AI Analysis and Prediction of Pollution Levels in Industrial Wastewater: Case Study of Al-Zafaraniya Area, Al-Rusafa District/Baghdad

Objective: This study aims to assess the quality of treated industrial wastewater before discharge into the Tigris River, by measurement of physical and chemical parameters such as pH, temperature, and total suspended solids (TSS) and pollutant levels such as sulfates, chlorides, phosphates, and nitrates. A model was also developed based on artificial intelligence techniques to predict future wa-ter quality variation due to environmental and climatic conditions. Methodology: Monthly data for the years 2021 to 2023 were gathered and statistically compared using significant difference tests at a probability level of P≤0.05. Various prediction models were used to forecast future trends in pol-lutant levels up to 2026, by comparing performance metrics like mean absolute error (MAE), root mean square (RMSE), and coefficient of determination (R²). The results showed significant varia-tions in pH levels, ranging from 7.00 to 8.87, within permissible limits according to environmental standards, temperature of the water changed according to the seasons, with the highest (32-34°C) in July and August, and the lowest (15.3-19.0°C) in January. There were significant variations in chemical concentrations, with TSS, chlorides, phosphates, and nitrates' concentrations higher than global permissible limits, which compromised the environment of the aquatic ecosystem. Statistical analysis showed that there was a positive relationship between temperature rise and higher concen-trations of some of the pollutants such as phosphates and nitrates, which are eutrophication-promoting. The model was extremely accurate in predicting water quality changes with a coefficient of determination (R²) of ≈ 0.90, indicating the predictive power of the model. Conclusion: Indus-trial wastewater treated water contains pollutants beyond tolerable levels, necessitating more inten-sive pre-discharge treatment technologies. Climatic conditions have strong influences on water quality, which needs more stringent efforts to monitor and adjust for seasonal changes. Predictive models can be used to guide water resource management practices and prevent pollution, especially with temperature and pollutant concentration rises expected through 2026.Improving industrial treatment processes, such as reverse osmosis and ion exchange, is recommended to reduce sulfates, phosphates, and chlorides levels before releasing water into natural environments.