Published online 9 August 2006
Published in J Environ Qual 35:1659-1667 (2006)
DOI: 10.2134/jeq2005.0294
© 2006 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Dependence of Ammonia Emissions From Housing on the Time Cattle Spent Inside
Sarah Gilhespya,*,
J. Webbb,
Andrew Rettera and
David Chadwicka
a Institute of Grassland and Environmental Research, North Wyke Research Station, Okehampton, Devon, EX20 2SB, UK
b ADAS Research, Wergs Road, Wolverhampton WV6 8TQ, UK
* Corresponding author (sarah.gilhespy{at}bbsrc.ac.uk)
Received for publication August 2, 2005.
 |
ABSTRACT
|
|---|
Buildings housing cattle contribute 19% (42 kt NH3–N yr–1) of total UK ammonia (NH3) emissions. In the UK there is not usually an abrupt switch from cattle being kept inside to when they are turned out to graze 24 h a day. Moreover, during the summer dairy cows return to the farm twice a day to be milked and may spend some time inside buildings. Hence, there is uncertainty over the treatment of the transitional and summer periods when inventorying NH3 emissions. The aim of this study was to measure, under controlled and replicated conditions, the relationship between the number of hours cattle spend in buildings and the NH3 emissions from those buildings. Our results indicate that NH3 emissions decrease as the proportion of the day cattle spend in the buildings decreases, although the trend is not linear. Daily emission rates from cattle housed for 2 h ranged from 1.6 to 6.2 g NH3–N lu–1 whereas emissions from cattle housed for 24 h ranged from 8.1 to 24.1 g NH3–N lu–1. To significantly reduce NH3 emissions in comparison to those from buildings where cattle are housed for 24 h, the occupancy would have to be reduced to no more than 6 h each day. Thus, the strategy of extending the grazing season by allowing cattle to graze for c. 4 to 12 h during the winter is unlikely to reduce NH3 emissions from buildings or overall.
Abbreviations: TAN, total ammoniacal-N • lu, livestock unit (500-kg live weight)
|
INTRODUCTION
|
|---|
AMMONIA (NH3) EMISSIONS from UK agriculture have been estimated at 226 kt NH3–N yr–1 (Misselbrook et al., 2000). This accounts for 80% of the total UK emissions of NH3 to the atmosphere. Ammonia volatilization is not only a cause for environmental concern, but represents a substantial loss of fertilizer N value. Environmental damage is inflicted following (long-range) transport and deposition through direct toxicity to plants (van der Eerden, 1982), changes in plant species composition of natural ecosystems (Heil and Diemont, 1983), and eutrophication and soil acidification (van Breemen and van Dijk, 1988). Of all the livestock classes, cattle are the largest contributors, accounting for 52% (118 kt NH3–N yr–1) of the total UK annual emissions. Furthermore, emissions from buildings housing cattle have been identified as contributing 19% (42 kt NH3–N yr–1) of the total. The estimate from housed cattle is confounded, because although cattle are normally housed from late October to early April in the UK, in practice there is no clear distinction between occupied and unoccupied buildings. For example, during the summer dairy cows return to the farm twice a day to be milked and will often spend some time, either before or after milking, inside the farm buildings. Furthermore, there is not usually an abrupt switch from cattle being kept inside to when they are turned out to graze 24 h a day. This transitional period and the summer period pose problems in respect to inventorying NH3 emissions from buildings housing cattle.
There is some evidence that NH3 emissions from buildings during the summer months, when cattle spend only a portion of the day inside, may be equal to or greater than emissions during the winter when cattle are housed all day. Phillips et al. (1998) found no significant difference (P < 0.05) between NH3 emissions in summer, when the cows were in the buildings for 4 h d–1, and winter when the same number of cows were housed all day. Similarly, Kroodsma et al. (1993) found that NH3 emissions from a cow house in May, where animals were grazing during the day (and brought in for milking twice a day), were greater than measured in previous months when cattle were housed continuously. The increase in emissions, despite a decrease in NH3 concentration within the cow house, was attributed to a complex interaction between increased temperature, ventilation rate, and grass intake.
There is also a need to more accurately assess the effect of extended-season grazing on NH3 emissions. Urine infiltrates soil rapidly whereas slurry or farmyard manure accumulates on the impermeable surfaces found in livestock buildings. Consequently, excreta voided by cattle when they are grazing produces less NH3 emission than the excreta voided by cattle in and around buildings. In the latter case emissions also occur during the storage and land spreading of manures removed from buildings (Jarvis and Ledgard, 2002). Current estimates (Misselbrook et al., 2000) indicate that only c. 10% of total emissions from cattle are from grazed pastures despite UK dairy and beef herds spending c. 6 mo of the year in pastures, during the night as well as during the day. Thus, total NH3 emissions are typically much less from cattle when grazing than when housed. A desk study (Webb et al., 2005) concluded that extended-season grazing by dairy cattle may reduce total UK agricultural NH3 emission by 1 to 2%. While this reduction may be small, it can be achieved at no cost to the industry,and application of the technique to beef cattle could lead to further reductions. However, the potential benefits of increasing the grazing period are heavily dependent on a reduction in emissions from unoccupiedbuildings. Additionally, while increasing the length of time cattle graze may reduce NH3 emissions and improve animal welfare, it may also increase losses of both nitrate (NO3–) and nitrous oxide (N2O) (Webb et al., 2005). Any benefits need to be gauged against additional N losses from grazing (via denitrification, leaching, and volatilization) as a result of extending the grazing period, together with the improved animal welfare.
There is a need, therefore, to quantify the interaction between the amount of time animals spend in livestock buildings and the size of NH3 losses from those buildings. A more extensive dataset is also needed to provide a robust estimate of seasonal differences in emission. The aim of this study was to measure, under controlled and replicated conditions, the relationship between the number of hours cattle spend in buildings and the resulting NH3 emissions from those buildings.
|
MATERIALS AND METHODS
|
|---|
Housing Facilities and Cattle Details
Four replicate polytunnels were built on concrete bases measuring 8.25 by 10.00 m (Asteraki et al., 1997). Each concrete base was set at a 1° slope leading to a storage vessel to allow leachate to be collected as and when necessary. The frames were made of 5-cm pregalvanized tube with hoops at 2-m spacing, spanning the solid concrete bases, giving each polytunnel a volume of 204 m3. A white polythene covering was used to cover the frames and reduce heat transfer into the polytunnels. The sides of each polytunnel were composed of a 1.2-m high net ‘skirt’ in 55% shade/windbreak netting. The polythene covers were designed to roll down over the net to seal the polytunnels when NH3 measurements were being made. The net ‘skirt’ provided additional ventilation when left open. The front and rear doors of each polytunnel were adjustable to various heights to aid ventilation. Cattle were held in pens erected at the center of each polytunnel measuring 4.5 by 4.0 m with access to feed via a feed barrier enclosing the front of the pen, and water from a trough situated outside of the pen on one side. The rear of each pen incorporated a communal lying area measuring 4.5 m by 1.5 m bedded with wood shavings to simulate a cubicle system. This provided each animal with 2.9 m2 of space (4.6 m2 including the lying area) and is within the allowance limits recommended in the Agricultural Notebook (McConnell, 1988).
Cattle (Bos taurus) were weighed before starting each experiment to match total liveweight in each polytunnel. Red Devon cattle were used during the autumn of 2002 and spring of 2003. Charolais x Friesian and Simmental beef cattle were used during spring 2002 and autumn 2003, respectively. Liveweight gain was not assessed as the cattle were housed for just two consecutive weeks during each experiment. The cattle were fed a diet of grass silage during 2002 and hay during 2003 ad libitum (which was consistent with the diet of the animals during each experiment) for a minimum of 2 wk before the commencement of each experiment. With housing periods as short as 2 h it was not practical to measure feed N intake accurately; thus, intakes were calculated from data collected during the 2-wk period before each experiment.
Ammonia Measurements
During NH3–measurement periods the polytunnels were sealed, apart from the rear doors, which were rolled down to within 30 cm of the concrete bases. A fan (labeled A in Fig. 1
) of 45 cm in diameter positioned centrally above each of the front doors was used to draw air through the polytunnels, resulting in an air change approximately every 3.5 min ensuring adequate ventilation for the cattle during the experiment. A hand-held anemometer (Airflow Instrumentation LCA 6000 VT, High Wycombe, Buckinghamshire, UK) was used to measure the exhaust fan flow rates from the exit of the fan housing on two occasions (beginning and end) during each sampling period to ensure that the flow rate was constant.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 1. Schematic diagram of ammonia measurement system. Arrows indicate direction of air flow. A, Single phase fan (2102/450 Woods air movement); B, aluminium honeycomb anti-swirl device (Airflow Developments Ltd., High Wycombe, Bucks.); C, 6-armed sampler; D, adsorption flask; E, water trap; F, Flow meter (Type GSD, G.A. Platon Ltd., Basingstoke, Hants); G diaphragm pumps (Type B100 SE, Charles Austen Pumps Ltd., Byfleet, Surrey); H, Remus 4G 1.6 gas meter (Schlumberger Gas systems, Basingstoke, Hants).
|
|
Adsorption flasks containing 30 mL of 0.02 M orthophosphoric acid were used to capture NH3 losses from the building. Ammonia concentrations at the inlet to each polytunnel were measured from a sampler positioned on a mast at a height of 2 m. Exhaust air was sampled from a 6-armed sampler (C, Fig. 1), which was positioned in the laminar airflow 30 cm downwind of an aluminum honeycomb anti-swirl device (B). Air from the incoming and exhaust air was drawn through adsorption flasks (Phillips et al., 2001) using a diaphragm pump (G). The adsorption flasks (D) were connected to the pump system using plastic airlines (3-mm i.d., 6-mm external diameter). The adsorption flask sampling from the exhaust air was positioned as close to the 6-armed sampler as possible. The rate of flow through the adsorption flasks was controlled by a flow meter (F) at approximately 4 L min–1. A gas meter (H) measured the total gas volume through each adsorption flask for any measurement period. Before passing through the pump and gas meter system, the air passed through a water trap (E) to protect the equipment from accidental contamination with orthophosphoric acid. The temperature inside each of the polytunnels was logged using Tinytalk data loggers (Gemini data loggers, Chichester, West Sussex, UK).
The NH4–N concentrations in the adsorption flasks (x, mg L–1) were determined using automated colorimetric analysis (Krom, 1980; Searle, 1984). Concentrations in the sampled air (C, mg m–3) were determined according to:
 | [1] |
where vl is the volume of liquid in the adsorption flask (L) and va (m3) the total volume of air passing through the adsorption flask. The NH3 emission rate (F, mg NH3–N h–1) from each polytunnel was then calculated according to:
 | [2] |
where Ci and Co (mg m–3) are the concentrations of NH3–N in the polytunnel inlet and exhaust air, respectively, and V (m3) is the total volume of air passing through the polytunnel in time t (h).
Adsorption flasks were changed after 2 and 4 h and every 6 h thereafter. The flux data were then converted to units of g NH3–N lu–1 d–1 (where lu is livestock unit, equivalent to 500 kg body weight), based on the mean initial liveweight of the animals in the polytunnels.
Polytunnel Validation Test
Before the study, NH3 gas (99.9%) was released within the polytunnels at a rate of 500 mL min–1, using a flow meter calibrated for NH3 gas, for 15 min through perforated tubing. Ammonia in the incoming and exhaust air was trapped using adsorption flasks for a period of 1 h after the start of the NH3 release following the procedure described above. The percentage recovery of the NH3 released was determined.
Treatments
Beef cattle, penned in groups of four, were housed for four different periods of time which constituted the experimental treatments: (i) 2 h, (ii) 6 h, (iii) 12 h and (iv) 24 h (where h represents hours spent indoors). The first two treatments were set up to represent the time cows spend indoors before and after milking during the summer. Some farms have indoor collecting yards, but on many farms much of the waiting period before milking will be spent outdoors on collecting yards; Treatments (i) and (ii) distinguish between these farms. Treatment (iii) was included to represent the housing period during an extended grazing season and Treatment (iv) to represent the standard winter housing period whereby cattle are housed all day. The treatments were allocated to the polytunnels in a Graeco-Latin square design, with four replicates of each treatment. Each part of the Graeco-Latin square took 48 h and the four replicates were completed during two consecutive weeks. The NH3 loss measurements commenced simultaneously with introduction of the cattle and continued for 24 h with cattle from Treatments (i) through (iv) being removed at appropriate times of the day. The NH3 loss measurements then continued for an additional 24 h to provide information on ‘background’ emissions from floors in the absence of cattle.
Each experiment was repeated during the spring and autumn of 2002 and 2003. During 2002 all polytunnels were scraped immediately after the last group of cattle were removed (i.e., after 24 h) to simulate commercial practice, and the communal lying area received fresh wood shavings. During 2003 scraping and bedding was performed concurrently while each group of cattle was removed from the house to represent best practice. Urease activity of the polytunnel floors was not assessed but assumed to be comparable with that found in commercial buildings due to occupation of the polytunnels by cattle for at least 6 mo of the year for the previous 7 yr.
Statistical Analyses
The Student's t test was used to test the actual NH3 recoveries measured during the polytunnel validation test against the null hypothesis of 100% recovery. Preliminary statistical analysis on the experimental data was performed using the analysis of variance procedure of GENSTAT (Lawes Agricultural Trust, 1993) incorporating a Graeco-Latin square design. This showed that the cattle and housing were not significant factors. Therefore, a Latin square approach was adopted to gain degrees of freedom. The Latin square design ensured that over the course of each experiment, each polytunnel and each group of cattle was used for all four treatments. Significant differences were further investigated using the Fisher's Protected Least Significance Difference (FPLSD) test at the 5 and 10% level of significance (Hsu, 1996). Regression analysis was done to explore any links between temperature and NH3 emission.
|
RESULTS
|
|---|
Nitrogen Intake and Excretion
The dry matter (DM) and total N content of the feed used during each experiment are summarized in Table 1. Dry matter intake largely reflected the mean initial liveweight of the cattle; 342 and 223 kg during the spring and autumn of 2002, and 318 and 394 kg during the spring and autumn of 2003, respectively. The total N intake ranged from 42.7 to 81.4 g head–1 d–1. The hay fed during the 2003 experiments had a typically greater DM of 89% compared with the silage (23.1%, 25.6%) fed during 2002.
It was considered impractical to measure excretion with cattle housed for as little as 2 h in one treatment. Thus, the total ammoniacal N (TAN) excreted by the cattle was estimated from the standard assumption that, for ruminants, TAN is 60% of total N excretion, i.e., that voided by the animals before losses take place (Webb, 2000). For fattening beef cattle of 450 kg live weight, annual N excretion was estimated at 68 kg N yr–1, with 47 kg yr–1 for animals of 250 kg live weight. The N and TAN excretion data for cattle used in this study were interpolated from these figures.
Polytunnel Validation Test
The mean NH3 recoveries are presented in Table 2. There was a significant difference in percentage recovery to 100% shown using a one-sample t test (P < 0.001); therefore, the NH3 flux data from each polytunnel was adjusted accordingly.
Temperature
The mean, maximum, and minimum polytunnel temperatures during each experiment are presented in Table 1. The overall minimum, maximum, and mean temperatures were highest, and very similar, during the two spring experiments. The maximum temperature reached inside the polytunnels did on occasion cause concern with regard to animal welfare but proved to be short lived. The lowest temperatures were recorded during the autumn 2003 experiment when temperatures frequently dipped below freezing inside of the polytunnels. Regression analysis failed to find a significant link between temperature and NH3 emission.
Ammonia Emissions
Overall Results
Mean exhaust fan flow rates during the study were 1.0, 1.0, 0.9, and 1.1 m3 s–1 for polytunnels 1 to 4, respectively, these being typical of those measured during previous experiments. Average emissions after both 24 and 48 h generally increased with increasing occupancy of the polytunnels (Tables 3 and 4). There was a significant decrease in average NH3 emissions from housing animals for 6 h or less compared with 24 h. The NH3 loss was not directly proportional to the amount of time cattle spent in the polytunnel, e.g., the NH3 emissions from cattle housed for 12 h were not 50% of the cattle housed for 24 h.
View this table:
[in this window]
[in a new window]
|
Table 3. Ammonia loss after 24 and 48 h. Experiments where the floor area was scraped after the 24-h animal group was removed from the building.
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Ammonia loss after 24 and 48 h. Experiments where the floor area was scraped immediately after each group of animals were removed from the building.
|
|
Emissions, when expressed as g N lu–1 d–1, were greater when calculated over 24 h of NH3 measurement than when calculated over 48 h of measurement, since while emissions continued after the animals were removed from the polytunnel, the rates of emission decreased over the subsequent 24 h. It follows that NH3 emissions as a percentage of the TAN excreted decreased with increased occupancy as the time available for depletion of the finite source of NH3 became limiting.
Commercial Scraping Regime (2002)
Ammonia emissions ranged from 3.0 to 24.1 g N lu–1 d–1 during the two experiments when scraping was performed after 24 h (Table 3). Emission was greatest during the spring 2002 experiment. The NH3 emission expressed as a percentage of the TAN excreted was much higher during this experiment when compared with the other experiments. Figure 2
illustrates that NH3 emission did not always decline immediately after removal of the animals from the house. Furthermore, scraping of the concrete floor after 24 h did not always result in an immediate reduction in emission.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 2. Ammonia emission rate during (a) spring 2002 and (b) autumn 2002. Dashed line shows when scraping occurred.
|
|
The emission from cattle housed for 2, 6, and 12 h equated to 26, 54, and 90%, respectively, of the emission from cattle housed for 24 h during spring 2002 after 24 h of measurement (Table 3), showing that the NH3 loss was not directly proportional to the amount of time cattle spent in the polytunnel. After 48 h of NH3 measurement, the mean emission from cattle housed during spring 2002 for 2, 6, and 12 h was less at 21, 44, and 79%, respectively, of the emission from the 24-h treatment as the emission rate declined on removal of the animals. A similar trend was seen during the autumn of 2002.
Best Practice Scraping Regime (2003)
Ammonia emissions ranged from 1.4 to 10.4 g N lu–1 d–1 (Table 4) during the two experiments when scraping was performed immediately after the animals were removed from the polytunnels. The NH3 loss was not directly proportional to the amount of time cattle spent in the polytunnel with, for example, the emission from cattle housed for 2, 6, and 12 h equating to 20, 40, and 96%, respectively, of the emission from cattle housed for 24 h during spring 2003 after 24 h of measurement. The emission increased as the animals spent more time in the polytunnel after both 24 and 48 h of measurement. This trend was not significant after 24 h of measurement during autumn 2003. The emission rates measured during the 2003 experiments are illustrated in Fig. 3
. The emission rate declines as would be expected, although not always immediately, on removal of the animals and scraping of the floor.
|
DISCUSSION
|
|---|
Comparison with Existing Literature
The results in this study were outside the range of those (29.0 to 51.0 g N lu–1 d–1) used to compile the UK Ammonia Emissions Inventory (UKAEI) for cattle on cubicles (Misselbrook et al., 2004). The emission factor (EF) for beef cattle on cubicles is estimated at 34.3 g N lu–1 d–1 (Misselbrook et al., 2000) and is greater than may be expected because it was derived from measurements on dairy cattle housed in cubicles (Demmers, 1997; Demmers et al., 1997), which would have a much greater N intake.
As the authors of the UKAEI discovered in compiling emission factors for beef cattle on cubicles, the data presented in existing literature is mainly focused on emissions from dairy cow houses or beef on litter or slat systems. This makes it difficult to compare the results of this study in context, but the results discussed below do allow for an indirect comparison. Groot Koerkamp et al. (1998) measured mean EF's for beef cattle on slats of 20.5, 21.6, and 8.9 g N lu–1 d–1, generally at the upper end of those measured in this study. There is evidence to suggest, however, that emissions from cattle on sloped solid floors (as used in this study) can be 50% of those from cattle on slats (Swierstra et al., 1995); broadly agreeing with the results of this study. Braam et al. (1997) also found emissions from a one-sided sloped floor to be up to 25% less than from a slatted floor. The reduction in emissions was attributed to the absence of a slurry store beneath the slats, which has been shown to have a significant contribution to the total emission (Kroodsma et al., 1993).
Time Spent Inside in Relation to NH3 Emission
Reducing the time cattle spend in buildings was found to lead to a disproportionate reduction in NH3 emissions from housed cattle. Cattle in all treatments were introduced to the polytunnels at 0900 h; thus, for the 2 to 12 h treatments the period in which the cattle were housed was predominately during daylight hours. However, emissions from cattle housed for 24 h would have encompassed a period of relative inactivity overnight and thus a proportionate relationship may not be expected.
A significant reduction in NH3 emission resulted when housing periods were reduced from 24 to 6 h or less per day. A recent desk study (Webb et al., 2005) concluded that extended-season grazing by dairy cattle may reduce total UK agricultural emissions by 1 to 2%. The results of this study suggest that unless animals are housed for less than 6 h d–1 this potential abatement may not be achieved. However, when seasonal fluctuations in temperature are taken into account, the situation becomes less clear cut. Each individual experiment in this study has shown that increasing the time that animals are housed increases NH3 emissions per se. However, the link between NH3 emissions and temperature is well established; for example, Monteny et al. (1998) found NH3 emissions from a cow house to increase by c. 50% as the mean temperature increased from 9 to 17°C. Furthermore, Phillips et al. (1998) and Kroodsma et al. (1993) found that NH3 emissions in the summer, when the cows were in the buildings for a proportion of the day, were greater than in winter when the cows were housed all day. Although, on an area basis the summer and winter emissions measured by Phillips et al. (1998) were almost identical over the 4- and 24-h housing periods; the area measured in the summer was only 56% of the area that was soiled in the winter.
To provide more information on seasonal effects, each experiment in this study was replicated in the spring and autumn. Emissions during the spring of 2002 were much greater than during the autumn, although not significantly different, as may be expected given that the temperature was much higher leading to an increased mass transfer of NH3 through the liquid/gas boundary (Elzing and Monteny, 1997). However, the initial liveweight of the animals used during the spring of 2002 and their resultant N intake was much higher than during the autumn, and this would have had a major influence on emission. Incidentally, in 2003, emissions during the autumn were marginally greater than those measured during the spring even though the temperatures were much lower in the autumn. It is possible that the greater initial liveweight and consequently N intake of the autumn cattle outweighed any effect of seasonal temperature. Thus, it is difficult to pinpoint a single reason for any variations during spring and autumn for each of the years because of too many confounding factors (e.g., cattle breed and liveweight), which also precludes any statistical analysis across the individual experiments.
Scraping Regime
For the same reasons as discussed above, statistical analysis was not performed across years and therefore scraping regimes. The ideal experiment, had resources permitted, would have incorporated both the two scraping regimes and the four experimental treatments requiring eight polytunnels. There is, however, some scope for comparison across the four experiments because the first 24 h of measurement of the 24-h treatment for all four experiments was identical in terms of the time the animals were housed and the point at which scraping occurred. Figures 2 and 3 show the variation in rates of NH3 emission from the 24-h treatment in all four experiments. A similar pattern of emission was seen during each experiment, but it must be noted that the size of emission was different for each year. The cumulative loss after 24 h reflects this with a loss of 24.1 and 13.6 g N lu–1 d–1 during the spring and autumn of 2002 and 8.1 and 10.4 g N lu–1 d–1 during the spring and autumn of 2003. This variation may be the result of the interactions of one or more factors making it difficult to decipher any potential benefit of ‘best practice’ compared with ‘commercial practice’ scraping.
Following removal of the animals and/or scraping, the emission rate on occasion temporarily increased before declining. This is contrary to the findings of two Dutch studies by Kroodsma et al. (1993) and Monteny et al. (1998), which showed a rapid decrease in NH3 emissions after dairy cows left the house. Clearly emissions will take place from soiled surfaces in the absence of animals, but emissions cannot continue indefinitely. It is possible that the increased activity associated with loading the cattle into the livestock trailer to remove them from the polytunnel temporarily increased emissions. Kroodsma et al. (1993) measured two peaks in emissions during the day, which was attributed to the fact that cows produce more feces and urine before and after the twice-daily milking. Emissions are also affected by scraping, which serves not only to remove urine and feces, but also to spread urine over an area larger than the initial puddle, resulting in both a positive and negative effect (Braam et al., 1997). This was also seen in a recent study at IGER (unpublished), where NH3 emission rates from beef cattle housed on a slurry-based system for a 6-mo winter housing period (in the IGER polytunnels) were not significantly different before and after daily scraping.
Incorporation of Results into Inventories
To make unbiased estimates of the cost-effectiveness of measures to reduce NH3 emissions, calculation of national NH3 emissions may also be made using a mass-flow approach (Menzi et al., 1998; Webb and Misselbrook, 2004). In these approaches EF for NH3 are calculated as a percentage of the TAN deposited within the emission source. The UK NARSES model (Webb and Misselbrook, 2004) uses the same EF for the winter housing period and the summer period when cattle enter buildings during the milking period. The results presented here do not suggest using a different EF but have highlighted the complexities and pitfalls of producing such an EF.
The replacement of the current EF for all cattle (31% of TAN) with the average of the four values reported here for occupancy of 6 h over a measurement period of 24 h (20% of TAN) would decrease the current estimate of NH3 emissions from UK livestock production by c. 2.57 kt NH3–N or 1.2% of the current total from livestock production. The main reason for this small decrease is that current data suggest 65% of dairy cows, the source of the majority of NH3 emissions from slurry-based cattle systems, congregate on outdoor unroofed areas when waiting to be milked rather than enter the buildings (Webb et al., 2001). Hence, any decrease in the EF will only apply to 35% of dairy cows.
|
CONCLUSIONS
|
|---|
Our results indicate that NH3 emissions decrease as the proportion of the day cattle spend in the buildings decrease, although the trend is not linear. To significantly reduce NH3 emission per se, when compared with 24 h occupancy, cattle should occupy buildings for no more than 6 h. Thus, the approach of spending a portion of the day, c. 4 to 12 h, grazing during the winter is unlikely to reduce NH3 emissions from buildings or overall.
|
ACKNOWLEDGMENTS
|
|---|
This research was funded by the UK Department for the Environment, Food, and Rural Affairs (DEFRA). IGER is supported by the Biotechnological and Biological Sciences Research Council (BBSRC).
|
REFERENCES
|
|---|
- Asteraki, E.J., R.A. Matthews, and B.F. Pain. 1997. Ammonia emissions from beef cattle bedded on straw. p. 343–347. In J.A.M. Voermans, and G.J. Monteny (ed.) Ammonia and odour control from animal production facilities. Nederlanse Vereniging Techniek Landbouw, Rosmalen, the Netherlands.
- Braam, C.R., M.C.J. Smits, H. Gunnink, and D. Swierstra. 1997. Ammonia emission from a double-sloped solid floor in a cubicle house for dairy cows. J. Agric. Eng. Res. 68:375–386.[CrossRef]
- Demmers, T.G.M. 1997. Ventilation of livestock buildings and ammonia emissions. Ph.D. thesis. University of Nottingham, Nottingham, UK.
- Demmers, T.G.M., V.R. Phillips, J.L. Short, L.R. Burgess, R.P. Hoxey, and C.M. Wathes. 1997. Validation of ventilation rate measurement methods and the ammonia emission from a naturally-ventilated UK dairy and beef unit. p. 219–230. In J.A.M. Voermans, and G.J. Monteny (ed.) Ammonia and odour control from animal production facilities. Nederlanse Vereniging Techniek Landbouw, Rosmalen, the Netherlands.
- Elzing, A., and G.J. Monteny. 1997. Modelling and experimental determination of ammonia emission rates from a scale model dairy-cow house. Trans. ASAE 40(3):721–726.
- Groot Koerkamp, P.W.G., H.M. Metz, G.H. Uenk, V.R. Phillips, M.R. Holden, R.W. Sneath, J.L. Short, R.P. White, J. Hartung, J. Seedorf, M. Schröder, K.H. Linkert, S. Pederson, H. Takai, J.O. Johnsen, and C.M. Wathes. 1998. Concentrations and emissions of ammonia in livestock buildings in Northern Europe. J. Agric. Eng. Res. 70:79–95.
- Heil, G.W., and W.H. Diemont. 1983. Raised nutrient levels change heathland into grassland. Vegetatio 53:113–120.[CrossRef][ISI]
- Hsu, J.C. 1996. Multiple comparisons theory and methods. Chapman & Hall, London.
- Jarvis, S.C., and S. Ledgard. 2002. Ammonia emissions from intensive dairying: A comparison of contrasting systems in the United Kingdom and New Zealand. Agric. Ecosyst. Environ. 92:83–92.[CrossRef]
- Krom, M.D. 1980. Spectrophotometric determination of ammonia: A study of a modified Berthelot reaction using salicylate and dichloroisocyanurate. Analyst 105:305–316.[CrossRef]
- Kroodsma, W., J.W.H. Huis in't Veld, and R. Scholtens. 1993. Ammonia emission and its reduction from cubicle houses by flushing. Livestock Prod. Sci. 35:293–302.[CrossRef]
- Lawes Agricultural Trust. 1993. Genstat 5, Release 3 Reference Manual. Oxford Univ. Press, Oxford, UK.
- McConnell, P. 1988. Part 3: Farm equipment; Chapter 22 Farm buildings. p. 562. In Halley, R.J. and R.J. Soffe (ed.) Primrose McConnell's The Agricultural Notebook, 18th ed., Butterworth & Co., London.
- Menzi, H., P.E. Katz, M. Fahrni, A. Neftel, and R. Frick. 1998. A simple empirical model based on regression analysis to estimate ammonia emissions after manure application. Atmos. Environ. 32(3):301–308.[CrossRef]
- Misselbrook, T.H., D.R. Chadwick, B.J. Chambers, K.A. Smith, J. Webb, T. Demmers, and R.W. Sneath. 2004. Inventory of Ammonia Emissions from UK Agriculture– 2002. Report to Defra of Project AM0127, DEFRA, London.
- Misselbrook, T.H., T.J. van der Weerden, B.F. Pain, S.C. Jarvis, B.J. Chambers, K.A. Smith, V.R. Phillips, and T.G.M. Demmers. 2000. Ammonia emission factors for UK agriculture. Atmos. Environ. 34:871–880.[CrossRef]
- Monteny, G.J., D.D. Schulte, A. Elzing, and E.J.J. Lamaker. 1998. A conceptual mechanistic model for the ammonia emissions from free stall dairy cow houses. Trans. ASAE 41(1):193–201.
- Phillips, V.R., S.J. Bishop, J.S. Price, and S. You. 1998. Summer emissions of ammonia from a slurry-based, UK, dairy cow house. Bioresour. Technol. 65:213–219.[CrossRef]
- Phillips, V.R., D.S. Lee, R. Scholtens, J.A. Garland, and R.W. Sneath. 2001. A review of methods for measuring emission rates of ammonia from livestock buildings and slurry or manure stores: II. Monitoring flux rates, concentrations and airflow rates. J. Agric. Eng. Res. 78(1):1–14.
- Searle, P.L. 1984. The Berthelot or Indophenol reaction and its use in the analytical chemistry of nitrogen. Analy. (Cambridge, UK) 109:549–568.
- Swierstra, D., M.C.J. Smits, and W. Kroodsma. 1995. Ammonia emission from cubicle houses for cattle with slatted and solid floors. J. Agric. Eng. Res. 62:127–132.[CrossRef]
- van Breemen, N., and H.F.G. van Dijk. 1988. Ecosystem effects of atmospheric deposition of nitrogen in the Netherlands. Environ. Pollut. 54:249–274.[CrossRef][Medline]
- van der Eerden, L.J.M. 1982. Toxicity of ammonia to plants. Agric. Environ. 7:223–235.
- Webb, J. 2000. Estimating the potential for ammonia emissions from livestock excreta and manures. Environ. Pollut. 111:395–406.[CrossRef]
- Webb, J., S.G. Anthony, L. Brown, H. Lyons–Visser, C. Ross, B. Cottrill, P. Johnson, and D. Scholefield. 2005. The impact of increasing the length of the cattle grazing season on emissions of ammonia and nitrous oxide and on nitrate leaching in England and Wales. Agric. Ecosyst. Environ. 105:307–321.[CrossRef]
- Webb, J., and T.H. Misselbrook. 2004. A mass-flow model of ammonia emissions from UK livestock production. Atmos. Environ. 38:2163–2176.[CrossRef]
- Webb, J., T.H. Misselbrook, B.F. Pain, J. Crabb, and S. Ellis. 2001. An estimate of the contribution of farm concrete yards to the UK inventories of ammonia, nitrous oxide and methane. Atmos. Environ. 35:6447–6451.[CrossRef]
TOP
ABSTRACT
INTRODUCTION
