Impacts of urban morphometric indices on ventilation

Abstract

Ventilation has shown strong sensitivity to urban geometrical heterogeneity; however, the quantitative connection with the urban morphometric indices, e.g. the packing densities (λf and λp for the frontal and plan area, respectively) and the building-height variability (σH), remains unclear. Using large-eddy simulation (LES), the dispersion of a pollutant scalar is investigated for turbulent boundary-layer (TBL) flows developed over real urban regions. While the time history of the domain averaged concentrations follow the familiar exponentially decaying function for all cases, the decay rates differ and there is a clear monotonic dependence on λf. The mean tracer age, τa, exhibits a strong linear scaling with λf as well (R2=0.8); however, an unambiguous relationship is hard to be discerned for the ventilation timescales versus the other morphometric indices (e.g. λp and σH). An effective canyon aspect ratio (AReff) is proposed for real urban districts, which allows ventilation to be quantified in a simpler manner as that for idealised canyons. The age spectrum, which measures the probability density function of tracer ages across multiple timescales, is analysed to reveal the mechanisms underlying the well-defined connection between ventilation and λf. Results are compared with that for an aligned building array of an equivalent canyon aspect ratio to AReff. The departures reflect the nonlinear effects that are not captured in idealised urban topography setups. The results could be useful for the parameterisation of ventilation in urban areas of complex geometrical profiles and may serve as a reference for urban layout design towards optimum active ventilation capability.