Using Fishery-related Data, Scientific Expertise and Machine Learning to Improve Marine Habitat Mapping in Northeastern Mediterranean Waters

Marine habitat mapping is an essential tool for better planning of conservation efforts and the sustainable management of marine activities. High spatial resolution in marine habitat maps is of utmost importance as it may encompass more detail in the imagery and potentially reveal important biotopes. This level of detail supports directing monitoring and analysis efforts for the effective implementation of the EU environmental policies and providing more relevant advice for robust decision-making under both sectorial policies (e.g. the Common Fisheries Policy) and more integrated ones (e.g. the Marine Spatial Planning). In this study, sea bottom type data, recorded during the national monitoring of commercial fishing vessel operations and fishery surveys in the Greek Seas, were used. These data were then assigned to EU EMODnet seabed habitats using local ecological knowledge. Two Machine Learning (ML) algorithms were trained on the entire national-scale dataset and subsequently applied to assess their performance in predicting habitat types in the Saronikos Gulf (regional scale), using various environmental factors as predictors. These algorithms were the Random Forest Classifier (RFC), and the Gradient Boosting Classifier (GBC), while the Borderline Synthetic Minority Oversampling Technique (B-SMOTE) was applied for handling the inherent data class imbalances. A validation dataset and georeferenced data from previous studies were considered for comparing the models' accuracies and predictive performance. Through this approach, the Saronikos Gulf was enriched by five more habitat types than those visualized in the EMODnet portal, while also filling habitat gaps in areas where no data existed. Results from the application of the RFC-BS model (62% accuracy, 0.51 Kappa score) were then used to address conservation planning commitments recently made by the Greek government; the vast majority of marine seabed priority habitats in the study area seem to fall outside the borders of the current Natura 2000 sites, which served as the baseline for the declared trawl bans in Greek waters, following the provisions of the EU Marine Action Plan.