The Professional Animal Scientist 24 (2008):1–22

Auditing and Assessing Air Qualityin Concentrated FeedingOperations1,2,3

N. A. Cole,*4 PAS, R. Todd,* B. Auvermann,† and D. Parker‡*USDA-ARS-Conservation and Production Research Laboratory, Bushland, TX 79012; †TexasAgricultural Experiment Station, Amarillo 79106; and ‡Department of Agricultural Sciences, WestTexas A&M University, Canyon 79016

ABSTRACT

The potential adverse effects of con-centrated animal feeding operations(CAFO) on the environment are a grow-ing concern. The air quality issues ofmost concern to CAFO vary but gener-ally include ammonia, hydrogen sul-fide, particulate matter, volatile organiccompounds, greenhouse gases, andodors. Air pollutants may be regulatedby federal and state laws or by nui-sance complaints. The United States En-vironmental Protection Agency, andpoultry, swine, and dairy industries re-cently agreed to the National Air Emis-sions Monitoring System to fund re-search on atmospheric emissions from

1Contribution from the USDA-ARS Conser-vation and Production Res. Laboratory,Bushland, TX 79012, in cooperation withthe Texas Agricultural Exp. St., Amarillo,and West Texas A&M Univ., Canyon.2Mention of trade names or commercialproducts in this article is solely for the pur-pose of providing specific information anddoes not imply recommendation or en-dorsement by the USDA.3Presented at the Annual ARPAS Sympo-sium Current and Future On-Farm Auditingand Assessment, San Antonio, TX, July,2007.4Corresponding author: nacole@cprl.ars.usda.gov

production farms in the United States.Air quality regulations may be based onactual emissions, atmospheric concentra-tions, or human perception, or via lim-iting the size or location of CAFO. Mea-suring the concentrations or emissionsof most air pollutants is expensive andcomplex. Because of spatial and tempo-ral variability, concentrations and emis-sions must be measured continuouslyover an extended period of time. Be-cause different methods or models cangive different results with the samedata set, a multitude of methods shouldbe used simultaneously to assure emis-sions are reasonable. The “best”method to measure concentrations andemissions will depend upon atmo-spheric concentrations, cost, facilitycharacteristics, objectives, and other fac-tors. In the future, requirements formonitoring of air emissions from CAFOwill probably increase. Reliable process-based models need to be developed sothat emissions of air pollutants can beestimated from readily obtained diet,animal, facility, and environmentalvariables. Auditors will need to betrained in a variety of disciplines includ-ing animal sciences, chemistry, engi-neering, micrometeorology, instrumenta-tion, modeling, and logic.

Key words: air quality, concentratedanimal feeding operations, disper-sion, modeling, regulation

INTRODUCTION

The potential adverse effects of an-imal feeding operations (AFO) onthe environment are a growing con-cern. The effects of concentrated ani-mal feeding operations (CAFO) onwater quality have been regulatedunder the Clean Water Act for manyyears. However, the effects of inten-sive and extensive livestock opera-tions on air quality have receivedless attention at least until recently.Even in rural communities, the gen-eral public has become less tolerantof the odors and dust emitted fromagriculture because of concernsabout health, quality of life, prop-erty values, and the environment.

In general, the air pollutants ofmost concern to livestock operationsinclude ammonia, hydrogen sulfide,particulate matter (PM), volatile or-ganic compounds (VOC), green-house gases (methane, nitrous oxide,and carbon dioxide), and odors orodorants. However, the predomi-nant air quality concerns of live-stock and poultry feeding operationswill vary with the location of the op-eration, the species reared, type ofoperation (enclosed or open-lot),and other factors.

To successfully audit and assess airquality at AFO, it may be necessaryto quantify gaseous and PM emis-

Cole et al.2

sions. Measuring atmospheric emis-sions is difficult and entails 2 majorchallenges: 1) measuring the atmo-spheric concentration, and 2) esti-mating the flux to the atmospherebased on direct measurement or ona flux model that describes or simu-lates the turbulent dispersion ofgases and particulates. Thus develop-ment of process-based models willbe needed to adequately monitor at-mospheric emissions from CAFO.

CURRENT RULES ANDREGULATIONS

The ultimate responsibility for airquality regulations resides with thefederal government. However, stateand local governments can also regu-late pollutants in some cases. In addi-tion, based on real, perceived, or po-tential quality-of-life issues, many airpollutants are “regulated” throughthe court system via injunctions andlaw suits. The permitting process is,in itself, a potential regulatory mech-anism for CAFO. Some pollutants(odor, PM) have short travel dis-tances and affect relatively small geo-graphic areas. Therefore, local or re-gional, rather than federal, regula-tions may be most appropriate. Inother cases, for example ammonia,the ultimate negative effects may beboth local (dry deposition on sensi-tive ecosystems) and longer range(formation of respirable particulatesnear urban areas; wet deposition);thus Federal regulations may bemore appropriate.

A small group of “criteria” pollut-ants {PM [both larger (PM10) andsmaller particulates (PM2.5)], ozone,carbon monoxide, nitrogen dioxide,sulfur dioxide, and lead} are regu-lated under The Clean Air Act of1970 and its amendments (EPA,1987). [The PM10 and PM2.5 are parti-cles having an aerodynamic equiva-lent diameter (AED) of less than 10and 2.5 �, respectively.] The Environ-mental Protection Agency (EPA) es-tablished a set of primary and sec-ondary National Ambient Air Qual-ity Standards (NAAQS) designed to

protect the public against adversehealth effects and to protect the en-vironment. Recent court cases haveestablished that ambient air qualitystandards apply not only to large,heterogeneous air sheds but also atthe property line of an individualsource (EPA, 2007a). As monitoringmethods improve and as the weightof scientific evidence increases, ambi-ent air quality criteria are to be mod-ified to accommodate the latestscience.

Although agriculture has receivedsome exclusions to the NAAQS inthe past, the EPA recently issued afinal rule amending the NAAQS regu-lation of PM10 to include agriculturalsources (EPA, 2004b). In addition,some states have initiated their ownair quality regulations that affectagriculture. For example, Californiais currently developing regulationsfor ozone precursors such as reactiveVOC and for ammonia emitted bylivestock, their manure, or both, andseveral states have adopted regula-tions on ammonia, hydrogen sul-fide, or odors. In general, current airquality regulations that relate toCAFO are based on actual emissionsor atmospheric concentrations ofpollutants (EPA, 1988, 1995). Morecomplex pollutants, such as odors,may be regulated based on humanperception (olfactometry or scent-ometry). In some states, regulationslimiting the size or location ofCAFO, or establishing separation dis-tances from CAFO to businesses orresidences, have been used as a regu-latory mechanism to decrease waterand air pollution from CAFO.

At the time of this writing, ammo-nia and hydrogen sulfide emissionsare not regulated by the EPA. How-ever, it is possible that in the futurethey may be regulated under theComprehensive Environmental Re-sponse, Compensation and LiabilityAct (also called “Superfund”; EPA,2007b) or Emergency Planning andCommunity Right-to-Know Act (alsocalled “Right to Know”; EPA, 2007c).Under these regulations, all facilitiesthat emit more than 45.4 kg (100 lb)

of ammonia or hydrogen sulfide perday must report the release to a Fed-eral National Response Center aswell as to local and state emergencyplanning committees (EPA, 2006).Until recently, it had been generallyassumed these regulations did notapply to agriculture, primarily basedon portions of the regulations thatlimit the response to a release of “anaturally occurring substance in itsunaltered form or altered solelythrough naturally occurring pro-cesses or phenomena from a loca-tion when it is naturally found areexempt” (US Code, 2005). In 2006the EPA concurred with a legal as-sessment by the National Cattle-men’s Association stating that CleanAir Act Title V permitting require-ments and the Comprehensive Envi-ronmental Response, Compensationand Liability Act or Emergency Plan-ning and Community Right-to-Know Act emergency release re-porting requirements do not applyto open-lot cattle operations (Drov-ers Journal, 2006).

In the late 1990s the EPA deter-mined that it did not have adequateair emissions data to determine po-tential air quality regulatory require-ments for AFO. Therefore, the EPA,poultry industry, swine industry,and dairy industry developed the An-imal Feeding Operations Air Compli-ance Agreement (ConsentAgreement or National Air EmissionsMonitoring System; EPA, 2005). Un-der the agreement, the industrieswill fund a 2-yr research project tomeasure emissions of ammonia, hy-drogen sulfide, PM [total suspendedparticles (TSP), PM10, and PM2.5],and VOC at designated productionfarms across the United States to es-tablish scientifically based measuresof these emissions (Federal Register,2005).

The monitoring and regulating ofatmospheric emissions is complex.Because emissions of different pollut-ants have varying spatial and tempo-ral effects, Halberg et al. (2005) rec-ommended that emissions linked toenvironmental effects with a local or

Assessing air quality 3

regional target be area-based (i.e.,emission per hectare), whereas thoselinked to global effects should beproduct-based (e.g., emission per ki-logram milk produced). Similarly,NRC (2003) recommended that airemissions with national or globalscope (ammonia, greenhouse gases)should be evaluated based on emis-sions per unit of production,whereas air emissions of local con-cern (hydrogen sulfide, PM, odor)should be based on the farm and onconcentrations at the farm boundaryor nearest occupied dwelling.

European RegulationsBecause of adverse effects on the

environment, many parts of Europehave established strict regulations onsome air pollutants, most notablyammonia (de Vries, et al., 2001;Oenema, 2004). Within the Euro-pean Union, targets for ammoniaemission have been established thatrequire a 0 to 43% (depending uponthe member state) decrease in ammo-nia emissions between the 1990benchmark and 2010. This has neces-sitated reduction in animal numbers,diet modifications, and modifica-tions in manure handling. Becausethe relationship between agriculturalactivities and their environmentalimpact is complex and difficult tomeasure directly, the EuropeanUnion has developed a series of indi-cators to provide information regard-ing the relationship between an ag-ricultural activity and its impact onthe environment. This is based on aneed to simplify complex phenom-ena and quantify their significance.Policy measures are designed andevaluated on a conceptual frame-work named DPSIR, which standsfor Driving forces (societal, market,and technology causes), Pressures(emissions of pollutants), State ofthe environment or recipient (i.e.,status or quality of the environ-ment), Impacts (effects on health,ecosystems, and agriculture) and Re-sponses (effect of governmental pol-icy on items above; Oenema, 2004;Halberg et al., 2005). Pressures in-

cluded in at least 1 of the some 35models used include emissions ofgreenhouse gases (methane, nitrousoxide), ozone-depleting gases, acidi-fying gases, and nutrifying sub-stances (ammonia; Halberg et al.,2005; EEA, 2006).

ENVIRONMENTAL AUDITS

The EPA (CFR, 2000) defines an en-vironmental audit as a “systematic,documented, periodic, and objectivereview of facility operations andpractices related to meeting environ-mental requirements.” Formal envi-ronmental audits and assessmentsare parts of an organization’s envi-ronmental management and pollu-tion prevention plans. The objec-tives of an audit include the follow-ing: 1) verifying compliance withenvironmental requirements, 2) eval-uating the effectiveness of in-placeenvironmental management sys-tems, and 3) assessing risks from reg-ulated and unregulated materialsand practices. In short, the objec-tives of an audit are to identify prob-lems, analyze the underlyingcause(s), and develop action plans tocorrect those causes. The EPA (2000)suggests that by conducting audits,an operation gains a better under-standing of where its facilities standcompared with specified criteria,such as regulations, managementgoals, or other legal requirements,and provides the organization witha list of what needs to be done tomeet specific goals.

Environmental audits may be assimple as a tour of facilities and re-view of records or as complicated asintensive air sampling and monitor-ing. Audit criteria may be manage-ment practices that benefit the envi-ronment or may be compliance re-quirements such as regulations. Forexample, if the goal is to test forcompliance with regulations, theaudit may provide information onwhether compliance has beenachieved or not, and if not, whatspecific measures are needed to com-ply with regulations. Environmental

audits should be performed by per-sons that possess a working knowl-edge of the regulations and have afamiliarity with the operations andpractices of the facility beingaudited. These 2 basic skills are a pre-requisite for adequately identifyingareas at the facility subject to envi-ronmental regulations and potentialregulatory violations (EPA, 2001).

The US EPA (EPA, 1996b, 1997;2000, 2001), several states (IWRC,2000), and the International Organi-zation for Standardization (ISO,1999, 2001, 2002) have developed re-sources to assist small businesses indesigning audit programs, includingaudit checklists, protocols, and soft-ware. A number of businesses andagencies also provide trainingcourses and certification for environ-mental auditing (BEAC, 1999; EPA,2000), most designed around theISO 14000 standards (ISO, 1999;EPA, 2004a). In 1993 the ISO beganwork on the “ISO 14001 — Stan-dards for Environmental Manage-ment Systems” which was subse-quently supplemented with “ISO19011 — Guidance for Quality and/or Environmental Management Sys-tems Auditing” (ISO, 2002). TheseISO standards are the basic frame-work around which an auditing pro-gram may be developed.

To encourage environmentalaudits, the EPA and several stateshave developed incentive programsthat encourage regulated entities tovoluntarily “police” themselves.Facilities that voluntarily discover,promptly disclose, and expeditiouslycorrect violations of federal environ-mental regulations can obtain elimi-nation of, or substantial reduction ofcivil penalties, criminal prosecution,or both (CFR, 2000). The policy wasdesigned to encourage greater com-pliance with federal laws and regula-tions by promoting a higher stan-dard of self-policing. The potentialcost savings associated with wasteand emission management and lia-bility risks are factors in motivatingindustries to establish proactive envi-ronmental programs that encourageauditing and pollution prevention.

Cole et al.4

Pollution prevention opportunityassessments (PPOA) can be used byenvironmental managers to identifyopportunities to change facility oper-ations in order to save money, in-crease worker safety and morale, anddecrease regulatory liability (EPA,1992). The PPOA can be used as atool for identifying and eliminatingthe underlying causes of complianceproblems. Compliance problems canbe resolved via a combination ofbest management practices (EPA,2007d), management changes, ortechnical modifications.

Accurately defining the objectivesand scope of an environmentalaudit is critical to ensure that theaudit achieves the desired results.The scope of an audit usually de-fines a specific procedure or area ofinvestigation and can be influencedby factors such as facility conditions,cost, staff availability, or other re-source constraints. An initial air qual-ity audit at an AFO might include alisting of air emissions of concern,their source(s), estimates of emis-sions or concentrations within anddownwind of the facility, and poten-tial effects of management practiceson the emissions. It should also iden-tify potential environmental or hu-man health problems and developschedules for remedial actions. Theaudit should include managementaudits, PPOA, and auditing stan-dards (EPA, 1997). Other factors thatshould be evaluated include organi-zational structure, environmentalcommitment, formality of environ-mental programs, internal and exter-nal communication programs, stafftraining and development, programevaluation, reporting and correctiveaction, environmental planning andrisk management, and the environ-mental protection program (EPA,1996b).

The purpose of any pollution pre-vention program is to prevent pollu-tion, not to collect data. Thereforethe simplest auditing system that fitsa facility’s needs is the best. How-ever, because the facility can notmanage what it does not measure,

emission inventories and air emis-sion analyses may both be im-portant components of an initial pol-lution prevention plan audit (EPA,1992). For the purposes of thismanuscript, we assumed that futureauditing and assessment of air qual-ity at AFO will revolve around devel-opment of nutritional and manage-ment practices to comply with fed-eral, state, and local air qualityregulations and to avoid nuisancelaw suits. Preliminary assessmentsand site inspections will be requiredto determine what, if any, regula-tions are broken, to describe the re-lease, to develop remediation prac-tices, and to evaluate the remedia-tion practices (EPA, 2001).

AIR QUALITY COMPONENTSOF LIVESTOCK OPERATIONS

Air quality components from AFOare formed by a complex set of mi-crobial, physical, and chemical pro-cesses that occur within the animaland the manure storage and pro-cessing system. Air quality compo-nents associated with livestock aregenerally more complex than thosefrom industrial sources because ofthe numerous biological processes in-volved. Unfortunately, there is littleinformation on air quality aroundlivestock operations. However, it isimperative that a better understand-ing of the effects of AFO on air qual-ity be developed to fairly regulatethem and to develop potential solu-tions to air quality concerns.

Significant variation occurs amongAFO in the air pollutants of mostconcern. These variations are a resultof differences in animals’ digestivesystems, the diets fed, the produc-tion systems, and manure storageand handling systems. For example,ammonia and odor emissions can bechanged 20 to 50% via modifica-tions of the diet of feedlot cattle(Cole et al., 2005, 2006; Archibequeet al., 2006; Todd et al., 2006), dairycattle (James et al., 1999; Frank etal., 2002), and swine (Colina et al.,

2001; Powers et al., 2006) with onlysmall effects on animal performance.

Animal production facilities alsovary in size, construction, operation,and location. In many cases, for ex-ample, litter in broiler houses, deep-pit storage in swine facilities, andthe pen surface and mounds inopen-lot beef cattle and dairies facili-ties, the animal facilities serve asboth the production unit and ma-nure storage site. Because of differ-ences in bacterial populations andnutrient content, each manure stor-age system produces different airquality components.

AmmoniaAmmonia is formed in manure

storage systems through the fermen-tation of nitrogenous compounds. Anumber of forms of nitrogen are ex-creted by animals, ranging fromcomplete proteins to urea. The ma-jor source in most situations is theurea in urine (or uric acid in poultryexcreta), which is rapidly convertedto ammonia via the bacterial ureaseenzyme in soil and feces. Ammonialoss appears to begin almost immedi-ately after urea is excreted, and itcontinues through manure han-dling, storage, and land application(Arogo et al., 2001). Ammonia emis-sions from retention ponds and la-goons appear to be a continuous pro-cess that occurs primarily from thepond surface (Ni, 1999). Therefore,factors that affect the surface affectammonia emissions.

In its gaseous form, atmosphericammonia can travel from rural to ur-ban areas and neutralize acid gasessuch as sulfates and nitrates (prod-ucts of fossil fuel burning) in the at-mosphere, converting these gasesinto small particulates (PM2.5) thatmay pose a potential health risk tosome individuals (Watson et al.,1998; Neas, 2000; McCubbin et al.,2002). Ammonia that travels down-wind may also be deposited via wetor dry deposition onto the soil orwater bodies and become a nutrientsource. In ecologically sensitive ar-eas, ammonia deposition may pro-

Assessing air quality 5

vide an oversupply of N for the na-tive flora, resulting in modificationsof the native ecosystem (Hutchinsonand Viets, 1969; Wolfe et al., 2003;Todd et al., 2004). From an eco-nomic perspective, ammonia lossesalso represent a loss of potentiallyvaluable N fertilizer.

Atmospheric ammonia concentra-tions at AFO vary greatly and thereappears to be a notable diel patternwith highest concentrations duringthe day and lowest concentrationsat night (Omland, 2002; Todd et al.,2005, 2007). Ammonia concentra-tions in open-lot feedyards rarely ex-ceed 3 ppm (Todd et al., 2005,2007); however, concentrationswithin animal houses can frequentlyexceed 25 ppm, the threshold limitvalue for worker safety in Denmark(Omland, 2002). Ammonia emis-sions from AFO may be affected bymany factors including diet (proteinquantity and degradability, carbohy-drate degradability, acid-base bal-ance), pen surface, retention pond,or lagoon conditions (total ammo-nium concentration, pH, tempera-ture, moisture, solids), weather, ven-tilation rate, manure storagemethod, and animal age (Dewes,1996; Ni, 1999; Ni et al., 1999; Coleet al., 2005, 2006; Todd et al., 2006,2007).

Some current ammonia emissionfactors used by the EPA are in doubtbecause incorrect assumptions weremade and because many values arebased on European data in which dif-ferent managing systems were used(Asman, 1992; Battye et al., 1994)that may not be applicable to Ameri-can production systems. In addition,a single emission factor for ammo-nia and many other pollutants is dif-ficult to justify because so many en-vironmental factors can affectemissions.

Hydrogen SulfideHydrogen sulfide forms in live-

stock operations primarily from an-aerobic fermentation by sulfate-re-ducing bacteria. In solution, sulfideions develop an equilibrium with hy-

drogen ions (Shurson et al., 2000).Under basic conditions (pH > 8),most reduced S exists in solution asHS− and S2− ions, and the quantityof free H2S is small. At pH < 8 theequilibrium shifts rapidly toward for-mation of unionized H2S: this isabout 80% complete at pH 7. Thus,in contrast to ammonia, H2S emis-sions from ponds tend to be greaterat lower pH. Typical pH for swine la-goons and feedlot retention pondsare in the range of 7 to 8.5. Hydro-gen sulfide emissions from lagoonsand retention ponds appear to occurepisodically when sufficient hydro-gen sulfide gas, produced from nutri-ents or sludge on the bottom of thepond, accumulates to overcome thesurface tension of the water and riseto the pond surface. Hydrogen sul-fide emissions from AFO are relatedto diet, surface and pond pH, tem-perature, and biological oxygendemand.

Reported hydrogen sulfide emis-sion rates from swine and dairy ma-nure storage tanks and anaerobic la-goons are highly variable (Parker etal., 2005a) ranging from 146 (Zahnet al., 2001) to 46,260 (Hobbs et al.,1999) �g/m2 per min. Hydrogen sul-fide concentrations downwind offeedyard pens in Nebraska (Koelschet al., 2004) and Texas (Rhoades etal., 2003) ranged from 0.003 to 0.13ppm; however, fewer than 1% ofmeasurements exceeded the state reg-ulated value of 0.1 ppm. Hydrogensulfide emissions from feedyard pensaveraged approximately 3.7 kg per1,000 head daily, and emissionsfrom retention ponds ranged from102 to 1,348 �g/m2 per min (0.54 to11.2 kg per 1,000 head daily;Rhoades et al., 2003).

At elevated concentrations, hydro-gen sulfide is a toxic gas, and thus,is regulated primarily by state andfederal regulations designed forworker safety and based on atmo-spheric concentrations, not totalemissions. Concentrations of 100ppm are considered immediatelydangerous to health (ATSDR, 2004).The Occupational Safety and Health

Administration Permissible ExposureLimit for hydrogen sulfide is 20 ppmfor a 10-min ceiling value (ATSDR,2004).

Odors and Volatile OrganicCompounds (VOC)

Odor is not regulated at the fed-eral level but is often a high-profileissue at the local level where it gener-ally falls under the purview of statenuisance laws (nuisance is generallydefined as any condition that unrea-sonably interferes with another’s en-joyment of life or property). Odor isthe result of human perception ofthe hundreds of gaseous odorants re-leased into the atmosphere.

Most odors from livestock opera-tions are the result of formation ofVOC via anaerobic fermentation oforganic wastes (Mackie et al., 1998).These VOC can be grouped into 5general classes of compounds (in-doles, phenols, acids, cresols, and di-sulfides) that produce a wide rangeof odors, from cleaning materials(phenols) to vinegar (acids) to fecalsmell (indoles, cresols, and disul-fides). When mixed, these com-pounds create odors that are unique.This class of compounds is very elu-sive because each behaves differentlyin the air, reacts with other chemi-cals in the air, and thus may traveldifferent distances. The major driv-ers for odor formation include pH,surplus moisture (which excludes ox-ygen), and warm temperatures. Odoremissions are affected by diet, penconditions, retention pond or la-goon chemistry, and other factors(Sweeten et al., 1983, 1995).

The concentrations of VOC gener-ally decrease with distance from thesource due to dispersion and interac-tion with other compounds in theair. For example, VFA may reactwith ammonia close to the source toform ammonium salts that are lessodiferous than the original VFA.Thus, few VFA are isolated in air fur-ther downwind (i.e., 800 m) of openfeedlots, whereas some phenolic andindolic compounds can be measuredup to several kilometers downwind

Cole et al.6

of the lot (Parker et al., 2007). Someodorous chemicals may also be car-ried by dust particulates. ReactiveVOC have been implicated as precur-sors in ground level ozone forma-tion and thus are coming under in-creased scrutiny in some ozone non-attainment areas such as California(Rabaud et al., 2003).

Particulates and DustParticulates and dust can have ad-

verse effects on visibility, quality oflife, and possibly human and animalhealth (Donham, 1986; MacVean etal., 1986; Barnes, 1994; Guarino etal., 1999; Cole et al., 2000; Neas,2000; McCubbin et al., 2002; Om-land, 2002). Dust particles also cancarry other potential pollutants. Thechemical composition of PM gener-ated from livestock operations hasnot been well characterized, but it isgenerally organic in nature, originat-ing from manure, feed, bedding orlitter, and animal dander. Some inor-ganic dust from roads and pen sur-faces is also present. Dusts from AFOfall into a number of classes basedon their size and characteristics. Un-til 1987 EPA regulated TSP. How-ever, in 1987 the EPA replaced itsTSP standards with ambient stan-dards for PM10 (EPA, 1987). In 2001,EPA added additional new ambientstandards for PM2.5 based on epide-miological studies that suggested aneffect of respirable particulate con-centrations on hospitalization andmorbidity rates in cities (EPA, 1996a;Neas, 2000). Currently the 24-h aver-age NAAQS for ambient PM10 is 150�g/m3 and for ambient PM2.5 is 65�g/m3.

In an effort to more appropriatelyregulate the coarse fraction of parti-cles, the EPA has recently proposedestablishing a “coarse fraction” PM(PM10-2.5) NAAQS (EPA, 2004b),which theoretically refers to themass fraction of particles between2.5 and 10 � AED. The intendedgoal of this proposed standard is toregulate the size fraction of particlesfor which the PM10 standard was ini-tially intended — i.e., inhalable

coarse particles. The proposed con-centration limit for the new PM10-2.5

standard is 75 �g/m3 on a 24-h-aver-age basis.

Typical agricultural emissions suchas those from feed mills and AFOare composed of particles with mass-median diameter (MMD) of about10 to 20 � or greater; moreover, lessthan 5% of particle mass is smallerthan 2.5 � (Sweeten et al., 1988;1998). (Mass median diameter refersto the AED at which 50% by massof the aerosol consists of particleswith AED > MMD and 50% by massconsists of particles with AED <MMD.) The concentrations of PM inopen-lot feedyards vary greatly withhighest dust concentrations at dawnand dusk, when animal activity isgreatest and air movement is themost stable (Sweeten et al., 1988).Similarly, PM emissions in swinefacilities are greatest during the dayand during the summer monthswhen ventilation rates are highest.Dust emissions from feed mills, alesser source, also vary greatly, de-pending on the commodities used,how they are handled and pro-cessed, and how the feeds aremixed.

Dust emissions are affected by theenvironment, pen surface conditions(moisture, manure depth), ventila-tion, time of day, and dietary factors(Van Wicklen and Yoder, 1988; VanWicklen et al., 1988; Phillips andThompson, 1989; Pearson andSharples, 1995; Auvermann, 2006).Emissions from open lots may be de-creased by increasing animal den-sity, collecting manure more fre-quently, sprinkling, and using wind-breaks (Auvermann et al., 2000,Auvermann and Romanillos, 2000),whereas emissions from confine-ment buildings can be decreased viaoil sprays (Pearson and Sharples,1995) or by impaction air damsdownstream of ventilation exhaustfans (Bottcher et al., 1998).

Dust from AFO may also give riseto nuisance complaints, which aretypically regulated at the state or lo-cal level. Visibility may be signifi-

cantly reduced on roadways nearopen-lot AFO during peak PM emis-sion events or when the atmosphereis thermally stable, creating a traffichazard and associated civil liability.Because of potentially confoundingfactors, the effects of PM on the em-ployee health, animal health andperformance, or both are still un-clear (Curtis et al., 1975; Blanc,1999). However, Wyatt et al. (2007)recently identified a cellular mecha-nism by which feedyard and swinedust extracts are thought to induceinflammatory responses in lungcells.

Pathogens and Endotoxins

Airborne pathogens or endotoxinscan potentially affect the health oflivestock, employees, and neighbors(Cole et al., 2000; Omland, 2002).The pathogens in AFO with thegreatest potential to infect humansinclude some species of Escherichiacoli, Salmonella, Listeria, Campylo-bacter, Staphylococcus, Clostridia, andCryptosporidium. The concentrationsand predominant genera of airbornebacteria are affected by housing sys-tem, ventilation, feeding practices,animal species, animal age, and man-agement practices (Chang et al.,2001). Endotoxins are highly resis-tant to radiation and temperatures,and thus are often present in the airof confinement buildings (Eduard,1997a,b; Zhang et al., 1998) andopen feedlots (Purdy et al., 2004).Pathogens and endotoxins may befree in the air or carried on dust par-ticles. Few living gram-negativepathogenic bacteria have been cul-tured in open-lot feedyard air (Purdyet al., 2002; Wilson et al., 2002).This may be due to rapid killing ofgram-negative organisms by radia-tion and desiccation. However,pathogens have been cultured fromair within swine and poultry build-ings (Eduard, 1997a,b; Predicala etal., 2000; Zucker et al., 2000; Changet al., 2001). Although not well stud-ied, it is highly probable that AFO at-mospheres also contain fungal anti-

Assessing air quality 7

gens and mycotoxins (Eduard,1997a,b).

Greenhouse Gases

Carbon dioxide, methane, and ni-trous oxide are the 3 primary green-house gases associated with livestockproduction and manure handling.Greenhouse gases trap long-wave ra-diation near the earth’s surface, creat-ing a “greenhouse” effect thatwarms the atmosphere; thus theypose an environmental impact,rather than a direct health or nui-sance impact. However, there arefew direct measurements of thesegases from livestock facilities to helpdefine emission rates.

Carbon dioxide, methane, and ni-trous oxide gases are formedthrough numerous processes includ-ing fermentation, aerobic and anaer-obic respiration, or enteric methano-genesis. Most manure storage sys-tems use anaerobic storage and thusrelease greenhouse gases (Sharpe andHarper, 1999). Greenhouse gas lossesfrom manure and enteric fermenta-tion are highly variable and highlydependent upon temperature, diet,and management (Johnson andJohnson, 1995; NRC, 2003).

There is little information on ni-trous oxide emissions from AFO,and the data that are available varygreatly (Singurindy et al., 2007).However, research shows that theamount of nitrous oxide generatedand emitted from soil depends onsoil temperature and on the quan-tity of nitrate, carbon, water, and ox-ygen in the soil (Berges and Crutzen,1996). Nitrous oxide emissions fromdairy farms originate about equallyfrom 3 categories: manure manage-ment, feed production, and the rede-posited volatilized N and leached N(Berges and Crutzen, 1996; Brown etal., 2002).

DISPERSION PROCESSES

Emission and dispersion of gaseswithin the atmosphere are con-trolled by the concentration of the

gas, wind speed, type of surface, andatmospheric stability. Dispersionmodels can be used to assess the im-pact of livestock operations on dow-nwind areas and may also be usedto establish setback distances. How-ever, there is no general consensuson the best dispersion model to usefor AFO, and comparisons of differ-ent models have reported large (5-to 200-fold) differences among theavailable models used to establishset-back distances (Piringer andSchauberger, 1999; Chaoui and Brug-ger, 2007).

Gases or particulates in the air aretransported by the wind. As withother items, the faster the wind ve-locity, the greater the ability of theair to transport gases and PM withinthe atmosphere. The velocity of thewind increases with height abovethe earth’s surface. In addition togradients in wind speed, there arealso gradients in air temperaturewith height. Variations in windspeed profiles and air temperaturegradients combine to affect the mix-ing and diffusion of gases and PM inthe air; therefore these 2 forces im-pact how gases and particulates areemitted, dispersed, and transportedin the air (Chen et al., 1998). Thepattern of air movement is affectedby terrain, buildings, obstructions,plants, and other objects. Thus, dis-persion around agricultural areas isnormally complex. As the surfaceroughness increases, the rate of mix-ing increases and the rate at whichmaterials are moved in the air is en-hanced. Typical wind directions andvelocity vary throughout the yearand from specific locations.

CURRENT PROCEDURES TOASSESS, MEASURE, OR

AUDIT AIR POLLUTANTS

A number of factors complicate airmonitoring at CAFO. In some cases,air quality regulations are based onatmospheric concentrations, and inother cases they are based on actualemissions. Because environmental

conditions vary greatly, atmosphericconcentrations and ambient emis-sions are usually not highly corre-lated, especially from open-lotfacilities.

Determination of the concentra-tions of ammonia, hydrogen sulfide,PM, VOC, and odorants from AFOoften requires highly sophisticatedand expensive equipment and con-siderable labor. Because of large spa-tial and temporal variability, concen-tration measurements should betaken over extended periods of timeand include all the annual seasons.For data regarding atmospheric con-centrations and emissions to havemaximum value, the facility, ani-mals, diets, management, andweather should be adequately de-scribed.

There is considerable controversyabout the best method(s) to estimateemissions of different air pollutantsfrom AFO, and these estimates canbe difficult and prone to errors. Onemethod normally used by regulatorsto estimate emissions from a specificlocation is to multiply an emissionfactor (normally presented as thequantity of gas or particulate loadper unit animal per unit time: EPA,1986, 1988, 1995, 1996a) by thenumber of animals at the site. Ingeneral, emission factors are onlycrude averages and do not apply tospecific locations or environmentalconditions (Misselbrook et al., 2000).The National Research Council(NRC, 2003) criticized the emission-factor approach for estimating AFOemissions and recommended a pro-cess-based modeling approach to re-place it so that weather conditions,management practices, and manurehandling technologies could be ex-plicitly acknowledged on a site-spe-cific basis.

Other methods for estimatingemission rates include mass balance,micrometeorology, flux chambers,and models. These methods requireseveral components, including agood technique for accurately mea-suring the concentration of the gasor PM in the atmosphere, measure-

Cole et al.8

ments of atmospheric flow and sta-bility, and thorough documentationof the livestock operation (e.g., num-ber of animals, area, management,diet, age, type, health, and housingtype) and environmental conditions.Preferably, a multitude of methodsshould be used to estimate emissionsand, when feasible, for examplewith ammonia-N, a complete nutri-ent balance should be calculated forthe facility to assure the values arereasonable.

Air quality observations that havebeen made in research studies aregenerally very short in duration andrepresent a specific site at a specifictime, rather than a source area overan extended time period. Somelong-term data that can be used tohelp guide agricultural managementand air quality can be taken fromthe National Atmospheric Deposi-tion Program (NADP, 2007) and theClean Air Status and Trends Network(EPA, 2007e), which use a numberof stations across the United Statesto document wet and dry depositionof nitrates, sulfates, ammonium, andother nutrients (Demerjoian, 2000).

MEASURING ATMOSPHERICCONCENTRATIONS OF

POLLUTANTS

AmmoniaAtmospheric ammonia concentra-

tions in AFO can range from < 1ppm (open-lot feedlots and dairyfarms) to > 100 ppm (poultry andswine houses). A number of activemethods are currently available tomeasure atmospheric ammonia con-centrations including gas washing(Figure 1), annular denuders, open-path or tuned diode lasers (Figure2), Fourier-transformed infra-redspectroscopy (FTIR), ultraviolet dif-ferential optical absorbance spec-trometry (UVDOAS), and chemilu-minescence (Phillips et al., 2000,2001; Mount et al., 2002; Todd etal., 2005, 2006; Marti et al., 2007).A number of passive samplers arealso available (Rabaud et al., 2001;

Figure 1. Impingers with portable pump used to sample air for ammonia downwindof a 50,000 head feedyard. A 2-� Teflon prefilter is used to remove dust particles.The first impinger contains 0.1 M sulfuric acid and the second contains deionizedwater to trap acid fumes to prevent damage to the portable pump. A passive ammoniasampler (Rabaud et al., 2001) is also attached.

Welch et al., 2001; Scholtens et al.,2003). Each method has its own ad-vantages and disadvantages. For ex-ample, compared with many othermethods, gas washing is relativelyinexpensive, but very labor inten-sive. In addition, gas washing, de-nuders, and passive samplers pro-vide average ammonia concentra-tions over an extended samplingperiod, normally 1 to 4 h; whereasFTIR, UVDOAS, lasers, and chemilu-minescence can give near-real-timeconcentrations and almost continu-ous (i.e., 5-s average) readings.Open-path lasers, UVDOAS, andFTIR also have the advantage of pro-viding an average concentrationover an extended sampling path of50 to 500 m. To assure that valuesare accurate, in general, the moremechanized methods and passivesamplers should be calibratedagainst gas washing or standardizedgases or both. In our experience,the ammonia concentration in stan-

dardized gases can differ consider-ably from the designated concentra-tion. Therefore, we believe evenstandardized gases should be cali-brated using gas washing.

Ammonia readily adsorbs tomany surfaces; therefore, any sam-pling lines must be as short as possi-ble to avoid loss of ammonia fromthe sample. Because AFO tend tohave relatively high dust concentra-tions, methods to measure ammo-nia and other gaseous contaminantsmust be either unaffected by thedust, or a method to remove thedust, such as a cyclone or prefilter,must precede the detector. Whenfilters are used, Teflon is preferredover more adsorbent materials; how-ever, tentative data at the USDA-ARS Conservation and ProductionResearch Laboratory (N. A. Cole, un-published data) suggest that the fil-ter material has little, if any, effecton measured ammonia concentra-tions when using gas washing andshort (< 10 cm) sampling lines.

Assessing air quality 9

Figure 2. Open-path ammonia lasers measuring ammonia concentrations over a100-m path at a 50,000 head feedyard. Gas washing bottles are set up at 20-mintervals to calibrate the lasers.

Hydrogen Sulfide

A number of instruments areavailable for measuring hydrogensulfide concentrations. One fre-quently used method is the Jeromemeter (Figure 3; Arizona Instru-ments LLC, Tempe, AZ). The instru-ment uses a gold film sensor to ad-sorb reduced sulfur; then thechange in resistivity of the film sur-face is measured and converted tohydrogen sulfide concentration.The instrument detects reduced Sconcentrations from 2 ppb by vol-ume to 50 ppm by volume. Becausethe instrument measures total re-duced S rather than hydrogen sul-fide specifically, it will also detectother S-bearing compounds such asdimethyl sulfide and methyl mer-captan (Winegar and Schmidt.1998). High concentrations of am-monia, sulfur dioxide, and water va-por do not appear to interfere withJerome meter readings.

A frequently used laboratorymethod to determine atmospheric

hydrogen sulfide is the TEI Model45C hydrogen sulfide monitor(Thermo Electron Corp, Franklin,

Figure 3. A Jerome meter with inlet airsampling port to the top.

MA.). For use in the field it must beplaced inside a protective housing.The instrument catalytically con-verts hydrogen sulfide to sulfur diox-ide and then analyzes the sulfur di-oxide concentration using a pulsedfluorescence analyzer. The instru-ment has a range of about 0.1 to100 ppm by volume. The open-pathUVDOAS (Marti et al., 2007) canalso be used to measure hydrogensulfide.

Particulate MatterAgricultural dusts have much

larger particle sizes than urban or in-dustrial emissions. Because large par-ticles tend to dominate in agricul-tural dusts, the validity of PM10 andPM2.5 concentrations determined atAFO using EPA-approved samplers(known as Federal Reference Meth-ods, FRM) developed for urban airsampling has been questioned (Auv-ermann et al., 2000; Wang et al.,2005; Buser et al., 2007a,b). TheFRM for PM10 and PM2.5 are filter-based, gravimetric samplers thatwere designed for use in urban andindustrial settings in which ambientPM tends to be dominated by fineparticles (i.e., less than 10 � AED).The FRM do not measure particlesizes directly, but rely on particleaerodynamics as well as sampler-spe-cific geometrics, airflow paths, andairflow rates to separate particlesinto 2 size fractions. Theoretically,particles larger than the sampler’s“cutpoint” are captured in an iner-tial pre-separator, and particlessmaller than the cutpoint passthrough the pre-separator and arecollected on a filter (micro quartzfor PM10 and Teflon for PM2.5). Theincrease in the weight of the filtersfrom before to after the samplingperiod is divided by the volume ofair sampled to obtain the PM con-centration. However, PM samplersthat rely on inertial pre-separationdo not perform perfectly. Somelarger particles will penetrate thepre-separator to the filter (over-sam-pling), and the pre-separator willalso collect some of the particles

Cole et al.10

that are smaller than the cutpoint(under-sampling). When FRM sam-plers are deployed in areas in whichambient PM tends to be dominatedby fine particles, the errors associ-ated with over-sampling and under-sampling offset one another, andthe resulting measurements are veryclose to reality. However, when thesamplers are deployed in agricul-tural settings in which the ambientPM is much larger than the sam-pler’s design cutpoint, the effect ofthe over-sampling error is magni-fied and the under-sampling errordoes not compensate for that bias.The result is an upward bias of themeasured concentration that in-creases in magnitude as the differ-ence between sampler cutpoint andambient particle size increases (Auv-ermann et al., 2000; Buser et al.,2007a,b). Because of the over-sam-pling bias of FRM samplers, Buser etal. (2007a) recommended thatwhen used in agricultural settings,FRM monitoring for PM10 or PM2.5

needs to be augmented by concur-rent and collocated TSP monitoringwith ensuing particle size distribu-tion analysis to determine the truefractions of the populations of PMcaught on the TSP sampler filters.

Odors and VOCStandardized methods for mea-

surement of odors from AFO havenot been developed (Hobbs et al.,1995). In addition, measurement ofodorants and VOC is difficult be-cause the concentrations are nor-mally very low (ppb or ppt in air),the compounds may adhere tosome surfaces, and because odor-ants will react with other chemicalsin the air to produce new chemi-cals (Parker et al., 2005a,b, 2007;Miller and Woodbury, 2006).When studying odor and its effectson people living near CAFO, 4 char-acteristics are typically used: 1) fre-quency or how often the odor oc-curs, 2) intensity or concentrationof the odor, 3) duration or howlong the odor is present, and 4) of-fensiveness or hedonic tone

(Sweeten, 1995; Redwine and La-cey, 2000). (Not all authors usethese terms to mean precisely thesame thing, therefore the readershould be cautious in interpretingothers’ data, analysis, and con-clusions.)

Olfactometry using human panel-ists has long been used as amethod to quantify odors, as thehuman nose can often detect odorsbelow the detection levels of cur-rent analytical equipment (CEN,1999; ASTM, 2001; Parker et al.,2007), and continues to be one ofthe primary methods of odor assess-ment for animal feeding operations(Jones et al., 1992; Zhu et al., 1999;Lim et al., 2001; Galvin et al.,2003; Gay et al., 2003; Bicudo etal., 2004). One of the difficultieswith olfactometry is the inherentvariability between odor panelists(Sweeten et al., 1983; Sweeten,1995; Clanton et al., 1999).

Olfactometry can be conductedin the laboratory or in the field,and there are specialized instru-ments for both. In field olfactome-try, the human panelist uses a por-table scentometer for diluting theambient air prior to presentation tothe nose. By opening and closingholes of different sizes, the humanpanelist can control the dilutionsof clean air to odorous air.

For laboratory olfactometry, theair sample must be transportedfrom the field to the laboratory forpresentation to a human panel typi-cally consisting of 4 to 8 people.Samples are collected in the fieldusing a vacuum apparatus in bagsconstructed of relatively nonadsor-bent materials such as Tedlar,Teflon, or Melinex. One difficultywith sampling in plastic bags isthat odorants can adsorb to thesides of the bags and greatly affectthe odor concentration (Koziel etal., 2005; Perschbacher-Buser et al.,2006). The bag is connected to alaboratory olfactometer, which di-lutes the odorous air with clean airscrubbed through an activated car-bon filter. The odor “detection

threshold” (DT) is usually deter-mined using triangular forced-choice olfactometry, where theodor is compared with 2 otherclean air presentations, and thepanelist is asked to pick which oneis different. The DT is a measure ofthe ratio of dilutions of clean air toambient (odorous) air at which50% of the human panelists canjust detect the presence of an unrec-ognized odor (Sweeten, 1995;ASTM, 2001). Detection thresholdis dimensionless and commonly re-ported as odor units. In addition tothe DT, the “recognition thresh-old” (RT) can be used to character-ize odor strength. The RT is theconcentration at which a panelistcan describe the odor. For a givenpanelist, the RT will always have agreater concentration than the DT.There is often a poor correlation be-tween field olfactometry and labora-tory olfactometry (Sheffield et al.,2004), probably because of the in-herent variability in sample collec-tion methods and among panelists.

Because human olfactometryalone does not provide the scien-tific information needed to re-search odor abatement methods,the scientific community is now us-ing measurement technologies,such as gas chromatography-massspectrometry (GC/MS) and proton-transfer-reaction mass spectrometryto quantify chemical compoundspresent in odor samples (Sunessonet al., 2001; Wright et al., 2005a,b;Filipy et al., 2006; Shaw et al.,2007). With GC/MS, VOC presentin ambient air samples are typicallyconcentrated by adsorption usingsolid phase microextraction fibersor sorbent tubes. In the laboratory,the fiber or tube is heated to volatil-ize the VOC into the GC column.The VOC separate based on theirmolecular weight and polarity andare quantified by a detector. Al-though flame ionization detectorsare sometimes used, MS detectorsare often used to better identify thevarious chemicals.

A recent technology improve-ment in odor science is a combina-

Assessing air quality 11

Figure 4. A gas chromatograph-mass spectrophotometer-olfactometer.

tion of the GC/MS with an olfac-tometry port (GC/MS-O; Microana-lytics Inc., Round Rock, TX; Figure4). Between the separation columnand the GC/MS unit, a portion ofthe column exhaust is diverted to asniffing port at which the operatorcontinually observes the relative in-tensity and hedonic tone of theodor associated with each volatilecomponent exiting the columnover time using a touch-screenmonitor. This results in a so-called“aromagram” (Wright et al.,2005a,b; Figure 5).

Microbes and Endotoxins

Bioaerosols of AFO are a complexmixture of live and dead microor-ganisms, their products, and otheraeroallergens. Unfortunately, nostandardized methods have beendeveloped to sample or analyze bi-oaerosols at AFO, making compari-son across experiments or locationsdifficult.

Microorganisms in the atmo-sphere are often sampled using 2-stage or 6-stage Andersen biologicalcascade impactors (Andersen Sam-

pler Inc., Atlanta, GA). Petri dishescontaining the desired selective me-dia are placed in each stage to“trap” viable microorganisms. Eachstage of the Andersen sampler con-tains various size holes designed tomimic portions of the human respi-ratory system in which inhaled par-ticles may deposit (6-stage sampler:stage 1 = nasal passages and si-nuses, stage 2 = pharynx, stage 3 =trachea and primary bronchi, stage4 = secondary bronchi, stage 5 =terminal bronchi, and stage 6 = al-veoli; 2-stage sampler: stage 1 =nonrespirable, stage 2 = respirable).However, the sampling efficiency isaffected by airflow rate (Stewart etal., 1995) and length of samplingtime (Folmsbee et al., 2000). Othermethods used to assess airborneconcentrations of microorganismsinclude all-glass impingers andNuclepore filtration and elution(Thorne et al., 1992; Chang et al.,2001). The optimal samplingmethod appears to be dependentupon the purpose of the sampling,expected bioaerosol concentra-tions, organism of interest, and en-

vironmental conditions (Thorne etal., 1992).

Culture methods have routinelybeen used for measurement of air-borne microorganisms (Eduard,1997a,b; Purdy et al., 2002, 2004,2007). However, nonculturable mi-croorganisms, which can poten-tially cause adverse health effects,can not be measured via culturemethods. In addition, aerosoliza-tion of gram-negative microorgan-isms can affect their culturability(Heidelberg et al., 1997). Thus,plate culturing and counting tech-niques may not provide an accu-rate description of atmospheric bac-terial burdens. In addition, sam-pling times are frequently short (5min or less). More “hardy” organ-isms can be collected using filtersand quantified by nonculturablemethods (Eduard, 1997b). The useof molecular techniques that cangive more accurate estimates of pos-sible exposure to pathogens is in-creasing.

Bacterial endotoxins (the heat-sta-ble lipopolysaccharide in the cellenvelopes of gram-negative bacte-ria) are usually measured by biologi-cal assays based on the reaction ofLimulus amoebocyte lysate with thelipopolysaccharide (Eduard,1997b). Although the precision ofassays performed within individuallabs is good, there can be signifi-cant differences in values reportedfrom different labs on the samesamples (Reynolds et al., 2002).Cole et al. (2000) reported that adifference of up to 17-fold in appar-ent airborne endotoxin concentra-tions could be obtained using differ-ent sampling, storage, extraction,and analysis methods.

Greenhouse GasesMethane is routinely measured

using a gas chromatographequipped with a flame ionizationdetector with tunable diode lasers(Sharpe and Harper, 1999; McGinnet al., 2006), or with FTIR (Amonet al., 2001). Nitrous oxide is nor-mally measured via gas chromatog-

Cole et al.12

Figure 5. An example chromatogram and the corresponding aromagram produced with the gas chromatograph-mass spectrome-ter with an olfactometry port. This sample was taken downwind of a fish meal plant and was used to identify the primaryodorants trimethylamine and dimethylsulfide. Note the small odor peaks associated with the large chromatogram peaks from5.6 to 24.4 min, in contrast to the enlarged image which shows large odor peaks associated with the very small chromatogrampeaks in the first 2.7 min.

raphy using an electron capture de-tector (Berges and Crutzen, 1996),but has also been measured usingtunable diode lasers (Brown et al.,2000, 2002; Kulling et al., 2001) orFTIR (Griffith and Galle, 2000;Amon et al., 2001). These tech-niques have been used for ambientair samples, for grab samples cap-tured in canisters, and for samplesfrom flux chambers. Because rumi-nants may produce a considerablequantity of methane via ruminalfermentation, methane emissions

from cattle facilities need to be par-titioned between the enteric andmanure or lagoon fermentation.

General

In general there is no one “best”method to measure most air pollut-ants. The optimal method will bedetermined by cost, labor availabil-ity, objectives, the environment,the facility characteristics, concen-trations of the pollutant to be mea-sured, and other factors.

MEASURING EMISSIONS OFPOLLUTANTS

Quantifying gaseous emissionsfrom AFO entails 2 major chal-lenges: 1) measuring the concentra-tion of the gas of interest, and 2) ob-taining an estimate of flux from thesurface to the atmosphere based ei-ther on direct measurement or on aflux model that describes or simu-lates the turbulent dispersion ofgases. The method used for measur-

Assessing air quality 13

ing emissions will vary dependingupon the type of emission and hous-ing system (open-lot vs. lagoon;area source vs. ventilated buildingor point source).

Flux Chambers and WindTunnels

Many studies have used fluxchambers or wind tunnels or bothto directly measure flux of gasesfrom ground-level area sources.However, flux chambers also havesignificant limitations, the primaryproblem being their effect(s) on theemitting surface. In agreement witha number of previous studies,Rhoades et al. (2005) and Cole et al.(2007) noted that ammonia emis-sions from a simulated retentionpond and feedlot surface increasedwith increasing air exchange rate.Air turnover rates of approximately15 chamber volumes/min were re-quired to obtain flux rates equiva-lent to undisturbed ammoniasources. Similarly, Rhoades et al.(2003) noted that ammonia and hy-drogen sulfide emissions from feed-yard pen surfaces, estimated usingflux chambers, were as low as 1% ofthose determined using 2 dispersionmodels. Sommer et al. (2004a)noted that emissions of methane,carbon dioxide, and nitrous oxidemeasured from manure stockpilesusing a static chamber were only 12to 22% of flux measured using a mi-crometeorology method. Many fluxchambers routinely used to measureflux have air exchange rates of 0.5chamber volumes/min or less (Kien-busch, 1986). In addition, emissionsof some pollutants, such as ammo-nia, are concentrated in small areas(i.e., urine spots); therefore manymeasurements (> 100) must oftenbe made to account for spatial vari-ability (Cole et al., 2007). Thus,most emission estimates based onflux chambers must be viewed withcaution. It appears that flux cham-bers can be used to effectively parti-tion emissions from different seg-ments of the AFO (manure stock-piles, lagoon, or pen surface)

(Sommer et al., 2004a) and to mea-sure treatment differences (Mei-singer et al., 2001).

Micrometeorological MethodsMicrometeorological methods to

determine gaseous emissions to theatmosphere are advantageous be-cause they do not interfere with theprocesses of emissions and they canintegrate emissions over large areas(McGinn and Janzen, 1998; Fowleret al., 2001; Harper, 2005). Gener-ally speaking, micrometeorologicalmethods rely on concentration mea-surements in the site of interest andcharacterization of the atmospherenear the ground. They have beensuccessfully applied to crops (Den-mead et al., 1978; Harper andSharpe, 1995; Rana and Mastrorilli,1998), open-lot AFO (Hutchinson etal., 1982; McGinn et al., 2003; Toddet al., 2005, 2007; Flesch et al.,2007) and AFO lagoons (Harper etal., 2000; Harper 2005).

The only micrometeorologicalmethod that directly measures tur-bulent transfer is the eddy covari-ance method (Fowler et al., 2001;Meyers and Baldocchi, 2005). Thismethod requires very fast measure-ments (10 to 20/s) of the vertical ve-locity of turbulent eddies and gasconcentration. Sonic anemometersare used to measure the vertical ed-dies and fast-response instrumentssuch as tunable diode lasers areused to simultaneously measure theconcentration of the gas of interestof those eddies. Alternatively, in therelaxed eddy accumulation method,air from up-eddies and down-eddiesis segregated, usually with de-nuders, and after a period of accu-mulation, concentration is mea-sured using ion chromatography orother methods (Ham and Baum,2007).

Mass balance methods accountfor the amount of a gas that passesacross the upwind edge of an emit-ting surface and the amount thatpasses across the downwind edge,so that the difference is the amountemitted. Mass balance methods as-

sume that source strength is homo-geneous, airflow is fully turbulent,and that the boundaries of the sys-tem (i.e., plume dimensions) are de-fined. The surface is considered thelower boundary, and the upperboundary (i.e., top of the plume) isdefined as the height where gas con-centration equals background con-centration. The Integrated Hori-zontal Flux method uses measure-ments of wind speed and gasconcentration profiles to calculatethe horizontal flux (the product ofwind speed and concentration) atvarious heights (Wilson et al., 1983;Wilson and Shum, 1992); integ-rating the horizontal fluxes givesthe vertical flux. Typically, circularplots are used to simplify the deter-mination of upwind source area(Yang et al., 2003; Todd et al.,2006), although the method can beused with strip sources (Denmead etal., 1977), irregularly shaped fields(Flesch et al., 2002; Laubach andKelliher, 2004), or finite volumes(Denmead et al., 1998).

A variant of the mass balancemethod is the box model. Box mod-els have been used to measure am-monia flux from beef and dairy op-erations (Freitas et al., 1997; Ash-baugh et al., 1998) and from swineoperations (McCulloch et al., 1998).Ammonia flux from a 50,000 headbeef cattle feedyard, measured usingthe box model of Ashbaugh et al.(1998) gave emissions similar to theflux-gradient, backward Lagrangianstochastic (bLS; see later text), andN-balance (see later text) methods(N. A. Cole and R. W. Todd, unpub-lished data).

The aerodynamic flux-gradient(FG) method treats turbulent flux asanalogous to molecular diffusion(McGinn and Janzen, 1998; Fowleret al., 2001; Harper, 2005). Flux of agas is the product of the verticalconcentration gradient of the gasand an eddy diffusivity, which var-ies with wind speed, atmosphericstability (Fowler et al., 2001), anddistance from the surface (Thom,1975). The FG method requires pro-

Cole et al.14

Figure 6. A 10-m micrometeorology tower at a 50,000-head feedyard used to deter-mine ammonia emissions using the flux-gradient method. Temperature, wind speed,and ammonia concentrations are determined at heights of 0.5, 1, 2, 4, 6, 8, and 10 m.

file measurements of gas concentra-tion, wind speed, and air tempera-ture (Figure 6). The FG method alsoassumes that there is horizontal uni-formity of airflow, that horizontalconcentration gradients are negligi-ble, and that vertical flux is con-stant with height (Thom, 1975;Harper, 2005). In situations of dis-turbed flow, such as those encoun-tered at AFO, these assumptionsmay be violated and the FG methodcould underestimate flux (Wilson etal., 2001).

Dispersion Models

More complex dispersion modelsare based on a description of the re-lationship between a source of a gasand a downwind receptor or point

(Harper, 2005) using assumptionsabout turbulent flow (Wilson et al.,2001). Sometimes, source strengthof a gas is known and the disper-sion model is used to predict con-centrations at a specified distancedownwind, and in other cases thereverse is true. Gaussian plume mod-els are an example of this type ofdispersion model, in which empiri-cal parameters describe the 3-dimen-sional spread of a plume of gasfrom its source. The bLS model esti-mates flux of a gas by taking con-centration of a gas measured at apoint downwind of an emittingsource and modeling the trajector-ies of thousands of gas particlesbackward to the emitting source(Flesch et al., 1995). Advantages ofthe bLS model include a small num-

ber of required inputs (gas concen-tration, wind speed and direction,atmospheric stability, and definedsource area; Flesch et al., 1995; Lau-bach and Kelliher, 2005; Sommer etal., 2005).

The EPA provides a list of pre-ferred air quality models (CFR,2007). Until recently, the EPA andmost state pollution regulatory agen-cies used the Industrial Source Com-plex Short Term Version 3 model astheir regulatory model. However,this has recently been replaced withthe American Meteorological Soci-ety-EPA Regulatory Model withPlume Rise Model Enhancements(AERMOD-PRIME). Both areGaussian models, the accuracy ofwhich in agricultural situations hasbeen debated (Beychok, 1996). La-grangian stochastic models are of-ten used by researchers as alterna-tives to Gaussian models. Using thesame data set, these models givefeedyard ammonia emission ratesthat differ by as much as 50%(Faulkner et al., 2006). Thus, in gen-eral, flux estimates obtained usingone model should not be used topredict downwind concentrationsusing a different model.

Nutrient BalanceFor some gaseous emissions, such

as ammonia from dry-lot pen sur-faces, it appears that a total N-bal-ance for the AFO (simply N in feedand water minus N retained by ani-mals and in manure) can give rea-sonable estimates of ammonia emis-sions because most of the gaseous Nlosses are as ammonia, rather thanN2O, N2, or NOx (Todd et al., 2005,2007). Using the ratio of N to a non-volatilizable mineral (usually P) inthe diet and in “aged” manure(combination of feces and urine)from the pen surface can also givereasonable estimates of ammonialosses from dry-lots, as long as freshurine spots are not sampled. Harperet al. (2000) reported that a consid-erable portion of the volatile Nlosses from swine lagoons were asN2; thus, a N-balance of N entering

Assessing air quality 15

and leaving lagoons may over-esti-mate ammonia losses.

Flux Measurements fromEnclosed Housing(Point Sources)

A number of differing methodshave been used to estimate emis-sions from enclosed animal houses(Monteny and Erisman, 1998; Doreet al., 2004; Scholtens et al., 2004).In general, emissions are deter-mined by multiplying the measuredpollutant concentrations within thehouse or in the air stream leavingthe building by the volumetric flowrate (Zhu et al., 1999; Xin et al.,2003). Livestock buildings may beventilated in a number of mechani-cal (negative pressure, positive pres-sure, or neutral) and natural ways.Measurement of ventilation rates,and thus flow rates, is difficult andpotentially prone to errors (Bottcheret al., 1996). Ventilation rate is af-fected by a number of factors in-cluding fan performance, weatherconditions, and building environ-ment. An alternative approach is tomeasure the ambient concentra-tions upwind and downwind of thebuilding and back-calculate theemission rate using dispersion mod-els (Flesch et al., 2005).

Tracer MethodsSome studies have used atmo-

spheric tracers to assist in measur-ing gaseous emissions. The tracerused most frequently is probably sul-fur hexafluoride (SF6; Kaharabata etal., 2000; McGinn et al., 2006).Known quantities of the tracer arereleased from the area(s) of interest,upwind of the sampler, to mimicflux of the gas of interest. The con-centration of the gas of interest andthe tracer are determined down-wind of the emission site. The emis-sion of the gas of interest is thencalculated by multiplying the trueemission of the tracer by the ratioof the gas of interest and the tracergas. For accurate determinations,the tracer must disperse in a man-

ner similar to the gas of interest.Tracer methods are usually limitedto use in relatively small source ar-eas (Harper, 2005). Tracers can alsobe used to determining ventilationrates from enclosed buildings.

Statistical, Empirical, andProcess-Based Models

A number of empirical, statistical,and process-based (mechanistic)models have been developed to esti-mate emissions from AFO and ma-nured fields. Most of these modelshave been concerned with ammo-nia emissions, possibly because thechemical and physical factors affect-ing ammonia formation and emis-sion are better understood thanother emissions. In general, statisti-cal models are based on experimen-tal data derived from monitoringemissions at a specific facility; thusthey are often site-specific, may con-tain a high degree of uncertainty,and a potentially large error may oc-cur when applied to other sites. Em-pirical models are frequently de-rived from more controlled experi-ments and are often limited toconditions under which they weredeveloped. Mechanistic models de-scribe emission processes via chemi-cal and physical transformations,transfer, and equilibria within themodeled system.

As a first step in developing mod-els to predict atmospheric emissionsof some pollutants, models describ-ing the effects of diet and manage-ment factors on nutrient excretion(Baldwin et al., 1987; de Boer et al.,2002; Monteny et al., 2002; Fox etal., 2004; Kebreab et al., 2004;Burgos et al., 2005) or enteric fer-mentation (Wilkerson et al., 1995;Rossi et al., 2001; Mills et al., 2003;Garnsworthy, 2004) have been de-veloped. Once the route (urine vs.feces), chemical form (urea, volatilefatty acids, organic, inorganic, etc.),and quantity of nutrient excretionis modeled, it becomes necessary topartition the nutrients to differentstorage or treatment locations (pensurface, manure pit, lagoon, reten-

tion pond, etc.). The chemical andphysical processes that occur duringmanure storage and treatment mustthen be modeled.

Ni (1999) reviewed a number ofmechanistic models developed to es-timate ammonia losses from swinelagoons. Most of these models re-quire only measurements of pH, to-tal ammonia-N, temperature (waterand air) and wind speed to estimateammonia flux. However, Ni (1999)noted that different ammoniumand ammonia dissociation con-stants, Henry’s constant, and masstransfer coefficients were used inthe models. The mechanistic modelof DeVisscher et al. (2002) devel-oped to estimate ammonia emissionfrom swine lagoons also appears topredict ammonia emissions frombeef cattle retention ponds with rea-sonable accuracy; however, a statisti-cal model they developed fromswine lagoon data was very inaccu-rate when used to estimate ammo-nia emissions from beef cattle reten-tion ponds in Texas (N. A. Cole andR.W. Todd, unpublished data). Addi-tional models of ammonia emissionfrom anaerobic lagoons (Liang etal., 2002; Rumburg et al., In press-b), manure pits (Olesen and Som-mer, 1993; Zhang et al., 1994),swine houses (Aarnink and Elzing,1998), dairy houses (Elzing andMonteny, 1997; Rumburg et al., Inpress-a), broiler litter (Carr et al.,1990), compost (Paillat et al., 2005),following field application of ma-nures (Genermont and Cellier,1997; Menzi et al., 1998; Sommerand Olesen, 2000; Guo et al., 2001;Wu et al., 2003), and from wholefarms (Pinder et al., 2004; Zhang etal., 2004) have also been developed.

Several models have been devel-oped to predict odor emissions andodor dispersion from AFO and ma-nured fields, as well as downwindodor concentrations (Janni, 1982;Carney and Dodd, 1989; Smith,1993; Guo et al., 2001; Schulte etal., 2007). Most are empirical andare based primarily on meteorologi-cal conditions or manure handling

Cole et al.16

methods. Henry et al. (2007) devel-oped an empirical model to predictodor emissions from feedyard reten-tion ponds that was based on airtemperature, effluent inflow vol-ume, and days since last inflowevent.

To our knowledge, few modelshave been developed to estimateother AFO emissions such as dust,hydrogen sulfide, and greenhousegases. Auvermann (2003) developeda preliminary mechanistic model topredict fugitive PM emissions frombeef cattle feedyard surfaces, and Ra-zote et al. (2006) reported prelimi-nary values of some parameters inthat model using a benchtop hoof-action simulator. Sommer et al.(2004b) developed a simple modelto predict methane and nitrous ox-ide emissions from livestock ma-nures and the EPA (1990) has usedthe rumen model of Baldwin et al.(1987) to estimate enteric methaneemissions from ruminants.

THE FUTURE

While an awareness of air qualityissues is a good start, it is notenough. It is imperative that every-one associated with livestock pro-duction and its potential impact onair quality begin to address the is-sue. The effects of air pollutionfrom AFO on quality of life, landvalues, and the ability of communi-ties to attract and maintain educa-tional, industrial, and medical facili-ties also need to be determined(Cole et al., 2000). The USDA-Natu-ral Resources Conservation Serviceis in the process of developing anAgricultural Air Quality and Atmo-spheric Change Planning Tool(Johnson et al., 2007) to assist Natu-ral Resources Conservation Serviceemployees and producers in identi-fying important air quality concernsand provide management informa-tion to help mitigate air quality im-pacts and emissions.

Training AuditorsIn the future, requirements for

monitoring of air emissions from

livestock and poultry operationswill vary from state to state and willneed to be adapted for the specifictype of operation. For example, intypical open-lot beef cattle feed-yards, ammonia emissions from re-tention ponds are less than 5% oftotal ammonia emissions (Flesch etal., 2007), whereas in swine opera-tions, ammonia losses from the la-goon may be 30% or more of totalammonia losses (Doorn et al.,2002). Based on the National AirEmissions Monitoring Systemagreement, it is probable that pro-ducers will be responsible for atleast some of the costs of any airquality auditing and monitoringprogram. However, government in-centives, similar to the Environmen-tal Quality Improvement Program,may be developed to help livestockproducers reduce air emissions. Ob-viously, the monitoring of emis-sions, air quality, or both at everyAFO is not practical. Therefore, pro-cess-based or empirical models willneed to be developed so that emis-sions or concentrations of air pollut-ants can be estimated from readilyobtainable variables such as dietcharacteristics, animal characteris-tics, facility design, and environ-mental conditions. When monitor-ing is required, it should be con-ducted over extended periods sothat affected times and places canbe determined. To be accurate andfair, auditors (or preferably teams ofauditors) will need to be trained ina variety of disciplines including an-imal sciences, chemistry, engi-neering, micrometeorology, instru-mentation, mathematical modeling,and logic. If air quality auditingand monitoring becomes requiredin the future, it is highly probablethat consulting firms will developto fill this void.

IMPLICATIONS

Air quality concerns relating toanimal feeding operations will con-tinue to grow in the future. Futurerequirements for auditing and moni-

toring of air emissions from live-stock and poultry operations willvary from state to state and willneed to be adapted for specific oper-ations. Producers will probably beresponsible for at least some of thecosts of any air quality monitoringprogram. However, government in-centives may be developed to pro-vide technical assistance to helplivestock producers reduce air emis-sions. Because the monitoring of airquality at every livestock operationis not practical, process-based mod-els will need to be developed sothat emissions and concentrationsof air pollutants can be easily esti-mated. When monitoring is re-quired, it should be conducted overextended periods of time withproper methods. Teams of auditorswill need to be trained in a varietyof disciplines to assure accurateresults.

LITERATURE CITEDAarnink, A. J. A., and A. Elzing. 1998. Dy-namic model for ammonia volatilization inhousing with partially slatted floors for fat-tening pigs. Livest. Prod. Sci. 53:153.

Amon, B., T. Amon, J. Boxberger, and C.Alt. 2001. Emissions of NH3, N2O and CH4from dairy cows housed in a farmyard ma-nure tying stall (housing, manure storage,manure spreading). Nutrient Cycling Agroe-cosyst. 60:103.

Archibeque, S. L., D. N. Miller, H. C. Freetly,and C. L. Ferrell. 2006. Feeding high-mois-ture corn instead of dry-rolled corn reducesodorous compound production in manureof finishing beef cattle without decreasingperformance. J. Anim. Sci. 84:1767.

Arogo, J., P. W. Westerman, A. J. Heber, W.P. Robarge, and J. J. Classen. 2001. Ammo-nia in animal production — a review. Paper# 01-4089, Am. Soc. Agric. Eng. Int. Meet-ing, Sacramento, CA.

Ashbaugh, L., N. Freitas, T. James, and R.Flocchini. 1998. Ammonia emissions from alarge dairy in California’s San Joaquin Val-ley. In Emission Inventory: Living in aGlobal Environment. Proc. Air Waste Man-age. Assoc. Specialty Conf., New Orleans,LA.

Asman, W. A. H. 1992. Ammonia Emissionsin Europe: Updated Emission and EmissionVariations. Natl. Inst. Public Health and En-viron. Protection. Bilthoven, the Neth-erlands.

ASTM. 2001. E697-91: Standard practice fordetermining odor and taste thresholds by

Assessing air quality 17

force-choice concentration series method oflimits. In Annual Book of ASTM Standards.American Society of Testing and Materials.Philadelphia, PA.

ATSDR. 2004. Public Health Statement: Hy-drogen Sulfide CAS # 7783-06-4. Agency forToxic Substances and Disease Registry.http://www.atsdr.cdc.gov. Accessed, Aug.,2005.

Auvermann, B. W. 2003. A mechanisticmodel of fugitive emissions of particulatematter from cattle feedyards, part I. Intro-ductory evaluation. p. 257 in Proc. 3rd Int.Conf. on Air Pollution from Agric. Opera-tions. Raleigh, NC.

Auvermann, B. W. 2006. Recent develop-ments in air quality from dairies and cattlefeedyards. p. 110 in Proc 21st SouthwestNutr. Manage. Conf., Tempe, AZ.

Auvermann, B. W., C. B. Parnell, B. W.Shaw, and A. Romanillos. 2000. A fresh lookat particle size distribution and the perfor-mance characteristics of ambient particlematter samplers. p. 242 in Proc. 2nd Int.Conf. on Air Pollution from Agricultural Op-erations, Des Moines, Iowa. Am. Soc. Agric.Eng., St. Joseph, MI.

Auvermann, B. W., and A. Romanillos.2000. Effect of stocking density manipula-tion on fugitive PM10 emissions from cattlefeedyards. Proc. Soc. Eng. Agric. Conf., Ade-laide, South Australia.

Baldwin, R. L., J. France, D. E. Beever, M.Gill, and J. H. M. Thornley. 1987. Metabo-lism of the lactating cow. III. Properties ofmechanistic models suitable for evaluationof energetic relationships and factors in-volved in the partition of nutrients. J. DairyRes. 54:133.

Barnes, P. J. 1994. Review article: Air pollu-tion and asthma. Postgrad. Med. J. 70:319.

Battye, R., W. Battye, C. Overcash, and S.Fudge. 1994. Development and selection ofammonia emission factors — Final Report.EPA Contract # 68-D3-0034. U.S. EPA, Of-fice of Research and Development, Washing-ton, DC.

BEAC. 1999. Standards for the ProfessionalPractice of Environmental, Health andSafety Auditing. Board of Environmental,Health, Safety Auditor Certifications, Alta-monte Springs, FL. www.beac.org AccessedJuly, 2007.

Berges, M. G. M., and P. J. Crutzen. 1996. Es-timates of global N2O emissions from cattle,pig and chicken manure, including a discus-sion of CH4 emissions. J. Atmos. Chem.24:241.

Beychok, M. R. 1996. Air-dispersion model-ing and the real world. Environ. Solut.9:24.

Bicudo, J. R., C. J. Clanton, D. R. Schmidt,W. Powers, L. D. Jacobson, and C. L. Teng-man. 2004. Geotextile covers to reduce odorand gas emissions from swine manure stor-age ponds. Appl. Eng. Agric. 20:65.

Blanc, P. D. 1999. Husker days and fevernights. Counting cases of organic dust toxicsyndrome. Chest 116:1157.

Bottcher, R. W., G. R. Baughman, and J. T.Magura. 1996. Field measurement of fanspeed and power use in poultry houses. J.Appl. Poult. Res. 5:56.

Bottcher, R. W., K. M. Keener, G. R. Baugh-man, R. D. Munilla, and K. E. Parbst. 1998.Field and model evaluations of windbreakwalls for modifying emissions from tunnelventilated swine buildings. Paper # 98-4071,Am. Soc. Agric. Eng. Int. Mtg, Orlando, FL.

Brown, H. A., C. Wagner-Riddle, and G. W.Thurtell. 2000. Nitrous oxide flux from soliddairy manure in storage as affected by watercontent and redox potential. J. Environ.Qual. 29:630.

Brown, H. A., C. Wagner-Riddle, and G. W.Thurtell. 2002. Nitrous oxide flux from asolid dairy manure pile measured using amicrometeorological mass balance method.Nutrient Cycling Agroecosyst. 62:53.

Burgos, S. A., P. H. Robinson, J. G. Fadel,and E. J. DePeters. 2005. Ammonia volatil-ization potential: Prediction of urinary ureanitrogen output in lactating dairy cows.Agric. Ecosyst. Environ. 111:261.

Buser, M., C. B. Parnell, B. W. Shaw, and R.E. Lacey. 2007a. Particle matter sampler er-rors due to the interaction of particle sizeand sampler performance characteristics:Ambient PM10 samplers. Trans. Am. Soc.Agric. Biol. Eng. 50:229.

Buser, M., C. B. Parnell, B. W. Shaw, and R.E. Lacey. 2007b. Particle matter sampler er-rors due to the interaction of particle sizeand sampler performance characteristics:Background and theory. Trans. Am. Soc.Agric. Biol. Eng. 50:221.

Carney, P. G., and V. A. Dodd. 1989. A com-parison between predicted and measuredvalues for the dispersion of malodors fromslurry. J. Agric. Eng. Res. 44:67.

Carr, L. E., F. W. Wheaton, and L. W. Doug-lass. 1990. Empirical models to determineammonia concentrations from broilerchicken litter. Trans. ASAE 33:1337.

CEN. 1999. Determination of odor intensityusing dynamic serial dilution olfactometry.Standard CEN/TC264. European Committeefor Standardisation, Brussels, Belgium.

CFR. 2000. Environmental ProtectionAgency Incentives for self-policing: Discov-ery, Disclosure, Correction and Preventionof Violations; Notice. Code of Federal Regu-lations 65:19618. www.epa.gov/compliance/incentives/auditing/auditpolicy.html Ac-cessed July, 2007.

CFR. 2007. Environmental ProtectionAgency Protection of Environment. Code ofFederal Regulations. Title 40. Part 51. Appen-dix W. Guideline on air quality models.http://www.access.gpo.gov/cgi-bin/cfrassemble.cgi?title-200740 Accessed July2007.

Chang, C. W., H. Chung, C. F. Huang, andH. J. J. Su. 2001. Exposure of workers to air-borne microorganism in open-air swinehouses. Appl. Environ. Microbiol. 67:155.

Chaoui, H., and M. Brugger. 2007. Compari-son and sensitivity analysis of setback dis-tance models. Am. Soc. Agric. Biol. Eng.Publ. # 701P0907cd. Proc. Int. Symp. AirQual. Waste Manage. Agric., Broomfield,CO.

Chen, Y. C., D. Bundy, and S. Hoff. 1998.Modeling the variation of wind speed withheight for agricultural source pollution con-trol. ASHRAE Trans. 104:1.

Clanton, C. J., D. R. Schmidt, R. E. Nicolai,P. R. Goodrich, L. D. Jacobson, K. A. Janni,S. Weisberg, and J. A. Buckel. 1999. Dy-namic olfactometry variability in determin-ing odor dilutions-to-threshold. Trans. ASAE42:1103.

Cole, D., L. Todd, and S. Wing. 2000. Con-centrated swine feeding operations and pub-lic health: A review of occupational andcommunity health effects. Environ. HealthPerspect. 108:685.

Cole, N. A., R. N. Clark, R. W. Todd, C. R.Richardson, A. Gueye, L. W. Greene, and K.McBride. 2005. Influence of dietary crudeprotein concentration and source on poten-tial ammonia emissions from beef cattle ma-nure. J. Anim. Sci. 83:722.

Cole, N. A., P. J. Defoor, M. L. Galyean, G.C. Duff, and J. F. Gleghorn. 2006. Effects ofphase-feeding of crude protein on perfor-mance, carcass characteristics, serum urea ni-trogen concentrations, and manure nitro-gen of finishing beef steers. J. Anim. Sci.84:3421.

Cole, N. A., R. W. Todd, D. B. Parker, andM. B. Rhoades. 2007. Challenges in usingflux chambers to measure ammonia emis-sions from simulated open feedlot pen sur-faces and retention ponds. Am. Soc. Agric.Biol. Eng. Publ. # 701P0907cd. Proc. Int.Symp. Air Qual. Waste Manage. Agric.,Broomfield, CO.

Colina, J. J., A. J. Lewis, P. S. Miller, and R.L. Fisher. 2001. Dietary manipulation to re-duce aerial ammonia concentrations in nurs-ery pig facilities. J. Anim. Sci. 79:3096.

Curtis, S. E., C. R. Anderson, J. Simon, A. H.Jensen, D. L. Day, and K. W. Kelley. 1975.Effects of aerial ammonia, hydrogen sulfideand swine in-house dust on rate of gain andrespiratory-tract structure in swine. J. Anim.Sci. 41:735.

de Boer, I. J. M., M. C. J. Smits, H. Mollenh-orst, G. van Duinkerken, and G. J. Mon-teny. 2002. Prediction of ammonia emis-sions from dairy barns using feed character-istics. Part I. Relation between feedcharacteristics and urinary urea concentra-tion. J. Dairy Sci. 85:3382.

Demerjoian, K. L. 2000. A review of na-tional monitoring networks in NorthAmerica. Atmos. Environ. 34:1861.

Denmead, O. T., L. A. Harper, J. R. Freney,D. W. T. Griffith, R. Leuning, and R. R.

Cole et al.18

Sharpe. 1998. A mass balance method fornon-intrusive measurements of surface-airtrace gas exchange. Atmos. Environ.32:3679.

Denmead, O. T., R. Nulsen, and G. W. Thur-tell. 1978. Ammonia exchange over a corncrop. Soil Sci. Soc. Am. J. 42:840.

Denmead, O. T., J. R. Simpson, and J. R.Freney. 1977. A direct field measurement ofammonia emission after injection of anhy-drous ammonia. Soil Sci. Soc. Am. J.41:1001.

De Visscher, A., L. Harper, P. Westerman, Z.Liang, J. Arogo, R. Sharpe, and O. VanCleemput. 2002. Ammonia emissions fromanaerobic swine lagoons: Model develop-ment. J. Appl. Meteorol. 41:426.

de Vries, W., H. Kros, O. Oenema, G. J. Re-insds, and J. W. Erisman. 2001. Animal pro-duction impacts on nitrogen emissions toair and ground water: A Dutch case with aEuropean perspective. Int. Anim. Agric.Food Sci. Conf. http://www/fass.org/fass01/pdfs/dvries.pdf. Assessed May, 2007.

Dewes, T. 1996. Effect of pH, temperature,amount of litter and storage density on am-monia emissions from stable manure. J.Agric. Sci. 127:501.

Donham, K. J. 1986. Hazardous agents in ag-ricultural dust and methods of evaluation.Special Issue: Health Effects of OrganicDusts in the Farm Environ. Am. J. Ind. Med.10:205.

Doorn, M. R. J., D. F. Natschke, S. A. Thor-neloe, and J. Southerland. 2002. Develop-ment of an emission factor for ammoniaemissions from US swine farms based onfield tests and application of a mass balancemethod. Atmos. Environ. 36:5619.

Dore, C. J., B. M. R. Jones, R. Sholtens, J. W.H. Huis in’t Veld, L. R. Burgess, and V. R.Phillips. 2004. Measuring ammonia emis-sion rates from livestock buildings and ma-nure stores — part 2: Comparative demon-strations of three methods on the farm.Atmos. Environ. 38:3017.

Drovers Journal. 2006. EPA sides with NCBAon Clean Air Act. Dec. 2006, pg 27.

Eduard, W. 1997a. Exposure to non-infec-tious microorganism and endotoxins in agri-culture. Ann. Agric. Environ. Med. 4:179.

Eduard, W. 1997b. Measurement methodsand strategies for non-infectious microbialcomponents in bioaerosols at the work-place. Analyst 121:1197.

EEA. 2006. EEA Report 2/2006 – Integrationof environment into EU agriculture pol-icy — the IRENA indicator-based assessmentreport. European Environment Agency, Co-penhagen. http://reports.eea.europa.eu/eea_report_2006_2/en Assessed May, 2007.

Elzing, A., and G. J. Monteny. 1997. Ammo-nia emission in a scale model of a dairy-cowhouse. Trans. ASAE 40:713.

EPA. 1986. AP-42: Compilation of Air Pollu-tion Emission Factors Vol. I: Stationary

Point and Area Sources. Research TrianglePark, NC.

EPA. 1987. National Ambient Air QualityStandards. 40 CFR Part 70. http://www.epa.gov/air/caa Accessed July, 2007.

EPA. 1988. Compilation of Air PollutionEmission Factors — Vol. I. Stationary Pointand Area Sources. Research Triangle Park,NC.

EPA. 1990. Anthropogenic Methane Emis-sions in the United States: Estimates for1990. EPA 430-R-93-003.

EPA. 1992. Facility Pollution PreventionGuide. EPA-600-R-92-088. www.epa.state.oh.us/opp/tanbook/fppgbgn.html AccessedJuly 2007.

EPA. 1995. Interim AP-42: Emission Factorsfor Agricultural Food Industry. Research Tri-angle Park, NC.

EPA. 1996a. Air Quality Criteria for Particu-late Matter. EPA 600-P-95-001aF, Vol. III. p.12-31. Washington, DC.

EPA. 1996b. Generic Protocol for Conduct-ing Environmental Audits of Federal Facili-ties. EPA 300-B-96-102B.

EPA. 1997. Environmental Audit ProgramDesign Guidelines for Federal Agencies. EPA300-B-96-011.

EPA. 2000. The Small Business Source Bookon Environmental Auditing. EPA 233-B-00-003. www.epa.gov/sbo/pdfs/auditbook_500.pdf Accessed July, 2007.

EPA. 2001. Protocol for conducting environ-mental compliance audits under the Emer-gency Planning and Community Right-to-Know Act and CERCLA Section 103. EPA305-B-01-002. www.epa.gov/complinace/in-centives/auditing/ Accessed July, 2007.

EPA. 2004a. ISO14001:2004. EnvironmentalManagement Systems — Requirements withGuidance for Use. Int.Organization for Stan-dardization. www.iso.org/iso/iso_catalogueAccessed July, 2007.

EPA. 2004b. Use of a Performance Based Ap-proach to Determine Data Quality Needs forthe PM-coarse (PMc) Standard. http://www.epa.gov/ttn/amtic/files/ambient/pm25/casac/app3dqo.pdf Assessed May,2007.

EPA. 2005. Animal Feeding Operations Con-sent Agreement and Final Order. http://www.epa.gov/EPA-AIR/2005/July/Day-12/a13672.htm Accessed July, 2007.

EPA. 2006. List of Lists: Consolidated List ofChemicals Subject to the Emergency Plan-ning and Community Right-To-Know Act(EPCRA) and Section 112(r) of the Clean AirAct. EPA 550-B-01-003. http://www.epa.gov/ceppo Accessed July 2007.

EPA. 2007a. Animal Feeding Operations —Compliance and Enforcement. http://www.epa.gov/agriculture/anafocom.html Ac-cessed July, 2007.

EPA. 2007b. Superfund. Cleaning Up the Na-tions Hazardous Wastes Sites. http://

www.epa.gov/superfund/index.htm Ac-cessed July, 2007.

EPA. 2007c. Emergency Planning and Com-munity Right-To-Know Act (EPCRA). http://www.epa.gov/agriculture/lcra.html AccessedOctober, 2007.

EPA. 2007d. Animal Feeding Operations -Best Management Practices (BMPs). http://www.epa.gov/agriculture/anafobmp.htmlAccessed July, 2007.

EPA. 2007e. Clean air status and trends net-work (CASTNET). http://www.epa/castnet.Accessed July, 2007.

Faulkner, W. B., J. J. Powell, J. M. Lange, B.W. Shaw, R. E. Lacey, and C. B. Parnell.2007. Comparison of dispersion models forammonia emissions from a ground levelarea source. Trans ASABE (in review). p 586in Proc. Workshop in Agric. Air Quality:State of the Science, Potomac, MD, June 5–8, 2006.

Federal Register. 2005. Animal feeding opera-tions consent agreement and final order.Fed. Regist. 70:4958.

Filipy, J., B. Rumburg, G. Mount, H. West-bery, and B. Lamb. 2006. Identification andquantification of volatile organic com-pounds from a dairy. Atmos. Environ.40:1480.

Flesch, T. K., J. H. Prueger, and J. L. Hat-field. 2002. Turbulent Schmidt numberfrom a tracer experiment. Agric. For. Mete-orol. 111:299.

Flesch, T. K., J. D. Wilson, L. A. Harper, andB. P. Crenna. 2005. Estimating gas emissionfrom a farm using an inverse-dispersiontechnique. Atmos. Environ. 39:4863.

Flesch, T. K., J. D. Wilson, L. A. Harper, R.W. Todd, and N. A. Cole. 2007. Determin-ing ammonia emissions from a cattle feed-lot with an inverse dispersion technique.Agric. For. Meteorol. 144:139.

Flesch, T. K., J. D. Wilson, and E. Yee. 1995.Backward-time Lagrangian stochastic disper-sion models and their application to esti-mate gaseous emissions. J. Appl. Meteorol.34:1320.

Folmsbee, M., K. Strevett, K. Stafford, andC. Evenson. 2000. The effect of samplingtime on the total efficiency of the Andersenmicrobial sampler: A field study. J. AerosolSci. 31:263.

Fowler, D., M. Coyle, C. Flechard, K. Har-greaves, E. Nemitz, R. Storeton-West, M. Sut-ton, and J. Erisman. 2001. Advances in mi-crometeorological methods for the measure-ment and interpretation of gas and particlenitrogen fluxes. Plant Soil 228:117.

Fox, D. G., L. O. Tedeschi, T. P. Tylutki, J.B. Russell, M. E. Van Amburgh, L. E. Chase,A. N. Pell, and T. R. Overton. 2004. The Cor-nell net carbohydrate and protein systemmodel for evaluating herd nutrition and nu-trient excretion. Anim. Feed Sci. Technol.112:29.

Assessing air quality 19

Frank, B., M. Persson, and G. Gustafsson.2002. Feeding dairy cows for decreased am-monia emission. Livest. Prod. Sci. 76:171.

Freitas, N., T. James, and L. Ashbaugh.1997. Preparation and sampling techniquefor a passive ammonia diffusion sampler. p591 In Ammonia and Odour Control for An-imal Production Facilities. J. A. M. Voer-mans and G. Monteny, ed. Dutch Society ofAgricultural Engineering (NVTL), Rosmalen,The Netherlands.

Galvin, G., K. D. Casey, S. A. Lowe, N. A.Hudson, M. A. Atzeni, and E. J. McGahan.2003. Spatial variability of odor emissionsfrom anaerobic piggery lagoons in Queens-land. In Proc. Air Pollutions from Agricul-tural Operations-III, Proc. 3rd Int. Conf., Re-search Triangle Park, NC. Am. Soc. Agric.Eng. St. Joseph, MI.

Garnsworthy, P. C. 2004. The environmen-tal impact of fertility in dairy cows: A model-ing approach to predict methane and ammo-nia emissions. Anim. Feed Sci. Technol.112:211.

Gay, S. W., D. R. Schidt, C. J. Clanton, K. A.Janni, L. D. Jacobson and S. Weisberg. 2003.Odor, total reduced sulfur, and ammoniaemissions from animal housing facilitiesand manure storage units in Minnesota.Appl. Eng. Agric. 19:347.

Genermont, S., and P. Cellier. 1997. A mech-anistic model for estimating ammonia vola-tilization from slurry applied to bare soil.Agric. For. Meteorol. 88:145.

Griffith, D. W. T., and D. Galle. 2000. Fluxmeasurements of NH3, N2O, and CO2 usingdual beam FTIR spectroscopy and the flux-gradient technique. Atmos. Environ.34:1987.

Guarino, M., A. Caroli, and P. Navarotto.1999. Dust concentration and mortality dis-tribution in an enclosed laying house.Trans. ASAE 42:1127.

Guo, H., L. D. Jacobson, D. R. Schmidt, andR. E. Nicolai. 2001. Calibrating Inpuff-2model by resident-panelists for long-dis-tance odor dispersion from animal produc-tion sites. Appl. Eng. Agric. 17:859.

Halberg, N., H. M. G. van der Werf, C. Bas-set-Mens, R. Dalgaard, and I. J. M. de Boer.2005. Environmental assessment tools forthe evaluation and improvement of Euro-pean livestock production systems. Livest.Prod. Sci. 96:33.

Ham, J. M., and K. A. Baum. 2007. Measur-ing ammonia fluxes from cattle feedlots us-ing time-averaged relaxed eddy accumula-tion. Am. Soc. Agric. Biol. Eng. Publ. #701P0907cd. Proc. Int. Symp. Air Qual.Waste Manage. Agric., Broomfield, CO.

Harper, L. A. 2005. Ammonia: MeasurementIssues. p. 345 in Micrometeorology in Ag-ricultural Systems — Agronomy Monograph47. J. L. Hatfield and J. M Baker, ed. ASA-CSSA-SSSA, Madison, WI.

Harper, L. A., and R. R. Sharpe. 1995. Nitro-gen dynamics in irrigated corn: Soil-plant ni-

trogen and atmospheric ammonia transport.Agron. J. 87:669.

Harper, L. A., R. R. Sharpe, and T. B. Parkin.2000. Gaseous nitrogen emissions from an-aerobic swine lagoons: Ammonia, nitrousoxide and dinitrogen gas. J. Environ. Qual.29:1356.

Heidelberg, J. F., M. Shahamat, M. Levin, I.Rahman, G. Stelma, C. Grim, and R. R. Col-well. 1997. Effect of aerosolization on cultur-ability and viability of gram-negative bacte-ria. Appl. Environ. Microbiol. 63:3585.

Henry, C. G., P. J. Watts, and P. J. Nicholas.2007. A feedlot pond odor emission model.Am. Soc. Agric. Biol. Eng. Publ. #701P0907cd. Proc Int. Symp. Air Qual.Waste Manage. Agric., Broomfield, CO.

Hobbs, P. J., T. H. Misselbrook, and T. R.Cumby. 1999. Production and emissions ofodours and gases from aging pig waste. J.Agric. Eng. Res. 72:291.

Hobbs, P. J., T. H. Misselbrook, and B. F.Pain. 1995. Assessment of odours from live-stock wastes by photoionization detector,and electronic nose, olfactometry and gaschromatography-mass spectrometry. J.Agric. Eng. Res. 60:137.

Hutchinson, G. L., A. R. Mosier, and C. E.Andre. 1982. Ammonia and amine emis-sions from a large cattle feedlot. J. Environ.Qual. 11:288.

Hutchinson, G. L., and F. G. Viets, Jr. 1969.Nitrogen enrichment of surface water by ab-sorption of ammonia volatilized from cattlefeedlots. Science 166:514.

ISO. 1999. ISO 14001 “ Environmental Man-agement Systems. A Complete Implementa-tion Guide. International Organization forStandardization. www.iso.org/iso/iso_catalogue Accessed July, 2007.

ISO. 2001. ISO 14015 — EnvironmentalManagement – Environmental Assessmentof Sites and Organizations. International Or-ganization for Standardization. www.i-so.org/iso/iso_catalogue Accessed July,2007.

ISO. 2002. ISO19011 — Guidance for Qual-ity and/or Environmental Management Sys-tem Auditing. www.iso.org/iso/iso_catalogue. Accessed July, 2007.

IWRC. 2000. Iowa Environmental AuditTraining Manual — CD-ROM. Iowa WasteReduction Center. www.iwrc.org/pubs/man-uals.htm Accessed July, 2007.

James, T., D. Meyer, E. Esparza, E. J. Depe-ters, and H. Perez-Monti. 1999. Effects of di-etary nitrogen manipulation on ammoniavolatilization from manure from Holsteinheifers. J. Dairy Sci. 82:2430.

Janni, K. A. 1982. Modeling dispersion ofodorous gases from agricultural sources.Trans. ASAE 25:1721.

Johnson, K. A., and D. E. Johnson. 1995.Methane emissions from cattle. J. Anim. Sci.73:2483.

Johnson, G., G. Zwicke, S. O’Neill, and R.Vining. 2007. An agricultural air qualityplanning tool for conservation planners.Am. Soc. Agric. Biol. Eng. Publ. #701P0907cd. Proc. Int. Symp. Air Qual.Waste Manage. Agric., Broomfield, CO.

Jones, M., P. J. Watts, S. C. Lott, R. W.Tucker, and R. J. Smith. 1992. A mobile dy-namic olfactometer for feedlot odor studies.Paper # 92-4515. Am. Soc. Agric. Eng., St. Jo-seph, MI.

Kaharabata, S. K., P. H. Schuepp, and R. L.Desjardins. 2000. Estimating methane emis-sions from dairy cattle housed in a barn andfeedlot using an atmospheric tracer. Envi-ron. Sci. Tech. 34:3296.

Kebreab, E., J. A. N. Mills, L. A. Crompton,A. Bannink, J. Dijkstra, W. J. J. Gerrits, andJ. France. 2004. An integrated mathematicalmodel to evaluate nutrient partition indairy cattle between the animal and its envi-ronment. Anim. Feed Sci. Technol.112:131.

Kienbusch, M. R. 1986. Measurement of gas-eous emission rates from land surfaces usingan emission isolation flux chamber: UsersGuide. EPA/600/8-86/008.

Koelsch, R. K., B. L. Woodbury, D. E. Stenb-erg, D. N. Miller, and D. D. Schulte. 2004.Total reduced sulfur concentrations in thevicinity of beef cattle feedlots. Appl. Eng.Agric. 20:77.

Koziel, J. A., J. P. Spinhirne, J. D. Lloyd, D.B. Parker, D. W. Wright, and F. W. Kuhrt.2005. Evaluation of sample recovery of mal-odorous livestock gases from air samplingbags, SPME, Tenax TA sorbent tubes, andsampling canisters. J. Am. Waste Manage.Assoc. 55:1147.

Kulling, D. R., H. Menzi, T. F. Krober, A. Nef-tel, F. Sutter, P. Lischer, and M. Kreuzer.2001. Emissions of ammonia, nitrous oxideand methane from different types of dairymanure during storage as affected by dietaryprotein content. J. Agric. Sci. 137:235.

Laubach, J., and F. M. Kelliher. 2004. Mea-suring methane emission rates of a dairycow herd by two micrometeorological meth-ods. Agric. For. Meteorol. 125:279.

Laubach, J., and F. M. Kelliher. 2005. Mea-suring methane emission rates of a dairycow herd (II): Results from a backward-La-grangian stochastic model. Agric. For. Mete-orol. 129:137.

Liang, Z. S., P. W. Westerman, and J. Arogo.2002. Modeling ammonia emission fromswine anaerobic lagoons. Trans. ASAE45:787.

Lim, T. T., A. J. Heber, J. Q. Ni, A. L. Sutton,and D. T. Kelly. 2001. Characteristics andemission rates of odor from commercialswine nurseries. Trans. ASAE 44:1275.

Mackie, R. I., P. G. Stroot, and V. H. Varel.1998. Biochemical identification and biologi-cal origin of key odor components in live-stock waste. J. Anim. Sci. 76:1331.

Cole et al.20

MacVean, D. W., D. K. Franzen, T. J. Keefe,and B. W. Bennett. 1986. Airborne particleconcentration and meteorological condi-tions associated with pneumonia incidencein feedlot cattle. Am. J. Vet. Res. 47:2676.

Marti, M., E. de Haro, R. E. Sheffield, andM. Chahine. 2007. Emissions from a diarywastewater storage pond, manure pro-cessing area, and composting yard in south-central Idaho. Am. Soc. Agric. Biol. Eng.Publ. # 701P0907cd. Proc. Int. Symp. AirQual. Waste Manage. Agric., Broomfield,CO.

McCubbin, D. R., B. J. Apelberg, S. Roe, andF. Divita. 2002. Livestock ammonia manage-ment and particulate-related health benefits.Environ. Sci. Technol. 36:1141.

McCulloch, R. B., G. S. Few, G. C. Murray,and V. P. Aneja. 1998. Analysis of ammo-nia, ammonium aerosols and acid gases inthe atmosphere at a commercial hog farmin eastern North Carolina, USA. Environ.Pollut. 102(Suppl. 1):263.

McGinn, S. M., T. K. Flesch, L. A. Harper,and K. A. Beauchemin. 2006. An approachfor measuring methane emissions fromwhole farms. J. Environ. Qual. 35:14.

McGinn, S. M., and H. H. Janzen. 1998. Am-monia sources in agriculture and their mea-surement. Can. J. Soil Sci. 78:139.

McGinn, S. M., H. H. Janzen, and T. Coates.2003. Atmospheric ammonia, volatile fattyacids, and other odorants near beef feedlots.J. Environ. Qual. 32:1173.

Meisinger, J. J., A. M. Lefcourt, and R. B.Thompson. 2001. Construction and valida-tion of small mobile wind tunnels for study-ing ammonia volatilization. Appl. Eng.Agric. 17:375.

Menzi, H., P. E. Katz, M. Fahrni, A. Neftel,and R. Frick. 1998. A simple empiricalmodel based on regression analysis to esti-mate ammonia emissions after manure ap-plication. Atmos. Environ. 32:301.

Meyers, T. P., and D. D. Baldocchi. 2005.Current micrometeorological flux methodol-ogies with applications in agriculture. p.381 in Micrometeorology in Agricultural Sys-tems — Agronomy Monograph 47. J. L. Hat-field and J. M Baker, ed. ASA-CSSA-SSSA,Madison, WI.

Miller, D. N., and B. L. Woodbury. 2006. Asolid-phase microextraction chamber meth-ods for analysis of manure volatiles. J. Envi-ron. Qual. 35:2383.

Mills, J. A. N., E. Kebreab, C. M. Yats, L. A.Crompton, S. B. Cammell, M. S. Dhanoa, R.E. Agnew, and J. France. 2003. Alternativeapproaches to predicting methane emissionsfrom dairy cows. J. Anim. Sci. 81:3141.

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. Dem-mers. 2000. Ammonia emission factors forUK agriculture. Atmos. Environ. 34:871.

Monteny, G. J., and J. W. Erisman. 1998.Ammonia emissions from dairy cow build-

ings: A review of measurement techniques,influencing factors and possibilities for re-duction. Neth. J. Agric. Sci. 46:225.

Monteny, G. J., M. C. J. Smits, G. van Du-inkerken, H. Mollenhorst, and I. J. M. deBoer. 2002. Prediction of ammonia emis-sions from dairy barns using feed character-istics. Part II. Relation between urinary ureaconcentration and ammonia emission. J.Dairy Sci. 85:3389.

Mount, G. H., B. Rumburg, J. Havig, B.Lamb, H. Westberg, D. Yonge, K. Johnson,and R. Kincaid. 2002. Measurement of atmo-spheric ammonia at a dairy using differen-tial optical absorption spectroscopy in themid-ultraviolet. Atmos. Environ. 36:1799.

NADP. 2007. National atmospheric deposi-tion program. http://nadp.sws.uiuc.edu. Ac-cessed July, 2007.

Neas, L. M. 2000. Fine particulate matterand cardiovascular disease. Fuel Process.Technol. 65-66:55.

Ni, Q. 1999. Mechanistic models of ammo-nia release from liquid manure: A review. JAgric. Eng. Res. 72:1.

Ni, J. Q., C. Vinckier, J. Coenegrachts, andJ. Hendricks. 1999. Effect of manure on am-monia emission from a fattening pig housewith partly slatted floor. Livest. Prod. Sci.59:25.

NRC. 2003. Air Emissions from Animal Feed-ing Operations: Current Knowledge, FutureNeeds. Natl. Acad. Press, Washington, DC.

Oenema, O. 2004. Governmental policiesand measures regulating nitrogen and phos-phorus from animal manure in Europeanagriculture. J. Anim. Sci. 82(Suppl. E):E196.

Olesen, J. E., and S. G. Sommer. 1993. Mod-elling effects of wind speed and surfacecover on ammonia volatilization fromstored pig slurry. Atmos. Environ.27A:2567.

Omland, O. 2002. Exposure and respiratoryhealth in farming in temperate zones – A re-view of the literature. Ann. Agric. Environ.Med. 9:119.

Paillat, J.-M., P. Robin, M. Hassouna, and P.Leterme. 2005. Predicting ammonia and car-bon dioxide emissions from carbon and ni-trogen biodegradability during animal wastecomposting. Atmos. Environ. 39:6833.

Parker, D. B., Z. Buser, E. Caraway, M.Rhoades, M. Olsen, K. Casey, J. Spears, andG. Green. 2007. VOC concentrations down-wind of beef feedlots during intense odourevents. 14th Int. Union Air Pollution Pre-vention Environ. Protection Assoc. WorldCongress, Brisbane, Australia.

Parker, D., K. Casey, A. Cole, J. Sweeten, B.Auvermann, M. Rhoades, S. Mukhtar, R.Todd, C. Rogers, J. Silva, T. Moss, C. Ponce,R. Pulikanti, and J. Glidden. 2005a. Hydro-gen sulfide emissions from open-lot beef cat-tle feedyard pens and ponds. Presentationto Industry Advisory Committee of theUSDA-CSREES Federal Air Qual. Initiative.

Parker, D. B., M. B. Rhoades, G. L. Schuster,J. A. Koziel, and Z. L. Perschbacher. 2005b.Odor characterization at open-lot beef cattlefeedyards using triangular forced-choice ol-factometry. Trans. ASAE 48:1527.

Pearson, C. C., and T. J. Sharples. 1995. Air-borne dust concentrations in livestock build-ings and the effect of feed. J. Agric. Eng.Res. 60:145.

Perschbacher-Buser, Z. L., D. B. Parker, G.Green, and M. B. Rhoades. 2006. Recoveryof malodorous livestock odors and gasesfrom Tedlar sampling bags using olfactome-try and thermal desorption tubes. p. 979 inProc. Workshop on Agric. Air Qual.: State ofthe Science, Potomac, MD.

Phillips, P. A., and B. K. Thompson. 1989.Respirable dust in fan and naturally venti-lated hog barns. Trans. ASAE 32:1807.

Phillips, V. R., D. S. Lee, R. Scholtens, J. A.Garland, and R. W. Sneath. 2001. A reviewof methods for measuring emission rates ofammonia from livestock buildings andslurry or manure stores, Part 2: Monitoringflux rates, concentrations and airflow rates.J. Agric. Eng. Res. 78:1.

Phillips, V. R., R. Scholtens, D. S. Lee, J. A.Garland, and R. W. Sneath. 2000. A reviewof methods for measuring emission rates ofammonia from livestock buildings andslurry or manure stores, Part 1: Assessmentof basic approaches. J. Agric. Eng. Res.77:355.

Pinder, R. W., N. J. Pekney, C. I. Davidson,and P. J. Adams. 2004. A process basedmodel of ammonia emissions from dairycows: Improved temporal and spatial resolu-tion. Atmos. Environ. 38:1357.

Piringer, M., and G. Schauberger. 1999.Comparison of a Gaussian diffusion modelwith guidelines for calculating the separa-tion distance between livestock farming andresidential areas to avoid odour annoyance.Atmos. Environ. 33:2219.

Powers, W., S. Bastyr, R. Angel, T.Applegate, and B. Kerr. 2006. Effects of re-duced crude protein on gaseous emissionsand swine performance. p. 15 in Proc. Work-shop on Agric. Air Quality. Potomac, MD.North Carolina State Univ., Raleigh.

Predicala, B. Z., R. G. Maghirang, J. E. Ur-ban, K. S. Ocfemia, R. D. Goodband, and M.L. Peterie. 2000. Characterization of particu-late matter in naturally and fan-ventilatedswine barns. Proc. 2nd Int. Conf on Air Pol-lution from Agric. Operations, Des Moines,IA. Am. Soc. Agric. Eng., St Joseph, MI.

Purdy, C. W., R. N. Clark, and D. C. Straus.2007. Analysis of aerosolized particulates offeedyards located in the Southern HighPlains of Texas. Aerosol Sci. Tech. 41:497.

Purdy, C. W., W. C. Rice, D. C. Straus, S. C.Wilson, and D. B. Parker. 2002. The air-borne culturable microbial ecology of sevenfeedyards in the high plains of Texas. p. 26(Abstr.) in Proc. 21st Annu. Conf. Am.Assoc. for Aerosol Res. Charlotte, NC.

Assessing air quality 21

Purdy, C. W., D. C. Straus, D. B. Parker, S.C. Wilson, and R. N. Clark. 2004. Compari-son of the type and number of microorgan-isms and concentration of endotoxin in theair of feedyards in the Southern GreatPlains. Am. J. Vet. Res. 65:45.

Rabaud, N. E., S. E. Ebler, L. L. Ashbaugh,and R. G. Flocchini. 2003. Characterizationand quantification of odorous and non-odor-ous volatile organic compounds near a com-mercial dairy in California. Atmos. Environ.37:933.

Rabaud, N. E., T. A. James, L. L. Ashbaugh,and R. G. Flocchini. 2001. A passive samplerfor the determination of airborne ammoniaconcentrations near large-scale animal facili-ties. Environ. Sci. Tech. 35:1190.

Rana, G., and M. Mastrorilli. 1998. Ammo-nia emissions from fields treated with greenmanure in a Mediterranean climate. Agric.For. Meteorol. 90:265.

Razote, E. B., R. G. Maghirang, B. Z. Predi-cals, J. P. Murphy, B. W. Auvermann, J. P.Harner, III, and W. L. Hargrove. 2006. Labo-ratory evaluation of the dust-emission po-tential of cattle feedlot surfaces. Trans.ASABE 49:1117.

Redwine, J., and R. Lacey. 2000. A summaryof state odor regulations pertaining to con-fined animal feeding operations. p. 33 inProc. Air Pollution from Agricultural Opera-tions. Am. Soc. Agric. Eng., St. Joseph, MI.

Reynolds, S. J., P. S. Thorne, K. J. Donham,E. A. Croteay, K. M. Kelly, D. Lewis, M.Whitmer, D. J. J. Heederik, J. Douwes, I.Connaughton, S. Koch, P. Malmberg, B. M.Larsson, and D. K. Milton. 2002. Compari-son of endotoxin assays using agriculturaldusts. Am. Ind. Hyg. Assoc. J. 63:430.

Rhoades, M. B., D. B. Parker, B. Auvermann,N. A. Cole, and Z. Perschbacher-Buser. 2005.Factors affecting emission measurementswith surface isolation flux chambers. ASAEPaper No. 054026. Proc. Annu. MeetingAmer. Soc. Agric. Eng., Tampa, FL.

Rhoades, M. B., D. B. Parker, and B. Dye.2003. Measurement of hydrogen sulfide inbeef cattle feedlots on the Texas HighPlains. ASAE Paper No. 034108. Proc. Annu.Meeting Am. Soc. Agric. Eng., Las Vegas,NV.

Rossi, F., P. Vecchia, and F. Masoero. 2001.Estimate of methane production from ru-men fermentation. Nutr. Cycling Agroeco-systems 60:89.

Rumburg, B., G. H. Mount, J. Filipy, B.Lamb, H. Westberg, D. Yonge, R. Kincaid,and K. Johnson. Measurements and model-ing of atmospheric flux of ammonia fromdairy cow housing. Atmos. Environ. (Inpress-a)

Rumburg, B., G. H. Mount, D. Yonge, B.Lamb, H. Westberg, M. Neger, J. Filipy, R.Kincaid, and K. Johnson. Measurements andmodeling of atmospheric flux of ammoniafrom an anaerobic dairy waste lagoon.Atmos. Environ. (In press-b)

Scholtens, R., C. J. Dore, B. M. R. Jones, D.S. Lee, and V. R. Phillips. 2004. Measuringammonia emission rates from livestockbuildings and manure stores — part 1: De-velopment and validation of external tracerratio, internal tracer ratio, and passive fluxsampling methods. Atmos. Environ.38:3003.

Scholtens, R., M. J. M. Wagemans, and V. R.Phillips. 2003. Improved passive flux sam-plers for measuring ammonia emissionsfrom animal houses, Part. 2: performance ofdifferent types of sampler as a function ofangle of incidence of airflow. BiosystemsEng. 85:227.

Schulte, D. D., M. R. Modi, C. G. Henry, R.R. Stowell, D. P. Billesback, S. J. Hoff, and L.D. Jacobson. 2007. Modeling odor disper-sion from a swine facility using AERMOD.Am. Soc. Agric. Biol. Eng. Publ. #701P0907cd. Proc. Int. Symp. Air Qual.Waste Manage. Agric., Broomfield, CO.

Sharpe, R. R., and L. A. Harper. 1999. Meth-ane emissions from an anaerobic swine la-goon. Atmos. Environ. 33:3627.

Shaw, S. L., F. M. Mitloehner, W. Jackson,E. J. DePeters, J. G. Fadel, P. H. Robinson, R.Holzinger, and A. H. Goldstein. 2007. Vola-tile organic compound emissions from dairycows and their waste as measured by pro-ton-transfer-reaction mass spectrometry. En-viron. Sci. Technol. 41:1310.

Sheffield, R., M. Thompson, B. Dye, and D.Parker. 2004. Sniff and tell: Study evaluatesfield-based odor assessment devices andmethods for several odor sources. Water En-viron. Fed. 11:8.

Shurson, J., M. Whitney, and R. Nicolai.2000. Nutritional manipulation of swinediets to reduce hydrogen sulfide emissions.http://www.manure.umn.edu/research/sulfide_emissions.html Accessed August,2007.

Singurindy, O., M. Molodovskaya, B. K.Richards, J. Warland, and T. S. Steenhuis.2007. Emission of nitrous oxide from NewYork state dairy farms. Am. Soc. Agric. Biol.Eng. Publ. # 701P0907cd. Proc. Int. Symp.Air Qual. Waste Manage. Agric., Broomfield,CO.

Smith, R. J. 1993. Dispersion of odours fromground level agricultural sources. J. Agric.Eng. Res. 54:187.

Sommer, S. G., S. M. McGinn, and T. K.Flesch. 2005. Simple use of the backwardsLagrangian stochastic dispersion techniquefor measuring ammonia emission fromsmall field-plots. Eur. J. Agron. 23:1.

Sommer, S. G., S. M. McGinn, X. Hao, andF. J. Larney. 2004a. Techniques for measur-ing gas emissions from a composting stock-pile of cattle manure. Atmos. Environ.38:4643.

Sommer, S. G., and J. E. Olesen. 2000. Mod-elling ammonia volatilization from animalslurry with trail hoses to cereals. Atmos. En-viron. 34:2361.

Sommer, S. G., S. O. Petersen, and H. B.Moller. 2004b. Algorithms for calculatingmethane and nitrous oxide emissions frommanure management. Nutr. Cycling Agroe-cosystems 69:143.

Stewart, S. L., S. A. Grinshpun, K. Willeke,S. Terzieva, V. Ulevicius, and J. Donnelly.1995. Effect of impact stress on microbial re-covery on an agar surface. Appl. Environ.Microbiol. 61:1232.

Sunesson, A. L., J. Gullberg, and G. Blom-quist. 2001. Airborne chemical compoundson dairy farms. J. Environ. Monit. 3:210.

Sweeten, J. M. 1995. Odor measurementtechnology and applications: A state-of-the-art review. p. 213 in Proc. 7th Int. Symp.Agric. Food Processing Wastes, Am. Soc.Agric. Eng., St. Joseph, MI.

Sweeten, J. M., T. Kantor, J. D. King, S.Amosson, and W. Rauh. 1995. Feedlot sur-face condition — coal ash surfacing vs. con-trol. Texas Agric. Ext. Serv. Result Demon-stration Handbook for Wheeler County, Col-lege Station, TX.

Sweeten, J. M., C. B. Parnell, R. S. Ether-edge, and D. Osborne. 1988. Dust emissionsin cattle feedlots. In Stress and Disease inCattle. Vet. Clin. North Am. Food Anim.Pract. 4:557.

Sweeten, J. M., C. B. Parnell, B. W. Shaw,and B. W. Auvermann. 1998. Particle sizedistribution of cattle feedlot dust emission.Trans. ASAE 41:1477.

Sweeten, J. M., D. L. Reddell, A. R. McFar-land, R. O. Gauntt, and J. E. Sorel. 1983.Field measurement of ambient odors with abutanol olfactometer. Trans. ASAE 26:1206.

Thom, A. S. 1975. Momentum, mass, andheat exchange of plant communities. p. 57in Vegetation and the Atmosphere. Vol. 1,Principles. J. L. Monteith, ed. AcademicPress, London, UK.

Thorne, P. S., M. S. Kiekhaefer, P. Whitten,and K. J. Donham. 1992. Comparison of bi-oaerosol sampling methods in barns hous-ing swine. Appl. Environ. Microbiol.58:2543.

Todd, R. W., N. A. Cole, and R. N. Clark.2006. Reducing crude protein in beef cattlediets reduces ammonia emissions from arti-ficial feedyard surfaces. J. Environ. Qual.35:404.

Todd, R. W., N. A. Cole, and R. N. Clark.2007. Ammonia emissions from open lotbeef cattle feedyards on the southern HighPlains. Proc. 16th Annu. International Emis-sion Inventory Conference. U.S. Environ-mental Protection Agency, Raleigh, NC.

Todd, R. W., N. A. Cole, L. A. Harper, T. K.Flesch, and B. H. Baek. 2005. Ammonia andgaseous nitrogen emissions from a commer-cial beef cattle feedyard estimated using theflux-gradient method and N:P ratio analysis.Proc. Symp. State of the Science: Animal Ma-nure and Waste Management. San AntonioTX. http://www.cals.ncsu.edu/waste_mgt/natlcenter/center.htm Accessed May, 2006.

Cole et al.22

Todd, R. W., W. Guo, B. A. Stewart, and C.Robinson. 2004. Vegetation, phosphorus,and dust gradients downwind from a cattlefeedyard. J. Range Manage. 57:291.

Van Wicklen, G. L., and M. F. Yoder. 1988.Respirable particle concentrations in natu-rally-ventilated broiler housing. Trans. ASAE31:1794.

Van Wicklen, G. L., M. F. Yoder, and B. D.David. 1988. Respirable aerosol concentra-tions in enclosed laying houses. Trans. ASAE31:546.

Wang, L., M. Buser, C. B. Parnell, B. W.Shaw, R. E. Lacey, and S. Capareda. 2005.Analysis of PSD of the PM in downwindplume and its impact on ambient PM10over-sampling errors for agricultural emis-sions. ASAE Paper No. 054046. Annu. Meet-ing Am. Soc. Agric. Eng., Tampa, FL.

Watson, J. G., E. M. Fujita, J. C. Chow, B.Zielinska, L. W. Richards, W. Neff, and D.Dietrich. 1998. Northern front range airquality study final report. Coop. Inst. forRes. in the Atmosphere, Colorado StateUniv., Fort Collins.

Welch, D. C., T. G. M. Demmers, J. J. Colls,and C. M. Wathes. 2001. Validation of thecollection efficiency of passive ammoniaflux samplers. ASAE Paper No. 701P0201.Proc. 6th Int. Symp. Livest. Environ., Louis-ville, KY.

Wilkerson, V. A., D. P. Casper, and D. R.Mertens. 1995. The prediction of methaneproduction of Holstein cows by several equa-tions. J. Dairy Sci. 78:2402.

Wilson, J. D., V. R. Catchpoole, O. T. Den-mead, and G. W. Thurtell. 1983. Verifica-tion of a simple meteorological method forestimating the rate of gaseous mass transferfrom the ground to the atmosphere. Agric.For. Meteorol. 29:183.

Wilson, J. D., T. K. Flesch, and L. A. Harper.2001. Micro-meteorological methods for esti-mating surface exchange with a disturbedwindflow. Agric. For. Meteorol. 107:207.

Wilson, S. C., J. Morrow-Tesch, D. C. Straus,J. D. Cooley, W. C. Wong, F. M. Mitlohner,and J. J. McGlone. 2002. Airborne microbialflora in a cattle feedlot. Appl. Environ. Mi-crobiol. 68:3238.

Wilson, J. D., and W. K. N. Shum. 1992. Are-examination of the integrated horizontalflux method for estimating volatilisationfrom circular plots. Agric. For. Meteorol.57:281.

Winegar, E. D., and C. E. Schmidt. 1998.Analysis using a Jerome 631-X portable hy-drogen sulfide sensor: Laboratory and fieldevaluation. Rep. to Arizona InstrumentsLLC, Tempe.

Wolfe, A. P., A. C. Van Gorp, and J. S.Baron. 2003. Recent ecological and biogeo-chemical changes in alpine lakes of RockyMountain National Park (Colorado USA): aresponse to anthropogenic nitrogen deposi-tion. Geobiology 1:153.

Wright, D. W., D. K. Eaton, L. T. Nielson, F.W. Kuhrt, J. A. Koziel, and D. B. Parker.2005a. Improved GC-olfactometry based,malodor assessment of swine CAFOs utiliz-ing novel air sampling technologies. PaperNo. 492, Proc. Air Waste Manage. Assoc.Symp., Minneapolis, MN.

Wright, D. W., D. K. Eaton, L. T. Nielsen, F.W. Kuhrt, J. A. Koziel, J. P. Spinhirne, andD. B. Parker. 2005b. Multidimensional gaschromatography-olfactometry for identifica-tion and prioritization of malodors fromconfined animal feeding operations. J.Agric. Food Chem. 53:8663.

Wu, J., D. L. Nofziger, J. G. Warren, and J.A. Hattey. 2003. Modeling ammonia volatil-ization from surface-applied swine effluent.Soil Sci. Soc. Am. J. 67:1.

Wyatt, T. A., R. E. Slager, J. DeVasure, B. W.Auvermann, M. L. Mulhern, S. VonEssen, T.Mathisen, A. A. Floreani, and D. J. Romb-erger. 2007. Feedlot dust stimulation of in-terleukin-6 and 8 requires protein kinase C-

epsilon human bronchial epithelial cells.Am. J. Physiol. Lung Cell. Mol. Physiol.293:L1163.

Xin, H., Y. Liang, A. Tanaka, R. S. Gates, E.F. Wheeler, K. D. Kasey, A. J. Heber, J. Q.Ni, and H. Li. 2003. Ammonia emissionsfrom U.S. poultry houses. Part I — Measure-ment system and techniques. p. 106 inProc. 3rd Int. Conf on Air Pollution fromAgric. Operations. Raleigh, NC.

Yang, Z., H. Niimi, K. Kanda, and Y. Sugab.2003. Measurement of ammonia volatiliza-tion from a field, in upland Japan, spreadwith cattle slurry. Environ. Pollut. 121:463.

Zahn, J. A., J. L. Hatfield, Y. S. Do, and A. A.DiSpirito. 2001. Functional classification ofswine manure management systems basedon effluent and gas emission characteristics.J. Environ. Qual. 30:635.

Zhang, R. H., D. L. Day, L. L. Christianson,and W. P. Jepson. 1994. A computer modelfor predicting ammonia release rates fromswine manure pits. J. Agric. Eng. Res.58:223.

Zhang, R., Z. Wang, J. Fadel, T. Rumsey, J.Arogo, H. Xin, and G. Mansell. 2004. Devel-opment of an improved process based am-monia emission model for agriculturalsources. Prepared for Lake Michigan Air Di-rectors Consortium, Des Plaines, IL.

Zhang, Y., A. Tanaka, J. A. Dosman, A. Sen-thilselvan, E. M. Barber, S. P. Kirychuk, L. E.Holfeld, and T. S. Hurst. 1998. Acute respira-tory response of human subjects to air qual-ity in a swine building. J. Agric. Eng. Res.70:367.

Zhu, J., L. Jacobson, D. Schmidt, and R. Ni-colai. 1999. Daily variations in odor and gasemissions from animal facilities. Appl. Eng.Agric. 16:153.

Zucker, B. A., S. Trojan, and W. Muller.2000. Airborne gram-negative bacterial florain animal houses. J. Vet. Med. B 47:37.