Satellite-based CH4 emission monitoring based on the Community Inversion Framework (CIF): application to the Tropics

Atmospheric inversions are used by the global greenhouse gas (GHG) monitoring community for the estimation of GHG emission budgets both at global and regional scales. The Community Inversion Framework (CIF) brings novel inversion capabilities for emission monitoring, especially based on satellite data. The system is designed to enable easily deployable inversions using a large variety of inputs, with one of the key aspects of interest being the smooth integration of satellite data (from existing platforms such as S5P TROPOMI, or new ones as GOSAT-GW and S5) and their comparison to transport model simulations. To cover the needs of the inversion community, CIF is compatible with multiple inversion algorithms (variational and ensemble-based) and chemistry and transport models. These integrated features in a single system facilitate inter-comparison analyses and transparent and reliable policy-relevant reporting, consistent with WMO-promoted guidelines for inversion use in the UNFCCC context. 

In the present study, we showcase a satellite-based inversion system over tropical regions using the CIF-CHIMERE inversion setup. Domains of interest include Africa, India and Southeast Asia, and South America, which contribute approximately 14%, 23%, and 16%, respectively, to the total annual CH₄ emissions worldwide, according to the Global Methane Budget (Saunois et al., 2025). Thus, our system covers more than 50% of methane emissions worldwide. Still, CH₄ emission estimates over these regions remain largely uncertain due to the scarcity of observational data, as well as the complexity and diversity of emission processes. We use satellite CH₄ total column mixing ratios observation from TROPOMI that provides extensive spatial coverage over the Tropics to better constrain emissions in these regions. 

Inversion results are highly dependent on transport patterns, observational coverage, uncertainties, and reliable prior information. In this study, we assess the system capabilities in monitoring fluxes using a low-cost Monte-Carlo approach. We highlight potential and remaining gaps for future systematic application for emission monitoring.