Published online 16 October 2007
Published in J Environ Qual 36:1585-1590 (2007)
DOI: 10.2134/jeq2007.0167
© 2007 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
Quantifying Ammonia Emissions from a Cattle Feedlot using a Dispersion Model
S. M. McGinna,*,
T. K. Fleschb,
B. P. Crennac,
K. A. Beauchemina and
T. Coatesa
a Agriculture and Agri-Food Canada, 5403–1 Ave South, Lethbridge, Alberta, Canada T1J 4B1
b Dep. of Earth and Atmospheric Sciences, Univ. of Alberta, Edmonton, Canada
c Thunder Beach Scientific, Nanaimo, British Columbia, Canada
* Corresponding author (mcginns{at}agr.gc.ca).
Received for publication April 3, 2007.
 |
ABSTRACT
|
|---|
Livestock manure is a significant source of ammonia (NH3) emissions. In the atmosphere, NH3 is a precursor to the formation of fine aerosols that contribute to poor air quality associated with human health. Other environmental issues result when NH3 is deposited to land and water. Our study documented the quantity of NH3 emitted from a feedlot housing growing beef cattle. The study was conducted between June and October 2006 at a feedlot with a one-time capacity of 22,500 cattle located in southern Alberta, Canada. A backward Lagrangian stochastic (bLS) inverse-dispersion technique was used to calculate NH3 emissions, based on measurements of NH3 concentration (open-path laser) and wind (sonic anemometer) taken above the interior of the feedlot. There was an average of 3146 kg NH3 d–1 lost from the entire feedlot, equivalent to 84 µg NH3 m–2 s–1 or 140 g NH3 head–1 d–1. The NH3 emissions correlated with sensible heat flux (r2 = 0.84) and to a lesser extent the wind speed (r2 = 0.56). There was also evidence that rain suppressed the NH3 emission. Quantifying NH3 emission and dispersion from farms is essential to show the impact of farm management on reducing NH3–related environmental issues.
Abbreviations: bLS, backward Lagrangian Stochastic • DM, dry matter
|
INTRODUCTION
|
|---|
LIVESTOCK manure that is exposed to air is the main source of atmospheric ammonia (NH3), accounting for about 65% of all terrestrial NH3 emissions (National Research Council, 2002). Ammonia is produced from the anaerobic digestion of the protein found in livestock feed (Barth and Polkowski, 1974). In the rumen of cattle, NH3 is used by rumen microbes to generate microbial protein. However, if there is too much rumen-degradable protein, or if there is a lack of energy (carbohydrates) for bacterial growth, the excess NH3 is absorbed from the rumen into the blood, and is excreted as urea in urine. About half of the total nitrogen (N) excreted by cattle is lost as urinary urea (Moiser et al., 1973), which is rapidly volatilized as NH3.
Upward of 50 to 55% of the total N fed to cattle in a feedlot is lost to the atmosphere as NH3 (Eghball and Power, 1994; Bierman et al., 1999, Todd et al., 2005; Flesch et al., 2007). Erickson et al. (1998) estimated the total volatilized component as a residual of the N balance, and reported that 52 to 74% of the excreted N in manure was lost from feedlot pens. In most situations, protein is fed to livestock close to, or above the requirement for growth and maintenance because underfeeding protein can reduce weight gain. Reducing protein intake, without reducing animal performance, to reduce N excretion is a potential means of limiting NH3 emissions from livestock (Paul et al., 1998; Todd et al., 2006).
Once NH3 is in the atmosphere, a portion is deposited back to the surface while the remaining NH3 reacts with atmospheric acids to form fine aerosols. The formation of these fine aerosols can be an air quality issue (McCubbin et al., 2002; Erisman and Schaap, 2004) especially in confined air sheds. Ammonia is the precursor to the formation of fine particulate matter (PM), largely in the form of particulate ammonium (NH4+). The production of high concentrations of airborne PM poses a health risk to people (Popendorf et al., 1985) and animals (MacVean et al., 1986). In confined air sheds, high PM concentrations cause visibility degradation (Barthelmie and Pryor, 1998). The deposition of NH3 to land can also create problems due to acidification of soil (van Breemen and van Dijk, 1988), and elevated soil N concentrations causing changes in species diversity of natural ecosystems (Sutton et al., 1993; Todd et al., 2004). In addition, the loss of NH3 to the atmosphere is a loss of valuable N from the manure that would otherwise be used by crops if applied as a fertilizer.
The emission of NH3 from beef cattle has been reported for a beef feedlot in Colorado as 47 µg NH3 m–2 s–1 during the day in summer (Hutchinson et al., 1982), and in Texas emissions ranged from 55 to 93 µg NH3 m–2 s–1 over the entire day during the summer (average 70 µg NH3 m–2 s–1; Todd et al., 2005). Winter emissions in the Texas study were half the summer values. The average summer emission rate in this latter study translates to 85 g NH3 head–1 d–1. Todd et al. (2006) further calculated emissions over the entire year as 15 kg NH3 head–1, similar to the EPA annual emission factor of 11.4 kg NH3 head–1 yr–1. There is considerable uncertainty regarding the EPA emission factor since the few measurements used were highly variable (National Research Council, 2002).
Despite the large quantity of NH3 emitted from feedlots, the difficulty associated with measuring large-scale emissions has resulted in a limited number of measurements since the early work of Hutchinson et al. (1982). However, whole-farm measurements are required to deliver emission factors that reflect the real management practices on commercial farms. It is recognized that the only method capable of such measurements are micrometeorological approaches that are non-intrusive and therefore are more inclined to give accurate emission factors (McGinn, 2006). One of the simplest and most flexible of these micrometeorological methods is the "inverse-dispersion" technique. Here one uses an atmospheric dispersion model to "backtrack" a source emission rate from a downwind gas concentration (e.g., Flesch et al., 2004; McGinn et al., 2006). This has the advantage of requiring only a single concentration measurement and basic wind information, with substantial freedom to choose convenient measurement locations.
Beef cattle production accounts for about 88% of the Canadian cattle population (Stats Canada, 2006). The amount of NH3 volatilized from the large amount of excreted manure is influenced by the type of animal management. Within the beef cattle industries, several types of management systems exist that need to be considered when developing emission factors. In a typical operation, calves born in the winter/spring are kept on pasture throughout the summer and weaned in the late fall (cow-calf operation). The weaned cattle may be moved to feedlots and fed a high forage diet for 80 to 100 d (backgrounding operation). The cattle are transitioned over a period of 2 to 4 wk to a high grain diet, which they are fed for about 130 d, gaining 1.4 kg d–1 in weight (finishing operation) before being slaughtered. In some operations, the backgrounding operation is extended throughout the winter and the cattle are reintroduced to pasture in the coming spring. These cattle typically undergo a shorter finishing period (<80 d). The handling of manure accumulated in the feedlot generally occurs twice a year in the spring and fall, where the manure is removed and either stockpiled or applied directly to the field.
The objective of our study was to quantify the emission of NH3 from a finishing operation feedlot during summer and understand the variability in these emissions for Canadian operations.
|
Material and Methods
|
|---|
Study Site
A commercial feedlot was used in our study, which was located in southern Alberta 65 km from the city of Lethbridge, Canada. During the measurement campaign (1 June to 2 October), the facility contained approximately 22500 head of cattle ranging in weight from 350 to 600 kg. The diet contained about 90% rolled barley grain on a dry matter (DM) basis, whole crop barley silage, corn silage, and a mineral and vitamin supplement. The analyzed crude protein content of the diet was 12.8% ± 0.04 (DM basis; Table 1
). The entire feedlot extended over an area 800 by 760 m, and the area occupied by the pens holding cattle was 433500 m2. The remaining area included lagoons, a dugout (source of clean water), alleys, roads, and silage pits and buildings in the yard area (Fig. 1
). The feedlot was the sole point source of NH3 for many kilometers in all directions. The topography of the site was flat with no tree shelter belts in the area. However, the feedlot pens had 3-m high fences aligned along the north-south perimeter of each pen. This configuration was typical of the feedlots in this region that require shelter from the harsh winter conditions and high winds.
bLS Inverse-Dispersion Technique
Atmospheric dispersion models can be used when a surface area emits a tracer gas at a uniform but unknown rate F (g m–2 s–1), assuming an average tracer concentration C (g m–3) is measured somewhere in the plume of dispersing gas. One can predict the ratio of the tracer concentration (above background) to the emission rate, (C/F)sim, for that measurement location and calculate
 | [1] |
where Cb is the background concentration (g m–3). This is the basis of the inverse-dispersion technique.
A number of different dispersion models could provide (C/F)sim. Flesch et al. (2004) describe how a bLS (particle tracking) model is well suited for "ideal surface layer problems" where Monin-Obukhov similarity theory (MOST) describes the wind near the ground. In these cases the wind statistics needed to predict (C/F)sim can be inferred from the friction velocity u* (m s–1), the Obukhov stability length L (m), the surface roughness length z0 (m), and the average wind direction ß (degrees from north). These meteorological properties can be measured with a three-dimensional sonic anemometer.
Flesch et al. (2007) discuss how the bLS inverse-dispersion technique can be applied to a feedlot. They consider the differences between the typical feedlot environment and an ideal surface layer problem, and how the choice of measurement location can minimize the effects of these complexities in the bLS emission calculation. We closely followed their example in our feedlot application.
Field Measurements
Near the middle of the feedlot a tower was erected on which a NH3 open-path laser (GasFinder, Boreal Laser Inc., Spruce Grove, Alberta, Canada) was mounted at a height of 3.4 m above the ground. A retro-reflector was located 312 m to the south of the laser at the same height. Co-located with the NH3 laser (at 5.85 m) was a three-dimensional sonic anemometer (CSAT3, Campbell Scientific, Logan, UT). The sampling rate of the laser was 5 s and that of the sonic anemometer was at 10 Hz (10 times a second). Air temperature (acoustic) was also available from the sonic anemometer. Air pressure was recorded using 5-s sampling intervals.
The laser was checked periodically in the feedlot by comparing against a second NH3 laser (same type). This latter laser was calibrated in the laboratory using a 3-m long polyvinyl chloride (PVC) tube (10.16-cm diameter) having a reflector (tape) at one end. The PVC tube was sealed except for an inflow and exhaust port through which a known concentration of NH3 was released. The correction for the laser used for our in-feedlot measurements was 0.695 ± 0.067 of the true NH3 concentration. The correction was needed to adjust for change in the manufacturer's sensor calibration.
Data Collection
The wind and heat flux statistics needed for our bLS technique (to give u*, L, z0, and ß) are calculated in 15-min intervals by the datalogger (CR10X, Campbell Scientific, Logan, UT). The laser signal was also connected to the datalogger and the laser concentration data were averaged over the same 15-min interval. The background concentration of NH3 in our study was estimated as 10 ppbv.
The bLS dispersion model WindTrax was used to calculate (C/F)sim for each 15-min observation (Thunder Beach Scientific, Nanaimo, Canada). This software combines the bLS model described by Flesch et al. (2004) with an interface where sources and sensors are mapped. In the bLS model thousands of trajectories are calculated upwind of the laser path for the prevailing wind conditions. The important information is contained in the trajectory intersections with ground ("touchdowns") and one computes
 | [2] |
where N is the number of computed trajectories, w0 is the vertical velocity (m s–1) at touchdown, and the summation covers only touchdowns within the source (the units in F are g m–2 s–1 in Eq. [2]; hereafter we multiply the aerial emission rate by the source area and report F as an area-integrated emission rate with units of kg h–1). The touchdowns map the concentration "footprint," i.e., the ground area where emissions influence concentration.
The 15-min averaged NH3 concentrations and wind statistics were input into the WindTrax model, and a time series of 15-min average feedlot emissions was calculated. These data were filtered (Table 2
) to remove emissions coinciding with periods when the atmospheric conditions are often associated with inaccurate dispersion model calculations (Flesch et al., 2007). Three criteria were used to remove periods of potential inaccuracy in the Monin-Obukhov similarity theory: (i) removed periods where u* 
0.15 m s–1 (low wind conditions), (ii) where |L| 10 m (strongly stable/unstable atmosphere), (iii) where z0 
0.9 m (unrealistically high value). In addition we also ignored periods when the wind direction was likely to lead to unrepresentative sonic wind observations. This occurred during easterly winds, which had the combined effect of placing the sonic anemometer in the lee of the tower structure and windbreak fence. The potential is thus for the sonic to be in a region of disturbed wind flow when the winds were from the east.
Two additional criteria specific to the open-path laser were also used to eliminate periods associated with poor performance of the laser instrument. When the return light level of the laser beam dropped below the recommended minimum value of 1800 (no units) the accuracy of the concentration measurement decreased. This inaccuracy is expressed as a decrease in the r2 value (normally above 0.95), which is an output from the laser unit that shows how well the spectra from the open path match the spectra of the reference cell.
Statistical Analysis
The difference between NH3 emission on sequential days, where there was 1 d with rain and one without, were compared using a paired Student t test. Significant differences were reported for P < 0.05.
|
Results and Discussion
|
|---|
Over the study period, 1307 15-min intervals of data were collected after the data were filtered. Before this filtering, periods with no data collection existed due to misalignment of the laser with the retroreflector, low power levels, planned shutdowns for calibrations of equipment, and poor weather.
Ammonia Concentration
The NH3 concentration (calculated using a conversion factor of 1 ppbv = 0.7 µg NH3 m–3) along the open path within the feedlot was on average 560 ± 226 µg NH3 m–3. The concentration range for the 15-min averaged data was 46 to 1730 µg NH3 m–3. There was a tendency for higher concentrations under lower wind speeds where the slope of the linear fit was –56.1 µg NH3 m–3 per m s–1 (for a given emission rate, the concentration will be inversely proportional to windspeed). However, as expected, there was considerable scatter in this relationship (r2 = 0.07) due to the influence of other factors such as temperature on NH3 emissions. Including sensible heat flux (the energy exchanged between the surface and air) still did not dramatically reduce the variability associated with the predictive model (r2 increased to 0.15).
Ammonia Emission
The average NH3 emission from the feedlot during the study was 84 ± 43 µg m–2 s–1. This value is higher than the average reported by Todd et al. (2005) of 70 µg m–2 s–1 at a Texas feedlot during the summer. The higher value may be related to the quantity of manure accumulated in the pens in the two studies. Flesch et al. (2007) reported an average NH3 emission over two summer periods of 150 g head–1 d–1, which is similar to our study average emission of 140 g head–1 d–1. When calculating the average diel emissions during the study, we found emissions ranged from 36 µg NH3 m–2 s–1 at approximately 0400 h to 129 µg m–2 s–1 at about 1400 h (Fig. 2
). This relative diel range, with afternoon emissions more than tripling the early morning level, is lower than that found by Flesch et al. (2007) for a Texas feedlot (ten-fold change from 360 to 36 µg m–2 s–1). This may be related to the relative differences in temperature between the northern and southern sites, or to diet differences.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 2. Diel pattern generated from the 15-min interval data that were averaged over all days for ammonia emission and the standard deviation of the emission.
|
|
Coinciding with the peak daytime value was a peak in the sensible heat flux as determined from the sonic anemometer data with an eddy covariance technique. The coefficient of determination (r2) for the best fit linear line of emission to sensible heat flux was 0.84 (Fig. 3
). The fit of emission to wind speed was not as evident (r2 = 0.56).
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 3. Correlation between average hourly ammonia emission and sensible heat flux density (top), and wind speed (bottom) using data from the entire study period.
|
|
A second smaller peak in NH3 emissions was evident around 2200 h (Fig. 2). It is speculated that this secondary peak may be related to the observed increase in cattle activity at dusk. The increased activity may have generated urine excretion and hence NH3 volatilization. This same evening peak was seen by Flesch et al. (2007) in Texas, and was also attributed to increased animal activity at sunset.
Rain was shown to impact the magnitude of the daytime maximum NH3 emissions. On 21 September there was no rainfall and the daytime maximum emission peaked at 174 µg m–2 s–1 (Fig. 4
). During the early morning on 22 September, approximately 2.1 mm of rain fell, which coincided with a reduced NH3 emission during the subsequent daytime period. Unlike the previous dry day, there was no mid-day peak in NH3 emissions on 22 September and the average daily NH3 emission was 50 ± 12 µg m–2 s–1 compared to the previous day's value of 82 ± 36 µg m–2 s–1, a significant (P < 0.05) reduction of 38% of the daily NH3 emission. The effect of rain was not evident on 23 September, which averaged 83 µg NH3 m–2 s–1 and was significantly different (P < 0.05) from the previous wet-day emission. The substantial mid-day peak in NH3 emission (Fig. 4) on 23 September was larger than that before the rain on 21 September, perhaps related to increased surface evaporation.
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 4. The diel ammonia emission pattern on 21 September, which was a day with no rain (top), 22 September when rain occurred at mid-day (middle), and 23 September with no rain (bottom).
|
|
Other studies have shown similar results where wetting the manure reduced NH3 volatilization. For example, McGinn and Sommer (2007) reported that 6 mm of irrigation on beef cattle manure applied to the soil suppressed NH3 emissions by between 21 and 52% over a period of several days. The reason for this is that NH3 has a high affinity for absorption in water. Other studies have also reported suppressed NH3 emissions when the feedlot ground was wet followed by enhanced emissions later as the surface then dried (Hutchinson et al., 1982; Todd et al., 2005; Flesch et al., 2007).
The regression of NH3 emissions against sensible heat for 21 and 23 September (dry days) was considerably different than that found for 22 September (wet day). In the former case the coefficient of determination was 0.71 and 0.45 on the two dry days, and nonexistent (<0.01) for the wet day. This supports the findings that a wet manure pad suppresses NH3 emissions. The lower value of r2 on 23 September may be related to the continued suppression throughout the early part of the day.
Nitrogen Fraction Lost as Ammonia
The dry matter intake (DMI) per head per day was estimated to be 9.0 kg and the crude protein (CP) of the diet was 12.8%, which resulted in 0.184 kg of fed N head–1 d–1:
For the entire feedlot of 22500 head, the total N consumed per day was 4140 kg. Given the average emission of 84 µg NH3 m–2 s–1 (69 µg NH3–N m–2 s–1) and the pen area of 433500 m, it follows that an average of 2591 kg of N was lost each day from the whole feedlot. This represents about 63% of the N in the ration. This is higher than the 45% loss found by Todd et al. (2005) and similar to the 63 to 65% loss found by Flesch et al. (2007) for a Texas feedlot, and within the range of 52 to 74% loss given by Erickson et al. (1999) in Nebraska. Cole et al. (2005) reviewed work by Erickson and Klopfenstein (2001a, 2001b) and reported that total N emissions were 60% of the N fed during the summer, and 40 to 50% during the winter.
|
Conclusions
|
|---|
The typical cattle feedlot is a significant source of atmospheric NH3. Our study shows that daily emissions in summer averaged just over 3 t of NH3 from a 22500-head feedlot, or approximately 140 g head–1 d–1 for a finishing feedlot operation. It is speculated that the larger emissions rates reported in our study are related to the accumulated manure within the pen typical of the manure management practices found in southern Alberta. In the local region of the study, the one-time capacity has been regulated to 650000 head and the daily emission during summer from this region is expected to be about 91 t of NH3. Although this is a large quantity of NH3, not all is retained in the atmosphere since dry deposition to the local surfaces occurs. Further work is required to understand the net emission of NH3 from such large agricultural sources to accurately estimate the national NH3 emission inventory.
|
ACKNOWLEDGMENTS
|
|---|
This research was funded by the National Agro-Environmental Standards Initiative, the GAPS project of Agriculture and Agri-Food Canada, and the Canadian Foundation for Climate and Atmospheric Sciences (CFCAS).
|
NOTES
|
|---|
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
|
REFERENCES
|
|---|
- Barth, C.L., and L.B. Polkowski. 1974. Identifying odorous components of stored dairy manure. Trans. ASAE 17:737–741.[Web of Science]
- Barthelmie, R.J., and S.C. Pryor. 1998. Implications of ammonia emissions for fine aerosol formation and visibility impairment: A case study from the Lower Fraser Valley, British Columbia. Atmos. Environ. 32:345–352.
- Bierman, S., G.E. Erickson, T.J. Klopfenstein, R.A. Stock, and D.H. Shain. 1999. Evaluation of nitrogen and organic matter balance in the feedlot as affected by level and source of dietary fiber. J. Anim. Sci. 77:1645–1653.[Abstract/Free Full Text]
- Cole, N.A., R.N. Clark, R.W. Todd, C.R. Richardson, A. Gueye, L.W. Greene, and K. McBride. 2005. Influence of dietary crude protein concentration and source on potential ammonia emissions from beef cattle manure. J. Anim. Sci. 83:722–731.[Abstract/Free Full Text]
- Eghball, B., and J.F. Power. 1994. Beef cattle feedlot manure management. J. Soil Water Conserv. 49:113–133.
- Erickson, G.E., T.J. Klopfenstein, and C.T. Milton. 1998. Effects of two levels of dietary protein on nitrogen excretion and waste management in the feedlot. p. 67–72. In Proc. Animal production systems and the environment: An international conference on odor, water quality, nutrient management, and socioeconomic issues. Iowa State Univ., Ames, IA. 19–22 July.
- Erickson, G., T. Klopfenstein, T. Milton, and D. Herold. 1999. Effects of matching protein to requirements on performance and waste management in the feedlot. 1999 Nebraska Beef Cattle Report. Univ. Nebraska-Lincoln, Lincoln, NE. Available at http://ianrpubs.unl.edu/beef/report/mp71-24.htm (verified 12 July 2007).
- Erickson, G.E., and T.J. Klopfenstein. 2001a. Managing N inputs and the effect on losses following excretion in open-dirt feedlots in Nebraska. ScientificWorldJournal 1(S2):830–835.[CrossRef][Medline]
- Erickson, G.E., and T.J. Klopfenstein. 2001b. Nutritional methods to decrease N losses from open-dirt feedlots in Nebraska. ScientificWorldJournal 1(S2):836–843.[CrossRef][Medline]
- Erisman, J.W., and M. Schaap. 2004. The need for ammonia abatement with respect to secondary PM reductions in Europe. Environ. Pollut. 129:159–163.[CrossRef][Medline]
- Flesch, T.K., J.D. Wilson, L.A. Harper, B.P. Crenna, and R.R. Sharpe. 2004. Deducing ground-air emissions from observed trace gas concentrations: A field trial. J. Appl. Meteorol. 43:487–502.[CrossRef]
- Flesch, T.K., J.D. Wilson, L.A. Harper, R.W. Todd, and N.A. Cole. 2007. Determining ammonia emissions from a cattle feedlot with an inverse dispersion technique. Agric. For. Meteorol. 144:139–155.[CrossRef]
- Hutchinson, G.L., A.R. Mosier, and E.E. Andre. 1982. Ammonia and amine emissions from a large cattle feedlot. J. Environ. Qual. 11:288–293.[Abstract/Free Full Text]
- McCubbin, D.R., B.J. Apelberg, S. Roe, and F. Divita. 2002. Livestock ammonia management and particulate-related health benefits. Environ. Sci. Technol. 36:1141–1146.[Medline]
- McGinn, S.M. 2006. Measuring greenhouse gas emissions from point sources in agriculture. Can. J. Soil Sci. 86:355–371.
- McGinn, S.M., T.K. Flesch, L.A. Harper, and K.A. Beauchemin. 2006. An approach for measuring methane emissions from whole farms. J. Environ. Qual. 35:14–20.[Abstract/Free Full Text]
- McGinn, S.M., and S.G. Sommer. 2007. Ammonia emissions from land-applied beef cattle manure. Can. J. Soil Sci. 87:345–352.
- MacVean, D.W., D.K. Franzen, T.J. Keefe, and B.W. Bennett. 1986. Airborne particle concentration and meteorological conditions associated with incidence in feedlot cattle. Am. J. Vet. Res. 47:2676–2682.[Web of Science][Medline]
- Moiser, A.R., C.E. Andre, and F.G. Viets, Jr. 1973. Identification of aliphatic amines volatilised from cattle feedyard. Environ. Sci. Technol. 7:642–644.
- National Research Council. 2002. The scientific basis for estimating air emissions from animal feeding operations. National Academy Press, Washington, DC.
- Paul, J.W., N.E. Dinn, T. Kannangara, and L.J. Fisher. 1998. Protein content in dairy cattle diets affects ammonia losses and fertilizer nitrogen value. J. Environ. Qual. 27:528–534.[Abstract/Free Full Text]
- Popendorf, W., K.J. Donham, D.N. Easton, and J. Silk. 1985. A synopsis of agricultural respiratory hazards. Am. Ind. Hyg. Assoc. J. 46:154–161.[Web of Science][Medline]
- Stats Canada. 2006. CANSIM. Available at http://cansim2.statcan.ca/ (verified 12 July 2007).
- Sutton, M.A., C.E.R. Pitcairn, and D. Fowler. 1993. The exchange of ammonia between the atmosphere and plant communities. Adv. Ecol. Res. 24:301–391.
- Todd, R.W., N.A. Cole, and R.N. Clark. 2006. Reducing crude protein in beef cattle diet reduces ammonia emissions from artificial feedyard surfaces. J. Environ. Qual. 35:404–411.[Abstract/Free Full Text]
- Todd, R.W., N.A. Cole, L.A. Harper, T.K. Flesch, and B.H. Baek. 2005. Ammonia and gaseous nitrogen emissions from a commercial beef cattle feedyard estimated using the flux-gradient method and N/P ratio analysis. In P.J. Nowak (ed.) Proc. State of the Science: Animal manure and waste management, National Center for Manure and Animal Waste Management, San Antonio, TX. 5–7 Jan. 2005. Available at http://www.cals.ncsu.edu/waste_mgt/natlcenter/sanantonio/Todd.pdf (verified 12 July 2007).
- Todd, R.W., W. Guo, B.A. Stewart, and C. Robinson. 2004. Vegetation, phosphorus, and dust gradients downwind from a cattle feedyard. J. Range Manage. 57:291–299.[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]
TOP
ABSTRACT
INTRODUCTION
