Anthropogenic Emissions Estimated Using Surface Observations and Their Impacts on PM2.5 Source Apportionment over the Yangtze River Delta, China
Shuzhuang Feng,Fei Jiang,Hengmao Wang,Yang Shen,Yanhua Zheng,Lingyu Zhang,Chenxi Lou,Weimin Ju
DOI: https://doi.org/10.1016/j.scitotenv.2022.154522
2022-01-01
Abstract:Source-tagged source apportionment (SA) has advantages for quantifying the contribution of various source regions and categories to PM2.5; however, it is highly affected by uncertainty in the emission inventory. In this study, we used a Regional multi-Air Pollutant Assimilation System (RAPAS) to optimize daily SO2, NOx and primary PM2.5 (PPM2.5) emissions in the Yangtze River Delta (YRD) in December 2016 by assimilating hourly in-situ measurements. The CMAQ-ISAM model was implemented with prior and posterior emissions respectively to investigate the impacts of optimizing emissions on PM2.5 SA in the YRD megalopolis (YRDM) and three megacities of Shanghai, Nanjing, and Hangzhou in the YRDM. The results showed that RAPAS significantly improved the simulations and reduced the emission uncertainties of the different pollutants. Compared with prior emissions, the posterior emissions in the YRD decreased by 13% and 11% for SO2 and NOx respectively, and increased by 24% for PPM2.5. Compared with SA using prior emissions, the contributions from Hangzhou, northern Zhejiang, and areas outside of the YRD to the YRDM increased. The local contributions from the YRDM, Nanjing and Shanghai decreased by 1.8%, 9.7%, and 2.3%, respectively, whereas that of Hangzhou increased by 5.6%. The changes in the daily local contributions caused by optimizing emissions ranged from -18.0% to 23.6%. Generally, under stable weather conditions, the local contribution changed the most, whereas under unstable weather conditions, the contribution from upwind areas changed significantly. Overall, with optimized emissions, we found in Nanjing, Shanghai, and Hangzhou, local emissions contributed 18.2%, 39.6% and 36.8% of their PM2.5 concentrations, respectively; long-range transport from outside the YRDM contributed 59.2%, 48.1%, and 48.2%, respectively. This study emphasizes the importance of improving emission estimations for source-tagged SA and provides more reliable SA results for the main cities in the YRD, which will contribute to pollution control in these regions.