Spatiotemporal variability in long-term population exposure to PM2.5 and lung cancer mortality attributable to PM2.5 across the Yangtze River Delta (YRD) region over 2010–2016: A multistage approach

Hong Wang,Jiawen Li,Meng Gao,Ta-Chien Chan,Zhiqiu Gao,Manyu Zhang,Yubin Li,Yefu Gu,Aibo Chen,Yuanjian Yang,Hung Chak Ho
DOI: https://doi.org/10.1016/j.chemosphere.2020.127153
IF: 8.8
2020-10-01
Chemosphere
Abstract:<p>The Yangtze River Delta region (YRD) is one of the most densely populated regions in the world, and is frequently influenced by fine particulate matter (PM<sub>2.5</sub>). Specifically, lung cancer mortality has been recognized as a major health burden associated with PM<sub>2.5</sub>. Therefore, this study developed a multistage approach 1) to first create dasymetric population data with moderate resolution (1 km) by using a random forest algorithm, brightness reflectance of nighttime light (NTL) images, a digital elevation model (DEM), and a MODIS-derived normalized difference vegetation index (NDVI), and 2) to apply the improved population dataset with a MODIS-derived PM<sub>2.5</sub> dataset to estimate the association between spatiotemporal variability of long-term population exposure to PM<sub>2.5</sub> and lung cancer mortality attributable to PM<sub>2.5</sub> across YRD during 2010–2016 for microscale planning. The created dasymetric population data derived from a coarse census unit (administrative unit) were fairly matched with census data at a fine spatial scale (street block), with R<sup>2</sup> and RMSE of 0.64 and 27,874.5 persons, respectively. Furthermore, a significant urban-rural difference of population exposure was found. Additionally, population exposure in Shanghai was 2.9–8 times higher than the other major cities (7-year average: 192,000 μg·people/m<sup>3</sup>·km<sup>2</sup>). More importantly, the relative risks of lung cancer mortality in high-risk areas were 28%–33% higher than in low-risk areas. There were 12,574–14,504 total lung cancer deaths attributable to PM<sub>2.5</sub>, and lung cancer deaths in each square kilometer of urban areas were 7–13 times higher than for rural areas. These results indicate that moderate-resolution information can help us understand the spatiotemporal variability of population exposure and related health risk in a high-density environment.</p>
environmental sciences
What problem does this paper attempt to address?