Water-quality monitoring in urban lakes is of paramount importance due to the direct implications for ecosystem health and human well-being. This study presents a novel approach to predicting the Water Quality Index (WQI) in an urban lake over a span of two decades. Leveraging the power of Machine Learning (ML) algorithms, we developed models that not only predict, but also provide insights into, the intricate relationships between various water-quality parameters. Our findings indicate a significant potential in using ML techniques, especially when dealing with complex environmental datasets. The ML methods employed in this study are grounded in both statistical and computational principles, ensuring robustness and reliability in their predictions. The significance of our research lies in its ability to provide timely and accurate forecasts, aiding in proactive water-management strategies. Furthermore, we delve into the potential explanations behind the success of our ML models, emphasizing their capability to capture non-linear relationships and intricate patterns in the data, which traditional models might overlook.