Scalar dispersion from street canyons

Janet Barlow and Stephen Belcher


Transport of pollution and heat out of streets into the boundary layer above is not currently understood and so fluxes cannot be quantified. Scalar concentration within the street is determined by the flux out of it and so quantifying fluxes for turbulent flow over a rough urban surface is essential. We are working on experimental approaches to the problem consisting of windtunnel experiments and full-scale experiments.

Windtunnel Experiments

We have developed the naphthalene sublimation technique to measure transfer from a 2D street canyon in a windtunnel for the case of flow perpendicular to the street. Some questions we have been looking at:

A windtunnel model for quantifying fluxes in the urban boundary layer

Barlow, J.F. and Belcher, S.E.(2002) Boundary Layer Meteorology, vol. 104, 131-150

The windtunnel at Reading University

Dimensions of working section = 232x232x1500mm. The floor of the windtunnel was covered with Lego to increase surface roughness. In addition to this, a fence of Lego was placed at the entrance to the working section to generate a logarithmic windprofile upstream of the street canyon model. A single hot-wire anemometer was used to measure windspeed U at the top of the boundary layer.

Plan and elevation view of street canyon model

The street canyon model consisted of two aluminium blocks which spanned the width of the windtunnel with 3mm thick plate between them to represent the street itself. The height of the blocks was 12.5mm, and the width of the street was varied between 50mm and 6.25mm.

The street was covered in a layer of thermal contact paste and then naphthalene powder. Naphthalene sublimes at room temperature, so the vapour represented a scalar source at street level.The method involved running the windtunnel at a fixed speed for 10 minutes, measuring the mass change of naphthalene on the street (to give the flux) and its temperature (to calculate source concentration of naphthalene vapour above the street). The plate was divided into three, such that only the central portion (length 0.28L, where L=length of street = 232mm) was weighed - this was to avoid measuring transfer in the vicinity of the windtunnel walls where flow is distorted.

Linear relationship found between transfer velocity and windspeed

For the range of windspeeds studied (4 to 14m/s), a linear relationship was found between transfer velocity wT and windspeed U. The dimensionless transfer coefficient wT/U represents the ventilation efficiency of the canyon. Observed values were between 0.0015 and 0.0027 and, for the case where H/W tends to zero, values were in the same range as estimates of transfer from a flat plate, giving confidence that the technique yields accurate values for street canyon scalar transfer.

Variation of ventilation efficiency with street aspect ratio H/W

The street aspect ratio (ratio of building height to street width, H/W) was varied and the transfer velocity measured for a range of windspeeds for each aspect ratio. Experiments A and C used upstream surfaces of low roughness (for experiment C, the height of the building was varied) whilst high roughness was used for experiment B.

The transfer coefficient, or ventilation efficiency, wT/U varied with aspect ratio, reaching a maximum in the wake interference regime (0.3< H/W <0.65). However, when upstream roughness was increased (experiment B), the maximum reduced, suggesting that street ventilation is less sensitive to H/W when the flow is in equilibrium with the urban surface.

back to the top

Full-scale experiments

In February 2000, sonic anemometers were used by Reading University to measure windflow in a typical street in Birmingham as part of the PUMA campaign. The site was chosen for being in an area of "street canyon" type roughness, i.e. long straight streets with houses of equal height on both sides . Initial results have shown an increase in shear stress with height within the street. Further analysis includes looking at the correlation between turbulence above and below roof level, and spectral analysis to establish the time- and lengthscales on which turbulence is acting.

A larger measurement campaign is planned for Easter 2001, to take place in Salford. The aim is to formulate a more complete picture of flow in and above an urban area, taking into account the inhomogeneity of the surface. This will involve using the Lidar which has been developed at Salford University for use in urban areas to measure windflow above the urban area. At the same time, sonic anemometers will be used to measure windflow in and above a typical urban street.

Experiments will also be carried out at Hall Farm near Reading University, using an idealised canyon. Previous experiments by Louka et al (2000) established that there was an intermittent, turbulent circulation within the canyon and a shear layer at roof level.

back to the top