关键词:
摘要: Particulate matters (PM) at the pedestrian level significantly raises health impacts in compact urban environment of Hong Kong. A detailed investigation fine-scale spatial variation pedestrian-level PM is necessary to assess risk pedestrians outdoor environment. However, collection data difficult Kong due limited amount roadside monitoring stations and complicated context. In this study, we measured variability three most representative commercial districts using a backpack environmental measuring unit. Based on measurement data, 13 types geospatial interpolation methods were examined for mapping PM2.5 PM10 with group building geometrical covariates. Geostatistical modelling was adopted as basis PM. The results show that original cokriging exponential kernel function provides best performance mapping. Using features covariates slightly improves performance. study also imply fine-scale, localized pollution emission sources heavily influence exposure