A 2-year study of streamflow and water-quality characteristics in Wind Cave National Park was performed by the U.S. Geological Survey in cooperation with the National Park Service. During this study, streamflow and water-quality data were collected for three of the park's perennial streams (Cold Spring, Beaver, and Highland Creeks) from January 2002 through November 2003. The potential influence of parking lot runoff on cave drip within Wind Cave also was investigated by collecting and analyzing several time-dependent samples from a drainage culvert downstream from the parking lot and from Upper Minnehaha Falls inside the cave following a series of simulated runoff events. The primary focus of the report is on data collected during the 2-year study from January 2002 to November 2003; however, data collected previously also are summarized. Losing reaches occur on both Beaver and Highland Creeks as these streams flow across outcrops of bedrock aquifers within the park. No streamflow losses occur along Cold Spring Creek because its confluence with Beaver Creek is located upstream from the outcrop of the Madison aquifer, where most streamflow losses occur. Physical properties, major ions, trace elements, nutrients, bacteria, benthic macroinvertebrates, organic (wastewater) compounds, bottom sediment, and suspended sediment are summarized for samples collected from 2 sites on Cold Spring Creek, 2 sites on Beaver Creek, and 1 site on Highland Creek. None of the constituent concentrations for any of the samples collected during 2002-03 exceeded any of the U.S. Environmental Protection Agency drinking-water standards, with the exception of the Secondary Maximum Contaminant Level for pH, which was exceeded in numerous samples from Beaver Creek and Highland Creek. Additionally, the pH values in several of these same samples also exceeded beneficial-use criteria for coldwater permanent fisheries and coldwater marginal fisheries. Water temperature exceeded the coldwater permanent fisheries criterion in numerous samples from all three streams. Two samples from Highland Creek also exceeded the coldwater marginal fisheries criterion for water temperature. Mean concentrations of ammonia, orthophosphate, and phosphorous were higher for the upstream site on Beaver Creek than for other water-quality sampling sites. Concentrations of E. coli, fecal coliform, and total coliform bacteria also were higher at the upstream site on Beaver Creek than for any other site. Samples for the analysis of benthic macroinvertebrates were collected from one site on each of the three streams during July 2002 and May 2003. The benthic macroinvertebrate data showed that Beaver Creek had lower species diversity and a higher percentage of tolerant species than the other two streams during 2002, but just the opposite was found during 2003. However, examination of the complete data set indicates that the quality of water at the upstream site was generally poorer than the quality of water at the downstream site. Furthermore, the quality of water at the upstream site on Beaver Creek is somewhat degraded when compared to the quality of water from Highland and Cold Spring Creeks, indicating that anthropogenic activities outside the park probably are affecting the quality of water in Beaver Creek. Samples for the analysis of wastewater compounds were collected at least twice from four of the five water-quality sampling sites. Bromoform, phenol, caffeine, and cholesterol were detected in samples from Cold Spring Creek, but only phenol was detected at concentrations greater than the minimum reporting level. Concentrations of several wastewater compounds were estimated in samples collected from sites on Beaver Creek, including phenol, para-cresol, and para-nonylphenol-total. Phenol was detected at both sites on Beaver Creek at concentrations greater than the minimum reporting level. Bromoform; para-cresol; ethanol,2-butoxy-phosphate; and cholesterol were detected