Introduction
The survey, conducted on October 29, 2024, utilized DJI M350 RTK UAV and RadSys Zond Aero 500 GPR to collect 16 survey lines, each 50 meters long, with 1.0-meter line spacing.
The test site is located at the premises of Al Lisaili RC Flying Field, UAE. Access to the site was arranged by Dynatech Innovation, UAE.
Data processing and target detection were performed in Geolitix, a cloud-based platform for GPR processing, and SPH Engineering’s GeoHammer. To maintain precise altitude above the ground, a True Terrain Following system with radar altimeter and SkyHub onboard computer was used. Flight planning and execution were carried out using SPH Engineering's UgCS (Universal Ground Control System) software.
Access the processed data in the Geolitix demo »»»
Data was collected by Mussab Abdelrahim1, with the assistance of Aleksejs Sjomins2. Data processing and the report were provided by Maikls Andriksons2.
1 Dynatech Innovations, UAE
2 SPH Engineering SIA, Latvia
Disclaimer:
The term "detected" here means that the data interpreter has identified a strong enough signal that warrants further examination or action. Interpretations are subjective and will change depending on the experience and knowledge of the interpreter.
Neither SPH Engineering SIA nor Radar Systems Inc. makes any claims or warranties that detection of the same or similar targets is guaranteed under conditions other than at a particular test site using the same or different hardware, software, and workflow.
This report is provided "as is" and is intended to demonstrate the capabilities of the system described herein and provide some guidance for planning underground water pipe detection surveys.
Methods
The test site is an artificial lawn located at the premises of Al Lisaili RC Flying Field, UAE. (Location on Google Maps - Figure 1).
Data Processing
The initial GPR data processing was performed in GeoHammer, where unnecessary flight paths were removed, which simplifies the data set and improves accuracy. The cleaned GPR profiles were then uploaded to the Geolitix processing software. Consistent filters, gains, and corrections were applied uniformly to all profiles. A total of nine processing steps were completed in Geolitix to achieve the final processed data, as detailed in Table 1.
The dielectric permittivity of the medium was estimated using an average value of approximately 6, typical for dry sand. The hyperbolic reflection measurement method was not utilized due to the inability to accurately calculate wave propagation through the air (i.e., the distance from the Ground-penetrating radar (GPR) to the ground surface) and the medium (dry sand). As a result, the average dielectric permittivity value was adopted to ensure a reasonable approximation, ensuring consistent data interpretation. [5]
| Process | Applied values | Explanation | |
|---|---|---|---|
| Time cut | Remove below (40 ns) | The data below 1.5 m is unusable, so there isn't a reason to keep it; therefore, it is removed. | |
| Time zero correction | Method | Set travel time | The so-called “dead” time is removed as the signal is internal to the radar system and not related to the survey. The signal start time is set to ground level. |
| Travel time (ns) | 12 | ||
| Frequency filter | Method | Bandpass | Reduces background and instrument noise. |
| Cutoff freq. (lower) | 80 MHz | ||
| Cutoff freq. (upper) | 805 MHz | ||
| Filter order | 80 | ||
| Constant scale | 13.3 dB | Applies gain that facilitates further use for manual gain. | |
| Dewow | Sample window | 31 | Removal of inductive effects made the antenna known as “wow”. |
| Manual gain |  |
Allows the application of a gain curve for specific depth. | |
| Background subtraction | Method | Remove average | Eliminates horizontal lines in the data due to the antennas ringing and enhances the targets' visibility. |
| Traces window | 100 | ||
| Time range | 0 – 32 ns | ||
| Migration (FOR SLICING) |
Method | Hyperbolic summation | In order to create slice maps, migration is used to collapse the energy in hyperbolic reflections into a point. |
| Number of traces | 45 | ||
| Velocity source | Project velocity | ||
| Hilbert transform (FOR SLICING) |
No parameters used | Replacing each trace with its envelope, removing its negative component. This is done after migration! | |
Data visualization and results
After applying the corrections, nearly all GPR profiles displayed hyperbolas at a consistent depth parallel to the ground. Each hyperbola was individually selected for further data interpretation, as shown in Figure 2. Once the GPR data was processed in Geolitix, several maps were generated using QGIS 3.34 software.
For comparison, a splice map was created to estimate the depth of the underlying water pipes (Figure 3). To get a better understanding, a polyline interpretation of the identified hyperbolas was generated, with a 0.4-meter buffer zone added to approximate the potential pipe layout.
Results and Conclusion
The application of airborne ground-penetrating radar (GPR) has demonstrated significant potential for detecting underground utilities, such as plastic water pipes, saving time and effort compared to traditional methods. Typically, plastic pipes are challenging to detect with GPR due to their low dielectric contrast with the surrounding soil, which often results in weak or indistinct signals. However, in this survey, the detection of plastic pipes from the irrigation distribution network was notably unusual and suggests either particularly favorable soil conditions, optimal survey parameters, or a combination of both.
The analysis conducted using splice mapping, hyperbola fitting, and polyline interpretation produced comparable results, highlighting the reliability of these methods for utility detection.
Estimating the depth of underground utilities, however, is a complex process influenced by several factors, including flight altitude, geological settings, and the material properties of the utilities. When a permittivity value of 6 is assumed, the estimated depth of the pipes ranges between 0.4 and 0.7 meters.
References
[1] Prego F.J., Solla M., Puente I., Arias P. 2017. Efficient GPR data acquistion to detect underground pipes. NDT & E International. Volume 91, Pages 22-31.
[2] Kanemitsu T., Morifuji Y., Kubota K. 2024. Estimation of buried pipe depth in an artificial soil tank using ground-penetrating radar and moisture sensor. Journal of Applied Geophysics. Volume 220.
[3] Seyfried D., Jansen R., Schoebel J. 2014. Shielded loaded bowtie antenna incorporating the presence of paving structure for improved GPR pipe detection. Journal of Applied Geophysics. Volume 111, Pages 289-298.
[4] Zhou Y., W. L. Lai Wallace. 2024. Characterization of leakage signatures in buried water pipes by ground-penetrating radar(GPR) and instantaneous frequency analysis. Journal of Tunneling and Underground Space Technology. Volume 153.
[5] Guillemoteau J., Bano M., Dujardin J-R. 2012. Influence of grain size, shape and compaction on georadar waves: example of an Aeolian dune. Geophysical Journal International, Volume 190, Pages 1455-1463.