JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 20 February 2008
Published in J Environ Qual 37:542-550 (2008)
DOI: 10.2134/jeq2006.0395
© 2008 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dusenbury, M. P.
Right arrow Articles by Wallander, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dusenbury, M. P.
Right arrow Articles by Wallander, R.
Agricola
Right arrow Articles by Dusenbury, M. P.
Right arrow Articles by Wallander, R.
Related Collections
Right arrow Air Pollution
Right arrow Soil Fertility and Productivity
Right arrow Tillage

Nitrous Oxide Emissions from a Northern Great Plains Soil as Influenced by Nitrogen Management and Cropping Systems

M. P. Dusenburya, R. E. Engela,*, P. R. Millera, R. L. Lemkeb and R. Wallandera

a Dep. of Land Resources and Environmental Sciences, Montana State Univ., P.O Box 173120, Bozeman, MT 59717-3120
b Agriculture and Agri-Food Canada, 51 Campus Drive, Saskatoon, SK S7N 5A8

* Corresponding author (rengel{at}montana.edu).

Received for publication September 22, 2006.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary and Conclusions
 REFERENCES
 
Field measurements of N2O emissions from soils are limited for cropping systems in the semiarid northern Great Plains (NGP). The objectives were to develop N2O emission-time profiles for cropping systems in the semiarid NGP, define important periods of loss, determine the impact of best management practices on N2O losses, and estimate direct N fertilizer-induced emissions (FIE). No-till (NT) wheat (Triticum Aestivum L.)-fallow, wheat-wheat, and wheat-pea (Pisum sativum), and conventional till (CT) wheat-fallow, all with three N regimes (200 and 100 kg N ha–1 available N, unfertilized control); plus a perennial grass-alfalfa (Medicago sativa L.) system were sampled over 2 yr using vented chambers. Cumulative 2-yr N2O emissions were modest in contrast to reports from more humid regions. Greatest N2O flux activity occurred following urea-N fertilization (10-wk) and during freeze–thaw cycles. Together these periods comprised up to 84% of the 2-yr total. Nitrification was probably the dominant process responsible for N2O emissions during the post-N fertilization period, while denitrification was more important during freeze–thaw cycles. Cumulative 2-yr N2O-N losses from fertilized regimes were greater for wheat-wheat (1.31 kg N ha–1) than wheat-fallow (CT and NT) (0.48 kg N ha–1), and wheat-pea (0.71 kg N ha–1) due to an additional N fertilization event. Cumulative losses from unfertilized cropping systems were not different from perennial grass-alfalfa (0.28 kg N ha–1). Tillage did not affect N2O losses for the wheat-fallow systems. Mean FIE level was equivalent to 0.26% of applied N, and considerably below the Intergovernmental Panel on Climate Change mean default value (1.25%).

Abbreviations: CT, conventional till • FIE, fertilizer-induced emissions • IPCC, Intergovernmental Panel on Climate Change • NGP, northern Great Plains • NT, no-till • WFPS, water-filled pore space


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results and Discussion
 Summary and Conclusions
 REFERENCES
 
RECENT concern about the build-up of greenhouse gases in the atmosphere has stimulated interest in management practices that sequester carbon in agricultural soils. For dryland cropping systems in the Great Plains, no till (NT) and annual cropping practices have been identified as management practices that promote carbon sequestration (Campbell et al., 2001; Eve et al., 2002). Although adoption of best management practices may provide partial offsets to the atmospheric build-up of CO2, a more complete inventory of other soil-emitted gas is believed to be important to understanding the impact on greenhouse gas mitigation (Mosier et al., 1998; Six et al., 2004). Nitrous oxide is often the gas of greatest interest because it has approximately 300 times the global warming potential of CO2, and agriculture has been identified as its primary anthropogenic source by the Intergovernmental Panel on Climate Change (IPCC, 2001). Inputs of N to agricultural soils from commercial N fertilizer applications, organic manures, biological N2 fixation, and green manures or crop residues have been identified as major contributors to N2O emissions from agriculture (Eichner, 1990).

Published reports on N2O emissions from croplands in the northern Great Plains (NGP), including annual emission losses, seasonal patterns of emissions, and the effect of tillage and cropping sequences are currently very limited. A few reports have come from the Canadian provinces of Alberta and Saskatchewan. Lemke et al. (1999) reported annual N2O emissions ranging up to 4.0 kg N ha–1 yr–1 from dryland cropping systems in the Parkland Region of Alberta, while mean annual N2O emissions were less than 0.5 kg N ha–1 yr–1 from drier regions in Saskatchewan (Lemke et al., 2005). Emissions tended to be similar or lower under NT compared to CT at several sites in Alberta (Lemke et al., 1999; Lemke et al., 2002). In northwest Nebraska, tillage during fallow was found to increase N2O emission by almost 100% (Kessavalou et al., 1998). In contrast, results from more humid regions have generally shown greater emissions under NT than CT systems (Goodroad et al., 1984; Robertson et al., 2000).

The degree to which N fertility influences N2O emissions is also not well documented in the NGP. Regional estimates of fertilizer contributions to N2O emissions in the NGP often utilize IPCC methodology, which assumes that 1.25% ( ± 1.0) of all N inputs, including fertilizer N, will be lost directly as N2O (IPCC, 1997). However, due to the limited data available to provide emission factors (IPCC, 1997), this approach does not account for differences in crop type, soils, or climate. Recent investigations indicate that IPCC methodology may overestimate N2O emissions from fertilizer for the semiarid NGP (Lemke et al., 2005). Further research is needed to determine the impact of best management practices on N2O emissions in the semiarid NGP, and to establish the veracity of the IPCC default value for this region.

Wheat is the dominant crop in the NGP. Historically, wheat-fallow rotations were widely utilized by producers in the semiarid areas of the NGP. More recently, conservation tillage systems, in particular NT, have grown in popularity and facilitated adoption of more intensified and/or diversified cropping systems including annual wheat and wheat-pea rotations. Therefore, the objectives of this study were (i) to develop N2O emission vs. time profiles for four cropping systems and a perennial grass-alfalfa system adapted to the NGP (ii) to define important periods of N2O loss (iii) determine the impact of best management practices (i.e., NT and annual cropping) on N2O losses and (iv) to provide estimates of direct N fertilizer-induced N2O emission (FIE) losses and contrast these with predicted N2O losses using IPCC methodology.


   Materials and Methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results and Discussion
 Summary and Conclusions
 REFERENCES
 
Site Description and Experimental Design
Nitrous oxide emission measurements were conducted over 2 yr (14 Apr. 2004 to 15 Apr. 2006) at the Montana State University– Arthur Post Farm in Bozeman, MT (45°40' N, 111°09' W). The study was initially established in the fall of 2002 as part of a long-term study to examine the impact of cropping systems on soil C levels. The soil at the field site is classified as an Amsterdam silt loam (fine-silty, mixed, superactive, frigid Typic Haplustolls) with 88 g kg–1 sand, 825 g kg–1 silt, 86 g kg–1 clay, pH 7.2, and 9.0 g kg–1 organic carbon in the surface 0.10 m. The soil is characterized as well drained. The study included four cropping systems, including winter wheat-fallow CT (or wheat-fallow CT), winter wheat-fallow NT(or wheat-fallow NT), winter wheat-spring wheat NT (or wheat-wheat NT), and winter wheat-spring pea NT (or wheat-pea NT). In addition, there was an alfalfa (Medicago sativa L.)- perennial grass system consisting of western wheatgrass (Pascopyrum smithii (Rydb.) A. Löve), slender wheatgrass (Elymus trachycaulus (Link) Gould ex Shinners), and green needlegrass (Nassella viridula (Trin.) Barkworth). Cropping systems and perennial grass system main plots (21.3 m x 7.6 m) were replicated four times in a random complete block design. The four cropping system treatments were managed as single-phase rotations, or with only one phase of the system present in a season. Cropping system main plots were divided longitudinally into subplots representing low (unfertilized control), moderate (available N = 100 kg N ha–1), and high N fertility (available N = 200 kg ha–1) regimes beginning in spring 2004 to create a split-plot design.

Fertilization Practices, Soil Sampling and Analysis, and Seeding
Fertilizer N (urea) was applied to cropping systems subplots during phases of the system where spring or winter wheat was grown to produce targeted levels of available N in the moderate and high regime, or 100 and 200 kg N ha–1, respectively. The available N pool was estimated from the sum of soil NO3–N (0–60 cm) + fertilizer N for wheat-fallow (CT and NT), and wheat-wheat, and soil NO3–N (0–60 cm) + fertilizer N + pulse credit for wheat-pea. The pulse credit was assessed at 20 kg N ha–1 following recent work by Miller et al. (2006). The 200 kg ha–1 rate for the high available N regime was established based on soil fertility guidelines for wheat in the state of Montana. These recommendations indicate approximately 50 kg available N is required to produce 1 Mg of grain. Given the yield potential at this site was 4.0 Mg ha–1, the available N requirement became 200 kg N ha–1.

Soil NO3–N levels (0–0.60 m) for defining N applications in spring 2004 were based on samples collected on March 20. At that time, the study was in the winter wheat phase following a 1 Sept. 2003 seeding. Hence, the resulting urea-N surface broadcast application (13 Apr. 2004) was made to an established stand of winter wheat. Soil NO3–N levels for defining N applications in the spring 2005 (wheat-wheat NT only) and fall 2005 (all systems) were based on samples collected approximately 2 wk before seeding (14 Apr. 2005 and 30 Sept. 2005). During these two seasons, urea-N was applied on the seeding dates in a band 25 mm below and 50 mm to the side of the seed-row.

Soil sampling for NO3–N analysis and defining fertilizer N levels was conducted by collecting two composite cores for each affected cropping system x N regime subplot. The cores were divided into 0.0–0.15, 0.15–0.30, and 0.30–0.60 m depth increments, then dried (50°C) and ground ( < 2 mm) before chemical analysis. Nitrate-N analysis was performed using a flow-injection analyzer (Lachat, Milwaukee, WI) following extraction with KCl (Mulvaney, 1996). Soil NO3–N levels were averaged for each cropping system x N regime subplot. For this reason, the differential in fertilizer N rate applied to the moderate and high N rates within a cropping system was not always equal to 100 kg N ha–1; and the fertilizer N rates applied to high N wheat-pea NT were greater than applied to the high N wheat-wheat NT in spring 2004 and fall 2005.

Wheat and peas seeding operations were performed with direct-seeding equipment (0.25-m row width) equipped with disc-openers for minimal disturbance. Conventional tillage was performed with discs and cultivators equipped with sweeps. Harvest operations were performed with small-plot combines, and subsamples of grain were analyzed for total N via an automated dry combustion analyzer (Leco Corporation, St. Joseph, MI).

Gas Sample Collection and Analysis
Gas sampling was conducted in all cropping systems using vented chamber techniques (Hutchinson and Mosier, 1981). Chambers (0.50-m long, 0.20-m wide and 0.15-m high) were positioned between crop rows and driven 0.05 m into the soil to produce a 10 L headspace. Chambers were placed in all subplots (one chamber per plot). Chambers were left uncovered except during the periods when gas samples were being collected. Sample collection and analysis followed the protocols of Lemke et al. (1998). Gas samples were collected from the headspace during the early to mid-afternoon (1300–1500 h) after 1 h. Samples were drawn from the headspace using a 25-mL syringe and then transferred to pre-evacuated 13 mL Exetainers (Labco International Inc., Houston, TX 77210–4346). The concentration of N2O in the sample exetainer was determined using a gas chromatograph (GC) equipped with a 63Ni electron capture detector. A Varian 3400 (Varian Inc., Palo Alto, CA) GC configured with a 30-m (0.32-mm i.d.) Carbon Plot capillary column (Agilent technologies, Santa Clara, CA) was used during the first year. The carrier gas was Ar/CH4 (95:5) with a flow rate of 4.3 mL min–1 through the column via a split-flow injector. The injector, column, and ECD temperatures were 30°C, 200°C, and 360°C, respectively. A Varian 3800 GC with a CombiPal autosampler, was used during the second year of the study. This system was configured with a 1-mL sample loop, 0.5-m long (3.2-mm i.d.) Hayesep N (80/100) backflush column (VICI Valco Instruments Corp. Houston, TX), and 2-m (3.2-mm i.d.) Hayesep D (80/100) analytical column. The carrier gas was Ar/CH4 (95:5) with a flow rate of 60 mL min–1 through the column. The injector, column oven, and ECD temperatures were 40°C, 120°C, and 360°C, respectively.

Nitrous oxide flux was estimated from the concentration change in the chamber headspace over the 60-min collection period. Changes in concentration over time were assumed to be linear and were calculated by subtracting the time-zero (or background) concentration from the final concentration. Time-zero concentrations were calculated from a series of ambient air samples collected during each sampling event. The mean of these samples was used as the time-zero concentration (Izaurralde et al., 2004). Before doing these analyses a number of multiple time-point sampling tests were performed at this study site, and similar sites, to validate the linearity of N2O flux over 1 h. The results showed flux vs. time relationships to be linear in most instances ( > 80%). Given that N2O concentration did not always change linearly with time, the flux estimates reported in this study may have slightly (probably < 10%) underestimated fluxes compared to a multiple time-point sampling scheme.

Sampling was initiated in April 2004 immediately following the thaw (early April) and continued year-round. Samples were collected approximately twice weekly during the early spring and following N fertilization events when the potential for large N2O emissions was great due to high soil-water contents and available N substrate (fertilizer, mineralized N). Sampling frequency was reduced (e.g., once weekly) during the summer and/or during period where emission activity was low, or near background levels. Nitrous oxide emissions are episodic in nature and abrupt changes in weather (e.g., periods of high rainfall), tillage events, and N applications can be followed by periods of high emissions. Therefore, the sampling schedule and frequency was augmented to meet the objective of capturing and recording the most important periods of N2O emissions. A total of 120 sampling events were captured over the course of this 2-yr investigation.

Ancillary Variables
Air temperatures outside and inside the chambers and below the soil surface (20- mm depth) was recorded continuously with temperature sensors and data-loggers. Soil cores (0–80 mm) were collected at all gas sampling events to determine water content, except on a few dates during the winter when the ground was frozen and we were unable to penetrate the surface. Gravimetric water content was determined, and then multiplied by bulk density to determine volumetric soil water. The ratio of volumetric soil water to total porosity was used to calculate soil water-filled pore space (WFPS). Total porosity was determined as (1– Db/Ds) where Db is the soil bulk density (Mg m–3) and Ds is soil particle density assumed to be 2.65 Mg m–3. Surface soil samples (0–0.15 m) were collected periodically during the spring of 2005 and analyzed for NO3–N as described previously.

Data Analysis and Interpretation
The arithmetic mean of N2O concentrations in the gas samples across the four replicates was used to estimate emission fluxes for the individual sampling dates. Nitrous oxide flux vs. time profiles were then developed for each cropping system x N regime.

Fertilizer was not applied to the moderate N regime within the wheat-fallow systems (CT and NT) in the spring of 2004 as soil NO3–N levels were > 100 kg N ha–1. Hence, the wheat-fallow main plots contained only two subplots (high N regime, and low or unfertilized regime) for the initial 18 mo of this study. Moderate and low N regimes were not differentiated until 30 Sept. 2005, following N fertilization. Gas emissions for the low and moderate N regimes over the initial 18 mo of this investigation (15 Apr. 2004– 19 Sept. 2005) were equivalent and based on measurement from one set of four chambers.

Nitrous oxide losses were calculated using the approach of Bronson and Mosier (1993). Mid-afternoon N2O fluxes were assumed to be constant for the 24-h periods. Cumulative fluxes for each subplot, or experimental unit, were then determined by linearly interpolating data points between each successive sampling event and integrating the underlying area. Direct N fertilizer induced emissions (FIE) were estimated using the approach of Eichner (1990), or by subtracting the emissions of the control (or unfertilized low N regime) from the fertilized areas (moderate and high regimes). This was done within each of the main plots. The net value was expressed as a percentage, or fraction, of N applied to calculate the fertilizer loss coefficient. Fertilizer-induced N2O emissions were not corrected for estimated losses of NH3. Ammonia losses were assumed to be small as a subsurface band application was used in 2005, and rainfall after the surface broadcast application in 2004 was believed sufficient to have allowed for fertilizer urea-N incorporation.

Analysis of variance (ANOVA) of N2O emission data for the four cropping systems was performed using the PROC Mixed procedure (SAS Institute, 1988). Block and block x cropping system factors were treated as random variables. Contrasts were performed using the pdiff option in SAS. Differences with a P < 0.05 were considered significant. An exploratory analysis of the N2O emission results was performed to determine the normality of data sets. The Proc Univariate Normal procedure of Statistical Analysis System Institute (SAS Institute, 1988) was used in this analysis to examine the normality of the residuals. Data sets of cumulative N2O emissions found to not pass the Shapiro-Wilk (W ≤ 0.05) normality test were log10 transformed to correct this problem. An ANOVA was then performed on the transformed data, but tabulated data are shown in the original units (untransformed) to aid interpretation.


   Results and Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
Summary and Conclusions
 REFERENCES
 
Cumulative precipitation over the 2 yr of this investigation (15 Apr. 2004 to 14 Apr. 2006) was 823 mm, or similar to the long-term average (Fig. 1 ). April, May, and June typically comprise the wettest 3-mo period at this location. Precipitation amounts for this period equaled 161 mm in 2004 and 165 mm in 2005, or approximately 40% of the annual total and consistent with historical averages. Monthly precipitation amounts in 2004 were near normal with the exception of the dry months of November and December. In 2005, October and November were characterized by well above normal precipitation, while precipitation in May was much below normal.


Figure 1
View larger version (28K):
[in this window]
[in a new window]

  		Fig. 1. Cumulative monthly precipitation for the 2-yr study and 40-yr average. Year totals presented in parentheses.

 
Perennial Grass–Alfalfa Emissions
Nitrous oxide flux from the perennial grass-alfalfa system was established as background, or control levels, as it was unfertilized and no biomass was removed. Also, they were most representative of undisturbed grassland conditions or conservation reserve program (CRP) even with the presence of alfalfa. Emissions flux vs. time profiles (Fig. 2 ) indicate losses were generally < 5 µg N2O-N m–2 h–1 with the exception of thaw events during the winter and early spring. The higher N2O fluxes during these periods were likely a result of denitrification as soil WFPS equaled 100%. Emission fluxes averaged 1.4 ± 0.6 ( ± 2{sigma}) µg m–2 h–1 for the 2 yr excluding the thaw events. Emissions above this interval (i.e., 2.0 µg m–2 h–1) were considered as elevated emission rates above background. Cumulative N2O emission losses from this perennial grass-alfalfa system were equivalent to 0.14 kg N ha–1 yr–1. Nitrous oxide emissions from CRP have been reported at 0.50 kg N ha–1 in Colorado with the greatest emission activity occurring during winter denitrification events (Mosier et al., 1997). Hence, the cumulative losses we observed were somewhat lower than reported in Colorado, while the pattern of emission activity was similar.


Figure 2
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 2. (a.) Nitrous oxide flux vs. time profile and for a dryland perennial grass-alfalfa system, and (b) soil water-filled pore space (WFPS) (± S.E.) and soil temperature 20 mm below surface vs. time.

 
Cropping Systems Emissions Profiles
Nitrous oxide emissions for the four cropping systems (Fig. 3–6GoGoGo ) were episodic and responsive to periods of high soil WFPS and availability of N substrate (soil or fertilizer) over the 2 yr of the investigation. Nitrous oxide flux varied considerably over the subplot (cropping system x N regime) replicates and CV's exceeded 100% on approximately half of the sampling dates. Although the N2O flux vs. time profiles differed for the cropping systems and N regimes, there were several commonalities evident over the 2 yr of this study. First, N2O fluxes were elevated above background following N fertilization and during freeze–thaw cycles in the winter and early spring. Second, the combined winter-spring thaw cycles and 10-wk post-fertilization periods accounted for the majority of emissions. Third, all cropping systems exhibited significant periods of low, near background N2O emission activity equivalent to fluxes observed from the perennial grass-alfalfa system.


Figure 3
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 3. (a) Nitrous oxide flux vs. time profile and for winter wheat–fallow under conventional tillage and as affected by N fertility (only low and high N regimes presented for clarity) and (b) soil water-filled pore space (WFPS) (mean of N levels ± S.E.) and soil temperature 20 mm below surface vs. time. Downward facing arrows and text indicate date, rate (kg N ha–1), and method of N fertilization. Upward facing arrows indicate tillage events.

 

Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 4. (a) Nitrous oxide flux vs. time profile and for winter wheat–fallow under no till and as affected by N fertility (only low and high N regimes presented for clarity, and (b) soil water-filled pore space (WFPS) (mean of N levels ± S.E.) and soil temperature 20 mm below surface vs. time. Downward facing arrows and text indicate date, rate (kg N ha–1), and method of N fertilization.

 

Figure 5
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 5. (a) Nitrous oxide flux vs. time profile and for wheat–wheat under no till and as affected by N fertility (only low and high N regimes presented for clarity), and (b) soil water-filled pore space (WFPS) (mean of N levels ± S.E.) and soil temperature 20 mm below surface vs. time. Downward facing arrows and text indicate date, rate (kg N ha–1), and method of N fertilization.

 

Figure 6
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 6. (a) Nitrous oxide flux vs. time profile and for wheat–pea under no till and as affected by N fertility (only low and high N regimes presented for clarity), and (b) soil water-filled pore space (WFPS) (mean of N levels ± S.E.) and soil temperature 20 mm below surface vs. time. Downward facing arrows and text indicate date, rate (kg N ha–1), and method of N fertilization.

 
Post-Fertilization Activity
Nitrogen fertilization was perhaps the single most important event that stimulated an elevation in N2O flux. The elevation in emission activity occurred within a week following fertilization, and peaked after approximately 2–4 wk. The duration of elevated flux above background ( > 2.0 µg N2O-N m2 h–1) for spring applications in 2004 (April 15) and 2005 (April 14) was approximately 10 wk, but extended somewhat longer than this for the fall application in 2005 (September 30). Overall, emission losses over the 10-wk post-N fertilization period increased with N regime (i.e., low N < moderate N < high N), but within a specific year, cropping system, or year x cropping system the magnitude of enhanced emission activity varied considerably. An example of temporal or season-to-season variability is provided by the results from the wheat-wheat NT system (Table 1 ). Nitrous oxide losses during the 10-wk post-fertilization were much lower (P < 0.01) in spring 2004 compared to 2005 (spring and fall) for the wheat-wheat NT system, even after consideration was given to the fertilizer N rate applied. The low emissions in spring 2004 may have resulted from better synchrony of the fertilizer N applications with wheat nutrient demand. Nitrogen fertilizer was broadcast applied to an established stand of winter wheat in April 2004, while in April and September 2005 fertilizer was band applied at seeding to spring and winter wheat, respectively. Previous research has shown spring applied N fertilizer was more available to winter wheat than fall applications (Vaughan et al., 1990), and that N2O emissions from N fertilizer could be minimized by applying N fertilizer in the spring to winter wheat (Hao et al., 2001). Drier soil moisture conditions in spring 2004 (average soil WFPS at sampling events = 48%) vs. spring 2005 (soil WFPS = 59%) and fall 2005 (soil WFPS = 69%) may have further contributed to the reduced N2O emissions from this period.


View this table:
[in this window]
[in a new window]

  		Table 1. Cumulative N2O–N emission losses observed over a 10-wk period following N fertilization of no-till winter wheat–spring wheat.

 
The majority of N2O losses during the 10-wk period following fertilization were probably a result of nitrification, except perhaps in the fall 2005. Soil WFPS rarely exceeded 70% in the spring following N fertilization. Previous research has indicated that aerobic microbial activity, such as nitrification peaks at WFPS = 60%, while anaerobic processes, such as denitrification, become dominant at WFPS > 80% (Linn and Doran, 1984). Denitrification may have played an important role in N2O emissions following fertilization during fall 2005. The cool soil temperatures (3.8°C at 20-mm depth) and above normal precipitation (127.2 mm, or 211% of the 40-yr average) resulted in soil WFPS conditions which were frequently > 70%. Further, these conditions may explain, at least in part, the somewhat prolonged period of elevated emission activity beyond 10 wk.

Freeze–Thaw Cycles
Some of the highest single event N2O fluxes occurred during freeze–thaw cycles in February 2005 (Cycle 1), December 2005 (Cycle 2), and March 2006 (Cycle 3). Emissions followed a rise in air temperature above 0°C, which triggered snowmelt and resulted in saturated conditions near the soil surface. During these periods (3 to 8 wks), substantial N2O emissions were observed even though soil temperatures 20 mm below the surface were < 0°C. Emissions remained elevated during spring thaw events (Cycles 1 and 3) until soil temperature increased to > 0°C, sub-soils thawed, and soil moisture drained or evaporated. Emissions were particularly high during Cycle 3 of the fertilized wheat-pea system. Although, N2O fluxes during the growth cycle of peas (14 Apr. to 23 July 2005) was low (discussed below), decomposition of their N-rich residues during the wet fall of 2005 may have resulted in more available N substrate for denitrification. This is consistent with Kaiser et al. (1998), who observed winter and spring thaw emissions were greatest in soils amended with crop residues with narrow C to N ratios.

Losses during the freeze–thaw cycles were likely attributed to denitrification as soil WFPS during the winter and early spring was frequently > 85% (Linn and Doran, 1984). The freeze-thaw action may have further facilitated denitrification due to the sudden release of carbon from frost-killed soil organisms and disintegrating soil aggregates (Christensen and Tiedje, 1990), and release of N2O trapped beneath the frozen soil layer (Burton and Beauchamp, 1994). Also, very high N2O fluxes from soils have been observed at temperatures near freezing due to inhibition of nitrous oxide reductase (Holtan-Hartwig et al., 2002) leading to a wide N2O/N2 production ratio.

Periods of Low Activity
Low, or background, fluxes were defined as those within two standard deviations of the perennial grass-alfalfa system mean (1.4 ± 0.6 µg m–2 h–1), or < 2.0 µg m–2 h–1. Low emission activity was observed in the summer (all cropping systems), fallow phases of the wheat-fallow (CT and NT) systems, and pulse phase of the wheat-pea system. In addition, emissions were low in the unfertilized cropping systems year-round, except for the freeze–thaw cycles. The reasons for periods of low activity can be traced to the extensive time the soil surface remained at < 50% WFPS, and/or an absence of available N substrate for nitrification.

Soil WFPS averaged 37% and 39% for the summer of 2004 and 2005, respectively. Even after significant rainfall ( > 50 mm) events WFPS returned to < 60% after only 48 h due to the excellent drainage characteristics of this field site. Emissions during the fallow phases (wheat-fallow systems) and pulse phases (wheat-pea system) were minimal, probably due to low rates of soil N mineralization and minimal rates of N2O production during N fixation, respectively. Our results were consistent with Lemke et al. (2002) who found pulse crops contributed minimally to cumulative N2O emissions. Substrate was also limiting in unfertilized cropping systems. As an example, soil NO3–N levels (0–0.15 m depth) in the unfertilized wheat-wheat system were ≤ 3.0 mg kg–1 on all dates sampled (1 Apr. 2005, 24 May 2005, 17 June 2005, 24 June 2005, 8 Sept. 2005, and 14 Nov. 2005, and 14 Apr. 2006). Dobbie et al. (1999) noted that emissions of N2O are very much reduced when soil NO3–N is < 5 mg kg–1.

Cumulative Nitrous Oxide Emissions (2 yr)
Cumulative N2O emissions over 2 yr were significantly (P < 0.05) affected by N regime, cropping systems, and cropping systems x N regime (Table 2 ). The combined 10-wk post-N fertilization period plus freeze–thaw cycles accounted for 56, 64, 84, and 78% of the emissions for the wheat-fallow CT, wheat-fallow NT, wheat-wheat NT, and wheat-pea NT systems N fertilized systems (moderate and high N average), respectively, illustrating the importance of these two periods to total emission losses. Nitrous oxide emissions were similar across the unfertilized wheat-fallow, wheat-wheat, wheat-pea systems, and losses averaged 0.15 kg N2O-N ha–1 yr–1, or nearly identical to the losses for the perennial grass-alfalfa system (0.14 kg N2O-N ha–1 yr–1). Cumulative 2-yr emissions in fertilized systems (moderate and high N) were greatest in wheat-wheat NT system, followed by wheat-pea NT, and then wheat-fallow systems (NT and CT). Higher emissions under wheat-wheat NT can be attributed in great part to the considerable emission activity observed following N fertilization in spring 2005. In contrast, urea-N was not applied during the second phase of the wheat-fallow CT NT and wheat-pea NT systems, and emission losses for these systems were considerable lower. Cumulative 2-yr emissions for fertilized wheat-pea NT were somewhat higher than wheat-fallow CT NT because of the considerable activity observed during the spring thaw of 2006 (freeze-thaw Cycle 3).


View this table:
[in this window]
[in a new window]

  		Table 2. Cumulative 2-yr N2O–N losses, fertilizer induced emissions (FIE), fraction of emissions derived from fertilizer, and fraction of fertilizer N lost as N2O for four cropping systems at three N fertility regimes.

 
Cumulative 2-yr N2O emissions were not different between CT and NT wheat-fallow cropping systems. Tillage only affected emission activity for brief periods (1–2 sampling events) following disturbance. The comparatively dry climate and excellent drainage characteristics at this field site resulted in similar surface moisture contents between wheat-fallow NT and CT for much of the year, thus minimizing the effect of tillage on N2O losses. Overall, cumulative 2-yr N2O emissions (0.21 to 1.31 kg N2O-N ha–1 or 0.10 to 0.65 kg N2O-N ha–1 yr–1) were modest in contrast to those reported for more humid regions such as eastern Canada (Gregorich et al., 2005) and the midwestern USA (Cates and Keeney, 1987; Venterea et al., 2005). Our results were within the range reported for other semiarid regions (i.e., 0.18–0.77 kg N ha–1 yr–1) by Mosier et al. (2003), Lemke et al. (1999), and Kessavalou et al. (1998) in Colorado, Alberta, and Nebraska, respectively.

Fertilizer-Induced Emissions
Fertilizer induced emissions increased with N regime (Table 2), were correlated with N rate (r = 0.74 P < 0.01) (Table 2), and comprised a large fraction of the 2-yr total. Fertilizer induced emissions were also affected by cropping system. The highest FIE occurred in the wheat-wheat system as a result of an additional fertilizer event applied to this rotation in spring 2005. The significance of the cropping system x N regime interaction was a result of the small difference in emissions observed between the moderate and high N regime for wheat-wheat NT compared to the other systems. As noted earlier, there was considerable variability across the replicates in this study. Mean emissions for the wheat-wheat NT- moderate N regime were inordinately influenced by high fluxes observed in 1 subplot during the spring of 2005 (or 10-wk post N fertilization period). Emissions from this one subplot were equivalent to 1.11 kg N2O-N ha–1, or 5-times greater than the mean of the other replicates (0.22 kg N2O-N ha–1) for this period. If observations from this one chamber were removed from our analysis, the 2-yr total emissions and FIE would have been equivalent to 0.86 and 0.56 kg N2O-N ha–1, respectively.

Fertilizer Nitrogen Loss Coefficient and Intergovernmental Panel on Climate Change Default Value
The fraction of fertilizer lost as N2O (or fertilizer loss coefficients) ranged from 0.08 to 0.45% (Table 2). The largest coefficients were observed in the wheat-wheat system. Fertilizer loss coefficients were similar among wheat-fallow (CT and NT) and wheat-pea. Overall, our results show a very modest rate of emission losses from fertilizer N compared to more humid regions (Eichner, 1990), but similar to results from other semiarid regions in Saskatchewan (Lemke et al., 2005) and inner Mongolia (Li et al., 2001). Due to the comparatively dry climate of the NGP, emission losses are confined to a very few specific periods, i.e., post-N fertilization and freeze–thaw cycles in the winter-early spring. The remainder of the year, emissions were near background levels similar to what might be found from grasslands. The authors suggest a lower, more conservative coefficient of 0.23% ± 0.05 (or ± 1 SE) provides a more realistic estimate of N fertilizer induced losses for this region. This coefficient is based on a mean of the four cropping systems. Although, coefficients of fertilizer N loss were significantly higher for the wheat-wheat system, there was considerable spatial and temporal variability associated with our chamber-based flux measurements. This precludes us from having sufficient confidence to suggest a separate coefficient for the wheat-wheat NT system at this time.

Calculated percentages, or fractions, of fertilizer N lost as N2O are considerably below the IPCC mean default value, or at the lower end of the IPCC range (i.e., 0.25 to 2.25%). Fertilizer induced emission losses in Table 2, and the associated loss coefficients were not adjusted for direct losses of fertilizer N as NH3(g). The IPCC assumes these losses to be equivalent to 10% of N fertilizer inputs. Hence, the fertilizer loss coefficients presented need to be adjusted upward, ÷ 0.9, to contrast directly with the IPCC default value. Accordingly our results indicate that mean IPCC default value will overestimate fertilizer-induced losses of N2O by a factor of 4.7 if we adjust our reported coefficient to 0.26% (0.23%/0.9).

Dinitrogen Oxide–Nitrogen Loss–Grain Nitrogen Harvest Relations
Contrasting the effect of management practices on N2O losses in the context of greenhouse gas accounting is important, but comparisons should also be made from the viewpoint of N2O intensity. Expressing emissions in terms of intensity allows them to be assessed in the broader context of production. One way to do this for cropping systems with dissimilar plant species is to contrast the amount of N2O loss to the amount of N harvested (Izaurralde et al., 2004). In this study N2O intensities are lower for the wheat-fallow (NT and CT) and wheat-pea NT vs. the wheat-wheat NT system (Table 3 ). Although, total grain yield was highest from the annually cropped wheat systems and yields were improved by N fertility, the increase in yield was not in proportion to the increase in N2O emissions. Thus, inclusion of a pulse crop may represent an opportunity to intensify cropping systems in the NGP (a recommended best management practice) without significantly increasing N2O intensity (as compared to wheat-fallow systems).


View this table:
[in this window]
[in a new window]

  		Table 3. Cumulative N2O–N (14 Apr. 2004 to 30 Sept. 2005) loss expressed as a percentage of grain N removed for four cropping systems at two N fertility levels.

 

   Summary and Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary and Conclusions
REFERENCES
 
Cropping system N2O flux vs. time profiles developed in this semiarid climate indicated the majority of flux activity occurred over a 10-wk period following N fertilization and during freeze–thaw cycles. Nitrous oxide fluxes were near perennial grass-alfalfa background levels (2.0 µg m–2 h–1) during the summer (all cropping systems), and fallow phases of the wheat-fallow (CT and NT) systems, and pulse phases of the wheat-pea system. Cumulative N2O losses were modest, ranging from 0.10 to 0.65 kg N2O-N ha–1 yr–1. No-till did not result in elevated emission losses over CT for the wheat-fallow system. In semiarid climates, adoption of wheat-fallow NT to replace wheat-fallow CT for C sequestration would appear to have, at least in the short-term, a minimum impact on emissions. Intensifying cropping by moving to continuous wheat, or cereal grain, systems may result in somewhat greater N2O emissions if annual inputs of fertilizer N are required to maintain productivity. Adopting a cereal-pulse system would be preferred under these circumstances since overall N2O emissions and N2O intensity were comparable to the wheat-fallow systems.

The estimated fraction of fertilizer lost as N2O was considerably below the IPCC 1.25% default value mean. A revised coefficient of 0.23% ± 0.05 (or ± 1 SE) is proposed, or 0.26% if a 10% loss of fertilizer N as NH3(g) is assumed. Although, FIE appear to be modest in semiarid climates, fertilizer application strategies that minimize residual soil NO3–N during the periods of highest emissions (i.e., 10 wk post-fertilization and freeze–thaw cycles) may further reduce overall N2O losses. For example, spring top-dress N applications to winter wheat may be preferable to band applications at seeding (provided NH3(g) losses are minimal) as result of the better synchrony of N application with crop demand and minimization of residual soil NO3–N during the winter and early spring months.


   ACKNOWLEDGMENTS
 
The authors wish to express their thanks to the USDA- National Research Competitive Grants Program for supporting our work on trace gas emissions from soils (Air quality program: 2004-35112-14233; Soil and soil biology program: 2004-35107-14951).


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary and Conclusions
 REFERENCES
 
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.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary and Conclusions
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dusenbury, M. P.
Right arrow Articles by Wallander, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dusenbury, M. P.
Right arrow Articles by Wallander, R.
Agricola
Right arrow Articles by Dusenbury, M. P.
Right arrow Articles by Wallander, R.
Related Collections
Right arrow Air Pollution
Right arrow Soil Fertility and Productivity
Right arrow Tillage

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome