Effects of restoration and reflooding on soil denitrification in a leveed Midwestern floodplain.

River floodplains have the potential to remove nitrate from water through denitrification, the anaerobic microbial conversion of nitrate to nitrogen gas. An important factor in this process is the interaction of river water with floodplain soil; however, many rivers have been disconnected from their historic floodplains by levees. To test the effect of reflooding a degraded floodplain on nitrate removal, we studied changes in soil denitrification rates on the Baraboo River floodplain in Wisconsin, USA, as it underwent restoration. Prior to this study, the site had been leveed, drained, and farmed for more than 50 years. In late fall 2002, the field drainage system was removed, and a gate structure was installed to allow controlled flooding of this site with river water. Soil moisture was extremely variable among zones and months and reflected local weather. Soil organic matter was stable over the study period with differences occurring along the elevation gradient. High soil nitrate concentrations occurred in dry, relatively organic-poor soil samples and, conversely, all samples with high moisture soils characterized by low nitrate. We measured denitrification in static cores and potential denitrification in bulk samples amended with carbon and nitrogen, one year before and two years following the manipulation. Denitrification rates showed high temporal and spatial variability. Static core rates of individual sites ranged widely (from 0.00 to 16.7 microg N2O-N x [kg soil](-1) x h(-1), mean +/- SD = 1.10 +/- 3.02), and denitrification enzyme activity (DEA) rates were similar with a slightly higher mean (from 0.00 to 15.0 microg N2O-N x [kg soil](-1) x h(-1), 1.41 +/- 1.98). Denitrification was not well-correlated with soil nitrate, organic matter content, or moisture levels, the three parameters typically thought to control denitrification. Static core denitrification rates were not significantly different across years, and DEA rates decreased slightly the second year after restoration. These results demonstrate that restored agricultural soil has the potential for denitrification, but that floodplain restoration did not immediately improve this potential. Future floodplain restorations should be designed to test alternative methods of increasing denitrification.     

Land use change and nitrogen enrichment of a Rocky Mountain watershed.

Headwater ecosystems may have a limited threshold for retaining and removing nutrients delivered by certain types of land use. Nitrogen enrichment was studied in a Rocky Mountain watershed undergoing rapid expansion of population and residential development. Study sites were located along a 30-km transect from the headwaters of the Blue River to Lake Dillon, a major source of drinking water for Denver, Colorado. Ground water in residential areas with septic systems showed high concentrations of nitrate-N (4.96 +/- 1.22 mg/L, mean +/- SE), and approximately 40% of wells contained nitrate with delta15N values in the range of wastewater. Concentrations of dissolved inorganic nitrogen (DIN) in tributaries with residential development peaked during spring snowmelt as concentrations of DIN declined to below detection limits in undeveloped tributaries. Annual export of dissolved organic nitrogen (DON) was considerably lower in residential streams, suggesting a change in forms of N with development. The seasonal delta15N of algae in residential streams was intermediate between baseline values from undeveloped streams and stream algae grown on wastewater. Between 19% and 23% of the annual N export from developed tributaries was derived from septic systems, as estimated from the delta15N of algae. This range was similar to the amount of N export above background determined independently from mass-balance estimates. From a watershed perspective, total loading of N to the Blue River catchment from septic and municipal wastewater (2 kg x ha(-1) x yr(-1)) is currently less than the amount from background atmospheric sources (3 kg x ha(-1) x yr(-1)). Nonetheless, nitrate-N concentrations exceeded limits for safe drinking water in some groundwater wells (10 mg/L), residential streams showed elevated seasonal patterns of nitrate-N concentration and ratios of DIN to total dissolved phosphorus, and seasonal minimum concentrations of nitrate-N in Lake Dillon have increased exponentially to 80 microg/L over the last decade from an initial value near zero. Results suggest that isotopic ratios in autotrophs can be used to detect and quantify increases in N enrichment associated with land use change. The biotic capacity of headwater ecosystems to assimilate increases in inorganic N from residential development may be insufficient to prevent nitrogen enrichment over considerable distances and multiple aquatic ecosystems downstream.     

Denitrification across landscapes and waterscapes: a synthesis.

Denitrification is a critical process regulating the removal of bioavailable nitrogen (N) from natural and human-altered systems. While it has been extensively studied in terrestrial, freshwater, and marine systems, there has been limited communication among denitrification scientists working in these individual systems. Here, we compare rates of denitrification and controlling factors across a range of ecosystem types. We suggest that terrestrial, freshwater, and marine systems in which denitrification occurs can be organized along a continuum ranging from (1) those in which nitrification and denitrification are tightly coupled in space and time to (2) those in which nitrate production and denitrification are relatively decoupled. In aquatic ecosystems, N inputs influence denitrification rates whereas hydrology and geomorphology influence the proportion of N inputs that are denitrified. Relationships between denitrification and water residence time and N load are remarkably similar across lakes, river reaches, estuaries, and continental shelves. Spatially distributed global models of denitrification suggest that continental shelf sediments account for the largest portion (44%) of total global denitrification, followed by terrestrial soils (22%) and oceanic oxygen minimum zones (OMZs; 14%). Freshwater systems (groundwater, lakes, rivers) account for about 20% and estuaries 1% of total global denitrification. Denitrification of land-based N sources is distributed somewhat differently. Within watersheds, the amount of land-based N denitrified is generally highest in terrestrial soils, with progressively smaller amounts denitrified in groundwater, rivers, lakes and reservoirs, and estuaries. A number of regional exceptions to this general trend of decreasing denitrification in a downstream direction exist, including significant denitrification in continental shelves of N from terrestrial sources. Though terrestrial soils and groundwater are responsible for much denitrification at the watershed scale, per-area denitrification rates in soils and groundwater (kg N x km(-2) x yr(-1)) are, on average, approximately one-tenth the per-area rates of denitrification in lakes, rivers, estuaries, continental shelves, or OMZs. A number of potential approaches to increase denitrification on the landscape, and thus decrease N export to sensitive coastal systems exist. However, these have not generally been widely tested for their effectiveness at scales required to significantly reduce N export at the whole watershed scale.     

Denitrification in nitrate-rich streams: application of N2:Ar and 15N-tracer methods in intact cores.

Rates of benthic denitrification were measured using two techniques, membrane inlet mass spectrometry (MIMS) and isotope ratio mass spectrometry (IRMS), applied to sediment cores from two NO3(-)-rich streams draining agricultural land in the upper Mississippi River Basin. Denitrification was estimated simultaneously from measurements of N2:Ar (MIMS) and 15N[N2] (IRMS) after the addition of low-level 15NO3- tracer (15N:N = 0.03-0.08) in stream water overlying intact sediment cores. Denitrification rates ranged from about 0 to 4400 micromol N x m(-2) x h(-1) in Sugar Creek and from 0 to 1300 micromol N x m(-2) x h(-1) in Iroquois River, the latter of which possesses greater streamflow discharge and a more homogeneous streambed and water column. Within the uncertainties of the two techniques, there is good agreement between the MIMS and IRMS results, which indicates that the production of N2 by the coupled process of nitrification/denitrification was relatively unimportant and surface-water NO3- was the dominant source of NO3- for benthic denitrification in these streams. Variation in stream NO3- concentration (from about 20 micromol/L during low discharge to 1000 micromol/L during high discharge) was a significant control of benthic denitrification rates, judging from the more abundant MIMS data. The interpretation that NO3- concentration directly affects denitrification rate was corroborated by increased rates of denitrification in cores amended with NO3-. Denitrification in Sugar Creek removed < or = 11% per day of the instream NO3- in late spring and removed roughly 15-20% in late summer. The fraction of NO3- removed in Iroquois River was less than that of Sugar Creek. Although benthic denitrification rates were relatively high during periods of high stream flow, when NO3 concentrations were also high, the increase in benthic denitrification could not compensate for the much larger increase in stream NO3- fluxes during high flow. Consequently, fractional NO3- losses were relatively low during high flow.     

Testing the field of dreams hypothesis: functional responses to urbanization and restoration in stream ecosystems

As catchments become increasingly urban, the streams that drain them become increasingly degraded. Urban streams are typically characterized by high-magnitude storm flows, homogeneous habitats, disconnected riparian zones, and elevated nitrogen concentrations. To reverse the degradation of urban water quality, watershed managers and regulators are increasingly turning to stream restoration approaches. By reshaping the channel and reconnecting the surface waters with their riparian zone, practitioners intend to enhance the natural nutrient retention capacity of the restored stream ecosystem. Despite the exponential growth in stream restoration projects and expenditures, there has been no evaluation to date of the efficacy of urban stream restoration projects in enhancing nitrogen retention or in altering the underlying ecosystem metabolism that controls instream nitrogen consumption. In this study, we compared ecosystem metabolism and nitrate uptake kinetics in four stream restoration projects within urban watersheds to ecosystem functions measured in four unrestored urban stream segments and four streams draining minimally impacted forested watersheds in central North Carolina, U.S.A. All 12 sites were surveyed in June through August of 2006 and again in January through March of 2007. We anticipated that urban streams would have enhanced rates of ecosystem metabolism and nitrate uptake relative to forested streams due to the increases in nutrient loads and temperature associated with urbanization, and we predicted that restored streams would have further enhanced rates for these ecosystem functions by virtue of their increased habitat heterogeneity and water residence times. Contrary to our predictions we found that stream metabolism did not differ between stream types in either season and that nitrate uptake kinetics were not different between stream types in the winter. During the summer, restored stream reaches had substantially higher rates of nitrate uptake than unrestored or forested stream reaches; however, we found that variation in stream temperature and canopy cover explained 80% of the variation across streams in nitrate uptake. Because the riparian trees are removed during the first stage of natural channel design projects, the restored streams in this study had significantly less canopy cover and higher summer temperatures than the urban and forested streams with which they were compared.     

Denitrification mitigates N flux through the stream-floodplain complex of a desert city

The Indian Bend Wash (IBW) flood-control project relies on a greenbelt to carry floods through Scottsdale, Arizona, USA. The greenbelt is characterized by a chain of shallow artificial lakes in a larger floodplain of irrigated turf, which has been protected from encroaching urban development. As such, this urban stream-floodplain complex can be divided into three subsystems: artificial lakes, channelized stream segments, and floodplain. We conducted experiments to evaluate which, if any, of these subsystems were important sites of denitrification, and to explore factors controlling denitrification rates. Denitrification enzyme activity (DEA) bioassays were conducted on sediments from eight lake and six stream segments as well as soil samples from eight floodplain transects. Mass-specific potential denitrification rates were significantly higher in lakes than in streams or floodplains. Nutrient limitation bioassays revealed that nitrate (NO3-) limited denitrification in lake sediments, a surprising finding given that NO3(-)-rich groundwater additions frequently raised lake NO3(-) concentration above 1 mg N/L. Experiments on intact lake cores suggested that denitrification was limited by the rate NO3(-) diffused into sediments, rather than its availability in overlying water. Floodplain denitrification was limited by water content, not NO3(-) or C, and irrigation of soils stimulated denitrification. We constructed a N budget for the IBW stream-floodplain complex based on our experimental results. We found that both lakes and floodplains removed large quantities of N, with denitrification removing 261 and 133 kg N ha(-1) yr(-1) from lake sediments and floodplain soils, respectively, indicating that lakes are hotspots for denitrification. Nevertheless, because floodplain area was >4.5 times that of lakes, floodplain soils removed nearly 2.5 times as much N as lake sediments. Given the desert's low annual precipitation, a finding that floodplain soils are active sites of denitrification might seem implausible; however, irrigation is common in urban landscapes, and it elevated annual denitrification in IBW. Based on our results, we conclude that construction of artificial lakes created hotspots while application of irrigation water created hot moments for denitrification in the stream-floodplain complex, demonstrating that management decisions can improve the ability of urban streams to provide critical ecosystem services like N retention.     

Denitrification and the nitrogen budget of a reservoir in an agricultural landscape.

Denitrification is an important process in aquatic sediments, but its role has not been assessed in the N mass balance of upper-Midwestern (USA) reservoirs that receive large agricultural riverine N inputs. We used a 4400-ha reservoir to determine the role of denitrification in the N mass balance and effectiveness in reducing downstream transport of NO(3-)N. Sediment denitrification was (1) measured monthly (March 2002-March 2003) at eight sites in the Lake Shelbyville reservoir in central Illinois using the acetylene inhibition, chloramphenicol technique, (2) scaled to the overall reservoir and compared to N not accounted for in a mass balance, and (3) estimated indirectly using long-term (1981-2003) mass balances of N in the reservoir. Denitrification rates in the reservoir were high during spring and early summer of 2002, when maximum NO(3-)N concentrations were measured (10-14 mg NO(3-)N/L). We estimated that denitrification for the year was between 2580 and 5150 Mg N. Missing N from the mass balance was 3004 Mg N, suggesting that sediment denitrification was the sink. Areal rates of sediment denitrification in the reservoir ranged from 62 to 225 g N x m(-2) x yr(-1), with rates a function of both denitrification intensity (microg N x g dry mass x h(-1)) and the overall mass of sediment present. From 1981 to 2003 the average NO(3-)N inlet flux was 8900 Mg N/yr. About 58% of the total NO(3-)N input was removed, and annual NO(3-)N removed as a percentage of inputs was significantly related to reservoir retention time (average = 0.36 yr for the 23 years, range = 0.21-0.84 yr). By scaling denitrification in Lake Shelbyville to other reservoirs in Illinois, we estimated a sink of 48900 Mg N/yr. When combined with estimated in-stream denitrification, 60900 Mg N/yr was estimated to be removed by sediment denitrification. This reduces riverine export from Illinois to the Gulf of Mexico, where the flux during the 1990s was about 244000 Mg N/yr, and illustrates the importance of reservoir denitrification as an N sink in Midwestern agricultural landscapes.     

Greenhouse gas fluxes in southeastern U.S. coastal plain wetlands under contrasting land uses

Whether through sea level rise or wetland restoration, agricultural soils in coastal areas will be inundated at increasing rates, renewing connections to sensitive surface waters and raising critical questions about environmental trade-offs. Wetland restoration is often implemented in agricultural catchments to improve water quality through nutrient removal. Yet flooding of soils can also increase production of the greenhouse gases nitrous oxide and methane, representing a potential environmental trade-off. Our study aimed to quantify and compare greenhouse gas emissions from unmanaged and restored forested wetlands, as well as actively managed agricultural fields within the North Carolina coastal plain, USA. In sampling conducted once every two months over a two-year comparative study, we found that soil carbon dioxide flux (range: 8000-64 800 kg CO2 x ha(-1) x yr(-1)) comprised 66-100% of total greenhouse gas emissions from all sites and that methane emissions (range: -6.87 to 197 kg CH4 x ha(-1) x yr(-1)) were highest from permanently inundated sites, while nitrous oxide fluxes (range: -1.07 to 139 kg N2O x ha(-1) x yr(-1)) were highest in sites with lower water tables. Contrary to predictions, greenhouse gas fluxes (as CO2 equivalents) from the restored wetland were lower than from either agricultural fields or unmanaged forested wetlands. In these acidic coastal freshwater ecosystems, the conversion of agricultural fields to flooded young forested wetlands did not result in increases in greenhouse gas emissions.     

Assessing stream restoration effectiveness at reducing nitrogen export to downstream waters

The degradation of headwater streams is common in urbanized coastal areas, and the role these streams play in contributing to downstream pollution is a concern among natural resource managers and policy makers. Thus, many urban stream restoration efforts are increasingly focused on reducing the downstream flux of pollutants. In regions that suffer from coastal eutrophication, it is unclear whether stream restoration does in fact reduce nitrogen (N) flux to downstream waters and, if so, by how much and at what cost. In this paper, we evaluate whether stream restoration implemented to improve water quality of urban and suburban streams in the Chesapeake Bay region, USA, is effective at reducing the export of N in stream flow to downstream waters. We assessed the effectiveness of restored streams positioned in the upland vs. lowland regions of Coastal Plain watershed during both average and stormflow conditions. We found that, during periods of low discharge, lowland streams that receive minor N inputs from groundwater or bank seepage reduced in-stream N fluxes. Furthermore, lowland streams with the highest N concentrations and lowest discharge were the most effective. During periods of high flow, only those restoration projects that converted lowland streams to stream-wetland complexes seemed to be effective at reducing N fluxes, presumably because the design promoted the spillover of stream flow onto adjacent floodplains and wetlands. The observed N-removal rates were relatively high for stream ecosystems, and on the order of 5% of the inputs to the watershed. The dominant forms of N entering restored reaches varied during low and high flows, indicating that N uptake and retention were controlled by distinctive processes during different hydrological conditions. Therefore, in order for stream restoration to effectively reduce N fluxes exported to downstream waters, restoration design should include features that enhance the processing and retention of different forms of N, and for a wide range of flow conditions. The use of strategic designs that match the dominant attributes of a stream such as position in the watershed, influence of groundwater, dominant flow conditions, and N concentrations is crucial to assure the success of restoration.     

Temporal changes in C and N stocks of restored prairie: implications for C sequestration strategies

The recovery of ecosystem C and N dynamics after disturbance can be a slow process. Chronosequence approaches offer unique opportunities to use space-for-time substitution to quantify the recovery of ecosystem C and N stocks and estimate the potential of restoration practices for C sequestration. We studied the distribution of C and N stocks in two chronosequences that included long-term cultivated lands, 3- to 26-year-old prairie restorations, and remnant prairie on two related soil series. Results from the two chronosequences did not vary significantly and were combined. Based on modeling predictions, the recovery rates of different ecosystem components varied greatly. Overall, C stocks recovered faster than N stocks, but both C and N stocks recovered more rapidly for aboveground vegetation than for any other ecosystem component. Aboveground C and N reached 95% of remnant levels in only 13 years and 21 years, respectively, after planting to native vegetation. Belowground plant C and N recovered several decades later, while microbial biomass C, soil organic C (SOC), and total soil N recovered on a century timescale. In the cultivated fields, SOC concentrations were depleted within the surface 25 cm, coinciding with the depth of plowing, but cultivation apparently led to redistribution of soil C, increasing SOC stocks deeper in the soil profile. The restoration of prairie vegetation was effective at rebuilding soil organic matter (SOM) in the surface soil. Accrual rates were maintained at 43 g C x m(-2) x yr(-1) and 3 g N x m(-2) x yr(-1) in the surface 0.16 Mg/m2 soil mass during the first 26 years of restoration and were predicted to reach 50% of their storage potential (3500 g C/m2) in the first 100 years. We conclude that restoration of tallgrass prairie vegetation can restore SOM lost through cultivation and has the potential to sequester relatively large amounts of SOC over a sustained period of time. Whether restored prairies can retain the C apparently transferred to the subsoil by cultivation practices remains to be seen.     

Floodplain restoration enhances denitrification and reach-scale nitrogen removal in an agricultural stream

Streams of the agricultural Midwest, USA, export large quantities of nitrogen, which impairs downstream water quality, most notably in the Gulf of Mexico. The two-stage ditch is a novel restoration practice, in which floodplains are constructed alongside channelized ditches. During high flows, water flows across the floodplains, increasing benthic surface area and stream water residence time, as well as the potential for nitrogen removal via denitrification. To determine two-stage ditch nitrogen removal efficacy, we measured denitrification rates in the channel and on the floodplains of a two-stage ditch in north-central Indiana for one year before and two years after restoration. We found that instream rates were similar before and after the restoration, and they were influenced by surface water NO3- concentration and sediment organic matter content. Denitrification rates were lower on the constructed floodplains and were predicted by soil exchangeable NO3- concentration. Using storm flow simulations, we found that two-stage ditch restoration contributed significantly to NO3- removal during storm events, but because of the high NO3- loads at our study site, < 10% of the NO3- load was removed under all storm flow scenarios. The highest percentage of NO3- removal occurred at the lowest loads; therefore, the two-stage ditch's effectiveness at reducing downstream N loading will be maximized when the practice is coupled with efforts to reduce N inputs from adjacent fields.     

Nitrogen dynamics in an Australian semiarid grassland soil

We conducted a four-week laboratory incubation of soil from a Themeda triandra Forsskal grassland to clarify mechanisms of nitrogen (N) cycling processes in relation to carbon (C) and N availability in a hot, semiarid environment. Variation in soil C and N availability was achieved by collecting soil from either under tussocks or the bare soil between tussocks, and by amending soil with Themeda litter. We measured N cycling by monitoring: dissolved organic nitrogen (DON), ammonium (NH4+), and nitrate (NO3-) contents, gross rates of N mineralization and microbial re-mineralization, NH4+ and NO3- immobilization, and autotrophic and heterotrophic nitrification. We monitored C availability by measuring cumulative soil respiration and dissolved organic C (DOC). Litter-amended soil had cumulative respiration that was eightfold greater than non-amended soil (2000 compared with 250 microg C/g soil) and almost twice the DOC content (54 compared with 28 microg C/g soil). However, litter-amended soils had only half as much DON accumulation as non-amended soils (9 compared with 17 microg N/g soil) and lower gross N rates (1-4 compared with 13-26 microg N x [g soil](-1) x d(-1)) and NO3- accumulation (0.5 compared with 22 microg N/g soil). Unamended soil from under tussocks had almost twice the soil respiration as soil from between tussocks (300 compared with 175 microg C/g soil), and greater DOC content (33 compared with 24 microg C/g soil). However, unamended soil from under tussocks had lower gross N rates (3-20 compared with 17-31 microg N x [g soil](-1) d(-1)) and NO3- accumulation (18 compared with 25 microg N/g soil) relative to soil from between tussocks. We conclude that N cycling in this grassland is mediated by both C and N limitations that arise from the patchiness of tussocks and seasonal variability in Themeda litterfall. Heterotrophic nitrification rate explained >50% of total nitrification, but this percentage was not affected by proximity to tussocks or litter amendment. A conceptual model that considers DON as central to N cycling processes provided a useful initial framework to explain results of our study. However, to fully explain N cycling in this semiarid grassland soil, the production of NO3- from organic N sources must be included in this model.     

A Method for Assessing Hydrologic Alteration within Ecosystems

Hydrologic regimes play a major role in determining the biotic composition, structure, and function of aquatic, wetland, and riparian ecosystems. But human land and water uses are substantially altering hydrologic regimes around the world. Improved quantitative evaluations of human-induced hydrologic changes are needed to advance research on the biotic implications of hydrologic alteration and to support ecosystem management and restoration plans. We propose a method for assessing the degree of hydrologic alteration attributable to human influence within an ecosystem. This method, referred to as the "Indicators of Hydrologic Alteration," is based upon an analysis of hydrologic data available either from existing measurement points within an ecosystem (such as at stream gauges or wells) or model-generated data. We use 32 parameters, organized into five groups, to statistically characterize hydrologic variation within each year. These 32 parameters provide information on ecologically significant features of surface and ground water regimes influencing aquatic, wetland, and riparian ecosystems. We then assess the hydrologic perturbations associated with activities such as dam operations, flow diversion, groundwater pumping, or intensive land-use conversion by comparing measures of central tendency and dispersion for each parameter between user-defined "pre-impact" and "post-impact" time frames, generating 64 Indicators of Hydrologic Alteration. This method is intended for use with other ecosystem metrics in inventories of ecosystem integrity, in planning ecosystem management activities, and in setting and measuring progress toward conservation or restoration goals.     

Biogeochemistry of a temperate forest nitrogen gradient

Wide natural gradients of soil nitrogen (N) can be used to examine fundamental relationships between plant-soil-microbial N cycling and hydrologic N loss, and to test N-saturation theory as a general framework for understanding ecosystem N dynamics. We characterized plant production, N uptake and return in litterfall, soil gross and net N mineralization rates, and hydrologic N losses of nine Douglas-fir (Pseudotsuga menziesii) forests across a wide soil N gradient in the Oregon Coast Range (U.S.A.). Surface mineral soil N (0-10 cm) ranged nearly three-fold from 0.29% to 0.78% N, and in contrast to predictions of N-saturation theory, was linearly related to 10-fold variation in net N mineralization, from 8 to 82 kg N.ha(-1) x yr(-1). Net N mineralization was unrelated to soil C:N, soil texture, precipitation, and temperature differences among sites. Net nitrification was negatively related to soil pH, and accounted for <20% of net N mineralization at low-N sites, increasing to 85-100% of net N mineralization at intermediate- and high-N sites. The ratio of net: gross N mineralization and nitrification increased along the gradient, indicating progressive saturation of microbial N demands at high soil N. Aboveground N uptake by plants increased asymptotically with net N mineralization to a peak of approximately 35 kg N.ha(-1) x yr(-1). Aboveground net primary production per unit net N mineralization varied inversely with soil N, suggesting progressive saturation of plant N demands at high soil N. Hydrologic N losses were dominated by dissolved organic N at low-N sites, with increased nitrate loss causing a shift to dominance by nitrate at high-N sites, particularly where net nitrification exceeded plant N demands. With the exception of N mineralization patterns, our results broadly support the application of the N-saturation model developed from studies of anthropogenic N deposition to understand N cycling and saturation of plant and microbial sinks along natural soil N gradients. This convergence of behavior in unpolluted and polluted forest N cycles suggests that where future reductions in deposition to polluted sites do occur, symptoms of N saturation are most likely to persist where soil N content remains elevated.     

Macroinvertebrate production and food web energetics in an industrially contaminated stream.

This study examines secondary production and periphyton-invertebrate food web energetics at two sites in an industrially contaminated, nutrient-enriched stream. Secondary production data and data from the literature were used to calculate potential amounts of mercury transferred from periphyton to chironomid larvae and into terrestrial food webs with emerging adults. The nutritional quality of periphyton was characterized using energy content, chlorophyll a, protein, ash-free dry mass (AFDM), and percentage of organic matter. Chironomid larvae (Orthocladiinae: Cricotopus spp.) comprised 96% of all macroinvertebrates collected from stones at the two sites. Cricotopus production was extremely high: production was 59.5 g AFDM x m(-2) x yr(-1) at the site upstream of a 1-ha settling basin and 32.4 g AFDM x m(-2) x yr(-1) at the site below the basin. Apparent differences in annual secondary production were associated with reduced organic content (i.e., nutritional quality) of the periphyton matrix under different loading of total suspended solids. The periphyton matrix at both sites was contaminated with inorganic (Hg(II)) and methyl (MeHg) mercury. The amount of Hg(II) potentially ingested by Cricotopus was calculated to be 49 mg Hg(II) x m(-1) x yr(-1) at the upstream site and 19 mg Hg(II)x m(-2) x yr(-1) at the downstream site. Mercury ingestion by Cricotopus at the downstream site was calculated to be 2% of the estimated annual deposition of particulate-bound Hg(II) to the stream bed. Emergence of adult Cricotopus was calculated to remove 563 microg Hg(II)x m(-2) x yr(-1) from the stream at the upstream site and 117 microg Hg(II) x m(-2) x yr(-1) at the downstream site, which amounted to 4.1 g Hg(II)/yr for the 2.1-km reach of stream included in this study. The ratio of metal export in emergence production to surface area for the study stream was 10 to 10(3) times higher than ratios calculated for lakes using data from the literature. This study is the first well-documented example of extremely high aquatic insect production in an industrially contaminated, nutrient-enriched stream, and it highlights the application of production measurements to examine the role of aquatic insect production in the trophic transfer of energy and persistent contaminants in aquatic food webs and into terrestrial food webs.     

Raising groundwater differentially affects mineralization and plant species abundance in dune slacks.

The experience with restoring high water levels (i.e., rewetting) within restoration ecology is limited, and information on changes in soil nutrient supply is scarce. A reduction in nutrient supply is needed to restore the desired oligotrophic vegetation. We determined the effects of restoration of high water levels on decomposition and net carbon (C), nitrogen (N), and phosphorus (P) mineralization rates in wet dune slacks and its consequences for the relative abundance of eutrophic vs. oligotrophic species in the vegetation. This was done by analyzing these variables for valleys that experienced a large groundwater rise vs. valleys that had a small groundwater rise but the same current water level. In addition, the influences of underlying factors (waterlogging, vegetation dieback, and soil dynamics prior to groundwater rise) were separated in a transplantation experiment. Short-term effects of large groundwater rise were a massive dieback of vegetation, increased thickness of the fermentation layer, increased microbial decomposition activity, increased C mineralization, and decreased net N mineralization. Net P mineralization was not affected. The relative abundance of oligotrophic vs. eutrophic species was greater at large groundwater rise. Changes in decomposition and mineralization by large groundwater rise were, however, not caused by the vegetation dieback, but due to previous soil conditions. Soils experiencing waterlogged conditions for 3-4 years or more prior to large groundwater rise had lower C and higher net N mineralization rates at waterlogged conditions than soils that had experienced aerobic conditions, presumably due to differences in labile soil C contents. Contrary to expectations induced by previously determined nutrient pulses and measured vegetation dieback, large groundwater rise resulted in lower soil nutrient supply rates and more oligotrophic vegetation. If these trends continue on the longer term, restoration of high water levels may be effective in restoration ecology to establish oligotrophic, wet vegetation in dune slacks.     

Twenty years of stream restoration in Finland: little response by benthic macroinvertebrate communities

The primary focus of many in-stream restoration projects is to enhance habitat diversity for salmonid fishes, yet the lack of properly designed monitoring studies, particularly ones with pre-restoration data, limits any attempts to assess whether restoration has succeeded in improving salmonid habitat. Even less is known about the impacts of fisheries-related restoration on other, non-target biota. We examined how restoration aiming at the enhancement of juvenile brown trout (Salmo trutta L.) affects benthic macroinvertebrates, using two separate data sets: (1) a before-after-control-impact (BACI) design with three years before and three after restoration in differently restored and control reaches of six streams; and (2) a space-time substitution design including channelized, restored, and near-natural streams with an almost 20-year perspective on the recovery of invertebrate communities. In the BACI design, total macroinvertebrate density differed significantly from before to after restoration. Following restoration, densities decreased in all treatments, but less so in the controls than in restored sections. Taxonomic richness also decreased from before to after restoration, but this happened similarly in all treatments. In the long-term comparative study, macroinvertebrate species richness showed no difference between the channel types. Community composition differed significantly between the restored and natural streams, but not between restored and channelized streams. Overall, the in-stream restoration measures used increased stream habitat diversity but did not enhance benthic biodiversity. While many macroinvertebrates may be dispersal limited, our study sites should not have been too distant to reach within almost two decades. A key explanation for the weak responses by macroinvertebrate communities may have been historical. When Fennoscandian streams were channelized for log floating, the loss of habitat heterogeneity was only partial. Therefore, habitat may not have been limiting the macroinvertebrate communities to begin with. Stream restoration to support trout fisheries has strong public acceptance in Finland and will likely continue to increase in the near future. Therefore, more effort should be placed on assessing restoration success from a biodiversity perspective using multiple organism groups in both stream and riparian ecosystems.     

Nitrogen cycling in microbial mats: rates and patterns of denitrification and nitrogen fixation

Spatial and temporal variations in nitrogen fixation and denitrification rates were examined between July 1991 and September 1992 in the intertidal regions of Tomales Bay (California, USA). Microbial mat communities inhabited exposed mudflat and vegetated marsh surface sediments. Mudflat and marsh sediments exhibited comparable rates of nitrogen fixation. Denitrification rates were higher in marsh sediments. Nitrogen fixation rates were lowest during January at both sites, whereas highest rates occurred during summer and fall. Denitrification rates were highest during fall and winter months in marsh sediments, while rates in mudflat sediments were highest during summer and fall. In mudflat sediments, nitrogen fixation and denitrification rates, integrated over 24 h, ranged from 6 to 79 mg N m^-1 d^-1 and 1 to 10 mg N m^-2 d^-1, respectively. Rates of denitrification represented between 6 and 20% of nitrogen fixation rates during the day, but exceeded or were equivalent to nitrogen fixation rates at night. The highest integrated rates of both nitrogen fixation and denitrification occurred during July, whereas, the highest percent loss occurred during spring when denitrification rates amounted to 20% of nitrogen fixation rates during the day. Over an annual cycle, inputs of fixed N to mudflat communities occurred exclusively during daylight. These results underscore the importance of determining integrated diel rates of both nitrogen fixation and denitrification when constructing N budgets. Using this approach, it was shown that microbial denitrification can represent a significant loss of combined nitrogen from mats on daily as well as monthly time scales.     

Effects of urbanization and urban stream restoration on the physical and biological structure of stream ecosystems

Streams, as low-lying points in the landscape, are strongly influenced by the stormwaters, pollutants, and warming that characterize catchment urbanization. River restoration projects are an increasingly popular method for mitigating urban insults. Despite the growing frequency and high expense of urban stream restoration projects, very few projects have been evaluated to determine whether they can successfully enhance habitat structure or support the stream biota characteristic of reference sites. We compared the physical and biological structure of four urban degraded, four urban restored, and four forested streams in the Piedmont region of North Carolina to quantify the ability of reach-scale stream restoration to restore physical and biological structure to urban streams and to examine the assumption that providing habitat is sufficient for biological recovery. To be successful at mitigating urban impacts, the habitat structure and biological communities found in restored streams should be more similar to forested reference sites than to their urban degraded counterparts. For every measured reach- and patch-scale attribute, we found that restored streams were indistinguishable from their degraded urban stream counterparts. Forested streams were shallower, had greater habitat complexity and median sediment size, and contained less-tolerant communities with higher sensitive taxa richness than streams in either urban category. Because heavy machinery is used to regrade and reconfigure restored channels, restored streams had less canopy cover than either forested or urban streams. Channel habitat complexity and watershed impervious surface cover (ISC) were the best predictors of sensitive taxa richness and biotic index at the reach and catchment scale, respectively. Macroinvertebrate communities in restored channels were compositionally similar to the communities in urban degraded channels, and both were dissimilar to communities in forested streams. The macroinvertebrate communities of both restored and urban degraded streams were correlated with environmental variables characteristic of degraded urban systems. Our study suggests that reach-scale restoration is not successfully mitigating for the factors causing physical and biological degradation.     

Habitat-mediated foraging limitations drive survival bottlenecks for juvenile salmon

Realistic population models and effective conservation strategies require a thorough understanding of mechanisms driving stage-specific mortality. Mortality bottlenecks for many species occur in the juvenile stage and are thought to result from limitation on food or foraging habitat during a "critical period" for growth and survival. Without a way to account for maternal effects or to measure integrated consumption rates in the field, it has been virtually impossible to test these relationships directly. Hence uncertainties about mechanisms underlying such bottlenecks remain. In this study we randomize maternal effects across sites and apply a new method for measuring consumption integrated over weeks to months to test the hypothesis that food limitation drives early-season juvenile mortality bottlenecks in Atlantic salmon (Salmo salar). Using natural signatures of geologically derived cesium (133Cs), we estimated consumption rates of >400 fry stocked into six streams. Two to four weeks after stocking, consumption was extremely low across sites (0.005 g x g(-1) x d(-1)) and was predicted to be below maintenance rations (i.e., yielding negative energy balances) for the majority of individuals from five of six sites. However, consumption during this time was positively correlated with growth rates and survival (measured at the end of the growing season). In contrast, consumption rates increased in mid- (0.030 g x g(-1) x d(-1)) and late (0.035 g x g(-1) x d(-1)) seasons, but juvenile survival and consumption were not correlated, and correlations between growth and consumption were weak. These findings are consistent with predictions of a habitat-based bioenergetic model constructed using the actual stream positions of the individual fish in the present study, which indicates that habitat-based models capture important environmental determinants of juvenile growth and survival. Hence, by combining approaches, reducing maternal effects and controlling initial conditions, we offer a general framework for linking foraging with juvenile survival and present the first direct consumption-based evidence for the early season bottleneck hypothesis.