Search

It is essential to investigate the morphological factors that contribute to air pollution's spatial distribution using mobile monitoring data, and to regulate them at the urban planning level. However, mobile monitoring data are unstable and more difficult to model under real-world atmospheric circumstances. This work assesses the nonlinear relationship between spatial distribution of air pollutants and building morphological indicators in a high-density city based on mobile monitoring and machine learning. By conducting a vehicle-mounted mobile monitoring experiment, we establish spatial distribution data sets for PM 2.5 and PM 10 on three typical regions in Huangpu District, Shanghai. 9 indicators of urban morphology are derived, including green view index and sky view factor, using semantic segmentation and deep learning on street-view images. Correlation analysis demonstrates that the difficulty lies in implementing linear modeling methods. The performances of six machine learning algorithms for predicting the spatial variability of pollutants are compared. The result shows that neural networks have the highest performance for repidly predicting pollutant diffusion levels in conceptual designs. • Spatial distribution of pollutants is strongly related to urban morphology. • Long-term mobile monitoring of pollutants in high-density urban blocks has been conducted. • Semantic segmentation-based indicators such as green view index have been investigated. • Machine learning is better at identifying non-linear patterns in pollutants distribution. • An open neural network model has optimal performance and allows for ongoing improvement. [ABSTRACT FROM AUTHOR]

With increasing urbanization, urban air pollutants are becoming more and more relevant to human health. Here, combined with meteorological observation data, a numerical simulation of typical urban blocks in Shanghai was carried out to understand the spread of air pollutants caused by road traffic sources (ground–level and viaduct–level). Firstly, we analyzed the wind environment characteristics. Then, we quantitatively analyzed the pollutant distribution profiles and the contributions of two pollutant sources (PSV). Finally, we analyzed seven urban morphological parameters based on ventilation efficiency indices. Results revealed the following. (1) Ventilation patterns within the architectural complex are determined by local geometry; (2) Pollutants released at ground level were dominant when the Z–plane < 8 m high, and pollutants released from the viaduct source were 0.8–6.1% higher when the Z–plane ≥ 8 m high; (3) From ground level to a height of 60 m, the spatially–averaged normalized concentration (C*) tended to decrease gradually with distance from the source. C* increased irregularly with an increase in distance between 60 m and 86 m. Above 86 m, C* tended to increase linearly; (4) Vertical profiles of C* around buildings were building–specific, and their rate of change was inconsistent with height increases. In general, the correlations between C* and VRw, and between C* and KEturb were larger on the windward side of PSV upstream buildings than on the leeward side. Buildings downstream of the PSV showed the opposite situation; (5) At pedestrian level, the seven urban morphological parameters had no significant correlation with VRw, Cir*, and Czs*. [ABSTRACT FROM AUTHOR]