Carbon Dioxide Emissions U.S. Carbon Dioxide Emissions in 1999: Effects of Weather and High Levels of Nuclear and Hydroelectric Power Generation A preliminary analysis by EIA indicates that U.S. energy-related carbon dioxide emissions in 1999 could have been higher by as much as 29 million metric tons carbon equivalent if weather patterns had been normal, and if electricity generation from nuclear and hydroelectric power plants had not been higher than normal. As in 1998, weather played a key role in moderating energy-related carbon dioxide emissions in 1999. A warmer-than-average winter for the United States as a whole meant less consumption of heating fuels. And, although the warmer-than-average weather continued throughout the summer months, demand for air conditioning in 1999 still was lower than in 1998, when the summer months were even hotter. Analysis of the underlying weather trends indicates that in 1999 the warm winter (heating degree-days 6.4 percent below normal) meant that carbon dioxide emissions were lower by 14.7 million metric tons carbon equivalent than they would have been with normai winter weather. The warm summer (cooling degree-days 3.5 percent above normal), on the other hand, added about 3.1 million metric tons carbon equivalent above the level expected with normal summer weather. The net effect of the deviations from normal weather was that carbon emissions were lower by 11.6 million metric tons carbon equivalent than they would have been if weather patterns had been normal in 1999. Electricity generation from U.S. nuclear power plants reached an all-time high of 725.0 billion kilowatthours in 1999. Although it is difficult to define an "average' or "typical" year for nuclear power generation in the United States, the average for 1997, 1998, and 1999 was about 674 billion kilowatthours, and so that value was used as the baseline for this analysis. No new nuclear facilities have come on line in recent years, but the output from existing facilities has increased. If nuclear power generation in 1999 had remained at its 3-year average level, it is estimated that carbon dioxide emissions from fossil-fired generating plants would have been higher by 11.7 million metric tons carbon equivalent. At 293.7 billion kilowatthours, hydroelectric power generation in 1999, while down from its high of 337.2 billion kilowatthours in 1997, still was about 8.6 percent above what would have been expected in a year of average precipitation levels. If hydroelectric power generation in 1999 had been at the average level, carbon dioxide emissions from fossil-fired generating plants would have been higher by 5.2 million metric tons carbon equivalent. Total generation from nuclear and hydroelectric power plants in 1999 was about 1,018.7 billion kilowatthours, accounting for 32.0 percent of electric utility generation. By comparison, their combined total in 1990 was 856.8 billion kilowatthours, or 30.5 percent of electric utility generation, and in 1998 it was 978.1 billion kilowatthours or 30.4 percent of the total generation by electric utilities. *The analysts described here is based on data contained in Energy Information Administration, Short-Term Energy Outlook, DOE/EIA-0202(2000/1S) (Washington, DC, April 2000). Energy data used elsewhere in this report are from more recent sources and, therefore, may differ from the data cited here. glass; primary metals; petroleum products; and foodtogether account for approximately two-thirds of total industrial energy-related carbon dioxide emissions. In 1999, two of those six industries grew by less than 1 percent (food by 0.7 percent and primary metals by 0.8 percent); two grew by slightly more than 1 percent (paper by 1.1 percent and petroleum by 1.3 percent); and only two grew by more than 2 percent (chemicals by 2.1 percent and stone, clay and glass by 2.9 percent). Output from the less energy-intensive industries grew rapidly (e.g., computer equipment by 57.2 percent and semiconductors and related components by 47.4 percent). Because the less energy-intensive industries use relatively little energy to produce goods, even large increases in their output are associated with only minor increases in energy consumption and related carbon dioxide emissions. Transportation sector energy demand is largely driven by income growth, fuel prices, and fuel economy trends. Propelled by gross domestic product (GDP) growth of 4.1 percent in 1999 and real disposable income growth of 3.2 percent, transportation energy-related carbon dioxide emissions increased by 2.9 percent, from 481.9 million metric tons carbon equivalent in 1998 to 496.1 million metric tons carbon equivalent in 1999. The increase in emissions accompanied large percentage increases in consumption of distillate fuel, aviation fuel, and motor gasoline. Although net generation of electricity increased by 1.7 percent in 1999, total carbon dioxide emissions from the electric power sector increased by only 1.0 percent, from 608.5 million metric tons carbon equivalent in 1998 to 614.3 million metric tons carbon equivalent in 1999. The growth rate in emissions was less than the growth rate in net generation in part because 54.2 billion kilowatthours of the net increase (59.8 billion kilowatthours) came from nuclear power plants, which produce essentially no carbon dioxide emissions. Nuclear electricity generation was 10.5 percent higher in 1999 than in 1998. Nonfuel uses of fossil fuels, principally petroleum, sequestered 89.8 million metric tons carbon equivalent in 1999-4.8 million metric tons carbon equivalent (5.7 percent) more than in 1998. The major fossil fuel products that sequester carbon include liquefied petroleum gas (LPG), feedstocks for plastics and other petrochemicals, and asphalt and road oils. It is estimated that, of the amount of carbon sequestered in the form of plastic, about 3.5 million metric tons carbon equivalent was emitted as carbon dioxide from the burning of plastic components in municipal solid waste. Emissions of carbon dioxide from non-energyconsuming industrial processes contributed 0.6 million metric tons carbon equivalent to the 1999 increase in emissions (Table 4). Emissions from cement production processes (excluding the energy portion) rose from 10.7 to 10.9 million metric tons carbon equivalent, while emissions from natural gas flaring rose from 3.9 to 4.2 million metric tons carbon equivalent. Energy Consumption The consumption of energy in the form of fossil fuel combustion is the largest single contributor to greenhouse gas emissions in the United States and the world. Of total 1999 U.S. carbon dioxide emissions, 98 percent. or 1,510.8 million metric tons carbon equivalent, resulted from the combustion of fossil fuels. This figure represents an increase of 1 percentage point over 1998 levels. In the short term, year-to-year changes in energy consumption and carbon dioxide emissions tend to be dominated by weather, economic fluctuations, and movements in energy prices. Over longer time spans, changes in energy consumption and emissions are influenced by other factors such as population shifts and energy consumers' choice of fuels, appliances, and capital equipment (e.g. vehicles, aircraft, and industrial plant and equipment). The energy-consuming capital stock of the United States-cars and trucks, airplanes, heating and cooling plants in homes and businesses, Carbon Dioxide Emissions steel mills, aluminum smelters, cement plants, and petroleum refineries-change slowly from one year to the next, because capital stock is retired only as it begins to break down or becomes obsolete. The Energy Information Administration (EIA) divides energy consumption into four general sectoral categories: residential, commercial, industrial, and transportation.20 Emissions from electric utilities, which provide electricity to the end-use sectors, are allocated in proportion to the electricity consumed in each sector (Table 5) Emissions from independent power producers and industrial cogenerators are included in the industrial sector estimates. EIA is in the process of moving the data on electricity generated in the industrial sector into a combined electric power sector that includes electric utilities, nonutility generators, and industrial cogenerators. When that process is completed, this report will follow the same protocol. In the interim, this report provides, below, a separate preliminary estimate of emissions from the entire electric power sector for the 1990 to 1999 time period 21 20 Energy consumption (and greenhouse gas emissions) estimates by end-use sector are less reliable than aggregate estimates, because in many cases EIA survey respondents attribute consumption to a particular economic sector by reference to the tariff or tax status of the buyer rather than definite knowledge of the buyer's activities. For example, a shopping mall that consumes enough electricity to be granted an "industrial" electricity tariff may be categorized as an industrial rather than a commercial customer. 21 Because most EÍA data sources include thermal energy from industrial cogenerators in estimates of nonutility carbon dioxide emissions, the estimates in this report do not agree with other ELA publications (see for example Table 12.7 of the Annual Energy Review 1999). In Table 10 of this report, the emissions attributable to thermal energy consumption have been estimated and removed from the emissions of nonutility generators in order to isolate emissions related only to electricity generation. The estimates for the electricity component for nonutility power producers should be considered preliminary for all years. Carbon Dioxide Emissions amount (193.4 million metric tons carbon equivalent) 22 Since 1990, residential electricity-related emissions have grown by 1.9 percent annually. In contrast, emissions from the direct combustion of fuels (primarily natural gas) have increased by 0.7 percent per year, to 96.7 million metric tons carbon equivalent. Total carbon dioxide emissions from the residential sector increased by 0.4 percent in 1999 (Table 6). Year-to-year, residential sector emissions are heavily influenced by weather. For example, in 1996, a relatively cold year, carbon dioxide emissions from the residential sector grew by 5.9 percent over 1995. In 1997, they declined by 0.4 percent due to warmer weather. Since 1990, growth in carbon dioxide emissions attributable to the residential sector has averaged 1.5 percent per year. As a result, residential sector emissions in 1999 were 35.9 million metric tons carbon equivalent higher than in 1990, representing 22 percent of the total increase in US. energy-related carbon dioxide emissions since 1990. Long-term trends in residential carbon dioxide emissions are heavily influenced by demographic factors, living space attributes, and building shell and appliance efficiency choices. For example, the movement of populations into the Sunbelt tends to increase summer air conditioning consumption and promote the use of electric heat pumps, which increases indirect emissions from electricity use. Growth in the number of households, resulting from increasing population and immigration, contributes to more residential energy consumption. Commercial Sector Commercial sector carbon dioxide emissions, at 243.5 million metric tons carbon equivalent, account for about 16 percent of total energy-related carbon dioxide emissions, of which almost three-quarters (182.6 million metric tons carbon equivalent) is the sector's pro-rated share of electric utility emissions. Although commercial sector emissions largely have their origin in the space heating and cooling requirements of structures such as office buildings, lighting is a more important component of commercial energy demand than it is in the residential sector. Thus, although commercial sector emissions are strongly affected by the weather, they are affected less than residential sector emissions. In the longer run, because commercial activity is a factor of the larger economy, emissions from the commercial sector are more affected by economic trends and less affected by population growth than are emissions from the residential sector. Emissions attributable to the commercial sector's pro-rated share of electricity consumption declined by 1.4 percent in 1999, while emissions from the direct combustion of fuels (dominated by natural gas, as in the restdential sector) increased by 2.6 percent. Overall, carbon dioxide emissions related to commercial sector activity declined by 0.4 percent-from 244.5 to 243.5 million metric tons carbon equivalent-between 1998 and 1999 (Table 7), as the reduction in electricity-related emissions outweighed the increase in emissions form the direct combustion of fuels. Since 1990, commercial emissions growth has averaged 1.8 percent per year--the largest growth of any energy-use sector-and commercial sector carbon dioxide emissions have risen by a total of 35.8 million metric tons carbon equivalent, or 22 percent of the total increase in U.S. energy-related carbon dioxide emissions. Transportation Sector Transportation sector emissions, at 496.1 million metric tons carbon equivalent, accounted for one-third of total energy-related carbon dioxide emissions in 1999. Almost all (98 percent) of transportation sector emissions result from the consumption of petroleum products, particularly, motor gasoline (60 percent of transportation sector emissions), diesel fuel or "middle distillates (20 percent), jet fuel (13 percent), and residual oil or heavy fuel oil, largely for maritime use (4 percent). Motor gasoline is used primarily in automobiles and light trucks, and middle distillates are used in heavy trucks, locomotives, and ships. Emissions attributable to the transportation sector grew by 2.9 percent. from 481.9 million metric tons carbon equivalent in 1998 to 496.1 million metric tons carbon equivalent in 1999 (Table 8). Fuel-use patterns and related emissions sources in the transportation sector are different from those in the other energy-use sectors. By far the largest single source of emissions, motor gasoline, at 299.1 million metric tons carbon equivalent, grew by 2.1 percent. The highest rates of growth were for jet fuel emissions (which grew by 3.1 percent, from 64.2 to 66.3 million metric tons carbon equivalent) and distillate fuel emissions (which grew by 3.8 percent, from 96.4 to 100.1 million metric tons carbon equivalent). Of the total growth in U.S. energy-related carbon dioxide emissions in 1999, 91.9 percent (14.2 million metric tons carbon equivalent) was accounted for by the transportation sector. Since 1990, carbon dioxide emissions attributable to energy use in the transportation sector have grown annually at a rate of 1.6 percent, increasing by a total of 64.3 million metric tons carbon equivalent and Sectoral (residential, commercial, and industrial) carbon dioxide emissions are based on the share of total electric power sector carbon dioxide emissions that can be attributed to each sector. The shares are based on the percentage of total electric utility sales purchased by each sector. All nonutility carbon dioxide emissions are allocated to the industrial sector. representing 39.6 percent of the growth in energy. related carbon dioxide emissions from all sectors. Industrial Sector Industrial sector emissions, at 481.2 million metric tons carbon equivalent, accounted for about 32 percent of total U.S. energy-related carbon dioxide emissions in 1999. In terms of fuel shares, electricity purchased from electric utilities was responsible for 37.3 percent of total Industrial sector emissions (179.5 million metric tons carbon equivalent), natural gas for 29.4 percent (141.6 million metric tons carbon equivalent), petroleum for 21.7 percent (104.2 million metric tons carbon equivalent), and coal for 11.3 percent (54.5 million metric tons carbon equivalent). Generally, industrial sector emissions are strongly affected by the growth of the economy. Estimated carbon dioxide emissions related to energy consumption in the industrial sector increased by only 0.2 percent in 1999-from 480.2 to 481.2 million metric tons carbon equivalent (Table 9)-despite GDP growth of 4.1 percent. Since 1990, growth in carbon dioxide emissions attributable to industrial sector energy consumption has averaged 0.6 percent per year. As a result, total energy-related industrial emissions in 1999 were 5.8 percent (26.3 million metric tons carbon equivalent) higher than they were in 1990, despite a much larger economy. The increase in industrial sector emissions from 1990 to 1999 represents 16.2 percent of the total growth in U.S. energy-related carbon dioxide emissions. In 1999 the six most energy-intensive industry groups, which together account for approximately two-thirds of total industrial energy-related carbon dioxide emissions, grew less rapidly than the overall economy (4.1 percent) or the manufacturing component of industrial production (4.3 percent). The 1999 growth rates for the six energy-intensive industries (Figure 2) were 0.8 percent (primary metals). 2.1 percent (chemicals). 1.1 percent (paper). 2.9 percent (stone, clay and glass). 1.3 percent (petroleum), and 0.7 percent (food) 23 Thus, all the energy-intensive industries grew less rapidly than the economy as a whole. The box on page 18 discusses the importance of the energy-intensive industries to overall emissions, using data from ELA's 1994 Manufac turing Energy Consumption Survey (MECS). A new survey of manufacturers was conducted by EIA in 1998, but the results are not yet available. What makes the low growth in industrial sector emissions even more remarkable is that this sector includes emissions from independent (nonutility) power producers and industrial cogenerators, which provide electrical Carbon Dioxide Emissions energy to the grid that is consumed in other sectors of the economy. Thus, the inclusion of carbon dioxide emissions resulting from the generation of electricity by independent power producers tends to overstate the amount of carbon dioxide attributable to the industrial sector, implying that actual emissions growth related to electrical and thermal energy use in the industrial sector may be lower than estimated here. A contributing factor to the low growth in industrial sector carbon dioxide emissions is the erosion of the older energy-intensive (and specifically coal-intensive) indus trial base. For example, coke plants consumed 38.9 million short tons of coal in 1990 but only 27.9 million short tons in 1999. Additionally, other industrial coal consumption has declined from 76.3 million short tons in 1990 to 68.0 million short tons in 1999. As a result, carbon dioxide emissions attributable to industrial coal use have declined by 13.9 million metric tons carbon equiva lent or about 20 percent since 1990, helping to offset an increase of 23.3 million metric tons carbon equivalent in natural-gas-related carbon dioxide emissions (also about 20 percent) over the same time period. Because natural gas is more than 40 percent less carbon-intensive than coal, the substitution of natural-gas-fired output for coal-fired output has further contributed to the modest growth in industrial emissions. Declines in residual and distillate fuel use have resulted in an additional decrease in emissions of 5 million metric tons carbon equivalent between 1990 and 1999. 23U.S. Federal Reserve Board, "G17 Historical Data: Industrial Production and Capacity Utilization," web site www.federalreserve.gov/ release/G17/download.htm. Carbon Dioxide Emissions Energy-Related Carbon Dioxide Emissions in Manufacturing Manufacturing, which accounts for 80 percent of Industrial energy consumption, also accounts for 80 percent of industrial energy-related carbon emissions. In 1994, two industries, petroleum and chemicals, emitted over 40 percent of the energy-related carbon in manufacturing. The next four largest emitters (primary metals, paper, food, and the stone, glass, and clay products industry) produced the other 40 percent of the energy-related carbon emissions from manufacturing (see figure). Total Energy-Related Carbon Dioxide Emissions for Selected Manufacturing Industries, 1994 0 Petroleum and Coal Products (81.9) Chemicals and Allied Products (78.3) Primary Metals Industries (64.5) Paper and Allied Products (31.6) Food and Kindred Products (24.4) Stone, Clay and Glass Products (21.6) All Other Manufacturing Industries (69.5) 10 20 30 40 50 60 70 80 90 Source: Energy Information Administration, Form EIA-846, *1994 Manufacturing Energy Consumption Survey," and Form EIA-810, "Monthly Refinery Report" (1994). The carbon intensity of energy use is the amount of carbon emitted per unit of energy used. Both the mix of energy sources used and the uses of energy affect carbon intensity. Overall, manufacturing industries had a carbon intensity of 17.16 million metric tons per quadrillion Btu in 1994; however, the carbon intensities of the various industries differed markedly. The petroleum industry and the chemical industry both convert energy sources into products, such as petrochemical feedstocks and plastics. Only part of the carbon content is emitted to the atmosphere, the rest being sequestered in the product (see Table A2 in Appendix A). As a result of the nonfuel use of energy sources, the petroleum and chemical industries had lower than average carbon intensities in 1994, 12.91 and 14.69 million metric tons per quadrillion Btu. respectively. The paper industry uses wood byproducts extensively, yielding a carbon intensity of 11.87 million metric tons per quadrillion Btu in 1994. The carbon emissions from wood combustion are considered to be zero, because the carbon emitted has been recently sequestered and the regrowing of the trees will re-sequester the emitted carbon. In contrast, the primary metals industry, which relies heavily on carbon-intensive coal and, to a lesser extent, on electricity (much of which is generated using coal), had an overall carbon intensity of 26.19 million metric tons per quadrillion Btu in 1994. The food industry and the stone, metal, and glass industry both had carbon intensities that were slightly above the average for all manufacturing industries. However, the carbon intensities for these two industries, and for the primary metals industry, were comparable to those for "All Other Manufacturing Industries" (25.45 million metric tons per quadrillion Btu). Electricity use is the main source of carbon dioxide emissions in industries outside the top six emitters. Other manufacturing includes industries such as textile mill products, furniture and fixtures, leather and leather products, fabricated metal products, industrial machinery, and electric and electronic products. Their collective carbon intensity was 25.45 million metric tons per quadrillion Btu in 1994. "Appendixes for this report are available on web site www.eia.doe.gov/oiaf/1605/ggrpt/index.html. Electric Power Industry In the data presented by sector above, carbon dioxide emissions from electric utility fuel use are assigned to the energy-consuming sectors. However, because the electric power industry is changing and the "electric utility" designation is becoming less relevant, this section and the data in Table 10 present an estimate of carbon dioxide emissions for the entire electric power industry. By this accounting, emissions from the electric power industry as a whole made up 41 percent of total U.S. energy-related carbon dioxide emissions in 1999. Carbon dioxide emissions from the electric power Industry, despite a strong economy, increased by only 1.0 percent, from 608.5 million metric tons carbon equivalent in 1998 to 614.3 million metric tons carbon equivalent in 1999 (Table 10).24 Although 1999 was 24 The emissions estimates given are for both electric utility and nonutility power producers. No attempt has been made, however, to subtract nonutility emissions from the industrial sector by fuel. Therefore, the estimate should be considered separately from the preceding emissions estimates by sector. A report produced by the U.S. Department of Energy and the U.S. Environmental Protection Agency, Carbon Dicode Emissions from the Generation of Electric Power In the United States (Washington, DC, July 2000), used more preliminary data and a different methodology, resulting in an estimated emissions growth rate of 1.4 percent from 1998 to 1999, |