Trends and gaps in ai‐driven predictive models for coastal water quality: A bibliometric study

The continuing decline in coastal water quality due to anthropogenic activities and climate change calls for more effective monitoring and prediction. This study examines trends in the use of artificial intelligence (AI) in modeling coastal water quality through bibliometric analysis and a systematic review of 30 selected publications. Methods include standardized literature searches, screening of relevant documents, bibliometric analysis, and data extraction for systematic review. Results show a significant increase in publications since 2020, dominance of machine learning algorithms such as random forest, SVM, and CNN, and application across various water quality parameters (e.g., chlorophyll-a, dissolved solids, E. coli bacteria). The co-author network revealed leading research groups (e.g., Uddin et al.) and international collaborations. The keyword network highlighted a focus on “water quality,” “machine learning,” “coastal waters,” and “remote sensing.” From the systematic review, AI-based models have proven capable of predicting water quality parameters with high accuracy (e.g., R² ≈ 0.88 for chlorophyll-a; Chen et al., 2024), but there are gaps in AI application (e.g., limited studies integrating deep learning with biological data, dataset constraints, and the need for models that can explain their results). In conclusion, AI research for coastal water quality is developing rapidly but still requires further development, particularly in the integration of multivariate data and model interpretability aspects.