Samples collected and analyzed by the Oklahoma Department of Agriculture, Food, and Forestry from 1999 to 2001 determined that nitrate exceeded the U.S. Environmental Protection Agency maximum contaminant level for public drinking-water supplies of 10 milligrams per liter as nitrogen in 79 monitoring wells at 35 swine licensed-managed feeding operations (LMFO) in Oklahoma. The LMFOs are located in rural agricultural settings where long-term agriculture has potentially affected the ground-water quality in some areas. Land use prior to the construction of the LMFOs was assessed to evaluate the types of agricultural land use within a 500-meter radius of the sampled wells. Chemical and microbiological techniques were used to determine the possible sources of nitrate in water sampled from 10 wastewater lagoons and 79 wells. Samples were analyzed for dissolved major ions, dissolved trace elements, dissolved nutrients, nitrogen isotope ratios of nitrate and ammonia, wastewater organic compounds, and fecal coliform bacteria. Bacteria ribotyping analysis was done on selected samples to identify possible specific animal sources. A decision process was developed to identify the possible sources of nitrate. First, nitrogen isotope ratios were used to define sources as animal, mixed animal and fertilizer, or fertilizer. Second, wastewater organic compound detections, nitrogen-isotope ratios, fecal coliform bacteria detections, and ribotyping were used to refine the identification of possible sources as LFMO waste, fertilizer, or unidentified animal or mixtures of these sources. Additional evidence provided by ribotyping and wastewater organic compound data can, in some cases, specifically indicate the animal source. Detections of three or more wastewater organic compounds that are indicators of animal sources and detections of fecal coliform bacteria provided additional evidence of an animal source. LMFO waste was designated as a possible source of nitrate in water from 10 wells. The source of waste in water from five of those wells was determined through ribotyping, and the source of waste in water from the remaining five wells was determined by detections of three or more animal-waste compounds in the well samples. LMFO waste in the water from wells with unidentified animal source of nitrate does not indicate that LMFO waste was not the source, but indicated that multiple animal sources, including LMFO waste, may be the source of the nitrate.