Lo Presti - Et - Al - KL - ICCE - 2017 - Tirana
Lo Presti - Et - Al - KL - ICCE - 2017 - Tirana
Lo Presti - Et - Al - KL - ICCE - 2017 - Tirana
Use in practice of seismic tests according to Eurocodes
Diego Carlo Lo Presti1, Stefano Stacul1, Ilaria Giusti1, Claudia Meisina2, Maria
Giuseppina Persichillo2, Edoardo Pagani3, Massimiliano Siviero3, Ervis Uruci3, Ermanno
Pagani3
1Department of Civil and Industrial Engineering, University of Pisa, Pisa, Italy
2Department of Earth and Environmental Sciences, University of Pavia, Pavia, Italy
3Pagani Geotechnical Equipment, Calendasco (PC) Italy
Abstract
Seismic tests consist in the measurement of propagation velocity of body or surface
waves into the subsoil or at the contact soil/air or soil/water. Surface waves are dispersive by
nature and propagate along the contact surface of two media having very different properties.
Surface waves velocities can be measured at ground surface or at the sea floor. Body wave
velocities are measured in hole and therefore require a single or multiple holes. A cost effective
way of conducting such a test is to push into the soil the receivers as in the case of Seismic
CPT (SCPT) or Seismic DMT (SDMT). Measurements require a source, single or multiple
receivers, trigger and data acquisition system. This KN paper gives some details about testing
procedures and focuses on the practical use of seismic measurements in Civil Engineering.
More specifically, the following applications are considered: assessment of seismic action at a
given site, definition of impedance function for dynamic soil-structure interaction, in situ
assessment of damping ratio. A comparative case that was conducted at the University of Pavia
campus, is shown.
Keywords: Seismic tests, seismic actions, Eurocode, impedance functions, damping
ratio.
1 INTRODUCTION
It is well known that the analysis of the complex dynamic soil-foundation-superstructure
interaction can be simplified studying separately [1]:
1. the free-field response;
2. the kinematic interaction;
3. the inertial interaction.
The first step provides an estimate of the seismic motion variation during the propagation
of the seismic waves from the bedrock to the ground surface, in terms of amplitude and
frequency content, due to the presence of a weaker overlaying soil deposit. Such an analysis
neglects the influence of both the foundation and the superstructure and only requires the wave
velocity profile and the soil parameters (according to the selected constitutive model). The
other two steps also require the foundation impedance functions.
2
It is worth noticing that free – field seismic response analyses are also used for seismic
microzonation studies.
In any case, assessment of the soil volume which is relevant for the study problem and
the velocity – wave profile within such a soil volume are fundamental. In the following, the
measurement methods are shown as well as the parameters that can be obtained from the
seismic measurements, in the light of Eurocode 8 prescriptions. The paper also shows a
comparative study case of seismic measurements that were conducted in the campus of the
University of Pavia, Italy.
Seismic tests are conventionally classified into borehole (invasive) and surface (non-
invasive) methods. They are based on the propagation of body waves [compression (P) and/or
shear (S)] and surface waves [Rayleigh (R)], which are associated to very small strain levels
(i.e. less than 0.001 %) [2]. Assuming a linear elastic response, the following relationships
allow to compute the small-strain deformation characteristics of the soil from the measured
body wave phase velocities:
Go Vs2 (1)
M o V p2 (2)
where: Go , M o = small strain shear and constrained modulus respectively; = mass density;
Vs , V p = velocity of shear and compression waves respectively; = Poisson ratio.
The above relationships hold for elastic isotropic media. Moreover, in the case of
saturated porous media the measured P wave velocity corresponds to the compression wave of
the first kind [3-4] that is strongly influenced by the pore fluid. In this case the above equations
are no longer valid and must be replaced with the corresponding ones of poroelasticity theory.
Seismic tests may also be used to determine the material damping ratio by measuring the
spatial attenuation of body or surface waves:
V
Do ( Do 10%) (4)
2f
where Do = small-strain material damping ratio; , V = attenuation coefficient and
velocity, respectively, of P, S or R waves and f = frequency.
Material damping measurements are difficult because they require accurate
measurements of seismic wave amplitude and accurate accounting of the effects of geometric
(radiation) attenuation [5].
Even at strains less than the linear threshold strain, soils have the capability not only of
storing strain energy (elastic behaviour) but also of dissipating it over a finite period of time
(viscous behaviour) [6]. This type of behaviour can accurately be modelled by the theory of
linear viscoelasticity. An important result predicted by this theory is that soil stiffness and
material damping are not two independent parameters, but they are coupled due to the
phenomenon of material dispersion [7].
3
Lai and Rix (1998) [8], Lai et al. (2001) [9], Rix et al. (2001) [10] and Lai et al. (2002)
[11] developed implies rigorous approaches for a simultaneous estimate of the velocity of
propagation of seismic waves and material damping ratio.
2.1 Borehole Methods
The most widely used borehole methods in geotechnical engineering are Cross Hole
(CH), Down Hole (DH), Suspension PS logging (PS) [12] tests. Strictly speaking, the Seismic
Cone (SCPT) and Seismic Dilatometer (SDMT) tests are not borehole methods, but they are
based on the same principle. Their popularity is due to the conceptual simplicity. The
measurement of the travel time of P and/or S waves, travelling between a source and one or
more receivers is determined from the first arrival of each type of wave. Current practice of
borehole methods is covered by many comprehensive works [1; 13-17]. In the following, only
some aspects of the borehole methods are briefly summarized. In particular, the focus is placed
on emphasising the importance of respecting these testing procedures:
- good mechanical coupling between receiver, borehole casing (if used) and surrounding
soil must be guaranteed. A distinct advantage of the SCPT and SDMT is that good coupling is
virtually assured. With conventional cased and grouted boreholes, good coupling is less certain
and, more importantly, is difficult to verify. The need for good coupling is particularly
important for attenuation measurements, which require accurate amplitude data;
- a check of the borehole verticality with an inclinometer is also highly recommended in
order to determine accurately the length of wave travel path in CH tests;
- it is important to generate repeatable waveforms with the desired polarity and
directivity. This allows receivers to be oriented in such a way to optimise the measurement of
a particular wave type, the use of reversal polarity to make the identification of wave arrivals
easier, and measurements along different directions to infer structural and stress-induced
anisotropy as explained below;
- in down-hole measurement, the use of two of receivers located at a fixed distance apart
[18] can increase the accuracy and the resolution because the true interval method for data
interpretation can be implemented;
- dedicated portable dynamic signal analysers and computer-based data acquisition
systems allow more sophisticated data processing methods. Thanks to these enhancements, it
is now possible to routinely use cross correlation (time domain) or cross power spectrum
(frequency domain) techniques to estimate travel times instead of subjective identification of
the first arrivals in the time histories. In addition, as multi-channel data acquisition systems
become more common, the logical extension will be to use arrays of receivers and array-based
signal processing (seismic tomography).
Generally, the shear wave velocity profiles inferred from various borehole tests are in
good agreement (see the example in Figure 1). However, SCPTs generally provide values of
the shear wave velocity slightly larger than those inferred from down-hole or cross-hole tests.
4
Figure 1 Shear wave velocity in Po river sand [19]
Some researchers stated that the differences between the velocities that were inferred
from CH and DH tests could be attributed to soil heterogeneity and anisotropy. The following
considerations explain why anisotropy cannot be responsible for these differences. In CH tests,
S waves propagate in the horizontal direction with vertical particle motion ( S hv ) . This type of
wave is generated by conventional mechanical source. Use of electromagnetic source (which
is not usual) can generate waves that are polarized in the horizontal plane ( S hh ) . In DH tests,
propagation of the S wave is sub vertical with horizontal particle motion ( S vh ) . In a continuous
medium, the Vsvh and Vshv shear wave velocities are the same and a unique value of the shear
modulus ( Gvh G hv ) is expected. Figure 2 shows that Vsvh = Vshv (i.e. Gvh G hv ) in the case of
laboratory tests on reconstituted sample of Fujinomori clay [20]. Measurements of the
propagation velocity Svh and Shv waves were performed by means of Bender Elements (BE).
Similar results were obtained in the case of reconstituted sands by Stokoe et al. (1991)[21], Lo
Presti & O' Neill (1991)[22] and Bellotti et al. (1996)[23]. Hence, different values of shear
wave velocity from CH and DH tests are most likely due to soil heterogeneity, different volume
of soil that was interested by the ray paths, as well as intrinsic scatter of experimental
measurements [14].
5
250 150
200 120
Vs , Vs [m/s]
Stresses [kPa]
150 90
hv
100 60
vh
50 30
0 0
0 2 4 6 8 10 12 14 16 18 20 22
Axial strain ea [%]
Figure 2 Vsvh and Vshv measured with BE during drained CLTX test on Fujinomori clay
Figure 3 Small strain stiffness anisotropy: field versus laboratory data (modified after
[23])
The assessment in situ of inherent and stress induced elastic anisotropy is possible by
measuring the velocity of propagation of both S hv and S hh waves in CH tests [24-26]. S hh
waves. propagate in the horizontal direction with particle motion polarized in the
complementary horizontal direction. This additional information enables the evaluation of the
Ghh / Gvh ratio, which is a function of inherent and stress-induced anisotropy. Figure 3
summarises some field and calibration chamber data. Figure 3 indicates that, for the considered
granular soils, the inherent anisotropy (inferred at K c 1 ) causes a 20% to 25% increase in
G hh over Gvh . The influence of stress induced anisotropy is apparent for other values of K c .
Efforts have been done to inferring the small-strain damping ratio, Do from borehole
tests. The current methods are based on measures of the spatial attenuation between two or
more receivers. The most widely used methods include:
6
a) The spectral ratio method [27][28] is based on the following assumptions which
hold only in the far field: i) the amplitude of the body waves decreases in proportion
to r 1 , where r is the distance from the source, due to geometric attenuation and ii)
the soil-receiver transfer function can be considered identical for both receivers.
Based on the above assumptions, the damping ratio can be computed by means of
the following equation:
lnA1 ( f ) r1 / A2 ( f ) r 2
D( f ) (5)
( f )
where: r1 and r2 are the distances from the source of a pair of receivers,
A1 ( f ) and A2 ( f ) are the amplitude spectra at the two receivers and ( f ) is the phase
difference between the two receivers.
b) The spectral slope method, originally developed for downhole measurements [29]
[30] differs from the spectral ratio method because it assumes that material damping
is frequency independent and that it is not necessary to define the law for geometric
attenuation. The attenuation constant, defined as the ratio of attenuation coefficient
to frequency k / f , represents the spectral slope, i.e. the slope of the spectral
ratio vs. frequency curve:
ln A1 ( f ) / A2 ( f )
k (6)
f ( r 2 r1 )
therefore the material damping can be computed using the following expression:
D( f ) lnA1 ( f ) / A2 ( f ) (7)
f 2 t( f )
Both methods require signal processing prior to interpretation to isolate direct arrivals
and frequency ranges. They provide damping values in the bandpass range of the filter.
Khawaja (1993) [31] and Fuhriman (1993) [28] recommend performing crosshole tests
with four boreholes, in order to obtain stable values of damping with the spectral ratio method.
They suggested placing the source in the outer boreholes, in order to propagate waves in both
forward and reverse directions, and the receivers in the two central boreholes. The spectral ratio
method with combined directions provides stable values of damping and avoids the extreme
case of negative damping values [32]. The main concerns with the application of these methods
is the accuracy in measuring wave attenuation. Use of combined directions and of a maximum
distance (source – receiver) of about 8 m should compensate the low repeatability of geophones
in terms of amplitude. On the other hand the use of calibrated accelerometers should be
preferable especially in the case of short distances between receivers. Campanella & Stewart
(1990) [32] studied the applicability of the above methods to the downhole SCPT's. They found
that the spectral slope method provides more realistic values of material damping. However, in
downhole tests, wave amplitudes are also affected by reflection/transmission phenomena at the
interfaces between layers and by ray path divergence: these phenomena make more complicate
the interpretation of the particle motion amplitude.
Examples of damping measurements with the spectral ratio and spectral slope methods
for reconstituted Ticino sand can be found in Puci & Lo Presti, (1998) [33]. The results are
from seismic tests, performed with miniature geophones embedded in large-size calibration
chamber specimens. Figure 4 [33] compares the damping ratio values obtained in the case of
7
reconstituted Ticino sand from laboratory tests (RCT) and those inferred from the spectral ratio
and spectral slope methods applied to calibration chamber seismic tests. In this very controlled
experiment, the seismic methods yield values of the material damping ratio that generally agree
with laboratory values. Measured damping ratios are plotted vs. the corresponding
consolidation stresses which have a great influence on the results.
Other approaches to measure the material damping ratio include the rise time method,
based on the experimental evidence that a seismic wave signal broaden with distance because
of material damping, and the waveform matching method. However, at the present time, none
of the available borehole methods to measure material damping ratio appears to be robust
enough for routine use in geotechnical engineering practice.
10
Ticino Sand
g 10-4%
f 1000÷3000Hz
Damping ratio: D [%].
g 10-3%
f 80÷140Hz
Seismic Tests (Spectral Ratio Method: Dr=45%)
Seismic Tests (Spectral Slope Method:Dr=45%)
Resonant Column (Dr=45%)
Resonant Column (Dr=85%)
0.1
100 1000 10000 100000
Consolidation Stress: 'v ∙ 'h [kPa]∙[kPa]
Figure 4 Damping ratio from laboratory and geophysical seismic tests on reconstituted
sands [33].
2.2 Surface Methods
Surface methods are non-invasive field techniques that are executed from the ground
surface of a soil deposit or from the sea floor, hence they do not require drilling of boreholes
or insertion of probes. They include seismic refraction, high-resolution reflection and surface
wave methods. Seismic refraction and reflection methods are not widely used for near-surface
site characterisation, particularly for S-wave velocity profiling. This is partially due to the fact
that there are situations (stiffer-over-softer layers; hidden layers) where the seismic refraction
method cannot be reliably applied [34]. High-resolution reflection, on the other hand, does not
suffer such limitations, however it requires very intensive data processing.
Advantages of surface methods are mainly related to their non-invasive nature. They are
more economical and can be performed more rapidly than borehole methods. Furthermore, in
sites like solid waste disposals and landfills, due to environmental concerns, surface methods
can be the only choice for geotechnical investigations. Another peculiar aspect of surface
methods is related to the volume of soil involved in the test, which is much larger than in
borehole methods. As a result, surface methods are particularly useful if the average properties
of a soil deposit are to be assessed as in the case of ground response analyses.
8
In the following, the discussion on surface methods will focus exclusively on Rayleigh
wave methods mainly because of their relevance in near-surface site characterization.
It is well known that, in the case of an elastic isotropic medium, the shear wave and
Rayleigh wave velocities are correlated as follows, depending on the Poisson ratio:
0.862 1.14
VR Vs (8)
1
Early surface wave methods employed laborious field procedures to measure the
dispersion curve (i.e. a plot of Rayleigh phase velocity vs. frequency) and crude inversion
techniques to obtain the S-wave profile from the experimental dispersion curve [35]. Stokoe
and his co-workers (i.e. [36][37]) re-invented engineering surface wave testing by taking
advantage of portable dynamic signal analysers, to efficiently measure the dispersion curve,
and of the widespread availability of high-speed computers, to implement theoretically-based
robust inversion algorithms. Actually, the Spectral Analysis of Surface Waves (SASW) method
was replaced by the Multichannel Analysis of Surface Waves (MASW) which uses an array of
geophones instead of only two. MASW method uses either impulsive sources such as hammers,
steady-state sources like vertically oscillating hydraulic or electro-mechanical vibrators that
sweep through a pre-selected range of frequencies, typically between 5 and 200 Hz [38], as
well as passive source [39-41]. R-waves are detected by a geophone array. Usually 24
geophones with an inter - geophone distance X in between 1.5 and 5 m are used (Figure 5).
The source can be external (Figure 5) or internal to the geophone array. Different impulse, with
different frequency content, can be generated in order to sample different subsoil depths. The
signals at the receivers are digitised and recorded by a dynamic signal analyser.
The use of a multi-station testing setup can introduce several advantages in surface wave
testing. In this case, the motion generated by an impact source is detected simultaneously at
several receiver locations and the corresponding signals are analysed as a whole (i.e. in both
the time and space domains) using a double Fourier Transform. It can be shown [42] that the
composite dispersion curve can be easily extracted from the location of the spectral maxima in
the frequency-wavenumber domain in which the original data are transformed. Using this
technique, the evaluation of the experimental dispersion curve becomes straightforward;
furthermore, the procedure can be easily automated [43].
The experimental dispersion curve is used to obtain the shear wave velocity profile via a
process called inversion. A theoretical dispersion curve is calculated for an assumed vertically
heterogeneous layered soil profile using one of several available algorithms [44-49]. The
theoretical dispersion curve is then compared with the corresponding experimental curve and
the “distance” between the two curves is used as a basis of an iterative process consisting of
updating the current soil profile until the match between the two curves is considered
satisfactory. The soil profile may be updated manually by trial and error or using an automated
minimisation scheme based on an unconstrained or constrained inversion algorithm [8]. When
a satisfactory agreement between theoretical and experimental dispersion curves is attained
(Figure 6a), the final shear wave velocity profile (Figure 6b) is taken as representative of the
site conditions.
9
Seismograph
Impulsive
Source
1 2 3 n
D X X
Figure 5 MASW configuration
6 00 ex p erim e ntal
la st iteratio n
pha se velocity, m/s
5 00 first iteratio n
4 00
3 00
2 00
1 00
0 10 20 30 40 50 60 70
fre qu en cy, Hz
Figure 6a Theoretical and experimental dispersion curve
shear wave velocity, m/s
100 200 300 400 500 600 700
0
10
depth ,m
15
20
CH test
25
SASW test
30
Figure 6b Shear wave velocity profile
For a successful application of MASW testing, it is recommended to observe the
following guidelines:
in choosing the relative spacing between source and receivers, attention should be
placed to minimize near-field effects and spatial aliasing. In this context, the near-
field is defined as a region close to the source where the magnitude of the body wave
10
components of the wave field are of comparable magnitude to the surface wave
components. Efforts should be made to eliminate or minimize near-field effects
unless they are explicitly accounted for during the inversion process [50][51]. In
normally dispersive media, the body wave field is significant until D/ exceeds about
0.5, hence the nearest receiver should be located at least one-half wavelength from
the source:
D / 2 (9)
This recommendation is consistent with other studies of the influence of near-field
effects, but more strict requirements are necessary for inversely dispersive
stratigraphies [52][40]. It is also important to limit the distance between receivers
to avoid spatial aliasing, a simple criterion is given by:
X / 2 (10)
the length of the receiver array must be sufficiently large, if the stiffness profile at
great depth has to be estimated. A rule of thumb is that the survey length must be as
long as about 3 times the maximum depth of interest. This requirement may not be
compatible with the space available at the site. Moreover, massive sources are
needed to get good quality signals with long testing arrays, causing an increase of
testing time and cost;
it is important to account for multiple modes of surface wave propagation, especially
in irregular, inversely dispersive soil profiles [53][40]. Currently several approaches
are used to account for multiple modes. Individual, modal dispersion curves can be
calculated and compared with the experimental dispersion curve during the inversion
process. Unfortunately, the use of only two receivers in the traditional SASW
method prohibits resolving individual modes in the experimental dispersion curve;
only the effective velocity representing the combination of several modes can be
determined. Also using a multi-station approach the individual modes cannot be
separated if a relatively short receiver array is used, as required by engineering
practice [54]. Thus, it must be assumed that the experimental curve represents an
individual mode, usually the fundamental mode. This approach is satisfactory only
in normally dispersive profiles. Another approach is to calculate the effective
velocity directly and use it as the basis of the inversion. Lai and Rix (1998)[8] have
developed an efficient procedure based on the normal mode solution to calculate the
effective velocity as well as closed-form partial derivatives required for inversion.
Finally, it is possible to numerically simulate the SASW test using Green’s functions
that calculate the complete wave field [50]. This approach is computationally
expensive, in part because the partial derivatives must be calculated numerically, but
it accurately models the actual field procedure used in SASW tests;
for the inversion of the experimental dispersion curve, it is essential to use
theoretically-based inversion algorithms. Prior to the widespread availability of
high-speed computers, simple empirical inversion techniques were used.
Furthermore, in recent years, there have been attempts to develop simple methods
based on parametric studies and regression equations. These methods have limited
usefulness and are likely to yield erroneous results. It is remarked that the rapidly
increasing power of personal computers makes it possible to use theoretically-based
inversion methods routinely;
the non-linear inversion of the experimental dispersion curve is inherently ill-posed
with the consequence that the solution (i.e. the S-wave profile) is not unique. This
11
problem can be overcome with the recourse of two strategies [8]. First, a priori
information about the soil profile can be used to limit the range of possible solutions.
Second, additional constraints such as smoothness and regularity (e.g. [55]) may be
imposed on the solution.
3 ENGINEERING PARAMETERS
Soil parameters for Civil Engineering applications are mainly (Vs)30 and impedance
functions. The (Vs)30 parameter was suggested by Ordaz and Arciniegas in 1992 [56] to
account for stratigraphic amplification. Many Technical Codes, including Eurocodes adopt
such a parameter to define the response spectrum (i.e. seismic action) at the soil – deposit
outcrop, in a simplified way. Such an approach is applicable in the case of simplified
stratigraphic profiles. Eurocode 8 part 1, defines two shapes of the response spectrum (low and
moderate seismicity) at rock outcrop (reference spectra – class A). Moreover, Eurocode 8
defines four classes (B C, D, E) in terms of simplified stratigraphic profiles and of (Vs)30 range.
Different shapes of the response spectra and different amplification factors (S) are prescribed
for these classes of seismic subsoil. Eurocode 8 also adopts an importance factor but does not
consider simplified approaches to account for topographic amplification.
The Italian Building Code [57] (NTC 2008) represents the Italian implementation of
Eurocodes. Some peculiarities are worth noticing.
The response spectra are defined, for each prescribed exceedance probability within a
reference period (i.e. for a given return period), starting from site-dependent parameters.
In particular, the reference period is inferred from the life-time and importance of the
considered construction/structure, while the return period is obtained by the following
equation:
PR
TR (11)
ln1 pL
Where PR is the reference period and pL is the exceedance probability.
Therefore, such a procedure implicitly incorporates an importance factor.
The site-dependent parameters are listed below and were obtained at the nodes of a square grid
of 0.05° size, covering the whole Italian territory. The seismic hazard parameters were obtained
by using a probabilistic approach:
ag = maximum free-field acceleration for a given return period and for a rigid reference
site, with horizontal topographical surface;
F0 = maximum spectral amplification factor for a rigid reference site, with horizontal
topographical surface (the minimum value for F0 is 2.2);
TC* = is used to determine the period above which the spectral velocity is constant.
The elastic response spectrum shape is then defined according to the following expressions:
T 1 T
S e ( T ) a g S F0 1 0 T TB (12a)
TB F0 TB
S e ( T ) a g S F0 TB T TC (12b)
T
Se ( T ) ag S F0 C TC T TD (12c)
T
12
T T
Se ( T ) ag S F0 C 2D TD T (12d)
T
Where S (=SS x ST) is the result of the product among two coefficients that take into
account for the subsoil site class (SS coefficient – stratigraphic amplification) and for the
topographic conditions (ST coefficient – topographic amplification); TC (=TC* x CC) is the
corrected period at which the spectral acceleration initiates to decrease and above which the
spectral velocity is constant. The corrected period is obtained as the product between TC* and
the coefficient CC (CC depends on the subsoil site class), TB (=TC/3) is the period above which
the spectral acceleration is constant and TD (= 4.0 x ag +1.6) is the period above which the
spectral displacement is constant.
The ST coefficient can assume values of 1.0 – 1.2 – 1,4. The SS coefficient is computed
according to equations of the following type:
S S 1.4 0.4 F0 a g 1.2 (13)
(for class B subsoil. Different numerical coefficients and upper limit are prescribed for different
soil classes).
The subsoil classes (Table 1) are identified on the basis of the a simplified profile and of
the parameter Vs30, that is computed in the following way.
30
Vs 30 (14)
hi
V
s ,i
Where hi and Vs,i are respectively the thickness of the generic sub-layer i and the
corresponding shear wave velocity.
Table. 1 Subsoil site classes according to the Italian Building Code (NTC 2008) [57]
Site Class Vs30[m/s]
A Vs30 800
B 360 Vs30 < 800
C 180 Vs30 < 360
D Vs30 < 180
E Subsoil of class C or D with thickness 5- 20 m, overlying class A bedrock
The simplified procedure is not applicable in the case of complex stratigraphic profiles
or irregular topographic conditions.
In these cases, the response spectrum is inferred from seismic response analyses which
require the knowledge of the shear wave velocity profile, as well as the dynamic characteristics
of the subsoil (at least shear modulus and damping ratio). In addition an appropriate selection
of accelerograms on rock outcrop is necessary (see as an example [58] and [59]).
The impedance is defined as the ratio between the harmonic excitation (force/moment)
to the resulting displacement/rotation. As force and displacement are not in phase, the
impedance is a complex expression depending on frequency. For a massless single degree of
freedom system the impedance can be written in the following way:
K ( ) k i c (15)
13
Where: = circular frequency; c = viscosity coefficient.
In practice the problem reduces to the definition of the real (static) part. More generally,
a 6x6 impedance matrix should be defined in order to account for different types of motion
(vertical translation, torsion, sliding/rocking).
The problem can be simplified in the following way:
neglect the embedment i.e. the sidewall contact [60];
neglect the so called trench effect [60];
assume a circular foundation of equivalent perimeter and radius (R).
Therefore the vertical stiffness is given by the following equation:
4G R
kv (16)
( 1 )
Where: G = shear modulus; R = equivalent radius of the foundation; = Poisson ratio
(assume 0.2).
The shear modulus can be inferred from the average shear wave velocity of the soil
volume relevant for the considered problem. As for the non – linearity it could be possible to
refer to Eurocode 8 part 5:
Go Vs2 G / Go 0.8 – 0.5 – 0.35
when respectively ag = 0.1 – 0.2 – 0.3 (17)
More accurate estimate of impedance can be obtained by referring to the Gazetas (1983)
[60] recommendations. In any case an estimate of the static stiffness is necessary (i.e. of shear
wave velocity from seismic in situ tests).
On February 2017 SCPT were conducted at the campus of the University of Pavia. At
the same location Down – Hole test (DH) was performed inside a borehole. P and S wave
measurements from these two types of tests were compared. Moreover SCPT measurements
were used to obtain the small strain damping ratio.
According to the borehole stratigraphic log, the first 3 meters consist of man – made soil
of various nature (very hard and coarse material). The natural deposit consists of a sequence of
sands with different percentages of silt and gravel. Ground water table was located at 12.2 m
depth from ground level.
SCPT were performed by means of a Pagani TG63 – 200 penetrometer. As for the first
3 meters, a preliminary dynamic penetration was carried out (a sort of pre-boring).
Two different types of source were used for SCPT:
a sledgehammer of 10 kg with a special anvil (drop height of about 1.8 m)
a manual hammer of 5 kg. This hammer was used to hit the aluminium blocks
from right or left. The blocks were kept well in contact to the soil by the
penetrometer - legs.
In principle, the first type of source mainly produced PV and SV waves while the second
type of source manly produced PV and SH waves. In any case, the generated wave field is
usually quite complex.
The seismic module of SCPT was equipped with a pair of triaxial accelerometers. The
relative distance between the accelerometers was 0.5 m. Therefore, two waveforms were
14
recorded for each hit by the data acquisition system. Test interpretation was carried out by
means of the cross-correlation method. In other words, the travel time between the two
accelerometers was computed from the time delay which maximised the cross – correlation
function between the two recorded waveforms. Butterworth filters were applied only to the
shear wave signals.
Figure 7 SCPT set-up at the University of Pavia campus. A) Pagani TG63 – 200
penetrometer; B) manual hammer of 5 kg hitting the aluminium blocks; C) sledgehammer.
A 3D geophone pack was used for DH test together with a manual hammer. First arrival
time of P and S wave was obtained by manual pick-up on the dromocrome plot.
Figure 8 Waveform records (SCPT at Pavia site): a) sledge –hammer; b) manual
hammer
Figure 8 shows typical examples of recorded waveforms (SCPT). Figure 8a shows the
simultaneous measurements of the two accelerometers. In this case the sledge hammer was
used as source. Figure 8b instead shows the waveform recorded by a single accelerometer when
the manual hammer is used as source. In this case the polarity inversion of shear wave is shown.
15
Figure 9 SV vs. SH
Figure 9 compares the shear wave velocities obtained, at the same depth, by using
different sources (sledge hammer and manual hammer). The values are comparable and suggest
that, for the considered case, the effects of anisotropy are negligible. Indeed it was postulated
that by using the sledgehammer SV are generated, while the manual hammer mainly generates
SH.
The measured S and P wave velocities are compared in Figure 10a. Figure 11 shows the
dromocrome from DH and SCPT. The comparison has been done for depths in between 3 and
16 m. Indeed it was not possible to measure P wave at greater depths by SCPT because of the
too low signal to noise ratio.
Figures 10 and 11 can be commented as follows:
Vs from the two types of measurements are comparable;
Vp are also comparable;
Vp (from both SCPT and DH) increase for depth greater than 11 m (i.e. below
the water table depth) and remains quite constant with values ranging in between
1700 – 1900 m/s. It could be argued that, at depth greater than 11 m, the P wave
velocity of sound in water was measured.
The spectral slope method was used to determine the small strain damping ratio from
SCPT. In particular only S wave signals were used. The success of the method mainly depended
on the following experimental aspects:
a single hit for generating the two signals;
use of accelerometers, instead of geophones, with an increased repetitiveness of signal
amplitude measurement.
16
Figure 10 a) P and S wave velocities (SCPT ad Down-Hole data) and b) Damping ratios
(SCPT data) at the University of Pavia campus
17
Figure 11a Dromocrome (DH)
Figure 11b Dromocrome (SCPT)
Figures 12a to 12c show the Spectral Ratio (Napierian Logarithm scale) vs. the frequency
for different frequency intervals. Data shown in Figures 12a to 12c refer to the measurements
at 9.5 m depth.
Figure 12a ln(SR) vs. Frequency (natural frequency interval)
18
Figure 12b ln(SR) vs. frequency (frequency interval corresponding to the maximum Fourier
Spectrum)
Figure 12c ln(SR) vs. frequency (full frequency range)
19
Figure 13 shows the Fourier spectra of the two waveforms.
A damping ratio of 2.1 % was obtained by considering a frequency interval in between 0
and 15 Hz. Such a frequency interval contains the natural frequency of the tested soil. Values
of damping ratio ranging in between 2.1 and 4 % were obtained. Higher values of the damping
ratio were obtained by considering the frequency interval containing the maximum of the
response spectra. Under these conditions damping ratio values in between 1.3 and 7.9 % were
obtained (see Figure 10b) for the two series of damping values with depth). These results
suggest that damping ratio values increase with frequency (i.e. Maxwell type damping).
5 CONCLUDING REMARKS
The paper reviewed the capabilities of seismic tests in the light of Eurocodes. More
specifically, the advantages and capabilities of in situ seismic piezocone tests (SCPT) were
shown in general terms and with specific reference to a comparative experimental study case.
Experimental results suggest that a lower noise to signal ratio could be obtained by using
geophones. Indeed, DH tests are carried out by using 3D geophones. Seismic module of
piezocone could be equipped both with accelerometers and geophones. Obviously, in the case
of the seismic piezocone small size instruments should be selected. The use of accelerometers
gives the advantage of repetitive measurements of amplitude which, in turn, offers a better
interpretation in terms of attenuation.
REFERENCES
[1] Kramer, S.L. (1996) Geotechnical Earthquake Engineering, Prentice Hall, Inc., Upper
Saddle River, New Jersey, 653 pp.
[2] Woods R.D. 1978 Measurement of dynamic soil properties. Earthquake Engineering and
Soil Dynamics, Pasadena CA. 1: 91-179.
20
[3] Biot, M.A. (1956a) Theory of Propagation of Elastic Waves in a Fluid-Saturated Porous
Solid.”, I. Lower Frequency Range, J. Acoust. Soc. Am., 28: 168-178.
[4] Biot, M.A. (1956b) Theory of Propagation of Elastic Waves in a Fluid-Saturated Porous
Solid. II. Higher Frequency Range, J. Acoust. Soc. Am., 28; 179-191.
[5] Rix, G.J., Lai C.G., Spang A.W. Jr (2000) In situ measurement of damping ratio using
surface waves, J. Geotech. and Geoenvir. Eng., ASCE, 126: 472-480.
[6] Ishihara, K. (1996) Soil Behaviour in Earthquake Geotechnics. Oxford Science
Publications, Oxford, UK, pp. 350.
[7] Aki K., Richards P.G. (1980) Quantitative seismology: theory and methods - 2 vol.,
Freeman, S. Francisco.
[8] Lai, C. G. and Rix, G. J. (1998) Simultaneous inversion of Rayleigh phase velocity and
attenuation for near-surface site characterization. Technical Report GIT-CEE/GEO-98-
2, Georgia Institute of Technology.
[9] Lai, C.G., Pallara O., Lo Presti, D.C.F., and Turco, E. (2001) Low-Strain Stiffness and
Material Damping Ratio Coupling in Soils. Advanced Laboratory Stress-Strain Testing
of Geomaterials. TC29 Summary Book. 15th ICSMGE, Istanbul, Turkey. August 27-31,
2001. Ed. Tatsuoka, F., Shibuya, S. and Kuwano.
[10] Rix, G.J., Lai C.G., Foti S. (2001) Simultaneous measurement of surface wave dispersion
and attenuation curves, accepted for publication Geotech. Testing J. 24(4): 350–358.
[11] Lai C.G., Rix G.J., Foti S., Roma V. (2002) “Simultaneous Measurement and Inversion
of Surface Wave Dispersion and Attenuation Curves”, Soil Dynamics and Earthquake
Engineering, 22, 9-12: 923-930.
[12] Nigbor R.L. & Imai T. (1994) The Suspension P-S Velocity Logging Method. Proc. XIII
ICSMFE New Delhi, India TC # 10, pp. 57-61.
[13] Auld B. (1977) Cross-Hole and Down-Hole Vs by Mechanical Impulse. Journal GED,
ASCE. 103 (GT12): 1381-1398.
[14] Stokoe K.H. II & Hoar, R.J. (1978) Variable Affecting In Situ Seismic Measurements.
Proc. of the Conference on Earthquake Engineering and Soil Dynamics, ASCE,
Pasadena, CA.2: 919-939.
[15] Woods, R.D. and Stokoe, K.H. II (1985) Shallow Seismic Exploration in Soil Dynamics.
Richart Commemorative Lectures, ASCE. 120-156.
[16] Woods, R.D. (1991) Field and Laboratory Determination of Soil Properties at Low and
High Strains. Proc. of 2nd International Conference on Recent Advances in Geotechnical
Earthquake Engineering and Soil Dynamics, St. Louis, Missouri, SOA1.
[17] Woods, R.D. (1994) Borehole Methods in Shallow Seismic Exploration. Proc. of XIII
ICSMFE New Delhi, India TC # 10: 91-100.
[18] Patel, N.S. (1981) Generation and Attenuation of Seismic Waves In Downhole Testing.
M. Sc. Thesis GT81-1, Department of Civil Engineering, University of Texas at Austin.
[19] Jamiolkowski M. and Lo Presti D. (1998) Geotechnical Characterization of Gravelly
Deposits. 13th SEAGC, Taipei.
[20] Lo Presti D., Pallara O., Jamiolkowski M., Cavallaro A. (1999) Anisotropy of small strain
stiffness of undisturbed and reconstituted clays. Proc. IS Torino 99, Balkema, 1: 3-10.
21
[21] Stokoe, K.H. II, Lee, J.N.K. & Lee, S.H.H. (1991) Characterization of Soil in Calibration
Chambers with Seismic Waves. Proc. ISOCCT1, Postdam, NY
[22] Lo Presti, D.C.F. & O'Neill, D.A. (1991) Laboratory Investigation of Small Strain
Modulus Anisotropy in Sand. Proc. of ISOCCT1, Postdam, NY.
[23] Bellotti R., Jamiolkowski M., Lo Presti D.C.F. & D.A. O'Neill (1996) Anisotropy of
Small Strain Stiffness in Ticino Sand. Geotechnique. 46(1): 115-131.
[24] Jamiolkowski M. & Lo Presti D.C.F. (1991) Anisotropy of Soil Stiffness at Small Strain.
Panel Discussion to Session 1, 9th ARC on SMFE, Bangkok, Tailandia.
[25] Mitchell, J.K., Lodge, A.L., Coutinho, R.Q., Kayen, R.E., Seed, R.B., Nishio, S. &
Stokoe, K.H. II, (1994) In Situ Test Results from four Loma Prieta Earthquake
Liquefaction Sites: SPT, CPT, DMT and Shear Wave Velocity. EERC Report No.
UCB/EERC - 94/04.
[26] Fioravante V., Jamiolkowski M., Lo Presti D.C.F., Manfredini G. & Pedroni S. (1998)
Assessment of coefficient of earth pressure at rest from shear wave velovity
measurements. Géotechnique, 48(4): 1-10.
[27] Mok, Y.J. (1987) Analytical and Experimental Studies of Borehole Seismic Methods.
Ph.D. Thesis, Univ. of Texas at Austin, Austin, TX.
[28] Fuhriman, M.D. (1993) Cross-Hole Seismic Tests at two Northern California Sites
Affected by the 1989 Loma Prieta Earthquake. M. Sc. Thesis, The University of Texas
at Austin.
[29] Redpath, B.B., Edwards, R.B., Hale, R.J. & Kintzer, F.C. (1982) Development of Field
Techniques to Measure Damping Values for Near Surface Rocks and Soils. Prepared for
NSF Grant No. PFR-7900192.
[30] Redpath B.B. & Lee R.C. (1986) In-Situ Measurements of Shear-Wave Attenuation at a
Strong-Motion Recording Site. Prepared for USGS Contract No. 14-08-001-21823.
[31] Khawaja A.S. (1993) Damping Ratios from Compression and Shear Wave
Measurements in the Large Scale Triaxial Chamber. M. Sc Thesis, The University of
Texas at Austin.
[32] Campanella, R.G. & W.P. Stewart (1990) Seismic Cone Analysis Using Digital Signal
Processing for Dynamic Site Characterization. 43rd Canadian Geotechnical Engineering
Conference, Quebec City.
[33] Puci I & Lo Presti D.C.F. (1998) Damping Measurement of Reconstituted Granular Soils
in Calibration Chamber by means of Seismic Tests. Submitted to the 11 European
Conference on Earthquake Engineering.
[34] Reynolds J.M. (1997) An introduction to applied and environmental geophysics. J. Wiley
& sons, New York
[35] Jones R.B. (1958) In-situ measurement of the dynamic properties of soil by vibration
methods, Geotechnique, 8 (1): 1-21
[36] Nazarian S. (1984) In situ determination of elastic moduli of soil deposits and pavement
systems by Spectral-Analysis-of-Surface waves method, PhD Diss., Un. of Texas at
Austin
22
[37] Stokoe KH II, Mok Y.S., Lee N. & Lopez R. (1989) In Situ Seismic Methods. Recent
Advances in Testing, Understanding and Applications. Proc. XIV CGT, Politecnico di
Torino, Department of Structural Engineering.
[38] Rix, G.J. (1988) Experimental Study of Factors affecting the Spectral Analysis of Surface
Waves Method. Ph. D. Thesis. The University of Texas at Austin, 315 pp.
[39] Horike M. (1985) Inversion of phase velocity of long-period microtremors to the S-wave-
velocity structure down to the basement in urbanized areas”, J. Phys. Earth, 33: 59-96
[40] Tokimatsu K. (1995) Geotechnical Site Characterisation using Surface Waves. Proc. IS
Tokyo 1995, Balkema, 1333-1368
[41] Zywicki D., Rix G.J. (1999) Frequency-wavenumber analysis of passive surface waves,
Proc. Symp. on the Appl. of Geophysics to Environm. and Eng. Problems, Oakland, pp.
75-84
[42] Tselentis G.A., Delis G. (1998) Rapid assessment of S-wave profiles from the inversion
of multichannel surface wave dispersion data, Annali di Geofisica, 41: 1-15
[43] Roma V., Hebeler G., Rix G., Lai C.G.: (2002) Geotechnical soil characterization using
fundamental and higher Rayleigh modes propagation in layered media , XII European
Conference on Earthquake Engineering, London 9-13 Settembre 2002
[44] Haskell N.A. (1953) The dispersion of surface waves on multilayered media, Bullettin of
the Seismological Society of America, 43 (1): 17-34
[45] Thomson W.T. (1950) Transmission of elastic waves through a stratified solid medium”,
J. Applied Physics, 21 (1): 89-93
[46] Kausel E., Roesset J.M. (1981) Stiffness matrices for layered soils, Bullettin of the
Seismological Society of America, 71 (6): 1743-1761
[47] Chen X. (1993) A systematic and efficient method of computing normal modes for
multilayered half-space, Geophys. J. Int., 115: 391-409
[48] Hisada Y. (1994) An efficient method for compunting Green's functions for a layered
half-space with sources and receivers at close depths, Bullettin of the Seismological
Society of America, 84 (5): 1456-1472
[49] Hisada Y. (1995) An efficient method for compunting Green's functions for a layered
half-space with sources and receivers at close depths (part 2), Bullettin of the
Seismological Society of America, 85 (4): 1080-1093
[50] Roësset, J.M., Chang, D.W., Stokoe, K.H. II, (1991) Comparison of 2-D and 3-D Models
for Analysis of Surface Wave Tests, 5th International Conference on Soil Dynamics and
Earthquake Engineering, Karlsruhe, Germany, pp. 111-126.
[51] Ganji, V., Gucunski, N., and Nazarian, S., (1998) Automated Inversion Procedure for
Spectral Analysis of Surface Waves, Journal of Geotechnical and Geoenvironmental
Engineering, 124(8): 757-770.
[52] Sànchez-Salinero I. (1987) “Analytical investigation of seismic methods used for
engineering applications”, PhD Diss., Un. of Texas at Austin
[53] Gucunski N., Woods R.D. (1991) Use of Rayleigh modes in interpretation of SASW test,
Proc. 2th int. Conf. Recent Advances in Geot. Earthq. Eng. and Soil Dyn.-S.Louis, pp.
1399-1408
23
[54] Foti S., Lancellotta R., Sambuelli L., Socco L.V. (2000) Notes on fk analysis of surface
waves Annali di Geofisica, 43(6): 1199-1210
[55] Constable S.C., Parker R.L., Constable C.G. (1987) Occam's inversion: a pratical
algorithm for generating smooth models from electromagnetic sounding data,
Geophysics, 32 (3): 289-300
[56] Dobry R. & Iai S. (2000) Recent Developments in Understanding of Earthquake Site
Response and Associated Seismic Code Implementation. Geoeng 2000, Melbourne, 19-
24 November 2000. Vol. 1 (Invited Papers): 186-219.
[57] Norme Tecniche per le Costruzioni, NTC (2008). Italian Building Code. DM, 14, 2008.
[58] Lo Presti, D. C., Lai, C. G., & Puci, I. (2006). ONDA: Computer code for nonlinear
seismic response analyses of soil deposits. Journal of geotechnical and
geoenvironmental engineering, 132(2): 223-236.
[59] Lo Presti D.C.F., Stacul S. (2017) - ONDA (One-dimensional Non-linear Dynamic
Analysis): ONDA 1.4 User's Manual Version 1.4. DOI: 10.13140/RG.2.2.32409.83043
[60] Gazetas, G. (1983). Analysis of machine foundation vibrations: State-of-the-art, Soil
Dynamics and Earthquake Engrg., 3(1): 2-42.
24