Evaluation of impacts of trees on PM2.5 dispersion in urban streets

Reducing airborne particulate matter (PM), especially PM2.5 (PM with aerodynamic diameters of 2.5 mu m or less), in urban street canyons is critical to the health of central city population. Tree-planting in urban street canyons is a double-edged sword, providing landscape benefits while inevitably resulting in PM2.5 concentrating at street level, thus showing negative environmental effects. Thereby, it is necessary to quantify the impact of trees on PM2.5 dispersion and obtain the optimum structure of street trees for minimizing the PM2.5 concentration in street canyons. However, most of the previous findings in this field were derived from wind tunnel or numerical simulation rather than on-site measuring data. In this study, a seasonal investigation was performed in six typical street canyons in the residential area of central Shanghai, which has been suffering from haze pollution while having large numbers of green streets. We monitored and measured PM2.5 concentrations at five heights, structural parameters of street trees and weather. For tree-free street canyons, declining PM2.5 concentrations were found with increasing height. However, in presence of trees the reduction rate of PM2.5 concentrations was less pronounced, and for some cases, the concentrations even increased at the top of street canyons, indicating tree canopies are trapping PM2.5. To quantify the decrease of PM2.5 reduction rate, we developed the attenuation coefficient of PM2.5 (PMAC). The wind speed was significantly lower in street canyons with trees than in tree-free ones. A mixed-effects model indicated that canopy density (CD), leaf area index (LAI), rate of change of wind speed were the most significant predictors influencing PMAC. Further regression analysis showed that in order to balance both environmental and landscape benefits of green streets, the optimum range of CD and LAI was 50%-60% and 1.5-2.0 respectively. We concluded by suggesting an optimized tree-planting pattern and discussing strategies for a better green streets planning and pruning. (C) 2014 Elsevier Ltd. All rights reserved.