Restoring clean air to cities using tree cover

Restoring ecosystem services using vegetation is a time honored first step in restoration that has special appeal in cities. Tree planting is touted as a way to filter particulate matter (PM) from the air, thereby reducing the incidence of respiratory disease. Understanding small-scale variation in [PM] logically precedes functional landscape design to achieve this goal. We developed a short term monitoring protocol to predict return periods of local pollution events. We used this to compare exposure risk on rooftops where official monitors are placed with locations 2m above the sidewalk more typical for pedestrian exposure. In the South Bronx, [PM2.5] at 2m was 1.5 X rooftop concentrations. A 60 X 10⁶ m^-3 event has a 20 min. return period at street level while the maximum event observed on a roof was only 40 X 10⁶ m^-3. By lumping particles into the regulatory class PM2.5, and by expressing concentrations by mass rather than counts, official measurements further underestimate the contribution of the more injurious ultrafine PM to the total load. To address the health impact, we aspirated PM samples onto filters adjacent to our counters, challenged RAW cell cultures with filter extracts, and used ELISA to quantify the induction of the proinflamatory cytokine TNFα caused by exposure to the particles. Sidewalk samples elicited 10X greater expression than rooftop samples while samples collected in a park showed no variation from curbside to park core. To assess trees’ effect on [PM], we ran a diurnal transect through a suburban forest canopy at the eddy flux tower maintained by the Baltimore Ecosystem Study. Above canopy concentrations were consistently lower than either within or below. Our findings have far-reaching implications for both air quality measurement and the practicality of using trees as air filters for PM2.5.