With the continuous renewal of urban greening, pollen released by allergenic tree species has become a prominent environmental issue affecting residents’ health. However, existing research still lacks city-wide, rapidly replicable methods for identifying allergenic tree species and assessing exposure risks. Taking Beijing’s central urban districts as a case study, this research establishes a method for the automated identification of allergenic tree species and the assessment of pollen exposure risks based on high-resolution satellite imagery. This study coupled tree species distribution results derived from model inference with population density per unit area to delineate three tiers of exposure risk zones. Subsequently, these risk zones were overlaid with the road network within the study area to determine the distribution of roads with low, medium, and high exposure risk. Public transport stop locations were then introduced as a proxy variable for areas of high population mobility. Lorenz curves and Gini coefficients were calculated to quantify the spatial equity of pollen exposure risk. The results indicate that the model reliably identifies target tree species, with approximately 117,000 valid targets. Exposure risks exhibit significant clustering characteristics and can form continuous expansions along road networks. Incorporating population factors shows minimal change in risk concentration, suggesting pollen exposure risk is primarily driven by the spatial clustering of allergenic tree species and their accessibility within road networks. This risk is highly correlated with the spatial distribution patterns and accessibility characteristics of allergenic tree species, rather than being solely determined by population size. This study provides foundational data and methodological support for urban tree species identification, pollen exposure risk management, and optimised greening configurations.
Xu et al. (Sat,) studied this question.