Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Changes in the global climate system have been linked to anthropogenic emissions of greenhouse gases (GHG) (IPCC 2007). In 2008, 8.4% of the total GHG emissions reported in Canada resulted from agricultural activities (Environment 2010). Nitrous oxide (N2O) from agricultural soils accounted for 50% of the anthropogenic GHG emissions from Canadian agroecosystems. In addition to being a greenhouse gas, N2O is also responsible for stratospheric ozone depletion (Crutzen and Ehhalt 1977). Nitrogen fertilizer, the major source of agricultural N2O emissions, has been instrumental in increasing yield from arable cropping systems over the past 50 years (Matson et al. 1997; Tillman et al. 2002). Potato is the dominant arable crop in Atlantic Canada. Large fertilizer N inputs, frequently 200 kg N  ha−1 or more, are applied to the potato crop to achieve the tuber yield and size distribution demanded by the potato processing industry (Zebarth et al. 2003; Zebarth and Rosen 2007). These large N inputs raise concerns as to the potential for increased N2O emissions (Zebarth et al. 2009a).

The potato production system is an interesting one from the perspective of N2O emissions. As with other ridge/furrow management production systems, soil management practices cause distinct distributions of soil physical, chemical and biological characteristics over relatively short distances. The potato ridge, formed as a result of tillage, provides a low bulk density, well-aerated environment suitable for the formation of tubers (Fig. 14.1). In Atlantic Canada, the ridge is commonly formed when the plants are 15–30 cm tall, but there is a trend towards earlier ridge formation that can occur anytime from planting to emergence. The majority of the fertilizer application in Atlantic Canada is as a band application at planting placed adjacent to the seedpieces in the potato ridge. The potato furrows by contrast have a greater soil bulk density and therefore less aerobic environment where carbon and nitrate (NO −3 ) supplies are reduced relative to the ridge. In Atlantic Canada the majority of the A horizon, the horizon in which organic carbon accumulates, is used to form the ridge. The furrow consists of a thin layer of A horizon (maximum of 5 cm) that remains after tillage overlying a B horizon or in some cases C horizon. In Atlantic Canada the soils tend to be rather shallow soil with low pH and frequently have compact subsurface horizons making the distinction between ridge and furrow even more dramatic. The furrow is also a region of water accumulation as a result of the runoff of precipitation. This substantial variation in soil physical and chemical properties over a short distance has the potential to result in markedly different soil N processes and microbial communities. In this chapter we will use the results of studies we have conducted in rain-fed potato production in Atlantic Canada to examine the controls on N2O emissions from potato production systems and strategies for managing these emissions.

Fig. 14.1
figure 1

The ridge-forming operation results in the formation of a low bulk density ridge that contains higher NO −3 due to banded fertilizer application, and higher organic carbon concentrations because most of the C-rich surface horizon is located in the ridge, compared with the furrow. The hilling operation can occur at different times, but is commonly done in eastern Canada when the plants are 15–30 cm high

Full size image

2 Factors Controlling N2O Emissions

The biological production of N2O in agricultural soils occurs primarily as a result of the processes of nitrification and denitrification (Firestone and Davidson 1989; Mosier 1998). The conditions that favor these processes are very different. Nitrification, the oxidation of ammonium (NH +4 ) to NO −3 , occurs as a result of the activity of autotrophic organisms in aerobic soils. Denitrification is a respiratory process whereby heterotrophic organisms respire oxides of N in the absence of O2, producing N2 as the final product. As a result, denitrification generally occurs in moist soils where O2 supply is reduced. For this, an energy source, primarily C, is required by the microorganisms for the process to proceed. In humid climates such as those found in Atlantic CanadaFootnote 1, denitrification is the predominant source of N2O (Firestone and Davidson 1989; Granli and Bøckman 1994; Mosier 1998) and will be the primary focus of our discussion here. Since denitrification is a respiratory process the primary factors that control denitrification relate to the supply of preferred terminal electron acceptor (O2 availability or aeration), the demand for terminal electron acceptors (carbon availability) and the relative supply of alternate terminal electron acceptors, primarily NO −3 availability. Other factors (e.g., temperature, pH, texture) also influence the denitrification process (Firestone and Davidson 1989; Granli and Bøckman 1994) but are less impacted by management practices in the potato production system.

2.1 Aeration

Denitrification commonly occurs at water-filled pore space (WFPS) of 0.70 m3 m−3 and greater (Bateman and Baggs 2005). The rate of gaseous diffusion through the soil depends on water content and bulk density (Ball et al. 1999, 2008). Soils with greater bulk density have smaller pore volume. Similarly, greater soil water content reduces air-filled porosity. Both factors result in reduced and less continuous air-filled porosity, decreasing the rate of O2 movement through the soil (Focht 1992). Increasing the percentage of WFPS has been shown to increase denitrification and N2O emissions (Ruser et al. 1998, 2006; Ball et al. 1999, 2008; Flessa et al. 2002).

In C-rich environments, increased microbial activity can also result in reduced availability of O2. For example, Gillam et al. (2008) found that even in relatively well-aerated soils (WFPS 0.60 m3 m−3), increased C availability can substantially increase N2O emissions in soils through its influence on O2 consumption. In comparison, N2O emissions were negligible at WFPS below 0.45 m3 m−3 due to enhanced aeration, or above 0.75 m3 m−3 because restricted gaseous diffusion limits the diffusion of N2O away from the site of denitrification thereby favoring the further reduction of N2O to N2, and consequently C availability has a more limited effect on N2O emissions under these conditions.

The WFPS varies in ridges and furrows. It is consistently greater in the furrow than in the ridge (Burton et al. 2008a; Dandie et al. 2008; Haile-Mariam et al. 2008). This difference is due in large part to differences in soil bulk density. For example, soil bulk density for 0–15 cm depth in the ridge and furrow were commonly 1.0 and 1.2 Mg m−3, respectively, in the study of Burton et al. (2008a). Greater WFPS in the furrow can also reflect the redistribution of water after rainfall events, and the greater loss of water from the ridge due to transpiration (Zebarth and Milburn 2003). The greater total pore volume and preponderance of large pores in the ridge results in a soil in which the pores are largely filled with air. The presence of continuous air-filled porosity greatly enhances the movement of O2 into the soil (Focht 1992). The enhanced supply of O2 to the microbial community suppresses anaerobic processes such as denitrification (Robertson and Tiedje 1987; Focht 1992; Granli and Bøckman 1994). This is consistent with reports of increased N2O emissions in furrows that were compacted as a result of wheel traffic (Ruser et al. 1998; Thomas et al. 2004) and with increased denitrification in furrows compared with ridges (Dandie et al. 2008).

2.2 Carbon Availability

Denitrification is a respiratory process where nitrogen oxides (NO −3 , NO −2 , NO and N2O) are utilized as terminal electron acceptors by microorganisms in the absence of oxygen. Organic C substrates serve as a source of energy for heterotrophs in this process (Beauchamp et al. 1989). Carbon also has an indirect influence on denitrification by increasing the consumption of O2 during aerobic respiration stimulating the formation of anoxic sites within the soil (Smith 1980; Hojberg et al. 1994). Sources of soil C inputs include soil organic matter, the potato rhizosphere, residues from the preceding crop, senescing leaves and roots from the current crop, and organic amendments added to the soil. In most cases soil management results in all of these sources being concentrated in the potato ridge, either as a result of ridge formation or as a result of the location of the potato plant. The greater concentration of the organic carbon in the potato ridge results in Greater microbial activity in these locations and greater N2O and CO2 emissions (Fig. 14.2; Haile-Mariam et al. 2008).

Fig. 14.2
figure 2

Synchrony of nitrous oxide (open symbols; g N2O-N ha−1 d−1) and carbon dioxide (closed symbols; kg CO2-C ha−1 d−1) emissions in the 2002 and 2003 growing seasons from a potato crop in New Brunswick (Burton et al. 2008a)

Full size image

Management and climatic conditions influence the quantity, quality and availability of organic carbon compounds. Management may directly influence carbon availability as a result of carbon addition (manure, compost or crop residues), altered quality of the organic substrates (Zebarth et al. 2009b, c) or the introduction of a plant (Sey et al. 2010). Management may also indirectly influence carbon availability as a result of soil disturbance such as tillage, which results in a flush of carbon metabolism (Alvarez et al. 2001), or as a result of plant desiccation with herbicide (Tenuta and Beauchamp 1996; Flessa et al. 2002). Climatic events such as the rewetting of dry soil following precipitation (Birch 1958) or the thawing of a frozen soil also result in a flush of microbial activity (Christensen and Tiedje 1990) and N2O emissions (Burton and Beauchamp 1994).

Carbon availability is one of the primary controllers of N2O emissions (Firestone and Davidson 1989; Granli and Bøckman 1994). The availability of C sources in the soil influences the magnitude of the denitrification process as well as the products of the denitrification process, specifically the N2O molar ratio (i.e. N2O:N2  +  N2O) (Baggs et al. 2000). Gillam et al. (2008) determined that N2O emissions only increased with NO −3 addition when C was also added; NO −3 addition alone did not necessarily stimulate N2O emissions. Increasing the availability of C substrates has been shown to increase (Dendooven et al. 1996) or decrease (Weier et al. 1993) the N2O molar ratio. Miller et al. (2008) found that the relationship between the supply of, and demand for, terminal electron acceptors (TEAs), as determined by the relative availability of C and NO −3 , influenced the amount of denitrification and the N2O molar ratio for both simple (glucose) and complex C sources (plant residues).

The form of C also has a significant influence on denitrification and N2O emissions. Gillam et al. (2008) found ground red clover was equivalent to glucose in its ability to stimulate N2O emissions per unit of available C added, but the greater C:N ratio of barley straw resulted in less CO2 and N2O emissions (Gillam et al. 2008). Similarly, Miller et al. (2008) found that C from ground red clover was equivalent to glucose C in stimulating the microbial population. Furthermore, Miller et al. (2008, 2009) found N2O emissions to be linearly related to C utilization by microbes as indicated by soil respiration across a wide range of carbon sources (glucose, plant residues, manures).

2.3 Nitrate Availability

Nitrate has been found to limit denitrification when NO −3 concentrations are low (< 5–10 mg N  kg−1 soil; Ryden 1983), a situation not commonly found in soils cropped to potato. In agricultural soils where NO −3 often accumulates to  >  10 mg N  kg−1 soil, the availability of NO −3 was found to have no effect on denitrification but a significant effect on the N2O molar ratio (Gillam et al. 2008). The influence of NO −3 on the N2O molar ratio is due to NO −3 being a more favorable electron acceptor than is N2O (Betlach and Tiedje 1981; Cho et al. 1997), and therefore its presence in soil suppresses both the expression (Firestone et al. 1979) and activity (Betlach and Tiedje 1981) of nitrous oxide reductase enzyme. The N2O molar ratio is influenced by the relative availability of N-oxides and C, as simple or complex sources (Weier et al. 1993; Miller et al. 2008). Concentrations of NO −3 -N ranging from 100 to 300 kg N  ha−1 soil can inhibit N2O reductase activity (Blackmer and Bremner 1978; Weier et al. 1993), reducing the conversion of N2O to N2 and thereby making N2O the principle product of denitrification (Blackmer and Bremner 1978; Firestone et al. 1979; Weier et al. 1993).

In potato production systems in Atlantic Canada, Dandie et al. (2008) found that in the ridge, the magnitude of N2O emissions was comparable with the denitrification rate suggesting that most gaseous emissions occurred as N2O. In contrast in the furrow, the denitrification rate was greater than N2O emissions, suggesting there was significant conversion of N2O to N2. This was attributed to the N fertilizer being concentrated in the potato ridge as a result of banded fertilizer application, and the increased NO −3 concentration decreasing N2O reduction, increasing the N2O molar ratio of denitrification.

2.4 Denitrifier Community

Denitrifiers are mainly aerobic heterotrophs (utilize C as an energy source) and are not necessarily controlled by the presence of anaerobic conditions or NO −3 availability in the soil (Murray et al. 1990). Several microcosm studies using agricultural soils from potato production systems investigated the effect of organic C availability and source under completely anoxic conditions. Miller et al. (2008, 2009) found that C from plant residues or manures applied at 0–500 mg C  kg−1 soil did not increase the abundance of broad groups of denitrifiers but did increase denitrification rate and N2O emissions. In contrast, the abundance of a broad group of denitrifiers, N2O emissions and denitrification all increased in soil amended with plant residues at 1,000 mg C  kg−1 soil (Henderson et al. 2010). Plant residues with different C/N ratios increased similarly the abundance of a broad group of denitrifiers (Henderson et al. 2010). These findings suggest that abundance of broad groups of denitrifiers increase when availability of organic carbon is greater than 1000 mg C  kg−1 soil but that the source of organic carbon had little influence on growth.

Nitrate and reduced intermediate products induce denitrification gene expression under anoxic conditions (Saleh-Lakha et al. 2009). Denitrification gene expression in a smaller group of denitrifiers, Pseudomonas mandelii and related species (Dandie et al. 2007) increased with nitrate addition in soil without organic C amendment (Shannon et al. 2011). In contrast, there was no increase in denitrification gene expression in response to soil NO −3 addition following organic amendment even despite there being an increase in denitrification rate (Shannon et al. 2011).

It is not clear whether the abundance of the denitrifier community and/or their denitrification gene expression play a role in controlling denitrification and N2O emissions. Several studies found no relationship between abundance of denitrifiers (Dandie et al. 2007; Miller et al. 2008, 2009; Henderson et al. 2010; Shannon et al. 2011) or denitrification gene expression (Henderson et al. 2010; Shannon et al. 2011) and N2O emissions or denitrification rate. These findings suggest that the denitrification activity was decoupled from the abundance and denitrification gene expression of denitrifiers and that control over the magnitude of denitrification might be occurring primarily at the level of the denitrifying enzymes. Absence of relationships between abundance and gene expression could also be explained from the lack of information as to which denitrifier species are active in a particular environment and from limitations in the tools used to quantify the denitrifiers.

3 Temporal and Spatial Variation in N2O Emissions

3.1 Temporal Variation

The interaction between the controlling factors discussed above result in highly episodic N2O emissions. Despite this episodic nature, seasonal patterns in N2O emissions are frequently observed. Following spring thaw and during the early part of the growing season soil moisture conditions, the release of organic carbon and the availability of NO −3 often combine to result in large N2O emissions (Christensen and Tiedje 1990). This has often been observed in potato production systems in Atlantic Canada (Fig. 14.2) but is not always the case. N2O emissions are, however, seldom monitored during the non-cropping period (November to May). In studies that have examined the non-cropping period in temperate climates, the emissions can be significant in the autumn and spring periods (Wagner-Riddle and Thurtell 1998; Ruser et al. 2006). This may be particularly true in the autumn to spring period following potato production where residual NO −3 after harvest is large (Zebarth et al. 2003).

The location, timing and intensity of C availability combine to determine the magnitude of N2O emissions (Ball et al. 2008). This can be seen in the similarity in patterns of N2O and CO2 emissions (Fig. 14.2). The spring flush of CO2 and N2O emissions is often attributed to the release of C following soil thawing (Christensen and Tiedje 1990; Burton and Beauchamp 1994) or as a result of tillage operations (Alvarez et al. 2001). The rewetting of dry soil can also result in a flush of respiration and has been shown to result in N2O emission events (Zebarth et al. 2008a). Snowden (2010) demonstrated that the preceding crop significantly affected denitrification and N2O emissions from the potato crop. While there were no significant differences in N2O or CO2 emissions during the growth of the non-potato rotational crops, the subsequent potato crop had greater rates of denitrification and N2O emissions where the preceding crops were forages (red clover, Italian ryegrass). This effect was attributed to the quality of the forage crop as a substrate driving denitrification. Nitrate exposure (a temporally integrated measure of NO −3 concentration), CO2 emissions and WFPS were able to explain 60% of the variation in cumulative growing season N2O emissions from the potato crop (Snowden 2010).

3.2 Spatial Variations in N2O Emissions and Denitrification

The spatial distribution of water is one the primary factors that govern where denitrification and N2O emissions occur. The movement of water falling on the soil as precipitation through the soil profile and along landscape features results both in the transport of C and N compounds, particularly NO −3 , as well as influencing the water content and therefore aeration status of the soil environment. Consideration of the influence of hydrology on the spatial distribution of N2O emissions must be addressed at several scales: the ridge-furrow complex, the soil profile, and the landscape.

3.2.1 Ridge-Furrow Complex

Water is not uniformly distributed in the potato ridge-furrow complex. Water applied to a potato field, whether as rainfall or irrigation, tends to accumulate in the furrow due to runoff from the ridge and leaf-drip from the outer foliage system (Saffigna et al. 1976). There is also greater soil wetting at the base of the plant due to stemflow. Zebarth and Milburn (2003) reported lower soil water content in the ridge than in the furrow in a potato field, a finding attributed to greater loss of water from the ridge by transpiration because no difference in water content was measured when no plants were present.

The architecture of the potato ridge-furrow complex is important in influencing both denitrification and N2O emissions. The furrow locations systematically maintain a greater percentage of water-filled pores than do ridges (Table 14.1). The greater WFPS results in the impeded diffusion of O2 into the soil, favoring denitrification (Table 14.1), and impeded diffusion of N2O away for the site of denitrification, resulting in a lower N2O molar ratio. The lower N2O molar ratio is also favored by the low NO −3 concentration in the furrow because fertilizer N is banded in the ridge. By comparison, the ridge has lower WFPS due to lower soil water content and soil bulk density, and consequently is less favorable for denitrification (Table 14.1). Despite the greater potential of gas exchange, denitrification does occur in the ridge due to the presence of organic matter rich soil and the potato rhizosphere resulting in a large rate of O2 consumption. In addition, the presence of large concentrations of NO −3 in the ridge ensure that when O2 supply cannot meet demand, denitrification will result in N2O being the primary end product and the subsequent reduction of N2O to N2 will be limited due to the greater substrate affinity of the nitrate reductase compared to the N2O reductase (Betlach and Tiedje 1981).

Table 14.1 Influence of row location (ridge, furrow) on cumulative growing season N2O ­emissions, cumulative CO2 emissions, mean denitrification rate and mean water-filled pore space (WFPS) as influenced fertilizer N management (N0  =  0 N; N200  =  200 kg N  ha−1 at planting; N120  +  80  =  120 kg N  ha−1 at planting plus 80 kg N  ha−1 at final ridge formation in a potato crop located in New Brunswick (Burton et al. 2008a)
Full size table

The end result of this architecture and the placement of N fertilizer in the ridge is that the potato ridge has a lower rate of denitrification, but greater N2O emissions, compared with the furrow (Table 14.1; Burton et al. 2008a; Dandie et al. 2008). Where fertilizer N is broadcast, resulting uniform distribution of NO −3 , greater N2O emissions occur from the furrow, particularly compacted furrows, and over-all N2O emissions from the field are greater (Ruser et al. 1998).

The distinct soil environments within the ridge-furrow complex also result in distinguishable changes in the abundance and diversity of the soil denitrifier community. Distinct differences in denitrifier community abundance (Dandie et al. 2008) and diversity (Wertz et al. 2009) were measured between the ridge and furrow, and also between locations within the ridge that were adjacent to or more distal from the plant. However the spatial distribution of the denitrifier communities within the ridge-furrow complex did not match the spatial distribution of denitrification activity. Thus, as indicated above, denitrifier abundance is de-coupled from denitrification activity. This shows that soil environmental conditions influence the activity of denitrifier enzymes and that greater denitrification is not necessarily the result of greater gene abundance (Dandie et al. 2008; Wertz et al. 2009).

3.2.2 Soil Profile

The ridge-furrow complex is hydrologically connected to the underlying soil as a result of the water draining through the soil profile. While the most biologically active region of the soil occurs at the soil surface, microbial processes also occur in the sub-surface. Water not only influences the potential for denitrification through its impacts on aeration, but also as the primary means by which C and NO −3 are delivered to the sub-surface. Examination of the N2O concentration in the soil profile of a potato – barley – red clover rotation revealed that the greatest concentrations of N2O were often not found in the biologically active surface layer but at depth (Fig. 14.3). Fall application of solid swine manure resulted in an increase in N2O concentration throughout the profile and in particular at 60 cm depth. This result points to the role of leaching of C and N, even if applied at the soil surface, in influencing the magnitude and location of N2O production (Burton and Beauchamp 1994) and denitrification (Paul et al. 1997).

Fig. 14.3
figure 3

Influence of N source on the N2O concentration of the soil profile of a potato–barley–red clover rotation. Sampling was done in the spring prior to the potato phase of the rotation following fall manure application but prior to spring manure applications

Full size image

In Atlantic Canada many potato producers have installed tile drainage as a means of removing excess water from their fields during the spring period. Examination of N2O emitted in tile drainage water indicated that dissolved N2O emissions could be as great in some years as those coming from the soil surface (Fig. 14.4; Burton et al. 2008b). The relative importance of surface emissions relative to dissolved N2O was a function of preceding crop and precipitation. In 2002, the year following potato production, large quantities of residual NO −3 and a wet autumn combined to favor the production of N2O in the sub-surface and its loss in tile-drainage water (Fig. 14.4). Lower soil NO −3 accumulation and reduced precipitation in the barley and red clover phases of the rotation resulted in lower dissolved N2O emissions.

Fig. 14.4
figure 4

Annual emissions of N2O dissolved in tile drainage water (drainage) or emitted from the soil surface (soil surface) in a soil cropped to barley (2002), red clover (2003), and in the spring period following red clover plow-down (2004)

Full size image

3.2.3 Landscape

The influence of water on the distribution of denitrification and its end products extends beyond the soil profile. In soil landscapes, lower positions in the landscape have greater water contents for longer periods of time. This has the potential to result in greater denitrification and N2O emissions. The degree to which this potential is realized is a function of the architecture of the system. As an example, consider the N2O emissions from a soil landscape located in a single field in New Brunswick where soils vary primarily in terms of internal drainage (Fig. 14.5). In this landscape, a catena of soils exist along the slope with the Research Station Association being located in the uppermost slope positions and the Fundy Association being located at the lowest positions. In 2002, a large precipitation event in July resulted in a period of very large N2O emissions and these were greatest in the lower slope positions. The remainder of the year had little to no N2O emissions and there was no difference among slope positions. Similar patterns were seen in the rate of denitrification with greatest denitrification occurring on the July sampling date and at the lower slope position (Fig. 14.6). In contrast, lower precipitation and the lack of a large growing season rainfall event in 2003 resulted in lower rates of N2O emissions and no clear pattern in N2O emissions (Fig. 14.5) or denitrification (Fig. 14.6) with slope position. This emphasizes the situational nature of the controls of the denitrification process and its products.

Fig. 14.5
figure 5

Influence of precipitation on the timing of N2O emission events as a function of landscape. The Research Station soil association is located at the top of the slope and Fundy at the bottom. The relative internal drainage of the associations decrease in the order Research Station  >  Fredericton  >  Lincoln Road  >  Fundy

Full size image
Fig. 14.6
figure 6

Denitrification rate measured using the acetylene blockage method as influenced by soil association in a landscape study under potato production in New Brunswick: (a) 2002, (b) 2003

Full size image

4 Implications of Management on N2O Emissions

4.1 Crop Species

One of the most profound influences on soil biology and its activity is the choice of crop species grown. Crop species differ in the nature and extent of the plant root system and its associated rhizosphere, fertility requirements, residue quality (C/N ratio, lignin content), and soil and water management practices. These conditions also influence the potential for N2O emissions. Potato is often grown in rotation with other crops, often in 2- or 3-year rotations. The potential for N2O production during the non-potato phases should also be considered in the overall N2O emission profile of the potato production system and may offer an opportunity to reduce the N2O emissions from this production system. Snowden (2010) demonstrated that the preceding crop has an impact on the N2O emissions occurring during potato growth and attributed this to differences in the C/N ratio and quantity of crop ­residue returned and the subsequent impacts on C and NO −3 availability. An examination of the N2O emissions associated with each phase of a potato–barley–red clover ­rotation indicated that N2O emissions were generally related to the accumulation of NO −3 in the soil (Fig. 14.7; Burton 2005). Nitrate exposure, an integrated measure of NO −3 concentration over time, explained 65% of the variation in cumulative growing season N2O emissions across the three cropping seasons (Fig. 14.7). Periods of large emissions were generally associated with periods of NO −3 accumulation as a result of N fertilization or the plow-down of N-rich red clover residues. One exception to this generalization was large N2O emissions in the spring of the red clover phase that was attributed to precipitation events. Selection of rotational crops that maintain small soil NO −3 concentrations through plant uptake or smaller rates of N application will generally result in smaller N2O emissions. For potatoes grown in 2-year rotations, differences in cumulative N2O emissions during the potato phase could be explained primarily by the effect of the preceding crop on soil nitrate availability (measured as nitrate exposure). The exception was where the preceding crop was a legume or non-legume forage crop, where N2O emissions were greater than would be expected based on nitrate exposure alone, a finding attributed to increased N2O emissions early in the growing season due to enhanced respiration (Snowden 2010).

Fig. 14.7
figure 7

Relationship between nitrate exposure and cumulative growing season N2O emissions as measured in a potato-barley-red clover crop rotation located at the Crops and Livestock Research Station, Charlottetown, PE (Burton 2005)

Full size image

4.2 Fertilizer N Management

Emissions of N2O can be influenced by the rate, timing, placement and form of fertilizer N applied. In general, N2O emissions increase as a function of N application rate, and this has also been reported for potato production (Ruser et al. 2001). This concept forms the basis of the IPCC coefficient for N2O emissions associated with fertilizer application that assumes that 1% of the applied N fertilizer is emitted as N2O on an annual basis (IPCC 2007).

While research is limited on the effect of fertilizer N rate on N2O emissions in potato production in Atlantic Canada, greater N2O emissions were measured for potatoes grown with 200 kg N  ha−1 applied than with an unfertilized control due to increased soil NO −3 concentration (Burton et al. 2008a). This concept is also demonstrated in the relationship between nitrate exposure and cumulative growing season N2O emissions (Fig. 14.7). Nitrate exposure increases with increasing N rate (Burton et al. 2008a; Zebarth et al. 2008a) and also results in increased N2O emissions. In other crop species, N2O emissions increased with increasing fertilizer N rate on barley in New Brunswick, and the increase in emissions per unit of N applied was greater for N rates above the optimal application rate (Zebarth et al. 2008a). However, fertilizer N application does not necessarily increase N2O emissions in all cases. Application of additional N pre-plant or at sidedress to corn in New Brunswick did not significantly increase N2O emissions compared with a treatment that received starter fertilizer N only (Zebarth et al. 2008b).

Emissions of N2O are highly sporadic and event driven. It seems reasonable to expect management practices that avoid the accumulation of NO −3 , in the presence of available organic C under restricted aeration, will result in reduced N2O emissions. The timing of N fertilizer application is one such example. Split application of NH4NO3 has been shown to result in reduced N2O emissions (Burton et al. 2008a), but this only occurred in a year when there was significant precipitation between the initial application of N fertilizer and the split application. Thus, while split application of N fertilizer has the potential to reduce N2O emissions and increase N use efficiency, this practice will only do so in the years where the conditions are favorable for N losses during the timing between planting and the split application of N.

The placement of N fertilizer in the potato ridge is a common practice in Atlantic Canada. Earlier in this chapter we described why this results in an increase in the N2O molar ratio, resulting in more N2O emissions from the ridge than if no NO −3 was present. Other studies have shown that the broadcast of N fertilizer results in an increase in the N2O molar ratio in both ridge and furrow (Ruser et al. 1998). Placement of the N fertilizer in the potato ridge lowers the soil NO −3 concentration in the furrow, the area of most active denitrifying activity (Table 14.1; Burton et al. 2008a) and therefore results in lower overall N2O emissions.

Controlled release N fertilizer products have been shown to increase nitrogen-use efficiency (Shoji et al. 2001) and reduce N2O emissions (Hyatt et al. 2010). Plant N uptake can be greater with controlled release N fertilizer products than ­conventional products, particularly on sandy soils (Shoji et al. 2001; Wilson et al. 2009; Ziadi et al. 2011). Hyatt et al. (2010) reported reduced N2O emissions with application of a polymer-coated urea product compared with conventional fertilizer N products in irrigated potato production on sandy soils in Minnesota. Controlled-release products are most likely to be most beneficial in reducing N2O emissions when the rates of application are reduced to reflect the increased efficiency of delivery of the controlled release product.

4.3 Influence of Tile Drainage

Tile-drainage influences N2O emissions in several ways. The most obvious is by reducing the extent and duration of water accumulation within the soil profile. Greater water accumulation increases the potential for anoxic processes such as denitrification, potentially resulting in greater N2O production. Alternatively, greater water accumulation may also result in a reduced N2O molar ratio under very wet conditions, potentially reducing N2O emissions. As a result, the introduction of tile drainage may increase or decrease N2O emissions. The relative importance of these two opposing influences will vary from year-to-year and from site-to-site.

The less obvious implication of tile drainage is its role in transporting NO −3 and dissolved organic carbon into the soil subsurface and for the production of N2O and its loss as dissolved N2O in tile drainage water. As indicated earlier (Fig. 14.4) the loss of N2O dissolved in tile drainage water could be similar to, or much less than, surface emissions depending on management practices and climatic conditions.

4.4 Application of Manure

Animal manures and other organic nutrient sources are increasingly applied as nutrient amendments in potato rotations and as a means of increasing soil organic matter content. These amendments have the potential to increase denitrification and/or N2O emissions as a result of both increased N and C availability. A study of the form of manure (solid vs. liquid) and the timing of manure application prior to potato production (autumn vs. spring), applied at equivalent rates, demonstrated that neither form of manure nor timing of application influenced N2O emissions compared with NH4NO3 as an N source (Table 14.2). Greater differences were observed between the various phases of the crop rotation (i.e., crop species) than the manure application

Table 14.2 Cumulative N2O emissions as influenced by location (ridge vs. furrow), nitrogen source and timing of application in (a) 2001-potato; (b) 2002-barley under-seeded to red clover; (c) 2003-red clover; and (d) Spring 2004 (following fall manure application and plow-down of red clover). Note that manure was applied in the fall 2000/spring 2001 and again in fall 2003/spring 2004 (Burton et al. 2005)
Full size table

5 Conclusions

The basic biology of the denitrification process is well understood and the environmental factors that control denitrification have been well studied under controlled conditions. The supply of electron acceptors (O2, NO −3 ) in combination with the demand for electron acceptors (C availability) determines the extent of denitrification and N2O molar ratio of the products of denitrification. In the potato production system the formation of a potato ridge results in two distinct environments – a more aerobic ridge and a less aerobic furrow. In Atlantic Canada the banding of fertilizer in the ridge results in increased N2O emissions from the ridge but reduced emissions from the furrow resulting in lower overall emissions. As a result the N2O emissions from potato production in Atlantic Canada during the growing season are relatively modest, frequently less than the assumed value for annual emissions of 1% of applied fertilizer N application used by the IPCC (2007). Furthermore, the split application of N fertilizer has the potential to reduce N2O emissions in years when there is rainfall in the early part of the growing season. The potential for N2O loss during the non-growing season and/or the loss of N2O in tile drainage water has yet to be adequately assessed. Annual N2O emissions are usually related to N application rate but are also influenced by other management factors such as the timing of N fertilizer application, manure application and the preceding crop. In general, management practices that reduce the accumulation of NO −3 in the soil will likely result in reduced N2O emissions. This underscores the need for better tools to manage nitrogen in potato production (Chaps. 13 and 14) and effective monitoring tools (nitrate exposure).