Accurate ammonia (NH3) emission inventories are essential for assessing environmental impacts. Currently, commonly used bottom-up inventories contain substantial uncertainties, particularly in those less-developed regions such as South America and Africa. This study presents a global NH3 emission inversion in 2022 by assimilating Infrared Atmospheric Sounding Interferometer (IASI) retrievals into the GEOS-Chem model at 2° × 2.5° resolution. The inversion results indicate that the prior inventory, constructed from the Community Emissions Data System (CEDS) and the Global Fire Emissions Database (GFED), generally captures NH3 emissions' spatial distribution but systematically underestimates emission intensity in South America and Africa. Posterior emissions, defined as the optimized ammonia emission estimates obtained by assimilating satellite observations into the prior inventory, reveal the upward trend in South America, where growing season fluxes exceed 0.2 g/m2 per month and peak values surpass 0.5 g/m2 per month, corresponding to relative increases of approximately 100–200% compared to prior estimates. Africa also exhibits pronounced increases, with maximum monthly emissions exceeding 0.5 g/m2 per month, representing approximately a twofold enhancement compared to prior estimates (0.2–0.3 g/m2 per month). The posterior inventory also captures seasonal variability more realistically, with approximately 100% increases during peak emission periods in both regions.
Chen et al. (Mon,) studied this question.