Source contribution analysis of PM2.5 using Response Surface Model and Particulate Source Apportionment Technology over the PRD region, China

Identifying the emission source contributions to PM2.5 is essential for a sound PM2.5 pollution control policy. In this study, we conduct a comparative analysis of PM2.5 source contributions over the Pearl River Delta (PRD) region of China using two advanced source contribution modeling techniques: Re-sponse Surface Model (RSM) and Particulate Source Apportionment Technology (PSAT). Our compara-tive analyses show that RSM and PSAT can both reasonably predict the contribution of primary PM2.5 emission sources to PM2.5 formation due to its linear nature. For the secondary PM2.5 formed by the nonlinear reactions among PM2.5 precursors, however, our study shows that PSAT appears to have limi-tations in quantifying the nonlinear contribution of PM2.5 precursors to emission reductions, while RSM seems to better address the nonlinear relationship among PM2.5 precursors (e.g., PM2.5 disbenefits due to local NOx emission reductions in major cities with high NOx emissions). The pilot study case results show that for the ambient PM2.5 in the central cities (Guangzhou, Shenzhen, Foshan, Dongguan, and Zhongshan) of the PRD, the regional source emissions contribute the most by 42-66%; the dust emissions are the top contribution sources (29-34% by RSM and 27-31% by PSAT), and the mobile sources are listed as the secondary contributors accounting for 16-25% by RSM and 19-30% by PSAT among the anthropogenic emission sources. The city-scale cooperation on emission reductions and the enhancement of dust and mobile emission control are recommended to effectively reduce the ambient PM2.5 concentration in the PRD.(C) 2021 Elsevier B.V. All rights reserved.

Automated summary

Identifying the contribution of emission sources to PM2.5 formation is crucial for an effective pollution control policy. A comparative analysis of PM2.5 source contributions in the Pearl River Delta region of China shows that both the Response Surface Model (RSM) and Particulate Source Apportionment Technology (PSAT) can reasonably predict the contribution of primary PM2.5 emission sources due to their linear nature. However, for secondary PM2.5 formed by nonlinear reactions, PSAT has limitations in quantifying the contribution of PM2.5 precursors, while RSM better addresses the nonlinear relationship among precursors.