Differences in Nitrogen Cycling Characteristics in a Natural and Reclaimed Wetland: Implications for Nitrogen Load Mitigation.

Enter abstract here

Former natural wetlands are often “reclaimed” in an attempt to enhance particularly valuable ecosystem functions (e.g., nutrient and sediment removal from surface waters, wildlife conservation, and groundwater recharge). The effectiveness of reclamation is seldom documented after the manipulation is in place.  Halfway Creek Marsh Complex (HWC), Holmen, Wisconsin, was recently reclaimed by the US Fish and Wildlife Service from agricultural use for water fowl habitat and sediment retention from storm flows from the Creek, a tributary of the Upper Mississippi River. The upstream portion of HWC was recently reclaimed from agricultural use and managed to receive peak storm flows through a single gated culvert.  The downstream portion of HWC is an unmodified natural marsh (NM) that intercepts floodwaters throughout its length.     The NM is dominated by Phalaris on drier sediments and diverse vegetation on water saturated sediments.  Wet areas of the NM contain greater soil total N (1.5 %) and total C (22 %) than the RM (TN: 0.5 %, TC: 9.5 %). We evaluated rates of sediment/soil nitrification, denitrification, N and C and moisture to determine how well the RM performed as a processor of N relative to NM.  Over 3.5 years we also made detailed estimates of sediment and nutrient loads as Creek waters flowed through the HWC.  During the wettest summer (2004 discharge: 1.6 x 10^-2 km³ y^-1) denitrification (enzyme activity: DEA) was greatest in water-saturated areas of the NM (2.2 μg-N cm² hr^-1) and lowest in the RM (0.5 μg-N cm² hr^-1); while nitrification rates were highest in the drier areas (0.35 μg-N cm² hr^-1), whereas the RM was lowest (0.3 μg-N cm² hr^-1).  The summer of 2005 was drier than in 2004 (1.0 x 10^-2 km³ y^-1) and DEA was higher in the RM (8.1 μg-N cm² hr^-1) than in the NM (6.2 μg-N cm² hr^-1) and all sites were higher than in 2004.  Nitrification rates were highest in the dry pond area of the RM (0.70 μg-N cm² hr^-1), and the NM was similar to 2004 (0.18 μg-N cm² hr^-1).  The RM captured between 0.6 (2006: 5 mton) and 6.2 (2004: 573 mton) percent of the transported sediment load while the NM captured between 64 (2006: 534 mton) and 70 (2005: 1698 mton) percent of the sediment load.  The RM retained between 0.1 (2006: 24 kg) and 4.9 (2004: 2,021 kg) percent of the total nitrogen (TN) load while the NM captured between 8.2 (2006: 2610 kg) and 15 (2005: 5526 kg) percent of the TN load.  Correlations between DEA and soil percent N (r=0.27, p=0.015) and percent C (r=0.28, p=0.013) suggest that the wet NM is better suited for microbial removal of nitrates than the RM. While the denitrification potential was quite high in portions of the RM, reduced sediment C, and lack of NO3 loading appears to limit N-removal capacity relative to the NM.  Furthermore, management strategies of the RM are such that to enhance wildlife diversity limit greater sediment and nitrogen removal from Halfway Creek.