Uncertainties associated with Lagrangian transport in greenhouse gas flux estimates

Atmospheric Lagrangian particle dispersion models (LPDMs) are commonly combined with Bayesian inversion/optimization methods to infer emission fluxes across spatial scales from local to global. These tools are central to monitoring greenhouse gases, especially CO₂, CH₄, and N₂O. However, uncertainties in flux estimates arise from multiple sources: prior flux information, representation of the background atmospheric composition, statistical model choices (including hyperparameters and error covariance assumptions), and errors in atmospheric transport. In this presentation, we describe current uncertainty quantification activities linked to ongoing projects (e.g. EYE-CLIMA). We will discuss the use of meteorological ensemble simulations to assess transport related uncertainty and explore connections with dynamical systems tools and common assumptions such as Gaussian errors. Emphasis will be placed on high-resolution transport modelling applications.