Comparison of Moderate Resolution Imaging Spectroradiometer (MODIS) and Multiangle Imaging Spectroradiometer (MISR) Aerosol Optical Depth Correlation with Particulate Matter (PM¬2.5) Concentrations at the Manila Observatory (14.635 N, 121.078 E)

HANZ VLADIMIR CABANES^1*, IRWIN ANGELO AMAGO¹, and NOFEL LAGROSAS^1,2

¹Physics Department, Ateneo de Manila University, Loyola Heights, Quezon City, Philippines

²Manila Observatory, Ateneo de Manila University, Loyola Heights, Quezon City, Philippines

hanzcabanes@yahoo.com

In this paper, a correlation analysis for both Moderate Resolution Imaging Spectroradiometer aerosol optical depth and Multiangle Imaging Spectroradiometer aerosol optical depth with ground-based PM_2.5 concentration measurements at the Manila Observatory for January to December 2011. The resulting analysis showed that MISR AOD had a better correlation (R²=0.7360) than MODIS AOD (R²=0.6793). Using this analysis, it was observed that the MISR AOD readings have a tendency to overestimate the ground-measured PM_2.5 concentration values while the MODIS AOD underestimates the PM_2.5 ground-measured data. Moreover, given the higher R² value, the results affirm those of previous studies, which say that MISR AOD is more accurate in determining PM_2.5 concentrations than MODIS AOD [1]. However, MODIS is known to have a better temporal resolution because it is onboard both TERRA and AQUA satellites whereas MISR is only onboard the TERRA satellite. To couple the individual advantages of these two data sets, the multivariate regression approach was used with the MODIS and MISR AOD values as predictors to the ground-measured PM_2.5 values. The resulting equation from the analysis was PM_2.5c=52.94MISR+37.10MODIS+15.33 where the PM­_2.5c is the computed PM_2.5 value, and MISR and MODIS represent the corresponding MISR AOD and MODIS AOD values, respectively. The computed PM_2.5 values from this equation yielded an absolute error of 21.85% and an R² of 0.7249 compared to the actual values. The resulting error value is lower than the 30-40% range often reported in literature, which affirms the utility of the multivariate regression method in relating satellite-derived data with ground-measured data [1, 2]. The analysis presented in this paper foregrounds further work geared towards the utilization of satellite data as a surrogate for ground-measured data.

References

[1]     Engel-Cox, J.A.; Hoff, R.M.; Haymet, A.D.J., “Recommendations on the Use of Satellite Remote-Sensing Data for Urban Air Quality,” Journal of Air & Waste Management Association. 54, 1360-1371, 2004.

[2]     Gupta, P., et al, “Satellite remote sensing of particulate matter and air quality assessment over global cities.” Atmospheric Environment 40, 5880-5892, 2006.

Presentation

Open publication - Free publishing - More aerosol

Open publication - Free publishing - More physics