Impacts of urban-scale building height diversity on urban climates: A case study of Nanjing, China

• The relationship between building height and urban climates was explored. • The urban-scale simulation was carried out. • The energy consumption of HRB caused by UHI is about 1.2–4.5%. • The UHI intensity of HRB is 0–0.7 °C higher than that of MRB and LRB. • The suggested building height is set at 19–50 m in the rapid urbanization process. The rapid development of urbanization is accompanied by the diversification of building heights. It has caused cities to face problem of non-uniform distribution of urban climate parameters (e.g., temperature, velocity), may further leading to urban heat island (UHI) and increasing of building energy consumption etc. Thus, it is of great significance to explore the associated relationships between building height and urban climates from the urban scale by CFD simulation. There is almost a gap on urban scale simulation for both research and engineering because of the complex building diversity and expensive simulation cost. K-means clustering and low-dimensional method based on image recognition were utilized in this work to build the urban geometric model, further improving simulation efficiency. A case study of Nanjing city was selected (about 144 km 2 containing 140 thousand buildings). The urban-scale simulation method was verified by experimental data. The numerical results show that the energy consumption of high-rise buildings (HRB) caused by UHI is about 1.2–4.5 %. The UHI intensity of HRB is 0–0.7 °C higher than that of middle-rise buildings (MRB) and low-rise buildings (LRB), while the velocities are similar around three different buildings height. In order to make the pedestrian area (at the height of 1.7 m from the ground) with pleasant urban climates in rapid urbanization process, it is suggested to set the building height between 19 and 50 m. The findings can be beneficial to alleviate the problems of UHI and energy consumption, which have important implications for urban optimization design.