Modelagem espacial da exposição humana às fontes de poluição do ar no Distrito Federal : o uso e ocupação do solo como variável preditora

Abstract

Air pollution sources are associated with harmful effects on human health. Cardiorespiratory diseases occurrence are indicators of human exposure to air pollutants. The measurement of this exposure is used as collaborative toll for sources control and management in order to promote quality of life to population. Among the methods for measurement (monitoring station, remote sensing, land use models, biomonitoring, portables devices), the challenge is related to the equilibrium between the aim of the measurement and the cost. Faced with this operational and economic challenge, most of the urban centers still have a gap to measure human exposure to air pollution, for example, the Federal District (FD). The method which represents the land use models is prominent for the challenges in urban areas as the FD, and also for urban areas which already have a monitoring system consolidated. However, the land use models are in initial process of research, which need more development. For instance, it is still unclear the understanding of the spatial relationship between land use, air pollution sources and human health. Therefore, this research tested the hypothesis that in the FD human exposure to air pollution sources can be estimated by classification and measurement of land use. It was considered as land use in this study three categories: vehicular traffic, conventional land use, and land use related to urban typology-Urban Structure Types (UST). It was addressed seven specific objectives in this research: 1) to estimate emissions from the roads; 2) to create a geodatabase for health data; 3) to assess the relationship between health and vehicle emissions; 4) to evaluate spatial patterns of cardiorespiratory diseases; 5) to identify high-priority areas to measure human exposures to air pollution sources; 6) to estimate the cardiorespiratory disease risk related to land use; and 7) to evaluate the relationship between UST and cardiorespiratory diseases risk. The methods are as follows: bottom-up to calculate vehicular emissions; Modeling in Geographic Information System (GIS) for geodatabase creation; Geostatistical analysis with Global Moran's I, Getis-Ord General G, semivariogram analysis and k-function; Analytic Hierarchy Process (AHP) for decision making and values judgment, and; Ordinary Least Square regression (OLS) and quantile regression for assessment of the relationship between sources and human health. The findings are as follows: the light vehicles presented a different (cluster) spatial patters of air pollution emissions; Spatial patterns of cardiorespiratory diseases are clustered; There are no air pollution monitoring stations in areas which were identified with high priorities levels to measure human exposures; the number of light vehicles was associated with hospital admissions risk (risk = 6 admissions; 95% CI: 2.6; 14.6); A 2,500 m increase in highways was associated with a 46% increase in cardiorespiratory diseases; A 6,000 m increase in streets and avenues was associated with a 51% increase in cardior...