Vous êtes sur la page 1sur 10

Aerosol and Air Quality Research, 13: 333342, 2013

Copyright Taiwan Association for Aerosol Research


ISSN: 1680-8584 print / 2071-1409 online
doi: 10.4209/aaqr.2012.06.0154

Application of Trajectory Clustering and Source Apportionment Methods for


Investigating Trans-Boundary Atmospheric PM10 Pollution

Shuiyuan Cheng1*, Fang Wang1,2, Jianbing Li3, Dongsheng Chen1, Mingjun Li1, Ying Zhou1,
Zhenhai Ren4
1
College of Environmental & Energy Engineering, Beijing University of Technology, Beijing 100124, China
2
Institute of Environmental Engineering, Beijing General Research Institute of Mining & Metallurgy, Beijing 100070, China
3
Environmental Engineering Program, University of Northern British Columbia, Prince George, British Columbia V2N 4Z9,
Canada
4
Chinese Research Academy of Environmental Sciences, Beijing 100012, China

ABSTRACT

A modeling framework was proposed to investigate the impact of trans-boundary air pollutant transport on regional air
quality. This was based on a combination of the HYSPLIT trajectory model, the CAMx air quality model, and the MM5
meteorological model. The examination of atmospheric PM10 pollution in Guangzhou within the Pearl River Delta (PRD)
region of southern China was used as a case study. The HYSPLIT and MM5 models were used to qualitatively identify the
dominant PM10 pollutant transport pathways that led to PM10 pollution events in Guangzhou, with five clusters of air mass
trajectories being examined. The emission source contribution through each transport pathway to Guangzhous PM10
concentration was then quantified using a MM5-CAMx modeling system. The results illustrated that the trans-boundary
PM10 transport played a critical role in the formation of PM10 pollution events in Guangzhou, with a mean contribution
ratio of nearly 49%. In particular, two air mass trajectory clusters that originated from Guangzhous surrounding regions
were found to be the main pollutant transport pathways, and three surrounding cities (Foshan, Dongguan and Huizhou) had
a total emission contribution of nearly 30% to Guangzhous PM10 concentration through these two pathways. The emissions
from these three cities also accounted for 70 to 94% of the total trans-boundary contributions from Guangzhous nine
surrounding cities through the five transport pathways. As a result, in order to improve Guangzhous air quality, coordinated
effort is required to reduce emissions in both Guangzhou itself and its three surrounding cities. It is expected that the
presented modeling approach can be applied to air quality studies in many other regions.

Keywords: Air pollution; CAMx/PSAT model; Cluster analysis; Emission source contribution; HYSPLIT model.

INTRODUCTION deteriorated air quality (Wei et al., 2007). For example,


haze has been frequently observed in the PRD region due
The Pearl River Delta (PRD) region is located at the Pearl to the high level of air particulates (Tan et al., 2009a). As
River estuary where the river enters the South China Sea. the geographical center of the PRD region, Guangzhou is
With many developed cities on both sides of the Pearl the largest city in southern China. In recent years, the air
River, the PRD region is the manufacturing and distribution pollution problem in Guangzhou has been of concern due to
center of southern China. This region has undergone rapid the adverse effects it may result in. Air pollution can decrease
urbanization during the past two decades due to a significant atmospheric visibility, change meteorological conditions, and
influx of migrants. The increasing population and rapid negatively impact human health (Sun et al., 2006; Chan and
industrialization of this area have resulted in seriously Yao, 2008; Tie and Cao, 2009; Xiao et al., 2011; Lin et al.,
2012). Particularly, the atmospheric visibility in Guangzhou
has been significantly reduced with a reported decline rate
of 0.3 km per year during the past four decades (Huang et al.,
*
Corresponding author. Tel.: 86-10-67391656; 2008). The increasing air particulate matter levels have also
Fax: 86-10-67391983 been linked to a significant increase in lung cancer occurrence
E-mail address: chengsy@bjut.edu.cn in Guangzhou over the past few years (Tie et al., 2009). As

Corresponding author. Tel.: +1250 9606397; a result, there is a pressing need to understand and control
Fax: +1250 9605845 air pollution in Guangzhou.
E-mail address: li@unbc.ca It has been recognized that air pollution is not only a
334 Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013

local but also a regional problem (Oh et al., 2011). Some Trajectory Calculation and Clustering
pollutants can be transported regionally over hundreds or Generally, the qualitative identification of air pollutant
even thousands of kilometers (Bergin et al., 2005; Wang et transport pathways can be conducted through air trajectory
al., 2010; Zheng et al., 2010). Guangzhou is surrounded by clustering by grouping similar trajectories in terms of air
many large cities in the PRD region, including Foshan, mass movement (Abdalmogith and Harrison, 2005). This
Dongguan, Shenzhen, and Hong Kong. Consequently, the method has been widely applied to identify homogeneous
trans-boundary transport of air pollutants from these cities groups of air mass transport patterns that affect air quality
may greatly contribute to Guangzhous air pollution problem. in urban regions (Prapat and Nguyen, 2007; Wang et al.,
For example, Wang et al. (2005) examined the different 2010). In this study, air mass trajectory clustering was used
characteristics of air pollutant transport among the cities in to reveal the relationship between the atmospheric transport
the PRD region and found that Guangzhous air pollution can pathways and the PM10 pollution levels in Guangzhou in
be significantly affected by emissions from its surrounding the autumn of 2006 (September 1 to November 30), which
regions. Lee et al. (2007) reported that the enrichment of is usually the most polluted season. An overview of air
atmospheric heavy metals in Hong Kong and Guangzhou pollution, meteorology, and topography within the PRD
was closely associated with the air mass from the north and region and Guangzhou can be found in a previous study
the northeast that originates from northern China. The (Chan and Yao, 2008). The HYSPLIT model is the most
effect of such trans-boundary air pollutant transport may widely used air trajectory model for establishing source-
hamper the efforts that Guangzhou has made to address its receptor relationships over long distances (Wang et al.,
air pollution problem. In order to develop an effective air 2010), and was thus utilized in this study. The version 4.8
quality management strategy in Guangzhou, it is of critical of this model was applied to calculate the 24-hour back
importance to identify the dominant air pollutant transport trajectories (one in each hour in a day) at an altitude of 30
pathways and quantify the effect of trans-boundary air m above the sea level in Guangzhou (23.16N, 113.23E)
pollutant transport on the air pollution in the city. However, (Draxler and Hess, 1998). The selection of 24-hour back
there have been very limited studies so far to address this trajectories was because the objective of this study was to
issue. Particularly, the PM10 pollution has been recognized focus on regional air pollutant transport in the PRD region
as a serious air quality problem in Guangzhou and in the instead of long-distance transport from far-away areas. The
PRD region (Tan et al., 2009a, b). Although some previous altitude of the starting point for trajectory calculation was
works have revealed the existence of trans-boundary air based on the fact that the sampling height of the PM10
pollutant transport from the surrounding areas to Guangzhou monitoring station in Guangzhou was located at 30 m above
(Wang et al., 2005, Lee et al., 2007; Zheng et al., 2010), the the sea level.
dominant transport pathways and their quantitative effects The MM5 meteorological model was used to provide the
on atmospheric PM10 pollution in Guangzhou still remain meteorological data fields required to run the HYSPLIT
poorly characterized. model. A detailed description of the MM5 model can be
The objective of this study was to investigate the impact found in Dudhia et al. (2004). In this study, the MM5 model
of trans-boundary air pollutant transport on the atmospheric was configured using a two-level nested modeling domain
PM10 pollution in Guangzhou through both qualitative and on a Lambert map projection centered at (23N, 113.5E).
quantitative modeling approaches. The PM10 pollution in The outer modeling domain had a spatial resolution of 27
the autumn of 2006 was analyzed as a case study. The km 27 km covering most of southeastern China, and was
HYSPLIT (Hybrid Single-Particle Lagrangian Integrated established with a dimension of 100 100 grid cells. The
Trajectory) model (Draxler and Hess, 1998) and the MM5 inner domain had a spatial resolution of 9 km 9 km covering
meteorological model (Dudhia et al., 2004) were used to all of the cities within the PRD region, and was set up with
track the air masses and analyze the dominant transport a dimension of 58 44 grid cells (Fig. 1). Vertically, 35
pathways, while a k-means clustering algorithm was applied layers from the ground level to an altitude of 15 km were
to group various air mass trajectories into different transport designed with uneven spacing, among which 20 layers were
pathways. The quantification of emission source contribution distributed within a height of 3 km from the ground level in
to the ambient PM10 pollution in Guangzhou from each city order to provide specific planetary boundary information.
in the PRD region through each trans-boundary transport The initial conditions (IC) and boundary conditions (BC) for
pathway was then analyzed using the CAMx air quality the MM5 model were obtained from the NCEP data every 6
model. The MM5 model was used to provide meteorological h with a 1 1 resolution. The vertical configuration used
inputs for the CAMx model, and the particulate matter in the HYSPLIT model was the same as that used in the
source apportioning technology (PSAT) implemented within MM5 model. Based on the results of HYSPLIT model, the
CAMx model was used to quantify the emission source air mass trajectories were assigned to distinct clusters
contribution. Consequently, the effects of trans-boundary according to their moving speed and direction using a k-
air pollutant transport through different transport pathways means clustering algorithm (Hartigan and Wong, 1979). For
on Guangzhous PM10 pollution were characterized. This the k-means algorithm, the results may change when using
would provide a sound basis for developing effective air different number of clusters. Since the concern of this study
quality management strategy in the study region. was the pollutant transport pathway, the clustering result
with the smallest possible cluster number should be selected
METHODOLOGY (Wang et al., 2010). In this study, various pre-selected
Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013 335

Guangzhou
Zhaoqing

Huizhou

Foshan Dongguan

Shenzhen

Jiangmen Zhongshan
Hong Kong South China Sea

Zhuhai Macau

0 10 20 40 80 D2
Kilometers D1

Fig. 1. The administrative divisions of the PRD region (the yellow dots represent the locations of point emission sources, the
blue flags represent the automatic meteorological observation stations, and the red towers represent the air quality monitoring
stations).

number of clusters (e.g., 3, 4, 5, 6, and 7) were tested, and organic aerosol, crustal fine, other fine, crustal coarse, and
it was found that 5 clusters can best represent the other coarse). The source apportionment can be calculated
meteorological characteristics of the transport pathways in in parallel to the main program in CAMx for calculating
Guangzhou. As a result, this number was selected to be the pollutant concentrations, which means that selecting PSAT
expected number of air mass trajectory clusters. A more does not affect the CAMx simulation results (Koo et al.,
detailed clustering procedure by using the k-means algorithm 2009; Huang et al., 2010).
can be found in Wang et al. (2010). After identifying the dominant air pollutant transport
pathways to Guangzhou using the HYSPLIT model, the
CAMx Model with PSAT emission contribution to Guangzhous PM10 concentration
The atmospheric PM10 concentration in Guangzhou and through each transport pathway was then calculated using
its surrounding areas was simulated using the Comprehensive the CAMx model with PSAT module. Similar to the
Air Quality Model with Extensions (CAMx). CAMx is a HYSPLIT model, the meteorological data fields required to
publicly available Eulerian photochemical dispersion model run the CAMx model were provided by the MM5 modeling
that allows for the integrated assessment of gaseous and results. The simulation domain for the CAMx model was
particulate air pollutants over many scales (Environ, 2006). the same as the inner domain for the MM5 model, but was
One of the unique features of CAMx is its mass tracking designed with 5 grids less in each of the two horizontal
module called particulate matter source apportioning directions in order to minimize the side effects at the
technology (PSAT) for studying source apportionment from boundary of the meteorological modeling domains (Wang
different emission source categories and regions (Dunker et et al., 2010). The 35 vertical layers for the MM5 model
al., 2002; Yarwood et al., 2005; Ward et al., 2012). It can were also collapsed into 12 layers for the vertical domain
conduct source apportionment for particulate matter (PM) of the CAMx model, with 9 of them being distributed
species, including sulfate, particulate nitrate, ammonium, within an altitude of 3 km from the ground level (Wang et
particulate mercury, secondary organic aerosol, and six al., 2010). As a result, the CAMx modeling domain was set
categories of primary PM (e.g., elemental carbon, primary up with a dimension of 48 34 grid cells, covering all of
336 Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013

the administrative divisions in the PRD region. A total of sources (such as residential fuel consumption, waste burning,
11 emission source areas were defined, including nine cities and biomass burning). A total of 2292 point emission sources
(e.g., Guangzhou, Huizhou, Foshan, Dongguan, Jiangmen, (Fig. 1) were collected in this study, including industrial
Zhongshan, Zhaoqing, Zhuhai, and Shenzhen), one special and non-industrial stationary equipment or processes. The
administration region (i.e., Hong Kong), and all of other emission data of area sources were obtained at a county-
areas outside the boundary of the PRD region within the level. Table 1 lists the PM10 emission rates from both point
CAMx modeling domain, as shown in Fig. 2. sources and area sources in the PRD region in 2006. Fig. 1
illustrates that the air pollutant emissions in the PRD region
Emission Data were mainly distributed over the central-southern areas. For
The locations of the nine cities within the PRD region areas outside the PRD region, emission inventories were
are shown in Fig. 1. An emission inventory for the year taken from those prepared by Streets et al. (2003). The
2006, with high temporal and spatial resolution, was compiled biogenic and VOCs emission data sets were obtained from
for the PRD region using a bottom-up investigation the GEIA emission inventories (Global Emission Inventory
approach. A seasonal coefficient was used to adjust the Activity). The Sparse Matrix Operator Kernel Emissions
annual emission inventory into different seasons in order to (SMOKE) model was used to convert the emission inventory
provide emission inventory for air quality simulation within data into the formatted emission files required by the CAMx
the study period of September 1November 30 in 2006. model (Houyoux and Vukovich, 1999; Borge et al., 2008).
The emission inventory covers the major emission sources
that were divided into six categories based on the main Meteorological and Air Quality Monitoring Data
emission types in the PRD region, including power plant, A PM10 monitoring station at Luhu park is located in the
industrial, traffic, biogenic, VOC product-related, and other urban area of Guangzhou and was used to represent the

2 1
5
3
4

9
7
6 10

0 10 20 40 80
Kilometers

Fig. 2. Division of the air quality modeling domain into 11 emission source areas, with area 11 (not shown) representing all
other areas along the boundary of the PRD region that were not included in areas 110 (1: Guangzhou; 2: Zhaoqing; 3:
Foshan; 4: Dongguan; 5: Huizhou; 6: Jiangmen; 7: Zhongshan; 8: Zhuhai; 9: Shenzhen; 10: Hong Kong).

Table 1. PM10 emissions from different sources within the PRD region in 2006.
City Area sources (t/yr) Point sources (t/yr) Total (t/yr)
Guangzhou 77792.47 113152.46 190944.93
Shenzhen 40514.27 6598.73 47112.99
Zhuhai 13404.32 6347.79 19752.11
Foshan 57683.36 49721.19 107404.55
Jiangmen 44277.74 40856.29 85134.03
Dongguan 48606.19 56521.65 105127.84
Zhongshan 20766.36 9759.03 30525.39
Huizhou 26940.35 11808.52 38748.87
Zhaoqing 26563.37 17084.08 43647.45
Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013 337

citys PM10 pollution characteristics. A Beta Gauge method observation values of PM10 concentration ranged from 24 to
was used at this station to measure the PM10 concentration 36%, with an average value of 30%. The Pearson correlation
at a sampling height of 9 m above ground level (i.e., 30 m coefficient ranged from 0.58 to 0.72, with an average value
above the sea level). The hourly PM10 concentration data of 0.64. These results indicate that the MM5-CAMx modeling
was also obtained from 15 separate air quality monitoring performance was generally satisfactory and acceptable (Eder
stations (e.g., 12 in PRD region and 3 in Hong Kong, as and Yu, 2006).
shown in Fig. 1) for evaluating the modeling performance.
The 3-D first-guess meteorological fields, from 2006, for Characteristics of Various Air Pollutant Transport
the MM5 model were obtained from the Global Troposphere Pathways
Analyses datasets provided by the US National Center for Fig. 3 presents the five clusters of air pollutant transport
Environmental Prediction, and were available every 6 h with pathways to Guangzhou obtained from the HYSPLIT model.
1 1 resolution. The required topography data for the It can be found that the air mass of cluster 1 came from the
MM5 model was obtained from the Digital Elevation Model eastern coastal areas of the PRD region, mainly Dongguan,
from USGS with a spatial resolution of 30 s, and the required Huizhou, Shenzhen, and Hong Kong, and moved westward
land-use data for the MM5 model was obtained from the to arrive in Guangzhou. This transport pathway occurred
USGSs 1-km NOAA/AVHRR satellite data sets. The for 636 hours out of the total simulated 2184 hours in the
observed meteorological data, from 2006, was obtained from autumn of 2006, with an occurrence frequency of 29%.
the Guangdong Provincial Meteorological Administration The air mass of cluster 2 came mainly from Foshan city in
through its network of automatic meteorological observational the western part of the PRD region, and its occurrence
stations as shown in Fig. 1. The meteorological observation frequency was 20%. Clusters 3, 4 and 5 were mainly from
data was applied for the 4DDA meteorological data the northeastern areas of Guangdong province, and had a
assimilation into MM5 (Dudhia et al., 2004). total occurrence frequency of 51%.
Table 3 lists the monitored mean PM10 concentration in
RESULTS AND DISCUSSIONS Guangzhou during the occurrence of each transport pathway
and the corresponding meteorological conditions during the
Simulation Performance of the MM5-CAMx Modeling three autumn months of 2006. It can be seen that clusters 1
System and 2 were associated with higher PM10 concentrations in
The simulation performance of the MM5-CAMx modeling Guangzhou, while the other three clusters were corresponding
system is of critical importance for its effective application to lower PM10 concentrations. Fig. 4 presents the variation
to air quality management. In this study, the hourly PM10 of the observed hourly PM10 concentration in Guangzhou
concentrations in the first vertical layer of the CAMx and the occurrence of different transport pathways from
modeling domain during the study period was simulated September 1 to November 30, in 2006. It is evident that the
using the MM5-CAMx modeling system. The simulations PM10 pollution in Guangzhou was closely related to air
results in the grids representing the 16 air quality monitoring mass trajectory clusters 1 and 2. The occurrence of these
stations in the PRD region were then compared with the two transport pathways was generally corresponding to the
related observations at these monitoring stations. Table 2 increased and high-level PM10 concentrations in Guangzhou.
lists the relative modeling errors and the Pearson correlation It can be found from Fig. 3 that clusters 1 and 2 were
coefficients under the significance level of 0.01. It can be associated with a shorter transport distance when compared
found that the relative error between the simulation and to the other three clusters. Their air masses mainly moved

Table 2. Simulation error of the hourly PM10 concentration and the Pearson correlation coefficient.
City (monitoring station) Relative modeling error (%) Pearson correlation coefficient
Guangzhou (Luhu park) 26 0.72
Guangzhou (Wanqingsha) 25 0.70
Guangzhou (Tianhu) 28 0.65
Shenzhen 29 0.65
Zhuhai 31 0.66
Foshan (Shunde) 32 0.60
Foshan (Huijingcheng) 30 0.61
Jiangmen 30 0.63
Dongguan 27 0.68
Zhongshan 24 0.71
Huizhou (Jinguowan) 31 0.62
Huizhou (Xiabu) 30 0.60
Zhaoqing 31 0.59
Hong Kong (Quanwan) 36 0.58
Hong Kong (Tamen) 32 0.61
Hong Kong (Tung Chung) 34 0.60
338 Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013

26 o N 26 o N

25 o N (b) 25 o N

24 o N 24 o N
o
30 N
23 o N 23 o N

(a) 22 o N 22 o N
28 o N
21 o N 21 o N

20 o N 20 o N
26 o N
Cluster1
19 o N o 19 o N
110 E 112 o E 114 o E 116 o E 118 o E

24 o N
26 o N

(c) 25 o N
22 o N
24 o N

20 o N 23 o N

22 o N

18 o N 21 o N
108 o E 111 o E 114 o E 117 o E 120 o E 123 o E
20 o N
Cluster2
19 o N
110 o E 112 o E 114 o E 116 o E

Fig. 3. The simulated back trajectories arriving at Guangzhou in the autumn of 2006, (a) five clusters of trajectories (cyan
for cluster 1, blue for cluster 2, green for cluster 3, red for cluster 4, and yellow for cluster 5), (b) trajectories in cluster 1,
(c) trajectories in cluster 2.

Table 3. Monitored mean PM10 concentration in Guangzhou during the occurrence of each transport pathway and the
corresponding meteorological conditions in autumn 2006.
Transport pathway PM10 concentration (mg/m3) Wind speed (m/s) Temperature (C) Relative humidity (%)
Cluster 1 0.081 0.8 24.2 74.4
Cluster 2 0.080 0.7 27.0 74.6
Cluster 3 0.061 1.7 23.8 58.4
Cluster 4 0.060 1.0 24.1 65.0
Cluster 5 0.061 1.0 23.7 68.7

through the highly industrialized and densely populated was associated with weaker wind speed since the windy
areas within the PRD region, such as Dongguan, Huizhou, weather is conducive to the dispersion of pollutants. In
and Foshan. The emissions from these cities were transported summary, the atmospheric PM10 pollution in Guangzhou
to Guangzhou with air mass movement, thus causing elevated had a direct relationship with the air mass transport pathway,
pollution levels in Guangzhou as shown in Fig. 4. and the air mass trajectory clusters 1 and 2 (Fig. 3) represent
The results in Table 3 indicate that the emissions from the main trans-boundary transport pathways that influenced
Guangzhous surrounding areas had a significant contribution Guangzhous PM10 concentrations. During their occurrence
to its PM10 pollution levels through the trans-boundary in the autumn of 2006, the average PM10 concentration in
transport effect. Similar observations were found in a haze Guangzhou was 0.081 mg/m3, which is 32.7% higher than
formation study in Guangzhou (Tan et al., 2009b). Moreover, that (i.e., 0.061 mg/m3) during the occurrence of the other
it can be found from Table 3 that the PM10 pollution in three transport pathways.
Guangzhou was closely related to its meteorological
conditions. The occurrence of clusters 1 and 2 was observed Trans-Boundary Emission Contribution from Each
to be associated with relatively lower wind speed, higher Surrounding City
temperature, and higher relative humidity than the other Fig. 5 presents the calculated average ratios of PM10
three air mass trajectory clusters. This illustrates that the emission contributions to the hourly PM10 concentration in
dispersion and transport of PM10 was affected by the Guangzhou from the local emission sources and the
meteorological parameters (wind speed, wind direction, surrounding cities, as well as from the initial conditions (ICs)
relative humidity, and temperature). The heavier air pollution and boundary conditions (BCs) during the autumn of 2006.
Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013 339

The ICs represent the gridded initial pollutant concentration


fields in the gridded modeling domain, and the BCs represent
the gridded concentration fields on the lateral faces of the
grid boundary (e.g., north, south, east, west, top). Both ICs
and BCs are important contributions to the simulated air
pollutant concentrations. The system default initial conditions
and boundary conditions were used in the modeling setup.
The emission source contribution ratio was defined as the
ratio of the simulated PM10 concentration in Guangzhou
after removing that emission source from the inventory to
the simulated PM10 concentration in Guangzhou using the
original emission inventory (Cheng et al., 2007). It can be
found that the mean contribution ratio of the local emission
source to Guangzhous PM10 concentration in the study period
was 51%, implying that the mean trans-boundary emission
contribution ratio was 49% (i.e., including contributions
from Guangzhous nine surrounding cities, the other areas
outside the boundary of the PRD region within the CAMx
modeling domain, and the ICs and BCs). This result is
similar to the findings in a previous SO2 source apportionment
study which showed that the mean annual contribution to
the SO2 concentration in Guangzhou from its surrounding
areas could reach as high as 50% (Wang et al., 2005). It is
evident from the simulation results that trans-boundary
emission contributions played a critical role in the formation
of PM10 pollution events in Guangzhou. It can be found from
Fig. 5 that the major trans-boundary emission contributors
included Huizhou (8%), Foshan (6%), and Dongguan (8%),
with a total emission contribution ratio of 22%, which
accounts for 79% of the total trans-boundary contributions
from the nine surrounding cities of Guangzhou in the PRD
region. The emission contribution ratios of the other six
cities in the PRD region were ranked in descending order
as Shenzhen, Zhaoqing, Zhongshan, Jiangmen, Hong Kong
and Zhuhai. The areas outside the boundary of the PRD
region accounted for an emission contribution ratio of 18%.

Emission Contribution from Each Surrounding City


through Each Transport Pathway
The results of the trans-boundary emission contributions
were in agreement with the air trajectory analysis results
described above. The trajectory analysis indicated that
clusters 1 and 2 represented the major transport pathways.
In order to further investigate the impact of different transport
pathways on Guangzhous PM10 pollution, the contributions
from different emission source areas to Guangzhous PM10
concentration through each transport pathway were simulated
using the MM5-CAMx modeling system. Table 4 presents
the modeling results. It was found that through the transport
pathways of air mass trajectory clusters 3, 4, and 5, about
26% of the PM10 concentration in Guangzhou could be
attributed to the emissions from the areas outside the PRD
region. However, this number was reduced to about 11%
through the transport pathways of clusters 1 and 2. Through
cluster 1, the upwind areas (including Huizhou, Dongguan,
Shenzhen, and Hong Kong) accounted for a total emission
contribution ratio of 31%. Through cluster 2, the upwind
areas (including Foshan, Zhaoqing and Dongguan) accounted
for 28% of emission contribution to Guangzhous PM10
340 Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013

IC and BC, 3
Other areas, 18
Guangzhou, 51
Hong Kong, 1
Shenzhen, 2

Zhuhai, 0
Zhaoqing, 1

Zhongshan, 1

Dongguan, 8

Jiangmen, 1

Foshan, 6

Huizhou, 8

Fig. 5. Calculated emission source contribution ratio (%) to PM10 concentration in Guangzhou.

Table 4. Emission contribution ratio (%) to Guangzhous PM10 concentration through each transport pathway.
Transport pathway
Emission source areas
Cluster1 Cluster2 Cluster3 Cluster4 Cluster5
Guangzhou 52 53 52 46 55
Huizhou 12 2 5 13 5
Foshan 2 19 1 2 2
Jiangmen 0 4 0 0 0
Dongguan 12 4 4 7 5
Zhongshan 1 2 1 1 1
Zhaoqing 0 5 0 0 0
Zhuhai 0 0 0 0 0
Shenzhen 5 0 0 2 0
Hong Kong 2 0 0 0 0
Other areas 12 10 25 26 26
IC and BC 3 2 11 3 7

concentrations. It can be found from Table 4 that the trans- categories may be required and is worth of further
boundary contribution to Guangzhous PM10 concentration examination. This would help characterize the contribution
from its nine surrounding cities was 34, 36, 11, 25, and from different emission categories within various emission
13% through air mass trajectory clusters 1, 2, 3, 4, and 5, cities to the PM10 concentration in Guangzhou through each
respectively. It can also be found that the emissions from transport pathway, while the proposed modeling approach
Foshan, Dongguan, and Huizhou greatly affected is still appropriate for the calculation of such emission
Guangzhous air quality. Through the transport pathways contribution.
of clusters 1 and 2, their total emission contribution to
Guangzhous PM10 was nearly 26%. Among the total trans- CONCLUSIONS
boundary contributions within the PRD region, the emissions
from these three cities accounted for 77, 70, 91, 88, and In this study, air trajectory clustering and source
92% through the five transport pathways, respectively. apportionment technology were used to examine the
Such information is valuable for developing effective air atmospheric PM10 pollution problem in Guangzhou within
quality management strategy in the study region. It implies the Pearl River Delta region of China. By combining with
that the improvement of air quality in Guangzhou requires the MM5 meteorological model, the HYSPLIT model was
a coordinated effort between Guangzhou and its surrounding used for trajectory clustering, and the impact of air mass
areas, especially Foshan, Dongguan, and Huizhou. In order transport pathways on the PM10 pollution in Guangzhou
to obtain more detailed information for effective decision was analyzed using the CAMx model with the PSAT
making, the source apportionment of different emission module. Five air pollutant transport pathways were identified,
Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013 341

and were represented by five air mass trajectory clusters. Chan, C.K. and Yao, X.H. (2008). Air Pollution in Mega
Among them, two air mass trajectories that originated from Cities in China. Atmos. Environ. 42: 142.
Guangzhous surrounding cities in the PRD region, with a Cheng, S.Y., Chen, D.S., Li, J.B., Wang, H.Y. and Guo,
total occurrence frequency of 49%, were observed to be the X.R. (2007). The Assessment of Emission-source
main transport pathways that greatly affect Guangzhous Contributions to Air Quality by Using a Coupled MM5-
PM10 pollution. It was found from the simulation results ARPS-CMAQ Modeling System: A Case Study in the
that nearly 49% of the PM10 concentration in Guangzhou Beijing Metropolitan Region, China. Environ. Modell.
was due to trans-boundary emission contributions. Softw. 22: 16011616.
Particularly, the cities of Foshan, Huizhou, and Dongguan Draxler, R.R. and Hess, G.D. (1998). An Overview of the
represented the most important trans-boundary emission HYSPLIT_4 Modelling System for Trajectories,
contributions to Guangzhous PM10 concentration. Their Dispersion, and Deposition. Aust. Meteorol. Mag. 47:
contributions accounted for 77, 70, 91, 88, and 92% of the 295308.
total trans-boundary contributions from all of Guangzhous Dudhia, J., Gill, D., Manning, K., Wang, W. and Bruyere,
surrounding cities through the five air pollutant transport C. (2004). PSU/NCAR Mesoscale Modeling System
pathways. Therefore, in addition to emission reduction Tutorial Class Notes and Users Guide: MM5 Modeling
within Guangzhou itself, the regulation of emissions within System Version 3, National Center for Atmospheric
the three surrounding cities is important for improving Research, U.S.A.
Guangzhous air quality. Since there have been few Dunker, A.M., Yarwood, G., Ortmann, J. P. and Wilson, G.
applications of HYSPLIT, MM5 and CAMx models within M. (2002). Comparison of Source Apportionment and
a general modeling framework to both qualitatively and Source Sensitivity of Ozone in a Three-dimensional Air
quantitatively examine the emission contributions to regional Quality Model. Environ. Sci. Technol. 36: 29532964.
air pollution problems, the proposed modeling approach in Eder, B. and Yu, S. (2006). A Performance Evaluation of
this study has the potential to be applied to other urban or the 2004 Release of Models-3 CMAQ. Atmos. Environ.
regional air pollutant emission investigations and decision 40: 48114824.
analysis studies. Environ (2006). Users Guide for the Comprehensive Air
Quality Model with Extensions (CAMx), Version 4.40.
ACKNOWLEDGEMENTS ENVIRON International Corporation, Noavto, California,
USA.
This research was supported by the Natural Sciences Hartigan, J.A. and Wong, M.A. (1979). A k-means
Foundation of China (No. 51038001), the Beijing Science Clustering Algorithm. J. Appl. Stat. 28: 100108.
and Technology Project (D09040903670801) of the Beijing Houyoux, M. and Vukovich, J. (1999). Updates to the
Municipal Science & Technology Commission, and the Sparse Matrix Operator Kernel Emissions (SMOKE)
Ministry of Environmental Protection Special Funds for Modeling System and Integration with Models-3, Presented
Scientific Research on Public Causes (No. 200909008 & at The Emissions Inventory: Regional Strategies for the
201209003). The authors would also like to thank the Future Conference, Air & Waste Management Association,
Natural Science Foundation of Beijing (No.8092004), the Raleigh, NC.
Beijing NOVA Program of China (No. 2009B07), the Huang, J., Wu, D., Huang, M.H., Li, F., Bi, X.Y., Tan,
Innovation Team Project of Beijing Municipal Education H.B. and Deng, X.J. (2008). Visibility Variations in the
Commission (PHR201007105), as well as the Cultivation Pearl River Delta of China during the Period of 1954-
Fund of the Key Scientific and Technical Innovation Project, 2004. J. Appl. Meteorol. Sci. 19: 6170.
Ministry of Education of China (708017) for supporting Huang, Q., Cheng, S.Y., Li, Y.P., Li, J.B., Chen, D.S. and
this work. The authors are also grateful to the anonymous Wang, H.Y. (2010). An Integrated MM5-CAMx Modeling
reviewers for their insightful comments. Approach for Assessing PM10 Contribution from Different
Sources in Beijing, China. J. Environ. Inform. 15: 4761.
REFERENCES Koo, B., Wilson, G.M., Morris, R.E., Dunker, A.M. and
Yarwood, G. (2009). Comparison of Source
Abdalmogith, S.S. and Harrison, R.M. (2005). The Use of Apportionment and Sensitivity Analysis in a Particulate
Trajectory Cluster Analysis to Examine the Long-range Matter Air Quality Model. Environ. Sci. Technol. 43:
Transport of Secondary Inorganic Aerosol in the UK. 66696675.
Atmos. Environ. 39: 66866695. Lee, C.S.L., Li, X.D., Zhang, G., Li, J., Ding, A.J. and
Bergin, M.S., West, J.J., Keating, T.J. and Russell, A.G. Wang, T. (2007). Heavy Metal Concentrations and Pb
(2005). Regional Atmospheric Pollution and Isotopic Composition in Urban and Suburban Aerosols
Transboundary Air Quality Management. Annu. Rev. of Hong Kong and Guangzhou, South China - Evidence
Environ. Resour. 30: 137. of the Long-range Transport of Air Contaminants. Atmos.
Borge, R., Lumbreras, J. and Rodriguez, E. (2008). Environ. 41: 432447.
Development of a high-resolution emission inventory for Lin, M., Tao, J., Chan, C.Y., Cao, J.J., Zhang, Z.S., Zhu, L.H.
Spain using the SMOKE modelling system: a case study and Zhang, R.J. (2012). Regression Analyses between
for the Years 2000 and 2010. Environ. Modell. Softw. Recent Air Quality and Visibility Changes in Megacities
23: 10261044. at Four Haze Regions in China. Aerosol Air Qual. Res.
342 Cheng et al., Aerosol and Air Quality Research, 13: 333342, 2013

12: 10491061. PM10 Transport Pathways Using HYSPLIT, MM5-CMAQ


Oh, M.S., Lee, T.J. and Kim, D.S. (2011). Quantitative and Synoptic Pressure Pattern Analysis. Environ. Modell.
Source Apportionment of Size-segregated Particulate Softw. 25: 927934.
Matter at Urbanized Local Site in Korea. Aerosol Air Wang, S.L., Zhang, Y.H., Zhong, L.J., Li, J.L. and Yu, Q.
Qual. Res. 11: 247264. (2005). Interaction of Urban Air Pollution among Cities
Prapat, P. and Nguyen, T.K.O. (2007). Assessment of in Zhujiang Delta. China Environ. Sci. 25: 133137.
Potential Long-range Transport of Particulate air Pollution Ward, T., Trost, B., Conner, J., Flanagan, J. and Jayanty,
Using Trajectory Modeling and Monitoring Data. Atmos. R.K. M. (2012). Source Apportionment of PM2.5 in a
Res. 85: 317. Subarctic Airshed - Fairbanks, Alaska. Aerosol Air Qual.
Streets, D.G., Bond, T.C., Carmichael, G.R., Fernandes, S.D., Res. 12: 536543.
Fu, Q., He, D., Klimont, Z., Nelson, S.M., Tsai, N.Y., Wei, X.L., Li, Y.S., Lam, K.S., Wang, A.Y. and Wang,
Wang, M.Q., Woo, J.H. and Yarber, K.F. (2003). An T.J. (2007). Impact of Biogenic VOC Emissions on a
Inventory of Gaseous and Primary Aerosol Emissions in Tropical Cyclone-related Ozone Episode in the Pearl River
Asia in the Year 2000. J. Geophys. Res. 108: 8809. Delta Region, China. Atmos. Environ. 41: 78517864.
Sun, Y.L., Zhuang, G.S., Tang, A.H., Wang, Y. and An, Xiao, Z.M., Zhang, Y.F., Hong, S.M., Bi, X.H., Jiao, L.,
Z.S. (2006). Chemical Characteristics of PM2.5 and PM10 Feng, Y.C. and Wang, Y.Q. (2011). Estimation of the
in HazeFog Episodes in Beijing. Environ. Sci. Technol. Main Factors Influencing Haze, Based on a Long-term
40: 31483155. Monitoring Campaign in Hangzhou, China. Aerosol Air
Tan, J.H., Duan, J.C., Chen, D.H., Wang, X.H., Guo, S.J., Qual. Res. 11: 873882.
Bi, X.H., Sheng, G.Y., He, K.B. and Fu, J.M. (2009a). Yarwood, G., Wilson, G. and Morris, R. (2005).
Chemical Characteristics of Haze during Summer and Development of the CAMx Particulate Source
Winter in Guangzhou. Atmos. Res. 94: 238245. Apportionment Technology (PSAT) - Final Report.
Tan, J.H., Duan, J.C., He, K.B., Ma, Y.L., Duan, F.R., Chen, ENVIRON International Corporation, Noavto, California,
Y. and Fu, J. M. (2009b). Chemical Characteristics of USA.
PM2.5 during a Typical Haze Episode in Guangzhou. J. Zheng, J.Y., Zhong, L.J., Wang, T., Louie, P.K.K. and Li,
Environ. Sci. 21: 774781. Z.C. (2010). Ground-level Ozone in the Pearl River Delta
Tie, X.X. and Cao, J.J. (2009). Aerosol Pollution in China: Region: Analysis of Data from a Recently Established
Present and Future Impact on Environment. Particuology Regional Air Quality Monitoring Network. Atmos.
7: 426431. Environ. 44: 814823.
Tie, X.X., Wu, D. and Brasseur, G. (2009). Lung Cancer
Mortality and Exposure to Atmospheric Aerosol Particles
in Guangzhou, China. Atmos. Environ. 43: 23752377. Received for review, June 20, 2012
Wang, F., Chen, D.S., Cheng, S.Y., Li, J.B., Liu, M.J. and Accepted, August 31, 2012
Ren, Z.H. (2010). Identification of Regional Atmospheric

Vous aimerez peut-être aussi