Dutch modelling helps Dunkirk predict bathing water pollution before it reaches the coast
Dutch water expertise is helping Dunkirk, France, better anticipate water quality risks along its popular coastline. Deltares and Delft University of Technology (TU Delft) are working with French partners (IXSANE, ULCO University, and Dunkirk Urban Community) on FEWS-DK’Eau, an operational prototype that combines real-time data, hydraulic modelling and water quality forecasting to enable better decision making on bathing water quality.
Located on the northern French coast, Dunkirk is a complex area for water management. Much of the city and its surroundings are low-lying and intersected by an extensive network of canals, sewers and hydraulic structures. At the same time, its sandy coastline is home to several popular bathing sites.
During heavy rainfall, Dunkirk’s sewer systems can overflow, releasing untreated water into canals and eventually the sea. While these discharges can help reduce flood risk, they can temporarily affect bathing water quality.
This challenge prompted the Urban Community of Dunkirk (Communauté urbaine de Dunkerque, CUD) and its partners to design FEWS-DK’Eau: a forecasting and alert system that combines information on water levels, water flows and pollution risks.
From data to operational decisions
At the heart of the system is Delft-FEWS, Deltares’ platform for operational forecasting and water management. Deltares developed the Delft-FEWS application for Dunkirk, integrated its underlying models and helped translate complex modelling results into information that water managers can use in practice. The system combines several Dutch modelling tools. D-Flow FM calculates water levels and flows through the canal network, while D-Real Time Control models the operation of gates, weirs and pumps. D-Water Quality simulates how pollutants from sewer overflows move towards the sea. These models are combined with operational data, including rainfall observations and weather forecasts from Météo-France, tidal information and data on water levels, pumps and gates. Every few hours, the system automatically processes new data and generates updated forecasts of hydraulic conditions. When sewer overflow conditions are shown, the system can identify periods when the need to discharge water into the sea coincides with elevated pollution levels – helping water managers assess when bathing water may be a risk. Users can also explore different scenarios, for example to see how changing pump settings or reducing inflows could affect water quality at the outlet.
Combining Dutch and French expertise
Close cooperation between Dutch and French organisations since 2022 led to the creation of the system. CUD initiated the project and will host the system on its own cloud infrastructure. Deltares supplies expertise in operational forecasting, hydraulic and water quality modelling and model integration. TU Delft acts as technical adviser on sewer dynamics and control. On the French side, the Ixsane consultancy firm supports local stakeholder engagement and co-creation and is designing a separate tool for managing sewer dynamics. Université du Littoral Côte d’Opale shares expertise in stakeholder analysis and engagement. As users of the system, local waterway, polder and port authorities are also involved. This combination of Dutch technical expertise and French local knowledge was important in designing a suitable tool for Dunkirk’s particularly complex water system.
Managing uncertainty
One of the distinguishing features of FEWS-DK’Eau is its ability to provide useful operational information even where data is limited. Rather than presenting forecasts as certainties, the system enables water managers to compare scenarios and make decisions while taking uncertainty into account. The project has also led to the further development of Delft-FEWS itself, including a new transformation module and new modules for importing observations and weather forecasts directly from the Météo-France. The second phase of FEWS-DK’Eau is now approaching completion. It has expanded the geographical area covered, integrated real-time data and improved both the hydraulic and water quality models. Training will help CUD build the capacity to operate and maintain the system. By combining Dutch modelling and forecasting expertise with French knowledge of local water systems and operational needs, FEWS-DK’Eau demonstrates how international collaboration can turn water data into practical tools for protecting both urban water systems and bathing water quality.
