Ground penetrating radar (GPR) is an electromagnetic “echo sounding” method. It is commonly used for a great many tasks where information about the top 1cm to several metres is needed. Pavement, bridge deck, and concrete structures can be investigated with very high frequency devices that use signals of 900 - 1200 x 10^6 cycles per second (ie 900-1200 megahertz). Large buried objects, voids and karst formations, sedimentary structure and other similar targets down to 10m depth may be imaged using systems with signals of 12 - 500 megahertz. Also, the internal and basal structure of glaciers as deep as several kilometres can be studied using specialized systems operating at 1.0 - 25 megahertz.
In this exercise we will learn about field operations and use of results for a typical small scale GPR survey. Data are nearly always gathered along lines. Sometimes many closely spaced lines are surveyed and interpreted in terms of 3D structure, but more typically, individual lines of data are collected, plotted and interpreted in terms of 2D structures under the line.
The task we are addressing is to characterize in as much detail as possible the structures, objects and material types under a single short line. Specific questions that will be asked are:
• What are the lateral position of distinct buried objects?
• What are the depths to these buried objects?
• Can the size of buried objects be determined?
• What is the topography of the overburden-basement interface?
• What are the electrical conductivity and dielectric permittivity of uniform ground?
• Can we make qualitative judgements about the ground’s characteristics? For example, we could ask “is there some difference between materials under each end of a line?”
Q1). Fill in the blanks of the following paragraphs
Setting up GPR acquisition parameters is similar to seismic surveying except that GPR time scales are in seconds x 10^( _____(a)_____) [called __________(b)__________], whereas seismic signals are more commonly recorded in units of seconds x 10^( ____(c)______) [called __________(d)__________]. The other major differenceis that the GPR energy source and signals are ___________(e)_________(energy type) where as seismic signals involve __________ signals (energy type again). This means that the principle physical properties affecting GPR signals are _________(f)___________ and _________(g)___________, whereas the principle physical properties affecting seismic signals are _________(h)___________ and _________(i)___________. Also, concerns regarding noise and interfering targets (above and below ground) are different for these two survey types. One example of a source of noise for GPR work is _________(j)___________ and one example of a source of noise for seismic work is __________(k)__________.
As for many geophysical surveys, the key design parameters are line spacing and station spacing. If lines are perpendicular to a roughly 2D target, one of these (namely _________(l)___________) can be fairly large. However _________(m)___________ (the other) along a line can, and should be, tighter than for many other surveys in order to ensure images are as easy to interpret as possible. Look at the data we are working on (click here to open in a new window). These GPR echo traces were gathered at ____(n)______ cm intervals along the survey line.
Instrument configuration usually involves _____(o)_____ transmitter(s) and _____(p)_____ receiver(s) for each measurement. These are moved together along the line.
GPR data are gathered in the so-called common offset configuration. For each measurement the source and receiver are the same distance apart. In figures above, the survey configuration is illustrated on top and resulting data are shown beneath. How far apart were the transmitter and receiver for our data set? _____(q)_____ cm.
To convert echo travel time into depth, the _____(r)_____ of EM waves in the ground is needed. It can be estimated by gathering a so-called common mid point data set (also known as a wide-angle radar reflection, or WARR, measurement).
Q2) Once a location has been chosen for this measurement, describe how the transmitter and receiver are moved so as to record this type of data set.
Q3) The hyperbolic diffraction pattern observed when a GPR is passed over a buried body can be explained by an equation relating T^2 to x^2 in the form:
The use of this relationshiop is known as T^2-X^2 analysis. In the equation above, t is the two way travel time of a EM wave from the GPR to a burried body and back, and x is the horizontal offset of the transmittor/receiver midpoint from the point directly above the burried body. Derive the above equation using d=vt and find the expressions for A and B. You can also write B in terms of t_0, which is the 2-way travel time when the GPR unit is directly over the body.
Q4) Fill in the blanks in the following paragraph:
GPR results are presented as echo traces along a survey line. The vertical axis is __________(s)__________, and the horizontal axis is _________(t)___________). Simple processing is sometimes applied to smooth traces, average traces, or to provide other simple enhancements, but results are always presented either as a set of variable area wiggle traces (as we have done for this lab), or as colour (or gray-scale) images showing signal strength as colour instead of as wiggly lines.
Q5) Examine the data provided. How many data sets are there?
Q6) What was done differently to produce Plot 2, compared to plot 1?
Q7) What was done differently to produce Plot 3, compared to plot 1?
Q8) What is the range of the time scale in the data? Over what range of times does the data appear to contain useful geophysical data?
The common mid point (CMP) data set shown here was obtained to determine the velocity in the top layer.
Q9) The data on the plot above was collected using the common midpoint survey method. The data presented on the data page was collected using the common offset survey method. Write a brief description of these two methods.
The first sloping arrival is the air wave. Confirm that this pattern has the speed of EM signals in air. You can use peaks because first arrivals are not easy to discern. Use the closest possible approximation to a straight line even though the arrival is clearly not ideally consistent. You should get 0.3m/ns, plus or minus 20% or so.
Q10) Print the image above. Draw the line on this image showing how you estimated the air velocity.
Q11) Draw a new line on the image showing how you can estimate the velocity in the top material using the second arrival. Again, you must use the best straight line that you can approximate.
Velocity is related approximately to relative dielectric constant via 
where C is the speed of light in air.
of air is 1, and of water is 80, and the values for most geologic materials varies between roughly 1 and 15. Water is the most important factor in determining dielectric constant, and hence the velocity of GPR signals.
Q12) What does the relative dielectric constant of the top layer appear to be?
Velocity is much less dependent upon electrical conductivity. However, the rate at which signal energy decays as it propagates within the soil is controlled by conductivity. We can not obtain quantitative estimates directly, but we can observe where signals penetrate deeper and where they appear to penetrate poorly - ie where they are rapidly attenuated.
Q13) What is the greatest depth from which signals appear to be returning? Do not consider hyperbolic or slanting patterns. You should base your estimate on the unprocessed results (data plot 1).
Q14) Does this maximum depth of investigation appear to vary along the line?
Now we can begin to interpret the data.
Q15) Determine location, 2-way travel times and probable depths for a the:
a) buried pipe
b) Concrete utilities caseing (a second target generating a rather complicated “1-sided” signature)
c) the top of the till “basement”.
Remember that all 3 data figures show the same data, but each has had different processing steps applied. Interpretations should be made by considering all 3 figures. When estimating depth use the velocity you obtained above. Also recall that the image shows two way travel times for signals going directly from instrument to reflector and back.
Q16) Print the data plots. On any one of the 3 figures mark, and label, the locations you have chosen for the three features from Q13. Sketch the feature with your best guess for it’s size and location.
Q17) Compare the signal “character” from 40 - 100 ns over the first 4 metres of the line and from 40 - 100 ns over the last 3 metres of the line.
Q18) What do these “scattered signals” suggest about the consistency of materials in the subsurface under this line?
Q19) Identify the hyperbolic signal that you beleive to be caused by the pipe. Use the equation you derived in Q3 to determine the depth to the top of the pipe and the wave velocity in the top layer. Please show/explain how you came to your answers.
Q20) To finish with this data set, we should reflect upon how useful GPR was for learning about different aspects of the field site. For each geoscience task choose one value only. This is subjective, and a range of possible answers may be possible, but your choices should at least be consistent. For each task write brief a comment on why you chose the value you did.
