Using Satellite Column Observations with a Bayesian Inversion System for constraining the GHG budget over India

The source-sink estimation of greenhouse gases and the accurate quantification of their flux distributions are a major scientific challenge of our times. Although statistically developed inventory-based carbon emissions improve understanding of emissions, they are uncertain and occasionally, they do not reflect actual emissions, especially at finer scales. By using GHG observations in conjunction with numerical atmospheric models, GHG budgets can be constrained. However, the sparseness in the in-situ observation networks of GHGs, makes this a challenging problem, especially in certain regions like India. However, satellite remote sensing can fill this data gap. This study applies the Bayesian inversion system, FLEXINVERT (Thompson and Stohl 2014), coupled with a Lagrangian Particle Dispersion Model, FLEXPART (Stohl et al 1998), to assimilate TROPOMI satellite observations of CH₄ and constrains the CH₄ fluxes over India domain. This inverse modeling system uses the Source-Receptor Relationship derived from Flexpart as the Transport operator while minimizing the cost function for optimization.  The system will use various prior fluxes and initial concentration fields to test the set-up over Indian domain, which have strong fluxes of CH₄ and are not well constrained by the existing ground-based measurement networks. The results will be compared with the limited in-situ observations available for the region.