ground penetrating radar survey design · what is ground penetrating radar? ground penetrating...

Ground Penetrating Radar

Survey Design

© Copyright 1992-1999, Sensors & Software Inc.

SENSORS & SOFTWARE INC.1091 BREVIK PLACE

MISSISSAUGA, ON L4W 3R7CANADA

PHONE: (905) 624-8909FAX: (905) 624-9365

E-MAIL: [email protected] Site: www.sensoft.on.ca

http://www.sensoft.on.ca

Survey DesignSurvey DesignSurvey DesignSurvey DesignSurvey Design Senosrs & SoftwareSenosrs & SoftwareSenosrs & SoftwareSenosrs & SoftwareSenosrs & Software

Senosrs & SoftwareSenosrs & SoftwareSenosrs & SoftwareSenosrs & SoftwareSenosrs & Software Survey DesignSurvey DesignSurvey DesignSurvey DesignSurvey Design

TTTTTaaaaabbbbble ofle ofle ofle ofle of Contents Contents Contents Contents Contents

1. Wha1. Wha1. Wha1. Wha1. What is Grt is Grt is Grt is Grt is Ground Pound Pound Pound Pound Penetrenetrenetrenetrenetraaaaating Rting Rting Rting Rting Radar?adar?adar?adar?adar? ................................................................................................................................................................................................................................. 11111

2. Pr2. Pr2. Pr2. Pr2. Proboboboboblem Deflem Deflem Deflem Deflem Definitioninitioninitioninitioninition ........................................................................................................................................................................................................................................................................................................................................................................ 33333Question 1: What is the target depth?Question 1: What is the target depth?Question 1: What is the target depth?Question 1: What is the target depth?Question 1: What is the target depth? ......................................................................................................................................................................................................................................................................... 33333Question 2: What is the target geometry?Question 2: What is the target geometry?Question 2: What is the target geometry?Question 2: What is the target geometry?Question 2: What is the target geometry? ........................................................................................................................................................................................................................................... 33333Question 3: WhaQuestion 3: WhaQuestion 3: WhaQuestion 3: WhaQuestion 3: What art art art art are the tare the tare the tare the tare the targgggget electrical pret electrical pret electrical pret electrical pret electrical properoperoperoperoperties?ties?ties?ties?ties? ....................................................................................................................................... 33333Question 4: What is the host material?Question 4: What is the host material?Question 4: What is the host material?Question 4: What is the host material?Question 4: What is the host material? .................................................................................................................................................................................................................................................................... 33333Question 5: WhaQuestion 5: WhaQuestion 5: WhaQuestion 5: WhaQuestion 5: What is the surt is the surt is the surt is the surt is the survvvvveeeeey eny eny eny eny envirvirvirvirvironment likonment likonment likonment likonment like?e?e?e?e? ............................................................................................................................................................................... 33333

3. Evaluating GPR Suitability3. Evaluating GPR Suitability3. Evaluating GPR Suitability3. Evaluating GPR Suitability3. Evaluating GPR Suitability ............................................................................................................................................................................................................................................................................................................ 44444

4. Reflection Survey Design4. Reflection Survey Design4. Reflection Survey Design4. Reflection Survey Design4. Reflection Survey Design ................................................................................................................................................................................................................................................................................................................. 777774.1 Selecting Oper4.1 Selecting Oper4.1 Selecting Oper4.1 Selecting Oper4.1 Selecting Operaaaaating Fting Fting Fting Fting Frrrrrequencequencequencequencequencyyyyy ........................................................................................................................................................................................................................................................................................ 777774.2 Estimating the Time Window4.2 Estimating the Time Window4.2 Estimating the Time Window4.2 Estimating the Time Window4.2 Estimating the Time Window ................................................................................................................................................................................................................................................................................................................. 888884.3 Selecting The Sampling Interval4.3 Selecting The Sampling Interval4.3 Selecting The Sampling Interval4.3 Selecting The Sampling Interval4.3 Selecting The Sampling Interval ........................................................................................................................................................................................................................................................................................ 999994.4 Selecting Station Spacing4.4 Selecting Station Spacing4.4 Selecting Station Spacing4.4 Selecting Station Spacing4.4 Selecting Station Spacing ........................................................................................................................................................................................................................................................................................................................... 10101010104.5 Selecting Antenna Separation4.5 Selecting Antenna Separation4.5 Selecting Antenna Separation4.5 Selecting Antenna Separation4.5 Selecting Antenna Separation ........................................................................................................................................................................................................................................................................................ 11111111114.6 Survey Grid and Coordinate System4.6 Survey Grid and Coordinate System4.6 Survey Grid and Coordinate System4.6 Survey Grid and Coordinate System4.6 Survey Grid and Coordinate System ........................................................................................................................................................................................................................................... 12121212124.7 Selecting Antenna Orientation4.7 Selecting Antenna Orientation4.7 Selecting Antenna Orientation4.7 Selecting Antenna Orientation4.7 Selecting Antenna Orientation ........................................................................................................................................................................................................................................................................................ 1313131313

5. CMP/WARR V5. CMP/WARR V5. CMP/WARR V5. CMP/WARR V5. CMP/WARR Velocity Sounding Designelocity Sounding Designelocity Sounding Designelocity Sounding Designelocity Sounding Design........................................................................................................................................................................................................ 1414141414

6. Summary6. Summary6. Summary6. Summary6. Summary.................................................................................................................................................................................................................................................................................................................................................................................................................................... 1616161616

References References References References References.................................................................................................................................................................................................................................................................................................................................................................................................................................... 1717171717

TTTTTABLE 1PrABLE 1PrABLE 1PrABLE 1PrABLE 1Properoperoperoperoperties ofties ofties ofties ofties of Common Geologic Ma Common Geologic Ma Common Geologic Ma Common Geologic Ma Common Geologic Materialsterialsterialsterialsterials ................................................................................................................... 1818181818

Survey DesignSurvey DesignSurvey DesignSurvey DesignSurvey Design Senosrs & SoftwareSenosrs & SoftwareSenosrs & SoftwareSenosrs & SoftwareSenosrs & Software

1

Senosrs & Software Survey Design

1. What is Ground Penetrating Radar?

Ground penetrating radar (or GPR for short) isthe general term applied to techniques which em-ploy radio waves, typically in the 1 to 1000 MHzfrequency range, to map structure and featuresburied in the ground (or in man-made structures).Historically, GPR was primarily focused onmapping structures in the ground; more recentlyGPR has been used in non-destructive testingof non-metallic structures. The applications arelimited only by the imagination and availabilityof suitable instrumentation.

The concept of applying radio waves to probethe internal structure of the ground is not new.�The successful application of these techniques,however, is still in its infancy. Without doubtthe most successful early work in this area wasthe use of radio echo sounders to map the thick-ness of ice sheets in the Arctic and Antarcticand sound the thickness of glaciers.

Work with GPR in non-ice environments startedin the early 1970�s. Early work focused on per-mafrost soil applications, Annan and Davis(1976). As an understanding of strengths andweaknesses of the method became apparent, itsapplication areas broadened as described byDavis and Annan (1989) and Scaife and Annan(1991). Applications in other areas are describedby Morey (1974), Benson et al (1984) andUlriksen (1982).

Radar systems can be deployed in three basicmodes which are referred to as reflection, ve-locity sounding and transillumination. Thesemodes are depicted in Figure 1. The most com-mon mode of operation is single-fold, fixed-off-set reflection profiling as illustrated in Figure 2.This mode of operation gives rise to data suchas shown conceptually in Figure 3. An examplesection from good radar terrain is presented inFigure 4.

Figure 1: Illustration of the three basic modes ofGPR operation.

Figure 2: Schematic illustration of common-offsetsingle-fold profiling.

2

Survey Design Senosrs & Software

Estimation of velocity versus depth is conductedfrequently with GPR systems; one of the sim-plest and most common methods is common-midpoint (CMP) sounding such as depicted inFigure 5. By varying antenna spacing and iden-tifying the time move out versus antenna sepa-ration for the various EM wavefronts (Figure6), radar wave velocity in the ground can be es-timated.

These two most common forms of GPR survey-ing are the subject of this paper on survey de-sign. Multifold reflection surveys (Fisher et al,1992), which are in essence a merger of the ba-sic CMP sounding and reflection mode, as wellas transillumination gathers which form the ba-sis of GPR tomography (Olsson et al, 1987) de-pend on the same principles discussed here.

In the following, a number of simple guidelinesare given to aid with design of a GPR survey.The reader should note that common sense mustprevail. The rules provided are based on simpli-fying more complicated relationships. Rules-of-thumb simplify a problem for expediency! Theuser desiring to become knowledgeable withGPR is encouraged to question the underlyingassumptions; this will lead to true understand-ing of GPR.

Figure 4: 100 MHz GPR section mapping bedrockdepth beneath sand and graveloverburden.

Figure 3: Format of a GPR reflection section withradar events shown for features asdepicted in Figure 2.

Figure 5: Procedure for conducting a CMP velocitysounding.

Figure 6: Illustration of CMP sounding ray pathsand idealized event arrival time versusantenna separation and display.

3


2. Problem Definition

The most important step in a ground penetrat-ing radar (GPR) survey is to clearly define theproblem. This step is not unique to radar butcommon to all geophysical techniques althoughoften overlooked in the urge �to rush off andcollect data�. There are five fundamental ques-tions to be answered before deciding if a radarsurvey if going to be effective.

Question 1: What is the targetdepth?

The answer to this question is usually the mostimportant. If the target is beyond the range ofGPR in ideal conditions then GPR can be ruledout as a viable method very quickly.

Question 2: What is the targetgeometry?

The target to be detected should be qualified asaccurately as possible. The most important tar-get factor is target size (i.e., height, length,width). If the target is non- spherical, the targetorientation (i.e., strike, dip, plunge) must bequalified.

Question 3: What are the targetelectrical properties?

The relative permittivity (dielectric constant) andelectrical conductivity must be quantified. Inorder for the GPR method to work, the targetmust present a contrast in electrical propertiesto the host environment in order that the elec-tromagnetic signal be modified, reflected, orscattered.

Question 4: What is the hostmaterial?

The host material must be qualified in two ways.First the electrical properties of the host mustbe defined. The relative permittivity and elec-trical conductivity have to be evaluated. Sec-ond, the degree and spatial scale of heterogene-ity in the electrical properties of the host mustbe estimated. If the host material exhibits varia-tions in properties which are similar to the con-trast and scale of the target, the target may notbe recognizable in the myriad of responses (com-monly referred to as volume scattering and clut-ter) generated by the host environment.

Question 5: What is the surveyenvironment like?

The GPR method is sensitive to the surround-ings in which measurements are made. Twoimportant factors are the presence of extensivemetal structures and of radio frequency electro-magnetic sources or transmitters. Another as-pect of the survey environment is accessibility.Can the equipment and the operator get to thearea of interest safely and economically? Arethere any unusual conditions or hazards (heat,cold, wet, toxic contamination, explosive atmo-sphere)?

In general, there are few environments whereradar cannot be used but available instrumenta-tion may not be suited to the specific situation.

4


3. Evaluating GPR Suitability

Prediction of whether GPR will �work� for theproblem at hand is not clear cut. In general it iseasier to rule out situations where radar is to-tally unsuitable than to state with confidence thatradar will be successful. Again, this is not aunique feature of the GPR method but is a factof life with all geophysical methods. GPR tendsto have more mystery because people have notnormally had as much experience with it as withsome other methods.

There are some basic tools which assist the GPRuser in the decision making process. The twomost important are the radar range equation, andnumerical simulation techniques. Some ex-amples are described by Annan and Chua (1988).The radar range equation (RRE for short) doesa basic allocation of available power against allthe loss mechanisms to yield a �yes/no� answeron whether a target will return sufficient powerto be detectable. The RRE has to simplify theproblem at hand; therefore, the results are goodguides, not absolute predictors of success or fail-ure. The basic steps of the radar range equationare depicted in Figure 7. Example results of anautomated program to carry out these calcula-tions are shown in Figure 8. Nomograms forspecific systems and targets can also be gener-ated such as shown in Figure 9.

Numerical simulation techniques (NST forshort) are not well developed for GPR. Simpleprograms for flat layered earth structures arecommercially available and are instructive touse. More complex 2 and 3-dimensional mod-eling programs are not available. The basic con-cepts for plane (flat) layered earth modeling areshown in Figure 10 accompanied by an examplesynthetic generated by the commercial Sensors& Software Inc. Synthetic Radargram program.

Answering the question �Will GPR work?� isneither easy nor exact. Addressing the follow-ing three questions will certainly help in antici-pating the answer.

Figure 7: Block diagram of radar range equation.

5


Question 1: Is the target withinthe detection range of the radar

irrespective of any unusualtarget characteristics?

The way to answer this question is to calculateor measure the host attenuation coefficient. Us-ing the radar range equation and the system per-formance factor (example in Figure 8), computethe maximum range that a reflector of the an-ticipated target type can be detected. If the tar-get is at a depth greater than this range, radarwin not be effective. A conservative rule-of-thumb is to state that radar will be ineffective ifthe actual target depth is greater than 50% ofthe maximum range.

Commercial radar systems can typically affordto have a maximum of 60 dB attenuation asso-ciated with conduction losses. A rough guide topenetration depth is

G ��RU��G ��

PD[ PD[<

��

a s

where a is attenuation in dB/m and a is conduc-tivity in mS/m. These equations are not univer-sal but are applicable when attenuation is mod-erate to high ( < 0. 1 dB/m) which is typical ofmost geologic settings.

Question 2: Will the targetgenerate a response detectable

above the background clutter? Inother words, does the target

have sufficient contrast inelectrical properties and is it

physically enough to reflect orscatter a detectable amount of

energy?

Power reflectivity is estimated using the expres-sion

Figure 8: Example listing of a radar range equationanalysis of a problem.

Figure 9: Radar range equation nomogramexample.

Figure 10: Illustration of a synthetic radargramprogram to predict a GPR response.

6


3U

. .

. .

+RVW 7DUJHW

+RVW 7DUJHW

�

-

-

Two conservative rules-of-thumb for predictingsuccess are as follows. First, the electrical prop-erties of the target should be such that the powerreflectivity be at least 0.01. (Note that a metaltarget is equivalent to KTargetà∞.) Second theratio of target depth to smallest lateral target di-mension should not exceed 10:1.

Question 3: Is there anything thatprecludes use of radar?

One example would be a radio transmitter lo-cated at the site. Another example would be atunnel lined with metal mesh to prevent looserock from falling. In the first case external sig-nals may saturate the sensitive receiver electron-ics. In the later, all the radar signal would bereflected at the tunnel wail and none would pen-etrate into the tunnel wall.

If the above questions can be answered in a posi-tive manner then there is a reasonable chanceGPR will work. The above conditions are posedin a conservative manner and intended to err onthe pessimistic side. More detailed analyses canemploy RRE and NST techniques. In general itis almost impossible to obtain reliable estimatesof all of the parameters involved in RRE andNST; these procedures are most effectively usedas part of a sensitivity analysis. The conclusionsdrawn win be fuzzy but informed.

As with all predictions nothing beats a real fieldtrial and if practical should be an integral com-ponent in survey design optimization. Unfortu-nately, financial constraints usually are a realand limiting factor.

7


4. Reflection Survey Design

The most common mode of GPR surveying iscommon-offset, single-fold reflection profiling.In such a reflection survey, a system with a fixedantenna geometry is transported along a surveyline to map reflections versus position.

There are seven parameters to define for a com-mon-offset, single-fold GPR reflection survey.These are the frequency, the time window, thetime sampling interval, the station spacing, theantenna spacing, the line location and spacing,and the antenna orientation.

4.1 Selecting OperatingFrequency

Selection of the operating frequency for a radarsurvey is not simple. There is a trade off be-tween spatial resolution, depth of penetration andsystem portability. As a rule, it is better to tradeoff resolution for penetration. There is no use inhaving great resolution if the target cannot bedetected!

The best way to approach the problem is definea generic target type (i.e., point target, roughplanar target, or specular target) and specify adesired spatial resolution, x. The initial fre-quency estimate is then defined by the formula.

I

��

[ .

0+]=

K is the relative permittivity (or dielectric con-stant) of the host material.

Using this frequency and the radar range equa-tion (see Figure 7 and 8), the ability to penetrateto sufficient depth to detect the target can thenbe evaluated. If penetration is insufficient, re-duce the frequency until adequate penetrationis achieved. Note that there is a practical limitplaced on this process by available instrumen-tation and the electrical properties of the host.

A simple guide is to use the following tablewhich is based on the assumption that the spa-tial resolution required is about 25% of the tar-get depth.

8


Depth (m) Center Frequency (MHz) 0.5 1000 1.0 500 2.0 200 5.0 100

10.0 50 30.0 25 50.0 10

The above are values based on practical experi-ence. Since every problem requires carefulthought, the above values should only be usedas a quick guide and not a replacement forthoughtful survey planning.

4.2 Estimating the Time Window

The way to estimate the time window requiredis to use the expression

: ��[�'HSWK

9HORFLW\=

where the maximum depth and minimum ve-locity likely to be encountered in the survey areaare used. The above expression increases theestimated time by 30% to allow for uncertain-ties in velocity and depth variations.

If no information is available on the electricalproperties, Table 1 provides a guide for first es-timates of the velocities of common geologicmaterials.

9


4.3 Selecting The Sampling Interval

The function relationship is

W

��

�I

=

where f is the center frequency in NM and t istime in ns.

In some instances it may be possible to increasethe sampling interval slightly beyond what isquoted but this should only be done when datavolume and speed of acquisition are at a pre-mium over integrity of the data. In any eventthe sampling interval should never be more than2 times that quoted here.

One of the parameters utilized in designing ra-dar data acquisition is the time interval betweenpoints on a recorded waveform. The samplinginterval is controlled by the Nyquist samplingconcept and should be at most half the period ofthe highest frequency signal in the record. Formost ground penetrating radar antenna systems,the bandwidth to center frequency ratio is typi-cally about one. What this means is that the pulseradiated contains energy from 0.5 times the cen-ter frequency to 1.5 times the center frequency.As a result the maximum frequency is around1.5 times the nominal center frequency of theantenna being utilized.

Based on the assumption that the maximum fre-quency is 1.5 times the nominal antenna centerfrequency, the data should be sampled at a ratetwice this frequency. For good survey design itis better that one uses a safety margin of two.As a result the sampling rate should be approxi-mately six times the center frequency of the an-tenna being utilized. Based on this analysis thefollowing table summarizes suitable samplingintervals versus operating frequency.

Antenna MaximumCenter Frequency Sampling Interval

(MHz) (ns)

10 16.720 8.3 50 3.3100 1.67200 0.83500 0.33

1000 0.17

10


4.4 Selecting Station Spacing

The selection of spacing between discrete radarmeasurements (see Figure 2) is closely linkedto the center operating frequency of the anten-nas and to the dielectric properties of the sub-surface materials involved. In order to assure theground response is not spatially aliased, theNyquist sampling intervals should not be ex-ceeded. The Nyquist sampling interval is onequarter of the wavelength in the host materialand expressed as

QF

�I .

��

I .��LQ�P�

[= =

where f is the antenna center frequency (in NM)and K is the relative permittivity of the host. Ifthe station spacing is greater than the Nyquistsampling interval, the data will not adequatelydefine steeply dipping reflectors. In areas of flatlying reflectors, this criterion can be compro-mised.

The spatial interval of measurement is clearlyillustrated by the example sections shown inFigure 11 and 12. The 50 MHz data in Figure11 were collected with a station spacing of 3mwhich is considerably larger than the computedNyquist interval of 0.5 to 0.75. The data in Fig-ure 11 clearly define the strong relatively flatlying reflectors. The steeply dipping events arealiased and appear as �hash� on the section. Theshown section sampled at a 0.5m station inter-val shown in Figure 12 clearly defines the steeplydipping features.

There are practical trade-offs to be made in se-lection of station interval. From a practical view-point, data volume and survey time are reducedby increasing the station interval. From a datainterpretation standpoint, adhering to theNyquist sampling interval is very important.There is also nothing to be gained by spatial

Figure 11: GPR reflection section from a deltaticenvironment obtained using 50 MHzantennas at a 3m station spacing.

Figure 12: Same section as Figure 11 but stationinterval reduced to Nyquist samplinginterval of 0.5 m.

oversampling. The sampling interval is extremelyimportant as this example indicates and should becarefully weighed in the survey design process.

11


4.5 Selecting Antenna Separation

Increasing the antenna separation also increasesthe reflectivity of flat lying planar targets whichcan sometimes be advantageous.

If little is known about the survey area, a saferule-of-thumb is set S equal to 20% of the targetdepth. In practice, a small antenna spacing isoften used because operational logistics usuallydemand simplicity of operation. Depth resolu-tion of targets decreases as antenna separationincreases although this factor is small until Sapproaches half the target depth.

Figure 13: Variation in antenna patterns as relative permittivity of the ground (K2

) changes. (Upper medium isK

1=1.)

Most GPR systems employ separate antennasfor transmitting and receiving (commonly re-ferred to as bistatic operation) although the an-tennas may be housed in a single module withno means of varying the antenna separation. Theability to vary the antenna spacing can be a pow-erful aid in optimizing the system for specifictypes of target detection. To maximize targetcoupling, antennas should be spaced such thatthe refraction focusing peak in the transmitterand receiver antenna patterns point to the com-mon depth to be investigated. Since the antennapattern peaks at the critical angle of the air-earthinterface as illustrated in Figure 13, (Annan etal, 1975 and Smith 1984). An estimate of opti-mum antenna separation is given by the expres-sion

V��'HSWK

�. ��=

-

12


4.6 Survey Grid and Coordinate System

An important aspect of survey design is estab-lishment of a survey grid and coordinate sys-tem. The use of a standardized coordinate sys-tem for position recording is very important; thebest data in the world are useless if no one knowswhere they came from.

Generally, survey lines are established which runperpendicular to the trend of the features underinvestigation in order to reduce the number ofsurvey lines. Line spacing is dictated by the de-gree of target variation in the trend direction. Ifisolated small targets are sought, the line spac-ing should be less than the radar footprint illus-trated in Figure 14.

The selection of survey line location and orien-tation should be made such as to analyze targetdetection. All survey lines should be orientedperpendicular to the strike of the target if the

target has a preferred strike direction. In attempt-ing to cover an area to map a feature such asbedrock depth, the survey fines should be ori-ented perpendicular to the bedrock relief and linespacing should be selected to adequately samplealong-strike variations without aliasing. In situ-ations where strike is known and the structure2-dimensional, a very large spacing betweenlines can be employed. If there is no two dimen-sionality to the structure, then line spacing mustbe the same as the station spacing to assure thatthe ground response is not aliased. Needless tosay, when nx is a fraction of a meter, a tremen-dous amount of data has to be collected to fullydefine a 3-dimensional structure.

Figure 14: Simplified GPR footprint concept whereshaded zone depicts area illuminated atdepth.

13


4.7 Selecting Antenna Orientation

Figure 15: Illustration of trhe various modes for antenna deployment. (E field assumed aligned along the antennaaxis.)

The last factor and seldom discussed factor tobe considered is the antenna orientation. In gen-eral, the antennas used for GPR are dipolar andradiate with a preferred polarity. The antennasare normally oriented so that the electric field ispolarized parallel to the long axis or strike di-rection of the target. There is no optimal orien-tation for an equi-dimensional target. In someinstances, it may be advisable to collect two datasets with orthogonal antenna orientations in or-der to extract target information based on cou-pling angle. If the antenna system is one whichattempts to use a circularly polarized signal theantenna orientation becomes irrelevant. Sincemost commercial systems employ polarized an-tennas, orientation can be important. The vari-ous arrangements of antenna deployment are il-lustrated in Figure 15. The most commonly usedis the one designated �PR-BD�.

14


5. CMP/WARR Velocity Sounding Design

Figure 16: CMP sounding data with 2 modes ofantenna polarization. (Data acquired at500 m position on reflection sectionshown in Figure 4.

The CMP (common mid-point) and WARR(wide angle reflection and refraction) soundingmodes of operation are the electromagneticequivalent to seismic refraction and wide anglereflection. CMP/WARR soundings are used toobtain an estimate of the radar signal velocityversus depth in the ground by varying the an-tenna spacing at a fixed location and measuringthe change of the two-way travel time to the re-flections as illustrated in Figures 5 and 6.

In the CMP sounding, both the antennas aremoved apart about a fixed location. In a WARRsounding, one antenna is held fixed while theother is moved away. In early GPR systems uti-lizing metallic cables, the WARR approach ofsounding was preferred because cable handlingwas a major concern in obtaining data free ofsystem artifacts. With modem systems such asthose employing fiber optics cables, the CMPapproach is the standard mode of operation sincethe reflected signal is more likely to come froma fixed spatial location rather than moving abouton the surface of a reflector as occurs with aWARR sounding.

Optimal CMP/WARR soundings are obtainedwhen the electric fields of the antennas are par-allel and the antennas are moved apart along aline which is perpendicular to the electric fieldpolarization (the �PR-BD� configuration in Fig15). This configuration gives the widest angu-lar coverage of a subsurface reflector. In addi-tion, close coupling of the antennas to the groundshould be maintained in order to maximize re-flection energy detectable at angles beyond thecritical angle of the air-earth interface.

The procedure for a CMP/WARR sounding issimple. A reflector is normally identified froma reflection section. A point on the ground sur-face is selected which is over the reflector. An-

tennas are then positioned over the target pointwith minimal separation. The initial spacing isusually nx, the Nyquist station interval selectedfor reflection profiling. Data are then acquiredat antenna separations which increase as inte-ger multiples of nx. If the CMP mode is used,both antennas are moved out from the centerpoint in steps of nx /2. If WARR mode is used,one antenna is moved out in step intervals of nx.The maximum separation in a CMP/WARRsounding is usually 1 to 2 times the reflectordepth. If the ground attenuation is high, the sig-nals may die out before the maximum separa-tion is reached.

The reflection arrival times should have a hy-perbolic dependence (to first order) on antennaseparation. Example data sets are shown in Fig-ure 16 for both the PR-BD and PLEF antennaconfigurations.

Analysis of the move out hyperbola of time ver-sus separation permits estimation of propaga-tion velocity and target depth. The basic inter-

15


pretation procedure is �Tl-X2�, analysis com-monly used in early seismic reflection interpre-tations. In simple terms, a plot of travel timesquared versus antenna separation squared yieldsa straight line relationship whose slope gives avelocity estimate and whose time intercept yieldsa depth estimate (see Figure 17). Computerizedschemes of varying degrees of complexity arenow commonly used to do this type of analysissuch as the velocity stack shown in Figure 18.

CMP/WARR soundings provide a measure ofsignal attenuation in the ground although to datethe estimation is more qualitative than quantita-tive. In addition, at sites with a large amount ofsurface clutter, CMP/WARR soundings can aidin separating above and below ground reflec-tions.

Figure 17: T2 - X2 analysis of CMP/WARR sounding.

Figure 18: Example of moveout velocity stacking of aPR-BD data set (partially presented inFigure 16.)

16


6. Summary

The preceding survey design guidelines providea basis for survey planning. As mentioned ear-lier, the use of common sense and a logicalthought process are needed to conduct high qual-ity GPR surveys. The more planning that goesinto a survey, the higher the likelihood of suc-cess and the easier the interpretation of the re-sults.

17


References

Annan, A.P. and Chua, L.T., 1988. Ground Penetrating Radar Performance Predictions, Ground Pen-etrating Radar (JA Pilon, Editor), Geological Survey of Canada. Paper 90-4.

Annan, A.P. and Davis, J.L, 1976. Impulse Radar Soundings in Permafrost. Radio Science, Vol. 11,p. 383-394.

Annan, A.P., Waller, W.M., Strangway, D.W., Rossiter, J.R., Redman, J.D., and Watts, R.D., 1975.The Electromagnetic Response of a Low-Loss, 2-Layer Dielectric Earth for Horizontal Electric Di-pole Excitation: Geophysics V. 40 p. 285-298.

Benson, R.C., Glaccum, RA and Noel, M.R., 1984. Geophysical Techniques for Sensing BuriedWastes and Waste Migration. US EPA Contract No. 68-03-3053. Environmental Monitoring Sys-tems Laboratory. Office of R&D. US EPA, Las Vegas, Nevada 89114. 236p-

Davis, J.L., and Annan, A.P., 1989. Ground Penetrating Radar for High-Resolution Mapping of Soiland Rock Stratigraphy, Geophysical Prospecting, 37 p. 531-551.

Fisher, E., McMechan, G.A., and Annan, A.P., 1992 (in press) Acquisition and processing of wide-aperture ground penetrating radar data: Geophysics

Morey, R.M., 1974. Continuous Subsurface Preflag by Impulse Radar. Proceedings of EngineeringFoundations Conference on Subsurface Exploration for Underground Excavations and Heavy Con-struction. Henniker, N.H. p. 213-232.

Olsson, O., Falk, L., Forslund, O., and Sandberg, E.,1987. Crosshole Investigations - Results fromBorehole Radar Investigations. Stripa Project TR 87-11. SKB, Stockholm, Sweden.

Scaife, J.E., & Annan, A.P., 1991. Ground Penetrating Radar - A Powerful, High Resolution Tool forMining Engineering and Environmental Problems; Paper presented at 93rd CIM Annual GeneralMeeting, Vancouver, B.C., April 29 - May 1, 1991.

Smith, G.S., 1984. Directive Properties of Antennas for Transmission into a Material Half-Space.IEEE Trans. Antennas propagate, vol AP-32, p. 232-246.

Ulriksen, C.P.F., 1982. Application of Impulse Radar to Civil Engineering. Unpublished Ph.D. The-sis, Dept. of Engs Geol., U. of Technology, Lund, Sweden, 175 p.

18


TABLE 1

Properties of Common Geologic Materials

Typical Relative Permittivity, Electrical Conductivity, Velocity andAttenuation Observed in Common Geologic Materials

K s v a MATERIAL (mS/M) (m/ns) (dB/m)

Air 1 0 0.30 0

Distilled Water 80 0.01 0.033 2xl0-3

Fresh Water 80 0.5 0.033 0.1

Sea Water 80 3xl03 .01 103

Dry Sand 3-5 0.01 0.15 0.01

Saturated Sand 20-30 0.1-1.0 0.06 0.03-0.3

Limestone 4-8 0.5-2 0.12 0.4-1

Shales 5-15 1-100 0.09 1-100

Silts 5-30 1-100 0.07 1-100

Clays 5-40 2-1000 0.06 1-300

Granite 4-6 0.01-1 0.13 0.01-1

Dry Salt 5-6 0.01-1 0.13 0.01-1

Ice 3-4 0.01 0.16 0.01

ground penetrating radar survey design · what is ground penetrating radar? ground penetrating...

Documents