Soil texture is known to have an inuence on the physical and
biological processes that produce N2O emissions in agricultural
elds, yet comparisons across soil textural types are limited by
considerations of time and practicality. We used the DayCent
biogeochemical model to assess the eects of soil texture on
N2O emissions from agriculturally productive soils from four
counties in Wisconsin. We validated the DayCent model using
eld data from 2 yr of a long-term (approximately 20-yr) cropping
systems trial and then simulated yield and N2O emissions from
continuous corn (Zea mays L.) and corn–soybean (Glycine max
L.) cropping systems across 35 Wisconsin soil series classied
as either silt loam, sandy loam, or loamy sand. Silt loam soils
had the highest N2O emissions of all soil types, exhibiting 80
to 158% greater mean emissions and 100 to 282% greater
emission factors compared with loamy sand and sandy loam
soils, respectively. The model predicts that for these soils under
these cropping systems, denitrication constituted the majority
of the N2O ux only in the silt loam soils. However, across all soil
textures, locations, and years, denitrication explained the most
variation (74–98%) in total N2O emissions. Our results suggest
that soil texture is an important factor in determining a range of
N2O emission characteristics and is critical for estimating future
N2O emissions from agricultural elds.
Simulated Eects of Soil Texture on Nitrous Oxide Emission Factors
from Corn and Soybean Agroecosystems in Wisconsin
Richard Gaillard,* Benjamin D. Duval, William R. Osterholz, and Christopher J. Kucharik
I U S, N2O emissions
constitute the majority (64%) of greenhouse gas emissions
from agriculture (Larsen et al., 2007) and are principally
driven by N fertilizer applications in row cropping systems (Grace
et al., 2011). Previous studies have calculated N2O emissions
from Midwest agriculture using the Intergovernmental Panel
on Climate Change (IPCC) Tier I emission factor methodol-
ogy (Grace et al., 2011; Larsen et al., 2007), which estimates eld
N2O–N emissions as a percent of total N applied in chemical
and organic forms (IPCC, 2006). e currently recommended
emission factor for N-fertilized row cropping systems is 1% of
total N applied with an uncertainty range of 0.3 to 3%. However,
some studies of corn systems in the Midwest have shown that the
IPCC default emission factor range signicantly underestimates
N2O ux (McSwiney and Robertson, 2005; Parkin and Kaspar,
2006). In contrast, other studies have observed emission factors
within the recommended IPCC default range (e.g., Hoben et
e IPCC Tier II emission factor methodology recommends
that, where available, regional data and assessments be used to
improve regional emission factors in replacement of the default
values (IPCC, 2006), which may resolve some of the inconsis-
tencies in Tier I estimations. Nitrous oxide emissions are directly
related to the availability of N for microbial nitrication and
denitrication and are inuenced by soil texture and moisture
conditions (Conrad, 1996; Groman, 1991; Robertson and
Groman, 2015). Lesschen et al. (2011) demonstrated that
these inuences were signicant for Tier II emission factors in
European agriculture by examining fertilizer rate and source, soil
texture, land use, and precipitation. In the Midwest, McSwiney
and Robertson (2005) and Hoben et al. (2011) showed N2O
emissions from corn elds increase disproportionately with
increasing N application rate (i.e., exponentially), and Hoben
et al. (2011) found a similar relationship between N application
rate and emission factor. However, the eect of soil texture on
N2O emission factors has yet to be studied in the Midwest.
Abbreviations: CC, continuous corn rotation; CSB, corn–soybean rotation; EF,
emission factor; IPCC, Intergovernmental Panel on Climate Change; SOC, soil
R. Gaillard, Nelson Institute for Environmental Studies, Univ. of Wisconsin-Madison,
Science Hall, Madison WI 53706; B.D. Duval, Dep. of Biology, New Mexico Institute
of Mining and Technology, 801 Leroy Pl., Socorro, NM 87801; W.R. Osterholz, Dep.
of Agronomy, Univ. of Wisconsin-Madison, 1575 Linden Dr., Madison, WI 53706;
C.J. Kucharik, Dep. of Agronomy and Nelson Institute Center for Sustainability and
the Global Environment, Univ. of Wisconsin-Madison, 1575 Linden Dr., Madison, WI
53706. Assigned to Associate Editor Stephen Del Grosso.
Copyright © American Society of Agronomy, Crop Science Society of America, and
Soil Science Society of America. 5585 Guilford Rd., Madison, WI 53711 USA.
All rights reserved.
J. Environ. Qual. 45:1540–1548 (2016)
Supplemental material is available online for this article.
Received 29 Mar. 2016.
Accepted 13 July 2016.
*Corresponding author (firstname.lastname@example.org).
Journal of Environmental Quality
ATMOSPHERIC POLLUTANTS AND TRACE GASES
• Soil texture inuences agricultural N2O emissions and emission
• Soil texture inuences absolute and relative rates of nitrication
• Understanding soil texture eects is vital to estimating N losses
Published September 8, 2016
Journal of Environmental Quality 1541
e eects of soil texture on N2O emissions are well docu-
mented. Pore size, a function of soil texture, and O2 availability
interact to control the relative contributions of nitrication and
denitrication to total N2O ux (Bateman and Baggs, 2005;
Maag and Vinther, 1996; Khalil et al., 2004). In nely textured
soils, where increased water content and smaller particle size
increase the likelihood of anoxic microsites, emissions from
denitrication are typically greater relative to nitrication (Del
Grosso et al., 2000; Tiedje, 1988). Fine-textured soils have also
been shown to produce greater total N2O emissions than coarse-
textured soils (van Groenigen et al., 2004). In addition to greater
total emissions, Lesschen et al. (2011), using the dataset from
Stehfest and Bouwman (2006), found a similar eect of soil tex-
ture on emission factor, where clay soils had emission factors 1.5
times that of sandy soils.
Although several studies have collected N2O ux measure-
ments in the Midwest (Burzaco et al., 2013; Cates and Keeney,
1987; Goodroad et al., 1984; Hoben et al., 2011; Jarecki et al.,
2008; McSwiney and Robertson, 2005; Parkin and Kaspar,
2006; Omonode et al., 2011; Osterholz et al., 2014; Venterea
et al., 2011), few have explored relationships between edaphic
variables and emission factors. In the case of soil texture, the lack
of analysis may be due to diculty in establishing experimental
gradients in soil texture under controlled conditions and at the
eld scale. However, the development of process-based com-
puter models may provide a way to evaluate variables that inu-
ence N2O emissions along experimental gradients that would be
materially impractical to establish.
e DayCent biogeochemical model (Del Grosso et al., 2001;
Parton et al., 2001) has been frequently validated for estimating
eld, regional, and national scale agricultural N2O emissions
(Abdalla et al., 2010; Chamberlain et al., 2011; Del Grosso et al.,
2006, 2008; Jarecki et al., 2008; Parton et al., 1998). Jarecki et al.
(2008) simulated N2O emissions from a corn eld in Iowa and,
while nding that DayCent tended to underestimate peak daily
N2O ux, showed that the model predicted cumulative emis-
sions similar to measured emissions. To assess the eects of soil
texture on N2O emissions and emission factors in Wisconsin, we
used DayCent to simulate cumulative N2O emissions from 35
dierent soil series managed as corn and corn–soybean (Glycine
max L.) agroecosystems over 19 yr.
is study aims to assess the eects of soil texture on N2O
emissions for common row crops and soils in Wisconsin and sug-
gest soil texture–based estimates of N2O emission factors. We
rst validated DayCent using data from a long-term cropping
systems experiment in Wisconsin and then simulated data from
four counties in Wisconsin. e objectives of this study were (i)
to generate soil texture–based estimates of N2O emissions and
emission factors for common cropping systems in Wisconsin
and (ii) to evaluate the contributions of nitrication and deni-
trication to N2O ux across a range of soil textures. We hypoth-
esized that higher N2O emissions in ner-textured soils would
result in increased emission factors.
Materials and Methods
DayCent Model Description
DayCent is a biogeochemical process-based model that has
been used to determine patterns of soil C and N2O ux from
agroecosystems (Abdalla et al., 2010; Del Grosso et al., 2005,
2006; Jarecki et al., 2008) and uses a daily time step to simu-
late the eects of environmental variables on the production
of N gases in ecosystems. DayCent simulates nitrication and
denitrication separately, which allowed us to investigate how
changes in soil texture inuence the contribution of each process
to N2O emissions. Nitrication and denitrication submodels
in DayCent are driven by four factors: N substrate availability,
soil water content, temperature, and heterotrophic respiration
(Parton et al., 2001). Soil texture directly inuences N2O pro-
duced through these processes by regulating gas diusivity (i.e.,
O2 availability), which DayCent calculates as a function of tex-
ture and soil water content.
Inputs used to parameterize the DayCent model are soil physi-
cal and chemical properties, weather, latitude, and management
such as tillage, timing and amount of fertilization, and planting
and harvesting events. Soil parameters are soil depth, bulk density,
pH, and percentages of sand, silt, and clay. Weather data consisting
of daily minimum temperature, maximum temperature, and total
precipitation was used to drive the DayCent simulations.
Model Parameterization and Validation
To parameterize the DayCent model for use in Wisconsin,
we generated simulation data (soil organic carbon [SOC], yield,
N2O) for comparison with eld data collected at the Wisconsin
Integrated Cropping Systems Trial (WICST), which is an ongo-
ing long-term agricultural trial designed to evaluate six common
Wisconsin cropping systems (Jokela et al., 2011; Posner et al.,
2008; Sanford et al., 2012). e site is located at the University
of Wisconsin-Madison Arlington Agricultural Research Station
in Arlington, WI (43°18¢ N, 89°20¢ W). e soil at the valida-
tion site is classied as a Plano silt loam (ne-silty, mixed, super-
active, Mesic Typic Argiudoll), with a depth greater than 1 m,
is well drained with slopes of 0 to 2%, and is representative of
productive silt loam soils in Wisconsin.
To stabilize soil C and N in the model, we ran a 2000-yr
spin-up that simulated historical land use and vegetative cover
for Wisconsin (Posner et al., 1995) (Supplemental Table S1).
Maximum harvest index for corn was increased from 0.5 (1970–
1989) to 0.6 (1990–2011) to reect increases in corn yield over
the course of the late 20th Century. We increased production
of alfalfa by reducing optimum temperature for production and
the maximum temperature for production from 25 to 22°C and
from 40 to 37°C, respectively. ese temperature values were
within the ranges reported by Undersander et al. (2011). e
initiation date of the last spin-up phase (alfalfa–alfalfa–alfalfa–
corn) was also adjusted to better simulate WICST soil C stocks
observed in 1989 (Sanford et al., 2012).
We then parameterized the model for two WICST cropping
systems: a continuous corn rotation (CC) and a reduced tillage
corn–soybean rotation (CSB). Typical management practices
(planting date, timing of cultivation and fertilization, and har-
vest date at WICST) were used to parameterize the schedule of
events for each system (Supplemental Table S2). Corn for both
CC and CSB was managed with inorganic N fertilizer and a fall
chisel plow, with no mechanical weed control. No-till soybeans
were simulated without preplant tillage, no mechanical cultiva-
tion, and no N fertilizer addition. We made no further adjust-
ments to parameters for crop physiology for the simulation
1542 Journal of Environmental Quality
of WICST (1993–2011). To simulate soil C loss observed by
Sanford et al. (2012) in 2009, we extended the eects of culti-
vation on SOM decomposition through the months of May
to August, which increased SOM decomposition rates with-
out physical disruption. To simulate the rapid accumulation of
early-season N2O emissions reported by Osterholz et al. (2014),
we removed the respiration restraint on denitrication during
the months of April, May, and June. e model uses heterotro-
phic respiration as a proxy for labile carbon availability; thus,
the removal of the restraint eectively leaves only NO3
ability and O2 diusion as limits on denitrication. We relied
on comparison with observed grain yields from a 15-yr period
at WICST (1997–2011) and eld measurements of N2O for
model validation. A more detailed description of the dataset
used for N2O validation is described by Osterholz et al. (2014).
DayCent Experimental Simulations
To cover a range of soil types, four counties were selected as
representative of major Wisconsin ecoregions (Omernik et al.,
2000): (i) Columbia County, Southeastern Till Plain; (ii) Grant
County, Driless Area; (iii) Marathon County, North Central
Hardwood Forests; and (iv) Waushara County, Central Sands.
ese counties also host agricultural research stations managed
by the University of Wisconsin system for which consistent,
long-term weather data were available (Extension Sta, 2016).
ese data were used to drive simulations for soils located within
From the representative counties, a total of 35 soil series
(Supplemental Table S3) were selected for DayCent simulations
based on two criteria: (i) the soil series was designated as either
“prime farmland” or “farmland of statewide importance” by the
USDA (this limited the selection of soils to those suited for grow-
ing corn and soybeans) and (ii) the series covered at least 2000 ha
within the county. Each of the 35 soils was designated as loamy
sand, sandy loam, or silt loam. To control for the eect of N appli-
cation rate on N2O emissions and emission factors, a single annual
N application rate (157 kg N ha-1) was used that was within the
recommended range for each soil type (Laboski et al., 2006). e
two systems (CC and CSB) were simulated on each soil type with
history and management as described by Osterholz et al. (2014).
Irrigation was not simulated because most row crop agriculture
in Wisconsin is rain fed. For example, the most widely irrigated
county in our study, Waushara, only held 25% irrigated corn
acres and 17% irrigated soybean acres in 2012 (uick Stats Sta,
2016). We validated experimental simulation results by compar-
ing DayCent simulated county-level yields (Mg grain ha-1) with
county-level yields reported by USDA–NASS (uick Stats Sta,
2016). To obtain a simulated county-level yield from our modeled
results, we calculated the arithmetic mean of yields from all simu-
lated soils within each county.
Analysis of N2O Emissions
DayCent output for N2O emissions is expressed as g N2O–N
m-2 d-1 produced as a result of nitrication or denitrication.
For validation with WICST data, simulated emissions were
compared with the linearly interpolated cumulative emissions
over the growing seasons reported by Osterholz et al. (2014).
In contrast, annual N2O emissions for the experimental simu-
lations were the sum of all yearly ux. Annual emissions from
the corn–soybean (CS) rotation were calculated as the average
of emissions from CS–corn and CS–soybean simulated in each
year. is produced a total system average to reect the mean
annual contribution, per unit land area, of emissions from an
agronomic system and allowed us to evaluate annual dierences
between the two cropping systems (CC and CS). We also calcu-
lated emission factors for each simulation year where corn was
present. e IPCC (IPCC, 2006) recommendations suggest
that an emission factor (here denoted by EF) be calculated as:
N O-N fertilized plot N O-N unfertilized plot
N O EF Napplied to plot
To obtain simulated emissions from unfertilized plots, we
ran identical simulations for each system without N fertilizer
additions. Emission Factors were not calculated for soybean
We tested the dierences between observed and simulated
mean yields using paired t tests at a = 0.05. Preliminary analy-
sis of DayCent-simulated N2O emissions failed tests for equal
variance, where the Levene’s test exceeded p = 0.05 (R Core
Team, 2015), indicating unequal variances between soil textures.
To account for unequal variance, we used a rank transforma-
tion (Conover and Iman, 1981) of N2O emissions to construct
a linear model that included the xed eects of county, soil
texture, cropping system, and all interactions. We then con-
ducted ANOVA on the rank-transformed data to determine the
inuence of soil texture on N2O emissions characteristics (kg
N2O–N ha-1 and emission factor). We considered the inuence
of the “county” variable to represent the eect of climate on N2O
emissions because simulations in respective counties were driven
by county-specic weather (Supplemental Table S4). Pairwise
testing of signicant categorical variables within the linear
model was conducted using Tukey’s HSD multiple comparison
procedure (R Core Team, 2015), which isolated signicant pair-
wise dierences. Statistical signicance was determined using a
Results and Discussion
Soil organic C followed the long-term trend of SOC loss
reported by Sanford et al. (2012). Simulated SOC values
decreased from 1989 to 2009 and were within 10% of observed
values for CC and CSB rotations in both 1989 and 2009 (Fig.
1a). Comparison of observed and simulated mean grain yields
diered by rotation (Fig. 1b). ere was no signicant dierence
between observed and simulated mean yields for continuous
corn (df = 14; t = -0.19; p = 0.85) or soybean (df = 14; t =
-1.38; p = 0.19). However, mean simulated yield for corn grown
in rotation with soybean was signicantly lower than observed
values (df = 14; t = 3.06; p = 0.004).
DayCent closely approximated the accumulation of growing
season N2O–N emissions across both systems (CC and CSB)
and growing seasons (2010 and 2011) reported by Osterholz
et al. (2014) (Fig. 2). For the CC rotation, DayCent simulated
Journal of Environmental Quality 1543
99 and 78% of the total observed emissions for 2010 and 2011,
respectively. Simulations of N2O emissions from CSB diered
by crop phase, with DayCent simulating 101 and 95% of corn
emissions for 2010 and 2011, respectively, and 124 and 79% of
soybean emissions for 2010 and 2011, respectively. e perfor-
mance of DayCent was comparable with previous studies, which
found DayCent simulated cumulative emissions were similar to
observed cumulative emissions (Abdalla et al., 2010; De Gryze
et al., 2010; Del Grosso et al., 2008; Jarecki et al., 2008; Scheer
et al., 2014).
DayCent captured the observed dierence in cumulative
N2O emissions from corn for the years 2010 and 2011. In the CC
rotation, observed emissions were 3.88 and 2.74 kg N2O–Nha-1,
where DayCent simulated 3.87 and 2.15 kg N2O–N ha-1 in
2010 and 2011, respectively. For corn in the CSB rotation,
observed emissions were 3.71 and 2.04 kg N2O–N ha-1, where
DayCent simulated 3.78 and 1.94 kg N2O–N ha-1 in 2010 and
2011, respectively. Accumulated precipitation at the Arlington
research station for the growing season (April–September) was
862 mm in 2010, compared with 447 mm in 2011. We would
expect that a very wet year would increase soil water content
and would thus limit O2 availability, increase denitrication,
and support an increase in N2O uxes (Bollmann and Conrad,
1998; Dobbie and Smith, 2003; Khalil et al., 2004; Mathieu et
We were only able to simulate observed periods of high N2O
ux by removing the respiration restraint on denitrication,
which suggests that the model assumption that CO2 respiration
accurately represents labile C constraints on denitrication may
not be correct. Our results demonstrate that, when CO2 respira-
tion was not limiting, the DayCent model accurately simulated
N2O emissions from corn and corn–soybean systems on an agri-
culturally productive soil in the Midwest.
e range of N2O emissions we observed on our simulated
soil series (Table 1) was similar to ranges reported by eld
studies on cropping systems under similar management in
the Midwest (Table 2). Across all 35 simulated soil series, we
Fig. 1. Simulated results from DayCent compared with (a) soil organic
C values (Mg C ha-1) reported by Sanford et al. (2012) and (b) 15-yr
average of harvested grain as reported by Osterholz et al. (2014). Bars
represent SE. CC, continuous corn; CSB, corn–soybean rotation.
Fig. 2. Comparison of cumulative emissions observed by Osterholz
et al. (2014) and simulated by DayCent for two growing seasons
(2010, 2011) and for two cropping systems (CC, continuous corn; CSB,
1544 Journal of Environmental Quality
found mean annual N2O emissions ranged from 1.31 to 6.78
kg N2O–N ha-1. Parkin and Kaspar (2006) reported observed
values outside our range of simulated values, although the higher
emissions reported in Iowa may be due to greater N fertilizer
application because we simulated 157 kg N ha-1, whereas 202 kg
N ha-1 was applied in Iowa. Emissions of N2O have been shown
to increase nonlinearly with N application rate (Hoben et al.,
2011; McSwiney and Robertson, 2005), and these dierences
in N fertilizer application rates could explain the dierence in
mean N2O emissions between our simulated results and those
reported by Parkin and Kaspar (2006).
Simulated mean yields (1997–2011) at the county level were
comparable with county-level yields reported by USDA–NASS
(uick Stats Sta, 2016). With the exception of Waushara
County (df = 1; F = 20.45; p < 0.0001), there was no signicant
dierence between simulated and reported mean county-level
yields (Fig. 3a). Lower simulated yields in Waushara County
may be due to our exclusion of irrigation events from DayCent
Table 1. Nitrous oxide emissions and N2O emission factors for 35 simulated soil series on two cropping systems (continuous corn and corn–soy-
County Soil Texture
Area-scaled emissions Emission factor
CC† CSB‡ CC CSB
—————— k g N2O–N ha-1 —————— ————————— % —————————
Columbia 3.30 (0.13)§ 2.59 (0.10) 1.33 (0.05) 1.65 (0.06)
Dodge silt loam SL 6.46 (0.53) 4.50 (0.32) 3.38 (0.22) 4.11 (0.18)
Friesland ne sandy loam SaL 2.28 (0.34) 1.68 (0.05) 0.73 (0.06) 0.96 (0.04)
Grellton ne sandy loam SaL 2.59 (0.39) 2.12 (0.06) 0.31 (0.04) 0.48 (0.06)
Griswold silt loam SL 2.01 (0.20) 2.08 (0.27) 0.70 (0.04) 1.04 (0.08)
Joy silt loam SL 2.87 (0.43) 2.40 (0.25) 0.87 (0.16) 1.11 (0.17)
Lapeer ne sandy loam SaL 1.92 (0.26) 1.53 (0.13) 0.67 (0.04) 0.92 (0.04)
McHenry silt loam SL 1.89 (0.17) 1.89 (0.21) 0.65 (0.03) 0.93 (0.07)
Okee loamy ne sand LSa 1.91 (0.28) 1.54 (0.18) 0.64 (0.03) 1.13 (0.08)
Oshtemo loamy sand LSa 1.83 (0.26) 1.39 (0.03) 0.62 (0.03) 0.91 (0.04)
Plano silt loam SL 4.82 (0.48) 3.82 (0.39) 2.15 (0.12) 2.64 (0.20)
St. Charles silt loam SL 5.59 (0.49) 4.27 (0.42) 2.56 (0.15) 2.76 (0.20)
Saybrook silt loam SL 4.89 (0.48) 3.77 (0.38) 2.10 (0.12) 2.48 (0.18)
Grant 5.05 (0.13) 3.81 (0.12) 3.02 (0.08) 3.25 (0.09)
Arenzville silt loam SL 5.46 (0.30) 4.10 (0.27) 3.34 (0.16) 3.57 (0.18)
Chaseburg silt loam SL 5.42 (0.27) 3.99 (0.24) 3.46 (0.17) 3.74 (0.18)
Downs silt loam SL 4.89 (0.28) 3.69 (0.24) 2.76 (0.13) 2.88 (0.15)
Dubuque silt loam SL 4.14 (0.25) 3.25 (0.22) 2.52 (0.12) 2.76 (0.14)
Fayette silt loam SL 6.77 (0.33) 4.82 (0.30) 4.44 (0.22) 4.63 (0.22)
Seaton silt loam SL 1.91 (0.16) 1.83 (0.16) 0.60 (0.03) 0.87 (0.07)
Tama silt loam SL 6.78 (0.40) 4.89 (0.34) 3.98 (0.20) 4.12 (0.22)
Marathon 2.41 (0.04) 2.06 (0.04) 1.32 (0.03) 1.60 (0.03)
Chetek sandy loam SaL 1.58 (0.08) 1.46 (0.08) 0.73 (0.03) 1.04 (0.05)
loam SL 2.30 (0.09) 1.99 (0.09) 1.14 (0.06) 1.35 (0.08)
Freeon silt loam SL 2.48 (0.10) 2.15 (0.11) 1.27 (0.06) 1.61 (0.08)
Kennan sandy loam SaL 1.56 (0.08) 1.43 (0.08) 0.73 (0.03) 1.03 (0.05)
Loyal silt loam SL 3.32 (0.13) 2.71 (0.13) 2.04 (0.12) 2.36 (0.13)
Magnor silt loam SL 3.23 (0.13) 2.60 (0.13) 1.98 (0.12) 2.15 (0.12)
Marathon silt loam SL 3.04 (0.12) 2.52 (0.12) 1.78 (0.10) 2.09 (0.11)
Mosinee sandy loam SaL 1.60 (0.08) 1.47 (0.08) 0.72 (0.04) 1.01 (0.05)
Mylrea silt loam SL 3.25 (0.13) 2.62 (0.13) 1.98 (0.12) 2.17 (0.12)
Rietbrock silt loam SL 2.60 (0.10) 2.25 (0.11) 1.39 (0.07) 1.74 (0.09)
Rosholt sandy loam SaL 1.58 (0.08) 1.46 (0.08) 0.71 (0.03) 1.02 (0.05)
Waushara 1.45 (0.04) 1.41 (0.05) 0.62 (0.01) 1.10 (0.02)
Billett sandy loam SaL 1.53 (0.09) 1.54 (0.13) 0.65 (0.03) 1.16 (0.05)
Boyer loamy sand LSa 1.31 (0.06) 1.28 (0.07) 0.59 (0.03) 1.03 (0.04)
Hortonville ne sandy
loam SaL 1.64 (0.09) 1.55 (0.11) 0.65 (0.03) 1.09 (0.05)
Okee loamy sand LSa 1.41 (0.08) 1.37 (0.10) 0.62 (0.03) 1.12 (0.04)
Richford loamy sand Lsa 1.36 (0.07) 1.33 (0.09) 0.61 (0.03) 1.09 (0.04)
Four-county mean 3.10 (0.06) 2.49 (0.05) 1.57 (0.03) 1.88 (0.03)
† Continuous corn.
‡ Corn–soybean rotation.
§ Values are means with SE in parentheses.
Journal of Environmental Quality 1545
simulations. Irrigated corn and soybean acres would have expe-
rienced lower water stress and likely achieved higher yields
than those simulated by nonirrigated corn systems in DayCent.
However, because only 25% of 2012 harvest corn acres were
irrigated in Waushara County, our results represent the major-
ity of corn and soybean acres. Other studies have used county-
level average soil characteristics to drive regional simulations
and validated results with county-level yields (Del Grosso et al.,
2006; Jarecki et al., 2015). Our approach captured 29 to 62%
of total land area within each county (Supplemental Table S3)
and demonstrates that simulating all agriculturally signicant
soils of land area ³2000 ha produces favorable county-level
Only the main eects of soil texture and cropping system sig-
nicantly inuenced N2O emissions (Supplemental Table S5).
We did not nd any eect of climate on emissions, perhaps due
to the relative similarity of the climatic conditions between each
county (Supplemental Table S4). Mean emissions for CC across
all simulated soils were 3.10 and 2.49 kg N2O–N ha-1 yr-1 for
the CSB rotation (Table 3). System dierences were mostly due
to the lack of applied N fertilizer in soybean years causing a lack
of fertilizer-induced emissions, thus reducing mean system emis-
sions of N2O (Osterholz et al., 2014).
e ne-textured silt loam soils had signicantly higher
emissions (80–158% greater) than emissions from the coarse-
textured loamy sand and sandy loam soils (Table 3). Mean values
for the coarse-textured soils ranged from 1.38 to 1.81 kg N2O–N
ha-1 yr-1, with no signicant dierence between emissions from
loamy sand or sandy loam soils. Nitrous oxide emissions were
higher from sandy soils than values reported by van Groenigen
et al. (2004), who reported 0.24 kg N2O–N ha-1 from a sandy
soil (typic endoaquoll; 2% clay, 23% silt, 75% sand) in e
Netherlands cultivated for corn silage and fertilized with 150 kg
N ha-1. However, to our knowledge, values for N2O ux in the
Midwest have not been reported for soils similar to those we have
simulated. Our results agree with previous work that has shown
soil texture to have signicant eects on N2O emissions and that
ne-textured soils are more likely to emit greater amounts of
N2O than coarse-textured soils (Bouwman et al., 2002; Stehfest
and Bouwman, 2006). It has also been reported that in a tropical
climate, clay soils produce greater N2O emissions than silt soils
(Weitz et al., 2001). Conversely, it has been reported that in a
narrower range of soil texture (13–20% clay, 0.7–2.3% sand),
increasing clay content is associated with lower N2O emissions
(Gu et al., 2013). We simulated soils with sand content rang-
ing from 7 to 84% and with clay content ranging from 6 to 25%
(Supplemental Table S3), and conicting results suggest that the
eect of texture on N2O emissions may be a matter of the range
of texture investigated.
Additionally, we might expect that our simulated cumula-
tive emissions, which include all yearly ux, tend to be higher
than growing-season values from similar systems reported in the
literature. Of the studies reported in Table 2, only Parkin and
Kaspar (2006) collected N2O ux throughout the year, and in
that case only one measurement was taken for each month over
the winter period (December–January). Although the mecha-
nisms of N2O ux from frozen and thawing soil are not currently
well understood, emissions under these conditions can be signi-
cant (de Bruijn et al., 2009; Risk et al., 2014; Tatti et al., 2014).
DayCent simulates over-winter ux events due to soil warming,
although further work is required to quantify these emissions
and whether the model processes accurately represent proposed
Soil texture signicantly inuenced N2O emissions from
both nitrication and denitrication (Fig. 4). In loamy sand and
sandy loam soils, nitrication constituted 73 and 71% of total
N2O emissions, respectively, whereas in silt loam soils nitrica-
tion contributed 24% to total N2O emissions. Emissions from
nitrication were signicantly dierent between all three soil
textures (df = 2; F = 65.49; p < 0.0001), with emissions of 1.00,
1.13, and 1.06 kg N2O–N ha-1 from loamy sands, sandy loams,
and silt loam soils, respectively. ere were also signicant dif-
ferences in denitrication between all three textures (df = 2; F
= 371.47; p < 0.0001), with emissions of 0.36, 0.45, and 2.11
kg N2O–N ha-1 from loamy sands, sandy loams, and silt loam
soils, respectively. Other studies have found that denitrication
can produce between 52 and 100% of N2O emissions from a
ne, mesic Typic Hapludalf loamy soil in Michigan (Opdyke
et al., 2009; Ostrom et al., 2010), which is higher in sand con-
tent (40–60% sand) than silt loams from our study (7–33%
sand). e soil water dynamics that exert a signicant inuence
on the relative contributions of nitrication and denitrication
Table 2. Nitrous oxide emissions as reported by studies conducted on corn systems in the Midwest.
Study State Management N fertilizer application N2O emissions
kg N ha-1kg N2O–N ha-1
Cates and Keeney (1987) Wisconsin conventional tillage 181–237 3.6–5.2
Parkin and Kaspar (2006) Iowa chisel plow 202 10.2–11.3
no-till 202 7.87–11.3
Jarecki et al. (2008) Iowa chisel plow 168 4.26
Omonode et al. (2011) Indiana chisel plow 258–296 7.25
moldboard plow 258–296 5.61
no-till 258–296 3.37
Venterea et al. (2011) Minnesota conventional tillage 146 0.63
no-till 146 0.75
Burzaco et al. (2013) Indiana conventional tillage 22–202 1.55–3.52
Osterholz et al. (2014) Wisconsin conventional tillage 185 3.88
reduced tillage 157 3.71
1546 Journal of Environmental Quality
to total N2O ux (Bateman and Baggs, 2005; Mathieu et al.,
2006; Stevens et al., 1997) are strongly dependent on soil tex-
tural properties. Gu et al. (2013) found no signicant associa-
tion between soil texture and potential denitrication, although
our conicting results may be explained by the range of soil
textures studied. However, it is still unclear at which combina-
tion of soil textural properties nitrication or denitrication
produces the majority of total N2O ux. Across all soil textures,
N2O produced via denitrication explained more variation in
N2O emissions than N2O produced via nitrication (Fig. 5).
Emissions of N2O–N from nitrication in DayCent are calcu-
lated as 2% of total nitried NO3
-–N (Parton et al., 2001), and
values for nitrication N2O were typically bounded within the
range of 0.5to2.5kgN2O–N ha-1. Denitrication appears to
drive variation in emissions across textures and suggests that it
is an important factor to consider for N2O mitigation strategies.
Further study is required to determine the relationship between
soil textural properties and the contribution of nitrication and
denitrication to total N2O ux, yet our results suggest that the
relationship exists and is signicant in determining the magni-
tude of N2O emissions.
e main eects of system and soil texture, as well as the two-
way interactions between system and climate (i.e., county) and
system and soil texture, had a signicant eect on N2O emission
factors (Supplemental Table S5). Emission factors for CSB were
consistently higher than for CC (Table 3). To our knowledge, no
rotation eect on emission factors has been observed. Osterholz
et al. (2014) did not observe increased area-scaled emissions
on corn following soybean, although less N was applied rela-
tive to corn following corn. e range of emission factors sim-
ulated by DayCent were comparable to the range of emission
Fig. 3. Comparison of 1997 to 2011 average grain yields reported by
Quick Stats Sta, 2016 and simulated by DayCent for (a) corn and
(b) soybean across four counties (Columbia, Grant, Marathon, and
Waushara). Error bars represent SE.
Table 3. Simulated N2O emissions and N2O emission factors by soil
texture (loamy sand, sandy loam, silt loam) and cropping system.
Area-scaled emissions Emission factor
CC† CSB CC CSB‡
— — k g N2O–N ha-1 yr-1 —— ————— % —————
3.10 (0.06)§ 2.49 (0.03) 1.57 (0.03) 1.88 (0.03)
Loamy sand 1.56 (0.08)a 1.38 (0.04)a 0.62 (0.01)a 1.060 (0.01)a
Sandy loam 1.81 (0.07)a 1.59 (0.04)a 0.66 (0.01)b 0.970 (0.01)b
Silt loam 3.92 (0.08)b 3.07 (0.05)b 2.11 (0.04)c 2.390 (0.04)c
† Continuous corn.
‡ Corn–soybean rotation.
§ Values are means with SE in parentheses. Dierent letters indicate
means are signicantly dierent.
Fig. 4. Simulated relative contribution of nitrication and denitrication
to total N2O emissions across three soil textures (LSa, loamy sand; SaL,
sandy loam; SL, silt loam). Dierent letters represent signicant dier-
ences between soil textures.
Journal of Environmental Quality 1547
factors (2–7%) from corn systems in the Midwest reported by
McSwiney and Robertson (2005). Emission factors from our
simulated sandy soils (0.62–1.06) were higher than the 0.07
reported by van Groenigen et al. (2004), which further suggests
that emission factors from sandy soils under corn agriculture
need additional study.
Our results agree with other studies that have demonstrated
signicant variability in emission factors (Leip et al., 2011;
McSwiney and Robertson, 2005). Other attempts to quantify
this variability have assessed the impact of environmental fac-
tors such as water-lled pore space (Dobbie and Smith, 2003),
precipitation (Lesschen et al., 2011), and N fertilization rate
(Mosier et al., 2006). Although we were able to isolate the con-
tributions of nitrication and denitrication to total N2O emis-
sions, neither was signicantly correlated with emission factor.
e dierence in mean emission factors between the soil textures
may be explained by sand content (Del Grosso et al., 2006), but
further study is required to isolate drivers of variability in emis-
sion factors due to soil characteristics. Additionally, Shcherbak
et al. (2014) found that % soil C and soil pH exerted a signicant
inuence on emission factors. Our results agree with these nd-
ings because ne-textured soils, which typically support higher
SOC levels (Oades, 1988), had signicantly higher simulated
emission factors than coarse-textured soils (Table 3). us, we
suggest that soil texture be included in the calculation of regional
emission factors for Midwest cropping systems. Additionally, it
should be examined whether the eect of soil texture could be
integrated with the ndings of Shcherbak et al. (2014) and the
nonlinear relationship between N fertilization rate and N2O
emissions (Hoben et al., 2011; McSwiney and Robertson, 2005)
to improve the estimation of N2O emission factors. For example,
soil texture–based mean N2O emission factors could be adjusted
according to N rate, generating regional inventories by integrat-
ing a static environmental condition (soil texture) with a variable
management practice (N rate).
We found that soil texture exerted a signicant inuence
on simulated N2O emissions and emission factors from agri-
culturally productive soils across Wisconsin and that N2O
emissions and emission factors were greatest in silt loam soils.
Using DayCent, we found that denitrication emissions were
a signicant predictor of variation in total N2O emissions but
were unable to conrm a similar eect on emission factor. We
conclude that soil texture is an important consideration for
estimations of N2O emissions from agricultural elds and have
reported ranges for expected N2O emissions and emission fac-
tors for common agricultural soils and cropping systems in
Wisconsin. Future research should evaluate our ndings in com-
parison with eld data, investigate the factors that may inuence
interannual variability in denitrication rates, and further rene
an understanding of the relationship between soil texture, N2O
emissions, and emission factors.
Abdalla, M., M. Jones, J. Yeluripati, P. Smith, J. Burke, and M. Williams. 2010.
Testing DayCent and DNDC model simulations of N2O uxes and assess-
ing the impacts of climate change on the gas ux and biomass production
from a humid pasture. Atmos. Environ. 44(25):2961–2970. doi:10.1016/j.
Bateman, E.J., and E.M. Baggs. 2005. Contributions of nitrification and denitrifi-
cation to N2O emissions from soils at different water-filled pore space. Biol.
Fertil. Soils 41(6):379–388. doi:10.1007/s00374-005-0858-3
Bollmann, A., and R. Conrad. 1998. Inuence of O2 availability on NO and N2O re-
lease by nitrication and denitrication in soils. Glob. Change Biol. 4(4):387–
Bouwman, A.F., L.J.M. Boumans, and N.H. Batjes. 2002. Modeling global annual N2O
and NO emissions from fertilized elds. Global Biogeochem. Cycles 16(4):28-
de Bruijn, A.M.G., K. Butterbach-Bahl, S. Blagodatsky, and R. Grote. 2009. Model
evaluation of dierent mechanisms driving freeze–thaw N2O emissions. Agric.
Ecosyst. Environ. 133(3-4):196–207. doi:10.1016/j.agee.2009.04.023
Burzaco, J.P., D.R. Smith, and T.J. Vyn. 2013. Nitrous oxide emissions in Mid-
west US maize production vary widely with band-injected N fertilizer
rates, timing and nitrapyrin presence. Environ. Res. Lett. 8(3):035031.
Cates, R., and D. Keeney. 1987. Nitrous oxide production throughout the year from
fertilized and manured maize elds. J. Environ. ual. 16:443–447. doi:10.2134/
Chamberlain, J.F., S.A . Miller, and J.R. Frederick. 2011. Using DAYCENT to quantify
on-farm GHG emissions and N dynamics of land use conversion to N-managed
switchgrass in the Southern U.S. Agric. Ecosyst. Environ. 141(3-4):332–341.
Conover, W.J., and R.L. Iman. 1981. Rank transformations as a bridge between para-
metric and nonparametric statistics. Am. Stat. 35(3):124–129.
Conrad, R. 1996. Soil microorganisms as controllers of atmospheric trace gases (H2,
CO, CH4, OCS, N2O and NO). Microbiol. Rev. 60:609–640.
De Gryze, S., A. Wolf, S.R. Kaa, J. Mitchell, D.E. Rolston, S.R. Temple, J. Lee, and
J. Six. 2010. Simulating greenhouse gas budgets of four California cropping sys-
tems under conventional and alternative management. Ecol. Appl. 20(7):1805–
Del Grosso, S.J., A.D. Halvorson, and W.J. Parton. 2008. Testing DAYCENT
model simulations of corn yields and nitrous oxide emissions in irrigated till-
age systems in Colorado. J. Environ. ual. 37(4):1383–1389. doi:10.2134/
Del Grosso, S., A. Mosier, W. Parton, and D. Ojima. 2005. DAYCENT mo del analysis
of past and contemporary soil N2O and net greenhouse gas ux for major crops
in the USA. Soil Tillage Res. 83(1):9–24. doi:10.1016/j.still.2005.02.007
Fig. 5. Correlation between source of emissions (nitried kg N2O–N
ha-1, denitried kg N2O–N ha-1) and total emissions (kg N2O–N ha-1)
across two cropping systems (CC, continuous corn; CSB, corn–soybean
rotation) by soil texture (loamy sand, sandy loam, silt loam).
1548 Journal of Environmental Quality
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and
D.S. Schimel. 2001. Simulated interaction of carbon dynamics and nitrogen
trace gas uxes using the DAYCENT model. In: M. Shaer, S. Hansen, and L.
Ma, editors, Modeling carbon and nitrogen dynamics for soil management. CRC
Press, Boca Raton, FL. p. 303–332.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, D.S. Ojima, A.E. Kulmala, and S.
Phongpan. 2000. General model for N2O and N2 gas emissions from soils
due to denitrication. Global Biogeochem. Cycles 14(4):1045–1060.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.K. Walsh, D.S. Ojima, and P.E. orn-
ton. 2006. DAYCENT national-scale simulations of nitrous oxide emissions
from cropped soils in the United States. J. Environ. ual. 35(4):1451–1460.
Dobbie, K., and K. Smith. 2003. Nitrous oxide emission factors for agricultural soils in
Great Britain: e impact of soil water-lled pore space and other controlling vari-
ables. Glob. Change Biol. 9:204–218. doi:10.1046/j.1365-2486.2003.00563.x
Extension Sta. 2016. UW extension Ag weather. http://agwx.soils.wisc.edu/uwex_
agwx/weather/index (accessed 15 Mar. 2016).
Goodroad, L., D. Keeney, and L. Peterson. 1984. Nitrous oxide emissions from agri-
cultural soils in Wisconsin. J. Environ. ual. 13(4):557–561. doi:10.2134/
Grace, P.R., G. Philip Robertson, N. Millar, M. Colunga-Garcia, B. Basso, S.H.
Gage, and J. Hoben. 2011. e contribution of maize cropping in the Midwest
USA to global warming: A regional estimate. Agric. Syst. 104(3):292–296.
Groman, P. 1991. Ecology of nitrication and denitrication in soil evaluated at scales
relevant to atmospheric chemistry. In: J. Rogers and W. Whitman, editors, mi-
crobial production and consumption of greenhouse gases: Methane, nitrogen
oxides and halomethanes. American Society for Microbiology, Washington,
DC. p. 201–217.
Gu, J., B. Nicoullaud, P. Rochette, A. Grossel, C. Hénault, P. Cellier, and G. Richard.
2013. A regional experiment suggests that soil texture is a major control of N2O
emissions from tile-drained winter wheat elds during the fertilization period.
Soil Biol. Biochem. 60:134–141. doi:10.1016/j.soilbio.2013.01.029
Hoben, J.P., R.J. Gehl, N. Millar, P.R. Grace, and G.P. Robertson. 2011. Non-
linear nitrous oxide (N2O) response to nitrogen fertilizer in on-farm
corn crops of the US Midwest. Glob. Change Biol. 17(2):1140–1152.
IPCC. 2006. 2006 IPCC guidelines for national greenhouse gas inventories. IPCC,
Jarecki, M.K., J.L. Hateld, and W. Barbour. 2015. Modeled nitrous oxide emissions
from corn elds in Iowa based on county level data. J. Environ. ual. 44(2):431–
Jarecki, M.K., T.B. Parkin, A.S.K. Chan, J.L. Hateld, and R. Jones. 2008. Compari-
son of DAYCENT-simulated and measured nitrous oxide emissions from a corn
eld. J. Environ. ual. 37(5):1685–1690. doi:10.2134/jeq2007.0614
Jokela, W., J. Posner, J. Hedtcke, T. Balser, and H. Read. 2011. Midwest cropping sys-
tem eects on soil properties and on a soil quality index. Agron. J. 103(5):1552–
Khalil, K., B. Mary, and P. Renault. 2004. Nitrous oxide production by nitrication
and denitrication in soil aggregates as aected by O2 concentration. Soil Biol.
Biochem. 36(4):687–699. doi:10.1016/j.soilbio.2004.01.004
Laboski, C.A.M., J.B. Peters, and L. Bundy. 2006. Nutrient application guidelines for
eld, vegetable, and fruit. Cooperative Extension Publishing, Madison, WI.
Larsen, J., T. Damassa, and R. Levinson. 2007. Charting the Midwest: An inven-
tory and analysis of greenhouse gas emissions in America’s heartland. World
Resources Institute, Washington, DC.
Leip, A., M. Busto, and W. Winiwarter. 2011. Developing spatially stratied N2O emis-
sion factors for Europe. Environ. Pollut. 159(11):3223–3232. doi:10.1016/j.
Lesschen, J.P., G.L. Velthof, W. de Vries, and J. Kros. 2011. Dierentiation of nitrous ox-
ide emission factors for agricultural soils. Environ. Pollut. 159(11):3215–3222.
Maag, M., and F. Vinther. 1996. Nitrous oxide emission by nitrication and denitrica-
tion in dierent soil types and at dierent soil moisture contents and tempera-
tures. Appl. Soil Ecol. 4(1):5–14. doi:10.1016/0929-1393(96)00106-0
Mathieu, O., C. Hénault, J. Lévêque, E. Baujard, M.-J. Milloux, and F. Andreux. 2006.
uantifying the contribution of nitrication and denitrication to the nitrous
oxide ux using 15N tracers. Environ. Pollut. 144(3):933–940. doi:10.1016/j.
McSwiney, C.P., and G.P. Robertson. 2005. Nonlinear response of N2O ux to incre-
mental fertilizer addition in a continuous maize (Zea mays L.) cropping system.
Glob. Change Biol. 11(10):1712–1719. doi:10.1111/j.1365-2486.2005.01040.x
Mosier, A.R., A.D. Halvorson, C.A. Reule, and X.J. Liu. 2006. Net global warming
potential and greenhouse gas intensity in irrigated cropping systems in northeast-
ern Colorado. J. Environ. ual. 35(4):1584–1598. doi:10.2134/jeq2005.0232
Oades, J.M. 1988. e retention of organic matter in soils. Biogeochemistry 5(1):35–
Omernik, J.M., S.S. Chapman, R.A. Lillie, and R.T. Dumke. 2000. Ecoregions of Wis-
consin. Trans. Wisc. Acad. Sci. Arts Lett. 88:77–103.
Omonode, R.A., D.R. Smith, A. Gál, and T.J. Vyn. 2011. Soil nitrous oxide emis-
sions in corn following three decades of tillage and rotation treatments. Soil
Sci. Soc. Am. J. 75(1):152–163. doi:10.2136/sssaj2009.0147
Opdyke, M.R., N.E. Ostrom, and P.H. Ostrom. 2009. Evidence for the predomi-
nance of denitrication as a source of N2O in temperate agricultural soils based
on isotopologue measurements. Global Biogeochem. Cycles 23:GB4018.
Osterholz, W.R., C.J. Kucharik, J.L. Hedtcke, and J.L. Posner. 2014. Seasonal ni-
trous oxide and methane uxes from grain- and forage-based production sys-
tems in Wisconsin, USA. J. Environ. ual. 43(6):1833–1843. doi:10.2134/
Ostrom, N.E., R. Sutka, P.H. Ostrom, A.S. Grandy, K.M. Huizinga, H. Gandhi, J.C.
von Fischer, and G.P. Robertson. 2010. Isotopologue data reveal bacterial de-
nitrication as the primary source of N2O during a high ux event following
cultivation of a native temperate grassland. Soil Biol. Biochem. 42(3):499–506.
Parkin, T.B., and T.C. Kaspar. 2006. Nitrous oxide emissions from corn-soybean systems
in the Midwest. J. Environ. ual. 35(4):1496–1506. doi:10.2134/jeq2005.0183
Parton, W.J., M. Hartman, D. Ojima, and D. Schimel. 1998. DAYCENT and its land
surface submodel: Description and testing. Global Planet. Change 19(1-4):35–
Parton, W., E. Holland, S. Del Grosso, M. Hartmann, R. Martine, A. Mosier, D. Ojima,
and D. Schimel. 2001. Generalized model for NOx and N2O emissions from
soils. J. Geophys. 106(15):17403–17419. doi:10.1029/2001JD900101
Posner, J.L., J. Baldock, and J.L. Hedtcke. 2008. Organic and conventional production
Systems in the Wisconsin Integrated Cropping Systems Trials: I. Productivity
1990-2002. Agron. J. 100(2):253–260. doi:10.2134/agrojnl2007.0058
Posner, J.L., M.D. Casler, and J.O. Baldock. 1995. e Wisconsin Integrated Cropping
Systems Trial: Combining agroecology with production agronomy. Am. J. Al-
tern. Agric. 10(03):98–107. doi:10.1017/S0889189300006238
uick Stats Sta. 2016. uick Stats 2.0. quickstats.nass.usda.gov (accessed 15 Mar.
R Core Team . 2015. R: A language and environment for statistical computing. R Foun-
dation for Statistical Computing, Vienna, Austria.
Risk, N., C. Wagner-Riddle, A. Furon, J. Warland, and C. Blodau. 2014. Comparison
of simultaneous soil prole N2O concentration and surface N2O ux measure-
ments overwinter and at spring thaw in an agricultural soil. Soil Sci. Soc. Am. J.
Robertson, G., and P.M. Groman. 2015. Nitrogen transformations. In: E.A. Paul, edi-
tor, Soil microbiology, biochemistry, and ecology. 4th ed. Academic Press, Burl-
ington, MA. p. 421–446.
Sanford, G.R., J.L. Posner, R.D. Jackson, C.J. Kucharik, J.L. Hedtcke, and T.-L. Lin.
2012. Soil carbon lost from Mollisols of the north central U.S.A. with 20 years
of agricultural best management practices. Agric. Ecosyst. Environ. 162:68–76.
Scheer, C., S.J. Del Grosso, W.J. Parton, D.W. Rowlings, and P.R. Grace. 2014. Modeling
nitrous oxide emissions from irrigated agriculture: Testing DayCent with high-
frequency measurements. Ecol. Appl. 24(3):528–538. doi:10.1890/13-0570.1
Shcherbak, I., N. Millar, and G.P. Robertson. 2014. Global metaanalysis of the nonlin-
ear response of soil nitrous oxide (N2O) emissions to fertilizer nitrogen. Proc.
Natl. Acad. Sci. USA 111(25):9199–9204. doi:10.1073/pnas.1322434111
Stehfest, E., and L. Bouwman. 2006. N2O and NO emission from agricultural elds
and soils under natural vegetation: Summarizing available measurement data and
modeling of global annual emissions. Nutr. Cycl. Agroecosyst. 74(3):207–228.
Stevens, R.J., R.J. Laughlin, L.C. Burns, J.R.M. Arah, and R.C. Hood. 1997. Mea-
suring the contributions of nitrication and denitrication to the ux of
nitrous oxide from soil. Soil Biol. Biochem. 29(2):139–151. doi:10.1016/
Tatti, E., C. Goyer, M. Chantigny, S. Wertz, B.J. Zebarth, D.L. Burton, and M. Fil-
ion. 2014. Inuences of over winter conditions on denitrication and nitrous
oxide-producing microorganism abundance and structure in an agricultural soil
amended with dierent nitrogen sources. Agric. Ecosyst. Environ. 183:47–59.
Tiedje, J.M. 1988. Ecology of denitrication and dissimilatory nitrate reduction to am-
monium. In: A.J.B. Zehnder, editor, Environmental microbiology of anaerobes.
John Wiley & Sons, New York. p. 179–244.
Undersander, D., M.H. Hall, P. Vassalotti, and D. Cosgrove. 2011. Alfalfa germination
and growth. Univ. of Wisconsin, Madison.
van Groenigen, J.W., G.J. Kasper, G.L. Velthof, A.V.D.P. Dasselaar, and P.J. Kuik-
man. 2004. Nitrous oxide emissions from silage maize elds under dierent
mineral nitrogen fertilizer and slurry applications. Plant Soil 263:101–111.
Venterea, R.T., M. Bijesh, and M.S. Dolan. 2011. Fertilizer source and tillage eects on
yield-scaled nitrous oxide emissions in a corn cropping system. J. Environ. ual.
Weitz, A., E. Linder, S. Frolking, P. Crill, and M. Keller. 2001. N2O emissions from
humid tropical agricultural soils: Eects of soil moisture, texture and ni-
trogen availability. Soil Biol. Biochem. 33(7-8):1077–1093. doi:10.1016/