A study of urban heat island effect using ground-level stationary observation stations in Hong Kong

Abstract

The Urban Heat Island (UHI) is a phenomenon in which the urban temperature is higher than that of the non-urban areas. Previous studies from other countries showed that an UHI phenomenon exists in urbanized cities. In Hong Kong, instead of only few study using fixed stationary stations on UHI analysis, past effort have been concentrated on using traverse and remote sensing method to study the UHI. This dissertation therefore describes the annual, seasonal and diurnal UHI variability in Hong Kong based on hourly data of temperatures, wind speed and relative humidity from five selected fixed stations at urban and non-urban areas over the period 2001-2011.

UHI intensity is calculated by the temperature difference between the urban and non-urban areas excluding the global warming factors and micro-climatic effects. The results showed that, although there is no significant increasing and decreasing trend in annual UHI variation, the maximum annual UHI intensity is increasing and the nocturnal UHI is proved significantly higher than that its daytime counterpart. The seasonal UHI variation is higher in summer or autumn and winter depends on the choice of non-urban stations due to different geographic settings and land use. The diurnal UHI variation revealed that the strongest UHI intensity was observed in late afternoon or early morning while the weakest one occurred in afternoon. Among the study groups, Hong Kong International Airport experiences the highest UHI intensity. It is probably due to the geographical settings, unique land use and significant anthropogenic heat. On the other hand, in using Ta Kwu Ling as non-urban stations, the wind speed showed a significant negative correlation with UHI intensity while the relative humidity showed a positive one in diurnal UHI variation.

The results of the study indicated that the choice of non-urban station poses a significant impact to the UHI intensity calculation especially in seasonal and diurnal variations. Further study on choosing representative non-urban station for UHI intensity calculation is recommended.