Impairment of urban freshwater by eutrophication, in particular from excess nonpoint nitrogen (N) and phosphorus (P) pollution, remains a pressing issue in the management of urban watersheds. Specific nutrient sources and transport pathways are not particularly well understood in urban landscapes, in part due to the complexity of the human-modified land surface and drainage network. While much work has focused on nutrient loading by stormwater, baseflow in storm drains, which results from groundwater seepage into storm drains or outflow from drain-connected surface waters, may make a substantial contribution to water and nutrient loading of downstream waters, yet has rarely been addressed in previous studies. We investigated the relative importance of baseflow versus stormflow for N and P flux through the drainage networks of six urban sub-watersheds in St. Paul, MN, USA. Samples collected during baseflow periods at lake outlets, streams, springs, and small drains within and near the monitored sub-watersheds were analyzed for dissolved inorganic carbon (DIC), chloride (Cl-), and oxygen isotope ratios of water (δ18O) to investigate the relative contribution of surface water and groundwater sources to the storm drain networks during dry weather periods.
Results/Conclusions
Baseflow in St. Paul storm drains contributed substantially to total warm season (May – October) water yields (27% to 66% across sites), with the highest fractions in sub-watersheds with buried streams and drain-connected surface water. Fractions of total nutrient yields due to baseflow were also significant, especially for N (33% to 68% across sites), with smaller yields for P (8% to 34% across sites). Nitrate (NO3-), DIC, Cl-, and δ18O measurements suggested that all sites were dominated by groundwater inputs during baseflow, but that contribution from surface water sources varied. Tracer measurements from springs and lake outlets across the watershed explained the lower NO3-, Cl-, DIC, and δ18O observed at sites with upstream connections to lakes and ponds relative to groundwater-dominated sites. Additionally, elevated Cl- at several sites suggested the presence of shallow groundwater contaminated by winter road salt applications. These results imply that consideration of infiltrated nutrients and ions is necessary in the development of stormwater management practices in watersheds with high water tables and deep, complex storm drain networks that intersect shallow aquifers. Management of N in particular will require a focus on source reduction through improved land management and consideration of surface water connections to storm drains.