Bartsch, Will Melvin2012-08-092012-08-092012-06https://hdl.handle.net/11299/130934University of Minnesota M.S. thesis. June 2012. Advisors: Richard P. Axler and George E. Host. 1 computer file (PDF); ix, 66 pages, appendices A-H.The St. Louis River drains an area of 9,412 km2 (3,634 mi2) and empties into the western arm of Lake Superior. Shortly downstream from the Fond du Lac Reservoir, the river opens up into a 48.5 km2 freshwater estuary that separates Minnesota and Wisconsin. A GIS-based anthropogenic stressor gradient was developed to characterize the anthropogenic stressors within the watershed. The components of the stressor gradient were: road density, point-source pollution permit density, population density, percent agricultural land and percent developed land. Water quality sampling was conducted at 26 sites in the estuary in both nearshore areas and above the mouths of the associated tributaries during multiple flow regimes in 2010-2011. Additional data were analyzed from 34 upper watershed sites sampled in 2009-2010. The stressor gradient was shown to be significantly, positively correlated (p<0.1) with TSS, turbidity, TP, NO2-/NO3--N, DIN, dissolved oxygen saturation, pH, specific electrical conductivity, chloride , sulfate, E. coli, and hardness in the upper watershed. In the estuary it was significantly, positively correlated with NO2-/NO3--N, NH4+-N, DIN, and chloride at multiple flow regime and location combinations. The strength of the correlations of the stressor gradient and water quality was generally improved by the removal of the less relevant agricultural component. Sediment-related parameters were either not correlated with the stressor gradient in the estuary or anomalously were negatively correlated. This was found to be due to the non-uniform distribution of more erodible soils. Soil K factor (an erosivity index from SSURGO) was significantly, positively correlated with the sediment-related parameters. Although it was originally designed to help stratify sampling programs across a gradient of stress and identify reference areas for restoration projects, the stressor gradient was shown to have some predictive power for multiple water quality parameters.en-USNatural Resources ManagementThesis or Dissertation