The effect of street-level greenery on walking behavior: Evidence from Hong Kong

Abstract

Accumulating evidence shows that urban greenspaces have great health benefits, but establishing a causal relationship is difficult. It is often hypothesized that walking and physical activity are mediators in the relationship between urban greenspaces and health outcomes. Furthermore, most urban greenspace-physical activity studies have focused on parks rather than on landscaped streets, even though the latter are the most popular places for physical activity. The lack of research attention for landscaped streets is largely due to the fact that street greenery is difficult to measure, especially at eye level. Using readily available Google Street View images, we developed methods and tools to assess the availability of eye-level street greenery. A two-layered study was developed that 1) examined the association between urban greenspaces and the odds of walking (versus not walking) for 90,445 participants in the Hong Kong Travel Characteristics Survey and 2) carried out sensitivity analysis of the association between urban greenspaces and total walking time for a subset of 6770 participants. Multilevel regression models were developed to reveal the associations between street greenery and walking behaviors while controlling for sociodemographic characteristics and other activity-influencing built environment factors, taking into account the inherent clustering within the data. The results showed that both street greenery and the number of parks were associated with higher odds of walking; street greenery but not parks was associated with total walking time. Our results suggest that walking behavior is at least as strongly affected by eye-level street greenery as by parks. They also implicitly support the health benefits of urban greenspaces via walking and physical activity. With the large sample size, our findings pertain to the entire population of Hong Kong. Furthermore, the use of Google Street View is a sound and effective way to assess eye-level greenery, which may benefit further health studies.