Estimating methane sources and sinks by assimilating satellite data in a global atmospheric inverse system.

Methane (CH4) is the second most important greenhouse gas, contributing to approximately 30% of the additional greenhouse effect since 1750. Its varied sources and relatively short lifetime in the atmosphere (~9 years) offer interesting mitigation opportunities. To develop practical strategies for mitigating climate change, precise quantification of methane fluxes and a better understanding of its spatial distribution and biogeochemical cycling are imperative. The observations currently used to infer methane sources and sinks face limitations affecting calculation accuracy. Surface stations measuring CH4 are sparse and notably absent in major emitting regions. In contrast, satellite-derived data, while providing broader coverage, present systematic errors and estimate atmospheric composition with an accuracy range of 1-10%. Additionally, passive satellite shortwave infrared (SWIR) measurements exhibit higher sensitivity near surface emission sources but are less effective in high latitude regions. Conversely, passive satellite thermal infrared (TIR) measurements have a higher sensitivity between the free troposphere and the stratosphere.Current worksare currently being developed to integrate TIR and SWIR to obtain consolidated CH4 information on the vertical atmospheric profile. This studyaims on improving methane flux estimates using the top-down approach, which integrates observations, flux priors, and an atmospheric chemical transport model utilizing Bayesian methodology. This will be perfomed on the inversion system developed at the LSCE (Community Inversion Framework CIF) using the global transport model LMDz. We analyze the information provided by different observing systems (TIR, SWIR and surface network) at the global scale and for a period between June 2018 and June 2020. In a first step, the sensitivity of the fluxes to the observations is estimated. In a second step, Observing System Simulation Experiments are performed to evaluate the performance of the different observations system to retrieve the target fluxes. Considering both steps, observing systems are chosen to provide the best information in terms of sensitivity and spatial representation (vertical and horizontal).