AAAR 35th Annual Conference October 17 - October 21, 2016 Oregon Convention Center Portland, Oregon, USA
Abstract View
On the Applicability of Aerosol Optical Depth Retrievals as a Proxy for Surface Particulate Matter in India
Karen Xia, DANIEL WESTERVELT, Columbia University
Abstract Number: 488 Working Group: Remote and Regional Atmospheric Aerosols
Abstract India is home to some of the world’s poorest air quality, with ambient concentrations of fine-mode particulate matter (PM2.5) often exceeding World Health Organization daily health guidelines by as much as an order of magnitude. However, the spatio-temporal variability of air pollution in India is not well known. While PM2.5 monitoring stations exist in some major cities and at U.S. Embassies in India, the amount of available data across the Indian subcontinent is somewhat sparse, especially in comparison to networks in the U.S. or Europe. The use of satellite observations has been proposed as a potential source of air quality data where ground-based measurements are lacking, since satellites can provide more complete spatial coverage than possible with ground-based monitoring. We aim to evaluate the relationship between available ground-based PM2.5 measurements and satellite aerosol optical depth (AOD). In particular, we compare the spatial and temporal variability in high resolution MODIS AOD over India to that detected by ground-based PM2.5 and PM10 measurements across India. We correlate daily and monthly PM2.5 and PM10 with high-resolution MODIS aerosol optical depth retrievals and for individual cities and larger regions in India. We also evaluate the spatial and temporal coherence of the two datasets (PM and AOD) using an empirical orthogonal function (EOF) approach. Initial results for New Delhi indicate that daily PM-AOD temporal correlations for a full year are as high as 0.8 in non-monsoon months but as low as 0.3 during monsoon season, when clouds prevent accurate AOD retrievals. We also find large spatial heterogeneity in both AOD and PM2.5 concentrations, whereas temporal variability is more consistent and is controlled by the monsoon. Our results will contribute to better understanding of the spatial and temporal variability of air pollution in India and will improve the accuracy of future retrievals of satellite-derived PM2.5 products.