480356 Improving the Estimates of Particulate Matter Impact on Human Health through Remote Sensing

Monday, November 14, 2016
Grand Ballroom B (Hilton San Francisco Union Square)
Wan Nurlaila Mat Desa, Chemical and Biochemical Engineering, University of Iowa, Iowa City, IA and G. R. Carmichael, Department of Chemical & Biochemical Engineering, University of Iowa, Iowa City, IA

Improving the Estimates of Particulate Matter Impact on Human Health through Remote Sensing


Wan Nurlaila Mat Desa1, Gregory Carmichael1,2

1 Department of Chemical and Biochemical Engineering, University of Iowa

2 Centre for Global and Regional Environmental Research


Exposure to particulate matter severely affects human health as approximately 7 million people die from air pollution each year. There is an increasing need for accurate documentation of particulate matter in the atmosphere in order to provide more accurate estimates of exposure, and to document effectiveness of strategies to mitigate the adverse impact of pollution on human health. Currently surface based observations of particulate matter are done at only a limited number of locations within a city. Satellite based observations of aerosol optical depth (AOD) provide the possibility of adding more information on the spatial distribution of PM, which would increase the accuracy of exposure estimates. However the relationship between AOD and PM2.5 are complicated. This study examines how well aerosol behaviour can be predicted using atmospheric remote sensing data in Eastern Asia. To achieve the objective of the study, the relationship between PM2.5, PM10, and AOD was examined using observational data, taken from the KORUS-AQ campaign where samples from Gwangju station were selected for analysis. The relationships found in the observations were also compared to those predicted using the Weather Forecasting and Research (WRF) model.

Linear correlation between PM2.5 and AOD were determined from linear regression method. For each set of data obtained from Weather Forecasting and Research model (WRF-Chem) and observational (ground station and AERONET), plots of PM2.5:AOD to AOD values were constructed to give distribution slope (m) and correlation coefficient from the relationship of . The correlation of determination, R2 obtained ranged from 0.167 to 0.530, while the slope ranged from -142.1 to -53.7. Linear regression function of observation PM to AERONET data showed the strongest correlation, with R2 of 0.530, while model to model analysis only yielded 0.167. The direct comparison between the computed PM to AOD from yielded the lowest R2 value. Strong correlation between observed PM and AERONET implies that measured AOD is a fair indicator of the PM concentration in Gwangju, compared to computed AOD using the WRF model. Values obtained from model simulations were higher compared to ground observations, indicating that more efforts are needed to improve the model capabilities. While this is the first step in improving the prediction of ambient particulate matter, the next step should incorporate space-based measurements that can provide broader spatial coverage.

Extended Abstract: File Not Uploaded
See more of this Session: Undergraduate Student Poster Session: Environmental
See more of this Group/Topical: Student Poster Sessions