Influences of Uncertainties in the STT Flux on Modeled Tropospheric Methane

Methane (CH4) is the second most important greenhouse gas. At the global scale, inverse modeling is usually applied to infer CH4 fluxes on the surface. In the inverse model, the chemistry transport model (CTM) is used to link CH4 fluxes to its concentration in the atmosphere. Any uncertainties in the transport and chemical processes modeled by the CTM can lead to biases in the inferred fluxes. Therefore, diagnosing transport processes in the CTM is important for improving the accuracy of inferred fluxes. It is well-known that the inverse model-calculated total columns of CH4 contain latitude-dependent biases when compared to observations. Previous studies revealed that these biases mostly occur in the troposphere. However, we demonstrate in the present study that the model biases in the troposphere can originate in the stratosphere, especially in the lowermost stratosphere, and propagate downward to the troposphere. The propagated biases in the simulations of the atmospheric model TM5-CAMS are estimated to be about 25 ppb in the mid-troposphere above the northern high-latitudes and about 2 ppb over the tropics. At the surface the propagated biases are estimated to be about 20 and 7 ppb in the above two regions, respectively. In addition, the propagated biases display an important zonal asymmetry in the troposphere, especially over the Tibetan-Plateau, Southeast Asia and tropical South America. It is recommended that some correction functions should be considered in inverse modeling that use surface observations only.