Compositional data analysis of element concentrations of simultaneous size segregated PM measurements

This preliminary study presents the application of compositional data analysis on element concentrations of size-segregated PM simultaneous measurements. There is a growing interest in particulate matter (PM) due to its impact on human health, air quality and global climate change. PM is a mixture of particles suspended in the air which differ in size, chemical composition and emission sources. The assessment of the chemical composition and size distribution of PM in relation to its possible emission sources is a starting point to plan actions aimed at mitigating the levels of PM in the environment and to protect public health. Selected chemical elements have been linked to specific sources of PM including mineral matter, sea-spray, and fuel-oil combustion. However, the identification of a set of elements useful in the discrimination of specific natural and anthropogenic sources of mineral matter has proven to be problematic, since they can have the same range of chemical elements in common. This study considers PM element concentrations of a typical suburban background site with (indust days) and without (non-dust days) the contribution from a Saharan dust event. The selected elements were Al, Si, Ca, Fe, Ti, Mg, Sr, commonly interpreted as related to mineral matter. The element concentrations of PM related to in-dust and non-dust days have been converted into two compositional data sets based on percentage weight. The compositional data analysis provides evidence that the two compositional data sets are statistically distinct. This outcome shows that the Saharan dust event (in-dust days) together with local sources of mineral matter (non-dust days) can determine the chemical composition of PM. Therefore, compositional data analysis allows the study of environmental sites effected by natural sources of mineral matter (e.g. Saharan dust).