Estimating moisture changes in concrete using GPR velocity analysis: potential and limitations

Chloride-induced corrosion of steel is a key problem for reinforced concrete buildings such as bridges. Here, the moisture conditions are of major interest because they are closely related to chloridization and its tempo-spatial variability. Ground-penetrating radar (GPR) is commonly used for extensive non-destructive imaging of small-scale structural defects in concrete. To identify the potential and the limitations of GPR for practical investigation of reinforced concrete buildings, a longterm GPR monitoring experiment on a well-defined reinforced concrete specimen characterized by a typical rebar geometry has been performed under laboratory-like conditions. The GPR data analyzed in this study have been acquired before and after a three-week immersion of the specimen in water using a geometric setup and antenna system as commonly applied in concrete inspection. Before analyzing the data in detail, we apply a typical imaging processing flow and demonstrate the potential of our GPR data to image the spatial location of the rebar structures. To assess the moisture content of the specimen and its tempo-spatial variability, our study focuses on developing and performing an automatic migration-based velocity analysis, which also allows to estimate uncertainties in the derived velocities. Although analyzing single diffraction hyperbolas is not sufficient to resolve moisture changes in the range of a few percent, we are able to identify a mean temporal trend of increasing moisture content related to the immersion in water at two depth levels of the specimen. The observed trends and the estimated change of concrete moisture content are in good agreement with independent reference measurements. Our results demonstrate the resolvable limit of identifying moisture changes in typical reinforced concrete using GPR diffraction velocity analysis.