Chem. Senses 26: 327-331,
2001
© Oxford University Press 2001
SYMPOSIUM: AChemS XXI Symposium |
An Environmental Nuisance: Odor Concentrated and Transported by Dust
North Carolina State University, Department of Biological and Agricultural Engineering, Box 7625, Raleigh, NC 27695-7625, USA
Correspondence to be sent to: Robert W. Bottcher, North Carolina State University, Department of Biological and Agricultural Engineering, Box 7625, Raleigh, NC 27695-7625, USA. e-mail: robert_bottcher{at}ncsu.edu
Abstract
Intensive swine production generates odorous emissions which flow from the buildings housing the animals. High ventilation rates bring in fresh air, remove heat and moisture and enhance pork productivity. Numerous compounds contribute to the uniquely offensive odors from swine facilities, including fatty acids, amines, aromatics and sulfur compounds. Dust particles, which originate predominantly from feces and feed, can adsorb and concentrate odorants in swine facilities. In addition, organic particles can decay and generate odorous compounds. Odorants can exist in much higher concentrations in the dust particles than in equivalent volumes of air. Thus, inhalation of odorous dust and deposition of the dust particles in the mucus overlying the olfactory mucosa are likely responsible for some odor-related complaints by swine farm neighbors. Accurate prediction of odor transport and dispersion downwind from swine farms may require models of dust dispersion and correlation between dust and odorant levels. Unfortunately, many approaches to estimating odor impact currently incorporate filtering of air to remove particulate matter before sensing by humans or electronic sensors. Accelerated progress in understanding this and other ‘real world’ odor control problems will require methodological innovations that allow quantification of odor in response to air streams containing vapor and particulate phases.
Introduction
Odor emissions from the production buildings of intensive livestock operations (ILOs) may pose a large problem in many areas of the country. These emissions arise, in part, from the need to ventilate the production buildings by moving in large amounts of fresh air to force out excess heat, moisture, ammonia and dust from the indoor air space. ILOs generate such large quantities of biological heat, moisture and waste products that dilution ventilation is the most economical and therefore predominant method of removing indoor air contaminants from the buildings. Ventilation rates needed for animal well-being in hot weather are large, typically 70–80 air changes/h (ACH), which is 
10-fold the rate typical for most other types of buildings.
Although ventilation of livestock and poultry buildings improves animal productivity and well-being, the emission of odorants in the vapor or particulate from the buildings contributes significantly to odor problems. Schiffman et al. found that neighbors of swine farms suffered more mood disturbances and negative emotions than a control population of rural residents (Schiffman et al., 1995
). The present paper reviews factors affecting the emission of swine building odorants, interactions of odorants with dust and implications for measurement of swine odors. The desired outcome is to stimulate smell researchers to develop odor measurement methods and exposure studies to more accurately account for particulate phase odorants.
Attempts to measure swine building odors
Several methods have been applied to the measurement of livestock odors, the most popular being olfactometry and scentometry. Olfactometry involves human panelists sniffing clean air and odorous air at various well-controlled dilution ratios. The scentometer is a device which allows a human sniffer to inhale air into the nose through a chamber containing odorous air that is mixed with air cleaned by a charcoal filter (Huey et al., 1960). The size of the aperture allowing odorous air entry determines the dilution ratio. Researchers in Texas developed and evaluated dynamic olfactometers for ambient odor measurements and were able to compare olfactometer and scentometer measurements (Sorel et al., 1983; Sweeten et al., 1983). More recently, olfactometry has been applied to the measurement of swine odors at Iowa State University (Bundy et al., 1993), the University of Minnesota (Clanton et al., 1999) and Duke University (Schiffman and Williams, 1999).
Although quantifying odors and air pollutant concentrations are important in judging air quality, the emission rate of the pollutants is also significant in determining potential impacts downwind. The emission rate of an air pollutant from a building is the product of the air pollutant concentration in the exhaust air multiplied by the airflow rate. For example, the emission rate of dust from pigs measured in a European study involving the UK, Netherlands, Denmark and Germany averaged 0.031 mg/s/pig at an average pig weight of 73 kg (Takai et al., 1998). This rate is equivalent to 2.7 kg/day/1000 head; since many typical swine buildings house 1000 pigs, this also represents the dust emission rate per such building. The same types of emission rate calculations can be made for other air pollutants, such as ammonia and other odorous compounds, and is also being made for odors using measures of odor strength.
Since swine odors tend to be offensive to humans, the unpleasantness of the odor is an important parameter. However, the prevailing approach to quantifying swine odor strength has involved determining when panelists can barely detect the odor. This is the detection threshold or dilution ratio, which is measured in odor units (OU) (Chen et al., 1999). The concentration of odor at the detection limit has been defined to be 1.0 OU/m3 (Sneath and Clarkson, 2000), so that odor emissions can be expressed in odor units per second (OU/s) or odor units per second per animal unit (AU), where 1 AU = 500 kg animal weight (OU/s/AU). For example, Heber and Ni estimated odor emissions from swine finishing buildings to range from 5 to 36 OU/s/AU (Heber and Ni, 1999). Although this approach is contentious, i.e. many people have objected to the notion of ‘odor units per cubic meter’ as being physically meaningless, the measurement remains useful in some circumstances.
Although using odor units to quantify odor emissions has become popular with agricultural odor researchers, the same researchers have indicated considerable variation between laboratories in odor measurements and the methods used to quantify odor perception. This may reflect variations in the detection abilities of people as ‘odor sensing instruments’, for which little information is apparent in the literature. This problem is compounded by a marked lack of consensus in the operationalization and explicit definition of odor threshold. These issues may explain why researchers attempting to correlate odor measurements for swine building air samples using olfactometry with objective methods have not been very successful. For example, Gralapp et al. were able to correlate electronic nose and GC measurements for swine building air samples fairly well (r2 > 0.8), but not electronic nose and olfactometry (r2 < 0.1) or GC and olfactometry (r2 < 0.2) (Gralapp et al., 2000). Researchers in the chemical senses are in the best possible position to help address these deficiencies.
Complexities of the stimulus
Swine odors are typically caused by a remarkably complex mixture of odorants which can occur in the gaseous, liquid or solid (e.g. dust) phase. More than 150 compounds have been identified as contributing to swine manure odors (Eaton, 1996). Relevant odorants include alcohols, aldehydes, amines, ammonia, carbonyls, esters, indoles, mercaptans, organic acids, phenols and sulfides (Miner et al., 1975; Barth et al., 1984). Of these compounds, several researchers have identified low molecular weight volatile fatty acids, phenol, p-cresol, mercaptans, indole and skatole as contributing more to the odors (Schaefer, 1977; Spoelstra, 1980; Williams, 1984). Due to the number of compounds involved, removal of one or more individual compounds may not improve the odor quality. Additional complicating factors affecting swine building odor measurements are humidity, dust and variations in the movement of air pollutant plumes by wind.
Many people have observed dramatic increases in odor intensity and offensiveness from ILOs after rainfall or when humidity increases. Classen et al. correlated odor intensity of samples collected at the inlet to an experimental biofilter with ambient humidity and obtained a correlation coefficient of 0.91, with odor intensity increasing with humidity (Classen et al., 2000). This effect may be due to humidity altering the distribution of odorants into the vapor versus particulate phase, the rates of enzymatic reactions producing odorants, the adsorption of odorants onto dust particles and even the delivery of odorants to the human nasal mucosa. Hence, measurements of odor concentration and emission rates using odor units are presumably greatly influenced by ambient humidity, when in fact the total odorant emission rates may not be as sensitive to humidity. When air quality regulators measure odors downwind of swine farms in response to complaints or to determine a need for an odor control program their conclusions may be dramatically affected by ambient humidity. A conclusion that odors are not problematical based on measurements taken on a dry day may therefore be inaccurate.
Another major source of inaccuracy in odor measurement is dust. Dust in intensive animal housing is primarily composed of feed components and dried fecal material, but can also contain dander (hair and skin cells), molds, pollen, grains, mites, insect parts, mineral ash and, with floor-reared poultry, litter and feathers (Carpenter, 1986; Donham et al., 1986; Heber et al., 1988). Several researchers have established a strong link between dust and odors (Barth et al., 1984). Fecal particles can emit odors and organic dust particles can adsorb odorous compounds, including ammonia. Hammond et al. extracted odorous compounds, including acids, phenols and carbonyls, from dust collected from air in a swine building and concluded that the concentrations of some odorants, such as p-cresol and butyric acid, are of the order of 107 times greater in aerosols than equal volumes of air (Hammond et al., 1981). Although the data indicated that there may be five times more odorant molecules in the gas phase than are adsorbed on aerosols, removal of the aerosols reduced odors, leading Hammond et al. to suggest that aerosol particles amplify odors by concentrating odorous compounds and by deposition on the olfactory mucosa (Hammond et al., 1981). Hartung reviewed the relevant literature and noted that at least 60 compounds from different groupings were identified in dust from animal houses (Hartung, 1986). Volatile fatty acids and indolic/ phenolic compounds contributed to the strong animal house odor, with the main components being acetic acid and p-cresol, respectively (Hartung, 1986).
With dynamic olfactometry (Hobbs et al., 1995; Chen et al., 1999; Schiffman and Williams, 1999) air samples are presented to trained odor panelists over a series of dilution ratios. This method provides the detection threshold that may be used to characterize odor concentration in OU/m3. Unfortunately, the method typically involves filtering dust from the air samples before dilution and delivery to the panelists to avoid fouling of the apparatus by dust. A recent European standard for dynamic olfactometry specifically excludes odors from particulates (Sneath and Clarkson, 2000). Hence, the portion of odor responses due to dust-borne odorants is presumably not accurately represented. Regulators measuring odors using methods which filter dust particles may underestimate the problem facing a farm neighbor and effective dust control methods may be unfairly judged to be ineffective at odor control.
Another potential source of error is plume movement and dispersion by wind; as plumes of polluted air travel from a swine farm, their movement and dispersion are affected by wind shear (wind velocity profile), local circulations due to irregular terrain, turbulence in the wind and stability of the atmosphere (a tendency to resist vertical mixing, often greatest in the early morning) (ASME, 1979). Regulators should therefore not be expected to arrive at a site during atmospheric conditions identical to those associated with nuisance odor levels, but making measurements during stable atmospheric conditions with the wind approaching from a suspected odor source should help address errors due to plume behavior.
Opportunities to improve measurements
Despite the problems affecting measurement of odor parameters and emissions, some promising approaches have been developed. Scentometry allows measurement on site, which can eliminate errors due to changes in air samples occurring during transport to odor laboratories and also does not filter out dust.
Another approach which can directly incorporate dust effects involves static headspace sampling, where air in contact with an odorous liquid or solid in a vessel is sniffed by humans. Marin compared detection thresholds from humans sniffing the static headspace above eight different odorous liquids with gas chromatography olfactometry (GCO), which involves humans sniffing odorants as they are analyzed by the gas chromatograph (Marin, 1999). This work found no statistical difference in threshold values between the two measurement methods for most compounds. This result may be helpful in that some methods of preparing odorant samples for headspace sampling can incorporate dust.
Cloth swatches which had been placed in air streams or odorous environments for specified periods of time have been used in a headspace application by inserting exposed swatches into glass containers for later sniffing by trained panelists (Miner and Licht, 1980; Schiffman and Williams, 1999). Aspirating the swatches by pulling airflow through them for differing durations allows adjustment of the odor intensity of the swatches (Miner and Licht, 1980). Since this method impregnates dust particles in the swatches, it may facilitate measurement of odors having a significant dust-borne component. Zhang et al. correlated odor intensity of aspirated cloth swatches with the amount of odorous air passed through the swatches and observed a need for moist air conditions to avoid rapid odor losses (Zhang et al., 1999). Although cloth swatches can apparently incorporate dust-borne odorants, the fabric can adsorb volatile odorants as well as dust particles. Accurate measurement of both the vaporous and particulate odors may therefore require more sophisticated techniques.
Development of standardized odor sources may also facilitate calibration of odor measurement methods and quantification of errors. For vaporous odorants synthetic mixtures designed to simulate the odor of swine manure have been developed for use in odor research. Kim-Yang et al. specified a mixture of 11 odorants which produces realistic swine odors (Kim-Yang et al., 1999). Such mixtures are useful for calibrating gas chromatographs and odor panels and may be applicable to human exposure studies. A synthetic swine dust was produced by Wathes et al. from swine feces, feed, and straw which was dried and milled (Wathes et al., 1999); the dust is remarkably odorous and realistic. Like the synthetic manure slurry for producing vaporous odorants, such an artificial dust has potential applications for calibrating measurement methods and conducting human exposure studies. It can also be used in training workshops to emphasize the odorous nature of the dust.
Needs for chemoreception research
The production and transport of odorants from swine farming operations have a significant impact on the well-being of farm neighbors and the viability of swine operations. Improved methods of quantifying odorants in the particulate and vapor phases and apportioning the relative strength of odors from particulates and vapors will be needed before the effectiveness of dust control approaches can be accurately measured. Modifications of olfactometry and aspirated fabric swatch techniques may enable such apportioning. As concern about physical health effects of air pollutants from swine farms intensifies, correlations between particulate phase odorants and respiratory effects of the particulates would be helpful in characterizing the importance of dust control. Exposure studies to determine short-term physiological effects of low levels of swine farm vapors and dust on humans would assist in developing such correlations.
Conducting such studies in environmental exposure chambers is problematical due to the need to avoid contamination of air distribution equipment by odorous dust and difficulties in generating representative mixtures of odorants from manure or synthetic odorants. However, the recent development of standardized synthetic odorant mixtures and artificial pig dust may facilitate such laboratory studies. Conducting exposure studies on or near swine farms is also problematical due to the tendency of humans to be desensitized by swine farm odors on the farm and the need to provide control (clean air) exposures while in an inherently polluted environment. The same factors responsible for intense odor sensations from exposure to dusty air in the field, i.e. the concentration of odorants on particles and difficulty in cleaning air and surfaces contaminated with such dust, makes it difficult to quantify cognitive responses to dusty air samples in laboratory settings. Chemoreception scientists are well placed to facilitate development and evaluation of odor control technologies by addressing these concerns.
Conclusions
Dust particles in swine buildings may be responsible for a considerable portion of odorant emissions from the buildings and odor perceptions by downwind neighbors of swine farms. A reduction in organic dust levels in swine building airflows has been associated with odor reduction. Methods of measuring swine odors from air samples which include dust generally require filtering of the dust to avoid instrument problems, so methods to adequately account for dust-borne odors are needed. In many cases odor control will require a reduction in dust emissions from buildings. Chemoreception scientists can help address these issues by developing improved odor measurement techniques which better account for particulate phase odorants and apportion vaporous and particulate odors and odorants and by conducting exposure studies which quantify human physiological responses to low levels of odorous dust and vapors representative of conditions downwind from swine farms.
Acknowledgments
Appreciation is extended to all of the swine producers who have cooperated in evaluating odor control systems. We also appreciate support from the Dr Susan Schiffman and the Duke University Taste and Smell Laboratory, Dr Mike Williams and the North Carolina State University Animal and Poultry Waste Management Center, the North Carolina Agricultural Research Service and the North Carolina Cooperative Extension Service. This research is part of USDA Regional Research Project S-291, ‘Systems for Controlling Air Pollutant Emissions and Indoor Environments of Poultry, Swine, and Dairy Facilities’.
References
ASME (1979) Behavior of airborne effluents. In Martin, J.R. (ed.), Recommended Guide for the Prediction of the Dispersion of Airborne Effluents, 3rd edn. The American Society of Mechanical Engineers, New York, Ch. II, pp. 16–37.
Barth, C.L., Elliot, L.F. and Melvin. S.W. (1984) Using odor control technology to support animal agriculture. Trans. Am. Soc. Agric. Engng, 27, 859–864.
Bundy, D.S., Liu, Q. and Hoff, S.J. (1993) An olfactometer for livestock odor measurement, ASAE paper no. 93-4522. American Society of Agricultural Engineers, St Joseph, MI.
Carpenter, G.A. (1986) Dust in livestock buildings—review of some aspects. J. Agric. Engng Res., 33, 227–241.
Chen, Y., Bundy, D.S. and Hoff, S.J. (1999) Using olfactometry to measure intensity and threshold dilution ratio for evaluating swine odor. J. Air Waste Manag. Assoc., 49, 847–853.[Medline]
Clanton, C.J., Nicolai, R.E. and Larson, V.J. (1999) Dynamic olfactometer variability in determining odor dilutions-to-threshold. Trans. Am. Soc. Agric. Engng, 42, 1847–1852.
Classen, J.J., Young, J.S., Bottcher, R.W. and Westerman, P.W. (2000) Design and analysis of a pilot-scale biofiltration system for odorous air. Trans. Am. Soc. Agric. Engng, 43, 111–118.
Donham, K.J., Scallon, L.J., Popendorf, W., Treuhaft, M.W. and Roberts, R.C. (1986) Characterization of dusts collected from swine confinement buildings. Am. Ind. Hyg. Assoc. J., 47, 404–410.[Web of Science][Medline]
Eaton, D.L. (1996) Swine Waste Odor Compounds. Pioneer Hi-Bred International, West Des Moines, IA.
Gralapp, A.K., Powers, W.J. and Bundy, D.S. (2000) Comparison of olfactometry, gas chromatography, and electronic nose technology for measurement of indoor air from swine facilities, ASAE paper no. 004045. American Society of Agricultural Engineers, St Joseph, MI.
Hammond, E.G., Fedler, C. and Smith, R.J. (1981) Analysis of particle-borne swine house odors. Agric. Environ., 6, 395–401.
Hartung, J. (1986) Dust in livestock buildings as a carrier of odours. In Nielsen, V.C., Voorburg, J.H. and l’Hermite, P. (eds), Odour Prevention and Control of Organic Sludge and Livestock Farming. Elsevier Applied Science, New York, pp. 321–331.
Heber, A.J. and Ni, J.Q. (1999) Odor emission from a swine finishing facility with a surface-aerated lagoon, paper no. 994129 presented at the 1999 ASAE/CSAE-SCGR International Meeting. American Society of Agricultural Engineers, St Joseph, MI.
Heber, A.J., Stroik, M., Faubion, J.M. and Willard, L.H. (1988) Size distribution and identification of aerial dust particles in swine finishing buildings. Trans. Am. Soc. Agric. Engng, 31, 882–887.
Hobbs, P.J., Misselbrook, T.H. and Pain, B.F. (1995) Assessment of odours from livestock wastes by a photoionization detector, an electronic nose, olfactometry and gas chromatography–mass spectrometry. J. Agric. Engng Res., 60, 137–144.
Huey, N.A., Broering, L.C., Jutze, G.A. and Gruber, C.W. (1960) Objective odor pollution control investigations. J. Air Pollut. Control Assoc., 10, 441–445.[Medline]
Kim-Yang, H., Wu, J., Masten, S. and Yokoyama, M. (1999) A new method for evaluating and validating odor remediation techniques. In Proceedings of the 1999 Animal Waste Management Symposium, North Carolina State University, Raleigh, NC, pp. 316–320.
Marin, A.B. (1999) A comparison of GC Olfactometry CharmTM to a static headspace method for measuring odor detection thresholds. In Impressions on Olfaction and Taste, Proceedings of the XXI Meeting of the Association for Chemoreception Sciences, Sarasota, FL. AChemS, Tallahassee, FL, p. 56.
Miner, J.R. and Licht, L.A. (1980) Fabric swatches as an aid in livestock odor evaluations. In Livestock Waste: A Renewable Resource, Proceedings of the 4th International Symposium on Livestock Wastes. American Society of Agricultural Engineers, St Joseph, MI, pp. 302–305.
Miner, J.R, Kelly, M.D. and Anderson, A.W. (1975) Identification and measurement of volatile compounds within a swine building and measurement of ammonia evolution rates from manure covered surfaces. In Managing Livestock Wastes, Proceedings of the 3rd International Symposium on Livestock Wastes. American Society of Agricultural Engineers, St Joseph, MI.
Schaeffer, J. (1977) Sampling, characterization, and analysis of malodor. Agric. Environ., 3, 121–128.
Schiffman, S.S. and Williams, C.M. (1999) Evaluation of swine odor control products using human odor panels. In Proceedings of the 1999 Animal Waste Management Symposium, North Carolina State University, Raleigh, NC, pp. 110–118.
Schiffman S.S., Sattely Miller, E.A., Suggs, M.S. and Graham, B.G. (1995) The effect of environmental odors emanating from commercial swine operations on the mood of nearby residents. Brain Res. Bull., 37, 369–375.[Web of Science][Medline]
Sneath, R.W. and Clarkson, C. (2000) A standard that ensures repeatable odour measurements. In Air Pollution from Agricultural Operations: Proceedings of the Second International Conference. American Society of Agricultural Engineers, St Joseph, MI, pp. 170–179.
Sorel, J.E., Gauntt, R.O., Sweeten, J.M., Reddell, D.L. and McFarland, A.R. (1983) Design of a 1-butanol scale dynamic olfactometer for ambient odor measurements. Trans. Am. Soc. Agric. Engng, 26, 1201–1205, 1216.
Spoelstra, S.F. (1980) Origin of objectionable odorous components in piggery wastes and the possibility of applying indicator components for studying odor development. Agric. Environ., 5, 241–260.
Sweeten, J.M., Reddell, D.L., McFarland, A.R., Gauntt, R.O. and Sorel, J.E. (1983) Field measurement of ambient odors with a butanol olfactometer. Trans. Am. Soc. Agric. Engng, 26, 1206–1216.
Takai, H., Pedersen, S., Johnsen, J.O., Metz, J.H.M., Groot Koerkamp, P.W.G., Uenk, G.H., Phillips, V.R., Holden, M.R., Sneath, R.W., Short, J.L., White, R.P., Hartung, J., Seedorf, J., Schroder, M., Linkert, K.H. and Wathes, C.M. (1998) Concentrations and emissions of airborne dust in livestock buildings in northern Europe. J. Agric. Engng Res., 70, 59–77.
Wathes, C.M., Demmers, T.G.M., Boddington, I.J. and Crook, S. (1999) Specifications, manufacture, and generation of an artificial pig dust. In International Symposium on Dust Control in Animal Production Facilities, Congress Proceedings. Danish Institute of Agricultural Sciences, Horsens, Denmark, pp. 56–66.
Williams, A.G. (1984) Indicators of piggery slurry odor offensiveness. Agric. Wastes, 10, 15–36.
Zhang, Q., Bulley, N.R., York, R., Lau, A. and Baidoo, S. (1999) Using cloth swatches for odor measurement, paper no. 994145 presented at the 1999 ASAE/CSAE-SCGR International Meeting. American Society of Agricultural Engineers, St Joseph, MI.
Accepted November 30, 2000
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||