 This research used artificial intelligence models to estimate water quality indexes in the Hudson River. The models were trained on Landsat 80LITRS images and four different algorithms, M5 model tree, MT, multivariate adaptive regression spline, Mars, gene expression programming, GEP, and evolutionary polynomial regression EPR. The models were tested with actual measurements of water quality parameters, WQPs, taken from the river. The results showed that the Mars model had the highest accuracy in predicting water quality indexes. This research demonstrates the potential of using artificial intelligence models to monitor water quality in real time. This article was authored by Mohammed Najafzadi and Sajad Bassyrian.