Modeling the Spatial-Temporal Variability of NO2 and Its Relationship to the Prevalence of Low Birth Weight in Georgia
Title:  Modeling the Spatial-Temporal Variability of NO2 and Its Relationship to the Prevalence of Low Birth Weight in Georgia
Author:   
co-Authors: 
Organization: 
Personal profile:  Login required for author information!
Conference presentation:  The 19th International Conference on GeoInformatics, Shanghai, China
Abstract:  Air quality has become a great concern to the public, researchers, and policy-makers, as extensive research has demonstrated that air pollutants harm the health of humans. Despite the significant spatial variation of air quality across Georgia, there are only a very small number of ground stations installed in the state and most of them are located around the Atlanta area. The sparseness of ground data has limited to a large degree our understanding of air quality in terms of its spatial-temporal distribution and effects to public health, because the traditional method of estimating regional air quality has been through spatially interpolating point air quality data collected by monitoring stations. Recent advance in geospatial technologies, particularly in remote sensing, has offered unprecedented opportunities to model air pollutant distribution with fine resolution and spatial continuity. In this research project, we take advantage of the technological advancement to investigate the spatial-temporal variability of air quality in Georgia primarily based on remotely sensed data, and link the distributional patterns of air pollutant NO2 to public health problems from a perspective of environmental epidemiology. Air quality data sensed by the Ozone Monitoring Instrument (OMI) onboard the Earth Observing System (EOS) satellite of AURA will be collected, processed, and then compared to the ground measurement. Spatial analytical techniques including hotspot analysis, spatial autocorrelation, and spatial-temporal cluster analysis will be employed to characterize spatial patterns while time series analysis will be conducted to address seasonality and year-to-year trend of air quality. Low birth weight (LBW) is chosen as a public health outcome of interest because it is a significant health issue in the state as well as in the country and the risk of LBW has recently been found to be associated with ambient air quality. County-level LBW rates will be spatially regressed upon air quality to determine a numerical relationship between the two. Results gleaned from this preliminary study will be used for more refined discovery studies well suited for publication and external grant applications.
Full-text paper link: