Due to increasing concerns about odorous air pollution, the USEPA and the National Institute on Deafness and Other Communication Disorders (NIDCD) cosponsored a workshop at Duke University in 1998 to assess our current state of knowledge regarding the health effects of ambient odors (see Schiffman et al., 2000). Special emphasis was placed on potential health issues associated with odorous emissions from animal manures and other biosolids. To address this issue, workshop participants defined levels of odor exposure to clarify the intensities associated with potential health effects (see Table 2). Participants concluded that at least three mechanisms exist by which ambient odors may produce health symptoms in communities with odorous manures and biosolids. In Mechanism 1, symptoms can be induced by exposure to odorants (compounds with odor properties) at levels that also cause irritation or other toxicological effects. That is, irritation—rather than the odor—is the cause of the health symptoms, and odor (the sensation) simply serves as an exposure marker. An example is ammonia with an odor threshold of 0.8 ppm (v/v) and an irritation threshold of 4 to 8 ppm (v/v). At concentrations of 4 to 8 ppm and above, odor is merely coincident with the more relevant irritative process, and health symptoms are more likely caused by irritation rather than "odor-induced." In Mechanism 2, health symptoms can occur at odorant concentrations that are above odor thresholds but are not irritating, which typically occur with exposure to certain odorant classes such as sulfur-containing compounds (for example, hydrogen sulfide, H2S). The odor threshold for H2S ranges from 0.5 to 30 ppb (v/v) for 83% of the population while the irritant threshold ranges from 2.5 to 20 ppm (v/v). Six community studies (Jaakkola et al., 1990, 1991; Haahtela et al., 1992; Kilburn and Warshaw, 1995; Legator et al., 2001; Campagna et al., 2000) have reported that exposure to H2S at nonirritant concentrations is associated with health symptoms. In Mechanism 3, the odorant is part of a mixture that contains a copollutant (such as a pesticide or bacterial endotoxin) that is fundamentally responsible for the reported health symptom. Workshop participants emphasized the importance of using objective biomarkers to determine if health complaints constitute health effects. In addition, participants also concluded that far better technologies for mitigating odor are necessary to reduce any potential health effects.
Evidence for Mechanism 1: Irritation Rather than the Odor Causes the Health SymptomsTo understand Mechanism 1, it is necessary to describe the basics of odor physiology. Odors are sensations that occur when compounds (called odorants) stimulate receptors in the nasal cavity. Odorants can induce sensations in two ways: (i) interaction with odorant receptors in the olfactory epithelium in the top of the nasal cavity and (ii) stimulation of free nerve endings in the nose, throat, and lungs at elevated concentrations. When volatile compounds activate odorant receptors, signals are transmitted via the olfactory nerve (first cranial nerve) to the olfactory bulb and ultimately to the brain. The odor sensations that are induced by this process are described by adjectives such as floral, fruity, earthy, fishy, fecal, and urinous. When odorous compounds also activate free nerve endings in the upper and lower respiratory system (via the trigeminal and vagus nerves respectively), sensations such as irritation, tickling, burning, stinging, scratching, prickling, and itching are induced. For Mechanism 1, irritancy occurs at a concentration above—but within an order of magnitude of—the odor threshold. That is, concentration at which irritancy is first detected is between 3 and 10 times higher than the concentration at which odor is first detected. Examples of odorous compounds in the home or office that become irritants at concentrations somewhat above their odor thresholds include ammonia, chlorine, camphor, menthol, alcohol, and formaldehyde (for example, from building products) as well as acrolein, acetaldehyde, and organic acids (for example, from cigarettes). Thus, at concentrations at or above the irritant threshold, both odor and irritant sensations occur simultaneously. Odor is merely coincident with the more relevant irritative process, and health symptoms are more likely caused by irritation rather than "odor-induced." Odor sensations are simply a warning that potential health symptoms can occur at elevated concentrations.
Sensory irritation can be induced by a single odorous compound above its irritant threshold or by the aggregate effect of low concentrations of compounds (although each individual chemical constituent is below its irritant threshold concentration) (Cometto-Muñiz and Cain, 1992; Cometto-Muñiz et al., 1997, 1999; Korpi et al., 1999). Agonistic effects can even occur when subthreshold concentrations of multiple individual volatile organic compounds (VOCs) combine to produce odor and noticeable sensory irritation. When irritant compounds or mixtures come in contact with the upper and/or lower airway, many systemic responses can occur including: (i) altered respiratory rate, depending on the primary level of irritation (upper versus lower); (ii) reduced respiratory volume; (iii) increased duration of expiration; (iv) contraction of the larynx and bronchi and increased bronchial tone; (v) increased nasal secretion, inflammation, and nasal airflow resistance; (vi) lacrimation or tearing; (vii) alterations in spontaneous body movements; (viii) increased epinephrine secretion; (ix) peripheral vasoconstriction and increased blood pressure; and (x) sneezing (Allison and Powis, 1976; Angell and Daly, 1969; Alarie, 1973; Nielsen, 1991).
Repeated exposure to odorous irritants can induce chronic respiratory disorders including asthma (Andersson et al., 2003; Tarlo and Liss, 2003; Luo et al., 2003; Yang et al., 2003). The potential induction of asthma is of special concern because its prevalence has increased 75% in the entire population (and 160% in children under the age of five) from 1980 to 1994 (Mannino et al., 1998). Asthma prevalence in rural children is comparable with that found in large cities of the U.S. Midwest (Chrischilles et al., 2004). The elevated vulnerability to environmental exposures in young children is due to the fact that they breathe more air per pound of body weight than adults (Etzel, 2003; American Academy of Pediatrics, 1993). Older adults are also vulnerable to air pollution exposures due to age-related impaired function of the lung (Kelly et al., 2003; National Academy of Sciences, 2002). Direct health care costs for asthma in the United States total more than $8.1 billion annually; indirect costs (lost productivity) add another $4.6 billion for a total of $12.7 billion (American Lung Association, 2002).
Evidence for Mechanism 2: Health Symptoms Occur at Odorant Concentrations that Are Not Irritating
Health complaints frequently occur from odorous emissions that are below irritant thresholds, especially when the odor is unpleasant (Schiffman et al., 2000, 2001). An example is the gas H2S, which smells like "rotten eggs" at low concentrations. The odor threshold for H2S ranges from 0.5 to 30 ppb (v/v) for 83% of the population while the irritant threshold ranges from 2.5 to 20 ppm (v/v). Thus, the mean odor threshold for H2S (and other sulfur-containing compounds and organic amines) tends to be three to four orders of magnitude (that is, 103 and 104 times) below the level that causes irritation or classical toxicological symptoms. Yet six community investigations have found that exposure to low levels of H2S or other reduced sulfur compounds cause health effects: (i) two studies in communities near paper mills in South Karelia, the southeastern part of Finland (Jaakkola et al., 1990; Haahtela et al., 1992); (ii) northern Finland studies of respiratory infections in children (Jaakkola et al., 1991); (iii) neurobehavioral studies near a refinery (Kilburn and Warshaw, 1995); (iv) studies in Odessa, Texas, and Puna, Hawaii (Legator et al., 2001); and (v) studies near the IBP meat packing plant in Nebraska (Campagna et al., 2000). Furthermore, two of these community studies (Jaakkola et al., 1990; Kilburn and Warshaw, 1995) reported health effects from an average daily exposure to 10 (to 11) ppb H2S (v/v).
The mechanisms responsible for health complaints to an unpleasant odor in the absence of irritation are not well understood, but several factors appear to be involved. First, humans are genetically coded such that pleasant and unpleasant (for example, H2S) odors activate different parts of the brain. Noninvasive functional neuroimaging techniques including positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) have shown that there is regional specialization in the brain based on odorant hedonic values (Fulbright et al., 1998; Zald and Pardo, 1997; Birbaumer et al., 1998). Brain structures that are activated by unpleasant experiences are preferentially stimulated when smelling H2S. Thus, aversion to unpleasant odors for the human species appears to have an evolutionary basis and is hence biologically developmentally driven. That is, there appears to be a biological imperative based on anatomy of the nervous system that alerts humans to avoid certain unpleasant odors associated with potentially unsafe food and air (similar to the gag reflex from tasting something excessively sour or bitter, or the reflex action of withdrawing the hand after accidentally touching something hot). Second, exquisite sensitivity of the nose to hydrogen sulfide gas (H2S) may be a protective mechanism to prevent dysregulation of normal H2S metabolism. Hydrogen sulfide gas is produced endogenously during metabolism of sulfur-containing amino acids, and it functions as a neuromodulator in the brain as well as a regulator of the tone in smooth muscle (Kimura, 2000; Hosoki et al., 1997). A small increase in sulfide levels less than twofold greater than endogenous values is lethal (Warenycia et al., 1989). Even small changes in the brain may affect behavior (see Reiffenstein et al., 1992). Third, unpleasant odors can modulate breathing patterns and thus can potentially affect health and well-being. The RD50 values (concentrations that induce a 50% decrease in respiratory rate) for a random sample of unpleasant smelling compounds were much lower than for pleasant smelling compounds (Gift and Foureman, 1998, as reported by Schiffman et al., 2000). Furthermore, if the odors are strong, shallow and irregular breathing can occur due in part to the fact that sniff volume is inversely proportional to the concentration of the odorant (Laing, 1983; Schiffman et al., 2000). Fourth, exposure to malodors may cause or exacerbate illnesses because they impair mood and induce stress. Many studies have shown that unpleasant odors including H2S impair mood (Ehrlichman and Bastone, 1992; Schiffman et al., 1995; Kilburn and Warshaw, 1995). For example, residents living near large-scale hog operations were found to have increased levels of tension, depression, anger, fatigue, and confusion as measured by the profile of mood states (POMS) when malodors were present (Schiffman et al., 1995). This mood impairment may be due in part to the fact that the exposure to malodor was involuntary. Mood impairment and stress have been associated with development of coronary artery disease, chronic hypertension, and structural changes of the heart in some studies (Karasek et al., 1981; Johnson and Hall, 1988; Schnall et al., 1990). Finally, conditioned or learned associations may play a role in perceptions and health symptoms induced by malodors (Shusterman, 1992; Simon et al., 1990; Dalton and Wysocki, 1996; Karol, 1991). For example, if an unpleasant odor has previously been associated with flu or allergic symptoms, the odor alone may subsequently recreate these symptoms in the absence of flu virus or allergy.
Evidence for Mechanism 3: A Copollutant in an Odorous Mixture Is Responsible for the Reported Health Symptom
Odorant mixtures may contain (i) nonodorous copollutants such as nitrogen dioxide (NO2) and/or carbon monoxide (CO), (ii) particulates, or (iii) toxicants from mold that are the actual cause of health effects. Odors can arise from incomplete combustion of fuel with oxygen (Schiffman et al., 2000). However, the harmful effects of the combustion may be due to odorless components such as NO2 and/or CO. Particulate exposure also elevates the incidence of respiratory symptoms and can increase the risk of respiratory and cardiovascular morbidity including increased hospital admissions or emergency room visits for asthma or other respiratory problems. Health effects can begin to occur when ambient particles smaller than a 10 µm fall between 30 and 150 µg m–3 (Committee of the Environmental and Occupational Health Assembly of the American Thoracic Society, 1996). Particulates in indoor air can arise from stoves, fireplaces, chimneys, tobacco smoke, hair, skin, molds, and pollen. Sources of particulates in outdoor air can arise from motor vehicles, industrial facilities, residential wood burning, and outdoor burning. In rural communities, particulates are also emitted from intensive animal operations and include manure, molds, pollen, grains, feathers, endotoxin, and feed dust. A recent study suggests that adverse effects of particulates are augmented by the presence of an odorous compound (Donham and Cumro, 1999).
Sustainable Agriculture Necessitates Mitigation of Odorous Aerial Emissions
One of the main conclusions from the workshop at Duke University sponsored by the USEPA and National Institute on Deafness and Other Communication Disorders (NIDCD) (see above) was that sustainable animal agriculture necessitates the development of technologies for reducing odorous emissions to blunt potential human health effects. During the past decade, trends in animal production agriculture have been toward intensive industrial systems in which less than 10% of the feed for the animals is produced within the production (or farm) unit. While intensive systems are effective at addressing the world's escalating demand for affordable meat products, their effect on both human health and the environment will determine the future of animal agribusiness in many parts of the world. The environmental issues are often geographically specific but, in general, include animal manure management; production-associated consumption of limited water resources; and aerial emissions including ammonia, hydrogen sulfide, methane, nitric oxide, nitrous oxide, volatile organic compounds (VOCs), endotoxins, exotoxins, particulate matter, and odorants (Williams, 2002). Particulates and odor emissions are of particular importance, especially because of the potential effects that these components have on human health (Schiffman et al., 2000).
North Carolina represents a state in the United States in which much activity has occurred over the past decade relative to pork production agriculture and serves as a model for the rapid growth of the industry, associated environmental issues, and efforts to develop new technology to address the issues. Between 1991 and 1997 the swine inventory in the state increased by approximately 300% from 2.7 million head to approximately 10 million head. However, since 1997 the number of facilities and the number of animals has remained stable due, in part, to a state-mandated moratorium on development of new facilities that use traditional waste management treatment processes. Expansion or new facilities can only occur with the implementation of "innovative" or "environmentally superior" technologies.
Technologies for Mitigating Aerial Emissions
In North Carolina a research, development, and demonstration initiative is underway to identify technologies capable of addressing aerial emission concerns and other environmental effects associated with concentrated swine production operations. The initiative is sponsored through agreements between the Attorney General of North Carolina and Smithfield Foods and Premium Standard Farms to develop "environmentally superior technologies" (EST) for implementation onto farms located in North Carolina that are owned by these companies (Williams, 2002, 2003a, 2003b). Swine waste treatment technology development under these agreements includes a covered in-ground anaerobic digester, a sequencing batch reactor, an upflow biological aerated filter system, mesophilic and thermophilic anaerobic digesters, energy recovery systems, greenhouse vegetable production system, solid separations systems, constructed wetlands system, nitrification–denitrification systems, soluble phosphorus removal systems, belt manure removal systems, gasification system to thermally convert dry manure to a combustible gas stream for liquid fuel recovery, ultrasonic plasma resonator system, manure solids conversion to insect biomass for value-added processing into animal feed protein meal and oil system, reciprocating water technology system, and a dewatering–drying–desalinization system.
Technology Descriptions
Descriptions and process flow diagrams for most of these systems have been published elsewhere (Williams, 2002, 2003a, 2003b; Havenstein, 2003). General mechanisms of how these technology processes may reduce odor emissions are enumerated in Table 3. Environmental performance analysis for these technologies includes an integrated program approach in which each is systematically analyzed for emissions of odor (Schiffman et al., 2003). Following are overview summaries for some of the candidate EST technologies in which odor remediation data have been procured to date.
Covered In-Ground Anaerobic Digester and Nitrification Biofilter
This system, located on the Julian Barham Farm in Johnson County, North Carolina, is comprised of an impermeable high-density polyethylene cover over an earthen lined digester that operates under ambient temperature conditions. Liquid manure from approximately 4000 sows housed in six buildings is conveyed to the digester. Biogas that is produced during the anaerobic digestion is extracted and conveyed to a generator where electricity is produced for use on the farm. Treated effluent from the digester flows into a storage pond, some of which is further treated in trickling nitrification biofilters. The nitrified effluent from the biofilters is used to flush the six swine buildings or for fertilization of tomato plants in greenhouses located on the farm. An aerial view of the treatment system is shown in Fig. 1 .
Solids Separation and Reciprocating Wetland
This technology is located on the Corbett Farm 2 in Duplin County, North Carolina. The reciprocating wetland component represents a wastewater treatment process developed by the Tennessee Valley Authority's (TVA) Environmental Research Center. The reciprocating wetlands are comprised of two cells (basins), filled with aggregate media, which alternately drain and fill on a recurrent basis. The draining and filling cycles create aerobic, anaerobic, and anoxic conditions within the cells, providing both biotic and abiotic treatment processes to provide nitrification, denitrification, and phosphorus removal. The liquid manure entering the cells is previously processed through a belowground settling tank for solids separation. An aerial view of the treatment system is shown in Fig. 2 .
Upflow Biological Aerated Filter System
This technology system, designed and operated by Ekokan LLC, was housed on Murphy-Brown Farm 93, located in Bladen County, North Carolina. The system treated wastewater from five hog buildings containing approximately 800 finishing pigs each. The wastewater was initially processed through a solids separation unit to remove course solids. Subsequently, the wastewater was treated through first- and second-stage aerated upflow biofilters connected in series (two units, four biofilters total). Each biofilter contained plastic fixed media providing surface area for a biofilm of microorganisms. Under aerobic conditions the bacteria catabolized the organic compounds in the wastewater resulting in reduced biological oxygen demand (BOD) and odorants as well as conversion of ammonia to nitrate nitrogen (nitrification). An aerial view of the treatment system is shown in Fig. 3 .
Answers to all your Biology Questions
