Informing Receptor-based PM2.5 Source-apportionment Using a Chemical Transport Model

Abstract:

Receptor models quantify the contributions of different sources of PM_2.5 based on the observed spatiotemporal variability of its chemical constituents at the receptor site; this information can guide emission reduction measures to improve air quality. However, the configuration of the receptor models and the interpretation of their results rely heavily on the qualitative judgement of the user, and the sources of secondary components in PM_2.5 cannot be distinguished. As such, the accuracy of receptor-based source-apportionment is largely unknown and may be subject to the experience of the user and the meteorological conditions of the case of interest. In this study, we generated a synthetic PM_2.5 chemical composition dataset over Northern China using the regional chemical transport model, WRF-CMAQ. We applied the receptor model Multilinear Engine 2 (ME-2) to the synthetic dataset and compared the source-apportionment result to the outcomes from sensitivity tests using WRF-CMAQ. We discussed the following issues: (1) the spatial characteristics of the ME-2 model’s source-apportionment results and the impact of meteorology, (2) the strategies to quantitatively attribute secondary particle mass to specific emission sources, and (3) the strategies to interpret the ME-2 source-apportionment outcome objectively and quantitatively. Our result helps improve the objectiveness and accuracy of receptor-based source-apportionment of PM_2.5.