Efficient use of a Lagrangian particle dispersion model for atmospheric inversions using satellite observations of column mixing ratios

Satellite instruments for measuring atmospheric column mixing ratios have improved significantly over the past couple of decades, with increases in pixel resolution and accuracy. As a result, satellite observations are being increasingly used in atmospheric inversions to improve estimates of emissions of greenhouse gases (GHGs), particularly CO2 and CH4, and to constrain regional and national emission budgets. However, in order to make use of the increasing resolution in inversions, the atmospheric transport models used need to be able to represent the observations at these finer resolutions. Here, we present a new and computationally efficient methodology to model satellite column average mixing ratios with a Lagrangian particle dispersion model (LPDM) and calculate the Jacobian matrices describing the relationship between surface fluxes of GHGs and atmospheric column average mixing ratios, as needed in inversions. The development will enable a more accurate representation of satellite observations (especially high-resolution ones) via the use of LPDMs and, thus, help improve the accuracy of emission estimates obtained by atmospheric inversions. We present a case study using this methodology in the FLEXPART (FLEXible PARTicle dispersion model) LPDM and the FLEXINVERT inversion framework to estimate CH4 fluxes over Siberia using column average mixing ratios of CH4 (XCH4) from the TROPOMI (TROPOspheric Monitoring Instrument) instrument aboard the Sentinel-5P satellite. The results of the inversion using TROPOMI XCH4 are evaluated against results using ground-based observations.