Particulate matter (PM) has been observed to produce negative health impacts in humans^1,2. However, these health impacts can vary depending on several factors, including their sources^3,4. One approach to source apportioning PM is to determine its elemental composition, which can then be used as data inputs for receptor models that can characterise distinct source fingerprints. This project aims to identify a protocol for conducting elemental analysis of PM using Laser Ablation Inductively-Coupled Plasma Mass Spectrometry (LA-ICP-MS), which can serve as a high-throughput, low-cost alternative to analytical techniques employed for this purpose. Data from this analysis is then processed through a combination of the factor analysis receptor model Positive Matrix Factorisation (PMF), atmospheric measurements and HYSPLIT-derived back trajectories to estimate the contributions of different sources of PM.

PM_2.5 samples (as determined by Scanning Electron Microscope, SEM, analysis) were collected from Garden Island, Western Australia between April and May 2018. SEM was also used to identify spatial distributions of particles on sample papers, which was used to determine the ideal proportion of sample to be scanned to representatively detect the particles present. It was found that LA-ICP-MS analysis was able to reflect atmospheric conditions of the sampling periods. For example, samples that showed strong contributions from a sea spray source (i.e. those high in elements like Na, Cl and Br) corresponded with westerly winds from the Indian Ocean and low radon concentrations (as radon is a reliable tracer of terrestrial air). Conversely, sample that showed strong contributions from terrestrial sources (i.e. those high in elements like Al, Fe and Ti) aligned with winds from the mainland and high radon concentrations. A biomass burning source was also identified, with back trajectories showing that particles in these samples were transported by air that passed over regions where burning had occurred.