Att. I - Geophysics Report.pdf

AVELLINO DEPARTMENT

Via Campo di Fiume, 13 – 83030 Montefredane (AV) Tel.: +39 0825 24353 - Fax.: +39 0825 248705

SOCOTEC ITALIA Srl – P.Iva 01872430648 - Capitale sociale 7.144.000,00 euro Sede Legale: Via Bariola, 101-103 - 20020 Lainate (MI) Tel.: +39 02 9375 0000 - Fax: +39 02 9375 0099 www.socotec.it

Seismic investigations and GPR surveys - Area A1, Aviano

Air Base, Italy

Emissione Data Elaborato Verificato Approvato

Rev. 01 30/06/2020

Paola Luiso Donato Fiore Massimo De Iasi

Marco Cardamone

Pietro Boni

INFRASTRUCTURE

Index

1. Introduction

2. Seismic refraction (SRT)

2.1 Methodology and basic principles of the method

2.2 Influence factor of the wave velocity

2.3 Equipment

2.4 Description of geophysical investigation methods

2.5 Acquisition scheme and data processing

3. Ground Penetrating Radar (GPR)

3.1 Methodology and basic principles of the method

3.2 The propagation of electromagnetic waves

3.2.1. Electromagnetic waves reflection

3.2.2. Electromagnetic waves penetration

3.2.3. Propagation velocity

3.3 Equipment

3.4 Acquisition scheme and processing

4. Results

4.1 Geolithological framework

4.2 Refraction seismic interpretation key

4.3 Refraction seismic results

4.4 Soil type classification

4.5 Seismo-stratigraphyc model

4.6 Site model

4.7 GPR survey interpretation

5. Bibliography

1. Introduction

The following report provides the results of the geophysical investigations conducted in a portion of Area

A1 at Aviano Air Base, Friuli-Venezia Giulia, Italy on June 2020 (Figure 1). The geophysical investigations included the execution of 9 seismic prospections with P waves and S waves and a Ground Penetrating

Radar (GPR) surveys. The purpose of the surveys was the geotechnical characterization of the shallow underground materials and local utilities location.

Fig. 1 - Surveys location of the investigated area.

Table 1 below summarizes the coordinates of the geophysical survey lines conducted:

Line ID Length (m) UTM – WGS84 33T UTM – WGS84 33T

SRT1 120 312915.00 mE 5104323.00 mN 312970.00 mE 5104431.00 mN

SRT2 96 312900.00 mE 5104336.00 mN 312907.00 mE 5104431.00 mN

SRT3 96 312873.00 mE 5104425.00 mN 312875.00 mE 5104328.00 mN

SRT4 72 312900.00 mE 5104324.00 mN 312921.00 mE 5104390.00 mN

SRT5 72 312880.00 mE 5104370.00 mN 312951.00 mE 5104380.00 mN

SRT6 72 312876.00 mE 5104419.00 mN 312937.00 mE 5104455.00 mN

SRT7 72 312881.00 mE 5104554.00 mN 312922.00 mE 5104496.00 mN

SRT8 60 312870.00 mE 5104496.00 mN 312870.00 mE 5104436.00 mN

SRT9 120 312838.00 mE 5104442.00 mN 312883.00 mE 5104553.00 mN

Table 1 - Summary and coordinates of the geophysical surveys.

The main scope of the seismic refraction survey is supporting the new Area A1 Access Control Point (ACP) engineering design. The surveys allowed the reconstruction of the shallow subsurface geology from the stratigraphic and geotechnical point of view. Ground penetrating radar (GPR) is an electromagnetic method used for shallow applications. GPR works on electromagnetic properties (dielectric permittivity, electrical conductivity and magnetic permeability) in order to define the physical boundaries of shallow structures (anthropic and natural features) of the subsurface materials.

2. Seismic refraction (SRT)

2.1 Methodology and basic principles of the method

According to ASTM D 5777 – “Standard Guide for Using the Seismic Refraction Method for Subsurface

Investigation” standard, the Method defines the analysis as: “Measurements of the travel time of a compressional (P) and Shear (S or Sh) waves from a seismic source to a geophone(s) are made from the land surface and are used to interpret subsurface conditions and materials. This travel time, along with distance between the source and geophones, is interpreted to yield the depth to refractors (refracting layers). The calculated seismic velocities of the layers are used to characterize some of the properties of natural or artificial subsurface materials”. Thus, the purpose of a seismic geophysical study is to measure

P and Sh waves velocity values as well as to calculate dynamic properties of geological deposits constituting the investigated volumes, defining their geometrical and stratigraphical relationships. To properly reconstruct the site ground model, geophysical surveys are normally planned concurrently with other exploration techniques (i.e. boreholes, trenches, pits, etc.), to allows a complete cross analysis of collected data. The velocities assumed by the P and S waves when propagating in a medium will depends both on the medium physical properties (stiffness, density, saturation) as well as by its homogeneity degree.

There are two type of seismic body waves, P waves and S waves (also known respectively as Primary and

Secondary waves), which propagates through materials with different velocity. Primary waves (P-waves) are compressional body waves that pass through any type of material (including fluids). Furthermore, these waves travel faster than other seismic waves; for this reason, P waves are the first detected at seismograph stations. Secondary waves (S-waves) are shear waves detected at seismograph stations after the arrivals of the P-waves. These waves propagate only through solid materials as fluids do not support shear stresses and, in addition, they can also show polarization phenomena, being classified as Sh-waves when polarized in the horizontal plane and Sv-waves if polarized in the vertical plane. Each of these different types of waves, induces different motions to the particles constituting the medium through which the waves are travelling.

Seismic refraction uses acoustic waves generated by a hammer or by weight drops (or some other comparable method) able to put impulsive energy into the ground. The detection of seismic wavefronts generated in a known point (source point), is done by positioning an acquisition array on the surface, consisting of many transducers (geophones) able to detect such waves arrival. The “first break time” at each geophone is registered by a recording unit called “seismograph”. From the first break times and source-geophone distances, the seismic velocities for each of the different crossed layers can be determined analyzing the first break times curves at each transducer (travel-time curve). Analyzing the time paths and hence the seismic waves velocity, can be traced back to the geometrical, mechanical and elastic characteristics of lithotypes in the investigated areas.

The present seismic refraction surveys were performed for the acquisition of P and Sh waves through a multi shots energization.

The wavefronts can be of any shape or have spherical surfaces. However, the Huygens principle

(fundamental in the study of any disturbance wave) states that "at any time, any point of a wave front of any shape can be considered as a point source of circular waves (or if the perturbation is spherical in three dimension)”. Waves propagation can be interpreted as a continuous generation of circular waves which interfere with each other giving rise to a resulting wave macroscopically observable. The rays normal to the wave fronts are called "seismic rays" and indicate the direction of the motion propagation. When a ray reaches a heterogeneity on its path, such as a lithological contact, the incident ray is transformed into several new rays. The angles that the incident rays, the reflected rays and refracted rays create with the normal to the contact surface between two materials are related to the speeds of propagation within these two mediums and described by Snell's law.

There is a particular value of the incidence angle called "critical angle" particularly significant for seismic refraction because the refracted ray propagates along the separation surface of the two mediums ( =

90°) and acts as a source for a new wavefront and new secondary rays (Fig. 2). Is possible to determine the critical angle by calculating the inverse of the sine of the value of the speed of the first layer / second layer velocity.

Fig. 2 - Illustrative scheme of seismic waves

Reporting in a simple x-y graph, for each distance of each receiver, the arrival time of the first impulse read on the seismograms, the points are arranged in a series of alignments and the first passes through the origin: each line is called "travel-time curve" and is representative of the propagation velocity of the elastic waves in the medium above a refractor (a separation surface between two materials with different elastic behavior). The angular coefficient of the travel-time curve passing through the origin represents the inverse of the propagation velocity of the elastic waves in the first layer (V0); the angular coefficients of the other travel-time curve represent the inverse velocity of the underlying layers (from the most superficial to the deepest). The intersection between two travel-time curves is called "knee point" and represents the point where the waves refracted by two adjacent refractors, or the direct waves and the waves refracted by the first refractor, arrive simultaneously. In a refraction seismic survey, it will be possible to identify a number of “knee points” equal to the number of refractors investigated (Fig.3)

Fig. 3 - Example scheme of the travel-time curve in a 2-layer subsoil.

If the layers are not very thick and characterized by increasing velocity with depth (as normally occurs in unconsolidated soils), the wave paths will be curved, with concavity facing upwards, and also the corresponding travel-time curve will be a curve. In real cases, it is assumed that the effect of direct waves, in terms of vibrations amplitude, is predominant compared to that of indirect waves, so much to be able to neglect the latter and consider the medium, limited to the area comprising the source and the receiver, as a homogeneous, isotropic and non-confined.

Seismic refraction survey provides with velocities trends of compressional P-waves crossing the materials constituting the subsoil and these velocities are good indicators of soil or rock type. Each type of soil or rock has a wide range of P wave velocities (Fig. 4a) and albeit often these ranges overlap each other, they are generally greater for:

Denser rocks than lighter rocks;

Older rocks than younger rocks;

Igneous rocks than sedimentary rocks;

Solid rocks than rocks with cracks or fractures;

Moreover, seismic refraction survey provides also with velocities trends of shear S-waves crossing the materials constituting the subsoil. This is the most commonly used parameter in shallow geophysics, measured along with the Vp, for calculating dynamic parameters in the elastic range of soil behavior. The measurement of shear properties of the solid skeleton is allowed, as it is the only one absorbing shear strengths unlike the fluids (water, air or other) in which shear stresses are not transmitted. With reference to geotechnical investigations shear wave velocities provide information on low-strain dynamic properties of a given material, each of which present typical S wave velocities range as reported in Figure 4b below

Fig. 4 – P and S Wave velocity in different Lithology & Material (Stanford Rock Physics Laboratory-Gary Mavko)

2.2 Influence factor of the wave velocity

The first experiments on the variation of the rocks elastic properties began between 1950 and 1960. At that time many researchers (Wyllie et al., 1956; 1962; 1963; 1985) studied the changes in wave velocity elasticity as a function of saturation, porosity, clay content and fracturing. In detail:

• Saturation: in conditions of total saturation, the velocity of the P waves is controlled by the liquid

(which is incompressible), if it is more than the solid skeleton because the velocity of the S waves is not affected by this parameter (the water does not support shear). For partial saturations, the velocity of the

P waves is lower than the one obtainable at maximum saturation but, at the same time, it is also lower than the velocity obtained for dry conditions.

• Porosity and clay content: the velocity decrease is associated with an increase in the clay content and can be compared with the decrease in velocity associated with an increase in porosity. The clay particles generate a microporosity composed of small pores of 1 µm, considerably reducing the pore size.

A large part of the velocity variation due to the change in clay content is ascribable to the significant amount of microporosity present in clay minerals: a clear change in the properties of the rock matrix occurs when a small amount of clay is added.

• Fracturing: the fracturing degree of a rock formation is linked to the rock’s hydrogeological properties, such as porosity and permeability. Some studies highlight a decrease in the compression waves (VP) and shear velocities (VS) and an increase in the coefficient of attenuation α with increasing fracture density. Thus, the P and S wave velocities decrease with increasing fracture parameters, while the Vp/Vs ratio increases with decreasing fracture parameters.

Thus, the seismic velocities measured on rocks or land will often be influenced by the simultaneous effect of all factors.

2.3 Equipment

The equipment used for this type of survey consists of the following parts:

Energizing system;

Reception system;

Trigger;

Data acquisition system.

In detail:

Energizing system: this system must be able to generate high-frequency elastic waves with high energy.

A 5 kg impact weight was used to generate the P and S waves.

Reception system: for the survey, an electromagnetic geophones with vertical oscillation moving coil and a characteristic 10 Hz frequency, were used. Electromagnetic geophones with horizontal oscillation moving coil and a characteristic 8 Hz frequency, were also used (Fig. 5b and 5c).

Trigger: consists of an electrical circuit that is closed when the energization starts. The generated pulse, sent to the acquisition system, allows to set the reference zero time: t0 for the calculation of the path times of the generated waves.

Data acquisition system: the GEODE seismograph produced by GEOMETRICS (Figure 5d), was used. It has an acquisition resolution equal to 24 bits (Delta Sigma Technology). The seismograph’s other technical characteristics are summarized in Table 2. This system can digitally record on each channel, the waveforms and to store them on a minimum dynamic 24-bit mass memory. It is connected to each geophones and to the trigger sensor, recording the vibrations starting from the impulse sent by the trigger on a specific monitor and display them in the form of traces.

Fig. 5 - Geophones and seismograph; a) simplified scheme of the geophone for P wave; b) geophone for P wave with characteristic frequency of 4.5 Hz; c) geophone for Sh wave with characteristic frequency of 8 Hz; d) GEODE seismograph (Geometrics).

A/D Conversion: 24 bit (Crystal Semiconductor sigma-delta converters)

DynamicRange: 144 dB (system), 110 dB (instantaneous, measured) at 2ms, 24dB.

Distortion: 0.0005% @ 2 ms, 1.75 to 208 Hz.

Bandwidth: 1.75 Hz to 20 kHz

Common Mode Rejection: > -100 dB at <=100 Hz, 36 dB.

Crosstalk: -125 dB at 23.5 Hz, 24 dB, 2 ms.

Noise Floor: 0.2uV, RFI at 2 ms, 36 dB, 1.75 to 208 Hz.

Stacking Trigger Accuracy: 1/32 of sample interval.

Maximum Input Signal: 2.8 VPP.

Input Impedance: 20 kOhm, 0.02 uf.

Preamplifier Gains: dB, or 0dB. 24 or 36 dB

Anti-alias Filters: -3 dB at 83% of Nyquist frequency, down 90 dB.

Pre-trigger Data: Up to 4,096 Samples.

Sample Interval: 0.02, 0.03125, 0.0625, 0.125, 0.25, 0.5, 1.0, 2.0, 4.0, 8.0, 16.0 ms.

Record Length: 16,000 samples standard

Delay: 0 to 9999 ms in 1 ms steps.

Data Transmission: Ethernet connections

Table 2 - Technical characteristics of GEODE seismograph.

2.4 Description of geophysical investigation methods

The main steps of the seismic refraction survey can be summarized as follows:

Acquisition:

o On site set up of the seismic profiles: following preparatory activities on site (tracks preparation, vehicles clearing, etc.) proceed to set up all the equipment required to perform the acquisition

(seismographs, geophones, seismic cables, energizing sources, etc.), positioning it with a GPS according the planned line acquisition array.

o Data recording: all the seismic signals are acquired according to the technical specifications for the profile’s acquisition (profile length, coverage, energizing shots interdistance, inter-geophonic distance, etc.). Testing of the S/N ratio to ensure data accuracy was also conducted. The data were recorded in a storage unit.

o Location survey: all geophones and shots point coordinates were then positioned within the correct coordinate system: WGS 84.

Pre - Processing o Quality control: the first operation was to check the acquired data accuracy in term of signal/noise ratio and amplitude.

o Pre-processing: consisting in all the operation of staking and filtering of the raw data. All the acquired signals for each shot position, were summed up (staking) in order to increase the signal (constant) while reducing the noise (random).

Processing o Project definition: all the data were loaded in the processing software database making sure of defining all the geographical coordinates (x, y, z) for each shot position and for each geophone.

o Picking: detecting for each seismic trace of each shot the “first break” of the seismic waves. The data are then plotted in a time–distance graph from which the velocities and the relative depths of different layers crossed by the seismic waves were calculated.

o Tomographic Inversion: The first break data were then used for tomographic inversion; then the seismic sections were divided in tomograms (cells with dimensions depending on survey geometry and with a single velocity value assigned), and a ray-tracing software was used to iteratively calculate velocity values for each cell.

2.5 Acquisition scheme and data processing

The seismic surveys collected both P and S waves. For each of the 9 acquisition, 24 channels (geophones) were used (Fig. 6). All spreads were characterized by:

- Acquisition time: 1 s;

- Sampling interval: 0.125 μs.

The 9 tomographies were acquired as follows:

Fig. 6 – acquisition phases of SRT investigations

The data analysis was divided into two phases: Pre-processing and processing.

Pre-processing

During fieldworks acquisitions the main noise was represented by wind, vehicle passing and electromagnetic interference. For these reasons, all the data referred to a single shot location, are subjected to the stacking process to increase the signal/noise ratio. For the S waves, polarity inversion was checked for each point.

Processing

The first processing operation was the survey geometry setting with all topographic information defined for each receiver (geophone) and shot’s position. All the seismic traces were checked to detect the first break of the seismic waves. The inversion processing was carried out using the RAYFRACT Software

(Intelligent Resources Inc., Canada), allowing the processing of the surface seismic survey data, both with

P and S waves. RAYFRACT allows both the reconstruction of the geometry of the refractors with the traditional refraction seismic restitution, and the creation of detailed models of underground velocities with the most advanced tomographic techniques mainly used for deep structures. The interpretation method used is based on the creation of an initial model obtained using first break times processing based on the "wave fronts" (Brückl 1987; Jones and Jovanovich, 1985) and the Plus-minus (Hagedoorn 1959) methods. These methods are based on a regression of the first break times (Brückl 1987) and can be considered an optimization algorithm of the GRM. A calculation code for the tomographic modeling, using a "ray-tracing" algorithm with curved rays and least squares calculation based on the Delta-tv method

(Gebrande and Miller 1985), was used.

3. Ground Penetrating Radar (GPR)

3.1 Methodology and basic principles of the method

The GPR method is based on the reflection of electromagnetic waves (e.m.) in the shallow subsoil. A transmitting antenna introduces pulses of electromagnetic energy of very short duration (nanoseconds) into the investigated material. A receiving antenna detect each reflection of the e.m. signal caused by the material change.

The Georadar system can normally be equipped with antennas with a central frequency from 40 to 2,500

MHz and acquisition band from 10 to 5,000 MHz, reaching variable depths depending on the type of material and the specific antenna used. Furthermore, the position transducer (optical encoder) allows precise locations of the detected anomalies on the ground.

3.2 The propagation of electromagnetic waves

Through the propagation of electromagnetic waves in the soil, the following parameters are evaluated:

• Reflection;

• Attenuation;

• Penetration;

• The propagation velocity.

3.2.1. Electromagnetic waves reflection

An electromagnetic wave reflection occurs when the propagation medium shows a variation (or inhomogeneity) of the electromagnetic characteristics, as happens in correspondence with a change in the medium or presence of a target. The reflection capacity of a target is expressed by the S.E.R. (Radar

Equivalent Surface), whose evaluation is very complex, being a function of multiple parameters:

• target shape;

• target size;

• target material;

• background materials;

• frequency and polarization of the e.m. wave;

• target distance;

The attenuation of the electromagnetic wave, whose energy decreases progressively with increasing depth, is the main cause of limitation of the investigation depth; it is expressed in decibels/meter and originates from various effects:

• the conductivity of the medium;

• resonance phenomena of water molecules, or other molecules present in the subsoil;

• the geometric attenuation: during propagation, the electromagnetic energy is distributed on a wave front whose surface increases as you move away from the antenna, effectively decreasing the overall energy that affects single targets.

3.2.2. Electromagnetic waves penetration

The electromagnetic wave penetration is closely related to attenuation; it is highly dependent on the soil nature, thus on its conductivity. There are soils characterized by low conductivity and good homogeneity, in which the penetration can be very high (up to tens of meters), such as compact rock, ice, dry sand, etc.

However, in other common soil types (clayey, sandy and gravel nature), the presence of water and the salts dissolved in such soils limits the penetration to a few meters; in particular, the penetration can be very reduced for water-saturated soils.

3.2.3. Propagation velocity

The propagation velocity estimate is essential to correctly measure the targets depth; it is a function of the dielectric constant of the medium, and can vary by almost 10 times depending on the nature of the soil; the propagation velocity is expressed in m/s; the propagation velocity in vacuum (and in air) is equal to 3 x 108 m/s. For GPR survey the propagation velocity is often expressed in cm/ns. Typical propagation velocity for soil are around 7-12 cm/ns. In this case, an average velocity of 10 cm/ns was chosen.

3.3 Equipment

For the utilities mapping two types of antenna were used. The investigation depth was about 2 m b.g.l.

Two different antenna types were used for the survey:

The STREAM C antenna, allowing a complete coverage of the area, was used on asphalt areas;

The Hi-Mod 200-600 MHz antenna (Fig.7), easily to manage, was used on green areas. It enables a high resolution survey of shallower depths, using 600 MHz antennas, while guaranteeing a great depth range with the 200 MHz. This array of antenna is used to:

Correlate information coming from several data channels;

Accurately reconstruct the path of underground utilities in complex situations;

Ensure accuracy and productivity.

The presence of two200 MHz and 600MHz antennas in the same set helps the operator to locate pipes and cables by providing the proper frequency for a specific search parameter. This provides a real-time display of deep and shallow antennas on the same screen.

Fig. 7 – 200/600 MHz antenna

STREAM C, is an innovative georadar system produced by IDS Italia (Fig.8): In general, the STREAM

C is able to speed up and improve:

The identification of potential war devices;

The mapping of the concrete thickness of the bituminous conglomerate;

Mapping of underground utilities;

STREAM C is made of a system of 32 antennas (600 MHz): 24 type H and 8 type V, equipped with position sensors making not necessary to work at the same distance from the surface to be investigated; this allows to reach an investigation depth of about 3.5 m (in optimal propagation conditions) and to obtain a very rapid and accurate dataset. The H-V antennas configuration, with a dedicated electric power supply, allows, at the same frequency, a greater penetration than the old radar instruments; moreover, the two different dipole orientations allow to detect objects of any geometry and orientation more effectively.

Fig. 8 – GPR STREAM C

3.4 Acquisition scheme and processing

GPR survey was conducted through a series of parallel lines; the regular link was not followed if there were green areas or obstacles, such as sidewalks or buildings. The GPR lines were spaced by 1 m each in order to obtain total coverage of the volume with the road surface as top; the acquisitions were georeferenced through the use of precision GPS and total station.

The stages of acquisition (paved surfaces and green areas) through georadar methodology are shown in

Fig. 9.

Fig. 9 – GPR acquisition

The immediate result of the investigation for each individual survey line, is a continuous section of the subsoil/investigated material, in which the x-axis corresponds to the path followed by the antenna on the surface (in meters), and the y-axis to the time of round trip necessary for the electromagnetic waves to reach horizons or reflective objects. The reflection intensities of the signal are rendered with a color

"palette" (Fig. 10) whose gradation is proportional to the intensity of the signal itself. The interpretation requires the determination of the average velocity of the electromagnetic waves in the soils/materials investigated, and the normalization of the scans compared to the horizontal meters travelled along the lines (in this case automatic because the antenna is equipped with a precision optical encoder).

Fig. 10 - Radar amplitude display scale

The interpretation, that is the identification of the depth and exact location on the map at an appropriate scale, requires the determination of the average velocity of the electromagnetic waves in the soils/materials investigated and the normalization of the scans compared to the horizontal meters travelled along the lines. The georadar data acquired on two-dimensional sections are processed through a special calculation program (GRED 3D - IDS, Italy).

GPR data, displayed directly after acquisition, generally contains "noise" or "sound tail (reverberations)" or "interference (extraneous reflections)". To eliminate noise and correct the horizontal and vertical scales of the raw data, pre-processing it is necessary.

The following pre-processing filters are applied to the acquired data:

Move start time: eliminates the portion of the signal between the air and the investigated medium;

Background removal: removes continuous components along the horizontal direction of the radargram;

Vertical and horizontal bandpass filters: attenuate the frequencies of each sample constituting the radargram, outside a chosen window;

Smoothed gain: equalizes the power along a radargram for an assigned shift window.

It should be noted that the georadar measure, in general, is affected, working in optimal conditions, by an instrumental error of 2% in the targets depth evaluation ; in some cases, presence of welded mesh acts as a disturbing factor to the natural propagation of the signal. In this instance, a precaution error of

± 5% is used.

4. Results

The seismic refractive investigations were carried out to characterize the investigation area, to obtain information on the characteristics and geometries of the subsurface materials. Geophysical surveys were also interpreted thanks to the information deriving from soil boring investigations, as shown in Figure 1.

A detailed GPR survey was conducted to identify and locate underground utilities.

4.1 Geolithological framework

Before proceeding to the description and the interpretation of the geophysical surveys, an excerpt of the area geological map (1:100.000 scale - ISPRA, sheet n° 24) is shown in Fig. 11. From a structural point of view, the area is not affected by fault and folds; however, the study area is characterized by recent fluvioglacial sediments.

Fig. 11 – Geological map of the area with its legend. The red borders indicate the study area.

The Quaternary deposits are characterized by stratum debris and fluvioglacial sediments. The deposits consist mainly of coarse and loose materials, limestones or dolomitic limestones, with extended lenses of clayey sands. Sometimes, there are impermeable clay layers to which resurgence areas relate. Below the

Quaternary deposits, a Cenozoic sequence is found. It is characterized by grey silty marls, with alternation of arenites and grey silts with heterometric conglomerates. A succession of calcareous breccias and conglomerates with fossil remains follows underneath, with presence of coarse glauconite arenites.

The basement is made of Mesozoic deposits characterized by stratified limestones with frequent emergence structures, sometimes with breccias, residual clays and stromatolites, followed by massive limestone strata.

4.2 Refraction seismic interpretation key

As showed in Figure 4, seismic velocity has a high variability due to the many influence factors involved.

Fractures, saturation and porosity can also occur together to modify velocity in different ways. This behavior causes a large overlap of velocity range for different lithologies. This can be useful, together with a direct survey, to understand some general characteristic for a particular lithology ( for example: high velocity for a sand may mean some cementation and a low velocity for a rock con be related to an high level of fractures).

In the discussion of the seismic results below, for each seismic line it is described the velocity distribution that, together with bibliographic references information and soil borings data, has been used to provide a stratigraphic model of the study area.

4.3 Refraction seismic results

Refraction seismic survey show a general distribution of velocity for the survey area. The maximum velocity reached 3,000 m/s, for P waves, and 1,500 m/s, for S waves (see Attachment 1 - Graphs). Such velocities were found at a depth between 25 and 30 m below ground level (b.g.l.). Soil borings S1, S4, and

S7 were used to limit the obtained results to 10 m b.g.l..; bibliography data were used to investigate the area below that depth (see Section 5).

All the seismic survey shows a gradual increase in velocity with depth, although very high velocity values are never achieved. The plane-parallel of layers’ stratification is sometimes interrupted by an undulation of the layers, as shown in Drawings 2 and 3 of Attachment 1. A simple sketch is shown on Fig. 12.

Fig. 12 - Example of layer depth change. In this Figure it is shown the P wave at SRT1.

In the first few meters from ground surface, the tomographic images show particularly low P and S wave velocity values (1,600 m/s and 600 m/s, respectively), in accordance with the completed soil boring stratigraphy, with presence of sandy and silty gravel until about 10 m b.g.l.. Below that depth, the seismic survey shows a gradual increase of seismic velocity suggesting either an intensification in the sandy and silty gravel compaction degree, or a change in lithology. The absence of boreholes deeper than 10 meters b.g.l. did not allow to confirm this hypothesis.

Details of each seismic line are provided in the paragraphs below:

SRT 1: the seismic tomography in P and S waves (Attachment 1 - Drawings 2 and 3) reaches a depth of about 30 – 33 m b.g.l.. The tomography show an increase in the seismic velocities (P and S) with depth. A marked increasing of seismic velocities is present at about 10 m of depth, where the P velocity increases from 1,200 to 1,800 m/s and the S velocity from 600 to 800 m/s. This increase is less noticeable in the first meters of the seismic survey, from 0 to 40 m from the starting point of the investigation, suggesting the presence of a non-homogeneous thicknesses of the shallow layer. The first meters from ground level are characterized by the presence of sandy silty gravel, as confirmed by soil boring S1. Below 10 m b.g.s., the higher P and S velocity values suggest the presence of a different lithology. Based on geological map in Figure 11, it is possible to speculate the presence of the Scaglia Rossa Formation, characterized by a low compaction degree.

SRT 2: the seismic tomography in P and S waves (Attachment 1 - Drawings 4 and 5) reaches a depth of about 25 – 28 m b.g.l.. The seismic velocities are characterized by slow increasing with depth.

As shown in paragraph 2.2, this suggests the presence of high porosity gravel. Also, the seismic tomography shows that the layers are plain-parallel and, at about 15/16 m b.g.l., the presence of the Scaglia Rossa Formation can be inferred.

SRT 3: the seismic tomography in P and S waves (Attachment 1 - Drawings 6 and 7) reaches a depth between 25 to 30 m b.g.l.. The seismic results are very similar to the SRT2 ones: the P wave velocity increases faster with depth than the S wave velocity, reaching 2,400 m/s. This suggests the presence of high porosity gravel. The presence of more compacted lithology, such as the Scaglia

Rossa Formation, is again present below 28 m b.g.l..

SRT 4: the seismic tomography in P and S waves (Attachment 1 - Drawing 8), reaches a depth of about 13 to 15 m b.g.l.. The seismic surveys show a very low values of P and S velocities (P reaching

1,400 m/s and S velocity 500 m/s); this can be related to a very low compacted sandy gravel.

SRT 5: the seismic tomography in P and S waves (Attachment 1 - Drawing 9), reaches a depth of about 20 – 25 m b.g.l.. The seismic surveys show an increase of the seismic velocities of both P and S and the presence of plain-parallel layers. Also, in the first 5 m b.g.l., the seismic tomography shows the existence of two areas (the red ellipses in Drawing 9), characterized by high compaction degree.

SRT 6: the seismic tomography in P and S waves (Attachment 1 - Drawing 10), reaches a depth of about 13 – 15 m b.g.l.. The seismic surveys confirm the presence of very low values of P and S velocities (1,700 m/s and S 700 m/s, respectively), suggesting the presence of sandy gravel characterized by a very low compaction degree.

SRT 7: the seismic tomography in P and S waves (Attachment 1 - Drawing 11), reaches a depth of about 17 to 20 m b.g.l.. The seismic surveys confirm the presence of very low values of P and Sh velocities (1,600 m/s and 750 m/s, respectively), suggesting the presence of sandy gravel characterized by a very low compaction degree. Also, in final portion of the seismic tomography investigated depths, between 40 and 70 m, an increase in the thickness of the shallower materials is noticeable; considering the isoline Vs = 600 m/s, its depth rises from 3 to 13 m from point A to point B of the survey.

SRT 8: the seismic tomography in P and S waves (Attachment 1 - Drawing 12), reaches a depth of about 12 – 15 m b.g.l.. The seismic surveys confirm the presence of very low values of P and S velocities (1,200 m/s and 750 m/s, respectively), suggesting the presence of sandy gravel characterized by a very low compaction degree.

SRT 9: the seismic tomography in P and S waves (Attachment 1 - Drawings 13 and 14), reaches a depth of about 30 to 37 m b.g.l.. When compared to the other seismic profiles, SRT 9 shows the highest values of S waves recorded: 1,400 m/s at 36 m b.g.l.. This could suggest the existence of a more compact portion of Scaglia Rossa Formation at that depth.

4.4 Soil type classification

A series of MASW investigations were conducted along each refraction seismic line to define the soil type according to the NTC_2018 guidelines. The MASW method identifies S wave velocity by measuring the surface waves arriving at each geophones on the surface. S velocity waves are obtained from the Rayleigh ones. MASW surveys provide a 2D Vs model attributable to the seismic line central point as follows:

MASW 1 (SRT1): the Vs values increase from 370 to 530 m/s at about 9 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 530 to 1,200 m/s) is identified at about 26 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 448 m/s.

MASW 2 (SRT2): the Vs values increase from 370 to 490 m/s at about 10 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 490 to 890 m/s) is identified at about 25 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 378 m/s.

MASW 3 (SRT3): the Vs values increase from 440 to 540 m/s at about 11 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 540 to 860 m/s) is identified at about 24 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 483 m/s.

MASW 4 (SRT4): the Vs values increase from 340 to 530 m/s at about 10 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 530 to 950 m/s) is identified at about 24 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 404 m/s.

MASW 5 (SRT5): the Vs values increase from 400 to 650 m/s at about 11 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 650 to 890 m/s) is identified at about 23 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 490 m/s.

MASW 6 (SRT6): the Vs values increase from 480 to 600 m/s at about 13 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 600 to 880 m/s) is identified at about 22 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 510 m/s.

MASW 7 (SRT7): the Vs values increase from 470 to 610 m/s at about 10 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 610 to 920 m/s) is identified at about 23 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 535 m/s.

MASW 8 (SRT8): the Vs values increase from 470 to 640 m/s at about 12 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 640 to 890 m/s) is identified at about 23 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 531 m/s.

MASW 9 (SRT9): the Vs values increase from 460 to 710 m/s at about 8 m depth, suggesting an increase of the gravel compaction degree. The greatest Vs increase (from 710 to 1200 m/s) is identified at about 28 m b.g.l., suggesting the lithology change from gravel to the Scaglia rossa formation. The soil category is B and the Vseq is 481 m/s.

Table 3, below, provides a summary of soil categories and their Vs equivalent:

ID Vseq (m/s) Soil category

SRT 1 448 B

SRT 2 378 B

SRT 3 483 B

SRT 4 404 B

SRT 5 490 B

SRT 6 510 B

SRT 7 535 B

SRT 8 531 B

SRT 9 481 B

Table 3 - Summary of the soil categories and Vs equivalent.

4.5 Seismo-stratigraphyc model

Using a velocity log of Vp and Vs (values deriving from the seismic refraction survey), located in the middle of each seismic line, a seismo-stratigraphyc model was developed for a better description of the local soil properties. From the average velocity value of each layer, the dynamic module for the site was also calculated.

The relationship among compressional and shear waves velocity and density were used for describing the in-situ dynamic properties of the investigated geological deposits. Compressional and shear wave velocities can be derived from the seismic surveys, while the density can be obtained from correlations, in-situ tests or from laboratory tests.

In particular, the parameters are:

Young's modulus or linear elasticity modulus (E), corresponding to the ratio between the stress applied to a body and the relative change in length.

Shear modulus or stiffness modulus (G), given by the ratio between shear stress and angular deformations. It expresses the ability of the body's resistance to change shape; it’s value is zero in water.

Incompressibility or bulk (K), expressed as the ratio between the total pressures and the relative volume changes.

Density ρ is calculated with an empirical formula ρ= Vp*0.002+16, derived from the Gardner relation. Gardner et al. (1974) conducted a series of empirical studies and determined the following relationship between velocity and density:

where α is = 0.31 when V is in m/s and β is 0.23.

This formula, widely used in oil exploration, is well suited for soil with average compaction. However, it strongly underestimate the soil density in case of loose or surface soils. For this reason, it is preferable using its Modified Version. However, in the specific case of the study area, both formulas provide very similar results.

In the table below (Table 4), summarizes the physical tie between all the involved values.

PARAMETER

MEASURE

UNIT

Vp Compressional wave velocity m/s

Vs Shear wave velocity m/s E Young's Modulus MPa G Shear Modulus MPa K Bulk Modulus MPa

Density (in situ) kN/m^3

PARAMETER

FORMULA

Shear wave velocity G = Vs2

Young's Modulus E=2Vs2(1+) Bulk Modulus K = (Vp2 -4/3)Vs2

Table 4 - Determining modules and ratio for typical velocities of earth materials from seismic survey

The following table summarizes all the values calculated for the investigated site:

Site n° layer Base depth (m)

P Waves S Waves

Poisson Young (MPa) Share (MPa) Bulk (MPa) (kN/m3)

NTC 2018

Soil type

Velocity (m/sec)

Velocity (m/sec) site 1 layer 1 3 697.76 280.70 0.40 384.73 137.06 664.19 17.40

B layer 2 9 930.93 369.95 0.41 687.53 244.46 1222.02 17.86 layer 3 26 1880.23 652.10 0.43 2405.96 840.29 5865.43 19.76 layer 4 30 2502.81 1205.95 0.35 8240.98 3054.90 9084.79 21.01 site 2 layer 1 3 470.90 227.71 0.35 236.72 87.84 258.55 16.94

B layer 2 10 681.50 318.74 0.36 479.80 176.40 571.21 17.36 layer 3 25 1994.95 513.98 0.46 1546.71 528.08 7251.49 19.99 layer 4 30 2580.45 825.72 0.44 4163.74 1442.77 12166.76 21.16 site 3 layer 1 3 630.40 342.00 0.29 521.46 201.89 416.76 17.26

B layer 2 11 876.39 370.98 0.39 679.63 244.32 1037.76 17.75 layer 3 24 1679.85 561.37 0.44 1753.57 610.09 4649.64 19.36 layer 4 30 2170.02 843.64 0.41 4085.22 1447.66 7647.91 20.34 site 4 layer 1 3 520.37 299.87 0.25 383.51 153.23 257.13 17.04

B layer 2 10 641.85 353.10 0.28 552.96 215.49 424.71 17.28 layer 3 24 1236.74 529.01 0.39 1435.17 516.98 2136.27 18.47 layer 4 30 1588.01 855.47 0.30 3636.32 1403.37 2964.61 19.18 site 5 layer 1 3 746.77 354.15 0.35 594.56 219.41 683.00 17.49

B layer 2 11 999.68 399.14 0.41 805.86 286.75 1416.46 18.00 layer 3 23 1866.27 598.33 0.44 2038.32 706.41 5930.88 19.73 layer 4 30 2327.06 857.43 0.42 4316.83 1518.46 9160.00 20.65 site 6 layer 1 3 664.07 408.85 0.19 692.14 289.65 377.96 17.33

B layer 2 13 797.04 453.11 0.26 911.18 361.22 636.07 17.59 layer 3 22 1115.46 596.76 0.30 1687.39 649.24 1402.71 18.23 layer 4 30 1660.02 870.15 0.31 3834.31 1462.85 3373.48 19.32 site 7 layer 1 3 500.15 289.52 0.25 355.69 142.50 235.26 17.00

B layer 2 10 698.35 405.22 0.25 711.98 285.66 467.55 17.40 layer 3 23 1003.05 630.48 0.17 1679.75 715.74 857.28 18.01 layer 4 30 1461.61 913.02 0.18 3722.92 1577.45 1939.31 18.92 site 8 layer 1 3 497.97 309.24 0.19 385.58 162.54 204.75 17.00

B layer 2 9 730.27 401.22 0.28 721.69 281.07 556.41 17.46 layer 3 23 1111.83 619.80 0.27 1784.55 700.06 1319.31 18.22 layer 4 30 1532.49 892.93 0.24 3778.97 1520.10 2450.66 19.06 site 9 layer 1 4 680.99 327.27 0.35 502.03 185.96 557.20 17.36

B layer 2 8 839.69 394.58 0.36 747.76 275.26 879.54 17.68 layer 3 27 1383.15 702.33 0.33 2455.43 925.67 2355.98 18.77 layer 4 30 1824.78 1230.38 0.08 6444.17 2974.62 2576.81 19.65

SITE

AVERAGE

layer 1 3 601.043 315.479 0.292 450.714 175.565 406.089 17.202

B layer 2 10 799.523 385.114 0.333 699.821 263.402 801.304 17.599 layer 3 24 1474.613 600.460 0.360 1865.204 688.063 3529.887 18.949 layer 4 30 1960.805 943.855 0.304 4691.497 1822.463 5707.148 19.922 Table 5 – Seismo-stratigraphyc model value

The orange portion of the table presents the average results for the site.

4.6 Site model

The silty gravel down to 10 m b.g.l. represents the quaternary cover characterized by fluvioglacial sediments, as shown in Fig. 11. This formation, as shown in Attachment 1 - Drawing 9, sometimes contains lens of materials characterized by different compaction degree. The first 3 meters clearly show a different behavior due to a different compaction degree. Below 10 m b.g.l., the formation was not clearly identified from a litho-stratigraphic point of view. The velocity value suggest a compact sediment or a highly fractured rock. Considering the dynamic of deposition of the fluvioglacial sediments, it can be assumed that this could be a transitional zone, where deposition and erosion processes occurred. The material of this layer can be assumed being a mix of both.

From bibliography and seismic data, it can be assumed that below 24-25 m b.g.l. the “Scaglia Rossa

Formation” can be found. It is characterized by stratified limestones with frequent raised structures, sometimes with breccias, residual clays and stromatolites.

According to the NTC 2018, the soil category identified using the MASW surveys, is B.

4.7 GPR survey interpretation

The GPR survey results are shown in a separate AutoCAD map. Figure 13 shows Area A1 and the investigated area.

The quality of the signal was partially compromised due to the soil humidity from the intense and extended rain that occurred immediately before survey; nevertheless, the utilities identification and location was successfully achieved.

Fig. 13 – Aviano Air Base Area A1. The red line highlights the investigated area.

The utilities interception was often interrupted due to the presence of construction sites, shrubs, or mud, as shown on Fig. 14.

Fig. 14 – Examples of obstacles encountered during the acquisition.

The utilities identified by the STREAM C and the 200/600 antennas are shown on the CAD file using lines with different colors: magenta lines for the utilities identified using the 200/600 antenna; green and cyan lines for the utilities identified using the STREAM C antenna according to the following:

Green lines identify utilities located at a depth greater than 10 cm b.g.l.;

Cyan lines identify shallow utilities located at a depth less than 10 cm b.g.l..

The CAD file shows the location of manholes, soil borings and the areas that could not be investigated due to various obstacles (Fig. 14). Figure 15 provides the legend used in the CAD file.

Fig. 15 – Legend used in the CAD file

The main gaps in the utilities surveys were:

Natural gas pipe: detected just for a few meters from Building 108 to Building 110. It is possible that the pipe deepens and the presence of very damp soil did not allow to further detect its path.

Water main line: located on the west side of the natural gas pipe. It has been impossible to follow its direction due to the presence of the construction site, as shown in Fig. 14.

Natural gas pipe located on the east side of Building 103: the pipe was located in the muddy area where a manhole was present (Fig. 14). For this reason, further identification was not possible.

Communication lines located from Building 106 to Building 175: the lines are part of an area characterized by trees and shrubs. For this reason, work was not possible in those areas.

Electrical line located from Building 106 to Building 175: the line is located in a driveway with reinforced concrete and rebars outcropping that prevented the proper acquisition in this area.

Water service lines along Buildings 108 and 103: the radargrams did not identify such lines, probably due to the great soil humidity or their depths.

Sewer from Building 108 to Building 110: the radargrams did not identify such lines, probably due to the great soil humidity or their depth.

Signals for some of the identified pipes were lost when trying to connect them to their respective manholes due to the absence of radar waves reflection because of the presence of large quantities of water inside the manholes.

Additional details are provided in the CAD file attached.

5. Bibliography Biot M. A., 1956. Theory of propagation of elastic waves in a fluid-saturated porous solid. Journal of the Acoustic

Society of America, 28, 2, 168-191.

Biot M. A., 1962. Mechanism of deformation and acoustic propagation in porous media. Journal of Applied Physics, 33, 4, 1482-1498.

Brückl E., 1987. The interpretation of travel-time fields in refraction seismology. Geophysical prospecting, 33, 9, 973-992. https://doi.org/10.1111/j.1365-2478.1987.tb00855.x.

Gebrande H., Miller H., 1985. Refraktionsseismik. In: Angewandte Geowissenschaften - Band II, Bender, F., ed.:

226–260. Stuttgart: F. Enke Verlag.

Hagedoorn J. G, 1959. The plus-minus method of interpreting seismic refraction sections. Geophysical Prospecting, 7, 2, 158-182. https://doi.org/10.1111/j.1365-2478.1959.tb01460.x.

Jones G.M. and Jovanovich D.B., 1985. A ray inversion method for refraction analysis. Geophysics, 50, 11, 1701- 1720. https://doi.org/10.1190/1.1441861.

Wyllie M. R., Gregory A. R. & Gardner G. H., 1956. Elastic wave velocities in heterogeneous and porous…