Progress 09/01/00 to 08/31/04

Outputs
This project addressed the interactions of roots, nitrate and soil redox processes in forested wetlands. Two major efforts were undertaken. First, extended severe drought during the first two years of the project prevented anoxic conditions from developing in the soils of the selected field sites. Therefore, soil samples were used in laboratory cultures to explore the relationships among N additions, nitrate reduction, and iron redox chemistry. Either ammonium sulfate or urea was added, and the cultures were then incubated under anoxic conditions. Over the course of several such experiments, using different levels of N addition, with and without buffers to control pH, nitrate was never observed but a small production of nitrite occurred after 7-14 days incubation. This nitrite rapidly disappeared, and was accompanied by the production of Fe(II) and depletion of Fe(III). Our current hypothesis is that these results suggest a novel process in the nitrogen cycle, in which ammonium oxidation occurs anaerobically and is linked to iron reduction, and the nitrite produced in then denitrified. This result, if verified, would help explain a variety of field-based observations in the literature which are currently difficult to explain. In particular, the presence of active denitrifier communities in soil horizons in which nitrate is absent and in which nitrification is prevented by micro-aerobic or anaerobic conditions could reflect an iron-driven production of nitrite. In addition, the presence of redoximorphic soil features (low chomas, etc.) in lower soil horizons in which nitrification is an unlikely process could be explained by iron reduction mediated by ammonium reduction. A paper reporting these results is now in press in Soil Biology and Biochemistry. Secondly, a field-based study was implemented after the drought. Two sites Assumpink Wildlife Management Area, with similar plant communities and soil profiles, were used. In each site, 12 plots (4 m x 4 m) were delimited; all understory (herbs and shrubs) vegetation were removed from 6 of the plots in each site. Cores from each plot were analyzed for root biomass and soil chemistry. Slow-release ammonium sulphate fertilizer was added to half of each plot type (vegetated and unvegetated). Plots were instrumented with piezometers and groundwater wells, and groundwater levels and direction of flow and soil moisture were monitored on a bi-weekly to monthly basis. Plant samples have been taken for nutrient content, and root screens were installed to monitor root abundance in each plot. Nitrogen cycling rates were measured on four dates (once per season). Analyses of the data are being finalized and papers prepared showing that the presence of roots reduces the amount of Fe reduction, but that the rates of N cycling (mineralization, denitrification) is not affected by the presence/absence of vegetation. Another paper is in preparation, reporting the root biomass distributions and relationship to the above-ground vegetation.

Impacts
The project illustrates the importance of maintaining forest vegetation in wetlands being used for nitrogen removal and water quality protection. The presence of roots through the mineral portion of the soil affects both the dynamics of nitrogen, and the linked redox chemistry that controls other pollutant-immobilizing processes. Moreover, if our preliminary data suggestive of a new process in the nitrogen cycle are validated, this will be a major contribution to the understanding of the fate and transport of nitrogen in the environment. It will also be important in helping to explain several aspects of wetland soils that are not well understood given the conventional understanding of nitrification-denitrification processes.

Publications

• Clement, J-C., Shrestha, J, J. G. Ehrenfeld and P. R. Jaffe. 2005. Ammonium oxidation coupled to dissimilatory reduction of iron under anaerobic conditions in wetland soils. Soil Biology and Biochemistry (in press).

Progress 01/01/03 to 12/31/03

Outputs
This project addressed the interactions of roots, nitrate and soil redox processes in forested wetlands. Two major efforts have been undertaken. First, extended severe drought during the first two years of the project prevented anoxic conditions from developing in the soils of the selected field sites. Therefore, soil samples were used in laboratory cultures to explore the relationships among N additions, nitrate reduction, and iron redox chemistry. Either ammonium sulfate or urea was added, and the cultures were then incubated under anoxic conditions. Over the course of several such experiments, using different levels of N addition, with and without buffers to control pH, nitrate was never observed but a small production of nitrite occurred after 7-14 days incubation. This nitrite rapidly disappeared, and was accompanied by the production of Fe(II) and depletion of Fe(III). Our current hypothesis is that these results suggest a novel process in the nitrogen cycle, in which ammonium oxidation occurs anaerobically and is linked to iron reduction, and the nitrite produced in then denitrified. This result, if verified, would help explain a variety of field-based observations in the literature which are currently difficult to explain. In particular, the presence of active denitrifier communities in soil horizons in which nitrate is absent and in which nitrification is prevented by micro-aerobic or anaerobic conditions could reflect an iron-driven production of nitrite. In addition, the presence of redoximorphic soil features (low chomas, etc.) in lower soil horizons in which nitrification is an unlikely process could be explained by iron reduction mediated by ammonium reduction. We are currently pursuing additional funding to test these observations using both 14C and 15N experiments, and attempting to produce a pure culture that produces these results. Secondly, a field-based study was implemented after the drought. Two sites Assumpink Wildlife Management Area, with similar plant communities and soil profiles, were used. In each site, 12 plots (4 m x 4 m) were delimited; all understory (herbs and shrubs) vegetation were removed from 6 of the plots in each site. Cores from each plot were analyzed for root biomass and soil chemistry. Slow-release ammonium sulphate fertilizer was added to half of each plot type (vegetated and unvegetated). Plots were instrumented with piezometers and groundwater wells, and groundwater levels and direction of flow and soil moisture were monitored on a bi-weekly to monthly basis. Plant samples have been taken for nutrient content, and root screens were installed to monitor root abundance in each plot. Nitrogen cycling rates were measured on four dates (once per season). Laboratory processing of samples is continuing, and should be completed by July 2004. Measurements of iron chemistry (Fe(III) and Fe(II)) are being made concurrently. The data are being analyzed to determine the relationship of N cycling data to both N additions, the presence of roots, and to iron chemistry. Preliminary analysis of the data suggests that the presence of roots reduces the amount of Fe reduction.

Impacts
The project illustrates the importance of maintaining forest vegetation in wetlands being used for nitrogen removal and water quality protection. The presence of roots through the mineral portion of the soil affects both the dynamics of nitrogen, and the linked redox chemistry that controls other pollutant-immobilizing processes. Moreover, if our preliminary data suggestive of a new process in the nitrogen cycle are validated, this will be a major contribution to the understanding of the fate and transport of nitrogen in the environment. It will also be important in helping to explain several aspects of wetland soils that are not well understood given the conventional understanding of nitrification-denitrification processes.

Publications

• Ehrenfeld, J. G., J-C. Clement, J. Shrestha and P. Jaffe. 2003. Effects of roots and nitrates on redox chemistry of wetland soils. Ecological Society of America Annual Meeting, August, 2003.

Progress 01/01/02 to 12/31/02

Outputs
The purpose of this project is to determine whether the roots of woody plants in forested riparian wetlands control the redox status of the soil, and thereby affect the ability of the wetland to remove nitrogen through denitrification processes. We have hypothesized that there are a variety of mechanisms through which roots may affect the availability of nitrate for denitrification and/or affect the oxygen status of the soil, thereby poising the redox system at or below the level at which denitrification takes place. In order to test these ideas, we have established a field experiment and are supplementing the data from the field with laboratory/greenhouse studies. Two study sites were established in Assumpink Wildlife Refuge, a state-owned wildlife management area. Both sites are hardwood wetlands adjoining the Assumpink Creek; one site is immediately downgradient of a field managed alternately in corn and soybeans; the other site is in an agricultural watershed but is bounded by roads which separate it from mixed agricultural/low-density residential land-use. In both sites, 12 plots, each 4 m x 4 m, have been established. Half the plots were cleared of all above-ground vegetation during the first growing season (2001), and have been kept clear. Prior to removing the vegetation, the plant community was sampled to determine initial species composition and stem densities. A set of piezoemeters and a well screened to 1 m were installed in each plot, in order to follow water levels and determine water flow directions. Following vegetation removal, four soil cores were removed from each plot. The cores (two sets of two replicates) were taken to 60 cm, and were sliced into 5-cm segments in the top 30 cm and 10 cm segments in the bottom 30 cm. Each segment is being analyzed for: organic matter, pH, extractable nitrogen, total soluble nitrogen, texture, root biomass, root length, and Fe(III) and Fe(II). All laboratory analyses have been completed except for the root biomass/root length; we expect these analyses to be completed by March 2003. Data analysis will focus on correlations between root biomass, above-ground plant community structure, and soil redox status as modeled from the soil chemistry. Because of the severe drought last summer, water tables were below the wells (although soil properties, including gley colors and pronounced mottling indicated that water levels are frequently within the top 30 cm ). We therefore decided to set up greenhouse microcosms to explore the root-denitrification/Fe reduction system, as soil saturation could not be relied on in the field. Soils were excavated and greenhouse systems set up, but due to problems with seed germination, these studies are still in progress. Containers have been established, however, without plants, to explore the relationship between nitrate additions and the responsiveness of the Fe reduction system. We will be presenting the results of these experiments at the Ecological Society of America annual meeting this summer.

Impacts
Because riparian forested buffers are now a "best management practice" for the removal of excess nitrogen from nonpoint sources, it is important to know whether the density of roots, reflecting the vegetative structure, affects the capacity of the wetlands to perform this function. The results should allow us to make recommendations about the importance of managing riparian vegetation in order to maximize nitrogen removal.

Publications

• No publications reported this period

Progress 01/01/01 to 12/31/01

Outputs
The purpose of this project is to investigate the relationship between root biomass and growth in forested wetlands and the biogeochemistry of the soils, with an emphasis on redox chemistry. We have proposed that there are numerous pathways of feedback between plant roots and the nitrogen dynamics and redox status of wetland soils, and that these feedback may have a major effect on the retention/denitrification/transmission of N to surface waters. We have located two study sites in the Assumpink Wildlife Management Area (a state-owned reserve). Each contains mature hardwood forested wetland adjacent to the Assumpink Creek, but the sites are situated about two miles apart; thus they can be considered independent replicates. In each site, we have established 12 plots distributed across the areas of wetland soils (criterion: clayey texture, chroma < 2 and presence of redoximorphic features), each 4 m x 4 m. In each plot, we have censussed the vegetation, established a monitoring well to record water table levels, and taken initial soil samples to document pH, soil organic matter, extractable nitrogen fractions, N mineralization rates, and iron chemistry (concentrations of FeIII and FeII), in samples taken at 5 cm increments to a depth of 60 cm. In a replicate set of samples from each plot, we are isolating the root biomass, and using imaging software to determine the root length and surface area in each 5 cm increment. In one-half of the plots at each site, all above-ground vegetation was removed in fall, 2001, and the plots trenched to sever roots (in order to create root-free treatments). Because of the severe and continuing drought in the region, the study areas have not been subjected to the expected hydrological regime (based on the soil and vegetation properties and proximity to the streams). In order to effectively test the hypotheses initially proposed, we have decided to conduct a greenhouse-based microcosm experiment, rather than proceed with the field-based experiments that were initially proposed, so that we can maintain wetland hydrology. We are currently collecting 30 cm cores in plastic liners for this experiment. Nitrate at several concentrations will be applied to the microcosms, half of which will be planted with red maple seedlings. The microcosms will be maintained at near-saturated conditions (as would be expected for normal field conditions). We will monitor N fractions, denitrification, iron chemistry and redox potential to determine redox status of the soils, and determine the abundance and distribution of roots at the conclusion of the experiment. We will also continue to monitor the field plots for root biomass, N fractions and denitrification, iron chemistry and redox status, and will use correlation analysis to relate the soil chemistry to root abundance. Hopefully, the drought will end and wetland hydrology will become re-established.

Impacts
We expect the project to have impact on the evaluation of forested wetlands for water quality protection. By knowing the relationship of root biology to the capacity of the site to remove N, it will be possible to use surveys of vegetation density to realistically estimate water quality protection functions. Since such relationships are widely used in wetland management programs (e.g., hydrogeomorphic assessment methods), the results should be widely applicable.

Publications

• No publications reported this period