Context of the assignment Turbidity plumes created by dredging operations can cause impact on sensitive receivers such as coral reefs, sea grass or cooling water intakes. That is why regulators set turbidity limits around dredging works. If you can predict where the plume will travel, at what intensity, and when, you can plan the operation to stay within limits while keeping the project running. Good plume forecasting is essential for an optimal balance between environmental protection and operational efficiency. There currently is a knowledge gap in what the optimum prediction method is in low current environments. This assignment considers a recent dredging operation in Sohar, Oman. The location is especially interesting due to the relatively low tidal currents, causing other drivers to dominate the turbidity plume dynamics, such as waves and wind. We are looking to investigate different methods to improve turbidity forecasts for similar locations. Assignment The scope is flexible and shaped around your interests. A potential main question is: How can dredging-induced turbidity dispersion in low current environments best be predicted ... In the context of a project in execution (early phase)? Using Machine Learning? Potential sub questions can be: Which processes drive plume dispersion when tidal currents do not dominate? What is the initial area of mixing (the near-field phase of the plume)? How can machine learning improve predictions? - as a parallel track running alongside the physics-based model? - to capture processes the model does not include, such as waves or dredging tracks? - what quantity of measurement records is needed before ML starts to add value? How can ensemble forecasting contribute to turbidity predictions? How should uncertainty in the input parameters be accounted for? Available case study Sohar project in Oman, executed in 2025. Available data: Hydrodynamic model: operationally implemented, with daily model runs and turbidity forecasts Turbidity data: monitored continuously by in-situ buoys Meteo forecasts: wind and wave-height predictions used during the works Production data: dredged quantities, cycle times, overflow durations (to be confirmed what can be shared) Your qualities and experience You can move ahead with this assignment if you have: You are enrolled in a university program, preferably in the field of Hydraulic Engineering, Offshore Engineering, Applied Mathematics, or another field linked to the dredging sector You are interested in dredging and environmental impact. You have an affinity with at least one of the following: numerical modelling, programming and/or machine learning A good command of written and spoken English and/or Dutch.