Causal Graph Discovery for Explainable Insights on Marine Biotoxin Shellfish Contamination

Harmful algal blooms are natural phenomena that cause shellfish contamination due to the rapid accumulation of marine biotoxins. To prevent public health risks, the Portuguese Institute of the Ocean and the Atmosphere (IPMA) regularly monitors toxic phytoplankton in shellfish production areas and temporarily closes shellfish production when biotoxins concentration exceeds safety limits. However, this reactive response does not allow shellfish producers to anticipate toxic events and reduce economic losses. Causality techniques applied to multivariate time series data can identify the variables that most influence marine biotoxin contamination and, based on these causal relationships, can help forecast shellfish contamination, providing a proactive approach to mitigate economic losses. This study used causality discovery algorithms to analyze biotoxin concentration in mussels Mytilus galloprovincialis and environmental data from IPMA and Copernicus Marine Environment Monitoring Service. We concluded that the toxins that cause diarrhetic and paralytic shellfish poisoning had more predictors than the toxins that cause amnesic poisoning. Moreover, maximum atmospheric temperature, DSP toxins-producing phytoplankton and wind intensity showed causal relationships with toxicity in mussels with shorter lags, while chlorophyll a (chl-a), mean sea surface temperature and rainfall showed causal associations over longer periods. Causal relationships were also found between toxins in nearby production areas, indicating a spread of biotoxins contamination. This study proposes a novel approach to infer the relationships between environmental variables to enhance decision-making and public health safety regarding shellfish consumption in Portugal.