The reason it matters is simple. Every time contractors cut into concrete slabs, there’s a chance of hitting something unexpected. A severed post-tension cable can compromise structural capacity and create serious safety hazards. A cut electrical conduit can electrocute the operator. A damaged water line can flood an active floor. A U.S. industry study estimated the average cost of a single utility strike approximately $56,000 when repairs, downtime, and emergency response are included. That figure does not account for injury or loss-of-life risk.
GPR is the standard non-destructive method for scanning concrete before cutting, coring, drilling, or anchoring. It is one of the few methods that works in real time, from one side of the slab. It transmits electromagnetic pulses into the concrete and records the reflections that bounce back from embedded objects, producing a subsurface profile that shows what is inside and how deep it sits.
GPR concrete scanning
GPR sends short electromagnetic pulses into concrete at a specific center frequency. When the pulse hits a boundary between two materials with different dielectric properties, part of the energy reflects back to the antenna. The system records the amplitude and the two-way travel time of each reflection.
The key variable is dielectric contrast. Concrete has a relative dielectric permittivity (ε) of roughly 6 to 8 when dry and cured. Steel rebar has effectively infinite conductivity, so it reflects nearly all the incoming energy. PVC conduit, air voids, and water-filled cavities also differ enough from the surrounding concrete to produce detectable reflections, but with varying strengths.
The system plots these reflections as a B-scan: a cross-sectional view where the horizontal axis is the antenna's position along the surface and the vertical axis is depth (derived from two-way travel time). Individual targets like rebar or conduit appear as hyperbolic arcs. The shape and brightness of the hyperbola help a trained operator infer the likely target type and approximate depth, but the signal alone does not provide definitive material identification.
What the dielectric constant means for your scan
The dielectric constant controls the speed of the radar wave through the concrete, and the strength of reflections at material boundaries.
In dry, fully cured concrete (ε ≈ 6 to 8), the radar wave travels at roughly 110 to 130 mm/ns. In wet or freshly poured concrete (ε ≈ 12 to 15), that speed drops to around 80 to 90 mm/ns. This matters because GPR calculates depth from travel time. If you calibrate your system assuming dry concrete but the slab is saturated, your depth readings will be off by 30% or more.
Freshly poured concrete generally has a high dielectric constant and that value decreases during curing as free moisture is lost due to evaporation. These values are approximate and vary with mix design, ambient humidity, and curing method. Elevated interior slabs cured for years may sit at 5 to 7. Slab-on-grade structures typically settle around 6 to 8, but can be higher where moisture remains. Water tanks, reservoir walls, and any concrete with ongoing moisture exposure may remain in the 12 to 15 range while that exposure persists.
Depth calibration on every scan is not optional. The most reliable method is to find a target at a known depth (an exposed rebar end, a core hole, a verified construction drawing) and calibrate the dielectric from the measured travel time. Without calibration, typical GPR depth accuracy is ±15 to 20%. With proper calibration against a known target, depth accuracy can often approach ±5% under favorable conditions. SPH Engineering's GPR calculator can estimate detectability for a given antenna frequency, target depth, and material type before you even get to the site.
What GPR Can Detect in Concrete
GPR concrete scanning identifies a range of embedded objects and structural features. A trained operator can typically locate the following, though detection confidence varies from site to site.
Rebar (reinforcing steel bars)
The highest-contrast target in concrete. Steel reflects virtually all GPR energy, producing strong, well-defined hyperbolas. GPR can map rebar position, spacing, depth (cover thickness), and the orientation of the reinforcing grid.
What GPR cannot reliably determine is bar diameter. Estimating rebar size from signal amplitude alone is unreliable because amplitude depends on cover depth, concrete condition, and surrounding reinforcement density as much as bar size. If you need diameter, you need to expose the bar or use a complementary method.
Post-tension cables
Both bonded and unbonded post-tension cables produce strong reflections. These are critical to identify before coring because severing a stressed cable can release high stored force and compromise structural capacity.. On a GPR scan, post-tension cables often appear as continuous linear reflections running across the slab, which distinguishes them from the regular grid pattern of conventional rebar.
Electrical conduit (metallic and non-metallic)
Metal conduit reflects strongly, similar to rebar. PVC and plastic conduit produce weaker reflections because the dielectric contrast with concrete is smaller, but they are still detectable in most conditions, particularly when the conduit contains wiring (the copper acts as a secondary reflector). Empty plastic conduit at greater cover depths or in heavily reinforced slabs can become difficult to distinguish.
Voids and delaminations
Air voids create a strong dielectric contrast with concrete (air ε ≈ 1 vs. concrete ε ≈ 6 to 8). Large voids, greater than approximately 50 mm diameter with 1 GHz antenna, are typically visible. Thin delaminations are more challenging and depend on the gap width and whether moisture has infiltrated. Water-filled voids can produce stronger reflections than air-filled ones, because the dielectric contrast between water (ε ≈ 80) and concrete is extreme.
Concrete slab thickness
GPR can measure total slab depth by detecting the reflection from the bottom of the concrete (the concrete-to-subbase interface). This works best when there is a clear dielectric contrast at the bottom surface, such as concrete on compacted gravel or soil. Concrete-on-concrete interfaces, such as an overlay on an existing slab, are harder to resolve because the dielectric contrast is minimal, but in certain scenarios it can be achieved.
Concrete overlay and cover thickness
Measuring the depth of the first rebar layer gives you concrete cover. Measuring the reflection from a material change near the surface gives you overlay thickness. Both are standard GPR concrete scanning outputs used for structural assessment.
Dowel bars and tie bars
In pavement and joint applications, GPR can locate dowel bars and assess their alignment, spacing, and depth. This is a standard application in highway and airport pavement evaluation.
Mesh reinforcement (welded wire)
Detectable, though the smaller wire gauge produces weaker reflections than standard rebar and can be missed if the resolution is low. In slabs with both mesh and rebar layers, the mesh can sometimes create a semi-continuous reflector that partially obscures deeper targets.
What GPR cannot reliably do in concrete
No technology provides complete visibility in all conditions, and understanding GPR's limits is just as important as knowing what it can and cannot do.
Identify specific utility types from the signal alone
GPR tells you something is there and how deep it is. It does not tell you whether a conduit carries electrical wiring, water, or gas. That determination comes from construction documents, field context, and sometimes operator experience. The GPR signal from a metal water pipe and a metal electrical conduit of the same diameter at the same depth will look identical.
Determine rebar diameter with precision
Amplitude-based diameter estimation exists in some software tools, but it is influenced by too many variables (cover depth, concrete moisture, adjacent reinforcement) to be treated as a reliable measurement. If you need bar diameter, you need to expose the bar or use complementary methods.
See through heavily congested reinforcement
When rebar is densely spaced (less than approximately 75 mm center-to-center), the top layer of steel reflects so much energy that very little signal penetrates deeper layers. This creates a "shadowing" effect that effectively blinds the GPR to anything below the first reinforcement mat.
This is one of the most common practical limitations on real job sites. Heavily reinforced bridge decks, transfer slabs, and structural walls all suffer from it and it is the primary reason X-ray still has a role in concrete scanning although the GPR method is much cheaper, does not have limitations due to strict licensing as X-ray method has and doesn’t have necessary nighttime closure.
Scan through metal deck or metal formwork
If the bottom (or top) of the concrete has a continuous metal layer, GPR cannot penetrate it. The metal reflects nearly all of the signal. Concrete on metal decks is generally opaque to GPR.
Scan freshly poured or saturated concrete
Concrete that is fresh, highly moist, or insufficiently cured can have a dielectric constant high enough to severely attenuate the GPR signal. The radar pulse may scatter and reflect near the surface rather than penetrating to useful depth. The same applies to slabs that are continuously wet: pool decks, water treatment structures, any concrete exposed to ongoing moisture intrusion. Schedule GPR scans during dry conditions and wait until the concrete has cured sufficiently.
Detect very small targets at depth
As a rule of thumb, targets substantially smaller than about half the antenna wavelength . For a 1.6 GHz antenna in concrete (wavelength ≈ 70 to 75 mm), the minimum detectable target is roughly 35 to 40 mm under favorable near-surface conditions, growing larger with depth as the signal weakens. Small fiber optic ducts, thin PVC sleeves and hairline cracks are often below the practical detection threshold.
GPR antenna frequency selection for concrete
Frequency selection is one of the most important decisions in concrete scanning. Higher frequencies sharpen the image at the cost of penetration depth. Go lower and you reach deeper but lose the fine detail.
For most concrete scanning, you are working with antenna frequencies between 1 GHz and 2.6 GHz. The trade-off in practice:
- 2.0 to 2.6 GHz (high frequency): Best resolution. Detects small targets (rebar, small conduit, thin delaminations) in slabs up to roughly 300 to 400 mm thick. This is the standard range for structural concrete inspection, rebar mapping, and pre-core scanning on typical building slabs and walls. Most handheld concrete scanners operate here.
- 1.0 to 1.6 GHz (mid frequency): Moderate resolution with deeper penetration, typically reaching 500 to 800 mm under favorable dry conditions. Better suited for thicker structural elements like bridge decks, retaining walls, and foundation elements. You sacrifice some fine detail for the ability to see deeper targets.
- 500 to 1000 MHz (lower frequency): Penetration in dry concrete can approach 1 to 1.5 meters at the lower end of this frequency range, but resolution drops significantly. Useful for measuring total slab thickness in heavy structural elements or locating large embedded objects (major conduit runs, large voids, sub-slab utilities) in thick foundation mats. Not suitable for detailed rebar mapping. SPH Engineering's Zond Aero 500 and Zond Aero 1000 operate at 500 and 1000 MHz, respectively, and can be deployed on a terrestrial cart or mounted on a drone, making them relevant for scanning thicker concrete elements and large structural assessments where lower frequency, larger footprint GPR is appropriate..
In practice, use the highest frequency that still reaches your target penetration depth. There is no benefit to running a 500 MHz antenna on a 200 mm slab when a 2 GHz antenna will give you far sharper results. For most pre-core/pre-drill work, a 1.6 to 2.0 GHz system covers the vast majority of applications. Step down to lower frequencies only when the structural element exceeds 500 mm or when you need deeper penetration and can accept lower resolution. Keep in mind that dense reinforcement can still block deeper targets or layers.
Modes of GPR concrete scanning: line scans vs. area scans
GPR concrete scanning can be performed in two modes. The right choice depends on what information you need.
Line scan (B-scan)
The antenna is moved in a straight line across the surface. The result is a single cross-sectional profile showing targets along that line. Line scans are fast and effective for quick checks before individual core holes or anchors. They answer the question: "Is there anything directly in the path of my drill at this specific point?"
But line scans have a significant limitation. A target that runs parallel to the scan direction may not produce a clear hyperbola because the antenna never crosses it at an angle. Experienced operators always scan in at least two perpendicular directions at each location.
Area scan (C-scan / depth slice)
The operator collects multiple parallel line scans in two perpendicular directions over a grid. Software combines these into horizontal depth slices, showing a plan-view map of all targets at a given depth. Tools like GeoHammer apply background removal and gain adjustment to produce interpretable depth slices from raw scan data. This produces a much more complete picture than individual line scans and is the standard approach for larger scan areas, structural assessment, and documentation.
Area scans reveal rebar grid patterns, conduit routing, void locations, and slab thickness variations across the entire scanned area. They are essential when the goal is to produce a full reinforcement map or to find safe zones for multiple penetrations across a floor section.
The trade-off is time. A line scan takes seconds, and a full area scan over a 2 m × 2 m grid can take 15 to 30 minutes depending on scan spacing. For pre-drill spot checks, line scans are usually sufficient. For renovation planning, structural review, or post-tension cable mapping, area scans are worth the investment.
Scanning vertical surfaces and ceilings
GPR is not limited to horizontal floor slabs. The technology works on any concrete surface where you can maintain antenna contact, such as walls, columns, beams, and the underside of elevated slabs.
The physics are the same. The antenna still requires physical contact (or very close proximity) to the concrete surface to couple the signal effectively. The practical challenges are ergonomic. Holding a scanner against a ceiling slab while maintaining consistent contact pressure and scan speed takes practice and, in many cases, a second operator to monitor the display.
Applications include locating rebar in concrete columns before cutting openings, mapping conduit in walls before installing anchors, and scanning the underside of a floor to plan core locations from below. Some manufacturers offer extension poles and remote display tablets to make overhead scanning less physically demanding.
GPR vs. X-ray for concrete scanning
X-ray (radiography) is the other major technology used to image the interior of concrete. The comparison matters because project teams often need to decide between them.
GPR advantages over X-ray
GPR is single-sided. You only need access to one surface. X-ray requires access to both sides of the element (source on one side, film or detector on the other). On a slab-on-grade, the underside is buried. On a wall shared with an occupied space, X-ray requires evacuating the area behind it due to radiation exposure. GPR typically requires no safety zone (if the survey area isn’t hazardous), no radiation license, no evacuation. It can be operated during active construction.
GPR is also faster. A skilled operator can often clear a core location in under a minute with a line scan. X-ray requires setup, film or detector placement, exposure, and development or readout, taking 10 to 30 minutes per exposure depending on whether the system is film-based or digital.
X-ray advantages over GPR
X-ray can produce a radiographic shadow image of every object embedded in the concrete. It is less affected by rebar density the way GPR is. If you need to see through a heavily reinforced transfer slab with multiple rebar layers and post-tension cables, X-ray will often resolve multiple layers that GPR may only show partly. X-ray can also sometimes differentiate materials more clearly: the shadow of a copper pipe looks different from a PVC pipe, which looks different from rebar.
When to use which
For most pre-drill and pre-core work on accessible slabs, walls, and ceilings, GPR is the primary tool because it is faster, safer in occupied areas, requires no evacuation, and provides results in real time. Reach for X-ray when GPR cannot provide sufficient confidence, particularly in heavily congested areas where shadowing limits visibility, or when you need to positively identify the type of embedded object.
In practice, the majority of concrete scanning is performed with GPR. X-ray is reserved for high-stakes situations or verification when GPR results are ambiguous.
Factors that affect GPR scan quality in concrete
Even with the right equipment and frequency, several site conditions influence the quality of your GPR data.
Concrete moisture content is the single biggest variable. Wet concrete attenuates the radar signal, reduces penetration depth, and can produce misleading reflections near the surface. Fresh or insufficiently cured concrete or concrete in continuous contact with water will significantly degrade scan quality. If possible, wait for dry conditions.
Surface condition matters because GPR relies on antenna coupling. Rough, spalled, or heavily textured concrete surfaces reduce the energy transferred into the slab. Smooth, clean surfaces produce the best coupling. Thin coatings (paint, epoxy, tile) generally do not affect the scan. Thick toppings (mortar screeds, thick adhesive layers) may introduce additional reflections that complicate interpretation.
Reinforcement density creates the shadowing problem described earlier. In lightly reinforced slabs (150 to 200 mm rebar spacing), GPR can usually see through the first layer to detect deeper targets. In heavily reinforced sections (75 mm or less), the first layer acts as a near-continuous reflector that blocks deeper penetration. There is no equipment setting that fixes this. It is a physics limitation.
Concrete aggregate type can influence signal propagation. Conventional limestone and granite aggregates are relatively favorable for GPR. Certain aggregates with high iron content or slag-based aggregates can increase signal attenuation. This is uncommon, but worth noting for specialty concrete mixes.
Temperature has a minor effect under normal conditions. Extremely cold concrete (below freezing) can actually improve GPR penetration slightly because frozen pore water has a lower dielectric constant than liquid water. A niche consideration, not a typical scan variable.
Interpreting GPR concrete scan data
GPR interpretation in concrete is not automatic. The raw data is a grayscale image of electromagnetic reflections, not a photograph of the embedded objects.
- Hyperbolic reflections are the signature of point targets: rebar, conduit cross-sections, small objects. The apex of the hyperbola marks the lateral position of the target. The depth of the apex (from two-way travel time, corrected for dielectric constant) gives the target depth. The shape of the hyperbola, specifically how quickly the tails spread, gives information about the radar wave velocity in the concrete, which can be used to refine depth calibration.
- Linear reflections running parallel to the scan direction indicate targets aligned with the antenna path. Typically a conduit run, post-tension cable, or construction joint. These show up as bright horizontal bands rather than hyperbolas because the antenna stays above the target for the entire pass.
- Flat, continuous reflections across the full scan width usually indicate a material interface: the bottom of the slab, a concrete-to-subbase boundary, a bond line between an overlay and the original slab, or the top of a metal deck.
- Amplitude differences can support material interpretation, but they do not definitively identify materials. Steel produces the strongest reflections. PVC and plastic can produce weaker to moderate reflections, mainly due to contrast of the medium the pipes are located in. Air voids produce reflections that are strong but with a characteristic polarity reversal (the reflected wave flips phase because the radar wave moves from a higher-dielectric material to a lower one). Water-filled voids produce extremely strong reflections with no polarity reversal.
The critical skill is distinguishing real targets from artifacts. Ringing (multiple reflection echoes from a single strong reflector), side reflections (energy returning from targets not directly below the antenna), and coupling artifacts (noise from the antenna-surface interface) all appear in GPR data. Processing software can apply migration algorithms to collapse hyperbolic tails into point targets, cleaning up the display. But software does not replace the operator's judgment on ambiguous targets.
Operator training and experience matter as much as equipment quality. The same GPR system in the hands of a novice operator and an experienced geophysicist will produce very different levels of useful information. If you are building GPR interpretation skills, SPH Engineering offers a self-paced GPR Basics course covering data acquisition, scan types, and interpretation fundamentals.
Relevant standards and guidelines for GPR concrete scanning
GPR concrete scanning is governed by several industry standards that define equipment requirements, procedures, and reporting.
- ASTM D6432 (Standard Guide for Using the Surface Ground Penetrating Radar Method for Subsurface Investigation) provides the general framework for GPR survey methodology, including equipment specifications, data collection procedures, and limitations.
- ASTM D6087 (Standard Test Method for Evaluating Asphalt-Covered Concrete Bridge Decks Using Ground Penetrating Radar) is specific to bridge deck evaluation but contains relevant methodology for concrete scanning with GPR.
- ACI 228.2R (Report on nondestructive Test Methods for Evaluation of Concrete in Structures) covers GPR alongside other NDT methods (ultrasonic pulse velocity, impact echo, rebound hammer) and provides guidance on when to use each method and how to interpret results in a structural assessment context.
- AASHTO (American Association of State Highway and Transportation Officials) provides guidance on GPR application to highway pavements and bridge decks. Consult the current edition, as designations have been updated over the years.
For concrete cutting and coring safety specifically, most jurisdictions and trade organizations reference the Post-Tensioning Institute (PTI) guidelines, which require scanning before any penetration of post-tensioned slabs.
When GPR is not enough: complementary methods
GPR has blind spots. Knowing when to bring in another method is part of the job.
- Electromagnetic (EM) covermeters are simpler, lower-cost tools that detect ferromagnetic materials (steel rebar) and measure cover depth. They work well for quick rebar cover checks but cannot detect non-metallic targets: PVC conduit, plastic pipes, voids. They also struggle with accuracy when multiple rebar layers are present. EM covermeters complement GPR by providing a fast cross-check on rebar cover measurements.
- X-ray radiography, as discussed above, is the fallback when GPR cannot see through dense reinforcement or when positive material identification is required.
- Impact echo is an acoustic method that detects delaminations, voids, and slab thickness by analyzing the resonant frequency of stress waves. It is better than GPR at detecting thin delaminations and honeycombing in concrete because it responds to mechanical (not electromagnetic) property differences. Impact echo is often recommended for delamination mapping. It works, but the learning curve is steep compared to GPR for most operators, and coverage rates are much slower.
- Ultrasonic pulse velocity (UPV) measures the speed of sound through concrete, which correlates with concrete quality, density, and the presence of internal defects. It is a complementary method for structural assessment, but it does not provide the object-locating capability of GPR.
- Concrete coring and exposure is the only way to get definitive information. When GPR, X-ray, and other methods still leave uncertainty about what is inside the slab, extracting a core or chipping away cover to expose the target is the final verification step.
In practice, most concrete scanning projects rely on GPR as the primary method and call in supplementary techniques only when results are inconclusive or the stakes justify the additional cost and time.
Where lower-frequency GPR fits in concrete work
Most concrete scanning discussion focuses on handheld scanners in the 1.6 to 2.6 GHz range. But there is a class of concrete scanning problems where those frequencies are not enough.
Thick foundation mats, mass concrete pours, bridge piers, dam faces, and retaining walls can exceed 500 mm to over a meter in thickness. A 2 GHz scanner will not penetrate to the back face. Sub-slab utility detection, where you need to see through the concrete and into the ground below, is another case. Large-area structural assessment is another case, especially where scanning a 10,000 m² warehouse floor with a handheld unit is tedious and impractical.
For these applications, GPR systems in the 500 to 1000 MHz range become relevant. The Zond Aero 500 and Zond Aero 1000 operate at 500 and 1000 MHz respectively and can be deployed on a terrestrial cart for floor and pavement scanning, or mounted on a drone for large-area coverage where site geometry and regulations allow. The same Zond Aero platform can support both cart-based and drone-based surveys, reducing the need for separate systems.Data from these surveys can be processed in GeoHammer for field QC (background removal, gain adjustment, quick depth slices) and in Prism2 or Geolitix for full processing including migration, filtering, and 3D visualization. For construction site applications where GPR is one of several survey technologies deployed on the same drone platform, SPH Engineering’s UgCS and SkyHub handle mission planning and sensor integration.
The single most common mistake
Failing to calibrate the dielectric is one of the largest avoidable sources of depth error in concrete scanning.
An operator who picks the right frequency, scans in two directions, maintains good surface contact, and reads the position of hyperbolas correctly will still report wrong depths if the dielectric constant is off. On many job sites, it is left unchecked. The system ships with a default value (often ε = 5 or ε = 9), and that default gets used whether the slab is a 30-year-old elevated interior or a 6-week-old slab-on-grade sitting on damp soil.
Calibrate against a known target on every job. If no target is available, at minimum verify the dielectric by measuring slab thickness at a free edge and comparing the GPR reading to a tape measurement. It takes two minutes in favorable cases and reduces the risk that every depth estimate on the job is systematically wrong.
Scanning concrete with GPR before cutting, coring, or drilling is more than a precaution. It is a standard of care. The cost of a scan is usually small compared with the cost of severing a post-tension cable, cutting a live electrical conduit, or breaching a pressurized water line in an occupied building.