Development of a high-resolution integrated emission inventory of air pollutants for China

Abstract

Constructing a highly resolved comprehensive emission dataset for China is challenging due to limited availability of refined information for parameters in a unified bottom-up framework. Here, by developing an integrated modeling framework, we harmonized multi-source heterogeneous data, including several up-to-date emission inventories at national and regional scales and for key species and sources in China to generate a 0.1° resolution inventory for 2017. By source mapping, species mapping, temporal disaggregation, spatial allocation, and spatial–temporal coupling, different emission inventories are normalized in terms of source categories, chemical species, and spatiotemporal resolutions. This achieves the coupling of multi-scale, high-resolution emission inventories with the Multi-resolution Emission Inventory for China (MEIC), forming the high-resolution INTegrated emission inventory of Air pollutants for China (INTAC). We find that INTAC provides more accurate representations for emission magnitudes and spatiotemporal patterns. In 2017, China's emissions of sulfur dioxide (SO2), nitrous oxides (NOx), carbon monoxide (CO), non-methane volatile organic compounds (NMVOCs), ammonia (NH3), PM10 and PM2.5 (particulate matter), black carbon (BC), and organic carbon (OC) were 12.3, 24.5, 141.0, 27.9, 9.2, 11.1, 8.4, 1.3, and 2.2 Tg, respectively. The proportion of point source emissions for SO2, PM10, NOx, and PM2.5 increases from 7 %–19 % in MEIC to 48 %–66 % in INTAC, resulting in improved spatial accuracy, especially mitigating overestimations in densely populated areas. Compared with MEIC, INTAC reduces mean biases in simulated concentrations of major air pollutants by 2–14 µg m−3 across 74 cities, compared against ground observations. The enhanced model performance by INTAC is particularly evident at finer-grid resolutions. Our new dataset is accessible at http://meicmodel.org.cn/intac (last access: 15 April 2024) and https://doi.org/10.5281/zenodo.10459198 (Wu et al., 2024), and it will provide a solid data foundation for fine-scale atmospheric research and air-quality improvement.