Abstract
The use of non-invasive methodologies is becoming essential for archaeological research, and ground penetrating radar is one of the most important techniques to obtain high resolution information. In this paper we present the analysis of a full 3-D GPR dataset integrated with a high-resolution photogrammetric survey acquired in a Roman archaeological site located in Aquileia (Northeast Italy) within the partially excavated area known as “Fondo Pasqualis”. We evaluated the importance of dense and accurate data collection and of processing of the GPR signal for characterization of the archaeological features. We further discuss the parametrization and the applicability of GPR attributes, in particular amplitude-based and coherence attributes, to better identify and characterise the archaeological buried targets. Furthermore, autopicking procedures for isosurfaces mapping were critically evaluated with the objective of detecting complex structures. The final interpretation of all the GPR features, with the support of digital terrain modelling and orthophotos from unmanned aerial vehicles, guided the archaeologists to open and excavate newly selected areas, which revealed interesting structures and contributed to the understanding of the historical events that characterized the Aquileia city.
Highlights
Ground penetrating radar (GPR) is probably the most popular geophysical technique for archaeological exploration [1] mainly due to its resolution, which is higher than that of the other available techniques, and its high versatility [2]
There are several other available geophysical techniques used for archaeology, in particular electrical resistivity tomography (ERT) [3], magnetometry [4], infrared thermography [5], and multi-spectral surveys [6], but GPR has demonstrated its wide versatility since it has been applied at different steps of archaeological prospecting and at various scale levels, assuring resolution higher than that of the other methods
A velocity of 7 cm/ns produced a sufficient imaging of the circular structure, while the semi-circular one cannot be detected; the opposite occurs for a velocity of 12 cm/ns, demonstrating how it is difficult to obtain a reliable interpretation when the velocity field is not correct and when spatial sampling is not adequate [44]
Summary
Ground penetrating radar (GPR) is probably the most popular geophysical technique for archaeological exploration [1] mainly due to its resolution, which is higher than that of the other available techniques, and its high versatility [2]. GPR surveys encompassed just a few single separated common offset (CO) profiles [10], later, GPR was improved with mono- or bi-directional grids of CO profiles In the latter acquisition scheme, there is a higher spatial resolution and some data redundancy, in turn allowing in some cases for the gathering of information even when complex and superimposed structures are present. Time- or depth-slices represent a rapid and efficient way to provide planar comprehensive views of the anomalies’ patterns They are essential, especially for the analysis of large areas, but they have intrinsic constraints in the case of great subsurface complexity such as when there are different superimposed structures or non-regular features. As a matter of fact, present standards of 3-D GPR in archaeological prospection are based on “pseudo 3-D” volumes obtained by combining a series of 2-D profiles [16,17], so the expression “2.5-D” seems to be more appropriate than the “3-D” for that specific acquisition and processing approach [18]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
