Energy-Related Carbon Dioxide Emissions

Carbon dioxide is the most abundant anthropogenic (human-caused) greenhouse gas in the atmosphere. Atmospheric concentrations of carbon dioxide have been rising at a rate of about 0.6 percent annually in recent years, and that growth rate is likely to increase. As a result, by the middle of the 21st century, carbon dioxide concentrations in the atmosphere could be double their pre-industrialization level (see box on page 90).

Carbon dioxide is the most abundant anthropogenic (human-caused) greenhouse gas in the atmosphere. Atmospheric concentrations of carbon dioxide have been rising at a rate of about 0.6 percent annually in recent years, and that growth rate is likely to increase. As a result, by the middle of the 21st century, carbon dioxide concentrations in the atmosphere could be double their pre-industrialization level (see box on page 90).
Because anthropogenic emissions of carbon dioxide result primarily from the combustion of fossil fuels for energy, world energy use has emerged at the center of the climate change debate. In the IEO2008 reference case, world carbon dioxide emissions are projected to rise from 28.1 billion metric tons in 2005 to 34.3 billion metric tons in 2015 and 42.3 billion metric tons in 2030. 19 From 2004 to 2005, total energy-related carbon dioxide emissions from the non-OECD countries grew by 6.6 percent, while emissions from the OECD countries grew by less than 1 percent. Consequently, annual emissions from the non-OECD countries currently exceed total annual emissions from the OECD countries, and the difference is growing ( Figure 75). In addition, the projected average annual increase in non-OECD emissions from 2005 to 2030 (2.5 percent) is five times the increase projected for the OECD countries (0.5 percent). In 2030, non-OECD emissions, projected at 26.8 billion metric tons, exceed the projection for OECD emissions by 72 percent. The IEO2008 reference case projections are, to the extent possible, based on existing laws and policies. The projections for carbon dioxide emissions could change significantly if existing laws and policies aimed at reducing the use of fossil fuels, and thus greenhouse gas emissions, changed.
The relative contributions of different fossil fuels to total energy-related carbon dioxide emissions have changed over time. In 1990, emissions

What Will It Take To Stabilize Carbon Dioxide Concentrations?
Currently, world energy-related carbon dioxide emissions are increasing at a rate of about 2.1 percent per year. Carbon dioxide concentrations, on the other hand, are rising by only about 0.6 percent per year (see figure below). There are two major reasons for the difference: •First, the base from which growth in the atmospheric carbon dioxide concentration is calculated is much larger than the base from which increases in annual emissions are calculated. Before the industrial revolution, the weight of carbon dioxide in the atmosphere was about 2,163 billion metric tons, a and in the early stages of industrialization the concentration increased slowly-at a rate of about 0.04 percent per year.
•Second, the Earth's oceans and soils absorb carbon dioxide. Over time, about 42 percent (at current emission rates, between 11 and 12 billion metric tons) of the net carbon dioxide emitted through the burning of fossil fuels and deforestation has been absorbed by the planet and has not accumulated in the atmosphere. The other 58 percent has been added to the atmospheric balance. One of the uncertainties in projecting future concentrations is whether the same absorption ratio will hold for future emissions.
In pre-industrial times, the concentration of carbon dioxide in the atmosphere was about 280 parts per million (ppm). The atmospheric concentration of carbon dioxide at present is about 380 ppm, and according to the IEO2008 reference case projections, by 2030 it would be about 450 ppm. b If the growth of world carbon dioxide emissions continues unabated, the concentration of carbon dioxide in the Earth's atmosphere could reach 560 ppm by the middle of the 21st century.
Many possible actions beyond those currently projected in the business-as-usual baseline would be needed to stabilize the atmospheric concentration of carbon dioxide at a level below 560 ppm (still double the pre-industrial level). There is no unique path for achieving any stabilization goal. In addition, a number of "wild cards" could alter the relationship between emissions rates and atmospheric concentrations-such as the Earth's capacity to absorb carbon, which some scientists believe could be diminished by global warming. Each of the options outlined below could be expected to mitigate 1 billion metric tons or more annually by 2030, relative to the IEO2008 reference case projection. It is beyond the scope of this analysis to project either the upper bound or the economic cost of each option.
•Reductions in energy demand growth. Reducing the growth of energy demand in residential and commercial buildings would require adoption of more energy-efficient lighting systems (such as compact fluorescent bulbs and, eventually, light-emitting diodes) and of more efficient heating, cooling, and refrigeration systems, as well as energy-efficient building shell retrofits and new construction. In the transportation sector, it would require more fuelefficient vehicles and more use of public transit and telecommuting. In the industrial sector, more combined heat and power and more efficient processes would be needed to lower energy demand per unit of industrial output.
•Increases in nuclear electricity generation. According to the World Nuclear Association, the achievement of 740 gigawatts of installed nuclear electricity capacity by 2030-36 percent more than projected in the IEO2008 reference case-is possible. If additional nuclear power displaced only coal, such an increase would achieve a reduction of about 1 billion metric tons annually by 2030.
(continued on page 91) a Scientists typically measure carbon dioxide concentrations by the weight of the carbon only, because some carbon exchanges (fluxes) do not involve carbon dioxide. For this analysis, however, the weight of carbon dioxide is used for consistency with the rest of the chapter. b The concentration levels calculated here are based only on energy-related carbon dioxide. Taking into account other sources of carbon dioxide and concentrations of other heat-trapping gases, total greenhouse gas concentrations will be somewhat higher.

What Will It Take To Stabilize Carbon Dioxide Concentrations? (Continued)
•Increased use of nonhydropower renewables for electricity generation in the OECD economies. For nonhydropower renewables to provide 20 percent of the electricity consumed in the OECD economies in 2030, the use of renewables would have to increase by an average of 7.4 percent annually from 2010 to 2030, as compared with the 2.5-percent average increase in the IEO2008 reference case. The increase would yield 1 billion metric tons of abatement annually by 2030.
•Increased use of hydropower and nonhydropower renewables for electricity generation in the non-OECD economies. Assuming that there are more opportunities for hydropower expansion in the non-OECD economies than in the OECD economies, if the combined use of hydropower and nonhydropower renewables in non-OECD countries grew by 3.5 percent per year from 2020 to 2030, as compared with 1.3 percent in the IEO2008 reference case, 1 billion metric tons of carbon dioxide emissions would be avoided annually by 2030.
•Increased use of renewable fuels for transportation. If new technologies were employed to minimize carbon dioxide emissions from input fuels and indirect emissions of other greenhouse gases, so that an additional 20 quadrillion Btu of biofuels was consumed in the transportation sector, assuming a life-cycle savings of 80 percent in carbon dioxide emissions compared to conventional petroleum, 1 billion metric tons of carbon dioxide emissions could be avoided by 2030.
•Carbon capture and storage. It is unlikely that significant amounts of carbon capture and storage will be implemented before 2020. When the technology does become available commercially, its application to about 250 gigawatts of coal-fired generation capacity with a 90-percent removal rate would result in the mitigation of 1 billion metric tons of carbon dioxide emissions annually. The IEO2008 reference case does not include carbon capture and storage. Although there are some small projects in pilot phases around the world, the assumption is that without binding constraints on carbon dioxide emissions throughout the projection period there would be no economic incentive to engage in carbon capture and storage.
•Anthropogenic sequestration. The latest assessment by the Intergovernmental Panel on Climate Change estimates that about 3.7 billion tons carbon dioxide equivalent per year is sequestered by anthropogenic activity, including projects such as reforestation and other land-use programs. A 27-percent increase in such activity by 2030 would represent an emissions reduction of 1 billion metric tons.
For many of the options listed above, the magnitude of the required changes relative to the reference case projections points to the difficulty of achieving stabilization at an atmospheric concentration that is at or below twice preindustrial levels. The effectiveness of reductions in electricity demand as a way to decrease carbon dioxide emissions depends on the fuel mix, the efficiency of generation, and the resultant carbon intensity of electricity supply (carbon dioxide emitted per kilowatthour of generation). For example, because coal-fired generation is more carbon-intensive than natural-gas-fired generation, achieving a given level of reduction in carbon dioxide emissions would require a smaller cut in coal use than the cut in natural gas use that would be required for the same reduction in emissions. Similarly, as the overall carbon intensity of electric power production declines, larger reductions in electricity demand will be needed to achieve a given level of emission abatement (see figure below).
Over time, increases in the efficiencies of generation technologies, such as new natural gas combined-cycle generation, will mean that demand reductions avoid smaller amounts of carbon dioxide emissions. With the average efficiency of electricity generation improving over time, the 2030 reference case intensity of 0.48 metric tons carbon dioxide per megawatthour of electricity supplied is lower than the 2005 historical carbon intensity of 0.56 metric tons per megawatthour supplied. As a result, if more non-carbon-emitting electricity supply is added, such as nuclear and renewables, the demand reduction requirement for the same amount of carbon dioxide emissions savings increases over time.
(continued on page 92) $ $ $ $ $ Coal's share of world carbon dioxide emissions grew from 39 percent in 1990 to 41 percent in 2005 and is projected to increase to 44 percent in 2030. Coal is the most carbon-intensive of the fossil fuels, and it is the fastest-growing energy source in the IEO2008 reference case projection, reflecting its important role in the energy mix of non-OECD countries-especially China and India. In 1990, China and India together accounted for 13 percent of world carbon dioxide emissions; in 2005 their combined share had risen to 23 percent, largely because of strong economic growth and increasing use of coal to provide energy for that growth. In 2030, carbon dioxide emissions from China and India combined are projected to account for 34 percent of total world emissions, with China alone responsible for 28 percent of the world total.
The Kyoto Protocol, which requires participating "Annex I" countries to reduce their greenhouse gas emissions collectively to an annual average of about 5 percent below their 1990 level over the 2008-2012 period, entered into force on February 16, 2005. Annex I countries include the 24 original OECD countries, the European Union, and 14 countries that are considered "economies in transition." 20 As of December 3, 2007, 174 countries and the European Commission had ratified the Kyoto Protocol; however, only the Annex I countries that have ratified the Protocol are obligated to reduce or limit their carbon dioxide emissions. The United States has not ratified the Protocol; and although both China and India have ratified it, neither is subject to emissions limits under the terms of the treaty.
Although the Protocol is technically "in force," it would have an effect on only one year of the IEO2008 forecast, namely, 2010. The IEO2008 projections do not explicitly include the impacts of the Kyoto Protocol, because the treaty does not indicate the methods by which ratifying parties will implement their obligations.
Further, although some countries have passed laws intended to implement the goals of the Kyoto Protocol, it is difficult to interpret those laws in the IEO2008 reference case. Many of the Kyoto goals are being met by "Kyoto mechanisms," such as reforestation, which are not reflected in the projections. Additionally, greenhouse gases other than carbon dioxide often are the least expensive to reduce, and those reductions may account for a larger proportion of some countries' Kyoto goals.
In the IEO2008 projections only energy-related carbon dioxide emissions are calculated; estimates of other greenhouse gas emissions are not included.
Finally, the participants have been unable to agree on a second commitment period or on any actions that might occur after 2012. Until those issues are resolved, it will be difficult to project the effects of the Kyoto Protocol through 2030. 21 There are signs that concerns about global climate change are beginning to affect the world fuel mix. In recent years, many countries have begun to express new interest in expanding their use of non-carbon-emitting nuclear power, in part to stem the growth of greenhouse gas emissions. The IEO2008 reference case projection for electricity generation from nuclear power in 2030 is almost 4 percent higher than the IEO2007 projection, which in turn is 10 percent higher than the IEO2006 projection. The changes reflect a generally more favorable 92 Energy Information Administration / International Energy Outlook 2008

What Will It Take To Stabilize Carbon Dioxide Concentrations? (Continued)
There are wide ranges of estimates both for the marginal cost levels required to achieve various reduction levels and for the corresponding impacts on GDP.
Policies to achieve emission abatements can have a large effect on the cost estimates, as can the rate of development of low-or non-carbon technologies. Specific questions that would have to be answered in order to estimate costs include: •Are all greenhouse gases included in the analysis? Are emissions credits freely traded?
•Is nuclear power allowed to grow at a rapid pace?
•Are biomass and other renewable technologies allowed to penetrate rapidly?
•What discount rates are used for future costs and benefits?
• perception of nuclear power as an alternative to carbonproducing fossil fuels for electricity generation.

Reference Case Carbon Dioxide Emissions
In the IEO2008 reference case, world energy-related carbon dioxide emissions are projected to grow by an average of 1.7 percent per year from 2005 to 2030 (Table 12). For the OECD, annual increases in carbon dioxide emissions are projected to average 0.5 percent, from 13.6 billion metric tons in 2005 to 14.4 billion metric tons in 2015 and 15.5 billion metric tons in 2030.
The highest rate of increase in annual emissions of carbon dioxide among the OECD countries is projected for Mexico, at 2.1 percent per year (Figure 77). Mexico is projected to have the highest GDP growth rate among the OECD countries, and much of its growth is expected to come from energy-intensive industries. For all the other OECD countries, annual increases in carbon dioxide emissions are projected to average less than 1.5 percent. South Korea, which still is industrializing, is the only OECD country other than Mexico for which the average is projected to be greater than 1 percent. Japan's emissions are projected to decrease by an average of 0.2 percent per year from 2005 to 2030, and for OECD Europe an average annual increase of 0.4 percent per year is projected.
Although the United States has not ratified binding emissions constraints, recent changes in U.S. environmental laws and regulations (in addition to other factors) have lowered the projections for carbon dioxide emissions relative to earlier estimates. 22 In the IEO2007 reference case, U.S. emissions were projected to grow by an average of 1.1 percent per year from 2005 to 2030. In the IEO2008 reference case, in contrast, the projected annual growth rate is 0.5 percent over the same period, leading to a 14-percent lower projection for energyrelated carbon dioxide emissions in 2030 in IEO2008 compared with IEO2007 ( Figure 78).
For the non-OECD countries, total carbon dioxide emissions are projected to average 2.5-percent annual growth Energy Information Administration / International Energy Outlook 2008  ( Figure 79). The highest growth rate among the non-OECD countries is projected for China, at 3.3 percent annually from 2005 to 2030, reflecting the country's continued heavy reliance on fossil fuels, especially coal, over the projection period. China's energy-related emissions of carbon dioxide are projected to exceed U.S. emissions by almost 15 percent in 2010 and by 75 percent in 2030. The lowest growth rate among the non-OECD countries is projected for Russia, at 0.9 percent per year. Over the projection period, Russia is expected to expand its reliance on indigenous natural gas resources and nuclear power to fuel electricity generation, and a decline in its population is expected to slow the overall rate of increase in energy demand.
By fuel, world carbon dioxide emissions from the consumption of liquid fuels and other petroleum are projected to grow at an average annual rate of 1.2 percent from 2005 to 2030. The average growth rates for the OECD and non-OECD countries are projected to be 0.3 percent and 2.2 percent per year, respectively ( Figure  80). The highest rate of growth in petroleum-related carbon dioxide emissions is projected for China, at 3.5 percent per year, as its demand for liquid fuels increases to meet growing demand in the transportation and industrial sectors. The United States is expected to remain the largest source of petroleum-related carbon dioxide emissions throughout the period, with projected emissions of 2.8 billion metric tons in 2030-still 34 percent above the corresponding projection for China.
Carbon dioxide emissions from natural gas combustion worldwide are projected to increase on average by 1.7 percent per year, to 8.7 billion metric tons in 2030, with the OECD countries averaging 1.0 percent and the non-OECD countries 2.4 percent ( Figure 81). Again, China is projected to have the most rapid growth in emissions, averaging 5.5 percent annually; however, China's emissions from natural gas combustion amounted to only 0.1 billion metric tons in 2005, and in 2030 they are projected to total only 0.4 billion metric tons, or less than 5 percent of the world total. The growth in U.S. emissions from natural gas use is projected to average 0.1 percent per year, but the projected level of 1.2 billion metric tons in 2030 is triple the projection for China.
Total carbon dioxide emissions from the combustion of coal throughout the world are projected to increase by 2.0 percent per year on average, from 11.4 billion metric tons in 2005 to 18.8 billion metric tons in 2030. Total coal-related emissions from the non-OECD countries have been greater than those from the OECD countries since 1987, and in 2030 they are projected to be more than 2.5 times the OECD total ( Figure

Carbon Dioxide Intensity Measures Emissions per Dollar of GDP
In all countries and regions, energy-related carbon dioxide intensities-expressed in emissions per unit of economic output-are projected to improve (decline) over the projection period as all world economies continue to use energy more efficiently. In 2005, estimated carbon dioxide intensities were 461 metric tons per million dollars of GDP in the OECD countries and 529 metric tons in the non-OECD countries (Table 13). 23 Fossil fuel use in the non-OECD countries is projected to increase strongly over the projection period; however, their economic growth is expected to be even stronger. As a result, non-OECD carbon dioxide intensity is projected to decline by an average of 2.6 percent per year, from 529 metric tons per million dollars of GDP in 2005 to 274 metric tons per million dollars of GDP in 2030. In particular, China, with a relatively high projected rate of growth in emissions (3.3 percent per year), has an even higher projected growth rate for GDP (6.4 percent). As a result, its emissions intensity falls from 693 metric tons per million dollars in 2005 to 334 metric tons in 2030.
For all the OECD countries, average carbon dioxide intensity in 2030 is projected to be 296 metric tons per million dollars. OECD Europe is projected to have the lowest carbon dioxide intensity among the OECD economies in 2030, at 241 metric tons per million dollars, followed by Mexico at 247 metric tons and Japan at 262 metric tons. (Mexico's relatively low carbon dioxide intensity results in large part from its projected 3.9percent annual GDP growth rate, the highest among the OECD countries.

Emissions per Capita
Another measure of carbon dioxide intensity is emissions per person. Carbon dioxide emissions per capita in the OECD economies are significantly higher (about fourfold in 2005) than in the non-OECD economies ( Figure 83). If non-OECD countries consumed as much energy per capita as the OECD countries, the projection for world carbon dioxide emissions in 2030 would be much larger, because the non-OECD countries would consume almost four times more energy than the current reference case estimate of 409 quadrillion Btu. Further, given the expectation that non-OECD countries will rely heavily on fossil fuels to meet their energy needs, the increase in carbon dioxide emissions would be even greater.
Among the non-OECD countries, Russia has the highest projected increase in carbon dioxide emissions per capita in the IEO2008 reference case, from 12 metric tons per person in 2005 to 17 metric tons in 2030 ( Figure 84 and    Other factors that can affect carbon dioxide emissions per capita include climate (in general, more energy is used per capita for heating in colder climates than is used for cooling in warmer climates) and population density (densely populated countries use less energy per capita for transportation). For example, Canada has a relatively cold climate with a low population density, and its carbon dioxide emissions in 2005 are estimated at 19.5 metric tons per capita, whereas Japan has a more temperate climate and a much higher population density, and its emissions in 2005 are estimated at 9.6 metric tons per capita-about half the rate for Canada.

Alternative Macroeconomic Growth Cases
Economic growth is the most significant factor underlying the projections for growth in energy-related carbon dioxide emissions in the mid-term, as the world continues to rely on fossil fuels for most of its energy use. Accordingly, projections of world carbon dioxide emissions are lower in the IEO2008 low economic growth case and higher in the high economic growth case.
In the high growth case, world carbon dioxide emissions are projected to increase at an average rate of 2.1 percent annually from 2005 to 2030, as compared with 1.7 percent in the reference case. For the OECD countries, the projected average increase is 0.9 percent per year; for the non-OECD countries, the average is 2.9 percent per year.
In the low growth case, world carbon dioxide emissions are projected to increase by 1.3 percent per year, with averages of 0.2 percent per year in the OECD countries and 2.1 percent per year in the non-OECD countries (compared with 0.5 percent and 2.5 percent, respectively, in the reference case). Total emissions worldwide are projected to be 38.4 billion metric tons in 2030 in the low growth case and 46.6 billion metric tons in the high growth case-21 percent higher than projected in the low growth case (Figure 86). The projections for emissions by fuel show similar variations across the cases.

Alternative Price Cases
The projections for carbon dioxide emissions in the IEO2008 low and high price cases ( Figure 87 as compared with the reference case projection (42.3 billion metric tons), total carbon dioxide emissions are projected to be higher in the low price case (43.4 billion metric tons) and lower in the high price case (40.1 billion metric tons). Thus, there is an 8-percent difference between the projections in the two alternative world oil price cases, as compared with a 21-percent difference between the alternative macroeconomic growth cases.
In the alternative price cases, world oil and natural gas prices are affected more strongly than coal prices. As a result (and in the absence of policies to limit the use of coal), in the high price case both liquids and natural gas lose global market share to coal relative to the reference case projection. In the IEO2008 reference case, coal's share of total energy use is projected to increase to 29 percent in 2030; in the high price case, its share increases to 30 percent; and in the low price case, its share drops to 27 percent in 2030.
Prices have the greatest impact on world liquids consumption and the associated carbon dioxide emissions. In the high price case, where nominal world oil prices reach $186 per barrel in 2030, nations choose alternative fuels over liquids wherever possible, so that liquidsrelated emissions total 13.1 billion metric tons in 2030, down from 14.9 billion metric tons in the reference case. In the low price case, world oil prices decline to $69 per barrel in 2030, substantially lower than the $113 per barrel projected in the reference case and providing little economic incentive for nations to turn to other forms of energy. Consequently, liquids-related emissions in 2030 in the low price case, at 16.2 billion metric tons, are 1.3 billion metric tons higher than projected in the reference case.
The impact of high prices on natural gas use is smaller than the impact on liquids consumption, but a similar trend away from natural gas to other fuels, particularly coal, is projected. In the high price case, world carbon dioxide emissions from natural gas combustion in 2030 total 8.3 billion metric tons, down from 8.7 billion metric tons in the reference case. In the low price case, naturalgas-related emissions in 2030 are projected to total 9.2 billion metric tons.