Environmental Science and Engineering Seminar

Satellite measurements of greenhouse gases can improve our understanding of the global carbon cycle in a changing climate, but quantifying the uncertainty of the fluxes estimated using these observations is critical. The optimal fluxes, associated uncertainties, and information content given atmospheric observations can be generated by analytical minimization of a Bayesian cost function. I will discuss two methodological advancements for analytical inversions. First, the specification of background concentrations given by boundary conditions (BCs) at the edges of the inversion domain is necessary for regional analyses of greenhouse gas fluxes. But, biases in the BCs propagate to biases in the optimized fluxes. We develop a theoretical framework to explain, predict, and correct the influence of BCs on optimized fluxes as a function of the prior and observing system error statistics and of model transport. Second, we estimate monthly CO2 fluxes and uncertainties on a multiscale grid using atmospheric CO2 column observations from the Orbiting Carbon Observatory-2 (OCO-2) instrument. The multiscale grid preserves the native 4° x 5° resolution in areas with high observational density and goes to coarser resolution elsewhere. We also demonstrate a method for estimating the uncertainty for the biosphere, oceans, and fossil fuels. The analysis is developed to be reproducible by expanding the Integrated Methane Inversion into the Integrated Carbon Inversion using the new GEOS-Chem carbon simulation.