Md. Abdullah Al Mamun Hridoy, Matteo Bodini, Munshaibur Rahman Mahin, Petra Schneider, Paolo Pastorino, Chiara Bordin, Md. Abdullah Al Mamun, Leonardo Goliatt
Aquaculture water quality exhibits time-dependent dynamics, making accurate short-term forecasting essential for proactive farm management. This study presents a leakage-safe and interpretable machine-learning framework for forecasting total dissolved solids (TDS) from high-frequency aquaculture sensor time series. The proposed approach integrates lag-based feature engineering with an expanding-window walk-forward validation protocol (19 folds) to ensure realistic time-forward evaluation and to avoid information leakage. Under a leakage-safe lag-only specification that excludes redundant conductivity predictors, tree-based ensemble learning emerged as the most robust solution. XGBoost achieved the highest forecasting accuracy, yielding a mean MAE of 0.314 ± 0.482 mg/L and RMSE of 1.596 ± 4.206 mg/L across walk-forward folds. Residual diagnostics based on ACF/PACF and Ljung–Box testing indicated no significant remaining autocorrelation, confirming that predictive skill is not driven by residual serial dependence. SHAP-based interpretation revealed that TDS dynamics are primarily governed by ionic-strength-related signals, whereas temperature and pH contribute marginally. By combining leakage-safe validation, ensemble forecasting, and explainable inference, this work advances an operational early-warning and decision-support framework for sustainable aquaculture water-quality management. © 2026 The Authors.
Faculty of Fisheries, Sylhet Agricultural University, Sylhet, 3100, Bangladesh; Department of Pathophysiology and Transplantation, University of Milan, Via Francesco Sforza 35, Zonda Pavilion, 2° floor, Milan, 20122, Italy; Department of Computer Science and Engineering, Shahjalal University of Science and Technology, Sylhet, 3114, Bangladesh; Department Water, Environment, Civil Engineering and Safety, Magdeburg-Stendal University of Applied Sciences, Breitscheidstr. 2, Magdeburg, D-39114, Germany; Experimental Zooprophylactic Institute of Piedmont, and Aosta Valley Via Bologna, Liguria, Turin, 148 – 10154, Italy; Department of Computer Science, UiT The Arctic University of Norway, 6050, Langnes, Tromsø, 9037, Norway; Department of Fish Health Management, Laboratory of Fish Diseases Diagnosis and Pharmacology, Faculty of Fisheries, Sylhet Agricultural University, Sylhet, Bangladesh; Faculty of Veterinary Medicine, Universitas Brawijaya, East Java, Malang, 65151, Indonesia; Department of Computational and Applied Mechanics, Federal University of Juiz de Fora, Juiz de Fora, 36036-900, Brazil
