4 - Potential management practices and technologies for reducing nitrous oxide emission from agriculture

Summary

Nitrous oxide emissions from pastoral agriculture arise as a result of the soil processes denitrification and nitrification. Mitigation options aim to reduce the rates of these processes and ensure that nitrogen is emitted as harmless N2 rather than N2O. Denitrification and nitrification are affected by many soil and climatic factors, which hampers the evaluation of potential mitigation options. As a result, experimental evidence of the impact of mitigation options on N2O emissions is limited. The main findings on nitrous oxide mitigation options are:

  • Management practices that reduce the amount of excreta N yield the biggest reduction in N2O emissions.
  • Reducing animal numbers is financially unsustainable.
  • Diet manipulation and changing winter management practices have the potential to each reduce N2O emissions by approximately 6%. These practices may, however, have a negative impact on methane emissions and carbon storage in soils.
  • Based on current N fertiliser usage, N2O emissions could be reduced by 1-3% by reducing the amount, timing and type of N fertilisers used. However, the impact of these mitigation options could increase with time, with an expected increase in N fertiliser use.
  • Altering soil conditions by liming, improving drainage and avoiding compaction have the potential to reduce N2O emissions by 4.5%, 2.5% and 3% respectively. Improving drainage may, however, increase carbon oxidation from soils.

4.1 - Introduction

Nitrous oxide emissions from agriculture are a result of the biological soil processes denitrification and nitrification. Denitrification is the stepwise reduction of soil nitrate to gaseous nitrogen compounds, with N2O being one of the intermediate products (Haynes & Sherlock 1986):

Nitrification is the biological oxidation of soil ammonium to soil nitrite and nitrate, with N2O being produced as a by-product (Haynes 1986):

Mitigation options should aim to reduce the N2O emission rate of these processes per unit of soil surface by,

  • Manipulating process rates
  • ensuring that during denitrification, N is emitted as N2 rather than N2O:

4.1.1 - Manipulating process rates

At the processes level, denitrification and nitrification are both affected by a range of soil variables (Fig. 4.1), of which mineral N availability and soil aeration are generally considered the main factors (Mosier et al 1996). These so-called direct or 'proximal' regulators are in turn affected by more indirect or 'distal' environmental or management factors. The potential to manipulate direct regulators of denitrification and nitrification varies for each factor. For example, options for influencing soil temperature are very limited, whereas soil mineral N content could be influenced, for example, by adapting N fertiliser practices and/or the return of animal excreta to the soil. Many of the proposed mitigation options focus on reducing the availability of mineral N for nitrification and/or denitrification (Table 4.1), while some mitigation practices focus on soil drainage or irrigation management to improve aeration and avoid prolonged periods of wet soil conditions.

4.1.2 - Ensuring N is emitted as N2 rather than N2O

The ratio at which N2O and N2are produced during denitrification can range from less than 5% to more than 50%, and largely depends on environmental conditions (Stevens et al 1997). It is generally considered that the N2O:N2 ratio increases with higher NO3- concentration, and decreases with higher soil pH, organic C, soil water content and temperature (Fig. 4.1).

Figure 4.1. Schematic diagram of factors affecting N2O emission from agricultural soils (adapted from Tiedje 1988 and de Klein et al (2001)). Shaded boxes represent biological processes. + positive effect; - negative effect; n/a not applicable

Figure 4.1

4.2 - Overview of current operational options

Potential mitigation options generally focus on reducing the availability of mineral N for nitrification and/or denitrification and aim to increase the N efficiency of agricultural systems by optimising the plant's ability to compete with N loss processes (Table 4.1).

Table 4.1. Proposed management options for reducing N2O emissions from agricultural systems (after Cole et al 1996; AEA Technology Environment 1998; Oenema et al 1998; Velthof et al 1998).

1. Improve livestock production management

  • Re-use manure in plant production
  • Reduce or limit stocking rate
  • Increase production per animal, if linked with an equivalent decrease in animal numbers
  • Increase efficiency of excreta N (restricted grazing to avoid urine and dung patches during high-risk periods)
  • Reduce N concentration in animal excreta (diet manipulation, lower N inputs)
  • Breed livestock with higher N use efficiency

2. Improve synthetic and organic N fertiliser management/match N supply with crop demand

  • use soil/plant testing to determine fertiliser needs
  • optimise split application schemes
  • match N application to reduced production goals
  • use controlled release fertilisers and nitrification inhibitors
  • place fertiliser below surface
  • use foliar feed fertilisers
  • match fertiliser type to seasonal conditions

3. Improve grassland/crop management

  • Optimise irrigation and drainage (prevent large groundwater fluctuations or flooding).
  • Maintain soil pH above 5
  • Use N fixing crops
  • Breed cultivars that improve N use efficiency
  • Maintain residue N on production site
  • Minimise fallow periods by growing cover crops
  • Prevent soil compaction

This will not only lower the required total N input while maintaining production levels, but will also reduce nitrate leaching and ammonia volatilisation and thus indirect nitrous oxide emissions. In livestock production systems, the largest inefficiency of the nitrogen cycle is caused by the relatively poor conversion of dietary N into products. The retention of N in meat, wool or milk generally ranges from 3 to 25% of the N intake (Whitehead 1995). As a result, large quantities of N are recycled within these systems via animal excreta. For a typical New Zealand dairy or hill country sheep farm, Haynes & Williams (1993) reported N retention values of 16% and 8%, respectively. In many overseas systems, where livestock is generally housed for 5 to 12 months of the year, N2O emissions can be reduced by improved utilisation of N excreta collected during housing (Cole et al 1996). In addition, advanced fertilisation techniques that improve the N efficiency of synthetic fertilisers, e.g. the use of slow release fertilisers, nitrification inhibitors or matching the type of fertiliser to seasonal conditions, are also considered to have good potential for reducing N2O emission. The improved utilisation of both excreta and fertiliser N can reduce the direct emissions of N2O from these sources, as well as reduce the requirement for fertiliser N inputs to maintain production levels.

The success of the management options listed above in reducing N2O emissions largely depends on the rate at which they are adopted, which could be increased by enforcement and/or encouragement through policy measures. In recent EU reports on options to reduce N2O emissions (AEA Technology Environment 1998), proposed policy measures included:

  • N use quotas to limit the rates of fertiliser N application
  • Limits on the timing of fertiliser and manure application
  • Reduced price support for product
  • Provision of direct subsidies for marginal land

On an individual farm basis, proposed measures included:

  • Rationalisation of fertiliser N use by accounting for N applied as manure
  • Switching from winter to spring cultivars
  • Applying N fertiliser below the economic optimum level
  • Use of slow release nitrogen fertilisers

Options that were not further considered included:

  • Soil pH management, because on most farms liming already occurs
  • Irrigation management, due to lack of data
  • Fertiliser efficiency management, because many of the practices and technologies are already adopted
  • Fertiliser tax rates, because this measure is already in place in some countries and high tax rates are required to achieve a relatively small gain

In Canada, improving the efficiency of N fertiliser is one of the main N2O mitigation strategies (CHEMinfo, 1998). In 1996, direct emissions from synthetic fertiliser accounted for about 13% of the total N2O emissions from agriculture (Desjardins & Keng 1999), with over 90% of the synthetic fertiliser being used in arable systems.

In contrast, New Zealand's agriculture is dominated by pastoral livestock systems. Pastures are generally grazed year-round and obtain N mainly via biological fixation through clover, rather than through synthetic fertiliser N. As a result, many of the proposed mitigation options in overseas studies are likely to be of limited value. In fact, the use of white clover to replace fertiliser N inputs with biologically fixed N, has been suggested as an option for reducing N2O emissions in European pastoral agriculture (Jarvis et al 1996; Velthof et al 1998). However, Jarvis et al (1996) also noted that N losses depend largely on the total N input into a system, rather than the form of N input. Although the N content of animal excreta returned to pasture might be higher in a grass/clover system due to higher N concentration in grass/clover compared to grass herbage, all-grass systems generally require large amounts of fertiliser N which forms an additional source of N2O. Because of the generally lower production levels and stocking rates of grass/clover pastures compared to highly fertilised grass swards, total N losses from the former system are generally substantially lower (Jarvis et al 1996). Ruz-Jerez et al (1994) measured N2O emissions from New Zealand grass/clover and N fertilised grass sward receiving 400 kg N ha-1 year-1 (Table 4.2). Both the total N2O emissions and the N2O production per ton of dry matter produced were higher for the N-fertilised grass sward, despite the higher dry matter production of this system (Ruz-Jerez et al 1991).

Table 4.2 Annual N2O emission and dry matter production from a grass/clover sward and a N fertilised grass sward (receiving 400 kg N ha-1 year-1) in New Zealand (Ruz-Jerez et al 1991; 1994).

  Grass/Clover N-fertilised grass
N2O emission (kg N2O-N ha-1 year-1) 1.3 5.3
Dry matter (t DM ha-1 year-1) 11.7 17.0
N2O emission per ton DM (kg N2O-N t-1 DM) 0.11 0.31

Application of the IPCC methodology to New Zealand shows that the recycling of the biologically fixed N by the grazing animal is the single largest source of N2O (over 50% of the total emissions). The second largest source of N2O is indirect emissions, contributing about 35% of the total emissions (de Klein et al 2001). However, about 85% of these indirect emissions also originate from excreta N deposited by grazing animals, and management options that reduce the amount of N recycled by the grazing animal, or increase the efficiency of this recycled N, are likely to have the biggest impact on reducing N2O emissions from agriculture. Synthetic fertilisers currently contribute about 10% of New Zealand's total N2O emissions, including both direct (7% of total) and indirect emissions (3% of total). Although their contribution is relatively small, management practices that increase the utilisation of N from synthetic fertilisers should also be considered. Some of the advanced fertilisation techniques that are suggested overseas, e.g. the use of slow release fertilisers, foliar feed fertilisers, placing fertiliser below the surface or matching the type of fertiliser to seasonal conditions, are likely to be of limited relevance in New Zealand, where urea is by far the most commonly used nitrogen fertiliser because of its lower cost compared to other types of N fertiliser. The use of nitrification inhibitors could however, be a potential mitigation option.

Management options that ensure N from denitrification is emitted as N2 rather than N2O, will also be applicable to New Zealand.

4.3 - Mitigation options relevant to New Zealand pastoral systems

Mitigation options and management practices for reducing N2O emissions from New Zealand pastoral agriculture can be divided into four categories, with the following impact:

  • reduce the amount of N recycled by the grazing animal
  • increase the efficiency of recycled N
  • increase the utilisation of N from synthetic fertilisers
  • ensure N from denitrification is emitted as N2 rather than N2O

This section summarises the mitigation options and management practices that are most relevant to New Zealand pastoral farming. For each option, a brief justification is given, followed by the assumptions that are used to calculate the impact of each option on the total N2O inventory. The references in brackets behind each option refer to Figure 4.2. (p23), which presents the results of these calculations.

The assumptions used in the calculations are based on the (limited) information that is currently available in the national and international literature. Where appropriate, references to these publications are included.

4.3.1 - Reduced amount of N recycled by the grazing animal

Reduce or limit stock numbers (Stock numbers -5% reduction)

At face value, a reduction in stock numbers appears to be an effective way to reduce the amount of excreta N produced, and initial estimates of N2O and CH4 emissions indeed suggested a decrease in these emissions due to the reduction in sheep numbers between 1990 and 1999. However, recent studies have shown that in that same period the production of excreta N per sheep increased, due to an increase in lambing % and slaughter weight per head (Clark 2001; Sherlock et al 2001). Although sheep numbers declined between 1990 and 1999 by about 18%, the total N2O production from sheep excreta decreased by only 7%, due to an increase in the excreta N production per head of about 13%. Therefore, a reduction in stock numbers will only be an effective mitigation option if the excreta N production per animal remains the same or increases at a slower rate than the stock numbers decline. Conversely, a management practice that will increase the production per animal, will only reduce N2O emissions if it coincides with a reduction in animal numbers that is larger than the increase in production per animal.

For the purpose of this report, this N2O mitigation option was evaluated by assuming that stock numbers reduced by 5% compared to the 1999 levels, while the production per animal remained constant.

Reduce N concentration in animal excreta (Diet manipulation - dairy cattle only)

The N concentration of grass/clover pasture in New Zealand is typically higher than that of N-fertilised grass pasture in Europe and this could be the main factor for higher estimates of N excretion rates from NZ animals compared to the IPCC default values (Sherlock et al 2001). In 1999, sheep, dairy cows and beef cattle contributed 41, 34 and 25% of the total excreted N from these 3 main animal species, respectively. It is estimated that by 2010 these relative contributions will be about 31, 45 and 24% (Sherlock et al 2001).

In the dairy industry, fertiliser N is commonly used to provide additional feed when demand exceeds supply. Replacement of this fertiliser N-boosted grass with maize silage to reduce the amount of N in excreta has been suggested as a management practice to reduce nitrate leaching (Ledgard et al 2001). Here we evaluate diet manipulation as a N2O mitigation option by assuming that the excreta N production from dairy cows was reduced by 10% through substitution of fertiliser N-boosted grass with maize silage (Jarvis et al 1996; S.F. Ledgard personal communication). For a 100 ha farm, receiving 100 kg N/ha/yr as N fertiliser, the estimated dry matter (DM) production from this fertiliser is 1000 kg DM/ha (Ledgard et al 2001), i.e. 100 t DM for the entire farm, requiring 10,000 kg N. To produce the same amount of DM from maize would require 5 ha of land in maize at 20 t DM/ ha (Yamoah et al 1998). The average N fertiliser requirement for maize is about 200 kg N/ha , i.e. 1000 kg N to produce 100 t DM. It was therefore also assumed that, by substituting N fertiliser boosted grass with maize silage the N input through fertiliser use in the dairy industry was reduced by 90%. As a result, the total N fertiliser use, and the associated N2O emission, was reduced by 60%.

4.3.2 - Increased efficiency of N recycled by the grazing animal

Reduce urine and dung patches from dairy and beef cattle (Cattle winter management).

N2O emissions from animal excreta are likely to be highest during the wet autumn/winter period (de Klein et al 2001). If dairy and beef cattle were to be kept on feed-pads during these high-risk periods, the excreta collected and re-utilised as effluent, emissions could be reduced as N2O emissions for urine and dung are higher than for effluent, which has been applied to the soil properly (Oenema et al 1997). This management practice could also reduce nitrate leaching losses by 45-55% (de Klein & Ledgard, 2001), although ammonia volatilisation is likely to increase. This mitigation option was assessed by assuming that direct emissions from dairy excreta were reduced by 25%, nitrate leaching was reduced by 40%, and ammonia volatilisation was increased by 100% (de Klein & Ledgard 2001; Ledgard et al 2001).

Re-use dairy effluent in forage production (Dairy effluent utilisation).

This management option applies to the dairy industry only where effluent that has been collected in the farm dairy during milking can be re-applied to pasture. Land application of farm dairy effluent is increasingly common, and it is estimated that currently about half of the farmers nation-wide have adopted this practice. A typical effluent application rate is about 150 kg N/ha/yr, compared to the average N fertiliser rate of about 75 kg N/ha/yr (Ledgard et al 2000). On an individual farm, re-utilising the farm dairy effluent can thus reduce the fertiliser use by half of the effluent N applied. It was therefore assumed that if all dairy farmers adopted this management practice, the current N fertiliser use could be reduced by 25%.

Optimise drainage (Improved drainage).

Results of the NzOnet pilot study (Barton et al 2000) showed that N2O emissions from urine amended pasture on poorly or imperfectly drained soils are much higher than from free draining soils. The effect of improving drainage of poorly and imperfectly drained soils on N2O emissions is unknown, but results from a recent laboratory study suggest that N2O emissions from soil kept under 'poorly drained' conditions could be about 5 times as high as emissions from the same soil under 'free draining' conditions (C. de Klein unpublished. data). However, improved drainage is likely to increase nitrate leaching and thus indirect N2O emissions. Scholefield et al (1993) found that nitrate leaching from a clay soil increased 3 fold due to artificial drainage. However, the relative increase in nitrate leaching due to improved drainage depends on both N input and soil texture (Scholefield et al 1991). Although field data are limited, in his dynamic N model Scholefield et al (1991) suggested that on average nitrate leaching would increase by 100% due to improved drainage. It was assumed here that optimising drainage in poorly and imperfectly drained soils (i.e. 26% of total pastoral area), will reduce the emission factor for excreta and fertiliser for these soils by 25%, but will double nitrate leaching losses.

Breed cultivars that improve N use efficiency (High sugar grass).

At the Institute of Grassland and Environmental Research in Aberystwyth, UK, studies on feeding dairy cows high sugar grasses showed that these grasses resulted in reduced N excretion rates to the environment (IGER 2001). These grasses also contained less plant protein but due to a better balance between energy and protein supply animal performance was not affected. Under some conditions these grasses even increased milk yield. The results suggested that the high sugar grasses reduced the feed N loss to the environment by about 24%.

In New Zealand, dairy cows are grass/clover fed with approximately 25% of the diet N from clover (based on average clover content of pasture of 17%; N content of clover of 5%; and N content of mixed herbage of 3.55%). Substituting normal grasses with high sugar grasses will therefore have a smaller effect on excreta N rates than in a grass-only system. Here we evaluate the effect of feeding dairy cows high sugar grass by assuming the excreta N rate of dairy cows is reduced by 15%, and that the same N fertiliser input is required to grow the high sugar grasses.

It should be noted, however, that a poor survival rate of new plants could limit its usefulness as a mitigation tool, and that traditional soil cultivation practices will release large quantities of carbon from the soil (Crush et al 1992).

Prevent soil compaction (Avoid compaction)

R.A. Carran and P. Theobald (unpublished data) recently measured N2O emission rates of up to 9.8 kg N2O-N ha-1 day-1 on a severely pugged, fertilised pasture in Palmerston North. Several overseas studies have also reported on the effect of compacted soil on N2O emission (McTaggart et al 1997; Sitaula et al 1997; Abbasi & Adams 1999). For example, McTaggart et al (1997) found that N2O emissions from compacted grassland soil were more than twice those from uncompacted soil. Oenema et al (1997) suggested that although data on the effect of soil compaction on N2O emissions is limited, treading by cattle can easily enhance N2O emissions from grassland soil by a factor of two.

The prevention of soil compaction as a N2O mitigation option is evaluated by assuming that:

  • NZ soils are susceptible to pugging and compaction for on average 4 months per year,
  • dairy cows and beef cattle treading is avoided on poorly and imperfectly drained soils,
  • and as a result, the N2O emission factor for excreta N on these soils is reduced by 50%.

4.3.3 - Increased efficiency of N from synthetic fertiliser

Use nitrification inhibitors or adjust timing of N fertiliser (Nitrification Inhibitors; Fertiliser timing)

Nitrification inhibitors can affect N2O emissions in two ways:

  1. by delaying the nitrification process and thus N2O emissions from nitrification, and
  2. by delaying the formation of nitrate and thus reducing N2O from denitrification. Several overseas studies have shown that the use of a nitrification inhibitor can reduce N2O emissions from urea fertiliser by on average 50% (de Klein et al 2001).

Nitrous oxide emissions following fertiliser application are highest in wet soils. In addition, a short sharp burst of N2O often occurs shortly after (heavy) rainfall. Emissions appear to be high during the first 24 to 48 h after rainfall that follows a relatively dry period. This is probably due to the enzyme responsible for reducing N2O into harmless N2 gas not surviving well in dry conditions (de Klein et al 1999; Dendooven & Anderson 1994). It is expected that N2O emissions from nitrogen fertiliser can be reduced by delaying the timing of application to avoid periods of (heavy) rainfall or wet soils. Although the effect of this management practice on N2O emissions has not been studied in detail, based on our understanding of the effect of soil water and/or rainfall on emission patterns following fertiliser application (Ruz-Jerez et al 1994; Velthof et al 1998), it was assumed that N2O emissions could be reduced by up to 50%.

4.3.4 - Ensure N from denitrification is emitted as N2 rather than N2O

Optimise liming practice (Lime management)

Recent studies have suggested that increasing soil pH through liming could be an effective tool for reducing N2O emissions, as at higher soil pH, N2O is likely to be reduced to N2 (Stevens and Laughlin 1997; van der Weerden et al 1999). However, the research data currently available are inconclusive and sometimes contradictory. Work by van der Weerden et al. (1999) suggests that maintaining soil pH at 6.5 could reduce emissions, while Stevens et al (1998) suggested that N2O flux is maximum at pH 6.5 and minimum at pH 6.0 and 8.0. Ellis et al (1998) suggested that N2O flux decreases when the pH decrease from 6.1 to 3.3. Wang et al (1997) measured N2O emissions from limed (soil pH in top 5 cm: 6.4) and unlimed soil (soil pH 5.4) and found higher N2O emissions from the limed soil when the soil was wet, but lower emissions from the limed soil under dryer conditions. Clearly, more work is required to accurately determine the impact of pH management on N2O emissions. However, if pH management is proven to be a promising option, it is likely to have the biggest impact by targeting this management practice to soil with the highest N2O emissions factors, i.e. the poorly and imperfectly drained soils (Sherlock et al 2001). Although there is still great uncertainty about the impact of pH management on N2O emissions, for the purpose of this report it was assumed that pH of poorly and imperfectly drained soils is increased by 0.5 pH units and, as a result, the emission factor for excreta and fertiliser on these soils is reduced by 15%. Although the potential could perhaps be higher, it provides an illustration of how N2O emissions are affected at that reduction rate.

It should also be noted here that the most commonly used liming agent is calcium carbonate, which, if completely decomposed, would yield CO2. However, under the soil pHs generally encountered in New Zealand soils, HCO3- is likely to be the main reaction product. This HCO3- can leach from the soil system and ultimately end up in the oceans. Since oceans are considered to be CO2 sinks, the amount of CO2 emitted from lime applications could be limited. Furthermore, liming could also increase pasture production and the increased dry matter yield may offset some, or all, of the CO2 that could potentially occur from calcium carbonate applications. The use of liming agents that would not directly affect CO2 emission could also be explored. More research on the effect of various liming agents on CO2 emissions is warranted.

4.4 - Potential of mitigation options to reduce nitrous oxide emissions

A summary of the estimated effect of various mitigation options on the total N2O emission from agriculture is given in Figure 4.2. The assumptions used for these calculations are presented in section 4.3., but a summary of the assumptions is given in Table 4.3. These mitigation options either impact on activity data (e.g. stock numbers, excreta N per head, fertiliser use) or on emission factors.

Table 4.3. A summary of mitigation options and assumptions used to calculate their impact on the N2O inventory for 1999.

Option Assumptions
Stock numbers Animal stock numbers reduced by 5%
Diet Manipulation The amount excreta N from dairy cows is reduced by 10% because they are fed maize silage instead of N fertiliser boosted grass
Cattle winter management N2O emission are reduced by 25%, nitrate leaching reduced by 40%, and ammonia volatilisation increased by 100%, by keeping dairy and beef cattle on feed pads during autumn/winter
High sugar grass The amount excreta N from dairy cows is reduced by 15% by using high sugar grass
Dairy effluent utilisation N fertiliser use is reduced by 25% through increased utilisation of farm dairy effluent
Nitrification inhibitor N2O emissions from N fertiliser are reduced by 50% by using a nitrification inhibitor
Fertiliser timing N2O emissions from N fertiliser are reduced by 50% by avoiding fertiliser applications during wet periods or after (heavy) rainfall
Lime management The N2O emission factor of poorly and imperfectly drained soil is reduced by 15% by increasing the pH of these soils by 0.5 units
Improved drainage The N2O emission factor of poorly and imperfectly drained soil is reduced by 25%, but nitrate leaching is doubled
Avoid compaction For 4 months of the year, the N2O emission factor of dairy and beef cattle excreta on poorly and imperfectly drained soil is reduced by 50%

Figure 4.2

Figure 4.2. Estimated reduction in total national N2O emission for 1998 inventory based on various mitigation options.

As expected, management options that reduce the amount of excreta N returned to pasture (stock numbers, diet manipulation, winter management, high sugar grasses) are likely to have the biggest impact on N2O emissions. The effect of management practices to reduce the N2O emissions from fertiliser use (manure utilisation, nitrification inhibitors, fertiliser timing) is limited, due to the relatively low amounts of N fertiliser used in New Zealand. Their impact could, however, increase with time, with an expected increase in N fertiliser use. The results further suggest that management practices that impact either on the N2O-to-N2 ratio (lime management) or on the N2O emission factor of poorly and imperfectly drained soils (improved drainage, avoid compaction) could reduce N2O emissions by about 2.5 to 5%. It should be noted, however, that the estimations presented in Figure 2 are based on limited research data and that more work is required to verify the assumptions used (section 4.3).

The impact of these mitigation options on N2O emissions estimated for 2010 is given in Table 4.4. These calculations are based on the predicted N2O emissions from dairy, sheep and beef for 2010 as presented by Sherlock et al (2001) under scenario 4, and an estimated N fertiliser use of about 275,000,000 kg N/yr. For each mitigation option the same assumptions were used as above (Table 4.3),

Table 4.4. Predicted direct and indirect N2O emissions (Gg/yr) from dairy, beef and sheep and from N fertiliser use for 2010, based on various mitigation options using the assumptions listed in Table 4.3, compared to current estimates of N2O emissions for 1990 and 2010. The estimates are given for various adoption rates for the different mitigation options.

Year   N2O emission (Gg/yr) % increase since 1990      
1990   42.5            
2010 BAU* 57.2 35          
    Adoption rate       At 100% adoption rate
  Mitigation option 25% 50% 75% 100%   % increase since 1990 % decrease from BAU*
2010 Stock numbers 56.5 55.9 55.3 54.7   29 4
  Diet Manipulation 55.6 54.0 52.4 50.8   20 11
  Cattle winter management 56.1 55.0 54.0 52.9   24 7
  High sugar grass 56.3 55.5 54.7 53.9   27 6
  Dairy effluent utilisation 57.0 56.9 56.7 56.5   33 1
  Nitrification inhibitor 56.6 56.0 55.4 54.7   29 4
  Fertiliser timing 56.6 56.0 55.4 54.7   29 4
  Lime management 56.5 55.8 55.1 54.4   28 5
  Improved drainage 56.4 55.6 54.9 54.1   27 5
  Avoid compaction 56.7 56.1 55.6 55.1   30 4

* BAU, Business as usual

The results suggest that manipulating the diet of dairy cows could yield one of he highest reductions in N2O emissions. Replacement of fertiliser N-boosted grass with maize silage will not only reduce the total amount of excreta-N returned to pasture, it is likely to reduce the N concentration in urine, rather than in dung (van Vuuren and Meijs 1987). In other words, a larger proportion of N will be excreted as dung. Although New Zealand currently uses the same N2O emissions factor for urine and dung, there is limited overseas evidence that N2O emissions from urine could be higher than those from dung (Yamulki et al 2000). If so, diet manipulation will not only reduce N2O emission by reducing the total amount of excreta-N returned, but also by partitioning less N into urine N, which has the largest emission factor. However, more work is required to determine the relative contributions of dung and urine to N2O emissions in New Zealand. It should also be noted that converting 5% of pasture land into maize cropping, will result in a 5% increase in stocking rate on the remaining pasture land if stock numbers remain constant. Alternatively, if the stocking rate is not increased, a 5% increase in per cow production is required to maintain production levels. The increased release of soil carbon due to cropping should also be taken into account.

The results further suggest that cattle winter management also has potential as an N2O mitigation option. With this management practice, animal excreta are collected and stored during autumn/winter and returned to pasture in spring. Adoption of this management practice could have implications for the New Zealand N2O inventory calculation. Currently, New Zealand uses the IPCC default value for the N2O emission factor for manure application (1.25%), but the New Zealand country specific emission factor for animal excreta returned during grazing is 1%, which is lower than the IPCC default. Research results show, however, that N2O emissions for urine and dung are generally higher than for manure application (de Klein et al 2001), particularly if the latter has been applied to the soil properly (Oenema et al 1997). At the moment, this apparent anomaly does not have a big impact on the inventory calculations, as direct N2O emissions from animal manure applications only contribute about 2% of the total N2O emission (de Klein et al 2001). However, if the cattle winter management mitigation option to reduce N2O emissions is widely adopted, it seems prudent to refine the emission factor for manure applied to soil for the New Zealand situation.

A Systems Approach

The mitigation strategies presented in Figure 4.2 are evaluated as individual options, as it is outside the scope of this study to evaluate 'packages' of measures. However, simultaneous implementation of two or more of the options evaluated is likely to yield the greatest decrease in N2O emissions. For example, the dairy diet manipulation option and the cattle winter management could be combined by feeding the cattle maize silage while on the feed pad. However, evaluation of this 'package' of measures requires detailed system analysis, including simulation of farmer behaviour to ensure that the rules or measures achieve the desired effect. Further investigation into 'packages' of measures is recommended, including their impact on a whole farm systems basis.

4.5 - Effects of these options on other environmental parameters

Many of the proposed management options aim to increase N use and N efficiency. They can broadly be considered to fall into two groups:

  • those that reduce the amount of N deposited on pastures from dung and urine by focusing on the source of this N, the ruminant, and
  • those that try to reduce the specific rate of N2O emissions by better N management.

Manipulating the diet of ruminants (e.g. high sugar grasses, greater use of maize) should also have a beneficial impact on methane emissions (see Section 5.2.5). However, cultivation releases large quantities of carbon from the soil (Crush et al 1992) and any savings in N2O and methane will have to be balanced against increases in direct CO2 emissions. Collecting and storing manure during winter yields one of the highest reductions in N2O emissions but methane emissions from this stored excreta also have to be considered. Compared to enteric sources, methane emissions from dung are insignificant because dung is deposited directly onto pastures and undergoes an aerobic breakdown. Emissions from stored excreta are considerably higher.

Decreasing the specific rate of N2O release from land by increasing the efficiency of N use should also reduce nitrate leaching and/or ammonia volatilisation. Increased nitrate levels in rivers and lakes is seen as a problem in New Zealand (e.g. Lake Taupo) and any management practices that help to reduce N leaching are viewed favourably. Improving drainage may have some negative consequences for soil carbon. Saggar et al (2001) have argued that soil carbon levels in New Zealand pastures have changed little over the last 30-50 years but it is noted from their data that at two locations where the soils have been drained, soil carbon levels have more than halved.

In European agriculture, reductions in N2O emissions are often achieved through the adoption of policy measures aimed at reducing nitrate leaching (Kroeze 1998; Velthof et al 1998). In most of these cases, a 'package' of measures is proposed, and their impact is evaluated on a whole farm systems basis to ensure:

  1. that there are no adverse effects on other environmental parameters, and
  2. that the policy rule or measure achieves the desired effect.

A recent farm systems study in the Taupo catchment illustrates where a rule did not achieve the desired effect (Thorrold et al 2001). One of the options for reducing nitrate leaching from dairy farms was to winter cows off between 1 June and 31 August. Based on the understanding of the N cycle, this option was expected to have a significant impact. However, when the system analysis was carried out by a farm systems expert to simulate farmer behaviour, the results showed that the impact of this option on nitrate leaching was much less than anticipated. Within the rule of off-wintering, the farmer optimised farm profit by using an increased stocking rate and increased feed utilisation (including grazing the farm out in autumn). This increased feed utilisation, especially the high autumn utilisation and led to substantial leaching of the N deposited as urine in the autumn.

It is apparent that any measures aimed at reducing N2O emissions cannot be viewed in isolation. A key behavioural factor in any new technology or practice is the level of farmer adoption. In this study a range of adoption scenarios have been used. Measures need to be evaluated on a whole farm basis and this evaluation needs to cover both behavioural and technical aspects. The development of such evaluation tools is a priority.

4.6 - Cost/benefit analysis

The N2O mitigation options that appear most promising include 'stock number reduction', 'dairy diet manipulation' and 'cattle winter management'. Since these practices affect the management of farms as a whole, a cost/benefit analysis should include a systems analysis of all financial implications of these practices, which is beyond the scope of this report. However, some of the financial implications of these management options are discussed based on a recent farm systems study for the Taupo catchment (Finlayson and Thorrold 2001). Finlayson and Thorrold (2001) estimated the costs of management options to reduce nitrate leaching on pastoral farms. The farm systems that were evaluated included sheep & beef farms at various stocking rates (8, 12 and 16 SU/ha), a 'current' dairy system using 50 kg N fertiliser/ha/yr, a dairy system in which N boosted grass was replaced with maize silage (Diet Manipulation), and a dairy system where the cows where wintered on a feed pad and no N fertiliser was used (Winter Management). These farm systems are similar to the ones evaluated in the current report. The results of Finlayson and Thorrold (2001) give a good indication of the costs of these management practices.

The results indicated that the Farm Trading Surplus (FTS1) of the sheep & beef farms decreased dramatically with decreasing stocking rate: 486, 255 and 20 $/ha for stocking rates of 16, 12 and 8 SU/ha respectively. In other words, a 25% reduction in stock numbers reduced the FTS by 48%, while a 50% reduction in stock numbers resulted in a 96% reduction in FTS. Finlayson and Thorrold (2001) also estimated that the capital value of land and improvements of both sheep & beef farms (>5 SU/ha) and dairy farms would reduce between 35 and 40%, if stock numbers were reduced by 20%. The reduction in capital value of land and improvements was slightly lower (29-37%) if it was assumed that farmers could extract some capital ($200/ha) from their property as a result of the extensification. For sheep & beef farms with already very low stocking rates (<5 SU/ha), there was no reduction in the capital value of land and improvements following a 20% reduction in stocking rate. For the Taupo catchment as a whole, this resulted in a capital cost of $45-54m following a 20% reduction in stock numbers. These finding show that a reduction of livestock numbers is a costly option for NZ farmers.

For the dairy farm systems, the Farm Trading Surplus of the 'diet manipulation' and the 'winter management' options were $512/ha and $495/ha, respectively, compared with an FTS of $540/ha for the 'current' system. Thus, the FTS of these management options was reduced by 5 and 8%, respectively. In addition, a capital cost of $400/cow was estimated for the 'winter management' system. For an average dairy farm (100 ha; 2.5 cows/ha), this would result in a capital investment of $100,000. However, de Klein (2001) suggested that a dairy winter management system with a feed pad, could increase pasture production due to an increase in the efficiency of the excreta nitrogen, and also as a result of reduced sward damage from animal treading and compaction. Based on an estimated 5% increase in production for an average New Zealand farm, the return on capital investment was estimated at about 5-9% (de Klein 2001).

4.7 - Conclusions

Management practices that reduce the amount of excreta N produced yield the biggest reduction in N2O emissions.

  • Reducing the amount of excreta N produced by reducing animal numbers is expensive and financially unsustainable.
  • Diet manipulation and cattle winter management show potential as N2O mitigation options, particularly if these where combined as a 'package' of measures.

Management practices to reduce N2O emissions from N fertiliser have a limited impact due to the relatively low usage of N fertiliser in New Zealand.

  • However, their impact is likely to increase with increasing N fertiliser use.

Management practices that impact either on the N2O-to-N2 ratio or on the N2O emission factor of poorly and imperfectly drained soils also show potential as N2O mitigation options

  • Predictions for 2010 suggest that lime management and improved drainage could yield reasonable reductions in N2O emissions.
  • However, the impact of these mitigation options could not be fully estimated due to a lack of comprehensive information.

It is recommended that detailed system analyses be carried out to investigate the effect of 'packages' of measures on N2O emissions.

 

1 Returns after stock purchases and miscellaneous expenditure that is available for principal repayments, management wages, tax on profit and return on investment.

Previous PageTable Of ContentsNext Page

Contact for Enquiries

MAF Information Services
Pastoral House
25 The Terrace
PO Box 2526
Wellington, NEW ZEALAND

Fax: +64 4 894 0721
Contact this person

 
Print this page Print this page




WebSite survey