Limitations of stream restoration for nitrogen retention in agricultural headwater streams

Limitations of stream restoration for nitrogen retention in agricultural headwater streams

Ecological Engineering 60 (2013) 224–234 Contents lists available at ScienceDirect Ecological Engineering journal homepage: www.elsevier.com/locate/...

1MB Sizes 1 Downloads 53 Views

Ecological Engineering 60 (2013) 224–234

Contents lists available at ScienceDirect

Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng

Limitations of stream restoration for nitrogen retention in agricultural headwater streams Weigelhofer Gabriele a,b,∗ , Nina Welti a,1 , Thomas Hein b,a a

WasserCluster Lunz, Dr Carl Kupelwieser Promenade 5, A-3293 Lunz/See, Austria University of Natural Resources and Life Sciences, Institute of Hydrobiology and Aquatic Ecosystem Management, Max Emanuelstr. 17, A-1180 Vienna, Austria b

a r t i c l e

i n f o

Article history: Received 21 March 2013 Received in revised form 2 July 2013 Accepted 6 July 2013 Available online 15 August 2013 Keywords: Agricultural streams Restoration Ammonium uptake Potential denitrification Ammonium release

a b s t r a c t High nutrient loading and channelization reduce the nutrient retention capacity of agricultural streams and lead to increases in nutrient downstream transport. The aim of the current study was to study the effects of channel reconfiguration and riparian reforestation on the nitrogen retention capacity of eutrophic agricultural headwater streams. In addition, we investigated the role of stream sediments as a nitrogen sink or source for the stream ecosystem. We compared two restored reaches with two morphologically pristine and four channelized reaches in an agricultural catchment in the north-east of Austria regarding in-stream ammonium uptake, wholereach retention of dissolved inorganic nitrogen, potential denitrification enzyme activity, and sedimentary ammonium release. Restored and pristine reaches exhibited significantly shorter ammonium uptake lengths (330 m) and larger mass transfer coefficients (2.7 × 10−5 m s−1 ) than channelized reaches (2500 m and 1.1 × 10−5 m s−1 , respectively). Increased ammonium uptake was positively correlated with increased transient storage in restored and pristine reaches. Total DIN retention was slightly, though not significantly higher in restored sections (average rates 0.06 g DIN m−2 h−1 ) and showed signs of temporal nitrogen saturation in all reaches. In general, sediments were characterized by small grain sizes (0.04–0.31 mm), high ammonium (60–215 ␮g g−1 DW), and low nitrate concentrations (0.4–5.7 ␮g g−1 DW). Ammonium was released from sediments of all reaches below concentrations of 100 ␮g NH4 + -N L−1 in the overlying water column which shows the high potential of nutrient-rich sediments to act as an internal ammonium source for the stream ecosystem. Potential denitrification was lowest in sediments of restored reaches and significantly increased after nitrate amendment to 3–26 mg N m−2 h−1 . The study reveals that stream sediments, which are loaded with nutrient-rich soil from the agricultural catchment, may limit the effects of stream restoration in agricultural streams. In order to improve the nutrient retention capacity of agricultural streams, reach-scale restoration measures have to be combined with measures in the catchment which reduce nutrient and soil inputs to streams. © 2013 Elsevier B.V. All rights reserved.

1. Introduction

Abbreviations: ˛, transient storage exchange coefficient; A, cross-sectional area; As, transient storage zone area; D, dispersion coefficient; DIN, dissolved inorganic nitrogen; DEA, denitrification enzyme activity; DO, dissolved oxygen; GPP, gross primary production; ER, ecosystem respiration. ∗ Corresponding author at: WasserCluster Lunz, Dr Carl Kupelwieser Promenade 5, A-3293 Lunz/See, Austria. Tel.: +43 7486 200 60 43; fax: +43 7486 200 60 20. E-mail addresses: [email protected], [email protected] (W. Gabriele), [email protected] (N. Welti), [email protected] (T. Hein). 1 Present address: University of Queensland, School of Civil Engineering, Brisbane St. Lucia 4072, Australia. 0925-8574/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ecoleng.2013.07.057

The intensification of agriculture during the last century has impacted the integrity and function of streams considerably (Bernhardt and Palmer, 2011; Dodds and Oakes, 2008; Hancock, 2002; Riseng et al., 2011). An increasing proportion of the landscape has been turned into arable land, thereby altering the vegetation, soil properties, and the hydrologic regime of the catchment and depriving streams of their natural riparian buffer zones (Gordon et al., 2008; Verhoeven et al., 2006). Elevated nutrient concentrations in soils and groundwater, resulting from excessive fertilizer application, decrease the water quality and lead to the eutrophication of agricultural streams (Dodds and Oakes, 2008; Kronvang et al., 2008; Oenema et al., 2005).

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

In addition to changes in the catchment, many agricultural streams have been transformed into straight, trapezoid-shaped channels designed for efficient flood removal. The reduced hydrological and morphological diversity, the deployment of instream-structures (e.g. through the removal of debris dams), and enhanced water velocities decrease the water residence time and reduce the contact area with biogeochemically reactive surfaces, thus diminishing in-stream nutrient retention (Baker et al., 2012; Bernot et al., 2006; Bukaveckas, 2007; Ensign and Doyle, 2005; Hancock, 2002; Roberts et al., 2007). The loss of this vital ecosystem service is especially problematic in catchments where streams are already heavily loaded with nutrients (Beaulieu et al., 2009; Gücker and Pusch, 2006; Kronvang et al., 2008; Lefebvre et al., 2007; Oenema et al., 2005; Ranalli and Macalady, 2010; Riseng et al., 2011). As nutrient concentrations in the water column increase, instream nutrient uptake approaches saturation (Bernot and Dodds, 2005; Bernot et al., 2006; Dodds et al., 2002; Earl et al., 2006). Thus, elevated nutrient inputs from agricultural catchments amplify the negative effects of channelization on the nutrient retention capacity of agricultural streams. Due to the lack of buffer zones, large amounts of nutrients enter the stream in form of eroded soil particles via overland flow (Birgand et al., 2007; Hamilton, 2012; Hancock, 2002). Accumulations of organic-rich soil in the stream channel alter the sediment structure, restrict the hyporheic water exchange, and lead to oxygen depletion in the sediments, thereby changing the nitrogen cycling in the hyporheic zone (Gordon et al., 2008; Hancock, 2002; Lefebvre et al., 2004; Teufl et al., 2012). In the absence of oxygen, mineralization and nitrification becomes restricted, while denitrification and dissimilatory reduction of nitrate to ammonium may be enhanced (Arango et al., 2007; Harrison et al., 2012). Due to high organic matter contents and small grain sizes, sediments of agricultural streams have the potential to remove substantial amounts of stream water nitrate via denitrification, if the surface water is in contact with denitrifying sites in the sediments (Arango et al., 2007; Birgand et al., 2007; Lefebvre et al., 2004). In contrast to nitrate, ammonium is usually accumulated in the anaerobic sediments of agricultural streams (Lefebvre et al., 2004; Teufl et al., 2012) and may be released into the water column through diffusion, bioturbation, or sediment mobilization (Birgand et al., 2007; O’Brien et al., 2012). Sediments of agricultural streams may hence act as sinks or sources for different nitrogen species, depending on concentration gradients at the water–sediment interface, sediment structure, and the duration and frequency of contact between stream water and sediment particles (Birgand et al., 2007). As the last two factors are determined largely by the hydrology and morphology of the stream, channelization and channel reconfiguration may affect the sink-source function of sediments for nitrogen considerably (Hancock, 2002; Harrison et al., 2012; Lefebvre et al., 2004). During the last decades, channel reconfiguration has been increasingly used to reverse stream degradation, restore the morphological and hydrological heterogeneity, re-connect streams with their riparian zones, and initiate the rehabilitation of stream functions and ecosystem services (Baker et al., 2012; Bernhardt and Palmer, 2011; Bukaveckas, 2007; Craig et al., 2008). Particularly in headwater streams, channel reconfiguration is also expected to restore the natural capacity of in-stream nitrogen uptake and storage (Craig et al., 2008). However, in agricultural catchments, land-cover conversion and the on-going intensive agricultural land use may significantly restrict the effectiveness of channel reconfiguration to improve in-stream nitrogen retention (Bernhardt and Palmer, 2011). Legacies in groundwater and soils may hold nutrient background concentrations at an elevated level for decades, thereby protracting an effective reduction of nutrient inputs to stream channels (Gordon et al., 2008; Hamilton, 2012; Oenema

225

et al., 2005; Verhoeven et al., 2006). Considering the continued nutrient loading of surface and subsurface waters, the effects of channel reconfiguration on nitrogen spiralling and in-stream nitrogen uptake is a key question for the management of agricultural streams (Baker et al., 2012; Ranalli and Macalady, 2010). In the present study, we investigated the potential of stream restoration to increase the nitrogen retention capacity of eutrophic agricultural headwater streams. Restoration measures included channel reconfiguration and riparian reforestation. In specific, we aimed to answer the following questions: (1) What is the effect of stream restoration on nitrogen uptake at the reach scale? (2) Does restoration affect the denitrification potential and net ammonium flux in stream sediments? For that purpose, we compared two restored headwater reaches with two morphologically pristine and four channelized reaches within an intensively used agricultural catchment in the north-east of Austria. In stream uptake of nitrogen was measured via shortterm additions of ammonium, which is readily taken up by the biota (Birgand et al., 2007; Gücker and Pusch, 2006). Whole-reach nitrogen retention capacity was estimated via the net export of dissolved inorganic nitrogen during low water level. We also measured denitrification potential and potential ammonium flux from sediments of different reaches exposed to a gradient of nitrate and ammonium concentrations. The study was based on the following hypotheses: (1) Due to increased hydrologic retention, restored and morphologically pristine meandering reaches will exhibit increased in-stream ammonium uptake compared to channelized reaches, showing shorter uptake lengths and higher uptake velocities. (2) As a result of increased nitrogen uptake, restored and morphologically pristine reaches will show a reduced net export of dissolved inorganic nitrogen during low water level compared to channelized reaches. (3) Sediments of restored and morphologically pristine reaches will show lower potential ammonium release rates than those of channelized reaches due to coarser grain sizes and lower sedimentary ammonium contents. Ammonium release will increase with decreasing stream water concentrations. (4) Due to coarser grain sizes, sediments of restored and morphologically pristine reaches will also show a lower denitrification potential than channelized reaches. The denitrification potential will increase with increased stream water nitrate concentrations. 2. Study area The Weinviertel in the north east of Austria is one of the most productive agricultural regions of the country. Intensive agriculture prevails the catchment with cultivation of grain (wheat, barley, rye), wine, and sugar beet (Weigelhofer et al., 2012). The study area belongs to the Molasse Zone and the Vienna Basin, characterized by gravel, sand, clay, and marly siltstone. Average annual precipitation is between 500 and 600 mm. Until the 18th and 19th centuries, the Weinviertel was predominantly marshland with small, meandering, and often intermittent streams. With the intensification of agriculture, large parts of the original landscape were transformed into arable land. Most headwater streams were excavated and straightened, turning them into deeply incised channels designed for a fast removal of surface water

226

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

from fields. In addition, a dense network of drainage pipes and ditches was established around the channel systems. In most cases, reed, grass, and other herbaceous plants form the dominant bank vegetation (Gaitzenauer, 2012; Weigelhofer et al., 2012). Crop fields usually extend to the upper bank margins. After ploughing in September, most fields are uncovered until April. Due to this agricultural practice, the lack of riparian buffer zones, and the overall high erosion potential of the chernozem and pseudogley soils, most streams are characterized by high accumulations of fine, organic-rich soil particles in the channel. Sediments usually become anoxic below 3–10 cm depth, indicating a largely reduced hyporheic water exchange (Teufl et al., 2012). Water quality is further strained by partly insufficient waste water treatment, occasional inputs of untreated sewage, nitrate loaded groundwater or drainage water, and diffuse inputs of nitrate and ammonium from the fields after fertilization in spring (Gaitzenauer, 2012). For the study, we selected eight 200-m-long 1st–2nd order stream reaches (Strahler, 1957) draining 3–30 km2 large catchments dominated by cropland (>85% of catchment area). Four reaches represented deforested, channelized types characterized by homogenous, steep, and V-shaped banks and straightened stream courses (Sinuosity index <1.01). Two of those reaches, Herrnbaumgarten II (Hbg II) and Stuetzenhofen II (Stu II), featured narrow and incised channels (“incised reaches”). The other two reaches, Stronsdorf II (Str II) and Herrnbaumgarten I (Hbg I), were characterized by slightly broadened channels with dense reed stocks growing in the margin areas of the channel bed (“broadened reaches”). The two restored reaches were located at Stronsdorf stream and Stuetzenhofen stream upstream of the channelized sites (“restored reaches”). Stuetzenhofen I (Stu I) had been restored after regulation activities at the beginning of the 20th century, and conserved and continuously fostered since then. Stronsdorf I (Str I) had been restored in 1993. Both reaches were characterized by structurally diverse, meandering channels (Sinuosity index 1.08–1.25) bordered by a small riparian forest (amongst others, Alnus glutinosa (L.) Gaertn., Fraxinus excelsior L., Sorbus aucuparia L., Sambucus nigra L., Crataegus monogyna Jacq.). We also included two morphological pristine sites (Hipples stream, Hi, and Herbertsbrunn stream, Hbb; “pristine reaches”) in our study to set a benchmark for the restored sites regarding nitrogen retention. Pristine reaches had been preserved from stream regulation in the past and, thus, featured remnants of the original riparian vegetation (amongst others, A. glutinosa L., F. excelsior L., Salix sp., S. nigra L.,) as well as structurally diverse meandering stream channels. In contrast to the visually uniform slopes of the channelized reaches, both pristine and restored reaches exhibited a distinct longitudinal step-pool pattern. Besides, the number of woody debris accumulations on the channel bed was higher in pristine and restored reaches (15–60 per 100 m stream length) than in channelized ones (0–3 per 100 m; Teufl et al., 2012).

3. Methods 3.1. In-stream ammonium uptake, DIN retention, transient storage, and stream metabolism To quantify in-stream ammonium uptake and transient storage, we conducted 3–5 short term additions of ammonium chloride (NH4 Cl) and a conservative tracer (sodium chloride, NaCl) at constant rate at each study site between April and September 2009 and 2010 according to the protocol of the Stream Solute Workshop (1990). In short, ammonium and sodium chloride were injected simultaneously into each of the 200 m long study reaches for

2–3 h using a peristaltic pump. Target ammonium concentrations at plateau were 2–4 times the background concentrations. We recorded electrical conductivity at the head and the bottom of each reach at 10 s intervals using a Hach Lange HQ40d conductivity metre. After reaching plateau conditions, we took triplicate water samples every 40 m downstream of the injection point (5 transects, 15 samples in total). Filtered water samples (GFF-filters) were analyzed for NH4 + -N, NO3 − -N, and NO2 − -N concentrations using standard colorimetric methods (APHA, 1998). Uptake parameters were calculated from the longitudinal decline of the ammonium concentrations during plateau conditions. After correction against chloride concentrations, we estimated ammonium uptake lengths, uptake rate coefficients, mass transfer coefficients, and uptake rates via a first order uptake regression curve following the protocol of the Stream Solute Workshop (1990). The ammonium uptake length (m) is an estimate for the average travel distance of an ammonium molecule before it is removed from the water column. The uptake rate coefficient (s−1 ) describes the uptake on a volumetric basis. The mass transfer coefficient (m s−1 ) represents the mean uptake velocity at which a nutrient is removed from the water column. Net dissolved inorganic nitrogen (DIN) retention was estimated as follows:

DIN (mg m−2 h−1 ) =

Cin × Qin − Cout × Qout A

where Cin and Cout are the background concentrations of total DIN at the head and the bottom of each reach (mg L−1 ), Qin and Qout are the discharge (L h−1 ), and A is the stream bottom area (m2 ). Hydrologic retention describes the temporary detainment of water and solutes in stream channel dead zones (e.g. side pools, eddies) or the hyporheic zone, summed up under the term transient storage zone (Runkel and Bencala, 1995). We estimated transient storage parameters (transient storage zone As, dispersion coefficient D, transient storage exchange coefficient ˛) by fitting the one-dimensional solute transport model OTIS-P to the observed conductivity break-through curves from the salt injections (Runkel, 1998). The relative extension of the transient storage zone was calculated as the ratio of As to the cross-sectional area A of the stream (As/A). After each nutrient uptake measurement, we measured bankful and channel width, water depth, and current velocity. Furthermore, we took 5 samples of the sediment surface (diameter 5 cm, depth 1 cm) at each of the 5 transects to determine benthic chlorophylla concentrations (25 samples in total per reach and date). We extracted 3–5 g wet sediments of each surface sample overnight with 90% acetone and determined the supernatant spectrophotometrically at 665 nm after centrifugation (2500 rpm, 20 min) (Jeffrey and Humphrey, 1975). In addition, we estimated whole stream metabolism based on single-station diurnal dissolved oxygen (DO) curves (Bott, 1996). During each nutrient uptake measurement, we measured DO concentrations in 5 min intervals in the middle of each reach over 72 h using a DO metre with a data logger (Driesen + Kern, O2 Log550). To account for the O2 exchange between stream water and atmosphere, we estimated reaeration coefficients based on nightly DO profiles via the oxygen delta method (Mc Bride and Chapra, 2005). Gross primary production (GPP) in mg O2 m−2 h−1 was calculated as the area under the DO change curves corrected for gas exchange. Ecosystem respiration (ER) in mg O2 m−2 h−1 was determined by multiplying the average hourly night-time respiration rate by 24 (Bott, 1996).

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

3.2. Potential ammonium release from sediments At each stream reach, we took 15 samples of the sediment surface in September 2012. Samples were taken by inserting a small PVC tube into the sediment surface (depth 1 cm, diameter 3 cm), placing a plate underneath the tube, lifting the tube, and transferring the sample into a small petri dish. Due to the compactness of the sediment surface, sampling could be accomplished without disrupting the surface layer. During transport, the petri dishes with the water-saturated samples were tightly closed and kept in the dark at 10 ◦ C. In the lab, the petri dishes with the sediment samples were carefully placed in 100 mL plastic jars filled with 60 mL filtered stream water of different NH4 + -N concentrations (15–120 ␮g NH4 + -N L−1 ). Sampling tubes, petri dishes, and jars fitted closely into each other so that disruption of the sediment structure during handling could be avoided. The jars were closed loosely to facilitate oxygen exchange between the air and the headspace in the jar. On average, the DO concentrations in the overlying water column decreased from 7 mg L−1 to 5 mg L−1 during the experiments. We kept the jars gently shaken in the dark at 20 ◦ C for 8–10 h. Despite careful handling, we could not avoid partly resuspension of clay particles from the sediment surface when placing the petri dishes into the jars. This disruption led to a first high release of NH4 + -N to the overlying water column of up to 500 ␮g NH4 + -N L−1 . To exclude this uncontrollable first release, we took the NH4 + -N concentrations in the overlying water column after re-settlement of the particles (3 h) as starting point of the experiment. We analyzed the initial and final NH4 + -N, NO2 − -N, and NO3 − -N concentrations in the overlying water column according to APHA (1998). We used subsamples to determine the initial NH4 + -N, NO2 − -N, and NO3 − -N concentrations in the sediments. Sedimentary NH4 + -N was determined colorimetrically after 1 M KCl extraction, and NO2 − -N and NO3 − -N concentrations after H2 O extraction (SSSA Book Series 5, 2005). 3.3. Potential denitrification enzyme activity Triplicate samples of the sediment surface (5 cm depth) were taken randomly at each study site using a PVC corer (internal diameter 5 cm). Each sample was a homogenized mixture of 5 sediment cores to provide a representative sample of the sampling location. Denitrification enzyme activity (DEA) was measured in the lab via the acetylene (C2 H2 ) block technique according to Smith and Tiedje (1979). Ten grams subsets of sediment samples (fresh weight) with 5 mL sterilized MilliQ water were filled into 100 mL serum flasks which were made anoxic by flushing the flask headspace with N2 . The flask contents were incubated with 10% (v/v) acetylene and kept shaken in the dark at 25 ◦ C. After 4 h incubation, 10 mL gas was extracted from the headspace with a gas-tight syringe and stored in pre-evacuated 10 mL glass vials at 5 ◦ C in the dark until analysis by gas chromatography with 63 Ni electron capture detector (HP 5890II GC). We measured the potential denitrification enzyme activity (DEApot ) by adding 1 mg C g−1 sediment as glucose and 1, 2.5, or 5 mg N g−1 sediment as KNO3 to the sediment slurry before the incubation. Control flasks were left unamended (DEAcont ). In order to determine the proportion of N denitrified as N2 O during the assays, all samples were measured under the same conditions but without acetylene (N2 Ocont and N2 Opot ) and analyzed by gas chromatography with AGILENT 6890N (Santa Clara, USA) connected to an automatic sample-injection system (DANI HSS 86.50, Headspace-sampler, Cologno Monzese, Italy). The efficiency of the denitrification was expressed as ratio of N2 O:DEA. After the experiments, we analyzed the sedimentary nitrate, nitrite, and ammonium concentrations of the samples (SSSA Book Series 5,

227

2005). In addition, we estimated the sedimentary organic matter content via the ash-free dry weight of the samples after combustion at 450 ◦ C for 4 h. 3.4. Statistics Data were tested for normality and homogeneity of variance with Kolmogorov–Smirnov and Levene tests. Depending on the distribution, parametric or non-parametric tests were performed. We analyzed differences among reaches in hydrologic retention, instream nutrient uptake, and stream metabolism via Kruskal–Wallis and Mann–Whitney-U tests. We used MANOVA and ANOVA followed by Tamhane post hoc tests to find effects of stream type, carbon, and nitrate addition on DEA and N2 O production. ANOVA was applied to compare characteristics of sediments used in the experiments. Correlations among hydromorphology, water quality, and in-stream nutrient uptake as well as among potential denitrification, sedimentary ammonium exchange, and sediment characteristics were identified by Spearman’s or Pearson’s correlation. All statistical analyses were performed using SPSS 15.0 for Windows (SPSS Inc., Chicago, USA, 2006). 4. Results 4.1. Hydromorphology, transient storage, and stream metabolism Discharge during the nutrient uptake measurements did not differ significantly among stream reaches (Kruskal–Wallis, p > 0.05, n = 31; Table 1). Regarding hydromorphology, we observed significant differences among reach types only for bankful width and mean current velocity (Kruskal–Wallis, p < 0.05, n = 31). Bankful width was largest in broadened reaches (2.0 ± 0.2 m) and smallest in incised reaches (0.8 ± 0.04 m). Mean current velocity was significantly higher in incised reaches (0.14 m s−1 ) than in the others (0.08–0.09 m s−1 ). Due to the high variability, transient storage parameters did not differ significantly among reach types. Mean As/A was largest in restored reaches and smallest in broadened reaches (Table 1). Neither ammonium nor nitrate or nitrite background concentrations showed significant differences among reach types during the nutrient uptake measurements (Kruskal–Wallis, p > 0.05, n = 31; Table 1). The range of dissolved oxygen (DO) concentrations was generally lower in reaches with riparian forests. Pristine and restored reaches yielded minimum DO concentrations of 8.9 and 9.2 mg O2 L−1 during night and maximum DO concentrations of 9.5 and 9.7 mg O2 L−1 during the day, respectively. Average DO concentrations at incised and broadened reaches ranged from 5.3 mg O2 L−1 at night to 11.5 mg O2 L−1 during day. Most reaches were heterotrophic during the nutrient uptake measurements, featuring GPP/ER ratios of <1. GPP rates as well as chlorophyll-a concentrations were significantly lower in pristine reaches than in the others (U-test, p < 0.05, n = 31; Table 1). The highest GPP rates were measured in restored reaches. GPP increased significantly with chlorophyll-a content (Spearman, p < 0.01, n = 31). Average ecosystem respiration was lowest in restored reaches and highest in broadened reaches which differed significantly from each other (U-test, p < 0.05, n = 31; Table 1). We found no correlations between nutrient background concentrations and stream metabolism. 4.2. In-stream ammonium uptake and DIN retention Ammonium uptake parameters pooled the four reach types in two groups. Restored and pristine reaches exhibited similar ammonium uptake characteristics but differed significantly from incised

228

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

Fig. 1. Ammonium uptake lengths, mass transfer coefficients, uptake rate coefficients, and uptake rates for morphological different reach types in the Weinviertel. Shown are median, 10, 25, 75, and 90% quartiles (n = 7–9 per reach type). Significant differences are indicated by capital letters.

and broadened reaches in NH4 uptake lengths (Kruskal–Wallis test, p < 0.001, n = 31) and mass transfer coefficients (p = 0.04). Mean NH4 uptake lengths amounted to 330 m for pristine and restored reaches and 2500 m for incised and broadened reaches (Fig. 1). Mean mass transfer coefficients were about 2.7 × 10−5 m s−1 in pristine and restored reaches and 1.1 × 10−5 m s−1 in incised and broadened reaches (Fig. 1). Neither uptake rate coefficients nor uptake rates differed significantly among reach types. Ammonium uptake rates were highest in pristine reaches and lowest in restored reaches (Fig. 1). Ammonium uptake lengths correlated negatively with As/A (p = 0.006, Spearman, n = 31) and positively with mean water velocity (p = 0.02). Ammonium uptake rates increased significantly with ammonium background concentrations (p = 0.02). We also observed a non-significant increase in uptake lengths and decrease in uptake velocities with increasing ammonium concentrations. Uptake parameters did not correlate with stream metabolism (gross primary production, ecosystem respiration) or benthic chlorophyll a concentrations. Net DIN retention was not significantly different among reach types (Table 1). Restored reaches showed the highest net DIN retention, yielding maximum rates of 0.32 g DIN m−2 h−1 . Maximum DIN retention rates of other reaches ranged between 0.1 g DIN m−2 h−1 for broadened and 0.12 g DIN m−2 h−1 for pristine and incised reaches. At all sites, we could observe an increase in DIN concentrations along the study reaches at one or two sampling dates, independent of season or discharge.

4.3. Potential ammonium release from sediments Grain sizes of the sediment surface were significantly smaller in broadened reaches (ANOVA, p < 0.05, n = 48) than in the others which did not differ (Table 2). Nitrate and nitrite concentrations of the incubated sediments were similar, but ammonium concentrations differed among reaches, albeit not significantly (ANOVA, p > 0.05; Table 2). Sedimentary NH4 was highest in sediments of broadened reaches and lowest in those of incised reaches. Sediments of different reach types showed similar results when exposed to different ammonium concentrations, irrespective of the sediment structure and nitrogen content. At initial concentrations of less than 100 ␮g NH4 + -N L−1 in the overlying water column, we observed an average release between 1 and 2 mg NH4 + -N m−2 h−1 from all sediments (Fig. 2). Above stream water concentrations of 200 ␮g NH4 + -N L−1 , all sediments showed a slight uptake of ammonium of up to 1.4 mg NH4 + -N m−2 h−1 . We observed a general nitrate uptake during the experiments in all sediments but those of incised reaches, independent of the initial nitrate concentrations in the overlying water column. Initial nitrate concentrations were 0.1–2 mg NO3 − -N L−1 . Mean uptake rates ranged between 0.4 mg NO3 − -N m−2 h−1 (restored reaches) and 0.7 mg NO3 − -N m−2 h−1 (broadened reaches). At incised reaches, we observed a mean nitrate release of 0.5 mg NO3 − -N m−2 h−1 . 4.4. Potential denitrification enzyme activity Sediments used for the denitrification experiments differed significantly among reach types (ANOVA, p < 0.05, n = 27). The smallest

Table 1 Hydrology, water chemistry, metabolism, and retention of dissolved inorganic nitrogen (DIN) during the nutrient addition experiments (mean ± standard deviation; n = 31). Q, discharge; As/A, relative extension of transient storage zone; D, dispersion coefficient; ˛, transient storage exchange coefficient; Chl-a, chlorophyll a; GPP, gross primary production; ER, ecosystem respiration. Type

Q (L s−1 )

As/A

Hi Hbb Str I Str II Hbg I Hbg II Stu I Stu II

Pristine 9.6 ± 9.2 Pristine 0.9 ± 0.5 Restored 6.5 ± 5.9 Broadened 11.7 ± 7.6 Broadened 5.3 ± 0.5 Incised 3.7 ± 2.4 Restored 6.2 ± 4.5 Incised 9.3 ± 4.1

0.24 0.21 0.54 0.10 0.08 0.18 1.51 0.11

a

D (m2 s−1 ) ± ± ± ± ± ± ± ±

0.2 0.2 0.5 0.1 0.1 0.2 1.3 0.1

0.11 0.06 0.07 0.06 0.09 0.07 0.16 0.14

± ± ± ± ± ± ± ±

0.10 0.03 0.07 0.03 0.06 0.04 0.02 0.06

˛ (10−4 s−1 ) 1.63 0.21 3.16 6.35 2.96 1.92 2.08 5.55

± ± ± ± ± ± ± ±

2.4 0.2 3.9 4.0 3.4 2.2 1.8 7.8

NO2 − -N (␮g L−1 ) 20.4 61.8 120.1 71.8 74.3 72.5 39.07 50.46

± ± ± ± ± ± ± ±

8.1 59.8 177.2 60.5 51.3 41.2 23.8 20.9

NO3 − -N (␮g L−1 ) 5869.6 1788.7 4136.8 7606.4 3200.3 3307.4 4905.6 4764.2

± ± ± ± ± ± ± ±

1073 221 996 6289 743 590 2504 1464

NH4 + -N (␮g L−1 ) 52.5 107.2 83.2 157.0 85.6 65.4 42.0 48.4

± ± ± ± ± ± ± ±

28.5 85.3 102.5 104.0 42.8 36.5 31.5 19.4

Chl-a (g m−2 ) 0.02 0.02 0.21 0.23 0.10 0.29 0.18 0.02

± ± ± ± ± ± ± ±

GPP (g O2 m−2 d−1 )

0.01 0.01 0.34 0.32 0.02 0.41 0.23 0.00

0.1 0.1 1.4 0.6 1.1 0.5 1.1 1.1

± ± ± ± ± ± ± ±

0.0 0.0 0.6 0.6 0.9 0.4 0.6 0.3

ER (g O2 m−2 d−1 ) 2.1 1.6 1.8 3.4 3.1 2.1 1.8 4.8

± ± ± ± ± ± ± ±

DIN (g m−2 h−1 ) −0.02 −0.01 −0.07 0.01 −0.02 0.003 −0.04 −0.03

0.7 0.9 0.3 1.7 0.2 1.4 0.1 5.3

± ± ± ± ± ± ± ±

0.07 0.00 0.02 0.1 0.03 0.01 0.04 0.03

Hi, Hipples stream; Hbb, Herbertsbrunn stream; St, Stronsdorf stream; Hbg, Herrnbaumgarten stream; Stu, Stuetzenhofen stream.

Table 2 Nutrient concentrations and grain sizes of sediments used for potential denitrification and potential ammonium release measurements for each reach type (mean ± standard deviation; n = 27 and 48, respectively). Q50 , mean grain size; OM, organic matter. Potential denitrification −

−1

NO2 -N (␮g g Pristine Restored Incised Broadened

5.8 0.6 2.6 2.5

± ± ± ±

0.5 0.2 0.6 1.1

DW)

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

Sitesa

Potential ammonium release −

−1

NO3 -N (␮g g 5.7 0.4 2.3 3.0

± ± ± ±

3.2 0.2 0.9 2.4

DW)

−1

+

NH4 -N (␮g g 102.8 56.4 70.2 156.8

± ± ± ±

48.4 22.0 25.4 61.2

DW)

Q50 (mm) 0.21 0.31 0.24 0.07

± ± ± ±

0.0 0.1 0.1 0.1

NO2 − -N (␮g g−1 DW)

OM (%) 25.5 4.3 6.8 11.1

± ± ± ±

2.3 0.1 0.5 3.9

1.6 0.9 1.3 1.1

± ± ± ±

1.1 0.4 1.0 0.5

NO3 − -N (␮g g−1 DW) 2.4 3.5 1.2 3.8

± ± ± ±

1.9 4.7 1.3 6.4

NH4 + -N (␮g g−1 DW) 93.3 128.4 74.8 215.0

± ± ± ±

125.1 109.5 85.8 248.0

Q50 (mm) 0.19 0.27 0.25 0.04

± ± ± ±

0.1 0.1 0.3 0.0

229

230

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

5. Discussion Streams draining intensively used cropland have lost vital ecosystem functions and services (Gordon et al., 2008). Key stressors are channelization and nutrient loading which both reduce the nutrient retention capacity and increase nutrient downstream transport (Baker et al., 2012; Bernot et al., 2006; Earl et al., 2006; Gücker and Pusch, 2006; Verhoeven et al., 2006). The question for the management of agricultural streams is whether the reduction of just one stressor, namely channelization, can improve the situation despite the persistence of the second stressor. The main aim of the current study was to investigate the effects of channel reconfiguration and riparian reforestation on the nitrogen retention capacity of eutrophic agricultural headwater streams. 5.1. Effects of stream restoration on the uptake of ammonium Fig. 2. Changes in NH4 + -N concentrations in the overlying water column (in mg NH4 + -N m−2 h−1 ) when sediments of different reaches are exposed to different initial NH4 + -N concentrations in the overlying water column.

grain sizes could be found at broadened reaches which differed significantly from the others (Tamhane, p < 0.05, n = 27; Table 2). Pristine reaches featured the highest organic matter and nitrite concentrations in the sediments (p < 0.05). Sediments of restored reaches showed significantly lower organic matter and nitrate concentrations than the others (p < 0.05). Nitrite and ammonium concentrations were lowest, while grain sizes were largest at these sites. Reach type, carbon addition, and nitrate addition had highly significant effects on the denitrification enzyme activity (MANOVA, p < 0.01, n = 152). DEAcont rates in control flasks were about 0.2–1.7 mg N m−2 h−1 (5.9–62.2 ng g−1 DW h−1 ; Fig. 3). The amendment of carbon led to a significant increase in potential denitrification to 0.5–7.7 mg N m−2 h−1 (17.9–290.1 ng g−1 DW h−1 ; ANOVA, p < 0.01, n = 36). The effects of nitrate were much higher than those of carbon. Nitrate amendment significantly increased DEApot rates at all reach types to 3–26 mg N m−2 h−1 (160–1115 ng g−1 DW h−1 ; ANOVA, p < 0.001, n = 152). Increases were highest in sediments of broadened reaches and lowest in sediments of restored reaches (Fig. 3). While the addition of 1 mg NO3 − -N already led to significant increases in DEApot rates (Tamhane, p < 0.05), higher amounts of nitrate did not affect DEApot rates further. Carbon addition to already nitrate amended treatments did not increase DEApot rates. DEAcont rates differed significantly only between pristine reaches and others (Tamhane, p < 0.05, n = 18; Fig. 3). Nitrate and carbon enriched sediments revealed highly significant differences between broadened reaches and the others (p < 0.01) as well as between restored reaches and pristine or incised reaches (Tamhane, p < 0.01, n = 62). The efficiency of potential denitrification was lowest in sediments of broadened reaches without C or N amendment giving an N2 Ocont :DEAcont ratio of 0.64 (0.62, 0.42, and 0.15 for incised, restored, and pristine reaches, respectively). The N2 Opot :DEApot ratio was significantly decreased at all sites to 0.02–0.09 by the addition of nitrate (Tamhane test, p < 0.01, n = 152; Fig. 3). Carbon addition had no significant effects on N2 Opot :DEApot ratios. Maximum DEApot rates (amendment with 1 mg C and >1 mg NO3 − -N) decreased significantly with increasing mean grain size of the respective sediment sample (Pearson, p < 0.01, n = 63) and increased, albeit non-significantly, with the organic matter content (Pearson, p = 0.055, n = 63). DEAcont rates also decreased with increasing mean grain size (p < 0.01).

As expected, we found a significantly higher uptake of ammonium in both morphologically pristine and restored reaches than in channelized sections, resulting in significantly shorter uptake lengths and higher mass transfer coefficients. Nutrient uptake is partly a function of the hydrologic retention of the reach which determines the contact time of solutes with biogeochemically reactive surfaces (Baker et al., 2012; Ranalli and Macalady, 2010). The discharge-dependent nutrient uptake length reflects the hydrologic component of in-stream nutrient uptake best (Stream Solute Workshop, 1990). In our study, uptake lengths correlated positively with the extension of the transient storage zone which was about 4–5 times greater in pristine and restored meandering reaches than in channelized sections. Small grain sizes and extended anoxic zones below the sediment surface, as observed in our study streams (Teufl et al., 2012), are indicators for a restricted hyporheic water exchange (Hancock, 2002). Therefore, we assume that the increased transient storage in the meandering reaches was mainly due to surface hydrologic retention caused by frequent debris dams and pools in the channel (Weigelhofer et al., 2012). Beside longer water residence times, enhanced channel heterogeneity leads to local increases in the hydraulic resistance which, in turn, may increase the diffusion of stream solutes to biogeochemically reactive zones and enlarge the effective surface area for nutrient uptake (Dodds et al., 2002). Increased nutrient uptake resulting from increased surface transient storage was also found by Baker et al. (2012), Bukaveckas (2007), Ensign and Doyle (2005), and Roberts et al. (2007). Beside the physical aspect of retention, nutrient uptake is determined by the efficiency of the benthic community which is largely a function of nutrient supply vs. demand (Birgand et al., 2007; Earl et al., 2006; Gücker and Pusch, 2006; Ranalli and Macalady, 2010). The mass transfer coefficient is standardized for stream morphology and hydraulics and thus describes nutrient uptake efficiency (Stream Solute Workshop, 1990). Both restored and pristine reaches showed significantly larger numbers of wood accumulations on the channel bed than deforested channelized reaches (Teufl et al., 2012). As wood is characterized by higher carbon-to-nitrogen ratios than aquatic plants or microorganisms (Dodds et al., 2004), its decomposition requires additional nitrogen sources, thereby increasing the nutrient demand of microbial decomposers (Craig et al., 2008; Bernot and Dodds, 2005; Roberts et al., 2007). Therefore, we expected an increased ammonium uptake efficiency in pristine and restored reaches compared to channelized sites. Although our results met our expectations, differences in mass transfer coefficients among reaches were much smaller than for uptake lengths. We thus

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

231

Fig. 3. Potential denitrification rates (mg N m−2 h−1 ) for pristine, restored, incised, and broadened reaches amended with carbon (+C), nitrate (+N), and both carbon and nitrate (+C + N) versus non-amended controls (Ctl). Shown are median, 10, 25, 75, and 90% quartiles, and outliers (n = 12–21 per reach type).

assume that the increased ammonium uptake in restored and pristine reaches was rather the result of increased hydrologic retention than of increased nutrient demand. Bukaveckas (2007) also observed positive effects of channel morphology on nutrient transport, but not on uptake efficiency in a naturalized creek. Although ammonium uptake represents an important component of in-stream nitrogen cycling, it may be an inadequate indicator for the retention of chronic nitrogen loads (Bernot and Dodds, 2005; Earl et al., 2006). Besides, unlabelled nutrient additions tend to underestimate nutrient uptake at high background concentrations due to saturation effects (Mulholland et al., 2002; Riis et al., 2012). In our study, net retention of dissolved inorganic nitrogen did not differ significantly among reaches. Differences in net DIN export were mainly due to reductions in stream water nitrate, yielding mean DIN retention rates about 4–10 times higher than mean ammonium uptake rates. In the presence of ammonium, nitrate assimilation is reported to be low (Birgand et al., 2007; Gücker and Pusch, 2006). Therefore, we assume that reach-scale denitrification may constitute the key mechanism for nitrogen removal in our agricultural streams. Occasional net increases of DIN along the study reaches reveal that temporary nitrogen saturation may occur at all sites, probably due to short-term increases in nutrient inputs (e.g. after fertilization or ploughing) or seasonally dependent low biological activities. Nitrogen saturation or near-saturation is a phenomenon frequently found in agricultural streams (Bernot et al., 2006; Dodds et al., 2002; Gücker and Pusch, 2006).

5.2. Effects of stream restoration on the sink-source function of the sediments So far, sedimentary ammonium release has mainly been reported from constructed or natural wetlands (Erler et al., 2010; Kadlec et al., 2005; O’Brien et al., 2012). Our experiments demonstrate that organic-rich sediments of agricultural streams may also show a high potential for ammonium release. We assume that ammonium release in our experiments was mainly due to increased diffusion from ammonium-rich interstitial water to ammonium-poor surface water (Birgand et al., 2007; O’Brien et al., 2012). Decreases in DO concentrations during the incubations indicate that small amounts of ammonium may have also been derived from microbial mineralization of organic matter in the sediments. In addition, anoxic conditions have probably occurred within the tightly packed sediment samples during the incubations. Thus, dissimilatory nitrate reduction (DNRA), which can be a relevant source of ammonium in nutrient-rich anoxic sediments (Crenshaw et al., 2010; Lefebvre et al., 2004), could have caused part of the ammonium increase in our experiments. Due to their intense linkage with the agricultural catchment and their spatial restriction, restored reaches showed similar sediment characteristics to the others. This may be one reason why we did not observe any significant differences in ammonium release rates among reaches. Besides, ammonium release seems to depend more on the concentration gradient between interstitial and stream water than on the actual ammonium content of the sediments.

232

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

Ammonium was released from sediments only below a threshold of 100 ␮g NH4 + -N L−1 in the overlying water column. Such concentrations are common in the Weinviertel in summer (Table 1). During those times, sediments may act as a relevant internal ammonium source for the stream ecosystem which compensates possible deficits in the water column and supplies the benthic community with sufficient amounts for their metabolism. Further research is needed about in situ ammonium release in order to understand its significance for the whole-reach nitrogen retention in eutrophic agricultural streams. Potential denitrification rates highlighted the significance of an increased contact of surface water with the sediment compartment for the nutrient cycling in our study streams. While denitrification was low in control flasks, nitrate enrichment enhanced denitrification rates by a factor of 10–100. Together with the low sedimentary nitrate concentrations, this shows the high nitrate demand of the microbial community in these predominantly anoxic sediments. Potential denitrification also significantly increased after glucose addition. Several studies indicate that denitrification is influenced not only by the quantity, but also by the quality of the organic matter (Dodla et al., 2008; Morley and Baggs, 2010; Welti et al., 2012). Considering the predominantly terrestrial origin of the organic matter in our sediments, it is thus not surprising that the microbial community reacted to the addition of a readily available carbon source like glucose. Potential denitrification rates measured with the acetylene block technique cover a wide range in literature, from a few nanograms to more than 1000 ng N g−1 DW h−1 (Arango et al., 2007; Harrison et al., 2012; Klocker et al., 2009; Lefebvre et al., 2004; Martin et al., 2001). Our results from both amended and control flasks fall within this range and are also similar to in situ rates measured by Mulholland et al. (2008) for agricultural streams. Although denitrification may constitute a potential sink for nitrate, its role for whole-reach nitrogen retention is controversial. While several studies stress the significance of in-stream denitrification as substantial nitrate sink (Harrison et al., 2012; Klocker et al., 2009; Lefebvre et al., 2004), some also question its potential to counterbalance excess nitrate inputs in agricultural catchments (Böhlke et al., 2004; Klocker et al., 2009; Martin et al., 2001; Mulholland et al., 2008; Ranalli and Macalady, 2010). In our study, mean potential denitrification amounted to less than 1–5% of whole-reach DIN retention which, in turn, comprised only 2–10% of total DIN input from upstream reaches. The sharp boundary between the oxic, nitrate-saturated water column and the anoxic, nitrate-depleted sediments in our study streams may restrict nitrate removal via denitrification to local and temporal hot-spots of reduced surface flow (e.g. within debris dams) or enhanced hyporheic water exchange (e.g. in step-pool-sequences or after floods; McClain et al., 2003). The effects of channel reconfiguration on in-stream denitrification may be partly antagonistic. In general, stream restoration is expected to stimulate denitrification by creating favourable conditions for denitrifying bacteria through the supply of additional carbon sources and the increase in hydrologic residence time and hyporheic water exchange (Arango et al., 2007; Harrison et al., 2012; Klocker et al., 2009). In our DEA assays, stream restoration affected potential denitrification through an increase in grain sizes, resulting in lowest DEA rates in restored reaches. Our results correspond to the findings of Harrison et al. (2012) who measured lower denitrification rates in coarse-grained riffles than in finegrained pools. If stream restoration leads to an increase in grain sizes, we might, therefore, assume a decrease in sediment denitrification rates. However, increased grain sizes enable an enhanced water exchange between nitrate-rich surface water and denitrifying sites in the sediments, thereby increasing the potential space

Fig. 4. Scheme of the hypothetical three-dimensional linkage of agricultural streams with adjacent ecosystems in their pristine (A) and degraded state (B) as well as after stream restoration (reach-scale; C), and application of Best Management Practices (BMPs) in the catchment (catchment-scale; D). x, longitudinal connectivity; y, lateral connectivity with the riparian zone; z, vertical connectivity with the hyporheic zone.

and time for denitrification and, in turn, promoting reach-scale denitrification. 5.3. The potential of stream restoration to improve nitrogen retention Agricultural streams are exposed to multiple human impacts on the catchment- and reach-scale which substantially change their three-dimensional linkage with adjacent ecosystems (Fig. 4; Gordon et al., 2008; Hancock, 2002). Land cover conversion and stream regulation turn headwater streams from their originally retentive, laterally interactive pristine state into transport systems decoupled from their riparian zone (Fig. 4A and B). Channel reconfiguration in conjunction with riparian reforestation may partly reverse or, at least, impede this development and induce a partial recovery of some of the former ecosystem functions (Fig. 4C). As shown by the current study, the increased longitudinal channel heterogeneity may decrease water and solute transport for the benefit of improved hydrologic retention. In addition, an intensified contact between stream water, riparian zone, and biogeochemically reactive surfaces in the channel may stimulate the riparian and in-stream nitrogen cycling, thus binding readily available dissolved nitrogen in more complex and stable compounds (e.g. dead or living biomass) or even removing it from the stream ecosystem (Birgand et al., 2007; Earl et al., 2006; Verhoeven et al., 2006). However, restoration measures at the reach-scale cannot compensate for deficits in the catchment (Bernhardt and Palmer, 2011), especially if they are spatially restricted like in the Weinviertel region (Teufl et al., 2012). If conditions in the catchment remain unchanged, streams will largely maintain their human-assigned transport function, thus limiting the potential of stream restoration to improve in-stream nitrogen retention. Our study reveals that one reason for this persistence lies in soil-loaded, nutrient-rich sediments. Stream sediments constitute a compartment essential for in-stream nutrient uptake even in streams where diffusion and hyporheic exchange are naturally limited due to small grain sizes

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

(Fig. 4A). Excess soil inputs from the catchment clog sediments and inhibit the exchange between interstices and stream water, thus cutting off large parts of the microbial community from the nutrient supply and depriving agricultural streams of an important part of their natural retention zone (Hancock, 2002). In addition, nutrient-loaded stream sediments may act as an internal eutrophication source for the stream ecosystem and lead to an increasing saturation of the aquatic system. If ecosystem managers aim for reducing nutrient loading of streams in agricultural catchments and restore part of their original nutrient retention function, a combination of measures on both the reach-scale and the catchment-scale is necessary (Bernhardt and Palmer, 2011; Craig et al., 2008; Verhoeven et al., 2006). Best management practices (BMPs) in the catchment, like reduction of fertilizer applications, conservation tillage, and the use of lowerosive plants, may effectively minimize erosion in the catchment and reduce sediment and nutrient delivery to stream channels (Kronvang et al., 2008; Schoumans et al., 2011). The rehabilitation of riparian zones and floodplain wetlands over long stretches of the stream restores important buffer zones for the stream ecosystem which retain nutrients before they enter the channel (Verhoeven et al., 2006). At last, channel reconfiguration improves the hydrologic retention capacity of the stream, increases the contact between nutrients and metabolic reactive surfaces, and thus facilitates in-stream nutrient uptake (Baker et al., 2012; Craig et al., 2008; Dodds et al., 2002).

Acknowledgments This study was funded by the European Regional Development Fund (European-Territorial-Cooperation Austria-Czech Republic 2007–2013), the Government of Lower Austria, and the Austrian Ministry of Environment. We thank two anonymous reviewers for their valuable comments on an earlier version of the article.

References APHA, 1998. Standard Methods for the Examination of Water and Wastewater, 20th ed. American Public Health Association, Washington, DC. Arango, C.P., Tank, J.L., Schaller, J.L., Royer, T.V., Bernot, M.J., David, M.B., 2007. Benthic organic carbon influences denitrification in streams with high nitrate concentration. Freshw. Biol. 52, 1210–1222. Baker, D.W., Bledsoe, B.P., Mueller Price, J., 2012. Stream nitrate uptake and transient storage over a gradient of geomorphic complexity, north-central Colorado, USA. Hydrol. Process. 26, 3241–3252. Beaulieu, J.J., Arango, C.P., Tank, J.L., 2009. The effects of season and agriculture on nitrous oxide production in headwater streams. J. Environ. Qual. 38, 637–646. Bernhardt, E.S., Palmer, M.A., 2011. River restoration: the fuzzy logic of repairing reaches to reverse catchment scale degradation. Ecol. Appl. 21, 1926–1931. Bernot, M.J., Dodds, W.K., 2005. Nitrogen retention, removal, and saturation in lotic ecosystems. Ecosystems 8, 442–453. Bernot, M.J., Tank, J.L., Royer, T.V., David, M.B., 2006. Nutrient uptake in streams draining agricultural catchments of the midwestern United States. Freshw. Biol. 51, 499–509. Birgand, F., Skaggs, R.W., Chescheir, G.M., Gilliam, J.W., 2007. Nitrogen removal in streams of agricultural catchments – a literature review. Crit. Rev. Environ. Sci. Technol. 37, 381–487. Böhlke, J.K., Harvey, J.W., Voytek, M.A., 2004. Reach-scale isotope tracer experiment to quantify denitrification and related processes in a nitrate-rich stream, midcontinent United States. Limnol. Oceanogr. 49, 821–838. Bott, T.L., 1996. Primary productivity and community respiration. In: Hauer, F.R., Lamberti, G.A. (Eds.), Methods in Stream Ecology. Academic Press, San Diego, pp. 533–556. Bukaveckas, P.A., 2007. Effects of channel restoration on water velocity, transient storage, and nutrient uptake in a channelized stream. Environ. Sci. Technol. 41, 1570–1576. Craig, L.S., Palmer, M.A., Richardson, D.C., Filoso, S., Bernhardt, E.S., Bledsoe, B.P., Doyle, M.W., Groffman, P.M., Hassett, B.A., Kaushal, S.S., Mayer, P.M., Smith, S.M., Wilcock, P.R., 2008. Stream restoration strategies for reducing river nitrogen loads. Front. Ecol. Environ. 6, 529–538. Crenshaw, C.L., Grimm, N.B., Zeglin, L.H., Sheibley, R.W., Dahm, C.N., Pershall, A.D., 2010. Dissolved inorganic nitrogen dynamics in the hyporheic zone of reference

233

and human-altered southwestern U.S. streams. Fundam. Appl. Limnol. Arch. Hydrobiol. 176, 391–405. Dodds, W.K., Lopez, A.J., Bowden, W.B., Gregory, S., Grimm, N.B., Hamilton, S.K., Hershey, A.E., Marti, E., McDowell, W.H., Meyer, J.L., Morall, D., Mulholland, P.J., Peterson, B.J., Tank, J.L., Valett, H.M., Webster, J.R., Wollheim, W., 2002. N uptake as a function of concentration in streams. J. N. Am. Benthol. Soc. 21, 206–220. Dodds, W.K., Marti, E., Tank, J.L., Pontius, J., Hamilton, S.K., Grimm, N.B., Bowden, W.B., McDowell, W.H., Peterson, B.J., Valett, H.M., Webster, J.R., Gregory, S., 2004. Carbon and nitrogen stoichiometry and nitrogen cycling rates in streams. Oecologia 140, 458–467. Dodds, W.K., Oakes, R.M., 2008. Headwater influences on downstream water quality. Environ. Manage. 41, 367–377. Dodla, S.K., Wang, J.J., DeLaune, R.D., Cook, R.L., 2008. Denitrification potential and its relation to organic carbon quality in three coastal wetland soils. Sci. Total Environ. 407, 471–480. Earl, S.R., Valett, H.M., Webster, J.R., 2006. Nitrogen saturation in stream ecosystems. Ecology 87, 3140–3151. Erler, D.V., Eyre, B.D., Davison, L., 2010. Temporal and spatial variability in the cycling of nitrogen within a constructed wetland: a whole-system stable-isotopeaddition experiment. Limnol. Oceanogr. 55, 1172–1178. Ensign, S.H., Doyle, M.W., 2005. In-channel transient storage and associated nutrient retention: evidence from experimental manipulations. Limnol. Oceanogr. 50, 1740–1751. Gaitzenauer, C., 2012. Nutrient Retention Along Stream Banks of Different Morphology. University of Natural Resources and Life Sciences, Vienna, pp. 52 (Master thesis). Gordon, L.J., Peterson, G.D., Bennett, E.M., 2008. Agricultural modifications of hydrological flows create ecological surprises. Trends Ecol. Evol. 23, 211–219. Gücker, B., Pusch, M.T., 2006. Regulation of nutrient uptake in eutrophic lowland streams. Limnol. Oceanogr. 51, 1443–1453. Hamilton, S.K., 2012. Biogeochemical time lags may delay responses of streams to ecological restoration. Freshw. Biol. 57, 43–57. Hancock, P.J., 2002. Human impacts on the stream-groundwater exchange zone. Environ. Manage. 29, 763–781. Harrison, M.D., Groffman, P.M., Mayer, P.M., Kaushal, S.S., 2012. Microbial biomass and activity in geomorphic features in forested and urban restored and degraded streams. Ecol. Eng. 38, 1–10. Jeffrey, S.W., Humphrey, G.F., 1975. New spectrophotometric equations for determining chlorophylls a, b, c1 and c2 in higher plants, algae and natural phytoplankton. Biochem. Physiol. Plants 167, 191–194. Kadlec, R.H., Tanner, C.C., Hally, V.M., Gibbs, M.M., 2005. Nitrogen spiralling in subsurface-flow constructed wetlands: implications for treatment response. Ecol. Eng. 25, 365–381. Klocker, C.A., Kaushal, S.S., Groffman, P.M., Mayer, P.M., Morgan, R.P., 2009. Nitrogen uptake and denitrification in restored and unrestored streams in urban Maryland, USA. Aquat. Sci. 71, 411–424. Kronvang, B., Andersen, H.E., Børgesen, C., Dalgaard, T., Larsen, S.E., Bøgestrand, J., Blicher-Mathiasen, G., 2008. Effects of policy measures implemented in Denmark on nitrogen pollution of the aquatic environment. Environ. Sci. Policy 11, 144–152. Lefebvre, S., Clement, J.-C., Pinay, G., Thenail, C., Durand, P., Marmonier, P., 2007. 15 NNitrate signature in low-order streams: effects of land cover and agricultural practices. Ecol. Appl. 17, 2333–2346. Lefebvre, S., Marmonier, P., Pinay, G., 2004. Stream regulation and nitrogen dynamics in sediment interstices: comparison of natural and straightened sectors of a third-order stream. River Res. Appl. 20, 499–512. Martin, L.A., Mulholland, P.J., Webster, J.R., Valett, H.M., 2001. Denitrification potential in sediments of headwater streams in the southern Appalachian Mountains. J. N. Am. Benthol. Soc. 20, 505–519. Mc Bride, G.B., Chapra, S.C., 2005. Rapid calculation of oxygen in streams: approximate delta method. J. Environ. Eng. 131, 336–341. McClain, M.E., Boyer, E.W., Dent, C.L., Gergel, S.E., Grimm, N.B., Groffman, P.M., Hart, S.C., Harvey, J.W., Johnston, C.A., Mayorga, E., McDowell, W.H., Pinay, G., 2003. Biogeochemical hot spots and hot moments at the interface of terrestrial and aquatic ecosystems. Ecosystems 6, 301–312. Morley, N., Baggs, E.M., 2010. Carbon and oxygen controls on N2 O and N2 production during nitrate reduction. Soil Biol. Biochem. 42, 1864–1871. Mulholland, P.J., Tank, J.L., Webster, J.R., Bowden, W.B., Dodds, W.K., Gregory, S.V., Grimm, N.B., Hamilton, S.K., Johnson, S.L., Marti, E., McDowell, W.H., Merriam, J.L., Meyer, J.L., Peterson, B.J., Peterson, B.J., Wollheim, W.M., 2002. Can uptake length in streams be determined by nutrient addition experiments? Results from an interbiome comparison study. J. N. Am. Benthol. Soc. 21, 544–560. Mulholland, P.J., Helton, A.M., Poole, G.C., Hall, R.O., Hamilton, S.K., Peterson, B.J., Tank, J.L., Ashkenas, L.R., Cooper, L.W., Dahm, C.N., Dodds, W.K., Findlay, S.E.G., Gregory, S.V., Grimm, N.B., Johnson, S.L., McDowell, W.H., Meyer, J.L., Valett, H.M., Webster, J.R., Arango, C.P., Beaulieu, J.J., Bernot, M.J., Burgin, A.J., Crenshaw, C.L., Johnson, L.T., Niederlehner, B.R., O’Brien, J.M., Potter, J.D., Sheibley, R.W., Sobota, D.J., Thomas, S.M., 2008. Stream denitrification across biomes and its response to anthropogenic nitrate loading. Nature 452, 202–205. O’Brien, J.M., Hamilton, S.K., Kinsman-Costello, L.E., Lennon, J.T., Ostrom, N.E., 2012. Nitrogen transformations in a through-flow wetland revealed using wholeecosystem pulsed 15 N additions. Limnol. Oceanogr. 57, 221–234. Oenema, O., van Liere, L., Schoumans, O., 2005. Effects of lowering nitrogen and phosphorus surpluses in agriculture on the quality of groundwater and surface water in the Netherlands. J. Hydrol. 304, 289–301.

234

W. Gabriele et al. / Ecological Engineering 60 (2013) 224–234

Ranalli, A.J., Macalady, D.L., 2010. The importance of the riparian zone and in-stream processes in nitrate attenuation in undisturbed and agricultural watersheds – a review of the scientific literature. J. Hydrol. 389, 406–415. Riis, T., Dodds, W.K., Kristensen, P.B., Baisner, A.J., 2012. Nitrogen cycling and dynamics in a macrophyte-rich stream as determined by a 15 N-NH4 release. Freshw. Biol. 57, 1579–1591. Riseng, C.M., Wiley, M.J., Black, R.W., Munn, M.D., 2011. Impacts of agricultural land use on biological integrity: a causal analysis. Ecol. Appl. 21, 3128–3146. Roberts, B.J., Mulholland, P.J., Houser, J.N., 2007. Effects of upland disturbance and instream restoration on hydrodynamics and ammonium uptake in headwater streams. J. N. Am. Benthol. Soc. 26, 38–53. Runkel, R.L., Bencala, K.E., 1995. Transport of reacting solutes in rivers and streams. In: Singh, V.P. (Ed.), Environmental Hydrology. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 137–164. Runkel, R.L., 1998. One Dimensional Transport with Inflow and Storage (OTIS): A Solute Transport Model for Streams and Rivers, U.S. Geological Survey WaterResources Investigation Report 98-4018. Bechmann, M., Gascuel-Odoux, C., Hofman, G., Kronvang, B., Litaor, M.I., Lo Porto, A., Newell-Price, P., Rubæk, G., 2011. In: Schoumans, O.F., Chardon, W.J. (Eds.), Mitigation options for reducing nutrient emissions from agriculture. A study amongst European member states of Cost action 869. Alterra report 2141, Wageningen, UR.

Smith, M.S., Tiedje, J.M., 1979. Phases of denitrification following oxygen depletion in soil. Soil Biol. Biochem. 11, 261–267. SSSA Book Series 5, 2005. Methods of Soil Analysis, Part 3 – Chemical Methods, 4th ed. Soil Science Society of America Inc., American Society of Agronomy Inc., Madison, WI, USA. Strahler, A.N., 1957. Quantitative analysis of watershed geomorphology. Trans. Am. Geophys. Union 38, 913–920. Stream Solute Workshop, 1990. Concepts and methods for assessing solute dynamics in stream ecosystems. J. N. Am. Benthol. Soc. 9, 95–119. Teufl, B., Weigelhofer, G., Fuchsberger, J., Hein, T., 2012. Effects of bank and channel morphology on the sediment quality of agricultural low-order streams. Environ. Sci. Pollut. Res., http://dx.doi.org/10.1007/s11356-012-1135-2. Verhoeven, J.T.A., Arheimer, B., Yin, C., Hefting, M.M., 2006. Regional and global concerns over wetlands and water quality. Trends Ecol. Evol. 21, 96–103. Weigelhofer, G., Fuchsberger, J., Teufl, B., Welti, N., Hein, T., 2012. Effects of riparian forest buffers on in-stream nutrient retention in agricultural catchments. J. Environ. Qual. 41, 373–379. Welti, N., Bondar-Kunze, E., Mair, M., Bonin, P., Wanek, W., Pinay, G., Hein, T., 2012. Mimicking floodplain reconnection and disconnection using 15 N mesocosm incubations. Biogeosciences 9, 4263–4278.