Influence of urban design factors on summertime urban heat island intensity : on-site measurement of pocket parks in high-rise high-density environment in Hong Kong

Abstract

Due to urbanization and climate change, the UHI effect has become a modern endemic threat to human health and energy consumption, especially in highly populated compact urban areas like Hong Kong. One way in which a city can address this problem is to increase the amount of green spaces, which act as “urban lungs.” Unlike lower-density cities, however, the scope for expanding green open space (referred to as “pocket parks”) within Hong Kong’s urban fabric is limited, and in any case, the city’s green lungs cannot be distributed optimally due to the complex urban fabric. The research therefore is conducted to examine the scope for re-configuring pocket parks and surrounding building forms to ameliorate summer UHI intensity in Hong Kong. Previous research on urban climate and urban design has focused on the fundamental thermophysics knowledge stage and hasn’t successfully transferred to applied science. In order to fulfill part of the knowledge gap, this study endeavors to propose climate-sensitive urban design guidelines through investigating the relationships between summer UHI intensities and 34 urban design factors, focusing on land-use intensity, built form, space enclosure, street-building dimension and vegetation ratio, which are either well known by professionals in design practices, or are more accurate for addressing complex design features in real urban environments. Statistical analysis based on field measurement is adopted. Climate measurements and site survey in 12 research sites with pocket parks in Hong Kong were conducted, in which 195 points were investigated. A spatial size of 400 m × 400 m was adopted in defining site boundary. Trend analysis, bivariate regression analysis and multiple regression analysis were conducted to answer three sub questions respectively. Six sub data sets were separately examined in order to control the influence from factors that were beyond urban design scope. Two data sets were identified as the most representative data and then used in multiple regression analysis. Results show that the UHI intensity in nighttime (mean value: 2.39℃) in Hong Kong is much higher than during the daytime (mean value: 0.1 ℃). Ten parks are cooler than their surrounding urban streets for both daytime and nighttime. The tree ratio of each park is highly related to the air temperature difference between the insides of parks and on the surrounding streets within each site, but not the UHI intensity between different sites. Twenty-one variables are significantly related with daytime UHI, while twelve variables are related with nighttime UHI. The R2 values of daytime regression models range from 0.524 to 0.780, while nighttime models range from 0.543 to 0.588. Nighttime UHI measurements are highly affected by anthropogenic heat, which dilutes the influence of physical design features. It is found that site-level design features have higher explanatory power on the UHI intensities than point level, although point-level features can also explain some of the air temperature variation within the sites. Design implications based on these analyses are proposed in the thesis.