The Use of GIS in Assessing Exposure to Airborne Pollutants

Sammanfattning: GIS (Geographical Information System) has become a valuable tool in air pollution exposure assessment studies due to its abilities to accurately model and estimate exposure for large populations and areas. The general aim of the work presented in this thesis was to investigate the factors that affect the validity and precision in the assessment of human exposure to airborne pollution using GIS. The first study examined the association between levels of nitrogen dioxide (NO2) and two socio-economic indices, and the influence of differences in the size and characteristics of study area on such associations. In the second study a GIS-based dispersion model was used to evaluate the spatial and temporal scale effects on modelled levels of nitrogen oxides (NOx) for urban and rural settlements in Scania. In the third study, modelled outdoor NO2 levels were compared with measured levels, and the relations between personal exposure and measured and modelled outdoor levels were examined. It was also investigated whether the accuracy between the two measurements could be improved by adopting a dynamic approach, i.e. by incorporating outdoor levels at the workplace during working hours. The findings showed that the study area size and socio-economic determinants, as well as the spatial and temporal resolution, must be chosen carefully. Modelled and measured outdoor NO2 levels agreed well, but outdoor levels at the residential address proved to be a poor predictor for personal exposure during a one-week observation period, which illustrates the importance of conducting more detailed time-activity-based exposure assessment studies. Spatial patterns of children's blood lead levels and proximity effects of two sources of lead exposure, a lead smelter and major roads, were identified in the fourth study using GIS. In conclusion, GIS techniques can be successfully applied in exposure assessments of air pollution, but factors that may affect the validity and precision of the results must be carefully considered.