Accurately measuring denitrification in stochastic floodplains, particularly the leveed and unleveed reaches of the Mississippi River basin, requires innovative experiments. To replicate hydraulic variability ranging from overland flooding to groundwater exfiltration in floodplain wetlands, I incubated sediment cores collected from four field sites across the Dogtooth Bend of the middle Mississippi River; pairing novel deep injection (Graphic Abstract Fig. A, left) with traditional surface delivery (Graphic Abstract Fig. A, right) of both oxic and anoxic Mississippi River water. In sandy sediments with unconstrained flux of nutrients, denitrification more than doubled across a range from 192 to 429 mg N m-2 day-1 in a linear anoxic-injection hierarchy of anoxic deep > anoxic surface > oxic deep > oxic surface treatments. In contrast, for incubations in diffusion-limited clay sediments, injection type made no difference; however, in anoxic conditions denitrification rates were as high as 435 mg N m-2 day-1 compared to oxic incubations at 187 mg N m-2 day-1. This methodology reveals the magnitude of diverse denitrification rates spanning different hydrologic conditions (Abstract Fig. B) and the mediation of denitrification by sediment type. These findings provide quantified bounds to inform resource management decisions regarding what areas should be selected for protection or hydrologic reconnection to best facilitate nutrient processing services like denitrification under varying hydrologic conditions.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

At Crabby Creek, an urbanized watershed in northeast Chester County, Pennsylvania, an NaCl tracer test was conducted in 2010 to assess changes in hyporheic flow from a 2009 tracer test around the same stream restoration J-Hook. This project compares the 2009 and 2010 tracer test breakthrough curves and geophysical time-lapse resistivity surveys. This project also compares elevation cross sections and tile probing from 2009 and 2010, both measured upstream and downstream from the J-Hook. To confirm areas of lingering tracer seen in the time-lapse resistivity profiles, sediment cores using the freeze core method were taken to measure pore water for tracer. This project also measured diurnal temperature flux through the streambed at several locations along the sample site to model vertical water and heat flux. The breakthrough graphs constructed from the conductivity of the well water samples shows similar hyporheic flow characteristics from 2009 to 2010. The time-lapse resistivity profiles show an area of lingering tracer upstream from the J-Hook in 2010 that is similar in shape and location to an area upstream from the J-Hook in the 2009 profiles. However, an area of lingering tracer downstream from the J-Hook present in 2009 as a round feature on the profile is now a thin linear feature. The freeze cores show tracer present in the pore water after the end of the tracer injection in the stream sediment, confirming areas of lingering tracer seen in the time-lapse resistivity profiles. The grain size analysis of the freeze cores and the comparison to the 2009 cores taken at Crabby Creek show similar grain size distribution upstream from the J-Hook. Downstream from the J-Hook the grain size analysis shows a redistribution of sediment. Upstream from the J-Hook the tile probe shows both shallower and deeper bedrock, a redistribution of sediment but no net erosion. Downstream from the restoration structure, however, the tile probe data show a sediment loss of 20 cm. Elevation cross section surveys from 2009 and 2010 confirm what the tile probing found, a loss of sediment downstream but not upstream from the J-Hook. Temperature modeling of heat flux through the sediment shows that the diurnal temperature distribution can be accounted for without vertical flux. Thus, the immobile regions upstream and downstream from the J-Hook seem to be related to sediment distribution rather than hydrologic gradient differences. The significance of this study shows the need to use multiple techniques to characterize the immobile zone as a part of hyporheic flow. The immobile zone is an important area of chemical reactions in the streambed. At Crabby Creek the central J-Hook inhibits net erosion patterns upstream from the structure, allowing for the continued presence of an immobile zone. Downstream from the central J-Hook the erosion of the streambed sediment led to a decrease in size and location of the immobile zone. The disturbance of sediment around restoration structures influences the development of a healthy hyporheic flow and needs to be studied for future restoration of impaired streams and riparian corridors. / Geology

The Wissahickon Creek is an urban stream that runs through Montgomery and Philadelphia Counties and discharges to the Schuylkill River in Philadelphia. A majority of stream segments in the Wissahickon watershed are considered impaired by the USEPA due to sediment and nutrients. Total Maximum Daily Loads (TMDLs) were implemented in 2003 for nutrients (NO3-, PO43-, NO2-, and CBOD5) and siltation. A new TMDL for total phosphorus (TP) was proposed in 2015, despite minimal data on the effectiveness of the 2003 TMDLs. This new proposal was met with concern, suggesting more data must be collected to better understand impairment in the Wissahickon Creek. The purpose of this research was to study turbidity and nutrient responses to storm events, as storm events are known to contribute significant loads of both sediment and nutrients. Twelve sites were chosen for high frequency turbidity and water level monitoring along the Wissahickon Creek and one of its main tributaries, Sandy Run. These sites were selected around three of the major wastewater treatment plants (WWTPs) to determine the relative roles of WWTPs and overland flow as sources of turbidity and nutrients during storm events. The upstream site and first downstream site at each WWTP were monitored for nutrients during storms using high frequency loggers and ISCO automatic samplers. Stream assessments were done at each site to characterize in-stream physical parameters, bank vegetation, and algae cover. High frequency turbidity data suggests that the turbidity is locally sourced, as turbidity peaks at the same time as water level, or within an hour or two, at all sites regardless of storm size. Comparisons of the turbidity response with in-stream parameters and land cover helped determine that the main factor driving the turbidity response is discharge, although bank topping and impervious cover, particularly roads, may increase turbidity responses at some sites. Similarities in nutrient, turbidity, and conductivity responses upstream and downstream of the WWTPs strongly suggest that overland flow, not WWTP effluent, is the major source of nutrients and sediment during storm events. Finally, a strong relationship between total phosphorus and high turbidity suggests that only during high discharge events is there a significant increase in TP in the Wissahickon Creek. Results from this research identify the source of turbidity and nutrients to the Wissahickon Creek during storms as primarily coming from overland flow, that the primary factor controlling the turbidity response is discharge, with some secondary influence from over-banking and the contribution of roads to land use, and a close link between TP concentrations and sediment during storms in the stream. / Geology