Prehistoric mines are often too large and too deep for conventional archaeological excavations. Non‐destructive and minimally invasive methods of prospection can help to overcome these limits. Our case study of a Late Bronze Age opencast mine (ca. 1050 to 780 BC) shows the potential of geophysical prospection methods combined with core drillings. For the reconstruction of this mine, we combined electrical resistivity and induced polarization (IP) tomography, seismic refraction tomography (SRT) and ground penetrating radar (GPR). The geophysical data were collected based on an orthogonal grid of 10 longitudinal and transverse profiles, laid out over an area of ~330 × 300 m. The profiles allowed a three‐dimensional interpolation of the geological units, the mining dumps, the mining areas and the residual mineralization. Additionally, two deep cores were drilled to ground‐truth the geophysical prospection results. They provided information about the stratification at intersections of the measurement grid, and this proved crucial for validating the interpreted geophysical profiles. Each geophysical method applied provided different information for the reconstruction of the site: the electrical resistivity tomography offered the best clues as to the locations of the geological units and the dumps, the seismic refraction tomography visualized the transition between the dump or backfill layers and the underlying bedrock, and the IP measurements revealed residual mineralization. The georadar measurements, on the other hand, did not contribute to the interpretation owing to the limited depth of penetration. Based on the combination of borehole and geophysical data, it was possible to develop a hypothetical model of an open‐pit mine for copper ore that developed in three phases (mines A–C) during the Late Bronze Age. Without the control provided by the core drillings, one of the mining areas (mine A) could not have been correctly identified in the geophysical prospections.

Assessing the structure of a building with non-invasive methods is an important problem. One of the possible approaches is to use GeoRadar to examine wall structures by analyzing the data obtained from the scans. However, so far, the obtained data have to be assessed manually, relying on the experience of the user in interpreting GPR radargrams. We propose a data-driven approach to evaluate the material composition of a wall from its GPR radargrams. In order to generate training data, we use gprMax to model the scanning process. Using simulation data, we use a convolutional neural network to predict the thicknesses and dielectric properties of walls per layer. We evaluate the generalization abilities of the trained model on the data collected from real buildings.

The archaeological prospection focused on a multiphase fortified settlement. It was conducted as an integrated survey using airborne laserscanning (ALS) data, aerial photography, magnetics and GPR. The results show among other things that restricted military areas provide outstanding conditions for the preservation of archaeological sites.

Multiple antenna GPR data provide excellent results when using 3d migration. 3d migration must be performed without erroneous measurements. 3d migration must be sensitive to the different channels. Wall structures along the measurement lines can then be visualized.