Connecting Landscape Nitrogen Loads to Groundwater Nitrate Concentrations

Elevated nitrate concentrations in drinking water are linked with significant health risks, with recent studies showing links between cancer and adverse birth outcomes at concentrations far lower than the US EPA standard (10 mg/L NO₃-N). Because of this risk, understanding the connections between landscape nitrogen loading and concentrations in groundwater is critical. Identifying groundwater nitrogen hotspots can help target landscape management to avoid nitrogen pollution in drinking water wells. Additionally, groundwater is an important, often slow, pathway of nitrogen to streams and lakes where it contributes to eutrophication and harmful algal blooms. Here, we combine a novel groundwater chemistry dataset from drinking water wells with sources of N inputs to the landscape, N loads to groundwater, land use, and soil characteristics to provide a new view of the nitrogen landscape. We utilize a dataset of over 7000 well chemistries collected by county health departments in Michigan, USA. Using this dataset, we visualize patterns in space and time in groundwater nitrate levels. We connect the surface sources to the subsurface through SENSMap, the Spatially Explicit Nutrient Source Map, which maps 7 sources of nitrogen inputs at 30 m resolution across the US Great Lakes Basin (ca. 2010) using broadly available data, remote sensing products, and statistical methods. Groundwater source sheds, or capture areas, for each well were delineated using a state-wide groundwater model for wells with NO₃-N concentrations sampled since 2000. By combining well nitrate concentrations with groundwater source shed characteristics using statistical and machine learning methods, we provide new insight about the effects of land management practices on subsurface nitrate concentrations.