Haimei Wang

Hainan Institute of Biotechnology, Haikou, 570206, Hainan, China
Author    Correspondence author
International Journal of Marine Science, 2026, Vol. 16, No. 5   
Received: 20 Jul., 2026    Accepted: 24 Aug., 2026    Published: 06 Sep., 2026

This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Accurate prediction of fish growth is essential for improving aquaculture productivity, optimizing feeding strategies, and achieving sustainable management of aquatic systems. Water quality parameters are closely associated with fish physiological processes, metabolic activity, and growth performance; however, the complex and nonlinear interactions among environmental factors make traditional prediction approaches insufficient for dynamic aquaculture conditions. This paper systematically reviews data-based approaches for predicting fish growth based on water quality parameters, focusing on data acquisition, feature construction, machine learning modeling, and practical applications in precision aquaculture. Key environmental variables, including temperature, dissolved oxygen, pH, salinity, ammonia nitrogen, nitrite nitrogen, and organic matter, are integrated with biological growth indicators to establish predictive models. Various statistical and artificial intelligence methods, such as multiple linear regression, random forest, support vector machines, gradient boosting algorithms, and deep learning models, provide effective tools for capturing complex relationships between environmental fluctuations and fish growth dynamics. Model interpretation approaches, including feature importance analysis and explainable artificial intelligence, further clarify the contribution of individual water quality factors to growth variation. A case-study framework demonstrates how continuous water quality monitoring combined with machine learning can identify critical environmental drivers, predict growth trends, and support adaptive aquaculture management. Overall, data-driven growth prediction models provide a powerful approach for transforming water quality information into actionable decisions, enabling intelligent feeding regulation, environmental control, and sustainable fish production.

Fish growth prediction; Water quality parameters; Machine learning; Precision aquaculture; Data-driven modeling