SOURCES AND TECHNICAL NOTES FOR ATMOSPHERE AND CLIMATE.

Data Table 14.1
CO2 Emissions from Industrial Processes, 1992

Source: Carbon Dioxide Information Analysis Center (CDIAC), Environmental Sciences Division, Oak Ridge National Laboratory, 1992 Estimates of CO2 Emissions from Fossil Fuel Burning and Cement Manufacturing Based on the United Nations Energy Statistics and the U.S. Bureau of Mines Cement Manufacturing Data, ORNL/CDIAC-25, NDP-030 (an accessible numerical database) (Oak Ridge, Tennessee, September 1995).

This table includes data on industrial additions to the carbon dioxide (CO2 ) flux from solid fuels, liquid fuels, gas fuels, gas flaring, and cement manufacture . CDIAC annually calculates emissions of CO2 from the burning of fossil fuels and the manufacture of cement for most of the countries of the world. Estimates of total and per capita national emissions do not include bunker fuels used in international transport because of the difficulty of apportioning these fuels among the countries benefiting from that transport. Emissions from bunker fuels are shown separately for the country where the fuel was delivered.

CDIAC calculates emissions from data on the net apparent consumption of fossil fuels (based on the World Energy Data Set maintained by the United Nations Statistical Division), and from data on world cement manufacture (based on the Cement Manufacturing Data Set maintained by the U.S. Bureau of Mines). Emissions are calculated using global average fuel chemistry and usage.

Although estimates of world emissions are probably within 10 percent of actual emissions, individual country estimates may depart more severely from reality. CDIAC points out that the time trends from a consistent, uniform time series should be more accurate than the individual values. Each year, CDIAC recalculates the entire time series from 1950 to the present, incorporating its most recent understanding and the latest corrections to the database. As a result, the car-bon emissions estimate data set has become more consistent, and probably more accurate, each year.

Emissions of CO2 are often calculated and reported in terms of their content of elemental carbon. CDIAC reports them in that way. For this table, their values were converted to the actual mass of CO2 by multiplying the carbon mass by 3.664 (the ratio of the mass of carbon to that of CO2 ).

Solid, liquid, and gas fuels are primarily, but not exclusively, coals, petroleum products, and natural gas, respectively. Gas flaring is the practice of burning off gas released in the process of petroleum extraction, a practice that is declining. During cement manufacture, cement is calcined to produce calcium oxide. In the process, 0.498 metric ton of CO2 is released for each ton of cement production. Total emissions consist of the sum of the CO2 produced during the consumption of solid, liquid, and gas fuels, and from gas flaring and the manufacture of cement.

Combustion of different fossil fuels releases CO2 at different rates for the same level of energy production.
Burning oil releases about 1.5 times the amount of CO2 released from burning natural gas; coal combustion releases about twice the CO2 of natural gas.

It was assumed that approximately 1 percent of the coal used by industry and power plants was not burned, and an additional few percent were converted to nonoxidizing uses.
Other oxidative reactions of coal are assumed to be of negligible importance in carbon budget modeling. CO2 emissions from gas flaring and cement production make up about 3 percent of the CO2 emitted by fossil fuel combustion.

These data from CDIAC represent the only complete global data set of CO2 emissions. Individual country estimates, based on more detailed information and a country-specific methodology, could differ. An experts meeting, convened by the Organisation for Economic Co-Operation and Development (OECD) in February 1991, has recommended (Estimation of Greenhouse Gas Emissions and Sinks, OECD, Paris, August 1991) that when countries calculate their own emissions of CO2 , they use a more detailed method when these data are available. Such data are currently available for only a few countries (see Data Table 14.6). CDIAC s method has the advantage of calculating CO2 emissions from a single common data set available for all countries.

Data Table 14.2
Other Greenhouse Gas Emissions, 1991

Sources: Land use change: Food and Agriculture Organization of the United Nations (FAO), Forest Resources Division, Forest Resources Assessment 1990: Global Synthesis (FAO, Rome, 1995).

Methane (CH4 ) from municipal solid waste: Jean Lerner, personal communication (National Aeronautics and Space Administration Goddard Space Flight Center, Institute for Space Studies, May 1989); and H.G. Bingemer and P.J.
Crutzen, The Production of CH4 from Solid Wastes, Journal of Geophysical Research, Vol. 92, No. D2 (1987), pp.
2181 2187.

Methane (CH4 ) from coal mining: David W. Barns and J.A.
Edmonds, An Evaluation of the Relationship Between the Production and Use of Energy and Atmospheric Methane Emissions (U.S. Department of Energy, Office of Energy Research, Carbon Dioxide Research Program, No. TR047, April 1990); and the World Energy Council (WEC), 1994 Survey of Energy Resources (WEC, London, 1995).

CH4 from oil and gas production and distribution: David W.
Barns and J.A. Edmonds, An Evaluation of the Relationship Between the Production and Use of Energy and Atmospheric Methane Emissions (U.S. Department of Energy, Office of Energy Research, Carbon Dioxide Research Program, No. TR047, April 1990); Carbon Dioxide Information Analysis Center (CDIAC), Environmental Sciences Division, Oak Ridge National Laboratory, 1991 Estimates of CO2 Emissions from Fossil Fuel Burning and Cement Manufacturing Based on the United Nations Energy Statistics and the U.S. Bureau of Mines Cement Manufacturing Data, ORNL/CDIAC-25, NDP-030 (an accessible numerical database) (Oak Ridge, Tennessee, September 1995); American Gas Association (AGA), Natural Gas and Climate Change: The Greenhouse Effect, Issue Brief 1989 7 (AGA, Washington, D.C., June 14, 1989); A.A. Makarov and I.A. Basmakov, The Soviet Union: A Strategy of Energy Development with Minimum Emission of Greenhouse Gases (Pacific Northwest Laboratory, Richland, Washington, 1990); and S. Hobart, David Spottiswoode, James Ball et al., Methane Leakage from Natural Gas Operations (The Alphatania Group, London, 1989).

CH4 from wet rice agriculture: FAO, FAOSTAT-PC (FAO, Rome, 1995); Elaine Mathews, Inez Fung, and Jean Lerner, Methane Emission from Rice Cultivation: Geographic and Seasonal Distribution of Cultivated Areas and Emissions, Global Biogeochemical Cycles, Vol. 5, No. 1 (March 1991), pp.
3 24; and U.S. Environmental Protection Agency (U.S. EPA), Office of Policy, Planning, and Evaluation, International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994), pp. 3-1 3-25.

CH4 from livestock: Jean Lerner, Elaine Mathews, and Inez Fung, Methane Emissions from Animals: A Global High- Resolution Data Base, Global Biogeochemical Cycles, Vol.
2, No. 2 (June 1988), pp. 139 156; FAO, FAOSTAT-PC (FAO, Rome, 1993); U.S. EPA, Office of Policy, Planning, and Evaluation, International Anthropogenic Methane Emissions:
Estimates for 1990 (EPA 230-R-93-010, January 1994), pp. 2- 1 2-44.

CO2 and CH4 are the two most important greenhouse gases, and both are still uncontrolled. This data table provides estimates of annual emissions of CO2 from land use change (i.e., deforestation) and CH4 emissions by source. Nitrous oxide, tropospheric ozone, and chlorofluorocarbons (CFCs) are also important to the greenhouse effect, but are difficult to estimate especially the potent CFCs that are rapidly being controlled at national levels by international agreement. Tropospheric ozone has an average lifetime measured in hours and is a product of particular chemical processes involving the precursors CH4 , carbon monoxide, nitrogen oxides, and nonmethane hydrocarbons in the presence of sunlight. Nitrous oxide emissions by country have proved difficult to estimate, in part because significant emissions are poorly understood. Production estimates and emission parameters for CFC-11 and CFC-12 are not available, but production in developing countries will cease on January 1, 1996 (except for some residual production not to exceed 15 percent of 1986 production, to supply developing countries), and recycling mandates have already been imposed.

The Organisation for Economic Co-Operation and Development (OECD) hosted an experts meeting in February 1991 on greenhouse emissions (a final report was published in August 1991, Estimation of Greenhouse Gas Emissions and Sinks (OECD, Paris)) to discuss methodologies that countries could use to estimate their own inventories of greenhouse gases (other than CFCs) and to point to areas requiring further research. Although these discussions served to illuminate and define the methods used here, the final published recommendations were directed toward informing governments as to what data they could collect and what kind of basic country-specific (and even ecosystem-specific) research is required if they are to assess their contribution to greenhouse gas emissions in an international context. The final report of the OECD experts meeting included additional suggested data sets and methods not fully discussed or validated during the meeting (e.g., a suggested source of deforestation data).

The estimates of emissions in this data table can be controversial, but are believed to be accurate estimates of the relative magnitudes of emissions and also the best possible, given the available data sets. The World Resources Institute (WRI) welcomes independent estimates of anthropogenic emissions of greenhouse gases from the countries of the world (see Data Table 14.6). The methods used here were chosen to maximize the use of the available international data so the estimates would be comparable among countries. The international data set on any subject is limited, and so these estimates are also limited. Until most of the countries of the world publish their own independent estimates based on common methods and scientifically valid parameters global comparisons will require the use of methods based on the least-common data set. Common methods and parameters were used between countries unless differing, but explicit and published, parameters were available that covered all countries. For example, estimates of CH4 emissions from coal mining were based on published data on the differing CH4 content of various coals and their production around the world. More complex calculations that might have been possible for one or two data-rich countries are inappropriate for the world as a whole and were not attempted even for those few countries that might have sufficient (and uncontroversial) data. An alternative accounting of national greenhouse gas emissions has recently been published (Susan Subak, Paul Raskin, and David Von Hippel, National Greenhouse Gas Accounts: Current Anthropogenic Sources and Sinks (Stockholm Environment Institute, Boston, 1992)) for the year 1988. This approach generally produced results similar to those reported in past volumes of this report and generally followed similar methodologies. WRI has adopted Subak et al. s methodological refinement for estimating CH4 emissions from solid wastes. Another study, by the U.S. EPA Office of Policy, Planning, and Evaluation, International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994), provided still more estimates and helped to refine estimates of CH4 from wet rice cultivation and livestock.

Carbon dioxide emissions from land use change are based on FAO estimates of deforestation and forest biomass for tropical countries. The burning of biomass, per se, does not necessarily contribute to the CO2 flux. Fire is a natural process, and as long as burning and growth are in balance, there is no net movement of carbon from biomass to the atmosphere. Deforestation, however is defined as the conversion of land from forest to other uses. Carbon released in this process will not be replaced.

The carbon density used to estimate these releases was 45 percent of biomass, and biomass estimates were taken from average forest densities for whole countries as reported by the FAO. These CO2 emission estimates explicitly include shifting cultivation and the diversion of forest fallow to permanent clearing. They are also consistent and global in scope. They are the most complete estimates available, but are subject to modification should better data become available. Individual countries question these FAO estimates. See Sources and Technical Notes for Data Table 14.1 for further information.

Although, in principle, emissions from land use change should include other gases emitted from the burning of forest land, as well as from the burning of grassland, the conversion of grassland to cropland, the creation of wetlands, and the burning of crop and animal residues, the international data sets needed to estimate these emissions do not exist (OECD experts meeting report). Except for CO2 emissions from deforestation, then, emissions from biomass burning in general are not available. Grasses or trees that grow back after a fire merely recycle the carbon and do not contribute CO2 to long-term greenhouse heating.

WRI subtracted elemental carbon permanently sequestered in the soil (an estimated 5 percent of the biomass carbon) and also subtracted the weight of carbon contained in sawlogs and veneer logs (FAO, FAOSTAT-PC (FAO, Rome, 1993)), produced in each tropical country from CO2 releases calculated from land use change. Carbon was estimated as making up 45 percent of the weight of these wood products.
This step was taken to approximate the amount of carbon sequestered from the global carbon cycle by the production of durable wooden goods in each country. This is only an estimate because portions of other forest products are also sequestered (e.g., books in libraries, pit props, utility poles), and portions of saw and veneer logs are consumed (e.g., wastewood, disposal of plywood sheets used in concrete form building). This should lead to a small underestimate of total CO2 emissions because it includes logs from areas not counted as deforested. The methods for estimating emissions from land use change, suggested at the OECD experts meeting, require data and research into processes that do not yet exist. The method used here parallels that found in the work of R.A. Houghton, R.D.
Boone, J.R. Fruci et al., The Flux of Carbon from Terrestrial Ecosystems to the Atmosphere in 1980 Due to Changes in Land Use: Geographic Distribution of the Global Flux, Tellus, Vol. 39B, No. 1 2 (1987), pp. 122 139, which has been peer reviewed.

Choices must be made regarding the exact parameters to use in these calculations, but the deforestation and carbon density measures used here are the best general data available. The parameters used for this calculation were based on consistent definitions and common data sources.
Even if slightly lower values were used for deforestation and biomass per area, the magnitude of carbon emissions would remain about the same. These estimates are thus a good first approximation to current (i.e., circa 1991) emissions that result from land use change. There is some suggestion that northern temperate and boreal forest areas are net sinks for atmospheric carbon, but this, too, is controversial.

The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994), p. ES-9) estimated that the sources of CH4 shown in this data table solid waste, coal production, oil and gas production, wet rice agriculture, and livestock together make up about 72 percent of total global anthropogenic emissions. The remaining sources are less tractable to reasonable national- level estimates and include biomass burning, liquid wastes, livestock manure, and minor industrial sources.

CH4 emissions from municipal solid waste were calculated by multiplying the 1993 urban population by per capita emission coefficients developed for each country by H.G.
Bingemer and P.J. Crutzen in The Production of CH4 from Solid Wastes, Journal of Geophysical Research, Vol. 92, No. D2 (1987), pp. 2,181 2,187; and by S.D. Piccot et al., in Evaluation of Significant Anthropogenic Sources of Radiatively Important Trace Gases (Office of Research and Development, U.S. EPA, Washington, D.C., 1990), cited in OECD, Estimation of Greenhouse Gas Emissions and Sinks (OECD, Paris, August 1991), taking into account the proportion landfilled and its degradable organic carbon content. R.J. Cicerone and R.S. Oremland, Biogeochemical Aspects of Atmospheric Methane, Global Biogeochemical Cycles, Vol. 2, No. 4 (December 1988), pp. 299 327, suggest a likely range for annual world emissions from landfills of 30 million to 70 million metric tons. The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994)) estimates total emissions from solid waste at 57 million metric tons using a method substantially similar to that recommended by the OECD, or between 19 million and 39 million metric tons using a specialized regression model. The method used in this table parallels that recommended at the OECD experts meeting.

CH4 from coal mining was estimated using information on the average CH4 content of anthracite and bituminous coals, subbituminous coals, and lignite mined (WEC) in each country of the world. This latter data set is updated only every 3 years, and so the most recent year for which the necessary data are available is 1993. Less detailed data sets are available, but are inadequate to the task. This estimate assumed that 100 percent of the CH4 in extracted coal was emitted, although this is a slight exaggeration. CH4 is emitted from mines in larger quantities than would be accounted for by the CH4 content of the coal removed although in the long run, the CH4 in an extractable deposit of coal will be emitted, on average, at the rate that it is mined. CH4 trapped within the rock is released by mining, and this is one of the hazards of underground coal mining. Cicerone and Oremland (Aspects of Atmospheric Methane ) show a likely range of 25 million to 45 million metric tons of CH4 emitted annually in the course of mining coal. The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions:
Estimates for 1990 (EPA 230-R-93-010, January 1994)) reports its best estimate of total emissions from the coal fuel cycle as between 24.4 million and 39.6 million metric tons, and its global average estimate as between 19.4 million and 57 million metric tons. No data set exists that would allow internationally comparable estimates using a methodology suggested by the OECD in its report on the experts meeting.

Substantial quantities of CH4 are released to the atmosphere in the course of oil and gas production and distribution. CH4 vented in the course of oil production is estimated at 25 percent of the amount that is flared (Gregg Marland, CDIAC (personal communication), 1990). Estimates of CO2 from gas flaring in Data Table 14.1 also include gas that is vented (see also Barns and Edmonds, p. 3.9). CH4 emissions from natural gas production were estimated at 0.5 percent of production (Barns and Edmonds, pp. 3.2 3.3).
Recent estimates are that no more than 1 percent of CH4 is lost through leakage from distribution systems in the United States (AGA, Natural Gas and Climate Change: The Greenhouse Effect ), and no more than 1.7 percent in the former Soviet Union (Makarov and Basmakov, The Soviet Union: A Strategy for Energy Development with Minimum Emission of Greenhouse Gases ), although careful surveys have not been done. There is reason to believe that pipeline leaks in the former Soviet Union are grossly understated although the volume of gas produced in the former Soviet Union is sometimes mistakenly overstated but no other estimates exist. For these estimates, the U.S. experience was extended to Western Europe, Canada was counted as half the U.S. rate, and the Soviet estimate was used for Central Europe and the developing world because their situations were thought to be similar (S. Hobart et al., Methane Leakage from Natural Gas Operations ). Cicerone and Oremland (Aspects of Atmospheric Methane ) suggest a likely range of 25 million to 50 million metric tons of CH4 emitted because of leaks associated with natural gas drilling, venting, and transmission. The OECD experts meeting developed a general conceptual model of how to estimate emissions from these production and distribution systems, but it was unable to identify data on the factors leading to emissions or any individual data source for this purpose. The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions:
Estimates for 1990 (EPA 230-R-93-010, January 1994)) estimates the total emissions from oil and gas production and natural gas processing, transport, and distribution as between 30.3 million and 65.9 million metric tons.

CH4 produced from the practice of wet rice agriculture was calculated based on the area of rice production (as reported by the FAO, Agrostat-PC , FAO, Rome, 1993), subtracting those areas devoted to dry (upland), tidal, and deepwater (floating) rice production in each country or, in the case of China and India, in each province (Dana G. Dalrymple, Development and Spread of High-Yielding Rice Varieties in Developing Countries, Bureau of Science and Technology, U.S. Agency for International Development, Washington, D.C., 1986; and Robert E. Huke, Rice Area by Type of Culture:
South, Southeast, and East Asia, International Rice Research Institute, Los Banos, Laguna, Philippines, 1982).
This estimate follows the method suggested in the OECD experts meeting report and calculates the number of days of rice cultivation and the percentage of total rice area in each crop cycle by country or, in the case of China and India, by province (Elaine Mathews, Inez Fung, and Jean Lerner, Methane Emissions from Rice Cultivation: Geographic and Seasonal Distribution of Cultivated Areas and Emissions, Global Biogeochemical Cycles, Vol. 5, pp.
3 24).

There are many different studies of CH4 emissions from wet rice agriculture. In the past, many of these studies had been criticized because they had been undertaken on temperate rices grown in North America or Europe. Recently published studies based on similar rigorous methods from subtropical China have dispelled some of that criticism.
Studies using similar methodologies from India suggest lower emissions. Country-specific emission factors used here are from a review of the extant literature published by the U.S.
EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994), pp. 3 22).
Emission factors for all other countries were derived from the OECD experts meeting report, which recommended using a range of emissions found in a study of China (0.19 to 0.69 gram of CH4 per square meter per day; H. Schtz, W. Seiler, and H. Rennenberg, presented by H. Rennenberg at the International Conference on Soils and the Greenhouse Effect, August 14 18, 1989, Wageningen, the Netherlands, reported at the OECD experts meeting). The estimate here used the midpoint of that range (0.44 gram of CH4 per square meter per day), assuming that this range is an unbiased estimate of the normally distributed range of CH4 emissions.
Alternate estimates are possible. Where known, the amount of rainfed wet rice agriculture used emission factors that were 60 percent of those reported for irrigated rice.

A 2-year study in the subtropical rice bowl of China (Szechuan province) produced an estimated median flux (from some 3,000 flux estimates) of about 1.2 grams of CH4 per square meter per day, and a mean flux of 1.39 grams of CH4 per square meter per day (M.A.K. Khalil et al., Methane Emissions from Rice Fields in China, Environmental Science and Technology, Vol. 25, No. 5, pp. 979 981). Studies in Europe and North America seem to support the range suggested at the OECD experts meeting. (See the sources for more information.) In general, estimates of CH4 flux are based on a technique that captures CH4 produced anaerobically before the growth of the rice plant as well as the bulk of the CH4 produced that is transported through the rice plant throughout the growing period. Growing periods, temperature, and the type of rice cultivar, fertilizers, and possibly pesticides could influence methanogenesis. In the tropics, using modern varieties of rice, sufficient fertilizer, and adequate water, two or even three rice crops per year are possible.

Rice cultivation uses common techniques in both temperate and tropical climes even if the cultivars are not well adapted. The preparation of the impoundments wherein wet rice is grown the creation of a hardpan overlain by soft anaerobic muck creates similar environmental and chemical regimes wherever it occurs. Nonetheless, variations in water quality, soil, ambient temperature, precision of water control, and presence of cultivated algae or fish could also affect the total flux of CH4 .

Wet rice agriculture is practiced under four main water regimes: irrigated (52.8 percent of the world s total rice area), rainfed (similar to irrigated, 22.6 percent of the total), deep water (often dry in the early part of the season, may be planted to floating rice, 8.2 percent of the world s rice area), and tidal (3.4 percent of the total area). Cicerone and Oremland (Aspects of Atmospheric Methane ) suggest a likely range of 60 million to 170 million metric tons for CH4 emissions associated with wet rice agriculture. The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions:
Estimates for 1990 (EPA 230-R-93-010, January 1994)) estimates total global CH4 emissions at 65 million metric tons.

CH4 emissions from domestic livestock were calculated using FAO statistics on animal populations and published estimates of CH4 emissions from each type of animal. The animals studied included cattle and dairy cows, water buffalo, sheep, goats, camels, pigs, and caribou. P.J.
Crutzen, I. Aselmann, and W. Seiler ( Methane Production by Domestic Animals, Wild Ruminants, Other Herbivorous Fauna, and Humans, Tellus, Vol. 38B (1986), pp. 271 284) estimated CH4 production from animals on the basis of energy intake under several different management methods for several different feeding regimes. These findings were extended and further refined by the U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994)). These emission coefficients were then assigned to each country by animal type, based on the specifics of that country s animal husbandry practices and the nature and quality of feed available. Cicerone and Oremland s Aspects of Atmospheric Methane shows a likely range of 65 million to 100 million metric tons of CH4 emissions from enteric fermentation in domestic animals. The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994)) estimates total global emissions of 79.8 million metric tons. Alternate methods of estimation, such as a complex modeling method suggested in the OECD report, are not yet possible because of the lack of basic data.

A major anthropogenic source of CH4 , unaccounted for here, are emissions consequent to the burning of biomass.
Extensive biomass burning, especially in the tropics, is believed to release large amounts of CH4 . Cicerone and Oremland (Aspects of Atmospheric Methane ) put the likely range of those emissions at 50 million to 100 million metric tons. The OECD experts meeting elaborated on the absence of data needed for countries to estimate CH4 emissions from biomass burning. The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994)) estimates emissions from biomass burning at 48 million metric tons. In addition, it estimates CH4 emissions from liquid wastes at 35 million metric tons and emissions from livestock manure at 14 million metric tons.

Other natural sources of CH4 include wetlands, methane hydrate destabilization in permafrost, termites, freshwater lakes, oceans, and enteric emissions from other animals.
Natural sources account for an estimated 25 percent of all CH4 emissions. Cicerone and Oremland (Aspects of Atmospheric Methane ) estimate likely ranges of CH4 emissions at 100 million to 200 million metric tons from natural wetlands, 10 million to 100 million metric tons from termites, 5 million to 25 million metric tons from the oceans, 1 million to 25 million metric tons from fresh water, and possibly 5 million metric tons (potentially rising to 100 million metric tons if temperatures increase in the high arctic) from methane hydrate destabilization.
The U.S. EPA Office of Policy, Planning, and Evaluation (International Anthropogenic Methane Emissions: Estimates for 1990 (EPA 230-R-93-010, January 1994)) citing J.
Lelieveld and P.J. Crutzen ( Methane Emissions into the Atmosphere: An Overview, in A.R. van Amstel, ed., Methane and Nitrous Oxide, Methods in National Emissions Inventories and Options for Control (Proceedings of an International IPCC Workshop, 3 5 February 1993, Amersforrt, Netherlands, RIVM, Bilthoven, Netherlands, 1993, pp. 17 25)) estimate wetland sources at 125 million metric tons; termite emissions, 30 million metric tons; fresh water and the oceans, 15 million metric tons; and methane hydrate destabilization, 5 million metric tons.

Data Table 14.3
Atmospheric Concentrations of Greenhouse and Ozone-Depleting Gases, 1970 94

Sources: Carbon dioxide: Charles D. Keeling, Scripps Institution of Oceanography, Carbon Dioxide Information Analysis Center (CDIAC), Environmental Sciences Division, Oak Ridge National Laboratory, Atmospheric CO2 Concentrations Mauna Loa Observatory, Hawaii, 1958 1994, ORNL/CDIAC-25, NDP-001/R5 (an accessible numerical database) (Oak Ridge, Tennessee, September 1995); and C.D. Keeling and T.P. Whorf, Atmospheric CO2 records from sites in the SIO sampling network, in T.A. Boden, D.P. Kaiser, R.J. Sepanski et al., eds., Trends 93: A Compendium of Data on Global Change (ORNL/CDIAC-65, CDIAC, Oak Ridge, Tennessee, 1994), pp. 16 26. Other trace gases: CDIAC, Environmental Sciences Division, Oak Ridge National Laboratory, ORNL/CDIAC-25, DB- 1001 (an accessible numerical database); the Internet (ALE/GAGE Monthly Readings at Cape Grim, Tasmania); and originally R.G. Prinn et al., Atmospheric CFC-11 (CCl3 F), CFC-12 (CCl2 F2 ), and N2 O from the ALE-GAGE network, in T.A. Boden et al., eds., Trends 93: A Compendium of Data on Global Change (ORNL/CDIAC-65, CDIAC, Oak Ridge, Tennessee, 1994), pp. 396 420.

The trace gases listed here affect atmospheric ozone, contribute to the greenhouse effect, or both. Carbon dioxide (CO 2 ) accounts for about half the increase in the greenhouse effect and is emitted to the atmosphere by natural and anthropogenic processes. See the Technical Notes for Data Tables 14.1 and 14.2 for further details.

Atmospheric CO2 concentrations are monitored at many sites worldwide; the data presented here are from Mauna Loa, Hawaii (19 32 N, 155 35 W). Trends at Mauna Loa reflect global trends, although CO2 concentrations differ significantly among monitoring sites at any given time. For example, the average annual concentration at the South Pole in 1988 was 2.4 parts per million (ppm) lower than at Mauna Loa.

Annual mean values disguise large daily and seasonal variations in CO2 concentrations. The seasonal variations are caused by photosynthetic plants storing larger amounts of carbon from CO2 during the summer than in the winter.
Some annual mean values were derived from interpolated data.

Data are revised to correct for drift in instrument calibration, hardware changes, and perturbations to background conditions. Details concerning data collection, revisions, and analysis are contained in C.D. Keeling et al., Measurement of the Concentration of Carbon Dioxide at Mauna Loa Observatory, Hawaii, Carbon Dioxide Review: 1982, W.C. Clark, ed. (Oxford University Press, New York, 1982).

Data for all other gases are from values monitored at Cape Grim, Tasmania (45 41 S, 144 41 E) under the Atmospheric Lifetime Experiment (ALE) and Global Atmospheric Gases Experiment (GAGE). Although gas concentrations at any given time vary among monitoring sites, the data reported here reflect global trends. Cape Grim generally receives unpolluted air from the Southeast and is the ALE/GAGE station with the longest, most complete data set. Air samples were collected 4 times daily for ALE and 12 times daily for GAGE. The annual values shown here are averages of monthly values calculated by CDIAC. Missing values were interpolated.

Carbon tetrachloride (CCl 4 ) is an intermediate product in the production of CFC-11 and CFC-12. It is also used in other chemical and pharmaceutical applications and for grain fumigation. Compared with other gases, CCl4 makes a small contribution to the greenhouse effect and to stratospheric ozone depletion.

Methyl chloroform (CH 3 CCl 3 ) is used primarily as an industrial degreasing agent and as a solvent for paints and adhesives. Its contribution to the greenhouse effect and to stratospheric ozone depletion is also small.

CFC-11 (CCl 3 F), CFC-12 (CCl 2 F 2 ), and CFC-113 (C 2 Cl 3 F 3 ) are potent depletors of stratospheric ozone. Together, their cumulative effect may equal one fourth of the greenhouse contribution of CO2 .

Total gaseous chlorine is calculated by multiplying the number of chlorine atoms in each of the chlorine-containing gases (carbon tetrachloride, methyl chloroform, and the chlorofluorocarbons) by the concentration of that gas.

Nitrous oxide (N 2 O) is emitted by aerobic decomposition of organic matter in oceans and soils, by bacteria, by combustion of fossil fuels and biomass (fuelwood and cleared forests), by the use of nitrogenous fertilizers, and through other processes. N2 O is an important depletor of stratospheric ozone; present N2 O levels may contribute one twelfth of the amount contributed by CO2 toward the greenhouse effect.

Methane (CH 4 ) is emitted through the release of natural gas and as one of the products of anaerobic respiration.
Sources of anaerobic respiration include the soils of moist forests, wetlands, bogs, tundra, and lakes. Emission sources associated with human activities include livestock management (enteric fermentation in ruminants), anaerobic respiration in the soils associated with wet rice agriculture, and combustion of fossil fuels and biomass (fuelwood and cleared forests). CH4 acts to increase ozone in the troposphere and lower stratosphere; its cumulative greenhouse effect is currently thought to be one third that of CO2 , but on a molecule-for-molecule basis, its effect, ignoring any feedback or involvement in any atmospheric processes, is 11 to 30 times that of CO2 .

Data Table 14.4
World CO2 Emissions from Fossil Fuel Consumption and Cement Manufacture, 1950 92

Source: Carbon Dioxide Information Analysis Center (CDIAC), Environmental Sciences Division, Oak Ridge National Laboratory, 1992 Estimates of CO2 Emissions from Fossil Fuel Burning and Cement Manufacturing Based on the United Nations Energy Statistics and the U.S. Bureau of Mines Cement Manufacturing Data, ORNL/CDIAC-25, NDP-030 (an accessible numerical database) (Oak Ridge, Tennessee, September 1995).

CDIAC calculates world emissions from data on the global production of fossil fuels (based on the World Energy Data Set maintained by the United Nations Statistical Office) and from data on world cement manufacturing (based on the Cement Manufacturing Data Set maintained by the U.S. Bureau of Mines). Emissions are calculated using global average fuel chemistry and usage. These data account for all fuels including bunker fuels not accounted for in Data Table 14.1, which are also shown separately. For further information, see the Technical Notes for Data Table 14.1.

Data Table 14.5
Common Anthropogenic Pollutants, 1980 93

Sources: Sulfur and nitrogen emissions: Hilde Sandnes and Helge Styves, Calculated Bud-gets for Airborne Acidifying Components in Europe 1985, 1987, 1988, 1989, 1990, and 1991 (Co-Operative Programme for Monitoring and Evaluation of the Long-Range Transmission of Air Pollutants in Europe (EMEP), The Norwegian Meteorological Institute, Technical Report No.
97, 1992), pp. 11 14. Sulfur, nitrogen, carbon monoxide, particulate matter, and volatile organic compounds: Economic Commission for Europe (ECE), Impacts of Long-Range Transboundary Air Pollution (ECE Air Pollution Studies 8, United Nations, New York, 1992), pp. 4 5; and Organisation for Economic Co-Operation and Development (OECD), OECD Environmental Data Compendium 1995 (OECD, Paris, 1995).

Emissions of sulfur in the form of sulfur oxides and nitrogen in the form of its various oxides together contribute to acid rain and adversely affect agriculture, forests, aquatic habitats, and the weathering of building materials. Sulfate and nitrate aerosols impair visibility.
These data on anthropogenic sources should be used carefully. Because different methods and procedures may have been used in each country, the best comparative data may be time trends within a country.

Sulfur dioxide (SO2 ) is created by natural as well as anthropogenic activities. High concentrations of SO2 have important adverse health effects, and there is particular concern about its effects on the health of young children, the elderly, and people with respiratory illnesses (e.g., asthma). SO2 in the presence of moisture contributes to acid precipitation as sulfuric acid.

Anthropogenic nitrogen oxides (NOx ) come mainly from industrial sources and contribute to the creation of photochemical smog and the production of tropospheric ozone an important greenhouse gas. All oxides of nitrogen also contribute to acid precipitation, in the form of nitric acid.

This data table combines data from EMEP, ECE, and OECD to compile as complete a picture as possible of sulfur and nitrogen emissions. EMEP is an activity of the 1979 Convention on Long-Range Transboundary Air Pollution. Data on sulfur and nitrogen emissions are submitted to EMEP and ECE by parties to the 1985 protocol on SO2 emissions and the 1988 protocol on nitrogen oxide emissions. Parties to these protocols are required to submit preliminary estimates of sulfur and nitrogen emissions by May of the year following the year being estimated, with final estimates due within a year after that. In the event that official data are missing, EMEP interpolates between years for which official data exist. In the event that this is not possible, EMEP will use its own or others emission estimates.

OECD polls its members on emissions with questionnaires that are completed by the relevant national statistical service or designee. OECD does not have any independent estimation capability.

EMEP and ECE report emissions in terms of the elemental content of sulfur, whereas OECD reports its emissions in terms of tons of oxides of sulfur. EMEP and ECE emission estimates were converted to their weight in SO2 . EMEP and OECD report nitrogen emissions in terms of nitrogen dioxide.
Please consult the sources for further information.

This data table also reports OECD data for carbon monoxide and particulate matter emissions and combines both EMEP and OECD data to describe the emissions of volatile organic compounds. Differences in definition can limit the comparability of these estimates.

Carbon monoxide (CO), is formed both naturally and from industrial processes, including the incomplete combustion of fossil and other carbon-bearing fuels. Automobile emissions are the most important source of CO, especially in urban environments. CO interferes with oxygen uptake in the blood, producing chronic anoxia leading to illness or, in the case of massive and acute poisoning, even death. CO also scavenges hydroxyl radicals that would otherwise contribute to the removal of methane a potent greenhouse gas from the atmosphere.

The health effects of particulate matter (PM) are in part dependent on the biological and chemical makeup and activity of the particles. Heavy metal particles or hydrocarbons condensed onto dust particles can be especially toxic. PM arises from numerous anthropogenic and natural sources.
Among the anthropogenic sources are combustion, industrial and agricultural practices, and the formation of sulfates from SO2 emissions.

In the presence of sunlight, volatile organic compounds (VOCs) are, along with oxides of nitrogen, responsible for photochemical smog. Anthropogenic emissions of VOCs arise in part from the incomplete combustion of fuels or the evaporation of fuels, lubricants, and solvents, as well as from the incomplete burning of biomass. These data combine VOC emission data from OECD with VOC data from EMEP.

Data Table 14.6
Inventories of National Greenhouse Gas Emissions, 1990

Sources: Intergovernmental Negotiating Committee for a Framework Convention on Climate Change, Matters Relating to Commitments First Review of Information Communicated by Each Party Included in Annex I of the Convention (United Nations General Assembly, A/AC.237/81, New York, 7 December 1994); and Intergovernmental Negotiating Committee for a Framework Convention on Climate Change, Matters Relating to Commitments First Review of Information Communicated by Each Party Included in Annex I of the Convention (United Nations General Assembly, A/AC.237/WP/1, New York, 6 February 1995).
Poland: Ministry of Environmental Protection, Natural Resources, and Forestry, National Report to the First Conference of the Parties to the United Nations Framework Convention on Climate Change (Ministry of Environmental Protection, Natural Resources, and Forestry, Warsaw, 1994).
Italy: Ministry of Environment and Ministry of Industry, National Programme for the Limitation of Carbon Dioxide Emissions to the 1990 Levels by the Year 2000 (Ministry of Environment and Ministry of Industry, Rome, 1994).

As part of their responsibilities under the Framework Convention on Climate Change, each party listed in Annex I of the Convention must submit (within 6 months of the Convention entering into force) information that includes inventories of national emissions of greenhouse gases other than those controlled by the Montreal Protocol for the Protection of the Ozone Layer. The first of these inventories came due on September 21, 1994. Parties were asked to use guidelines created by the Intergovernmental Panel on Climate Change when preparing their inventories of 1990 emissions, so as to enhance comparability. These inventories are, in fact, detailed estimates of emissions and not inventories as the word is commonly understood.
Estimates o f other gas emissions at the national level (e.g., those contained in Data Table 14.5) have been shown to be highly labile due to changes in the understanding of the underlying data, changes in the methods used in estimation, and even changes in the extent of the phenomenon under study. Variations of 30 percent or 40 percent from one estimate of a particular year s emissions to a later estimate of the same year s emissions are not unheard of.

Additional information on the gases contained in this table, and their sources, can be found in the Sources and Technical Notes to Data Table 14.2.

Although estimates of carbon dioxide (CO2 ) emissions from fossil fuel use and industrial processes are similar in this table and in Data Table 14.1, these inventories differ more extensively from estimates of emissions found in Data Table 14.2. In Data Table 14.2, no attempt was made to estimate sinks for CO2 as shown in this data table for land use change. Differences also exist between the two data tables in the estimates of methane (CH4 ) emissions. These two data tables provide different views of the same phenomenon and together give a better picture than either could alone. Data Table 14.2 provides a global picture with similar methods and a common source for estimation parameters. In contrast, this data table provides national estimates of each nation s own emissions. These inventories use nationally specific details and parameters to come up with these inventory estimates.

CO2 emissions from fossil fuels include emissions from combustion and other industrial processes. CO2 emissions from land use change are estimates of the emissions associated with the clearing of land or increases in forest cover or forest biomass. Net emissions sums energy use and negative or positive emissions from forest growth.

Methane emissions from oil and gas systems include both emissions from combustion and emissions from venting and leakage from oil and gas production and distribution systems. CH4 emissions from livestock include both enteric fermentation and animal waste. Other agricultural sources include wet rice agriculture, CH4 released from soils, and the burning of agricultural waste and grazing lands. Waste includes emissions from landfills and other includes emissions from industrial processes and land use change. See the Technical Notes to Data Table 14.2 for further information.

Nitrous oxide (N2 O) is another potent greenhouse gas that is difficult to model. In descending order of importance, the primary sources of N2 O are agriculture, industry, and energy use for transport.