Chapter 5 - The Measurement of Agricultural Methane and Nitrous Oxide Emissions

Summary

Methods of measuring agricultural methane and nitrous oxide emissions are discussed and evaluated. While calorimetry has been the `gold standard' for measurement of methane output by ruminants, it has significant disadvantages in measuring outputs to simulate the grazing animal. Sulphur hexafluoride tracer technology is the method of choice for the grazing animal.

Nitrous oxide emissions from soils are most frequently measured by chamber techniques. A modification using underground permeable pipes to sample fluxes is being tested at present.

Micrometeorological methods are likely to have principal value in monitoring emissions at a whole farm-scale or larger.

Modeling will continue to be the principal method of estimating emissions for inventory purposes. The benefits of pooling effort in model development should continue to be explored.

The importance of improving the methods for the measurement of methane and nitrous oxide emissions has been highlighted in Chapter 4. Improving the effectiveness and efficiency of measurement is a key to providing credible models that can estimate changes in emission levels in response to the adoption of abatement strategies, and to verify that abatement is occurring.

5.1 Measurement of Methane Emission

For the development of an accurate inventory, or to implement mitigation procedures, it is very important that there be confidence in the accuracy of the methane measurement technology. Many methods have been employed to estimate the methane emitted from ruminant animals and these are outlined below.

5.1.1 Calorimeter/respiration chamber

The classical standard for ruminant methane measurement by nutritionists is the respiration chamber, or calorimeter, and thousands of measurements have been made over the past 150 years, covering a wide range of diets and intake levels. There are many different designs of calorimeters (Blaxter, 1962), but the most common in use today is the open circuit calorimeter. Air is passed through a chamber containing an animal, gas flows and the difference in gas concentrations between inlet and outlet are measured, and the uptake of oxygen and the output of carbon dioxide and methane are calculated. The predominant use of calorimeters has been to measure gaseous exchange as part of energy balance measurements, methane loss being a necessary part of this procedure. Data collected in such feed evaluation experiments has been invaluable in describing relationships between methane emission and diet quality and quantity (e.g. Blaxter & Clapperton 1965; Moe & Tyrell, 1979).

While this technique is satisfactory for measuring methane emission from dried diets, there are difficulties in deriving values that are applicable to the grazing ruminant. Grazing animals select their diet, maximum feed intakes in a chamber are considerably lower than in grazing animals, fresh pasture plants continue to respire in the chamber, calorimeters are expensive to operate and the impact of variations in pasture management can not be addressed. Other techniques are thus preferred for grazing animals, although if possible they should be calibrated against a calorimeter.

5.1.2 Tracer gas techniques

A method for measuring methane emission in the field, known as the sulphur hexafluoride (SF6) tracer technique, was developed at Washington State University by Johnson et al (1994). A calibrated source of SF6 is placed in the rumen per os prior to an experiment, the animal's expired breath is sampled, and the ratio of methane to SF6 determined. The source of SF6 is a permeation tube, and the rate of release of SF6 is controlled by a permeable TeflonTM membrane held in place by a porous stainless steel frit and a locking nut. Each tube is calibrated at 39°C by regular weighing for a period prior to insertion into the rumen. The tubes for sheep, typically 35 mm length by 11 mm in external diameter, are made from brass rods and weigh about 32 g. Each test animal is fitted with a halter, which supports an inlet tube that is placed so that its opening is close to the nose. The inlet tube leads via a capillary tube and shut-off valve to a PVC collection yoke that is fitted over the neck and strapped to the halter. The collection yoke is evacuated prior to use, and the rate at which air is sampled from near the animal's nose is determined by the length and internal diameter of the capillary tube. The yoke is easily isolated for daily changing by means of a shut-off valve and quick connect fittings.

Yoke volumes are typically 1.7 and 2.5 litres for sheep and cattle respectively, and the capillary system is designed to deliver half this volume during the collection period, usually 24 hours. An identical apparatus is placed upwind each day to collect an integrated background air sample. The methane emission rate (QCH4) is calculated as:

QCH4 = QSF6 x ([CH4 sample] - [CH4 ambient])/([SF6 sample] - [SF6 ambient])

where QSF6 is the calibrated rate of permeation from the SF6 tube and [CH4] and [SF6] are concentrations in the collection yokes and background concentrations. Details of tube calibration and behaviour in sheep have been described by Ulyatt et al (1999) and Lassey et al (2001).

The validity of the SF6 technique has been checked in comparisons with respiration chamber measurements. Johnson et al (1994) compared 55 measurements using the SF6 technique with 25 chamber measurements of cattle, and showed that while the SF6 estimates were 0.93 of those in the chamber, the difference was not significant. Pinares-Patiño (2000) in New Zealand found in one experiment with 10 sheep fed chaffed lucerne hay that methane production estimated from SF6 was 0.95 chamber emission.

Similarly, Boadi et al (2002) compared estimates of methane production using the SF6 tracer technique (137.4 l/d) with an open circuit hood calorimeter (130.0 l/d) using yearling beef heifers and found no significant difference (P=0.24). This is what might be expected, given that Murray et al (1976) estimated that greater than 98% of combined rumen and hind gut methane production is excreted via the mouth. However, a number of workers have found difficulty in matching chamber and SF6 results. In two experiments with sheep, Pinares-Patiño (2000) obtained results that were extremely variable. McCrabb and Baker (cited Ulyatt et al, 1999) measured methane production from five Friesian calves fed Rhodes grass (Chloris gayana) hay, using both SF6 and a confinement-type respiration chamber. The estimate of methane production made with the respiration chamber (7.7 ± 0.67 l/h) was twice that (P<0.005) estimated using SF6, either in pens (4.1 ± 0.35 l/h) or in the chamber (4.0 _ 0.46 l/h). Clearly there is a need to confirm that SF6 reliably reflects respiration chamber estimates of methane production.

The reliability of the SF6 technique under field conditions was assessed in the New Zealand work, where gas sampling using the SF6 technique was attempted on several hundred sheep and cow days. The percentage of successful collections was 95% for sheep and 90% for cows. Further, estimates of methane production with the SF6 technique (Johnson et al, 1994; Lassey et al, 1997; Baker, unpublished data) give values between 5-11% of gross energy intake. This is within the range that can be calculated using the IPCC (1995) methodology, which is based on indirect calorimetry. Despite the variation imposed by the various factors noted above, the SF6 tracer technique and gas collection system appears to be robust.

SF6 has also been used successfully as a tracer to estimate the total methane emission from all the cattle in a barn (Kaharabata & Schuepp, 2000).

Ethane (C2H6) has also been used as a marker to estimate methane emission (Moate et al, 1997; Mbanzamihigo et al, 2002) using essentially the same principle as SF6. The major difference in the use of the two tracers is that ethane has to be bubbled from an external source into the rumen, and so is not suitable for use in grazing experiments.

5.1.3 Alternative methods for measuring methane emission in the field

There is a range of meteorological techniques that have been employed to try and validate predictions of greenhouse gas inventories. These can generally be classified as "bottom up", or direct measurement of emissions from a known number of animals at the ground level, or "top down" techniques that infer land-based or area-based emissions from their atmospheric signature (Beswick et al, 1998; Denmead et al, 2000).

5.1.3.1 Bottom up techniques

The SF6 tracer technique, described above, is a bottom up methodology that is often used to validate the more indirect techniques.

Two different versions of mass balance techniques have been tested recently: a portable wind tunnel and a non-intrusive enclosure technique. Lockyer and Jarvis (1995) and Lockyer (1997) described a system in which air was drawn across animals enclosed in a 4.3 x 9.9 m polythene-clad tunnel placed over pasture. Various numbers of sheep and calves were enclosed for up to 10 days, and methane emission was estimated to be on average 13-14 g/d for sheep and 74.5 g/d for calves. In both studies methane emission declined with time, probably in response to declining feed availability given the very high stocking rates employed (470 sheep or calves per ha with two animals in the tunnel and 2 818 sheep/ha with 12 sheep in the tunnel). The method is not suited to evaluating differences between imposed experimental treatments.

Denmead et al (1998), Harper et al (1999) and Leuning et al (1999) described a variant of the mass balance approach in which animals were fenced in a 22 x 22 m enclosure, and gas was sampled from many ports on a framework up to 3.5 m high surrounding the enclosure. Wind speeds were measured from anemometers at the same levels as the sample ports, and from these airflow across each boundary could be measured every 30 seconds. The advantage of this technique is that it can accommodate changes in wind direction. This technique was used by Leuning et al (1999) to measure methane emission for five days from 14 sheep grazing a grass and legume pasture. Seven of these animals were used concurrently to measure methane emission using the SF6 tracer technique. Methane concentration from the sample ports was measured on line by high precision Fourier transform infra-red spectroscopy. The daily mean values for the two techniques were similar: 11.7 g/day for the SF6 technique and 11.9 g/day for the mass balance measurements. With this technique, a very high stocking rate (289 sheep/ha) is required to achieve a differential in gas concentration that can be measured. It is also not suitable for experiments where treatments need to be evaluated.

5.1.3.2 Top down techniques

A range of meteorological techniques has been developed to infer areal emissions from their atmospheric signature (Denmead et al, 2000). These vary in scale from flux gradient analyses designed to measure at the paddock scale, to boundary layer techniques that integrate fluxes over larger areas of the landscape (Beswick et al, 1998; Denmead et al, 2000; Wratt et al, 2001).

5.1.3.2.1 Flux gradient techniques

Judd et al (1999) used a micrometeorological flux gradient technique to estimate methane fluxes for five days across a paddock grazed by sheep. The experimental site was on a flat coastal plain in the Manawatu with a reliable prevailing westerly wind. Samples of air were drawn from two heights (3.8 and 1.2 m) on a tower sited on the downwind boundary of the experimental area. Wind speed and direction were also measured from the tower. Four three ha paddocks upwind of the tower were stocked at 20 sheep/ha, and 11 of these sheep were used to estimate methane emission for the five days of the campaign using the SF6 tracer technique as described by Lassey et al (1997). Thus, the conditions in the upwind measurement footprint were as close as possible to normal farming practice. The methane emission estimated by this technique was 19.5 ± 4.8 g/day, which compared well with the SF6 tracer measurements of 19.4 ± 4.2 g/day. A similar measurement system was described by Denmead et al (2000) who found that good agreement was obtained with inventory predictions, but that the error was too high for detection of small changes that might be important for inventory, regulatory or animal science experimental purposes. The method of Judd et al (1999) is inflexible in that it requires a large fetch of undisturbed air on the upwind side of the sampling tower, it can only be used on rainless days when the wind is in one direction, and it can be affected by the movement of animals within the measurement footprint. It measures from groups of animals and thus is not suited to evaluating differences between treatments.

5.1.3.2.2 Boundary layer techniques

Variations of the boundary layer technique have been used to estimate methane emissions over larger land areas. Basically, vertical profiles of gas concentration are determined through the depth of the atmospheric boundary layer, and this data is used in various modeling techniques to infer emissions over a specified land area. Measurements within both the convective and nocturnal boundary layers have been made with this technique. In one series of experiments in Australia, a 22 m tower and an aircraft were used to collect samples of gas at different heights within the convective boundary layer, and gas budgeting techniques were used to estimate methane emissions (Denmead et al, 2000; Griffith et al, 2002). In another application in New Zealand (Lassey et al, 2000b; Wratt et al, 2001; Gimson et al, 2002) air samples were collected by light aircraft from two columns of air, one upwind and another towards the downwind boundary of the target site.

Both regional budget and source-oriented meteorological and mesoscale dispersion modeling techniques were used to estimate land-based methane emissions. Night time measurements utilising the gas concentrating effect of the nocturnal boundary layer have also been made, usually by taking gas samples at various heights up a profile using a balloon (Denmead et al, 2000; Harvey et al, 2002), or sampling from a tower (Griffith et al, 2002). Harvey et al (2002) have also proposed an isotope dilution/mass balance technique for use in conjunction with the nocturnal boundary layer method.

In a limited number of experiments methane emission has been measured simultaneously at one site with a variety of techniques from the soil, from grazing animals, by micrometeorological flux measurements, and by nocturnal and convective boundary layer measurement (Judd et al, 1999; Denmead et al, 2000). There was reasonable agreement between methods for estimating emission, although the variability increased with scale.

Denmead et al (2000) reviewed the strengths and weaknesses of a range of meteorological flux measurement techniques. Those described above provide estimates of methane emission that have reasonable agreement with inventory estimates. However, the error was generally too high for detection of small changes that might be important for inventory, regulatory or animal science research. There are advantages with these techniques in that they give an integrated net emission of all sources in their "foot print" and they are non-invasive in terms of farm management. These techniques are, however, inflexible in terms of the limited climatic conditions in which they can operate. Despite these caveats, the methods are at an early stage of development and give promise for the future.

5.1.4 Prediction with regression equations

A number of empirical regression equations, based on the results of calorimetric experiments, have been developed for predicting methane production. The two best known equations are (a) those of Blaxter and Clapperton (1965), which are based on data from cattle and sheep and have as independent variates energy digestibility at maintenance, gross energy intake and level of feeding (multiples of maintenance); and (b) the equation of Moe and Tyrell (1979) which are based on data from Holstein dairy cows and have as independent variates the intakes of non-fibre carbohydrate, hemicellulose and cellulose. These and a number of other equations were evaluated for accuracy of prediction of methane production for Holstein cows against a set of data compiled at the USDA, Beltsville by Wilkerson et al (1995). The conclusion was that the equation of Moe and Tyrell (1979) was the most accurate for the prediction of methane production from dry and lactating cows. However, the mean absolute error of prediction (11.0% of the mean) was still high. Further, the equation is probably limited in application to dairy cows confined indoors and fed the typical high concentrate type of diet used in the US.

Recently Yan et al (2000) published a specialised equation for dairy and beef cattle offered grass silage-based diets where the independent variates are digestible energy intake, the proportion of ADF in the diet, and the level of feeding above maintenance. The R2 was 0.89, indicating a reasonably good fit. Pelchen and Peters (1998) developed equations for sheep from 1 137 sets of data from the literature. There was a wide range of diet types, including a few from animals fed fresh herbage. By grouping the sets of data according to criteria such as digestibility (<65, 65-70, 70-75, >75) and crude fibre (<15, 15-20, 20-25, 25-30, >30) they derived regressions for predicting methane emission (g/d) with R2=0.7-0.85. Pelchen et al (1998) developed similar regression equations for predicting methane emission from dairy cows using 729 sets of data from the literature.

The equations described above are useful in examining the relationships between dietary constituents and methane emission, however they are not accurate enough for use in determining small differences between animals in experiments. Also, being derived predominantly from animals fed indoors on preserved rations, they do not satisfactorily measure methane emission from animals grazing fresh pasture.

5.1.5 Prediction with dynamic mathematical models

Dynamic mechanistic models of rumen digestion have been developed that will predict methane emission from an input of diet intake and composition. Ulyatt et al (1991) used the model of Baldwin et al (1987) to predict methane emission for a model of the New Zealand inventory. Ulyatt et al (1991) compared the Baldwin model predictions against predictions with the Blaxter and Clapperton (1965) and Moe and Tyrell (1979) regressions, using a standard set of input data. They found that while there was bias in the Baldwin model predictions, the bias was greater with the two regression equations. Benchaar et al (1998) challenged an updated version of the Baldwin et al (1987) model, a modified version of the Dijkstra et al (1992) dynamic model and the Blaxter and Clapperton (1965) and Moe and Tyrell (1979) regressions with 32 experimental rations with large variations in diet composition.

The Dijkstra model predicted methane production with an R2 of 0.71, compared to an R2 of 0.70 with the Baldwin model. However, the Baldwin model had a higher error of prediction (36.9 vs 19.9%). Predictions with the Blaxter and Clapperton (1965) and Moe and Tyrell (1979) regressions were poor, with R2=0.57 and 0.42 respectively. Mills et al (2001) also developed a dynamic methane model that predicted results from the literature with R2= 0.76. It appears that dynamic mechanistic models can predict methane production with better accuracy than the best of the available empirical regression equations. This might be satisfactory for inventory purposes where resources are limited, however these models are very complicated and require a skilled operator. Further they are not appropriate for evaluating differences between experimental treatments.

5.1.6 Choice of a method for grazing animals

In terms of accuracy, ability to measure at the individual grazing animal level, and usefulness in evaluating experimental treatments, there can be no doubt that the SF6 tracer technique is the method of choice. There are, however, issues with this technique that require attention:

  • There have been concurrent comparisons with the respiration chamber, but these have generally been less than convincing. Further work is required to absolutely confirm that SF6 is accurate.
  • The technique is reliant for its efficacy on one experiment with one diet (Murray et al, 1976) with regard to the proportion of large intestinal methane that is absorbed and excreted via the lungs.
  • The question of whether allowance should be made for excretion in the flatus, and if so how much, is still to be resolved.
  • Variation between animals in methane emission appears to be much higher with the SF6 technique than respiration chamber measurements (Boadi et al, 2002). Is this a true effect or is it an artefact of the SF6 procedure?
  • The SF6 measurement is relatively expensive to conduct.
  • All the above issues need addressing if the SF6 technique is to become universally accepted.

5.1.7 Conclusions

Although many methods have been employed to measure or calculate methane emissions from ruminants, the method of choice for grazing animals is the SF6 tracer technique. The other methods have too high an error to be able to detect the small differences that are important in inventory, regulatory or animal science research. Metereorological methods have the advantage that they can integrate over large land areas, but a lot more methodological development will be necessary before they can become accurate and routine. Further work is also needed on the SF6 technique: rigorous evaluation against the respiration calorimeter; confirmation of the proportion of methane that is excreted in the flatus; and evaluation of the high variability of the technique.

References

Baldwin RL, Thornley JHM & Beever DE (1987): Metabolism of the lactating cow. II - Digestive elements of a mechanistic model. Journal of Dairy Research, 54: 107.

Benchaar C, Rivest J, Pomar C & Chiquette J (1998): Prediction of methane production from dairy cows using existing mechanistic models and regression equations. Journal of Animal Science, 76: 617-627.

Beswick KM, Simpson TW, Fowler D, Choularton TW, Gallagher MU, Hargreaves KJ, Sutton MA & Kaye A (1998): Methane emissions on large scales. Atmospheric Environment, 32: 3283-3291.

Blaxter KL (1962): The Energy Metabolism of Ruminants. Hutchinson, London. 329p.

Blaxter KL & Clapperton JL (1965): Prediction of the amount of methane produced by ruminants. British Journal of Nutrition, 19: 511-522.

Boadi DA, Wittenberg KM & Kennedy AD (2002): Validation of the sulphur hexafluoride (SF6) tracer gas technique for measurement of methane and carbon dioxide production by cattle. Canadian Journal of Animal Science, 82: 125-131.

Denmead OT, Harper LA, Freney JR, Griffith DWT, Leuning R & Sharpe RR (1998): A mass balance method for non-intrusive measurement of surface-air trace gas exchange. Atmospheric Environment, 32: 3679-3688.

Denmead OT, Leuning R, Griffith DWT, Jamie IM, Esler MB, Harper LA & Freney JR (2000): Verifying inventory predictions of animal methane emissions with meteorological measurements. Boundary-Layer Meteorology, 96: 187-209.

Dijkstra J, Neal HDSC, Beever DE & France J (1992): Simulation of nutrient digestion, absorption and outflow in the rumen: model description. Journal of Nutrition, 122: 2239-2256.

Gimson NR, Lassey KR, Brailsford GW, Bromley AM & Uliasz M (2002): The determination of agricultural methane emission fluxes based on air sampling and advanced modeling techniques. In Proceedings of the Third International Symposium on Non-CO2 Greenhouse Gases: 1-6. Maastricht, Netherlands.

Griffith DWT, Leuning R, Denmead OT & Jamie IM (2002): Air-land exchanges of CO2, CH4 and N2O measured by FTIR spectrometry and micrometeorological techniques. Atmospheric Environment, 36: 1833-1842.

Harper LA, Denmead OT, Freney JR & Byers FM (1999): Direct measurements of methane emissions from grazing and feedlot cattle. Journal of Animal Science, 77: 1392-1401.

Harvey MJ, Brailsford GW, Bromley AM, Lassey KR, Mei Z, Kristament IS, Reisinger AR, Walker CF & Kelliher FM (2002): Boundary-layer isotope dilution/mass balance methods for measurement of nocturnal methane emissions from grazing sheep. Atmospheric Environment, 36: 4663-4678.

Johnson K, Huyler M, Westberg H, Lamb B & Zimmerman P (1994): Measurement of methane emissions from ruminant livestock using an SF6 tracer technique. Environmental Science and Technology, 28: 359-362.

Judd MJ, Kelliher FM, Ulyatt MJ, Lassey KR, Tate KR, Shelton ID, Harvey MJ & Walker CF (1999): Net methane emissions from grazing sheep. Global Change Biology, 5: 647-657.

Kaharabata SK & Schuepp PH (2000): Estimating methane emissions from dairy cattle housed in a barn and feedlot using an atmospheric tracer. Environmental Science and Technology, 34: 3296-3302.

Lassey KR, Ulyatt MJ, Martin RJ, Walker CF & Shelton ID (1997): Methane emissions measured directly from grazing livestock in New Zealand. Atmospheric Environment, 31: 2905-2914.

Lassey KR, Gimson NR, Wratt DS, Brailsford GW & Bromley AM (2000): Verifying agricultural emissions of methane. In Non-CO2 Greenhouse gases: Scientific Understanding, Control and Implementation: 107-114. (Eds) van Ham J et al. Kluwer, Netherlands.

Lassey KR, Walker CF, McMillan AMS & Ulyatt MJ (2001): On the performance of SF6 permeation tubes used in determining methane emission from grazing livestock. Global Change Science, 3: 367-376.

Leuning R, Baker SK, Jamie IM, Hsu CH, Klein L, Denmead OT & Griffith DWT (1999): Methane emission from free-ranging sheep: a comparison of two measurement methods. Atmospheric Environment, 33: 1357-1365.

Lockyer DR (1997): Methane emissions from grazing sheep and calves. Agriculture, Ecosystems and Environment, 66:11-18.

Lockyer DR & Jarvis SC (1995): The measurement of methane losses from grazing animals. Environmental Pollution, 90: 383-390.

Mbanzamihigo L, Fievez V, da Costa Gomez C, Piattoni F, Carlier L & Demeyer D (2002): Methane emissions from the rumen of sheep fed a mixed grass-clover pasture at two fertiliser rates in early and late season. Canadian Journal of Animal Science, 82: 69-77.

Moate PJ, Clarke T, Davis LH & Laby RH (1997): Rumen gases and bloat in grazing dairy cows. Journal of Agricultural Science, 129: 459-469.

Moe PW & Tyrell HF (1979): Methane production in dairy cows. Journal of Dairy Science, 62: 1583-1586.

Murray RM, Bryant AM & Leng RA (1976): Rates of production of methane in the rumen and large intestine of sheep. British Journal of Nutrition, 36: 1-14.

Pelchen A & Peters KJ (1998): Methane emissions from sheep. Small Ruminant Research, 27: 137-150.

Pelchen A, Peters KJ & Holter JB (1998): Prediction of methane emissions from lactating dairy cows. Archiv furTierzucht, 41: 553-563.

Pinares-Patiño C (2000): Methane Emission from Forage-fed Sheep: A Study of Variation between Animals. PhD thesis, Massey University. 226p.

Ulyatt MJ, Betteridge K, Costall D, Knapp J & Baldwin RL (1991): Methane Production by New Zealand Ruminants. Report prepared for the Ministry for the Environment. 27p.

Ulyatt MJ, Baker SK, McCrabb GJ & Lassey KR (1999): Accuracy of the SF6 tracer technology and alternatives for field measurements. Australian Journal of Agricultural Research, 50: 1329-1334.

Wilkerson VA, Casper DP & Mertens DR (1995): The prediction of methane production of Holstein cows by different equations. Journal of Dairy Science, 78: 2402-2414.

Wratt DS, Gimson NR, Brailsford GW, Lassey KR, Bromley AM & Bell MJ (2001): Estimating regional methane emissions from agriculture using aircraft measurements of concentration profiles. Atmospheric Environment, 35: 497-508.

Yan T, Agnew RE, Gordon FJ & Porter MG (2000): Prediction of methane energy output in dairy and beef cattle offered grass silage-based diets. Livestock Production Science, 64: 253-263.

5.2 Measuring Nitrous Oxide Emission from Soil

Emissions of nitrous oxide from soil are difficult to measure accurately because of the small amounts emitted, the large variation which occurs over the soil surface, and the marked changes which take place with time. The variability is due mainly to the variation in the factors controlling metabolism of nitrogen across the landscape such as the microorganisms, substrate for the organisms, water, temperature, oxygen and pH.

5.2.1 Chamber methods

Most of the information on rates of nitrous oxide emission from soils has been obtained by placing simple, relatively small chambers on the soil surface, and measuring the increase in nitrous oxide concentration within the chamber after a set time (Mosier, 1989; IAEA, 1992; Smith & Arah, 1992; Ryden & Rolston, 1983; Granli & Böckman, 1994; Fowler et al, 1997). A variety of chambers have been designed and labelled as closed, open, mega, and vented (Berges & Crutzen, 1996; Meixner et al, 1997; Sibbetsen & Lind, 1993; Smith et al, 1994; Velthof et al, 1997). In addition automated closed chamber methods have been developed (Meyer et al, 2001; Smith & Dobbie, 2001) which enable the determination of fluctuations in nitrous oxide fluxes not seen by manual sampling.

The nitrous oxide concentration in the chamber is usually determined by taking samples with a syringe at particular times and analysing them by gas chromatography using a Ni63 electron capture detector operated at a temperature between 300 and 400 ºC (Mosier & Mack, 1980). The concentration has also been monitored by employing an infrared gas analyser (Denmead, 1979) and Fourier transform infrared (FTIR) spectroscopy in a closed loop (Meyer et al, 2001).

Enclosure methods have the advantage of being able to detect small fluxes, are inexpensive, and are useful for short-term process studies. They have many disadvantages, including changes to the microclimate, but the main one is that measurements are obtained over a small area. When coupled with the large spatial variability in nitrous oxide emission, this makes it difficult to obtain meaningful values for emission from a field (Matthias et al, 1978; IAEA, 1992; Livingston and Hutchinson, 1995; Smith et al, 1995). The problem can be overcome to some extent by using more chambers and increasing their size (Smith et al, 1994).

Sherlock et al (2002) describe a closed chamber technique which is being used within NzOnet to measure nitrous oxide emission from pasture soils. Microplots are formed by pushing steel cylinders (24 cm diameter) into the soil. Cylindrical gas-tight lids, having a headspace height of 10 cm, and fitted with rubber septa for sampling, are attached to the steel cylinders during gas emission measurements, but are left open between measurements. The changes in headspace concentrations are used for calculating the flux of nitrous oxide. Gas samples are collected with 50 ml syringes at 0, 10 and 20 minutes after lid closure. The gas samples are injected via a 10-port sampling valve into a carrier stream of nitrogen, to a gas chromatograph equipped with a 63Ni electron-capture detector and a stainless steel column (4 m long, 3 mm internal diameter) packed with Poropak Q (80/100 mesh). Detector and column temperatures are 350oC and 20oC respectively. In general, fluxes are calculated using the logarithmic equation described by Hutchinson and Mosier (1981).

5.2.2 Micrometeorological methods

Micrometeorological methods are also employed for measuring nitrous oxide emissions from field soils, although less frequently than chamber methods. They are extremely useful in that they integrate emissions over large areas, and can assess the effect of rainfall, temperature and wind speed on emission (Fowler & Duyzer, 1989). As with the chamber methods, a variety of micrometeorological methods are available.

The mass balance method is non-disturbing, does not require a large fetch or uniform land area, has a simple theoretical basis, and can be used to measure gas emission from small plots. The idea is to calculate gas production from a particular area by measuring the difference between the rates at which gas is transported into and out of an area. Denmead et al (2000) used such a method to determine the nitrous oxide resulting from the grazing of sheep on a lucerne (Medicago sativa)-ryegrass (Lolium rigidum) pasture in a test plot, 22 x 22 m. Measurements of atmospheric nitrous oxide concentrations were made with Fourier transform infrared (FTIR) spectroscopy. They found that the average emission over eight days was 1.87 g N2O-N per head per day, which corresponded to 13.9% of the nitrogen excreted by the animals. Brown et al (2002) measured nitrous oxide emission from a solid dairy manure pile with a mass balance method using a tunable diode laser trace gas analyser which provided high precision concentration difference measurements. Loss from the pile was 0.42 g N m-2 day-1.

Wagner-Riddle et al (1996) used a flux-gradient method to measure fluxes of nitrous oxide from a bare soil. In this technique, changes in atmospheric nitrous oxide concentrations with height above the surface are measured. They used a tunable diode laser system for rapid measurement of nitrous oxide concentrations at only two heights, switching between them two every four seconds. Denmead et al (2000), Griffith and Galle (2000) and Griffith et al (2002) used the same approach, but measured concentrations with FTIR spectroscopy.

Griffith et al (2002) measured vertical profiles of nitrous oxide every 30 minutes from the ground up to 22 m so they could use a nocturnal boundary layer method to measure nitrous oxide fluxes. In this method the rate of change in mass storage in the 0-22 m layer is combined with fluxes measured at 22 m to estimate surface fluxes. Night time fluxes of nitrous oxide were 2_3.2 ng N m-2 s-1 which were in good agreement with chamber measurements and inventory estimates based on stocking rates in the region. For comparison day time rates were 17_48 ng N m-2 s-1.

A second night time micrometeorological method has been proposed for measuring the nitrous oxide emission rate from grazed pasture (Kelliher et al, 2002). . This method used Fourier-transform infrared spectroscopy to simultaneously monitor concentrations of nitrous oxide and carbon dioxide over a 97 m long, open-air absorption path at a height of 3 m in the stable boundary layer. On calm and clear nights, the formation of an inversion layer trapped surface gas emissions and led to a build-up of N2O and CO2 concentrations near ground level. The ratio of these concentrations was combined with the more readily measured CO2 emission rate to calculate an area-integrated N2O emission rate. The increases in nitrous oxide and carbon dioxide concentrations in the stable boundary layer concentration were highly correlated (r2=0.83, n=201).

References

Brown HA, Wagner-Riddle C & Thurtell GW (2002): Nitrous oxide flux from a solid dairy manure pile measured using a micrometeorological mass balance method. Nutrient Cycling in Agroecosystems, 62: 53-60.

Berges MG & Crutzen PJ (1996): Estimates of global N2O emissions from cattle, pig and chicken manure, including a discussion of CH4 emissions. Journal of Atmospheric Chemistry, 24: 241-269.

Denmead OT (1979): Chamber systems for measuring nitrous oxide emissions from soils in the field. Soil Science Society of America Journal, 43: 89-95.

Denmead OT, Leuning R, Jamie I & Griffith DWT (2000): Nitrous oxide emissions from grazed pastures: measurements at different scales. Chemosphere: Global Change Science, 2: 302-312.

Fowler D & Duyzer JH (1989): Micrometeorological techniques for the measurement of trace gas exchange. In Exchange of Trace Gases between Terrestrial Ecosystems and the Atmosphere: 189-207. (Eds) Andrae MO & Schimel DS. John Wiley & Sons, Chichester.

Fowler D, Skiba U & Hargreaves KJ (1997): Emissions of nitrous oxide from grasslands. In Gaseous Nitrogen Emissions from Grasslands: 147-164. (Eds) Jarvis SC & Pain BF. CAB International, Wallingford, UK.

Granli T & Bøckman OC (1994): Nitrous oxide from agriculture. Norwegian Journal of Agricultural Science: Supplement No. 12. 128p.

Griffith DWT, Leuning R, Denmead OT & Jamie IM (2002): Air-land exchanges of CO2, CH4 and N2O measured by FTIR spectrometry and micrometeorological techniques. Atmospheric Environment, 36: 1833-1842.

Griffith DWT & Galle B (2000): Flux measurements of NH3, N2O and CO2 using dual beam FTIR spectroscopy and the flux gradient technique. Atmospheric Environment, 34: 1087-1098.

Hutchinson GL & Mosier AR (1981): Improved soil cover method for field measurement of nitrous oxide fluxes. Soil Science Society of America Journal, 45: 311-316.

IAEA (1992): Manual on Measurement of Methane and Nitrous Oxide Emissions from Agriculture. International Atomic Energy Agency, Vienna. 91p.

Kelliher FM, Reisinger AR, Martin RJ, Harvey MJ, Price SJ & Sherlock RR (2002): Measuring nitrous oxide emission rate from grazed pasture using Fourier-transform infrared spectroscopy in the nocturnal boundary layer. Agricultural and Forest Meteorology, 111: 29-38.

Livingston GP & Hutchinson GL (1995): Enclosure-based measurement of trace gas exchange: application and sources of error. In Biogenic Trace Gases: Measuring Emissions from Soil and Water: 14-51. (Eds) Matson PA & Harriss RC. Blackwell Science, Oxford.

Matthias AD, Yarger DN & Weinbeck RS (1978): A numerical evaluation of chamber methods for determining gas fluxes. Geophysical Research Letters, 5: 765-768.

Meixner FX, Fickinger Th, Marufu L, Mukurumbira L, Makina E, Nathaus FJ & Serca D (1997): Biogenic emission of nitric oxide from a subtropical grassland: Miombo-type Savanna (Marondera, Zimbabwe). In Gaseous Nitrogen Emissions from Grasslands: 37-46. (Eds) Jarvis SC & Pain BF. CAB International, Wallingford, UK.

Meyer CP, Galbally IE, Wang YP, Weeks IA, Jamie IM & Griffith DWT (2001): Two automatic chamber techniques for measuring soil-atmosphere exchanges of trace gases and results of their use in the Oasis field experiment. In CSIRO Atmospheric Research Technical Paper: No. 51. 30p.

Mosier AR (1989): Chamber and isotope techniques. In Exchange of Trace Gases Between Terrestrial Ecosystems and the Atmosphere: 7-21. (Eds) Andreae MO & Schimel DS. John Wiley & Sons, Chichester.

Mosier AR & Mack L (1980): Gas chromatographic system for precise, rapid analysis of N2O. Soil Science Society of America Journal, 44: 1121-1123.

Ryden JC & Rolston DE (1983): The measurement of denitrification. In Gaseous Loss of Nitrogen from Plant Soil Systems: 91-132. (Eds) Freney JR & Simpson JR. Martinus Nijhoff & Dr W Junk Publishers, The Hague.

Sherlock RR, Sommer SG, Khan RZ, Wood CW, Guertal EA, Freney JR, Dawson CO & Cameron KC (2002): Emission of ammonia, nitrous oxide and methane from pig slurry applied to a pasture in New Zealand. Journal of Environmental Quality, 31: 1491-1501.

Sibbesen E & Lind AM (1993): Loss of nitrous oxide from animal manure in dungheaps. Acta Agricultural Scandinavica, Section B: Soil Plant Science, 43: 16-20.

Smith KA & Arah JRM (1992): Measurement and modeling of nitrous oxide emissions from soils. Ecological Bulletins, 42: 116-123.

Smith KA & Dobbie K (2001): The impact of sampling frequency and sampling times on chamber-based measurements of N2O emissions from fertilised soils. Global Change Biology, 7: 933-945.

Smith KA, Scott A, Galle B & Klemedtsson L (1994): Use of a long-path infrared gas monitor for measurement of nitrous oxide from soil. Journal of Geophysical Research, 99: 16,585-16,592.

Smith KA, Clayton H, Taggart IP, Thomson PE, Arah JRM & Scott A (1995): The measurement of nitrous oxide emissions from soil by using chambers. Philosophical Transactions of the Royal Society, Series A, 351: 327-338.

Velthof GL, Jarvis SC & Oenema O (1997): Spatial variability of nitrous oxide fluxes from grasslands: differences between soil types. In Gaseous Nitrogen Emissions from Grasslands: 173-179. (Eds) Jarvis SC & Pain BF. CAB International, Wallingford, UK.

Wagner-Riddle C, Thurtell GW, King KM, Kidd GE & Beauchamp EG (1996): Nitrous oxide and carbon dioxide fluxes from a bare soil using a micrometeorological approach. Journal of Environmental Quality, 25: 898-907.

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