Gaseous mercury emissions from urban surfaces: Controls and spatiotemporal trends

The spatial and temporal variability of Hg emissions from urban paved surfaces was assessed through repeated measurements under varying environmental conditions at six sample sites in Toronto, Ontario, Canada. The results show significant spatial variability of the Hg emissions with median values ranging from below detection limit to 5.2 ng/m2/h. Two of the sites consistently had higher Hg emissions (on several occasions >20 ng/m2/h) than the other 4, which were equivalently low (maximum emission: 2.1 ng/m2/h). A surrogate measure of the pavement Hg concentrations was obtained during each day of sampling through the collection of street dust. The median street dust concentration also showed significant spatial variability (ranging from 9.6 to 44.5 ng/g). Regression analysis showed that the spatial variability of the Hg emissions was significantly related to the street dust concentrations. Controlled experiments using Hg amended street dust confirmed the relationship between Hg surface concentration and emission magnitude. Within a given sample site, Hg emissions varied temporally and multiple regression analysis showed that within-site variability was significantly influenced by changes in solar radiation with only a minor effect from surface temperature. Controlled experiments using shade cloths confirmed that solar radiation can have a large influence on the magnitude of Hg emissions within a given site. The emissions measured in Toronto were contextualized through comparison sampling in Austin, Texas. The Hg emissions measured in Austin were within the range detected in Toronto and also showed significant correlation with Hg street dust concentrations between sites. To provide a holistic assessment of Hg emissions from urban environments, samples were also collected from other common urban surfaces (soil, roofs, and windows). Soils consistently had higher emissions than all the other surfaces (7.3 ng/m2/h, n = 39).