Studies are still required to understand how rural/marine wind remove ground-level pollutants released uniformly in street networks of high-rise urban areas. The link between building height variability and pollutant removal process still remains unclear. Several idealized urban-like neighbourhoods made of 9-row and 18-row small-scale high-rise square arrays (building width B = street width W, building packing density λ_p = 0.25) were first numerically studied with a parallel approaching wind and neglecting thermal effects. Normalized pollutant transport rates and pedestrian purging flow rate were applied to quantify the contribution of pollutant removal by mean flow and turbulent diffusion and their net purging capacity.

Results show that the prediction of isothermal turbulent flows agreed generally well with wind tunnel data. For 9-row arrays with building height variations (standard deviation of 0–57.1%) and the same average canopy height (H₀ = 2.33W), pollutant removal mainly depends on mean flows. Larger standard deviations tend to induce better pedestrian ventilation. In comparison to small and large standard deviations, medium values of 14.3–42.9% may experience smaller purging capacity by horizontal mean flows but significantly enhance that by vertical mean flows. For arrays with uniform heights, lowering aspect ratios (H/W = 2.33 and 2.67–1.5) or increasing street lengths (9-row to 18-row) may enhance the contribution of removing pollutants by turbulent diffusions across canopy roofs which may be similarly important as that by mean flows. Although further investigations are still required, this paper clarifies the relationship between building layouts, height variability and removal potential of ground-level pollutants in high-rise urban-like geometries.