Regional source apportionment of trace metals in fine particulate matter using an observation-constrained hybrid model

Abstract Trace metals in fine particulate matter (PM 2.5 ) are of significant concern environmental chemistry due to their toxicity and catalytic capability. An observation-constrained hybrid model is developed resolve regional source contributions trace other primary species PM . In this method, (PPM from the Community Multiscale Air Quality (CMAQ) Model at each monitoring location improved align better with observation data by applying source-specific scaling factors estimated a unique regression model. The adjusted PPM predictions chemical speciation then used generate profiles species. Finally, spatial distributions produced multiplying CMAQ deduced profiles. applied Pearl River Delta (PRD) region, China using daily collected multiple stations 2015 8 metals, elemental carbon, organic 10 Compared three previous methods, new predicts 13 smaller errors 16 less biases 10-fold cross validation analysis. determined study also show good agreement those literature. shows that during PRD Cu mainly area sources (31%), industry sector (27%), power generation (20%), an annual average concentration as high 50 ng m −3 some districts. Meanwhile, major contributors Mn (40%), emission outside (23%) (17%), leading mean level around Such information essential assessing epidemiological impacts well formulating effective control measures protect public health.