Long-term Remote Vibration Monitoring for Scour Detection of a Masonry Arch Bridge

Frequent heavy rainfall events caused by global warming are increasing the scour risk of bridges, including historical masonry bridges. Structural Health Monitoring (SHM) technologies have been widely adopted for detecting bridge scour. Among these, monitoring the changes in the rocking-mode frequency of bridge piers has shown promise. However, vibration-based scour detection for masonry arch bridges has not been fully explored. One major challenge lies in the difficulty of monitoring the rocking mode of masonry arch bridges due to weak vibration signals and high ambient noise. Despite these challenges, recent advances in sensor technology enable precise measurement of weak vibrations, while stochastic operational modal analysis methods, such as the Fast Bayesian FFT approach, can address uncertainty in a probabilistic framework. This study presents a long-term vibration-based scour monitoring of an in-service historical masonry arch bridge in Northern Ireland, UK. High-precision tri-axial accelerometers were installed on the upstream spandrel wall piers to capture vibration responses especially the rocking mode. The bridge’s natural frequencies were identified using the Fast Bayesian FFT method, and their daily variations were analysed. Differences in frequency characteristics between daytime and nighttime were observed, and correlations between the identified frequencies and water levels were examined to enhance the reliability of natural frequency as an index for scour assessment.