The effect of tree planting on traffic pollutant dispersion in an urban street canyon using large eddy simulation with a recycling and rescaling inflow generation method

The influence of various tree planting configurations on tracer gas dispersion in an urban street canyon is studied numerically, using OpenFOAM Large Eddy Simulation (LES). A recycling and rescaling turbulent inflow generation method is implemented and validated against a canonical backwards-facing step case. The street canyon simulations are based on the CODASC experiments, where the tracer gas was emitted from line sources at street level. The effect of tree stand density on the pollutant dispersion is studied for the first time using LES. Performance metric analysis of the CODASC simulations shows that the influence of tree crown porosity, and tree stand density, on the concentration field within the canyon can be captured by LES. The simulations show that tree crowns reduce the effectiveness of the canyon vortex at ventilating the street, and enhance the mean tracer gas concentration at pedestrian level on the leeward wall. A novel feature of the research arises from concentration probability density functions at pedestrian height in the canyon. Pedestrians can be exposed to pollutant concentrations almost three times the mean value at the leeward wall. The results have implications for urban greening strategies in city streets with heavy traffic.