INVERSION OF THE GLOBAL METHANE CYCLE USING CHANCE-CONSTRAINED PROGRAMMING - METHODOLOGY AND RESULTS

This paper presents a new methodology for constraining sources and sin ks in mass balances and applies it to a global model of methane. Sourc es and sinks are treated as random variables that are constrained by i sotopic mass balance and ice-core data. A chance constrained programmi ng formulation is used to perform inverse calculations and determine c onstrained bounds on sources of methane. Budgets consistent with atmos pheric concentrations, flux measurements, and priors on model variable s and parameters are derived. It is observed that global budgets of me thane presented in the literature have greater uncertainty ranges than allowed by the constraints on its global cycle. The analysis shows th at emissions from rice paddies and wetlands may be lower than presente d in previous budgets, a result that is in agreement with recent estim ates based on flux measurements. Ranges for other anthropogenic source s such as biomass burning, and fossil and landfill emissions are also obtained. Sensitivity analysis is used to determine the key uncertain parameters in the global methane budget.