Hyperbolic diffractions in Ground Penetrating Radar (GPR) data are caused by a variety of subsurface objects such as pipes, stones, or archaeological artifacts. Supplementary to their location, the propagation velocity of electromagnetic waves in the subsurface can be derived. In recent years, it was shown that deep learning tools can automatically detect hyperbola in radargrams using data measured over urban infrastructure, which are relatively clear. In contrast, in this study, we used an archaeological dataset with diverse underground structures. In the first step we used the deep learning network RetinaNet to detect hyperbola automatically and achieved an average precision of 0.58. In the next step, 10 different approaches for hyperbola fitting and thus velocity determination were applied. The derived information was validated with manually determined velocities and apex points. It was shown that hyperbola extraction by using a threshold and a column connection clustering (C3) algorithm followed by simple hyperbola fitting is the best method, which had a mean velocity error of 0.021 m/ns compared to manual determination. The average 1D velocity-depth distribution derived in 10 ns intervals was in shape comparable to the manually determined one, but had a systematic shift of about 0.01 m/ns towards higher velocities.

We present a case study of a bog showing how an integrated approach of multi-method geophysical sounding and local soil sampling can be used to identify, differentiate, and map organic sediments. Our study is based on ground-penetrating radar (GPR), electrical resistivity tomography (ERT) and shear-wave seismic (SH seismic) profiling applied to sediments of the former Lake Duvensee (northern Germany), nowadays a bog. This is a well-known locality for remains from the Mesolithic hunter-gatherers’ occupation that has been attracting archaeological and geoarchaeological research for100 years. The bog is embedded in low conductive glacial sand and is characterized by layers of different gyttja sediments (detritus and calcareous). The present study was conducted in order to identify the bog morphology and the thickness of the peat body and lake sediments, in order to understand the basin evolution. To validate the geophysical results, derived from surface measurements, drilling, soil analyses as well as borehole guided wave analysis of electromagnetic waves and Direct-Push (DP-EC) have been carried out and used for comparison. It turned out that each method can distinguish between sediments that differ in grain size, particularly between peat, lake sediments (gyttjas and mud) and basal glacial sand deposits. GPR is even able to separate between strongly and weakly decomposed peat layers, which is also clear considering resistivity variations in the ERT computation. From the association between geophysical properties and sediment analysis (e.g., water content and organic matter) different gyttjas were distinguished (coarse and fine) and seismic velocity was correlated to bulk density. Moreover, GPR and SH-wave seismics present different resolutions, confirming that the latter allows measurements, which are more focused on determining the extension of basal sand deposits, the depth of which is difficult to reach with GPR. Representative values of electrical resistivity, dielectric permittivity, and shear wave velocity have been determined for each sediment type and are therefore available to complete the investigation of wetland environments. Fine grained lake sediments were difficult to differentiate by the applied methods. This could be a result of high ionic concentration within the permanent groundwater body, partly masking the sediment properties.

Integration of multiple geophysical data is a key practice to reduce model uncertainties and enhance geological interpretations. Electrical resistivity models resulting from inversion of marine magnetotelluric (MT) data, often lack depth resolution of lithological boundaries and distinct information for shallow model parts. This is due to the diffusive nature of electromagnetic fields, enhanced by deficient data sampling and model regularization during inversion. Thus, integrating data or models to constrain layer thicknesses or structural boundaries is an effective approach to derive better constrained and more detailed resistivity models. We investigate the different impacts of three cross‐gradient coupled constraints on 3D MT inversion of data from the Namibian passive continental margin. The three constraints are (a) coupling with a fixed structural density model; (b) coupling with satellite gravity data; (c) coupling with a fixed gradient velocity model. Here, we show that coupling with a fixed model (a and c) improves the resistivity model the most. Shallow conductors imaging sediment cover are confined to a thinner layer in the resulting resistivity models compared to the MT‐only model. Additionally, these constraints help to suppress vertical smearing of a conductive anomaly attributed to a fracture zone, and clearly show that the seismically imaged Moho is not accompanied by a change in electrical resistivity. All of these observations help to derive an Earth model, which will form the basis for future interpretation of the processes that lead to continental break‐up during the early Cretaceous.

The Baltic Sea basins, some of which only submerged in the mid-Holocene, preserve Stone Age structures that did not survive on land. Yet, the discovery of these features is challenging and requires cross-disciplinary approaches between archeology and marine geosciences. Here, we combine shipborne and autonomousunderwater vehicle hydroacoustic data with up to a centimeter range resolution, sedimentological samples, and optical images to explore a Stone Age megastructure located in 21 m water depth in the Bay of Mecklenburg, Germany. The structure is made of 1,673 individual stones which are usually less than 1 m in height, placed side by side over a distance of 971 m in a way that argues against a natural origin by glacial transport or ice push ridges. Running adjacent to the sunken shoreline of a paleolake (or bog), whose youngest phase was dated to 9,143 ±36 ka B.P., the stonewall was likely used for hunting the Eurasian reindeer (Rangifer tarandus) during the Younger Dryas or early Pre-Boreal. It was built by hunter–gatherer groups that roamed the region after the retreat of the Weichselian Ice Sheet. Comparable Stone Age megastructures have become known worldwide in recent times but are almost unknown in Europe. The site represents one of the oldest documented man-made hunting structures on Earth, and ranges among the largest known Stone Age structure in Europe. It will become important for understanding subsistence strategies, mobility patterns, and inspire discussions concerning the territorial development in the Western Baltic Sea region.

In this study, we investigate GPR measurements in freshwater of less than 5 m at four different locations to derive rules of thumb in terms of depth penetration, resolution, and material contrasts of the method for 200 and 400 MHz antennas under field conditions. The objective is to improve the attractiveness of the method for archaeological issues in water, as there are hardly any studies on this subject so far. The depth penetration of 2–4 m is negligibly influenced by the choice of the 200 or 400 MHz antenna. Organic material in the water column also does not affect the water depth but offers new fields of applications for mapping and volume estimation of biomass in lakes with GPR. The horizontal resolution in the cm range in the direction of the profile and in the dm range across the profile could not be improved by the narrow antenna radiation pattern of 0.1 for materials of <80% porosity to the water column and for materials of 45% porosity to water-saturated sand. Large reflection coefficients also show, for example, granite to sand and gyttja to wood. The water column has a considerable effect on the data quality of the 400 MHz antenna from a depth of 2 m due to the antenna ringing. Furthermore, multiples must be expected in a water column <0.5 m. The method can especially complement the common geophysical methods of seismics and geoelectrics to exclude material ambiguities. The major advantage is the simple setting of the land equipment in the water.