Ground Penetrating Radar tracing Improvised Explosive Devices and Unexploded and Military Ordnance

Abstract

The need to locate and identify buried explosive devices has grown in recent decades. Geophysical techniques, particularly ground-penetrating radar (GPR), are widely used for their speed and non-destructive investigation of the subsoil (Núñez-Nieto et al., 2014). However, detection remains challenging, as factors such as electrical conductivity, dependent on soil characteristics and climatic conditions, as well as object properties strongly affect performance (Abeynayake & Tran, 2016; Ebrahim et al., 2018; Rodrigues et al., 2022). Therefore, improving detection and identification methods is essential to making them more expeditious (Brito-da-Costa et al., 2022), more accurate, and reliable. This study analyses the effectiveness of GPR in detecting buried explosive devices in Portuguese soil.

For this purpose, 15 inert explosive devices (projectiles, grenades, mines, fuses, and an improvised explosive device) were buried at different depths (10, 20, 30, 40, 50, 60 and 70 cm) and orientations (horizontal and vertical). Data were collected using a GPR system (Noggin SmartCart Sensors&Software Inc. 250 and 500 MHz) at three different moments: twice at the end of July 2025 (before and after burial), with the 250 MHz antenna, and a third time in early September 2025 with both antennas. The readings were then processed using ReflexW software to create 3D models of the study site.

The results showed that GPR detected most buried objects despite the conditions. In the second sweep of July, 12 of the 15 targets were detected (250 MHz antenna), while in the third sweep (September), 12 were detected with the 250 MHz antenna and 14 with the 500 MHz antenna. Only one device (60mm Mortar Grenade) was not identified in either scan with the 250 MHz antenna, probably due to its small size (24.5 cm high and 6 cm diameter) and the electromagnetic properties of its steel casing. The coastal artillery shell was the only target not detected by the 500 MHz antenna, probably because the scan line did not pass exactly over the axis of the target, which was oriented vertically. Clearer and wider hyperbolas were observed in September, especially with the 500 MHz antenna, while the 250 MHz antenna showed greater penetration. In the September scans, increased background noise was also noted, becoming more pronounced with depth, making the data interpretation more difficult. This observation may be associated with external factors (e.g., changes in soil conditions) that may have occurred between sweeps. Regarding target orientation, small variations in the shape and intensity of the responses were observed, but without differences marked enough to confirm a clear trend.

GPR proved to be a promising tool for detecting buried explosive devices, even when in different orientations and depths. Nevertheless, it is essential to continue investigating the influence of these and other factors to refine the technique and increase the reliability of the results.