Abstract
Some air pollution datasets contain multiple variables with a range of measurement units, and combined analysis using positive matrix factorization (PMF) can be problematic but can offer benefits through the greater information content. In this work, a novel method is devised and the source apportionment of a mixed unit dataset (PM10 mass and number size distribution, NSD) is achieved using a novel two-step approach to PMF. In the first step the PM10 data are PMF-analysed using a source apportionment approach in order to provide a solution which best describes the environment and conditions considered. The time series G values (and errors) of the PM10 solution are then taken forward into the second step, where they are combined with the NSD data and analysed in a second PMF analysis. This results in NSD data associated with the apportioned PM10 factors. We exemplify this approach using data reported in the study of Beddows et al. (2015), producing one solution which unifies the two separate solutions for PM10 and NSD data datasets together. We also show how regression of the NSD size bins and the G time series can be used to elaborate the solution by identifying NSD factors (such as nucleation) not influencing the PM10 mass.