MERMAID

Eutrophication is a major environmental issue with many negative consequences, such as hypoxia and harmful cyanotoxin production. Monitoring coastal eutrophication is crucial, especially for island countries like the Republic of Cyprus, which are economically dependent on the tourist sector. Additionally, the open-sea aquaculture industry in Cyprus has been exhibiting an increase in recent decades and environmental monitoring to identify possible signs of eutrophication is mandatory according to the legislation. In this project, deep learning models are created and adjusted for maritime hotspots near Cyprus. In particular, two different types of artificial neural networks (ANNs) are developed based on in situ data collected from stations located in the coastal waters of Cyprus to model the eutrophication phenomenon.

First, the self-organizing map (SOM) ANN examines several water quality parameters’ (specifically water temperature, salinity, nitrogen species, ortho-phosphates, dissolved oxygen, and electrical conductivity) interactions with the Chlorophyll-a (Chl-a) parameter. The SOM model enables us to visualize the monitored parameters’ relationships and to comprehend complex biological mechanisms related to Chl-a production. For example, the SOM-based clustering verified the oligotrophic nature of Cypriot coastal waters and the good water quality status.

Second, feed-forward ANN models are also developed for predicting various parameters of interests, including the Chl-a levels, surface dissolved oxygen (DO) levels, and dissolved inorganic nitrogen (DIN) levels. Using the developed ANN models, sensitivity analysis was also performed to measure the impact of each input parameter on the ANN’s target parameter (i.e., Chl-a, DO, DIN). The sensitivity analysis results revealed that salinity and water temperature are the most influential parameters on Chl-a production, while the water temperature is the most influential factor for DO. Moreover, the sensitivity analysis results of the feed-forward ANN captured the winter upwelling phenomenon that is observed in Cypriot coastal waters.

The created ANNs allowed us to comprehend the mechanisms related to eutrophication and hypoxia regarding the coastal waters of Cyprus and can act as useful management tools regarding eutrophication control and the prevention of hypoxia. Overall, the impact of the models is twofold: (1) to produce restoration scenarios, derived from additive or synergistic parameters interactions, based on the application of sensitivity analysis; and (2) to act as a warning tool, able to predict a possible harmful algal bloom.