GC43D-1057: How a new uncertainty estimate of global carbon dioxide emissions from fossil fuel consumption affects our understanding of the source/sink nature of the terrestrial biosphere
Authors: Robert J Andres, Tom Boden
Author Institutions: Oak Ridge National Laboratory, Oak Ridge, TN, USA
A new evaluation of the uncertainty associated with the global total of carbon dioxide emissions from fossil fuel consumption and cement production is based on approaching the emissions estimates from a different perspective than the approach used by Marland and Rotty (1984, Tellus 36B: 232-261). This new evaluation is based on quantifying the qualitative national error classes shown in Andres et al. (1996, GBC 10:419-429) and assuming that the energy data for each nation are independent of each other nation’s data. Global total emissions for the years 1950 to 2011 are calculated from the sum of emissions from individual nations plus an additional term which accounts for emissions not included in national totals (e.g., bunker fuels, Andres et al., 2012, Biogeosci. 9:1845-1871). The uncertainty on this global total is then calculated from square root(Î£((emissions*uncertainty)^2)) where the summation is done over all nations. The result of this calculation is the uncertainty expressed in terms of mass in a given year. Calculations are now underway for the 1950-2011 time series. Based on previous calculations, for the years 1950 to 2010, the 2 Ïƒ uncertainty on global fossil fuel emissions was 2.6 to 4.8% (41 Tg C in 1950 and 415 Tg C in 2010, respectively). To examine one implication of this new uncertainty estimate, it is propagated into the rest of the global carbon cycle via a five-component box model. For simplicity, this uncertainty is accommodated entirely within the terrestrial biosphere. We conclude that for most years since 1950, the propagated uncertainty results in a terrestrial biosphere that may have acted as either a source or sink of carbon.