Intercomparison of Models to Estimate Methane Emissions From Rice Agriculture Using Common Data Sets
Rice agriculture is a major source of anthropogenic methane and a likely target of greenhouse gas mitigation efforts. The major factors that determine emissions are for the most part known and include agricultural management practices and climate. Despite this, the range of global source strength estimates remains large (~25 to 100 Tg CH4 y-1). Assessing how changes in these factors influence paddy emissions is important in understanding current and future budgets of methane, impacts of reduction strategies, and for verification of national inventory reports. A number of methods were used to determine national and global inventories of paddy methane including process models and empirically-based statistical models. Currently emissions from these models are significantly different at multiple spatial scales. Uncertainties are introduced to emission estimates through both the input data that drives the models, and the means by which the models calculate flux. These sources of error are difficult to disentangle in current inventories. Here we report results from an intercomparison of models that simulate methane emissions from rice agriculture performed under common data constraints. The models include the Denitrification-Decomposition process model, an empirical statistical model developed at Portland State University, and the UNFCCC recommended method for countries to report national emissions. Common input data sets that serve the models were prepared on a 0.5°×0.5° spatial grid across monsoon Asia and include harvested rice areas from 48 different rice cropping systems, crop residue inputs to paddies, water management practices, nitrogen fertilizer rates, and climate data. The simulations were performed over the years 2003 to 2005 to coincide with available column methane abundances from the satellite-borne SCIAMACHY instrument, which are used as an observational check on model simulations. By eliminating one major source of disagreement in the modeling effort, we can better assess where and why current methods disagree and improve these to provide better estimates of rice methane emissions.
EOS, Transactions American Geophysical Union, Fall Meeting, Supplement
American Geophysical Union Publications
Butenhoff, C., Frolking, S., Li, C., Houweling, S., Milliman, T., Khalil, A. and Zhuang, Q. (2009), Intercomparison of Models to Estimate Methane Emissions From Rice Agriculture Using Common Data Sets, Eos Trans. AGU, 90(52), Fall Meet. Suppl., Abstract A53C-0283.