QUANTITATIVE EVALUATION OF SATELLITE AEROSOL DATA FOR MAPPING FINE PARTICULATE AIR POLLUTION IN THE CONTERMINOUS UNITED STATES

摘要

Introduction Existing studies have shown that fine aerosol particles have more adverse health effects than course particles and gases. Air quality monitoring has long relied on ground measurement at point monitoring si tes. Ground monitoring data often lacks spatially and temporally complete coverage. In recently years, repetitive and broad coverage capabilities of satellites allow atmospheric remote sensing to offer a unique opportunity to monitor air quality at continental, national and regional scales. The NASA Earth Observation System (EOS)'s MODIS (M oderate Resolution Imaging Spectrometer) aerosol optical depth (AOD) da ta could be used to improve ground measurement of fine particulate matter.Objectives This paper quantitatively examin ed the relationship between PM_(2.5) concentration and MODIS AOD data in the co nterminous U.S. and assessed the potential of using AOD data as an indicator of particulate air pollution. If an association should be found, AOD data would be used to map particulate matter concentration.Methodology Relationship between AOD and ground-based PM_(2.5) observations was examined at two levels. One used MODIS Level 2 data, and the other used Level 3 annual mean data. Daily MODIS Level 2 imag es covering the conterminous U.S. were compared with hourly PM_(2.5) data for the year 2004 with imaging time collated with PM_(2.5) measurement time within one hour and monitoring site locations matched to conjugate pixels. The Pearson's correlation value was calculated for each point containing at least 8 pairs of data values. For the year 2004, a geographically weighted regression (GWR) model was also fitted to examine the spatial variation of the relationship between PM_(2.5) and AOD using annual mean PM_(2.5) and MODIS Level 3 AOD. The possible association between AOD and PM_(2.5) was then used to predict annual mean PM_(2.5) values using AOD Level 3 annual mean values and the prediction accuracy was assessed against ground measurement.Results Significant positive correlations between PM_(2.5) and AOD were found to be to the east of the -100° longitude line. Eighteen of the twenty sites with highest correlation values (r>0.8) are in the east. GWR predicts well in the eastern U.S. and poorly in the west, as indicated by the map of local R square. The coefficient raster surface (AOD) exhibits regional variation. The relationship between PM_(2.5) and AOD is not spatially consistent (stationary) across the conterminous states. Eastern U.S. shows higher AOD coefficient values, while values in the west are lower. PM_(2.5) calculation for the east and the year 2003 using the linear regression equation fitted on the year 2004 data achieved an accuracy of 2.24 μm/m~3. Conclusions EPA PM _(2.5) ground data is positively related to MODIS AOD data in the eastern U.S. where the aerosol retrieval algorithm used the urban/industrial aerosol model. Urban-industrial aerosols are mainly from fossil fuel combustion in populated industrial regions and dominated by fine par ticles. The positive rela tionship in the east could exist in other regions in the world where the aerosol retrieval algorithm uses the urban/industrial aerosol model. Satellite aerosol optical depth data can be used as an alternative indicator of air quality for regions dominated by anthropogenic fine model aerosol particles.Introduction Fine particulate matter (PM) causes more visibility degradation and possibly more health problems than do gases. Air qua lity monitoring has long relied on ground measurement at individual monitoring sites. Ground monito ring data often lacks spatially complete coverage. In recently years, repetitive and broad coverage capabilities of satellites allow atmospheric remote sensing to offer a unique opportunity to monitor air quality at continental, national and regional scales. This project examined the relationship betweenPM_(2.5) concentration and aerosol optical de pth (AOD) measured by the NASA MODIS (Moderate Resolution