Assessing spatial variability of extreme hot weather conditions in Hong Kong: A land use regression approach

The number of extreme hot weather events have considerably increased in Hong Kong in the recent decades. The complex urban context of Hong Kong leads to a significant intra-urban spatial variability in climate. Under such circumstance, a spatial understanding of extreme hot weather condition is urgently needed for heat risk prevention and public health actions. In this study, the extreme hot weather events of Hong Kong were quantified and measured using two indicators – very hot day hours (VHDHs) and hot night hours (HNHs) which were counted based on the summertime hourly-resolved air temperature data from a total of 40 weather stations (WSs) from 2011 to 2015. Using the VHDHs and HNHs at the locations of the 40 WSs as the outcome variables, land use regression (LUR) models are developed to achieve a spatial understanding of the extreme hot weather conditions in Hong Kong. Land surface morphology was quantified as the predictor variables in LUR modelling. A total of 167 predictor variables were considered in the model development process based on a stepwise multiple linear regression (MLR). The performance of resultant LUR models was evaluated via cross validation. VHDHs and HNHs were mapped at the community level for Hong Kong. The mapping results illustrate a significant spatial variation in the extreme hot weather conditions of Hong Kong in both the daytime and nighttime, which indicates that the spatial variation of land use configurations must be considered in the risk assessment and corresponding public health management associated with the extreme hot weather. [Display omitted] •Applying land use regression (LUR) in the spatial estimation of extreme hot weather conditions.•Reducing spatial uncertainties in the heat-related health impact assessment.•Land surface morphology was integrated and identified as influential factors.•VHDHs and HNHs were mapped at the community level in Hong Kong.•Significant spatial variations were found under the extreme hot weather conditions.