[ACG35-P06] Temporal variations of the global CH4 sources estimated by mole fraction, carbon and hydrogen isotope ratios of atmospheric CH4, and an atmospheric chemistry transport model
キーワード：メタン、同位体、逆解析
There still remains a large uncertainty on the relative contributions of CH4 sources, broadly defined as biogenic (wetlands, rice paddies, ruminants, termites, and landfills), fossil fuel (coal, oil, natural gas, and geological seepage), and biomass burning, to the global CH4 budget (e.g., Saunois et al., 2016). Carbon and hydrogen isotope ratios of atmospheric CH4 (δ13C and δD) allow for better source apportionment and therefore help reduce the uncertainty. In this study, we performed inverse modeling of the observed atmospheric CH4 mole fractions to estimate surface CH4 fluxes for 1995–2013, using the NIES global atmospheric tracer transport model (NIES-TM) with a priori CH4 fluxes and CH4 sink fields. Forward simulations of the CH4 mole fraction, δ13C, and δD were further conducted by using the a posteriori (optimized by the inversion) CH4 fluxes. The δ13C and δD, thus obtained, were compared to those observed at two polar surface stations, Ny-Ålesund, Svalbard (78°55'N, 11°56'E; site code: NAL) and Syowa Station, Antarctica (69°00'S, 39°35'E; site code: SYO).
Variations of the atmospheric CH4 mole fraction simulated using the a posteriori CH4 fluxes reproduce the observational results fairly well, not only at the sites where the CH4 data were incorporated into the inversion (104 sites in the global, including NAL and SYO) but also at other independent sites, for example, in the Western Pacific. This suggests that the CH4 fluxes are well constrained by this inverse modelling at least regional to global scales. However, forward simulations of δ13C and δD using the a posteriori CH4 fluxes and the respective isotopic source signatures significantly underestimate the observed δ13C and δD values globally. It is indicated that the present a posteriori CH4 fluxes from biogenic sources and those from fossil fuel and/or biomass burning were overestimated and underestimated, respectively. By constraining the CH4 fluxes by δ13C and δD values observed at NAL and SYO, the agreements between simulated and observed CH4, δ13C, and δD are much improved not only at the two sites, but also in the Western Pacific. The relative contributions of biogenic, fossil fuel, and biomass burning sources to the global CH4 emissions are 62, 30, and 8% for 2003–2012. These values are not in complete agreement with the range of recent top-down estimates, but comparable to an estimate based on global atmospheric δ13C data (Schwietzke et al., 2016). Our model infers that biogenic and biomass burning CH4 emissions decreased in the first half of the 2000s and that biogenic CH4 emissions increased after 2006, which could be responsible for the complicated behavior in global atmospheric CH4 growth in recent decades, i.e., plateau in the early 2000s and re-rise after 2006.
Variations of the atmospheric CH4 mole fraction simulated using the a posteriori CH4 fluxes reproduce the observational results fairly well, not only at the sites where the CH4 data were incorporated into the inversion (104 sites in the global, including NAL and SYO) but also at other independent sites, for example, in the Western Pacific. This suggests that the CH4 fluxes are well constrained by this inverse modelling at least regional to global scales. However, forward simulations of δ13C and δD using the a posteriori CH4 fluxes and the respective isotopic source signatures significantly underestimate the observed δ13C and δD values globally. It is indicated that the present a posteriori CH4 fluxes from biogenic sources and those from fossil fuel and/or biomass burning were overestimated and underestimated, respectively. By constraining the CH4 fluxes by δ13C and δD values observed at NAL and SYO, the agreements between simulated and observed CH4, δ13C, and δD are much improved not only at the two sites, but also in the Western Pacific. The relative contributions of biogenic, fossil fuel, and biomass burning sources to the global CH4 emissions are 62, 30, and 8% for 2003–2012. These values are not in complete agreement with the range of recent top-down estimates, but comparable to an estimate based on global atmospheric δ13C data (Schwietzke et al., 2016). Our model infers that biogenic and biomass burning CH4 emissions decreased in the first half of the 2000s and that biogenic CH4 emissions increased after 2006, which could be responsible for the complicated behavior in global atmospheric CH4 growth in recent decades, i.e., plateau in the early 2000s and re-rise after 2006.