Ground Penetrating Radar (GPR)

The last blog on jargon was driven by frustration about how far we, in the GPR community, have drifted from the fundamentals.  Our devices and methods are clearly not readily understood by the general user community. Teaching new users makes the problem abundantly clear.  Let’s take an example of someone describing their measurement instrument as a 100 MHz GPR.  Unless you are very knowledgeable, you would get a totally wrong or misleading impression of the measurement system.  I take most umbrage with the use of the term ‘100 MHz’.  We have allowed a totally inappropriate terminology to worm its way into our lexicon to the point that the terminology describing GPR devices is no longer related to the physics of the measurement process.

When I first started into the world of GPR 50 years ago, the term ground-penetrating radar did not exist!  In fact, the term used was ‘impulse’ radar.  When I investigated the term ‘impulse’, I was led to another term ‘baseband’ radar.  Baseband was more obtuse and was jargon from the electrical engineering world.

Impulse radar made sense from the physics of emitting an impulse.  Further, the real goal of GPR is to measure the ‘impulse response’ of the ground.  Again, this is jargon, but there is both solid and intuitive understanding that a system’s response to an impulse is a powerful way of describing a system and is directly related to the physics.  We see the impulse response concept regularly in the everyday world.  A loud bang (‘acoustic’ impulse) can create a series of echoes that we can hear, and these are, in fact, the impulse response of the surroundings.  We see this in the movies with submarines using sonar (acoustic pulses traveling in water to detect other vessels) and most boats that have marine echo sounders that detect water depth and/or fish.

The term ground-penetrating radar (GPR) became formally accepted at the international GPR users meeting held in Ottawa in 1988.  At an evening discussion, the major players in the field discussed what we should collectively use to describe the field and the method and agreed on ‘ground-penetrating radar’.  Since then there have a few pockets of resistance but, by and large, the name has become accepted.  Hence, the term impulse radar morphed into ground penetrating radar.  Of course, there was the usual editorial nit-picking from journals that insisted that the term be ‘ground-penetrating radar’ and not the unhyphenated form ‘ground penetrating radar’.  (As a side note, the Ottawa meeting was not part of the now biennial GPR conferences but produced one of the earliest compendia on GPR which was published by the Geological Survey of Canada as GSA Paper 90-4 ed. J.A. Pilon)

Whenever I hear the term GPR, I immediately associate it with an impulse radar device.  The device will display signal amplitude versus time (which is a surrogate for depth in the ground) and spatial position.  The display is a crude cross-section image of the subsurface and, if one delves deeper, it is a way of displaying the impulse response of the ground.

What this leads to is the understanding that GPR measures signals versus time (ground system response echoes appear at various delay times after pulse emission).  The system is characterized by a pulse being emitted by the transmitter.  In the early days of GPR, impulse radar systems were characterized by “pulse width”.  Pulse width was the time duration of the excitation pulse.  Early systems were described as 10 ns or 2ns or 1 ns impulse radars (1 ns = 1 billionth of a second).   Pulse width (when multiplied by the speed of light 3 x10⁸ m/s) characterized the physical length of the emitted pulse and indicated its spatial resolution attainable in the impulse response.

In my mind, the preceding descriptions have a sound physical basis and can be explained to the average user fairly quickly.  As you can clearly see, there is a lot of jargon here already.  While we should minimize jargon, jargon is a necessary part of all fields of technology; it is acceptable if it carries clear meaning that is generally understandable to the community with minimal difficulty.

Now we have some understanding of the meaning of GPR, what does ‘100 MHz’ mean?  Explaining this term takes a lot of explaining that brings in more jargon.  Before digging in, note that ‘100 MHz GPR’ should really be interpreted as a 10ns pulse width impulse radar.

We have already talked about time (which we all somewhat know, but do not necessarily know the subtleties).  MHz is a measure of frequency, a new term. The term means a signal repeats 1 million times per second.  The terms time and frequency both need clear definition.  The true meaning and in-depth understanding are profound.  I have inserted Wikipedia definitions here for reference. Surprisingly, the term Hz, which is the abbreviation for hertz (one oscillation per second), is not mentioned.

The complexity of time and frequency are self-evident from the definitions.

Refer Definitions on Wikipedia:

Time

Frequency

There is a connection between 10 ns impulse and 100 MHz, although not apparent; the period of a 100 MHz signal is 10 ns. This simple relationship is not the true explanation.  More in-depth understanding leads to understanding the concepts of Fourier analysis.  Suffice it to say that an impulse can be decomposed into a superposition of sinusoidal signals with a wide swath of frequencies.  A superposition of frequencies spanning a spectral width of 100 MHz will create a pulse with a time duration of 10ns.  The correct interpretation of 100 MHz is a bandwidth of 100 MHz, not a frequency of 100 MHz.  (Note that this paragraph contains an immense amount of jargon and calls for fairly in-depth knowledge of mathematical physics.)

From this limited description, which really skips over an immense amount of important detail, the use of frequency to describe a GPR is definitely subject to misinterpretation, is not intuitive to the average user, and takes a considerable amount of explanation.  There are literally thousands of texts on the subjects of time, frequency and Fourier analysis.  Attempting to provide a clear explanation is for another blog where we will try to provide clearer terminology definitions.  One can see there is certainly a lot of attraction to just saying what is implied, a 10 ns impulse.

There is no doubt a need for jargon in all specialty fields.  The field of GPR is no exception. My big concern is that we have allowed complex and mostly unnecessary jargon, math complexity and technical-field specialty factors make the field of GPR murky and opaque to non-specialists.  I will plead guilty to allowing this happen; I can also get caught in the jargon because it is easy. While I try to check myself, I do worry that some people deliberately add unmeaningful terms just to sound good and satisfy their own desire to appear knowledgeable.  Let’s all make a commitment to become more transparent and clearer in our dialog.  Simpler really is better!