An inter-comparative evaluation of PKU-FUEL global SO2 emission inventory

Abstract

PKU-FUEL is a recently developed gridded global emission inventory for multiple air pollutants that uses a bottom-up approach. The inventory includes data collected monthly for the period of 1960 to 2014 and at a 0.1° × 0.1° latitude/longitude resolution. In an effort to evaluate and improve this emission inventory, the PKU-FUEL Sulfur Dioxide (SO2) emission inventory was compared to other currently available and widely used global SO2 emission inventories constructed based on bottom-up and top-down approaches, including CEDS and OMI-HTAP. While PKU-FUEL is capable of capturing SO2 emissions across the globe and particularly in Asia, it misses 41 industrial point sources globally, accounting for 9.3% of Ozone Monitoring Instrument (OMI) remote sensing-measured industrial point sources. Most of these missing point sources are identified in Latin America, the Middle East (~60%), and some remote places. To improve the PKU-FUEL SO2 inventory, we applied OMI-measured emissions to sources missing from PKU-FUEL. GEOS-Chem model simulations were performed to evaluate original and improved PKU-FUEL SO2 inventories against measured SO2 concentrations across the world. Results were further compared to GEOS-Chem modeled SO2 concentrations using the CEDS inventory. We show that the modeled SO2 concentrations determined using both CEDS and improved PKU-FUEL inventories to a large extent corroborate sampled data and that the improved PKU-FUEL performs better for those regions lacking monitoring data.