Ground penetrating radar applied to archaeology (2006-present)

High-resolution imaging of buried Roman remains by means of a modular ground penetrating radar array.

 Georadar-slice (depth: c. 0.60 - 0.65 m) and interpretation of Ammaia (Portugal)
Georadar-slice (depth: c. 0.60 - 0.65 m) and interpretation of Ammaia (Portugal)
Ground Penetrating Radar (GPR) prospection is, along with magnetometer prospection, earth resistance survey and prospection with sensors based on electromagnetic induction, one of the most widely used geophysical prospection techniques in archaeology. This method is based on the reflection of radar waves by buried archaeological structures. Measuring the time interval within which a wave returns at the surface, allows estimating the depth of the features.

As a consequence, the most important advantage of this technique is that the information is three-dimensional: one can also have an idea about the depth of the structures. Moreover, GPR prospection allows identifying underground structures with an unequalled spatial resolution. It is capable of detecting a broad range of archaeological structures: walls, floors, ditches, graves, voids, conduits etc. Measurements underneath paved surfaces (floors in houses or churches, asphalt of car parks) are also possible. An important drawback is the limited depth penetration when prospecting soils with a high clay content.

Since 2007 this technique has successfully been used by Dr Lieven Verdonck on archaeological sites from the Bronze Age until the Middle Ages. The employability of GPR has been demonstrated in several regions in Europe and beyond, including sandy Flanders (Koekelare, Aalter, Middelburg), Kent (Bishopsbourne, Petham, Ickham) and the United Arab Emirates (Mleiha). Nevertheless, the main focus of research is on rural and urban sites in the Mediterranean, dating from the Punic, Greek, Hellenistic and especially the Roman periods. Prospections were conducted in Bidnija (Malta), Ammaia (Portugal), Potentia and Trea (Italy), Mariana (Corsica), Thorikos and Koroneia (Greece). The research is embedded in landscape archaeological projects in which the GPR data are confronted with data from other kinds of archaeological prospection (e.g., field walking, remote sensing, topographic survey) and small-scale excavations. At the moment, the emphasis is on the investigation of two Roman towns in Lazio (Italy): Falerii Novi and Interamna Lirenas. This occurs within the project ‘Beneath the surface of Roman Republican cities: a novel application of landscape-scale ground-penetrating radar (GPR) survey in archaeology’ (2015–2017), in collaboration with the Faculty of Classics, University of Cambridge. The objective is to gain more understanding in the origins of Roman cities in Italy in the fourth–third centuries BC. The survey covers a total area of c. 60 ha.

The methodological perspective of the research includes among others the determination of the optimal GPR sample density. This is dependent on field conditions, although mostly a distance between the measurements of 10 cm in two perpendicular directions can be proposed as optimal. Furthermore,  experiments are being conducted regarding the semi-automatic interpretation of GPR data. Moreover, the benefits of multi-offset measurements (i.e., measurements with multiple distances between transmitter and receiver antennas) are being investigated.


Dr Lieven Verdonck