Atmosphere-based estimates of non-CO2 greenhouse gas emissions for the U.S. derived from 14CO2 during 2009-2012.

Atmospheric measurements of trace gases with ‘known’ emissions provide a means to derive emission magnitudes of other simultaneously measured trace gases, provided sources are co-located and co-varying. Here we consider atmospheric mixing ratio covariations in the fossil fuel derived component of observed CO₂ (C_ff; derived from high-precision measurements of the radiocarbon fraction of atmospheric CO₂) relative to more than 20 other anthropogenic trace gases including CO, CH₄, N₂O, SF₆, and halo- and hydro-carbons over large industrialized land areas. Pairing C_ff with boundary-layer concentration enhancements of these gases allows us to determine apparent emission ratios for each gas with respect to C_ff. When combined with sample-specific model-derived spatial footprints and the relatively accurate U.S. inventory of fossil fuel emissions (i.e., estimated uncertainty of ±10%), absolute emission rates for the correlate gases are derived. Here we will present U.S. annual emission magnitudes for select gases based on year-round measurements from tall towers and aircraft profiling sites in California, Texas, the mid-west, south-east and north-east for the 2009-2012 period. Statistically significant and coherent spatial and seasonal patterns in emission ratios and absolute emissions are determined for many gases based on these measurements. For HFC-134a and HCFC-22, results derived with this approach generally agree very well with an independent Bayesian-inversion based analysis of the larger number of samples collected and analyzed in our network, but that are not paired with C_ff measurements. We believe this approach provides reliable 'top down', observationally-based emission estimates for these gases, many of which influence climate, air quality and stratospheric ozone.