Characterization of the performance of a CityTree in an urban environment

Human activities have caused a huge increase in atmospheric emissions, mainly due to industrial, commercial and agriculture development, as well as the expansion of urban areas and with them the transportation system. Some of the most important anthropogenic primary aerosol sources in the cities are related to transport. A wide range of devices for removing PM and gaseous pollutants from urban environment have been developed in the last years. In the present work, the performance of a new remediation device of urban air pollution is reported. CityTree (CT) could be an innovative win-win solution for the urban environment. It is a freestanding unit, which contains combi planting of specific moss cultures, designed to bind and convert particulate matter (PM) and nitrogen dioxide into biomass. Three field campaigns were performed in order to characterize the performances of the CT in two different operative modalities: (1) active and (2) passive. In the first modality, an airflow was forced through the CT by a vent; in the second, the CT worked as a passive device on which atmospheric pollutants deposited in virtue of the high surface of the mosses. Filtration efficiency and deposition characteristics of the CT, in different meteorological conditions in a real environment, were measured at an urban street canyon in Modena (Italy). Measurements of air pollutants were performed at two different horizontal distances at both sides of the CT (in the front and in the rear of the panel) in order to determine concentration gradients and the corresponding deposition rates onto moss surfaces. Fluxes for a particular variable can be calculated from a gradient-flux relationship (Dyer, 1974), using turbulent diffusion coefficients. From deposition fluxes, the deposition velocity (Vd) can be obtained, normalising for average concentration. Furthermore, an average value of the measured variable at two sides of the CT were obtained, and from the difference, a CT Filtration Efficiency (FE) was calculated. Total particles (9 nm - 1 ?m) show a similar mean value of the deposition velocity for the front and rear side of the CT with, respectively, a mean of 0.26 ± 0.01 cm/s and 0.25 ± 0.01 cm/s. In order to avoid artefacts in the Vd calculation procedure of NO and NO2, which are both very reactive gases, the deposition velocity of NOx has been considered. In the roadside, it was 0.42 ± 0.02 cm/s, while in the rear side it measures 0.45 ± 0.02 cm/s. PM10, PM2.5 and PM1 deposition velocity show a very low decreasing behaviour ranging from 0.24 cm/s for PM10 to 0.21 cm/s for PM1 (mean values) on the street-side. Black carbon deposition velocity ranges between 0.65 cm/s and 0.68 cm/s in the front and rear side of the panel, respectively. In terms of particle number concentration, the CT filtration efficiency increases with increasing particle diameter in the range between 0.25 and 3.00 µm particle diameter, passing from 10% (0.25-0.28 µm size range) to 60% (2.50-3.00 µm). FE increase was observed for particles of smaller diameter (38±14%). In terms of the estimated aerosol mass, FE decreases passing from the PM10 (19±7%) size interval to PM1 (11±5%). Finally, the CT removed BC with a filtration efficiency around 17%. To compare the CT performance between the filtration and deposition mode, we calculated the amount of aerosol removed by the CT per unit of time in the two modes, assuming the same aerosol concentration levels. Results, clearly, shows that the aerosol removal efficiency of the CT is from 3 to almost 20 times higher in filtration than in deposition mode, according to the selected measured pollutant.