Spatial Interpolation Methodologies in Urban Air Pollution Modeling: Application for the Greater Area of Metropolitan Athens, Greece

Air pollution in urban environments has serious health and quality of life implications. A wide variety of anthropogenic air pollution sources increase the levels of background air pollutant concentrations, leading to the deterioration of the ambient air quality. Principal sources of urban air pollution are vehicular traffic, industrial activity and in general fossil fuel combustion, introducing a mixture of chemical components, particulate matter and biological material into the atmosphere. The deterioration of urban air quality is considered worldwide one of the primary environmental issues and current scientific evidence associate the exposure to ambient air pollution with a wide spectrum of health effects like cardiopulmonary diseases, respiratory related hospital admissions and premature mortality (Analitis et al. 2006; Ito et al., 2005; Samet et al., 2000). Direct measurements of sensitive population groups’ exposure to air pollution are scarce and therefore methods of accurate point and areal air quality estimations are prerequisite. This fact highlights the importance of generating accurate fields of air pollution for quantifying present and future health related risks. In the field of air pollution modeling, two different approaches have been adopted by the scientific community, differentiated by their applied fundamental principles. The first approach involves the numerical simulation of atmospheric dispersion based on the current understanding of physics and chemistry that govern the transport, dispersion and transformation of pollutants in the atmosphere. The modeling process typically requires a set of parameters such as meteorological fields, terrain information along with a comprehensive description of pollution sources. An alternative approach is based on statistical analysis of pollutant concentrations collected from air quality monitoring networks commonly deployed in urban areas. The reasoning of the statistical approach is that physical processes are likely to induce correlations in air quality data collected over space and time. Statistical models generate predictions by exploiting these spatio-temporal patterns, enabling the estimation of pollutant concentrations in unmonitored locations. The chapter’s main objective is to present and review the statistical spatial interpolation methodologies which are commonly employed in the field of air pollution modeling. An additional scope of the chapter is to compare and evaluate the accuracy of the interpolation methods for point estimations, using data from a real urban air quality monitoring network located at the greater area of metropolitan Athens in Greece.