Environmental, Socioeconomic and climatic factors have been affecting the dynamic system of Waste Water Treatment Plants (WWTP) in discharging very qualitative water for different purposes. The quality of the treated effluent is essential, and it is determined by the concentration level of dominant parameters like Chemical Oxygen Demand (COD). In this study, three Machine Learning (ML) models, namely the Hammerstein-Weiner Model (HW), Least Square Support Vector Machine (LSSVM) and Multiple Linear Regression (MLR), were employed. To improve the performance of the single model’s linear ensemble techniques, Weighted Average Ensemble (WAE) was employed to predict Chemical Oxygen Demand effluent (CODeff). For the prediction of the CODeff, three types of data were used, the first one being environmental data from the new Nicosia Waste Water Treatment Plant (NWWTP) M1, whereby the second was metrological data from the “National Aeronautics and Space Administration (NASA) (at 2m above the Earth’s surface) M2, a Hybrid data (M3) which was a combination of both the metrological and environmental data M1 and M2. Based on the results of the best single models, there is a significant improvement of the models by the ensemble techniques. The study shows the effect of the hybrid data combination and the impact of employing the ensemble technique in predicting COD in NWWTP.