Condition Assessment of Bridge Decks with Asphalt Overlay

Asphalt overlay is a common practice for concrete bridge deck rehabilitation. However, asphalt overlays inhibit visual inspection of concrete bridge deck and limit application of many traditional nondestructive testing (NDT) methods, including the popular chain-drag and sounding techniques. Undetected deck delamination and deterioration may eventually result in significant rehabilitation or replacement costs. Although significant effort has been done in NDT research for bridges in past decades, there is limited progress in NDT applications to asphalt overlaid bridge decks. Ground Penetrating Radar (GPR) is currently the only NDT method that can be used to evaluate a concrete bridge deck with asphalt overlays (American Society for Testing and Materials [ASTM] 2015). GPR can penetrate though the asphalt layer and uses the amplitude of GPR reflection signals from reinforcement bars or the bottom surface of the deck to indicate concrete deterioration. The GPR amplitude maps are usually used as a tool to classify the maintenance action level required on a bridge and to determine the sections of the bridge deck requiring repair. However, the accuracy of this method is affected by rebar depth, rebar diameter, and moisture condition within both the concrete deck and asphalt overlay. It has also been reported that GPR may provide a false indication of damaged concrete. Recent studies have concluded that GPR alone cannot provide sufficient information to determine the type and range of deteriorations in bridge decks. Interpretation of the GPR amplitude maps (contour plots) depends on the spatial changes within the data. The level of bridge deck deterioration is significantly influenced by the selected threshold value. However these values are usually selected based on experience, without a validation using the ground truth. When the condition of the bridge deck is uniform, i.e. fully sound, or fully deteriorated, the lack of contrast in the data creates another challenge in data interpretation. Therefore, ground truth and/or supplementary data from other NDT methods are needed to evaluate the accuracy of GPR results.