Context: Both urban two-dimensional (2D) and three-dimensional (3D) morphology can affect air and land surface temperature. While many studies have looked at the impact of horizontal morphology, few have explored the relationship between vertical morphology and temperature, especially at the neighborhood scale. Objectives: This study aims to answer two questions: (1) Does air temperature vary in neighborhoods with different morphology? (2) If so, how does the 2D (horizontal) and 3D (vertical) morphology affect air temperature? Methods: We examined the relationship between morphology and air temperature for 24 residential neighborhoods in Beijing, using correlation analysis, regression analysis, and structural equation modeling. Morphological indicators were derived from remotely sensed land cover and light detecting and ranging (LiDAR) point cloud data. Air temperature was continuously measured using HOBO data loggers during the summer of 2014. Results
Nighttime air temperature was higher in neighborhoods dominated by high-rise structures compared to neighborhoods dominated by low-rise structures suggesting that 3D morphology is more important than 2D morphology in predicting air temperature. The ratio of vegetation volume to building volume negatively correlated with average air temperature and daytime temperature, while the mean distance among adjacent buildings had a positive effect. Building height was the most important predictor of nighttime air temperature. The major determinants of air temperature in high-rise and low-rise neighborhoods were different. Conclusions: Both 2D and 3D morphology can affect air temperature in residential neighborhoods. Increasing vegetation volume relative to building volume and decreasing the distance among buildings can reduce daytime air temperatures.