© 2015 European Association of Geoscientists & Engineers 447
Near Surface Geophysics, 2015, 13, 447-453 doi:10.3997/1873-0604.2015031
*
t.shirzad@srbiau.ac.ir
Near-surface V
S
structure by inversion of surface
wave estimated from ambient seismic noise
Taghi Shirzad
1*
, Z. Hossein Shomali
2,3
, Mojtaba Naghavi
2
and Rahim Norouzi
4
1
Department of Physics, Islamic Azad University, Damavand Branch, Damavand 39715-194, Iran
2
Institute of Geophysics, University of Tehran, Tehran 14155-6466, Iran
3
Department of Earth Sciences, Uppsala University, 752 36 Uppsala, Sweden
4
Tehran Disaster Mitigation and Management Organization, Tehran 14636-13111, Iran
Received July 2014, revision accepted April 2015
ABSTRACT
In this study, 20 hours of ambient seismic noise recorded from a small-scale inter-station distance
was used to obtain near-surface shear wave velocity structures at a local test site in Tehran (Iran).
High-resolution group velocity dispersion curves using fundamental mode of surface waves were
calculated for all possible combinations of station pairs at frequencies ranging from 1 Hz to 25 Hz.
Unlike most previous studies regarding ambient seismic noise, which observe very little coherent
noise at frequencies larger than 1 Hz, the empirical Green’s functions were extracted using a root-
mean-square stacking method showing more coherent signals. Our results indicate that ambient
seismic noise is a viable technique at a frequency range of 1 Hz–25 Hz even when different sensor
types are present. One-dimensional V
SV
and V
SH
models from the near surface were then assessed
by inverting the calculated Rayleigh and Love waves’ dispersion measurements. We observed that
the calculated shear wave velocity model agrees with the available downhole model and shows three
distinct layers in the upper 25 m of the test site.
at different depths in a single borehole. Two conditions are
required to reliably estimate a V
S
profile. First, accurate travel-
time measurements are a fundamental aspect of any downhole
studies, and second, a reliable data reduction method that consid-
ers the ray path is important (Campanella and Stewart 1992). The
downhole method has some advantages such as low cost (only
needs one borehole), ease of operation, and the use of a simple
surface seismic source compared with the other downhole seismic
test methods, such as crosshole tests that need two boreholes.
Shapiro and Campillo (2004) showed that the cross-correla-
tion of ambient seismic noise can provide empirical Green’s
functions (EGFs) between receiver pairs. These results can be
applied to study medium structures, e.g., travel-time tomography
studies at regional scales as those performed by Shapiro et al.
(2005) and Yang et al. (2007), and in extracting V
S
profiles at
small scales, such as the work presented by Picozzi et al. (2009).
In general, previous ambient seismic noise studies used data with
frequency ranges of less than 1 Hz. An improved knowledge of
the shallow structure in metropolitan areas using inverted surface
wave group velocity dispersion curves extracted from ambient
seismic noise is a very powerful tool to calculate V
S
models
versus depth for near-surface structures. Groos and Ritter (2009)
explained that man-made sources affect the frequency spectrum
content of ambient seismic noise in Bucharest, Romania, for
frequency ranges of 0.04 Hz–45 Hz. Consequently, high levels of
INTRODUCTION
One of the fundamental interests in engineering, environmental,
and groundwater studies is the shear wave velocity (hereafter,
V
S
) structure of unconsolidated sediments and the V
S
contrast
between bedrock and overlying sediments (Xia, Miller, and Park
1999). These physical properties and the geometry of the bed-
rock–sediment interface directly influence seismic wave propa-
gation (Panza et al. 2002; Nunziata, Nisco, and Panza 2009).
Interfering P- and V
SV
-waves result in Rayleigh waves for a
given mode with longer periods (or lower frequencies) that pen-
etrates deeper compared with a mode with shorter periods (or
higher frequencies; Babuska and Cara 1991). Therefore, shorter
periods (or higher frequencies) are sensitive to the physical prop-
erties of near-surface layers. Furthermore, two types of veloci-
ties, known as phase and group velocities, characterize the basic
feature of surface wave dispersion in a given layered medium.
Consequently, the obtained dispersion curves of Rayleigh or
Love waves can be used to study the properties of the medium
(Aki and Richards 1980).
Downhole seismic tests are widely used to measure in situ
compression (P-wave) and shear wave (S-wave) velocity profiles
(e.g., Kim, Bang, and Kim (2004)). The downhole method meas-
ures the body wave travel time from the surface source to receivers