Nakamura Technique for Soil Characterization

ORIGINAL PAPER
Soil characterization of Bornova Plain (Izmir, Turkey)
and its surroundings using a combined survey of MASW
and ReMi methods and Nakamura’s (HVSR) technique
Eren Pamuk1,2
& Özkan Cevdet Özdağ3
& Mustafa Akgün4
Received: 7 December 2017 /Accepted: 24 April 2018
# Springer-Verlag GmbH Germany, part of Springer Nature 2018
Abstract
Structural damage which occurs during earthquakes is related to both the soil dynamic behavior attributes and soil response
spectrums. Therefore, soil characterization based of S-wave velocity (Vs) is one of the prime factors to estimate damage and
hazards. In this study, multi-channel analysis of surface waves (MASW) and refraction microtremor (ReMi) methods have been
applied to estimate Vs values of the area located east of Izmir Bay. Based on the estimated Vs values, two- and three-dimensional
shear-wave velocity and the maximum shear modulus maps at various depths down to 50 m were prepared. To study the
relationship between the predominant period of the soil and shear-wave velocity values, a horizontal-to-vertical (H/V) spectral
ratio method using Nakamura’s technique has been applied. Groundwater level, standard penetration test (SPT-N30) and Poisson
ratio values were also obtained from previous geotechnical borehole data in the study area. In addition, we calculated building
periods using the empirical relationship between height (or number of floors) of buildings and predominant period of the
buildings to examine soil-structure resonance. According to the National Earthquake Hazard Reduction Program (NEHRP;
1997) soil classification, the study site consists of B-, C- and F-type soils. Risk maps were created using dynamic properties
of the soil.
Keywords Soil classification . S-wave velocity (Vs) . Dynamic properties of soil . Soil-structure resonance . Bornova Plain
Introduction
In a seismically active areas, investigation of soil-structure in-
teraction due to strong earthquakes is one of the major tasks.
For this purpose, three main factors primarily responsible for
structural damage during an earthquake, e.g. earthquake source
parameters, local soil conditions and structure features, are an-
alyzed. Studies on structure damage due to earthquakes showed
that in any given area, even if the structural quality is the same,
the local soil attributes play an impactful role in the distribution
of the structural damage during an earthquake (Yalçınkaya
2010). Therefore, local soil behavior under dynamic earth-
quake loading needs to be thoroughly examined and identified
(Kramer 1996). The average S-wave velocity of the uppermost
30 m of a soil column (Vs30) values are mostly considered as
the basis for soil identification, and are calculated by the fol-
lowing formula in accordance with the following expression:
Vs30 ¼ 30= ∑N
i ðhi=Vi ) where hi and Vi denote the thickness
(in meters) and shear-wave velocity of the ith formation or
layer, in a total of N, existing in the top 30 m (Kanlı et al. 2006).
Many studies have been carried out to estimate the dynam-
ic properties of soils and soil characterization. Donohue et al.
(2004) calculated Gmax values using multichannel analysis of
surface waves (MASW) method for two soft clay sites in
Ireland. Pamuk et al. (2017) determined the shear-wave veloc-
ity structure and predominant period features of Tınaztepe in
İzmir using active–passive surface wave methods and single-
* Eren Pamuk
eren.pamuk@mta.gov.tr
1
The Graduate School of Natural and Applied Sciences, Dokuz Eylül
University, 35160 İzmir, Turkey
2
Department of Geophysical Research, MTA (General Directorate of
the Mineral Research & Exploration of Turkey),
06800 Ankara, Turkey
3
Dokuz Eylül University Aegean Implementation and Research
Center, 35430 İzmir, Turkey
4
Dokuz Eylül University Engineering Faculty Department of
Geophysical Engineering, 35160 İzmir, Turkey
Bulletin of Engineering Geology and the Environment
https://doi.org/10.1007/s10064-018-1293-7

station microtremor measurements. Carvalho et al. (2009) es-
timated P-wave velocity (Vp)-to-Vs ratios, Poisson’s ratio,
and a subsoil classification based on geophysical and
geotechnical parameters. Essien et al. (2014) determined
Poisson’s ratio using P- and S-wave measurements. Akin
and Sayil (2016) utilized active-source (MASW) and
passive-source (single-station microtremor) surface wave
methods for the soil dynamic characteristics in Trabzon,
Turkey. Cavallaro et al. (2006) described and compared the
results of in situ and laboratory investigations that were
carried out in order to determine the soil dynamic
characteristics in the city of Catania. Cavallaro et al. (2008)
used borings and dynamic in situ tests for site characterization
of the soil and deep site investigations in the city of Catania.
Castelli et al. (2016a) carried out site investigation for
determinig the soil geotechnical characteristics using down-
hole (DH) tests, dilatometer tests (DMT), MASW and
geotechnical borings in Sicily, Italy. Castelli et al. (2016b)
obtained seismic microzoning maps of the city of Catania in
terms of different peak ground acceleration at the surface and
in terms of amplification ratios for given values of frequency.
Ferraro et al. (2016) utilized boreholes, DH tests, seismic di-
latometer Marchetti tests (SDMT) and an MASW method in
the city centre of L’Aquila, Italy, for obtaining a detailed geo-
technical model. Caruso et al. (2016) combined traditional
tests for direct measurement of shear-wave velocity (DH,
cross-hole and SDMT) with indirect and less expensive tests
(MASW) for reliable geotechnical modeling used in seismic
response analysis. Cavallaro et al. (2016) conducted in situ
and laboratory investigations in order to obtain a
geotechnical model for a realistic seismic response analysis.
Cavallaro et al. (2017) utilized in situ investigations and lab-
oratory tests for geotechnical dynamic characteristics of the
foundation soil of the Augusta Hangar, Sicily, Italy. They pro-
vided a representative geotechnical model of the site where an
important historical building is located.
The aim of this study is to determine soil dynamic proper-
ties of the Bornova Plain and its surroundings using geophys-
ical and geotechnical data (Fig. 1a). At first, surface wave data
were collected using MASW, ReMi and single station
microtremor methods. The Vs30 values were calculated by
inverting phase velocity data obtained from a combined sur-
vey of MASW and ReMi. Vs30 values ranging from 180 to
1400 m/s were used to prepare a soil classification map of the
Bornova Plain according to the National Earthquake Hazard
Reduction Program (NEHRP 1997) soil classification. Based
on the one-dimensional (1D) velocity structure obtained from
inversion of the phase velocity data, as mentioned above, we
also created level maps of Vs values at depths of 5, 10, 20, 30,
40 and 50 m. Gmax values were calculated using Vs values at
different depths and rendered two- and three-dimensionally up
to a 50-m depth. In the next stage, single-station microtremor
data was evaluated according to the Nakamura method (1989)
and the predominant period of the soil was determined, which
was in the range of 0.45–1.6 s, indicating the area exhibits
lower predominant periods and less sediment thickness. In
the next step, previous geotechnical studies [standard penetra-
tion test (SPT-N), Poisson ratio, groundwater level (GL)] car-
ried out by Kıncal (2004) were examined. Next, risk maps
were created using geotechnical and geophysical data for the
study area. In the last stage, the predominant periods of the
high-rise buildings in the study area and hospitals and educa-
tional institutions in the risky areas were calculated with the
help of empirical relations between height (or number of
floors) of buildings and the predominant period of the build-
ings; thusly, the resonance condition was investigated.
Geology of the study area and its
surroundings
There are three tectonic belts in West Anatolia, around Izmir.
These belts are from the east to west; the Menderes Massif, the
Izmir–Ankara Zone and the Karaburun Belt (Bozkurt and
Oberhänsli 2001). The Menderes Massif consists of metamor-
phic rocks that reach to the early Eocene at their uppermost
levels. Located upon the Menderes Massif, the Izmir–Ankara
Zone is represented by sedimentary rocks settled on
Campanian–Danian-aged flysch facies and a mafic volcanic in-
tercalated matrix and a unit formed of limestone blocks longer
than 20 km, swimming in the matrix stretching in a wide region
from Manisa to Seferihisar. This unit is called the Bornova
Mélange; its limestone blocks and mega-blocks were carried to
a sedimentation environment during the precipitation of its ma-
trix and, consequently, complex contact structures, which exhibit
soft sediment deformations, were formed around the blocks. This
generalized stratigraphy was obtained by putting together the
measured sections of the limestone mega blocks, which is similar
to the carbonate pile that outcrops on the Karaburun Peninsula
(Erdoğan 1990). The foundation is formed by Upper Cretaceous
and old Bornova Mélange. Older limestone mega-olistoliths
compared to the mélange matrix are located in the matrix of
Bornova Mélange in a random order. The aforementioned lime-
stone is known as Işıklar limestone in Altındağ and its vicinity
(Özer and İrtem 1982). Bornova Mélange consists of platform-
type limestone which swims inside the matrix which is created
by the alternation of sandstone/shale chalk and diabase blocks
and pebble lens/channel fillings (Erdoğan 1990). Neogene-aged
lacustrine sediments come upon Bornova Mélange with angular
unconformity. Yamanlar ebonites cover the existing units uncon-
formably as well. A typical Bornova Mélange unit was segregat-
ed by the precipitated flysch facies between the Upper
Cretaceous and Paleocene; its matrix consists of alternating sand-
stone–shale , pebble lenses and limestone olistoliths (in various
ages and sizes) that are older than the matrix in which they are
located. The Bornova melange is in the oldest unit position in the
E. Pamuk et al.

study area. The aforementioned sedimentary rocks are pebbles,
argillaceous limestone and silicified limestone. Yamanlar ebo-
nites are represented by andesitic–dacitic massif lava, tuff, auto-
combined andesite and agglomerates within the field of the study.
Ebonites cover the regional Neogene sediment rocks unconform-
ably. Two faults with directions of NE-SW and E-W are discov-
ered in Izmir and its vicinity. It has been established that the
regular fault following an E-W direction is crossing the other
fault which is diagonally following a NE-SW direction (Kıncal
2004). Fundamentally being developed upon the same terrestrial
fillings, today’s alluvial plains near Izmir vary regarding their
geomorphologic formations. Deltas of the inner bay shores,
Balçova and Alsancak in the south and Karşıyaka in the north
are basic delta plains developed before mountain streams.
However, the Gediz Delta is a vast and complex geomorphologic
formation which collects water from the majority of the streams
in the Western Anatolia and is shaped by the Gediz River’s
alluvions. Despite originating from the shore, the Bornova
Plain in the east is not a typical delta plain, primarily due to the
lack of a large river carrying its water to the sea originating from
Bornova. In reality, the mountain stream watershed which de-
scends to Bornova is very close to the plain (Kayan 2000). The
geology of study area is shown in Fig. 1b.
Geophysical survey
A geophysical site characterization was conducted as part of
the Bornova Plain analysis. MASW and ReMi (64 sites),
ReMi and single-station microtremor (137 sites) measure-
ments were carried out in the study area (Fig. 1b).
Fig. 1 a) Representation of a site
location map of study area. b The
geological map of study area
together with geomorphology
(modified from Kıncal 2004; Uzel
et al. 2012)
Soil characterization of Bornova Plain (Izmir, Turkey) and its surroundings using a combined survey of MASW...

MASW method
MASW is a method of estimating the shear-wave velocity
profile from surface waves. It uses the dispersive properties
of Rayleigh waves for imaging the subsurface layers. In the
MASW method, surface waves can be easily generated by an
impact source (sledge hammer, etc.; Park et al. 1999).
MASW measurements were conducted at 64 sites. The
MASW system consisted of a 24-channel Geode seismograph
with 24 4.5-Hz geophones. The seismic waves were generated
by impulses from a hydraulic sledgehammer (100 lb) with
three shots. The data processing contains three steps. The first
is the preparation of a multichannel record, the second is
dispersion-curve analysis and the third is inversion (using a
least-squares approach).
The ReMi method
The ReMi process, developed by Louie (2001), has been
widely used to determine shear-wave velocity profiles
using ambient noise recordings. This array-analysis tech-
nique finds average surface-wave velocity over the length
of a refraction array. Also, 64 ReMi measurements were
carried out at the same locations as the MASW measure-
ments. For the ReMi measurements, eight records were
recorded at each site. The array lengths were 60 m and
120 m. In the ReMi interpretation and analysis, firstly, p–
f transformation, which is the basis of velocity spectral
analysis, takes a recorded section of multiple seismograms,
with seismogram amplitudes relative to distance and time
(x–t), and converts it to amplitudes relative to the ray pa-
rameter, p (the inverse of apparent velocity). Secondly, the
Rayleigh phase-velocity dispersion is selected. This analy-
sis only adds a spectral power ratio calculation for the
spectral normalization of noise records. The final step is
shear-wave velocity modeling. The modeling iterates on
phase velocity at each frequency.
Dispersion curves obtained by active (MASW) and
passive (ReMi) surface wave methods were combined to
enlarge the analyzable frequency range of dispersion and
improve the modal identity of the dispersion trends. High-
resolution Vs profiles were obtained by inverting the dis-
persion curve, and S-wave velocities were obtained from
the combined dispersion curves using the damped least-
squares method (Levenberg 1944; Marquardt 1963;
Fig. 2).
Fig. 2 Examples of Vs-depth cross-sections were obtained by inverting the combined dispersion curve
E. Pamuk et al.

Combined dispersion curves were used in the study to in-
crease the depth of the research and to identify the velocity
differences that occur within the soil in detail. According to
Vs-depth cross sections obtained from each measurement site,
sudden velocity differences are observed in a lateral and ver-
tical direction within the soil. These changes need to be con-
sidered for soil dynamic analysis studies. Based on the Vs30
distribution map which is made for soil type identification, it
has been observed that Vs30 values vary between 180 and
1400 m/s (Fig. 3).
When these velocity changes and the geological struc-
ture of the area are considered together, andesites and
Miocene pyroclastics north of the area of study and
Miocene-aged limestone in the south of area of the study
are observed to have higher Vs values compared to the rest
of the study area, and the threshold value of the bedrock’s
Fig. 3 Vs30 distribution values overlaid on the 3D topographic map of Bornova Plain and its surroundings
Fig. 4 Soil classification map of the study area according to NEHRP standards

Vs30 values are also observed to be higher than 760 m/s. In
spite of this, Vs30 values, especially in the areas nearby the
sea, are confirmed to change between 100 and 300 m/s,
and Vs30 values along the Bornova Plain are confirmed
Fig. 5 Average Vs distribution
maps of Bornova Plain and its
surroundings down to depths of 5,
10, 20, 30, 40 and 50 m depth
Fig. 6 Examples of H/V spectral
ratio for the study area (dashed
lines demonstrate the standard
deviation)
E. Pamuk et al.

to be lower than 500 m/s (Fig. 3). When evaluating the
NEHRP distribution map (Fig. 4), B, C and F soil types
are seen. We also created level maps of Vs values at depths
of 5, 10, 20, 30, 40 and 50 m (Fig. 5).
Single-station microtremor method [Nakamura
method (1989)]
The microtremor method is widely used for assessing the effect
of soil conditions on earthquake shaking. The horizontal-to-
vertical (H/V) spectral ratio was first introduced by Nogoshi
and Igarashi (1970). The H/V method is convenient and inex-
pensive for soil investigations. It was developed by Nakamura
(1989), who demonstrated that the ratio between horizontal and
vertical ambient noise records related to the fundamental fre-
quency and amplification of the soil beneath the site.
Microtremor observations were carried out at 137 sites in the
study area. All microtremor measurements were taken with the
Guralp Systems CMG-6TD seismometer with a sampling rate
Fig. 7 Predominant period distribution values overlaid on the 3D topographic map of Bornova Plain and its surroundings
Table 1 Relationship between Vs-Vp and density [P velocity values
were calculated using the equation Vp = Vs*1.74 (Catchings 1999)]
References Formula Material type
Destici (2001) ρ = 0.6*(Vs0.2
) Soil-bedrock
Keçeli (2009) ρ = 0.44*(Vs0.25
) Theoric
Komazawa et al. (2002) ρ = 0.7904*(Vs0.138
) ?
Uyanık (2002) ρ = 0.4*(Vp0.22
) Soil-bedrock
Uyanık and Çatlıoğlu (2015) ρ = 0.7*[(Vs*Vp)0.08
] Soil-bedrock
Fig. 8 a Geological map; b Vs10; c Gmax distribution maps at 10 m
overlaid on the 3D topographic map of Bornova Plain and its
surroundings

of 100 Hz. At each location, recording duration was approxi-
mately 30 min. To remove intensive artificial disturbance, all
signals were band pass-filtered in a band pass of 0.05–20 Hz.
Then, they were divided into 81.92-s-long windows and ta-
pered individually using Konno-Ohmachi smoothing. For each
window, the amplitude spectra of the three components were
computed using a fast Fourier transform (FFT) algorithm. As a
result, the average spectral ratio of H/V was thus calculated
(Fig. 6). Predominant period values were determined using
the H/V spectral ratio and they have been mapped (Fig. 7).
Microtremor measurements were processed and interpreted
using the GEOPSY software package (www.geopsy.org).
The predominant period values vary between 0.45 and
1.6 s where the regions have thick soil layers from rivers in
the Bornova Plain, and decreases gradually from the bay.
Decreasing periods are correlated with the increasing topog-
raphy in the north and south parts of the study area and indi-
cate a different geological unit (Fig. 7). Towards the eastern
part of the study area, the predominant period values shift
towards lower values. When maps of the predominant period
and Vs30 are compared, they match up to higher Vs values
where the period values are observed to be less than 1 s. The
period values are observed to be higher than 1 s and Vs30
values are observed to be much lower than 500 m/s, especially
in the areas which are Quaternary-aged and mostly consist of
soil layers that are thicker than 30 m.
Dynamic properties of the study area
Gmax studies; For geotechnical investigations, measurement of
the small strain shear modulus, Gmax, of a soil is very impor-
tant parameter. Gmax can be calculated from the shear-wave
velocity using the following equation:
Gmax ¼ ρ:Vs2*
100 ð1Þ
where Gmax = shear modulus (kg/cm2
), Vs = shear-wave ve-
locity (m/s) and ρ = density (gr/cm3
). The density values used
in modeling were calculated using S and P-velocity values and
formulas in Table 1.
Figure 8a, b shows that Vs10 and Gmax values at 10 m
depth. In addition, the calculated Gmax values was drawn
three-dimensionally until 50 m of depth with 3D Vs model
(Fig. 9a, b). Examining the Gmax distribution indicates the
Gmax values in the Bornova Plain are lower than 5000 kg/
cm2. As for the north and south parts of the study area, the
values are greater than 10,000 kg/cm2.
SPT-N studies were conducted on the 422 bores in the
study area (Kıncal 2004). When the SPT-N30 distribution
map was examined for 10 m of depth, it was observed that
these values vary between 0 and 60. Especially in the water-
front, these values are less than 20 and they increase towards
the east (Fig. 10b).
Fig. 9 a 3D Vs distribution map; b 3D distribution of Gmax of Bornova Plain and its surroundings up to a 50-m depth [the red rectangle shows that area
damaged during the 2003 Urla earthquake (Mw = 5.8)]
E. Pamuk et al.

Poisson ratio studies: In the S- and P-seismic refraction stud-
ies of Kıncal (2004) carried out on the 29 sites of the Bornova
Plain, the Poisson ratio has been calculated for 10 m of depth
(Fig. 10c). Poisson’s ratio (υ) can be calculated using Eq. 2;
v ¼
Vp2
−2Vs2
2 Vp2
−Vs2

!
ð2Þ
where Vp is P-wave velocity and Vs is S-wave velocity.
In the distribution map of the Poisson ratio at 10 m
of depth, it is observed that the areas with the values
between 0.3 and 0.5 probably have a loose soil layer.
The areas with values lower than 0.3 comprise hard and
firm soil.
GL studies; As a result of the measurements carried out in
the 422 boreholes, the GL map was prepared by Kıncal (2004;
Fig. 10d). Examining the GL distribution shows the GL in-
creases from west to east.
Predominant period values increase, as expected, on areas
that equaled the bay coast and in middle parts of sections in A-
A’ and B-B’ cross-sections. In C-C’, predominant period
values are higher than in other sections. In the A-A’ section;
lower Vs10 (100–200 m/s) and lower SPT-N30 (30) values
are obtained in middle parts of the A-A’ cross-section due to
the alluvial unit. GL values change from 1 to 8 m in the A-A’
cross-section. Poisson ratio values range from 0.15 to 0.42 in
the A-A’. Lower GL values are observed in middle parts of the
section, while lower Poisson ratio values are observed in the
southern part of the A-A’ cross-section. In the B-B’ cross-
section, Vs10 changes from 250 to 1300 m/s, while SPT-
N30 values are between 30 and 50 m. GL values are between
1 and 20 m in the B-B’, and Poisson ratio values range from
0.18 to 0.40 in the B-B’. Higher GL and lower Poisson ratio
values are obtained in 1–3 km of the B-B’ cross-section. In the
C-C’ cross-section, Vs10 changes from 150 to 550 m/s, while
SPT-N30 values are between 3 and 60 m. GL values are be-
tween 1 and 20 m in the C-C’ cross-section, and Poisson ratio
values range from 0.15 to 0.42. Higher GL and lower Poisson
ratio values are obtained in the east portion of the C-C’ cross-
section. Therefore, it is clearly seen that these values are gen-
erally compatible with each other (Fig. 11).
Soil-structure resonance studies
The main goal of resonance study is to determine the expected
resonance phenomena using the empirical relationship be-
tween the fundamental period of buildings and their height
(or floor number) during future earthquakes. We have calcu-
lated predominant periods of nine high-rise buildings using
empirical formulas in the study area. There are nine buildings
whose heights range from 68 to 216 m (i.e. from 17 to 50
floors), as listed in Table 2. The years of construction are from
2009 to 2017. The tallest building is the T8 and its height is
216 m, while the shortest building is T2 and its height is 68 m.
Regions where high-rise buildings exist have very low Vs
values. Moreover, the GL in this region is quite shallow
(Figs. 8 and 10). If the dominant period of the building and
the soil predominant period values are close to each other,
soil-structure resonance may occur. The predominant periods
calculated by the height-based area of Gallipoli et al. (2010;
T = 0.016 H; T = predominant building period, H = height of
building) varies between 1.1 and 3.5 s, while these values
calculated by the floor-based area of Navarro and Oliveira
(2008; T = 0.049 N; T = predominant period of building, F =
number of floors) change from 0.8 to 2.5 s. We calculated the
average of the predominant periods of the high-rise buildings
Fig. 10 a Fault zones in the study area with A-A’, B-B’ and C-C’ cross
sections; b SPT-N30; c Poisson ratio; d groundwater level distribution
maps at a 10 m overlaid on the 3D topographic map of Bornova Plain
and its surroundings

Fig. 11 a Geological cross sections; b predominant period; c Vs10; d groundwater level; e Poisson ratio; f SPT-N30 at 10-m depth on A-A’, B-B’ and C-
C’ cross sections
Table 2 Comparison of the high-rise buildings periods and soil predominant periods
Building
Building height (m) Number of floors Building period (sec) Soil predominant
period (sec)
NEHRP
soil type
Gallipoli
et al. (2010)
Navarro and
Oliveira (2008)
Average
period (sec)
T1 100 24 1.6 1.2 1.40 1.03 F
T2 68 17 1.1 0.8 0.95 1.13 F
T3 142 35 2.3 1.7 2.00 1.24 F
T4 200 45 3.2 2.2 2.70 1.20 F
T5 90 22 1.4 1.1 1.25 1.21 F
T6 216 48 3.5 2.4 2.95 1.22 F
T7 106 27 1.7 1.3 1.5 1.16 F
T8 100 27 1.6 1.3 1.45 1.28 F
T9 200 50 3.2 2.5 2.85 0.76 F
E. Pamuk et al.

obtained from different formulas in this study (Table 2). In
addition, predominant soil period values obtained from
microtremor measurements change from 1.0 to 1.6 s in the
high-rise building area. Figure 13 shows the predominant pe-
riod values of the buildings and the soil. We determined build-
ings T2, T5, T7 and T8 to potentially possess soil structure
resonance. Because periods of T4, T6 and T9 are higher than
the soil periods, these buildings are not significantly affected
by resonance.
Risk maps of the study area
Risk map I This map was formed by superimposing the area
with Vs30 values lower than 760 m/s and with soil predomi-
nant period values greater than 1 s (Fig. 12). The Urla earth-
quake on 10 April 2003 with a magnitude of Mw = 5.8 partly
damaged buildings with 8–9 floors in the Bornova Plain
(Kıncal 2004; Fig. 12). Figure 12 also shows the high-rise
buildings (towers) and health and education institutions. The
Fig. 12 Risk maps of the study
area
Fig. 13 Distribution of expected
resonance effect in the buildings
in Risk-1 area on soil
predominant period map. The
circle shows building periods.
Colour harmony signifies
probable resonance phenomena

dominant periods and the soil predominant period values, cal-
culated with the help of empirical relations, of the education
institutions, hospitals and towers risk map I are compared in
Fig. 13. The education institutions and the hospitals generally
have less than 10 floors. Therefore, it was not predicted that
most of these buildings would be affected by the resonance in
the area with soil predominant period values greater than 1 s.
But this is not the same for towers. Because the average pe-
riods of these buildings that vary between 17 and 50 floors is
between 0.95 and 2.95 s. The buildings that have the same or
similar color with the soil predominant period have resonance
risk (Fig. 13). The area including the buildings with eight or
nine floors that were damaged in the Urla earthquake are
completely in risk map I. In this area, the soil predominant
period values vary between 0.8 and 1.4 s. The dominant peri-
od values of the buildings with eight to nine floors are between
0.4 and 0.5 s with the Navarro and Oliveira (2008) relation-
ship. Then, it is possible to say that these buildings were not
exposed to the resonance effect during the Urla earthquake.
Risk map II Risk map II was prepared using the GL, Gmax,
Poisson ratio and SPT-N30 obtained for the 10-m depth of this
study and the study of Kıncal (2004). The considered values
for each of the data layers in risk map II are presented below;
the places where SPT-N30 ≤ 30; 0.3 ≤ Poisson ratio ≤ 0.5;
Gmax ≤ 3000 kg/cm2 and GL ≤ 10 values superimposed are
determined as the risk areas. Risk map II is included in risk
map I. Also, some of the buildings damaged during the Urla
earthquake are included in risk map II.
Conclusion
This study shows the Vs and predominant period characteris-
tics of different soil units in Bornova Plain. The Vs profiles
were determined by using an MASW and ReMi combined
survey at 64 sites. Also, predominant periods were determined
using Nakamura’s method at 137 sites. The predominant pe-
riod in the range is 0.45–1.6 s, which indicates the area ex-
hibits lower predominant periods and comprises less sediment
thickness (Figs. 5 and 7). The Vs30 values which change from
180 to 1400 m/s were used to create a soil classification map
of the Bornova Plain according to NEHRP (Fig. 3).
Soil classification results show that most parts of the region,
located in alluvial basin, have low shear-wave velocity values.
These values are within the range of 180–400 m/s and thus
classified the C and D categories according to NEHRP, gener-
ally. Some parts located on the north and the south part of the
study area have better soil conditions and have comparatively
high shear wave velocities in the range of 500–1400 m/s. Vs30
and soil classification maps were compared with predominant
distribution associated with the earthquake. According to re-
sults of comparing these parameters, in general, it is noticed
that there is a correlation between the Vs30 values and the pre-
dominant distribution of the region (Figs. 3 and 4).
According to Vs30 values, the bedrock is mostly dominant
in the regions south of the bay and north of the bay due to Vs30
values being higher than 760 m/s. These regions are classified
as B-type soil according to NEHRP regulations. C and F types
of soil are dominant in these regions according to NEHRP
regulations in other parts of the study area.
Based on the Vs level maps created for depths up to 50 m,
the soil thickness in the regions where C and F types are dom-
inant is more than 30 m (Fig. 5). The predominant period
values being higher than 1 s in these regions supports the notion
that the soil thickness of these regions is more than 30 m.
Distinctive soil bedrock models are suggested for soil de-
formation analysis in regions where the soil thickness is more
than 30 m and where there are sudden Vs differences in lateral
and vertical directions. Therefore, 2D or 3D soil engineering
bedrock models should be established for these regions.
Risk map- I almost covers the Bornova Plain. It can be seen
that there are many education institutions and hospitals within
the risk map 1 area. All of the buildings partially damaged in
the earthquake are located in the risk map. The existing Turkish
earthquake regulations CODE TE (2007) for buildings to be
built in this area are insufficient. Therefore, detailed studies are
required to calculate the in situ design spectrum. Resonance
risk exists for buildings with periods varying from 1 to 1.6 s
within this area. Seismic impedance must be determined from
the seismic bedrock (Vs 3000 m/s) to the surface in order to
calculate the earthquake effect on the surface within this area.
Risk map II is within the risk map I and some of the buildings
damaged by the earthquake are located within this area.
The number of floors of educational institutions and hospi-
tals is less than 10 (dominant period values of the buildings
between 0.4 and 0.5 s) in the risky areas. Therefore, it cannot
be said that these structures in risky areas carry a resonance
risk. However, some of the high-rise buildings carry a reso-
nance risk (Fig. 13). The resonance effect must be taken into
account when designing the buildings to be built in this area.
Therefore, it is proposed that the predominant period map be
used in the design of new buildings.
The GL in the study area is close to the surface, ranging
from 1 to 24 m, and is especially shallow in the western part of
the study area. Since these areas have risk of liquefaction dur-
ing an earthquake, these areas should be examined in detail.
The Poisson ratio, which indicates the water saturation level of
soil, varies from 0.3 to 0.5, especially in soft soil. The Poisson
ratio varies from 0.3 to 0.5 where the GL is 10 m or lower in
the study area. The Gmax values at a 10-m depth are less than
4000 kg/cm2
, in accordance with these values.
Acknowledgements This work was realized within the scope of Mr. Eren
Pamuk’s PhD thesis at Dokuz Eylul University, The Graduate School of
Natural and Applied Sciences. Microtremor measurements in this
E. Pamuk et al.

research were provided by DEU BAP (project no. 2015FEN032).
MASW and ReMi data in this research were provided by TUBITAK-
KAMAG (project no. 106G159).
References
Akin Ö, Sayil N (2016) Site characterization using surface wave
methods in the Arsin-Trabzon province. Environ Earth Sci
75:72. https://doi.org/10.1007/s12665-015-4840-6
Bozkurt E, Oberhänsli R (2001) Menderes Massif (western Turkey):
structural, metamorphic and magmatic evolution-a synthesis. Int J
Earth Sci 89(4):679–708
Caruso S, Ferraro A, Grasso S, Massimino MR (2016) Site response
analysis in eastern sicily based on direct and indirect Vs measure-
ments. 1st IMEKO TC4 International Workshop on Metrology for
Geotechnics, MetroGeotechnics 2016
Carvalho J, Torres L, Castro R, Dias R, Mendes-Victor L (2009) Seismic
velocities and geotechnical data applied to the soil microzoning of
western Algarve, Portugal. J Appl Geophys 68(2):249–258
Castelli F, Cavallaro A, Ferraro A, Grasso S (2016a) In situ and laboratory
tests for site response analysis in the ancient city of Noto (Italy). 1st
IMEKO TC4 International Workshop on Metrology for
Geotechnics, MetroGeotechnics 2016
Castelli F, Cavallaro A, Grasso S, Lentini V (2016b) Seismic microzoning
from synthetic ground motion earthquake scenarios parameters: the case
study of the city of Catania (Italy). Soil Dyn Earthq Eng 88:307–327.
(ISSN: 0267- 7261). https://doi.org/10.1016/j.soildyn.2016.07.010
Catchings RD (1999) Regional Vp, Vs, Vp/Vs, and Poisson’s ratios
across earthquake source zones from Memphis, Tennessee, to St.
Louis, Missouri. Bull Seismol Soc Am 89(6):1591–1605
Cavallaro A, Grasso S and Maugeri M (2006) Volcanic soil characterisation
and site response analysis in the city of Catania. Proceedings of the 8th
National Conference on Earthquake Engineering, San Francisco, 18- 22
April 2006, paper n°. 1290, vol 2, pp 835–844. (ISBN: 1-932884-25-4)
Cavallaro A, Ferraro A, Grasso S, Maugeri M (2008) Site response analysis
of the Monte Po Hill in the city of Catania. Proceedings of the 2008
Seismic Engineering International Conference Commemorating the
1908 Messina and Reggio Calabria Earthquake MERC EA’08,
Reggio Calabria and Messina, 8- 11 July 2008, pp. 240–251. (ISBN:
978-0-7354-0542-4/08; EISSN: 15517616; ISSN: 0094-243X), https://
doi.org/10.1063/1.2963841. AIP Conference Proceedings, Volume
1020, Issue PART 1, 2008, pp 583–594. (ISSN: 0094243X)
Cavallaro A, Grasso S, Ferraro A (2016) Study on seismic response
analysis in BVincenzo Bellini^ garden area by seismic dilatometer
Marchetti tests. Proceedings of the 5th International Conference on
Geotechnical and Geophysical Site Characterisation, ISC 2016
Cavallaro A, Castelli F, Ferraro A, Grasso S, Lentini V (2017) Site response
analysis for the seismic improvement of an historical and monumental
building: the case study of Augusta Hangar. Bulletin of Engineering
Geology and the Environment, November 2017. Springer, pp 1–32.
(ISSN: 1435-9529). https://doi.org/10.1007/s10064-017-1170-9
Code TE (2007) Specification for structures to be built in disaster areas
Ministry of Public Works and Settlement Government of Republic
of Turkey
Destici C (2001) Obtaining dynamic and static parameters with seismic
wave velocities, SDU MMF Geophysics Eng. Bachelor Thesis,
Isparta (Unpublished)
Donohue S, Long M, O’Connor P, Gavin K (2004) Use of multichan-
nel analysis of surface waves in determining Gmax for soft clay. In:
Proceedings 2nd. Int. Conf on Geotechnical Site Characterisation,
ISC, vol 2, pp 459–466)
Erdoğan B (1990) Stratigraphic features and tectonic evolution of the İzmir-
Ankara zone located between İzmir and Seferihisar. Turkish Association
of Petroleum Geologist (TPJD) Bulletin 2:1–20 (in Turkish)
Essien UE, Akankpo AO, Igboekwe MU (2014) Poisson’s ratio of surface
soils and shallow sediments determined from seismic compressional
and shear wave velocities. Int J Geosci 5:1540–1546. https://doi.org/
10.4236/ijg.2014.512125
Ferraro A, Grasso S, Maugeri M, Totani F (2016) Seismic response anal-
ysis in the southern part of the historic Centre of the City of L’Aquila
(Italy). Soil Dyn Earthq Eng 88:256–264
Gallipoli MR, Mucciarelli M, Šket-Motnikar B, Zupanćić P, Gosar A,
Prevolnik S, Herak M, Stipćević J, Herak D, Milutinović Z,
Olumćeva T (2010) Empirical estimates of dynamic parameters on
a large set of European buildings. Bull Earthq Eng 8(3):593–607
http://www.geopsy.org. Access Date 20102016
Kanlı Aİ, Tildy P, Pronay Z, Pınar A, Hemann L (2006) Vs30 mapping
aAnd soil classification for seismic site effect evaluation in dinar
region, Sw Turkey. Geophys J Int 165:223–235
Kayan I (2000) Morphotectonic units and alluvial geomorphology of the
İzmir area. Seismicity of Western Anatolia symposium, 10
Keçeli A (2009) Applied geophysics, chamber of geophysıcal engineers of
Turkey, No:9
Kıncal C (2004) Evaluation of geological units in İzmir inner bay and it’s
vicinity by using GIS and remote sensing systems in terms of engi-
neering geology Dokuz Eylül University, The Graduate School of
Natural and Applied Sciences, PhD Thesis, İzmir (in Turkish)
Komazawa M, Morikawa H, Nakamura K, Akamatsu J, Nishimura K,
Sawada S, Erken A, Önalp A (2002) Bedrock structure in
Adapazari, Turkey—a possible cause of severe damage by the
1999 Kociaeli earthquake. Soil Dyn Earthq Eng 22:829–836
Kramer SL (1996) Geotechnical earthquake engineering. Prentice-Hall,
Inc, Upper Saddle River
Levenberg K (1944) A method for the solution of certain non-linear
problems in least squares. Q Appl Math 2:164–168
Louie JN (2001) Faster, better: shear-wave velocity to 100 meters depth from
refraction microtremor arrays. Bull Seismol Soc Am 91:347–364
Marquardt D (1963) An algorithm for least-squares estimation of nonlin-
ear parameters. J Soc Ind Appl Math 11(2):431–441
Nakamura Y (1989) A method for dynamic characteristics estimation of
sub-surface using microtremor on the ground surface. Quarterly
Report of Railway Technical Research Institute 30:25–33
Navarro M, Oliveira CS (2008) Experimental techniques for assessment
of dynamic behaviour of buildings. In: Assessing and managing
earthquake risk. Springer, Netherlands, pp 159–183
NEHRP (1997) Recommended provisions for seismic regulations for
new buildings and other structures. FEMA-303 Prepared by the
Building Seismic Safety Council for the Federal Emergency
Management Agency Washington DC
Nogoshi M, Igarashi T (1970) On the propagation characteristics of
microtremors. J Seismol Soc Jpn 23:264–280
Özer S, İrtem O (1982) Geological setting, stratigraphy and facies char-
acteristics of the upper cretaceous limestones in the Işıklar-Altındağ
(Bornova-İzmir) area. Turkey Geology Bulletin 25:41–47
Pamuk E, Özdağ ÖC, Özyalın Ş, Akgün M (2017) Soil characterization
of Tınaztepe region (İzmir/Turkey) using surface wave methods and
Nakamura (HVSR) technique. Earthq Eng Eng Vib 16(2):447–458
Park CB, Miller RD, Xia J (1999) Multichannel analysis of surface
waves. Geophysics 64(3):800–808
Uyanık O (2002) Potential liquefaction analysis method based on shear
wave velocity. DEU The School of Natural and Applied Sciences
(PhD Thesis), İzmir, p 176
Uyanık O, Çatlıoğlu B (2015) Determination of density from seismic
velocities. Jeofizik 17:3–15 (in Turkish)
Uzel B, Sözbilir H, Özkaymak Ç (2012) Neotectonic evolution of an
actively growing superimposed basin in western Anatolia: the inner
bay of Izmir, Turkey. Turk J Earth Sci 21(4):439–471
Yalçınkaya E (2010) Why is the soil so important? UCTEA Chamber
Geophysics bulletin 63:77–80 (in Turkish)
View publication stats
View publication stats

Nakamura Technique for Soil Characterization

More Related Content

What's hot (20)

Similar to Nakamura Technique for Soil Characterization (20)

More from Ali Osman Öncel (20)

Recently uploaded (20)

Nakamura Technique for Soil Characterization