Progress 10/01/06 to 09/30/09

Outputs
OUTPUTS: Since May 2008 we have monitored stream nitrate-nitrogen (N), chloride, and discharge monthly at 19 stream stations within a total of seven streams in four RI watersheds. We will use elements of the SPARROW model to estimate the potential extent of in-stream denitrification. With field data we will calculate the nitrate flux between stations along a stream reach to determine in-situ nitrate-N removal rates. All samples collected through Sept 2009 have been analyzed and data calculations have commenced. In fall 2008 we harvested woody debris blocks that had been tethered to bricks and placed in streams in May 2008. We used these woody debris blocks in laboratory mesocosm experiments (2nd set in our lab) to assess the impact of woody debris on in-stream denitrification. In the lab we amended the plastic containers holding the wood blocks and stream water with 10 mg/L nitrate-N, reduced the ambient dissolved oxygen content to less than 2 mg/l with helium gas, sealed the mesocosm units with silicone and epoxy, and added acetylene/helium to the headspace via a septum in the lid. We took four sets of headspace samples per mesocosm over four days to analyze for N2O denitrification gas. A blank mesocosm containing stream water and a block cut at the same time but never installed at a stream location was also set-up. We sampled and analyzed the water in the mesocosms at the start and end of the incubation for nitrate-N and calculated nitrate-N removal for each mesocosm. Gas samples were analyzed for N2O; denitrification rates will be calculated. In summer 2009 we repeated the woody debris experiment in the stream with the highest rates of nitrate-N removal in 2008. We reinstalled 20 of the May 2008 cut wood blocks, tethered to bricks, into our East Greenwich, RI stream in June 2009. We added natural woody debris and a clay block treatment to the study. Natural woody debris from the stream was bundled and tethered to bricks and positioned adjacent to the cut blocks in the stream. To isolate the effects of the woody substrate from the effects of biofilms created during the stream incubation, we crafted clay blocks similar in size to the wood blocks, tethered them to bricks and placed them in the stream. In August of 2009 we harvested all units, placed them in plastic containers with stream water, and set up our mesocosm treatments in the lab. Half of the mesocosms were run with the acetylene inhibition method as in 2008 to quantify denitrification through the analysis of N2O. The remaining half were run with the paired-isotope technique with the addition of 15N-enriched nitrate-N and sampled for 15N2 and 15N2O gases. Water samples were taken for initial and ending nitrate-N concentration. We took four gas samples per mesocosm over the course of 18-24 hours. In June 2009 we cut down another red maple tree and cut it into blocks. We tethered these freshly cut wood blocks to bricks and put them in a new stream reach in PA. At the PA stream we also added natural woody debris bundles composed of sticks found in the same stream reach and clay bricks as in our RI stream. In August 2009 we ran the mesocosm experiments in PA, similar to the RI study. PARTICIPANTS: Participants of this project included Arthur J. Gold, principal investigator; Kelly Addy, responsible for field work, analysis performed at URI, and supervision of graduate and undergraduate students; and Suzanne Cox, graduate research assistant. The Rhode Island Agricultural Experiment Station (RI AES) provided the tuition waiver and assistantship salary for S. Cox in the fall of 2008 and spring of 2009. S. Cox was responsible for field work and GIS analyses. The RI AES also funded the undergraduate 2009 summer coastal fellowships for Lauren Creamer and Mat Lautenberger. Several undergraduates were involved in the project as general field and laboratory assistants. A. Gold incorporated in-stream dosing of tracers into a laboratory of his undergraduate course "Watershed Hydrology and Management." TARGET AUDIENCES: We incorporated the results of the project into our funded CSREES NIWQP project entitled "Advancing watershed N management at the local level: Incorporating stream reach ecosystem N sinks into a spatial decision support system." This project will target our outputs to State, local and regional decision makers along with NRCS staff. The CSREES funded National NEMO (Non-point Education for Municipal Officials) Program has reviewed our results and plans to incorporate suggestions and recommendations into their programming aimed at local watershed protection. Undergraduate students at URI were a target audience of our efforts. Two undergraduates had an experiential learning experience as URI Undergraduate Coastal Fellows working the summer in our laboratory on this project. Dr. Gold incorporated in-stream dosing of tracers and watershed nitrate sources and sinks into a laboratory of his undergraduate course "Watershed Hydrology and Management." PROJECT MODIFICATIONS: Due to the severe drought in the summer and fall of 2007, headwater streams in RI were at minimal flows or even dry prohibiting us from conducting the in-stream, reach-scale tracer denitrification studies in 4-5 headwater streams as initially proposed in year one. We were still able to obtain valuable data from our GIS analyses, site selection, and work on the effects of buried woody debris on denitrification. Rather than assessing in-stream denitrification with reach-scale tracer studies as outlined in our proposal, we began our assessment of the impact of woody debris on in-stream denitrification by installing a series of red maple woody debris blocks in six stream reaches. These blocks were harvested for a laboratory mesocosm experiment in August and November of 2008. The experiments were modified and redone in the summer of 2009.

Impacts
Through a random sampling of streams in GIS, we determined that headwater streams (first and second order streams) account for 70% of the length of the drainage network in RI. About 45% of the headwater stream length in RI was found to have comparable surficial geology, soils and land cover to the stream reach used in a previous in-stream nitrogen removal study in RI. The average first order stream length and second order stream length in RI was found to be 0.51 and 0.59 km, respectively. In order for in-stream denitrification to be a substantial sink (>20%) for watershed nitrogen in southern New England, denitrification uptake lengths (length of stream a nitrate molecule travels downstream before being denitrified) must be less than 2-5 km, approximately 5-10 fold longer than most headwater stream lengths in RI. A recently completed project on in-stream denitrification within an undisturbed stream reach in this area indicated an uptake length of at least 170 km, far in excess of this target. The extended uptake length resulted from a combination of high stream velocities (low retention time) in headwater streams and a relatively high concentration of nitrate-N (> 0.5 mg/l) compared to many other in-stream studies of denitrification. While we are still assessing our 1.5 yrs of stream flow data, nitrate-N and Cl concentration data at 19 stream stations indicate that while most stream locations do not have elevated rates of nitrate-N removal, at least 2-3 sites demonstrated significant rates of nitrate removal on more than a few instances. We will be determining the nitrate-N flux and mass removed at each stream reach at every point sampled which will be a robust data set to expand on our GIS date assessment. While we are still assessing our N2 and N2O data on our 2008 and 2009 mesocosms to quantify denitrification, our nitrate-N data in the mesocosms indicates that fresh woody debris can enhance nitrate-N removal rates in RI and PA streams. The PA streams with higher ambient nitrate-N rates had higher rates of nitrate-N removal. Headwater streams with impoundments and woody debris might have longer retention times and more fuel for denitrification than narrow reaches comprised primarily of riffles and runs without impoundments, wetlands, or woody debris. The uncertainties surrounding in-stream N removal warrant further investigation of settings with extended retention times and benthic interactions.

Publications

• Kellogg, D.Q., A.J. Gold, S. Cox, E. Wentz, K. Addy, J. Rozum, and P. Groffman. 2009. Targeting watershed nitrogen export at the local level: The role of landscape sinks. USDA-CSREES National Water Conference, St. Louis, MO.
• Greenwood, C., K. Addy, A. Gold, C. Sawyer, and B. Vinhateiro. 2008. Woody Debris and Nitrogen Cycling in Rhode Island Streams. Annual University of Rhode Island Undergraduate Coastal Fellows Ceremony, Kingston, RI.
• Kellogg, D.Q., A.J. Gold, P.M. Groffman, M.H. Stolt, and K. Addy. 2008. Riparian ground-water flow patterns using flownet analysis: Evapotranspiration-induced upwelling and implications for N removal. Journal of the American Water Resources Association 44:1024-1034.
• Gold, A., D.Q. Kellogg, S. Cox and K. Addy. 2009. Geospatial Approaches for Assessing Denitrification Sinks at the Local Level. NSF Research Coordination Network Workshop: Managing Denitrification in Human Dominated Landscapes, Narragansett, RI.

Progress 10/01/07 to 09/30/08

Outputs
OUTPUTS: We used Geographic Information System (GIS) ArcMap 9.1 software, data from the RIGIS Rhode Island GIS Data, and national hydrography data (NHDplus) to assess the proportion of first and second order streams in Rhode Island relative to their concurrent land use, surficial geology, slope and soil type. This information assessed the range of settings and conditions found within RI headwater streams and helped us select streams for field evaluation. Due to an unusually severe drought in 2007, most of the headwater stream reaches we had selected were dry. We then undertook an alternative set of field studies to examine denitrification capacity of woody debris. At two sites along Green Hill Pond, South Kingstown, RI, a nitrogen (N) bio-reactor structure, composed of a woody debris-filled trench parallel to the shore (3.6 m long x 2 m wide) extending 1.8 m into nitrate-enriched groundwater, had been constructed in the summer of 2004. We conducted tracer studies to examine the N removal capacity of woody debris. The study required the installation and monitoring of intensive well networks. We ran bromide dosing trials per site to capture and characterize the plumes. Approximately 2500 samples were collected and more than 1700 samples were analyzed. Finally in the fall of 2007 when stream levels started to rebound, we were able to return to our stream reach selections for stream monitoring and characterization. By May of 2008, we began to monitor stream nitrate-N, chloride, and discharge on a monthly basis at 19 reaches within a total of seven streams in four Rhode Island watersheds. With this information, we will be able to run the SPARROW model to determine the model's prediction of the significance of in-stream denitrification. We selected 3 pairs of sites each from within the same watershed for an assessment of the effect of woody debris on in-stream denitrification. Within each pair, we could compare a site with high versus low ambient nitrate-N and high versus low % shade. We cut a live red maple tree down in May of 2008 to create a set of woody debris blocks (25 cm X 4.5 cm X 2.5 cm) of common age, species and history. Ten woody debris blocks each tethered to a single brick, were installed at each site in June 2008. In August of 2008 we randomly harvested three brick/block units in plastic containers with stream water from two pairs of sites. In the laboratory we amended these units with 10 mg/L nitrate-N, reduced the ambient dissolved oxygen content to less than 2 mg/l by bubbling helium into the water, sealed these mesocosm units with silicone and epoxy, and added acetylene/helium to the headspace via a septum in the lid. We took four sets of headspace samples over the course of four days to analyze for the generation of N2 denitrification gases. A blank mesocosm containing stream water and a block cut at the same time but never installed at a stream location was set-up similarly. Another mesocosm run is planned for the fall of 2008. In addition we plan to remove additional blocks to quantify dry biomass and chlorophyll a on the blocks and use the extract from these harvests in acetylene block bioassays with stream sediments. PARTICIPANTS: Participants of this project included Arthur J. Gold, principal investigator; Kelly Addy, responsible for field work, analysis performed at URI, and supervision of graduate and undergraduate students; and Holly Meehan and Suzanne Cox, graduate research assistants. The Rhode Island Agricultural Experiment Station (RI AES) provided the tuition waiver and assistantship salary for H. Meehan in the spring of 2007 and for S. Cox in the fall of 2007, spring of 2008, and fall of 2008. H. Meehan and S. Cox were responsible for field work and GIS analyses. The RI AES also funded the undergraduate summer coastal fellowship for Marissa Kelly in 2007 and Colin Greenwood in 2008. Several undergraduates were involved in the project as general field and laboratory assistants. A. Gold incorporated in-stream dosing of tracers into a laboratory of his undergraduate course "Watershed Hydrology and Management." TARGET AUDIENCES: We incorporated the results of the project into a recently funded CSREES NIWQP project entitled: "Advancing watershed N management at the local level: Incorporating stream reach ecosystem N sinks into a spatial decision support system." This project will target our outputs to State, local and regional decision makers along with NRCS staff. The CSREES funded National NEMO (Non-point Education for Municipal Officials) Program has reviewed our results and plans to incorporate suggestions and recommendations into their programming aimed at local watershed protection. Undergraduate students at URI were a target audience of our efforts. One undergraduate had an experiential learning experience as a URI Undergraduate Coastal Fellow working the summer in our laboratory on this project. Dr. Gold incorporated in-stream dosing of tracers and watershed nitrate sources and sinks into a laboratory of his undergraduate course "Watershed Hydrology and Management." PROJECT MODIFICATIONS: Due to the severe drought in the summer and fall of 2007, headwater streams in RI were at minimal flows or even dry prohibiting us from conducting the in-stream, reach-scale tracer denitrification studies in 4-5 headwater streams as initially proposed in year one. We were still able to obtain valuable data from our GIS analyses, site selection, and work on the effects of buried woody debris on denitrification. Rather than assessing in-stream denitrification with reach-scale tracer studies as outlined in our proposal, we began our assessment of the impact of woody debris on in-stream denitrification by installing a series of red maple woody debris blocks in six stream reaches. These blocks were harvested for a laboratory mesocosm experiment in August of 2008. This experiment will be repeated in November of 2008.

Impacts
Through a random sampling of streams in GIS, we determined that headwater streams (first and second order streams) account for 70% of the length of the drainage network in RI. About 45% of the headwater stream length in RI was found to have comparable surficial geology, soils and land cover to the stream reach used in a previous in-stream nitrogen removal study in RI (Milliman 2007). The average first order stream length and second order stream length in RI was found to be 0.51 and 0.59 km, respectively. In order for in-stream denitrification to be a substantial sink (>20%) for watershed nitrogen in southern New England, denitrification uptake lengths (length of stream a nitrate molecule travels downstream before being denitrified) must be less than 2-5 km, approximately 5-10 fold longer than most headwater stream lengths in RI. A recently completed project (Milliman 2007) on in-stream denitrification within an undisturbed stream reach in this area indicated an uptake length of at least 170 km, far in excess of this target. The extended uptake length resulted from a combination of high stream velocities (low retention time) in headwater streams and a relatively high concentration of nitrate-N (> 0.5 mg/l) compared to many other in-stream studies of denitrification. In the summer of 2007 we found that a bio-reactor of woody debris removed virtually all the groundwater nitrate-N flowing through it. We estimated that groundwater nitrate-N removal from the N bio-reactor was 0.3-0.6 kg N/m shore/yr. While we are still awaiting N2 data on our mesocosms to quantify denitrification, virtually all the nitrate-N was removed from the stream water in all mesocosms over the course of four days in August of 2008. Even our "controls" with a woody debris block lost virtually all the nitrate-N during their incubation indicating that labile carbon from the woody blocks might control nitrogen removal, rather than the extent of biofilms generated on the woody debris during the period of in-stream exposure. In future mesocosms we will also include a stream water only control. Headwater streams with impoundments and woody debris might have longer retention times and more fuel for denitrification than narrow reaches comprised primarily of riffles and runs without impoundments, wetlands, or woody debris. The uncertainties surrounding in-stream N removal warrant further investigation of settings with extended retention times and benthic interactions.

Publications

• Kellogg, D.Q., A.J. Gold, P.M. Groffman, M.H. Stolt, and K. Addy. 2008. Riparian ground-water flow patterns using flownet analysis: Evapotranspiration-induced upwelling and implications for N removal. Journal of the American Water Resources Association 44:1024-1034.
• Addy, K. and A. Gold. 2008. Groundwater Nitrogen Removal Estimates for Nitrogen Barriers: Green Hill Pond, RI. Project Report for Town of South Kingstown, RI.
• Milliman, A.J., K. Addy, A.J. Gold, and D.Q. Kellogg. 2008. In-Stream Denitrification: Pilot Studies on a Headwater Stream. Proceedings of the USDA CSREES National Water Conference, Sparks, NV.
• K. Addy, A. Gold, M. Stolt, and S. Donohue. 2008. HTM Effects on Groundwater Denitrification along RI's Coast. Northeast Regional Cooperative Soil Survey Conference. Narragansett, RI.
• Kellogg, D.Q., T. Watson, K. Addy, A.J. Gold, M.H. Stolt, and P.M. Groffman. 2008. Ground-water Nitrate Removal Capacity of Riparian Zones in Urbanizing and Agricultural Watersheds. Proceedings of the American Water Resources Association's Summer Specialty Conference on Riparian Ecosystems and Buffers: Working at the Water's Edge.

Progress 01/01/07 to 12/31/07

Outputs
OUTPUTS: We used Geographic Information System (GIS) ArcMap 9.1 software, data available through the RIGIS Rhode Island GIS Data, and national hydrography data (NHDplus) to assess the proportion of first and second order streams in Rhode Island relative to their concurrent land use, surficial geology, slope and soil type. This information was compiled to assess the range of settings and conditions found within headwater streams in RI and help us select streams for the measurement of field parameters and in-stream denitrification. We generated a random sample of first and second order streams in RI using a random point generator. Eleven random points were buffered with a 1 km radius, and all first and second order stream lengths that fell within the buffer area were measured. The random sample included 54 first order and 19 second order streams. Headwater streams (first and second order streams) account for 70% of the length of the drainage network in RI. Approximately 45% of the headwater stream length in RI was found to have comparable surficial geology, soils and land cover to the stream reach used in a previous in-stream nitrogen removal study in RI (Milliman 2007). The average first order stream length and second order stream length in RI was found to be 0.51 and 0.59 km, respectively. In the spring of 2007, we began field reconnaissance and characterization of 4-5 headwater streams. Unfortunately, our efforts were stymied by an unusually severe drought. During much of the summer and fall of 2007, stream flow rates were at 20 year lows and virtually all the streams we were monitoring dried up for an extended period. Therefore, we could not obtain meaningful stage discharge relationships, collect stream chemistry data, or conduct in-stream, reach-scale tracer denitrification studies. To advance our understanding of the effects of woody debris on denitrification, we undertook a set of field studies on two sets of woody debris experiments. At each of two sites along Green Hill Pond, South Kingstown, RI, a nitrogen (N) bio-reactor structure, composed of a woody debris-filled trench parallel to the shore (plan view dimensions: 3.6 m long x 2 m wide) extending 1.8 m into nitrate-enriched groundwater, had been constructed in the summer of 2004. We conducted tracer studies to examine the nitrate removal capacity of woody debris. The study required the installation and monitoring of intensive well networks. We ran two to three separate bromide dosing trials per site to capture and characterize the plumes. Approximately 2,500 samples were collected and more than 1,700 samples were analyzed. We found that a bio-reactor of woody debris removed virtually all the groundwater nitrate-N flowing through it. We estimated that groundwater nitrate-N removal from the N bio-reactor was 0.3-0.6 kg N/m shore/yr. PARTICIPANTS: Participants of this project included Arthur J. Gold, principal investigator; Kelly Addy, responsible for field work, analysis performed at URI, and supervision of graduate and undergraduate students; and Holly Meehan and Suzanne Cox, graduate research assistants. The Rhode Island Agricultural Experiment Station (RI AES) provided the tuition waiver and assistantship salary for H. Meehan in the spring of 2007 and for S. Cox in the fall of 2007. H. Meehan and S. Cox were responsible for field work and GIS analyses. The RI AES also funded the undergraduate summer coastal fellowship for Marissa Kelly in 2007. Several undergraduates were involved in the project as general field and laboratory assistants. A. Gold incorporated in-stream dosing of tracers into a laboratory of his undergraduate course "Watershed Hydrology and Management." TARGET AUDIENCES: We incorporated the results of the project into a recently funded CSREES NIWQP project entitled: "Advancing watershed N management at the local level: Incorporating stream reach ecosystem N sinks into a spatial decision support system." This project will target our outputs to State, local and regional decision makers along with NRCS staff. The CSREES funded National NEMO (Non-point Education for Municipal Officials) Program has reviewed our results and plans to incorporate suggestions and recommendations into their programming aimed at local watershed protection. Undergraduate students at URI were a target audience of our efforts. One undergraduate had an experiential learning experience as a URI Undergraduate Coastal Fellow working the summer in our laboratory on this project. Dr. Gold incorporated in-stream dosing of tracers and watershed nitrate sources and sinks into a laboratory of his undergraduate course "Watershed Hydrology and Management." PROJECT MODIFICATIONS: Due to the severe drought in the summer and fall of 2007, headwater streams in RI were at minimal flows or even dry prohibiting us from conducting the in-stream, reach-scale tracer denitrification studies in 4-5 headwater streams as initially proposed in year one. We were still able to obtain valuable data from our GIS analyses, site selection, and work on the effects of buried woody debris on denitrification. We will complete objective 1 as outlined in the proposal characterizing 4-5 headwater streams. In addition, we plan to run the SPARROW model on data collected from these headwater streams to determine the model's prediction of the significance of in-stream denitrification. To overcome methodological constraints imposed by the drought in year one, we will modify our methods to assess the questions outlined in objective 2. Rather than assessing in-stream denitrification with reach-scale tracer studies, we will run a series of acetylene block sediment slurry assays which can be completed under variable flow and climatic conditions. For these assays, we will select two streams from which to collect sediments - one with high (~1 mg N/l) and one with low (<0.3 mg N/l) ambient nitrate concentration. We will collect sediments and run a series of assays in the summer of 2008, fall of 2008, winter of 2008, and spring of 2008. We will select two sampling locations per stream reach system based on sediment texture - one with sands to coarse sands and another with more loamy or silty sediments. In addition to running the sediments alone in the assays, we will also run a series of treatments with woody debris additives. Replicated sets of coarse and fine woody debris will be placed in three types of stream settings in the spring of 2008: 1) shaded flowing stream reach, 2) flowing stream reach under sparse canopy and 3) slow-moving pool under sparse canopy. Each time we sample sediments, we will also harvest a set of the introduced woody debris from each setting. In the laboratory, we will extract the biofilms to create assay additions. N2O will be measured using a gas stripping/trapping system connected to a Hewlett Packard Model 5890A gas chromatograph with a 2 m Poropak Q column and electron capture detector. We will also measure the chlorophyll a content of the woody debris biofilm to quantify microalgal biomass per setting in each stream. Using these methodologies, we will be able to determine the seasonality of in-stream denitrification in RI headwater streams and its relationship to biofilms grown on woody debris in different settings.

Impacts
In order for in-stream denitrification to be a substantial sink (>20%) for watershed nitrogen in southern New England, denitrification uptake lengths (length of stream a nitrate molecule travels downstream before being denitrified) must be less than 2-5 km, approximately 5-10 fold longer than most headwater stream lengths in RI. A recently completed project (Milliman 2007) on in-stream denitrification within an undisturbed stream reach in this area indicated an uptake length of at least 170 km, far in excess of this target. The extended uptake length resulted from a combination of high stream velocities (low retention time) in headwater streams and a relatively high concentration of nitrate-N (> 0.5 mg/l) compared to many other in-stream studies of denitrification. The uncertainties surrounding in-stream N removal warrant further investigation of settings with extended retention times and benthic interactions. Headwater streams containing impoundments or connected to ponds and swamps have longer retention times which will increase the potential for N removal than narrow reaches comprised primarily of riffles and runs without impoundments or wetlands. Methods beyond the constant rate injection method may be warranted to optimize assessments of the river/pond/reservoir settings that are likely to hold promise as N sinks in southern New England.

Publications

• Clough T.J., K. Addy, D.Q. Kellogg, B.L. Nowicki, A.J. Gold, and P.M. Groffman. 2007. Dynamics of nitrous oxide in groundwater at the aquatic-terrestrial interface. Global Change Biology 13:1528-1537.
• Milliman, A.J. 2007. In-stream Denitrification: Pilot studies and site characteristics on a headwater stream. M.S. Thesis, University of Rhode Island, Kingston, Rhode Island.
• Addy, K., A. Gold, M. Stolt, and P. Groffman. 2007. Groundwater Denitrification Below Filled Salt Marshes. Proceedings of the Estuarine Research Federation, page 1. Providence, RI.
• Gold, A., D.Q. Kellogg, K. Addy, P. Groffman, and L. Joubert. 2007. Managing Watershed Nitrogen Export at the Local Level: Accounting for Nitrogen Sinks. Proceedings of the Estuarine Research Federation, page 73. Providence, RI.
• Milliman, A.J., A.J. Gold, K. Addy, D.Q. Kellogg, and B. Nowicki. 2007. In-stream Denitrification: Pilot Studies and Site Characteristics. 2007 National Water Program Conference. USDA-CSREES.
• Kelly, M., K. Addy, and A. Gold. 2007. Groundwater Nitrogen Removal Estimates for a Nitrogen Barrier. Annual University of Rhode Island Undergraduate Coastal Fellows Ceremony, Kingston, RI.

Progress 10/01/06 to 12/31/06

Outputs
In the first three months of this project, we explored numerous first and second order streams and identified at least two streams that we will include in our summer 2007 reach-scale isotope tracer experiment to assess in-stream denitrification potential within 500 meter reaches of 4-5 streams. At both streams, we collected water samples for ambient nitrate, bromide, pH, DO, and temperature analysis. We also estimated average stream discharge at locations within both streams via a slug dosing of rhodamine several times in early October. Discharge at these sites ranged from 3.3 - 12 L/sec. As the fall progressed, discharge substantially increased; we expect the discharge to decrease substantially at these sites in the summer of 2007. At one of the sites, we sampled in-channel and riparian area characteristics over the entire 500 m stream reach. Every 30 m we recorded the stream habitat type, existence and width of emergent bars or channels, densiometer readings, wetted channel area measurements, circumference of woody debris, bank slope incision, diameter of channel substrate, and identification of any vegetation existing in the wetted channel. Riparian vegetation and soil measurements were taken at nine locations extending back 25 m along both sides of the stream.

Impacts
We hypothesize that in-stream denitrification will occur in areas with greater amounts of woody debris and with decreased hydraulic radii of the inundated channel. Outcomes of this research will contribute to better watershed management by improving the knowledge base for the selection of locales for individual and public investment of pollution control and restoration, thereby advancing stream/riparian restoration and management practices.

Publications

• No publications reported this period