Contribution of local climate zones to the thermal environment and energy demand

Abstract

Urban heat islands (UHIs) and their energy consumption are topics of widespread concern. This study used remote sensing images and building and meteorological data as parameters, with reference to Oke's local climate zone (LCZ), to divide urban areas according to the height and density of buildings and land cover types. While analyzing the heat island intensity, the neural network training method was used to obtain temperature data with good temporal as well as spatial resolution. Combining degree-days with the division of LCZs, a more accurate distribution of energy demand can be obtained by different regions. Here, the spatial distribution of buildings in Shenyang, China, and the law of land surface temperature (LST) and energy consumption of different LCZ types, which are related to building height and density, were obtained. The LST and energy consumption were found to be correlated. The highest heat island intensity, i.e., UHILCZ 4, was 8.17°C. The correlation coefficients of LST with building height and density were -0.16 and 0.24, respectively. The correlation between urban cooling energy demand and building height was -0.17, and the correlation between urban cooling energy demand and building density was 0.17. The results indicate that low- and medium-rise buildings consume more cooling energy.

Keywords: air temperature inversion; degree-days; energy consumption; neural network; urban heat island.