Development of Satellite-constrained Pollution Emissions for Improved Simulation of Global Tropospheric Composition

Emissions of short-lived air pollutants from large point sources or cluster of sources, including SO₂ and NO_x, can be estimated directly from satellite observations. Satellites have observed significant changes in air pollutant concentrations driven by anthropogenic emissions that can vary over short periods owing to, for example, the commissioning of new industrial facilities or the implementation of stricter emission controls. Air quality and climate models typically use “bottom-up” emission inventories compiled by aggregating activity data and emission factors, which do not always reflect recent changes, leading to inaccurate model results. In this presentation, we will discuss a new approach to merge satellite-derived emissions with a state-of-the-science global inventory to provide more accurate and timely emissions for improved analysis of atmospheric composition. This new approach starts from the development of the NASA Satellite-derived Emission (SatEm) database that quantifies SO₂ and NO_x emissions at enhanced accuracy and temporal resolution from the latest Ozone Monitoring Instrument (OMI) retrievals. We develop the phase I of SatEm, which focuses on refineries. We then perform intercomparison of SatEm and bottom-up emission estimates to improve our understanding of emissions from refineries. We finally merge SatEm with a leading global bottom-up inventory, the Community Emissions Data System (CEDS), to construct the phase I of hybrid emission inventory. This hybrid inventory combines the comprehensive nature of the bottom-up emissions data with the SatEm emissions database (magnitude, spatial location, annual time series, and seasonality) to provide accurate and timely estimates of emissions.