2009 Volume 87 Issue 4 Pages 635-663
We have used an AGCM (atmospheric general circulation model)-based Chemistry Transport Model (ACTM) for the simulation of methane (CH4) in the height range of earth’s surface to about 90 km. The model simulations are compared with measurements at hourly, daily, monthly and interannual time scales by filtering or averaging all the timeseries appropriately. From this model-observation comparison, we conclude that the recent (1990-2006) trends in growth rate and seasonal cycle at most measurement sites can be fairly successfully modeled by using existing knowledge of CH4 flux trends and seasonality. A large part of the interannual variability (IAV) in CH4 growth rate is apparently controlled by IAV in atmospheric dynamics at the tropical sites and forest fires in the high latitude sites. The flux amplitudes are optimized with respect to the available hydroxyl radical (OH) distribution and model transport for successful reproduction of latitudinal and longitudinal distribution of observed CH4 mixing ratio at the earth’s surface. Estimated atmospheric CH4 lifetime in this setup is 8.6 years. We found a small impact (less than 0.5 ppb integrated over 1 year) of OH diurnal variation, due to temperature dependence of reaction rate coefficient, on CH4 simulation compared to the transport related variability (order of ±15 ppb at interannual timescales). Model-observation comparisons of seasonal cycles, synoptic variations and diurnal cycles are shown to be useful for validating regional flux distribution patterns and strengths. Our results, based on two emission scenarios, suggest reduced emissions from temperate and tropical Asia region (by 13, 5, 3 Tg-CH4 for India, China and Indonesia, respectively), and compensating increase (by 9, 9, 3 Tg-CH4 for Russia, United States and Canada, respectively) in the boreal Northern Hemisphere (NH) are required for improved model-observation agreement.