About GPR


With a minimal amount of training and practice it is possible to become proficient at reading raw GPR data in less than a day.

First, it is essential to understand that the radar signal spreads in a fan shape when it is transmitted. Because of this, an object will be visible to the radar before and after the radar is directly over it. This is the reason that a point-shaped object will show up as a hyperbola (arc shape). Since the radar signal will always have the shortest time to travel when the antenna is directly over the target, the center line of the target will always be at the highest point of the hyperbola in the data.


An example of unprocessed 2D data of two pipes:



3D data can be represented in one of 3 ways: either a 3D alignment of 2D traces, one or more depth slices, or isosurfaces. A 3D alignment of 2D traces requires almost no post-processing thus requiring less time to produce and the least amount of assumptions regarding velocity variation. Depth slices require an accurate model of the velocity of the medium being investigated. The more consistent the materia is (i.e. concrete), the quicker and easier it is to achieve this. In some cases, velocity can be measured and sometimes, this can be worked out through a combination of educated guesses and trial and error. Depth slices tend to be good for modeling linear features such as rebar and conduit. Isosurfaces require the most amount of post-processing and filtering. The result can be good for modeling more complex features, but also have a tendency to filter out smaller and fainter features. Sometimes this is desirable, and sometimes it isn't.

Blah blah blah I'm a dirty tramp
Don't do this to me

A 3D alignment of 2D slices

A fiber optic line is clearly visible in the data set when a simple amplitude filter is applied.


Depth Slice

Isosurface Model


There are three main approaches to surveying with GPR the selection of which depends on the desired results and whether real-time results are required or if post processing is desired.

If the goal is to identify one or more specific targets, the easiest way to achieve this is to examine the site to be surveyed for any clues such as manholes, catch basins, valves, etc. If there are any, they can serve as an excellent starting location for the investigation. Typically, in this approach, one would move the radar across the medium being investigated until they detect the object on screen, they would then determine the precise center of the object and either mark this on the ground or log the GPS coordinates of the point. At this point, one would move over and repeat this process essentially tracing the target. This works best when real time results are required.

A mark out is generally accomplished by scanning a site on a grid pattern. When a target is observed, a mark is placed on the ground by the operator. This is usually in the form of a flag or spray paint mark. Since the radar data will reveal only that a target is in the earth which has a different composition than the surrounding material, it is not possible on only one pass to determine if the target is a rock or a utility or other type of target. The easiest way to differentiate them is to make another pass and see if the target continues or not. One can choose to either place marks at every target and connect the dots later or to immediately move over and make another pass on the target to determine if the target is linear or not.

3D data can be useful for more complex sites with many targets. The generation of 3D data requires that data be collected on a regular grid in perpendicular directions and also usually requires some degree of post-processing. The amount of post processing required increases as the uniformity of the medium being investigated decreases. Also, there is a practical correlation between the uniformity of the medium being investigated and the clarity of the images which can be expected to be produced through post-processing. Furthermore, usable 3D presntations usually require that data be collected on a much denser grid than is necessary with 2D data presentation. In many cases, the number of survey lines is doubled or quadrupled. For these reasons, 3D data tends to be used more often on smaller scale surveys of concrete floors and walls than it is on large scale ground surveys.


There are two basic configurations of the GPR: Cart and handheld. Handheld units are all in one units which contain either 1000 MHz or 2000 MHz antennas or both. Due to the characteristics of these antennas, they are ideal for scanning floors and walls, but generally do not offer sufficient depth to locate exterior buried utilities. Cart systems are modular and expandable. They can be used with a variety of antennas ranging from 100 MHz to 2000 MHz. They be configured to be handheld for walls and floors or cart-based. They can even be configured to interface to almost any available GPS unit.

  • Pull System - 100 MHz antenna
    Pull System - 100 MHz antenna
  • Cart System - 500 MHz Antenna
    Cart System - 500 MHz Antenna
  • Handheld System - 1000 & 2000 MHz Dual Frequency Concrete Scanner
    Handheld System - 1000 & 2000 MHz Dual Frequency Concrete Scanner


The Quantum Imager, Q series and Seeker SPR can be integrated with almost any type of GPS from handheld to RTK. For most applications, GPS positions are logged constantly while surveying. In the field, or at the office, as points of interest are identified, they are logged for future reference and can have numbers and descriptions assigned to them. These points can then be exported for use in spreadsheets, databases, GIS and CAD software. For certain specialized applications such as lakes and terrain with very irregular topography, GPS positioning can be used to trigger each scan made by the system. High accuracy RTK GPS systems are usually the type of GPS system which yields the best results for these applications.

  • software_googleearth
  • Geotag_screen
  • GPS_Mount_dirty
  • Ground Penetrating Radar Maps


Electromagnetic emissions from ground penetrating radar systems manufactured by US Radar Inc. do not constitute a safety or health hazard under normal operating conditions. The emissions are far below the 10mW/cm² (100W/m²) level specified by United States Occupational Safety and Health Administration (OSHA) regulations and similar regulations in other jurisdictions.

Here is the average power density data at 50mm:

Antenna Model

Average Power Density

(W/m² @ 5cm)

OSHA Spec.



< 0.001



< 0.001



< 0.001






< 0.001



< 0.001



< 0.001