Abstract: Particulate matter (PM) deposition on urban green enables the collection of particulate pollution from a diversity of contexts, and insight into the physico-chemical profiles of PM is key for identifying main polluting sources. This study reports on the morphological and elemental characterization of PM2-10 deposited on ivy leaves from five different environments (forest, rural, roadside, train, industry) in the region of Antwerp, Belgium. Ca. 40,000 leaf-deposited particles were thoroughly investigated by particle-based analysis using scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM/EDX) and their physico-chemical characteristics were explored for PM source apportionment purposes. The size distribution of all deposited particles was biased towards small-sized PM, with 32% of the particles smaller than 2.5 mu m (PM2.5) and median diameters of 2.80-3.09 mu m. The source type influenced both the particles' size and morphology (aspect ratio and shape), with roadside particles being overall the smallest in size and the most spherical. While forest and rural elemental profiles were associated with natural PM, the industry particles revealed the highest anthropogenic metal input. PM2-10 profiles for roadside and train sites were rather comparable and only distinguishable when evaluating the fine (2-2.5 mu m) and coarse (2.5-10 mu m) PM fractions separately, which enabled the identification of a larger contribution of combustion-derived particles (small, circular, Fe-enriched) at the roadside compared to the train. Random forest prediction model classified the source type correctly for 61-85% of the leaf-deposited PM. The still modest classification accuracy denotes the influence of regional background PM and demands for additional fingerprinting techniques to facilitate source apportionment. Nonetheless, the obtained results demonstrate the utility of leaf particle-based analysis to fingerprint and pinpoint source-specific PM, particularly when considering both the composition and size of leaf-deposited particles.

Abstract: As real-life experimental data on natural ventilation of atmospheric pollution levels in urban street canyons is still scarce and has proven to be complex, this study, experimentally evaluated the impact of an urban street canyon opening on local atmospheric pollution levels, during a 2-week field campaign in a typical urban street canyon in Antwerp, Belgium. Besides following up on atmospheric particulate matter (PM), ultrafine particles (UFPs) and black carbon (BC) levels, the magneto-chemical PM10 composition was quantified to identify contributions of specific elements in enclosed versus open street canyon sections. Results indicated no higher overall PM, UFP and BC concentrations at the enclosed site compared to the open site, but significant day-to-day variability between both monitoring locations, depending on the experienced wind conditions. On days with oblique wind regimes (4 out of 14), natural ventilation was observed at the open location while higher element contributions of Ca, Fe, Co, Ni, Cu, Zn and Sr were exhibited at the enclosed location. Magnetic properties correlated with the PM10 filter loading, and elemental content of Fe, Cr, Mn and Ti. Magnetic bivariate ratios identified finel-grained magnetite carriers with grain sizes below 0.1 μm, indicating similar magnetic source contributions at both monitoring locations. Our holistic approach, combining atmospheric monitoring with magneto-chemical PM characterization has shown the complex impact of real-life wind flow regimes, different source contributions and local traffic dynamics on the resulting pollutant concentrations and contribute to a better understanding on the urban ventilation processes of atmospheric pollution.

Abstract: Urban green works as a recorder of atmospheric PM. This paper reports on the utility of combining magnetic- and particle-based techniques to investigate PM leaf deposition as a bio-indicator of metal pollution. Ivy (Hedera helix) leaves were collected from five different land use classes, i.e. forest, rural, roadside, industrial, train. Leaf magnetic measurements were done in terms of saturation isothermal remanent magnetization (leaf SIRM), while ca. 40,000 leaf-deposited particles were analyzed through SEM/EDX to estimate the elemental composition. The influence of the different land use classes was registered both magnetically and in terms of metal content. Leaf area-normalized SIRM values ranged from 19.9 to 444.0 μA, in the following order forest < rural < roadside < industrial < train. Leaf SIRM showed to be significantly correlated (p < 0.01) with the content in Fe, Zn, and Pb, followed by Mn and Cd (p < 0.05), while no significant correlation was found with the metals Cr and Cu. Although presenting a similar metal content, roadside and train were magnetically very distinct. By exhibiting a very high content in Pb, and with an Fe content being comparable to the one observed at the forest and rural land uses, the industrial leaf-deposited particles showed to be mainly due to industrial activity. While SEM/EDX is a suitable approach for detailed particle analysis, leaf SIRM of ivy can be used as a rapid discriminatory tool for metal pollution. Their complementary use delivers further knowledge on land use classes reflecting different PM conditions and/or sources.