Unraveling Contributions from Combustion, Secondary, Traffic, and Dust Sources through Particle Mass Size Distribution Measurement

Source apportionment of ambient particulate matter (PM) across different size ranges is challenging, particularly for hourly data. This study utilized high-time-resolution particle mass size distribution (PMSD) data collected by an optical particle counter (OPC) and employed the normalized non-negative matrix factorization (N-NMF) algorithm to achieve enhanced separation of five distinct sources. Each source is characterized by unique PMSD profiles and temporal patterns, indicative of diverse pollution origins. Through comparative analysis with criteria pollutants and the chemical compositions of PM_2.5, the sources were identified as combustion, secondary, SO₂-related, traffic-related, and dust. The size distribution data facilitate the quantification of each source’s contribution across various particle sizes (e.g., PM_2.5, PM₁₀). Combustion sources and secondary aerosol contribute most to PM₁ (> 80%), while traffic-related contribute most to PM_1–2.5 (> 50%). Traffic-related contribute 85% to PM_2.5–10 and 99% to PM_10–20. The study delineates and discusses the methodological differences between this approach and traditional source apportionment methods based on the chemical compositions of PM. This methodology not only underscores the distinct impacts of these sources on different particle size ranges but also provides a novel framework for future receptor model source apportionment using low-cost OPC in environmental aerosol analysis.