Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
8862698 | Atmospheric Pollution Research | 2017 | 8 Pages |
Abstract
Air pollution has become more serious in many developing countries. Heavy particulate matter (PM) air pollution is a major threat to people's respiratory and cardiopulmonary health. It is an important problem for public health research to accurately estimate the spatial distribution of high PM concentrations from a limited number of monitoring stations. In this study, a maximum entropy (MaxEnt) model was adopted to obtain the probability distribution map of high PM10 concentrations. Daily PM10 concentration data from 19 air quality monitoring stations from the years 2008-2011 were collected. Land cover, road density, and meteorological data were selected as explanatory variables entered in the model. A receiver operating characteristic (ROC) analysis was used to evaluate the performance of the MaxEnt model. The area under the ROC curve (AUC) shows that the MaxEnt model fits well in the four year period. AUC is 0.78 in 2008, 0.79 in 2009, 0.81 in 2010, and 0.80 in 2011. A probability distribution map of high PM10 concentration indicates high human health risks in regions of Beijing in areas with dense roads and buildings. During the entire research period from 2008 to 2011, the distribution of high PM10 concentration is relatively stable and it indicates that the trend of high concentration has not changed significantly during the four years. Traffic and land cover are the two most important factors that can explain more than 80% variance of PM10 from 2008 to 2011.
Related Topics
Physical Sciences and Engineering
Earth and Planetary Sciences
Atmospheric Science
Authors
Zhi-Hong Zhang, Mao-Gui Hu, Jing Ren, Zi-Yin Zhang, George Christakos, Jin-Feng Wang,