ASTM D6432-11

Concepts—This guide summarizes the equipment, field procedures, and data processing methods used to interpret geologic conditions, and to identify and provide locations of geologic anomalies and man-made objects with the GPR method. The GPR uses high-frequency-pulsed EM waves (from 10 to 3000 MHz) to acquire subsurface information. Energy is propagated downward into the ground from a transmitting antenna and is reflected back to a receiving antenna from subsurface boundaries between media possessing different EM properties. The reflected signals are recorded to produce a scan or trace of radar data. Typically, scans obtained as the antenna(s) are moved over the ground surface are placed side by side to produce a radar profile.

The vertical scale of the radar profile is in units of two-way travel time, the time it takes for an EM wave to travel down to a reflector and back to the surface. The travel time may be converted to depth by relating it to on-site measurements or assumptions about the velocity of the radar waves in the subsurface materials.

Vertical variations in propagation velocity due to changing EM properties of the subsurface can make it difficult to apply a linear time scale to the radar profile (Ulriksen (32)).

Parameter Being Measured and Representative Values:

Two-Way Travel Time and Velocity—A GPR trace is the record of the amplitude of EM energy that has been reflected from interfaces between materials possessing different EM properties and recorded as a function of two-way travel time. To convert two-way times to depths, it is necessary to estimate or determine the propagation velocity of the EM pulses. The relative permittivity of the material (ε_r) through which the EM pulse propagates mostly determines the propagation velocity of the EM wave. The propagation velocity through the material is approximated using the following relationship (see full formula in Balanis (33)):

It is assumed that the magnetic permeability is that of free space and the loss tangent is much less than 1.

Table 1 lists the relative permittivities (ε_r) and radar propagation velocities for various materials. Relative permittivity values range from 1 for air to 81 for fresh water. For unsaturated earth materials, ε_r ranges from 3 to 15. Note that a small change in the water content of earth materials results in a significant change in the relative permittivity. For water-saturated earth material, ε_r can range from 8 to 30. These values are representative, but may vary considerably with temperature, frequency, density, water content, salinity, and other conditions.

If the relative permittivity is unknown, as is normally the case, it may be necessary to estimate velocity or use a reflector of known depth to calculate the velocity. The propagation velocity, V_m, is calculated from the relationship as follows:

Methods for measuring velocity in the field are found in 6.7.3. Note that measured velocities may only be valid at the location where they are measured under specific soil conditions. If there is lateral variability in soil and rock composition and moisture content, velocity may need to be determined at several locations.

Attenuation—The depth of penetration is determined primarily by the attenuation of the radar signal due to the conversion of EM energy to thermal energy through electrical conduction, dielectric relaxation, or magnetic relaxation losses. Conductivity is primarily governed by the water content of the material and the concentration of free ions in solution (salinity). Attenuation also occurs due to scattering of the EM energy in unwanted directions by inhomogeneities in the subsurface. If the scale of inhomogeneity is comparable to the wavelength of EM energy, scattering may be significant (Olhoeft (34)). Other factors that affect attenuation include soil type, temperature (Morey (35)), and clay mineralogy (Doolittle (36)). Environments not conducive to using the radar method include high conductivity soils, sediments saturated with salt water or highly conductive fluids, and metal.

Equipment—The GPR equipment utilized for the measurement of subsurface conditions normally consists of a transmitter and receiver antenna, a radar control unit, and suitable data storage and display devices.

Radar Control Unit—The radar control unit synchronizes signals to the transmitting and receiving electronics in the antennas. The synchronizing signals control the transmitter and sampling receiver electronics located in the antenna(s) in order to generate a sampled waveform of the reflected radar pulses. These waveforms may be filtered and amplified and are transmitted along with timing signals to the display and recording devices.

Real-time signal processing for improvement of signal-to-noise ratio is available in most GPR systems. When working in areas with cultural noise and in materials causing signal attenuation, time varying gain is necessary to adjust signal amplitudes for display on monitors or plotting devices. Filters may be used in real time to improve signal quality. The summing of radar signals (stacking) is used to increase effective depth of exploration by improving the signal-to-noise ratio.

Data Display—The GPR data are displayed as a continuous profile of individual radar traces (Fig. 2). The horizontal-axis represents horizontal traverse distance and the vertical-axis is two-way travel time (or depth). Data are commonly presented in wiggle trace display, where the intensity of the received wave at an instant in time is proportional to the amplitude of the trace (see Fig. 2), or as a gray scale of color scale display, where the intensity of the received wave at an instant in time is proportional to either the intensity of gray scale (that is, black is high intensity, and white is low intensity; see Fig. 3) or to some color assignment defined according to a specified color-signal amplitude relationship.

Antennas and Control Cables—The antennas used to transmit and receive radar signals are generally electric dipoles. A single-dipole antenna can be used to both transmit and receive signals in the monostatic mode. The bi-static mode uses separate antennas for transmitting and receiving. These antennas can be housed in a single enclosure where the distance between the two antennas are fixed, or in separate enclosures where the distance between the two antennas can be varied. The ability to vary the distance between the two antennas is helpful in optimizing the survey design for specific types of target detection.

Electromagnetic waves are three-dimensional vector fields where the orientation of the fields is described by the vector direction or polarization of the electrical and magnetic fields. Changing the polarization of a linearly polarized electric dipole antenna can cause maximum or minimum coupling to a scattering object. For example, alignment of the electric field axis (the long length of a dipole antenna) parallel to a pipe or wire will maximize the response of the pipe as a reflector scatterer, while a perpendicular alignment will minimize the pipe response. Typically, two antenna systems use the same orientation and polarization for both antennas, but sometimes the receive antenna will be oriented with its electric field perpendicular (orthogonal) to the transmit antenna, resulting in insensitivity to reflection from horizontal layers and linear features (like pipes) that are aligned to either antenna, but high sensitivity to off-alignment pipes.

Antennas are manufactured both with and without shielding (metal or high radar absorption material). Shielding reduces energy radiation from the sides and top of the antenna, which in turn reduces reflections from surface and above-ground targets. Low-frequency antennas (less than 100 MHz) are rarely shielded, whereas most high-frequency antennas are shielded.

The center frequency of commercially available antennas ranges from 10 to 3000 MHz. These antennas generate pulses which typically have 2 to 3 octaves of bandwidth. In general, lower-frequency antennas provide an increase in depth of penetration but have less resolution than higher-frequency antennas.

The selection of antenna frequency depends on the depth of penetration, spatial resolution, and system portability required for the study.

Limitations and Interferences:

General Limitations Inherent to Geophysical Methods:

A fundamental limitation of all geophysical methods lies in the fact that a given set of data cannot always be associated with a unique set of subsurface conditions. In most situations, surface geophysical measurements alone cannot resolve all ambiguities, and some additional information is required. Because of this inherent limitation in the geophysical methods, a GPR survey alone can not be considered a complete assessment of subsurface conditions. Properly integrated with other sources of knowledge or geophysical methods, GPR can be a highly effective, accurate, and cost-effective method of obtaining subsurface information.

In addition, all surface geophysical methods are inherently limited by decreasing resolution with depth.

Limitations Specific to the GPR Method:

The GPR method is site specific in its performance (depth of penetration and resolution), depending upon surface and subsurface conditions. Radar penetration of more than 30 m has been reported in geologic settings of water saturated sands (Morey (35); Beres and Haeni (2), Smith and Jol (38), Wright et al (1)), 300 m in granite, 2000 m in dry salt (Unterberger (39)), and 5400 m in ice (Wright et al (23)). More commonly, penetration is on the order of 1 to 10 m. Limitations are discussed in the following section.

Material Properties Contrast—Reflection coefficients quantify the amplitude of reflected and transmitted signals at boundaries between materials. Reflection coefficients depend on the angle of incidence, the polarization of the incident field, and the EM property contrast. In addition to having sufficient velocity contrast, the boundary between the two materials needs to be sharp. For instance, it is more difficult to see a water table in fine-grained materials than in coarse-grained materials because of the different relative thicknesses of the capillary fringe for the same contrast.

Attenuation—Radar signal attenuation is caused by the effect of electrical conductivity, dielectric and magnetic relaxation, scattering, and geometric spreading losses (Olhoeft (34)).

(1) Electrical Conductivity LossesElectrically conductive materials such as many mineralogic clays and free ions in solution attenuate the radar signal by converting EM energy to thermal energy (Olhoeft 34)).

(2) Dielectric Relaxation LossesRadar signals can also be attenuated by dielectric relaxation losses due to the rotational polarization of the liquid water molecule and chemical charge transfer processes on the surface of clay minerals. Attenuation due to dielectric relaxation losses arises from the conversion of EM energy to thermal energy (Olhoeft (34)).

(3) Geometric Scattering LossesScattering may be a dominant factor in signal attenuation when inhomogeneities in materials with grain sizes in the order of a radar wavelength (Table 2) are present (Olhoeft (34)).

Polarization—The type and alignment of polarization of the vector electromagnetic fields may be important in receiving responses from various scatterers. Two linear, parallel polarized, electric field antennas can maximize the response from linear scatters like pipes when the electric field (typically long axis of the dipole antenna) is aligned parallel with the pipe and towed perpendicular across the pipe. Similarly, alignment with the rebar in concrete will maximize the ability to map the rebar, but alignment perpendicular to the rebar will minimize scattering reflections from the rebar to see through or past the rebar to get the thickness of concrete. Similar arrangement may be made for overhead wires and nearby fences. Cross-polarized antennas (perpendicular to each other) minimize the response from horizontal layers.

Interferences Caused by Ambient, Geologic, and Cultural Conditions:

Measurements obtained by the GPR method may contain unwanted signals (noise) caused by geologic and cultural factors.

Ambient and Geologic Sources of Noise—Boulders, animal burrows, tree roots, or other inhomogeneities can cause unwanted reflections or scattering of the radar waves. Lateral and vertical variations in EM properties can also be a source of noise.

Cultural Sources of Noise—Aboveground cultural sources of noise include reflections from nearby vehicles, buildings, fences, power lines, lampposts, and trees. In cases where this kind of interference is present in the data, a shielded antenna may be used to reduce the noise.

(1) Scrap metal at or near the surface can cause interference or ringing in the radar data. The presence of buried structures such as foundations, reinforcement bars (rebar), cables, pipes, tanks, drums, and tunnels under or near the survey line may also cause unwanted reflections (clutter).

(2) In some cases, EM transmissions from nearby cellular telephones, two-way radios, television, and radio and microwave transmitters may induce noise on the radar record.

(3) Other Sources of NoiseOther sources of noise can be caused by the EM coupling of the antenna with the earth and decoupling of the antenna to the ground due to rough terrain, heavy vegetation, water on the ground surface, or other changes in surface conditions.

Summary—All possible sources of noise present during a survey should be noted so that their effects can be considered when processing and interpreting the data.

Alternate Methods—The limitations previously discussed may prohibit the effective use of the GPR method, and other methods or non-geophysical methods may be required to resolve the problem (see Guide D6429).

Note 1—The quality of the result produced by applying this standard is dependent on the competence of the personnel performing the work, and the suitability of the equipment and facilities used. Agencies that meet the criteria of Practice D3740 are generally considered capable of competent and objective testing/sampling/inspection/etc. Users of this standard are cautioned that compliance with Practice D3740 does not in itself assure reliable results. Reliable results depend on many factors; Practice D3740 provides a means of evaluating some of those factors.

TABLE 1 Approximate Electromagnetic Properties of Various Materials

Note—
d = function of density,
w = function of porosity and water content,
f = function of frequency,
t = function of temperature
s = function of salinity, and
p = function of pressure.

TABLE 2 Radar Wavelengths (metres) for Various Antenna Frequencies (f) and Relative Permittivities (ε_r)

FIG. 3 Generalized Diagram of a Pipe Signature: GPR Record (300 MHz) Showing a Hyperbola from a Buried Pipe, and Computation of Depth and Velocity from that Target (see 6.7.2.3(2b))