Assessing PM2.5 Exposures with High Spatiotemporal Resolution across the Continental United States

被引:419
作者
Di, Qian [1 ]
Kloog, Itai [1 ,4 ]
Koutrakis, Petros [1 ]
Lyapustin, Alexei [2 ]
Wang, Yujie [3 ]
Schwartz, Joel [1 ]
机构
[1] Harvard Univ, Dept Environm Hlth, TH Chan Sch Publ Heath, Boston, MA 02115 USA
[2] NASA, Goddard Space Flight Ctr, Code 613, Greenbelt, MD 20771 USA
[3] Univ Maryland, Baltimore, MD 21250 USA
[4] Ben Gurion Univ Negev, Dept Geog & Environm Dev, POB 653, IL-84105 Beer Sheva, Israel
关键词
AEROSOL OPTICAL DEPTH; LAND-USE REGRESSION; PARTICULATE AIR-POLLUTION; LONG-TERM EXPOSURE; ISOPRENE EMISSION; CARDIOVASCULAR-DISEASE; ULTRAFINE PARTICLES; HOSPITAL ADMISSIONS; MODIS; MORTALITY;
D O I
10.1021/acs.est.5b06121
中图分类号
X [环境科学、安全科学];
学科分类号
083001 [环境科学];
摘要
A number of models have been developed to estimate PM2.5 exposure, including satellite-based aerosol optical depth (AOD) models, land-use regression, or chemical transport model simulation, all with both strengths and weaknesses. Variables like normalized difference vegetation index (NDVI), surface reflectance, absorbing aerosol index, and meteoroidal fields are also informative about PM2.5 concentrations. Our objective is to establish a hybrid model which incorporates multiple approaches and input variables to improve model performance. To account for complex atmospheric mechanisms, we used a neural network for its capacity to model nonlinearity and interactions. We used convolutional layers, which aggregate neighboring information, into a neural network to account for spatial and temporal autocorrelation. We trained the neural network for the continental United States from 2000 to 2012 and tested it with left out monitors. Ten-fold cross-validation revealed a good model performance with a total R-2 of 0.84 on the left out monitors. Regional R-2 could be even higher for the Eastern and Central United States. Model performance was still good at low PM2.5 concentrations. Then, we used the trained neural network to make daily predictions of PM2.5 at 1 km x 1 km grid cells. This model allows epidemiologists to access PM2.5 exposure in both the short-term and the long-term.
引用
收藏
页码:4712 / 4721
页数:10
相关论文
共 80 条
[1]
Estimation of outdoor NOx, NO2, and BTEX exposure in a cohort of pregnant women using land use regression modeling [J].
Aguilera, Inmaculada ;
Sunyer, Jordi ;
Fernandez-Patier, Rosalia ;
Hoek, Gerard ;
Aguirre-Alfaro, Amelia ;
Meliefste, Kees ;
Bomboi-Mingarro, M. Teresa ;
Nieuwenhuijsen, Mark J. ;
Herce-Garraleta, Dolores ;
Brunekreef, Bert .
ENVIRONMENTAL SCIENCE & TECHNOLOGY, 2008, 42 (03) :815-821
[2]
Improving national air quality forecasts with satellite aerosol observations [J].
Al-Saadi, J ;
Szykman, J ;
Pierce, RB ;
Kittaka, C ;
Neil, D ;
Chu, DA ;
Remer, L ;
Gumley, L ;
Prins, E ;
Weinstock, L ;
MacDonald, C ;
Wayland, R ;
Dimmick, F ;
Fishman, J .
BULLETIN OF THE AMERICAN METEOROLOGICAL SOCIETY, 2005, 86 (09) :1249-+
[3]
[Anonymous], 1995, HDB BRAIN THEORY NEU
[4]
[Anonymous], 2015, NASA EOSDIS LAND PRO
[5]
[Anonymous], 2004, Neural Networks, DOI DOI 10.5555/541500
[6]
[Anonymous], 2012, OMI AURA MULTIWAVELE, DOI [10.5067/Aura/OMI/DATA3004, DOI 10.5067/AURA/OMI/DATA3004]
[7]
[Anonymous], 2009, ENVIRON HEALTH PERSP, DOI DOI 10.1289/ehp.0800108
[8]
[Anonymous], SPIE REMOTE SENSING
[9]
[Anonymous], ATMOS ENV IN PRESS
[10]
Application of the deletion/substitution/addition algorithm to selecting land use regression models for interpolating air pollution measurements in California [J].
Beckerman, Bernardo S. ;
Jerrett, Michael ;
Martin, Randall V. ;
van Donkelaar, Aaron ;
Ross, Zev ;
Burnett, Richard T. .
ATMOSPHERIC ENVIRONMENT, 2013, 77 :172-177