Machine Learning in management of precautionary closures caused by lipophilic biotoxins

TL;DR

This study tackles the risk management of lipophilic biotoxins in Galician mussel farming by building a data-driven approach to predict precautionary closures. It fuses 15 years of environmental data from the Vigo estuary with six classifiers, identifying k-Nearest Neighbour as the top performer, achieving a mean sensitivity of $97.34\%$, accuracy of $91.83\%$, and a Cohen's kappa of $0.75$. The work underscores the predictive value of prior-week status, Dinophysis acuminata concentration, and dissolved nutrients for lipophilic toxin risk, and demonstrates the potential to support expert decision-making when recent sampling is unavailable. Practically, the framework can guide closures and openings across multiple zones and may be extended to additional estuaries and variables, improving public health protection and industry resilience.

Abstract

Mussel farming is one of the most important aquaculture industries. The main risk to mussel farming is harmful algal blooms (HABs), which pose a risk to human consumption. In Galicia, the Spanish main producer of cultivated mussels, the opening and closing of the production areas is controlled by a monitoring program. In addition to the closures resulting from the presence of toxicity exceeding the legal threshold, in the absence of a confirmatory sampling and the existence of risk factors, precautionary closures may be applied. These decisions are made by experts without the support or formalisation of the experience on which they are based. Therefore, this work proposes a predictive model capable of supporting the application of precautionary closures. Achieving sensitivity, accuracy and kappa index values of 97.34%, 91.83% and 0.75 respectively, the kNN algorithm has provided the best results. This allows the creation of a system capable of helping in complex situations where forecast errors are more common.

Figures (8)

• Figure 1: Schematic representation of the machine learning-based system for predicting harmful algal bloom closures and aiding decision making in mussel farming. This graphic has been designed with resources from Flaticon.com
• Figure 2: Map of the production areas of cultivated molluscs in the Vigo estuary. Source: http://193.144.46.136/EstadoZonas/Default.aspx?tmapa=0
• Figure 3: Map of oceanographic stations located in the Vigo estuary. Source: http://www.intecmar.gal/Ctd/Default.aspx
• Figure 4: Distribution of closure episodes caused by HAB in mussel production areas in the Vigo Estuary (2016). Source: http://www.intecmar.gal/Informacion/biotoxinas/Evolucion/DiagramaBateas.aspx
• Figure 5: Summary table with the occurrence of the features after the feature selection processes. Where each point represents the likelihood of a variable being selected as an input feature for a particular production area.