HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (22) Citing Articles via Google Scholar Google Scholar Articles by Hefting, M. M. Articles by de Caluwe, H. Search for Related Content PubMed PubMed Citation Articles by Hefting, M. M. Articles by de Caluwe, H. GeoRef GeoRef Citation Agricola Articles by Hefting, M. M. Articles by de Caluwe, H. Related Collections Ground Water Quality Watershed and Landscape Processes Wetlands and Aquatic Processes Air Pollution Nitrogen

TECHNICAL REPORTS
Atmospheric Pollutants and Trace Gases

Nitrous Oxide Emission and Denitrification in Chronically Nitrate-Loaded Riparian Buffer Zones

Mariet M. Hefting^*, Roland Bobbink and Hannie de Caluwe

Section of Landscape Ecology, Dep. of Geobiology, Faculty of Biology, Utrecht Univ., P.O. Box 80084, 3508TB Utrecht, the Netherlands

^* Corresponding author (M.M.Hefting{at}bio.uu.nl)

Received for publication September 25, 2002.

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Riparian buffer zones are known to reduce diffuse N pollution of streams by removing and modifying N from agricultural runoff. Denitrification, often identified as the key N removal process, is also considered as a major source of the greenhouse gas nitrous oxide (N₂O). The risks of high N₂O emissions during nitrate mitigation and the environmental controls of emissions have been examined in relatively few riparian zones and the interactions between controls and emissions are still poorly understood. Our objectives were to assess the rates of N₂O emission from riparian buffer zones that receive large loads of nitrate, and to evaluate various factors that are purported to control N emissions. Denitrification, nitrification, and N₂O emissions were measured seasonally in grassland and forested buffer zones along first-order streams in the Netherlands. Lateral nitrate loading rates were high, up to 470 g N m^-2 yr^-1. Nitrogen process rates were determined using flux chamber measurements and incubation experiments. Nitrous oxide emissions were found to be significantly higher in the forested (20 kg N ha^-1 yr^-1) compared with the grassland buffer zone (2–4 kg N ha^-1 yr^-1), whereas denitrification rates were not significantly different. Higher rates of N₂O emissions in the forested buffer zone were associated with higher nitrate concentrations in the ground water. We conclude that N transformation by nitrate-loaded buffer zones results in a significant increase of greenhouse gas emission. Considerable N₂O fluxes measured in this study indicate that Intergovernmental Panel on Climate Change methodologies for quantifying indirect N₂O emissions have to distinguish between agricultural uplands and riparian buffer zones in landscapes receiving large N inputs.

Abbreviations: EF5-g, nitrous oxide emission factor from ground water and agricultural drainage water • PCA, principal component analysis

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
THE LOSS OF nitrogen compounds from agricultural environments to the shallow ground water and surface water has increased dramatically over the last decades (Van Eck, 1994; Olsthoorn and Fong, 1998). Although several agricultural measures are being taken to reduce the amount of excessive N in the soil profile after harvest, high leaching of nitrate still takes place, polluting the shallow ground water (Addiscot et al., 1991; Olsthoorn and Fong, 1998). This diffuse pollution of shallow ground water is a major cause of the eutrophication of freshwater environments in agricultural landscapes. Riparian buffer zones positioned between the terrestrial and aquatic environments are known to remove large amounts of N from shallow ground water (Peterjohn and Correll, 1984; Pinay and Decamps, 1988; Jordan et al., 1993; Hill, 1996; Groffman et al., 1996). The major mechanisms of N retention in riparian zones are plant uptake, denitrification, and microbial immobilization. Denitrification has been studied extensively in these ecosystems (Lowrance, 1992; Pinay et al., 1993, 2000; Schipper et al., 1993; Hanson et al., 1994; Jordan et al., 1998; Watts and Seitzinger, 2001). The predominant anaerobic environment rich in organic matter provides optimal conditions for denitrification and leads to a considerable denitrification potential. Consequently, denitrification has often been identified as the key N removal process in many riparian buffer zones. However, we have to consider the possibility that riparian zones may contribute to the enhanced greenhouse effect (Crutzen, 1981; Lloyd, 1995). Nitrous oxide (N₂O) can be an important intermediate of denitrification. Nitrous oxide is known as an effective greenhouse gas with a warming potential of approximately 300 compared with CO₂ and is also involved in the catalytic destruction of stratospheric ozone. The question then arises if riparian zones, used as buffers to protect freshwater ecosystems, are a solution to an environmental problem or if they partially substitute one environmental problem by another, that is, reducing water pollution but increasing greenhouse gas emissions. Although many studies have quantified denitrification in riparian buffer zones in agricultural landscapes, N₂O emissions from buffer zones receiving large nitrate loads have received surprisingly little attention (Groffman et al., 1998, 2000). Accordingly, the current recommended methodologies of the Intergovernmental Panel on Climate Change (1997, p. 4.107) to calculate N₂O emissions from ground water and agricultural drainage water do not account for possible high N₂O production in riparian buffer zones (Intergovernmental Panel on Climate Change, 1997, p. 4.107; Mosier et al., 1998).

The flux of N₂O from soils can be due to nitrification or denitrification (Stevens et al., 1997). Generally, denitrification and nitrifier denitrification (Wrage et al., 2001) are believed to be of major importance in the emission of N₂O (Knowles, 1982; Firestone et al., 1980; Robertson and Tiedje, 1987). Nitrous oxide is an obligate intermediate in denitrification and more than 5% of the gaseous end-product is N₂O, whereas N₂O production from nitrifying bacteria is less than 1% of the oxidized NH⁺₄ (Lipschultz et al., 1981). However, soil incubation studies have shown that nitrification can be the dominant process producing N₂O in aerobic soils (Stevens et al., 1997). Due to the predominantly wet conditions and the high nitrate and low ammonium inputs via the shallow ground water, nitrification would contribute little to N₂O emissions from riparian buffer zones. Nevertheless, high mineralization rates in aerobic parts of riparian buffer zones may enhance nitrification and it is thus also important to quantify the relative contribution of nitrification to the total N₂O flux.

There is considerable uncertainty about the ecosystem properties and environmental conditions affecting N₂O production. The N₂O production depends on a number of factors, such as nitrate concentration, anoxia, pH, temperature, organic matter availability, and microbial populations (Blackmer and Bremner, 1978; Firestone et al., 1980; Firestone and Davidson, 1989; Weier et al., 1993; Van Cleemput, 1998). The effect of the individual parameters on N₂O production by denitrifiers is quite well understood. In general, N₂O production is relatively higher under conditions that are suboptimal for denitrification. However, under field conditions there is a simultaneous and interactive influence of different factors, resulting in a great variability. It is therefore difficult to predict which circumstances are enhancing or limiting N₂O emissions. Insight into the ways in which soil variables control the absolute amounts of N₂O emitted is, however, crucial to determine risks of greenhouse gas emissions in zones with a high potential for N transformations.

In this study we quantified N₂O emissions from a forested and a grassland riparian buffer in an agricultural landscape with high nitrate loading. Specifically, this study aimed to (i) assess the significance of N₂O emissions from natural riparian buffer zones in agricultural landscapes; (ii) assess the relative role of denitrification and nitrification as sources of N₂O emission from riparian buffer zones receiving large loads of N from adjacent agricultural fields; (iii) reveal the environmental factors controlling N₂O emission from these ecosystems; and (iv) evaluate the emission factors measured in the perspective of the Intergovernmental Panel on Climate Change emission factors for indirect agricultural sources. Results from this study may contribute to the knowledge base for a future risk assessment of the emissions of N₂O from riparian buffer zones.

	MATERIALS AND METHODS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Research Sites
Two riparian buffer zones along first-order streams were selected on the sandy Pleistocene deposits in the eastern part of the Netherlands (51°25' N, 6°51' E). One of the buffer zones was covered with a natural alder [Alnus glutinosa (L.) Gaertn.] ("forested riparian zone") (Hefting and de Klein, 1998). The grassland buffer zone was dominated by reed meadow grass [Glyceria maxima (Hartm.) Holmb.]. Both riparian buffer zones border intensive agricultural fields planted with maize (Zea mays L.). Nitrate loading rates were high, with a lateral input flux of approximately 630 g N m^-2 yr^-1 into the forested buffer zone and a lateral input flux of 270 g N m^-2 yr^-1 into the grassland buffer zone in 1999 (Sabater et al., 2003). High nitrate loading rates in the forested zone hamper the N₂ fixation by Frankia strains in alder stands (Wolters et al., 1997).

Soils in the two riparian buffer zones were classified as Entisols at the upland–riparian boundary and Histosols close to the stream. In both riparian buffer zones a tertiary clay layer containing the mineral glauconite was found at 1 to 3 m below the soil surface. The study area is underlain by a glacial moraine that has gentle slopes facing west and east. These slopes consist mostly of impermeable glacial till covered by a thin layer of aeolien sand. Due to the position of the glacial till layer, infiltrated rainwater quickly drains toward the permanent streams. These first-order permanent streams have narrow streambeds (0.5–1.5 m) positioned in relatively wide valleys (20–50 m) created by the erosive force of the melting water. This typical geomorphologic situation created optimal conditions for riparian buffer zone research due to the confined layer and the relatively wide riparian buffer zone compared with the stream size.

Ground Water Monitoring
A grid of dipwell piezometers (five transects with at least three piezometers, at 10-m intervals) were installed in both study sites over an elevational gradient from the agricultural fields toward the stream. Piezometers were installed into the phreatic ground water with depth ranging from 0.5 to 3.5 m (depending on the depth of the impermeable clay layer). Tubes were perforated along the bottom 50 cm. Ground water levels were recorded fortnightly from mid-January to mid-November 2000, and water samples were taken seasonally for water quality analyses. Piezometers were emptied prior to water sampling to remove the standing water. Water samples were filtered in the laboratory using glass fiber filters (GF 52; Schleicher & Schuell, Dassel, Germany) before colorimetric analysis of NO^-₃ and NH⁺₄ using a continuous-flow autoanalyzer (SA-40; Skalar Analytical BV, Breda, the Netherlands). Water samples were kept at 4°C during transport and storage, and were analyzed for NO^-₃ and NH⁺₄ within 24 h of sampling.

Experimental Setup
Measurements on N process rates and controlling variables were performed in February, May, August, and November 2000. The riparian buffer zones were divided into three strips, which were each parallel to the stream. A strip that formed the boundary between the stream and the riparian buffer zone will hereafter be called the "stream border." The strip that was established approximately midway within the riparian buffer zone will be called the "intermediate strip." The third strip, indicated hereafter as "field border," was located just inside the buffer zone and downslope of the agricultural field. In each strip, eight permanent frames were installed upon which flux chambers could be attached for N₂O flux measurements. In line with the technical layout of the field analyzer, the frames of each strip were placed in two groups of two frames and one group of four frames. For statistical analysis these groups were interpreted as pseudo-replications and average values of emissions were used, resulting in three replicates per strip.

Because of the destructive sampling method, intact cores for the denitrification assay were collected within 1 m² of the position of the frames. For the denitrification assay, three cores were collected near each of the three groups of frames in each strip. The three cores from each sampling station were pooled, resulting in three replicates per strip, in analogy with the gas flux measurements.

Nitrous Oxide Emission Measurements
Fluxes of N₂O were measured between 0900 and 1600 h using vented, closed flux chambers with an inner diameter of 15.2 cm and a height of 24.2 cm. The flux chambers were attached to preinstalled, permanent frames to minimize disturbance of the soil structure. These frames did not hamper the ground water flow, because of their perforation and the dominant vertical ground water flow in the topsoil due to upwelling. Emissions were measured continuously with 2-min intervals, with two gas samples per flux chamber, over a period of approximately 1 h using a multisampler and a photoacoustic (spectroscopic) infrared gas analyzer (Brüel & Kjær, Nærum, Denmark). Emission rates were calculated from the increase in N₂O concentration over time using linear regression analysis. Besides N₂O emissions, measurements on CH₄ and CO₂ emissions were performed. Carbon dioxide measurements were used to correct for a possible interference between N₂O and CO₂. Soil temperature was measured immediately after the flux chamber was removed, at a depth of 10 cm, using an Eijkelkamp soil thermometer. Pore water was collected next to each flux chamber using Rhizon samplers (Rhizon SMS, 10 cm; Eijkelkamp Agrisearch Equipment, Giesbeek, the Netherlands). Pore water was analyzed for NH⁺₄ and NO^-₃ as described above.

Denitrification, Nitrification, and Soil Properties
"Actual" denitrification was measured using an intact core incubation method with acetylene inhibition (Yoshinari and Knowles, 1976; Ryden et al., 1987). Three 10-cm-deep cores with a diameter of 3.5 cm were collected near each ring, wrapped in perforated aluminum foil, and placed in 1-L preservation jars. The jars were closed with a glass lid containing a rubber gasket and fitted with two rubber septa. In the laboratory, jars were flushed with N₂ for 5 min, leaving the lid open to remove accumulated N₂O and to lower the oxygen content considerably to make conditions more comparable with the soil atmosphere. At the start of the incubation, jars were amended with acetone-free acetylene to bring the soil atmosphere concentration to 10 kPa (10% v/v) acetylene and 90 kPa air. Samples were incubated at average field temperature, and denitrification rates were calculated as the rate of N₂O accumulation in the head space between 1 and 4 h. Gas samples were analyzed directly via a gas chromatograph (Model 3300; Varian, Palo Alto, CA) equipped with an electron capture detector (ECD ⁶³Ni) and Porapak Q columns (2-m-long packed columns; Alltech Associates, Deerfield, IL). Nitrous oxide dissolved in water was taken into account by using the Bunsen coefficient (Wilhelm et al., 1977).

Following completion of the incubation experiments, soil cores from each jar were thoroughly mixed and large stones, roots, and twigs were removed. Soil NH⁺₄ and NO^-₃ contents were determined after extraction (1 h) of 20 g of fresh soil with 100 mL of 0.4 M KCl. After extraction, the pH of the soil suspension was measured using a pH meter (WTW Measurement Systems, Ft. Myers, FL). The suspension was filtered over a glass fiber filter and the extract was analyzed colorimetrically. Soil moisture content was determined gravimetrically after drying approximately 20 g of fresh soil at 105°C for at least 48 h. Organic matter content was determined by loss on ignition of dry (105°C) ground soil at 550°C for 2.5 h.

Net nitrification rates (0–20 cm) were estimated by measured changes in the NO^-₃–N content of soil extracts during 30-d incubations of largely undisturbed topsoil inside in situ buried polyethylene bags (Binkley and Hart, 1989; Eno, 1960).

Statistical Analysis
Both denitrification rates and N₂O fluxes were approximately lognormally distributed. Therefore, these data were log transformed before statistical analysis. Due to large differences in environmental conditions between strips within each riparian buffer zone, the variances of both soil properties and soil processes (i.e., denitrification, nitrification, N₂O emission) were not homogeneously distributed. Unequal variances were still observed after log transformation of the data. To test differences between strips within the buffer zones, nonparametric Kruskal–Wallis or Mann–Whitney U tests were used. In case a significant difference was found with the Kruskal–Wallis test, we used the nonparametric Tukey type multiple comparison Nemenyi test to distinguish differences between strips (Zar, 1998). Variables with a normal distribution and homogeneous variances were tested with analysis of variance (ANOVA) and Tukey post hoc tests. We used a principle component analysis to examine relations between N₂O emission and controlling soil factors. Statistical analysis were performed using SPSS 8.0 for Windows (SPSS, 1997).

	RESULTS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Ground Water Monitoring
Annual fluctuations in ground water levels were rather similar at the two study sites (Fig. 1) . The water table remained close to the soil surface (-15 to 0 cm) and hardly fluctuated throughout the year within the stream border and intermediate strips. Seasonal variations in ground water levels were only observed at the field boundary and agricultural upland sites.

View larger version (20K): [in this window] [in a new window]   Fig. 1. Water table levels relative to the soil surface in strips parallel to the stream in the forested and grassland riparian buffer zones. Water table level was recorded from 11 Jan. 2000 to 13 Nov. 2000. Values are means; n = 5.

View larger version (26K): [in this window] [in a new window]   Fig. 2. Nitrate concentrations in the shallow ground water in three strips parallel to the stream in the forested and grassland riparian buffer zones. Values are means with standard errors; n = 5.

Denitrification, Nitrification, and Nitrous Oxide Emissions
High denitrification activities were measured in both buffer zones, with average rates as high as 80 mg N m^-2 d^-1 (Fig. 3) . Denitrification rates did not differ significantly between the forested and the grassland buffer zones, and there was no significant effect of season. However, rates seemed to follow a seasonal trend, with higher rates in spring and summer. This trend was not significant for the whole riparian zone due to the high variation in the denitrification measurements (Fig. 3; Table 1) , but a significant seasonal effect was observed within the intermediate strip of the forested zone (Kruskal–Wallis, p = 0.036). Significant differences were observed for denitrification activity between the field border (with low rates) and the intermediate strip and the stream border (with much higher activities) (Fig. 3; Table 2) . Maximum denitrification rates were measured in the intermediate strip, particularly in local upwelling areas where ground water with high nitrate concentrations discharged to the surface. The spatial pattern of net nitrification rates was opposite the pattern of denitrification. In both riparian buffer zones nitrification differed significantly between the strips and nitrification was almost exclusively restricted to the field border, while rates were negligible in the intermediate and stream border strips (Fig. 4 ; Table 2). Net nitrification rates did not differ significantly between the forested and grassland buffer zones (Fig. 4; Table 1) and rates were very low compared with the rates of denitrification, except in the field border where nitrification and denitrification were in the same order of magnitude. No significant seasonal effects were found in the nitrification activity (Fig. 4; Table 1).

View larger version (22K): [in this window] [in a new window]   Fig. 3. Denitrification in strips parallel to the stream in the forested and grassland riparian buffer zones, and N2O emission in zones parallel to the stream in the forested and grassland riparian buffer zones. Values are means with standard errors; n = 3. Note differences in scale between the denitrification and N2O emission graphs.

View larger version (28K): [in this window] [in a new window]   Fig. 4. Net nitrification in strips parallel to the stream in the forested and grassland riparian buffer zones. Values are means with standard errors; n = 3.

Environmental Controls
A principal component analysis (PCA) on soil variables for both riparian buffer zones resulted in three components with eigenvalues larger than 1, which explained 76% of the total variance (Table 3) . A correlation bi-plot of the results from a standardized PCA on the soil variables for both riparian zones is given in Fig. 5A . Additional correlations of the soil N transformation processes with the main components are included. Cluster centroids (average score on each component, with standard errors) for the sites are presented in Fig. 5B. As can be seen from the bi-plot, some soil variables were strongly interrelated (arrows pointing in the same direction). The first component explained about 43% of the observed variance. Several soil variables were significantly correlated with this first component, namely soil moisture, organic matter, and pH, but also extractable NH⁺₄ (Table 3). The second component explained about 18% of the observed variance and the only "soil" variable that correlated strongly with this component was the nitrate concentration in the pore water (Table 3). The third component explained 15% of the variation and only correlated significantly with soil temperature (Table 3). This third component can be seen as influenced by the seasonal variation. It is clear from the cluster centroids that the grassland and forested buffer zones were generally separated by the second component (Fig. 5B), which was strongly associated with the nitrate concentration of the pore water (Fig. 5A). Nitrate concentrations in the pore water were significantly lower in the grassland buffer zone (mean = 2.8 mg L^-1) compared with the forested buffer zone (mean = 15 mg L^-1) (one-way ANOVA, F 19.534, p < 0.001). The three strips within the riparian buffer zones were clearly separated by the first component, and only partly by the second (forested stream border strip). The stream border and intermediate strips differed considerably in the forested buffer zone while in the grassland buffer zone the same strips were not significantly different.

View larger version (16K): [in this window] [in a new window]   Fig. 5. (A) Correlation bi-plot from the standardized principal component analysis (PCA) on soil variables. Components are given in Table 3. Correlations of the soil variables with the main axes are given by arrows. The term NO-3 is the nitrate concentration in soil water, OM is the organic matter content, MOIST is the moisture content, and exNO-3 and exNH+4 are the amounts of extractable nitrate and ammonium in the soil. Correlations with soil processes are given by dotted arrows and underlined abbreviations. The term LnDNT stands for denitrification, LnN2O is N2O emission, and NIT is the nitrification. (B) Correlation bi-plot from the standardized PCA with cluster centroids for the strips parallel to the stream in each of the two riparian buffers. The term F indicates the forest buffer zone, and G indicates the grassland buffer zone. Subscripts indicate the strip.

View larger version (22K): [in this window] [in a new window]   Fig. 6. (A) Denitrification and N2O emission versus soil moisture content in the two riparian buffer zones. (B) Net nitrification versus soil moisture content in the two riparian buffer zones.

	DISCUSSION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Because of the high nitrate loading found in shallow ground water in Dutch agricultural landscapes, we had suspected that riparian buffer zones protecting freshwater ecosystems could potentially contribute to atmospheric contamination by having high rates of N₂O emissions. Studies on N₂O emissions from seminatural ecosystems and temperate fen ecosystems and the few existing studies on riparian zones show emissions within a range of 0.1 to 5.3 kg N₂O-N ha^-1 yr^-1 (Weller et al., 1994; Augustin et al., 1996; Groffman et al., 1998; Skiba et al., 1998; Sozanska et al., 2002). The N₂O fluxes found in this study for the grassland buffer zone were within this range with 2 to 4 kg N ha^-1 yr^-1. However, N₂O fluxes found in the forested buffer zone were much higher with a yearly average emission of 20 kg N ha^-1 yr^-1. Particularly high fluxes were found in the intermediate and stream border strips in the forested buffer zone, a yearly average of 30 kg N ha^-1 yr^-1 and local spots exceeding rates of 100 mg N m^-2 d^-1. Fluxes up to 26.9 kg N ha^-1 yr^-1 were found by Merbach et al. (2001) in a drained alder swamp. Higher annual fluxes have only been measured in specific cases with direct fertilization, for instance on grazed fertilized peaty grasslands (36–42 kg N ha^-1 yr^-1; Velthof et al., 1996) or in subtropic fertilized mires with a maximum flux of 157 kg N ha^-1 yr^-1 (Duxbury et al., 1982). The relatively high emissions found in the forested zone clearly confirm the risk of "natural" riparian zones in nitrate-loaded agricultural landscapes as a significant source of greenhouse gas emission. This risk is, however, a relatively recent phenomenon, since historical records from 1944 indicate that this riparian zone used to be a low-productivity natural ecosystem with species-rich plant communities (Westhoff and Jansen, 1990).

In the literature, N₂O emissions are reported to be controlled by the availability of mineral N, soil temperature, and soil water content (Skiba et al., 1998; Van Cleemput, 1998; Heincke and Kaupenjohann, 1999). These controlling variables are universal, but operate in different combinations and in different orders of importance in both space and time (Skiba and Smith, 2000). In our study, the difference in N₂O fluxes between the two study sites could mainly be explained by nitrate availability. Although the input of nitrate via shallow ground water in the grassland zone was certainly not low, nitrate concentrations in the pore water of the topsoil were significantly lower in the grassland than in the forested buffer zone. Lower nitrate concentrations in the grassland buffer zone were probably caused by dilution of the nitrate-enriched shallow ground water with deeper seepage water and possibly by dilution with recharge water from the channel in dry summer periods (unpublished data, 1999; Sabater et al., 2003). Lower nitrate concentrations in the pore water limited N₂O emission rates in the grassland buffer zone, as indicated by the PCA analysis. In the forested buffer zone soil moisture content also played a role in controlling N₂O emissions. In many studies (Heincke and Kaupenjohann, 1999), a positive relation was found between moisture content and N₂O emissions. High water contents with increasingly anoxic conditions stimulate denitrification activity and thus facilitate N₂O production (Davidson and Firestone, 1988; Schnabel and Stout, 1994). On the other hand, high soil moisture content will also increase the residence time of N₂O in the soil, by restricting diffusion, and may consequently enhance the reduction of N₂O to nitrogen gas (Blicher-Mathiesen and Hoffmann, 1999; Jacinthe et al., 2000). Due to the microbial preference for the reduction of nitrate above N₂O, the further reduction of N₂O would only be prominent in soil solutions that are relatively low in nitrate (Davidson and Swank, 1986; Arah et al., 1991). This can partly explain the lower N₂O fluxes found in the grassland riparian zone.

Although N₂O fluxes from the forested riparian buffer zone were high, data from Rusch and Rennenberg (1998) indicate that our flux data underestimated N₂O emissions. They showed that alder trees, which were dominant in our forested riparian zone, could mediate N₂O emissions from the soil to the atmosphere by an efflux from the stem. This efflux was not taken into account with our flux chamber setup, while a possible plant-mediated transport of N₂O oxide via aerenchyma of reed meadow grass in the grassland buffer zone was included (Mosier et al., 1990; Yan et al., 2000).

Denitrification was most certainly the major source of N₂O emission from the wet strips in both riparian areas. The significantly lower denitrification activity in the field border was closely related to the lower soil moisture contents in this zone. The relatively higher N₂O fluxes compared with the denitrification found in the field border (Table 2) are consistent with results from Webster and Hopkins (1996) and may have been caused by a combination of two processes. First, the denitrification end-product is known to shift toward N₂O when the soil oxygen status is less favorable for denitrification (Reddy et al., 1989; Jacinthe et al., 2000). Second, an additional N₂O flux can occur from the nitrification activity in the aerobic field border strip (Stevens et al., 1997). Net nitrification rates measured in this zone were in the same order of magnitude as the low denitrification rates; however, net nitrification is the difference between nitrification, immobilization, and denitrification and can only be used as a "qualitative" measure. Gross nitrification rates can exceed net rates by an order of magnitude (Verchot et al., 2001). Burt et al. (1999) also found that nitrification is the major source of N₂O emission in aerobic soils with a relatively low absolute N₂O emission. The distinct pattern of spatially decoupled denitrification and nitrification activity observed in this study (Fig. 3, 4) was probably influenced by ground water level. As shown in Fig. 1, the average ground water level in the intermediate and stream border strips was continuously close to the soil surface, resulting in anaerobic soil conditions, conductive to denitrification, whereas the ground water level in the field border strip was lower, which resulted in aerobic soil conditions that stimulate nitrification. The relation between N process rates and soil moisture content also illustrates this pattern (Fig. 6). Even though nitrification might occur in aerobic spots in the wet soils, we presume that the nitrification activity is very limited and nitrification is an insignificant source of N₂O under these water-saturated conditions. This presumption is confirmed by the negative relation between nitrification and extractable NH⁺₄ (Fig. 5A), which is due to the accumulation of NH⁺₄ (originating from mineralization) under these wet soil conditions. These results are consistent with an Europe-wide study demonstrating the key role of the ground water table depth in soil N cycling processes in riparian zones (unpublished data, 1999; Pinay and Burt, 2001).

The seasonal trend in denitrification with higher rates in spring and summer in the wet strips (Fig. 3) did not correspond to the trends observed in studies by Burt et al. (1999) and Haycock and Pinay (1993), who found higher denitrification rates in autumn and winter under comparable climatic conditions. The observed difference in seasonal dynamics can probably be attributed to the higher availability of nitrate in the Dutch sites, decreasing the competition between denitrifiers and vegetation in summer, and the stable high ground water level in the Dutch sites providing permanent optimal conditions for denitrification in the topsoil (specifically in the intermediate strips). Consequently, denitrification rates will then be influenced by temperature following the Arrhenius equation (Maag and Vinther, 1996).

Apart from the hazardous N₂O emissions from buffer zones that receive high N inputs, and in spite of beneficial nitrate removal from the shallow ground water, concentrations will in some cases not be reduced strongly enough to prevent eutrophication of the surface waters. This phenomenon was observed in our forested riparian buffer zone with a rather high average nitrate concentration of 10 mg N L^-1 close to the stream. Thus, a realistic evaluation of the total environmental effect of riparian zones is needed. Results from this study indicate that riparian buffer zones that receive large nitrate loads may not fully protect the stream ecosystem.

Current methodologies of the Intergovernmental Panel on Climate Change (1997, p. 4.107) to calculate national N₂O emission from indirect (agricultural) sources do not account for N₂O production in riparian buffer zones. We expected that the emission factor from ground water (EF5-g) underestimates the indirect N₂O emission from riparian buffer zones in Dutch agricultural landscapes with high nitrate concentrations in the shallow ground water and suboptimal soil temperatures. The EF5-g calculated from this study, on the basis of N₂O flux measurements and the yearly incoming NO₃ flux in the ground water, ranges from 0.028 to 0.058 in the forested riparian zone and from 0.016 to 0.031 in the grassland riparian zone. The ranges found in this study are significantly higher than the proposed 0.015 (EF5-g). Due to the high solubility of N₂O in water, additional research is needed to distinguish between transported and locally produced N₂O. However, on the basis of our denitrification measurements, we conclude that there is a significant amount of N₂O production in riparian buffer zones and, depending on the surface area of riparian buffer zones within agricultural landscapes, these areas can significantly increase the indirect N₂O emissions. Groffman et al. (2000) already suggested that emission factors (EF5-g) based only on N₂O losses from supersaturated concentrations in ground water and agricultural drainage water are unrealistically low. On the other hand, Nevison (2000) has reevaluated the emission factor and proposed to reduce the EF5-g from 0.015 to 0.001 kg N₂O-N per kg N input because the Intergovernmental Panel on Climate Change agricultural source estimate now significantly overestimates the observed atmospheric increase. In accordance with Groffman et al. (2000) and Nevison (2000), we suggest that the Intergovernmental Panel on Climate Change inventory might be improved by separately considering emission factors for ground water flowing through riparian areas versus ground water under upland agricultural fields. As indicated by Groffman et al. (1998)( 2000) and Well et al. (2001), there is an urgent need for more data on N₂O emissions from riparian wetland buffer zones to adjust the existing EF5-g emission factor. The results from this study in nitrogen-stressed riparian zones clearly contribute to a more realistic basis for future N₂O emission inventories.

	CONCLUSIONS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
When nitrate loading in riparian buffer zones is high, N₂O is an important end-product of denitrification. In these cases N transformation by buffer zones results in an unfavorable shift from water pollution to an increase in greenhouse gas emission.

Until now, only the beneficial function of riparian zones on water quality improvement has received a lot of attention. To perform a full assessment of riparian wetland functioning, however, we have to evaluate the precise consequences of both forms of environmental pollution to determine the environmental risks.

	ACKNOWLEDGMENTS

 
The results reported here were collected as part of a European project on nitrate buffer zones, the NICOLAS (Nitrogen Control by Landscape Structures in Agricultural Environments) project, which was funded by the European Commission DG XII (H. Barth, Scientific Advisor), Grant ENV4-CT97-0395.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
H. de Caluwe, present address: Dep. of Experimental Plant Ecology, Univ. of Nijmegen, P.O. Box 9010, 6500GL, Nijmegen, the Netherlands.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 

• Addiscot, T.M., A.P. Whitmore, and D.S. Powlson. 1991. Farming, fertilizers and the nitrate problem. CABI Publ., Wallingford, UK.
• Arah, J.R.M., K.A. Smith, I. Crichton, and H.S. Li. 1991. Nitrous oxide production and denitrification in Scottish arable soils. J. Soil Sci. 42:351–367.
• Augustin, J., W. Merbach, W. Schmidt, and E. Reining. 1996. Effect of changing temperature and water table on trace gas emission from minerotrophic mires. J. Appl. Bot. 70:45–51.
• Binkley, D., and S.C. Hart. 1989. The components of nitrogen availability. Assessments in forest soils. Adv. Soil. Sci. 10:57–112.
• Blackmer, A.M., and J.M. Bremner. 1978. Inhibitory effect of nitrate on reduction of N₂O to N₂ by soil microorganisms. Soil Biol. Biochem. 10:187–191.
• Blicher-Mathiesen, G., and C.C. Hoffmann. 1999. Denitrification as a sink for dissolved nitrous oxide in a freshwater riparian fen. J. Environ. Qual. 28:257–262.[Abstract/Free Full Text]
• Burt, T.P., L.S. Matchett, K.W.T. Goulding, C.P. Webster, and N.E. Haycock. 1999. Denitrification in riparian buffer zones: The role of floodplain hydrology. Hydrol. Processes 13:1451–1463.
• Crutzen, P.J. 1981. Atmospheric chemical processes of oxides of nitrogen, including nitrous oxide. p. 17–44. In C.C. Delwiche (ed.) Denitrification, nitrification and nitrous oxide. John Wiley & Sons, New York.
• Davidson, E.A., and M.K. Firestone. 1988. Measurement of nitrous oxide dissolved in soil solution. Soil Sci. Soc. Am. J. 52:1201–1203.[Abstract/Free Full Text]
• Davidson, E.A., and W.T. Swank. 1986. Environmental parameters regulating gaseous nitrogen losses from two forested ecosystems via nitrification and denitrification. Appl. Environ. Microbiol. 52:1287–1292.[Abstract/Free Full Text]
• Duxbury, J.M., D.R. Bouldin, R.E. Terry, and R.L. Tate. 1982. Emissions of nitrous oxide from soils. Nature (London) 298:462–464.
• Eno, C. 1960. Nitrate production in the field by incubating the soil in polyethylene bags. Soil Sci. Soc. Am. J. 24:277–279.[Abstract/Free Full Text]
• Firestone, M.K., and E.A. Davidson. 1989. Microbial basis of NO and N₂O production and consumption in soil. p. 7–21. In M.O. Andreae and D.S. Schimel (ed.) Exchange of trace gases between terrestrial ecosystems and the atmosphere. John Wiley & Sons, Berlin.
• Firestone, M.K., R.B. Firestone, and J.M. Tiedje. 1980. Nitrous oxide from soil denitrification: Factors controlling its biological production. Science (Washington, DC) 208:749–751.[Abstract/Free Full Text]
• Groffman, P.M., A.J. Gold, and K.L. Addy. 2000. Nitrous oxide production in riparian zones and its importance to national emission inventories. Chemosphere 2:291–299.
• Groffman, P.M., A.J. Gold, and P.A. Jacinthe. 1998. Nitrous oxide production in riparian zones and groundwater. Nutr. Cycling Agroecosyst. 52:179–186.
• Groffman, P.M., G. Howard, A.J. Gold, and W.M. Nelson. 1996. Microbial nitrate processing in shallow groundwater in a riparian forest. J. Environ. Qual. 25:1309–1316.[Abstract/Free Full Text]
• Hanson, G.C., P.M. Groffman, and A.J. Gold. 1994. Denitrification in riparian wetlands receiving high and low nitrate inputs. J. Environ. Qual. 23:917–922.[Abstract/Free Full Text]
• Haycock, N.E., and G. Pinay. 1993. Groundwater nitrate dynamics in grass and poplar vegetated riparian buffer strips during the winter. J. Environ. Qual. 22:273–278.[Abstract/Free Full Text]
• Hefting, M.M., and J.J.M. de Klein. 1998. Nitrogen removal in buffer strips along a lowland stream in the Netherlands: A pilot study. Environ. Pollut. 102:521–526.
• Heincke, M., and M. Kaupenjohann. 1999. Effects of soil solution on the dynamics of N₂O emissions: A review. Nutr. Cycling Agroecosyst. 55:133–157.
• Hill, A.R. 1996. Nitrate removal in stream riparian zones. J. Environ. Qual. 25:743–754.[Abstract/Free Full Text]
• Intergovernmental Panel on Climate Change. 1997. Revised 1996 IPCC guidelines for national greenhouse gas inventories. Reference manual. Vol. 3. Organisation for Economic Cooperation and Development, Paris.
• Jacinthe, P.A., W.A. Dick, and L.C. Brown. 2000. Bioremediation of nitrate-contaminated shallow soils and waters via water table management techniques: Evolution and release of nitrous oxide. Soil Biol. Biochem. 32:371–382.
• Jordan, T.E., D.L. Correll, and D.E. Weller. 1993. Nutrient interception by a riparian forest receiving inputs from adjacent cropland. J. Environ. Qual. 22:467–473.[Abstract/Free Full Text]
• Jordan, T.E., D.E. Weller, and D.L. Correll. 1998. Denitrification in surface soils of a riparian forest: Effects of water, nitrate and sucrose additions. Soil Biol. Biochem. 30:833–843.
• Knowles, R. 1982. Denitrification. Microbiol. Rev. 46:43–70.[Free Full Text]
• Lipschultz, F., O.C. Zafiriou, S.C. Wofsy, M.B. McElroy, F.W. Valois, and S.W. Watson. 1981. Production of NO and N₂O by soil nitrifying bacteria. Nature (London) 294:641–643.
• Lloyd, D. 1995. Microbial processes and the cycling of atmospheric trace gases. Trends Ecol. Evol. 10:476–478.
• Lowrance, R. 1992. Groundwater nitrate and denitrification in a coastal plain riparian forest. J. Environ. Qual. 21:401–405.[Abstract/Free Full Text]
• Maag, M., and F.P. Vinther. 1996. Nitrous oxide emission by nitrification and denitrification in different soil types and at different soil moisture contents and temperatures. Appl. Soil Ecol. 4:5–14.
• Merbach, W., J. Augustin, and W. Gans. 2001. Nitrous oxide emission from fen mires in dependence of anthropogenic activities. J. Appl. Bot. 75:118–123.
• Mosier, A., C. Kroeze, C. Nevison, O. Oenema, S. Seitzinger, and O. Van Cleemput. 1998. Closing the global N₂O budget: Nitrous oxide emissions through the agricultural nitrogen cycle. OECD/IPCC/IEA Phase II development of IPCC guidelines for national greenhouse gas inventory methodology. Nutr. Cycling Agroecosyst. 52:225–248.
• Mosier, A., S.K. Mohanty, A. Bhadrachalam, and S.P. Chakravorti. 1990. Evolution of dinitrogen and nitrous oxide from the soil to the atmosphere through rice plants. Biol. Fertil. Soils 9:61–67.
• Nevison, C. 2000. Review of the IPCC methodology for estimating nitrous oxide emissions associated with agricultural leaching and runoff. Chemosphere 2:493–500.
• Olsthoorn, C.S.M., and N.P.K. Fong. 1998. The anthropogenic nitrogen cycle in the Netherlands. Nutr. Cycling Agroecosyst. 52:269–276.
• Peterjohn, W.T., and D.L. Correll. 1984. Nutrient dynamics in an agricultural watershed: Observations on the role of riparian forest. Ecology 65:1466–1475.[ISI]
• Pinay, G., V.J. Black, A.M. Planty-Tabacchi, B. Gumiero, and H. Decamps. 2000. Geomorphic control of denitrification in large river floodplain soils. Biogeochemistry 50:163–182.
• Pinay, G., and T.P. Burt. 2001. NICOLAS; Nitrogen Control by Landscape Structures, Final Report 1997–2000. Final report of the European Research Project 1997–2000 ENV4-CT97-0395. European Commission, Brussels.
• Pinay, G., and H. Decamps. 1988. The role of riparian woods in regulating nitrogen fluxes between the alluvial aquifer and surface waters: A conceptual model. Regulated Rivers Res. Manage. 2:507–516.
• Pinay, G., L. Roques, and A. Fabre. 1993. Spatial and temporal patterns of denitrification in a riparian forest. J. Appl. Ecol. 30:581–591.
• Reddy, K.R., W.H. Patrick, and C.W. Lindau. 1989. Nitrification–denitrification at the plant root–sediment interface in wetlands. Limnol. Oceanogr. 34:1004–1013.
• Robertson, G.P., and J.M. Tiedje. 1987. Nitrous oxide sources in aerobic soils; nitrification, denitrification and other biological processes. Soil Biol. Biochem. 19:187–193.
• Rusch, H., and H. Rennenberg. 1998. Black alder (Alnus glutinosa (L.) Gaertn.) trees mediate methane and nitrous oxide emission from the soil to the atmosphere. Plant Soil 201:1–7.
• Ryden, J.C., J.H. Skinnner, and D.J. Nixon. 1987. Soil core incubation system for the field measurement of denitrification using acetylene inhibition. Soil Biol. Biochem. 19:753–757.
• Sabater, S., A. Butturini, J.C. Clement, D. Dowrick, M.M. Hefting, V. Maitre, G. Pinay, C. Postolache, M. Rzepecki, and F. Sabater. 2003. Nitrogen removal by riparian buffers under various N loads along an European climatic gradient: Patterns and factors of variation. Ecosystems 6:20–30.
• Schipper, L.A., A.B. Cooper, C.G. Harfoot, and W.J. Dyck. 1993. Regulators of denitrification in an organic riparian soil. Soil Biol. Biochem. 25:925–933.
• Schnabel, R.R., and W.L. Stout. 1994. Denitrification loss from two Pennsylvania floodplain soils. J. Environ. Qual. 23:344–348.[Abstract/Free Full Text]
• Skiba, U.M., L.J. Sheppard, J. MacDonald, and D. Fowler. 1998. Some key environmental variables controlling nitrous oxide emissions from agricultural and semi-natural soils in Scotland. Atmos. Environ. 32:3311–3320.
• Skiba, U.M., and K.A. Smith. 2000. The control of nitrous oxide emissions from agricultural and natural soils. Chemosphere 2:379–386.
• Sozanska, M., U. Skiba, and S. Metcalfe. 2002. Developing an inventory of N₂O emissions from British soils. Atmos. Environ. 36:987–998.
• SPSS. 1997. SPSS 8.0 for Windows. SPSS, Chicago.
• Stevens, R.J., R.J. Laughlin, and J.P. Malone. 1997. Measuring the contributions of nitrification and denitrification to the flux of nitrous oxide from soil. Soil Biol. Biochem. 29:139–151.
• Van Cleemput, O. 1998. Subsoils: Chemo- and biological denitrification, N₂O and N₂ emissions. Nutr. Cycling Agroecosyst. 52:187–194.
• Van Eck, G. 1994. Nitrogen losses and nitrogen surpluses in Dutch agriculture. Rep. 3. Ministry of Agric., Nature Management and Fisheries, The Hague, the Netherlands.
• Velthof, G.L., S.C. Jarvis, A. Stein, A.G. Allen, and O. Oenema. 1996. Spatial variability of nitrous oxide fluxes in mown and grazed grasslands on a poorly drained clay soil. Soil Biol. Biochem. 28:1215–1225.
• Verchot, L.V., Z. Holmes, L. Mulon, P.M. Groffman, and G.M. Lovett. 2001. Gross vs net rates of N mineralisation and nitrification as indicators of functional differences between forest types. Soil Biol. Biochem. 33:1889–1901.
• Watts, S.H., and S.P. Seitzinger. 2001. Denitrification rates in organic and mineral soils from riparian sites: A comparison of N₂ flux and acetylene inhibition methods. Soil Biol. Biochem. 32:1383–1392.
• Webster, E.A., and D.W. Hopkins. 1996. Contributions from different microbial processes to N₂O emission from soil under different moisture regimes. Biol. Fertil. Soils 22:331–335.
• Weier, K.L., J.W. Doran, J.F. Power, and D.T. Walters. 1993. Denitrification and the dinitrogen nitrous oxide ratio as affected by soil-water, available carbon, and nitrate. Soil Sci. Soc. Am. J. 57:66–72.[Abstract/Free Full Text]
• Well, R., J. Augustin, J. Davis, S.M. Griffith, K. Meyer, and D.D. Myrold. 2001. Production and transport of denitrification gases in shallow groundwater. Nutr. Cycling Agroecosyst. 60:65–75.
• Weller, D.E., D.L. Correll, and T.E. Jordan. 1994. Denitrification in a riparian forest receiving agricultural discharges. p. 117–131. In W.J. Mitsch (ed.) Global wetlands: Old world and new. Elsevier, Amsterdam.
• Westhoff, V., and A.J.M. Jansen. 1990. Vegetatie gegevens uit de jaren veertig van Noordoost Twente. SWE 90.014. (In Dutch.) Kiwa, Nieuwegein, the Netherlands.
• Wilhelm, E., R. Battino, and R.J. Wilcock. 1977. Low-pressure of gases in liquid water. Chem. Rev. 77:219–262.
• Wolters, D.J., A.D.L. Akkermans, and C. Van Dijk. 1997. Ineffective Frankia strains in wet stands of Alnus glutinosa L. Gaertn. in the Netherlands. Soil Biol. Biochem. 29:1707–1712.
• Wrage, N., G.L. Velthof, V.M.L. Beusichem, and O. Oenema. 2001. Role of nitrifier denitrification in the production of nitrous oxide. Soil Biol. Biochem. 33:1723–1732.
• Yan, X., S. Shi, L. Du, and G. Xing. 2000. Pathways of N₂O emission from rice paddy soil. Soil Biol. Biochem. 32:437–440.
• Yoshinari, T., and R. Knowles. 1976. Acetylene inhibition of nitrous oxide by denitrifying bacteria. Biochem. Biophys. Res. Commun. 69:705–710.[ISI][Medline]
• Zar, J.H. 1998. Biostatistical analysis. 4th ed. Prentice Hall, Upper Saddle River, NJ.

Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2003 32: 1167-1172. [Full Text]  

This article has been cited by other articles:

				A. K. Sovik, J. Augustin, K. Heikkinen, J. T. Huttunen, J. M. Necki, S. M. Karjalainen, B. Klove, A. Liikanen, U. Mander, M. Puustinen, et al. Emission of the Greenhouse Gases Nitrous Oxide and Methane from Constructed Wetlands in Europe J. Environ. Qual., October 27, 2006; 35(6): 2360 - 2373. [Abstract] [Full Text] [PDF]

				A. Bedard-Haughn, K. W. Tate, and C. van Kessel Using Nitrogen-15 to Quantify Vegetative Buffer Effectiveness for Sequestering Nitrogen in Runoff J. Environ. Qual., November 1, 2004; 33(6): 2252 - 2262. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal