Manual on Subsurface Investigations (2019)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

35 C H A P T E R 4 Geophysical Investigations Introduction Geophysical investigations are used to estimate the physical properties of the subsurface by measuring, analyzing, and interpreting seismic, electrical, electromagnetic, gravitational, and magnetic fields measured at the ground surface or within boreholes. Investigations conducted from the ground surface typically provide information about the subsurface both laterally and to some depth; while most of the borehole investigations, with some exceptions, provide detailed information about subsurface materials only in the immediate vicinity of the borehole or between boreholes. In special applications, it is also possible to gather geophysical measurements of the earth over water or from airborne platforms. Geophysical methods are often useful during the initial phases of a site investigation program to efficiently gain an understanding of the overall subsurface conditions, including stratigraphy and the location and size of potential anomalies. The locations of subsequent borings and soundings can then be optimized to investigate areas of concern identified from the geophysical surveys. Geophysical methods are also useful to estimate the engineering properties of subsurface materials directly. For example, seismic methods may be used to directly measure the shear wave velocity (Vs) profile of near-surface soil and rock formations, which is required to determine the Site Class for seismic analyses (AASHTO 2011). Direct measurements of the electrical resistivity of soils via geophysical methods are useful for evaluating potential corrosion of steel pile foundations (Decker et al. 2008). Compared to more traditional forms of subsurface exploration (i.e., borings and soundings), geophysical methods offer several advantages (Wightman et al. 2003, Anderson et al. 2008, AASHTO 2017): • Because surface geophysical methods are noninvasive, they provide the ability to cover a large area in a time- and cost-effective manner to gain an understanding of the overall subsurface conditions. As noted above, this enables optimizing the locations of borings and soundings during subsequent phases of a subsurface exploration program or interpolating between existing borings and soundings. This is particularly true for linear projects (e.g., highway construction). • Geophysical methods are robust in the sense that they are based on fundamental physical principles with relatively little reliance on empiricism. In many cases, the methods used for geotechnical applications leverage the extensive experience gained with similar methods developed for resource (e.g., oil, gas) exploration. • Surface geophysical methods are also useful for sites where borings and soundings are difficult or impractical, such as gravel deposits or contaminated soils. The equipment used for many geophysical tests is highly portable, which may allow testing at sites that are not easily accessible (e.g., a heavily wooded area) using conventional drilling equipment. It is also important to recognize the limitations of geophysical methods to avoid using them in situations where there may be a low probability of success: • Geophysical methods are more likely to yield good results when (i) there is a large contrast in seismic, electrical, electromagnetic, gravitational, or magnetic properties between lithologic units or between an anomaly and the surrounding soils and rocks, and (ii) the subsurface features of interest are of sufficient size relative to their depth that they are within the limits of detection for a particular geophysical method.

36 • The interpreted subsurface conditions may not be unique for many geophysical methods; there may be multiple, physically plausible interpretations for the stratigraphy or location and size of anomalies that all yield the same measured geophysical response. For example, a structural low in bedrock topography; a small, air-filled void in the bedrock; or a larger, water-filled void in the bedrock may all produce the same magnitude of gravity anomaly. • Sites that have a stiff, surficial layer overlying a weaker layer or an electrically resistive layer over a conductive layer pose a challenge for many surface geophysical tests. For example, many seismic methods do not work well on concrete pavements because of the large stiffness of the pavement compared to the base and subgrade materials. In part because of these limitations and in part because geophysical methods are less familiar to many geotechnical engineers than conventional site investigation methods (e.g., SPT, CPT), it is essential that geophysical investigations be conducted by personnel who are trained and experienced in near-surface geophysics. Finally, in all but a few applications (e.g., reconnaissance-only investigations), the results of geophysical investigations should always be complemented by direct observation of subsurface conditions by means of borings, soundings, test pits, trenches, outcrops, and other geological information. This ground truth information will help ensure that interpreted subsurface conditions derived from geophysical methods are as accurate as possible. The combined use of a geophysical investigation with direct observation is a robust approach to developing an accurate ground model for a project. Planning a Geophysical Investigation There are a number of surface and borehole geophysical methods available to choose from when planning a geophysical site investigation. Selecting one or more methods for a project can be guided by answering the following questions (Anderson et al. 2008): • What are the physical properties of interest? • Which methods respond to the physical properties of interest? • Which methods can provide the required levels of detection and resolution for the subsurface features of interest? • Which methods can perform well given conditions at the project site? • Which methods provide complementary information to help improve interpretations based on the observed data? • What direct observations (e.g., borings or soundings) should be performed to constrain the interpretation of geophysical data? • Which methods are most cost effective, and is the overall geophysical investigation cost effective? Anderson et al. (2008) provide additional discussion and examples for each of these considerations. Surface Geophysical Methods This section summarizes surface geophysical methods—seismic, electrical, electromagnetic, and potential-field methods—that can be used for a variety of subsurface-investigation objectives for transportation-related projects as shown in Table 4-1. Further guidance is provided in ASTM D6429. For many applications, more than one method may be capable of achieving the objective. In these cases, an experienced geophysicist can evaluate what methods have the highest likelihood of success for a particular project based on site-specific conditions.

37 Table 4-1. Matrix of surface geophysical methods in relation to investigation objectives Objective Seismic Electrical and Electromagnetic Potential Field R ef ra ct io n an d R ef le ct io n Su rf ac e W av e R es is tiv ity El ec tr om ag ne tic G ro un d- Pe ne tr at in g R ad ar M ic ro gr av ity M ag ne to m et ry Se lf- Po te nt ia l Lithology and stratigraphy      Bedrock topography        Water table    Rippability  Shear wave velocity profile  Fault detection      Void and cavity detection       Subsurface fluid flow   Ferrous anomalies    Conductive anomalies     Corrosion potential  Sources: Fenning and Hasan (1995), USACE (1995), Sirles (2006), FHWA (2006), Anderson et al. (2008) There are many readily available references that describe these geophysical methods in detail (e.g., USACE 1995, Wightman et al. 2003, Sirles 2006, Anderson et al. 2008), and additional information is available in the ASTM guides and standards listed in Table 4-2. The focus in this section is to briefly summarize the key features of the surface geophysical methods that are used for transportation applications and provide examples of the results obtained for each. Table 4-2. ASTM guides and standards for surface geophysical investigations Geophysical Method ASTM Guide or Standard Standard Guide for Selecting Surface Geophysical Methods D6429 Seismic Refraction D5777 Seismic Reflection D7128 Electrical Resistivity D6431 Soil Resistivity G57 Frequency-domain electromagnetics D6639

38 Geophysical Method ASTM Guide or Standard Time-domain electromagnetic D6820 Ground-penetrating radar D6432 Microgravity D6430 Source: Sirles (2006) 4.3.1 Seismic Methods Seismic methods use measurements of the velocity of mechanical (i.e., stress) waves propagating through the ground to infer stratigraphy from contrasts in seismic velocity between layers and to evaluate the small- strain stiffness (i.e., modulus) of subsurface materials. Depending on the specific method, either body (i.e., compression and shear) waves or surface (i.e., Rayleigh) waves are used. Typical ranges of the compression (or P) wave velocity (Vp) and shear (or S) wave velocity (Vs) for earth materials are shown in Figure 4-1. 0 5,000 10,000 15,000 20,000 25,000 Compression Wave Velocity, Vp (ft/s) Water Clay Sand Till Weathered Rocks Intact Rocks Steel

39 Source: Bourbie et al. (1987), USACE (1995), Mavko et al. (1998), Santamarina et al. (2001), FHWA (2002) Figure 4-1. Typical ranges of compression (top) and shear wave velocity (bottom) The compression wave velocity is directly related to the constrained modulus of the material: = where M = the initial tangent constrained modulus = the total mass density of the material Similarly, the shear wave velocity is directly related to the shear modulus: = where G = the initial tangent shear modulus The compression wave velocity is greater than the shear wave velocity, with the ratio depending on the Poisson’s ratio of the material. Because the strain levels associated with the propagation of seismic waves through the ground are very small for most seismic methods, the constrained and shear moduli correspond to the initial tangent (i.e., maximum) stiffness. These simple, fundamental equations are the basis of one of the most attractive features of seismic methods⎯in situ measurements of compression and shear wave velocity may be used to directly measure the small-strain moduli of soil and rock, which are useful for evaluating deformations related to serviceability limit states. In saturated soils, the measured compression wave velocity often reflects the properties of the pore fluid in the voids (Allen et al. 1980). As such, measurements of compression wave velocity in saturated soils are 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 Shear Wave Velocity, Vs (ft/s) Clay Sand Till Weathered Rocks Intact Rocks Steel

40 often of limited value for determining the properties of the soil itself. In these cases, the shear wave velocity is a more useful quantity because it is mostly unaffected by the presence of fluid in the voids. 4.3.1.1 Seismic Refraction and Reflection When compression and shear waves encounter an interface between two dissimilar materials, some of the energy is reflected by the interface and some is transmitted (i.e., refracted). The characteristics of the reflected and transmitted waves are governed by the physics of wave propagation in layered media (e.g., Ewing et al. 1957). The seismic refraction test is based on measuring the travel time of critically refracted (i.e., angle of refraction = 90°) compression or shear waves along the interfaces between different lithologic formations. Seismic refraction data is collected using a linear array (spread) of vertical geophones to record first arrivals of compression waves generated from a weight drop or explosive source. Although seismic refraction tests can also be conducted using shear waves, it is less common. The geophones are connected via a geophone cable to an engineering seismograph. The production rate can be increased significantly by using a towed land streamer rather than placing the geophones individually. The land streamer consists of geophones attached to a sled or high-strength sleeve, and it allows the entire array to be moved simultaneously while maintaining a fixed spacing. Energy sources range from sledgehammers for shorter arrays and shallow depths to explosive sources for longer arrays and deeper depths. For each geophone spread, multiple source locations (shot points) at the ends and in the center of the spread are used; more shot points should be used if large lateral variations in subsurface stratigraphy are expected. Seismic refraction processing steps include filtering to reduce the effect of unwanted noise, selecting the first-arrival time for each geophone and shot point, and plotting the travel times vs. geophone offset. Several methods are available to interpret the data, including the simple intercept-time method, Generalized Reciprocal Method, and seismic refraction tomography. The Generalized Reciprocal Method (Palmer 1981) is widely used because it provides the depth and seismic velocities of refractors below each geophone location and can accommodate nonplanar refractors (i.e., layer interfaces) and lateral variations in seismic velocity within a single geophone spread. The most important limitation of these interpretation methods is the hidden layer problem that occurs when either (i) a low-velocity layer underlies a high-velocity layer (i.e., velocity reversals) or (ii) there are thin layers with insufficient velocity contrasts. Seismic refraction tomography (White 1989, Zhang and Toksoz 1998) is a more advanced interpretation method capable of calculating a two-dimensional (2D) velocity structure by minimizing the difference between predicted and observed first-arrival travel times. It is supplanting the intercept-time method and Generalized Reciprocal Method for geotechnical applications because the aforementioned hidden-layer problem is minimized. The most common application of the seismic refraction test is to evaluate the stratigraphy of various lithologic units, particularly the interface between soil and rock because of the contrast in seismic velocity between these two materials. An example is shown in Figure 4-2 of using the measured vertical and lateral variation of compression wave velocity to estimate the depth to bedrock. The black plus signs indicate the depth to bedrock observed in soil borings.

41 Source: Geosyntec Consultants, Inc. Figure 4-2. Example seismic refraction tomography test results Seismic refraction is also used to estimate the rippability of bedrock (Caterpillar Inc. 2000) based on the compression wave velocity using relationships like that shown in Figure 4-3. Source: Caterpillar, Inc. Figure 4-3. Estimated D8R ripper performance from seismic velocity The seismic reflection method is based on measuring the travel time of seismic waves that are reflected by the interface between different lithologic units. Geotechnical applications of seismic reflection (often called shallow or high-resolution seismic reflection) have benefited from the experience gained from the extensive use of the method for resource exploration. Seismic reflection data are also collected using a linear array of geophones, a seismograph with a corresponding number of channels, and an appropriate energy source. Like the seismic refraction test, land streamers are commonly used to expedite testing. Energy sources used include blank shotgun shells fired using a Buffalo gun, explosives, and swept- frequency sources (e.g., Vibroseis). The processing of shallow seismic reflection data is often complex and requires considerable skill and experience. Processing steps may include the following: • Filtering and scaling each record • Muting to remove ground roll (i.e., surface waves) • Making static corrections for elevation changes and variable thickness of near-surface zones

42 • Correcting for normal move-out • Migrating to correct for dip of subsurface reflectors Subsequent data interpretation can be subjective. Other data, such as boring logs, may be used to generate synthetic seismograms to correlate with measured seismic reflection data to check the reasonableness of interpreted sections. The shallow seismic reflection method (Steeples and Miller 1990) is less common for geotechnical site characterization than other seismic methods, in part due to cost and the extensive data processing requirements. However, many near-surface geophysical practitioners have been successful in using shallow seismic reflection to map depth to bedrock, locate faults, image abandoned mines, and identify areas prone to subsidence from sinkholes (e.g., Hunter et al. 1984, Steeples and Miller 1990, Miller and Steeples 2008, Pugin et al. 2009). An example of mapping the interface between soil and bedrock is shown in Figure 4-4. Figure 4-4(a) shows the processed depth section, while Figure 4-4(b) is the interpreted section identifying the bedrock surface. Note the borehole located at about 4,500 meters that correlates well with the seismic reflection section. A second example is shown in Figure 4-5 of a seismic reflection section along an interstate highway that was used to identify the location and lateral extent of sinkholes. The sinkholes are located at approximately stations 1340 and 1500. Source: Pugin et al. (2009) Figure 4-4. Seismic reflection section showing depth to bedrock

43 Source: Miller and Steeples (2008) Figure 4-5. Seismic reflection section showing sinkhole locations 4.3.1.2 Surface Wave Methods Surface wave methods use the dispersive nature of Rayleigh (i.e., surface) wave propagation in a vertically layered medium to develop 1D shear wave velocity profiles. Surface-wave tests may be conducted using an active source, such as a sledge hammer or dropped weight; it is also possible to use large, swept-frequency vibrators (e.g., Vibroseis) to generate low-frequency Rayleigh waves that enable deeper profiling. A linear array of vertical, low-frequency geophones is used to record data. Like seismic refraction and reflection, a land streamer can be used to expedite testing. Surface wave seismic data can be collected using passive sources of Rayleigh waves such as ambient noise. Passive sources generally produce energy at lower frequencies (e.g., 2 to 10 Hertz [Hz]) than most active sources and thus allow deeper profiling. However, passive tests may be impractical at some sites, especially those located in rural areas, because of insufficient ambient noise. For passive-source measurements, a 2D array of geophones is commonly used because the location of the source(s) of passive energy is usually unknown. An alternative approach, called the refraction microtremor (ReMi) technique (Louie 2001), uses a linear array of geophones. The passive ReMi technique assumes an omnidirectional wavefield that should be verified for each test (Foti et al. 2015). However, the ReMi method can also accommodate active collinear sources, similar to other active-source surface-wave methods. At many sites, it is advantageous to combine active and passive tests to obtain data over a broad range of frequencies.

44 The processing steps consist of (i) calculating an experimental dispersion curve that shows the variation of Rayleigh wave phase velocity with frequency (or wavelength) and (ii) performing an inversion of the experimental dispersion curve to obtain the profile. An example dispersion curve is presented in Figure 4-6. There are many variants of surface wave methods that have been developed, including the Spectral Analysis of Surface Waves (SASW) and Multichannel Analysis of Surface Waves (MASW) methods, among others. Typically, the differences between these variants include the type of source (active or passive), the shape of the receiver array (2d or linear), and the specific techniques used to calculate the experimental dispersion curve and perform the inversion to obtain the profile. Source: Geosyntec Consultants, Inc. Figure 4-6. Example Rayleigh-wave dispersion curve The inversion process results in a 1D profile. As noted previously, the profile is useful for seismic site response analyses (AASHTO 2011). Multiple 1D models generated from a series of array locations can be combined and presented as a 2D cross section of shear wave velocity. Because shear wave velocity is directly related to the stiffness of subsurface materials, 2D models are valuable in estimating the depth to rock and delineating loose zones in the subsurface. Figure 4-7 shows an example of using an MASW test to identify the top of a firm sand layer. Additional guidance on performing surface wave tests is available in Foti et al. (2015, 2018). Source: Geosyntec Consultants, Inc. Figure 4-7. Example 2D shear wave velocity cross section from MASW test

45 4.3.2 Electrical and Electromagnetic Methods Electrical and electromagnetic geophysical methods use the flow of electrical currents through the ground to evaluate subsurface characteristics. Electrical (or galvanic) methods commonly induce the currents via electrodes that are directly coupled to the ground and measure the resulting potential (i.e., voltage) difference via a separate pair of electrodes. But it is also possible to measure currents and potentials that occur naturally due to subsurface processes. Electromagnetic (or induction) methods use eddy currents that are induced in the ground by time-varying magnetic fields generated by an electrical current within coils that are not directly coupled to the ground. From these measurements, the vertical and lateral distribution of electrical resistivity (or its inverse– conductivity) can be calculated. Because the resistivity of earth materials is affected by mineralogy, porosity, chemistry of the pore fluids, and degree of saturation, electrical resistivity surveys can be used to define subsurface layering, locate cavities, and delineate the groundwater table. For example, clays tend to have low resistivities because of the presence of exchangeable cations in the pore fluids, while sands containing fresh water have higher resistivities. The resistivity of an earth material usually decreases as the moisture content of the material increases. The resistivity of selected earth materials is given in Figure 4-8. Source: ICE (1976), USACE (1995) Figure 4-8. Typical ranges of electrical resistivity 4.3.2.1 Resistivity Traditionally, an electrical resistivity array consists of four electrodes that are coupled to the ground. Two of the electrodes transmit an electrical current ( ) to the ground and the other two electrodes measure the change in potential ( ) in the earth materials between the current electrodes. The apparent resistivity ( ) is a function of the measured electrical impedance ( = ⁄ ): = where = the geometric factor that depends on the configuration and spacing of the electrodes Three configurations of four-electrode arrays have been commonly used in electrical resistivity surveys: (i) the Schlumberger array, (ii) the Wenner array, and (iii) the dipole-dipole array. These array configurations differ in their vertical and horizontal resolution; signal-to-noise ratio; susceptibility to 1 10 100 1,000 10,000 100,000 1,000,000 Resistivity, ρ (ohm-meters) Igneous rocks Metamorphic rocks Dense Limestone Porous Limestone Shale Sand, wet to moist Clay

46 electromagnetic coupling; and ease of field implementation, automated data acquisition, and interpretation. Additional array configurations are available for specific objectives (Zonge et al. 2005). The Wenner and dipole-dipole arrays readily adapt to modern automatic data acquisition systems employing multielectrode cables. The measured apparent resistivities are average resistivities of all the earth materials through which the electrical current flows. As the electrode spacing is increased, the electrical current flows through more material, and the apparent resistivities calculated from the field arrays represent averages over a larger volume. The measured apparent resistivities can be interpreted to evaluate subsurface features using several approaches of increasing complexity: • Pseudosections (profile plots of 2D apparent resistivity) are generated from the resistivity measurements. For relatively simple, nearly homogeneous profiles, the pseudosection gives a reasonable approximation of the vertical and lateral distribution of actual resistivities. Because computers have made more complex, accurate methods of interpretation possible, pseudosections are now primarily used for quality control during field acquisition. • Closed-form expressions are available that relate the apparent resistivity to the resistivity and thickness of individuals layers in a 1D, multilayer earth model. Algorithms have been developed that automatically adjust the resistivity and thickness of individuals layers until satisfactory agreement is obtained between the calculated and measured apparent resistivity values as a function of electrode spacing. • Electrical resistivity tomography or electrical resistivity imaging are based on using numerous combinations of four-electrode arrays from a large, multielectrode cable. For each four-electrode array, the calculated and measured pseudosections are compared, and the resistivities within a 2D or 3D model are adjusted until satisfactory agreement is obtained. The numerous combinations of arrays provide ample data to develop detailed models. An example of electrical resistivity imaging test results is shown in Figure 4-9. The low-resistivity materials are interpreted to be caused by (i) higher porosity due to weathering of carbonate bedrock and (ii) higher water content. Source: Geosyntec Consultants, Inc. Figure 4-9. Example electrical resistivity imaging results Induced polarization methods are closely related to resistivity methods, but they include an additional measurement called chargeability, which is a measure of the energy storage capacity of a material. The additional parameter aids with identifying anomalies and has proven effective for mapping landfills and contaminant plumes (Zonge et al. 2005).

47 4.3.2.2 Electromagnetic Methods Electromagnetic methods can be broadly divided into two groups: Frequency-domain electromagnetics (FDEM) methods and time-domain electromagnetic (TDEM) methods. In FDEM methods, a transmitter coil emits a sinusoidally varying current at a specific frequency. A receiver coil measures the secondary field generated by the induced eddy currents in the subsurface. The amplitude of the secondary field is usually expressed as a percentage of the primary field at the receiver. The phase shift caused by the time delay in the received field can also be measured. An alternative is to separate the received field into two components. The first component is in phase with the transmitted field, and the second component is 90° out-of-phase with the transmitted field. The in-phase component is sometimes called the real component, and the out-of-phase component is sometimes called the quadrature or imaginary component. The most common type of FDEM method for engineering applications is the terrain conductivity method (McNeill 1990). Terrain conductivity electromagnetic systems are instruments that use two loops or coils as transmitter and receiver, respectively. For shallow profiling, the two coils are located a fixed distance apart in a boom that is carried by one person. For deeper profiling, two people are needed: one person generally carries the transmitter coil, while a second person carries the second coil that receives the primary and secondary fields. Terrain conductivity meters are operated in both the horizontal and vertical dipole modes. These terms describe the orientation of the transmitter and receiver coils to each other and the ground, and each mode gives a significantly different response with depth. When used in the vertical dipole mode, the instruments are more sensitive to the presence of relatively conductive, steeply dipping structures; whereas in the horizontal dipole mode, the instruments are relatively insensitive to this type of structure and can give accurate measurement of ground conductivity near them. Because terrain conductivity meters read directly in apparent conductivity and most surveys using the instrument are done in the profile mode, interpretation is usually qualitative and used to identify anomalies. Any anomalous areas are investigated further with other geophysical techniques or borings and soundings. Information about the variation of conductivity with depth can be obtained by measuring two or more coil orientations or intercoil separations, or both, and using commercially available software to perform an inversion of the measured data to obtain a 1D profile of conductivity at the sounding location. Figure 4-10 shows the results of a terrain conductivity survey performed to identify areas of high conductivity corresponding to buried waste and contaminated soils prior to site development.

48 Source: Spotlight Geophysical Services, LLC Figure 4-10. Example terrain conductivity results A common TDEM resistivity sounding survey consists of a square transmitter coil laid on the ground and a receiver coil located in the center of the transmitter coil. The TDEM method measures the decaying secondary field induced in the subsurface by the current in the transmitter coil. By making measurement of the voltage out of the receiver coil at successively later times, measurement is made of the current flow and, thus, also of the electrical resistivity of the earth at successively greater depths. The measured apparent resistivity as a function of time can be interpreted using commercially available software to calculate a 1D profile of resistivity at the sounding location. 4.3.2.3. Ground-Penetrating Radar Ground-penetrating radar (GPR) uses an electromagnetic pulse transmitted from a radar antenna to probe the subsurface. The transmitted, high-frequency electromagnetic pulses are reflected from various interfaces within the ground, and the reflected pulse is detected by a receiver antenna. Reflecting interfaces may be soil horizons, the groundwater surface, soil-rock interfaces, man-made objects, or any other interface possessing a contrast in dielectric properties. GPR has found widespread use for transportation applications (Sirles 2006), and the following are common applications (Morey 1998): • Measuring pavement thickness • Locating voids beneath pavements • Evaluating delamination in bridge decks • Estimating depth to bedrock and the groundwater table

49 • Mapping underground utilities and other buried objects such as tanks and drums GPR antennas are designated by their center frequency and range from approximately 10 megahertz (MHz) for mapping deep subsurface stratigraphy to approximately 3,000 MHz for shallow rebar mapping. The antennas may be placed on the ground surface (i.e., ground-coupled) or suspended above the ground surface (i.e., air-launched). The latter is often mounted on a vehicle and used for pavement studies to allow surveys to be conducted at highway speeds. A typical mode of operation is the common-offset mode where the transmitter and receiver antennas are maintained at a fixed separation distance and moved along a survey line. Measurements are taken at specific distance or time intervals as needed. The reflected pulses are recorded as a function of time and may be converted to an approximate depth using an estimated velocity or the reflection time of a known feature such as a drain pipe. As the antennas are moved along the survey line, the output is displayed as a cross section or radar image of the subsurface. Resolution of interfaces and discrete objects is typically very good due to the short wavelengths associated with the electromagnetic pulses. However, the attenuation of the pulses in earth materials may be high, and depth penetration may be limited. This is particularly true in highly conductive soils such as clays where penetration may be limited to a few feet. In low-conductivity sands, the penetration depths are potentially much greater, up to approximately 150 ft (45 m). The majority of GPR applications use 2D sections that have been corrected for attenuation by applying a gain function. Further data processing, when needed, is similar to that used for seismic reflection data. Available processing steps include frequency filtering to remove noise, spatial filtering to remove horizontal noise, and migration to collapse diffractions and move dipping reflectors to their actual position. While 2D profiles are sufficient for most projects, 3D processing of GPR data collected along closely spaced lines can provide more accurate and quicker evaluation of specific targets, such as voids, and better visualization of subsurface features. Figures 4-11 and 4-12 present examples showing the use of a GPR test to identify the base of a fill layer and a void under a continuously reinforced concrete pavement. Source: Spotlight Geophysical Services, LLC Figure 4-11. Example GPR results showing base of fill layer

50 Source: Texas DOT Notes: CRCP: continuously reinforced concrete pavement Figure 4-12. Example GPR results showing pavement void detection 4.3.3 Potential Field Methods Examples of potential fields include gravitational, magnetic, and electrical fields. Several geophysical methods are based on measuring local changes in these potential fields for identifying subsurface features. 4.3.3.1 Microgravity Gravity methods involve measuring anomalies in the gravitational field of the Earth due to differences in the densities of materials in the subsurface. For example, an air-filled solution cavity has a different density than surrounding rock units, which would result in a gravity anomaly, or a low in the bedrock surface would appear as an anomaly when compared to the gravity readings in surrounding terrain. A gravity study to evaluate near-surface voids and loose zones is often referred to as a microgravity survey, indicating the magnitude of the gravity anomalies that are measured, typically in the microgal (µgal) range (1 µgal ≈ 10-9 g). Typical applications of microgravity surveys in transportation include mapping bedrock topography and identifying voids and loose zones caused by karst activity or underground mining. The resolution of a gravity survey depends on the sensitivity of the gravimeter and on the density contrast between different units in the subsurface. Microgravity surveys are conducted by taking measurements at specific stations along a line or at grid points. The stations should be spaced close enough each other to adequately define the subsurface conditions of interest. The elevation of each station must be obtained accurately, typically within ±1 in. (±2 cm) for a data accuracy on the order of a few µgals. The horizontal station accuracy is not as critical as elevation, and differential global positioning system (GPS) values are usually sufficient. The measurements are made by placing the gravity meter on the ground surface, leveling the meter (or use a self-leveling meter), and waiting for the reading to stabilize before recording the reading. At least once per hour, the gravity meter should be returned to a reference location (i.e., base station) to monitor temporal changes in the local gravitational field and check for instrument drift.

51 The processing steps for microgravity surveys include making corrections to the measured data for (i) instrument drift (using base station readings), (ii) Earth tides (using published models), (iii) latitude (using measured horizontal location), and (iv) elevation differences compared to a reference datum. The last process step in the list, comparing the elevation differences to a reference datum, is done in two steps: 1. A free-air correction to compensate for the change in gravity due to the elevation of a station above the reference datum 2. A Bouguer correction to compensate for the mass of soil or rock between the gravity station and the datum. The corrected value is the simple Bouguer anomaly and is the value typically used for data interpretation. Terrain corrections may also be required if the surrounding terrain is not relatively flat (Long and Kaufmann 2013). Processed microgravity data can be presented as a profile line (Figure 4-13) or, in the case of a grid study, as a contour map (Figure 4-14). Gravity anomaly sources can be modeled, if needed, but anomalies are often just used as a guide to select locations for drilling or excavation. Source: Spotlight Geophysical Services, LLC Figure 4-13. Example of a microgravity profile

52 Source: Spotlight Geophysical Services, LLC Figure 4-14. Example of a microgravity contour map 4.3.3.2 Magnetometry Magnetic surveys measure changes in the magnitude in the magnetic field of the Earth due to the presence of ferrous metal objects or by Earth materials that have high magnetic susceptibilities. Common targets for a magnetic survey include buried (ferrous) metal objects such as tanks, pipelines, and steel-cased wells; although the use of magnetometers has been replaced in some applications by electromagnetic induction metal detectors due to the higher lateral resolution of the metal detectors. Magnetic surveys can also be used to map the bedrock surface if the rock is known to have a high magnetic susceptibility. Examples of the latter include rocks that have unusually high magnetic susceptibilities (e.g., gabbro, basalt) as well as other types of rock that contain significant amounts of ferrous minerals (e.g., manganese, magnetite, hematite). Prior to conducting a magnetometer study, solar storm activity should be checked, as this may affect the Earth’s magnetic field. The National Oceanic and Atmospheric Administration (NOAA) Space Weather Prediction Center maintains a website that tracks and predicts solar activity. Magnetometer data can be collected at specific fixed stations but usually is recorded while the operator is walking along a survey line or grid using GPS for positioning. Data stations should not be located near any man-made object that can change the magnitude of the Earth's magnetic field (e.g., power transmission lines, automobiles, metal pipelines and fences, structural steel in roads and buildings). The operator should also avoid having metallic objects on their person (e.g., steel-toe boots, metal belt buckles). The total magnetic field is measured in units of nanoTesla (nT) by a single-sensor magnetometer and using a base station to record diurnal changes in the Earth’s magnetic field. The need for a base station can be eliminated by using a two-sensor gradiometer that measures the vertical or horizontal gradient of the field. The vertical gradient cancels out diurnal changes and resolves shallow metallic objects better than single-sensor magnetometer data because it measures the rate of change between the readings of the two sensors (the gradient) rather than the total field. Data processing steps will depend on the type of instrument used (i.e., single-sensor magnetometer or gradiometer). For both types, the data is initially filtered to remove high-frequency noise. For a single- sensor magnetometer, subtracting the base station readings from the readings at other stations removes the

53 time-varying changes in the magnetic field (and the value of the field at the base station location). Long- wavelength filtering (e.g., subtracting a best-fit polynomial) can be used to remove regional changes in the field over the area of the survey, producing a total field anomaly map. Due to the inclination and declination of the Earth’s magnetic field, magnetometer anomalies over a single magnetic object will appear dipolar, and the peak response will be offset laterally from the source of the anomaly. The reduction-to-pole processing available in commercial software corrects for inclination and declination and converts the dipole anomaly to a monopole anomaly that is centered over the anomaly. The use of a gradiometer eliminates much of the processing required for single-sensor data. However, the vertical gradient of the field, while highlighting the response from shallow ferrous objects, emphasizes the slope of the magnetic field so that the peak response is offset from the source of the anomaly. The results of a magnetometer survey can be presented as a profile line or as a contour map. The presented data should include the location of known metallic objects so that anomalies from these objects will not be mistaken as targets of interest. 4.3.3.3 Self-Potential The self-potential (SP) method, also called the spontaneous potential method, measures the naturally occurring voltage differences caused by water moving through a porous medium. A common geotechnical application for the SP method is to map seepage paths from dams and reservoirs (e.g., Jansen et al. 1994). Other applications include mapping groundwater flow around pumping wells or around faults. SP measurements are made using nonpolarizable (porous pot) electrodes that have a porous ceramic tip, contain a copper rod, and are filled with a conductive solution of copper-sulfate. A reference electrode is placed in a fixed location, while a rover electrode is moved from location to location. The two electrodes are connected to a high-impedance multimeter to measure the voltage difference between the two locations. Measurements are typically made by inserting the rover electrode in shallow hand-auger holes, 6 to 12 in. (15 to 30 cm) deep to contact moist soils. SP data are usually collected along parallel lines with measurements made at a fixed spacing such as 5 or 10 ft (1.5 to 3 m). In addition to variations caused by water seepage, measurements are affected by corroding metal, electrode polarization and drift, changes in moisture in the soil at the base station, and variation in natural telluric currents (Corwin 1990). Corrections must be made for these sources of noise. SP data can be plotted as a single curve or, if collected using a grid of lines, as a contour map. Typically, SP data are evaluated qualitatively, although quantitative methods exist (Corwin 1990). Borehole Geophysical Methods There are two general categories of borehole geophysical tests used for geotechnical subsurface explorations: (i) borehole-to-borehole and surface-to-borehole methods, and (ii) in-hole logging methods. The in-hole logging methods are often based on methods developed for resource exploration and adapted for near-surface applications. This section provides a summary of borehole geophysical methods that have applications for transportation-related projects as shown in Table 4-3. Further guidance is provided in the ASTM guides for borehole geophysical methods listed in Table 4-4.

54 Table 4-3. Matrix of borehole geophysical methods in relation to investigation objectives Objectives Se is m ic C ro ss ho le Se is m ic D ow nh ol e C al ip er R es is tiv ity Sp on ta ne ou s Po te nt ia l In du ct io n N at ur al G am m a G am m a- G am m a D en si ty N eu tr on P or os ity A co us tic T el ev ie w er Se is m ic L og gi ng 1 Lithology       Seismic wave velocity profile    Fracture location and characteristics   Density, porosity, and water content   Borehole diameter  Source: Paillet and Ellefsen (2005) Table 4-4. ASTM guides for borehole geophysical investigations Geophysical Method ASTM Guide Standard Guide for Conducting Borehole Geophysical Logging D5753 Mechanical Caliper D6167 Natural Gamma and Gamma-Gamma Density D6274 Electromagnetic Induction D6726 Neutron D6727 Crosshole Seismic Testing D4428 Downhole Seismic Testing D7400 Source: Sirles (2006) 4.4.1 Borehole-to-Borehole and Surface-to-Borehole Methods The seismic crosshole and downhole methods have been used in geotechnical engineering for many years to measure compression ( ) and shear wave velocity ( ) profiles, with the latter being of valuable for determining the site class (AASHTO 2011) or analyzing seismic site response. As noted previously, seismic velocities may also be used to calculate small-strain moduli in soil and rock.

55 4.4.1.1 Seismic Crosshole Method The seismic crosshole method is conceptually simple. Compression and shear waves are generated in one borehole, and the arrival of those waves is recorded in one or more additional boreholes by a geophone at the same depth. By moving the source and geophones(s) up and down the boreholes in unison, a profile of compression and shear wave velocity vs. depth can be generated. The boreholes are commonly 10 to 15 ft (3 to 4.5 m) apart and are typically cased to prevent collapse. Cement grout ensures good physical coupling between the casing and surrounding soil to enable seismic waves to be recorded accurately. Sources for compression wave surveys include electrical discharge sources and air guns for water-filled boreholes, and downhole hammers for both water-filled and dry boreholes. A source that clamps to the inside of the casing and generates a vertical shearing motion can be used for shear wave velocity surveys. One or more geophones are placed in the adjacent boreholes at the same depth as the source. Generally, measurements are repeated at 2-ft (0.6-m) depth intervals from the ground surface to 20 ft (6 m). Below 20 ft (6 m), measurements are usually taken at 5-ft (1.5-m) depth intervals to the maximum depth of the boreholes. Because the seismic wave velocities calculated from the observed travel times depend directly on the distance of travel between boreholes, it is critical to accurately measure this distance as a function of depth. The location of the top of each boring (i.e., top of casing) can be located by land surveying techniques; the borehole deviation can be measured using an inclinometer. The travel time of the seismic waves may be determined by identifying the initial arrival of compression and shear waves. Examples of recorded waveforms are shown in Figure 4-15, along with the selection of the arrival times of shear waves and the calculated travel time (Δt). Alternatively, the travel time can be determined via cross correlation; peaks in the cross-correlation function correspond to the time lag (i.e., travel time) between two or more geophones. The velocity is then determined using the calculated distance between boreholes at the corresponding depth. Although it is assumed that the measured wave velocity is for a seismic wave traveling directly between boreholes, it is important to consider that refraction through a higher-velocity layer above or below the source-receiver depth may produce the first arrival. One way to minimize the potential for refraction is to limit the distance between boreholes. Source: Geosyntec Consultants, Inc. Figure 4-15. Example seismic crosshole data

56 4.4.1.2 Seismic Downhole Method For seismic downhole measurements, a source on the ground surface is used to generate compression and shear waves, and arrivals are recorded by one or more geophones placed in a borehole. For tests conducted in soils, the borehole is usually cased to avoid collapse. Care must be taken in grouting the casing to prevent voids in the borehole annulus that can prevent the seismic waves from reaching the geophones. The source is located approximately 5 to 15 ft (1.5 to 4.5 m) from the borehole to reduce the potential for generating seismic waves in the casing itself. For shear wave measurements, a beam or plank in contact with the ground surface is struck to generate horizontally polarized shear waves. Automatic shear wave sources, such as the autoseis described in Chapter 5, may also be used. For compression wave measurements, a sledgehammer striking a metal plate on the ground surface is sufficient. The geophone is lowered down the borehole at 2-ft to 5-ft (0.6-m to 1.5-m) intervals. At each depth interval, the arrival of compression and shear waves is recorded. A limitation of the method is that the amplitude of the wave arrivals diminished with depth due to attenuation. If necessary, the records from multiple source activations can be added together (i.e., stacked) to increase the signal-to-noise ratio. Arrival times for the compression and shear waves are selected and used to calculate the corresponding wave velocity profile. If two or more geophones are used, the wave velocity may be calculated using a true- interval method where the travel time between the geophones is calculated for a single activation of the source. This is desirable because the two seismic waveforms should have similar shapes, making it easier to identify the arrivals and avoid the need for a source trigger. If only one geophone is available, a pseudo- interval travel time must be calculated as the difference between the source-to-receiver travel times when the single geophone is at two different depths. If the travel time is calculated over a small depth interval, the velocities are sensitive to errors in choosing arrival times. This may be overcome by using travel-time measurements at several adjacent depths to calculate an average velocity. The most common variant of the seismic downhole method in use is the seismic piezocone test described in Chapter 5. See Section 5.3.1.3 for further details. Figure 4-16 shows an example from a seismic piezocone test where seismic tests were performed at 3.3-ft (1-m) depth intervals to a total depth of approximately 260 ft (80 m). The red symbols indicate the first crossover point on each waveform, which is used to calculate the pseudo-interval travel time.

57 Source: ConeTec Figure 4-16. Example seismic downhole data 4.4.2 In-Hole Logging Methods In-hole logging methods use mechanical, electrical, electromagnetic, nuclear, acoustic (or sonic), and seismic measurements within a borehole (USACE 1995, Hearst et al. 2000, Paillet and Ellefsen 2005). Continuous borehole logs can be made by raising or lowering one or more tools in the borehole, and measurements made by the tools are plotted as a function of depth. With few exceptions, logs cannot be obtained from a cased section of borehole. Borehole logs can be correlated with samples taken from the borehole. Distinctive signatures of subsurface units can be identified on the logs and correlated between holes to generate detailed stratigraphic cross sections of a site. Borehole logging can be used to identify stratified sedimentary deposits such as sands, clays, and organic material; to identify rock units containing radioactive material; and to distinguish permeable sands from impermeable sands. 4.4.2.1 Mechanical Methods Caliper logs are logs of the mechanically or acoustically measured diameter of the borehole and represent one of the most useful and simplest techniques used in borehole geophysics. Mechanical calipers consist of a downhole probe with one or more feeler arms that contact the borehole walls and can detect irregularities on the walls as the probe is pulled up the hole. Mechanical calipers may be used in holes filled with air, water, or drilling fluid. Acoustic calipers consist of a probe usually containing transducers that emit acoustic waves and measure the travel times of the waves reflected from the borehole walls. Borehole diameters can be calculated if the seismic wave velocity in the borehole fluid is known. As such, acoustic calipers must

58 be used in boreholes that are filled with water or drilling fluid. Caliper logs may be used to detect fractures and are commonly used to correct other in-hole logs for borehole diameter effects. 4.4.2.2 Electrical and Electromagnetic Methods Borehole resistivity methods are similar to surface electrical surveys in that electrical current is imparted to subsurface formations, and voltage drops across the formations are measured. Single-point resistance logging consists of a single electrode in the borehole and a single electrode on the ground surface, both of which serve as current and potential electrodes. Normal-resistivity logging is a common multielectrode technique in which multiple current and potential electrodes are contained within the tool itself. The calculated apparent resistivity is sensitive to mineralogy, porosity, chemistry of the pore fluids, degree of saturation, and moisture content as discussed in Section 4.3.2. Spontaneous potential methods measure the electrical potentials established between formation fluids and the drilling fluid and the electrical potentials established at the boundaries of permeable subsurface layers. In both cases, the electrical potential is due to differences in the salinity across formational boundaries or between the formation fluids and drilling fluids. In many boreholes drilled for engineering purposes, natural formation waters are used as the drilling fluids and salinity differences between drilling fluids and formation fluids will not exist, limiting the usefulness of the spontaneous potential log. Similar to surface electromagnetic methods, induction logging uses a time-varying magnetic field generated by a current in a transmitting coil to induce eddy currents in the materials surrounding the borehole. In turn, the eddy currents create a secondary magnetic field that results in a voltage measured by the receiving coil. Induction logs measure conductivity, which is the reciprocal of resistivity. The measurement of conductivity usually is inverted to provide curves of both resistivity and conductivity. 4.4.2.3 Nuclear Methods Nuclear methods involve measuring natural gamma radiation in a formation or the backscatter of radiation as the result of bombardment of the formation by gamma radiation or neutrons. Natural gamma logs measure the natural radioactivity (due to the presence of potassium, thorium, and uranium) of geologic materials and are primarily used for lithologic identification. The natural radioactivity of a material (measured in counts per second) is proportional to the amount of clay minerals present, and, thus, the natural gamma log is ideal for identifying the presence of clay layers and seams. An example of a natural gamma log is shown in Figure 4-17. Also shown in Figure 4-17 are the results of the mechanical caliper and acoustic televiewer (discussed subsequently) logs.

59 Source: Geosyntec Consultants, Inc. Figure 4-17. Example mechanical caliper, natural gamma, and acoustic televiewer logs Gamma-gamma density logs use gamma radiation emitted from a source within the logging tool. As the gamma radiation passes through the formation, some of it is backscattered and measured by a detector in the tool. The degree of gamma radiation backscatter is directly proportional to the bulk density of a material. Neutron porosity logs are made with a source of neutrons in the probe and detectors that measure the backscatter from materials near the borehole. The degree of backscatter is proportional to the amount of hydrogen present, which is largely a function of the water content. In saturated materials, the water content is in turn a function of the porosity.

60 A disadvantage of gamma-gamma density and neutron porosity logs is that the radioactive sources present a health hazard and require permits and certification for transportation and handling. Some states also restrict their use when testing in drinking water aquifers. 4.4.2.4 Optical and Acoustic Televiewer Methods Optical and acoustic televiewers provide a continuous, 360° view of the borehole wall that allows rock mass discontinuities to be identified and characterized. Both devices can be oriented within the borehole so that the absolute orientation of features (e.g., bedding planes) can be measured. An optical televiewer (OTV) uses a camera to record high-resolution images of the borehole wall and includes lights for illumination. An OTV is best suited for dry boreholes or boreholes filled with clear water. Any conditions that produce cloudy or murky water or coatings on the borehole wall limit the usefulness of the OTV. If good images are obtained, it is possible to identify locations and orientations of joints, bedding planes, foliations, faults, shears, and other naturally occurring rock mass discontinuities. The acoustic televiewer (ATV) uses ultrasound pulses from a rotating sensor in an open, fluid-filled borehole to record the amplitude and travel time of the signals reflected at the (high-impedance) interface between fluids and the borehole wall. Because an ATV uses ultrasound rather than visible light, the borehole fluid is not required to be clear. Rock mass discontinuities in the wall of the borehole will change the amplitude of the reflected acoustic wave compared to the surrounding material. The method does not work well in soil because of the lack of a high-impedance boundary between the fluid and soil. ATV surveys are used to provide information regarding locations and orientations of joints, bedding planes, faults, shears, and other naturally occurring rock mass discontinuities. General geologic structure data derived from OTV and ATV data includes a structure log, arrow plot (tadpole), and stereonet plots (polar and rose). An example of travel time and ATV logs is shown in Figure 4-17. The corresponding interpretation of the data is shown in Figure 4-18, which includes the structure, tadpole, and 3D logs. Williams and Johnson (2004) provide a summary of optical and acoustic televiewers and describe how they may be combined and integrated with other in-hole logging methods.

61 Source: Geosyntec Consultants, Inc. Figure 4-18. Example interpretation of acoustic televiewer log 4.4.2.5 Seismic Methods The purpose of seismic logging methods is to measure the compression and shear wave velocity profiles using a tool containing both the source(s) and receivers suspended in a fluid-filled borehole. The borehole fluid provides the necessary coupling between the tool and the surrounding soil or rock. Compression and shear wave velocity are measured at frequent depth intervals (2 ft [0.6 m] or less) as the probe is lowered (or raised) in the borehole. In the geophysical literature, these methods are also often called acoustic or sonic methods. Herein, the term seismic is used to indicate that the primary use is measuring seismic wave velocity profiles. Seismic logging methods offer excellent resolution, and they are well suited for measuring seismic wave velocity profiles at great depths.

62 Early geophysical applications of seismic logging methods focused on measuring only the travel time of compression waves. Since the 1980s, it has been more common to use full waveform logging methods that compression, shear, and Stonely waves to be measured to gather more information. Stonely waves are surface waves that propagate along the interface between the borehole wall and borehole fluid. Full waveform methods usually use logging tools with multiple receivers to enable robust processing methods to be used to calculate seismic wave velocities and amplitudes from the recorded waveforms. A key distinction for geotechnical applications is the type of source used in the logging tool. Most early versions of the tool employed a monopole source, which generates a compression wave in the borehole fluid. Such devices are useful for evaluating fast formations, defined as follows: , > , where , = the shear wave velocity of the formation , = the compression wave velocity of the borehole fluid In a fast formation, it is possible to measure the shear wave velocity of the soil or rock directly because the high-frequency monopole P-wave partitions to P and S waves at the interface between the borehole fluid and borehole wall and are critically refracted along the borehole wall to one or more receivers. Geotechnical applications typically involve slow formations ( , < , ). In such cases, there is no critical refraction of shear waves, and, thus, it is not possible to directly measure the shear wave of the soil using a monopole source. Although the shear wave velocity may be inferred from an analysis of the dispersion of Stonely waves, it is preferable to use a dipole source in slow formations. A dipole source generates compression, shear, and flexural waves along the borehole, enabling direct measurement of the shear wave velocity. The most commonly used variant of seismic logging used for geotechnical site investigations is the P-S suspension log. Figure 4-19 shows an example of compression and shear wave velocity profiles obtained from seismic logging using a dipole-source tool.

63 Source: Geosyntec Consultants, Inc. Figure 4-19. Example results from seismic logging 0 2000 4000 6000 8000 Velocity (ft/sec) 1600 1200 800 400 0 D ep th (f t)

64 Chapter 4 References AASHTO. 2011. AASHTO Guide Specifications for LRFD Seismic Bridge Design. Second Edition, with 2012, 2014, and 2015 Interim Revisions. American Association of State Highway and Transportation Officials, Washington, DC. AASHTO. 2017. AASHTO LRFD Bridge Design Specifications. US Customary Units, Eighth Edition. American Association of State Highway and Transportation Officials, Washington, DC. Allen, F.A., F.E. Richart, Jr., and R.D. Woods. 1980. “Fluid Wave Propagation in Saturated and Nearly Saturated Sands.” Journal of Geotechnical Engineering, Vol. 106, No. GT3, pp. 235–254. Anderson, N., N. Croxton, R. Hoover, and P. Sirles. 2008. Geophysical Methods Commonly Employed for Geotechnical Site Characterization. Transportation Research Circular E-C130, Transportation Research Board, Washington, DC. Bourbie, T., O. Coussy, and B. Zinszner. 1987. Acoustics of Porous Media. Gulf Publishing Company. Caterpillar Inc. 2000. Handbook of Ripping. Twelfth Edition, Peoria, Illinois. Corwin, R.F. 1990. “The Self-Potential Method for Environmental and Engineering Applications.” in Geotechnical and Environmental Geophysics, Vol. I: Review and Tutorial. Ward, S.H., (ed)., Society of Exploration Geophysicists, Tulsa, OK, pp. 127–145. Decker, J.B., K.M. Rollins, and J.C. Ellsworth. 2008. “Corrosion Rate Evaluation and Prediction for Piles Based on Long- Term Field Performance.” Journal of Geotechnical Engineering, Vol. 134, No. 3, pp. 341–351. Ewing, W.M., W.S. Jardetzky, and F. Press. 1957. Elastic Waves in Layered Media, McGraw-Hill, New York. Fenning, P.J., and S. Hasan. 1995. “Pipeline Route Investigations Using Geophysical Techniques.” in Eddleston, M., S. Walthall, S. Cripps, and M.G. Culshaw, (eds), Engineering Geology of Construction, Geological Society Engineering Geology, Special Publication No. 10, pp. 229–233. FHWA. 2002. Subsurface Investigations – Geotechnical Site Characterization. Publication No. FHWA NHI-01-031, Federal Highway Administration, Washington, DC. FHWA. 2006. Soils and Foundations, Reference Manual – Volume I. Publication No. FHWA NHI-06-088, Federal Highway Administration, Washington, DC. Foti, S., C.G. Lai, G.J. Rix, and C. Strobbia. 2015. Surface Wave Methods for Near-Surface Site Characterization. CRC Press, Boca Raton, FL. Foti, S., F. Hollender, F. Garofalo, et al. 2018. “Guidelines for the Good Practice of Surface Wave Analysis: A Product of the InterPACIFIC Project,” Bulletin of Earthquake Engineering, Vol. 16, p. 2367. Hearst, J.R., P.H. Nelson, and F.L. Paillet. 2000. Well Logging for Physical Properties: A Handbook for Geophysicists. Geologists, and Engineers, Second Edition, Wiley and Sons, New Jersey. Hunter, J.A., S.E. Pullan, R.A. Burns, R.M. Gagne, and R.L. Good. 1984. “Shallow Seismic Reflection Mapping of the Overburden‐Bedrock Interface with the Engineering Seismograph—Some Simple Techniques.” Geophysics, Vol. 49, No. 8, pp. 1381–1385. ICE. 1976. Manual of Applied Geology for Engineers. Institution of Civil Engineers, London. Jansen, J., N. Billington, F. Snider, and P. Jurcek. 1994. “Marine SP Surveys for Dam Seepage Investigations: Evaluation of Array Geometries through Modeling and Field Trials.” Proceedings of 7th EEGS Symposium on the Application of Geophysics to Engineering and Environmental Problems, Boston, Massachusetts, pp. 1053–1071. Long, L.T., and R.D. Kaufmann. 2013. Acquisition and Analysis of Terrestrial Gravity Data. Cambridge University Press, New York, New York. Louie, J.N. 2001. “Faster, Better: Shear-Wave Velocity to 100 Meters Depth from Refraction Microtremor Arrays.” Bulletin of the Seismological Society of America, Vol. 91, No. 2, pp. 347–364. Mavko, G., T. Mukerji, and J. Dvorkin. 1998. The Rock Physics Handbook: Tools for Seismic Analysis in Porous Media. Cambridge University Press, Cambridge, United Kingdom. McNeill, J.D. 1990. “Use of Electromagnetics for Groundwater Studies,” Geotechnical and Environmental Geophysics. Vol. I: Review and Tutorial, Ward, S.H., (ed)., Society of Exploration Geophysicists, Tulsa, OK, pp. 191–218. Miller, R.D., and D.W. Steeples. 2008. High-Resolution Seismic-Reflection Imaging of I-70 Sinkholes, Russell County, Kansas. Kansas Department of Transportation, Open-File Report 2008-18, October.

65 Morey, R.M. 1998. Ground Penetrating Radar for Evaluating Subsurface Conditions for Transportation Facilities. National Cooperative Highway Research Program Synthesis 255, National Cooperative Highway Research Program, National Academy of Science, Washington, DC. Paillet, F.L., and K.J. Ellefsen. 2005. “Downhole Applications of Geophysics.” Near-Surface Geophysics, Butler, D.K., (ed). Society of Exploration Geophysicists, Tulsa, Oklahoma, pp. 439–471. Palmer, D. 1981. “An Introduction to the Generalized Reciprocal Method of Seismic Refraction Interpretation.” Geophysics, Vol. 46, No. 11, pp. 1508–1518. Pugin, A.J.M., S. Pullan, J.A. Hunter, and G.A. Oldenborger. 2009. “Hydrogeological Prospecting Using P- and S-Wave Landstreamer Seismic Reflection Methods.” Near Surface Geophysics, pp. 315–327. Santamarina, J.C., K.A. Klein, and M.A. Fam. 2001. Soils and Waves: Particulate Materials Behavior, Characterization, and Process Monitoring. Wiley and Sons, West Sussex, United Kingdom. Sirles, P. 2006. Use of Geophysics for Transportation Projects. National Cooperative Highway Research Program Synthesis 357, Transportation Research Board, Washington, DC. Steeples, D.W., and R.D. Miller. 1990. “Seismic Reflection Methods Applied to Engineering, Environmental, and Groundwater Problems.” in S.H. Ward, (ed), Geotechnical and Environmental Geophysics, Society of Exploration Geophysics, 63, 1–30. USACE. 1995. Geophysical Exploration for Engineering and Environmental Investigations. Engineer Manual 1110-1-1802, U.S. Army Corps of Engineers, Washington, DC. White, D. J. 1989. “Two-Dimensional Seismic Refraction Tomography.” Geophysical Journal International, Vol. 97, No. 2, pp. 223–245. Wightman, W.E., F. Jalinoos, P. Sirles, and K. Hanna. 2003. Application of Geophysical Methods to Highway Related Problems. Report FHWA-IF-04-021, Federal Highway Administration, Washington, DC. Williams, J.H. and C.D. Johnson. 2004. “Acoustic and Optical Borehole-Wall Imaging for Fractured-Rock Aquifer Studies.” Journal of Applied Geophysics, Vol. 55, Issue 1-2, pp. 151–159. Zhang, J., and M. Toksoz. 1998. “Nonlinear Refraction Traveltime Tomography.” Geophysics, Vol. 63, No. 5, pp. 1726– 1737. Zonge, K., J. Wynn, and S. Urquhart. 2005. “Resistivity, Induced Polarization, and Complex Resistivity.” in Near-Surface Geophysics, Butler, D.K., (ed), Society of Exploration Geophysicists, Tulsa, Oklahoma, pp. 265–300.