2. Impact Assessment: Methane Emissions

2.1 Cattle Characteristics
2.2 Feed Characteristics
2.3 Enteric Fermentation Emissions
2.4 Manure System Characteristics
2.5 Manure Facility Emissions
2.6 Total Methane Emissions from Livestock

This section presents the estimates of methane emissions from livestock due to both enteric fermentation and manure. These estimates were developed from the datasets provided as part of the project, in conjunction with emissions inventory methods presented in IPCC/OECD (1994) and data developed for USEPA (1994).

The principal analyses required to estimate methane emissions from these sources involves characterizing the livestock, including their feed intakes and the manner in which they are managed. Because cattle account for the majority of methane emissions, this analysis focuses on the proper description of these animals. Estimates of emissions from other livestock are based on factors presented in IPCC/OECD (1994). This section is organized as follows:

· Section 0 presents the cattle characteristics used to estimate enteric fermentation methane emissions.

· Section 0 presents the feed characteristics used to estimate enteric fermentation methane emissions.

· Section 0 presents the emissions factors and emissions estimates developed for cattle as well as the other domesticated livestock.

· Section 0 presents the characterization of livestock manure management systems for purposes of estimating methane emissions from manure.

· Section 0 presents the emissions factors and emissions estimates for manure management facilities.

· Section 0 presents the total emissions estimates and compares them with previously published estimates.

2.1 Cattle Characteristics

The purpose of this portion of the analysis is to develop the characteristics of cattle in each production system in each region that are required to estimate methane emissions from enteric fermentation. Based on the emissions algorithm presented in IPCC/OECD 1994 and USEPA (1994), methane emissions from a cattle population are estimated by first developing emissions factors for each animal type within the population, and then multiplying the emissions factors by the relevant number of each animal type in the overall population. Using the cattle population statistics provided for this investigation, two types of cattle were defined: dairy cows and other cattle. Dairy cows are defined as those cows giving milk that enters into the dairy production statistics. Most commonly, this is the milk that enters into commerce. Dairy cows do not include cows whose milk is completely consumed by calves (offspring) and/or by subsistence farmers. Other cattle includes all other cattle that are not dairy cows, including: young; beef cattle; draft animals; and bulls. Exhibits 1 and 2 summarize the cattle populations by animal type, region, and production system.

To estimate methane emissions from enteric fermentation, the feed intake of each animal type is required. Feed intake is estimated based on estimated “net energy” intake requirements, which are driven by the following production characteristics: weight (kg); feeding situation (stall fed, grazing, grazing over large areas); milk production (kg per day); growth rate (kg per day); pregnancy; and work performed (hours per day).

The following data were used to develop the characteristics needed to estimate net energy intake:

· Detailed production characteristics of cattle in OECD were taken from USEPA (1994). These data include all the necessary characteristics for estimating net energy intake requirements for cattle in this region.

· The net energy intake requirements for milk production were estimated for dairy cows using the data on milk production per cow per year in each region and production system.

· The net energy required for each feeding situation (stall fed, grazing, and grazing large areas) was estimated as a percentage of the net energy required for maintenance as described in USEPA (1994): no additional net energy required for stall fed; 17% additional required for grazing; and 37% additional required for grazing large areas. The basis for defining the feeding situation is described below.

· For the remainder of the net energy requirement (“other requirements”), the “Animal Unit Weighting Factor” (AUWF) was used to scale the intake requirements estimated for OECD to each of the other regions. The AUWF, defined for each region, converts animal population data to a common “animal unit” basis, principally taking into account feed requirements. Therefore, using the AUWF for scaling feed intake requirements makes the estimates consistent with the method used to estimate cattle populations across regions.²

² Scaling energy intake requirements with the AUWF is preferable to scaling the weights of the animals because the AUWF is a relative measure of feed energy intake required. Noting that most of the feed energy intake is associated with maintenance energy requirements (USEPA, 1994) which is proportional to weight^0.75, scaling weight using the AUWF would have distorted the relative feed intakes required among regions. One of the drawbacks of the AUWF is that it is not differentiated by production system, and only reflects differences by region. This lack of data by production system adds uncertainty to the estimates of methane emissions by production system.

· The “other” net energy requirement was divided into intake required for growth and intake required for the non-growth functions (maintenance, work, pregnancy) so that the different efficiencies of utilization of net energy for these functions could be reflected (see, e.g., NRC, 1984 and NRC, 1989). Dairy cows are assumed to be adults, so no growth requirement is estimated for them. A relatively small fraction of net energy is associated with requirements for growth among other cattle outside the OECD, estimated at about 4.0 percent based on the estimates in USEPA (1994).

· Dairy Cows: All dairy cows in mixed farming agriculture production systems (irrigated and rainfed) are assumed to be stall fed. This assumption underestimates feed intake requirements for these animal because some dairy cows are grazed on pastures for at least a portion of their feed intake. However, given the need to milk the animals daily for most of the year, grazing distances are probably fairly short, so that stall feeding is a reasonable assumption. Dairy cows in grazing production systems are assumed to be in the grazing feeding regime.

· Other Cattle: The feeding situations of other cattle were determined by analyzing the animal densities of the livestock production systems. Low animal densities indicate that there is a large area over which the animals may graze and high animal densities indicate that grazing may be limited, leading to stall-feed conditions. To assign feeding situation values to each production system in each region, the following analysis was performed:³

³ All cattle in the landless production system (LLR) were assigned the stall-fed feeding situation.

a) Cattle per hectare: The number of cattle per hectare of grazing land was estimated for each production system for each region by multiplying the cattle population by the regional AUWF and dividing the product by the appropriate number of hectares of grazing land. In the grazing production systems, both the dairy cow and other cattle populations were used in the calculation. Because dairy cows were assigned stall-fed status in the mixed farming production systems, only the non-dairy cattle populations were used to estimate this figure in these production systems.

b) Situation Boundaries: Using the estimates of cattle per hectare of grazing land, boundary values were selected that define feeding situations as a function of cattle density. Exhibit 3 displays the cattle density estimates and the boundary values. As shown in the exhibit, all region/production systems with cattle densities below 0.12 AU/ha are assigned “grazing large areas.” This assignment is made almost exclusively to grazing production systems. The “grazing pastures” feeding situation is assigned to the region/production systems with cattle densities between 0.12 and 0.45 AU/ha. Above 0.45 AU/ha and less than 1.0 AU/ha the feeding situation is estimated as a mixture of pasture and stall-fed conditions. The mixture is estimated by interpolating between the two boundary values as follows:

% Grazing = 100 x (1.0 - D)/(1.0 - 0.45)

where D is the animal density estimate in AU/ha. Using this equation, the system is assigned 100 percent grazing at the boundary condition of 0.45 AU/ha, and 0 percent grazing at the boundary condition of 1.0 AU/ha. The portion that is stall fed is 100 percent minus the percent grazing. For systems with animals densities over 1.0 AU/ha, the stall-fed feeding situation is assigned.

c) Assign Feeding Situation: Using the boundary values, a feeding situation is assigned to each region/production system. Exhibits 4 and 5 summarize the portion of cattle in each feeding situation by region and production system. As shown in the exhibits, most of the cattle are estimated to be grazing pastures or are stall fed. Only 16 percent of the cattle are estimated to be grazing large areas. Most of the cattle grazing large areas are in the grazing production systems, whereas most of the cattle grazing pastures are estimated to be in rainfed, mixed farming production systems. Most of the cattle in the mixed farming systems are stall-fed, however.

2.2 Feed Characteristics

Feed characteristics define the cattle’s nutritional profiles. For purposes of estimating methane emissions from enteric fermentation, two feed characteristics are required:

· Digestibility: Feed digestibility is required to convert digestible energy to gross energy feed intake. The net energy intakes (estimated above) are converted to digestible energy using the method described in USEPA (1994).⁴ The digestible energy is converted to gross energy by dividing by the digestibility (e.g., 0.60 for 60 percent).

⁴ As described in USEPA (1994), net energy is converted to digestible energy based on estimates of the utilization efficiency of energy for basic metabolic functions. Energy used for growth is given a lower efficiency of utilization than energy used for other functions.

· Methane Yield: The methane yield defines the portion of gross energy that is converted to methane in the rumen. The methane yield is known to be influenced by feed quality and level of feeding, and generally ranges from 4.0 to 9.0 percent, with values of 6.0 percent being typical for many feeding situations around the world.

· Base Assumptions by Production System: A set of base assumptions was developed for each production system describing feed digestibility and methane yield. These assumptions were then modified to accommodate the characteristics of the feed resources in specific regions. The base assumptions are as follows:

-- Dairy Cows: Dairy cows are typically provided the best available feed resources. The base assumption for feed digestibility is 60 percent for all dairy cows in all production systems. Given that the dairy cows generally receive the best available nutrition, the base assumption for methane yield is 6.0 percent for all dairy cows in all regions.

-- Other Cattle: The digestibility of feed for other cattle, many of which graze, depends on the climate and availability of feed supplements, such as grain. The following base assumption values are used:

a) Temperate Climate: The temperate climate areas have the best grazing and other forage resources and grain available for cattle. The production systems in temperate areas have a digestibility value of 60 percent and a methane yield value of 6.0 percent.

b) Humid Climate: The humid climate areas have lower quality feed resources as compared to temperate areas, and tend to have less grain available for cattle. The production systems in humid areas have a digestibility of 55 percent and a methane yield of 6.0 percent.

c) Arid Climate: The arid climate areas have the poorest quality feed resources for cattle, with forage often being more fibrous and less digestible. The grazing and rainfed production systems in arid climates are assigned a digestibility value of 50 percent and a methane yield of 6.5 percent. In irrigated systems, the forage quality and grain availability are expected to be better than in rainfed systems. Therefore, in the irrigated mixed farming production system in arid climates, a digestibility figure of 55 percent and a methane yield value of 6.0 percent are used.

d) Landless Ruminants: The landless ruminant production system refers primarily to the feedlot fed cattle in North America, Europe, and the Commonwealth of Independent States (CIS). These large scale feeding systems typically include a substantial amount of grain feeding, and consequently a digestibility of 75 percent is used. In the OECD countries (primarily in the USA), the level of feeding is also very high, contributing to a relatively low methane yield of 3.5 percent. The methane yield for this production system in Eastern Europe and the CIS is estimated at 6.0 percent.

· Regional Adjustments: The base assumptions for digestibility and methane yield are modified to take into account regional differences in feed resources. Although a wide range of adjustments may be desirable, limitations on available data limit the number of adjustments that could be implemented. The adjustments are as follows:

-- Dairy Cows: The feed resources provided to dairy cows in OECD countries tend to be of higher quality than the feed resources elsewhere. Therefore, the digestibility of the feed is adjusted to 65 percent for dairy cows in this region. No adjustment is made to the methane yield value.

-- Other Cattle: Better than average feed resources are available in the OECD region, and poorer quality feed resources are typically found in Sub-Saharan Africa (SSA) and parts of Asia. The regional adjustments are as follows:

a) OECD: The feed digestibility was increased to 60 percent for the grazing production systems, reflecting the use of supplements. For the arid grazing production system, the methane yield was reduced from 6.5 percent to 6.0 percent to reflect this higher quality of feed available. In the mixed farming production system, the digestibility was increased to 65 percent to reflect the common use of supplements. The methane yield in the arid rainfed production system was reduced to 6.0 percent from 6.5 percent.

b) Sub-Saharan Africa: The feed resources in SSA are generally poorer than in other regions. The digestibility was reduced by 5.0 percent in temperate and humid production systems, and the methane yield was increased to reflect the lower-quality feed generally available in this region.⁵

⁵ The lower protein content of the feeds available in this region increases the methane yield value (see USEPA, 1994).

c) Asia: Crop byproducts are a common source of forage for cattle in Asia, and in particular on the Indian subcontinent where large numbers of cattle are kept. The digestibility estimate is reduced by 5.0 percent in the humid mixed farming rainfed production system, and the methane yield is increased to 6.5 percent.

2.3 Enteric Fermentation Emissions

Enteric fermentation emissions for cattle are estimated by first estimating emissions factors per head from the characteristics defined above, and then by multiplying the emissions factors by the relevant cattle populations. Separate emissions factors are estimated for each region/production system. For animals other than cattle, emissions factors are taken from USEPA (1994). Using such emissions factors is reasonable because: (1) emissions factors do not vary among regions and production systems for other livestock to the extent they vary for cattle; and (2) the emissions from other livestock are much smaller than the emissions from cattle.

Exhibits 9 and 10 summarize the emissions factors for dairy cows and other cattle respectively. As shown in the exhibits, the emissions factors for dairy cows are larger than the emissions factors for other cattle because they are larger and require feed intake for lactation. OECD and Other Developed countries have the highest emissions factors while Asia, West and North Africa, and Sub-Saharan Africa have the lowest. Among dairy cows, the temperate mixed farming production system has the highest emissions factors, reflecting higher levels of milk production.

Using the emissions factors in these exhibits, and the population data presented in Exhibits 1 and 2, total methane emissions from enteric fermentation from cattle are estimated at 53.9 million tons. As shown in Exhibit 11, CSA and OECD have the highest emissions, accounting for over 50 percent of the total estimated emissions. The temperate rainfed mixed farming production system accounts for over 25 percent of total emissions. Other cattle account for 40.4 million tons of emissions, or about 75 percent of the total. Exhibits 12 and 13 present the emissions estimates separately for dairy cows and other cattle.

Emissions factors from USEPA (1994) are used to estimate emissions from buffalo, sheep and goats. Total emissions from cattle, buffalo, sheep, and goats are estimated at about 72 million tons and are presented by region and production system in Exhibit 14. Emissions from cattle account for about 75 percent of total emissions. Primarily due to its large population of cattle and buffalo, Asia has the highest regional emissions. Exhibit 15 shows the cattle emissions by region and by production system in a graphical form. Exhibit 16 displays total emissions by region and production system in a graphical form.

As discussed above, emissions per unit of product can be used to identify opportunities for reducing methane emissions from enteric fermentation. For example, although OECD and EE+CIS have the highest total methane emissions from dairy cows, these regions have relatively low emissions per unit of product produced.⁶ The highest emissions per unit of product are found in the grazing systems in SSA, Asia, and CSA. The number of dairy cows and the emissions from these cows is relatively small, however. The largest population of dairy cows with relatively high emissions per unit of product are found in the mixed farming production system in humid regions of Central and South America (MRH in CSA).

⁶ Emissions per unit of milk produced were estimated as the dairy cow emissions divided by the milk production estimates. These estimates do not include the emissions associated with the young dairy cattle, principally replacements for dairy cows that are culled or die. Therefore, these emissions estimates per unit of milk produced do not include all the emissions associated with producing milk.

A similar analysis of emissions per unit of meat production is shown in Exhibit 18. These emissions estimates are estimated as total cattle emissions divided by total beef and veal production. Note that emissions from dairy cows are included in these estimates because dairy cows and their offspring contribute to meat production. Because meat is not produced from cattle in many parts of Asia, this indicator is not a valid assessment of production efficiency in this region. Instead, indicators of methane emissions per unit of draft power produced should be used. Nevertheless, Exhibit 18 indicates that production systems in CSA (C5 and C6) and Sub-Saharan Africa (S3, S2, and S5) may hold promise for opportunities to reducing emissions by reducing emissions per unit product.

2.4 Manure System Characteristics

The purpose of this portion of the analysis is to develop the characteristics of manure management systems in each region and production system needed for estimating methane emissions. These characteristics can be divided into two main parts:

· Manure Characteristics: describing the quantity and methane-producing potential of the manure.

· Manure Management Facilities: describing how the manure is handled and the extent to which the methane-producing potential is achieved.

· The total amount of manure produced per day on a dry matter basis is estimated as the dry matter feed intake (kg per day) times the non-digestible portion of the feed (100% - digestibility in percent). To estimate dry matter intake, the gross energy intake estimate for the enteric fermentation analysis was used along with a feed energy density of 18.45 MJ per kilogram (dry weight basis).

· The volatile solids (VS) portion of the dry matter was estimated using a figure of 8 percent for the ash content of the manure, so that VS (kg per day) equals total manure times 92 percent.

· Dairy Cows:

-- Developed Countries: 0.24 m³ CH₄ per kg VS
-- Developing Countries: 0.13 m³ CH₄ per kg VS

· Other Cattle:

-- Developed Countries: 0.17 m³ CH₄ per kg VS
-- Developing Countries: 0.10 m³ CH₄ per kg VS
-- Feedlot fed cattle in OECD: 0.33 m³ CH₄ per kg VS.⁷

⁷ The manure from feedlot fed cattle, particularly in the U.S., has a higher methane potential due to the type of feed consumed and the high level of feeding.

The portion of the methane potential that is actually realized is driven by how the manure is managed. Manure handled in anaerobic liquid-based systems, such as ponds and tanks, will achieve a higher portion of its potential than manure handled in a dry form. For purposes of this analysis, all manure from grazing cattle is assumed to be handled in a dry form. The portion of manure from stall-fed cattle handled in dry and liquid form was estimated from data collected by Safley et al. (1992) and is shown in Exhibit 19. As shown in the exhibit, the majority of the manure is estimated to be handled in a dry form, thereby limiting the extent to which the manure’s methane potential is realized. Because data that describe manure management practices systematically across regions are generally lacking, this characterization is particularly uncertain. Safley et al. Indicate that additional effort is warranted to better characterize these practices.

The portion of the methane emissions potential that is achieved for each type of manure management facility is presented in USEPA (1994) and is listed by region in Exhibit 19. These values, referred to as the “methane conversion factors” (MCFs) are on the order of 1.0 to 2.0 percent for manure managed in a dry form. The higher values are appropriate for warmer and more humid climates. MCFs for liquid manure management facilities range from 10 to 90 percent depending on the facility type and climate (USEPA, 1994). Exhibit 19 lists the MCFs selected by region for cattle manure managed in a liquid form. The high MCF value of 65 percent selected for WANA is driven by the warm climate in this region. This value is not used in the analysis, however, because all the cattle manure in this region is estimated to be managed in a dry form. The average MCFs for dairy cows and other cattle are estimated using the regional MCF values and the portion of manure that is handled as dry and wet. As shown in the exhibit, the average MCFs vary from as low as 1.0 percent to 11.6 percent for stall-fed dairy cows in OECD.

2.5 Manure Facility Emissions

Manure facility emissions for cattle are estimated by first estimating emissions factors per head from the characteristics defined above, and then multiplying the emissions factors by the relevant cattle populations. As discussed above, separate characteristics, and hence emissions factors, are developed for dairy cows and other cattle. The regional and production system differences in feed intake and feed characteristics are also reflected, as are the differences in manure management conditions discussed in the previous section.

Exhibits 20 and 21 summarize the emissions factors for dairy cows and other cattle respectively. As shown in the exhibits, the emissions factors for dairy cows are larger than the emissions factors for other cattle because their manure is more often handled in a liquid form and because they are larger and require feed intake for lactation. OECD and Eastern European countries have the highest emissions factors while West and North Africa, Sub-Saharan Africa, and Central and South America have the lowest. Among dairy cows, the temperate mixed farming production systems have the highest emissions factors, reflecting the use of liquid manure management systems in these systems.

Using the emissions factors in these exhibits, and the population data presented in Exhibits 1 and 2, total methane emissions from cattle manure management facilities are estimated at 5.8 million tons. As shown in Exhibit 22, OECD and Eastern Europe have the highest emissions, accounting for 75 percent of the total estimated emissions. The temperate rainfed mixed farming production system accounts for over 50 percent of total emissions. Of the total, dairy cattle account for 3.1 million tons of emissions, or about 55 percent of the total. Exhibits 23 and 24 present the emissions estimates separately for dairy cows and other cattle.

Emissions factors from IPCC/OECD (1994) are used to estimate emissions from buffalo, sheep and goats. Total emissions from cattle, buffalo, sheep, and goats are estimated at about 6.6 million tons and are presented by region and production system in Exhibit 25. Emissions from cattle account for about 85 percent of total emissions.

2.6 Total Methane Emissions from Livestock

Exhibit 26 summarizes the estimates of total methane emissions from livestock by region and production system. Total emissions are estimated at about 78.5 million tons. The pattern of emissions across regions and production systems is very similar to the pattern for enteric fermentation emissions because manure facility emissions are relatively small. As shown in the exhibit, Asia, OECD, and Central and South America have the largest emissions, accounting for about 58 million tons of emissions, or nearly 75 percent of the total. As a group, the mixed farming, rainfed production systems have the largest emissions, accounting for about 40 million tons, or 50 percent of total emissions.

Based on recent changes in livestock populations, methane emissions from cattle are increasing at a rate of about 0.1 percent per year. The rates of increase based on population changes are higher for buffalo, 1.6 percent per year, and sheep/goats, 0.6 percent per year. Overall emissions based on population changes are growing at about 0.3 percent per year. These estimates of trends in emissions, particularly for cattle, are likely to be underestimates for two main reasons: (1) emissions per head are likely increasing in many regions as feed intake per head increases with increases in production levels per head (e.g., milk production per head and growth rate per head); and (2) increased reliance on confined animal production systems is leading to increased reliance on liquid-based manure management facilities (which have higher methane emissions per head).

The emissions estimates presented here are similar to recently published estimates, including those in USEPA (1994). The similarities are due, in part, to the fact that similar methods and assumptions are used in the two studies. There are, however, several aspects of this study that are independently-derived, including the following:

· Feeding Situation: The characterization of the feeding situations by production system was based on the information available for this study. As described above in Exhibit 4, approximately 40 percent of cattle are estimated to be stall fed, 44 percent grazing pasture, and 16 percent grazing large areas. Using independent data and assumptions, USEPA (1994) estimated 40 percent stall fed, 35 percent grazing pasture, and 25 percent grazing large areas. The analysis based on production systems presented here implies fewer cattle grazing large areas than estimated previously.

· Relative Size: The relative sizes of the cattle (i.e., live weight) by region were estimated independently in this study and the USEPA (1994) study. Relative sizes in this study were estimated using the Animal Unit Weighting Factor for each region. USEPA (1994) estimated animal sizes in each region based on selected data available for each region. The overall estimates of cattle live weight in this study are about 25 percent smaller than the estimates in USEPA (1994) which contributes to the emissions estimates being lower in this study.

· Feed Characteristics: This study includes feed characteristics by production system and eco-region, and consequently is more detailed than the feed characterization conducted in USEPA (1994).

In addition to the emissions from the livestock evaluated in this study, USEPA (1994) estimate enteric fermentation and manure facility methane emissions from camels, swine, horses, and mules/asses. These methane emissions, relatively small compared to the emissions from the ruminant livestock estimated in this study, are shown in Exhibit 28 to be about 10.1 million tons in 1990. The largest component of these emissions is methane from swine manure facilities, estimated at about 5.3 million tons. Using these estimates of emissions from these livestock, total global methane emissions from enteric fermentation and manure management facilities are estimated as 78.5 + 10.1 = 88.6 million tons per year.