Estimates of Monthly CO2 Emissions and Associated 13C/12C Values from Fossil-Fuel Consumption in the U.S.A.

Estimates of monthly CO2 emissions and associated 13C/12C values from fossil-fuel consumption in the United States are available for the years 1981-2002. These estimates were derived from values of fuel consumed, multiplied by their respective thermal conversion factors, and expressed in units of heat energy derived from the fuel. [Thermal conversion factors are given in Appendix A of each issue of Monthly Energy Review (DOE/EIA, monthly).] These energy values were then multiplied by their respective carbon dioxide emission factors and expressed in units of the mass of carbon emitted per unit of energy liberated by the oxidation of the carbon in the fuel. [Carbon dioxide emission factors are given in Table A of EPA (2002).] Product specific emissions factors were used for the myriad of petroleum products. Finally, the results were multiplied by a factor (EPA, 2002, Table A-14, except for natural gas) to account for carbon not oxidized during combustion but released as soot, ash, or long-lived hydrocarbons.

Several petroleum products have applications not related to energy production and are not oxidized immediately when the products are consumed; these include asphalt, waxes, petroleum coke, special naphtha, etc. In some cases (e.g., waxes) virtually no carbon is oxidized during use; in other cases (e.g., petroleum coke) a substantial fraction (e.g., estimated 50%) of the carbon is oxidized during use; and for some products (e.g. special naphtha), virtually all the carbon is oxidized within the time frame of interest. The oxidized fractions used to estimate the amounts of carbon dioxide released during consumption are given in Annex B of EPA 2002 (Table B-1).

Although C emissions occur largely in the form of CO2, results are presented here in terms of the mass of C only. (To convert to mass of CO2, multiply the mass of C by 44/12). The authors assume that any C emitted to the atmosphere as CO will be soon oxidized to CO2.

13C values for each month were also estimated. Finally, differences from other CDIAC CO2 emissions estimates (e.g., Marland et al., 2002) are explained in the documentation.

These estimates show an annual cycle of CO2 emissions, peaking during the winter months and reflecting natural gas consumption, and a semi-annual cycle of lesser amplitude, peaking in summer and winter and reflecting coal consumption, comprise the dominant features of the annual pattern. There were relatively constant emissions until 1987, followed by an increase from 1987-1989, a decrease in 1990-1991, and record highs during the late 1990s; emissions have declined somewhat since 2000.