2:35 PM - 2:55 PM
[GP-07] Urban tree root detection using GPR -- lessons from practical surveys and forward models
The use of GPR to detect tree roots is already present multiple times in the literature. However, many of the peculiarities of tree root detection have not been studied in detail.
Urban GPR studies are of particular interest since there is a stronger financial incentive for study, as tree roots can cause damage to existing infrastructure (by cracking roads or sidewalks, exacerbating pipe leaks, causing subsidence). At the same time, urban trees provide quantifiable environmental services, which suggests GPR could realistically be implemented in urban tree management strategies if roots can be reliably detected thusly.
While in principle the matter of detecting tree roots seems straightforward, in practice, this is not always the case. Here, we present direct measurements on root permittivity and forward FDTD models that show that the contrast between roots and surrounding soils is not always strong enough to guarantee detection. This could help explain why some roots elude detection, even coarser roots. The nature of paved surfaces may facilitate detection more than urban soils, but here too, forward models suggest it may not only be the root itself that is detected, but also the cracks it produces as it penetrates the paved surfaces.
Several practical surveys are presented confirming these ideas. Tree root surveys on paved surfaces are scarcer in the literature, despite being arguably more important, our work serves to complement some of these gaps in the literature. We present ERT as an unlikely but potentially efficient complement of GPR in urban tree root detection.
Since urban surveys also often have practical constraints in terms of data acquisition time (surveys can require closing off streets or sidewalks), we also offer a few brief best-practice indications for practical surveys.
Urban GPR studies are of particular interest since there is a stronger financial incentive for study, as tree roots can cause damage to existing infrastructure (by cracking roads or sidewalks, exacerbating pipe leaks, causing subsidence). At the same time, urban trees provide quantifiable environmental services, which suggests GPR could realistically be implemented in urban tree management strategies if roots can be reliably detected thusly.
While in principle the matter of detecting tree roots seems straightforward, in practice, this is not always the case. Here, we present direct measurements on root permittivity and forward FDTD models that show that the contrast between roots and surrounding soils is not always strong enough to guarantee detection. This could help explain why some roots elude detection, even coarser roots. The nature of paved surfaces may facilitate detection more than urban soils, but here too, forward models suggest it may not only be the root itself that is detected, but also the cracks it produces as it penetrates the paved surfaces.
Several practical surveys are presented confirming these ideas. Tree root surveys on paved surfaces are scarcer in the literature, despite being arguably more important, our work serves to complement some of these gaps in the literature. We present ERT as an unlikely but potentially efficient complement of GPR in urban tree root detection.
Since urban surveys also often have practical constraints in terms of data acquisition time (surveys can require closing off streets or sidewalks), we also offer a few brief best-practice indications for practical surveys.
Abstract password authentication.
The password was sent to the registered participants by e-mail in advance.