Challenges in linking soil health to edge-of-field water quality across the Great Lakes basin.

To better understand agricultural nutrient losses, we evaluated relationships between management (e.g., manure and tillage), soil health measurements, and resulting edge-of-field (EOF) surface water quality. This work was conducted before or early into conservation implementation at 14 Great Lakes Restoration Initiative EOF sites spanning Wisconsin, Michigan, Indiana, Ohio, and New York. Analyses of site characteristics (hydroclimate, management, catchment properties) along with 3 yr of soil health measurements (chemical, biological, and physical properties) showed EOF-nutrient export depended on both site and soil properties. A pattern emerged whereby sites not receiving manure and sites with manure defined opposite ends of several gradients for soil and water data. Sites receiving manure had increased microbial activity, organic matter (3.2 vs. 2.7%), and soil test phosphorus (P) (2.8 times more) relative to sites without manure. Suspended sediments (SS), total P (TP), and total nitrogen (TN) in EOF surface runoff varied over three to five orders. Multivariate analysis among sites showed covariant linkages between soil nutrients, soil C, microbial properties, and nutrients in EOF water. There were positive univariate relationships between water-extractable soil P and annual EOF-water concentrations and yields of orthophosphate, TP, TN, and SS (p < .01). Some soil physical properties (e.g., bulk density and infiltration) also covaried among sites but were not consistently related to runoff index or water yield variables. Given the observed among-site variability, we were not able to isolate desirable soil health signals on EOF surface water quality.

Well water in karst regions of northeastern Wisconsin contains estrogenic factors, nitrate, and bacteria.

Well water in karst regions is particularly susceptible to contamination by various nonpoint source pollutants such as nitrate, fecal bacteria, and endocrine disrupting chemicals (EDCs). This study analyzed 40 wells in heavily farmed karst areas of northeastern Wisconsin to determine whether these and other pollutants are present, and if so, whether their presence is (1) correlated with other contaminants and (2) exhibits seasonal variation. Nitrate, bacteria, and estrogenicity (indicating the presence of EDCs) were present in at least some of well water samples collected over the course of four time periods between the summers of 2008 and 2009. Although estrogenicity was greatest during the summer months, bacterial contamination was most prevalent during snowmelt. Levels of estrogenicity present in some well water samples approached a threshold concentration that is known to exert endocrine disruption in wildlife. Strong correlations between estrogenicity and other water quality parameters were not found.