Spatial mapping and dimensional distribution of PM oxidative potential in Terni (Central Italy)

Over the last few decades, numerous studies have showed how the exposure to ambient particulate matter (PM) is associated to negative effects on human health (Pope et al, 2004). Moreover, the evaluation of PM exposure and toxic responses has been focused on the study of mass concentration, chemical composition and dimensional distribution of the particles (Kelly et al, 2012). However, the mechanisms behind the health effects are still not fully understood because of the complexity of PM composition. During the last years, the study of the oxidative potential (OP) has been proposed as a relevant metric related to biological responses to PM exposure (Simonetti et al, 2018a). Oxidative potential (OP) is defined as a measure of the capacity of PM to oxidise target molecules, by generating ROS in environments without living cells. Several acellular assays for testing OP have been developed, such as acid ascorbic (AA), 2,7-dichlorofluorescenin (DCFH) and dithiothreitol (DTT) assay, but anyone has been identified as the most appropriate for interpreting PM oxidative potential results. Therefore, the best solution turned out to be an inter-comparison between the different acellular tests. In fact, the three OP assays have been deemed sensible toward different PM-selected components, coming from various typical urban and industrial emission sources and characterized by very different chemical compositions, which can be associated to different adverse health effects (Simonetti et al, 2018b). In this study we applied the three OP assays (OPAA, OPDTT and OPDCFH) to PM10 samples, previously chemically analysed (Massimi et al, 2017), collected at 23 different sampling sites in Terni (an urban and industrial hot-spot of Central Italy), by using innovative and very-low volume devices for PM sampling on membrane filters (HSRS - High Spatial Resolution Sampler; Fai Instruments, Fonte Nuova, Rome, Italy). The HSRS worked in parallel during a two-month winter monitoring period. The sampling sites have been chosen for spatially representing the main anthropic PM emission sources (i.e. vehicular traffic, rail network, power plant, steel plant, domestic and industrial biomass heating) and the samplers were located in order to cover the study area with around 1 km spatial resolution (Figure 1, upper panel; Massimi et al, 2019). In this study we aimed to assess the spatial variation of the three acellular assays in order to investigate the relationships between the different OP results and the contribution of the local emission sources to the total PM10. To our knowledge, the comparison of the three OP assays applied to PM10 spatially-resolved samples has never been undertaken so far. Furthermore, we applied the three OP assays to size-segregated PM samples collected by a multistage impactor (cut-sizes: 0.18, 0.32, 0.56, 1.0, 1.8, 3.2, 5.6, 10 and 18 μm) at three sampling sites (MA, CA and PR; Figure 1, upper panel), characterized by different strength of the main PM sources, in order to evaluate the different sensitivity of the three acellular assays toward fine and coarse particles. The spatial mapping of the OP values, obtained by using the ordinary kriging interpolation (spherical semivariogram model; ArcGis, ArcMap 10.3.1) are reported in Figure 1. The results showed how OPAA was particularly sensitive toward PM10 coming from the rail network, released by resuspension of particles formed through the abrasion of train brakes. On the contrary, OPDTT, OPDCFH were found to be more related to PM10 coming from industrial sources (steel plant at PR and carpentry at FA) and biomass burning (domestic biomass heating at BR). Finally, size distribution analyses of OP confirmed that OPAA was more sensitive toward coarse particles, mainly released by brake abrasion, while OPDTT and OPDCFH were found to be more sensitive toward fine particles, mainly released by combustion processes such as biomass burning and industrial processes. These results confirmed the different sensitivity of the three OP assays toward PM10 released by the local emission sources.