Changes in riparian and stream hydrology and biogeochemistry following storms in an agricultural watershed

Thursday, 26 January 2017
Ballroom II (San Juan Marriott)
Molly Welsh, SUNY College of Environmental Science and Forestry, Environmental Science, Syracuse, NY, United States, Philippe Vidon, SUNY College of Environmental Science and Forestry, Forest and Natural Resources Management, Syracuse, NY, United States and Sara McMillan, Purdue University, Agricultural and Biological Engineering, West Lafayette, IN, United States
Abstract:
Quantifying changes in stream and riparian biogeochemistry following rainfall events is critical for watershed management, as shifting water levels and soil redox conditions can lead to hot moments of greenhouse gas (GHG) emissions and nutrient export. We assessed shifts from baseline conditions in hydrology and biogeochemistry at 24 and 72 hours post-rainfall following storms of three different magnitudes — 1.65 cm, 3.15 cm, and 6.35 cm. In the riparian zone, we measured GHG fluxes (CH4, CO2, N2O) at the soil-atmosphere interface using static chambers, examined groundwater nutrient chemistry, created groundwater surface maps, and calculated hydraulic head gradients. In the stream, we obtained surface and subsurface water chemistry (NO3-, NH4+, PO43-, TDN, DOC) and calculated changes in vertical hydraulic gradients. Generally, following storm events, the near-stream zone shifted from a sink to source of CH4 while the upper riparian zone shifted from a sink to source of N2O. Post-storm CO2 emissions across the riparian zone were comparable to background fluxes. However, we observed shifts in the magnitude and direction (emission v. consumption) of GHG fluxes at different time points following rainfall owing to changes in water table height, soil moisture, and soil redox status based on landscape position. Subsurface water chemistry also influenced GHG fluxes; the upper riparian zone (at the edge of the agricultural field) had higher TDN and NO3- concentrations, which were both significantly correlated to N2O emissions. Release of high concentrations of PO43- from the riparian zone to the stream, likely due to reduction of phosphorus-bound iron following denitrification, also occurred during the two largest storms we studied. Additionally, high concentrations of NO3- in subsurface water below riffles indicates that turbulence in riffles may promote nitrification, which can be particularly important as elevated in-stream NH4+ concentrations were observed following storm events over 3 cm. Though we observed variation in shifts in the magnitude of upwelling and downwelling in pools, riffles, and runs, areas of downwelling of NO3--rich surface water into stream sediments rich in organic matter and DOC (such as runs) may function as hotspots for denitrification.