Journal of Environmental Quality
TECHNICAL
REPORTS
TECHNICAL
REPORTS
Atmospheric Pollutants and Trace Gases
Nitrous Oxide Emission from Riparian Buffers in Relation
to Vegetation and Flood Frequency
P. A. Jacinthe,* J. S. Bills, L. P. Tedesco, and R. C. Barr
The nitrate (NO3-) removal capacity of riparian zones is well
documented, but information is lacking with regard to N2O
emission from riparian ecosystems and factors controlling temporal
dynamics of this potent greenhouse gas. We monitored N2O fluxes
(static chambers) and measured denitrification (C2H2 block using
soil cores) at six riparian sites along a fourth-order stretch of the
White River (Indiana, USA) to assess the effect of flood regime,
vegetation type, and forest maturity on these processes. The study
sites included shrub/grass, aggrading (<15 yr-old), and mature
(>80 yr) forests that were flooded either frequently (more than
four to six times per year), occasionally (two to three times per
year), or rarely (every 20 yr). While the effect of forest maturity
and vegetation type (0.52 and 0.65 mg N2O-m-2 d-1 in adjacent
grassed and forested sites) was not significant, analysis of variance
(ANOVA) revealed a significant effect (P < 0.01) of flood regime
on N2O emission. Among the mature forests, mean N2O flux was
in this order: rarely flooded (0.33) < occasionally flooded (0.99) <
frequently flooded (1.72). Large pulses of N2O emission (up to 80
mg N2O-m-2 d-1) occurred after flood events, but the magnitude
of the flux enhancement varied with flood event, being higher after
short-duration than after long-duration floods. This pattern was
consistent with the inverse relationship between soil moisture and
mole fraction of N2O, and instances of N2O uptake near the river
margin after flood events. These results highlight the complexity
of N2O dynamics in riparian zones and suggest that detailed flood
analysis (frequency and duration) is required to determine the
contribution of riparian ecosystems to regional N2O budget.
Copyright © 2012 by the American Society of Agronomy, Crop Science Society
of America, and Soil Science Society of America. All rights reserved. No part of
this periodical may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information
storage and retrieval system, without permission in writing from the publisher.
J. Environ. Qual.
doi:10.2134/jeq2011.0308
Posted online 13 Dec. 2011.
Received 27 Aug. 2011.
*Corresponding author (pjacinth@iupui.edu).
© ASA, CSSA, SSSA
5585 Guilford Rd., Madison, WI 53711 USA
L
ocated at the interface between terrestrial and
aquatic ecosystems, riparian buffers act as natural filters
for a variety of waterborne pollutants, including nitrate
(NO3-), one of the most prevalent water contaminants in agricultural landscapes (Nolan and Stoner, 2000). In light of their
water quality protection benefits, the preservation and restoration of riparian buffers have been advocated as a cost-effective
approach to mitigating the transfer of nutrients to surface
water bodies (Mitsch et al., 2001; Dosskey et al., 2010).
The removal of NO3- in riparian zones is primarily via plant
uptake and denitrification—the microbial reduction of NO3into nitrous oxide (N2O) and dinitrogen (N2). Riparian vegetation influences the fate of NO3- directly by incorporation
of N into plant tissues and indirectly by the amount and quality of organic matter made available to denitrifiers (Firestone,
1982; Dosskey et al., 2010). While vegetation uptake is considered a temporary N sink, denitrification results in the complete removal of NO3- and, as a by-product, releases the N
gases N2O and N2 into the atmosphere. However, given the
implication of N2O in stratospheric ozone depletion and the
greenhouse effect (Prather et al., 2001; IPCC, 2007), enhanced
emission of N2O from riparian buffers is a concern and some
authors have questioned whether the water quality benefits
of riparian ecosystems are realized at the expense of air quality (Groffman et al., 1998; Dhondt et al., 2004). Noting the
imbalance between the host of N removal studies (see review
Martin et al., 1999) conducted in recent decades and the
paucity of information pertaining to N gases emission from
riparian ecosystems, Groffman et al. (1998) also stressed the
need for N2O flux measurements in these settings to reduce
the uncertainty of the proposed emission factor (N2O emission range of 0.05–2.5% of N intercepted) for determination
of indirect N2O emission in agricultural regions (IPCC, 1997)
and ultimately improve global N2O budgets.
Numerous studies have investigated the N2O production
capacity of riparian soils in the laboratory (Jacinthe et al.,
1998; Addy et al., 1999; Martin et al., 1999; Rotkin-Ellman
et al., 2004; Hunt et al., 2007), but few studies have reported
field-measured N2O fluxes in riparian zones (Ambus and
Christensen, 1995; Walker et al., 2002; Dhondt et al., 2004;
Kim et al., 2009). Therefore, at the field scale, we have limited
Dep. of Earth Sciences, Indiana Univ. Purdue Univ. Indianapolis, 723 W. Michigan
St., Indianapolis, IN 46202. Assigned to Associate Editor Philippe Vidon.
Abbreviations: ANOVA, analysis of variance; PVC, polyvinyl chloride; SOC, soil
organic C.
understanding of the factors (e.g., hydrology, soil, and land
use/land cover) controlling the dynamics of N2O emission
from riparian zones. Reported findings for the effect of vegetation on N2O emission are conflicting. Hopfensperger et
al. (2009) found a negative trend between percent vegetation
cover and N2O fluxes in forested riparian wetlands. Walker
et al. (2002) reported similar rates of N2O emission between
grazed and restored (<2 yr without grazing) riparian zones,
suggesting a limited effect of land use on emission. In contrast, Hefting et al. (2003) reported significantly higher N2O
emission in forested than in grass-covered riparian buffers.
Likewise, when moisture was not severely limiting, McLain
and Martens (2006) measured higher N2O emission rates
in semiarid riparian ecosystems, supporting the leguminous
mesquite trees (Prosopis velutina) compared with other vegetation communities. In almost all these past studies (with the
exception of McLain and Martens, 2006), N2O fluxes were
monitored infrequently (once every 2–3 mo) and for just a
few months. Thus, the reported results may not adequately
capture the temporal variability of N2O emission, a process
known to be highly episodic (McClain et al., 2003). In riparian ecosystems, this variability could be further amplified by
hydrologic events such as floods.
Depending on the geomorphology, riparian buffers are
periodically impacted by flood events, which contribute to
nutrient delivery and spatial distribution of materials across
the riparian zone. Even when connected to large river systems,
riparian areas can retain a substantial amount (?20%) of the
suspended matter transported in flood waters (Brunet et al.,
1994). Moreover, depending on their frequency and duration,
flood events can affect soil moisture regime and have both
long-term and short-term effects on N2O dynamics in riparian
buffers. In light of past studies demonstrating high denitrification potential in poorly drained soils (Hanson et al., 1994;
Ashby et al., 1998), it is reasonable to assume that, by altering
soil moisture level, frequent inundations could lead to the evolution of an active denitrifying community in riparian buffers,
ultimately resulting in increased N2O emission.
Aside from the intensity of production, composition of N
gases evolved (N2O vs. N2) is also of great interest. Conditions
that favor N2O conversion into N2 and minimize the N2O
mole fraction [N2O/(N2O+N2)] are considered less detrimental to air quality. Several studies have identified soil moisture,
pH, organic C, C:N ratio, and NO3- availability as the most
important controllers of the N2O mole fraction (Weier et
al., 1993; Hunt et al., 2007). During flood events, an initial
pulse of N2O production is expected but, as shown in studies
(Jacinthe et al., 2000; Elmi et al., 2005) investigating the effect
of soil saturation duration on the N2O mole fraction, if riparian buffers remain inundated for several days and the available
NO3- pool is depleted, N2O reduction to N2 could become
the dominant process (Körner and Zumft, 1989). Therefore,
the relative composition of N gases emitted from riparian buffers could vary substantially, depending on the magnitude and
duration of flood events.
Quantitative information pertaining to N2O emission
and its regulatory controls is critical to an overall assessment
of the environmental impact of riparian buffers. The present
study was undertaken in an attempt to determine whether the
water quality improvement of riparian buffers is achieved at
the expense of air quality. Specifically, our objectives were to
assess N2O emission in riparian buffers and examine the effects
of vegetation and flood frequency on N2O fluxes. We hypothesized that, under a similar flood regime, higher N2O emission
rates would be measured in riparian zones supporting mature
vegetation compared with areas supporting young and rapidly
growing vegetation. In addition to greater availability of soil
organic C (SOC) (to fuel denitrification) under the mature
compared with the young forest stands, this hypothesis is
also supported by reports of declining plant N uptake, hence
greater mineral N availability to denitrifiers in later stage of
vegetation succession (Boggs and Weaver, 1994; Davidson et
al., 2007).We further hypothesized that, due to periodic interactions with N-laden floodwaters from the adjacent stream and
greater prevalence of wet soil conditions, both the production
potential and emission rates of N2O would be greater in frequently flooded than in flood-protected riparian ecosystems.
Materials and Methods
Site Description
The study was conducted at the Lilly Arbor restored floodplain (39°46¢23²N, 86°11¢09²W), Southwestway Park
(39°39¢26²N, 86°14¢12²W), and McCormick’s Creek State
Park (39°17¢51²N, 86°44¢08²W), along a fourth-order stretch
(70 km) of the White River—from Indianapolis to Spencer in
south central Indiana, USA (Fig. 1). The White River drainage
area ranges from 423.4 × 103 ha near Indianapolis to 773.9
× 103 ha near Spencer. Mean (1940–1970) river discharge at
these locations is 38.6 and 79.3 m3 s-1, respectively (http://
waterwatch.usgs.gov).
The Lilly Arbor floodplain includes a shrub/grass vegetation community and woodlots established in 1999 as part of
site restoration. The bankfull discharge at this location is ?270
m3 s-1 (?3 m stage height, based on USGS rating curve for
a nearby gauging station) and flooding occurs occasionally
(two to three times per year). The Southwestway Park includes
a mature (>80 yr) secondary growth forest and an aggrading
forest established in the mid-1990s on farmland removed from
agriculture in 1984. Located adjacent to the river, the mature
forest is occasionally flooded (two to three times per year),
whereas the aggrading forest is protected from flooding by a
constructed levee. At the McCormick’s Creek State Park, two
tracts of mature forest were delineated—one tract that is rarely
flooded (once every 20 yr), due to its position on a second terrace, and a frequently flooded (four to six times per year) tract
that lies near the confluence of White River and McCormick’s
Creek (Fig. 1). The flow of water from McCormick’s Creek
(a second-order stream) into the White River is sometimes
impeded by high water levels in the White River, causing backwater inundation of the adjacent floodplains.
On the basis of flood frequency and vegetation characteristics, six study sites were selected. They included: occasionally flooded shrub/grass (S1), occasionally flooded aggrading
forest (S2) (<8 yr); rarely flooded aggrading forest (S3) (<15
yr), rarely flooded mature forest (S4) (>80 yr), occasionally
flooded mature forest (S5) (>80 yr), and frequently flooded
mature forest (S6) (>0 yr). Due to their landscape position
Journal of Environmental Quality • Volume 41 • January–February 2012
Fig. 1. The White River watershed in Indiana, USA (depicted by the gray area in the Indiana map insert). Short (<20 km long) tributaries of the White
River are not shown. The squares indicate locations of the study sites along the White River in south central Indiana. The triangle symbols represent locations of weather stations.
and interactions with the White River, the study sites can be
categorized as flood affected (S1, S2, S5, and S6) and flood
protected (S3 and S4).
At S1 site, vegetation consisted of small trees, primarily mulberry (Morus alba) and Siberian elm (Ulmus pumila), and various herbaceous species, including barnyard grass (Echinochloa
spp.), reed canary (Phalaris arundinacea), and goldenrod
(Salidago spp.). At the other sites (S2–S6), vegetation was
dominated by red maple (Acer rubrum), silver maple (Acer saccharinum), white oak (Quercus bicolor), sycamore (Platanus
occidentalis), and cottonwood (Populus L.). The understory
consisted of stinging nettle (Urtica dioica L.), Virginia wild rye
(Elymus virginicus), cutleaf coneflower (Rudbeckia lacinata L.),
and common green briar (Smilax rotundifolia).
Riparian soils are developed from glacial outwash and/or
alluvium deposits, and include predominantly the Genesee
(Fine-loamy, mixed, superactive, mesic Fluventic Eutrudepts)
and Sloan (Fine-loamy, mixed, superactive, mesic Fluvaquentic
Endoaquolls) series. The region’s climate is temperate humid
with mean annual temperature between 11 and 11.6°C, and
precipitation between 1050 and 1140 mm in Indianapolis
and Spencer, respectively (Fig. 2). Temperature and rainfall
data were obtained from the Indiana State Climate Office
(http://climate.agry.purdue.edu/climate/data_archive.asp).
River discharge data (Fig. 2c) were downloaded from USGS
gauging stations 3353000 (39°44¢14²N, 86°10¢08²W) in
Indianapolis and 3357000 (39°16¢52²N, 86°45¢44²W) near
Spencer (Fig. 1).
Nitrous Oxide Flux Measurements
Nitrous oxide flux was measured by the static chamber method
(Jacinthe and Dick, 1997) from August 2005 to June 2007
at sites S1 and S2, and from June 2006 to November 2007 at
the other sites. At each site, two study areas were delineated
(S2 had four study areas) to capture topographic variability
and each study area was instrumented with a set of four static
chambers distributed within a 4-m by 4-m quadrat. Chambers
consisted of a polyvinyl chloride (PVC) pipe (height: 30 cm,
diam: 15 cm) with a beveled end inserted 5 cm into the ground.
During measurement, the PVC pipe was closed with a lid fitted
with a gas sampling port. Once closed, air samples were withdrawn from the chamber headspace at 0, 30, and 60 min, and
transferred into pre-evacuated glass vials (5 mL) fitted with
gray butyl rubber septa (Microliter, Suwanee, GA). Sampling
generally took place between 1100 and 1400 h local time.
Samples were analyzed by gas chromatography to determine
N2O concentration. Daily flux of N2O (mg N2O-N m-2 d-1)
was calculated from the change in N2O concentration inside
the chamber over the 1-h measurement period (obtained by
Fig. 2. White River discharge and rainfall in the drainage basin during the study period (2005–2007). Normal monthly rainfall in Indianapolis (top
graph panel, a) and Spencer (middle panel, b) are shown in graph inserts. In the bottom (c) panel, the solid and dotted lines represent discharge (USGS
3353000, 39°44¢14²N, 86° 10¢08²W) in Indianapolis and gauge height (USGS 3357000, 39°16¢52²N, 86°45¢44²W) in Spencer, respectively. The latter
station only reports gauge height for the period of interest. Discharge data, reported for both stations in 1971, suggest the following relationship: log
QS = 1.07 + [(1.72QI)/(39.28 + QI)], r2: 0.86, P < 0.001, where QI and QS represent discharge (m3 s−1) in Indianapolis and Spencer, respectively.
linear regression) and chamber dimensions (volume: 3.53 ×
10-3 m3, ground area covered: 1.77 × 10-2 m2).
At the S1 and S2 sites, duplicate soil atmosphere samplers
were also installed near each set of static chambers. The soil
atmosphere sampler design is described in Jacinthe and Lal
(2004), and consisted of a PVC rod supporting cells of airpermeable silicone membrane (H-06411–82, Cole-Parmer,
Vernon Hills, IL) centered at 20-, 40-, 60-, 80-, and 100-cm
depth. Soil air composition was monitored on nine occasions
between fall 2005 and spring 2006. Air samples were stored in
evacuated glass vials until analyzed.
Detailed measurements of postflood N2O fluxes were made
at the S1 site on 25 Oct. 2006, and 7 and 30 Mar. 2007 (Fig.
2c). Fluxes were measured along a 35-m transect starting from
the river margin. Chambers were installed once water had
receded, allowing safe access, and 2 d before flux measurement.
At each sampling occasion, surface soil temperature (0–10 cm)
was measured with a portable soil thermometer (Cole Parmer,
Vernon Hills, IL). Composite soil samples were also collected
and brought to the laboratory in plastic bags for determination
of gravimetric soil moisture content (105°C, 72 h).
The mole fraction of N2O was computed as the ratio of
N2O production without acetylene (C2H2) to the rate of pro
duction in the presence of C2H2. Acetylene is a well-known
and widely used inhibitor of N2O reduction in soils and sediments (Ryden et al., 1979). Determination of N2O mole fraction was conducted in spring 2007 at all sites, using small soil
cores (5 cm diam, 7.5 cm long) as described in Hefting et al.
(2003). Intact cores were collected (eight cores per site) on two
occasions (7 and 30 Mar., at S1 and S2; 9 and 30 May, at S3
and S5; 24 Apr. and 24 May, at S4 and S6). At the S1 site, soil
cores were taken next to each of the static chambers along the
35-m transect described above. In the laboratory, cores were
transferred into canning jars (490 mL) with half of the cores
incubated without C2H2 and the other half incubated after a
1-h exposure to C2H2. Jar headspace was sampled at 0, 30, and
60 min to determine the rate of N2O accumulation. Results are
reported on the basis of dry weight of soils in the cores.
Soil Properties
Composite soil samples and intact cores were collected to determine surface soil (0–20 cm) properties (pH, texture, bulk density, organic C, total N). Soil pH was measured with a glass
electrode connected to an Orion pH meter (soil-to-water ratio
of 1:2). Particle size was determined by the hydrometer method
after oxidation of organic matter with H2O2 and dispersion of
Journal of Environmental Quality • Volume 41 • January–February 2012
soil with 5% sodium–hexametaphosphate solution. Soil cores
were oven dried for 72 h (105°C) and dry weight of earth materials (excluding pebbles and gravel) in each core was used to
compute soil bulk density. Finely ground (150 µm), air-dried
soil samples were analyzed for total C and N by dry combustion (850°C) on a Thermo Electron CHNS analyzer (Waltham,
MA). The SOC was computed as the difference between total C
and inorganic C (Loeppert and Suarez, 1996).
Nitrous Oxide Analysis
Soil air samples were analyzed for N2O using a Varian CP3800
(Palo Alto, CA) gas chromatograph equipped with an electron capture (ECD, 63Ni) detector. Operating conditions of
the gas chromatograph were as follows: carrier gas (UHP N2
at 60 mL min-1), oven temperature (90°C), detector temperature (300°C). The stationary phase consisted of a precolumn
(length: 0.3 m and internal diam.: 2 mm) and an analytical
column (length: 1.8 m and internal diam.: 2 mm) packed with
Porapak Q (80–100 mesh). Certified gas standards (0.1, 0.5,
and 1 mL N2O L-1) obtained from Alltech (Deerfield, IL) were
used for calibration.
Data Analysis
Data were analyzed using one-way analysis of variance
(ANOVA) to assess the effect of vegetation and flood regime
on N2O flux. In the analysis, vegetation type (shrub, grass),
forest maturity (aggrading, mature), or flood frequency (rare,
occasional, frequent) were used as the treatment factor and
study areas as pseudo replicates of treatment. The mature forests occurred under all three flood regimes, whereas the aggrading forests were associated with two flood regimes (occasionally
and rarely flooded). As a result, the effect of flood regime was
tested separately for the mature (S4, S5, and S6) and aggrading
forests (S2 and S3). Likewise, the effect of vegetation type was
assessed using data from the occasionally flooded sites supporting shrub/grass (S1) and forest (S2) vegetation. Data from the
flood-protected sites—S3 (aggrading) and S4 (mature)—were
used to test the effect of forest maturity. The ANOVA was performed using the general linear modeling procedure available
in SAS (SAS Institute, 2001). The procedure regression was
used to evaluate relationships between N2O fluxes and envi-
ronmental factors (soil temperature, moisture). Unless otherwise stated, statistical significance was determined at the 95%
confidence level.
Results
Soil and Environmental Conditions
Soil pH (mean: 7.6, range: 7–8.1) did not differ significantly
among the riparian sites (Table 1). Soil texture was coarser
(72% sand) at sites S1 and S2, than at other sites. Soil bulk density was higher (1.22 ± 0.09 g cm-3) and organic C lower at S3
(20.2 ± 5.1 g C kg-1), than at other sites (Table 1). Compared
with the other sites (Table 1), the shrub/grass-covered site (S1)
was generally warmer in the summer (23.7 vs. 20.1°C) and
cooler in the spring (8.5 vs. 12.4°C). Wet soil conditions were
generally more prevalent in the spring than during other seasons and, as expected, the frequently flooded site (S6) was the
wettest (Table 1).
Nitrous Oxide Fluxes and Nitrous Oxide Concentration in
Soil Pore Space
Overall, N2O fluxes (Fig. 3) were more variable at the floodaffected (range: -0.85-11.56 mg N2O-N m-2 d-1) than floodprotected (S3 and S4) riparian sites (range: -0.39-2.06 mg
N2O-N m-2 d-1). The ANOVA revealed significant (P < 0.011)
effect of flood regime on N2O fluxes, regardless of forest maturity (Fig. 4). Of the two aggrading riparian forests (S2 and S3),
N2O flux was significantly higher at the occasionally flooded
(S2: 0.65 ± 0.42 mg N2O-N m-2 d-1) than flood-protected
(S3: 0.26 ± 0.11 mg N2O-N m-2 d-1) site (Fig. 4). Likewise,
among the mature riparian forests (S4, S5, and S6), mean flux
of N2O increased significantly with flood frequency, averaging
0.33 ± 0.14, 0.99 ± 0.58, and 1.72 ± 0.99 mg N2O-N m-2 d-1,
respectively, at the rarely, occasionally, and frequently flooded
riparian sites (Fig. 4). At the flood-affected sites, N2O flux was
significantly related to soil moisture (r2: 0.27, P < 0.01) and
antecedent 10-d river discharge (r2: 0.34, P < 0.01). Regression
analysis showed no relationship between mean N2O emission
and SOC, or the C:N ratio of organic matter at the study sites.
Subtle differences in seasonal N2O flux patterns were
observed (Fig. 3a) between the adjacent shrub/grass (S1) and
Table 1. Surface (0–20 cm) soil properties at the riparian sites. Values are means (n = 4–8 measurements) with standard deviations in parentheses.
Riparian sites
Soil texture
Bulk density (g cm-3)
pH
Organic C (g C kg-1)
Inorganic C (g C kg-1)
Total N (g N kg-1)
Soil temperature (°C)
Soil moisture (g g-1 soil)
spring
summer
fall
spring
summer
fall
S1†
S2
S3
S4
S5
S6
Sandy loam
1 (0.1)
7.8 (0.4)
30.9 (3.1)
20.1 (2.3)
1.8 (0.1)
8.5 (5.5)
23.7 (1.2)
16.7 (1)
0.47 (0.19)
0.18 (0.07)
0.25 (0.03)
Sandy loam
1.1 (0.1)
8.1 (0.3)
37.2 (3.9)
15.6 (1.9)
2 (0.2)
13.2 (1.8)
22.7 (1.6)
16 (0.6)
0.43 (0.14)
0.23 (0.07)
0.27 (0.1)
Loam
1.2 (0.1)
7.8 (0.2)
20.2 (2.1)
12.3 (1.1)
1.5 (0.1)
11.2 (8.6)
18.9 (2.9)
11.7 (2.6)
0.31 (0.06)
0.24 (0.03)
0.26 (0.08)
Loam
1 (0.1)
7.3 (0.1)
33.1 (9.5)
3.4 (2.4)
2.7 (0.7)
10 (5.4)
18.9 (2.8)
14 (4.1)
0.34 (0.06)
0.33 (0.06)
0.45 (0.08)
Loam
1.1 (0.1)
7.7 (0.1)
34.3 (9.3)
20.3 (1.5)
2 (0.7)
15 (6.3)
19.7 (3.3)
11.9 (2.1)
0.3 (0.05)
0.23 (0.06)
0.26 (0.11)
Loam
1.1 (0.2)
7.3 (0.2)
22.6 (1.4)
9.5 (0.5)
1.7 (0.2)
12.8 (6.1)
20.2 (1.4)
14 (4.5)
0.59 (0.15)
0.45 (0.11)
0.43 (0.08)
† S1 = occasionally flooded, grass/shrub; S2 = occasionally flooded young forest; S3 = rarely flooded young forest; S4 = rarely flooded mature forest; S5
= occasionally flooded mature forest; S6 = frequently flooded mature forest.
Fig. 3. Daily fluxes of nitrous oxide at the riparian sites during the period of study. Error bars represent standard deviations of the mean (n = eight
to 16 measurements). Description of sites: S1 = occasionally flooded shrub/grass; S2 = occasionally flooded young forest; S3 = rarely flooded
young forest; S4 = rarely flooded mature forest; S5 = occasionally flooded mature forest; S6 = frequently flooded mature forest.
forested (S2) riparian sites. During fall/winter, emission rates
tended to be higher at S1 (0.41 ± 0.22 mg N2O-N m-2 d-1)
than at S2 (0.31 ± 0.32 mg N2O-N m-2 d-1), whereas the
opposite was observed in spring/summer (S1: 0.68 ± 0.54; S2:
0.99 ± 0.72 mg N2O-N m-2 d-1). Although these patterns may
reflect seasonal cycle in vegetation growth/death and nutrient
availability, the effect of vegetation type (grass vs. forest) was
not significant. During the study period (Fig. 4), N2O emission
averaged 0.52 ± 0.20 mg N2O-N m-2 d-1 under the shrub/grass
vegetation (S1). This rate was not significantly different than
the average emission rate (0.65 ± 0.42 mg N2O-N m-2 d-1) at
the adjacent afforested plots (S2).
Concentration of N2O in soil pore space ranged from
0.1 to 3.1 μL N2O L-1, but below-ambient concentration
(<0.3 mL N2O L-1) was measured in only 10% of the samples
(Fig. 5). Average On most sampling dates, limited variation in
N2O concentration with soil depth was noted (Fig. 5), but gradients of N2O concentration were observed in late fall (21 Nov.
2005) and late spring (24 Apr. 2006). The link between N2O
concentration in soil pore space and N2O emission was variable.
In the fall and winter, instances of elevated N2O in soil pore
space were identified at depth >40 cm, but these were not associated with increased N2O emission (Fig. 3 and 5). However, in
late spring, high N2O concentration in soil air was measured in
the 20-cm depth range and relatively high N2O emission rates
were also measured during that period (Fig. 3 and 5).
Post Flood Nitrous Oxide Emission
Above-normal precipitation (Fig. 2) between October 2006
and April 2007 (rainfall total: 813 mm, normal: 575 mm)
resulted in several flooding events which,
in turn, significantly impacted N2O
emission from the riparian buffers. In
general, post-flood emission rates were
nine to 42 times higher than in adjacent flood-protected buffers (Fig. 2). The
N2O emission peaks recorded at S5 and
S6 provided a good illustration of these
impacts (Fig. 3b–3c). River discharge
(250–490 m3 s-1) at S5 was above flood
stage 2–5 Mar. 2007 (Fig. 2c). A few days
later (9 Mar. 2007), the highest peak of
N2O emission (4.29 ± 0.23 mg N2O-N
m-2 d-1) at S5 was recorded (Fig. 3b).
Similarly, the most intense N2O emission rates at S6 (7.12 ± 2.21 mg N2O-N
m-2 d-1) were measured on 24 Apr. 2007,
following inundation of that site (Fig. 2c
and 3c).
Flood-associated N2O emissions
and their spatial variability were further
Fig. 4. Studywide average of nitrous oxide fluxes in relation to vegetation and flood frequency.
investigated at S1 (Fig. 6a–6c). PostError bars indicate standard deviations. Statistically significant differences are indicated if
flood N2O fluxes (range: -0.1-81.15)
adjacent bars are labeled with different letters (capital or lowercase). Description of sites: S1 =
averaged 46.68 ± 13.24, 2.85 ± 1.26, and
occasionally flooded shrub/grass; S2 = occasionally flooded young forest; S3 = rarely flooded
young forest; S4 = rarely flooded mature forest; S5 = occasionally flooded mature forest; S6 =
2.03 ± 0.79 mg N2O-N m-2 d-1 after the
frequently flooded mature forest.
Journal of Environmental Quality • Volume 41 • January–February 2012
Fig. 5. Seasonal variation in the concentration of nitrous oxide in soil pore space at riparian sites supporting shrub/grass (open circle) and forest
vegetation (filled circle). Each data point is the mean of four measurements.
Fig. 6. Nitrous oxide fluxes at a riparian site (shrub/grass vegetation) following flood events in fall 2006 and spring 2007. In the top graph panels
(a–c), N2O fluxes are represented by the vertical bars, whereas diamond symbols represent the mole fraction of N2O determined as the ratio of
N2O production in soil cores incubated without and with C2H2. Soil cores were extracted next to the static chambers deployed in the field. Note the
scale difference between graphs (a–c).
flood events that occurred in late October 2006, early March
2007, and late March 2007, respectively. Soil moisture and
temperature averaged 0.42, 0.55, and 0.58 g g-1 soil, and 9.1,
2.4, and 12.2°C, respectively, after these events (Fig. 6d–6f ).
The spatial distribution of N2O emission also varied; N2O
emission hotspots were located close to the river margin (<15
m) after the October 2006 flood but shifted away from the
margin (>20 m) after the spring 2007 events (Fig. 6a–6c).
Denitrification Capacity and Nitrous Oxide Mole Fraction
Across sampling dates and study sites, N2O production
averaged 4.12 µg N kg-1 h-1 when soil cores were incubated
without C2H2 (Table 2). Although N gas production was
generally higher at S1, difference between study sites was
not significant (P < 0.22). However, when cores were incubated in the presence of C2H2 (denitrification capacity), significant difference (P < 0.03) among sites was detected with
respect to total N (N2O+N2) gas production. The riparian
Table 2. Nitrous oxide production and mole fraction of N2O [N2O/(N2O+N2)] using soil cores incubated without and with acetylene (C2H2). Values are
means with standard deviation in parentheses. Soil cores (8 cores per site per occasion) were collected on 2 occasions at each riparian site between
March and May 2007.
Riparian
sites†
Soil moisture content
g H2O g soil
0.54 (0.14)
0.45 (0.14)
0.31 (0.04)
0.3 (0.02)
0.31 (0.03)
0.48 (0.13)
-1
S1
S2
S3
S4
S5
S6
Nitrous oxide flux‡
mg N2O-N m d
2.4 (0.5)
0.4 (0.1)
0.5 (0.3)
0.2 (0.1)
2.1 (0.8)
4.5 (3.7)
-2
-1
N2O production
without C2H2
N2O production
with C2H2
————— mg N kg soil h —————
7.98 (3.2)
77.21 (25.56)a§
5.7 (2.29)
36.34 (13.68)ab
3.72 (1.1)
10.32 (4.41)b
1.49 (0.83)
6.99 (5.99)b
2.18 (1.38)
12.96 (10.3)b
1.89 (1.19)
29.14 (13.93)ab
-1
N2O mole fraction
-1
0.10
0.16
0.36
0.21
0.17
0.06
† S1 = occasionally flooded, grass/shrub vegetation; S2 = occasionally flooded young forest; S3 = rarely flooded young forest; S4 = rarely flooded mature
forest; S5 = occasionally flooded mature forest; S6 = frequently flooded mature forest.
‡ Field-measured N2O fluxes with static chambers during the period March to May 2007.
§ Values followed by different letters are significantly different at P < 0.05.
area supporting shrub/grass vegetation (S1) exhibited the
highest N gas production rate (77.2 ± 25.6 mg N kg-1 h-1;
Table 2), more than twice the production capacity at the
adjacent forested site (S2). Overall, total N gas production
capacity was higher (41.9 vs. 8.65 mg N kg-1 h-1) and the
mole fraction of N2O was lower (0.12 vs. 0.28) at the floodaffected (S1, S2, S5, and S6) compared with the floodprotected (S3 and S4) riparian sites (Table 2). A negative
relationship (r2: 0.44, P < 0.02) was found between N2O
mole fraction and soil moisture content (Fig. 7).
Discussion
The importance of vegetation as a driver of N cycling processes
in riparian buffers has long been speculated, but reported
results are inconclusive and sometimes difficult to interpret
due to, among other reasons, failure to account for the legacy
of past land uses (Jacinthe et al., 1998; Addy et al., 1999)
and variation in hydrogeomorphic settings (e.g., water table
depth, topography, soil types) among study sites (Dosskey et
al., 2010). Given the similarity of soil and flood regime at
the adjacent grassed (S1) and forested (S2) riparian sites, the
present study allows for such an assessment without these confounding factors. Thus, the lack of a significant effect of vegetation (shrub/grass vs. forest) observed in the present study
Fig. 7. Relationship between the mole fraction of N2O and moisture
content of riparian soils.
indicates that vegetation type was not an important factor controlling N2O emission from the riparian buffers investigated.
The effect of vegetation on N2O dynamics is generally thought
to involve competition between plant roots and soil microbes
for mineral N during the growing season and increased availability of root-derived organic substrates to denitrifiers at the
onset of senescence (Silvan et al., 2005; Dannenmann et al.,
2007). However, the evidence (N2O emission and concentration in soil air) gathered in the present study provides little
support for these propositions. Our results contrasted with
past studies (Hefting et al., 2003; McLain and Martens, 2006;
van Haren et al., 2010) documenting linkages between N2O
production and vegetation type. van Haren et al. (2010) concluded that tree species was the most important predictor of
N2O fluxes in central Amazonian forests. Hefting et al. (2003)
reported N2O emission rates that were five to 10 times greater
in forested than in grass-covered riparian buffers. It is important to note that the difference observed in the latter study
(Hefting et al., 2003) may have also resulted from the higher
(2.3 times) N-loading rates at the forested than at the grassland
buffer zones. However, our results are in agreement with the
work of Addy et al. (1999) and Groffman et al. (2009) documenting limited difference in NO3--removal and N2O emission between forested and grassed riparian buffers. Clement
et al. (2002) also reported similar findings and concluded that
topography, rather than vegetation, is a more important driver
of denitrification in riparian buffers.
We also examined the significance of forest maturity by
comparing emission from the flood-protected riparian sites
supporting either aggrading (S3) or mature (S4) forest stands.
At both sites, N2O fluxes exhibited only moderate temporal variation (Fig. 3b–3c), further suggesting a weak linkage
between N2O emission and forest annual growth cycle (e.g.,
leaf out, senescence). Although the average N2O flux was
higher at the mature (S4) than at the aggrading forest (S3), difference was not statistically significant. Therefore, contrary to
our hypothesis, forest maturity was not a determining factor of
N2O emission at the study sites. Our results differ from those
of Davidson et al. (2007) and Ball et al. (2007) who reported
higher N2O emission in older than in aggrading forest stands.
While differences in water table depth at the study sites may
have also contributed to the results reported by Ball et al.
Journal of Environmental Quality • Volume 41 • January–February 2012
(2007), Davidson et al. (2007) ascribed their findings to a shift
in N cycling patterns with forest maturation, from a conservative N cycle to a leaky N cycle as forest stands age.
A primary motivation for this study was to document the
impact of flood regime on N2O dynamics in riparian buffers.
As hypothesized, the data collected suggest that flood regime
may have both long-term and immediate impacts on N2O production in riparian zones. Laboratory-assessed denitrification
activity was about four times greater in flood-affected than in
flood-protected riparian areas (Table 2). Likewise, the N2O
mole fractions clearly differentiate between these two groups of
riparian buffers (0.12 and 0.28, respectively) and showed excellent agreement with average values reported in the literature
for this parameter in flooded (0.08) and upland soils (0.37)
(see review by Schlesinger, 2009). Overall, mean N2O emission
from the flood-affected sites (0.98 ± 1.13 mg N2O-N m-2 d-1)
was three times greater than the average emission in floodprotected buffers. This holds true even if measurements made
during flooded periods were excluded (Fig. 3). These results
(denitrification capacity, N2O emission intensity) underscore
the significance of flood regime on the biogeochemistry of
riparian soils.
Our mean N2O emission from the flood-affected sites
was in the upper range of rates reported for riparian buffers in Arizona (0.06–0.32 mg N2O-N m-2 d-1, McLain and
Martens, 2006) and Maryland (0.13–0.24 mg N2O-N m-2 d-1,
Weller et al., 1994) but compared well with fluxes measured
in a riparian area adjacent to agricultural fields in Iowa (0.48–
1.1 mg N2O-N m-2 d-1, Kim et al., 2009). Higher rates of N2O
emission were reported for riparian areas affected by cattle grazing (6.5 mg N2O-N m-2 d-1, Walker et al., 2002) and by high
N loads from intensively managed croplands (0.56–5.4 mg
N2O-N m-2 d-1, Hefting et al., 2003).
Due to logistical limitations and site access difficulties,
immediate (<5 d) post-flood N2O flux measurements were
only possible in a few instances. At our frequently flooded site
(S6), the largest pulse of N2O was recorded in late April 2007
(Fig. 3c). During that period, the area was affected by a series of
floods due to abundant precipitation (Fig. 2b) and amplified by
the unique topography of the site, a low-lying floodplain near
the confluence of a large river and small stream. Because water
level in the White River remained high during that period (Fig.
2), the flow of water from McCormick’s Creek into the main
stem of White River became restricted. As a consequence, backwater flooding occurred, resulting in enhanced N2O emission
at S6 (Fig. 3c). Due to the extensive flooding, immediate access
to the site was not always possible. As a result of these logistical
difficulties, the true maximum N2O peak associated with these
flood events may not have been captured. Considering that the
bulk of N2O emitted from terrestrial ecosystems often occurs
within a small time period, our average N2O emission from
S6 may be underestimated due to our inability to fully capture
these hot moments (McClain et al., 2003).
Immediate (<5 d) post-flood N2O flux measurements were
made at the S1 site (Fig. 6). Although limited, the data provided important insights regarding the dynamics of N gases
in riparian buffers. As observed at S6, post-flood N2O emission was much more intense than during nonflood period, but
the magnitude of the N2O flux enhancement varied with flood
events (Fig. 6). While N2O emission after the March 2007
floods was four to five times the S1 site average (0.52 ± 0.41mg
N2O-N m-2 d-1), emission was nearly 90 times higher (46.7 ±
13.2 mg N2O-N m-2 d-1, Fig. 6) after the flood that occurred at
the end of October 2006. These results not only demonstrated
the significant impact that flood events may have on riparian
zone N2O budget but also highlighted the variability of these
impacts. At present, the factors regulating the intensity of N2O
emission during and immediately after inundation events are
not well characterized.
Although positive relationships between N2O fluxes and soil
temperature have been reported (Parkin and Kaspar, 2006; Ball
et al., 2007), in our study it is difficult to link the difference in
N2O emission intensity after the flood events to soil temperature. While low soil temperature (2.45 ± 1.39°C) may have contributed to the moderate post-flood N2O emission recorded in
early March 2007, it does not adequately explain the difference
in N2O emission during the other events (Fig. 6). For example,
soil temperature was in the same range in October 2006 (9.1
± 0.55°C) and at the end of March 2007 (12.17 ± 2.05°C),
yet the magnitude of post-flood N2O emission was markedly
different. It is conceivable that increased availability of mineral
N (from decomposition of freshly deposited litter by senescent
vegetation) may have also contributed to the higher N2O flux
enhancement observed in late October 2006. This explanation
would be consistent with the elevated N2O concentration in
soil air observed during the same period in 2005 (Fig. 5). Our
results suggest, however, that flood duration is the most likely
explanation for the difference in the magnitude of post-flood
N2O fluxes measured in this study.
Inspection of the White River discharge data (Fig. 2c) indicates that the late October 2006 flood was of short duration
(riparian area was inundated for <2 d). Considering the texture (sandy loam) and drainage characteristics (well drained)
of soils at the site (Table 1), soil saturation was probably brief.
As the flood waters receded and O2 began to penetrate into the
soil profile, conditions could still be optimal for denitrification
(soil moisture: 0.42 g g-1 soil) but probably not conducive to
the reduction of N2O into N2 (Körner and Zumft, 1989). In
contrast, when measurements were made after the late March
2007 flood, the riparian site was inundated for at least five
consecutive days (23–27 March) and also received copious
amounts of rainfall (100 mm in 15 d, Fig. 2a, 2c). Anaerobic
soil conditions may have developed, which would lead to
increased conversion of N2O into N2. Indeed, our assessment
of the N2O mole fraction (Table 2 and Fig. 6) confirmed that
N2O was a minor (<10%) portion of total N gas produced
after the March 2007 floods. Although soil redox information
would have strengthened this argument, this interpretation is
nonetheless well supported by the data collected in the present study and published reports (Weller et al.,1994; Dhondt et
al., 2004), suggesting that N2O emission accounts for a small
proportion of the N intercepted in riparian ecosystems where
standing water is often present (e.g., riparian wetlands).
In contrast to the vigorous post-flood N2O emission peaks,
several instances of N2O uptake were also observed in the wettest section of the riparian zone, suggesting conversion of N2O
into N2 (Fig. 6). Anaerobic conditions, NO3- depletion, and
restricted diffusion of N2O in saturated riparian soils are the
most likely factors regulating that conversion (Körner and
Zumft, 1989; Jacinthe et al., 2000). Thus, successful modeling
of N2O flux in riparian zones hinges on our understanding of
the temporal dynamics of the N2O to N2 conversion and controlling factors (Firestone, 1982; Körner and Zumft, 1989).
The present study has clearly shown that both the frequency
and duration of flood events are important factors to consider
when assessing N gas evolution in riparian zones. Specifically,
the data suggest that N2O is the dominant N gas produced
during short-duration flood events, but the proportion of N2O
diminishes with prolonged flooding. Since this research was
conducted in mostly well-drained riparian buffers, it remains
unclear if similar patterns would be observed in poorly drained
riparian soils receiving N-rich agricultural runoff. In addition
to field studies, future work could also include simulation of
flood events using mesocosms to better characterize the temporal variation of the magnitude and partitioning (between N2
and N2O) of flood-induced N gas production in riparian soils.
Summary and Outlook
The emission of N2O from terrestrial ecosystems is controlled
by a suite of well-documented soil and environmental factors,
including NO3-, organic C, denitrifiers population, soil temperature, and moisture (Firestone, 1982). Building on that
knowledge, the present study attempts to link N2O fluxes to
coarser-scale attributes of riparian ecosystems, such as vegetation type, forest maturity, and flood frequency. This research
approach should facilitate the extrapolation of the study results
to riparian ecosystems in the Midwest (United States) and elsewhere in the humid temperate region. This expectation is well
justified considering that, in most regional surveys, riparian
ecosystems are categorized and mapped primarily on the basis
of vegetation characteristics (Ffolliott et al., 2004; Palik et al.,
2004; Zaimes et al., 2007). However, the lack of a significant
effect of vegetation type and maturity on N2O emission suggests
that these parameters are inadequate for large-scale aggregation
of N2O emission from riparian buffers. Since flood regime has
emerged as the overriding driver of N2O dynamics in riparian
buffers, careful integration of hydrology and geomorphology
will be required to characterize inundation pattern (timing,
frequency, and duration of flood events) and ultimately derive
regional estimates of N2O emission from riparian ecosystems.
Acknowledgments
The authors thank the students Alice Enochs, April Herman, Brandon
Lewis, Andrew Schoering, and Codi Weiler for their help in the
collection and analysis of gas samples. Special thanks to the Center
for Earth and Environmental Sciences, Indianapolis Department of
Parks and Recreation, and Indiana Department of Natural Resources
for providing access to the study sites. The study was funded through a
2006 USGS 104(b) grant (Indiana Water Resources Research Center).
Financial support through USDA–NRI (2009-35112-05241) grant is
also acknowledged.
References
Addy, K.L., A.J. Gold, P.M. Groffman, and P.A. Jacinthe. 1999. Ground
water nitrate removal in subsoil of forested and mowed riparian buffer zones. J. Environ. Qual. 28:962–970. doi:10.2134/
jeq1999.00472425002800030029x
Ambus, P., and S. Christensen. 1995. Spatial and seasonal nitrous oxide
and methane fluxes in Danish forest ecosystems, grassland ecosystems,
and agro-ecosystems. J. Environ. Qual. 24:993–1001. doi:10.2134/
jeq1995.00472425002400050031x
Ashby, J.A., W.B. Bowden, and P.S. Murdoch. 1998. Controls on denitrification in riparian soils in headwater catchments of a hardwood forest in the Catskill mountains, USA. Soil Biol. Biochem. 30:853–864.
doi:10.1016/S0038-0717(98)00012-1
Ball, T., K.A. Smith, and J.B. Moncrieff. 2007. Effect of stand age on greenhouse gas fluxes from a Sitka spruce [Picea sitchensis (Bong.) Carr.] chronosequence on a peaty gley soil. Glob. Change Biol. 13:2128–2142.
doi:10.1111/j.1365-2486.2007.01427.x
Boggs, K., and T. Weaver. 1994. Changes in vegetation and nutrient pools during riparian succession. Wetlands 14:98–109. doi:10.1007/BF03160626
Brunet, R.C., G. Pinay, F. Gazelle, and L. Roques. 1994. Role of the floodplain
and riparian zone in suspended matter and nitrogen retention in the
Adour River, south-west France. Regul. Rivers Res. Manage. 9:55–63.
doi:10.1002/rrr.3450090106
Clement, J.C., G. Pinay, and P. Marmonier. 2002. Seasonal dynamics of denitrification along topohydrosequences in three different riparian wetlands.
J. Environ. Qual. 31:1025–1037. doi:10.2134/jeq2002.1025
Dannenmann, M., R. Gasche, A. Ledebuhr, T. Holst, H. Mayer, and H. Papen. 2007. The effect of forest management on trace gas exchange at
the pedosphere-atmosphere interface in beech (Fagus sylvatica L.) forests
stocking on calcareous soils. Eur. J. For. Res. 126:331–346. doi:10.1007/
s10342-006-0153-3
Davidson, E.A., C.J.R. Carvalho, A.M. Figueira, F.Y. Ishida, J.H.B. Ometto,
G.B. Nardoto, R.T. Saba, S.N. Hayashi, E.C. Leal, I.C.G. Vieira, and
L.A. Martinelli. 2007. Recuperation of nitrogen cycling in Amazonian
forests following agricultural abandonment. Nature 447:995–998.
doi:10.1038/nature05900
Dhondt, K., P. Boeckx, G. Hofman, and O. Van Cleemput. 2004. Temporal
and spatial patterns of denitrification enzyme activity and nitrous oxide
fluxes in three adjacent vegetated riparian buffer zones. Biol. Fertil. Soils
40:243–251. doi:10.1007/s00374-004-0773-z
Dosskey, M.G., P. Vidon, N.P. Gurwick, C.J. Allan, T.P. Duval, and R. Lowrance. 2010. The role of riparian vegetation in protecting and improving
chemical water quality in streams. J. Am. Water Resour. Assoc. 46:261–
277. doi:10.1111/j.1752-1688.2010.00419.x
Elmi, A., D. Burton, R. Gordon, and C. Madramootoo. 2005. Impacts of
water table management on N2O and N2 from a sandy loam soil in
southwestern Quebec, Canada. Nutr. Cycling Agroecosyst. 72:229–240.
doi:10.1007/s10705-005-2920-9
Ffolliott, P.F., M.B. Baker, L.F. DeBano, and D.G. Neary. 2004. Introduction.
p. 1–9. In M.B. Baker et al. (ed.) Riparian areas of the Southwestern
United States: Hydrology, ecology and management. CRC Press, Boca
Raton, FL.
Firestone, M.K. 1982. Biological denitrification. p. 289–326. In F.J. Stevenson
(ed.) Nitrogen in agricultural soils. ASA, Madison, WI.
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. doi:10.1023/A:1009719923861
Groffman, P.M., C.O. Williams, R.V. Pouyat, L.E. Band, and I.D. Yesilonis.
2009. Nitrate leaching and nitrous oxide flux in urban forests and grasslands. J. Environ. Qual. 38:1848–1860. doi:10.2134/jeq2008.0521
Hanson, G.C., P.M. Groffman, and A.J. Gold. 1994. Denitrification in riparian
wetlands receiving high and low groundwater nitrate inputs. J. Environ.
Qual. 23:917–922. doi:10.2134/jeq1994.00472425002300050011x
Hefting, M.M., R. Bobbink, and H. de Caluwe. 2003. Nitrous oxide emission
and denitrification in chronically nitrate-loaded riparian buffer zones. J.
Environ. Qual. 32:1194–1203. doi:10.2134/jeq2003.1194
Hopfensperger, K.N., C.M. Gault, and P.M. Groffman. 2009. Influence
of plant communities and soil properties on trace gas fluxes in riparian northern hardwood forests. For. Ecol. Manage. 258:2076–2082.
doi:10.1016/j.foreco.2009.08.004
Hunt, P.G., T.A. Matheny, and K.S. Ro. 2007. Nitrous oxide accumulation in soils from riparian buffers of a coastal plain watershed-carbon/
nitrogen ratio control. J. Environ. Qual. 36:1368–1376. doi:10.2134/
jeq2006.0255
IPCC. 1997. Revised 1996 IPCC guidelines for national greenhouse gas inventories. IPCC/OECD/IEA.UK Meteorological Office, Bracknell, UK.
IPCC. 2007. Climate change 2007: The physical science basis. Contribution
of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge Univ. Press, New York.
Jacinthe, P.A., and W.A. Dick. 1997. Soil management and nitrous oxide emissions from cultivated fields in Southern Ohio. Soil Tillage Res. 41:221–
235. doi:10.1016/S0167-1987(96)01094-X
Journal of Environmental Quality • Volume 41 • January–February 2012
Jacinthe, P.A., W.A. Dick, and L.C. Brown. 2000. Bioremediation of nitratecontaminated shallow soils and waters via water table management
techniques: Evolution and release of nitrous oxide. Soil Biol. Biochem.
32:371–382. doi:10.1016/S0038-0717(99)00163-7
Jacinthe, P.A., P.M. Groffman, A.J. Gold, and A. Mosier. 1998. Patchiness in microbial nitrogen transformations in groundwater in
a riparian forest. J. Environ. Qual. 27:156–164. doi:10.2134/
jeq1998.00472425002700010022x
Jacinthe, P.A., and R. Lal. 2004. Effects of soil cover on fluxes and depth
concentration of trace gases. Soil Sci. 169:243–259. doi:10.1097/01.
ss.0000126839.58222.0f
Kim, D.G., T.M. Isenhart, T.B. Parkin, R.C. Schultz, T.E. Loynachan, and
J.W. Raich. 2009. Nitrous oxide emissions from riparian forest buffers,
warm-season and cool-season grass filters, and crop fields. Biogeosciences
Discuss. 6:607–650. doi:10.5194/bgd-6-607-2009
Körner, H., and W.G. Zumft. 1989. Expression of denitrification enzymes
in response to the dissolved oxygen level and respiratory substrate in
continuous culture of Pseudomonas stutzeri. Appl. Environ. Microbiol.
55:1670–1676.
Loeppert, R.H., and D.L. Suarez. 1996. Carbonate and gypsum. p. 437–474.
In D.L Sparks (ed.) Methods of soil analysis part 3: Chemical methods.
SSSA, Madison, WI.
Martin, T.L., N.K. Kaushik, J.T. Trevors, and H.R. Whiteley. 1999. Review:
Denitrification in temperate climate riparian zones. Water Air Soil Pollut. 111:171–186. doi:10.1023/A:1005015400607
McClain, M.E., E.W. Boyer, C.L. Dent, S.E. Gergel, N.B. Grimm, P.M.
Groffman, S.C. Hart, J.W. Harvey, C.A. Johnston, E. Mayorga, W.H.
McDowell, and G. Pinay. 2003. Biogeochemical hot spots and hot moments at the interface of terrestrial and aquatic ecosystems. Ecosystems
6:301–312. doi:10.1007/s10021-003-0161-9
McLain, J.E.T., and D.A. Martens. 2006. Moisture controls on trace gas fluxes
in semiarid riparian soils. Soil Sci. Soc. Am. J. 70:367–377. doi:10.2136/
sssaj2005.0105
Mitsch, W.J., J.W. Day, J.W. Gilliam, P.M. Groffman, D.L. Hey, G.W.
Randall, and N.M. Wang. 2001. Reducing nitrogen loading to
the Gulf of Mexico from the Mississippi River Basin: Strategies to
counter a persistent ecological problem. Bioscience 51:373–388.
doi:10.1641/0006-3568(2001)051[0373:RNLTTG]2.0.CO;2
Nolan, B.T., and J.D. Stoner. 2000. Nutrients in groundwaters of the conterminous United States 1992–1995. Environ. Sci. Technol. 34:1156–
1165. doi:10.1021/es9907663
Palik, B., S.M. Tang, and Q. Chavez. 2004. Estimating riparian area extent
and land use in the Midwest. Gen. Tech. Rep. NC-248. USDA, Forest
Service, North Central Research Station, St. Paul, MN.
Parkin, T.B., and T.C. Kaspar. 2006. Nitrous oxide emissions from cornsoybean systems in the Midwest. J. Environ. Qual. 35:1496–1506.
doi:10.2134/jeq2005.0183
Prather, M., D. Ehhalt, F. Dentener, R. Derwent, E. Dlugokencky, E. Holland, I. Isaksen, J. Katima, V. Kirchhoff, P. Matson, P. Midgley, and M.
Wang. 2001. Atmospheric chemistry and greenhouse gases. p. 239–287.
In Houghton et al. (ed.) Climate change 2001: The scientific basis. Contribution of Working Group I to the Third Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge Univ. Press,
Cambridge, UK.
Rotkin-Ellman, M., K. Addy, A.J. Gold, and P.M. Groffman. 2004. Tree
species, root decomposition and subsurface denitrification potential in riparian wetlands. Plant Soil 263:335–344. doi:10.1023/
B:PLSO.0000047747.56753.f3
Ryden, J.C., L.J. Lund, J. Letey, and D.D. Focht. 1979. Direct measurement of denitrification loss from soils: II. Development and application of field method. Soil Sci. Soc. Am. J. 43:110–118. doi:10.2136/
sssaj1979.03615995004300010020x
SAS Institute. 2001. SAS System for Windows, Version 8.02. SAS Inst., Cary, NC.
Schlesinger, W.H. 2009. On the fate of anthropogenic nitrogen. Proc. Natl.
Acad. Sci. USA 106:203–208. doi:10.1073/pnas.0810193105
Silvan, N., E.S. Tuittila, V. Kitunen, H. Vasander, and J. Laine. 2005. Nitrate uptake by Eriophorum vaginatum controls N2O production in a
restored peatland. Soil Biol. Biochem. 37:1519–1526. doi:10.1016/j.
soilbio.2005.01.006
van Haren, J.L.M., R.C. de Oliveira, Jr., N. Restrepo-Coupe, L. Hutyra, P.B.
de Camargo, M. Keller, and S.R. Saleska. 2010. Do plant species influence soil CO2 and N2O fluxes in a diverse tropical forest? J. Geophys.
Res. 115:G03010. doi:10.1029/2009JG001231
Walker, J.T., C.D. Geron, J.M. Vose, and W.T. Swank. 2002. Nitrogen trace
gas emissions from a riparian ecosystem in southern Appalachia. Chemosphere 49:1389–1398. doi:10.1016/S0045-6535(02)00320-X
Weller, D.E., D.L. Correll, and T.E. Jordan. 1994. Denitrification in riparian
forests receiving agricultural discharges. p. 117–131. In W.J. Mitsch (ed.)
Global Wetlands: Old World and New. Elsevier, New York.
Weier, K., J. Doran, J. 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. doi:10.2136/
sssaj1993.03615995005700010013x
Zaimes, G., M. Nichols, and D. Green. 2007. Characterization of riparian
areas. p. 15–29. In G. Zaimes (ed.) Understanding Arizona’s riparian
areas. University of Arizona Extension Pub. AZ 1432. Univ. of Arizona
Extension, Tucson.