Winter brown carbon over six of China's megacities: light absorption, molecular characterization, and improved source apportionment revealed by multilayer perceptron neural network

Abstract. Brown carbon (BrC) constitutes a large fraction of organic carbon and exhibits strong light absorption properties, thus affecting the global radiation budget. In this study, we investigated the light absorption properties, chemical functional bonds, and sources of BrC in six megacities in China, namely Beijing, Harbin, Xi'an, Chengdu, Guangzhou, and Wuhan. The average values of the BrC light absorption coefficient and the mass absorption efficiency at 365 nm in northern cities were higher than those in southern cities by 2.5 and 1.8 times, respectively, demonstrating the abundance of BrC present in northern China's megacities. Fourier transform infrared (FT-IR) spectra revealed sharp and intense peaks at 1640, 1458–1385, and 1090–1030 cm−1, which were ascribed to aromatic phenols, confirming the contribution of primary emission sources (e.g., biomass burning and coal combustion) to BrC. In addition, we noted peaks at 860, 1280–1260, and 1640 cm−1, which were attributed to organonitrate and oxygenated phenolic groups, indicating that secondary BrC also existed in the six megacities. Positive matrix factorization (PMF) coupled with multilayer perceptron (MLP) neural network analysis was used to apportion the sources of BrC light absorption. The results showed that primary emissions (e.g., biomass burning, tailpipe emissions, and coal combustion) made a major contribution to BrC in the six megacities. However, secondary formation processes made a greater contribution to light absorption in the southern cities (17.9 %–21.2 %) than in the northern cities (2.1 %–10.2 %). These results can provide a basis for the more effective control of BrC to reduce its impacts on regional climates and human health.