© JTI – Institutet för jordbruks- och miljöteknik 2000
In accordance with the Copyright Act, it is forbidden to copy any part of this document without the expressed written permission of the copyright holder.
Print: JTI – Institutet för jordbruks- och miljöteknik, Uppsala 2000 ISSN 1401-4963
Nr 274
A review of methods for measuring
methane, nitrous oxide and odour emissions
from animal production activities
Contents
Förord... 5
Summary... 7
Sammanfattning ... 8
Introduction...9
Emissions of greenhouse gases ... 9
Odour emissions... 12
Measurement of greenhouse gases and odours ... 13
Abbreviations ... 13
Methods for measuring N2O and CH4 fluxes... 13
Static chambers ... 13 Advantages ... 14 Disadvantages ... 14 Dynamic chambers ... 15 Advantages ... 16 Disadvantages ... 16 Micrometeorological techniques ... 16 Advantages ... 17 Disadvantages ... 17 Tracer methods ... 18 Advantages ... 20 Disadvantages ... 21
Comparison of analytical methods ... 21
Gas chromatography ... 21
Infrared photoacoustic spectrometer – trace gas analyser... 22
Fourier transform infrared absorption spectroscopy ...22
Tuneable Diode Laser Absorption Spectroscopy... 23
Odour measurement ... 23 Electronic nose ... 23 Olfactometry... 24 Odour sampling ...24 Conclusions... 25 Plot experiments ... 25
Manure heaps and stores...25
Field scale measurements ...25
Animals and animal housing ...25
Odour measurement...26
Förord
JTI:s studier av gasformiga emissioner från lagring och spridning av stallgödsel har de senaste åren till stor del koncentrerats på ammoniak. International Panel on Climate Change (IPCC), där Sverige deltar, vill fokusera uppmärksamheten på växthusgaser och deras betydelse för klimatförändringar. Förutom koldioxid betraktas lustgas och metan som viktiga växthusgaser. Dessa växthusgaser har studerats mindre i Sverige än i andra länder, men är viktiga genom sin påverkan på klimatet. Enligt skattningar från SCB står jordbruket för mellan 60 och 65 % av de svenska utsläppen av metan och lustgas. Därför är det intressant för JTI att utreda potentialen för nya forskningsprojekt inom detta område. Idisslare och stallgödselhantering är två av de största källorna till utsläpp av metan och lustgas. Ett annat ämne som rönt allt större intresse på senare år är mätning av lukt. JTI har jobbat med detta under 1970-talet, men aktiviteterna har legat nere de senaste 15-20 åren. En kort redovisning av det som har hänt nyligen inom området inklu-deras här.
Rapporten, som representerar ett första steg i kompetensuppbyggnaden, beskriver metoder som används av andra forskare, särskilt utomlands, för att mäta emis-sioner av lustgas, metan och lukt från olika lantbruksaktiviteter. Arbetet är en litteraturstudie och har genomförts av bitr. forskningsledare Jim Greatorex vid JTI. Projektledare har varit forskningschef Johan Malgeryd. Rapporten redovisar de metoder som bedöms som mest aktuella för JTI att använda i potentiella projekt, t.ex. inventering av emissioner och utvärdering av åtgärder för att minska utsläpp av metan, lustgas och lukt från animalieproduktion.
Ultuna, Uppsala i november 2000
Lennart Nelson
Summary
The aim of this literature study is to review the currently available knowledge on methodologies for the determination of gaseous emissions from livestock pro-duction. The emphasis is on methods for measuring fluxes of nitrous oxide and methane, but odour emissions are also included to a smaller extent. A discussion of the analytical methods available for quantifying N2O and CH4 is included. Various techniques are considered for the measurement of emissions from field trial plots, larger field-scale experiments, manure stores, animal buildings and directly from the animals themselves. The choice of technique depends to a large extent on cost, level of accuracy required, and the scale and design of the experi-ments to be undertaken.
On field plot experiments it is recommended that automated, dynamic chambers are used since they are well suited to comparisons of different treatments, mini-mise disturbance to the soil environment and allow increased sampling frequency compared with manual chambers.
The use of sulphur hexafluoride based, tracer ratio methods has great potential. Such methods have been used to determine CH4 fluxes from free ranging rumi-nants, as well as emissions from local sources such as livestock buildings, manure stores and lagoons.
Field-scale experiments require the use of micrometeorological methods. The most cost-effective technique would be to use the gradient method, with analyses being made by Fourier transform infrared spectroscopy.
Odour emissions are currently characterised at JTI in terms of odour strength by use of the pyridine method. It should be recognised that draft CEN standards may need to be adopted in the future which make use of the threshold dilution method. Electronic noses are considered, but more work is required to ascertain whether reliable relationships exist between odour concentration and sensor response for a range of agricultural odours.
Sammanfattning
Syftet med litteraturstudien var att samla aktuell kunskap om metodik för att mäta gasemissioner från lantbruket, i första hand från djurproduktion. Studien fokuseras på metoder för att mäta emissioner av lustgas och metan, men även luktemissioner inkluderas i viss mån. En kort redovisning av de analytiska meto-der som används till kvantifiering av N2O och CH4ingår också.
Diverse metoder för mätning av emissioner från fältförsöksrutor, större fältskale-försök, stallgödsellager, djurinhysningar och direkt från djuren diskuteras. Vilken mätteknik som är mest lämplig beror på vad man ska mäta på och kravet på nog-grannhet samt försöksupplägg.
När det gäller parcellförsök i fält rekommenderas att automatiska dynamiska kammare används eftersom de passar bra vid en jämförelse mellan olika behand-lingar, ger mindre störningar i markmiljön och möjlighet till högre provtagnings-frekvens jämfört med manuella kammare.
Spårgasmetoder baserade på svavelhexafluorid har stora potentiella möjligheter och har använts för att bestämma CH4-flöden från frigående idisslare och lokala källor såsom djurstallar och stallgödsellager.
Mikrometeorologiska metoder är mest lämpliga vid mätning i större skala i fält. Den mest kostnadseffektiva mätmetoden är i detta fäll den s.k. gradientmetoden med analyser utförda med FTIR (Fourier Transform Infrared Spectroscopy). För närvarande mäter JTI luktstyrkan genom att jämföra dess intensitet med bestämda koncentrationer av gasen pyridin. Det finns dock ett utkast till en ny CEN-standard enligt vilken en utspädningsmetod ska användas för framtagning av s.k. tröskelvärden. I framtiden bör JTI anpassa metodiken till CEN-standarden. Ett annat sätt är att mäta lukt med hjälp av s.k. elektroniska näsor. Mer arbete behövs dock för att utreda om det finns säkra samband mellan luktstyrka och sensorernas respons för olika lukter i lantbruket.
Introduction
This report reviews literature concerning the measurement of N2O, CH4 and odour emissions from animal production activities. There is increasing interest in obtain-ing accurate measurements of N2O and CH4 because of their contribution to climate change through their role as greenhouse gases. Odour emissions are a continual source of complaint from people living close to animal production activities. It is therefore important to quantify odours in order to assess the effectiveness of odour reducing measures.
Emissions of greenhouse gases
During the last 150 to 200 years, human activity has increased the atmosphere’s content of carbon dioxide by 30 %, methane by 145 % and nitrous oxide by 15 % (Regeringen, 1998, after data from the International Panel on Climate Change). In Section 4.2.15 of a bill delivered to the Swedish Parliament in 1998 (Regeringen, 1998), the Swedish Government proposed an environmental quality objective entitled, “Limited influence on climate”. The purpose of this is that the concentra-tions of greenhouse gases in the atmosphere should be stabilised at levels where human impact will not have a harmful effect on climate systems. The objective is based upon global objectives established by the UN Framework Convention on Climate Change (SÖ, 1993).
The objective, “Limited influence on climate” means that measures must be con-centrated on stabilising the levels of carbon dioxide in the atmosphere at less than 550 ppm, and ensuring that levels of other greenhouse gases in the atmosphere do not increase. The Government’s assessment is that this objective needs to be supple-mented with sub-objectives relating to emissions of carbon dioxide and other green-house gases (including N2O and CH4).
Table 1. Estimated emissions of greenhouse gases in Sweden 1998 (Source: Miljöstatistiken, SCB Statistics Sweden) Emissions, kTonnes CO2 CH4 N2O NOX CO Energy industry 9 810 2 1 14 13 Production industry 12 270 6 3 51 51 Transport 21 140 16 2 142 767
Small scale combustion 9 650 10 0 38 126
Industrial processes 4 120 3 12 13
Solvents 300
Agriculture excl. combustion & transport 159 16
Landfill 61
The preceding table shows the estimated amounts of greenhouse gases emitted in Sweden from different human activities1. These amounts indicate that agriculture, excluding combustion and transport, is a significant source of both methane and nitrous oxide. It should be noted that the size of the greenhouse effect varies for different greenhouse gases. In order to compare different greenhouse gases, it is common to talk about their Global Warming Potential measured over a 100 year period (GWP100). GWP100 describes how a gas influences the climate in relation to carbon dioxide, which is arbitrarily given the value 1. The GWP100 values of methane and nitrous oxide are 21 and 310, respectively. Another way in which to express the GWP of a gas is as carbon dioxide equivalents, i.e., 1 tonne of methane has the same effect on the climate as 21 tonnes of carbon dioxide, etc.
Figure 1. Contribution of different sectors to Sweden’s greenhouse gas emissions in 1998, calculated as carbon dioxide equivalents (Source: Sweden’s official report to the Climate Convention, SCB Statistics Sweden).
The diagram shows that the overall contribution of agriculture to climate change in Sweden is smaller than that of transport and combustion. However, at 12 %, agriculture represents a significant source of climate gases which should be taken into consideration when exploring ways of reducing climate gas emissions. Addi-tionally, it should be remembered that nitrous oxide is a principal component of the agricultural emission of climate gases, and this not only represents a nutrient loss from plant producing systems, but is also known to deplete atmospheric ozone (McElroy et al., 1977).
Emission of climate gases is a global issue and therefore Sweden can only con-tribute to the achievement of its own environmental quality objectives. Therefore in addition to acting at the national level, the Swedish government is also actively involved in international forums to reduce emissions of climate gases. The EU has reached an agreement on the reduction of six different climate gases, includ-ing methane and nitrous oxide. The agreement is known as “the burden sharinclud-ing agreement” (COM(1999)230) and it commits EU countries to a combined
1 The quality of the data taken from Statistics Sweden is open to question since, according to Adolfsson (2000), very little experimental work has been undertaken to verify it.
Other 2% Combustion 45% Transport 28% Machinery 6% Industrial processes 7% Agriculture 12%
tion of the six gases by 8 % of 1990 levels. The target should be achieved under the period 2008 to 2012 (Regeringen, 1998).
The Government’s environmental quality objective, “Limited influence on climate” (Regeringen, 1998) recognises that the dominant sources of methane emission in Sweden are landfills and agriculture (primarily rumination and manure handling). Emission sources of nitrous oxide are not well studied and the most important global sources are believed to be storing and utilisation of manure, combustion of fuels and production of mineral fertilisers. The objective goes onto describe the dominant source of nitrous oxide emission in Sweden as combustion (Regeringen, 1998), although this appears to conflict with data from Statistics Sweden (see Table 1).
The Swedish parliament has already taken action to reduce methane emissions by 30 % before the year 2000 (Regeringens proposition 1992/93:179). According to Naturvårdsverket, considerable progress has been made in meeting this objective, at least with regard to methane emissions from landfills. Total emissions have fallen by 28 % since 1990, however, the objective remains as one of the Govern-ment’s environmental quality objectives (Regeringen, 1998).
Only limited data concerning emissions of nitrous oxide and methane under Swedish conditions are available, for example Svensson (1999); Svensson and Lindén (1998). However an inventory for nitrous oxide emissions from the UK has been constructed (Chadwick et al., 1999). The principal sources of N2O emis-sion from UK agriculture are spreading of mineral fertilisers (22.66 kt p.a.) and manure stores (5.61 kt p.a.). Manure spreading accounts for 1.12 kt p.a., whilst livestock buildings and outdoor livestock account for 4.92 kt p.a. and 3.96 kt p.a., respectively. Chadwick et al. (1999) state that they lack conclusive evidence about the effect on N2O emissions of injecting slurries rather than applying them to the soil surface. However, they point to a method of Kroeze (1995) for estimating N2O emission where a greater proportion of N2O (1.25 to 2.5 % of added N) is lost after injection than after surface application (0.2 to 1.25 % of added N). Methods suggested by Chadwick et al. (1999) for abating N2O emissions include:
• Switching from straw-based cattle systems to slurry based systems. The anaerobic environment in slurries would inhibit nitrate production and hence formation of N2O. This might occur at the expense of increased CH4 emissions unless preventative action was taken (e.g., covering stores and generating bio-gas).
• Reducing the crude protein to animals and supplementing with specific amino acids. Spreading the slurry from pigs fed with this diet has been shown to give reduced CH4 (as a consequence of the reduced volatile fatty acid content), denitrification and NH3 losses, whilst improving utilisation of slurry NH4+-N (Misselbrook et al., 1998).
• Restrictions on time and rate of application so as to avoid loading N when crop requirements are at a minimum and conditions are favourable for denitrifica-tion, e.g., in autumn and early spring.
In the case of methane, Bouwman (1990) states that this is the second most important greenhouse gas, contributing approximately 20 % to global warming. Atmospheric CH4 concentrations are increasing by about 1.1 % p.a., although the rate of increase was reported to have slowed in the 1990’s, with the primary causes being an increase in the numbers of ruminant livestock and increased production of wetland rice (Chadwick and Pain, 1997). Estimates made by Safley et al. (1992) indicate that CH4 emissions from livestock wastes may contribute up to 10% of the total anthropogenic emission of CH4.
Soils have been shown to act as sinks for CH4 and it is thought that they may consume between 1 and 10 % of total global CH4 emissions (Adamsen and King, 1993). Chadwick and Pain (1997) state that there is evidence for the inhibition of this soil CH4 sink by addition of nitrogen or through cultivation of previously undisturbed soil. Hansen et al. (1993) also found that the addition of fertiliser nitrogen led to a decrease in CH4 uptake by the soil. This was attributed to competition between NH3 and CH4 for the same active site on monooxygenase enzymes which catalyse the first oxidation step of CH4 and NH4+ in methano-trophs and nitrifiers. Sitaula et al. (1995) found that application of ammonium nitrate (30 kg N ha-1) reduced CH4 uptake to 85 % of an unfertilised control, while an application of 90 kg N ha-1 reduced the CH4 uptake rate to 62 %. Hansen et al. (1993) also found that soil compaction reduced soil uptake of CH4 due to restricted gas diffusion.
Potential actions to reduce CH4 emissions could be: Minimising soil compaction (Hansen et al., 1993); disturbing the surface of heavier soils, e.g., by harrowing, before applying slurry so as facilitate infiltration; spreading onto recently culti-vated land (Chadwick and Pain, 1997), or diluting the slurry before application (Schürer, 2000). Diluting the slurry can also reduce NH3 emissions but increases N2O emissions (Schürer, 2000). Emissions of CH4 from slurry stores can be reduced by use of covers (Sommer et al., 1999).
Odour emissions
Odour emissions from livestock production are a frequent source of complaint from the general public. Land spreading of slurries and manure, in particular, causes more complaints about odour than any other stage of livestock production (Phillips et al., 1991). Studies of odours are important, not only to assess the degree of odour nuisance, but also because some components are known to have ill effects on crops, animals and humans. Sulphurous odorants may present a health hazard whilst phenols and indoles are known to inhibit plant growth and cause respiratory stress in livestock (Hobbs et al., 1999).
Williams and Evans (1981) found great variability in the concentrations of differ-ent volatile fatty acids emitted during storage of piggery slurry. Over the storage period of 25 weeks, they found that odour offensiveness increased as the organic matter decomposed. The changes in emissions which reflect changes in the com-position of wastes during the decay process, can therefore present difficulties in estimating emissions during storage (Hobbs et al., 1999).
A range of potential options is available to minimise odour emissions. Injection of pig slurry into agricultural land has been found to reduce odour threshold values compared with surface application (Lindvall et al., 1972). Slurry maybe treated
prior to storage or spreading, e.g., by aeration to degrade odorous compounds (Burton et al., 1998). Odorous air may be treated by a variety of methods, such as biofilters, oxidation or scrubbing (Norén, 1987). Odours have been shown to reflect the amount and form in which nitrogen was presented to the animal as feed. Therefore manipulation of dietary protein can lead to a reduction of odour in the waste (Hobbs et al, 1996).
Measurement of greenhouse gases and odours
A range of methods are available which would be suitable for measuring N2O and CH4 produced from the various stages of animal production. However, a number of factors need to be considered in order to select the most appropriate technique, e.g., cost, level of accuracy required, and the scale and design of the experiments to be undertaken. This report discusses the various techniques available and their advantages and disadvantages, including applicability to various situations. This is followed by a comparison of the analytical methods which are most commonly used for determining N2O and CH4 fluxes.
JTI has worked with odour measurement since the early 1970’s. However, there have been a number of developments in the field, including the advent of electronic noses and new techniques for collecting odour samples, and a short review of these is included here.
Abbreviations
ECD Electron capture detector
FID Flame ionisation detector
FTIR Fourier transform infrared (spectroscopy)
GC Gas chromatography / Gas chromatograph
TCD Thermal conductivity detector
TDL Tuneable diode laser
TGA Trace gas analyser
SF6 Sulphur hexafluoride
Methods for measuring N
2O and CH
4fluxes
Static chambers
Most measurements of N2O (and CH4 for that matter) have been made using some kind of static chamber technique. An area, normally less than 1 m2, is enclosed by a chamber in order to raise the concentration of the gas emitted. The enclosure period is usually for a duration of 1 hour or less. Samples are taken from the cham-bers using syringes, evacuated bottles or gas bags. These systems can be prone to gas leakage. However, Swedish investigations, carried out by Klemedtsson, have used non-evacuated vials with gas tight membranes. During sampling the gas in
the flasks was flushed with the gas in the field chamber by a pump. One minute of pumping of the gas resulted in a 99.98 % exchange of the gas in the flask. The small volume of air in the flask (22 ml) was diluted in the headspace of the cham-bers (>1 litre). The advantage of this system is that the vial is not evacuated, thereby saving time and avoiding leakage of air into the vial prior to sampling. The gas samples are transported to the laboratory at normal pressure and can be stored for up to two moths without leakage (Freibauer, 2000). This method of collecting gases has also been used in Denmark for measuring N2O and CH4 emissions from manure heaps (Sommer et al., 1999).
After collection, the samples are most commonly analysed by GC, e.g., see Sitaula et al. (1992), but other instruments are available, such as the TGA as used by Velthof et al. (1997) or FTIR as used by Galle et al. (1994). FTIR gives greatly improved detection limits and has been exploited for taking measurements in large headspace volumes as found in mega-chambers (Smith et al., 1994a). Mega-chambers allow trace gas fluxes to be averaged over several tens of square metres. They typically consist of tent-like, tunnel shaped constructions with an extension of up to about 30 m.
Advantages
Closed chamber systems are well established, cheap and easy to use, but labour-intensive unless sampling is made automatically. According to Ellis (2000) such
an automated system is used by ADAS in England for N2O measurements. When
samples are taken manually, the intervals between measurements are usually in the order of days to one week in long-term studies. Automated enclosures can substantially enhance the temporal resolution of long-term measurements and have been constructed by various research groups, e.g., Loftfield et al. (1992), Motz (2000).
Due to the limited area covered by the chambers, this technique is well suited for process-oriented studies and factorial designs. Of particular interest are studies where effects of fertilisation and other cultivation practices on trace gas fluxes must be considered, with the assumption that the soil composition is identical in all treatments.
Static chambers are also suitable for studies of point sources, e.g., solid manure heaps and slurry stores, or excretal returns from grazing animals. One could also consider their use in experimental situations that do not meet the requirements of micrometeorological techniques. However, one must consider the question of spatial variability (Husted, 1994).
Disadvantages
Chambers without insulation may lead to significant temperature changes of the headspace during deployment. A change of 10 °C, which is quite realistic (Livingston and Hutchinson, 1995), will change the gas volume sampled by between 3 and 4 %.
The transport of gases within unvented chambers is easily impeded. Mass flow is particularly important in media of high permeability and/or when trace gas turnover occurs close to the surface-air interface (e.g., litter layers, sandy soils,
manure heaps); in such systems the use of vented chambers is particularly important (Freibauer, 2000).
During chamber deployment, the production, consumption and transport of the gas are sensitive to perturbations. Two-component chambers with a permanently installed base are preferable in order to minimise such effects.
Enclosures are impractical for tall canopies, although Jørgensen et al. (1997) successfully applied static enclosures (2 m × 2 m × 2 m to the measurement of N2O emissions from Miscanthus.
Most chamber methods are not suitable for studying dynamic events like rainfall or diurnal temperature fluctuations, since the deployment may rapidly interfere with the soil conditions.
Spatial variability is a major limitation for the accurate quantification of trace gas fluxes using chambers. Increasing the chamber area allows average emission rates to be taken over a given area, but the larger size is coupled with higher difficulty in handling and usually a lower number of replicates.
Dynamic chambers
Dynamic or open chambers (including wind tunnels) address both advective and diffusive transport of gases. In open chamber measurements, the chamber is flushed with ambient air and the gas flux is calculated from the concentration difference between incoming and outgoing air. Gas concentrations are analysed using infra red monitors or GC systems. To account for background concentra-tions it is necessary to continuously measure the greenhouse gas concentraconcentra-tions in the incoming air. Concentration differences between the inflow and outflow are small and therefore require very accurate measuring systems. In addition, the flow rate must be accurately determined, e.g., with mass flow meters.
Ahlgrimm and Breford (1998) describe a special open chamber for manure heaps that allows the maintenance of a range of natural conditions. The manure heap is placed in a three walled bunker. To prevent uncontrolled gas flow from the heap into the atmosphere, the walls are sealed by plastic foil and the top and the front are also covered by foil. A suction fan generates a permanent gas stream out of the system. Fresh air can enter the heap by small slits in the front cover close to the bottom. Ventilation inside the system avoids concentration gradients and may serve to simulate the effect of natural wind.
Sommer and Dahl (1999) describe a mobile chamber of height 1.6 m, width 2 m and length 4 m used to measure NH3, N2O, CH4 and CO2 emissions from deep litter. The chamber was constructed from marine plywood mounted on a steel frame such that one side was open. The open side was closed by fixing the cham-ber to a corresponding stationary wall mounted behind the manure heap. Air was drawn through the chamber by means of a steel tube attached to a ventilator. The flow rate was measured using a cup anemometer. Samples of N2O and CH4 were taken from the inlet and outlet of the chamber using syringes. Analyses were made by GC.
On slurry stores, floating chambers can be used. Peu et al. (1999) describe a PVC chamber (height 0.4 m, width 0.4 m and length 0.6 m) fixed within an expanded
polystyrene float covering an area of 0.96 m2. Clean inlet air was delivered by one membrane pump, and emitted air was withdrawn from the chamber by means of a second membrane pump. During calibration experiments with the chamber, the average recovery of N2O was 88 ± 7 %.
Priemé et al. (1996) Describe a megachamber/FTIR method for analysing CH4
emissions from a forest soil. CH4 measurements were made by placing 2 sets of mirrors in the megachamber, 25 m apart. Light from a glowbar IR source was passed 40 times between the mirrors, giving an optical path of 1 km, before being analysed by the FTIR spectrometer. Thus an IR spectrum of the air mass between the mirrors was obtained. By comparing the intensity of this spectrum against calibration spectra, the concentration of CH4 in the air mass could be determined.
Advantages
Like static chambers, dynamic chambers are well suited to process oriented studies. Unlike static chambers, they are less susceptible to problems associated with impediment of gas transport.
Disadvantages
Installation of ventilators inside the chamber might enhance air exchange through the entrance slit leading to a major underestimation of fluxes; e.g., Gut (1998) determined recovery rates of ca. 50% for NO measurements made with a dynamic chamber.
Many of the disadvantages described for static chambers also apply to dynamic chambers, e.g., problems arising from spatial variability, the difficulties of taking measurements on tall crop canopies and the inability to study dynamic events.
Micrometeorological techniques
Micrometeorological techniques measure the turbulent transfer of gases from the ground surface to the lower atmosphere. They are able to measure gaseous fluxes over a larger area than is possible with static or dynamic chambers, with the added advantage that they do not disturb the conditions at the soil or manure heap sur-face. Where field scale measurements are to be taken, micrometeorological tech-niques are suitable for integrating fluxes over areas between 0.1 and 1 km2, depending on the technique and sampling height chosen (Fowler and Duyzer, 1989). The limitations of these techniques are the requirements for expensive and sophisticated instrumentation, relatively level terrain, and complexity in calculat-ing fluxes (Bogner et al., 1997).
Among the micrometeorological approaches, the eddy covariance or eddy corre-lation technique is the most direct one for flux-measurement. It requires simulta-neous, high frequency measurement of the vertical air velocity and the concentra-tion of the target air constituent. Gas sensors are needed which can measure the target gas concentration with a time resolution of 10 Hz or better (Freibauer, 2000). For trace gas analysis, laser and infrared spectroscopy devices are used.
In contrast to eddy correlation, gradient techniques represent an indirect measure-ment of trace gas fluxes. In the atmospheric gradient method the transport of a trace gas due to the turbulent air movement is described in analogy to the molecular diffusion. Gradient measurements require continuous and simultaneous measurement of trace gas concentration, temperature and the horizontal wind velocity at various heights above the ground. The flux that is derived from the atmospheric concentration gradient measurements is representative for an area that is within the so called fetch of the measurement tower. A rule of thumb, for both gradient and eddy correlation measurements, is that the fetch required is about 100 times the height of the measuring tower.
Measurement of N2O and CH4 can be made using a variety of techniques. Hargreaves et al. (1994) compared 3 instruments for N2O measurement using a gradient method. The instruments were GC, FTIR and TDL. Samples for the GC were collected at 5 different heights in Tedlar gas bags, before transporting to the laboratory. Pumps drew samples from air inlets at 2 heights into the optical cells of the FTIR and TDL. Good agreement was found between the 3 instruments which confirmed the suitability of micrometeorological techniques for field based measurements. Smith et al. (1994b) also describe the use of a fast response TDL to measure N2O fluxes by an eddy correlation method. Results were similar to measurements taken using a gradient method.
Micrometeorological methods for CH4 measurement are not commonly applied
on open field crops in the EU (Freibauer, 2000). This is not because of methodo-logical or meteoromethodo-logical limitations as the eddy correlation method has been applied on landfills (Hovde et al.1995), swamp and natural ecosystems (Fowler et al.,1995) and rice paddy. The explanation is that CH4 emissions from open field crops (except paddy fields) are very marginal in the greenhouse gas balance.
Advantages
Micrometeorological techniques do not suffer from many of the problems encountered with chamber techniques, e.g., perturbation of the emission source, difficulties of taking measurements in crop canopies and the inability to study dynamic events.
Since micrometeorological techniques integrate the trace gas fluxes over larger areas they can eliminate, to a certain extent, the problems of spatial variability, and are suitable for inventory studies rather than for process-orientated studies. An advantage of the gradient technique over the eddy correlation technique is that it does not require instruments with a high measuring frequency. This can lead to a saving in equipment and maintenance costs.
Disadvantages
The experimental site needs to be flat (a constant slope can be accepted) and homogeneous for the entire fetch in all wind directions, or at least in the dominant wind direction, i. e., without any obstacles like trees, houses etc. Therefore the methodology is not applicable in uneven terrain and in landscapes containing various small structures.
The difference in mean trace gas concentrations determined between the ground level and higher levels, is typically very small and may lead to substantial analyti-cal error.
An important constraint with the eddy correlation method is that sensitive and rapid-response instruments are required to analyse the trace gases. For CH4 and N2O, such instruments are available (e.g., tuneable diode laser spectrometers) but are rather expensive. Hence, measurement of these trace gases is not so common and restricted to just four or five specialised laboratories in the EU. Also data acquisition requires a large amount of storage capacity - roughly 1 Mbyte per hour of measurement (Freibauer, 2000).
Tracer methods
Tracer methods rely on the simultaneous measurement of the concentrations of both the target gas and an inert tracer released at a known rate. The concentration ratios of the two gases, measured downwind, can then be related to the ratio of their fluxes. The most commonly used tracer gas is sulphur hexafluoride (SF6), although other tracers are known (Sneath, 2000). SF6 has the advantage of being inert, it has a low back ground concentration (ppt) and is easily detectable by an ECD on a GC (Sneath, 2000).
To ensure reliability of the technique, the conditions that have to be met are:
• the tracer must be released in a way that to a high degree resembles the emission of the target gas,
• the target gas and the tracer must be well mixed.
When the above are satisfied, the emission rate of the target gas can be obtained directly by the ratio method as:
QT = QSF6 (CT /CSF6)
Where: QT is the target gas flux rate, QSF6 is the tracer gas flux rate, CT is the measured concentration of target gas and CSF6 is the measured concentration of tracer gas (Johnson et al., 1994).
Tracer gas methods have been successfully used to measure gaseous emissions from a range of situations:
Direct measurement of CH4 from ruminant livestock
Approximately 98 % of CH4 generated in the digestive systems of ruminant live-stock is released in the breath (Murray et al., 1976). The tracer technique for measuring this emission was first documented by Johnson et al. (1994) and is described in detail by Westberg et al. (1998). The technique uses small, calibrated permeation tubes, charged with SF6, which are placed in the rumen using normal dosing procedures. To obtain samples of expired air, PVC tubular yokes are placed over the neck of the animal and connected through capillary tubing to a sampling device which can be attached immediately above the nostrils of the animal. Immediately before use, the sampling yokes are evacuated and then used to sample expired air from the animal. The rate of sampling is controlled by the length and diameter of the capillary tube, ensuring that negative pressure is
main-tained in the yoke throughout the whole sampling period. A custom built GC with an ECD was employed for analysis of the SF6. CH4 was analysed on a separate GC equipped with a FID (Westberg et al., 1998).
In validation measurements, Johnson et al. (1994) compared the methane emis-sions from heifers and steers using both the tracer and the more established cham-ber technique. With 55 tracer measurements and 25 chamcham-ber measurements, no significant differences (P <0.10) were encountered between the two techniques. Investigations were also carried out which showed that SF6 did not adversely affect the function of the rumen. It has also been reported that the replacement of N2 by SF6 in breathing air does not adversely affect lung function (Martin et al., 1972).
In an Australian study, CH4 emissions from free-range sheep were compared using the tracer technique and a micrometeorological mass-balance method (Leuning et al., 1999). The daily mean values obtained from the methods were very similar with 11.7 g CH4 d-1 from the tracer technique and 11.9 g CH4 d-1 from the mass-balance measurements. Judd et al. (1999) also compared the tracer tech-nique against a micrometeorological method for measuring CH4 from grazing sheep. At a stocking rate of 20 sheep ha-1, the tracer method gave an emission rate of 39 ± 9.6 mg m-2 d-1 which was close to the value of 46 mg m-2 d-1 obtained with the micrometeorological method.
According to Freibauer (2000) the use of SF6 for measurement of methane pro-duction in individual animals, as described above, has been restricted in the USA due to concern over SF6 residues in meat and milk animals. Whilst this restriction has not yet been applied in Europe and other parts of the world, it may be cause for concern and worthy of further consideration.
Measurement of area source emissions
The use of SF6 tracer methods for the measurement of emission rates from area sources, e.g., landfills or lagoons, may be carried out in 2 ways. One option is to approximate the source using a single point release of tracer gas and use disper-sion modelling to select a suitable location. Another option is to use a line source, or series of line sources, instead of a single point release. Eklund (1999) evaluated these 2 options in connection with a study to estimate greenhouse gas emissions from a wastewater treatment plant. Measurements were made using open path monitoring with detection by FTIR. The 2 methods gave almost identical results. However, errors increase with the dispersion modelling technique when target gas and tracer are not released from the same location.
In Sweden, an area release system has been developed in which 1 km of silicone tubing is laid out uniformly over the area to be investigated. The permeability of the silicon tubing allows SF6 to be released at a controlled rate; SF6 and the target gas can then be studied downwind of the release area. An SF6 method has been successfully used to measure methane emissions from landfills (Galle et al., 1999).
Determination of emissions from animal buildings, slurry stores and manure heaps
The trace gas technique is an established method for the determination of air exchanges and ventilation effectiveness in buildings (Kronvall, 1979; Gustafsson,
1993). In order to measure air exchange, the trace gas should meet the following criteria (Kronvall, 1979):
• Should be chemically stable in the environment studied
• Should not be present in normal air
• Must not be hazardous to human health at the concentrations analysed
• Must not be inflammable or explosive
• Must be possible to measure with a high degree of accuracy, even at low concentrations
• Should be readily available and inexpensive
Very few gases fulfil these requirements, the main gases used in studies of air exchange in buildings being argon, ammonia, carbon dioxide, helium, hydrogen, nitrous oxide, xenon and SF6. SF6 has the advantage of being detectable at very low concentrations below its threshold limit value, it is chemically inert and not present in normal building atmospheres. Marik and Levin (1996) have used a SF6 tracer ratio method to measure methane emissions from a cowshed housing 43 dairy cows. They also used the technique to determine CH4 emissions from a slurry storage tank. Marik and Levin (1996) describe their method as easy and reliable, and capable of obtaining mean emission rates from a whole dairy herd in its natural environment.
Gustafsson and Andersson (1998) determined the magnitude of the leakage of air from covered slurry stores by measuring the reduction with time of a SF6 tracer injected into the headspace. This was used to determine ammonia emissions from the stores, but the technique can easily be applied to the study of other gases. Sneath (2000) has also described the use of SF6 tracer methods to determine the emissions of N2O from manure heaps and poultry buildings.
Advantages
Tracer methods can be a valuable alternative to micrometeorological methods when the sources are limited in size and the micrometeorological conditions are unfavourable. The technique is also powerful for measuring emissions from local sources such as livestock buildings, manure stores and lagoons. In these applica-tions the condition of good mixing between the tracer and target gases can be met by performing the concentration measurements at a distance where the source can be regarded as a point source. Tracer methods also can be used indirectly to determine ventilation rates and to validate other methods like micrometeorologi-cal methods.
An SF6 tracer method can be used to take measurements of methane directly from individual livestock. This has the advantage that animals can be placed under normal grazing conditions and data can be obtained from individual animals thus allowing treatment comparisons to be made. The technique is equally applicable to stall fed situations where full enclosure methodology is not possible.
Under composting conditions, e.g., during manure storage, convective transport of gases occurs which may lead to unreliability in measurements made with chamber methods. In such cases, systems based on tracer methods could provide a solution.
Tracer methods in combination with FTIR can be used to measure emission rates of a range of gases including NH3 and CO2 as well as N2O and CH4. The calcula-tion of emission rates is also greatly simplified compared with other methods.
Disadvantages
Beever and Cammell (1997) report that the Johnson et al. (1994) method applied to cattle receiving feed with different amounts of added fat, did not give results which agree with the biochemical principles of fat utilisation in the rumen. It appears that small errors in estimation of SF6 release rate from the permeation tubes could lead to sizeable errors in the estimation of methane output. Ulyatt et al. (1999) confirm the need for careful attention to be paid to the calibration of the permeation tubes as well as gas collection procedures and gas analysis. However, if due care is paid to the details of the technique, then estimates of CH4 production are comparable with those obtained by respiration calorimetry.
The complexity of the SF6 technique on ruminants, as described by Westberg et al. (1998), may in itself prove to be a hindrance towards its selection. Also, SF6 is a very powerful greenhouse gas (Maiss and Brenninkmeijer, 1998). However, current industrial emissions of SF6 far exceed the total amounts released during scientific research.
Tracer ratio methods are restricted to situations with no interfering sources, i.e., where the plume of interest is not mixed with another nearby source. Also the target gas concentration must be sufficiently high to distinguish it from back-ground levels and to enable its measurement far enough down wind to ensure adequate mixing with the tracer gas.
Comparison of analytical methods
Gas chromatography
The most commonly used instrument for measuring N2O and CH4 is GC. It relies on the individual partitioning characteristics of different gases in the sample between a mobile phase (an inert gas such as He) and a stationary solid phase packed in a column. Thus the components in the gaseous mixture are separated, with each component being identified by its retention time on the column and quantified by a subsequent detector. A key part of a GC system is the detector. Three types of detector are commonly employed for measurement of greenhouse gases:
Thermal conductivity detector (TCD) – generally for CO2 but also CH4 in special cases.
Flame ionisation detector (FID) - sensitive to CH4.
Electron capture detector (ECD) - commonly used for N2O.
The detectors may be individually connected to GC systems, or fitted in combina-tion, thus allowing the simultaneous analysis of several gas species (Sitaula et al., 1992).
GC typically measures differences in N2O concentration down to 10 ppb (Fowler et al., 1997). Detection limits below 200 ppb are possible with CH4 (Crill et al., 1995).
Infrared photoacoustic spectrometer – trace gas analyser
The operating principle of the TGA is as follows (Beck-Friis, 2000): A gas sample is contained in a sealed cell and irradiated with chopped IR light of selected wave-length. The wavelength is specifically absorbed by the gas to be studied and is selected using filters. The energy absorbed by the gas leads to an increase in its temperature and pressure. Since the IR light is chopped, this causes a series of pressure pulses in the cell which are detected by microphones. The voltage generated by the microphones is proportional to the gas concentration in the cell. The TGA can measure 4 gas simultaneously, e.g., N2O, CH4, CO2, and NH3. TGA is less sensitive for N2O (detection limit 30 ppb) than GC, but theoretically more sensitive to CH4 (detection limit 100 ppb); however CH4 measurements are prone to interference, probably by water vapour (Beck-Friis, 2000). According to Velthof (2000) sensitivities for N2O detection in the field by TGA are acceptable (±10 to 20 % of the background value) and the instrument has the advantage of being portable. TGA is also able to give on-line measurements, e.g., Osada et al. (1998) made continuous measurements of N2O and CH4 emissions from pig units using a TGA instrument. Disadvantages of the TGA include cost (ca. 260000 SEK) and there is concern that N2O measurements may be interfered with by high concentrations of CH4 (Velthof, 2000). CO2 and water must be removed from air samples prior to measurement because they interfere with both N2O and CH4 measurements, although according to Yamulki and Jarvis (1999) the interferences are linear and can be corrected for.
Fourier transform infrared absorption spectroscopy
The FTIR principle involves Infrared light being split into two paths by a two-beam interferometer. When the two-beams are combined at an infrared detector, constructive and destructive interference produces a modulated signal which is a function of the optical path difference between the two beams. This so-called interferogram is converted into a spectrum by a complex Fourier transform. In FTIR spectroscopy the unique infrared absorption of different molecules are used to quantify their concentration. A number of gases of interest in climate change research can be uniquely and simultaneously determined, e.g. CH4, CO2, N2O, CO and H2O. With different types of mirror arrangements, long optical paths can be obtained yielding good sensitivity (Freibauer, 2000).
The greatly improved sensitivity of FTIR permits flux measurements to be made by micrometeorological techniques (Hargreaves et al., 1996). FTIR systems are not fast enough to be used in eddy correlation measurements but are very suitable for gradient measurements due to the high precision that can be obtained in differ-ence measurements. This is especially valuable when small gradients need to be measured on top of high background concentrations.
Disadvantages include the high cost of the equipment and the requirement of constant wind direction in the case of instruments operated in open path configu-ration. Also fog and rain may limit the measurements. FTIR is a highly sensitive technique, e.g., Gut et al. (1998) report a detection limit for N2O of 1 ppb.
Tuneable Diode Laser Absorption Spectroscopy
Determination of gas concentrations using tunable diode lasers (TDL) is based on the absorption of an infra-red laser beam as it travels along a path through the gas sample. As in any absorption spectroscopy method, the total absorption depends on the number of absorbing molecules in the beam’s path according to Beer’s law. The sensitivity of TDL based instruments depends on the path length and the strength of the absorption line, with highest detection sensitivities for gas species having strong absorption lines in the spectral region emitted by the laser (Freibauer, 2000). Typical laser emission line widths are small (~10-4 cm-1) relative to typical absorption line widths, and a high spectral
resolution can be achieved in resolving individual absorption lines at atmospheric and low pressures, without interference from other gases (Edwards et al., 1994). Atmospheric gases that have been measured using TDL spectroscopy include CO, CO2, N2O, NO, NO2, NH3 and CH4.
Due to its sensitive and fast response, TDL based instruments are ideally suited for the in situ measurement of trace gas concentrations using micrometeorological methods. Detection limits are possible in the low tens of ppt (parts per trillion) range for averaging times of between 5 and 30 minutes. While TDL absorption spectroscopy has the highest spectral resolution of any of the methods, its main limitation is the number of gas species which can be measured simultaneously with the same diode (Freibauer, 2000). The instrumentation is also the most expensive of those described in this report.
Odour measurement
Electronic nose
An electronic nose may be defined as an instrument which consists of an array of electronic chemical sensors with partial specificity and an appropriate pattern recognition system capable of recognising simple or complex odours (Gardner and Bartlett, 1994). A range of materials have been used in sensor arrays, includ-ing sintered metal oxides, catalytic metals, lipid layers, phthalocyanins, conduct-ing polymers and organic semi-conductors (Gardner and Bartlett, 1994). Sensors have been developed for a number of applications, mostly in the food industry, such as grading coffee blends and beans, detection of off-flavours in beers and freshness of meat (Misselbrook et al., 1997). Electronic noses have also been applied to the measurement of odours from livestock wastes (Hobbs et al., 1995; Misselbrook et al., 1997).
Misselbrook et al. (1997) describe electronic noses based on conducting polymer type sensors which were capable of responding to odours from cattle slurry applied to grassland. The concentration ranges measured were in the typical range for agricultural odours (50 to 10 000 odour units/m3) and the performance was better than that reported by earlier researchers. For example, Hobbs et al. (1995)
reported a polypyrrole based electronic nose to be sensitive to 60 000 odour units/m3. However, it is clear that more work is required to ascertain whether reli-able relationships exist between odour concentration and sensor response for a range of agricultural odours Misselbrook et al. (1997).
Olfactometry
Since 1970, JTI have collaborated with the Department of Psychology, University of Stockholm, on the measurement of odour emissions using olfactometry. Com-prehensive work on the development of the olfactometric method has been carried out by Berglund et al. (1986). The odour strength is determined by a panel of 9 people. The odorous air is presented to the panel in one of 2 hoods. The panel members are then instructed to compare their odour perception against known concentrations of pyridine gas presented in the second hood. Thus all odour strengths are expressed as pyridine concentration equivalents. The advantage of this method over the “threshold method” is that odour measurements are compa-rable between different experiments. The pyridine method is also less sensitive to differences in perception between panel members. However, one should be aware that a European Standard currently exists in draft form which defines a method for determination of odour concentration using dynamic olfactometry (CEN, 1998). The standard states that the concentration of a gaseous sample of odorants is determined by presenting a panel of selected and screened human subjects with that sample, varying the concentration by diluting with neutral gas, in order to determine the dilution factor at the 50 % detection threshold.
It is generally accepted that the most sensitive method for assessing odour quality is olfactometry. One should appreciate that the perception of odour is highly com-plex. For example, O’Neill & Phillips (1992) identified 168 compounds associ-ated with odours from livestock wastes. Non-olfactometric methods, which rely on the quantification of the concentration of different chemical components in the odour, do not characterise the resulting odour nor the human response to that odour (Hobbs et al., 1995). This information can be obtained by using olfacto-metry. The disadvantages of olfactometry are that the method is time consuming, labour intensive, has to be carried out in a specially designed laboratory often remote from the sampling site, and on-line measurements are not possible.
Odour sampling
Sampling of odour emissions from soil and crops (Burton et al. 1998, after Lockyer, 1984) can be made using portable, semi-cylindrical wind tunnels (2.0 m long x 0.5 m diameter). A fan is used to induce airflow over the soil applied with, e.g., manure. Air samples are taken with a stainless steel bellows pump and captured in Teflon FEP odour bags with a nominal volume of 60 l. An alternative approach to measuring odours and emissions has been developed at Silsoe Research Institute in England (Hobbs et al., 1999). In the so called “Odours and Emissions Chamber”, up to 200 l of liquid can be exposed to an enclosed atmosphere under controlled environmental conditions. The chamber consists of a “U” shaped, enclosed system of stainless steel ducting (0.5 m x 0.5 m internal section) whose ends are connected to a large Tedlar bag. The air in the chamber is pressurised by compressing the bag with a roller, thereby preventing
dilution of odours in the chamber by air leaking into the system. Air in the chamber is circulated at a controlled rate. The slurry sample is placed in the chamber in a 200 l tray equipped with an impeller for mixing at a controlled rate. Odour samples are drawn into 10 l Teflon FEP bags for subsequent olfactometric analysis. Other gases, e.g., CH4, CO2, and NH3 are measured on-line using an FTIR.
Conclusions
Plot experiments
In randomised block designs, e.g., when comparing different techniques for manure application, dynamic chamber techniques represent the best option. To increase sampling frequency and reduce labour costs, one could consider the use of automated, dynamic chambers such as those used at Hohenheim (Motz, 2000). These also include an air conditioning system to minimise temperature effects during coverage, as well as an automated sampling system to allow simultaneous measurements.
Gases could be analysed either by GC or by TGA. TGA offers the advantage of on-line measurement, and it may be possible to loan an instrument from the Department of Agricultural Engineering at SLU.
Manure heaps and stores
SF6 tracer ratio methodologies offer the most reliable measurement possibilities for emissions from local sources. In order to analyse samples of gases, either a GC or an FTIR spectrometer can be used. It may be necessary to modify existing GC equipment or purchase new equipment. FTIR will be more expensive, but offers the possibility to measure a variety of gas species including NH3.
Field scale measurements
A micrometeorological method using the gradient technique would be the
cheapest option since samples can be collected in gas bags and analysed by FTIR. The eddy correlation method requires the use of the more costly TDL.
Animals and animal housing
The SF6 tracer ratio method developed by Johnson et al. (1994) allows CH4 flux determinations to be made directly on individual ruminants, irrespective of their environment. The method therefore allows measurements to be taken on free-range animals. In livestock buildings, the method allows CH4 emissions from animals to be distinguished from CH4 emissions from manure. SF6 can also be used as a tracer to determine general gaseous emissions from livestock buildings.
Odour measurement
JTI currently uses the pyridine method to characterise odour strength. Whilst this method functions well and gives reliable results, it should be recognised that CEN standards (currently in draft format) may need to be adopted in the future.
Useful information on odour, and other gaseous releases, from livestock wastes could be gained from experiments carried out under controlled conditions in a so called odour and emissions chamber (Hobbs et al., 1999).
References
Adamsen A.P.S. & King G.M., 1993. Methane consumption in temperate and subarctic forest soil: rates, vertical zonation, and responses to water and nitrogen. Applied Environmental Microbiology, 59, 485-490.
Adolfsson R., 2000. Personal communication. Statistics Sweden, SCB, Box 24300, 104 51 Stockholm.
Ahlgrimm H.-J. & Breford J., 1998. Methanemissionen aus der Schweinemast. Landbauforschung Völkenrode 1/1998, 26-34.
Beck-Friis B., 2000. Personal communication. Sveriges lantbruksuniversitet, Institution för markvetenskap, Box 7014, 750 07 Uppsala.
Beever D.E. & Cammell S.B., 1997. Special measurement methodologies to determine methane output from ruminant livestock. Eds. Petten, M.K. and Freibauer, A. IER, University of Stuttgart, pp.22-32.
Berglund B., Berglund U. & Lindvall T., 1986. Psychophysical measurement of odorous air pollutants. VDI-Berichte No. 561/1986 pp. 511-520.
Bogner J., Meadows M. & Czepiel P., 1997. Fluxes of methane between landfills and the atmosphere: natural and engineered controls. Soil use and
management, 13, 268-277.
Bouwman A.F., 1990. Exchange of greenhouse gases between terrestrial eco-systems and the atmosphere. Pp. 61-127. In: A.F. Bouwman (ed.) Soils and the greenhouse effect. John Wiley & Sons, New York.
Burton C.H., Sneath R.W., Misselbrook T.H. & Pain B.F., 1998. The effect of farm scale aerobic treatment of piggery slurry on odour concentration,
intensity and offensiveness. Journal of Agricultural Engineering Research, 71, 203-211.
CEN, 1998. Draft European Standard: Air quality – Determination of odour concentration by dynamic olfactometry. Document
CEN/TC264/WG2/N222/e, release 98-05-10.
Chadwick D.R. & Pain B.F., 1997. Methane fluxes following slurry applications to grassland soils: laboratory experiments. Agriculture, Ecosystems and Environment, 63, 51-60.
Chadwick D.R., Sneath R.W., Philips V.R. & Pain B.F., 1999. A UK inventory of nitrous oxide emissions from farmed livestock. Atmospheric Environment, 33, 3345-3354.
Crill P.M., Butler J.H., Cooper D.J. & Novelli P.C., 1995. Standard analytical methods for measuring trace gases in the environment. In: Matson, P.A., Harriss, R.C. (eds.): Biogenic trace gases: Measuring emissions from soil and water. Methods in ecology series. Oxford, Blackwell Science. pp.164-205.
Edwards G.C., Neumann H.H., den Hartog G., Thurtell G.W. & Kidd G., 1994. Eddy correlation measurements of methane fluxes using a tunable diode laser at the Kinosheo Lake tower site during the Northern Wetlands Study
(NOWES). Journal of Geophysical Research, 99, 1511-1517.
Eklund B., 1999. Comparison of line- and point-source releases of tracer gases. Atmospheric environment, 33, 1065-1071.
Ellis S., 2000. Personal communication. Ministry of Agriculture, Fisheries and Food, UK. s.ellis@env.maff.gsi.gov.uk
Fowler D. & Duyzer J.H., 1989. Micrometeorological techniques for the measure-ment of trace gas exchange. In: Andreae, M.O. and Schimel, D.S (eds.) Exchange of trace gases between terrestrial ecosystems and the atmosphere. Wiley, Chichester, pp. 189-207.
Fowler D., Hargreaves K.J., Skiba U., Milne R., Zahniser M.S., Moncrieff J.B., Beverland I. & Gallagher M.W., 1995. Measurements of CH4 and N2O fluxes at the field and landscape using three different micrometeorological
techniques. Philosophical Transactions of the Royal Society, 351, 339-356. Fowler D., Skiba U. & Hargreaves K.J., 1997. Emissions of nitrous oxide from
grasslands. In: Gaseous nitrogen emissions from grasslands, editors Jarvis, S.C. and Pain, B.F. CAB International, Walingford, UK. ISBN 0-85199-192-0. Pp. 147-164.
Freibauer A., 2000. Personal communication. Institut für Energiewirtschaft und Rationelle Energiranwendung, University of Stuttgart.
Galle B., 1999. Development and application of methods based on DOAS and FTIR absorption spectroscopy for atmospheric research. Ph.D. Thesis, Chalmers University of Technology, Gothenburg.
Galle B., Klemedtsson L. & Griffith D.W.T., 1994. Application of a Fourier Transform IR system for measurement of nitrous oxide fluxes using micro-meteorological methods, an ultra large chamber system, and conventional field chambers. Journal of Geophysical Research, 99, 16575-16583.
Gardner J.W. & Bartlett P.N., 1994. A brief history of electronic noses. Sensors and Actuators B, 18-19, 211-220.
Gut A., 1998. Characterisation of the soil-atmosphere exchange flux of nitric oxide. Diss. ETH Nr. 12694
Gut A., Blatter A., Fahrni M., Lehmann B.E., Neftel A. & Staffelbach T., 1998. A new membrane tube technique (METT) for continuous gas measurements in soils. Plant and Soil, 198, 79-88.
Gustafsson G., 1993. Spårgasteknik som metod för att bestämma ventilations-effektivitet och spridning av luftföroreningar i djurstallar. Sveriges
Lantbruksuniversitet. Inst. För lantbruks byggnadsteknik. Rapport nr 84. Lund.
Gustafsson G. & Andersson M., 1998. Determination of air tightness and
ammonia emission from covered slurry storage tanks. Paper 98-B-039. AgEng International conference on agricultural engineering, Oslo, 24-27 August. Hargreaves K.J., Skiba U., Fowler D., Arah J.R.M., Weinhold F.G., Klemedttsson
L. & Galle B., 1994. Measurements of nitrous oxide emission from fertilised grassland using micrometeorological techniques. Journal of Geophysical Research, 99, D8, 16569-16574.
Hansen S., Mæhlum J.E. & Bakken L.R., 1993. N2O and CH4 Fluxes influenced by fertilization and tractor traffic. Soil Biology and Biochemistry, 25(5), 621-630.
Hargreaves K.J., Weinhold F.G. & Klemedttsson L., 1996. Measurements of nitrous oxide emission from agricultural land using micrometeorological methods. Atmospheric Environment, 30, 1563-1571.
Hobbs P.J., Misselbrook T.H. & Pain B.F., 1995. Assessment of odours from livestock wastes by a photoionization detector, electronic nose, olfactometry and gas chromatography – mass spectrometry. Journal of Agricultural Engineering Research, 60, 137-144.
Hobbs P.J., Pain B.F., Kay R.M. & Lee P.A., 1996. Reduction of odorous compounds in fresh pig slurry by dietary control of crude protein. Journal of Science, Food and Agriculture, 71, 508-514.
Hobbs P.J., Misselbrook T.H. & Cumby T.R., 1999. Production and emission of odours and gases from ageing pig waste. Journal of Agricultural Engineering Research, 72, 291-298.
Hovde D.C., Stanton A.C., Meyers T.P. & Matt D.R., 1995. Methane emissions from a landfill measured by eddy correlation using a fast response diode laser sensor. Journal of atmospheric chemistry, 20, 141-162.
Husted S., 1994. Seasonal variation in methane emission from stored slurry and solid manures. Journal of Environmental Quality, 23(3), 585-592.
Johnson K., Huyler M., Westberg H., Lamb B.& Zimmerman P., 1994. Measurement of methane emissions from ruminant livestock using a SF6 tracer technique. Environmental Science and Technology, 28, 359-362. Judd M.J., Kellier F.M., Ulyatt M.J., Lassey K.R., Tate K., Shelton I.D., Harvey
M.J. & Walker C.F., 1999. Net methane emissions from grazing sheep. Global Change Biology, 5, 647-657.
Jørgensen R.N., Jørgensen B.J., Nielsen N.E., Maag M. & Lind A.-M., 1997. N2O emissions from energy crop fields of Miscanthus "Giganteus" and winter rye. Atmospheric Environment, 31, 2899-2904.
Kroeze C., 1995. Comparison of inventory methods for estimating national emissions of nitrous oxide (N2O). Quarterly Journal of Hungarian Meteorological Service, 99 (3-4), 209-225.
Kronvall J., 1979. Mätningar och mätmetoder för lufttäthet. Statens råd för byggnadsforskning. Rapport T6:1979. Stockholm.
Leuning R., Baker S.K., Jamie I.M., Hsu C.H., Klein L., Denmead O.T. & Griffith D.W.T., 1999. Methane emission from free-ranging sheep: a comparison of two measurement methods. Atmospheric environment, 33, 1357-1365. Lindvall T., Norén O. & Thyselius L., 1972. Luktreducerande åtgärder vid
flytgödselhantering. Specialmeddelande 22, Swedish Institute of Agricultural Engineering, Uppsala.
Livingston G.P. & Hutchinson G.L., 1995. Enclosure-based measurement of trace gas exchange: applications and sources of error. In: Matson, P., R.C. Harris (eds.): Biogenic Trace Gases: Measuring emissions from soil and water. Blackwell Science, Oxford, UK, pp.14-51.
Lockyer D.R., 1984. A system for the measurement in the field of losses of ammonia through volatilisation. Journal of the Science of Food and Agriculture, 35, 837-848.
Loftfield N. S., Brumme R. & Beese F., 1992. Automated monitoring of nitrous oxide and carbon dioxide flux from forest soils. Soil Science Society of America Journal, 56, 1147-1150.
Maiss M. & Brenninkmeijer C.A.M., 1998. Atmospheric SF6: Trends, sources, and prospects. Environmental Science and Technology, 32, 3077-3086.
Marik T. & Levin I., 1996. A new tracer experiment to estimate the methane emissions from a dairy cow shed using sulphur hexafluoride (SF6). Global Biogeochemical Cycles, 10(3), 413-418.
Martin R., Zutter M. & Authonisen N., 1972. Journal of applied physiology, 33, 86-90.
McElroy M.B., Wofsy S.C. & Yung Y.L., 1977. The Nitrogen Cycle: Perturbations due to man and their impact on atmospheric N2O and O3. Philosophical Transactions of the Royal Society, 277, 159-181.
Misselbrook T.H., Hobbs P.J. & Persaud K.C., 1997. Use of an electronic nose to measure odour concentration following application of cattle slurry to
grassland. Journal of Agricultural Engineering Research, 66, 213-220. Misselbrook T.H., Chadwick D.R., Pain B.F. & Headon D.M., 1998. Dietary
manipulation as a means of decreasing N losses and methane emissions and improving herbage N uptake following application of pig slurry to grassland. Journal of Agricultural Science, 130, 183-191.
Motz I., 2000. Measuring of trace gases on field surfaces – modifying the Hohenheim measuring chamber. Landtechnik, 55, 122-123.
Murray B.R., Bryant A.M. & Leng R.A., 1976. Rates of production of methane in the rumen and large intestine of sheep. British Journal of Nutrition, 36, 1-14. Norén O., 1987. Odours from animal production. Pp.1-26. In: Animal production
and environmental health, editor Strauch, D. Elsevier Science Publishers, Amsterdam.
O’Neill D.H. & Phillips V.R., 1992. A review of the control of odour nuisance from livestock buildings. Journal of Agricultural Engineering Research, 53(1), 23-50.
Osada T., Rom H.B. & Dahl P., 1998. Continuous measurement of nitrous oxide and methane emission in pig units by infrared photoacoustic detection. Transations of the ASAE, 41(4), 1109-1114.
Peu P., Beline F. & Martinez J., 1999. A floating chamber for estimating nitrous oxide emissions from farm scale treatment units for livestock wastes. Journal of Agricultural Engineering Research, 73, 101-104.
Philips V.R., Pain B.F. & Klarenbeek J.V., 1991. Factors influencing the odour and ammonia emissions during and after the land spreading of animal slurries. Pp. 98-106. In: Odour and ammonia emissions from livestock farming, editors Nielsen V.C., Voorburg J.H. and L’Hermite P., Elsevier Applied Science, London and New York.
Priemé A., Christensen S., Galle B., Klemedtsson L. & Griffith D.W.T., 1996. Spatial variability of CH4 uptake in a Danish forest soil and its relation to different measurement techniques. Atmospheric Environment, 30, 1375-1379. Regeringen, 1998. Svenska miljömål. Miljöpolitik för ett hållbart Sverige.
Regeringens proposition (Government Bill) 1997/98:145. Section 4.2.15, “Begränsad klimatpåverkan” (“Limited influence on climate”), pp. 145-151. Safley L.M, Casada M.E, Woodbury R.W. & Roos K.F., 1992. Global methane
emissions from livestock and poultry manure. United States Environmental Protection Agency. Report No. 400/1-91-048.
Schürer E., 2000. Diluting of slurry. Landtechnik, 55, 126-127.
Sitaula B.K., Luo J. & Bakken L.R., 1992. Rapid analysis of climate gases by wide bore capillary gas chromatography. Journal of Environmental Quality, 21(3), 493-496.
Sitaula B.K., Bakken L.R. & Abrahamsen G., 1992. CH4 uptake by temperate forest soil: Effect of N input and soil acidification. Soil Biology and Biochemistry, 27(7), 871-880.
Smith K.A., Scott A., Galle B. & Klemedtsson L., 1994a. Use of a long-path infrared gas monitor for measurement of nitrous oxide from soil. Journal of Geophysical Research, 99, 16585-16592.
Smith K.A., Clayton H., Arah J.R.M., Christensen S., Ambus P., Fowler D., Hargreaves K.J., Skiba U., Harris G.W., Wienhold F.G., Klemedtsson L. & Galle B., 1994b. Micrometeorological and chamber methods for measurement of nitrous oxide fluxes between soils and the atmosphere: Overview and conclusions. Journal of Geophysical Research, 99, 16541-16548.
Sneath R., 2000. Personal communication. Silsoe Research Institute, Wrest Park, Silsoe, Bedfordshire, UK
Sommer S.G. & Dahl P., 1999. Nutrient and carbon balance during the
composting of deep litter. Journal of Agricultural Engineering Research, 74, 145-153.
Sommer S.G., Petersen S.O. & Søgaard H.T., 1999. Greenhouse gas emission from stored livestock slurry. Paper accepted for publication in Journal of Environmental Quality.
Svensson L., 1999. Konstgödselkväve till vårsådd spannmål. Teknik för lantbruket nr 73. JTI – Institutet för jordbruks- och miljöteknik, Uppsala, Sweden.
Svensson L. & Lindén B., 1998. Utnyttjande och förluster av kväve vid ytmyllning av flytgödsel. Teknik för lantbruket nr 65. JTI – Institutet för jordbruks- och miljöteknik, Uppsala, Sweden.
SÖ (Sveriges internationella överenskommelser), 1993. Klimatkonvention, (Sweden’s International Agreements) SÖ 1993:13. Utrikesdepartementet. Ulyatt M.J., Baker S.K., McCrabb G.J. & Lassey K.R., 1999. Accuracy of SF6
tracer technology and alternatives for field measurements. Australian Journal of Agricultural Research, 50, 1329-1334.
Velthof G.L., Jarvis S.C. & Oenema O., 1997. Spatial variability of nitrous oxide fluxes from grasslands: Differences between soil types. Gaseous nitrogen emissions from grasslands, editors Jarvis, S.C. and Pain, B.F. CAB International, Walingford, UK. ISBN 0-85199-192-0. Pp. 173-179. Velthof G.L., 2000. Personal communication. Plant Research International,
POSTBUS 14, 6700 AA WAGENINGEN, The Netherlands. + 31 317 475700. g.l.velthof@plant.wag-ur.nl.
Westberg H.H., Johnson K.A., Cossalman M.W. & Michal J.J., 1998. A SF6 tracer technique: methane measurement from ruminants. Revision 2 of report
prepared under work assignment 1-05 of US EPA contract 68-D4-0088. Washington State University, Pullman, Washington, USA.
Williams A.G. & Evans M.R., 1981. Storage of piggery slurry. Agricultural wastes, 3, 311-321.
Yamulki S. & Jarvis S.C., 1999. Automated chamber technique for gaseous flux measurements: Evaluation of a photoacoustic infrared spectrometer – trace gas analyzer. Journal of Geophysical Research, 104(D5), 5463-5469.
