For presentation at the Air _ Waste Management Associations 90th .doc

Document Sample
For presentation at the Air _ Waste Management Associations 90th .doc Powered By Docstoc
					Assessing Annual U.S. Broiler Chicken House Emissions
Paper #1292

Richard S. Gates, Kenneth D. Casey, Anthony J. Pescatore
University of Kentucky, 128 CE Barnhart Bldg. Lexington, KY 40546

Eileen F. Wheeler
The Pennsylvania State University, Agricultural Engineering Bldg. University Park, PA 16801

Hongwei Xin
Iowa State University, 100 Davidson Hall, Ames, IA 50011

Despite availability of recently acquired ammonia emissions baseline data for U.S. broiler
housing, the issue of estimating their contribution to an annual budget should be resolved. Broiler
operations are uniquely different than egg-producing layer operations, because the birds are
grown from day-old to market weight. Consumer demands and market forces determine actual
market weight which varies from “Cornish hens” weighing 1 kg to “roasters” weighing 4 kg.
Emission rate increases in a linear relationship with flock age from near zero at the start of the
flock to a maximum at the end, 28 to 56 days later. On a typical farm, 5 to 8 flocks are grown
each year depending on finished bird weight and market demand. The houses are empty for
seven to fourteen days between flocks while cleaning and maintenance in preparation for the next
flock is carried out. In addition, the houses may be empty for a further fourteen to twenty-one
days to allow for annual maintenance and litter removal. We present a model that takes into
account: broiler market weight, numbers slaughtered and ammonia emissions to compute annual
emissions estimates. The method can be readily applied and should provide for a more accurate
annual budget estimate than is currently available.

Meat birds raised for human consumption include chickens, turkeys and ducks. Poultry
consumption continues to grow in the USA, and worldwide. Per capita consumption of chicken
meat was 53.2 lbs/person in the U.S. in 2000. US poultry industries produced in excess of 22
million metric tons of broiler meat valued at over $13.4 billion in 20031. Poultry raised for meat
are generically referred to as “broilers” and these may be raised to different mature weights
depending on market forces such as consumer preferences, feed and fuel costs, and seasonal
demands. Broiler chickens are classified by mature weight (Table 1), with most numbers raised
being roasters, heavy broilers, or broilers. For purposes of this paper, we will utilize recent
numbers from Kentucky, where broiler production is an important contributor to agricultural
receipts, ranking second after horses. However, the classifications and systems are similar
throughout the broiler producing regions.

Table 1. Broiler chicken classifications and percentages in Kentucky, 2003.
 Classification         Mature         Days in-house           Number in      Annual
                        Weight,        (approximate)          typical house Numbers 106
                        kg (lb)                              (approximate) (KY, 2003)
 Broiler                2.0 (4.4)             36-44           27,000 – 29,000            58

 Heavy Broiler          2.3 (5.1)             46-51           23,000 – 25,000           128

 Roaster                3.1 (6.8)             55-65           16,500 – 19,000            89

Broilers are raised in houses with dimensions typically about 12m x 150m (40’ x 500’) and
stocked to achieve 29-31 kg m-2 (5.9-6.3 lbs ft-2) at market weight. Thus, actual bird numbers in
houses can vary substantially in otherwise identically sized houses. Bird stocking density is also
adjusted for seasonal influence, typically with fewer birds placed if birds will approach maturity
during hot weather conditions; other stocking adjustments include disease pressure, chick
availability and survivability characteristics. Recent field measurements of U.S. broiler houses
suggest that NH3 emissions are reasonably correlated with bird age and litter condition.2,3 Thus
NH3 emissions estimates based on as-designed bird numbers or animal units (1 AU=1000 kg), or
maximum AU capacity, can grossly over- or under-estimate actual annual emissions. This is
particularly an issue for growers who change market weights repeatedly according to their
contracts, and presents a challenge for estimation of annual NH3 emissions from this sector.

One reason for potential large NH3 emissions from poultry housing is the large volumes of
ventilation air used in thermal environment control of the facilities, and these are closely coupled
with weather events and size of birds (i.e. interior heat and moisture loads on the thermal
environment). A typical broiler house ventilation system design uses sidewall fans (92 cm, 36”)
and static pressure controlled eave inlets for cold and mild weather environmental control, and
end-to-end airflow with large inlets and fans (122 cm, 48”) for “tunnel” ventilation. During the
hottest weather, the ventilation system switches from using sidewall inlets and fans to the tunnel
ventilation mode, with a volumetric capacity of about 0.8 –1.2 m3 h-1 per kg market weight (1 –
1.5 cfm per lb). A typical Kentucky broiler house will have a total of 11 to 15 fans, with design
fan capacity of about 270,000 m3 h-1 (160,000 cfm). Supplemental heat is provided by gas-fired
furnaces, brooders, or radiant heaters. Some form of evaporative cooling is prevalent in southern
producing regions, using either open-cell evaporative pads at the air inlets and/or fogging nozzles
distributed inside the house.

In the next section, we outline a model for estimating annual emissions from broiler production
facilities. While this article is focused on NH3 emission, this approach is appropriate for other
constituents of interest, such as dust.

A linear relation has been recently reported between NH3 emission rate, bird age and litter
condition2,3. The slope on Equation 1 has a standard deviation of 0.0031 g NH3 bird-1 d-1.

Equation 1.

                                    age,      if used litter
ER  0.026  x, where x  
                           0, if 1  age  6; age  6, if new litter


ER, = emission rate, g NH3 bird-1 d-1

A model was developed to predict annual emission rate (AER, kg NH3 yr-1) per bird, per house
and per state population. Key assumptions that must be specified include stocking density,
mature weight and age, and litter condition.

To extend the model from a bird basis to a house-based AER, it is necessary to specify the
number of birds grown in a house, as well as the mean number of flocks per year, for each
category (roaster, heavy broiler, broiler). The annualized AER per house can then be estimated
from the model. To estimate state or national AER, per bird ER predictions from Equation 1 are
developed for each category of bird size. Bird number statistics are combined with these
predictions and totaled to get a state or regional estimate.

Cumulative daily ER over a full year, for broilers, heavy broilers, and roasters are plotted in
Figure 1a-1c. Figure 1a is a plot of mean values, using the slope in Equation 1, Figures 1b and 1c
are similar plots using ±3 standard deviations on the slope. The ER by bird for each class of bird,
for new and used litter, and as affected by uncertainty in the slope of Equation 1, are presented in
Table 2. Cumulative values at year’s end (i.e. AER) are listed in Table 3. As expected, AER is
greatest on used (“built-up”) litter, and varies from 240.6, 202.4 and 166.6 g NH3 bird-1 yr-1 for
roasters, heavy broilers and broilers, respectively.

By using Kentucky averages for birds placed by category (28,000, 24,000 and 18,000 birds per
flock for broiler, heavy broiler and roaster, respectively), and average number of flocks placed in
a house per year, annual house emissions can be estimate (Table 4). For the mean slope in
Equation 1, these values range from 4,331 to 4,858 kg NH3 house-1 yr-1. Perhaps surprising, the
heavy broiler category AER exceeds the roaster category AER, despite the fact that the latter is
grown to a heavier weight. This is because the time between flock placement and the relative
number of flocks in a year results in a greater number of days of larger birds in the house for the
heavy broiler category than for the roaster category. Critical to these estimates, however, is the
assumption that the same type of bird will be grown in a building for the entire year.

Table 2. Emissions Rate (ER) expressed per bird for three classes of broiler maturity.

      ER                  Roaster            Heavy Broiler                  Broiler
 (g NH3 bird
        )            New      Built-up      New       Built-up      New         Built-up
                     litter    litter       litter     litter       litter       litter
  Mean slope          33.2       41.5        24.6       31.9         15.5         21.3

    -3 sigma          21.3       26.7        15.8       20.5          9.9         13.7

   +3 sigma           45.0       56.3        33.4       43.2         21.0         28.9

Table 3. Annual Emissions Rate (AER) expressed per bird for three classes of broiler maturity.

    AER                  Roaster            Heavy Broiler                   Broiler
   (g NH3
 bird-1 yr-1)        New      Built-up      New       Built-up      New         Built-up
                     litter    litter       litter     litter       litter       litter
 Mean slope          191.5      240.6       154.8      202.4        120.4         166.6

 -3 sigma            123.0      154.6       99.4       130.0         77.3         107.0

 +3 sigma            260.0      326.7       210.1      274.8        163.4         226.1

Table 4. Annual Emissions Rate (AER) per house for three classes of broiler maturity.
    AER                   Roaster             Heavy Broiler                 Broiler
   (kg NH3
 house-1 yr-1)       (18,000 birds/flock)   (24,000 birds/flock)   (28,000 birds/flock)

                     New       Built-up      New      Built-up      New        Built-up
                     litter     litter       litter    litter       litter      litter
 Mean slope           3,447       4,331      3,714      4,858       3,371        4,664

 -3 sigma             2,214       2,782      2,386      3,120       2,165        2,995

 +3 sigma             4,680       5,881      5,043      6,595       4,576        6,332

An estimate of regional or state-wide AER can be obtained from this model by specifying the
relative number of each category of bird grown and the values per bird in Table 2. As an
example, using approximate 2003 Kentucky broiler data, Table 5 lists the estimated statewide
emissions, assuming 32, 46 and 21% of annual bird numbers (totaling 275 million birds, about
3.2% of the national supply) were raised as broilers, heavy broilers, and roasters, respectively.
This estimate does not account for houses in which different categories of broiler chickens were
grown during the year. The annual estimate for Kentucky is between 7,006 and 9,019 metric tons
NH3 yr-1, assuming a mean value for the slope in Equation 1.

Table 5. Annual Emissions Rate (AER) estimates for Kentucky, by category of meat-bird.

                        Roaster               Heavy Broiler                 Broiler

                   New        Built-up        New      Built-up      New       Built-up
                   litter      litter         litter    litter       litter     litter

  (metric tons      2,963       3,709         3,145      4,073        898        1,237
   NH3 yr-1)

     Total                   New Litter                              7,006
  (metric tons
   NH3 yr-1)                Built-up Litter                          9,019

According the most recent EPA National Emission Inventory4 for NH3, U.S. poultry (meat-bird
and egg producing birds) contributed an estimated 730,662 metric tons NH3 (664,238 tons) in
2002. The Kentucky values in Table 5 thus represent 1.2% to 1.5% of the national annual

Changing the slope in Equation 1 by ±3 standard deviations results in AER changes of ±36% for
all categories of birds grown, with the relative differences between the categories constant. The
effect of new vs. built-up litter varies by bird category, 26%, 31% and 38% for broiler, heavy
broiler and roaster, respectively. The concept of litter replacement after each flock could
conceivably reduce NH3 AER by a substantial amount; unfortunately, fresh litter is a high-
demand commodity, and used litter is typically land-applied annually. Replacing litter after each
flock would result in a 5 to 10-fold increase in litter costs, and a similar increase in volume of
litter to handle for land applications. Given the slim economic margins involved in broiler
chicken production, it is unlikely that this strategy can be cost-effective.

                                                        Figure 1a: Mean Values


Cummulative Emission (g
                      300           Roaster - new litter
                                    Heavy Broiler - new litter

                      200           Heavy Broiler
                      150           Broiler- new litter


                                0      50         100       150      200       250   300   350   400
                                                                  Julian Day

                                                     Figure 1b: Mean -3 sigma

Cummulative Emission (g

                          300       Roaster - new litter
                                    Heavy Broiler - new litter

                          200       Heavy Broiler
                          150       Broiler- new litter


                                0      50         100       150      200       250   300   350   400
                                                                  Julian Day

                                                     Figure 1c: Mean + 3 sigma

                                     Roaster - new litter
 Cummulative Emission (g

                          300        Roaster
                          250        Heavy Broiler - new litter

                                     Heavy Broiler
                                     Broiler- new litter
                          150        Broiler

                                0       50        100       150      200       250   300   350   400
                                                                  Julian Day

                                                    Figure 2a: Mean Values


Cumulative Emission (kg
                      6000        Roaster - new litter
     NH3/house)       5000        Roaster
                                  Heavy Broiler - new litter
                                  Heavy Broiler
                      3000        Broiler - new litter


                              0      50         100       150      200       250   300   350   400
                                                                Julian Day

                                                   Figure 2b: Mean -3 sigma

Cumulative Emission (kg

                      6000        Roaster - new litter

                                  Heavy Broiler - new litter
                      4000        Heavy Broiler
                      3000        Broiler - new litter


                              0      50         100       150      200       250   300   350   400
                                                                Julian Day

                                                  Figure 2c: Mean + 3 sigma

                      7000        Roaster - new litter
Cumulative Emission (kg

                                  Heavy Broiler - new litter
                      5000        Heavy Broiler

                      4000        Broiler - new litter



                              0      50         100       150      200       250   300   350   400
                                                                Julian Day

Perhaps a more realistic estimate of NH3 AER is to assume one litter clean-out annually. Under
this assumption, AER estimates are reduced about 5% compared to the built-up case. This
reduction arises from the lack of contribution to ER during the first week of each flock placed. In
this scenario, litter replacement is not a very significant abatement strategy.
The model predictions indicate the importance of management strategies which involve litter
management. Typically, ER is reduced when ventilation during cold weather is maintained to
control moisture, or more generally, when litter is managed to control pH and moisture content.
Lower moisture content prevents microbial activity from converting urea in feces to NH3, and
reduced pH prevents generated ammonium NH4 from becoming NH3.
This technique of estimating AER can be used as abatement strategies are developed and
quantified. It requires simply a suitable substitution of Equation 1. A key assumption involved in
this analysis include the neglect of any emissions from houses between flocks.

A model for estimating annual NH3 emissions (AER) from U.S. poultry operations was
developed to account for the varying market weight of broiler chickens. The model utilizes
recently obtained field data on NH3 emissions, and utilizes standard agricultural statistics
compiled by USDA’s National Agricultural Statistics Service. The model incorporates the
uncertainty in NH3 estimates, the condition of litter, and the varying market weight of birds
grown. The method can be adopted to other poultry building emissions, with suitable insertion of
the replacement to Equation 1 for NH3.

Results suggest that a 36% uncertainty in AER (as determined by a three-sigma estimate) is
expected across the three categories of birds grown (broilers, heavy broilers, and roasters). The
model predicts a 26 to 38% difference in AER between new vs. built-up litter. While litter
replacement after each flock may appear to be an attractive abatement strategy, the logistics and
economics would require significant changes in current production practices, and could have
undesirable and unintended environmental impacts from the sheer volume of low-N litter that
would have to be replaced.

1. USDA National Agricultural Statistics Service web site. See bulletin 994, found at (accessed January 2005)

2. Casey, K. D.; Gates, R. S.; Wheeler, E. F.; Xin, H.; Zajaczkowski, J. L.; Topper P. A.;.
   Liang, Y. 2004. Ammonia emissions from Kentucky broiler houses during winter, spring, and
   summer. In Proc. 97th Annual A&WMA Conference and Exhibition: Sustainable
   Development: Gearing Up for the Challenge, CD-ROM. Pittsburgh, Pa.: A&WMA

3. Wheeler, E.F.; Casey, K.D.; Zajaczkowski, J.L.; Topper, P.; Gates, R.S.; Xin, H.; Liang, Y.
   2004. Seasonal ammonia emission variation among twelve U.S. broiler houses. Paper No.
   044105. Presented at the International ASAE Mtg. Ottawa, Ontario. Aug. 1-4.

4. EPA. National Emission Inventory – Ammonia Emissions from Animal Husbandry. Found at (accessed January, 2005)


shenreng9qgrg132 shenreng9qgrg132 http://