Contribution of railway traffic to local PM10 concentrations in Switzerland

Abstract

Field measurement campaigns of PM10 and its elemental composition (daily sampling on filters) covering different seasons were performed at two sites near the busiest railway station of Switzerland in Zurich (at a distance of 10 m from the tracks) and at a site near a very busy railway line with more than 700 trains per day. At this latter site parallel samples were taken at 10, 36 and 120 m distances from the tracks with the aim to study the distance dependence of the railway induced PM10 concentrations. To distinguish the relatively small railway emissions from the regional background (typically 20–25 μg m−3), simultaneous samples were also taken at an urban background site in Zurich. The differences in PM10 and elemental concentrations between the railway exposed sites and the background site were allocated to the railway contribution. Small, however, measurable PM10 concentration differences were found at all sites. The elemental composition of these differences revealed iron as the only quantitatively important constituent. As a long-term average it amounted to approximately 1 μg m−3 Fe at a distance of 10 m from the tracks at all three sites. Assuming that iron was at least partly oxidised (e.g. in the form of Fe2O3) the contribution can amount up to 1.5 μg m−3. Emissions of copper, manganese and chromium from trains were also clearly identified. However, compared to iron these, elements were emitted in very low quantities. No significant contribution from rock material (calcium, aluminium, magnesium, sodium) was observed as might have been expected from erosion, abrasion and resuspension from the gravel below the tracks. Particle emissions from diesel exhaust were not considered as trains in Switzerland are operated nearly exclusively by electric locomotives. The railway, induced contribution to ambient PM10 decreased rapidly with increasing distance from the tracks. At a distance of 120 m this contribution dropped to only 25% of the contribution observed at 10 m distance.