Remote sensing of exposure to NO2; satellite versus ground based measurement in a large urban area

摘要

Background: Satellite remote sensing may be a powerful and useful tool for exploring spatial variability of air pollution within an urban area, particularly in countries or areas where direct measurement is lacking.Aims: To evaluate the ability of the Ozone Monitoring Instrument (OMI; satellite: Aura) to resolve urban-scale gradients in ground-level nitrogen dioxide (NO2) for a large urban area.Methods: We compared estimates of surface NO2 concentrations derived from OMI measurements and US EPA ambient monitoring stations. OMI provides daily afternoon (～13:30 local time) measurements of NO2 tropospheric column abundance. We used model-derived scaling factors (surface-to-column ratios, from GEOS-Chem and CAMx) to relate satellite column measurements to ground-level concentrations. We compared 4138 sets of paired data for 25 monitoring stations in the South Coast Air Basin of California for year-2005.Results: OMI measurements include more data gaps than the ground monitors (60% versus 5% of available data, respectively), owing to cloud contamination and imposed limits on pixel size. The spatial correlation between OMI columns and corrected in situ measurements is strong (r = 0.93 for annual average data), indicating that the within-urban spatial signature of surface NO2 is well resolved by the satellite sensor. Satellite-based surface estimates employing scaling factors from an urban model provide a reliable measure (annual mean bias: -13%; seasonal mean bias: ＜1% [spring] to -22% [fall]) of fine-scale surface NO2. OMI provides good spatial density in the study region (average area [km2] per measurement: 730 for the satellite sensor vs. 1,100 for the monitors).Conclusions: Satellite-derived estimates of NO2 from OMI provide a reliable measure of spatial variability in ground-level NO2 for a large urban area. Our findings suggest that OMI NO2 measurements are a useful tool for exploring urban air pollution, especially for locations without extensive ambient monitoring.