From October 1, 1999, through September 30, 2009, water-quality samples were collected, and discharge was measured at 13 streamgages within the Catskill and Delaware watersheds of the New York City water supply system. The Catskill and Delaware watersheds supply about 90 percent of the water needed by 9 million customers. On average, 59 water-quality samples were collected at each station during each year of the study and analyzed for major ions and nutrients. At six stations, suspended-sediment samples were collected during 2001–09, and turbidity samples were collected during 2003–09. Surficial geology exerted a strong influence on the water quality of streams in the region. Stations in the Cannonsville Reservoir watershed, which has a high percentage of glacial till, had circumneutral stream water, whereas stations in the Neversink Reservoir watershed, which has a high percentage of sedimentary bedrock outcrops, had acidic stream water. All stations showed significant decreases in stream water sulfate concentrations during the study period; however, only the most acidic watersheds showed decreases in hydrogen-ion concentration. Two of the most acidic stations, East Branch Neversink River northeast of Denning and Rondout Creek above Red Brook at Peekamoose also had significant decreasing trends in inorganic monomeric aluminum concentrations, a form of aluminum that is toxic to some aquatic biota at concentrations greater than 0.05 milligram per liter. Three stations in the Neversink Reservoir watershed had inorganic monomeric aluminum concentrations that commonly exceeded 0.05 milligram per liter during the study period. At the West Branch Neversink River at Winnisook Lake near Frost Valley station concentrations of inorganic monomeric aluminum exceeded 0.3 milligram per liter at the beginning of the study, but never exceeded that level during the last 2 water years of the study. The East Branch Neversink River northeast of Denning and Rondout Creek above Red Brook at Peekamoose stations also showed decreases in inorganic monomeric aluminum concentrations during the study. The reduction in inorganic monomeric aluminum concentrations were the result of reductions in stream acidity. The reductions in stream acidity were driven by reductions in sulfate concentrations in precipitation in response to emission regulations included in title IV of the Clean Air Act Amendments of 1990 (42 USC §7651).

Results indicated increasing trends in sodium and chloride concentrations for all stations with high road density relative to other stations included in the study, which could be a future water-quality concern in the region. The Town Brook watershed southeast of Hobart, the only study watershed that contained dairy farms, had a significant decreasing trend in total dissolved phosphorus concentration that may have been a result of agricultural best management practices implemented on farms by the Watershed Agricultural Program. The watershed with the second highest total phosphorus and total dissolved phosphorus concentrations was a completely forested, but previously agricultural, watershed (Town Brook tributary southeast of Hobart) that had not been actively farmed in about 80 years. The phosphorus concentrations at the Town Brook tributary southeast of Hobart station indicated that previously agricultural watersheds may continue to leach phosphorus to streams for many decades after farming has ceased.

At six of the study watersheds, samples of suspended-sediment and turbidity were also collected. The watersheds with the highest suspended-sediment concentrations and turbidity also had the strongest relations between discharge and suspended-sediment concentrations. In general, the relations between discharge and turbidity were not as strong as the relations between discharge and suspended-sediment concentrations. Results indicated strong relations between suspended-sediment concentrations and turbidity levels at each station; however, relations were less strong in the agricultural watersheds. Suspended-sediment concentrations appeared to decrease at the Stony Clove Creek below Ox Clove at Chichester station following a stream stabilization project completed during the study period. However, we were unable to directly attribute the decrease to the stabilization project; there were many complicating variables that made a direct attribution difficult, such as a series of large storms shortly after the stabilization project was completed and differences in flow conditions before and after the project. However, the results have led to additional monitoring within the watershed specifically designed to determine the effectiveness of stream stabilization projects for reducing suspended-sediment concentrations and turbidity in the upper Esopus Creek watershed, the primary source of water to the Ashokan Reservoir. Water quality in the Catskill and Delaware watersheds is generally improving, and although sodium and chloride concentrations increased at some of the stations from 1999 to 2009, the concentrations in 2009 were still well below U.S. Environmental Protection Agency drinking water standards.