Page 67 Vol 24, 4 2019

ParkSEIS-3D for 3D MASW

Surveys
Choon Park
Park Seismic LLC, Shelton, Connecticut
choon@parkseismic.com
Bio

Abstract Figure 1. A 3D visualization mode by ParkSEIS-3D of the shear-wave velocity (Vs)

data set obtained from a dedicated 3D MASW survey in Shelton, Connecticut.
Despite the potential value and efficacy of 3D surveys using
MASW, no software package has been able to handle all-in-one
3D data collection, analysis, and visualization. ParkSEIS-3D Although conventional 2D/1D surveys at different lines/points
provides the first such software package. It represents the latest can be cobbled together to achieve an output that provides
evolution of ParkSEIS, which has undergone continuous updates 3D analysis, a dedicated 3D survey requires well-defined
in underlying theoretical algorithms, empirical calibration, field logistics with properly designed shot and receiver (SR)
and graphical user interface (GUI), through field applications patterns. This ensures maximal effectiveness in surficial and
in several hundred different settings worldwide (conducted depth coverage while minimizing field operation efforts with
by Park Seismic LLC). While it is well-known that running a given set of field equipment. A proper software package
conventional 2D/1D surveys at multiple locations can ultimately should thus be able to provide such logistics for a potential 3D
produce and display a 3D volume of shear-wave velocity (Vs) project based not only on the size of 3D ground volume being
data by using multiple software packages, to date there has investigated but also the desired spatial sampling (bin) size for
been no dedicated software package that can handle optimum a given number of acquisition channels.
3D field operation, efficient data rearrangement, numerical analysis
Data analysis of a dedicated 3D survey requires special steps of
to create a 3D Vs data, and its visualization all in one package.
shot/receiver (SR) setup and data rearrangement into common-
This article provides an overview of the ParkSEIS-3D software mid-point (CMP) gathers. Location information of SR patterns
and its core theoretical properties and procedures. The most should be able to be entered using either Cartesian (*.txt) or
effective field logistics for dedicated 3D MASW surveys are GPS (*.gpx) coordinate files. Subsequent data sorting into CMP
described in terms of shot/receiver (SR) patterns. It describes gathers should take place with consideration of the applied
how corresponding field coordinates are conveniently set up by SR patterns, a bin size and the 3D ground volume. To further
importing either Cartesian coordinate (*.txt) or GPS (*.gpx) files, as simplify the process, it would be desirable for the dispersion
well as how streamlined graphical interfaces (GUI’s) incorporated and inversion analyses to proceed in a fully automated manner
at various critical steps of analysis ensure effective and user- to produce a 3D grid data set (*.G3D) at the end. A visualization
friendly 3D analysis. Automatic algorithms for dispersion and module should allow the user to freely display any part of the
inversion further simplify the procedure and ensure maximal subsurface from all directions and perspectives (Figure 1). The
consistency. Open-source vs. commercial software packages software would ideally also allow the 3D grid to be exported into
are also discussed briefly. a user-defined format for further use if necessary. ParkSEIS-3D
has been developed to provide all of these features.

Introduction
As MASW (Park et al., 1999) expands its areas of application, one
3D MASW − Data Acquisition
of the growing demands is the 3D survey. Although some earlier Common MASW surveys use a 1D linear receiver array (RA) to
pioneers executed such applications by running conventional generate a 2D shear-wave velocity (Vs) cross section through a roll-
2D surveys multiple times and displaying results by using along approach. Although surface waves spread into all azimuths,
separate visualization software packages (e.g., Miller et al., only the propagation along the path of the 1D RA is used to
2003; Suto, 2007; Park and Carnevale, 2009; Park and Taylor, 2010), generate a 1D velocity (Vs) profile being placed at the RA center.
there has so far been no dedicated software package that can Multiple such surveys can ultimately accomplish an equivalent
handle all-in-one 3D data collection, analysis, and visualization. 3D survey; e.g., by running parallel 2D lines. The most effective
That is, the coordination of “true 3D” field acquisition, data re- 3D survey, however, can take place only by using a 2D RA
arrangement for binning (Sheriff and Geldart, 1982), and 3D accompanied with a proper 2D shot pattern, similar to the 3D
visualization of final shear-wave velocity (Vs) data– all in one reflection survey. In this way, surface waves generated from
package (Figure 1). one shot point and propagating into multiple azimuths can be
 Page 68 Vol 24, 4 2019

a 1D velocity (Vs) profile to be placed at the corresponding CMP

location. Figure 2b shows all such CMP’s in red dots that are
created from the given shot-receiver configuration displayed in
Figure 2a and have six (6) or more shot-receiver pairs (traces)
sharing the same CMP. Each CMP point will have its own 1D
velocity (Vs) profile. It is noted that the CMP interval (10 m) is
only half the shot interval (20 m) and more CMP’s (96) are created
than the number of shots applied (55). This example illustrates
that a 3D survey can be simpler than multiple 2D surveys. In
addition, a CMP gather contains surface wave propagations from
multiple azimuths (Figure 2a), yielding a more realistic subsurface
sampling than the conventional 2D/1D survey. Further details
regarding the logistics for optimum shot patterns for a given 2D
RA will be expanded in a forthcoming article (Park, 2020).
ParkSEIS-3D includes a module that can generate a template
chart of shot and receiver patterns (similar to the ones in Figure 2)
based on acquisition-related parameters such as investigation
dimension, number of seismograph channels, takeout spacing
of seismic cable, and degree of desired spatial resolution. This
type of chart can be useful in designing potential 3D projects.

3D MASW − Data Analysis

The first step in 3D analysis is to set up spatial (X and Y)
configuration of shot/receiver (SR) patterns into each of the
acquired multichannel field records (i.e., SR setup). Encoding
these 2D coordinates during field acquisition is almost
prohibitive because most of engineering seismographs do not
provide such options. There are two ways to accomplish the SR
Figure 2. (a) Illustration of a dedicated 3D MASW survey with a stationary setup in ParkSEIS-3D. One is to use a Cartesian coordinate
24-channel 2D (X-Y) receiver array and a pattern of multiple shots. A common-mid-
point (CMP) is marked that has multiple (8) shot-receiver raypaths sharing the same system adopted during the field operation and therefore logged
point. (b) CMP's (red dots) that have six (6) or more traces created from the shot/ in the field notes. This information can be prepared in a text
receiver (SR) pattern in (a).
file (*.txt). Another is to use GPS coordinate files (e.g., *.gpx)
used to generate velocity (Vs) profiles at multiple locations. saved for all shot and receiver locations.
Efficiency of surface wave energy is therefore maximized. In
The next step is to sort field shot gathers into 2D (i.e., X-Y)
addition, if acquired field records are sorted into common-mid-
CMP gathers. All shot-to-receiver mid points are first mapped
point (CMP) gathers, the surficial sampling size (bin size) can be
on a fictitious chart for all possible combinations. Then, those
controlled not only by the pattern of receivers but also by the shot
points within a given bin size (e.g., 2 m x 2 m) are grouped
pattern. Controlling the shot pattern is a lot more convenient than
together as the same CMP. Next, only those CMP gathers are
controlling the pattern of whole receivers. This allows the bin size
selected for subsequent analysis that meet certain conditions
to become smaller than a receiver spacing used in the field. As a
such as an aperture size (e.g., 30 m), minimum number of
consequence, a relatively small number of acquisition channels
traces (e.g., 6), and source-receiver offset distributions. The
(e.g., 24 channels) can still be used for a dedicated 3D survey by
aperture size (D) is determined by the maximum investigation
using a stationary 2D RA of a long receiver spacing (e.g., 10 m)
depth being sought (Zmax); e.g., D = 2Zmax. These selected
to cover a relatively large area (e.g., 50 m x 50 m) with a relatively
CMP gathers then go through the normal MASW analysis
small (high) bin size (spatial resolution) (e.g., 2 m x 2 m).
sequence to generate 1D velocity (Vs) profiles being placed at
The general scheme of a dedicated 3D MASW data acquisition the corresponding CMP locations. Finally, a 3D grid data set is
is illustrated in Figure 2a. A 24-channel 2D RA is laid with a 20-m created by using an appropriate interpolation scheme. Once
longitudinal (West-East) and transverse (South-North) spacing. this 3D grid is prepared, it can be visualized in various ways (e.g.,
A pattern of multiple shots, among many other possibilities, is Figure 1) and exported into a user-defined format if necessary.
also displayed. To illustrate how a CMP gather is created, eight
ParkSEIS-3D provides a highly convenient graphical-user-
(8) different shot-to-receiver raypaths are marked in red lines,
interface (GUI) for shot/receiver (SR) setup, which can be
which have different shot-receiver offsets while sharing the
accomplished simply by importing two files prepared for receivers
same (common) midpoint (CMP). This 8-trace CMP gather can
and shot points in the form of either Cartesian (*.txt) or GPS (*.gpx)
go through the normal sequence of MASW analysis to generate
format (Figure 3a). Remaining steps of dispersion and inversion
 Page 69 Vol 24, 4 2019

followed by 3D grid generation proceed in a fully automated manner

without user intervention. The 3D grid can also be created from
multiple 2D and/or 1D surveys as illustrated in Figure 3b. Multiple 2D
cross sections can be imported by specifying surficial trajectories
either by using mouse or by importing location files (*.txt or *.gpx).
Multiple 1D profiles can also be added in the same way.

Algorithm and Graphical User

Interface (GUI)
The importance of GUI cannot be overstated, especially in applied
settings where the average users (e.g., civil engineers) are not
members of the technical community developing the methods (e.g.,
geophysicists). In ParkSEIS-3D, the shot/receiver (SR) setup is
accomplished through a convenient Graphic Wizard (GW) approach
that can avoid unnecessary confusion, due to for instance heavy
reliance on technical jargon.
There are two main steps in numerical analysis of MASW data:
dispersion and inversion. The purpose of dispersion analysis is to
extract a fundamental-mode (M0) [or an apparent mode (AM0), as
explained below] dispersion curve from field record. The following
inversion analysis automatically generates a 1D velocity (Vs) profile
by using the measured M0 curve. Correct extraction of M0 curve is
therefore critically important directly influencing the accuracy of the
profile. The proper extraction, however, requires operator’s ability
to accurately interpret energy patterns observed in the dispersion
image, which often requires theoretical knowledge and significant
observation experience with diverse field data sets. ParkSEIS-3D
Figure 3. Illustration of setting the shot/receiver (SR) patterns in ParkSEIS-3D (a) for
has incorporated an automatic algorithm (AUTO) similar to artificial a data set from a dedicated 3D survey by importing two separate files of receiver
intelligence (AI) that detects the correct M0 pattern in a mixture and shot patterns in either a Cartesian (*.txt) or a GPS (*.gpx) file, and (b) for multiple
data sets from conventional 2D/1D surveys by using mouse or importing location files
of complicated energy trends. The algorithm evolved through
(*.txt or *.gpx).
extensive self-learning process by using diverse field data sets
over the past two decades. is used depending on the velocity-inversion type. In consequence,
the inverse-velocity layers are more accurately handled than by the
ParkSEIS-3D uses two approaches for inversion of the measured traditional approach. In addition, the seismic modeling can always
dispersion curve; (1) fundamental-mode (M0) and (2) apparent- provide an ultimate inversion tool for any type of subsurface model.
mode (AM0) methods. Although there has been a great deal of
research and development in the higher modes inversion, software Figure 4 presents the algorithmic flowchart of ParkSEIS-3D
that takes full advantage of multi modes while efficiently handling that shows general procedures from both conventional multiple
all the associated complications has not yet been developed, at 2D/1D surveys and a dedicated 3D survey.
least to the level of commercial software package. This is because
of the modal characteristics of surface waves that are ultimately Discussion
determined by the velocity (Vs) profile of subsurface materials, the
The broader trend toward open-source software appears to be
unknown that we attempt to know through the MASW analysis.
influencing the geotechnical community as well. When available,
On the other hand, energy of the fundamental-mode (M0) surface
open-source software may be preferred among researchers and
waves almost always dominates the measured seismic wavefields.
students as it usually implements core theories, saves coding
In this sense, the traditional approach of the fundamental-mode
time, and provides room for customization.
(M0) inversion provides an excellent outcome under most common
overburden/bedrock settings. On the other hand, it is well known In many commercial settings, however, open-source software
that a velocity inversion, a layer of higher velocity (Vs) than the carries important risks. Proper calibration and optimization,
layer below, often results in an abnormal dispersion trend where established from real-world applications, and well-designed
phase velocities increase (instead of decrease) with frequencies. GUI are critical to ensure a reliable performance of the software
The trend usually appears to be one coherent mode, and therefore under diverse conditions. In this sense, applying open source
is called an apparent mode (AM0). Existence of AM0 trend is codes to commercial settings can often involve a great deal of
automatically detected in ParkSEIS-3D and a different inversion back-end customization, calibration, and modification especially
method is used; e.g., either mode-jump or Lamb-wave approach in terms of user interface as well as algorithmic optimization.
 Page 70 Vol 24, 4 2019

relatively large area (e.g., 100 m x 100 m) with a sufficient spatial

resolution (e.g., 2 m x 2 m) from one survey.
Diverse 3D visualization modes are available; i.e., cubic maps in
different directions and perspectives.
Output 3D velocity (Vs) grid data can be exported into a user-
defined format.

References
Miller, R.D., Anderson, T.S., Ivanov, J., Davis, J.C., Olea, R., Park, C.B.,
Steeples, D.W., Moran, M.L., and Xia, J. 2003, 3-D characterization
of seismic properties at the Smart Weapons Test Range, YPG [Exp.
Abs.]: Soc. Expl. Geophys., NSG 2.3.
Park, C. B., 2020, 3D MASW−field operation, data analysis and
visualization: SEG 90th Ann. Mtng., Houston, TX, October 11-16, in
preparation for submission.
Park, C.B., and Taylor, C., 2010, 3D MASW Characterization of Sinkhole:
A Pilot Study at USF Geology Park, Tampa, FL: Proceedings of the
Symposium on the Application of Geophysics to Engineering and
Environmental Problems (SAGEEP 2010), Keystone, Colorado, April 11-15.
Park, C.B., and Carnevale, M., 2009, 3D MASW: Proceedings of the
Symposium on the Application of Geophysics to Engineering and
Environmental Problems (SAGEEP 2009), Fort Worth, TX, March 29-April 2.
Park, C.B., Miller, R.D., and Xia, J., 1999, Multichannel analysis of surface
Figure 4. Algorithmic flowchart of ParkSEIS-3D for 3D MASW analysis from both
waves: Geophysics, v. 64, n. 3, pp. 800-808.
conventional 2D/1D surveys and a dedicated 3D survey.
Sheriff, R.E., and Geldart, L.P., 1982, Exploration seismology, vol 1:
For many engineering applications, the barriers to using open history, theory, and data acquisition: Cambridge University Press.
source materials can therefore be prohibitive, and may include
not only lack of in-house expertise but also time pressure and Suto, K., 2007, Multichannel analysis of surface waves (MASW) for
concerns about consistency in analysis results. An analogy would investigation of ground competence: an introduction, in “Engineering
be a home-made car with one engine sitting on a 4-wheel frame Advances in Earthworks,” Proceedings of the Sydney Chapter 2007
versus a commercially available car with all high-tech safety Symposium, Australian Geomechanics Society: 71-81.
and convenience features installed; most would-be drivers are
mainly interested in safe and reliable driving, less in the process
of tinkering with the parts and settings. Author Bio
Choon Park, Ph.D.
Conclusions Park Seismic LLC,
Founder/Principal Geophysicist
ParkSEIS-3D is a software package designed to enable 3D
MASW surveys (in addition to 1D and 2D surveys). It includes Shelton, Connecticut, USA
the following important features which may be particularly choon@parkseismic.com
important in enabling reliable all-in-one 3D survey and analysis:
Shot and receiver (SR) setup is conveniently accomplished through
Dr. Choon Park is lead author of the MASW technique published
Graphic-Wizard approach.
in GEOPHYSICS in 1999 while he was working at the Kansas
Coordinates of SR patterns are imported simply by preparing
Geological Survey (KGS). He is the author of the ParkSEIS
a Cartesian (*.txt) or a GPS (e.g., *.gpx) file.
software for MASW analysis. He is also author/co-author of
Field logistics for a potential 3D project are provided by a other near-surface seismic data-processing software packages
template chart of SR patterns that account for survey dimension,
developed at the KGS (SurfSeis, WinSeis, and Eavesdropper). Dr.
desired spatial sampling, and a given set of field equipment. Park founded Park Seismic LLC in 2007 to provide services and
Data rearrangement into CMP gathers takes place automatically Research and Development (R&D) in MASW method where he is
based on required depth of investigation and surficial dimension currently working as the principal geophysicist. He holds a B.S.
of survey area for given SR patterns applied in the field. in physics and science education from Seoul National University,
Dense shot patterns can be generated so that conventional an M.S. in Geophysics from Ohio University, and a Ph.D. from
field equipment (e.g., 24 channels) can be used to cover a the University of Kansas.