Methods and applications for quantitative assessment of uncertainty in atmospheric particulate matter source profiles

Abstract

Particulate Matter (PM), a primary atmospheric pollutant, is characterized by diverse emission sources and complex influencing factors. Particulate matter source profiles (PM-SPs) reported in different studies exhibited significant variability, and their uncertainty is not well documented, highlighting the urgent need for further research. To scientifically evaluate PM-SPs, this study begins with the construction process of PM-SPs, identifying key stages (sampling, weighing, and chemical analysis) that influence their uncertainty. The uncertainty components at each stage, as well as the comprehensive uncertainty, are then quantitatively assessed, resulting in the development of an assessment method that quantifies the uncertainty of PM-SPs. The industrial PM-SPs are evaluated as a case study, which indicate this full-chain method can quantitatively identify the uncertainties introduced at various stages. The sampling uncertainties are mostly below 40%, primarily resulting from errors in repeat sampling. The weighing uncertainties are generally minor, while the impact of the chemical analysis process varies significantly across different components. Among the fifty-three evaluated PM-SPs, 80% have average comprehensive uncertainties below 40%, with only five profiles exhibiting relatively high uncertainties. In practical applications, the uncertainty assessment of PM-SPs should be analyzed on a case-by-case basis, considering the importance of different components, with special attention given to marker components.