Urban parks have long been considered places of refuge from the bustle, noise and air pollutants of the surrounding city. Parks were often termed “the lungs of the city”, a metaphor which gained support as studies revealed the deposition of pollutants to leaf surfaces. However, emerging literature questions the long-accepted view that trees improve air quality in urban parks. Observations on air pollution in urban parks in Hong Kong, a compact city with dense high-rise buildings and limited land resources, show better environmental quality in park interiors compared with adjacent roadways. Pollutant decay profiles from roadside into park interiors were explored with analytical and numerical models. It is evident that decreases in pollutant concentrations largely arise from dispersion over the small spatial scales typical of urban parks, while deposition to foliage plays little role. Additionally, trees can reduce wind speed and potentially trap pollutants, so can increase localised concentrations.

Scarce land resources in dense cities means that small parks are an important leisure and amenity resource for the urban population. However, the short distances available for pollutant dilution potentially exposes users to air pollutants from nearby traffic. Nevertheless, appropriate park design could reduce such problems in crowded cities. Demographic observations in urban parks reveal a heterogeneous distribution of park users as they are constrained by park infrastructure and plantings. Overlapping the distribution of pollutants and users suggested that lower exposure can be achieved by thoughtful layout and design; particularly at the polluted border. Numerical simulations were used to assess the role of solid barriers (i.e. walls and earth berms) and porous barriers (i.e. trees and shrubs) within park borders in attenuating air pollution. Solid barriers reduced air pollutants noticeably in their shadow. A belt of dense conifers can trap pollutants in park borders and lower concentrations in interiors. However, more extensive plantings led to a decrease in wind velocity and hence an increase in pollutant concentrations. Tall and sparse trees tended to have little influence on airflow at the pedestrian level, which means they seem appropriate for small urban parks where wide barriers are not applicable and rapid ventilation should be encouraged. The source of pollution is another controlling factor. For example, the high concentrations just beyond bus stops observed through on-road measurements suggest that stops are better placed at the far-side of urban parks.

Greater awareness of the role of design in controlling pollutant exposure in urban parks could help reduce adverse effects on public health. Principles for achieving a healthier park environment can be obtained from this work, although there are conflicts with existing guidelines that need to be settled in thoughtful designs.