EN中文

Journal of Geographical Sciences >

2019 , Vol. 29 >Issue 2: 253 - 270

DOI: https://doi.org/10.1007/s11442-019-1595-0

Research Articles

Spatio-temporal evolution and the influencing factors of PM2.5 in China between 2000 and 2015

  • ZHOU Liang , 1, 2 ,
  • ZHOU Chenghu 2 ,
  • YANG Fan 3 ,
  • CHE Lei 4 ,
  • WANG Bo 5 ,
  • SUN Dongqi , 2, *
Expand
  • 1. Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou 730070, China
  • 2. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
  • 3. School of Geographic and Oceanographic Sciences, Nanjing University, Nanjing 210023, China
  • 4. College of Geography and Environment Sciences, Northwest Normal University, Lanzhou 730070, China
  • 5. Department of Geography, The University of Hong Kong, Hong Kong 999077, China
* Corresponding author: Sun Dongqi, Associate Professor, specialized in urban geography. E-mail:

Author: Zhou Liang, PhD and Associate Professor, specialized in environmental geography, urban geography and regional development. E-mail:

Received date: 2018-05-10

  Accepted date: 2018-10-22

  Online published: 2019-02-25

Supported by

The Strategic Priority Research Program of the Chinese Academy of Sciences, No.XDA19040401

China Postdoctoral Science Foundation, No.2016M600121

National Natural Science Foundation of China, No.41701173, No.41501137

The State Key Laboratory of Resources and Environmental Information System

Copyright

Journal of Geographical Sciences, All Rights Reserved
Fold

Abstract

High concentrations of PM2.5 are universally considered as a main cause for haze formation. Therefore, it is important to identify the spatial heterogeneity and influencing factors of PM2.5 concentrations for regional air quality control and management. In this study, PM2.5 data from 2000 to 2015 was determined from an inversion of NASA atmospheric remote sensing images. Using geo-statistics, geographic detectors, and geo-spatial analysis methods, the spatio-temporal evolution patterns and driving factors of PM2.5 concentration in China were evaluated. The main results are as follows. (1) In general, the average concentration of PM2.5 in China increased quickly and reached its peak value in 2006; subsequently, concentrations remained between 21.84 and 35.08 μg/m3. (2) PM2.5 is strikingly heterogeneous in China, with higher concentrations in the north and east than in the south and west. In particular, areas with relatively high PM2.5 concentrations are primarily in four regions, the Huang-Huai-Hai Plain, Lower Yangtze River Delta Plain, Sichuan Basin, and Taklimakan Desert. Among them, Beijing-Tianjin-Hebei Region has the highest concentration of PM2.5. (3) The center of gravity of PM2.5 has generally moved northeastward, which indicates an increasingly serious haze in eastern China. High-value PM2.5 concentrations have moved eastward, while low-value PM2.5 has moved westward. (4) Spatial autocorrelation analysis indicates a significantly positive spatial correlation. The “High-High” PM2.5 agglomeration areas are distributed in the Huang-Huai-Hai Plain, Fenhe-Weihe River Basin, Sichuan Basin, and Jianghan Plain regions. The “Low-Low” PM2.5 agglomeration areas include Inner Mongolia and Heilongjiang, north of the Great Wall, Qinghai-Tibet Plateau, and Taiwan, Hainan, and Fujian and other southeast coastal cities and islands. (5) Geographic detection analysis indicates that both natural and anthropogenic factors account for spatial variations in PM2.5 concentration. Geographical location, population density, automobile quantity, industrial discharge, and straw burning are the main driving forces of PM2.5 concentration in China.

Key words： air pollution; PM2.5; haze; spatio-temporal evolution; environmental influence; China

Cite this article

ZHOU Liang , ZHOU Chenghu , YANG Fan , CHE Lei , WANG Bo , SUN Dongqi . Spatio-temporal evolution and the influencing factors of PM2.5 in China between 2000 and 2015[J]. Journal of Geographical Sciences, 2019 , 29(2) : 253 -270 . DOI: 10.1007/s11442-019-1595-0

1 Introduction

PM2.5 is the most serious air pollutant in China, which shows typical regional and compound pollution characteristics. Concurrently with rapid urbanization and industrialization in China, energy consumption and car ownership have increased sharply. The airborne dust caused by urbanization, atmospheric pollution caused by coal burning, and emissions from automobile exhaust fumes have become more serious, which have led to a dramatic increase in total suspended particles (TSP) in the air (Hueglina et al., 2005; Stone, 2008; Huang, 2018). Haze has become increasingly frequent year after year, forming four geographic haze areas: the Huang-Huai-Hai Region, Yangtze River Delta, Sichuan Basin, and Pearl River Delta (Guo et al., 2011; Wu, 2012). China’s successive promulgation of a series of environmental protection laws and rules, long-term atmospheric prevention and control measures, and industrial and energy structural adjustment and upgrading during “the 11th Five-Year Plan” in 2006-2010 and “the 12th Five-Year Plan” in 2011-2015 have resulted in a distinct reduction in SO2, SOx, and dust. However, it is widely recognized that PM2.5 (particulate matter with an aerodynamic diameter no greater than 2.5 μm) has become a striking challenge for China’s atmospheric pollution prevention and control. PM2.5 is closely related to human activities inside the atmospheric boundary layer. PM2.5 lowers visibility and participates in chemical reactions in the atmosphere to generate new pollutants and severely affects human health. Medical studies have shown that PM2.5 enters the human respiratory system, which can result in various respiratory and cardiovascular diseases, an attenuation of lung function, destruction of the human immune system, and possible increases in risk of death in the exposed population (Dockery et al., 1994; Pope et al., 1995; Laden et al., 2000; Pope, 2000; Samet et al., 2000; Delfino et al., 2005; Laden et al., 2006; Franklin et al., 2008; Kioumourtzoglou et al., 2016). Long-term exposure to air pollution is responsible for the premature death of more than 1,250,000 persons annually, accounting for approximately 40% of such deaths in China (Wang et al., 2012). In 2013, the extreme concentration of PM2.5 in Beijing exceeded 1000 μg/m3, more than 40 times the health standard set by the World Health Organization (Cheng et al., 2011). The negative impact to residents’ physical and mental health in areas covered by smog cannot be estimated. In 2012, China promulgated a new Environmental Air Quality Standard (GB3095-2012) that lists PM2.5 as a regular key monitoring index. National monitoring of air quality conditions and the scope of pollution has continuously increased and the number of PM2.5 concentration monitoring points increased from 612 in 2013 to 1436 in 2016. Clearly, PM2.5 will be a key point in the prevention and control of air pollution in China in the future, and an important topic in international atmospheric environmental research.
At present, estimates of PM2.5 spatial concentrations and characteristics are generally determined from the following data and methods (Chu et al., 2015). Remote sensing images retrieval for aerosol optical depth (AOD) can estimate PM2.5 concentration, while analysis techniques include real-time data space interpolation of monitoring points, weighted regression models and mixed models. Based on these data sources and models, researchers have investigated the origin, genetic mechanism, spatial heterogeneity, transboundary transmission, health effects and coping mechanisms of PM2.5. For example, scholars have constructed a spatio-temporal distribution and derived relationships between PM10 and PM2.5 using a geographically temporally weighted regression model and cluster analysis (Yang et al., 2016; represented the spatial distribution of PM2.5 using satellite retrieval for remote measurement of aerosol optical depth (AOD) (Liu et al., 2005; Xue et al., 2015); established a list of PM2.5 discharge sources (Cao et al., 2011; Zhang and Cao, 2015) using an inventory-chemical mass balance model (Zhang et al., 2015), chemical mass balance method (Gramsch et al., 2006), and atmospheric diffusion model method (Austin et al., 2013); and revealed the spatial heterogeneity and cross-region transmission of PM2.5 concentration using a linear regression model and Comprehensive Air Quality Model Extensions (CAMx) air quality model for aerosol optical thickness (AOT) data (Wang et al., 2003; Xue et al., 2014). Additional studies have found that PM2.5 concentrations show distinctive seasonal changes, but also clear geospatial heterogeneity and spatial dependence (Chow et al., 2006; Gelencsér et al., 2007; Liu et al., 2009; Kloog et al., 2012; Beckerman et al., 2013; Lin et al., 2013). Furthermore, an economic growth mode with high energy consumption and a non-ecological urbanization mode are the main factors creating high PM2.5 in China and similar developing countries. Scholars have found that landform, meteorology, dust, transportation, biomass and coal burning were key factors affecting the spatial distribution of PM2.5 pollution. The large-scale spatio-temporal distribution of PM2.5 is mainly affected by global climatic change, landform and topography, population density, land utilization, economy, and traffic intensity (Charron and Harrison, 2005; Henderson et al., 2007; Merbitz et al., 2012; Gao et al., 2015), whereas small-scale spatio-temporal changes in PM2.5 are controlled by the distances from monitoring points to pollution sources (e.g., urban centers, bus stations, airports, factories) (Hoek et al., 2002).
In summary, PM2.5 studies have mainly focused on source analysis, pollution characteristics, and health evaluation, while largely neglecting the spatio-temporal evolution of PM2.5 and its driving forces. Furthermore, existing studies have focused on seasonal and spatial changes in PM2.5 concentration from case studies of international metropolises or pollution-sensitive cities, such as Los Angeles, London, and Beijing; and covered relatively short time spans (Bell et al., 2007). A comprehensive analysis of the spatio-temporal distribution characteristics, influencing factors, and driving forces of PM2.5 concentration based on large scales and long time frames has not been undertaken for several reasons. (1) It is difficult to obtain large-scale and long-term PM2.5 pollution data, and continuous monitoring data may not be available. For China, the National Atmospheric Environment Monitoring System has only included PM2.5 concentration in the monitoring index system since 2012, and no public national PM2.5 monitoring data is available prior to 2012. (2) In European and American countries with complete PM2.5 monitoring systems, PM2.5 pollution does not occur nationwide, but only appears in polluted island cities, such as Los Angeles, London and adjacent areas; large-scale, continuous pollution areas have not formed. From 2000 to 2015, China was in a period of rapid urbanization and industrialization. Based on average PM2.5 concentration data from 1999 to 2016 provided by the United States National Aeronautics and Space Administration (NASA), it is possible to analyze the spatial-temporal distribution and influencing factors of PM2.5 concentration in continuous polluted areas of China. These data should provide an accurate, macroscopic, and long time series, reflecting changes in PM2.5 pollution in China during this period. To a certain extent, this dataset also addressed the problem of missing macroscopic PM2.5 concentration monitoring data during the Chinese development from 2000 to 2012. Our findings can help forming policies to adjust energy structure, guide industrial layouts, and avoid risk of pollution in the next 10-20 years. Furthermore, such a dataset can be used to resolve PM2.5 cross-regional pollution problems, which take the administrative unit as the main body although there are isolated air pollution prevention model, and provide spatial decision references for national cross-regional pollution linkage governance.

2 Data and methods

2.1 Data sources

The research data were obtained from three sources. (1) PM2.5 concentrations from remote sensing retrieval. This research adopts the raster data of global atmospheric PM2.5 concentration from 1998 to 2016 published by NASA as basic research data (website) with a resolution of 0.1° (http://earthdata.nasa.gov). Because the aerosol optical depth (AOD) product from satellite remote sensing retrieval has advantages of low cost, wide spatial coverage and high simulation accuracy, and is an important index of ground PM2.5 concentration, it has been widely applied to remote monitoring of near-surface PM2.5. The high correlation between AOD determined by the MODIS/Terra AOD product and PM2.5 concentration has been verified by a variety of studies. The original data are three-year average values. To calculate data reliability and stability, this study adopted intermediate year substitution; for example, the average PM2.5 concentration from 2014 to 2016 is taken as the average PM2.5 concentration of 2015. (2) Basic geographic information data and spatial administrative boundaries were derived from 1:4 million Chinese basic geographic information data provided by the National Basic Geographic Information Center. Taking the full territory of the People’s Republic of China (including Mainland of China, Hong Kong, Macao and Taiwan) as the research area, the grid data was then extracted using the research area vector boundary as a mask. The average PM2.5 concentration was calculated for each year in each county-level administrative unit, which established the spatio-temporal database for PM2.5 concentration in China based on the county-level administrative region boundaries. (3) Socioeconomic data, car ownership, population density, and straw burning were obtained from the China City Statistical Yearbook, China Urban Construction Statistical Yearbook, China Region Statistical Yearbook, and China Rural Statistical Yearbook in each corresponding year from 2000 to 2016; some missing data were supplemented from data of corresponding provinces (autonomous regions) and municipalities.

2.2 Methodology

2.2.1 Gravity model
Tobler’s first law of geography considers that geographical things or properties are mutually related in terms of spatial distributions, i.e., clustering, random, and regularity. Specifically, neighbouring objects and properties are more related than distant objects, termed spatial autocorrelation (Wang et al., 2015). To thoroughly analyze the spatio-temporal distribution patterns and characteristics of PM2.5, this study first introduces the concept of center of gravity and a calculation method to reveal the PM2.5 spatial migration process. This gravity model is used as a representation of changing spatial cluster characteristics of PM2.5, and Moran’s I index is used as a representation of the spatial agglomeration characteristics of PM2.5. The coordinates X and Y of the center of gravity of PM2.5 pollution are calculated as:
$\overline{X}=\frac{\sum\limits_{i=1}^{n}{Wi\times Si\times Xi}}{\sum\limits_{i=1}^{n}{Wi\times Si}}$ and $\overline{Y}=\frac{\sum\limits_{i=1}^{n}{Wi\times Si\times Yi}}{\sum\limits_{i=1}^{n}{Wi\times Si}}$ (1)
where $\bar{X}$is the longitude of PM2.5 pollution gravity. $\bar{Y}$is the latitude of PM2.5 pollution gravity. n indicates the raster quantity and i indicates the raster number. Xi and Yi indicate the longitude and latitude of the geometric center of raster i, respectively. Si indicates the area of raster i, and Wi indicates the annual average PM2.5 concentration of raster i.
2.2.2 Spatial autocorrelation
This study applies global Moran’s index (Global Moran’s I) to test the global spatial autocorrelation of PM2.5 concentration. If the Global Moran’s I index is greater than 0, the research object has a positive spatial autocorrelation, and a larger value indicates a stronger spatial agglomeration of the observed PM2.5 value. When the Global Moran’s I index is less than 0, the PM2.5 concentration presents a negative spatial autocorrelation, and a smaller value indicates a stronger spatial dispersion of the observed value. The Global Moran’s I index is calculated as:
$I=\frac{\sum\limits_{i=1}^{n}{\sum\limits_{j=1}^{n}{Wij\left( xi-\bar{x} \right)}}\left( xj-\bar{x} \right)}{{{S}^{2}}\sum\limits_{i=1}^{n}{\sum\limits_{j=1}^{n}{Wij}}}$ (2)
$S=\frac{1}{n}\sum\limits_{i=1}^{n}{{{\left( xi-\bar{x} \right)}^{2}}}$ (3)
where n indicates the quantity of spatial units (county-level administrative units are adopted in this study). Xi and Xj are the annual average PM2.5 concentration of units i and j, respectively, and $\bar{X}$ indicates the average value of all units. Wij indicates the spatial weight matrix of units i and j; if there is a common edge between spatial units i and j, then Wij = 1, otherwise, Wij = 0. To test the significance of the Global Moran’s I index, the standardized normalization value of the Global Moran’s I index, Z(I), is defined as follows:
$Z\left( I \right)=\frac{\left[ I-E(I) \right]}{\sqrt{Var(I)}}$ (4)
where E(I) indicates the mathematical expectation of the Global Moran’s I index, and Var(I) indicates the variance in the Global Moran’s I index.
The Local Moran’s I index is applied to extract the local spatial autocorrelation of atmospheric PM2.5 pollution and identify spatial agglomeration and heterogeneity. For the spatial unit i, the Local Moran’s I index is defined as:
$Ii=\frac{\sum\limits_{i=1}^{n}{\sum\limits_{j=1}^{n}{Wij\left( xi-\bar{x} \right)}}\left( xj-\bar{x} \right)}{{{S}^{2}}}$ (5)
where Z(I) still indicates the significance level of the Local Global Moran’s I index, calculated from equation (4). Z(I) values are compared at different levels by dividing spatial units into four types of spatial autocorrelation based on threshold (p=0.05 in this study). When I is significantly positive and Z(I) > 0, it is termed a “high-high” type and indicates that the PM2.5 concentration of this unit and adjacent units are relatively high, i.e., a hot spot. When I is significantly positive and Z(I) < 0, it is termed a “low-low” type and indicates that the PM2.5 concentration of this unit and adjacent units are relatively low, i.e., a cold spot. When I is significantly negative and Z(I) > 0, it is termed a “high-low” type and indicated that high concentration of PM2.5 units are surrounded by low adjacent units. When I is significantly negative and Z(I) < 0, it is termed a “low-high” type and indicates that low concentration of PM2.5 units are surrounded by high adjacent units. A significantly positive I indicates that there is significantly local spatial positive autocorrelation and spatial clustering. A significantly negative I indicates that there is significantly local negative spatial autocorrelation and spatial dispersion.
2.2.3 Geographic detector
Geographic detection is a set of statistical methods to detect the spatial heterogeneity and reveal driving forces; it is an important method for detecting the causes and mechanism of spatial patterns of geographical factors and has been applied to the studies of disease risk detection, socio-economy, and eco-environment (Liu et al., 2012; Wang et al., 2017; Lindner, 2018). The factor detector in the model is used to identify a specific geographical factor and quantify the spatial distribution difference by comparing the total variance of this index in different types of areas and in the whole area. The model is defined as:
$PD,H=1-\frac{1}{n\sigma _{H}^{2}}\sum\limits_{i=1}^{n}{nD,i}\sigma \frac{2}{{{H}_{D,j}}}$ (6)
where PD,H indicates the explanatory power of the influencing factor of PM2.5; D indicates the factors influencing annual average PM2.5 concentration; n and σ2 indicate the overall sample quantity and variance of the research area, respectively; m indicates the number of categories for the factors; and nD,i indicates the number of D indices on category-i samples. PD,H ranges from 0 to 1, and a larger value indicates a stronger explanatory power of this factor for a change in PM2.5 concentration. When the value is 0, the classification factor is completely unrelated to the change in PM2.5 concentration. When the value is 1, the classification factor can completely explain the change in the spatial distribution of PM2.5 concentration.

3 The spatio-temporal evolution characteristics of PM2.5 pollution

3.1 Time series characteristics of PM2.5

Based on a comprehensive time analysis and Spearman rank correlation coefficient analysis, the annual average PM2.5 concentration in China rose steadily from 2000 to 2015, a trend that was not identified in the monitoring data (Figure 1). The annual average PM2.5 concentration increased from 21.84 μg/m3 in 2000 to 35.08 μg/m3 in 2006, with an average annual increase of 1.66 μg/m3; this was a significant rising trend in this stage (p=0.05). The PM2.5 concentration has generally fluctuated around 32.57 μg/m3 since 2006, and the significant upward trend leading to 2006 has stabilized to a certain degree. This indicates that 2006 was a turning point in PM2.5 annual average concentration change in China. This result is consistent with results published by China’s Ministry of Environmental Protection in 2007. The national ecological civilization construction, ecological supplement points, pollution survey and the achievement of environmental protection policies, the adjustment of industrial structure and improvements in energy efficiency have restrained the PM2.5 emissions to a certain extent. However, the annual average PM2.5 concentration in China still remained high. The raster of annual PM2.5 concentration in 2015 increased by 39.76% compared to 2000. Furthermore, 13.20% of rasters had annual average concentrations that approximately doubled, indicating that PM2.5 pollution continues to expand across the national land space.
Figure 1 Overall PM2.5 concentration trend in China from 2000 to 2015
Based on the annual average concentration limit of PM2.5 in the Ambient Air Quality Standard (GB3095-2012) of China (Samet et al., 2000), the annual average PM2.5 concentra-tion was divided into seven intervals and the area proportion of each interval was analyzed for the study period (Figure 2). Four important results were obtained. (1) The proportion of annual average PM2.5 concentration lower than 15 μg/m3 (the first-order concentration limit) decreased continuously from 28.95% in 2000 to 91.21% in 2015. (2) The proportion of an-nual average PM2.5 concentration higher than 35 μg/m3 (the second-order concentration limit) increased from 17.78% in 2000 to 32.89% in 2015. (3) The proportion of high-pollution areas with annual average PM2.5 concentration higher than 70 μg/m3 increased greatly from 0.04% in 2000 to 4.06% in 2015, and the increase was more than 100 times larger. (4) The high-pollution areas with annual average PM2.5 concentration higher than 100 μg/m3 emerged in large numbers from 2006 to 2007 and from 2010 to 2015, and the proportion in other years was lower than 0.50%. These results indicate that low-pollution areas with annual average PM2.5 concentration lower than 15 μg/m3 decreased continuously, while high-pollution areas with annual average PM2.5 concentration higher than 70 μg/m3 increased. High-pollution areas and extremely high-pollution areas showed a rapid expansion in the national space.
Figure 2 Variations in PM2.5 concentration by range in China from 2000 to 2015

3.2 Characteristics of spatio-temporal patterns in China

3.2.1 Spatial variations
Identifying the PM2.5 spatial evolution characteristics and distribution and exploring changes in PM2.5 concentrations in the national space have importance implications for controlling cross-regional linkage pollution in China. The primary spatial characteristics of PM2.5 in China for 2000-2015 were obtained by analyzing the PM2.5 grid data. In areas with annual average PM2.5 concentration higher than 100 μg/m3, the pollution was extremely serious (Figure 3), which only appeared in counties, such as Xinxiang and Yanjin in Henan Province, in 2000. However, serious pollution began to extend into and cover parts of northern Henan and southern Hebei provinces in 2008 and was distributed sporadically in the Fenhe-Weihe Basin and Sichuan Basin. High-pollution areas with annual average PM2.5 concentration higher than 70 μg/m3 were distributed continuously in the North China Plain, Fenhe-Weihe Basin and Sichuan Basin. The high-pollution area in the North China Plain was the largest and spread across the midstream and downstream plains of the Yangtze River. In addition to the densely populated and economically developed areas, high-pollution areas were distributed sporadically in the Tarim Basin. With the exceptions of Heilongjiang, Yunnan, Fujian, Taiwan, and Hainan, the annual average PM2.5 concentration in areas east of the Heihe-Tengchong Line generally exceeded 35 μg/m3; the annual average PM2.5 concentration in the most densely populated areas in China did not meet the secondary standard of the
Figure 3 PM2.5 spatial characteristics in China in specific years from 2000 to 2015
Ambient Air Quality Standard. Finally, in the areas west of the Heihe-Tengchong Line, with the exception of the Tarim Basin, which is subject to the influences of airborne dust from the Taklimakan Desert in spring, the annual average PM2.5 concentration is generally lower than 35 μg/m3. Areas with relatively serious PM2.5 pollution are mainly concentrated in North China and the Yangtze River Basin area. The PM2.5 pollution had close relationships with geographically low-lying plains and population density and economic activity.
3.2.2 Spatio-temporal evolution
To describe the spatial characteristics of PM2.5 concentration change and pollution, the county-level administrative division was used as the basic unit and grids were used to calculate detailed statistics for 2853 counties for four stages, 2000-2004, 2004-2008, 2008-2012 and 2012-2015. Changes in concentrations, i.e., rise (R) and decline (D), for the four stages were used to divide the variation in PM2.5 concentrations into 16 types of time sequences for evaluation (Table 1 and Figure 4). During the research period, the county-level units with continuous increases in PM2.5 accounted for 7.15% of the total units and were mainly distributed in Tibet and Northeast China, where PM2.5 pollution was not as serious. These regions had relatively good eco-environment, but the air quality deteriorated continuously, which is worthy of national attention. From 2000 to 2008, the county-level research units with continuous increases in degree of PM2.5 pollution accounted for 82.55% of all units, indicating PM2.5 pollution aggravation in most regions of China. This has been characterized as one of the fastest industrialization and urbanization periods in China. The distributions of time sequence trends showed clear spatial agglomeration; for example, D-R-D-R was concentrated in the Shaanxi, Gansu and Ningxia regions, whereas D-D-R-R was generally distributed in the Southern Xinjiang region. Only four counties showed continuously falling PM2.5, which were distributed in Jiuquan, Gansu Province.
Table 1 Distribution chart of 16 types of time sequences of PM2.5 concentration evolution in China
No. Change type Quantity % No. Change type Quantity %
1 D—D—D—D 4 0.17 9 R—D—D—D 38 1.59
2 D—D—D—R 5 0.21 10 R—D—D—R 71 2.97
3 D—D—R—D 11 0.46 11 R—D—R—D 120 5.02
4 D—D—R—R 1 0.04 12 R—D—R—R 68 2.85
5 D—R—D—D 45 1.88 13 R—R—D—D 811 33.93
6 D—R—D—R 13 0.54 14 R—R—D—R 801 33.51
7 D—R—R—D 36 1.51 15 R—R—R—D 190 7.95
8 D—R—R—R 5 0.21 16 R—R—R—R 171 7.15

Note: R indicates rise, and D indicates decline.

Figure 4 The evolution of PM2.5 concentrations for the 16 time sequences in China (see Table 1 for time sequence definitions)
3.2.3 Movement of center of gravity
The center of gravity movement for PM2.5 in China from 2000 to 2015 calculated from equation (5) showed significant trends. During the study period, the center of gravity was located at the juncture of Shaanxi and Henan provinces (Figure 5). However, the center of gravity moved southward at an average of 24.97 km annually from 2000 to 2004, showing a clear increase in PM2.5. The movement slowed after 2005 and the average annual movement distance decreased to 15.04 km. In 2001, the center of gravity for high concentration PM2.5 moved rapidly from southwest to northeast and was located in Shaanxi Province. It subsequently moved quickly to Henan Province with an average annual distance of 156.09 km from 2001 to 2003. After 2004, it returned to Shaanxi Province, but the movement distance rapidly dropped to 40.71 km after 2004, showing a relatively steady state. In contrast, the center of gravity for low concentration PM2.5 moved westward rapidly. The center of gravity was located in Inner Mongolia for a long time and showed a common direction with the high concentration center of gravity. The average annual movement distance before 2004 was 86.37 km, which decreased to 37.84 km from 2005 to 2015. Generally, the center of gravity for the overall and high concentration PM2.5 showed movement to the northeast, while the center of gravity for low concentration PM2.5 moved westward. All three centers of gravity moved rapidly before 2004 and tended to stabilize after 2005. This change is closely related to the enhancement of national ecological civilization construction, industrial transfer (eastern to central and western regions), and higher thresholds for environmental protection in eastern China. These results are in agreement with those presented in section 3.1, indicating that the PM2.5 pollution in the eastern region was higher than that in the western region and higher in the northern region than in the southern.
Figure 5 Changes in PM2.5 concentration center of gravity in China from 2000 to 2015

3.3 Spatial autocorrelation analysis

3.3.1 Global spatial autocorrelation
The Global Moran’s I index of the annual average PM2.5 concentration in each county-level unit from 2000 to 2015 was positive and passed the significance test (p=0.05). Therefore, the annual average PM2.5 concentration presented a significant positive spatial autocorrelation and showed an agglomerated spatial pattern. From the time sequence, the Global Moran’s I index reached a maximum in 2006 and then generally decreased year after year. This turning point is consistent with the turning point in annual average concentration, indicating that the spatial agglomeration of the annual average PM2.5 concentration first increased and then decreased, and the degree of spatial agglomeration reached its peak in 2006.
3.3.2 Local spatial autocorrelation
Based on local spatial autocorrelation analysis results (Figure 6), the county-level units presenting significant local spatial autocorrelation were divided into four types: high-high, low-low, high-low, and low-high. The high-high areas with high annual average PM2.5 concentration, i.e., the hot spots for PM2.5 pollution, were continuously distributed in the North China Plain, Fenhe-Weihe Basin, Sichuan Basin, and Jianghan Plain regions. The number of county-level administrative units contained in the hot spots was generally unchanged from 2000 to 2004, but increased greatly from 2006 to 2009, and then decreased in 2008. The low-low areas with low annual average PM2.5 concentration, i.e., the cold spots for PM2.5 pollution, were distributed continuously in the northern part of Northeast China, Taiwan, Yunnan, Qinghai-Tibet Plateau, Xinjiang, the northern part of Inner Mongolia, and other border areas. The number of county-level administrative units contained in the cold spots was similar to the number of hot spots in terms of change over time; generally, unchanged before the distinctive increase in 2012 and then dropped to a minimum in 2015. The other county-level administrative units did not present significant local spatial autocorrelation. During the research period, no county-level administrative units of high-low or low-high local spatial autocorrelation types were identified, indicating that the annual average PM2.5 concentration had strong local positive spatial autocorrelation characteristics.
Figure 6 Local spatial autocorrelation analysis results for PM2.5 in China from 2000 to 2015

4 Driving forces of PM2.5 pollution

Spatial variations in PM2.5 concentration in China were significant. However, the origins are complicated and the factors affecting changes in pollution concentrations are diverse and include natural factors, such as atmospheric circulation, volcano ash, forest fires, flying dust in bare desert areas, wind direction and frequency, and rainfall. Anthropogenic factors include industrial flue dust discharge, coal combustion, straw burning, exhaust from motor vehicles, flying dust from construction site. To analyze the factors contributing to changes in PM2.5 concentrations more comprehensively, we selected panel data for 287 prefecture-level cities in 2000, 2006, and 2011, including a total of 27,839 samples. We adopted the geographical detector method to detect characteristics of spatial differentiation and identify the driving forces. We chose the following 11 index factors as contributing to high PM2.5 concentrations in the data sources described in the methods section: natural geographical regionalization (X1), per capita GDP (X2), population density (X3), proportion of secondary industry (X4), proportion of built-up areas (X5), urban greening ratio (X6), urban residents’ car ownership (X7), sown area (X8), industrial flue dust discharge (X9), average energy consumption intensity (X10), average iron and steel output of lands (X11). We then calculated their degrees of influence on the spatial distributions of PM2.5 in 287 prefecture-level Chinese cities (Table 2).
Table 2 Geographical detection results for PM2.5 in China for 2000, 2006, and 2011
Detection indices 2000 2006 2011
P Q P Q P Q
Natural geographical regionalization (X1) 0.7047 0.0000 0.7447 0.0000 0.7196 0.0000
Per capita GDP (X2) 0.0077 0.9191 0.0062 0.9079 0.0068 0.8659
Population density (X3) 0.4320 0.0000 0.4372 0.0000 0.4120 0.0000
Proportion of the secondary industry (X4) 0.0984 0.0000 0.0665 0.0031 0.0917 0.0000
Proportion of built-up areas (X5) 0.0853 0.0030 0.0753 0.1033 0.1025 0.0282
Urban greening ratio (X6) 0.0280 0.1503 0.0625 0.0319 0.0359 0.1083
Urban residents’ car ownership (X7) 0.0259 0.8637 0.0913 0.0226 0.1074 0.0080
Sown area (X8) 0.1396 0.0557 0.1487 0.0000 0.1046 0.0000
Industrial flue dust discharge (X9) 0.0709 0.1537 0.0936 0.0000 0.0531 0.2766
Energy consumption intensity of lands (X10) 0.3109 0.0000 0.4124 0.0000 0.4143 0.0000
Average iron and steel output of lands (X11) 0.2869 0.0000 0.3373 0.0000 0.3217 0.0000

4.1 Natural factors affecting PM2.5 spatio-temporal evolution

The geographical detection results indicate that changes in PM2.5 concentrations are closely related to regional natural factors. Natural geographical regionalization (X1) has the most significant influence on PM2.5. In the three selected years (2000, 2006, and 2011), its detection and explanatory power P for PM2.5 were 0.7047, 0.7477, and 0.7196, respectively, indicating large-scale regional differentiation and significant influence from landform, climate, hydrology, soil and vegetation on the formation mechanism of PM2.5. Among the four areas suffering serious PM2.5 pollution, also recognized in section 3.2.1 in this paper, the change of PM2.5 in Taklimakan Desert was closely related to the regional atmospheric circulation and local sand-dust weather. Observations from 88 monitoring stations in Xinjiang from 2000 to 2011 show that the Taklimakan Desert and its southern edge were sand-dust weather frequently observed area with sand-dust days 2.7 times more in Southern Xinjiang than in the Northern (Jiang et al., 2013); therefore, PM2.5 concentrations showed spatial coupling characteristics that were higher in Southern Xinjiang than in the Northern. The overall increase in vegetation coverage and change in rainfall in Xinjiang also affected the change in PM2.5 in the Taklimakan Desert to a certain degree. Moreover, influenced by ocean currents, the southeast monsoon, warm and humid climate, high precipitation, dense population, and industrial agglomeration, the Pearl River Delta and island regions (e.g., coastal Fujian, Taiwan and Hainan) showed relatively low annual average PM2.5 concentration and were not severely polluted. Similarly, regions such as the Qinghai-Tibet Plateau, Inner Mongolia Plateau, and Yunnan-Guizhou Plateau with rugged landforms had few human activities; thus, the PM2.5 concentrations have remained relatively low.

4.2 Socio-economic factors affecting the spatio-temporal evolution in PM2.5

Increases in human activity has a strong impact on air pollution, including flying dust arising from urbanization, emissions due to more private cars, increases in energy consumption, coal-fired heating facilities due to sharp rises in population, and straw burning caused by agricultural production. Excluding Xinjiang, which is strongly affected by natural factors, the three regions with high PM2.5 concentration were generally consistent with the spatial distribution of population density in China. The geographical detection results indicate that the regions suffering serious PM2.5 pollution were mainly concentrated in densely populated regions, including the North China Plain, with Beijing-Tianjin-Hebei region as the center; Shandong Peninsula; Hunan-Hubei Plain; and Chengdu-Chongqing Basin. A quantitative analysis of each factor influencing PM2.5 concentration is as follows.
(1) Driven by urban construction. Based on geographical detection, the detected p values for urban development and construction influence on PM2.5 concentration in 2000, 2006, and 2011 were 0.4320, 0.4372, and 0.4120, respectively. Population density contributed the most to the changes in PM2.5 concentration, characterized by an inverted U-shaped pattern. The detected p values for the proportion of built-up areas (X5) in 2000, 2006, and 2011 were 0.0853, 0.0753, and 0.1025 respectively, and their contributions to the changes in regional PM2.5 concentration increased gradually with the acceleration in urban construction. The detected p values for the urban greening ratio (X6) in 2000, 2006, and 2011 were 0.0280, 0.0625, and 0.0359, respectively, and their contributions to changes in regional PM2.5 concentration were also characterized by an inverted U-shaped pattern. Urban greening can mitigate dust and relieve its effect on PM2.5 in the atmosphere to a certain degree; however, the influence was minor. Similarly, per capita GDP (X2) had a relatively small influence.
(2) Driven by industry and energy (coal) consumption. In 2000, 2006 and 2011, the detected p values for the proportion of secondary industry (X4) were respectively 0.0984, 0.0665, and 0.0917; those of the average energy consumption intensity of lands (X10) were respectively 0.3109, 0.4124, and 0.4143, and those of the average iron and steel output of lands (X11) were respectively 0.2869, 0.3373, and 0.3217. Both industry and coal consumption were significant factors in PM2.5 pollution (significance level = 0.01), indicating that emission of particulate pollutants in cities is the dominant source of pollution to most cities and regions. From 2000 to 2011, the total energy consumption in China increased by 2.39, and the proportion of coal to total energy consumption only decreased from 69.21% in 2000 to 68.42% in 2011; that is, coal remained the primary source of energy in China. In 2015, coal still accounted for 68% of all energy consumed with no substantial decline. Total coal consumption in high polluted areas, such as Shandong, Inner Mongolia, Shanxi, Hebei, Henan, and Jiangsu exceeded 200 million tons. From 2000 to 2011, the iron and steel industry consumed a huge quantity of coal resources in China, primarily in three regions, the Yangtze River Delta, North China Plain, and Sichuan Basin; these areas are iron and steel industry concentrated areas. In these regions, ten provinces and cities, including Hebei, Tianjin, Shandong, Shaanxi, Jiangsu, Liaoning, Shanghai, and Sichuan, host the majority of iron and steel industries in China. In 2011, the crude steel output of these ten provinces and cities was 4945 million tons, accounting for 70.44% of the national output in that year. Therefore, coal consumption and spatial distribution of iron and steel industries are key factors affecting changes in PM2.5 concentration.
(3) Driven by the exhaust emissions of motor vehicles. In 2000, 2006 and 2011, the detected p values for civil car ownership (X7) were 0.0984, 0.0665, and 0.0917, respectively, and car ownership had insignificant effects on PM2.5 concentration in 2000 and 2006. However, with the sharp increase in the number of cars after 2006, the effect of car ownership became significant in 2011. From 2000 to 2011, civil car ownership in China increased by 5.82 times and private car ownership increased by 11.72 times. The growth in car ownership was far higher than economic growth (4.77 times), per capita income (4.48 times), and highway mileage (2.45 times). In provinces and cities suffering serious haze pollution, e.g., Beijing, Hebei, Jiangsu, Zhejiang, and Shandong, the total numbers of civilian vehicles all exceed 4,000,000 per region.
(4) Driven by straw burning. In 2000, 2006 and 2011, the detected p values for the influence of sown area of farm crops (X8) on PM2.5 were 0.1396, 0.1487, and 0.1046, respectively. These were also characterized by an inverted U-shaped change pattern and the influence of straw burning on regional PM2.5 concentration reached a maximum in 2006. The total agricultural straw in China was 6×108 t, and 18.59% of straw was burnt in open space in rural areas. The pollution arising from straw burning was less than 5% of total PM2.5. However, the locations were agglomerated primarily in Eastern and Northern China regions with developed agriculture, and the time of burning was concentrated, with the peak occurring around October for 1-2 days. Therefore, PM2.5 arising from straw burning possibly accounted for 30% or 40% of the total PM2.5 in the air on specific days and had significant influence on regional air pollution (Lu et al., 2011).

5 Conclusions and implications

5.1 Conclusions

(1) Time sequence and Spearman rank correlation coefficient analysis shows that the annual average PM2.5 concentration in China was high throughout the study period. From 2000 to 2015, the PM2.5 concentration in China first increased rapidly and then became stable; 2006 was identified as a turning point for the overall change in PM2.5 concentration in China. However, the annual average PM2.5 concentration in China was high and presented a trend of obvious diffusion in national land space. In 2011, the proportion of annual PM2.5 concentration raster to total raster increased by 75.12% compared to 2000, and the raster with growth rates showing a doubling of annual average concentration accounted for 13.20%.
(2) Spatial analysis shows that northern and eastern China had higher concentrations than southern and western China; the Heihe-Tengchong Line is the significant dividing line. PM2.5 pollution concentration areas and extremely polluted areas showed clear expansion. The areas with average concentrations less than 15 μg/m3 decreased while areas with average concentrations more than 70 μg/m3 increased. Areas with high PM2.5 pollution concentrations were distributed in the North China Plain, and were related to low-altitudes. County-level administrative analyses indicate that PM2.5 pollution rose in 76.23% county-level administrative units from 2000 to 2006. The increase of PM2.5 pollution during this period was the general trend in most regions.
(3) The center of gravity for PM2.5 pollution in China is located in eastern Shaanxi; from 2000 to 2004, the center of gravity moved eastward at 24.97 km annually. After 2004, the center of gravity for high-pollution-concentration areas moved eastward, while the low-pollution center of gravity moved westward. The movement in opposite directions indicates that PM2.5 pollution in the eastern regions clearly increased in this period and eastern PM2.5 pollution increased more than western. Spatial autocorrelation analysis shows that the annual average PM2.5 concentration was strongly characterized by local spatial positive autocorrelation. The “high-high” PM2.5 agglomeration areas were distributed continuously in the North China Plain, Fenhe-Weihe Basin, Sichuan Basin and Jianghan Plain regions; “low-low” PM2.5 agglomeration areas were distributed in northern Inner Mongolia and Heilongjiang; the southeast coastal and island regions, e.g., Taiwan, Hainan, and Fujian; and the border areas, e.g., the Qinghai-Tibet Plateau and northern Xinjiang.
(4) An evaluation of the driving forces of PM2.5 pollution in China indicates that natural factors and human economic activities have both affected the spatial characteristics and concentrations of PM2.5. The influence of natural geographical division on PM2.5 is most significant, and the detected p values for the three stages of the study period were higher than 0.7240. In addition, atmospheric condition, population growth (population density), industrial emission, straw burning, energy consumption growth, increases in motor vehicle ownership increased, and increases in car exhaust over the short term are the main driving forces for changes in PM2.5 concentration in China.

5.2 Implications

Quantitatively identifying the spatial variation and regularity of PM2.5 concentrations and evaluating the driving forces and mechanism of changes in PM2.5 concentration are keys to developing the economy while protecting the environment. These efforts can relieve residents’ psychological fear that “haze is to be even more dreaded than tigers,” while also providing a scientific basis for regional atmospheric linkage prevention and control, a spatial configuration for heavy-polluting industries, designing urban landscapes (wind pipe, greenway), and adjusting industrial and energy structures. Due to data limitations, studying the large-scale evolution in PM2.5 concentrations began late in China. Existing studies have been primarily based on data from environmental monitoring locations and involved analyses of quarterly and daily changes using one-year segment data. This study used average PM2.5 concentration data from 1999 to 2016 provided by NASA to address the problems created by limitations in large-scale PM2.5 data from 2000 to 2015 in China, i.e., the limited monitoring stations in the central and western regions and limited research on regional information distortion. The spatial distribution patterns in PM2.5 concentrations presented here are consistent with previous studies (Zhang, 2015; Wang, 2015) and agree with the spatial characteristics of monitoring points in the sparse areas, although for a longer period. However, the factors affecting PM2.5 concentration in China are complicated. Natural factors include atmospheric circulation, extreme weather, landform, and regional transfer. Anthropogenic factors include industrial pollution, coal burning, motor vehicle emission, dust, biomass burning, car exhaust and waste burning, which are the most important driving forces. Due to the large gap between industrial structure, energy structure and consumption structure in various regions or cities, China is in a key period of economic transformation described as “adjusting structure, stabilizing growth and development in green.” Predicting the complexity of atmospheric pollutants to provide long-term administration includes key scientific issues, such as “reasons and control of atmospheric haze” and “the relationship between haze and health” that the country must address and are also the focus of future research and exploration.

The authors have declared that no competing interests exist.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Austin E, Coull B A, Zanobetti A et al., 2013. A framework to spatially cluster air pollution monitoring sites in US based on the PM2.5 composition.Environment International, 59(3): 244-254.Background: Heterogeneity in the response to PM2.5 is hypothesized to be related to differences in particle composition across monitoring sites which reflect differences in source types-as well as climatic and topographic conditions impacting different geographic locations. Identifying spatial patterns in particle composition is a multivariate problem that requires novel methodologies.Objectives: Use cluster analysis methods to identify spatial patterns in PM2.5 composition. Verify that the resulting clusters are distinct and informative.Methods: 109 monitoring sites with 75% reported speciation data during the period 2003-2008 were selected. These sites were categorized based on their average PM2.5 composition over the study period using k-means cluster analysis. The obtained clusters were validated and characterized based on their physico-chemical characteristics, geographic locations, emissions profiles, population density and proximity to major emission sources.Results: Overall 31 clusters were identified. These include 21 clusters with 2 or more sites which were further grouped into 4 main types using hierarchical clustering. The resulting groupings are chemically meaningful and represent broad differences in emissions. The remaining clusters, encompassing single sites, were characterized based on their particle composition and geographic location.Conclusions: The framework presented here provides a novel tool which can be used to identify and further classify sites based on their PM2.5 composition. The solution presented is fairly robust and yielded groupings that were meaningful in the context of air-pollution research. (C) 2013 Elsevier Ltd. All rights reserved.

DOI PMID

[2]
Beckerman B S, Jerrett M, Serre M et al., 2013. A hybrid approach to estimating national scale spatiotemporal variability of PM2.5 in the contiguous United States.Environmental Science & Technology, 47(13): 7233-7241.Airborne fine particulate matter exhibits spatiotemporal variability at multiple scales, which presents challenges to estimating exposures for health effects assessment. Here we created a model to predict ambient particulate matter less than 2.5 mu m in aerodynamic diameter (PM2.5) across the contiguous United States to be applied to health effects modeling. We developed a hybrid approach combining a land use regression model (LUR) selected with a machine learning method, and Bayesian Maximum Entropy (BME) interpolation of the LUR space-time residuals. The PM2.5 data set included 104,172 monthly observations at 1464 monitoring locations with approximately 10% of locations reserved for cross validation LUR models were based on remote sensing estimates of PM2.5, land use and traffic indicators. Normalized cross validated R-2 values for LUR were 0.63 and 0.11 with and without remote sensing, respectively, suggesting remote sensing is a strong predictor of ground level concentrations In the models including the BME interpolation of the residuals, cross validated R-2 were 0.79 for both configurations; the model without remotely sensed data described more line scale variation than the model including remote sensing. Our results suggest that our modeling framework can predict ground level concentrations of PM2.5 at multiple scales over the contiguous U.S.

DOI PMID

[3]
Bell M L, Dominici F, Ebisu K et al., 2007. Spatial and temporal variation in PM2.5 chemical composition in the United States for health effects studies.Environmental Health Perspectives, 115(7): 989-995.Although numerous studies have demonstrated links between particulate matter (PM) and adverse health effects, the chemical components of the PM mixture that cause injury are unknown. This work characterizes spatial and temporal variability of PM2.5(PM with aerodynamic diameter < 2.5 μm) components in the United States; our objective is to identify components for assessment in epidemiologic studies. We constructed a database of 52 PM2.5component concentrations for 187 U.S. counties for 2000–2005. First, we describe the challenges inherent to analysis of a national PM2.5chemical composition database. Second, we identify components that contribute substantially to and/or co-vary with PM2.5total mass. Third, we characterize the seasonal and regional variability of targeted components. Strong seasonal and geographic variations in PM2.5chemical composition are identified. Only seven of the 52 components contributed ≥ 1% to total mass for yearly or seasonal averages [ammonium (NH4+), elemental carbon (EC), organic carbon matter (OCM), nitrate (NO361), silicon, sodium (Na+), and sulfate (SO4261)]. Strongest correlations with PM2.5total mass were with NH4+(yearly), OCM (especially winter), NO361(winter), and SO4261(yearly, spring, autumn, and summer), with particularly strong correlations for NH4+and SO4261in summer. Components that co-varied with PM2.5total mass, based on daily detrended data, were NH4+, SO4261,OCM, NO3261, bromine, and EC. The subset of identified PM2.5components should be investigated further to determine whether their daily variation is associated with daily variation of health indicators, and whether their seasonal and regional patterns can explain the seasonal and regional heterogeneity in PM10(PM with aerodynamic diameter < 10 μm) and PM2.5health risks.

DOI PMID

[4]
Cao G L, Zhang X Y, Gong S L et al., 2011. Emission inventories of primary particles and pollutant gases for China.Atmospheric Environment, 45(37): 6802-6811. (in Chinese)This paper analyzes TOMS AOD at 500 nm (1980-2001), along with MODIS data (2000-2008) at 550 nm to investigate variations at one-degree grid over eight typical regions in China and the trends in AODs, temporally and spatially. In contrast to recently reported global decrease in AOD over global ocean beginning around 1990, we find there virtually exists no apparent AOD transition in China for that: firstly no notable upward tendencies in AOD during 1980-1992 for the relative low value (+0.001/decade), then during 1996-2001 a discernible ascending tendency with larger magnitude at 0.01/decade, and finally, since 2000, a weak upward trend with +0.004/decade. The large increases during 1996-2001 are presumably consequences of large increases in industrial activities and bear a strong resemblance to the long-term decreasing observations of incident solar radiation and cloud cover in China. Specifically, in late 1990's, only in Taklimakan Desert a negative trend with a maximum magnitude of -0.04/decade is detected. However, over regions such as Jingjinji and Pearl River Delta influenced by industrial activities, positive tendencies at +0.01/decade are observed.Seasonal patterns in the AOD regional long-term trend are evident. AODs exhibit generally similar seasonality and the summer dominates higher AOD value than the autumn. In particular, during the period 1980-2001, all the eight regions except Taklimakan Desert witness the maximum aerosols in winter while there is not such seasonality during the period 2000-2008. Geographically, we also document spatial patterns of AOD variations over China. Results reveal that no apparent upward trends in AOD (about 15% per decade) are observed in 1980's, while beginning 1990 till 2008, both data (TOMS and MODIS) are indicative of a significant AOD increase across China, especially in 1990's it is indeed the case, roughly in accordance with the overall trends at regional scale. (C) 2011 Elsevier Ltd. All rights reserved.

DOI

[5]
Charron A, Harrison R M, 2005. Fine (PM2.5) and coarse (PM2.5-10) particulate matter on a heavily trafficked London highway:  Sources and processes.Environmental Science & Technology, 39(20): 7768-7776.

[6]
Cheng S, Yang L X, Zhou X et al., 2011. Evaluating PM2.5 ionic components and source apportionment in Jinan, China from 2004 to 2008 using trajectory statistical methods.Journal of Environmental Monitoring, 13(6): 1662-1671.The mass concentrations and major chemical components of PM(2.5) in Jinan, Shandong Province, China from Dec. 2004 to Oct. 2008 were analyzed using backward trajectory cluster analysis in conjunction with the potential source contribution function (PSCF) model. The aim of this work was to study the inter-annual variations of mass concentrations and major chemical components of PM(2.5), evaluate the air mass flow patterns and identify the potential local and regional source areas that contributed to secondary sulfate and nitrate in PM(2.5) in Jinan. The annual mean concentrations of PM(2.5), sulfate and nitrate in 2004-2008 were almost the highest in the world. The most significant air parcels contributing to the highest mean concentrations of mass and secondary ions in PM(2.5) originated from the industrialized areas of Shandong Province. Clusters with a lower ratio of NO(3)(-)/SO(4)(2-) in PM(2.5) originated from the Yellow Sea, while a higher ratio was observed in the clusters passing through Beijing and Tianjin. PSCF modeling indicated that the provinces of Shandong, Henan, Jiangsu, Anhui and the Yellow Sea were the major potential source regions for sulfate, in agreement with the cluster analysis results. Regional and long-range transport of NH(4)NO(3) played an important role in the nitrate concentration of Jinan. By comparing the distributions of secondary sulfate and nitrate over three years, enhanced emission control management before and during the 29(th) Olympic Games led to a discernible decrease in source contributions from Beijing and its environs in 2007-2008.

DOI

[7]
Chow J C, Chen L W, Watson J G et al., 2006. PM2.5 chemical composition and spatiotemporal variability during the California regional PM10/PM2.5 air quality study (CRPAQS).Journal of Geophysical Research Atmospheres, 111(D10): 1-17.1] The 14-month-long (December 1999 to February 2001) Central California Regional PM10/PM2.5 Air Quality Study (CRPAQS) consisted of acquiring speciated PM2.5 measurements at 38 sites representing urban, rural, and boundary environments in the San Joaquin Valley air basin. The study's goal was to understand the development of widespread pollution episodes by examining the spatial variability of PM2.5, ammonium nitrate (NH4NO3), and carbonaceous material on annual, seasonal, and episodic timescales. It was found that PM2.5 and NH4NO3 concentrations decrease rapidly as altitude increases, confirming that topography influences the ventilation and transport of pollutants. High PM2.5 levels from November 2000 to January 2001 contributed to 5009000975% of annual average concentrations. Contributions from organic matter differed substantially between urban and rural areas. Winter meteorology and intensive residential wood combustion are likely key factors for the winter-nonwinter and urban-rural contrasts that were observed. Short-duration measurements during the intensive operating periods confirm the role of upper air currents on valley-wide transport of NH4NO3. Zones of representation for PM2.5 varied from 5 to 10 km for the urban Fresno and Bakersfield sites, and increased to 1509000920 km for the boundary and rural sites. Secondary NH4NO3 occurred region-wide during winter, spreading over a much wider geographical zone than carbonaceous aerosol.

DOI

[8]
Chu H J, Huang B, Lin C Y, 2015. Modeling the spatio-temporal heterogeneity in the PM10-PM2.5 relationship.Atmospheric Environment, 102(2): 176-182.61This study explores the spatio-temporal patterns of particulate matter (PM).61Spatial heterogeneity of the PM data is identified using fuzzy clustering.61PM10-PM2.5 relationship is modeled by GWR and GTWR.61GTWR provides spatio-temporal variations of the PM10-PM2.5 relationship.

DOI

[9]
Delfino R J, Sioutas C, Malik S, 2005. Potential role of ultrafine particles in associations between airborne particle mass and cardiovascular health.Environmental Health Perspectives, 113(8): 934-946.Numerous epidemiologic time-series studies have shown generally consistent associations of cardiovascular hospital admissions and mortality with outdoor air pollution, particularly mass concentrations of particulate matter (PM) ≤ 2.5 or ≤ 10 μm in diameter ( PM2.5, PM10). Panel studies with repeated measures have supported the time-series results showing associations between PM and risk of cardiac ischemia and arrhythmias, increased blood pressure, decreased heart rate variability, and increased circulating markers of inflammation and thrombosis. The causal components driving the PM associations remain to be identified. Epidemiologic data using pollutant gases and particle characteristics such as particle number concentration and elemental carbon have provided indirect evidence that products of fossil fuel combustion are important. Ultrafine particles < 0.1 μm (UFPs) dominate particle number concentrations and surface area and are therefore capable of carrying large concentrations of adsorbed or condensed toxic air pollutants. It is likely that redox-active components in UFPs from fossil fuel combustion reach cardiovascular target sites. High UFP exposures may lead to systemic inflammation through oxidative stress responses to reactive oxygen species and thereby promote the progression of atherosclerosis and precipitate acute cardiovascular responses ranging from increased blood pressure to myocardial infarction. The next steps in epidemiologic research are to identify more clearly the putative PM casual components and size fractions linked to their sources. To advance this, we discuss in a companion article (Sioutas C, Delfino RJ, Singh M. 2005. Environ Health Perspect 113:947-955) the need for and methods of UFP exposure assessment.

DOI PMID

[10]
Dockery D W, Pope CA, Xu X et al., 1994. An association between air pollution and mortality in six US cities.New England Journal of Medicine, 329(24): 1753-1759.

[11]
Franklin M, Koutrakis P, Schwartz P, 2008. The role of particle composition on the association between PM2.5 and mortality.Epidemiology, 19(5): 680-689.Background: Although the association between exposure to particulate matter (PM) mass and mortality is well established, there remains uncertainty about which chemical components of PM are most harmful to human health. Methods: A hierarchical approach was used to determine how the association between daily PM mass and mortality was modified by PM composition in 25 US communities. First, the association between daily PM and mortality was determined for each community and season using Poisson regression. Second, we used meta-regression to examine how the pooled association was modified by community and season-specific particle composition. Results: There was a 0.74% (95% confidence interval = 0.41%–1.07%) increase in nonaccidental deaths associated with a 10 μg/m06 increase in 2-day averaged PM mass concentration. This association was smaller in the west (0.51% [0.10%–0.92%]) than in the east (0.92% [0.23%–1.36%]), and was highest in spring (1.88% [0.23%–1.36%]). It was increased when PM mass contained a higher proportion of aluminum (interquartile range = 0.58%), arsenic (0.55%), sulfate (0.51%), silicon (0.41%), and nickel (0.37%). The combination of aluminum, sulfate, and nickel also modified the effect. These species proportions explained residual variability between the community-specific PM mass effect estimates. Conclusions: This study shows that certain chemical species modify the association between PM and mortality and illustrates that mass alone is not a sufficient metric when evaluating health effects of PM exposure.

DOI PMID

[12]
Gao M, Cao J, Seto E. A, 2015. A distributed network of low-cost continuous reading sensors to measure spatiotemporal variations of PM2.5 in Xi’an, China.Environmental Pollution, 199(4): 56-65.61A $15 portable PM sensor demonstrated high correlations with reference monitors.61The sensor can be deployed in high PM2.5 urban environments.61The sensor can improve spatiotemporal resolution of data from existing monitoring networks.

DOI PMID

[13]
Gelencsér A, May B, Simpson D et al., 2007. Source apportionment of PM2.5 organic aerosol over Europe: Primary/secondary, natural/anthropogenic, and fossil/biogenic origin.Journal of Geophysical Research Atmospheres, 112(D23): 1-12.1] On the basis of a 2-year comprehensive data set obtained within the CARBOSOL project, seasonal source apportionment of PM2.5 aerosol is attempted for five rural/remote sites in Europe. The approach developed combines radiocarbon measurements with bulk measurements of organic carbon (OC), elemental carbon (EC), and two organic tracers (levoglucosan and cellulose). Source types are lumped into primary emissions from fossil fuel combustion and biomass burning, bioaerosol, and secondary organic aerosol from precursors emitted by fossil and nonfossil sources. Bulk concentration ratios reported for these source types in the literature are used to estimate the source contributions which are constrained by measured radiocarbon concentrations. It has been found that while fossil-related sources predominate EC throughout the year at all sites, the sources of OC are primarily biogenic and markedly different between summer and winter. In winter biomass burning primary emission is the main source, with sizable additional contribution from fossil fuel combustion. In contrast, in summer secondary organic aerosol (SOA) from nonfossil sources becomes predominant (63 76% of TC), with some contribution of SOA from fossil fuel combustion. The results agree well with recent findings of other authors who established the predominance of biogenic SOA for rural sites in summer in Europe. An uncertainty analysis has been conducted, which shows that the main conclusions from this study are robust.

DOI

[14]
Gramsch E, Cereceda-Balic F, Oyola P et al., 2006. Examination of pollution trends in Santiago de Chile with cluster analysis of PM10 and ozone data.Atmospheric Environment, 40(28): 5464-5475.Because of the high levels of pollution that Santiago de Chile experiences every year in winter, the government has set up an air quality monitoring network. Information from this network is employed to alert people about the quality of air and to enforce several control strategies in order to limit pollution levels. The monitoring network has 8 stations that measure PM10, carbon monoxide (CO), sulphur dioxide (SO2) ozone (O-3) and meteorological parameters. Some stations also measure nitrogen mono- and dioxide (NOx), fine particles (PM2.5) and carbon. In this study we have examined the PM10 and O-3 data generated by this network in the year 2000 in order to determine the seasonal trends and spatial distribution of these pollutants over a year's period. The results show that concentration levels vary with the season, with PM10 being higher in winter and 03 in summer. All but one station, show a peak in PM10 at 8:00 indicating that during the rush hour there is a strong influence from traffic, however, this influence is not seen during the rest of the day. In winter, the PM10 maximum occurs at 24:00 h in all stations but Las Condes. This maximum is related to decreased wind speed and lower altitude of the inversion layer. The fact that Las Condes station is at a higher altitude than the others and it does not show the PM10 increase at night, suggest that the height of the inversion layer occurs at lower altitude. Cluster analysis was applied to the PM10 and O-3 data, and the results indicate that the city has four large sectors with similar pollution behavior. The fact that both pollutants have similar distribution is a strong indication that the concentration levels are primarily determined by the topographical and meteorological characteristics of the area and that pollution generated over the city is redistributed in four large areas that have similar meteorological and topographical conditions. (c) 2006 Published by Elsevier Ltd.

DOI

[15]
Guo J P, Zhang X Y, Wu Y R et al., 2011. Spatio-temporal variation trends of satellite-based aerosol optical depth in China during 1980-2008. Atmospheric Environment, 45(37): 6802-6811.This paper analyzes TOMS AOD at 500 nm (1980-2001), along with MODIS data (2000-2008) at 550 nm to investigate variations at one-degree grid over eight typical regions in China and the trends in AODs, temporally and spatially. In contrast to recently reported global decrease in AOD over global ocean beginning around 1990, we find there virtually exists no apparent AOD transition in China for that: firstly no notable upward tendencies in AOD during 1980-1992 for the relative low value (+0.001/decade), then during 1996-2001 a discernible ascending tendency with larger magnitude at 0.01/decade, and finally, since 2000, a weak upward trend with +0.004/decade. The large increases during 1996-2001 are presumably consequences of large increases in industrial activities and bear a strong resemblance to the long-term decreasing observations of incident solar radiation and cloud cover in China. Specifically, in late 1990's, only in Taklimakan Desert a negative trend with a maximum magnitude of -0.04/decade is detected. However, over regions such as Jingjinji and Pearl River Delta influenced by industrial activities, positive tendencies at +0.01/decade are observed.Seasonal patterns in the AOD regional long-term trend are evident. AODs exhibit generally similar seasonality and the summer dominates higher AOD value than the autumn. In particular, during the period 1980-2001, all the eight regions except Taklimakan Desert witness the maximum aerosols in winter while there is not such seasonality during the period 2000-2008. Geographically, we also document spatial patterns of AOD variations over China. Results reveal that no apparent upward trends in AOD (about 15% per decade) are observed in 1980's, while beginning 1990 till 2008, both data (TOMS and MODIS) are indicative of a significant AOD increase across China, especially in 1990's it is indeed the case, roughly in accordance with the overall trends at regional scale. (C) 2011 Elsevier Ltd. All rights reserved.

DOI

[16]
Henderson S B, Beckerman B, Jerrett M et al., 2007. Application of land use regression to estimate long-term concentrations of traffic-related nitrogen oxides and fine particulate matter.Environmental Science & Technology, 41(7): 2422-2428.

[17]
Hoek G, Brunekreef B, Goldbohm S et al., 2002. Association between mortality and indicators of traffic-related air pollution in the Netherlands: A cohort study.The Lancet, 360(9341): 1203-1209.Long-term exposure to participate matter air pollution has been associated with increased cardiopulmonary mortality in the USA. We aimed to assess the relation between traffic-related air pollution and mortality in participants of the Netherlands Cohort study on Diet and Cancer (NLCS), an ongoing study. We investigated a random sample of 5000 people from the full cohort of the NLCS study (age 55–69 years) from 1986 to 1994. Long-term exposure to traffic-related air pollutants (black smoke and nitrogen dioxide) was estimated for the 1986 home address. Exposure was characterised with the measured regional and urban background concentration and an indicator variable for living near major roads. The association between exposure to air pollution and (cause specific) mortality was assessed with Cox's proportional hazards models, with adjustment for potential confounders. 489 (11%) of 4492 people with data died during the follow-up period. Cardiopulmonary mortality was associated with living near a major road (relative risk 1·95, 95% Cl 1·09–3·52) and, less consistently, with the estimated ambient background concentration (1·34, 0·68–2·64). The relative risk for living near a major road was 1·41 (0·94–2·12) for total deaths. Non-cardiopulmonary, non-lung cancer deaths were unrelated to air pollution (1·03, 0·54–1·96 for living near a major road). Long-term exposure to traffic-related air pollution may shorten life expectancy.

DOI PMID

[18]
Huang, Y, Yan Q, Zhang C, 2018. Spatial-temporal distribution characteristics of PM2.5 in China in 2016,Journal of Geovisualization and Spatial Analysis, 2(2): 1-12.Conventional analysis of transportation demand is usually carried out using socioeconomic, travel, and land use attributes. Despite the effectiveness on travel demand forecasting, it is important to...

DOI

[19]
Hueglin C, Gehrig R, Baltensperger U et al., 2005. Chemical characterisation of PM2.5, PM10 and coarse particles at urban, near-city and rural sites in Switzerland.Atmospheric Environment, 39(4): 637-651.Daily PM2.5 and PM10 samples were taken from April 1998 to March 1999 at urban kerbside, urban background, near-city, and rural sites in Switzerland. The samples were analysed for mass, water soluble ions, trace elements, as well as elemental and organic carbon. The present paper focuses on the variation of element concentration between different site types and on the chemical mass closure of atmospheric particulate matter.Information on emission sources of trace elements is obtained by evaluation of the element abundances at sites that represent different pollution levels. The abundances of Ba, Ca, Ce, Cu, Fe, La, Mo, Mn, Pb, Sb, and Rh are gradually decreasing from urban kerbside to urban background, near-city and rural sites, indicating that road traffic is a main source of these elements. On the other hand, the abundances of Al, As, Cd, K, and V are similar for the different site types, which implies that emission sources are either spatially uniformly distributed (e.g. mineral dust), or there are no important regional emission sources and the ambient concentration of these elements might be dominated by long-range transport.When performing a mass closure, the annual average of the sum of aerosol chemical components was 22 27% for PM10 and 8 15% for PM2.5 lower than the PM mass. A drying procedure applied to a subset of PM10 samples and model calculations for PM2.5 samples according to their inorganic composition were used to estimate the contribution of retained water to the unaccounted mass at 50% RH. The obtained average water content was 10.6% for PM10 and 13 23% for PM2.5, clearly indicating that water is a major contributor to the unaccounted mass. Furthermore, a pronounced seasonal variation was observed with relatively lower water content in the colder season, indicating that the inorganic salts were mainly crystalline in winter, whereas they were probably dissolved during the rest of the year.

DOI

[20]
Jiang Y A, Chen Y, Zhao Y Z et al., 2013. Analysis on changes of basic climatic elements and extreme events in Xinjiang, China during 1961-2010.Advances in Climate Change Research, 4(1): 20-29.By using the observation data from 89 weather stations in Xinjiang during 1961 2010, this paper analyzed the basic climatic elements including temperature, precipitation, wind speed, sunshine duration, water vapor pressure, and dust storm in the entire Xinjiang and the subareas: North Xinjiang, Tianshan Mountains, and South Xinjiang. The results indicate that from 1961 to 2010 the annual and seasonal mean temperatures in the entire Xinjiang show an increasing trend with the increasing rate rising from south to north. The increasing rate of annual mean minimum temperature is over twice more than that of the annual mean maximum temperature, contributing much to the increase in the annual averages. The magnitude of the decrease rate of low-temperature days is larger than the increase rate of high-temperature days. The increase of warm days and warm nights and the decrease of cold days and cold nights further reveal that the temperature increasing in Xinjiang is higher. In addition, annual and seasonal rainfalls have been increasing. South Xinjiang experiences higher increase in rainfall amounts than North Xinjiang and Tianshan Mountains. Annual rainy days, longest consecutive rainy days, the daily maximum precipitation and extreme precipitation events, annual torrential rain days and amount, annual blizzard days and amount, all show an increasing trend, corresponding to the increasing in annual mean water vapor pressure. This result shows that the humidity has increased with temperature increasing in the past 50 years. The decrease in annual mean wind speed and gale days lessen the impact of dust storm, sandstorm, and floating dust events. The increase in annual rainy days is the cause of the decrease in annual sunshine duration, while the increase in spring sunshine duration corresponds with the decrease in dust weather. Therefore, the increase in precipitation indicators, the decrease in gales and dust weather, and the increasing in sunshine duration in spring will be beneficial to crops growth. Jiang, Y.-A., Y. Chen, Y.-Z. Zhao, et al., 2013: Analysis on changes of basic climatic elements and extreme events in Xinjiang, China during 1961 2010. Adv. Clim. Change Res.,4(1), doi: 10.3724/SP.J.1248.2013.020.

DOI

[21]
Kioumourtzoglou M A, Schwartz J, Weisskopf M et al., 2016. Long-term PM2.5 exposure and neurological hospital admissions in the Northeastern United States.Environmental Health Perspectives, 124(1): 23-29.Long-term exposure to fine particles (particulate matter ≤ 2.5 μm; PM2.5) has been consistently linked to heart and lung disease. Recently, there has been increased interest in examining the effects of air pollution on the nervous system, with evidence showing potentially harmful effects on neurodegeneration. Our objective was to assess the potential impact of long-term PM2.5exposure on event time, defined as time to first admission for dementia, Alzheimer’s (AD), or Parkinson’s (PD) diseases in an elderly population across the northeastern United States. We estimated the effects of PM2.5on first hospital admission for dementia, AD, and PD among all Medicare enrollees ≥ 65 years in 50 northeastern U.S. cities (1999–2010). For each outcome, we first ran a Cox proportional hazards model for each city, adjusting for prior cardiopulmonary-related hospitalizations and year, and stratified by follow-up time, age, sex, and race. We then pooled the city-specific estimates by employing a random effects meta-regression. We followed approximately 9.8 million subjects and observed significant associations of long-term PM2.5city-wide exposure with all three outcomes. Specifically, we estimated a hazard ratio (HR) of 1.08 (95% CI: 1.05, 1.11) for dementia, an HR of 1.15 (95% CI: 1.11, 1.19) for AD, and an HR of 1.08 (95% CI: 1.04, 1.12) for PD admissions per 1-μg/m3increase in annual PM2.5concentrations. To our knowledge, this is the first study to examine the relationship between long-term exposure to PM2.5and time to first hospitalization for common neurodegenerative diseases. We found strong evidence of association for all three outcomes. Our findings provide the basis for further studies, as the implications of such exposures could be crucial to public health. Kioumourtzoglou MA, Schwartz JD, Weisskopf MG, Melly SJ, Wang Y, Dominici F, Zanobetti A. 2016. Long-term PM2.5exposure and neurological hospital admissions in the northeastern United States. Environ Health Perspect 124:23–29;http://dx.doi.org/10.1289/ehp.1408973

DOI PMID

[22]
Kloog I, Nordio F, Coull B et al., 2012. Incorporating local land use regression and satellite aerosol optical depth in a hybrid model of spatiotemporal PM2.5 exposures in the Mid-Atlantic states.Environmental Science & Technology, 46(21): 11913-11921.Satellite-derived aerosol optical depth (AOD) measurements have the potential to provide spatiotemporally resolved predictions of both long and short-term exposures, but previous studies have generally shown moderate predictive power and lacked detailed high spatio- temporal resolution predictions across large domains. We aimed at extending our previous work by validating our model in another region with different geographical and metrological characteristics, and incorporating fine scale land use regression and nonrandom missingness to better predict PM2.5 concentrations for days with or without satellite AOD measures. We start by calibrating AOD data for 2000 2008 across the Mid-Atlantic. We used mixed models regressing PM2.5 measurements against day-specific random intercepts, and fixed and random AOD and temperature slopes. We used inverse probability weighting to account for nonrandom missingness of AOD, nested regions within days to capture spatial variation in the daily calibration, and introduced ...

DOI PMID

[23]
Laden F, Neas L M, Dockery D W et al., 2000. Association of fine particulate matter from different sources with daily mortality in six US cities.Environmental Health Perspectives, 108(10): 941-947.

DOI

[24]
Laden F, Schwartz J, Speizer F E et al., 2006. Reduction in fine particulate air pollution and mortality.American Journal of Respiratory and Critical Care Medicine, 173(6): 667-672.

DOI

[25]
Lindner A, Pitombo C S, 2018. A conjoint approach of spatial statistics and a traditional method for travel mode choice issues.Journal of Geovisualization and Spatial Analysis, 2(1): 1-13.Conventional analysis of transportation demand is usually carried out using socioeconomic, travel, and land use attributes. Despite the effectiveness on travel demand forecasting, it is important to...

DOI

[26]
Lin G, Fu J, Jiang D et al., 2013. Spatio-temporal variation of PM2.5 concentrations and their relationship with geographic and socioeconomic factors in China.International Journal of Environmental Research and Public Health, 11(1): 173-186.中国科学院机构知识库(CAS IR GRID)以发展机构知识能力和知识管理能力为目标，快速实现对本机构知识资产的收集、长期保存、合理传播利用，积极建设对知识内容进行捕获、转化、传播、利用和审计的能力，逐步建设包括知识内容分析、关系分析和能力审计在内的知识服务能力，开展综合知识管理。

DOI PMID

[27]
Liu Y, Paciorek C J, Koutrakis P et al., 2009. Estimating regional spatial and temporal variability of PM2.5 concentrations using satellite data, meteorology, and land use information.Environmental Health Perspectives, 117(6): 886-892.

DOI

[28]
Liu Y, Sarnat JA, Kilaru V et al., 2005. Estimating ground-level PM2.5 in the eastern using satellite remote sensing.Environmental Science & Technology, 39(9): 3269-3278.An empirical model based on the regression between daily PM2.5 (particles with aerodynamic diameters of less than 2.5 m) concentrations and aerosol optical thickness (AOT) measurements from the multiangle imaging spectroradiometer (MISR) was developed and tested using data from the eastern United States during the period of 2001. Overall, the empirical model explained 48% of the variability in PM2.5 concentrations. The root-mean-square error of the model was 6.2 g/m3 with a corresponding average PM2.5 concentration of 13.8 g/m3. When PM2.5 concentrations greater than 40 g/m3 were removed, model results were shown to be unbiased estimators of observations. Several factors, such as planetary boundary layer height, relative humidity, season, and other geographical attributes of monitoring sites, were found to influence the association between PM2.5 and AOT. The findings of this study illustrate the strong potential of satellite remote sensing in regional ambient air quality monitoring as an extension to ground networks. With the continual advancement of remote sensing technology and global data assimilation systems, AOT measurements derived from satellite remote sensors may provide a cost-effective approach as a supplemental source of information for determining ground-level particle concentrations.

DOI PMID

[29]
Liu Y S, Yang R, 2012. The spatial characteristics and formation mechanism of the county urbanization in China.Acta Geographica Sinica, 67(8): 1011-1020. (in Chinese)The spatial and temporal characteristics and the formation mechanism of the county urbanization in China since 1990 were analyzed systematically,using the methods including regional differences,transect and geography detectors. Results show that the temporal and spatial differences of the county urbanization were significant. The "herringbone" shape region pattern of high county urbanization was gradually highlighted,which were made by the counties along the north border and in eastern coastal areas. The county urbanization process of some regions were accelerated and enhanced,including Wuhan metropolitan region,Chengdu-Chongqing region and Guanzhong-Tianshui region. The low county urbanization level was maintained in Southwest China and Qinghai-Tibet Plateau regions. The differences of urbanization and the change rate of county urbanization were converged in China after 2000,but the rate has slowed down since 2000. The county urbanization trend of transects were significantly different,including Lianyungang-Lanzhou railway and Lanzhou-Urumqi railway transects,the Yangtze River transect,the border of north China transect,106 National Road transect,and the eastern coastal transect. There are many factors affecting county urbanization,mainly including economic development stage,the level of secondary and tertiary industries,rural net income per capita,population density,leading position of grain production,demographic statistics and special arrangements for counties. The high county urbanization in northern border regions was a typical type of statistical unrealistically high urbanization. In the future county urbanization development should follow the geographical differences,highlight its leading function,and adopt multiple urbanization development models such as promoting urbanization intensively in key urban economic development areas,separating urbanization in cropland and grain producing areas,migrating urbanization in ecological and water resource protection areas,suburban areas and urban-based urbanization and other leading county urbanization patterns.

DOI

[30]
Lu B, Kong S F, Han Bin, 2011. Inventory of atmospheric pollutants discharged from biomass burning in China continent in 2007.China Environmental Science, 31(2): 186-194. (in Chinese)In the present work,the total amounts of CH4,SO2,NOx,NH3,EC,OC,NMVOC,CO,CO2,TSP,PM10,PM2.5discharged from biomass burning in Chinese continent region were calculated with biomass consumption combined with emission factors.Spatial distribution of the pollutants and mass contribution of each type of biomass combustion were given.In general,in the year 2007,the total emissions of CH4,SO2,NOx,NH3,EC,OC,NMVOC,CO,CO2,TSP,PM10,PM2.5was 3332.7,335.3,951.3,7754.9,783.7,267.7,6049.6,76579.6,743743.7,7677.8,6668.9,4043.7kt,respectively.Sichuan,Anhui,Guangxi,Shandong,Henan and Jiangsu held high emission amounts while in Beijing,Tianjin,Hainan,Ningxia,Qinghai and Tibet,the pollutants amount was low.Straw and firewood burning were main emission sources,which share 93.8% 98.7% of the total emission.The main substances emitting pollutants differed in different regions as well as the emission intensity per unit area and per capita.

DOI

[31]
Merbitz H, Buttstädt M, Michael Set al., 2012. GIS-based identification of spatial variables enhancing heat and poor air quality in urban areas. Applied Geography, 2012, 33(4): 94-106.Due to anthropogenic climate change heat waves are expected to occur more frequently in the future, which might cause adverse health effects for urban population. Especially the combination of high temperatures and poor air quality impinges on the well-being of man. This accentuates the need for assessing the health risks of residents regarding air pollutants and anomalously high summer air temperatures. However, comprehensive information on the spatial and temporal distribution of temperature and particulate matter (PM) concentration in cities are presently difficult to obtain since only few measurement sites exist. In order to identify hot spots with high health risks for distinct groups of urban population, measurement campaigns were carried out, capturing the spatial distribution of temperature and PM concentrations in the City of Aachen, Germany (pop. 245,000). Several locations were selected to examine spatial influences such as topography, building density, vegetation and traffic on temperature and PM. The findings permit the detection of urban environmental variables that contribute to both temperature enhancement and poor air quality. Those variables were used as spatial predictors for the identification of possible hot spots inside and outside the area of field measurements. The zones of enhanced risks of high air temperature and PM levels were detected by means of GIS based geo-statistic modeling. These areas were mainly identified in the inner city, which is characterized by a dense building structure and heavy traffic. A chemical characterization of different PM fractions complements the GIS based investigations. The analysis of toxicologically relevant components provides information on air quality at urban, suburban and rural sites. The results of the chemical analyses support the results obtained from geo-statistical modeling. It reveals high concentrations of health relevant airborne species like metals and polycyclic aromatic hydrocarbons within the zone of enhanced risk for the coincidence of temperature stress and PM pollution.

DOI

[32]
Pope C A, 2000. Review: Epidemiological basis for particulate air pollution health standards.Aerosol Science & Technology, 32(1): 4-14.There are now more than 150 published epidemiologic studies of health effects of particulate air pollution and dozens of related literature reviews. This article explores the basic inferences currently being drawn from the literature regarding the epidemiologic evidence particulate pollution induced health effects. Although there is not a complete consensus of opinion, most reviewers conclude that the overall epidemiologic evidence suggests that particulate air pollution, especially fine combustion-source pollution common to many urban and industrial environments, is an important risk factor for cardiopulmonary disease and mortality. Most of the epidemiological effort has focused on effects of acute exposure, but effects of chronic exposure may be more important in terms of overall public health relevance. Some reviewers contend that long-term, repeated exposure likely increases the risk of chronic respiratory disease and the risk of cardiorespiratory mortality. There is more general (but still not unanimous) agreement that shortterm exposures to particulate pollution can exacerbate existing cardiovascular and pulmonary disease and increase the number of persons in a population who become symptomatic, require medical attention, or die.

DOI

[33]
Pope C A, Burnett R T, Thun M J et al., 2002. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution.Jama, 287(9): 1132-1141.Abstract CONTEXT: Associations have been found between day-to-day particulate air pollution and increased risk of various adverse health outcomes, including cardiopulmonary mortality. However, studies of health effects of long-term particulate air pollution have been less conclusive. OBJECTIVE: To assess the relationship between long-term exposure to fine particulate air pollution and all-cause, lung cancer, and cardiopulmonary mortality. DESIGN, SETTING, AND PARTICIPANTS: Vital status and cause of death data were collected by the American Cancer Society as part of the Cancer Prevention II study, an ongoing prospective mortality study, which enrolled approximately 1.2 million adults in 1982. Participants completed a questionnaire detailing individual risk factor data (age, sex, race, weight, height, smoking history, education, marital status, diet, alcohol consumption, and occupational exposures). The risk factor data for approximately 500 000 adults were linked with air pollution data for metropolitan areas throughout the United States and combined with vital status and cause of death data through December 31, 1998. MAIN OUTCOME MEASURE: All-cause, lung cancer, and cardiopulmonary mortality. RESULTS: Fine particulate and sulfur oxide--related pollution were associated with all-cause, lung cancer, and cardiopulmonary mortality. Each 10-microg/m(3) elevation in fine particulate air pollution was associated with approximately a 4%, 6%, and 8% increased risk of all-cause, cardiopulmonary, and lung cancer mortality, respectively. Measures of coarse particle fraction and total suspended particles were not consistently associated with mortality. CONCLUSION: Long-term exposure to combustion-related fine particulate air pollution is an important environmental risk factor for cardiopulmonary and lung cancer mortality.

DOI PMID

[34]
Pope C A, Dockery D W, Schwartz J, 1995. Review of epidemiological evidence of health effects of particulate air pollution.Inhalation Toxicology, 7(1): 1-18.This article summarizes epidemiological evidence of health effects of particulate air pollution. Acute exposure to elevated levels of particulate air pollution has been associated with increased cardiopulmonary mortality, increased hospitalization for respiratory disease, exacerbation of asthma, increased incidence and duration of respiratory symptoms, declines in lung function, and restricted activity. Small deficits in lung function, higher risk of chronic respiratory disease and symptoms, and increased mortality have also been associated with chronic exposure to respirable particulate air pollution. Health effects have been observed at levels common to many U.S. cites and at levels below current US. National Ambient Air Quality Standards. Although the biological mechanisms involved are poorly understood, recent epidemiological evidence supports the hypothesis that respirable particulate air pollution is an important risk factor for respiratory disease and cardiopulmonary mortality.

DOI

[35]
Samet J M, Dominici F, Curriero F C et al., 2000. Fine particulate air pollution and mortality in 20 U.S cities, 1987-1994. New England Journal of Medicine, 343:(24): 1742-1749.Air pollution in cities has been linked to increased rates of mortality and morbidity in developed and developing countries. Although these findings have helped lead to a tightening of air-quality standards, their validity with respect to public health has been questioned.We assessed the effects of five major outdoor-air pollutants on daily mortality rates in 20 of the largest cities and metropolitan areas in the United States from 1987 to 1994. The pollutants were particulate matter that is less than 10 microm in aerodynamic diameter (PM10), ozone, carbon monoxide, sulfur dioxide, and nitrogen dioxide. We used a two-stage analytic approach that pooled data from multiple locations.After taking into account potential confounding by other pollutants, we found consistent evidence that the level of PM10 is associated with the rate of death from all causes and from cardiovascular and respiratory illnesses. The estimated increase in the relative rate of death from all causes was 0.51 percent (95 percent posterior interval, 0.07 to 0.93 percent) for each increase in the PM10 level of 10 microg per cubic meter. The estimated increase in the relative rate of death from cardiovascular and respiratory causes was 0.68 percent (95 percent posterior interval, 0.20 to 1.16 percent) for each increase in the PM10 level of 10 microg per cubic meter. There was weaker evidence that increases in ozone levels increased the relative rates of death during the summer, when ozone levels are highest, but not during the winter. Levels of the other pollutants were not significantly related to the mortality rate.There is consistent evidence that the levels of fine particulate matter in the air are associated with the risk of death from all causes and from cardiovascular and respiratory illnesses. These findings strengthen the rationale for controlling the levels of respirable particles in outdoor air.

DOI PMID

[36]
Stone B, 2008. Urban sprawl and air quality in large US cities.Journal of Environmental Management, 86(4): 688-698.This study presents the results of a paper of urban spatial structure and exceedances of the 8-h national ambient air quality standard for ozone in 45 large US metropolitan regions. Through the integration of a published index of sprawl with metropolitan level data on annual ozone exceedances, precursor emissions, and regional climate over a 13-year period, the association between the extent of urban decentralization and the average number of ozone exceedances per year, while controlling for precursor emissions and temperature, is measured. The results of this analysis support the hypothesis that large metropolitan regions ranking highly on a quantitative index of sprawl experience a greater number of ozone exceedances than more spatially compact metropolitan regions. Importantly, this relationship was found to hold when controlling for population size, average ozone season temperatures, and regional emissions of nitrogen oxides and volatile organic compounds, suggesting that urban spatial structure may have effects on ozone formation that are independent of its effects on precursor emissions from transportation, industry, and power generation facilities.

DOI PMID

[37]
Wang H, Dwyer-Lindgren L, Lofgren K T et al., 2012. Age specific and sex-specific mortality in 187 countries, 1970-2010: A systematic analysis for the global burden of disease study 2010.The Lancet, 380(9859): 2071-2094.Estimation of the number and rate of deaths by age and sex is a key first stage for calculation of the burden of disease in order to constrain estimates of cause-specific mortality and to measure premature mortality in populations. We aimed to estimate life tables and annual numbers of deaths for 187 countries from 1970 to 2010. We estimated trends in under-5 mortality rate (children aged 0–4 years) and probability of adult death (15–59 years) for each country with all available data. Death registration data were available for more than 100 countries and we corrected for undercount with improved death distribution methods. We applied refined methods to survey data on sibling survival that correct for survivor, zero-sibling, and recall bias. We separately estimated mortality from natural disasters and wars. We generated final estimates of under-5 mortality and adult mortality from the data with Gaussian process regression. We used these results as input parameters in a relational model life table system. We developed a model to extrapolate mortality to 110 years of age. All death rates and numbers have been estimated with 95% uncertainty intervals (95% UIs). From 1970 to 2010, global male life expectancy at birth increased from 56·4 years (95% UI 55·5–57·2) to 67·5 years (66·9–68·1) and global female life expectancy at birth increased from 61·2 years (60·2–62·0) to 73·3 years (72·8–73·8). Life expectancy at birth rose by 3–4 years every decade from 1970, apart from during the 1990s (increase in male life expectancy of 1·4 years and in female life expectancy of 1·6 years). Substantial reductions in mortality occurred in eastern and southern sub-Saharan Africa since 2004, coinciding with increased coverage of antiretroviral therapy and preventive measures against malaria. Sex-specific changes in life expectancy from 1970 to 2010 ranged from gains of 23–29 years in the Maldives and Bhutan to declines of 1–7 years in Belarus, Lesotho, Ukraine, and Zimbabwe. Globally, 52·8 million (95% UI 51·6–54·1 million) deaths occurred in 2010, which is about 13·5% more than occurred in 1990 (46·5 million [45·7–47·4 million]), and 21·9% more than occurred in 1970 (43·3 million [42·2–44·6 million]). Proportionally more deaths in 2010 occurred at age 70 years and older (42·8% in 2010 vs 33·1% in 1990), and 22·9% occurred at 80 years or older. Deaths in children younger than 5 years declined by almost 60% since 1970 (16·4 million [16·1–16·7 million] in 1970 vs 6·8 million [6·6–7·1 million] in 2010), especially at ages 1–59 months (10·8 million [10·4–11·1 million] in 1970 vs 4·0 million [3·8–4·2 million] in 2010). In all regions, including those most affected by HIV/AIDS, we noted increases in mean ages at death. Despite global and regional health crises, global life expectancy has increased continuously and substantially in the past 40 years. Yet substantial heterogeneity exists across age groups, among countries, and over different decades. 179 of 187 countries have had increases in life expectancy after the slowdown in progress in the 1990s. Efforts should be directed to reduce mortality in low-income and middle-income countries. Potential underestimation of achievement of the Millennium Development Goal 4 might result from limitations of demographic data on child mortality for the most recent time period. Improvement of civil registration system worldwide is crucial for better tracking of global mortality. Bill & Melinda Gates Foundation.

DOI PMID

[38]
Wang J, Christopher S A, 2003. Intercomparison between satellite-derived aerosol optical thickness and PM2.5 mass: Implications for air quality studies.Geophysical Research Letters, 30(21): 1-4.ABSTRACT We explore the relationship between column aerosol optical thickness (AOT) derived from the Moderate Resolution Imaging SpectroRadiometer (MODIS) on the Terra/Aqua satellites and hourly fine particulate mass (PM2.5) measured at the surface at seven locations in Jefferson county, Alabama for 2002. Results indicate that there is a good correlation between the satellite-derived AOT and PM2.5 (linear correlation coefficient, R = 0.7) indicating that most of the aerosols are in the well-mixed lower boundary layer during the satellite overpass times. There is excellent agreement between the monthly mean PM2.5 and MODIS AOT (R > 0.9), with maximum values during the summer months due to enhanced photolysis. The PM2.5 has a distinct diurnal signature with maxima in the early morning (6:00 8:00AM) due to increased traffic flow and restricted mixing depths during these hours. Using simple empirical linear relationships derived between the MODIS AOT and 24hr mean PM2.5 we show that the MODIS AOT can be used quantitatively to estimate air quality categories as defined by the U.S. Environmental Protection Agency (EPA) with an accuracy of more than 90% in cloud-free conditions. We discuss the factors that affect the correlation between satellite-derived AOT and PM2.5 mass, and emphasize that more research is needed before applying these methods and results over other areas.

DOI

[39]
Wang J F, Li X H, George Christakos et al., 2010. Geographical detectors-based health risk assessment and its application in the neural tube defects study of the Heshun Region, China.International Journal of Geographical Information Science, 24(1): 107-127.Physical environment, man‐made pollution, nutrition and their mutual interactions can be major causes of human diseases. These disease determinants have distinct spatial distributions across geographical units, so that their adequate study involves the investigation of the associated geographical strata. We propose four geographical detectors based on spatial variation analysis of the geographical strata to assess the environmental risks of health: the risk detector indicates where the risk areas are; the factor detector identifies factors that are responsible for the risk; the ecological detector discloses relative importance between the factors; and the interaction detector reveals whether the risk factors interact or lead to disease independently. In a real‐world study, the primary physical environment (watershed, lithozone and soil) was found to strongly control the neural tube defects (NTD) occurrences in the Heshun region (China). Basic nutrition (food) was found to be more important than man‐made pollution (chemical fertilizer) in the control of the spatial NTD pattern. Ancient materials released from geological faults and subsequently spread along slopes dramatically increase the NTD risk. These findings constitute valuable input to disease intervention strategies in the region of interest.

DOI

[40]
Wang Z B, Fang C L, Xu G et al., 2015. Spatial-temporal characteristics of the PM2.5 in China in 2014.Acta Geographica Sinica, 70(11): 1720-1734. (in Chinese)Haze pollution in China has become a severe environmental problem for people's daily life as well as their health, among which PM2.5makes significant contribution to poor air quality. Satellite observations played a leading role in the recognition in the spatio-temporal variation of PM2.5nationally. However, based on the information and data obtained by satellites,the inversion method has limitations to truly reflect the spatio-temporal variation of PM2.5concentrations near ground level. Based on the observed PM2.5concentration data from 945 newly set-up air monitoring sites in 2014, our research reveals the spatio-temporal variations of PM2.5concentrations in China by using spatial statistical model. The results show that(1) in2014, the average PM2.5concentration in China was 61 g/m3. It had a periodical U-impulse type daily variation as well as a U-shaped monthly variation with a higher level in autumn and winter while a lower one in spring and summer.(2) Concentration of PM2.5in urban China shows a significant spatial differentiation and clustering pattern with spatial-periodic occurrences in north and south China.(3) The Hu-line(Hu Population Line) and Yangtze River are respectively the east-west and north-south boundaries which separate the high-value zone and the low-value zone of PM2.5concentrations in China. In 2014, the highly polluted cities by PM2.5were mainly distributed in the urban agglomerations(Central Henan, Harbin-Changchun,the Bohai Rim Region, the Yangtze River Delta, and the Middle Yangtze River), east of the Huline and north of the Yangtze River. The Beijing-Tianjin-Hebei urban agglomeration was the most severely polluted region all the year round. The southeast coastal region centered on the Pearl River Delta had good air quality in a stable manner.

DOI

[41]
Wu D, 2012. Hazy weather research in China in the last decade: A review.Acta Scientiae Circumstantiae, 32(2): 257-269.Systematic research on haze in China has been on-going for a decade.The rapid economic expansion and urbanization has led to worsening particulate matter(PM) pollution and more frequent poor visibility events.A rapid increase in the number of hazy days has been seen over eastern China.The culprit of hazy weather is fine PM pollution.Hazy weather is often associated with photochemical pollution and composition of aerosols responsible for haze is rather complicated.In recent years,hazy weather,due to its significant environmental impacts and climatic effects,has become a hot topic,attracting wide attention from scientific community,government departments and the public.This paper surveys the literature in Chinese in the recent decade to provide a review on the process of understanding haze and research progress in hazy weather.Future research directions are also discussed.

[42]
Xu W, He F, Li H et al., 2014. Spatial and temporal variations of PM2.5 in the Pearl River Delta.Research of Environmental Sciences, 27(9): 951-957.Fine particulate matter pollution( PM2. 5) in the Pearl River Delta is seriously threatening the ecological safety. In order to determine the temporal and spatial variations of the pollution,the geostatistic method was used to analyze concentration data collected from62 monitoring sites from September 2012 to August 2013. Qualitative analysis showed that: the base effect values ranged from 0. 12-0. 30;there was a less strong or strong spatial auto-correlation in the region; and,the variation of ρ( PM2. 5) was caused by regional structural factors complemented by random factors. Quantitative analysis showed that that the auto-correlation distance ranged from 51 to 85 kilometers along with direction and time influenced by meteorological conditions. The auto-correlation distance in the east-west direction ranged from 75 to 85 kilometers and was wilder than the others. The variation range and speed of ρ( PM2. 5) reached maximums( 0.34-0. 70,0. 14-0. 38 μg( m3·km)) in the north-south direction,while the minimums( 0. 25-0. 42,0. 13-0. 34 μg( m3·km)) appeared from the northeast to the southwest direction. This indicated that the variation in the north to the south direction was bigger than in other directions. The comprehensive heterogeneity index based on the combination of auto-correlation distance,variation and speed ranges changed from 0. 14 to 0. 54 with time,but remained at a middle level. Based on the different heterogeneity of each direction of PM2. 5in the Pearl River Delta,a rectangular regional division grid would work better than a square grid for monitoring site arrangement. The length and width of the grid may be twice the auto-correlation distance of the east to the west( 78 kilometers) and the south to the north( 56kilometers) directions.

DOI

[43]
Xue W, Wu W, Fu F et al., 2015. Satellite retrieval of a heavy pollution process in January 2013 in China. Environmental Science, 36, (3): 794-800. (in Chinese)By simulating vertical stratification data of PM2. 5with the third-generation air quality model CMAQ and high resolution relative humidity data with mesoscale meteorological model WRF,MODIS AOD data were revised by vertical and humidity correcting method,respectively. A linear regression model between revised AOD and PM2. 5was built,and the linear correlation coefficient was r= 0. 77( n = 57,P 0. 01). Based on this model,the average monthly concentrations of PM2. 5in 10 km resolution in January 2013 were firstly retrieved in the country,and the population exposure level was analyzed. The results showed that the areas where average monthly concentrations of PM2. 5was greater than 100 μg·m- 3and 200 μg·m- 3in January 2013 were 10. 99% and 1. 34% of the national terrestrial area,respectively,and the ratios of exposed population were as high as 45. 01% and 6. 31%,respectively.

DOI PMID

[44]
Xue W B, Fu F, Wang J N et al., 2014. Numerical study on the characteristics of regional transport of PM2.5 in China.China Environmental Science, 34(6): 1361-1368. (in Chinese)In order to study the pattern of regional transport of PM2.5 and its chemical components, we developed a transport matrix of PM2.5 and its chemical components from the 31provinces(source) to 333cities(receptor) by applying the Particulate Source Apportionment Technology(PSAT) of CAMx model. The regional contribution of ambient PM2.5and its key components, such as primary PM, sulfate, nitrate, and ammonium, are identified and quantified at region, province, and city level. The results indicate significant contribution of regional transport to ambient PM2.5 pollution in key regions and typical cities in Jing-Jin-Ji area. 22%, 37%, 28%, and 14% of ambient PM2.5in Jin-Jing-Ji, Yangtze River Delta, Pearl River Delta, Chengdu-Chongqing area, respectively, is contributed by emissions from outside region. Regional transport of pollutants contributes to more than 45% of annual average PM2.5 concentration in some provinces such as Hainan, Shanghai, Jiangsu, Zhejiang, Jilin and Jiangxi, and contributes to 37%, 42% and 33% of annual average PM2.5 concentration in Beijing, Tianjin, and Shijiazhuang City.

[45]
Yi H, Hao J, Tang X L et al., 2007. Atmospheric environmental protection in China: Current status, developmental.Energy Policy, 35(2): 907-915.Atmospheric environmental quality in China has been improving due to a variety of programs implemented by the Chinese government in recent decades. However, air pollution is still serious because of rapid socioeconomic development and increased energy consumption. Atmospheric environmental problems appear to be complex and regional in nature, and China's climate is aggravated by global climatic change. Air pollution originates from multiple sources and the effect on public human health will increase. The influence of acid rain in southern China will be long term, and the impact of climate change will rise. In order to reduce the adverse effects of air pollutants on the environment, the total number of emission sources from major industry, fine particle pollutants, SO2 emissions from power plants and the vehicle exhaust must be lowered and strictly controlled. The energy structure will affect the quality of the atmosphere for a long time. Increased energy efficiency, optimization of energy structure and the generation of a sustainable consumption and production patterns will provide opportunities to resolve regional and the global environmental problems.

DOI

[46]
Zhang Y, Cao F, 2015. Fine particulate matter (PM2.5) in China at a city level.Scientific Reports, 5: 1-11.

DOI

[47]
Zhang Y, Zhang W, Wang J et al., 2015. Establishment and application of pollutant inventory-chemical mass balance (I-CMB) model for source apportionment of PM2.5.Transactions of Atmospheric Sciences, 38(2): 279-284. (in Chinese)Aiming at the disability of receptor model in the source apportionment for secondary organic and inorganic aerosols in PM2. 5,a pollutant Inventory-based Chemical Mass Balance( I-CMB) model for source apportionment of PM2. 5is established,and the recent actual pollution data in Beijing is put into the model. Results show that the coal is the largest source( accounting for about 28. 06%) of PM2. 5in Beijing,followed by the vehicle( 19. 73%),dust( 17. 88%),industry( 16. 50%),food( 3. 43%),and plants( 3. 40%). Compared to conventional CMB( Chemical Mass Balance),application of I-CMB in source apportionment requires less accurate source profile and is more resistant to interference,and the result is more balanced and detailed. The I-CMB model meets the demands of PM2. 5reduction in China.

Options
Outlines

/

〈 〉