Improving satellite-driven PM2.5 models with Moderate Resolution Imaging Spectroradiometer fire counts in the southeastern U.S.

• Type of Material: Article
• Authors: Xuefei Hu, Emory University; Lance Waller, Emory University; Alexei Lyapustin, NASA Goddard Space Flight Center; Yujie Wang, NASA Goddard Space Flight Center; Yang Liu, Emory University
• Language: English
• Date: 2014-10-16
• Publisher: American Geophysical Union (AGU)
• Publication Version: Final Published Version
• Copyright Statement: © 2014. American Geophysical Union. All Rights Reserved.
• Final Published Version (URL): http://dx.doi.org/10.1002/2014JD021920
• Title of Journal or Parent Work: Journal of Geophysical Research: Atmospheres
• ISSN: 2169-897X
• Volume: 119
• Issue: 19
• Start Page: 11375
• End Page: 11386
• Grant/Funding Information: In addition, this publication was made possible by USEPA grant R834799.; This work was partially supported by NASA Applied Sciences Program (grants NNX09AT52G and NNX11AI53G, PI: Liu).
• Keywords: GROUND-LEVEL PM2.5; Meteorology & Atmospheric Sciences; Science & Technology; US; MODIS; UNITED-STATES; IMPACTS; AIR-QUALITY; RETRIEVALS; Physical Sciences; AEROSOL OPTICAL-THICKNESS
• Research Categories: Health Sciences, Public Health; Biology, Biostatistics; Biology, Bioinformatics

Abstract

Multiple studies have developed surface PM 2.5 (particle size less than 2.5 μm in aerodynamic diameter) prediction models using satellite-derived aerosol optical depth as the primary predictor and meteorological and land use variables as secondary variables. To our knowledge, satellite-retrieved fire information has not been used for PM 2.5 concentration prediction in statistical models. Fire data could be a useful predictor since fires are significant contributors of PM 2.5 . In this paper, we examined whether remotely sensed fire count data could improve PM 2.5 prediction accuracy in the southeastern U.S. in a spatial statistical model setting. A sensitivity analysis showed that when the radius of the buffer zone centered at each PM 2.5 monitoring site reached 75 km, fire count data generally have the greatest predictive power of PM 2.5 across the models considered. Cross validation (CV) generated an R 2 of 0.69, a mean prediction error of 2.75 μg/m 3 , and root-mean-square prediction errors (RMSPEs) of 4.29 μg/m 3 , indicating a good fit between the dependent and predictor variables. A comparison showed that the prediction accuracy was improved more substantially from the nonfire model to the fire model at sites with higher fire counts. With increasing fire counts, CV RMSPE decreased by values up to 1.5 μg/m 3 , exhibiting a maximum improvement of 13.4% in prediction accuracy. Fire count data were shown to have better performance in southern Georgia and in the spring season due to higher fire occurrence. Our findings indicate that fire count data provide a measurable improvement in PM 2.5 concentration estimation, especially in areas and seasons prone to fire events.