Abstract: A study on vertical variation of PM2.5 concentrations was carried out in this paper. Field measurements were conducted at eight different floor heights outside a building alongside a typical elevated expressway in downtown Shanghai, China. A back propagation neural network based on principal component analysis (PCA-BPNN), was applied to predict the vertical PM2.5 concentration and examined with the field measurement dataset. Experimental results indicated that the PCA-BPNN model provides reliable and accurate predictions as it can reduce the complexity and eliminate data co-linearity. Furthermore, this paper investigated the vertical distribution of PM2.5 and their relationship with traffic volume, weather and height by generalized additive model (GAM). These findings reveal the vertical distribution of PM2.5 concentration and the potential of the proposed model that will be applicable to predict the vertical trends of air pollution in similar situations.

Key words: urban elevated expressway, vertical variations, generalized additive model (GAM), principal component analysis (PCA), back propagation neural network (BPNN)