Article 1. Interpretation of GPR data (introduction).
One of the basic problems in gpr surveys is the data interpretation.
Often the customer of GPR does not have special knowledge in physics.
He wants to have a good user guide how to operate with device and examples of images for different kinds of underground objects.
He expects to gain basic skills in interpretation after some practice.
Unfortunately, it is not always so easy.
First, GPR image contains a lot of reflections from many underground objects, voids, grained soils, ... .
Also GPR image can include reflections from objects on the surface (iron fences, power lines, buildings, etc), especially with unshielded antennas.
In general, GPR image looks unlike to the real view of subsurface structures.
Second, the image quality depends on radar characteristics, antennas, probing parameters (antennas direction, step between points, density of tracks).
Third, GPR method does not give the full underground recognition itself because it is based on changes in electromagnetic properties only.
It is good to have additional information from geology, expected underground structures and objects, measurements obtained from other types of devices.
But there is a good news.
GPR can be used in many applications so a variety of GPR images is very big.
But if your tasks have specific area with restricted types of objects then it makes things easier.
For example, in concrete plates you have rebar, voids, cracks, plastic elements. You use very short antennas and coordinate sheet.
Another example - roadbed.
It does not mean that interpretation will always be easy. But after some practice in many cases you can do it good and in a short time.
Often 3D underground image is demanded. It means you need to assign coordinates to measurement points.
It can take considerable time because you cannot do this always in automatic mode. Inexpensive GPRS devices provide poor accuracy.
Grid of straight lines is not always possible. From time to time you must do measurements along curved lines.
If you want to view your measurement tracks on the map then you also must have assigned coordinates to the GPR data.
There are situations when it is demanded to make measurements with different antennas or directions on the same place.
Such GPR images can complement each other because you can see something on one image and do not see it on other image and vice versa.
For practical work it is important to have tools to make measurements and coordinate assignment quickly and convenient.
In this series of articles we do not take into account this aspect.
You can visualize GPR data along some track in different views. It can be 2D colored images in various palettes, sequence of narrow signals in points (wiggle),
forms of signals in separate points. You can visualize on 2D canvas points of local minimums and maximums (which show bounds of different soils or objects).
You can use different filters to transform GPR image.
All these manipulations help you to highlight underground structures and anomalies.
Remember that you can also see reflections and harmonic signals, not real objects.
Hardware tools have the primary value for GPR surveys.
Software tools also play important role in GPR surveys.
They provide convenience in data manipulations and transformations; help in recognition of underground structures;
calculate depths and diameters; make 3D views; create graphical presentations of GPR surveys in different forms and standards.
There are software tools for mathematical modeling of GPR. They help to view which GPR images will be created for different underground structures
according to the theory.
You can compare theoretical and real images. Also in models you can get better understanding for interpretation.
GPR image has time scale in the depth because it contains forms of reflected signals in different times.
The speed of signal propagation depends on the soils.
In common case you need to solve the inverse problem to get depth in meters.
It is very complex problem and it is unstable in solution.
In spite of the all difficulties the GPR survey is very practical and promising technology. Automatic recognition makes their first steps in this area.
Today interpretation of GPR images are made mostly manually by operator. Software products can include mathematical tools to make calculations.
For example, calculation of depth and electromagnetic permittivity by the markers placed on hyperbola image.
Also software products provide tools to put colored markers, draw bound lines for soils and objects.