carbon footprint and ammonia emissions of california beef ... · key words: ammonia, beef cattle,...

4641

© 2012 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2012.90:4641–4655 doi:10.2527/jas2011-4653

Key words: ammonia, beef cattle, carbon footprint, farm model, greenhouse gas, simulation.

ABSTRACT: Beef production is a recognized source of greenhouse gas (GHG) and ammonia (NH3) emis-sions; however, little information exists on the net emissions from beef production systems. A partial life cycle assessment (LCA) was conducted using the Inte-grated Farm System Model (IFSM) to estimate GHG and NH3 emissions from representative beef production systems in California. The IFSM is a process-level farm model that simulates crop growth, feed production and use, animal growth, and the return of manure nutrients back to the land to predict the environmental impacts and economics of production systems. Ammonia emis-sions are determined by summing the emissions from animal housing facilities, manure storage, fi eld applied manure, and direct deposits of manure on pasture and rangeland. All important sources and sinks of methane, nitrous oxide, and carbon dioxide are predicted from pri-mary and secondary emission sources. Primary sources include enteric fermentation, manure, cropland used in feed production, and fuel combustion. Secondary emis-sions occur during the production of resources used on the farm, which include fuel, electricity, machinery, fer-tilizer, and purchased animals. The carbon footprint is

the net exchange of all GHG in carbon dioxide equiva-lent (CO2e) units per kg of HCW produced. Simulated beef production systems included cow-calf, stocker, and feedlot phases for the traditional British beef breeds and calf ranch and feedlot phases for Holstein steers. An evaluation of differing production management strate-gies resulted in ammonia emissions ranging from 98 ± 13 to 141 ± 27 g/kg HCW and carbon footprints of 10.7 ± 1.4 to 22.6 ± 2.0 kg CO2e/kg HCW. Within the British beef production cycle, the cow-calf phase was respon-sible for 69 to 72% of total GHG emissions with 17 to 27% from feedlot sources. Holstein steers that entered the beef production system as a by-product of dairy production had the lowest carbon footprint because the emissions associated with their mothers were primarily attributed to milk rather than meat production. For the Holstein system, the feedlot phase was responsible for 91% of the total GHG emission, while the calf-ranch phase was responsible for 7% with the remaining 2% from transportation. This simulation study provides baseline emissions data for California beef production systems and indicates where mitigation strategies can be most effective in reducing emissions.

Carbon footprint and ammonia emissions of California beef production systems1

K. R. Stackhouse-Lawson,* C. A. Rotz,† J. W. Oltjen,* and F. M. Mitloehner*2

*Department of Animal Science, University of California, Davis 95616; †USDA-ARS, Pasture Systems and Watershed Management Research Unit, University Park, PA 16802

INTRODUCTION

Anthropogenic greenhouse gas (GHG) and ammo-nia (NH3) emission reporting from livestock production has increased in priority on political, social and eco-

nomic agendas. However, reporting accurate GHG and NH3 emissions is challenging due to the complexity of animal production systems and the regional ecological differences that drive biological processes.

Evaluation of gaseous emissions from animal pro-duction systems requires a whole system modeling approach (Stewart et al., 2009). Greenhouse gas emis-sions and the associated carbon footprint are often deter-mined through Life Cycle Assessment (LCA). An LCA includes all primary emissions during the production process and the secondary emissions that occur during the production of resources used in the production pro-cess (Rotz et al., 2010). Quantifi cation of these emis-sions can be assisted through simulation of the produc-

1 The authors appreciate the assistance of Larry Forero and Josh Davy UC Cooperative Extension advisors in Shasta and Tehama County, as well as George and Chris McArthur and Henry Giacomini Shasta county beef producers. Thanks also to anonymous feedlot own-ers and operators, and calf ranch owners and operators in California who contributed to developing our simulated production systems.

2 Corresponding author: [email protected] August 31, 2011.Accepted August 19, 2012.

Published January 20, 2015

Stackhouse-Lawson et al.4642

tion processes using a tool such as the Integrated Farm System Model (IFSM; USDA-ARS, 2011). This farm scale simulation integrates the interacting processes of pasture, rangeland and crop growth, crop harvest, feed storage, grazing, feeding, animal production, and manure handling to determine the long-term performance, envi-ronmental impact, and economics of production systems. This approach can help avoid practices that may reduce NH3 emissions while increasing GHG emissions and situ-ations where emissions may be reduced at the expense of productivity (Janzen et al., 2006).

The objective of this work was to use simulation modeling combined with LCA to compare the cradle-to-farm gate carbon footprint and NH3 emissions of the major beef production systems in California. Production phases included cow-calf, stocker, feedlot, Holstein calf ranch, and Holstein feedlot phases. The calculation of net emissions determined the relative contribution of each phase to the overall emissions and carbon footprint of the consumable product.

MATERIALS AND METHODS

Model description

Beef production systems were modeled according to industry typical production practices using the IFSM (Rotz et al., 2005; Rotz et al., 2011). The major processes of feed production, animal performance, and manure production and handling were simulated through time over 25 yr of weather to estimate daily and long term emissions. This software and additional information on the model are avail-able through the internet (USDA-ARS, 2011).

A cradle-to-farm gate LCA is used in IFSM to integrate emissions from primary and secondary GHG sources (Rotz et al., 2010; Rotz et al., 2011). Primary emissions are gases produced by the animal (enteric fermentation), emissions from feces and urine, and emissions during the production of feed. Secondary emissions are emitted during the pro-duction of resources used, which include machinery, fer-tilizer, fuel, electricity, and other important inputs used in the production process. Primary and secondary emissions are interlinked and dependent on the effi ciency of the cattle production system and the length of the animal’s produc-tive life. A carbon footprint is determined by totaling the net lifetime GHG emissions and dividing by lifetime pro-duction in kg of HCW.

A beef production system represents all phases nec-essary to produce a fi nished market beef animal. Beef production systems vary depending on region and in-clude 1) cow-calf and feedlot phases, 2) cow-calf, stock-er, and feedlot phases, 3) cow-calf and stocker phases (more unusual, only 1 to 2% of beef are fi nished on grass), and 4) calf ranch and feedlot phases for Holstein

cattle. Greenhouse gas emissions were simulated for each phase and totaled to obtain emissions per unit of fi nished beef. To account for the loss of animals, emis-sions for the cow-calf and stocker phases were increased to account for mortality in the subsequent phases. Emis-sions from replacement heifers, nursing calves, and bulls are included in the cow-calf phase emissions.

The system boundaries are beyond the physical form of each operational phase to include the produc-tion of feeds and other resources used to maintain the herds (Figure 1). Emissions from the production of feed crops are included in the carbon footprint whether the feed is produced on the ranch or purchased from an out-side farm. In the IFSM, all manure nutrients are assumed to be used in feed-crop production unless a portion is designated as exported from the farm. Manure from California beef feedlots is most often exported for use in organic vegetable operations; therefore, greenhouse gas emissions are reported with and without inclusion of emissions resulting from manure beyond the farm gate.

The 3 GHG gases emitted from beef cattle produc-tion are carbon dioxide (CO2), methane (CH4) and ni-trous oxide (N2O). Each has different global warming potentials which quantify the capacity of the gas to trap heat in the atmosphere. To standardize GHG emissions, the International Panel on Climate Change (IPCC, 2001) has established the global warming potential equivalence index to convert GHG to CO2e units. In our model, the conversion factors are 25 kg of CO2e/kg CH4 and 298 kg CO2e/kg N2O (IPCC, 2007).

Primary GHG emissions

Primary GHG emissions include all CO2, CH4, and N2O produced during the production of feed, mainte-nance of animals, and handling of manure. Process level simulation is used to predict animal and manure stor-age emissions; simpler relationships are used to estimate emissions from less important sources (Rotz et al., 2010, 2011).

Carbon Dioxide. The major sources of CO2 from beef production systems are animal respiration and mi-crobial respiration from land applied manure. Manure storage is also a small contributor. Rangeland forages, pastures grasses, and croplands assimilate CO2 from the atmosphere through photosynthetic fi xation during pe-riods of forage and crop growth and emit CO2 through plant and soil respiration and manure decomposition. On an annual basis rangeland forages, pasture grasses, and croplands are a net sink where plants assimilate more CO2 than is emitted (Chianese et al., 2009a).

To predict CO2 emissions from feed production (for rangeland, pasture, and crops), a carbon balance is deter-mined that considers all fl ows in and out of the land dur-

Carbon footprint and ammonia emissions in beef 4643

ing the production of forage and feed for the herd (Rotz et al., 2010). In the current study, the carbon balance also includes carbon from animal and manure sources. These sources are often ignored in GHG accounting (IPCC, 2001, 2007); however, when looking at the partial life cycle from cradle to farm gate, it is important to con-sider all carbon emission sources and sinks. Respired CO2, enteric CH4, and carbon converted to body mass are integral parts of the carbon balance that begins with photosynthetic fi xation by plants. Plants are consumed by the animals that release a portion of the carbon back to the atmosphere and use carbon for growth and muscle development. To predict the total C fi xed through pho-tosynthesis and emitted through autotrophic and hetero-trophic respiration, we include the net C assimilated in the feed consumed by the animals converted to CO2e.

Carbon dioxide emitted by the herd is predicted as a function of DMI and BW of 6 possible animal groups (Chianese et al., 2009a). The 6 animal groups are cows, nursing calves, young heifers, yearling replacement heifers, stocker cattle, and fi nishing cattle with animal characteristics defi ned by breed (Rotz et al., 2005). Ani-mals are fed to meet requirements for minimum forage, energy, and protein using requirements based upon the Cornell Net Carbohydrate and Protein System, level 1 (Fox et al., 2004). Animals on pasture can be fed an all forage diet where grain supplementation is used only in an emergency if cow BW loss reaches 0.9 kg/d or

stocker rate of BW gain drops to 10% of their desired rate of BW gain due to poor forage quality (Rotz et al., 2005). Carbon dioxide emitted from manure in drylot corrals and on pasture and rangeland is determined us-ing an empirical equation that incorporates effects of air temperature and manure surface area (Chianese et al., 2009b).

Operation of equipment releases CO2 when fossil fuel is burned (Rotz et al., 2010). The conversion used to determine CO2 from fossil fuel use is 2.64 kg of CO2/L of diesel fuel consumed (Wang, 2007). Fuel use is pre-dicted in IFSM through the simulation of equipment op-erations used in feed production and feed and manure handling (Rotz et al., 2011). For the present study, a pickup truck was added for general use in the cow-calf and stocker phases as this is typical for the industry.

Methane. Enteric fermentation is the largest con-tributor of CH4 from beef cattle (USEPA, 2009). Ma-nure storage and manure deposited on corral fl oors, pastures, and rangelands also contribute, but to a lesser extent (Chianese et al., 2009c). To predict enteric CH4, a model is used where methane emissions are a nonlinear function of metabolizable energy intake and the ratio of starch over ADF content in the diet (Mills et al., 2003). Total DMI for the animal group is the sum of the DMI for maintenance and gain (Rotz et al., 2005). The DMI for maintenance is the NEm requirement divided by the NEm of the diet and the DMI required for BW gain is

Figure 1. Primary and secondary emission sources for partial life cycle assessment of the carbon footprint of the California beef production system.


the NE required to meet the ADG goal divided by the NEg of the diet. Daily CH4 emissions are predicted for the 6 possible beef animal types based on their diets. The model responds with increased CH4 emissions from high fi ber diets and decreased CH4 emissions with diets high in starch (Rotz et al., 2010, 2011). This relation-ship, developed for dairy cattle, provided the only suit-able model for predicting enteric methane from cattle on very high or all forage diets. For high concentrate diets, predicted methane production was limited to a lower value similar to the IPCC (2006a) emission factor for feedlot cattle.

The cow-calf phase is maintained on rangeland or pasture all year long. Methane is predicted from pasture and rangeland using an average emission factor of 0.76 g of CH4/kg of feces DM (Chianese et al., 2009c). As such, the manure CH4 emission from grazing cattle is a function of daily manure excretion and the time spent on the pasture or rangeland each day (Rotz et al., 2010).

Feedlot animals are housed in dirt-fl oored drylot corrals where manure accumulates in a pack. To predict CH4 emissions from this manure pack, an adaptation of the tier 2 approach of the (IPCC, 2006a) is used, where the methane emission rate is a linear function of ambient outdoor temperature (Rotz et al., 2011). Stored manure also contributes to CH4 emissions and is predicted as a function of the total volatile solids in the storage and a methane conversion factor modeled as a linear function of ambient temperature (Rotz et al., 2011).

Nitrous Oxide. Cultivated croplands and pastures are the largest source of N2O emissions due to the ma-nure and fertilizer N inputs. Daily nitrous oxide emis-sions from croplands and pastureland are simulated as a function of soil texture, moisture content, and N content (Rotz et al., 2011).

Nitrous oxide is produced in lower amounts from drylot corral manure packs. These emissions are pre-dicted using the tier 2 approach from the IPCC (2006a) using an emission factor of 0.02 kg of N2O-N/kg of N excreted in the corral (Rotz et al., 2010). Nitrogen ex-creted is multiplied by the emission factor and the N to N2O conversion factor of 1.57 to obtain predicted N2O emission. Manure stored in stacks has an emission factor of 0.005 kg N2O-N/kg of N stored (IPCC, 2006a).

Secondary GHG emissions

Secondary emission sources include the produc-tion of electricity, fuel, machinery, fertilizer, pesticide, and purchased feed used in the maintenance of animals, handling of manure, and production of feed (Rotz et al., 2010). Emissions for the production of fuel and elec-tricity are set using the emission factors of 0.374 kg of CO2e/L of fuel and 0.73 kg of CO2e/kWh of electricity

(Wang, 2007). The amounts of fuel and electricity used are determined through the simulation of production processes by IFSM (Rotz et al., 2011).

Emissions associated with machinery include those during manufacture along with those for maintenance and repairs. The GHG emission factor for the production of machinery is 3.54 kg of CO2e/kg of machinery mass (Rotz et al., 2010). Machinery use and repairs are pre-dicted through the simulation of machinery operations (Rotz et al., 2011).

For our analysis, fertilizer, pesticide and seed use only apply to off-farm feed production. Fertilizer use is based upon the defi ned application rates and crop areas (Rotz et al., 2011). Emission factors for nitrogen, phos-phate, and potash production are 3.31, 1.03, and 0.87 kg of CO2e/kg, respectively (Wang, 2007). Pesticide use is based on typical requirements for producing each crop (Penn State, 2011) and the land area used for that crop. This emission is generally small and a set emission fac-tor of 22 kg of CO2e/kg of pesticide active ingredient is used (Wang, 2007). Emission from seed production is also very small (0.3 kg of CO2e/kg of seed; Rotz et al., 2011). User-assigned crop areas and typical seeding rates for each crop are used to predict seed use.

Emission factors were developed for the major feeds imported into the production systems. Steam fl aked corn makes up 80 to 85% of the concentrate diet fed to feedlot cattle in California. The corn, typically transported from the Midwest, is processed with steam during fl aking. A representative corn farm was simulated using IFSM with climatic data from Iowa to determine a typical GHG emission of 0.25 kg CO2e/kg DM of corn produced. We assumed a round trip distance traveled and the result-ing fuel use to transport grain from Iowa to California was 5,800 km and 2,300 L, respectively. Fuel use was multiplied by the fuel emission factors above and di-vided by the mass of corn DM transported to obtain a carbon footprint of 0.32 kg CO2e/kg DM. For the steam processing, we assumed gas use was 12.8 m3/t DM of corn (L. McKinney, Kansas State University, KS, per-sonal communication), which gave a footprint for steam processing of 0.03 kg CO2e/kg DM. The total carbon footprint or secondary emission assigned to steam fl aked corn used in California was 0.6 kg CO2e/kg DM.

Another major feed used in beef production is hay. Representative farm simulations with IFSM were again used to obtain a typical emission associated with alfalfa or stock quality hay production of 0.2 kg CO2e/kg DM. Considering that the transport was from within or near California, an emission associated with transport of 0.2 kg CO2e/kg DM was assigned giving a total emission of 0.4 kg CO2e/kg DM.

Other feeds included milk replacer for the Holstein calf ranch and cotton seed, fat and minerals primarily


for the fi nishing operations. No information was found on the GHG emissions associated with the production of milk replacer. A value of 12.5 kg CO2e/kg of milk pow-der was estimated based upon the production and drying of powdered milk (Ramirez et al., 2006). For all other feeds, a secondary emission value of 1.0 kg CO2e/kg DM was used. Considering the relatively small amounts of these other feeds fed, error in the estimation of this value had little effect on the overall carbon footprint of California beef production systems.

Ammonia emissions

Cattle are not effi cient users of protein N, and excess protein N is excreted primarily in urine as urea. Follow-ing excretion, ammonia emission from a manure surface involves 5 important processes: urea hydrolysis, disso-ciation, diffusion, aqueous-gas partitioning, and mass transport away from the manure surface to the atmo-sphere (Rotz et al., 2011). The 4 important NH3 sources are housing facilities, manure storage, fi eld applied ma-nure, and direct deposits on pasture or rangeland. Link-ing the models for each of the emission processes pre-dicts the emission rates from each of the major farm or ranch NH3 sources.

Fecal and urine excretions in a feedlot form a thin layer of manure across the soil surface where the emis-sion processes transform the N and control the NH3 emission rate (Rotz et al., 2011). Because the manure re-mains on this surface for many days, most of the ammo-niacal N is transformed and volatilized. When manure is scraped from the feedlot surface and stacked for storage or composting, remaining organic N slowly decomposes to form ammoniacal N. Daily emission rate is predicted as a function of the exposed surface area, ammoniacal N concentration, temperature, air velocity, and surface pH (Rotz et al., 2011).

Ammonia emission from pasture and rangeland is also predicted considering the dissociation, diffusion, aqueous-gas partitioning, and mass transport processes. Daily emissions are a function of the amount and form of N applied, ambient temperature, and precipitation (Rotz et al., 2011). Excreted N is determined by the diet and the portion of time each animal group spends on pasture or rangeland. Remaining N is available for fertilization of the growing forage.

Production Systems

California beef production. Beef cattle production in California is unique, when compared to beef produc-tion systems in other regions of the United States. Cali-fornia’s predominant rangeland is mediterranean where forage growth is supported by rains in the winter and

early spring. The state also has intermountain ranges where forage growth occurs during the late spring and early summer. Therefore, cattle graze the lower eleva-tion forage during the winter months and are then trans-ported to the intermountain ranges during the summer months, or are moved onto irrigated pasture during the late summer. California is also unique in that it is home to approximately 2 million dairy cows, 20% of the na-tion’s entire herd. Bull calves from the dairy sector enter the beef supply chain and make up a large portion of the feedlot cattle in the state. Additionally, California grows relatively little of the grain crops used in fi nishing cattle rations, and the cost of transport of these crops from the Midwestern US is high.

California beef cattle production typically consists of a 3-phase system: 1) cow-calf, 2) stocker, and 3) feed-lot. Each phase is often owned and operated independent of the other phases. The producers in the cow-calf phase maintain a breeding herd of cows, replacement heifers, and bulls. Cows are usually maintained extensively on rangelands or pasture, where they give birth to calves that are sold at weaning (typically around 205 d of age). The producers of cattle in the stocker phase purchase the weaned calf and add additional BW through graz-ing extensive rangelands. The animal exits the stocker phase and is sold into the feedlot at approximately 350 kg BW. Feedlots maintain large numbers of cattle and utilize high energy feed sources to increase BW gain and decrease time spent on feed. Cattle leave the feedlot to enter the food chain at approximately 550 kg and about 17 mo of age.

Holstein steers are managed with a 2-phase sys-tem: 1) calf ranch and 2) feedlot. At birth the bull calves are purchased by a calf ranch that specializes in rear-ing calves. The calves exit the calf ranch at 4 mo of age when they are sold to the feedlot. They remain in the feedlot until they weigh approximately 550 kg and are 15 mo of age.

Simulated ranches and farms. Representative farms or ranches were simulated for each phase of beef cattle production. Our production systems were defi ned through consultation from beef researchers, UC Cooperative Ex-tension advisors, and beef cattle producers. All simulations were conducted for a 20 to 25 yr period using recent his-torical weather for their location.

The cow-calf phase was simulated in Tehama Coun-ty, representative of Mediterranean climate rangeland and typical of many cow-calf operations in California. The herd was comprised of 600 Black Angus cows, 20 Black Angus bulls, 90 replacement heifers (15% re-placement rate), and 574 September born nursing calves (considering 4% twins and 8% mortality). The average lifespan of the cows was 8 yr with a calving interval of 365 d, thus each cow was expected to have 6 calves dur-


ing her life. Cull cow production was included by add-ing 125 kg SBW to the amount of animal mass produced per fi nished steer when calculating the footprint for the full system (Tables 4 and 5).

Cattle were produced on 2671 ha of rangeland and pasture with a medium sandy loam soil typical of this region (Table 1). Land use was determined based on a stocking rate of 4.5 ha/cow, which represented an aver-age between extensive winter and spring rangeland and irrigated spring and summer pasture. Rangeland forage consisted of annual and perennial grassland (Table 1). Cattle were fed primarily through grazing, but during times of poor forage growth, they were supplemented with poured protein supplement blocks (referred to as a protein tub) and stock quality hay. Nursing calves were creep fed a grain supplement and weaned at 210 d of age weighing 280 kg. All feed other than grassland was

purchased and brought onto the ranch. One pickup truck was used that consumed 4,450 L/yr of fuel in the activi-ties of the operation. Dry hay was stored inside a shed and a machine shed was on site.

The ranch for the stocker phase was simulated in Shasta County, representative of intermountain climate rangeland, and typical of many stocker operations in California. The herd of 1,000 Black Angus stocker cattle was implanted with a growth hormone on arrival. Cattle were maintained on 4047 ha of rangeland based on typi-cal stocking rates of about 4.0 ha/animal. Cattle were transported from the cow-calf phase in Gerber to McAr-thur, California in May (after weaning) and removed from the rangeland in September. Rangeland forage consisted of annual and perennial grassland produced on a medium sandy loam soil (Table 1). Cattle were fed primarily through grazing. During times of poor forage

Table 1. Characteristics of simulated beef production systems in California

Item Cow-calf Stocker Angus feedlot Holstein calves Holstein feedlot

Land and weatherWeather data location Gerber, CA McArthur, CA Kettleman City, CA Kettleman City, CA Meloland, CASimulation, yr 25 25 25 25 19Total land, ha 2,671 4047 17.8 17.8 17.8

Owned, ha 890 17.8 17.8 17.8Rented, ha 1,781 4047

Soil type Medium sandy loam Medium sandy loam Deep sandy loam Deep sandy loam Deep sandy loamFarm topography, % slope 15 to 25 > 25 0 to 3 0 to 3 0 to 3Soil phosphorous, mg/kg 30 to 50 30 to 50 50 to 100 50 to 100 50 to 100Crops Perennial grassland Perennial grasslandInitial sward DM, kg/ha 392 392Initial sward composition 60:40 60:40 ,cool-season grass:forbForage yield adjustment, % 90 90Fertilizer, % manure 100 100Grazing area summer, 1,781 4,047 spring, and fall, haGrazed forage 70 70 yield adjustment, %

Animal ManagementLabor, min/animal-d 0.4 0.1 0.1 2 0.1Number of animals, 600 1,000 5,000 15,000 5,000

Replacement heifers 90Breed Angus Angus Angus Holstein HolsteinCalving month September September September All year SeptemberWeaning age, d 210 210 210 60 60Days on feed 365 182 121 and 212 120 331Mortality, % 6.0 2.0 1.6 and 2.8 6.5 4.4Diet Grazing forage Grazing forage High grain Milk and starter grain High grain

fi nishing diet1 (d 1-60), high grain1 fi nishing diet1

(d 60-120)Supplemental feed Hay, protein tubs, Hay, protein tubs Cottonseed Cottonseed Cottonseed

grain for calvesAnimal housing None None Open lot Triplet hutches Open lot1Finishing diet varies but is approximately 75% steam fl aked corn, 20% hay, and 3% cottonseed on DM basis with added fat and minerals.


growth, they were supplemented with protein tubs and stock quality hay purchased and brought onto the ranch. Dry hay was stored inside a shed. One pickup truck was used consuming 2,300 L/yr of fuel in the activities of the stocker operation. Considering 2% mortality, 980 ani-mals were sold to the fi nishing phase.

Two simulations were performed for the Angus feedlot phase of the beef life cycle with cattle entering the feedlot from 1) a stocker system at 379 kg or 2) a cow-calf system at 280 kg (Table 2). Both simulations for the feedlot phase were simulated in Kern County, representative of mediterranean climate and typical of the predominant beef feedlot operations in California. The herd of 5,000 Black Angus cattle was implanted with a growth hormone on arrival. Mortality was 1.6% for simulation 1 and 3.2% for simulation 2. They were housed on 17.8 ha of land. Cattle were transported from McArthur (stocker phase, for simulation 1) or Gerber (cow-calf phase, simulation 2) to Kettleman City, Cali-fornia in September.

Cattle were fed a high concentrate fi nishing diet of approximately 75% (DM) steam fl aked corn, 20% al-falfa hay, 3% cottonseed, fat, and mineral with an iono-phore feed additive. All feed was purchased and brought onto the feedlot. Finished BW was set at 571 kg with HCW determined as 62% of FSBW. Cattle were housed in a drylot corral and manure was collected by scraping. Simulations were conducted to show the impacts of all manure being exported from the production system (typ-ical of California) compared to a scenario in which all manure remains within the production system. Machin-ery included a tractor, one large feed mixer, one medium box spreader, and 2 medium skid-steer loaders. Dry hay was stored inside a shed, concentrate feeds were stored in a commodity shed, and equipment was stored in a ma-

chine shed.The calf-ranch phase of the Holstein beef life cy-

cle was simulated in Kern County, representative of a mediterranean climate, and typical of many calf-ranch operations in California (Table 1). The herd consisted of 15,000 Holstein bull calves housed in triplicate calf hutches on 17.8 ha of land. After birth, the calves were transported 200 km to the calf ranch. Their carbon foot-print leaving the dairy farm was determined to be 112 kg CO2e per calf through simulation of dairy production systems using an economic allocation procedure (Rotz et al., 2010).

Calves were fed replacement milk and starter grain from d 1 to 60 and a high concentrate diet from d 60 to 120, which included cottonseed as a protein source. Energy used to mix replacement milk and clean feed-ing equipment was included as a secondary emission in the overall GHG calculation. Calves were weaned at 60 d of age with a mortality of 6.5%. Initial and fi nal BW was set at 42 and 127 kg, respectively. All feed was pur-chased and brought onto the ranch. Machinery included 2 tractors, 2 small feed mixers, 1 medium box spreader, and 2 medium skid-steer loaders. Replacement milk was stored inside a shed, other feeds were stored in a com-modity shed, and equipment was stored in a machine shed.

The Holstein feedlot phase was simulated in Impe-rial County, representative of desert climate, and typical of predominant Holstein feedlot operations in Califor-nia. The herd consisted of 5,000 Holstein steers. Ani-mals were implanted with a growth hormone on arrival, and mortality was 4.4%. Cattle were maintained on 17.8 ha of land for 331 d. Finished BW was set at 573 kg and HCW was calculated as 59% of FSBW. Cattle were transported from Kern County (calf-ranch phase) to Im-perial county. All other assumptions were similar to the Angus feedlot simulation.

Transportation of animals

Transportation of animals was not included in IFSM, so separate calculations were applied to determine emis-sions associated with this component. The round trip distance for transportation in the Angus production sys-tem that included the stocker phase was estimated at 2,350 km; without the stocker phase the distance was 2,150 km. For the Holstein system, the round trip dis-tance was 550 km. Fuel use was estimated based upon a typical truck load (50 animals) and fuel effi ciency (2.5 km/liter). Greenhouse gas emissions were calculated by multiplying the fuel use per animal transported by the emission factors for fuel production and combustion. This emission was increased an additional 40% to ac-count for embodied CO2e in the manufacture and main-

Table 2. Predicted performance for various beef produc-tion systems simulated with IFSM

MeasureAngus cow 1

Angus calf Stocker

Angusfeedlot

Angus feedlot (w/o

stocker phase)

Holstein calf

Holstein feedlot

DOF 2 365 212 182 121 212 121 334BW, kg

Initial - 40 283 379 280 42 139Final 625 280 384 571 571 127 573

ADG, kg/d - 1.1 0.5 1.4 1.4 0.7 1.3DMI, kg/d 8.9 2.7 6.6 10.0 9.4 2.6 3 8.0G:F - 0.41 0.08 0.14 0.15 0.27 0.16

1Heifers calve at 2 yr of age.2Days on feed.3Dry matter intake is calculated for the period after calves are weaned and

housed in group pens on a high grain feedlot diet.


tenance of the truck (Rotz et al., 2010).

Model uncertainty

All farm-level emission estimates have an uncer-tainty associated with their prediction. Net farm emission uncertainty is determined by defi ning the uncertainties of each component contributing to net GHG emissions. Proper statistical quantifi cation of biological system un-certainties requires large data sets that are not available. Therefore, the IPCC (2006b) has used expert opinion to estimate the uncertainty of emission factors used in pre-dicting GHG emissions, and this provides the best basis for estimating the uncertainty in predicting farm emis-sions. They report that their Tier 2 methodologies for CH4 emission from enteric fermentation and manure handling have an uncertainty of ±20%, and state that this uncer-tainty can be reduced by applying their methodologies to well defi ned conditions (IPCC, 2006a). Since our model predicts emissions for a herd with a specifi c animal type and feed composition, an uncertainty of ±10% is used to predict enteric methane emission along with ±20% for manure emissions (Chianese et al., 2009c). Similar to the IPCC (2006a), an uncertainty of ±50% is used for N2O emission from fi eld and manure surfaces (Chianese et al., 2009d). For fuel combustion and secondary emissions, an uncertainty of 20% is used. This assumes that our trans-port distances for animals and feed generally represent the average for production in California. No data exist on the uncertainty in predicting NH3 emissions of farm components, but preliminary evaluation of this emission component indicates that predictions are within ±20% for pasture, manure storage and feedlot surfaces (Rotz, un-published data). The uncertainties of the emissions from all farm components of each production phase are com-bined to give the overall uncertainty as the square root of the sums of squares of each emission component times its uncertainty (IPCC, 2006b).

Model evaluation of predicted animal emissions

To evaluate this new application of IFSM, predicted animal emissions were compared to measured data ob-tained through experiments conducted by Stackhouse et al. (2011). In these experiments, CH4, N2O, and CO2 emissions were recorded from 5 different animal types us-ing a whole animal respiration chamber (Table 3). The 5 animal types were: fi nished Angus and Holstein feedlot steers with BW of 540 and 530 kg, respectively, growing Angus and Holstein feedlot steers with BW of 350 and 337 kg, and Holstein calves with a BW of 150 kg.

For further evaluation of the enteric emission model as applied to beef cattle, simulated emissions were com-pared to those measured in a number of published stud-

ies. These studies included cows, heifers and steers on pasture or an all forage diet and heifers and steers fed a high concentrate diet. For the high concentrate diets, studies were selected that used diets similar to that used in our simulation where about 90% of the diet was con-centrate with corn normally being the primary concen-trate fed. For each comparison, animals of the reported type and size were simulated over 25 yr with the appro-priate diet, and the predicted mean emission was com-pared to that reported in the experimental study.

RESULTS

Model evaluation

Emissions predicted by IFSM were similar to those measured in whole animal respiration chambers accord-ing to BW and animal type (Table 3). Simulated CH4 emissions were within 1 to 11% of measured values across the 5 animal groups showing good agreement. Nitrous oxide emissions were very low for both simu-lated and measured data, and simulated emissions av-eraged 15% less than those measured. Differences be-tween measured and simulated CO2 emissions were greater with simulated values varying from 55% lower to 21% greater than those measured. The greatest differ-ence was for small Holstein animals where the model under predicted measured emissions. Measured CO2 emissions were not greatly different across large differ-ences in animal BW, which implies that the model more accurately represented typical animal CO2 emissions as a function of animal size.

When compared to emissions measured in other stud-ies, simulated enteric methane emissions were generally similar to those measured across a wide range in feed in-take (Table 4). In 7 of the 13 comparisons for animals on pasture or all forage diets, simulated emissions were

Table 3. The IFSM evaluation comparing predicted emissions from different animal types at different BW to emissions measured in a whole animal respiration cham-ber study by (Stackhouse et al., 2011)

Breed, BW, kg

Calorimetric emissions,

kg/animal/d MeasuredDMI,1,2

kg/d

Simulated emissions,

kg/animal/d SimulatedDMI,kg/d CH4 N2O CO2 CH4 N2O CO2

Angus, 540 0.100 0.0004 7.89 11.18 0.094 0.0005 8.22 10.40Angus, 350 0.068 0.0005 6.88 7.44 0.076 0.0003 5.72 7.84Holstein, 530 0.100 0.0004 7.11 11.32 0.101 0.0004 8.17 10.57Holstein, 337 0.076 0.0004 6.42 7.54 0.079 0.0003 5.67 8.07Holstein, 150 0.048 0.0003 5.41 3.19 0.040 0.0002 2.42 3.87

1Dry matter intake from the measured animals were taken from the day before the animals entered the chamber.

2Diet was similar across all animal types and was a high concentrate grain diet.


similar to those measured or the ranges in simulated data overlapped the measured ranges. In most cases though, the model tended to over-predict methane emission with values about 20% greater than the reported measured values. In a few of the experimental studies, DMI were exceptionally high or the CH4 emission was low in rela-tion to dry matter intake. For all other studies, the CH4 emission was within the range of 17 to 28 g CH4/kg of dry matter intake. For the simulated data, most values were in the range of 23 to 31 g CH4/kg DMI. Given the uncer-tainty in measuring dry matter intakes and CH4 emissions from grazing animals, the model was found to adequately represent this emission source.

For feedlot animals on high concentrate diets, simu-lated emissions compared well with reported measured values (Table 4). For both measured and simulated data, the enteric emission was around 9 g CH4/kg of dry mat-ter intake. The one exception was the study by Cooprider et al. (2011) where the reported data was not measured but predicted using mechanistic models of rumen func-tion. Over all studies, the comparisons supported that the model adequately represented enteric emissions over a wide range in animal types and diets.

Simulated production systems

Average simulated performance results, including ini-tial and fi nal BW, ADG, DMI and G:F for each phase is shown in Table 2. The Angus calf performance parameters were simulated to represent an average of the annual calf crop from the cow-calf operation. The stocker operation performance results were simulated to start with the wean-ing BW of the calf from the cow-calf phase. The Angus feedlot was simulated so that the initial BW in the feedlot was the fi nal BW in the stocker phase. An Angus feedlot without a stocker phase was a production system where the weaned calf was transferred directly to the feedlot for an 8 mo fi nishing period. Days on feed (DOF) refl ect the time animals spend in each phase. The predicted ADG, DMI, and G:F were the average over the DOF.

Greenhouse gas emissions and carbon footprints for the Angus and Holstein beef production systems are presented in Tables 5, 6, and 7. Emissions associated with the cow-calf phase included those from the cows, calves, replacements heifers, and bulls for 365 d, while the stocker and feedlot phases only accounted for the emissions from the single animal for the time the animal was in that particular phase. The cow-calf phase of the beef production system contributed the largest portion

Table 4. A comparison of measured and simulated enteric methane emissions from various beef animals on all forage and high concentrate diets

Experimental study Measurement method Animal type Body mass, kg

DMI, kg/d Emission rate, g/ d

Measured Simulated Measured Simulated

Grazing cattle, all forage dietHarper et al., 1999 Micrometeorological Bred heifers 436 8.3 8.2 230 240

mass differenceMcCaughey et al., 1999 SF6 tracer gas First calf heifers 511 9.7 to 11.4 9.6 262 to 304 299Jones et al., 2011 Fourier transform infrared Dry cows 496 10.2 to 10.7 10.2 130 293

spectrophotometer Lact. cows 515 13.1 to 14.0 11.0 230 307Pavo-Zuckerman et al., 1999 SF6 tracer gas Cows 514 - 7.6 to 10.6 168 to 240 217 to 290DeRamus et al., 2003 SF6 tracer gas Cows 551 to 631 - 8.1 to 11.1 120 to 255 204 to 275Pinares-Patino et al., 2003 SF6 tracer gas Cows 723 to 800 9.7 to 12.0 9.9 to 12.0 204 to 273 264 to 331McGinn et al., 2011 Point source dispersion Heifers 271 5.0 4.6 141 158DeRamus et al., 2003 SF6 tracer gas Heifers 295 to 400 7.8 to 9.9 7.5 to 9.6 125 to 176 175 to 240Pavo-Zuckerman et al., 1999 SF6 tracer gas Steers 333 - 7.1 140 to 201 210McCaughey et al., 1997 SF6 tracer gas Steers 398 13.2 to 14.9 8.2 173 to 219 236Boadi et al., 2002 SF6 tracer gas Steers 345 11.1 7.7 228 229Laubach et al., 2008 Micrometeorological Steers 325 - 6.1 161 209

mass differenceFinish cattle, high concentrate diet

Harper et al., 1999 Micrometeorological Bred heifers 436 9.1 9.1 70 82mass difference

Beauchemin and McGinn, 2005 Environmental chamber Heifers 452 6.8 6.8 62 62Boadi et al., 2004 SF6 tracer gas Steers 300 to 570 8.8 to 12.4 8.7 to 10.4 50 to 112 83 to 98Nkrumah et al., 2006 Calorimetric chamber Steers 508 9.6 to 11.6 10.1 90 to 123 91Cooprider et al., 2011 Enclosed pens Steers 600 9.9 9.9 136 to 1481 901Reported model predicted enteric emissions based upon the animals and feeding strategy used. Measured data included emissions from accumulated manure.


(68%) of GHG to the Angus beef system that included the stocker phase. When the stocker phase is used it con-

tributes the least (14%) of the 3 phases to total GHGs, while the feedlot phase contributed 17%. Transportation of cattle within the system contributed less than 1% of the total GHG emission. For the Angus beef production system that excluded the stocker phase, the cow-calf phase was again the largest contributor (72%) of total system GHG emission. The feedlot phase was respon-sible for 27% of total GHG. Following the standard pro-cedure of excluding biogenic CO2, the carbon footprint of the simulated Angus beef production system was 13.5

Table 5. Greenhouse gas emissions and carbon footprint for Angus beef production in California

Production component

Greenhouse gas emission per animal1

Gas,kg/ganimal/d

CO2e,2 kg/animal

Withoutbiogenic CO2,

kg/animal

Cow calfMethane 0.525 4,793 4,793Nitrous oxide 0.009 1,014 1,014Net biogenic carbon dioxide3 −2.45 −896 −Combustion carbon dioxide 0.093 34 34Secondary emissions 0.98 356 356Total4 5,301 6,197

StockerMethane 0.218 992 992Nitrous oxide 0.005 260 260Net biogenic carbon dioxide −1.077 −196 −Combustion carbon dioxide 0.033 6 6Secondary emissions 0.291 53 53Total5 1,115 1,311

FinishMethane 0.095 287 287Nitrous oxide 0.016 563 563Net biogenic carbon dioxide −7.40 -896 −Combustion carbon dioxide 0.116 14 14Secondary emissions 6.02 697 729Total6 713 1,593

Animal transport (full system)7 65 65Total production system8 7,392 9,416Farm gate carbon footprint9 10.6 ± 1.010 13.5 ± 1.2

kg CO2e/kg SBWkg CO2e/kg HCW11 17.7 ± 1.6 22.6 ± 2.01Expressed per animal exported to the next phase (i.e., for the cow calf

this is the number of calves produced, stockers are the number of stockers produced and fi nished are the number of animals fi nished). The total for each phase is adjusted by the mortality of the succeeding phases to obtain the total for the production system per fi nished animal.

2Converted using global warming indexes of 25 units of carbon dioxide equivalent (CO2e) per unit of methane and 298 units of CO2e per unit of nitrous oxide.

3Net exchange of CO2 with the atmosphere including, assimilation in crop and animal growth, respiration emission from animals and manure and as-similated CO2 converted to CH4.

4Includes the cows, calves, replacement heifers, and bulls for 365 d.5Based on a 182 d feeding period.6Based on a 121 d feeding period.7Includes emissions from fuel combustion and that associated with produc-

tion of the fuel and truck.8Total GHG in CO2e for the entire production systems adjusted per fi n-

ished animal.9Shrunk body weight (SBW) includes cull cows and bulls sold (125 kg

SBW per fi nished animal).10The uncertainty in full system predictions is determined from the esti-

mated uncertainties of each system component.11Hot carcass weight (HCW) determined as 62% of SBW for the steer and

50% SBW for the cow.

Table 6. Greenhouse gas emissions and carbon footprint for Angus beef production from weaning directly into the feedlot (without stocker phase) in California



Gas, kg/animal/d

CO2e,2kg/animal

Without biogenic CO2,

kg/animal

Cow calfMethane 0.525 4,793 4,793Nitrous oxide 0.009 1,014 1,014Net biogenic carbon dioxide3 −2.45 −896 −Combustion carbon dioxide 0.093 34 34Secondary emissions 0.98 356 356Total 4 5,301 6,197

FinishMethane 0.103 547 547Nitrous oxide 0.009 517 517Net biogenic carbon dioxide −7.099 −1,505 −Combustion carbon dioxide 0.108 23 23Secondary emissions 6.108 1,295 1,295Total 5 877 2,382

Animal transport (full system)6 57 57Total production system 6,410 8,841Farm gate carbon footprint 9.2 ± 0.97 12.7 ± 1.2

kg CO2e/kg SBW8

kg CO2e/kg HCW9 15.4 ± 1.5 21.2 ± 2.01Expressed per animal exported to the next phase, i.e. for the cow calf

this is the number of calves produced, stockers are the number of stockers produced and fi nished are the number of animals fi nished. The total for each phase is adjusted by the mortality of the succeeding phases to obtain the total for the production system per fi nished animal.


3Net exchange of CO2 with the atmosphere including, assimilation in crop and animal growth, respiration emission from animals and manure and as-similated CO2 converted to CH4.

4Includes the cows, calves, replacement heifers, and bulls for 365 d.5Based on a 212 d feeding period.6Includes emissions from fuel combustion and that associated with produc-

tion of the fuel and truck.7The uncertainty in full system predictions is determined from the esti-

mated uncertainties of each system component.8Shrunk body weight (SBW) includes cull cows and bulls sold (125 kg

SBW per fi nished animal).9Hot carcass weight (HCW) determined as 62% of SBW for the steer and

50% SBW for the cow.


± 1.2 kg CO2e/kg SBW or 22.6 ± 2.0 kg CO2e/kg HCW (Table 5). For the Angus beef production system without the stocker phase the carbon footprint was 12.7 ± 1.2 kg CO2e/kg SBW or 21.2 ± 2.0 kg CO2e/kg HCW (Table 6).

For the Holstein beef production system (Table 7), the majority of the emissions from the cow were account-ed for in dairy production, rather than beef production. The calf entered the system with 112 kg of CO2e, which accounted for the environmental cost of calf production while a part of the dairy production system. In the Hol-

stein beef system, the largest contribution (91%) of GHG emission was from the feedlot. The calf-ranch contributed 7% of the total system GHG, and transport contributed 2%. The carbon footprint for the Holstein system was 6.3 ± 0.9 kg CO2e/kg SBW or 10.7 ± 1.4 CO2e/kg HCW (Ta-ble 7). Including biogenic CO2 essentially provides a sys-tem expansion approach for allocating exported manure C from the feedlot phase to other production systems that utilize feedlot manure. This removes the manure C from this phase of our simulated beef production system and places it in the carbon footprint of the production system using the manure.

Biogenic CO2 represents the net exchange of CO2 with the atmosphere including, assimilation in crop and animal growth, respiration emission from animals, plants, soil and manure, and assimilated CO2 that is converted and emitted as CH4. Although standard LCA procedures ignore these sources, including them provides a more ac-curate assessment of the farm gate footprint when large amounts of carbon are exchanged with other production systems or subsequent parts of the beef production sys-tem. Including biogenic CO2 essentially provides a sys-tem-expansion approach for allocating exported manure C from the feedlot manure. This removes the manure C from this phase of our simulated beef production system and places it in the C footprint of the production system using the manure. For the Angus beef system without the stocker phase, including biogenic CO2 reduced the foot-print by 27% (Table 6). For the Holstein beef system, the carbon footprint was decreased by 58% by including bio-genic CO2 (Table 7).

The handling of manure was the major factor con-tributing to the difference between including and exclud-ing biogenic CO2. California feedlots sell the majority of their manure to other farms as fertilizer. Therefore, the reported carbon footprint values were calculated with 100% of the manure from feedlots exported. This assumed that the manure C became associated with the carbon footprint of another production system. To fur-ther evaluate this effect, we included additional simula-tions where all manure remained in the beef production systems. This implies that all manure goes back to land used to produce feed for beef animals. Carbon footprints were 19.5±1.7, 18.3±1.7, and 9.6±1.3 kg CO2e/kg HCW for the Angus beef system including the stocker phase, the Angus beef system without stockers, and the Hol-stein beef production system, respectively (Figure 2). This was a 10 to 14% decrease in the footprint com-pared to not considering biogenic CO2. The remaining difference includes CO2 assimilated in the C content of the animal and that assimilated in the C released as CH4.

Simulated NH3 emissions from the full production systems varied from 41.0 ± 5.4 to 49.8 ± 7.2 kg per fi n-ished animal (98 ± 13 to 141 ± 27 g/kg HCW) with the

Table 7. Greenhouse gas emissions and carbon footprint for Holstein beef production in California



Gas, kg/animal/d

CO2e,2kg/animal

Without biogenic CO2,

kg/animal

Calf RanchMethane 0.070 105 105Nitrous oxide 0 0 0Net biogenic carbon dioxide3 0.860 104 −Combustion carbon dioxide 0.030 4 4Secondary emissions4 0.930 113 113Total 5 326 222

FeedlotMethane 0.087 822 822Nitrous oxide 0.004 573 573Net biogenic carbon dioxide −6.14 −2,175 −Combustion carbon dioxide 0.098 33 33Secondary emissions 6.84 1,867 1,866Total 6 1,120 3,295

Animal transport (full system)7 73 73Total production system 1,534 3,601Farm gate carbon footprint 2.7 ± 0.48 6.3 ± 0.9

kg/CO2e kg SBW9

kg/CO2e kg HCW10 4.5 ± 0.6 10.7 ± 1.41Expressed per animal exported to the next phase (i.e., for the cow calf

this is the number of calves produced, stockers are the number of stockers produced and fi nished are the number of animals fi nished). The total for each phase is adjusted by the mortality of the succeeding phases to obtain the total for the production system per fi nished animal.


3Net exchange of CO2 with the atmosphere including, assimilation in crop growth and respiration emission from animals and manure.

4Includes 112 kg of CO2e from the production of the calf in the dairy in-dustry.

5Based on a 121 day feeding period.6Based on a 334 day feeding period.7Includes emissions from fuel combustion and that associated with produc-

tion of the fuel and truck.8The uncertainty in full system predictions is determined from the esti-

mated uncertainties of each system component.9Total GHG in CO2e for the entire production systems adjusted per fi n-

ished animal.10Hot carcass weight (HCW) determined as 59% of SBW.


lowest emission from the Angus system that included the stocker phase (Table 8). In the Angus beef produc-tion system, total NH3 emission increased by 21% when a stocker phase was not used as compared to that when the stocker phase was included (Table 8). In the Angus beef system that included the stocker phase, the cow-calf phase emitted 48% of the total. The feedlot phase emit-ted 44% with 8% from the stocker phase. In the Angus beef system that excluded the stocker phase, the largest contributor was the feedlot phase emitting 61%, while the cow-calf was only responsible for 39% of the total NH3 emission. In the Holstein beef production system, the feedlot phase emitted 96% of the total, while the calf ranch emitted 4%.

DISCUSSION

In the United States, CH4 from enteric fermentation in ruminant livestock species accounts for approximately 70% of GHG from livestock contributing 139 Tg CO2e/yr (USEPA, 2009). Beef cattle are the largest contributor of enteric CH4 producing nearly 72% of ruminant spe-cies (USEPA, 2009). This large contribution is due to the beef cattle population of approximately 90 million animals. The beef cattle industry spans the continental United States and is comprised of various management systems and farm and ranch sizes making the industry

diverse and complex. The complexity of this diverse in-dustry further complicates GHG emission reporting and suggests that individual life cycle evaluations of spe-cifi c regional beef production should be conducted to determine the true environmental impact of beef cattle. Beef cattle production in California differs greatly from production in the Midwestern states and thus must be studied independently to better understand the environ-mental impact of the California cattle industry.

Often, reports that attempt to quantify GHG contribu-tions from animal agriculture differ by region or scope and therefore conclude with dissimilar results. In 2006, the United Nations Food and Agriculture Organization (UN FAO) published a report titled “Livestock’s Long Shadow” estimating that livestock are responsible for ap-proximately 7,100 Tg CO2e/yr or 18% of the world’s an-thropogenic GHGs, which is a larger portion than all trans-portation (Steinfeld et al., 2006). In contrast, the USEPA estimates that agriculture contributes 413.1 Tg CO2e/yr or 5.8% of anthropogenic GHGs, and of that, livestock are responsible for 198 Tg CO2e/yr or 3.4% (USEPA et al., 2009; Pitesky et al., 2009). Although both reports use holistic LCA tools, the conclusions are different due to the regional inputs used. Therefore, it is important to evalu-ate the environmental impact of livestock systems within their specifi c geographical region of production.

Using input parameters characteristic of Califor-nia beef production, our LCA estimated that the carbon footprint of Angus beef production systems range from 12.7 to 13.5 kg CO2e/kg SBW or 21.2 to 22.6 kg CO2e/kg HCW. Our results are similar to the results from an LCA conducted on a simulated U.S. herd in 2003, with a reported carbon footprint ranging from 11.0 to 13.0 kg CO2e/kg SBW (Johnson et al., 2003). An LCA con-ducted on a simulated Midwestern herd reported results slightly higher at 14.8 kg CO2e/SBW (Pelletier et al., 2010), and this LCA did not include resource use and the secondary emissions associated with the production and maintenance of farm and ranch inputs. In comparison to a simulated Alberta beef herd with a carbon footprint of 22 kg CO2e/kg HCW (Beauchemin et al., 2010), our estimated footprint is also similar. Casey and Holden (2005) reported similar annual carbon footprints for an Irish suckler-beef production system at 13.0 kg CO2e/kg SBW; however, with a 2-yr grass fed fattening period per animal the carbon footprint would likely double. An LCA conducted in Japan concluded that the carbon foot-print was 36.4 kg CO2e/kg HCW (Ogino et al., 2007) which is much greater than our reported carbon foot-prints. This large difference is expected because 40% of the feed for the Japanese beef system is shipped from the U.S. The relatively small differences in carbon footprint estimates from different reports are due to differences in assumptions, approaches, and algorithms used as well

Table 8. Carbon footprint and ammonia emissions of each phase of beef production in California expressed per unit of fi nished HCW

Production system

Farm gate carbon footprint,

kg CO2e/kg HCW1

Ammonia emission

kg/animal g/kg HCW

Conventional AngusCow-calf 15.4 19.6 47Stocker 3.2 3.5 8Feedlot 3.8 17.9 43Feedlot, without 5.7 30.2 73 stocker phaseAnimal transport, 0.2 full systemTotal, with 22.6 ± 2.02 41.0 ± 5.4 98 ± 13 stocker phaseTotal, without 21.3 ± 2.0 49.8 ± 7.2 120 ± 17 stocker phase

HolsteinCalf ranch 0.7 2.0 6Feedlot 9.8 45.7 135Animal transport, 0.2 full systemTotal 10.7 ± 1.4 47.7 ± 9.0 141 ± 271Refer to tables 4-6 for calculations of CO2e and HCW.2The uncertainty in full system predictions is determined from the esti-

mated uncertainties of each system component.


as difference in climate and management. Often direct LCA comparison is diffi cult for these reasons.

The carbon footprint for the California Holstein beef production system was 6.3 kg CO2/kg SBW and 10.7 kg CO2/kg HCW. In comparison to the Angus beef produc-tion system, the Holstein system contributes 50 to 63% less GHG. This is because the breeding stock is not part of the Holstein beef system, but rather a part of the dairy production system. Therefore, emissions from the cow are primarily accounted for in the carbon footprint of milk rather than beef. The carbon footprint reported by Rotz et al. (2010) for 2 California dairy herds ranged from 0.47 to 0.57 kg CO2e/kg energy corrected milk (ECM). It is important that emissions are not double accounted from separate and yet interlinked animal production systems, such as dairy and Holstein beef production. In contrast to the Angus beef production system, the feedlot phase is responsible for the majority of GHG emitted from the Holstein beef production system.

Our LCA results suggest that within the California Angus beef production system, the cow-calf phase con-tributes 68 to 72% of total GHG emissions. The stocker phase contributes 14% and the Angus feedlot is respon-sible for 17 to 27% of total GHG depending on man-agement strategy. Our simulations suggest that the cow-calf phase contributes somewhat less of the total GHG emissions compared with the work of Beauchemin et al. (2010) who reported that the Canadian cow-calf system was responsible for 80% of total GHG emissions. John-son et al. (2003) reported that the cow-calf phase con-tributed 75% of the total GHG emission. Although there are minor differences, these studies all confi rm that the cow calf phase is the major contributor to the overall carbon footprint of traditional beef production.

The cow-calf phase differs from the stocker and feed-lot phases because of the maintenance of the large number of breeding stock required to produce the calves. Addi-tionally, replacement heifers are maintained for 2 yr be-fore producing their fi rst calf. The cow-calf phase in Cali-fornia is also maintained on rangeland or pasture where the high roughage diet increases CH4 production through enteric fermentation. As such, the large majority of GHG contribution from the beef production system is from the beef cows. Little research has pursued viable GHG miti-gation strategies for the cow-calf sector of the industry; however, emission reductions in this sector would likely have a greater impact on total GHG emission than mitiga-tion strategies for the stocker or feedlot phases.

In contrast to general perception, the feedlot phase contributes a relatively small amount of GHG to the beef production system (17 to 27%). This is due to the shorter duration of time that the cattle spend in the feedlot and to the high grain diet that is fed, which reduces enteric CH4 production. Overall, the contribution of GHG emissions

from the feedlot is primarily dependent on management of manure and the time duration that cattle spend in the feedlot.

Including a stocker phase in the California beef production system increases GHG emissions by 6%. Johnson et al. (2003) reported similar result with the stocker phase increasing total GHG emissions by 4%. The increased GHG emission from the stocker phase is the result of grazing rangeland forage, which increas-es CH4 emissions from enteric fermentation compared with feedlot diets. Additionally, during the stocker pe-riod, cattle have reduced ADG than animals that enter the feedlot directly after weaning. Thus, they may be harvested at a later age, producing more enteric CH4 and manure during their lifetime. However, grazing animals as stocker cattle reduces NH3 emissions from the beef production system by 18%, and this strategy may reduce production costs by requiring less high-priced corn for feed (Stackhouse-Lawson et al., 2012).

To date, no other studies have determined NH3 emis-sions from the entire beef production system, so our full system predictions cannot be compared with other data. The majority of measured NH3 fi eld emission data is re-ported from feedlots in desert climates. Measured NH3 emissions from Texas feedlots report daily emissions ranging from 85 to 150 g/animal (Flesch et al., 2007; Todd et al., 2008; Rhoades et al., 2010). Our simulated Angus feedlot NH3 emissions are reported for the entire time the animal remained in the feedlot and increased with DOF ranging from 17.9 to 30.2 kg/animal and 43 to 73 g/kg HCW, depending if the stocker phase was used. Simulated Holstein feedlot NH3 emissions were 47.7 kg/animal and 141 g/kg HCW. Thus, the average daily emissions from the simulated California feedlots were 148, 142, and 137 g/animal for the Angus feedlot system with and without the stocker phases and the Hol-stein feedlot, respectively. Our simulated results are well within measured NH3 emission ranges for beef feedlots. Differences can be expected due to climate differences between southern California and the Texas panhandle. Also, our simulation gives average values over many years of weather; whereas, the measured values repre-sent shorter periods of time.

Accurate representation of CO2 is challenging when conducting a partial LCA that stops at the farm-gate. For this type of LCA, the full C balance including consump-tion or use of the products is not included. This effectively shifts a portion of the carbon footprint from the consumer or user to the producer. To obtain a complete farm gate assessment, biogenic CO2 sources and sinks should be included. Biogenic CO2 represents the net exchange of CO2 with the atmosphere including, assimilation in crop and animal growth, respiration emission from animals, plants, soil and manure, and assimilated CO2 that is


converted and emitted as CH4. Including biogenic CO2 reduces the carbon footprint of the simulated California Angus beef production system by 22 and 27% with and without the stocker phases, respectively. For the Holstein beef production system, the carbon footprint is decreased by 58% by including biogenic CO2. Including biogenic CO2, along with a complete carbon balance, ensures that mitigation of GHG will be applied at points within the production system that give the greatest benefi t.

The effect of shifting the consumer’s portion of the total footprint onto the producer is relatively small since only minor amounts of C are assimilated in the meat. In California beef systems though, the assumption on manure use can have a substantial effect on the differ-ence between including and excluding biogenic CO2. California feedlots sell the majority of their manure to other farms as fertilizer. However, unless biogenic CO2 is accounted for in the system or manure is treated as a co-product in the LCA, the effect of exporting manure for use in another farming system is ignored, and the C assimilated in the manure remains as a component of the beef production system. Since this C is assumed to ultimately end up as CO2 in the atmosphere, the CO2 emission remains in the footprint of beef production. In California, the manure C that is used for crop production unrelated to livestock production should become associ-ated with the carbon footprint of that production system. This assumption on the end use of the manure results in a 10 to 14% decrease in the carbon footprint of beef production.

Our LCA demonstrates that the majority of GHG emissions from the California beef production system emanate from the cow-calf phase. Therefore, the cow-calf phase may deserve increased attention in the de-

velopment of effective mitigation strategies that most signifi cantly reduce GHG emissions without compro-mising productivity. Ammonia emissions are highest in the cow-calf and feedlot phases and increase with DOF. If the reduction of NH3 emissions becomes necessary, the most opportune target is the feedlot through changes in diet and manure handling. This regional level LCA provides CA beef producers with a carbon footprint bench mark unique to their management techniques and specifi c climatic region.

LITERATURE CITEDBeauchemin, K., A., H. H. Janzen, S. M. Little, T. A. McAllister, and S.

M. McGinn. 2010. Life cycle assessment of greenhouse gas emis-sions from beef production in western Canada: A case study. Agr. Syst. 103:371–379.

Beauchemin, K. A., and S. M. McGinn. 2005. Methane emissions from feedlot cattle fed barley or corn diets. J. Anim. Sci. 83:653–661.

Boadi, D. A. 2002. Effects of grain supplementation on methane produc-tion of grazing steers using the sulphur (SF6) tracer gas technique. Can. J. Anim. Sci. 82:151–157.

Boadi, D. A. 2004. Effect of low and high forage diet on enteric and ma-nure pack greenhouse gas emissions from a feedlot. Can. J. Anim. Sci. 84:445–453.

California Department of Food and Agriculture. 2008. California agri-culture production statistics 2009-2010: Livestock and dairy. Ac-cessed March 15, 2011. www.cdfa.ca.gov/statistics/PDFs/...2010-2011/5Livestock08_WEB.pdf

Casey, J. W., and N. M. Holden. 2005. The relationship between green-house gas emissions and the intensity of milk production in Ireland. J. Environ. Qual. 34:429–436.

Chianese, D. S., C. A. Rotz, and T. L. Richard. 2009a. Whole farm greenhouse gas emissions: A review with application to a Pennsyl-vania dairy farm. Appl. Eng. Agric. 25:431–442.

Chianese, D. S., C. A. Rotz, and T. L. Richard. 2009b. Simulation of carbon dioxide emissions from dairy farms to assess greenhouse gas reduction strategies. Trans. ASABE 52:1301–1312.

Chianese, D. S., C. A. Rotz, and T. L. Richard. 2009c. Simulation of methane emissions from dairy farms to assess greenhouse gas re-duction strategies. Trans. ASABE 52:1313–1323.

Chianese, D. S., C. A. Rotz, and T. L. Richard. 2009d. Simulation of nitrous oxide emissions from dairy farms to assess greenhouse gas reduction strategies. Trans. ASABE. 52:1325–1335.

Cooprider, K. L., F. M. Mitloehner, T. R. Famula, K. Kebreab, Y. Zhao, and A. L. Van Eenennaam. 2011. Feedlot effi ciency implications on greenhouse gas emissions and sustainability. J. Anim. Sci. 89: 2643–56.

DeRamus, H. A., T. C. Clement, D. D. Giampola, and P. C. Dickison. 2003. Methane emissions of beef cattle on forages: Effi ciency of grazing management systems. J. Environ. Qual. 32:269–277.

Flesch, T. K., J. D. Wilson, L. A. Harper, R. W. Todd, and N. A. Cole. 2007. Determining ammonia emissions from a cattle feedlot with an inverse dispersion technique. Agr. Forest Meteoro. 144:139–155.

Fox, D. G, L. O. Tedeschi, T. P. Tylutki, J. B. Russell, M. E. Van Am-burgh, L. E. Chase, A. N. Pell, and T. R. Overton. 2004. The Cornell net carbohydrate and protein system model for evaluating herd nu-trition and nutrient excretion. Anim. Feed Sci. Technol. 112:29–78.

Hamilton, S. W., E. J. DePeters, J. A. McGarvey, J. Lathrop, and F. M. Mitloehner. 2010. Greenhouse gas, animal performance, and bac-terial population structure responses to dietary monensin to dairy

Figure 2. Simulated carbon footprints of California beef production systems with and without considering biogenic CO2 and without considering the export of feedlot manure carbon from the production system. Standard practice is to exclude biogenic sources of CO2 and when biogenic CO2 is not considered, the export of manure has no impact on the carbon footprint. Does not include manure in the system; when biogenic CO2 is not considered manure has no impact on the production system.


cows. J. Environ. Qual. 39:1–9.Harper, L. A., O. T. Denmead, J. R. Freney, and F. M. Byers. 1999. Di-

rect measurements of methane emissions from grazing and feedlot cattle. J. Anim. Sci. 77:1392–1401.

International Panel on Climate Change (IPCC). 2001. Climate change 2001: The scientifi c basis. Contribution of working group 1 to the third assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

International Panel on Climate Change (IPCC). 2006a. Guidelines for national greenhouse inventories. Vol. 4: Agriculture, forestry and other land use. Accessed April 2, 2012. http://www.ipcc-nggip.iges.or.jp/public/2006gl/vol4.html.

International Panel on Climate Change (IPCC). 2006b. Guidelines for national greenhouse gas inventories. Vol. 1: General guidance and reporting. Accessed April 2, 2012. www.ipcc-nggip.iges.or.jp/public/2006gl/vol1.html.

International Panel on Climate Change (IPCC). 2007. Climate change 2007: The physical science basis. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Chapter 2, Changes in atmosphere constituents and in radioactive forcing. Accessed November 15, 2009. http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter2.pdf.

Janzen, H. H., D. A. Angers, M. Boehm, M. Bolinder, R. L. Desjardins, J. A. Dyer, B. H. Ellert, D. J. Gibb, E. G. Gregorich, B. L. Hel-gason, R. Lemke, D. Massé, S. M. McGinn, T. A. McAllister, N. Newlands, E. Pattey, P. Rochette, W. Smith, A. J. VandenBygaart, and H. Wang. 2006. A proposed approach to estimate and reduce net greenhouse gas emissions from whole farms. Can. J. Soil Sci. 86:401–418.

Johnson, D. E., H. W. Phetteplace, A. F. Seidl, U. A. Schneider, and B. A. McCarl. 2003. Management variations for us beef production sys-tems: Effects of greenhouse gas emissions and profi tability. Pages 953–961 in Proc. 3rd Int. Methane and Nitrous Oxide Mitigation Conf., Bejing China.

Johnson, K. A., and D. E. Johnson. 1995. Methane emissions from cat-tle. J. Anim. Sci. 73:2483–2492.

Jones, F. M., F. A. Phillips, T. Naylor, and N. B. Mercer. 2011. Methane emissions for Nagus beef cows selected for divergent residual feed intake. Anim. Feed Sci. Tech. 166-167:302–307.

Laubach, J., F. M. Kelliher, T. W. Knight, H. Clark, G. Molano, and A. Cavanagh. 2008. Methane emissions from beef cattle: A com-parison of paddock and animal scale measurements. Aust. J. Exp. Agric. 48:132–137.

McCaughey, W. P., K. Wittenberg, and D. Corrigan. 1997. Methane pro-duction of steers on pasture. Can. J. Anim. Sci. 77: 519–524.

McCaughey, W. P., K. Wittenberg, and D. Corrigan. 1999. Impact of pasture type on methane production by lactating beef cows. Can. J. Anim. Sci. 79:221–226.

McGinn, S. M., D. Turner, N. Tomkins, E. Charmley, G. Bishop-Hurley, and D. Chen. 2011. Methane emissions from grazing cattle using point-source dispersion. J. Environ. Qual. 40:22–27.

Mills, J. A. N., E. Kebreab, C. M. Yates, L. A. Crompton, S. B. Cam-mell, M. S. Dhanoa, R. E. Agnew, and J. France. 2003. Alternative approaches to predicting methane emissions from dairy cows. J. Anim. Sci. 81:3141–3150.

Nkrumah, J. D., E. K. Okine, G. W. Mathison, K. Schmid, C. Li, J. A. Basarab, M. A. Price, Z. Wang, and S. S. Moore. 2006. Relation-ship of feedlot feed effi ciency, performance, and feeding behavior with metabolic rate, methane production, and energy portioning in feedlot cattle. J. Anim. Sci. 84:145–153.

Ogino, A., H. Orito, K. Shimada, and H. Hirooka. 2007. Evaluating en-vironmental impacts of Japanese beef cow-calf system by the life

cycle assessment method. Anim. Sci. 78:424–432.Pavao-Zuckerman, M. A., J. C. Waller, T. Ingle, and H. A. Fribourg.

1999. Methane emissions of beef cattle grazing tall fescue pastures at three levels of endophyte infestation. J. Environ. Qual. 28:1963–1969.

Pelletier, N., R. Pirog, and R. Rasmussen. 2010. Comparative life cycle environmental impacts of three beef production strategies in the upper Midwestern United States. Agr. Syst. 103:380–389.

Penn State. 2011. Penn State Agronomy Guide. Cooperative Extension, The Pennsylvania State University. Available at: http://extension.psu.edu/agronomy-guide. Accessed 9 September 2011.

Pinares-Patiño, C. S., P. D’Hour, J. P. Jouany, and C. Martin. 2003. Ef-fects of stocking rate on methane and carbon dioxide emissions from grazing cattle. Agric. Ecosyst. Environ. 121:30–46.

Pitesky, M. E., K. R. Stackhouse, and F. M. Mitloehner. 2009. Clearing the air: Livestock’s contribution to climate change. Adv. Agron. 103:1–40.

Ramirez, C. A., M. Patel, and K. Blok. 2006. From fl uid milk to powder: energy use and energy effi ciency in the European dairy industry. Energy 31:1984–2004.

Rhoades, M. B., D. B. Parker, N. A. Cole, R. W. Todd, E. A. Caraway, B. W. Auvermann, D. R. Topliff, and G. L. Schuster. 2010. Con-tinuous ammonia emission measurements from a commercial beef feedyard in Texas. Trans. ASABE 53:1823–1831.

Rotz, C. A., D. R. Buckmaster, and J. W. Comerford. 2005. A beef herd model for simulation feed intake, animal performance, and manure excretion in farm systems. J. Anim. Sci. 38:231–242.

Rotz, C. A., F. Montes, and D. S. Chianese. 2010. The carbon footprint of dairy production systems through partial life cycle assessment. J. Dairy Sci. 93:1266–1282.

Rotz, C. A., M. S. Corson, D. S. Chianese, F. Montes, S. D. Hafner, R. Jarvis, and C. U. Coiner. 2011. Integrated farm system model: Reference manual, version 3.4. Pasture Systems and Watershed Management Research Unit, USDA Agricultural Research Ser-vice, University Park, PA.

Stackhouse, K. R., Y. Pan, Y. Zhao, and F. M. Mitloehner. 2011. Green-house gas and alcohol emissions from feedlot steers and calves. J. Environ. Qual. 40:899–906.

Stackhouse-Lawson, K. R., C. A. Rotz, J. W. Oltjen, and F. M. Mit-loehner. 2012. Growth promoting technology reduces the carbon footprint, ammonia emissions, and cost of California beef produc-tion systems. J. Anim. Sci. In press

Steinfeld, H., P. Gerber, T. Wassenaar, V. Castel, M. Rosales, and C. De Haan. 2006. Livestock’s long shadow: Environmental issues and options Food and agriculture Organization of the United Nations, Rome. ftp://ftp.fao.org/docrep/fao/010/a0701e/a0701e00.pdf

Stewart, A. A., S. M. Little, K. H. Ominski, K. M. Wittenberg, and H. H. Janzen. 2009. Evaluating greenhouse gas mitigation practices in livestock systems: An illustration of whole system approach J. Agr. Sci. 147:367–382.

Todd, R. W., N. A. Cole, N. R. Clark, T. K. Flesch, L. A. Harper, and B. H. Baek. 2008. Ammonia emissions from a beef cattle feedyard on the southern high plains. Atmos. Environ. 42: 6797–6805.

Wang, M. 2007. GREET version 1.8a. Argonne National Laboratory, Argonne, IL. Accessed January, 2008. http://www.transportation.anl.gov/software/GREET/.

U.S. Environmental Protection Agency (USEPA). 2009. Inventory of U.S. Greenhouse gas emissions and sinks: 1990-2007. Accessed December 10, 2010. http://epa.gov/climatechange/emissions/downloads09/GHG2007entire_report-508.pdf

USDA-ARS. 2011. Integrated Farm System Model. Pasture Systems and Watershed Management Research Unit, USDA Agricultural Research Service, University Park, PA. Accessed May 12, 2011. https://www.ars.usda.gov/Main/docs.htm?docid=8519.

carbon footprint and ammonia emissions of california beef ... · key words: ammonia, beef cattle,...

Documents