lable at ScienceDirect

Atmospheric Environment 86 (2014) 1e8

Contents lists avai

Atmospheric Environment

journal homepage: www.elsevier .com/locate/atmosenv

The development of seasonal emission factors from a Canadiancommercial laying hen facility

Robert J. Morgan, David J. Wood, Bill J. Van Heyst*

School of Engineering, University of Guelph, 50 Stone Rd. East, Guelph, Ontario N1G 2W1, Canada

h i g h l i g h t s

� A study was conducted at a layer facility in Wellington County, Ontario, Canada.� The average ammonia emission factor was 19.53 � 19.97 g day�1 AU�1.� Ammonia emissions were largely influenced by excreta cleanout times.� PM emissions were heavily influenced by the bird activity level and photoperiod.� The PM2.5/PM10 ratio was determined to be seasonally dependent.

a r t i c l e i n f o

Article history:Received 20 July 2013Received in revised form24 October 2013Accepted 20 December 2013Available online 30 December 2013

Keywords:AmmoniaParticulate matterEmission factorsPoultryLaying henOdor

* Corresponding author. Tel.: þ1 519 824 4120x536E-mail address: bvanheys@uoguelph.ca (B.J. Van H

1352-2310/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.atmosenv.2013.12.033

a b s t r a c t

Pollutants emitted from poultry housing facilities are a concern from a human health, bird welfare, andenvironmental perspective. Development of emission factors for these aerial pollutants is difficult due tovariable climatic conditions, the number and type of poultry, and the wide range of managementpractices used. To address these concerns, a study was conducted to develop emission factors forammonia and particulate matter over a period of one year from a commercial poultry laying hen facilityin Wellington County, Ontario, Canada.

Instruments housed inside an on-site mobile trailer were used to monitor in-house concentrations ofammonia and size fractionated particulate matter via a heated sample line. Along with a ventilationprofile, emission factors were developed for the facility. Average emissions of 19.53 � 19.97, 2.55 � 2.10,and 1.10 � 1.52 g day�1 AU�1 (where AU is defined as an animal unit equivalent to 500 kg live mass) forammonia, PM10, PM2.5, respectively, were observed. All emissions peaked during the winter months, withthe exception of PM2.5 which increased in the summer.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Intensive poultry operations can be a significant source ofharmful atmospheric pollutants such as ammonia (NH3) and size-fractionated particulate matter (PM). Ammonia, from a humanhealth and bird welfare perspective, is of interest due to the factthat it has been identified as a severe respiratory tract irritant(Anderson et al., 1964). Ammonia has also been designated as atoxic substance by Environment Canada (Environment Canada,2012a). PM2.5 (particulate matter with aerodynamic diameter of2.5 mmor less) inhalation has been linked to impaired lung functionand size fractions up to PM10, have been linked with an increasedrisk of mortality from long term exposure (Schwarze et al., 2006).

65.eyst).

All rights reserved.

Ammonia is also a precursor gas for the formation of ammoniumsalt aerosols, which contribute to the PM2.5 size fraction (Lin et al.,2012b).

The agriculture sector accounts for the majority of ammoniaemissions to the atmosphere in Canada, while only contributingrelatively small quantities of size fractionated particulate matter(Environment Canada, 2012b). The National Pollution Release In-ventory (NPRI) (Environment Canada, 2012b) uses constant emis-sion factors to estimate the quantities released from agriculture.Previous studies in broiler facilities (Wathes et al., 1997; GrootKoerkamp et al., 1998; Wheeler et al., 2006; Burns et al., 2007;2008; Gates et al., 2008; Roumeliotis et al., 2010a; b; Lin et al.,2012a), however, have shown that emissions of NH3 and PM areseasonally dependent in a temperate climate. There are limitedcorresponding seasonal studies with laying hens and hence the aimof the current study is to assess the seasonal behavior of NH3 andPM emissions from a commercial laying hen facility.

R.J. Morgan et al. / Atmospheric Environment 86 (2014) 1e82

The majority of laying hen facilities use battery cages althoughperchery and free range housing systems are becoming morecommon (Thiele and Pottguter, 2008). The two prominent manuresystems for caged birds are manure pits (deep pit) and automatedmanure belts. In the high rise deep pit configuration, the batterycages are situated over an open manure storage system while, inhouses with a manure belt system, a large belt is located beneatheach cage to conveymanure to a secondary storage location, usuallyoutside of the house. Currently, in the United States, 70% of layinghen facilities use the high rise with a deep pit system and theremaining 30% use the high rise with a manure belt configuration.Most new laying hen facilities, however, are opting for the manurebelt system (Xin et al., 2011).

Typically, facilities using deep pit systems have poorer airquality and emit more ammonia than facilities using a manure beltmanagement system (Green et al., 2009). The frequencywithwhichmanure belts are used to remove excreta greatly reduces theamount of ammonia emitted from the facility compared to high risesystems with a deep pit (Liang et al., 2005). In non-caged systems,the emission of ammonia and PM are typically higher than facilitiesusing a caged system (Xin et al., 2011).

Several studies have focused on developing emission factors forammonia and/or PM for laying hen facilities (for examples see:Hartung and Phillips, 1994; Phillips et al., 1995; Wathes et al., 1997;Groot Koerkamp et al., 1998; Yang et al., 2000; Keener et al., 2002;Jacobson et al., 2004; Nicholson et al., 2004; Heber et al., 2005;Liang et al., 2005; Lin et al., 2012b; Ni et al., 2012; Li et al., 2013;Wang-Li et al., 2013). Accurate and representative quantificationof atmospheric pollutants from a poultry production facility can bedifficult to obtain. This is due to variable climatic conditionsdepending on geographic location, the number and type of poultry,and the wide range of management practices that are employed.Management practices include feeding, watering and lighting reg-imens, manure removal and storage systems, and ventilation sys-tems. Commercial poultry production can thus be highly variableand studies should be conducted to include a diverse combinationof different poultry species, management practices, and geographiclocations. This will give governments, facility operators, and com-modity groups a scientifically sound interpretation of the behaviorof the pollutants emitted from poultry production facilities.

2. Materials and methods

The commercial laying hen facility used for the study waslocated in Wellington County of Ontario, Canada. The facility wascomprised of two identical barns that housed between 65,000 and70,000 laying hens per barn. Each two story barn had a caging areaof approximately 123 m by 12 m. On each floor, there were fourrows of cages running the length of the barn, and each row hadthree levels of caging. Approximately 8832 cages were located in-side each barn, each cage holding a maximum of 8 birds. Thisallowed for a maximum stocking density of 474 cm2 bird�1.

Periodically, throughout the production cycle, 64 birds wereweighed to obtain an average bird mass. Using the total number ofbirds and the calculated average bird mass, the total mass of thebirds in the housewas estimated. The total number of birds at week1 of the production cycle was 70,600 with an average of 40 mor-talities each week fromvarious causes. During the 34th week of theproduction cycle, 1776 birds were culled.

Mechanical ventilation was used to regulate the temperaturewithin the facility using: fourteen � 0.61 m, four � 0.91 m,twelve � 1.22 m, and six � 1.37 m diameter fans. All 0.61 mdiameter fans were variable speed, with the remainder being singlespeed on/off fans. Depending on the season, the indoor tempera-ture was maintained between 18.9 and 22.2 �C. The ventilation

controller used an eight stage ventilation program that varied theair flow of the facility from a cross ventilated system in the coolermonths to a hybrid tunnel ventilated system during the warmermonths. The maximum ventilation rate for the facility wasapproximately 158 m3 s�1.

The ventilation systemwas based on the difference between theaverage indoor temperature and the set point temperature. Theaverage indoor temperature was recorded using 8 evenly spacedtemperature probes. For the cross ventilation system, fresh air in-lets were located on either side of the facility using 3.05 m lengthbaffles running the full length of the barn. Under the hybrid tunnelventilation configuration, the fresh air inlets used were located atthe south end of the facility.

The lighting regimen began with 13.5 h of light per day startingat 06:00 for new flocks. After 20 weeks had elapsed, 15 min of lightwas added per week to a maximum of 16 h per day, typicallyoccurring by the 30th week of the 53 week production cycle. Feedwas delivered to feed trays by an automatic feeder auger, and waterby nipple drinkers. Birds were fed an industry standard diet forlaying hens. Manure belts were located underneath each level ofcages and were run twice a week on Tuesday and Friday. Manurecleanout times were consistently between 10:30 and 11:30. Floorswere swept by staff to remove dust on an as needed basis.

For this geographic region, data was collected during periodstypifying Fall, Winter, Spring, and Summer. Although the foursampling campaigns do not completely cover each season they aredeemed representative of the time period. The period fromNovember 6th, 2010 to December 15th, 2010 was considered theFall collection season, Winter data was collected from February19th, 2011 toMarch 20th, 2011 and the Spring collection period wasApril 26th, 2011 to May 11th, 2011. During the month of July (2011)the facility was depopulated and a new flock was started. Thisoccurred before the Summer data collection period which spannedAugust 9th, 2011 to September 12th, 2011.

3. Instrumentation

Ammonia concentrations were continuously monitored using achemiluminescent ammonia analyzer (Model 17I, Thermo ElectronCorporation, Franklin, MA, USA). A climate controlled mobile trailerunit was used to house the analyzer and support gases outside thefacility. The calibration was evaluated on a weekly basis with a25 ppm ammonia calibration gas balanced with air. A 5min logginginterval was used with a 10 s time constant. Sample air was drawnfrom the facility to the trailer through a heated sample line at121 �C (Model 0723-100, Clean Air Engineering Inc.) to preventcondensation within the air stream prior to entering the analyzer.

PM concentrations were continuously monitored using twoDustTrak aerosol analyzers (Model 8520, TSI Incorporated, Shore-view, MN, USA). Each analyzer was equipped with an inlet nozzleusing either 10 mm or 2.5 mm cut size. The analyzers were factorycalibrated using a standardized test dust (ISO 12103-1 A1 ArizonaRoad Dust), which differed significantly from that of the facility andhence dust samples were collected to adjust the calibration density.A bulk density test was used to determine the density of the dustspecific to the facility. A correction factor was developed using theratio of the measured density to the calibration density and wasapplied during the data analysis. The aerosol monitors were housedinside the facility collecting samples from the same location as theinlet for the heated sample line used for ammonia. Both PM mon-itors were set to a 5 min logging interval with a 15 s time constant.

A single sampling location was used throughout the study forboth the PM and NH3 monitors, located approximately 85 m fromthe south end of the facility on the west facing side of the building.The sampling location was located horizontally 1.0 m away from a

R.J. Morgan et al. / Atmospheric Environment 86 (2014) 1e8 3

0.61 m variable speed ventilation fan at a height of 2.5 m from theground. The nearest ventilation fan was activated during the firststage of the ventilation program thus ensuring that this fan wouldbe running during all months of the year. Sampling from thislocation was a compromise between recommended practices(Heber and Bogan, 2006) and the need for facility staff to operateuninhibited in the alley way.

A Flow Assessment Numeration System (FANS) (Model G4-54-03, University of Kentucky, Lexington, KY, USA) was used to mea-sure in-situ volumetric flow from the exhaust fans in the facility.The FANS unit operates by vertically traversing an array of sixpropeller anemometers over its cross sectional area. The DC voltagegenerated by the rotation of the propellers is linearly proportionalto the air velocity. Average flow rates were calculated using themeasured velocities across the fan and multiplying them by thecross sectional area of the FANS unit (Gates et al., 2004).

4. Emission factors calculation

Emission factors (EF, g day�1 AU�1) were calculated on a peranimal unit (AU, equivalent to 500 kg of live weight) basis as:

EF ¼hQh

PMwRð273þTÞ

iðCi � CoÞ

iM

��500 kgAU

�

where M is the mass of the birds in the house (kg), Ci is the indoorpollutant concentration (ppm), Co is the outdoor concentration(ppm), P is the barometric pressure (Pa), Mw is the molecularweight of ammonia (g mol�1), R is the universal gas constant(Pa m3 mol�1 K�1), T is the indoor temperature (�C), and Q is thefacility ventilation rate (m3 day�1) (Roumeliotis et al., 2010a).

5. Results and discussion

Table 1 summarizes the length of each sampling period as wellas the number of samples used in each. Table 2 summarizes theaverage seasonal and overall concentrations and emission factorsthat were developed for ammonia, PM10 and PM2.5. For ammonia,the highest concentrations and emission factors were observedduring the winter period. For PM10, the highest concentrations inthe barn were recorded for the fall period whereas the winterperiod had the highest emission factor. PM2.5 behaved quitedifferently in that the highest concentrations and emission factorswere both found in the summer period. Further discussions onthese observations are given below.

6. Seasonal temperatures and ventilation rates

Seasonal temperatures for this study were subject to fairly largefluctuations as illustrated in Fig.1 (panels a, c, e & g) which gives theambient and the indoor barn temperatures for each season. Theindoor temperature, on average, remains relatively constant, wherethe ambient temperature is highly dependent on the time of year.The lowest ambient temperatures were recorded during the Fall

Table 1Summary of Dates and Number of Data Points Included in Each Season.

Season Ammonia PM10

Sampling period Data points Dates included

Fall 11/26/2010e12/15/2010 437 11/12/2010e12/Winter 2/19/2011e3/14/2011 556 2/1/2011e3/2Spring 4/26/2011e5/11/2011 354 3/31/2011e5/2Summer 8/9/2011e9/12/2011 616 6/10/2011e6/3

(Fig. 1a) averaging �4.1 �C. The average ambient temperature forthe Winter was �3.5 �C (Fig. 1c). The Spring had an average tem-perature of 10.7 �C (Fig.1e). The highest outdoor temperatures wererecorded in the Summer averaging 19.4 �C (Fig. 1g).

Ventilation rates were largely a function of the difference be-tween the indoor temperature and the set point temperaturewithin the laying hen house. As the ventilation system drawsoutdoor air into the facility in an attempt to lower the indoor housetemperature, the ambient outside air temperature thus plays adominant role in the magnitude of the deviation from the set pointtemperature. Fig. 1 (panels b, d, f & h) illustrates the ventilationrates that were measured for the facility for the time periods givenabove. Fig. 1b and d indicated nearly identical ventilation rates forthe fall and winter periods. This is due to the very similar trends inoutdoor temperatures experienced for these time periods. As out-door temperatures began to rise during the spring, so did theventilation rates (Fig. 1f). Fig. 1h shows the ventilation rates for thesummer, when the highest ventilation rates were recorded. This isdue to the high temperature of the incoming fresh air combinedwith the heat released by the birds. The combination of these twofactors requires large ventilation rates to maintain appropriate in-door temperatures and also provide convective cooling for the birdswhen the house temperature exceeds the set point.

7. Ammonia concentrations and emission factors

Ammonia concentrations monitored inside the barn were sub-ject to large fluctuations due to a number of factors, primarilyexcreta removal and ventilation. The facility’s ventilation fluctuatedon both a diurnal and seasonal time scale where the ventilationrates were typically lower during the night on a diurnal basis andlower during the Winter on a seasonal basis.

The highest ammonia concentrations were observed during theWinter months, as illustrated by the sixteen day snap shot given inFig. 2a, reaching magnitudes as high as 35 ppm. Ventilation ratesfor the same time period are given in Fig. 2b and the resultingcalculated emission rates, in units of g s�1, are given in Fig. 2c. Dueto the relatively low ventilation rates during the Winter time, theconcentration and emission rate plots (Fig. 2a and c, respectively)follow nearly identical patterns. Also included in Fig. 2 are themanure cleanouts, given as the dashed vertical lines, which typi-cally occurred on Tuesdays and Fridays. Fig. 2 clearly demonstratesthat, as the bird excreta accumulates on themanure belts below thecages, the ammonia concentration rises until the cleanout periodafter which the ammonia concentrations and emission rates dropoff dramatically. The larger peaks for ammonia concentration andemission rate coincide with the Tuesday cleanout periods wherethe excreta has been allowed to accumulate for four days, asopposed to three days for the Friday cleanout.

The seasonal and overall average ammonia concentrations aresummarized in Table 2, in units of mg m�3 and ppm, which showsthat the Fall and Spring seasons had nearly half the averageammonia concentration of the Winter period. In contrast, theSummer period average ammonia concentration, due to the high

PM2.5

Data points Dates included Data points

13/2010 338 11/12/2010e12/13/2010 3382/2011 387 2/1/2011e3/1/2011 4380/2011 278 3/29/2011e5/24/2011 4120/2011 147 6/1/2011e9/16/2011 867

Table 2Emissions Summary of Ammonia, PM2.5, and PM10.

Season Ammonia PM10 PM2.5

Average barnconcentration

Average emission factor Average barnconcentration

Average emission factor Average barnconcentration

Average emission factor

mg m�3 [ppm] g day�1 AU�1 [g day�1 bird�1] mg m�3 g day�1 AU�1 [g day�1 bird�1] mg m�3 g day�1 AU�1 [g day�1 bird�1]

Fall 2.75 � 2.60 [4.39 � 4.13] 12.97 � 15.37 [0.04 � 0.05] 0.50 � 0.31 2.73 � 1.91 [0.0046 � 0.0032] 0.04 � 0.02 0.23 � 0.15 [0.0004 � 0.0003]Winter 5.55 � 4.51 [8.78 � 7.10] 28.89 � 23.13 [0.09 � 0.08] 0.49 � 0.43 2.82 � 2.42 [0.0050 � 0.0043] 0.06 � 0.03 0.30 � 0.19 [0.0005 � 0.0003]Spring 2.69 � 1.95 [4.28 � 3.07] 26.25 � 18.83 [0.09 � 0.07] 0.16 � 0.09 2.23 � 2.11 [0.0040 � 0.0038] 0.04 � 0.03 0.81 � 0.87 [0.0009 � 0.0009]Summer 0.24 � 0.14 [0.40 � 0.22] 8.13 � 7.32 [0.03 � 0.02] 0.14 � 0.08 2.51 � 2.08 [0.0045 � 0.0037] 0.08 � 0.04 2.46 � 2.04 [0.0034 � 0.0028]Overall 3.07 � 3.77 [4.61 � 5.88] 19.53 � 19.97 [0.06 � 0.05] 0.19 � 0.17 2.55 � 2.10 [0.0045 � 0.0037] 0.03 � 0.03 1.10 � 1.52 [0.0013 � 0.0010]

R.J. Morgan et al. / Atmospheric Environment 86 (2014) 1e84

ventilation rate, was less than 5% of the average Winterconcentration.

Fig. 3 illustrates the seasonal behavior of the ammonia emissionfactors (g day�1 AU�1). The emission factors for the Fall, Winter andSpring follow the same general trend with emission factorsincreasing with accumulation of excreta on the belts followed by asharp decrease after the belts are run. The Summer time emissionfactors behave somewhat differently in that their magnitudes arereduced with a much stronger diurnal pattern being observed. Dueto the higher ventilation rates required in the Summer to keep thebirds comfortable, the accumulated excreta forms an air-driedouter crust which lowers the production of ammonia and inhibitsdiffusion of ammonia from deeper within the pile (Xin et al., 2011).The diurnal pattern in the Summer time emission factors isattributed to the large swings in the ventilation from daytime highsto lower rates during the night.

Table 2 also summarizes the seasonal and overall averageammonia emission factors which indicate that the Winter andSpring periods had the highest average ammonia emission factorsfollowed by the Fall season. The lowest emission factors wererecorded during the Summer, due to the high ventilation ratedrying out the excreta and quickly exchanging the indoor air withfresh outdoor air. Fall and Spring periods also had similar ammonialevels in terms of concentration, but the Winter and Spring hadsimilar emission factors. The emission factors for Fall and Springdiffered due to the differences in ventilation rates for those timeperiods, despite their similar concentrations. For the Winter andSpring periods the emission factors were calculated to be verysimilar due to the higher ventilation rate observed during theSpring and the higher concentration during the Winter.

Fig. 1. Seasonal Temperatures (a, c, e & f) and Ventilation

Fall and Spring have similar levels in terms of concentration butthe calculated average emission factor for Fall is less than half of theSpring emission factor. This is primarily due to the ventilation rateduring the Fall being significantly lower than that of the Spring.Ambient temperatures during the Fall were also much lower thanthat observed during the Spring, resulting in the much lowerventilation rates.

A study conducted by Liang et al. (2005) demonstrated the effectthat manure belt cleaning frequency can have on ammonia emis-sion factors. The study was conducted at two separate laying henfacilities, both employing a manure belt system. One facility ranmanure belts on a daily basis and the other twice a week. Runningthe belts on a daily basis resulted in an overall emission factor of17.5 g day�1 AU, whereas running twice a week gave an overallemission factor of 30.8 g day�1 AU. The resulting overall emissionfactor for ammonia from this study is presented in Table 2. In thecurrent study, running themanure belts twice aweek resulted in anoverall emission factor that is slightly higher than that for the dailymanure removal but not as high as the facility running belts twice aweek in Liang et al. (2005).

8. Particulate matter concentration and emission factors

PM10 and PM2.5 concentrations in the facility varied on both adiurnal and seasonal time scale. This was primarily due to theseasonal fluctuations in ventilation rates and the daily duration ofexposure to light (photoperiod), which had a pronounced effect onthe bird activity level.

The highest concentrations of PM10 were observed during theWinter months, as illustrated by the six day snap shot given in

Rates (b, d, f & h) for the four sampling campaigns.

Fig. 2. Observed winter trends in (a) ammonia concentration (b) facility ventilationrate (c) ammonia emission rate where the dashed lines indicate manure cleanouts.

R.J. Morgan et al. / Atmospheric Environment 86 (2014) 1e8 5

Fig. 4a, reaching as high as 1.6 mg m�3. The ventilation rates forthe same duration are illustrated in Fig. 4b and the resultingemission rates in Fig. 4c. Due to the relatively low ventilationrates that were required during the Winter months, the con-centration and emission rate plots for PM10 follow very similarpatterns.

Also included in Fig. 4aec are the periods when the lights are offas indicated by the shaded bars. The corresponding increases inPM10 concentrations as well as the ventilation rates during thelights on periods demonstrate the large effect that the photoperiodand resulting bird activity have on PM10 emissions. During thelights on period, the bird’s activity level was high which results in agreater amount of food, dander, skin, feather, and dried excreta

Fig. 3. Ammonia emission factors for (a) Fall (November/December) (b) Winter

particles being released as well as re-entraining settled particlesthrough agitation by the motion of the birds. During the dark pe-riods, the bird activity level is typically very low which reduces thesources of PM and allows the already entrained PM10 to either beexhausted by the fans or settle out, thus greatly reducing theconcentration.

Table 2 presents a summary of seasonal and overall averagePM10 concentrations (mg m�3). Fall and Winter average concen-trations for PM10 are nearly identical, as are the average concen-trations for Spring and Summer. In comparison, the average Spring/Summer concentrations were approximately 30% of the magnitudeexperienced during the Fall/Winter, due to the much higherventilation rate in the Spring/Summer.

Fig. 5 illustrates the seasonal behavior of the calculated PM10emission factors (g day�1 AU�1). There is a strong diurnal signatureduring the Fall and Winter (Fig. 5a and b, respectively) due to lowventilation rates. However, during the Spring and Summer (Fig. 5cand d, respectively) the diurnal pattern is less pronounced due tothe higher ventilation rates required to maintain the indoor tem-perature within the bird’s comfort zone.

A summary of seasonal emission factors for PM10 are also pre-sented in Table 2. There is less than 20% variability in the calculatedaverage PM10 emission factors between the seasons. This is mainlydue to the fact that during the seasons with the lowest concen-trations, the ventilation rates were the highest and, during theseasons with the highest concentrations, the ventilation rates werethe lowest.

Similar to PM10, the Winter concentrations for PM2.5 are illus-trated in Fig. 3d using the same concentration scale as that used forPM10 to allow for a direct comparison of magnitudes. Unlike PM10,the PM2.5 concentrations did not display as strong a diurnal patternconnected to the timing of the photoperiods. This suggests thatPM2.5 is less dependent on bird activity and that the drop in con-centration during dark periods may not be as large because of theslower settling velocity of the smaller particles. This may also beevidence of another generation mechanism such as the formationof secondary aerosols, which contribute to the fine PM fraction(Roumeliotis et al., 2010a).

The ventilation rate for the same time period is presented inFig. 4e and the resulting emission rates are illustrated in Fig. 4f.Although the ventilation rate displayed a fairly strong diurnalpattern, the resulting PM2.5 emission rate did not.

The average seasonal concentrations and overall average con-centrations for PM2.5 are presented in Table 2. The observed con-centrations (mg m�3) of PM2.5 were highest during the Summer,followed byWinter, and lowest during the Fall and Spring. The highconcentration that was observed in the Winter season is attributedto the higher particle retention time due to the low ventilationrates, similar to that observed in PM10. The high concentrations

(February/March) (c) Spring (April/May) (d) Summer (August/September).

Fig. 4. Observed Winter trends in PM concentration (a & d), facility ventilation (b & e), PM emission rate (c & f) where shaded times represents periods when lights were off.

R.J. Morgan et al. / Atmospheric Environment 86 (2014) 1e86

during the Summer, however, is contrary to that observed for PM10and suggests that alternative PM2.5 formation mechanisms may bepresent that may be attributed to the warm ambient and in-housetemperatures. Higher temperatures promote the formation ofsecondary aerosols, which could contribute to the observed

Fig. 5. PM10 emission factors for (a) Fall (November/December) (b) Winter (F

increase in fine PM during the Summer. A study by Roumeliotiset al. (2010a) on a commercial broiler facility found that therewas a greater concentration of acidic gases in the barn air duringthe Summer that could react with ammonia to form fine PM, thuslending support for this hypothesis.

ebruary/March) (c) Spring (April/May) (d) Summer (August/September).

Fig. 6. PM2.5 emission factors for (a) Fall (November/December) (b) Winter (February/March) (c) Spring (April/May) (d) Summer (August/September).

R.J. Morgan et al. / Atmospheric Environment 86 (2014) 1e8 7

Fig. 6 gives the emission factors that were determined for thefour seasons for PM2.5. The emission factors for PM2.5 seem to behighly affected by outdoor temperature. In other words, during thecolder periods associated with the Fall and Winter, the emissionfactors are very low. With increasing temperatures during theSpring and Summer a marked increase in PM2.5 emission factorswas observed. A summary of seasonal emission factors for PM2.5 isalso presented in Table 2.

9. PM2.5/PM10 comparison

Fig. 7 depicts the ratio of PM2.5 to PM10 throughout an averageday for the four seasons. For each season, it can be observed that theratio increases at night, when bird activity is low. This is due to thelarger particles settling out, reducing the PM10 concentration.During the Summer, the ratio of PM2.5 to PM10 was much higher,thus supporting that an alternate generation mechanism for PM2.5is present that is connected to the warmer ambient conditionsduring that season.

The average ratios of PM2.5 to PM10 for Fall, Winter, Spring, andSummer are 0.13 � 0.07, 0.15 � 0.08, 0.25 � 0.08, and 0.55 � 0.19,respectively. Each of the seasonal average ratios were statisticallyand significantly different from the other seasons (p ¼ <0.001),

Fig. 7. Ratios of PM2.5 to PM10 during Fall, Winter, Spring, and Summer.

with the exception of the Fall andWinter (p¼ 0.0952), based on a t-test using the Satterthwaite’s approximation.

10. Conclusion

The average overall emission factor for ammonia was19.53 � 19.97 g day�1 AU�1, and was largely influenced by excretacleanout times. A drastic decline in indoor concentration ofammonia was observed after the manure belts were run, removingall excreta from the facility. Emissions were observed to be thehighest during the Winter months, due to the ammonia buildupinside the facility from low ventilation rates. Ammonia emissionswere shown to be lowest during the Summer months when theincrease in ventilation decreased the retention time of the indoorair and increased the drying and crusting of the excreta, therebydecreasing the microbial activity in the manure.

The average overall emission factors for PM10 and PM2.5 were2.55� 2.10 g day�1 AU�1 and 1.10� 1.52 g day�1 AU�1, respectively.PM emissions were heavily influenced by the bird activity level andphotoperiod. PM2.5 emissions varied seasonally, with high emis-sions in the Winter months due to low ventilation as expected butwith higher emissions during the Summer when both the venti-lation and concentrationwithin the facility were elevated. The ratioof PM2.5 to PM10 was seasonally dependent with the lowest in theWinter with an average ratio of 0.13 � 0.07 and highest during thesummer at 0.55 � 0.19. During the lights off periods, there was alsoan increase in the PM2.5 to PM10 ratio. This leads to the conclusionthat PM2.5 is less dependent on bird activity than PM10, and sug-gests another generation mechanism, such as the formation ofsecondary aerosols.

Acknowledgments

This study was assisted by funds provided from the Poultry In-dustry Council (PIC), Ontario Ministry of Agriculture, Food andRural Affairs (OMAFRA), Canadian Poultry Research Council, Na-tional Science and Engineering Research Council (NSERC), and theUniversity of Guelph. The laying hen facility was made available byLen Jewitt and the staff of BLT Farms Inc.

References

Anderson, D.P., Beard, C.W., Hanson, R.P., 1964. Adverse effects of ammonia onchickens including resistance to infection with Newcastle disease virus. AvianDis. 8, 369e378.

Burns, R.T., Xin, H., Gates, R.S., Li, H., Hoff, S.J., Moody, L.B., Overhults, D.G.,Earnest, J.W., 2007. Tyson Broiler Ammonia Emission Monitoring Project. Finalreport. Iowa State University, Ames, Iowa.

Burns, R.T., Li, H., Moody, L.B., Xin, H., Gates, R.S., Overhults, D.G., Earnest, J.W., 2008.Quantification of Particulate Emissions from Broiler Houses in the SoutheasternUnited States. ASABE Paper No. 701P0408. ASABE, St. Joseph, Mich.

R.J. Morgan et al. / Atmospheric Environment 86 (2014) 1e88

Environment Canada, 2012a. Toxic Substances List e Schedule 1. Available at:http://www.ec.gc.ca/lcpe-cepa/default.asp?lang¼En&n¼0DA2924D-1&wsdoc¼4ABEFFC8-5BEC-B57A-F4BF-11069545E434 (accessed 08.04.13.).

Environment Canada, 2012b. National Pollutant Release Inventory: Air PollutantEmissions Summaries and Trends. Available at: http://www.ec.gc.ca/pdb/websol/emissions/ap/ap_result_e.cfm?year¼2011&substance¼nh3&location¼CA&sector¼&submit¼Search (accessed 08.04.13.).

Gates, R.S., Casey, K.D., Xin, H., Wheeler, E.F., Simmons, J.D., 2004. Fan AssessmentNumeration System (FANS) design and calibration specifications. Trans. ASAE47 (5), 1709e1715.

Gates, R.S., Casey, K.D., Wheeler, E.F., Xin, H., Pescatore, A.J., 2008. U.S. broilerhousing ammonia emissions inventory. Atmos. Environ. 42 (14), 3342e3350.

Green, A.R., Wesley, L., Trampel, D.W., Xin, H., 2009. Air quality and bird health inthree types of commercial egg layer houses. J. Appl. Poult. Res. 18, 605e621.

Groot Koerkamp, P.W.G., Metz, J.H.M., Uenk, G.H., Phillips, V.R., Holden, M.R.,Sneath, R.W., Short, J.L., White, R.P., Hartung, J., Seedorf, J., Schröder, M.,Linkert, K.H., Pederson, S., Takai, H., Johnsen, J.O., Wathes, C.M., 1998. Concen-trations and emissions of ammonia in livestock buildings in northern Europe.J. Agric. Eng. Res. 70, 79e95.

Hartung, J., Phillips, V.R., 1994. Control of gaseous emissions from livestock buildingand manure stores. J. Agric. Eng. Res. 57, 173e189.

Heber, A.J., Bogan, B.W., 2006. The study and regulation of agricultural air quality inthe U.S.. In: Proceeding from the Workshop on Agricultural Air Quality: State ofthe Science, pp. 32e35. MD, USA.

Heber, A.J., Ni, J.-Q., Lim, T.T., Chervil, R., Tao, P.C., Jacobson, L.D., Hoff, S.J.,Koziel, J.A., Zhang, Y., Beasley, D.B., Sweeten, J.M., 2005. Aerial pollutantemissions from two high-rise layer barns in Indiana. Paper #04e1368. In:Presented at the 2005 ASAE Annual International Meeting, Minneapolis, MN.ASAE, St. Joseph, MI.

Jacobson, L.D., Hetchler, B.P., Johnson, V.J., Nicolai, R.E., Schmidt, D.R., Goodrch, P.R.,Heber, A.J., Ni, J.-Q., Lim, T.T., Tao, P.C., Hoff, S.J., Bundy, D.S., Huebner, M.A.,Zelle, B.C., Zhang, Y., McClure, J., Roberts, M., Koziel, J.A., Baek, B.H., Balota, A.,Spinhirne, J.P., Sweeten, J.M., Beasley, D.B., Baughman, G.R., Munilla, R., 2004.Preliminary NH3, H2S, and PM10 Data from Pig and Poultry Building for Six-StateProject. ASAE Paper #044156. American Society of Agricultural and BiologicalEngineers, St. Joseph, MI.

Keener, H.M., Elwell, D.L., Grande, D., 2002. NH3 emissions and N-balances for a 1.6million caged layer facility: manure belt/composting vs. deep pit operation.Trans. ASAE 45, 1977e1984.

Li, Q.-F., Wang-Li, L., Wang, K., Chai, L., Cortus, E.L., Kilic, I., Bogan, B.W., Ni, J.-Q.,Heber, A.J., 2013. The National air emissions monitoring study’s southeast layersite: part II. Particulate matter. Trans. ASABE 56 (3), 1173e1184.

Liang, Y., Xin, H., Wheeler, E.F., Gates, R.S., Li, H., Zajaczkowski, P., Topper, A.,Casey, K.D., Behrends, B.R., Burnham, Zajaczkowski, F.J., 2005. Ammonia

Emissions from U.S. Laying Hen Houses in Iowa and Pennsylvania, vol. 48(5).American Society of Agricultural and Biological Engineers, pp. 1927e1941.

Lin, X.J., Cortus, E.L., Zhang, R., Jiang, S., Heber, A.J., 2012a. Air emissions from broilerhouses in California. Trans. ASABE 55 (5), 1895e1908.

Lin, X.J., Cortus, E.L., Zhang, R., Jiang, S., Heber, A.J., 2012b. Ammonia, hydrogensulfide, carbon dioxide and particulate matter emissions from California high-rise layer houses. Atmos. Environ. 46, 81e91.

Ni, J.-Q., Chai, L., Chen, L., Bogan, B.W., Wang, K., Cortus, E.L., Heber, A.J., Lim, T.-T.,Diehl, C.A., 2012. Characteristics of ammonia, hydrogen sulfide, carbon dioxide,and particulate matter concentrations in high-rise and manure-belt layer henhouses. Atmos. Environ. 57, 165e174.

Nicholson, F.A., Chambers, B.J., Walker, A.W., 2004. Ammonia emissions from broilerlitter and laying hen manure management systems. Biosyst. Eng. 89, 175e185.

Phillips, V.R., Holden, M.R., White, R.P., Sneath, R.W., Demmers, T.G.M.,Wathes, C.M., 1995. Measuring and reducing gaseous and particulate air pol-lutants from UK livestock buildings. In: Proc. 7th Int. Symp. Agric. Food Process.Wastes, Chicago, IL. ASAE, St. Joseph, MI, pp. 241e251.

Roumeliotis, T.S., Dixon, B.J., Van Heyst, B.J., 2010a. Characterization of gaseouspollutant and particulate matter emission rates from a commercial broileroperation part I: observed trends in emissions. Atmos. Environ. 44, 3770e3777.

Roumeliotis, T.S., Dixon, B.J., Van Heyst, B.J., 2010b. Characterization of gaseouspollutant and particulate matter emission rates from a commercial broileroperation part II: correlated emission rates. Atmos. Environ. 44, 3778e3786.

Schwarze, P.E., Ovrevik, J., Lag, M., Refsnes, M., Nafstad, P., Hetland, R.B., Dybing, E.,2006. Particulate matter properties and health effects: consistency of epide-miological and toxicological studies. Hum. Exp. Toxicol. 25, 559e579.

Thiele, H., Pottguter, R., 2008. Management recommendations for laying hens indeep litter, perchery, and free range systems. Lohmann Inf. 43, 53e63.

Wang-Li, L., Li, Q.-F., Chai, L., Cortus, E.L., Wang, K., Kilic, I., Bogan, B.W., Ni, J.-Q.,Heber, A.J., 2013. The national air emissions monitoring study’s southeast layersite: part III. Ammonia concentrations and emissions. Trans. ASABE 56 (3),1185e1197.

Wathes, C.M., Holden, M.R., Sneath, R.W., White, R.P., Phillips, V.R., 1997. Concen-trations and emission rates of aerial ammonia, nitrous oxide, methane, carbondioxide, dust and endotoxin in UK broiler and layer houses. Br. Poult. Sci. 38,14e28.

Wheeler, E.F., Casey, K.D., Gates, R.S., Xin, H., Zajaczkowski, J.L., Topper, P.A., Liang, Y.,Pescatore, A.J., 2006. Ammonia emissions from twelve U.S. broiler chickenhouses. Trans. ASAE 49, 1495e1512.

Xin, H., Gates, R.S., Green, A.R., Mitloehner, F.M., Moore, P.A., Wathes, C.M., 2011.Environmental impacts and sustainability of egg production systems. Poult. Sci.90, 263e277.

Yang, P., Lorimor, J.C., Xin, H., 2000. Nitrogen losses from laying hen manure incommercial high-rise layer facilities. Trans. ASAE 43, 1771e1780.