EngD Defence Mishel Reyes Castillo | Development of a Dashboard for Management of Saltwater Intrusion in a Semi-closed Estuary

Development of a Dashboard for Management of Saltwater Intrusion in a Semi-closed Estuary

Mishel Reyes Castillo is an EngD student in the EngD programme Civil Engineering. Scientific supervisor is prof.dr. S.J.M.H. Hulscher and Co-supervisor is dr. V. Kitsikoudis from the Faculty of Engineering Technology.

The Haringvliet Estuary is managed to balance flood safety, freshwater availability and ecological connectivity. Under the Kierbesluit, the sluices are partially opened during flood tide to support fish migration. However, incoming saltwater can become trapped in bathymetric depressions and may spread towards freshwater intake areas under unfavourable conditions. Managing this trade-off requires guidance that accounts for uncertainty in environmental conditions and model predictions.

This EngD project developed a dashboard for saltwater management in the Haringvliet. It combines the existing operational protocol and analytical model with sensitivity and uncertainty analyses, an assessment of the model’s mixing representation, and an iterative stakeholder-centred design process.

The analyses identified Outer-delta salinity, Inlet volume and Ebb volume as the main factors influencing predicted saltwater behaviour. Outer-delta salinity had the strongest effect on the predicted Equilibrium Location, while Inlet and Ebb volumes strongly influenced the total salt mass. The analytical model was modified to represent differences in vertical mixing between bathymetric depressions. Although this improved the physical interpretation of mixing, it did not improve the overall prediction of Equilibrium Location, indicating that broader model recalibration is required.

The executable dashboard translates these results into uncertainty-aware information. It provides a recommended Inlet volume, an estimated probability of meeting the selected target, a confidence interval, operational warnings and a conservative safe recommendation. Users can compare up to four scenarios and explore changes in discharge, salinity, protocol settings and uncertainty assumptions.

Stakeholder feedback shaped the terminology, input workflow and layered interface. The final product is a working research prototype for scenario evaluation, discussion and learning. It has not been validated for real-time operation and is intended to support, rather than replace, expert judgement.