The main objective of this project is to leverage machine learning approaches -- more specifically, the combined use of hierarchical clustering and random forest (RF) classification -- to reveal regional patterns and drivers of nitrogenand phosphorus trends across the Chesapeake Bay watershed. This work involves three objectives: (1) use of hierarchical clustering to categorize nitrogen and phosphorus trends at the Nontidal Network stations into distinct clusters, (2) development of RF models to identify the most influential explanatory variables for the trend clusters, and (3) application of the RF models to predict trend clusters for the entire Bay watershed at the fine scale of river segments. Overall, this research provides new information to the Chesapeake Bay management community regarding regional patterns and drivers of nitrogen and phosphorus trends, including new information for the unmonitored areas.