The study of water is a crucial factor in the ecosystem. This study investigates an improved classification method for urban water by integrating data from Sentinel-1 (Synthetic Aperture Radar) and Sentinel-2 (Multispectral) from the European Space Agency's Sentinel constellation. The analysis leverages the power of the Random Forest Algorithm and Python's Sklearn library to create a classification map. Usually, water mapping extraction relies on a single data source, resulting in limitations in accuracy and the ability to differentiate water types. In this work, first, we performed independent classification of water using Normalized Difference Water Index (NDVI) and Radar Water Index (RWI). Secondly, to enhance classification accuracy, we incorporated a combination of each index, NDWI and RWI, with the Modified Normalized Difference Water Index (MNDWI) and Automated Water Extraction Index, and we identified the water classes exhibiting the highest prevalence. Thirdly, by using the proposed fusion method, we combined the Modified Normalized Difference Water Index and Automated Water Extraction Index with the fusion of Normalized Difference Water Index and Radar Water Index, and we compared the density variations among water types. The results showed that the effectiveness of the fusion approach significantly improved the classification accuracy, achieving a perfect 100% of overall accuracy and a perfect 100% of Kappa index in all the prediction elements. It also showed a classification result by class for the Random Forest Algorithm, where each water type was “present”, and it showed the agreement level by class for the Random Forest Algorithm. Some water types were classified as perfect, moderate, fair, slight, poor, very poor, or inexistent. These results provide valuable information for ecological efforts and future land management strategies.