Global Inversion of a Black Carbon Emissions based on FLEXPART modelling and a Bayesian inversion algorithm

Black carbon (BC) is a significant climate forcer and a major health hazard especially close to its sources. BC has both natural (e.g., wildfires) and anthropogenic sources (e.g., industry, traffic, oil and gas industries). Its distribution in the atmosphere is highly inhomogeneous. Understanding both the spatial distribution and the magnitude of BC emissions is critical for accurate climate modeling. However, emission inventories for BC are fraught with uncertainties, largely stemming from uncertainties in emission factors, which complicate global scale modeling efforts. Observational data, especially when obtained by different measurement techniques, which is essential for constraining emission estimates, carry their own uncertainties.

In this study, we present a global inversion of BC emissions over a 5-year period, using a combination of global observations, atmospheric modeling with FLEXPART, and a Bayesian inversion algorithm (FLEXINVERT). Our approach aims to reconcile uncertainties in both emissions and observations, providing a more robust estimate of BC distribution and sources. Even though presenting a global picture, we focus on Europe, an area with a high density of observation sites, enabling more precise emission estimates. The tropics and southern hemisphere have only sparse observations. Moreover, we highlight the significant role of wildfires as a source of BC, with implications for both local and global climate impacts.

Our findings contribute to improving the accuracy of BC inventories which can be used both for climate modelling and air quality assessments.