A31B-0038
Decadal Trends in Global CO Emissions as Inferred from Atmospheric Inversion
Wednesday, 16 December 2015
Poster Hall (Moscone South)
Yi Yin1, Frédéric Chevallier1, Philippe Ciais2, Gregoire Broquet1, Audrey Fortems Cheiney3, Isabelle Pison1 and Marielle Saunois1, (1)LSCE Laboratoire des Sciences du Climat et de l'Environnement, Gif-Sur-Yvette Cedex, France, (2)CNRS, Paris Cedex 16, France, (3)Institut National de l’Environnement Industriel et des RisqueS (INERIS), Verneuil-en-Halatte, France
Abstract:
Large negative trends of atmospheric CO concentrations in the recent decade are observed by both surface measurements and satellite retrievals over many regions globally. However, this decrease in CO concentration cannot be fully explained by current emission inventories. Here, we aim at attributing the observed CO concentration decline using an atmospheric inversion that simultaneously optimizes CO sources (from surface emissions and atmospheric oxidations of hydrocarbons) and CO sinks (through reaction with oxidant hydroxyl radical OH in the atmosphere) with observational constraints of CO and other chemically related trace gases. Satellite retrievals of CO total column from the latest product v6 of MOPITT sensor together with surface in-situ measurements of methane and methyl chloroform are assimilated for the period from 2002 to 2011. The optimized concentrations are evaluated with independent observations. Sensitivity tests of ACTM latitudinal/vertical resolution, prior OH fields, and additional observational constraints are discussed. The optimized CO budgets indicate that the CO concentration decline is primarily caused by a decrease in the CO emissions by 3% yr-1, whereas no significant trends are found at the global scale for the chemical CO sources and for the OH concentrations that regulate CO sinks. We highlighted some regional inconsistencies between bottom-up and top-down results. In contrast to bottom-up inventories, the inversion finds negative trends over China and positive trends over Africa.