Artificial Intelligence for mussels farm quality assessment and prediction

Mussel farming is one of the production sectors influenced by the pollutants in seawater, both of chemical and biological origin. Monitoring the impact of the pollutants on mussel farms is a crucial issue in coastal management. A computational approach to mitigate the coast connected to the in-situ monitoring and give the possibility to predict the water quality evolution concerning the coastal hydrodynamics and the known pollution source activities could be a convenient solution. However, although a coupled atmosphere-ocean numerical models workflow is a solution already made operational in diverse and different contexts, the prediction of bacteria contamination in farmed mussels, given the forecast of contaminant concentration, is still an open issue. In this paper, we introduce a novel methodology devoted to predicting the level of contamination given the pollutant concentration from Lagrangian models for transport and diffusion. We present the Artificial Intelligencebased water QUAlity Model (AIQUAM). AIQUAM adopts a computational approach based on High-Performance computer facilities and artificial intelligence to define the dynamics of pollutants in the proximity of mussel farms. We motivate the design and implementation of decision-making tools to support the local authorities in the management activities. Within the framework presented here, the mussel is modeled by the mussel-pollutant interaction time and the bio-accumulation phenomena in filtering organisms (mussels), which can result in hygienic-sanitary emergence deriving from the sale and consumption of potentially polluted products.