Interpretation of methane-ethane-acetylene-carbon monoxide correlations in atmospheric observations as constraints on sources of methane and chemical aging

I demonstrate in this thesis that quantitative interpretation of the observed correlations between methane (CH 4 ), ethane (C 2 H 6 ), acetylene (C 2 H 2 ), and CO with a global 3-D chemical transport model (GEOS-Chem) offers unique and previously unexploited constraints on methane emissions and chemical aging processes. Bottom-up global emission inventories of C 2 H 6 were developed and tested with GEOS-Chem model simulations of extensive datasets from surface in-situ and column stations as well as aircraft campaigns. No major model bias is seen except a 20-30% overestimate at European surface stations. I show that the U.S. EPA emission inventory (NEI-99) for the C 2 H 6 source from natural gas in the south-central U.S. is a factor of 7 too low. This bias would also affect national emission inventories for the fossil fuel source of CH 4 , and implies that a major U.S. contribution to greenhouse warming is underestimated. The CH 4 -C 2 H 6 -CO correlations were explored for the TRACE-P aircraft campaign (NW Pacific, spring 2001) to provide constraints on sources of CH 4 . The observed CH 4 concentration enhancements and CH 4 -C 2 H 6 -CO correlations in Asian outflow in TRACE-P are determined mainly by anthropogenic emissions from China and Europe, with only little contribution from tropical sources (wetlands and biomass burning). Matching both the observed CH 4 enhancements and the CH 4 -C 2 H 6 -CO slopes in Asian outflow requires increasing the East Asian anthropogenic source of CH 4 , and decreasing the European anthropogenic source, by at least 30% for both from the a priori inventory.

The global C 2 H 2 budget was quantified through GEOS-Chem evaluation with surface and aircraft observations. The C 2 H 2 -CO relationships in linear and log space from aircraft observations in different parts of the world were examined, and interpreted with simple models as well as with GEOS-Chem. I find that the slope of the linear regression of [C 2 H 2 ] vs. [CO] is not a robust measure of air mass aging because of the variability in background air. The log[C 2 H 2 ] vs. log[CO] slope measures the relative importance of dilution and chemistry in driving the air mass aging. GEOS-Chem sensitivity simulations show that the aging is dominated by dilution in fresh outflow but by chemical loss in remote air.