OOS2025-307, updated on 26 Mar 2025
https://doi.org/10.5194/oos2025-307
One Ocean Science Congress 2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Anticipating microbiological risks in coastal area through innovative prediction models
Maxime Paris, Arthur Tré-Hardy, Léo Seyfried, Samuel Dupont, and Charlotte Dupont
Maxime Paris et al.
  • Bioceanor, France (contact@bioceanor.com)

The increasing human pressure on coastal ecosystems, coupled with climate change, ocean acidification population migration along the coast, poses significant challenges to the management and preservation of water quality. Current practices for monitoring microbial pollution, such as Escherichia coli levels, rely heavily on sparse, manual sampling, which offers limited temporal and spatial insights. This approach fails to capture the dynamic nature of contamination events and their potential impacts on public health, aquaculture, and coastal biodiversity, leading to delayed actions to address this contamination.

BiOceanOr, through the PREMICE project (project supported by the French government through i-Lab, funded by ADEME), addresses this critical gap by developing a comprehensive solution for predictive modeling of microbiological risks in coastal waters. Anchored in cutting-edge machine learning and IoT technology, PREMICE provides a proactive solution for understanding and anticipating contamination events.

This solution, branded as "i-plage," integrates high-frequency, connected microbiological measurement tools, predictive algorithms based on physico-chemical and biological parameters, and a user-friendly visualization interface. These innovations empower local authorities and stakeholders to take informed, timely actions, enhancing both ecological and public safety.

Key technical advancements within the PREMICE project include:

  • Real-time physico-chemical monitoring: A network of sensors measures parameters (temperature, salinity, turbidity, dissolved oxygen, E. coli...) These data streams, combined with historical and satellite data, provide a robust foundation for understanding coastal water dynamics.
  • Predictive microbiological models: Leveraging machine learning, PREMICE generates short-term forecasts of E. coli contamination risks. This predictive capability considers environmental drivers, (weather and hydrodynamic patterns), to anticipate potential risks
  • Scalable visualization platform: The interface offers an intuitive experience, designed for both professional stakeholders and the public. It enables real-time communication of risks and supports decision-making processes, from beach safety alerts to usages management.

The project is conducted in collaboration with leading environmental institutions, water treatment experts and integrates expertise across biology, data science, and coastal engineering. By focusing on high-risk coastal areas, such as the estuary of the Var river in France, PREMICE contributes with local stakeholders (i.e. Métropole Nice Côte d’Azur) to a better understanding of pollution dynamics and offers a scalable model for international application.

Overall, PREMICE exemplifies the integration of advanced technology for environmental monitoring, offering a paradigm shift from reactive to anticipatory management. By providing actionable insights and promoting sustainable practices, this innovation supports global efforts to protect vulnerable coastal ecosystems in the face of escalating anthropogenic and climate-induced challenges.

How to cite: Paris, M., Tré-Hardy, A., Seyfried, L., Dupont, S., and Dupont, C.: Anticipating microbiological risks in coastal area through innovative prediction models, One Ocean Science Congress 2025, Nice, France, 3–6 Jun 2025, OOS2025-307, https://doi.org/10.5194/oos2025-307, 2025.