The influence of meteorology, urban topography and traffic on the variability in concentrations of a traffic-related pollutant in urban street canyons

摘要

A comprehensive study into the spatial and temporal variations in concentrations of a traffic-related pollutant was undertaken in two urban street canyons located in York, U. K. During the field experiment, investigations were carried out in order to determine the influence of meteorology, urban topography and traffic on the measured concentrations of a traffic-related pollutant. The pollutant measured in this study was carbon monoxide (CO). Results are presented from simultaneous and continuous measurements of the background (or reference) wind speed and direction, the in-canyon wind and turbulence fields, traffic characteristics and CO concentrations collected over a period of one month. The background wind was found to influence the development of in-canyon wind flow features, which became mechanisms for pollutant dispersion. Under certain background winds, evidence of across-canyon recirculating flows with horizontally- and vertically-aligned axes is presented. During these conditions, turbulence and traffic-related pollutants are likely to have been transported in the across-canyon re-circulating flows. During background wind orientated perpendicular to the street axis, the 15-minute mean concentrations of CO were a factor of 2 or 3 higher on the leeward (or upwind) side of the street canyons compared to the windward side. This was caused by the development of an across-canyon recirculating flow with a horizontally-aligned axis. Down draughts were measured on the windward side of the canyon during these conditions, while updraughts were measured on the leeward side. The wind direction at street level opposed the direction of the above-roof flow, causing the transport of CO towards the leeward side of the street canyon where concentrations increased. Evidence Results are presented from simultaneous and continuous measurements of the background (or reference) wind speed and direction, the in-canyon wind and turbulence fields, traffic characteristics and CO concentrations collected over a period of one month. The background wind was found to influence the development of incanyon wind flow features, which became mechanisms for pollutant dispersion. Under certain background winds, evidence of across-canyon re circulating flows with horizontally- and vertically-aligned axes is presented. During these conditions, turbulence and traffic-related pollutants are likely to have been transported in the across-canyon re circulating flows. During background wind orientated perpendicular to the street axis, the 15-minute mean concentrations of CO were a factor of 2 or 3 higher on the leeward (or upwind) side of the street canyons compared to the windward side. This was caused by the development of an across-canyon recirculatingf low with a horizontally-aligned axis. Downdraughts were measured on the windward side of the canyon during these conditions, while updraughts were measured on the leeward side. The wind direction at street level opposed the direction of the above-roof flow, causing the transport of CO towards the leeward side of the street canyon where concentrations in creased. Evidence of counter-rotating comer vortices with vertically-aligned axes are also presented and are thought to have been caused by along-canyon converging flows during certain background winds. However, across-canyon flow during these conditions also accounted for the higher concentrations of CO measured on the leeward side of the canyon, compared to the windward side. Background winds orientated oblique to the street axis were found to produce across-canyon rearticulating flows. On the other hand, parallel background winds produced channel flows, which had a 'flushing' effect, causing some of the lowest concentrations of CO to be measured during these conditions. Traffic-produced turbulence was investigated in the street canyon during the field experiment. Comparisons were made between calculated and measured turbulence parameters. The results indicate that the parameterisation performs reasonably well, particularly during weak perpendicular background winds when traffic-produced turbulence effects are likely to have been most dominant. The traffic was also investigated due to the implications of traffic characteristics on the variability in concentrations of CO. The influence of free-, unstable- and congested-traffic flows on measured CO concentrations was determined. The highest mean concentrations were measured during congested traffic conditions, when the emission levels are likely to have been elevated due to stop-start driving events. The combined influence of the background wind and traffic characteristics was investigated. Results are presented which show that the highest mean CO concentrations were measured during perpendicular background winds and during congested traffic conditions. The implications of using variable emission rates in a computational fluid dynamics (CFD) dispersion model were investigated in a sensitivity study. Results are also presented from the flow model study, which was conducted to help in the interpretation of the field experiment data.