To address the identified research needs and achieve the set objectives, the following actions have been taken. The project initially focused on exploring and evaluating various scour monitoring technologies. A critical part of this effort involved comparing different sensors, assessing their effectiveness, and identifying new strategies for both direct and indirect monitoring of scour risks. Based on this review, two case studies were developed at key bridges in Scotland (Figure 1). These case studies provided real-world settings for testing the selected monitoring technologies. The sensors deployed at these sites are part of an ongoing effort to collect data, which has been incorporated into an open-source dataset. This dataset continues to be updated, allowing for long-term monitoring of scour risks at these sites.
Furthermore, one of the case studies was numerically modelled as part of a Secondment activity, which enabled a deeper understanding of scour processes and their impacts during extreme events such as floods. The data collected from these sensors was integrated with advanced scour forecasting models to provide real-time estimations of local scour (Figure 2). This integration allowed for the creation of a dynamic tool capable of continuously updating scour risk estimates based on monitored flow properties. The use of advanced forecasting models ensures that the predictions take into account the variability in hydraulic conditions and their effect on scour, thereby improving the accuracy and reliability of the predictions.
The next step in the research was the development of a Bayesian Network model (Figure 3), which was designed to incorporate data from the monitoring tools and integrate forecasting results. This framework enables a more robust risk assessment by accounting for uncertainty and incorporating real-time observations, which are often overlooked in traditional methods. Additionally, it enables sensitivity analysis on various parameters, guiding the identification of optimal monitoring strategies. This capability ensures the efficient use of available resources and technology by pinpointing the most influential factors in scour risk and focusing efforts where they are most needed. Moreover, the Bayesian Network was employed to generate expert-judgment-based fragility curves for various types of bridge foundations and considering different hydraulic loads. These curves are essential for understanding the vulnerability of bridges, helping bridge operators and managers understand the potential impacts of scour and better adjust their strategies for flood events. Finally, the Bayesian Network functions as a powerful Decision Support System, providing a flexible and adaptive approach to scour risk management. It empowers decision-makers to make informed predictions based on a combination of historical data, current observations, and predictive models, allowing for more effective planning for flood events. Ultimately, this methodology assists policymakers in shaping environmental risk policies and making well-informed decisions during flood emergencies.
The study results have been presented at various international conferences, and several papers are currently in preparation or under submission for publication in journals. Regular meetings with stakeholders, including the bridge operator company, ensured that the risk assessment outcomes were effectively communicated, leading to improved flood risk management strategies. Participation in outreach activities, such as the Explorathon event, and engagement with local communities further contributed to raising awareness and fostering a better understanding of scour risks and flood management
