length of the receiver array for a given frequency, and the
resolution increase is normally less than 90%. A comparison of the resolution of
dispersion image generated from seismic record collected with the father source offset of
30 ft. (9.14m) at both Figures 5.18c and 5.18d is much better than in Figures 5.18a and
5.18B, and the double-ended arrow is 10% shorter than on the data acquired with the
nearest source offset of 10ft. (3.04m) as it shown in Figure 5.18B and 5.18D.
Analytically, it can be shown that the width of a dispersion curve depends on the
difference between the true phase velocity and the assumed phase velocity. If the
integrative energy rapidly decreases with the difference in velocity, the bandwidth would
be relatively narrow, giving a relatively well-defined velocity range that can be regarded
128
as the phase velocity. The relationship between the energy bandwidth and the phase range
of the harmonic waves facilitates an understanding of how the bandwidth varies with
phase velocity, frequency and source distance.
Figure 5.18. (A) The multichannel seismic record; (B) Corresponding overtone image in
which the fundamental mode dispersion is identified; (C) The multichannel seismic
record collected with doubled receiver spacing; (D) Corresponding overtone image in
which the fundamental mode dispersion is identified with visible distortion.
Figure 5.19 displays 4 dispersion images (overtones) deduced from the surface
wave records collected using various MASW array configurations. Figure 5.19A displays
dispersion curve extracted from fundamental mode of overtone image generated for the
seismic record collected with source offset of 10ft. (3.05m) and a receiver spacing of 2.5
ft. (0.76m), the longest Rayleigh wave wavelength of λmax = 1,663ft/sec/26 Hz = 64ft.,
the minimum phase velocity of CRmin = 827ft/sec and a difference between the
maximum and minimum phase velocities of ΔCR = 1,663ft/sec - 827ft/sec = 836ft/sec.
Therefore, the suggested optimum offset A would be 16ft.

129
Figure 5.19B displays dispersion curve extracted from fundamental mode of
dispersion image generated for the seismic record collected with source offset of 10ft.
(3.05m) and a receiver spacing of 5ft., the longest Rayleigh wavelength of λmax =
1,220ft/sec/31 Hz = 39ft., the minimum phase velocity of CRmin = 850 ft/sec and a
difference between the maximum and minimum phase velocities of ΔCR = 1,220ft/sec -
850ft/sec = 370ft/sec. Therefore, the suggested optimum offset A would be 22.3ft.
High level of Rayleigh waves coherency was noticed during comparison of
Figures 5.19a and 5.19b. It was observed that the data acquired with a configuration of
the first nearest offset of 10ft. (3.04m) and a receiver spacing of 2.5ft. (0.61m) considered
a better quality than that from the second nearest offset of 10ft. and a receiver spacing of
5ft. (1.52m). The continuity is even clearer for data in a rectangular window (especially
for frequencies lower than 30 Hz).
Figure 5.19C displays dispersion curve extracted from fundamental mode of
(9.14m) and a receiver spacing of 2.5ft. (0.61m), the longest Rayleigh wavelength of
λmax = 1,699ft/sec/25 Hz = 68ft. (20.72m), a minimum phase velocity of CRmin =
855ft/sec and a difference between the maximum and minimum phase velocities of ΔCR =
1,699ft/sec - 855ft/sec = 844ft/sec. Therefore, the suggested optimum offset A would be
17.2ft. (5.18m).
Figure 5.19D displays dispersion curve extracted from fundamental mode of
(1.52m), the longest Rayleigh wavelength of λmax = 1,546ft/sec/26 Hz = 59ft. (17.98 m),
a minimum phase velocity of CRmin = 904ft/sec and a difference between the maximum
130
and minimum phase velocities of ΔCR = 1,546ft/sec - 904ft/sec = 642ft/sec. Therefore, the
suggested optimum offset A would be 20.7ft. (6.01m).
After comparing all of the results, high level of Rayleigh waves coherency was
noticed in the data acquired from the nearest offset of 10ft. and a receiver spacing of
2.5ft. (0.61m) and the farthest offset of 30ft. (9.14m) and a receiver spacing of 2.5ft.
(0.61m) than with the farther offset of = 30ft. (9.14m) This is the most reliable method of
dispersion imaging with the lowest number of computational artifacts.
Figure 5.19. Dispersion images (overtones) deduced from the surface wave records
collected using various MASW array configurations (north to south orientation).
131
Similar analysis was performed for W-E oriented arrays (Figure 5.20). It was also
noticed that the overal quality of overtone images obtained by using west to east oriented
arrays are slightly poorer in terms of continuity of dispersion curve. This can be
attributed to the irregularity of the depth to top of rock. As the dominant solutuon
widening joints trend north-south in the study area, qulity of MASW data acquired
perpendicular to the structural trends is porrer due to the significant relief along the
geophone array.
Figure 5.20. Dispersion images (overtones) deduced from the surface wave records
collected using various MASW array configurations (west to east orientation).
132
5.3. COMPARISON OF 1-D SHEAR-WAVE VELOCITY PROFILES
The variable increases in shear-wave velocity with depth were attributed to
fracturing of the limestone bedrock. Generally, high shear-wave velocities are indicative
of limestone bedrock. Fracture zones and highly weathered bedrock are typically
attributed to the shear-wave velocities between 1,200 and 2,500ft/sec. Burlington-Keokuk
limestone in the study area has shear-wave velocities between 2,000 and 2,500 ft/sec, so
observed velocities may also be relevant to another limestone type with similar lithology.
Data collected using four MASW array configurations were analyzed. The first
and second data sets were acquired using the source offsets of 10ft. (3.04m) with a
receiver spacing of 2.5ft. (0.76m) and 5ft. (1.52m), respectively. The third and fourth data
sets were acquired using the source offset of 30 ft. (9.14m) with a receiver spacing of
2.5ft. (0.76m) and 5ft. (1.52m), respectively.
Figures 5.21 and 5.22 display the comparison of 1-D shear-wave velocity profiles,
deduced using the inversion technique from the seismic data sets collected with various
configurations of active MASW array. Table 5.3 lists the average Vs values for the top
4ft. (1.22m), 12.5ft. (3.81m), and 27.5ft. (8.38m) depths for one site.
Vs profiles were generated from seismic records acquired with different source
offsets and receiver spacings that were oriented from west to east. The 1-D shear-wave
velocity profiles obtained with different types of configuration of MASW array
configurations are shown in Figure 5.21. The variation of the Vs velocities at all depths
seems to be fairly consistent for all array configurations. The depth to the top of rock can
be identified at a depth of around 12.5ft. (3.81m) from the data collected with all array
configurations.
133
Table 5.3. Average Vs at ten 1-D shear-wave velocity profiles. No.1.
A comparison of 1-D shear wave velocity profiles generated up to a depth of
32.5ft. (9.9m) shows the following results:
a) All of the profiles start with a thin layer (from 1 to 4ft. [0.3 – 1.22m] thick) and
are characterized by average shear wave velocities between 1,070 and 1,240ft./s)
for the data collected with all array configurations. This high-velocity layer is
most probably observed due to the combination of stiff subsurface materials
(reddish-brown clay with admixed chert fragments, alternating from fine to coarse
grained with occasional stylolites and fossils).
b) There is a low velocity layer observed on depths of 11.5 - 12.5ft. (3.5 – 3.81m)
with corresponding shear wave velocities range between 1,230 and 1,250ft./s
which is confidently determined in the most profiles. This layer is attributed to the
top of rock (represented by Burlington-Keokuk limestone).
c) At a depth of approximately 27.5ft. (8.38m), corresponding 1-D shear wave
velocity values increase gradually until the depth of 32.5ft. (9.9m) from
1370ft/sec to 1729ft/sec in all MASW profiles.

134
d) It can also be observed from the Vs profiles that there is a slight decrease in the
velocities for the data sets collected using the farthest source offset of 30ft.
(9.14m) and longer receiver spacing of 5ft. (1.52m) at a depth of approximately
27.5ft. (8.38m). Decrease in velocities from 1,729ft/sec to 2,066ft/sec were
observed for the data sets collected using the nearest offset of 10ft. (3.04m) for
the data sets collected in west-east direction and approximately 1,370ft/sec to
1,404ft/sec for the data sets collected in north-south direction.
e) This decrease in velocity values may correspond to the presence of either clay or
saprolite formation above the bedrock. Also, the highest velocity found in the
inversion was around 1,200 to 2,500ft/sec which are indicative of bedrock, which
is characterized by calcareous shale with some limestone interbeds. A reasonable
correlation between the MASW data sets collected with opposite array orientation
was observed. Generally, extraction of dispersion curves from the overtone
images generated from seismic records collected in west-east or north-south
direction provided results of similar quality. That can be explained by the
heterogeneity of subsurface.
135
inversion technique from the seismic data sets collected with various configurations of
active MASW oriented west to east array.
inversion technique from the seismic data sets collected with various configurations of
active MASW oriented north to south array.
136
5.4. QUANTITATIVE ANALYSES
As a first step of quantitative analysis, three criterias were chosen, and include
continuity of dispersion curve, maximum depth of investigation, and reliability of the top
of rock estimate. The rating categories are briefly summarized in Table 5.4.
Table 5.4. Encoded categories for three criterias.
The continuity was analyzed on the basis of visual assessment of overtone images
and dispersion curves. The quality of the acquired surface wave records were evaluated in
terms of the resolution of the phase velocity spectrum, i.e., the sharpness of the amplitude
peaks observed at each frequency, the extractable frequency range and the continuity of
the fundamental mode high-amplitude band. The continuity ratings “Good”, “Fair”, and
“Poor” were assigned on the basis of visual assessment of overtone images and
dispersion curves, as shown in Fig. 5.23. To ease quantitative data analysis, each rating
was encoded numerically. Thus, number 30 was assigned to “good”, number 15 was
assigned to “fair”, and number 0 was assigned to “poor”.

137
Figure 5.23 Overtone images obtained from MASW data of (A) good quality, (B) fair
quality, and (C) poor quality.
Regarding to the maximum depth of MASW profiles, number 0 was assigned to
profiles that do not extend to a depth of 20ft., number 5 was assigned to profiles that do
extend to a depth of 30ft., and number 10 was assigned to profiles that extend to a depth
greater than 30ft.
The analysis of reliability of the top of rock estimate was based on the comparison
of ERT- and MASW-determined depth to top of bedrock. The difference between the two
was calculated in percent. Number Zero was assigned to profiles that gave the difference
more than 30%, number 5 was assigned to profiles that gave the difference in the interval
of 20 to 30%, and number of 10 was assigned to profiles that had difference less than
20%.
138
The quantitative data analysis was performed by summing up the ratings of tree
criterias. Thus, a number of 50 is the maximum possible number, and a number of Zero is
the minimum number. The data quality categories were developed based on the numbers
after summation. The categories are briefly summarized in Table 5.5. Using this
approach, all MASW profiles were evaluated in order to assign the data quality ratings.
Table 5.5. Data quality categories.
In an attempt to identify the MASW array configuration that gives the better data
quality, statistical analysis was performed. During the analysis, mean values were
calculated for each group of MASW array configuration. Table 5.6 summarizes statistics
for west to east direction arrays, and Table 5.7 summarizes statistics for north to south
direction arrays. As seen in Table 5.6, the highest mean value (30.73) was observed for
10ft. source offset and 2.5ft.geophone spacing array. This implies that this configuration
gave the better quality of MASW data. The lowest mean value (21.91) was observed for
30 ft. source offset and 5ft. receiver spacing array. It is interesting to note that according
to the statistical analysis, the shortest geophone spread (10 x 2.5ft.) gave the better data
quality, whereas the longer geophone spread (30 x 5ft.) gave the poorer quality. This
139
supports the previous statement about the effect of irregularity of the depth to top of rock
along short and long geophone arrays. According to Table 5.7, mean values calculated for
configurations direction north to south are slightly higher than those calculated for
configurations direction west to east (Table 5.6).
Table 5.6. Quantitative analysis performed for 136 MASW profiles acquired in west to
east direction.
Table 5.7. Quantitative analysis performed for 136 MASW profiles acquired in north to
south direction.
To analyze the frequency of occurrence of MASW profiles with good, fair, poor,
and severe quality, Figures 5.24 and 5.25 were generated. Figure 5.24 shows distribution
of MASW data acquired by the array oriented in west to east direction and Figure 5.25
shows MASW data acquired by the array oriented in north to south direction. According
140
to Figure 5.24, the largest number of good quality MASW profiles was acquired using
10ft. source offset and 2.5ft. receiver spacing. In contrast, the smallest number of good
quality MASW profiles was acquired by the configuration with 30 ft. source offset and
5ft. receiver spacing.
Figure 5.24 Histogram showing how often MASW data of good, fair, poor, and severe
quality were acquired using a specific array configuration (west to east direction).
Regarding to the north to south oriented arrays, the largest number of MASW
profiles with good data quality was acquired by the 10ft. source offset and 2.5ft. receiver
spacing array. It is interesting to note that no MASW profiles with good data quality were
acquired using the longest (30 x 5ft.) geophone spread. In contrast, this configuration (30
x 5ft.) gave the largest number of MASW profiles with severe data quality.
141
Figure 5.25 Histogram showing how often MASW data of good, fair, poor, and severe
quality were acquired using a specific array configuration (north to south direction ).
142
6. CONCLUSIONS
The main goal of this research was to develop optimum MASW configuration that
can be used in karst terrain to image subsurface to a depth of 30ft. (9m). To accomplish
the goal, MASW data were acquired with different parameters and in different directions.
Qualitative and quantitative analyses were performed in this study to assess the
performance of each configuration.
The following conclusions can be made on the basis of qualitative analysis:
 Continuity of dispersion curve must be the primary criteria in evaluating MASW
data quality. The dispersion is more continuous over a broad range of frequency
when MASW data are acquired parallel to the solution widened joints.
 As the solution widened joints trend north to south, MASW data acquired by the
north to south oriented arrays are of a better quality.
The following conclusions can be made on the basis of quantitative analysis:
 In addition to the continuity of dispersion curve, two criterias – maximum depth
of investigation and reliability to top of rock estimate – can be used to evaluate
overall quality of MASW data. Using three criterias, it was determined that
MASW data acquired by 10ft. source offset and 2.5ft. receiver spacing array
configurations are of a better quality.

143
7. RECOMMENDATIONS
This research has enhanced the understanding of acquisition of MASW data in
karst terrain. However, further studies are still required to improve acquisition,
processing and interpretation of seismic data in karst terrain. The following studies are
recommended:
 Choice of the method should be guided by the anticipated depth and size of
target(s), the nature of the background materials or the maximum depth of rock
surrounding the target(s), the reason for delineating the target(s), the desired
resolution of the target(s), and the size of the investigation area and the sources of
cultural interference in the investigation area.
 The values that are determined for optimum data acquisition and processing
should be taken as guidelines for all other parameters (i.e., it is important to
choose processing parameters that are appropriate for the depth target of the
investigation).
 For impact sources, such as the drop weight or vibrator should be used to enhance
the quality of the active MASW data in complex karst terrain.
 A quantitative evaluation of the resolution of surface wave methods should be
conducted to clearly delineate the examination capabilities not only for
researchers, but as well to associate engineers avoid error application of surface
wave methods.
 In addition, calculation efforts should be made to improve the efficiency of
intensive inversion algorithms using parallel programming and computing on
GPS or GIS.
144
BIBLIOGRAPHY
Abbiss, C.P. (1981). Shear wave measurements of the elasticity of the ground.
Geotechnique, 31, (1), pp91-104.
Advanced Geosciences, Incorporated. "Instruction Manual for the Super Sting with Swift
Automatic Resistivity and IP System." Austin, January 2006.
Aki, K., & Richards, P. G. (1980). Quantitative seismology: Theory and methods (Vol.
1). San Francisco, CA: W. H. Freeman and Co.
Anderson, N. (2010). Lectures in: Ground Penetrating Radar Applications in Engineering

and Environmental Studies. GE482. Rolla, Missouri, USA: Missouri University of
Science and Technology.
Anderson, N., 2012, Determination of Optimum “Multi-Channel Surface Wave Method”

Field Parameters: NUTC R292.
Auld, B. A., Acoustic Fields and Waves in Solids (Stanford University Press, 1973).
EWING, W.M., JARDETZKY, W. S., and PRESS, F., Elastic Waves in Layered
Media (McGraw-Hill, London, 1957), 380 pp.
Bacon, M., Simm, R and Redshaw, T. (2003): 3-D Seismic Interpretation; Cambridge
University press.
Baechle, G.T., G.P. Eberli, R.J. Weger, and J.L. Massafero, 2009, Changes in dynamic
shear moduli of carbonate rocks with fluid substitution: Geophysics, 74, E135-
E147.
Ballard, R.F. & McLean, F.G. 1975. Seismic field methods for in situ moduli. Proc. Conf.
on In Situ Measurement of Soil Properties. Spec. Conf. Geotech. Eng. Div.,
A.S.C.E., Vol 1, pp121-150.
Beaty, K. S., Schmitt, D. R., & Sacchi, M. (2002). Simulated annealing inversion of
multimode Rayleigh wave dispersion curves for geological structure. Geophysical
Journal International, 151(2), 622-631.
Boore, DM (2004). Estimating VS (30) (or NEHRP Site Classes) from shallow
velocity models (depths < 30m), Bull. Seismo. Soc. Am., 94(2):591–597.
http://parkseismic.com/SSC-Main.html.
Bolt, B. A. (1976). Nuclear Explosions and Earthquakes: The Parted Vail. San Francisco,
CA: W. H. Freeman and Co.
145
British Geological Survey (c) NERC 2013.
Buettner, M., Ramirez, A. and Daily, W. “Electrical Resistance Tomography for Imaging
the Spatial Distribution of Moisture in Pavement Sections” Structural Materials
Technology: An NDT Conference, San Diego, Feb. 20-23 1996 [abstract], 1996,
342-347.
Bullen, K.E., 1963, an Introduction to the Theory of Seismology: Cambridge University

Press, 381 p.
Cai, Y., Sangghaleh, A., & Pan, E. (April 01, 2015). Effect of anisotropic base/interlayer
on the mechanistic responses of layered pavements. Computers and Geotechnics,
65, 2, 250-257.
Chambers, J. E., Wilkinson, P. B., Penn, S., Meldrum, P. I., Kuras, O., Loke, M. H., &
Gunn, D. A. (June 01, 2013). River terrace sand and gravel deposit reserve
estimation using three-dimensional electrical resistivity tomography for bedrock
surface detection. Journal of Applied Geophysics, 93, 1, 25-32.
Choon B. Park, Richard D. Miller, and Hidetoshi Miura (2002). Optimum Field
Parameters of an MASW Survey.
Choon B. Park, Richard D. Miller, Jianghai Xia, and Julian Ivanov (2007) “Multichannel
Analysis of Surface Waves (MASW)-Active and Passive Methods”. The Leading
Edge.
Coskun, N. “The effectiveness of electrical resistivity imaging in sinkhole

investigations,” International Journal of Physical Sciences, 7(15), 2398 - 2405.
Coots, T. “Greene County Comprehensive Plan: Greene County Planning and Zoning,”
http://www.greenecountymo.org/web/Public_Information/files/Comp.pdf, 2007.
Choon B. Park, Richard D. Miller, and Hidetoshi Miura (2002) “Optimum Field
Parameters of an MASW Survey”.
Cornou, C., M. Ohrnberger, D. M. Boore, K. Kudo, and P.-Y. Bard, 2006, Derivation of
structural models from ambient vibration array recordings: Results from an
international blind test: Proceedings of the 3rd International Symposium on the
Effects of Surface Geology on Seismic Motion, paper NBT.
Daniels F. and Alberty R.A., 1966. Physical Chemistry. John Wiley and Sons, Inc. de
Groot-Hedlin, C. and Constable, S., 1990. Occam's inversion to generate smooth,
two dimensional models form magneto telluric data. Geophysics, 55, 1613-1624.
146
Dal Moro, G., Pipan, M., Forte, E., & Finetti, I. (2003). Determination of
Rayleigh wave dispersion curves for near surface applications in unconsolidated
sediments. In SEG International Exposition and Seventy-Third Annual
Meeting, 24-31 October 2003, Dallas, Texas (Vol. 22, pp. 1247–1250).
Dey A. and Morrison H.F. Resistivity modeling for arbitrary shaped two-dimensional
structures. Geophysical Prospecting 27, 1979, 106-136.
Denizman, C., 2003, Morphometric and spatial distribution parameters of karstic

depressions, lower Suwannee River basin, Florida: Journal of Cave and Karst
Studies, 65, 29-35.
Donohue, S., Forristal, D., & Donohue, L. A. (2013). Detection of soil compaction using
seismic surface waves. Soil and Tillage Research, 128, 54–60.
Doolittle, J.A., and M.E. Collins, 1998, Comparison of EM induction and GPR methods
in areas of karst: Geoderma, 85, 83-102.
Duffy, B.G., 2008, Development of multichannel analysis of surface waves (MASW) for
characterizing the internal structure of active fault zones as a predictive method of
identifying to distribution of ground deformation, M.Sc. Thesis, University of
Canterbury.
Elkrry, A., Nwokebuihe, S., Torgashov, E., Dera, A., Alotaibi, A., Anderson, N. (2015).
Site Specific Pavement Condition Assessment. EEGS SAGEEP 28 (Environmental
and Engineering Geophysical Society, the Symposium on the Application of
Geophysics to Engineering and Environmental Problems). Austin, TX. March,
2015.
Everett, M. E. (2013). Near-Surface Applied Geophysics. Cambridge: Cambridge

University Press.
Fellow, L. D., “Geology of Galloway Quadrangle Greene County Missouri,” Missouri

Geological Survey and Water Resources, 1970.
Forbriger, T., 2003a, Inversion of shallow-seismic wavefields: I. Wavefield

transformation: Geophys. J. Int., 153, 720-734.
Foti, S. (2000). Multistation methods for geotechnical characterisation using surface

waves. Ph.D. Thesis, Politecnico di Torino. 229p.
Gazetas, G. 1982. Vibrational characteristics of soil deposits with variable velocity. Int.
Jour. Numerical and Analytical Methods in Geomech., Vol 6, pp 1-20.
Gibson, Paul J, and Dorothy M George. Environmental Applications of Geophysical

Surveying Techniques. Hauppauge: Nova Science Publishers, 2003.
147
Gucunski, N., Woods, R.D., 1991. Instrumentation for SASW testing. In: Bhatia, S.K.,
Blaney, G.W. (Eds.), Recent Advances in Instrumentation, Data Acquisition and
Testing in Soil Dynamics, Geotechnical Special Publication, vol. 29. American
Society of Civil Engineers, pp. 1– 16.
Heisey, J. S., Stokoe II, K. H., Hudson, W. R., and Meyer, A. H. (1982), Determination
of in situ Shearwave Velocity from Spectral Analysis of Surface Waves, Research
Report No. 256-2, Center for Transportation Research. The University of Texas at
Austin, December, 277 pp.
Ivanov, J., Park, C. B., et al. (2001). Modal separation before dispersion curve extraction
by MASW method. SAGEEP Denver, Colorado.
Ivanov, J., Park, C. B., et al. (2005). "Analyzing and filtering surface wave energy by
muting shot gathers." Journal of Environmental and Engineering Geophysics 10:
307-322.
Ivanov, J., R.D. Miller, P. Lacombe, C.D. Johnson, and J.W. Jr. Lane, 2006, Delineating
a shallow fault zone and dipping bedrock strata using multichannel analysis of
surface waves with a land streamer, Geophysics, 71, 5, p. A39-A42. doi:
10.1190/1.2227521.
Ivanov, Julian, Choon Park, and Jianghai Xia. "MASW/Surfseis 2 Workshop." Fort
Worth: Kansas Geologic Survey, March 28, 2009.
Ivanov, J., R.D. Miller, S. Peterie, C. Zeng., J. Xia, and T. Schwenk, 2011, Multi-channel
analysis of surface MASW (MASW) of models with high shear-wave velocity
contrast, 2011 SEG Annual Meeting, September 18 - 23, 2011 , San Antonio,
Texas.
Jones, R.B. 1958 In-situ measurement of the dynamic properties of soil by vibration
methods. Geotechnique, vol. 8 (1), pp. 1-21.
Kearey, Philip, Micheal Brooks, and Ian Hill. An Introduction to Geophysical

Exploration. Williston: Blackwell Publishing, 2002.
Kansas Geological Survey “Multichannel Analysis of Surface Waves”, 2013

http:/www.masw.com/index.html
Keller G.V. and Frischknecht F.C., 1966. Electrical methods in geophysical prospecting.
Pergamon Press Inc., Oxford.
Kramer, S. L. (1996). Geotechnical Earthquake Engineering. Upper Saddle River, NJ:

Prentice-Hall.
148
Lin, C.-P., Chang, C.-C., & Chang, T.-S. (2004), the use of MASW method in the
assessment of soil liquefaction potential. Soil Dynamics and Earthquake
Engineering, 24 (9–10), 689–698.
Loke, M.H., “Electrical imaging surveys for environmental and engineering studies,” in
http://www.abem.se/support/downloads/case-studies/practical-guide-to-2d-
3dsurveys. 1999.
Loke, M.H. "Electrical Imaging Survey for Environmental and Engineering Studies."
2000.
Loke, M.H., 2011. Electrical resistivity surveys and data interpretation, in Gupta, H (ed.),
Solid Earth Geophysics Encyclopaedia (2nd Edition) “Electrical &
Electromagnetic” Springer-Verlag, 276-283.
Macfarlane, P.A., Healey, J.M., and Wilson, B.W., 2005 THE SOUTHEAST KANSAS
OZARK AQUIFER WATER SUPPLY PROGRAM PHASE 1 PROJECT
RESULTS, Kansas Geological Survey Open File Report 2005-15 15p.
(http://www.kgs.ku.edu/Hydro/Publications/2005/OFR05_15/aquifer.pdf. (Date
accessed: May 13, 2008).
Miller, R.D., Xia, J., and Park, C.B., 1999, Estimation of near-surface shear-wave
velocity by inversion of Rayleigh waves: Geophysics, v. 64, no. 3, p. 691-700.
(XIA-99-04)
Miller, R., Xia, J., Park, C., Davis, J., Shefchik, W., and Moore, L., 1999a, Seismic
techniques to delineate dissolution features in the upper 1000 ft at a power plant;
Technical Program with biographies, SEG, 69th Annual Meeting, Houston, Texas,
492-495.
Miller, R., Xia, J., Park, C., Ivanov, J., and Williams, E., 1999b, Using MASW to map
bedrock in Olathe, Kansas; Technical Program with biographies, SEG, 69th
Annual Meeting, Houston, Texas, 874-877.
Miller, R.D., J. Xia, and C.B. Park, 2005, Seismic techniques to detect dissolution
features (karst) at a proposed power-plant site, in, D.K. Butler, ed., near surface
geophysics: Society of Exploration Geophysicists, 663-679.
Milson, John. Field Geophysics. New York: John Wiley & Sons, 1996.
Missouri Department of Conservation 2005. Wildlife Code of Missouri, Rules of the

Conservation Commission. Issued March 1, 2005. Missouri Department of
Conservation, Jefferson City, MO.
Monroe, Watson H. A Glossary Karst Terminology. Geological Survey Water Paper

1899-K, 1970.
149
Muchaidze, I., “Imaging Karst Terrain Using Electrical Resistivity Tomography.” In

https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/26344/Muchaidze_
2008. Pdf? Sequence =1. 2008.
Muchaidze, I., 2009, Imaging in Karst Terrain Using Electrical Resistivity Tomography.
Master’s Thesis, Missouri Science and Technology University, 2009.
Myat, M., Wamweya, A., Kovin, O., Anderson, N., and Robison, J., 2008a, Application
of Electrical Resistivity Method in Steeply Dipping Karst Terrain: 11th
Multidisciplinary Conference on Sinkholes and the Engineering and
Environmental Impacts of Karst: Integrating Science and Engineering to Solve
Karst Problems, September 22-26, 2008, Ramada Conference Center Tallahassee,
Tallahassee, FL. 224.
Myat, M., Muchaidze, Y., Wamweya, A., Anderson, N., and Robison, J., 2008b,
Assessment of Karst Activity at Springfield Route 60 Study Site: 11th
Multidisciplinary Conference on Sinkholes and the Engineering and
Environmental Impacts of Karst: Integrating Science and Engineering to Solve
Nakano, M., Fuji, K., and Takeuchi, S. (1988), Rayleigh wave scatteringat wedge corners
with major wedge angles. In Scattering and Attenuation of Seismic Waves, Part I
(eds. Aki, K. and Wu R. S.), Pure Appl. Geophys. 128, 119–132.
Neidell, N.S., and Taner, M.T., 1971, Semblance and other coherency measures for
multichannel data: Geophysics, 36, 482–497.
Odum, J.K., R.A. Williams, W.J. Stephenson, D.M. Worley, C. von Hillebrandt-
Andrade, E. Asencio, H. Irizarry, and A. Cameron, 2007, Near-surface shear wave
velocity versus depth profiles, Vs 30, and NEHRP classifications for 27 sites in
Puerto Rico: USGS Open-File Report 2007-1174.
Okada, H. 2003. The microtremor survey method. Geophysical Monograph Series. No.
12. Society of Exploration Geophysicists, Tulsa, Oklahoma, USA.
O’Neill, A., 2003, Full-waveform reflectivity for modeling, inversion and appraisal of
seismic surface wave dispersion in shallow site investigations, Ph.D. thesis, The
University of Western Australia, Australia.
Park, C.B., Miller, R.D., and Xia, J., 1997, Multi-channel analysis of surface waves
(MASW) “A summary report of technical aspects, experimental results, and
perspective”: Kansas Geological Survey Open-file Report 97-10.
Park, C. B., R. D. Miller, and J. Xia, 1998, imaging dispersion curves of surface waves
on multi-channel record: 68th Annual International Meeting, SEG, Expanded
Abstracts, 1377–1380.
150
Park, C.B., XIA, J. and Miller, R.D. (1998b) Imaging dispersion curves of surface waves
on multichannel record: 68th Ann.
Park, C.B., R.D. Miller, and J. Xia, 1999, Multichannel Analysis of Surface Waves:
Geophysics, 64, 800-808.
Park, C.B., Miller, R.D., and Xia, J., 1999a, Multi-channel analysis of surface waves:
Geophysics, 64(3), 800-808.
Park, C.B., Miller, R.D., and Xia, J., 1999b, Detection of near-surface voids using surface
wave: Proceedings of the Symposium on the Application of Geophysics to
Engineering and Environmental Problems (SAGEEP 99), Oakland, CA, March 14-
8, 281-286.
Park, C.B., R.D. Miller, and J. Xia, 2000, Detection of higher mode surface waves over
unconsolidated sediments by the MASW method: Proceedings of the Symposium
on the Application of Geophysics to Engineering and Environmental Problems
(SAGEEP 2000), Arlington, Va., February 20-24, p. 1-9. (PAR-00-01).
Park, Choon B, Richard D Miller, Jianghai Xia, and Julian Ivanov. "Multichannel
Seismic Surface-Wave Methods for Geotechnical Applications." First International
Conference on the Application of Geophysical Methodologies to Transportation
Facilities and Infrastructure. St. Louis, 2000.
Park, C. B., Miller, R. D., & Xia, J. (2001). Offset and resolution of dispersion curve in
multichannel analysis of surface waves (MASW). In Proceedings of the
Symposium on the Application of Geophysics to Engineering and Environmental
Problems 2001 (p. SSM4). Denver, CO: Environment and Engineering
Geophysical Society.
Park, C.B., Miller, R.D., and Miura, H., 2002, Optimum field parameters of an MASW
survey [Exp. Abs.]: SEG-J, Tokyo, May 22-23, 2002. (PAR-02-03).
Park, Choon B. MASW - Horizontal Resolution in 2D Shear-Velocity (Vs) Mapping.

Open-File Report, Lawrence: Kansas Geologic Survey, 2005.
Park, C. B., Miller, R. D., Ryden, N., Xia, J., and Ivanov, J., 2005, Combined use of
active and passive surface waves: Journal of Engineering and Environmental
Geophysics (JEEG), 10, (3), 323-334.
Park, C. B., Miller, R. D., Xia, J., & Ivanov, J. (2007). Multichannel analysis of surface
waves (MASW) active and passive methods. The Leading Edge, 26 (1), 60–64.
151
Park, C. B., & Shawver, J. B. (2009). MASW Survey Using Multiple Sources Offsets. In
D. K. Butler (Ed.), Proceedings of the Symposium on the Application of
Geophysics to Engineering and Environmental Problems 2009 (pp. 15–19).
Denver, CO: Environment and Engineering Geophysical Society.
Park, C. B., & Carnevale, M. (2010). Optimum MASW survey—Revisit after a Decade
of Use. In D. O. Fratta, A. J. Puppala, & B. Muhunthan (Eds.), GeoFlorida 2010:
Advances in Analysis, Modeling & Design (pp. 1303–1312). Reston, VA:
American Society of Civil Engineers.
Park, C., 2013, MASW for geotechnical site investigation: The Leading Edge, June 2013,
v. 32, p. 656-662.
Park, C. B. (2015). Data acquisition. Retrieved 23 June, 2015, from

http://www.masw.com/DataAcquisition.html.
Press, F., 1966, Seismic velocities: in Handbook of physical constants, revised edition,
Clark, S.P., Jr. (ed.), Geological Society of America Memoir, 97, 97–173.
Richart, F. E., Hall, J. R., and Woods, R. D., 1970, Vibrations of soils and foundations:
Prentice-Hall, Inc.
Rix, G.J., and Leipski, A.E., 1991, Accuracy and resolution of surface wave inversion:
Recent Advances in Instrumentation, Data Acquisition and Testing in Soil
Dynamics: Geotechnical Special Publication No. 29, ASCE, 17-23.
Robinson, E.S. & Çoruh, C. (1988) Basic Exploration Geophysics. Wiley, New York.
Robison, J. and Anderson, N., Geophysical Investigation of the Delaware Avenue

Sinkhole, Nixa, Missouri: 11th Multidisciplinary Conference on Sinkholes and the
Engineering and Environmental Impacts of Karst: Integrating Science and
Engineering to Solve Karst Problems, September 22-26, 2008, Ramada
Conference Center Tallahassee, Tallahassee, FL, 2008.
Rosenblad, B. L. and Cheng‐ Hsuan, L. (2011). Influence of Poisson's Ratio on Surface

Wave Near‐ Field Effects. Geo‐ Frontiers 2011: Advances in Geotechnical
Engineering, ASCE.
Ryden, N. and Mooney, M. (2009). Analysis of surface waves from the light weight
deflectometer: Soil Dynamics and Earthquake Engineering, 29: 1134-1142.
Rene P, Maros N, Ivan D, David K (2012) “Determination of Cavities Using Electrical

Resistivity Tomography”. Contributions to Geophysics and Geodesy, 42:201-211.
Socco, L.V., and Strobia, C., 2004, Surface-wave method for near-surface
characterization: a tutorial: Near Surface Geophysics, 2, 165-185.
152
Socco, L. V., Foti, S., & Boiero, D. (2010). Surface-wave analysis for building near-
surface velocity models - Established approaches and new perspectives.
Geophysics, 75 (5), 75A83–75A102.
Stokoe II KH, Wright GW, James AB, Jose MR. Characterization of geotechnical sites
by SASW method. In: Woods RD, Ed. Geophysical characterization of sites, A. A.
Balkema/Rotterdam, 1994. p. 15-25.
Shade, K.A., 2003, Temporal analysis of floodplain deposition using urban pollution 153
stratigraphy, Wilson Creek, SW Missouri. Masters of Science Thesis, Department
of Geography, Geology, and Planning, Southwest Missouri State University,
Springfield, Missouri.
Silvester, P.P. and Ferrari R.L. Finite elements for electrical engmeers. 2nd. ed.
Cambridge University Press, 1990.
Sudha, K., M., Mittal, S., & Rai, J. (January 01, 2009). Soil characterization using
electrical resistivity tomography and geotechnical investigations. Journal of
Applied Geophysics, 67, 1, 74-79.
Sheriff, R. E., & Geldart, L. P. (1995): Exploration Seismology, Cambridge University

Press.
Shishay T. Kidanu, Evgeniy V. Torgashov, Aleksandra V. Varnavina, Neil L. Anderson

(May 23, 2016). ERT-based Investigation of a Sinkhole in Greene County,
Missouri. Journal of Geosciences, 2 (2): 99-115.
Schrott, L. and O. Sass, 2008, Application of field geophysics in geomorphology:

Advances and limitations exemplified by case studies: Geomorphology, 93, 55-73.
Shuang X, Lung, and Jianghal (2004). The Selection of Field Acquisition Parameters for
Dispersion Images from Multichannel Surface Wave Data: Pure and Applied
Geophysics, 161 (2004) 185–201.
Steeples, Don. "Near-Surface Seismology: A Short Course." Lawrence: Society of

Exploration Geophysicist, 1998.
Steeples, D. W., Baker, G. S., Schmeissner, C., 1999. Toward the Autojuggie: Planting
72 Geophones in 2 Sec. Geophysical Research Letters, 26(8): 1085–1088.
Surf-Seis. Multi-Channel Analysis of Surface Waves. 2006. Retrieved February 3, 2010

from: http://www.kgs.ku.edu/software/surfseis/masw.html. Steeples, Don. "Near-
Surface Seismology: A Short Course." Lawrence: Society of Exploration
Geophysicist, 1998.
153
Telford, W.M., Geldart, L.P., and Sheriff, R.E., 1990, Applied Geophysics, second
edition: Cambridge, Cambridge University Press, 770 p.
Tian, G. (April 01, 2003). Multichannel analysis of surface wave method with the
autojuggie. Soil Dynamics and Earthquake Engineering, 23, 3, 61-65.
Torgashov, E., Anderson, N., Li, M., Ismail, A., & Elkrry, A. (2012). Imaging in Karst
Terrain Using Electrical Resistivity Tomography and Subsurface Wave Method;
EEGS SAGEEP 25 (Environmental and Engineering Geophysical Society, the
Symposium on the Application of Geophysics to Engineering and Environmental
Problems. Tucson, AZ, 2012.
Tseng, Y. H., Kang, S. C., Chang, J. R., & Lee, C. H. (December 01, 2011). Strategies for
autonomous robots to inspect pavement distresses. Automation in Construction,
20, 8, 1156-1172.
Thierry, P., N. Debeblia, and A. Bitri, 2005, Geophysical and Geological characterization
of karst hazards in urban environments: application to Orleans (France): Bulletin
of Engineering Geology and the Environment, 64, 139-150.
Unklesbay, A.G. and Jerry D. Vineyard. Missouri Geology: Three Billion Years of
Volcanoes, Seas, Sediments, and Erosion. University of Missouri Press Columbia
and London, 1992.
United States Corps of Engineers. "EM 1110-1-1802: Geophysical Exploration for

Engineering and Environmental Investigations." Engineering Manual, Washington
D.C., 1995.
United States Corps of Engineers. "Geotechnical Investigations." Engineering Manual

No. 1110-1-1804, Washington D.C., 2001.
Vandike, James E., 1993, Groundwater Level Data for Missouri, Water Year 1991-
1992: Missouri Division of Geology and Land Survey, Water Resources Report No. 42,
96 p.
Vandike, James. Karst in Missouri, an Overview. Part 1. Missouri Department of Natural

Resources, 1997.
Veni, G., DuChene, H., Crawford, N.C., Groves, C.G., Huppert, G.H., Kastning, E.H.,
Olson, R. & Wheeler, B.J., 2001, Living with karst: a fragile foundation.
Environmental awareness series, American Geological Institute, 64 pp. + 1 pl.
Veni, George. Revising the karst map of the United States. Journal of Cave and Karst
Studies 64(1), 45-50, 2001.
154
Wadati, K. (1933). On the Travel Time of Earthquake Waves, Part II. Geophysical
Magazine, 7, 101–111.
What is Karst? http://water.usgs.gov/ogw/karst/pages/whatiskarst, 2012.
Xia, J., R.D. Miller, and C.B. Park, 1998, Construction of vertical seismic section of
near-surface shear wave velocity from ground roll [Exp. Abs.]: Soc. Explor.
Geophys./AEGE/CPS, Beijing, 29-33.
Xia, J., Miller, R.D., and Park, C.B., 1999, Estimation of near-surface shear-wave
velocity by inversion of Rayleigh wave: Geophysics, 64, 691-700.
Xia, J., Miller, R.D., and Park, C.B., 2000a, Advantage of calculating shear-wave
velocity from surface waves with higher modes: Technical Program with
Biographies, SEG, 70th Annual Meeting, Calgary, Canada, 1295-1298.
Xia, J., Miller, R.D., Park, C.B., Hunter, J.A., and Harris, J.B., 2000b, Comparing shear-
wave velocity profiles from MASW technique with borehole measurements in
unconsolidated sediments of the Fraser River Delta: Journal of Environmental and
Engineering Geophysics, v. 5, no. 3, 1-13.
Xia, J., Miller, R. D., Park, C. B., Hunter, J. A., Harris, J. B., & Ivanov, J. (2002).
Comparing shear-wave velocity profiles inverted from multichannel surface wave
with borehole measurements. Soil Dynamics and Earthquake Engineering, 22 (3),
181–190.
Xia, J., Miller, R. D., Park, C. B., & Tian, G. (2003). Inversion of high frequency surface
waves with fundamental and higher modes. Journal of Applied Geophysics, 52 (1),
45–57.
Xia, J., Chen, C., Li, P.H., and Lewis, M.J., 2004a, Delineation of a collapse feature in a
noisy environment using a multichannel surface wave technique: Geotechnique,
54(1), 17-27.
Xia, J. (2014). Estimation of near-surface shear-wave velocities and quality factors using
multichannel analysis of surface-wave methods. Journal of Applied Geophysics,
103, 140–151.
Xu, Y., J. Xia, and R. D. Miller, 2006, Quantitative estimation of minimum offset for
multichannel surface-wave survey with actively exciting source: Journal of
Applied Geophysics, 59, 117–125.
Yuan, D., & Nazarian, S. (1993). Automated Surface Wave Method: Inversion
Technique. Journal of Geotechnical Engineering, 119 (7), 1112–1126.
155
Yilmaz, Ö. (1987). Seismic Data Processing. Tulsa, OK: Society of Exploration

Geophysicists.
Yoon, S. and Rix, G.J. (2009). Near-field effects on array-based surface wave methods
with active sources: J. Geotech. Geoenviron. Eng., 135(3), 399-406.
Yixian Xu a, Jianghai Xia b, Richard D. Miller (2006). Quantitative estimation of

minimum offset for multichannel surface-wave survey with actively exciting
source. Journal of Applied Geophysics 59 (2006) 117–125.
Zeng, C., Xia, J., Miller, R. D., Tsoflias, G. P., & Wang, Z. (2012). Numerical
investigation of MASW applications in presence of surface topography. Journal of
Applied Geophysics, 84, 52–60.
Zhang, S. X., & Chan, L. S. (2003). Possible effects of misidentified mode number on
Rayleigh wave inversion. Journal of Applied Geophysics, 53 (1), 17–29.
Zhang, S.X., Chan, L.S., and Xia, J. (2004). The selection of field acquisition parameters
for dispersion images from multichannel surface wave data: Pure and Applied
Geophysics, 161: 1–17.
Zhou W, Beck BF, Stephenson JB (1999), “Investigation of ground water flow in karst
areas using component separation natural potential measurements”. Journal of
Environmental Geology, 37: 19-25.
156
VITA
Ghassan Alsulaimani was born in 1976, in Taif city, Saudi Arabia. Mr.
Alsulaimani began his collegiate studies in 1994 and received a Bachelor of Science in
Geological Engineering in 1998 from King Abdul-Aziz University (KAU) in Saudi
Arabia. He has worked for the Saudi Geological Survey since 1999 as an Engineer
Geologist. Meanwhile, he was involved in some projects related in the exploration of
industrial minerals and rocks. He also worked in the geohazard department for more than
four years on a project involving the stability analysis of rock slopes and rock falls.
In May 2014 he received his Master of Science from the Geological Engineering
from Missouri University of Science and Technology. Mr. Alsulaimani started the Ph.D.
program in Geological Engineering with an emphasis in Geophysics at Missouri
University of Science and Technology during the fall semester of 2014.
During his Ph.D. program, Mr. Alsulaimani worked as a research assistant. He
has been involved in several subsurface shallow geophysical projects. The geophysical
methods he has used include resistivity, seismic, electromagnetic, GPR, and metal
detector. Ghassan is interested in shallow-applied geophysics. He has presented at a
number of conferences as a primary author and as a co-author, and he has published two
scientific papers. He has also been involved in professional societies such as SEG, GSA,
AEG, EEGS and SCE. Mr. Alsulaimani has held several elected positions in student
organizations at Missouri University of Science and Technology, including vice president
of the Saudi Student Association (SSA) in 2013-2016, and ICRM in 2014.
In May 2017 Mr. Alsulaimani received his Ph.D. in Geological Engineering from
Missouri University of Science and Technology.

Acquisition of Active Multichannel Analysis of Surface Waves (MAS

Uploaded by

Copyright:

Available Formats

Acquisition of Active Multichannel Analysis of Surface Waves (MAS

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Acquisition of Active Multichannel Analysis of Surface Waves (MAS

Uploaded by

Copyright:

Available Formats

Scholars' Mine

Doctoral Dissertations Student Theses and Dissertations

Acquisition of active multichannel analysis of surface waves

Follow this and additional works at: https://scholarsmine.mst.edu/doctoral_dissertations

GHASSAN SALEM ALSULAIMANI

Presented to the Faculty of the Graduate School of the

MISSOURI UNIVERSITY OF SCIENCE AND TECHNOLOGY

In Partial Fulfilment of the Requirements for the Degree

Neil L. Anderson, Advisor

Ghassan Salem Alsulaimani

All Rights Reserved

tomography (ERT) data obtained for the same study site.

The comparative analysis indicated the profile configuration measurements have

I sincerely extend my gratefulness to my advisor, Dr. Neil L. Anderson, for the

I have availed with a well-grounded experience with the consistency of my research

objectives. He further provided me with encouragement not only to broaden my

experimentalist approach, but also to harness my independent thought. The

overwhelming contributions of my committee members, including Dr. J. David Rogers,

thanked for their generous support.

Zohair Nawab, for his support as well.

I would like to extend my gratitude and eternal thanks to my father Salem, my

ambitions with their love and encouragement. I am also exceedingly grateful to my

ABSTRACT ....................................................................................................................... iii

LIST OF ILLUSTRATIONS ............................................................................................. ix

LIST OF TABLES ........................................................................................................... xvi

1.1. OVERVIEW .........................................................................................1

1.2. RESEARCH OBJECTIVES .................................................................1

1.3. STRUCTURE OF THE DISSERTATION ...........................................2

1.4. LITERATURE REVIEW .....................................................................3

1.5. SIGNIFICANCE OF THE RESEARCH ..............................................5

2. GEOLOGY OF SOUTHWEST MISSOURI ...................................................... 7

2.1. INTRODUCTION ................................................................................7

2.2. OVERVIEW OF GEOLOGICAL AND STRATIGRAPHIC

2.2.1. Lower Mississippian of Lower Osagean............................. 11

2.2.2. Lower Mississippian of Uperer Osagean . .......................... 11

2.2.3. Overview of Structure and Faulting Based on the

2.3. FORMATION OF KARST FEATURES ...........................................14

2.3.1. Groundwater Recharge. ...................................................... 16

2.3.2. Sinkhole or Sink.................................................................. 16

2.3.3. Soluble Bedrock. ................................................................. 16

2.3.4. Natural Bridge or Tunnel. ................................................... 17

2.3.5. Losing Stream. .................................................................... 17

2.3.6. Cave. ................................................................................... 17

2.3.7. Spring. ................................................................................. 18

2.4. OVERVIEW OF SINKHOLE FORMATION PROCESS .................18

2.4.1. Collapse Sinkholes. ............................................................. 18

2.4.2. Solution-Sinkholes. ............................................................. 19

2.4.3. Cover-Subsidence Sinkhole. ............................................... 19

2.5. KARST TOPOGRAPHY IN MISSOURI ..........................................21

3. GEOPHYSICAL INVESTIGATIONS ............................................................. 23

3.1. MULTICHANNEL ANALYSIS OF SURFACE WAVE (MASW)..23

3.1.1. Overview. ............................................................................ 23

3.1.2. Basic Concept.. ................................................................... 24

3.1.3. Seismic Waves. ................................................................... 24

3.1.4. Body Waves. ....................................................................... 24

3.1.5. Surface Waves. ................................................................... 25

3.1.6. Rayleigh Waves in Homogeneous Elastic Half-Space ....... 27

3.1.7. Rayleigh Waves in Vertically Heterogeneous Elastic

3.2. MULTICHANNEL ANALYSIS OF SURFACE WAVES