Real-time monitoring and forecasting of dike strength

Real-time monitoring and forecasting provides useful information in early warning situations for emergency response as part of a modern (flood) risk management.This paper presents a case study of a coastal dike line, where multiple sensors are installed to measure in real-time the water level outside and inside the dike.The dike stability is calculated based on the inputs of the phreatic line and on the schematization of the subsoil.The resulting safety factor is a direct assessment of the dike strength in real-time.For a prediction of the dike performance, fragility curves are derived within a model-based probabilistic analysis for different failure mechanisms: overflow, wave overtopping, wave impact, wave erosion, piping, micro-and macro-stability are considered.They are combined in one overall fragility curve that represents the total probability of failure per dike cross-section as a function of the water level.By combining forecasted water levels and fragility curves, it is possible to get a prediction of the dike reliability.The two workflows of real-time monitoring and forecasting of dike strength are being integrated into the FEWS-DAM Live software system.This allows for the visualization of real-time and historical data of dike stability and probability of failures based on the forecasted water levels.The generated results provide precise information for the emergency response, such as location, timing and probability of failure of specific sections of the flood defense line.With the help of this information, emergency measures that apply to the flood defense line (e.g. starting from increased inspection intervals up to temporally dike enforcement) can be operationally planned, adapted to the situation and triggered.