City breathability in medium density urban-like geometries evaluated through the pollutant transport rate and the net escape velocity

Highlights

• Pollutant removal in UCL models of medium building density is studied.

• Pollutant transport rate (PTR) and net escape velocity (NEV) are used as ventilation indicators.

• NEV is smaller in UCLs with longer streets due to pollutant accumulation effect.

• Parallel wind and wall heating get bigger NEV than oblique wind and isothermal case.

• Building height variations enhance pollutant removal across street roofs.

Abstract

This paper investigates pollutant removal at pedestrian level in urban canopy layer (UCL) models of medium packing density (λ_p = λ_f = 0.25) using computational fluid dynamics (CFD) simulations. Urban size, building height variations, wind direction and uniform wall heating are investigated. The standard and RNG k−ε turbulence models, validated against wind tunnel data, are used. The contribution of mean flows and turbulent diffusion in removing pollutants at pedestrian level is quantified by three indicators: the net escape velocity (NEV), the pollutant transport rate (PTR) across UCL boundaries and their contribution ratios (CR).

Results show that under parallel approaching wind, after a wind-adjustment region, a fully-developed region develops. Longer urban models attain smaller NEV due to pollutant accumulation. Specifically, for street-scale models (∼100 m), most pollutants are removed out across leeward street openings and the dilution by horizontal mean flows contributes mostly to NEV. For neighbourhood-scale models (∼1 km), both horizontal mean flows and turbulent diffusion contribute more to NEV than vertical mean flows which instead produce significant pollutant re-entry across street roofs. In contrast to uniform height, building height variations increase the contribution of vertical mean flows, but only slightly influence NEV. Finally, flow conditions with parallel wind and uniform wall heating attain larger NEV than oblique wind and isothermal condition.

The paper proves that by analysing the values of the three indicators it is possible to form maps of urban breathability according to prevailing wind conditions and known urban morphology that can be of easy use for planning purposes.

Introduction

The increase of vehicle emissions in cities and the ongoing urbanization worldwide continue to deteriorate urban air quality within the urban canopy layer (UCL) [1], [2] which produces adverse effect to the health of people indoor and outdoor [3], [4]. Besides reducing pollutant emissions, improving urban ventilation through use of architectural modifications may help street-level pollutant dilution [5], [6], [7], [8], [9], [10], [11], [12], [13], [14].

Flow and pollutant dispersion within and above urban areas are commonly classified into four length scales, i.e. street-scale (∼100 m), neighbourhood-scale (∼1 km), city-scale (∼10 km) and meso-scale (∼1000 km) [15], [16], [17], [18]. The former three scales are micro-scale (∼100 m–10 km) for which the flow below building rooftops are explicitly solved. At this scale, due to pollutant accumulation effect, urban air quality depends upon their neighbourhoods and city-scale characteristics [15], [16]. Meso-scale modelling is usually employed to investigate regional pollutant transport in which urban areas are treated as roughness elements thus providing boundary conditions for smaller scale studies [17]. Within this framework flow and pollutant dispersion from street-scale to neighbourhood-scale have been widely investigated often coupling wind tunnel/field experiments with computational fluid dynamics (CFD) simulations [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [19], [20], [21], [22], [23], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [43], [44], [45]. CFD modelling has been rarely applied to city-scale (∼10 km) because it still requires too high computational costs to simulate urban airflows through thousands of buildings [18].

Street-scale and neighbourhood-scale studies usually disregard the larger-scale boundaries and emphasize the local parameters. For two-dimensional (2D) street canyon models, four flow regimes dependent on street aspect ratios (building height/street width, H/W) are reported [19], [20], [21], [22], [23], i.e. the isolated roughness flow regime (H/W < 0.3), the wake interference flow regime (0.3 < H/W < 0.67), the skimming flow regime with one main vortex (0.67 < H/W < 1.67), and multi-vortex flow regime (H/W > 1.67). For three-dimensional (3D) urban models, major urban morphological parameters are the building planar area index λ_p (i.e. the ratio between the planar area of buildings viewed from above and the total floor area) and the frontal area index λ_f (i.e. the ratio of the frontal area of buildings to the total floor area) [24]. 3D spare urban areas (for example λ_f = 0.0625) are more effective in removing pollutant [13], but have a lower efficiency of land utilization. Densely built-up urban areas usually results in poor ventilation conditions [7], [9], [10], [11], [12], [13], [14]. The most typical parameters of real urban areas are λ_p = λ_f = 0.25 [24]. As building packing densities are fixed, some other urban parameters are significant, including urban forms [8], [10], [11], urban size and building height variations [7], [8], [12], [29], ambient wind directions [5], [8], [14], [26], [27], [28], [29] etc. Thermal effect is another key factor. Field measurements showed temperature difference between air and building surfaces can reach up to 12–14 °C [31], [32]. If Richardson number (Froude number) is relatively large (small), the buoyancy force induced by air-wall temperature difference can affect or even dominate urban airflows [31], [32], [33], [34], [35], [36], [37], [38], [39], [40]. On the other hand, Giovannini et al. [46] showed using temperature data recorded in urban street canyons that when solar radiation is weak or absent, the temperature field remains mostly homogeneous.

In this context, fixing the medium building packing density (λ_f = λ_p = 0.25), we aim to quantify how urban sizes, building height variations, ambient wind directions and wall heating affect the capacity of removing pollutants at pedestrian level. To this purpose, pollutant transport rate (PTR) and its ratio (i.e. contribution ratio) (CR) [7] are applied to evaluate the relative contribution in pollutant removal by mean flows and turbulent diffusion across UCL boundaries. A new concept, the net escape velocity [41], is used to quantify the net capacity of pollutant dilution at pedestrian level.