Research Summary
S.No Reference Data and study area Key conclusions/Remarks
1. Wang and Christopher,2003 MODIS, 7 stations, Alabama Quantitative analysis with space and time collocated hourly PM2.5 and MODIS AOT. Demonstrated the potential of satellite data for PM2.5 air quality monitoring. (R=0.7)
2. Chu et al., 2003 AERONET, MODIS, PM10, 1 station, Italy Show relationship between PM10 and AOT. More qualitative discussion on satellite capabilities to detect and monitor aerosols globally. (R=0.82)
3. Hutchinson, 2003 MODIS AOT MAPS, MODIS Imagery, GEOS Imagery , PM2.5, Texas Shows potential of MODIS data in monitoring continental haze over land surface. No correlation analysis
4. Engel-Cox et al., 2004 MODIS, PM2.5 Continental United States First study, which present correlation analysis over entire USA and discuss difference in relationship over different regions. Qualitative and qualitative analysis over larger area, demonstrated spatial distribution of correlation. Range of R.
5. Hutchinson et al., 2004 MODIS AOT maps, Ozone, Eastern USA Used few MODIS AOT maps and discussed the hazy conditions, no correlation analysis, and more emphasis on ozone pollution.
6. Liu et al., 2004 MISR, GEOS-CHEM GOCART, USA First used MISR data for air quality study and have emphasis on seasonal and annual mean correlation analysis and forecasting. (R=0.78)
7. Engel-Cox et al., 2004 MODIS Recommendations to use satellite data into air quality applications. These data sets can add synoptic and geospatial information to ground-based air quality data and modeling.
8. Liu et al., 2005 MISR , GEOS-3 Meteorology, USA Regression model development and forecasting of PM2.5, model generated coarse resolution meteorological fields are used and focused only in Eastern United States. 48% explanation of PM2.5 variations.
9. Al-Saadi, J., et al., 2005 MODIS, USA More descriptive paper on IDEA program, which provides online air quality conditions from MODIS and surface measurements over several locations in the USA
10. Hutchinson et al., 2005 MODIS, Texas Correlation analysis in Texas. Correlation varies from 0.4 to 0.5 and long time averaging can make correlation greater than 0.9
11. Engel-Cox et al., 2005 MODIS, USA Potential of satellite data for monitoring transport of PM2.5 over state boundaries and event specific analysis.
12. Gupta et al., 2006 MODIS, Meteorology, Global 21 locations Correlation varies from 0.37 to 0.85 over different part of the world. Cloud fraction, relative humidity and mixing height information can improve relationship significantly. First study covered several global locations.
13. Engel-Cox et al., 2006 MODIS, LIDAR, USA Weak correlation can be significantly improved by using vertical aerosol information from LIDAR measurements.
14. van Donkelaar et al., 2006 MODIS, MISR, PM2.5, GEOS-CHEM, USA and Global Inter-comparison between MODIS and MISR over several locations in Canada and USA. R= 0.69 (MODIS) and R= 0.58 (MISR). Different approach used to calculate the fine mass concentration.
15. Koelemeijer et al., 2006 MODIS, PM2.5 and PM10, Europe Mainly focused on Europe. Correlation varies from 0.5 for PM10 to 0.6 for PM2.5. Use of boundary layer height in analysis improved the relationship.
16. Kacenelenbogen et al., 2006 POLDER, France Intercomparison between POLDER AOT and PM2.5 over 23 sites during April-October, 2003. Mean R value is 0.55 with maximum of 0.80.
17. Liu et al., 2006 MODIS, MISR, RUC Inter-comparison between MODIS and MISR in St. Louis area. MISR performed slightly better than MODIS in the region.
18. Gupta et al., 2007 MODIS, Sydney, Australia Impact of bushfires on local air quality has been studies using both ground and satellite measurements. The quantitative analysis shows up to 10 fold increments in surface level PM2.5 during fires.
19. Gupta and Christopher, 2008a MODIS, Birmingham, AL, USA Provide detailed assessment on satellite remote sensing of air quality. Issues like, MODIS AOT quality flags, cloud contamination, sampling bias, long term trends, AOT averaging has been discussed using almost 7 year data sets.
20. Kumar et al., 2007 MODIS (5km), Delhi, India Three months PM2.5 data from a field campaign were used. Correlation between PM2.5-AOT was 0.52(+/-)0.20
21. Kumar et al., 2008 MODIS (5km), Delhi, India Same as Kumar et al., 2007, plus more analysis on PM10-AOD relationships
22. Gupta and Christopher, 2008b MODIS, Southeast USA Long terms trends in air quality using satellite data could be affected due to sampling bias. Average bias value is about 2ugm-3 on monthly scale for SE USA.
23. Schaap et al., 2008 AERONET, MODIS, LIDAR, Netherlands Time varying relationship between AERONET and PM2.5 over single station where R values changes between 0.63 and 0.85. LIDAR data are used to cloud clear AERONET level 1.5 data sets.
24. Liu et al., 2008a MISR, USA Describe method of estimating PM2.5 mass and its major constituents using fractional AOD values from different aerosol types in MISR algorithm.
25. Liu et al., 2008b MISR, USA Method developed in Liu et al., 2008a is used for case study over EPA STN sites. Attempt to estimate SO4 and NO3 is made, which compares well with surface observations.
26. Martin and Canada, 2008 Review Mostly focused on gaseous air quality but also provide some review on particulate matter air quality from satellite observations.
27. Hutchinson et al., 2007 MODIS, LIDAR An attempt is made to improve AOT-PM2.5 relationship by refining MODIS AOT product, optimizing averaging area for MODIS pixels around surface station.
28. Paciorek et al., 2008 GASP, MODIS, MISR, USA Relationship between AOT derived from geostationary platform and PM2.5. Suggestion to calibrate GASP product for particulate matter applications. R values ranges from 0.41 to 0.51
SUNDAR  |  GUPTA  |  IDEA  |  AEROSOLS BLOG  |  SMOG BLOG  |  AIRE  |  AIRNOW  |  GIOVANNI  |  SITE ADMIN
Copyright © 2010 www.airqualityresearch.net. All rights reserved. | Designed and developed by Ambarish Mugalikar