Tip of the month - #2
Understanding the limits of detection of GPR for Utilities mapping
The use of ground penetrating radar (GPR) is required to fulfill the requirements of The Survey Association’s (TSA) Level 5 and 6 surveys and can optionally be used for Level 2,3 & 4 surveys. The American Society of Civil Engineers (ASCE) has also published guidance on the collection and depiction of subsurface utility information which includes references to GPR and other surface geophysical methods.
GPR is a non-invasive microwave scanning technique used alongside radio frequency location (RFL) systems typically to verify known and/or to detect unknown services. Despite the best efforts of equipment manufacturers and software publishers (often one and the same), there is no such thing as a black box GPR system that will detect all buried utilities in all conditions. The fundamental parameters determining the success of a GPR survey are: the transmitted signal frequency, which for most GPR systems requires a selection of an appropriate antenna; the level of signal scattering and attenuation in the ground; the target size; and, its actual depth of burial. The more heterogeneous the ground and/or the more conductive the soil, the less likely it is that a target buried at depth will be detectable. For narrow diameter utilities, only high frequency signal has the resolution to reflect back to the receiver antenna. But high frequency GPR signal attenuates relatively quickly in the ground. So small diameter utilities at depth will not be detectable.
TSA and the ASCE suggest that for depths up to 2m in ‘good ground conditions’, the diameter of detectability of a utility reduces by 1mm for each 10mm of depth. So a 200mm pipe can in theory be detected at 2m depth and a 25mm pipe at 0.25m depth. Our experience and the results of work carried out on our utilities test site confirm that these estimates are best case scenario only applicable in homogenous dry soils.
Figure (a) shows the trade-off between signal detectability (so-called signal to noise ratio – SNR), target size and depth of burial for utilities in silty soils. This shows that the limit of detection of a 100mm diameter pipe is 0.9m in homogenous soils. However, detectability is determined by burial setting so a higher SNR would often be required for detection which could reduce the depth of detection to just 0.6m.
Site interpretation (so-called real-time surveys) of GPR data to locate utilities is usually not recommended due to the complexities of interpreting the various influences on the data. However, TSA Level 2,3 & 4 surveys can have site interpreted GPR aspects added. In this case another limit of detection is the setup parameters of the survey display screen and the visability of a survey screen in different lighting conditions.
Figure (b) shows radargrams from incorrectly (top) and correctly (bottom) setup GPR systems. In the incorrectly setup case the targeted features (outlined) are all but invisible. The improved setup below, with double the scans per metre and a reduced range, reveals the subtle, but important, features sought.
A full and comprehensive GPR survey sufficient to issue a permit to dig must be done by a qualified and experienced geophysical technician with a comprehensive understanding of GPR principles of operation, GPR systems and the effects of site conditions on the limits of detection.
Next month…Understanding the limits of detection of RFL for Utilities mapping
Rate this article out of 1-5 (5 being most useful):
