Objective and subjective outdoor thermal comfort in Hong Kong's hot season

Abstract

Frequent visits to urban green spaces (UGS) are known to improve the general health of citizens. Creating a thermally comfortable environment is essential to encourage UGS usage. Shading is critical to thermal comfort in Hong Kong’s hot season. In this study, a field measurement was conducted to measure the cooling effects of two shading structures – tree and concrete shelter. The cooling magnitudes are compared in two thermal indices, namely PET and UTCI. These thermal indices serve to evaluate objective thermal comfort, but the thermal perception of people is dependent upon seasonality, geographical location and subjective judgement. Local studies are necessary to evaluate subjective thermal comfort; therefore, this study also evaluates the thermal perception of people through a concurrent meteorological measurement and questionnaire survey. A literature review was conducted to assist in the determination of four important thermal benchmarks that reflect people’s subjective thermal perception – neutral temperature, preferred temperature, neutral temperature range and acceptable temperature range. Previous studies showed that human can constantly adapt to the ever-changing thermal environment, which was reflected in the changing levels of neutral temperature across different studies. A global database of neutral temperature will be established to identify the trend and limit of such adaptation. The field measurement showed that the mean daytime cooling effects generated by the tree was 0.6 °C (air temperature), 3.9 °C (PET) and 2.5 °C (UTCI), which were higher than the shelter at 0.2 °C, 3.8 °C and 2.0 °C respectively. The differences were significant for air temperature and UTCI (p<0.001 and p<0.05 respectively, t-test) but not for PET (p=0.261). The tree’s mean daytime maximum cooling effects were 2.1 °C (air temperature), 18.8 °C (PET) and 10.3 °C (UTCI). In the subjective thermal comfort study, 427 valid questionnaires were collected. The acceptable temperature ranges of the sample in PET and UTCI were 21.3 to 39.5 °C and 22.7 to 38.8 °C respectively. The questionnaire data showed that 55.5 % of the respondents found it unacceptable to stay at the interview site for an extra hour although they felt acceptable during the interview. A new benchmark range – 1-hour 80 % acceptable temperature range – is therefore proposed to assess the usability of outdoor space in a more relevant time scale. The neutral temperature data in 19 international empirical studies were analysed using linear regression. Survey-period mean air temperature was found to be the best predictor of neutral temperature (R2=0.539, p=0.007). The temperature limit of thermal neutrality was between 12.9 °C and 32.3 °C in air temperature. The findings indicate that the adaptation effect has a worldwide pattern as neutral temperature always lies closely to background air temperature. However, global warming is likely to push air temperature further outside the upper neutrality limit in many parts of the world in the hot season. This study provides useful recommendations to urban planning and landscape design through a combination of on-site meteorological measurement and questionnaire survey. The literature review on thermal benchmark also addressed technical issues in data analysis in thermal comfort research.