-
at SciVerse ScienceDirect
Atmospheric Environment 76 (2013) 11e20Contents lists
availableAtmospheric Environment
journal homepage: www.elsevier .com/locate/atmosenvEmission
trends and source characteristics of SO2, NOx, PM10 andVOCs in the
Pearl River Delta region from 2000 to 2009Qing Lu a,b, Junyu Zheng
a,b,*, Siqi Ye a,b, Xingling Shen a,b, Zibing Yuan b,c, Shasha Yin
a,b
a School of Environmental Science and Engineering, South China
University of Technology, University Town, Guangzhou 510006, PR
Chinab Pearl River Delta Atmospheric Environmental Research Joint
Laboratory, Guangzhou 510006, PR ChinacAtmospheric Research Center,
Fok Ying Tung Graduate School, Hong Kong University of Science and
Technology, Nansha IT Park,Guangzhou 511458, PR China
h i g h l i g h t s< PRD emission trends were characterized
and validated from 2000e2009.< Variations in source
characteristics were investigated and analyzed.< SO2 emission
began to decrease from 2005, while PM10 emission decreased from
2007.< NOx and VOCs emissions exhibited upward trends during
2000e2009.< Immediate control is needed on marine emission
source in the PRD region.a r t i c l e i n f o
Article history:Received 6 March 2012Received in revised form22
October 2012Accepted 27 October 2012
Keywords:Emission estimationSource contributionControl
policySatellite dataGround observations* Corresponding author.
B4-514, School of Environeering, South China University of
Technology, SoutGuangzhou 510006, PR China. Tel./fax: 86 20
39380
E-mail address: [email protected] (J. Zheng
1352-2310/$ e see front matter 2012 Elsevier
Ltd.http://dx.doi.org/10.1016/j.atmosenv.2012.10.062a b s t r a c
t
Emission trends and variations in source contributions of SO2,
NOx, PM10 and VOCs in the Pearl RiverDelta (PRD) region from 2000
to 2009 were characterized by using a dynamic methodology, taking
intoaccount the economic development, technology penetration, and
emission control. The results indicatedthat SO2 emissions increased
rapidly during 2000e2005 but decreased significantly afterward.
NOxemissions went up consistently during 2000e2009 except for a
break point in 2008. PM10 emissionsincreased by 76% during
2000e2007 but started to decrease slightly in the following years.
VOCsemissions presented continuous increase during the study
period. Power plants and industrial sourceswere consistently the
largest SO2 and PM10 emission contributors. The on-road mobile
source was thelargest emission contributor for VOCs and NOx
emissions with decreasing contributions. The NOxcontribution from
power plants and industrial sources kept increasing. Worthy of
mention is that thenon-road mobile source is becoming an important
SO2 and NOx contributor in this region. Comparisonswith satellite
data, ground observations and national trends indicated that
emission trends developed inthis study were reasonable.
Implications for future air pollution control policies were
discussed.
2012 Elsevier Ltd. All rights reserved.1. Introduction
The PRD region, located in the southern coast of China,
coverscities of Guangzhou, Shenzhen, Zhuhai, Foshan, Dongguan,
Zhong-shan, Jiangmen, Huizhou and Zhaoqing (see Fig. 1). Benefited
fromthe implementation of Chinas reform and opening-up policies,
thePRD region has experienced rapid economic growth, with
thesurgingGrossDomestic Product (GDP)by280%, fuel consumptionby150%
and the population of passenger cars by 530% (GDPBS, 2001enmental
Science and Engi-h Campus, University Town,021.).
All rights reserved.2010), respectively, from 2000 to 2009
(Figs. 2 and 8). However,these dramatic growths have caused
serious, complex and regionalair pollution problems (Zhang et al.,
2008; Zheng et al., 2010). Themonitoring data showed that the
annual average number of hazedays was over 100 and the observed
highest ozone concentrationwas up to 0.45 mg m3 in the region
(GDEMC and HKEPD, 2005e2010; Deng et al., 2008), and the ozone
background concentrationsincreased by an average rate of 0.55 ppbv
yr1 during 1994e2007(Wang et al., 2009a). These findings indicated
that air pollutioncontrol in this region is challenging.
In order to improve air quality in the PRD region, both
nationaland local government agencies have made great efforts to
formu-late and issue various control measures and policies in the
pastdecade. These policies and control measures were summarized
in
mailto:[email protected]://crossmark.dyndns.org/dialog/?doi=10.1016/j.atmosenv.2012.10.062&domain=pdfwww.sciencedirect.com/science/journal/13522310www.elsevier.com/locate/atmosenvhttp://dx.doi.org/10.1016/j.atmosenv.2012.10.062http://dx.doi.org/10.1016/j.atmosenv.2012.10.062http://dx.doi.org/10.1016/j.atmosenv.2012.10.062
-
Fig. 1. The location of the PRD region and air quality
monitoring stations.
Table 1Emission source categorization in the PRD region.
Category Sub-category
Power plantsIndustrial sourcesIndustrial solvent useOn-road
mobile sources Heavy duty gasoline passenger cars
Heavy duty diesel passenger carsLight duty gasoline passenger
carsLight duty diesel passenger carsHeavy duty gasoline trucksHeavy
duty diesel trucksLight duty gasoline trucksLight duty diesel
trucksBusesTaxies
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e2012Table
S-1 in the Supplementary Material. These measures alreadyhelped
alleviate regional air pollution problems to some extent.However,
the monitoring data indicated that primary pollutantconcentrations
still remained at high levels, and secondary ozoneand haze
pollution episodes frequently happened (GDEMC andHKEPD, 2005e2010).
Therefore, further control measures andpolicies are still needed in
order to significantly improve the airquality in the PRD
region.
Due to the dramatic economic growth, the adjustment of
energystructure, and implementation of emission control measures,
it isexpected that source characteristics in this region have
greatlychanged during the past ten years. In the meantime, the
effec-tiveness and roles of implemented control measures and
policieshave yet to be reviewed or assessed in a systematic and
scientificmanner. In order to guide future control policy
formulation, there isa need for analyzing emission trends of
primary pollutants (SO2,NOx, PM10 and VOCs) and identifying
variations in source charac-teristics in the PRD region.
The main objectives of this paper are to characterize
emissiontrends of primary pollutants from 2000 to 2009 and to
assess theimpacts of control measures on source characteristics in
the PRDregion. A dynamic methodology, by considering economic
devel-opment, technology penetration, and emission control, was
used toestimate the emissions. The reliability of this analysis was
validatedby comparing emission trends with the satellite data and
groundobservations.MotorcyclesNon-road mobile sources Marine
Agriculture machineryConstruction machineryAirportRailroad
Non-industrial solvent use Personal domestic productArchitecture
surface coating
Biomass burningResidential fuel consumption2. Data and
methods
2.1. Methods for estimating emissions
Emission sources in eight major categories and twenty five
sub-categories were considered in this study, as listed in Table 1.
Thecategorization was based on the source classification in
theGuangdong-Hong Kong air pollution emission inventory
handbook(HG-JWGSDEP, 2008) and Zheng et al. (2009). Due to the
absence ofdetailed source-based activity data from 2000 to 2009, a
top-downapproach was used in this study.
By referring to the technology-based methodology for
analyzingnational emission trends (Lu et al., 2010; Zhang et al.,
2007; Leiet al., 2011), in this study, a dynamic approach,
consideringeconomic development, technology penetration, and
emissioncontrol, was used to characterize the emission trends in
the PRD
-
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e20 13region.
The emissions were estimated for different years by theEquation (1)
below:
En Xi;k;l
Ai;k;l;nXm
Xi;k;l;m;nEFk;l;m;n
Xj
hZj1 hj
i(1)
Where i, k, l, m, n, j represent the city, emission source, the
fuel orproduct type, the technology type, the year, control
technology,respectively. E represents regional emissions of SO2,
NOx, PM10 orVOCs, A stands for the activity level (such as fuel
consumption ormaterial production), Xm is for the proportion of
fuel or productionfor a sector that is consumed or produced by
technology m. Z is theproportion of the control technology j, hj is
the removal efficiency ofcontrol technology j. EF is the emission
factor. SO2 emission factorsof fuel combustion sources can be
calculated by Equation (2):
EF 2 S 1 SR (2)
Where S and SR represent the sulfur contents and sulfur
retentionin ash, respectively. Besides, emissions from on-road
mobile sour-ces were estimated by Equation (3):
En Xi
Pi;n Mi;n EFi;n
(3)
Where, i, n represent the vehicle class, the year, respectively.
P is thevehicle population, M is the annual mileages traveled.
Detailedmethods for estimating emissions from different sources
weredescribed in previous studies (Che et al., 2009) and summarized
inTable S-2 in the Supplementary Material. Table S-2 also listed
thedata sources by sectors.2.2. Activity data and emission
factors
2.2.1. Activity data processingGenerally, estimating a long term
emission trend is much more
challenging than just developing an annual emission inventory,
dueto the limitation of the availability, consistency and accuracy
inactivity data and emission factors (Zhang et al., 2007). In this
study,we referred much on official statistics for most activity
data (e.g.,fuel consumption, vehicle population, product output, as
shown inTable S-2 in the Supplementary Material). However, these
datasources were either lack of detailed activity data for some
emissionsources or inconsistent among different years, or missing
for someyears or emission sources, therefore, surrogate data have
to be usedwith certain level of data processing. Detailed activity
data pro-cessing for power plants, industrial combustion sources
and on-road mobile sources were presented in the following
sections.
The activity data of power plants and industrial
combustionsources at city level were collected from official
statistical reports(BSPRD, 2001e2010). However, in order to reduce
the effects ofdata gaps on emission trends, we mainly considered
three majorfuel types including coal, fuel oil and natural gas
consumed inpower plants and industrial combustion sources for all
cities duringthe study period. Since electricity generation has
strong associationwith power plant fuel consumptions, we used the
electricitygeneration data to estimate the missing fuel consumption
data bytaking into account changes in fuel structures and
improvements infuel efficiency in power plants. The gross
industrial output valuesavailable in the Guangdong Statistical
Yearbook (GDPBS, 2001e2010) were used as conversion factors to
estimate the missingdata in industrial combustion sources, by
considering the variationin energy intensity.
For on-road mobile sources, 11 vehicle types were
consideredincluding eight types of passenger cars and trucks (gross
weight:heavy or light duty, fueled by diesel or gasoline), buses,
taxis andmotorcycles. However, numbers of passenger cars and trucks
weretypically collected by gross weight without differentiation of
fueltypes in current official statistical yearbooks. In view of
this, weconducted a survey and reviewed previous studies (Che et
al., 2009)to estimate the ratios of gasoline to diesel vehicles
with differentgross vehicle weights, and then calculated the
numbers of eighttypes of passenger cars and trucks based on the
above ratios.
2.2.2. Determination of emission factors and control
efficienciesEmission factors and control efficiencies are greatly
influenced
by control measures, control technologies and emission
standards.In the past ten years, great efforts have been made to
reduceprimary air pollutant emissions, such as desulfurization for
powerplants and industrial sources and upgraded emission standards
formotor vehicles. In order to reflect the possible impacts from
thesemeasures, dynamic emission factors and control efficiencies
indifferent years or cities were used to estimate the emission
trends.In the following paragraphs, we introduced the approach
todetermine emission factors and control efficiencies for three
majorsources across years, including power plants, industrial
sources andon-road mobile sources.
Emission factors of on-road mobile sources were typically
esti-mated by using mobile emission estimation models with inputs
ofvehicle technology distribution, fuel types, fuel economy
andannual mileage traveled (He et al., 2005). Due to the lack of
detailedfleet and technology information from 2000 to 2009,
emissionfactors cannot be estimated by mobile emission estimation
modelsfor each year. In this study, we utilized the International
VehicleEmission (IVE) Model (UCR, 2008) to estimate 2007-based
motorvehicle mission factors in the PRD region with the use of PRD
localemission rates, vehicle emission standards, local ambient
condi-tions and other local data. The emission factors for other
years wereestimated based upon 2007-based emission factors by
taking intoaccount the schedule of upgrading vehicle emission
standards(including National 0, I, II and III) in the PRD region,
the differencebetween vehicle emission factors under different
emission stan-dards, and the vehicle numbers by types and
years.
Emission factors and control efficiencies of power plants
andindustrial sources were traditionally determined by the fuel
prop-erty, combustion equipment and removal technology. The
regionalaverage sulfur content (S) of coal were 0.89% in 2000
(Chen, 2001)and 0.80% in 2009 (PGGDP, 2010). Since no reliable data
wereavailable, interpolation values were used to calculate the
averagesulfur content (S) in each year during 2000e2009 (Lu et al.,
2010).SO2 removal efficiencies of power plants and industrial
sourceswere derived from Equation (4), due to lack of emission
controltechnology distribution and penetration data from 2000 to
2009
hn Rn=Rn En (4)
Where R is the regional SO2 removal amount from a
specificemission source, E is the pollutant discharge amount, h is
theaverage removal efficiency of SO2 control technology, n is the
year.The related data were collected from emission source census
data,official statistical reports (EPBGDP, 2001e2009; NBSC,
2003e2009)and power plant industrial reports (CAEPI-CDDRBK, 2008,
2009,2010). PM10 removal efficiencies were collected from the
Environ-ment Statistical Bulletin of Guangdong Province (EPBGDP,
2001e2009) and national studies (Zhang et al., 2006). Annual
averageremoval efficiencies trends were shown in Figs. 4, 5 and 9
andFig. S-4 in the Supplementary material.
NOx and PM10 emission factors of power plants and
industrialsources were obtained from field investigations conducted
onmajor point emission sources in the PRD region (HKEPD, 2011)
and
-
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e2014the
emission source census data in 2007 and 2009, with consid-ering the
effect of policy implementation on emissions. The NOxand PM10
emission factors trends of power plants and industrialsources used
in this study were shown in Fig. S-1 in theSupplementary Material.
With respect to VOCs emission factors,although field investigations
of key VOCs-related industries likeprinting, wood furniture
manufacturing, shoemaking, paint andcoating manufacturing, and
others were conducted in the PRDRegion, VOCs-related control and
process technology variationsduring the study period were not
available, emission factors wereassumed to be fixed in this study.
Detailed emission factors of majoremission sources were summarized
in Tables S-3e6 in theSupplementary Material.3. Results and
discussion
3.1. Emission trends in the PRD region
SO2, NOx, PM10 and VOCs emission trends from
anthropogenicsources in the PRD region from 2000 to 2009 were shown
in Fig. 2.SO2 emissions increased rapidly from 2000 to 2005 with
totalemissions increased by 72% while the GDP increased by
120%,mainly driven by the rapid growth of fossil fuel
consumptionwithout strict SO2 control. In response to the
implementation ofSO2 control measures for power plants and
industrial sectors, SO2emissions decreased significantly after 2005
and nearly halved in2009, compared to 2005, indicating the
effectiveness of controlmeasures adopted by governments in recent
years. This trend wasbasically consistent with Chinese national SO2
trends, in which thenational SO2 emissions increased by 53% from
2000 to 2006 andbegan to decrease after 2006 (Lu et al., 2010).
Emission trends of NOx and PM10 exhibited similar patterns
asshown in Fig. 2. During 2000e2009, the GDP in the PRD
regionincreased by 282%, while NOx and PM10 emissions increased
by96% and 66%, respectively. NOx emissions in the PRD region kepta
consistent growth from 2000 to 2009, with lower growth ratesin
recent years, while PM10 emissions kept increasing from 2000to 2007
with annual growth rates from 2% (in 2006) to 13% (in2005), but
slightly declined after 2007. The reason for differencesin growth
rates between emissions and GDP may be the imple-mentation of a
series of control measures on vehicles, powerplants and industrial
boilers by national and local governmentagencies in recent years,
and the increased GDP contributions ofnon-production sector, from
44% in 2001 to 50% in 2009 (GDPBS,0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
val
ue
GDP Fuel consumption SO2 NOx PM VOCs
Fig. 2. Trends in pollutant emissions, GDP and fuel consumption
(All data arenormalized to the year 2000).2001e2010). In comparison
with national NOx and PM10 emissiontrends, similar upward trend was
also identified in the nationalNOx emissions with an increase of
55% during 2001e2006 (Zhanget al., 2009), while the growth rates of
PM10 emissions in the PRDregion increased faster than that in the
national trend during2000e2005 (Zhang et al., 2009), probably due
to the more rapidlyincreasing GDP in this region.
VOCs emissions remained steadily increasing with annualgrowth
rates ranging from 2% to 10%. The stable growth can beattributed to
the significant increase of vehicle population and theincreasing
use of industrial solvent arising from rapid economicdevelopment.
Compared to other three primary pollutants, fluc-tuations of VOCs
emission growth rates were much smaller. Thiswas probably because
most of VOCs emission control measures orpolicies were targeted on
vehicle source with less focus on otheremission sources.
Additionally, the VOCs emission trend in the PRDregionwas similar
to the national onewith an upward trend (Zhanget al., 2009).3.2.
Variations in source characteristics
In this section, we discussed variations in source
characteristicsof SO2, NOx, PM10 and VOCs during the past decade
and identifiedpossible impacts of policies and control measures on
source char-acteristics. Besides, differences in source
contributions between thePRD region and the whole China were
analyzed.
3.2.1. SO2 emissionsFig. 3 presented variations in SO2 source
contributions from
2000 to 2009. Apparently, although power plants and
industrialsources showed declining trends in contributions, they
were stillmajor contributors, accounting for 92% in 2000 and 82% in
2009. Inaddition, the contribution of non-road mobile sources
showed anupward trend, accounting for 6% in 2000 and 15% in 2009 of
PRDregional SO2 emissions. The PRD regional SO2 source
characteristicswere similar to national ones, in which power plants
and industrialsources were major contributors with around 90% of
total nationalSO2 emissions since 2000 (Lu et al., 2010).
Figs. 4 and 5 showed the SO2 emission trends and related datafor
power plants and industrial sources. Generally, SO2 emissions
ofboth power plants and industrial sources shared similar
upwardtrends from 2000 to 2005, together with energy
consumption,electricity generation and the number of large
industrial enter-prises. However, although electricity generation
and the number of0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
Power plants Industrial sources
On-road mobile sources Non-road mobile sources
Biomass burning Residential fuel consumption
Fig. 3. Contribution trends of SO2 by categories (All data are
normalized to the year2000).
-
0
10
20
30
40
50
60
70
80
0.0
0.5
1.0
1.5
2.0
2.5
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
SO2
rem
oval
eff
icie
ncie
s (%
)
Nor
mal
ized
val
ue
SO emission Energy consumption
Electricity generation SO removal efficiencies
Fig. 4. Trends in SO2 emission from power plants and related
activity data (All dataexcept SO2 removal efficiencies are
normalized to the year 2000).
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
Marine Airport
Agriculture machinery Consturction machinery
Railroad Freight transfer volumes in ports
Fig. 6. Trends in SO2 emission from marine (non-road) source and
related activity data(All data are normalized to the year
2000).
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e20 15large
industrial enterprises kept increasing after 2005, the
energyconsumption clearly dropped after 2007, and SO2 emissions
fromboth power plants and industrial sources decreased
significantly.Installing and operating flue gas desulfurization
(FGD) facilities inpower plants and large industrial boilers were
main reasons for therapid decrease of SO2 emissions. Such
decreasing may also beattributed to the implementation of control
measures like shuttingdown of small and high-emitting power
generation units andindustrial boilers, limiting fuel sulfur
contents and increasing theproportion of clean energy consumption
(PGGDP, 2004). As shownin Figs. 4 and 5, there were higher removal
efficiencies in powerplants than industrial source, this was mainly
because FGD deviceswere required to install in all power plants
with strict supervision,while only required in larger industrial
boilers for industrial sour-ces, which led to relatively low
removal efficiencies for industrialsources on average.
Fig. 6 showed SO2 emission contributions and related
activitydata trends for non-road mobile sources. Among non-road
mobilesources, marine was the largest SO2 contributor, accounting
for 8%of total emissions on average and the contribution increased
by 12%per year. This was similar to the trend of freight transfer
volumes inports, arising from well-developed river systems and the
rapid0
10
20
30
40
50
60
70
0.0
0.5
1.0
1.5
2.0
2.5
3.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
SO2
rem
oval
eff
icie
ncie
s (%
)
Nor
mal
ized
val
ue
SO emission Energy consumption
Large industrial enterprises SO removal efficiencies
Fig. 5. Trends in SO2 emission from industrial sources and
related activity data (Alldata except SO2 removal efficiencies are
normalized to the year 2000). Note: Largeindustrial enterprises
refers to those with the annual main business income over 20million
RMB.development of waterway and marine transportation industries
inthis region. However, limited control measures were targeted
onnon-road mobile sources in this region at present.
3.2.2. NOx emissionsFig. 7 showed variations of NOx source
contributions from 2000
to 2009. Obviously, on-road mobile source was the
largestcontributor, although its contribution presented a
decreasing trend,from 41% in 2000 to 38% in 2009. In addition,
power plants andindustrial sources also made great contributions to
NOx emissions.Contributions of power plants fluctuated around
24e28% from2000 to 2009 while contributions of industrial sources
increasedslightly, from 17% in 2000 and 20% in 2009. The most
significantincrease was the non-road mobile source with doubled
emissionsand its contribution reaching 14% in 2009. In comparison
withnational NOx source characteristics (Ohara et al., 2007), the
largestNOx contributor was the on-road mobile source in the PRD
region,while nationally it was power plant source.
Fig. 8 showed trends in NOx emissions from on-road mobilesources
and vehicle population from 2000 to 2009. The vehiclepopulation
increased much faster than NOx emissions after 2005,0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
Power plants Industrial sources
On-road mobile sources Non-road mobile sources
Biomass burning Residential fuel consumption
Fig. 7. Contribution trends of NOx by categories (All data are
normalized to the year2000).
-
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
0.5
1.0
1.5
2.0
2.5
3.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
veh
icle
pop
ulat
ion
Nor
mal
ized
em
issi
on
Number of passenger cars Number of trucks NOx emission
Fig. 8. Trends in NOx emission from on-road mobile sources and
vehicle population(All data are normalized to the year 2000).
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
Marine Airport
Agriculture machinery Consturction machinery
Railroad Freight transfer volumes in ports
Fig. 10. Trends in NOx emission from marine (non-road) source
and related activitydata (All data are normalized to the year
2000).
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e2016due to
upgraded motor vehicle emission standards, such as theNational II
implemented in 2005 and National III implemented in2008 (PGGDP,
2009b). Additionally, measures like phase-out ofhigh-emitting
vehicles together with the popularization of cleanfuel vehicles
such as gas vehicles, hybrid vehicles and electricvehicles (PGGDP,
2004) have also contributed to the reduced NOxemission growth rates
under the pressure of rapid vehicle pop-ulation growth.
Fig. 9 showed the relationship between NOx emission trends
andrelated data of power plants from 2000 to 2009. There were
similarincreasing trends in NOx emissions from power plants to
those inenergy consumption and electricity generation before
2005.However, both NOx emissions and energy consumption dropped
in2006 while they resumed to decline after 2007. This may
beprobably because the deadline of shutting down small-scalethermal
power units was at the end of year 2005 (PGGDP, 2004),resulting in
the improvement of fuel efficiency and the reduced fuelconsumption
and NOx emission in 2006. Similar phenomenon wasfound in industrial
sources and details were provided in theSupplementary Material
(Fig. S-4).
Except for three largest contributors, non-road mobile sourcewas
becoming an important NOx contributor in the PRD regionduring the
study period. Among non-road mobile sources, asshown in Fig. 10,
marine was the major contributor, accounting for69% of total
non-road mobile sources emission on average with an20
30
40
50
60
70
80
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
PM
10re
mov
al e
ffic
ienc
ies
(%)
Nor
mal
ized
val
ue
NOx emission Energy consumption
Electricity generation PM emission
PM removal efficiencies
Fig. 9. Trends in NOx and PM10 emissions from power plants and
related activity data(All data except PM10 removal efficiencies are
normalized to the year 2000).increasing trend from 62% to 72%,
indicating marine source wasbecoming one of major NOx emission
sources in this region.
3.2.3. PM10 emissionsFig. 11 showed variations in PM10 source
contributions from
2000 to 2009. Power plants and industrial sources were
majorcontributors, with average contributions of 37% and 32%,
respec-tively. The contribution of industrial sources presented a
relativelyrapid growing trend (29% in 2000 and 36% in 2009), while
thecontribution of power plants kept relatively steady (34% in
2000and 38% in 2009). Besides, biomass burning and on-road
mobilesources were important PM10 contributors. There were
similarPM10 source characteristics in the PRD region, in comparison
withnational ones, in which power plants and industrial
sourcescontributed nearly 70% of total national PM10 emissions in
2000and 2005 (Lei et al., 2011).
Fig. 9 showed the relationship between PM10 emission trendsand
related data of power plants from 2000 to 2009. PM10 emis-sions
from power plants presented similar increasing trends withthose in
energy consumption and electricity generation before2005. However,
both PM10 emissions and energy consumptiondropped in 2006 while
they resumed to decline after 2007. This canbe attributed to the
measures of closing small and high-emit power0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
Power plants Industrial sources
On-road mobile sources Non-road mobile sources
Biomass burning Residential fuel consumption
Fig. 11. Contribution trends of PM10 by categories (All data are
normalized to the year2000).
-
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
Power plants Industrial sourcesIndustrial solvent use On-road
mobile sourcesNon-road mobile sources Non-industrial solvent
useBiomass burning Residential fuel consumption
Fig. 13. Contribution trends of VOCs by categories (All data are
normalized to the year2000).
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e20
17generation units, restricting coal-fired power plants, and
encour-aging the use of hydro-electric power and natural gas
(PGGDP,2004, 2009a). Similar trends were found in industrial
sources anddetails were provided in the Supplementary Material
(Fig. S-4). Itwas noteworthy that the emissions of PM10 did not
decline signif-icantly as the one of SO2 did. Generally, the two
trends weresupposed to be consistent since the installation of
desulphurizationsystem may lead to enhanced PM removal efficiency
(Zhao et al.,2008). In our case, the slight inconsistency might be
attributed tothe following reasons: (1) since the real removal
efficiencies weredetermined by not only the technology itself but
also the operatingconditions and managements (Zhang, 2005), the
removal efficien-cies of PM10 were lower than that of SO2 control,
partly due to lackof strict supervision and management since
currently PM10 emis-sion has not been taken into national
evaluation index system yet;(2) the wide use of low sulfur coals in
power plants and majorindustrial sectors in recent years also led
to the large amount of SO2emission reductions. The monitoring data
showed that thedecreasing rates of SO2 and PM10 concentrations from
2005 to 2009were 19% and 9% respectively (GDEMC and HKEPD,
2005e2010),indicating the PM10 emission trend developed in this
study wasreasonable, though there was inevitable uncertainty.
Fig. 12 showed trends in PM10 emission from the biomassburning
source from 2000 to 2009. Biomass burning includeddomestic biofuel
combustion, field burning of crop residues andforest fire (He et
al., 2011). The PM10 emission from biomassburning showed a
relatively steady trend, but its contributionsdecreased rapidly,
from 23% in 2000 to 13% in 2009.
It must be pointed out that road dust and construction
sourceswere important PM10 emission contributors in the PRD
region.However, due to the lack of detailed activity data and local
emissionfactors, analysis of PM10 emission trends from road dust
andconstruction sources were not made in this study. Further
investi-gations and studieswere needed for these two sources in the
future.
3.2.4. VOCs emissionsFig. 13 showed variations in VOCs source
contributions from
2000 to 2009. Obviously, the on-road mobile source was the
largestVOCs contributor, accounting for 57% of PRD regional VOCs
emis-sions on average with a slight declining trend from 58% to
53%,followed by the industrial solvent use source which accounted
for24% of regional total emissions on average, with an increasing
trendfrom 18% to 33% during the ten years. VOCs emission
contributionsof both non-industrial solvent use and biomass burning
sources0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
Domestic biofuel combustion Field burning of crop residues
Forest fire
Fig. 12. Trends in PM10 emission from biomass burning (All data
are normalized to theyear 2000).presented declining trends, from
10% to 7% and 12% to 5% respec-tively; while contributions from
power plants, non-road mobilesources and industrial sources kept
quite stable during the studyperiod.
Fig. 14 showed VOCs emission contributions from on-roadmobile
sources by vehicle types from 2000 to 2009. Apparently,motorcycles
and passenger cars were major vehicle types for VOCsemissions,
accounting for 56% and 29% of regional on-road mobilesources
emission on average, respectively. The number of motor-cycles
decreased by 7% from 2005 to 2009, due to the restriction
orprohibition on motorcycles within urban areas in most cities of
thePRD region (Che et al., 2011), with its contributions decreasing
from57% in 2005 to 45% in 2009. The number of passenger car
increasedby 530% during the ten years, while its emission
contribution justincreased from 17% in 2000 to 41% in 2009, due to
the imple-mentation of more strict vehicle emission standards.
As shown in Fig. 13, the VOCs emission from the
industrialsolvent use source has tripled during the ten years.
Although someVOCs-related industrial sectors were equipped with
VOCs gath-ering and treatment devices (e.g., activated carbon/water
adsorp-tion, catalytic combustion), the removal efficiency still
remained at0
1
2
3
4
5
6
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
veh
icle
pop
ulat
ion
Nor
mal
ized
em
issi
on
Motorcycles Passenger carsTrucks BusesTaxies Number of
motorcyclesNumber of passenger cars
Fig. 14. Trends in VOCs emission from on-road mobile sources and
vehicle population(All data are normalized to the year 2000).
-
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e2018a low
level at present due to the limitation of control
technology,ineffective operation, or less supervision.3.3.
Comparison with satellite data and ground observations
In order to validate the reliability of emission trend analysis,
NOxand PM10 emissions trends were compared with satellite data,
andSO2, NOx and PM10 emission trends were compared with
groundobservations, depending on data availability. Since the
PRDRegional Air Quality Monitoring Network was established in
2005,and started to operate in the next half year in 2005 (Zheng et
al.,2010), in order to keep the consistency of ground
observationdata used for comparison and to make the comparison
scientificallysound, in this study, we only utilized 2005e2009
ground obser-vations for comparisons. The NO2 column concentrations
used inthis study were data products by the Institute of
EnvironmentalPhysics (IUP), University of Bremen, from GOME
(2000e2002) andSCIAMACHY (2003e2009) satellites, using the
Differential OpticalAbsorption Spectroscopy-Method (DOAS) (Richter
et al., 2005). Thespatial resolutions of GOME and SCIAMACHY are 0.5
0.5 and0.125 0.125, respectively. The dry surface extinction
coefficient(SECdry) data with 2-km resolution were used to compare
withPM10 emission trends in this study. The SECdry data were
retrievedfromMODIS aerosol optical depth (AOD) data acquired
fromNASAsGoddard Earth Sciences Distributed Active Archive Center,
usingthe aerosol retrieval algorithms developed by Li et al.
(2003), withvertical distribution correction and relative humidity
correction (Liet al., 2005). The domain for retrieved satellite
datawas the same asthe emission domain used in this study.
Fig. 15 showed trends in NOx emissions and satellite-based
NO2column concentrations from 2000 to 2009. Generally, trends in
NOxemissions and satellite-based NO2 column concentrations
pre-sented broad agreement in temporal evolution. Both
presentedcontinuous growths (from 2000 to 2004, increased by 55%
and 82%,respectively; from 2005 to 2007, increased by 13% and 9%,
respec-tively) except that emission in 2008 showed a slight drop.
Thegrowth rates of emissions were lower than those of
satelliteobservations during 2000e2004, similar to the situation
over China(Zhang et al., 2007). However, the large discrepancy
betweenemission trends and satellite observations in 2004e2005 can
beattributed to the uncertainty in emission estimates, variability
inmeteorology, NOx injection height, and the increasing trend
ofsulfate aerosols (Zhang et al., 2007). As shown in Fig. 15, there
wasa good agreement between NOx emissions and ground
observations0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
NOx emissions Satellite-based NO columns
NOx ground observations
Fig. 15. Trends in NOx emissions, satellite-based NO2 column
concentrations, andground NO2 concentrations (All data are
normalized).during 2005e2009, indicating the reasonability of NOx
emissiontrend developed in this study.
Fig. 16 showed trends in PM10 emissions and
satellite-basedSECdry values from 2000 to 2009. Basically, PM10
emissions andsatellite-based SECdry values presented similar
patterns in temporalevolution, both of which increased at almost
the same rates from2000 to 2004 except that the SECdry value
dropped in 2002.Another discrepancy occurred during 2004e2006 when
PM10emissions continued to grow but the satellite data started
todecline slightly. Both emissions and the SECdry values
presenteddownward trends from 2007 to 2009. In comparison, a
generaldownward trend in PM10 emissions but an upward trend in
AODwas observed over the whole China in 2004e2009 (Lin et
al.,2010). The complex relationship between PM10 emissions and
thesatellite data was probably because the AOD generally has a
closerassociation with PM2.5 than with PM10 (Lin et al., 2010).
Besides,there was a good agreement between PM10 emission trend
andground observations during 2005e2009. These results
indicatedthat PM10 emission trend developed in this study was
reasonable,to some extent.
Fig. 17 showed trends in SO2 emission from 2000 to 2009
andground observations from 2005 to 2009. There was a very
goodagreement between emission and ground observations, bothshowing
significant downward trends after 2005. The big discrep-ancy in
2005 might be attributed to the fact that the PRD
regionalmonitoring network started the trial operation in the next
half yearof 2005, which may not be able to represent regional
averaged SO2concentrations over the whole year.
3.4. Implication for air pollution control policy
The implementation of control policies andmeasures did reduceSO2
and PM10 emissions in the PRD region, as shown in this
study.However, concentrations of primary and secondary pollutants
stillremain at high levels at present, indicating the importance
ofimplementing further control measures in this region. In
thissection, policy implications for future air pollution control
werediscussed based upon the identification of emission trends and
thecharacterization of source contributions from this study.
Although significant reductions in SO2 emissions were made inthe
past decade, strict control and management of SO2 emissions isstill
needed (Xu et al., 2009; Xu, 2011) for further decreasing
SO2concentrations and effectively controlling fine particulate
pollu-tions, a commonly concerned air pollution issue in China.
This0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
PM10 emissions Satellite-based SECdry values
PM10 ground observations
Fig. 16. Trends in PM10 emissions, satellite-based SECdry
values, and ground PM10concentrations (All data are
normalized).
-
0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Nor
mal
ized
em
issi
on
SO emissions SO ground observations
Fig. 17. Trends in SO2 emissions and ground SO2 concentrations
(All data arenormalized).
Q. Lu et al. / Atmospheric Environment 76 (2013) 11e20 19study
showed the effectiveness of current SO2 control measures onpower
plants and industrial sources. These measures should befurther
promoted under strict supervision and be targeted on othersources.
Specially, marine sources became the third SO2 contributorwith an
increasing trend but fewmeasures have been implementedin the past
decade, lowering sulfur contents in fuel of ships orvessels or
installing SO2 control devices is one of top priorities forreducing
marine SO2 emissions in the next few years.
Upgrading motor vehicle emission standards and phasing
outhigh-emitting vehicles alleviated the growth of NOx emissions in
thePRD region, to some extent. However, this study showed that
NOxemission still presented an upward trend from 2000 to
2009.National Ministry of Environmental Protection (MEP) listed
NOxcontrol as a top priority during the national Twelfth-Five plan.
Interms ofNOx source characteristics in thePRD region from this
study,priority measures for reducing NOx emissions in this region
mayinclude the wide use of low NOx burners and flue gas
denitrificationtechnology on power plants and industrial sources,
furtherupgrading vehicle emission standards, control of non-road
mobilesources especially the marine source. It must be pointed out
thatcautions need to be taken in reducing NOx emissions since there
arecomplicated non-linear relationships in NOx/VOC ratios for
ozoneformation in the PRD region (Zhang et al., 2008). Previous
studies(Wanget al., 2009b; Zhanget al., 2008) showed that ozone
formationwas typically under VOC-limited regime in urban areas of
the PRDregion, while most likely under NOx-limited regime in
surroundingrural areas. This implies that reducing NOx emissions
withoutsimultaneously controlling VOCs emissions with proper
ratios, viceversa, may lead to elevated ozone concentrations in
this region.
Although power plants and large-scale industrial boilers
havebeen equipped with particulate matter control devices in
recentyears, the regional PM10 emission still remained at
relatively highlevel. Such control devices should be further
installed inmiddle andsmall-scale industrial sources. Biomass
burning and heavy-dutydiesel vehicles are important PM10 emission
contributors in thisregion, more strict control measures should be
taken on thesesources. Road and construction dust sources made
large contribu-tions to PM10 emission in the PRD region, but have
not beeneffectively controlled. Future work should enhance the
control ofdust sources and extend target areas to both urban and
ruralregions.
VOCs emissions kept consistently increasing from 2000 to 2009in
the PRD region though restrictingmotorcycles in urban areas
andupgrading motor vehicle emission standards helped reduce
VOCsemissions from vehicles, to some extent. VOCs is an
importantprecursor for formations of ozone and secondary organic
aerosol(SOA). Thus, controlling VOCs emissions is critical to
significantlyalleviate ozone and fine particulate pollutions in
this region. Basedupon this study, VOCs emissions from industrial
solvent use sourcegrew rapidly, implying that control of VOCs
emissions from thissource in the PRD region is of great importance
in the future. Inaddition, control of vehicle emissions should
still be on the toppriority in reducing VOCs emissions in this
region, since vehiclesource is expected to be still the largest
VOCs contributor withrapidly increasing vehicle population in the
next five or ten years.
4. Summary and conclusions
Emission trends and variations in source contributions of
SO2,NOx, PM10 and VOCs in the PRD region from 2000 to 2009
werecharacterized by using a dynamic approach. The emission
trendresults showed that SO2 emissions increased rapidly during
2000e2005 but decreased significantly afterward. NOx emissions went
upconsistently during 2000e2009 except for a break point in
2008.PM10 emissions increased by 76% during 2000e2007 but started
todecrease slightly in the following years, and VOCs emissions
pre-sented a continuous increase during the study period. The
sourcecharacterization results showed that power plants and
industrialsources were consistently the largest contributors to SO2
and PM10emissions, the on-roadmobile sourcewas the largest NOx and
VOCscontributor, and industrial solvent use source was becoming
animportant VOCs emission source. Worthy of attention is that
non-road mobile sources gradually become important SO2 and
NOxemission contributors in the PRD region, which need
immediatecontrol actions on them.
In order to validate the emission trends, comparisons
withsatellite data and ground observations were made, where
appli-cable. The results showed that emission trends presented
broadagreements with the satellite data and ground observations,
indi-cating that emission trendsdeveloped in this studywere
reasonable.
Acknowledgments
This work was supported by National Natural Science Founda-tion
of China-Guangdong (NSFC-GD) Key Project (U1033001),International
Technology Cooperation Plan of Guangdong Province(2011B050300006)
and The Fundamental Research Funds for theCentral Universities,
South China University of Technology(2011ZZ0009).
Appendix A. Supplementary material
Supplementary material related to this article can be
foundonline at
http://dx.doi.org/10.1016/j.atmosenv.2012.10.062.
References
Bureau of Statistics of
Guangzhou/Shenzhen/Zhuhai/Foshan/Jiangmen/Dongguan/Zhongshan/Huizhou/Zhaoqing,
Guangdong Province (Ed.) (BSPRD),
2001e2010.Guangzhou/Shenzhen/Zhuhai/Foshan/Jiangmen/Dongguan/Zhongshan/Huiz-hou/Zhaoqing
Statistical Yearbook 2001e2010. China Statistics Press, Beijing
(inChinese).
Che, W.W., Zheng, J.Y., Zhong, L.J., 2009. Mobile source
emission characteristics andcontributions in the Pearl River Delta
region. Research of Environment Sciences22 (4), 456e461 (in
Chinese).
Che, W.W., Zheng, J.Y., Wang, S.S., Zhong, L.J., Lau, A., 2011.
Assessment of motorvehicle emission control policies using
Model-3/CMAQ model for the PearlRiver Delta region, China.
Atmospheric Environment 45 (9), 1740e1751.
Chen, P., 2001. China Coal Properties, Classification and
Utilization. ChemicalIndustry Press, Beijing (in Chinese).
Committee of Desulfurization and Dust Removal of Boiler Kiln of
China Associationof Environmental Protection Industry
(CAEPI-CDDRBK), 2008. An overview ofthe development in power plants
desulfurization and denitrification industryin 2007. China
Environmental Protection Industry 6, 18e21 (in Chinese).
http://dx.doi.org/10.1016/j.atmosenv.2012.10.062
-
Q. Lu et al. / Atmospheric Environment 76 (2013)
11e2020Committee of Desulfurization and Dust Removal of Boiler Kiln
of China Associationof Environmental Protection Industry
(CAEPI-CDDRBK), 2009. An overview ofthe development in power plants
desulfurization and denitrification industryin 2008. China
Environmental Protection Industry 7, 8e19 (in Chinese).
Committee of Desulfurization and Dust Removal of Boiler Kiln of
China Associationof Environmental Protection Industry
(CAEPI-CDDRBK), 2010. An overview ofthe development in power plants
desulfurization and denitrification industryin 2009. China
Environmental Protection Industry 6, 17e20 (in Chinese).
Deng, X.J., Tie, X.X., Wu, D., Zhou, X.J., Bi, X.Y., Tan, H.B.,
Li, F., Jiang, C.L., 2008. Long-term trend of visibility and its
characterizations in the Pearl River Delta (PRD)region, China.
Atmospheric Environment 42 (7), 1424e1435.
Environmental Protection Bureau of Guangdong Province (EPBGDP),
2001e2009.Environment Statistical Bulletin of Guangdong Province
(2001e2009). Availableat: http://www.gdep.gov.cn/zlkz/
(2012/02/15).
Guangdong Provincial Bureau of Statistic (Ed.) (GDPBS),
2001e2010. GuangdongStatistical Yearbook 2001e2010. China
Statistics Press, Beijing (in Chinese).
Guangdong Provincial Environmental Protection Monitoring Centre
(GDEMC), Envi-ronmental Protection Department of Hong Kong (HKEPD),
2005e2010. PearlRiver Delta Regional Air Quality Monitoring Network
Report (in Chinese). Avail-able at:
http://www.gdep.gov.cn/hjjce/hjxt/201010/t20101021_114994.html(2012/04/09).
He, K.B., Huo, H., Zhang, Q., He, D.Q., An, F., Wang, M., Walsh,
M.P., 2005. Oilconsumption and CO2 emissions in Chinas road
transport: current status,future trends, and policy implications.
Energy Policy 33 (12), 1499e1507.
He, M., Zheng, J.Y., Yin, S.S., Zhang, Y.Y., 2011. Trends,
temporal and spatial char-acteristics, and uncertainty in biomass
burning emissions in the Pearl RiverDelta, China. Atmospheric
Environment 45 (24), 4051e4059.
Hong Kong Environmental Protection Department (HKEPD), 2011.
Reports ofFeasibility Study on Major Industrial Air Pollution
Sources in the Pearl RiverDelta Region (Report).
Hong Kong-Guangdong Joint Working Group on Sustainable
Development andEnvironmental Protection (HG-JWGSDEP), 2008. Pearl
River Delta Regional AirQuality Management Plan Mid-term Review
Report.
http://www.legco.gov.hk/yr07-08/chinese/panels/ea/papers/ea0128cb1-666-4-ec.pdf.
Lei, Y., Zhang, Q., He, K.B., Streets, D.G., 2011. Primary
anthropogenic aerosolemission trends for China, 1990e2005.
Atmospheric Chemistry and Physics 11,931e954.
Li, C.C., Mao, J.T., Lau, K.H.A., Chen, J.C., Yuan, Z.B., Liu,
X.Y., Zhu, A.H., Liu, G.Q., 2003.Characteristics of distribution
and seasonal variation of aerosol optical depth
ineasternChinawithMODISProducts.Chinese ScienceBulletin48 (22),
2488e2495.
Li, C.C., Mao, J.T., Lau, K.H.A., Yuan, Z.B., Wang, M.H., Liu,
X.Y., 2005. Application ofMODIS satellite products to the air
pollution research in Beijing. Science inChina Series D Earth
Sciences 48 (Suppl. II), 209e219.
Lin, J.T., Nielsen, C.P., Zhao, Y., Lei, Y., Liu, Y., Mcelroy,
M.B., 2010. Recent changes inparticulate air pollution over China
observed from space and the ground:effectiveness of emission
control. Environmental Science & Technology 44,7771e7776.
Lu, Z., Streets, D.G., Zhang, Q.,Wang, S., Carmichael, G.R.,
Cheng, Y.F., Wei, C., Chin, M.,Diehl, T., Tan, Q., 2010. Sulfur
dioxide emissions in China and sulfur trends in EastAsia since
2000. Atmospheric Chemistry and Physics 10, 6311e6331.
National Bureau of Statistics of China (Ed.) (NBSC), 2003e2009.
Yangtze River Delta& Pearl River Delta and Hong Kong &
Macao Special Administrative Region andTaiwan Statistical Yearbook
2003e2009. China Statistics Press, Beijing (inChinese).
Ohara, T., Akimoto, H., Kurokawa, J., Horii, N., Yamaji, K.,
Yan, X., Hayasaka, T., 2007.An Asian emission inventory of
anthropogenic emission sources for the period1980e2020. Atmospheric
Chemistry and Physics 7, 4419e4444.
Peoples Government of Guangdong Province (PGGDP), 2004. Outline
Plan of theEnvironmental Protection in the Pearl River Delta Region
During 2004e2020.Available at:
http://www.gzepb.gov.cn/zwgk/gs/fzgh/200810/t20081010_54234.htm
(2011/10/11).
Peoples Government of Guangdong Province (PGGDP), 2009a.
Prevention andControl Measures of Air Pollution in the Pearl River
Delta Region of GuangdongProvince. Available at:
http://www.gdep.gov.cn/zcfg/zxfg/201008/t20100809_82410.html
(2011/04/09).
Peoples Government of Guangdong Province (PGGDP), 2009b. The
ImplementationPlan of Vehicle Exhaust Pollution Control in
Guangdong Province. Available
at:http://www.gdepb.gov.cn/flfg/dfxfg/201007/t20100713_76038.html
(2011/10/20).
Peoples Government of Guangdong Province (PGGDP), 2010. The
Clean Air ActionPlan of the Pearl River Delta Region of Guangdong
Province. Available at:
http://www.gdepb.gov.cn/wrfz/dqfz/xgwj/201003/P020100308608490907835.pdf(2011/02/03).
Richter, A., Burrows, J.P., N, H., Granier, C., Niemeier, U.,
2005. Increase intropospheric nitrogen dioxide over China observed
from space. Nature 437,129e132.
University of California at Riverside (UCR), 2008. IVE Model
Users Manual Version2.0. Available at: http://www.issrc.org/ive/
(2012/6/15).
Wang, T., Wei, X.L., Ding, A.J., Poon, C.N., Lam, K.S., Li,
Y.S., Chan, L.Y., Anson, M.,2009a. Increasing surface ozone
concentrations in the background atmosphereof southern China,
1994e2007. Atmospheric Chemistry and Physics Discussions9 (16),
10429e10455.
Wang, X., Zhang, Y., Hu, Y., Zhou, W., Lu, K., Zhong, L., Zeng,
L., Shao, M., Hu, M.,Russell, A.G., 2009b. Process analysis and
sensitivity study of regional ozoneformation over the Pearl River
Delta, China, during the PRIDE-PRD2004campaign using the CMAQ
model. Atmospheric Chemistry and Physics 9,26833e26880.
Xu, Y., Williams, H.R., Socolow, H.R., 2009. Chinas rapid
deployment of SO2scrubbers. Energy & Environmental Science 2,
459e465.
Xu, Y., 2011. Improvement in the operation of SO2 scrubbers in
Chinas coal powerplants. Environmental Science & Technology 45,
380e385.
Zhang, Q., Klimont, Z., Street, D.G., Huo, H., He, K.B., 2006.
Development ofanthropogenic emissions models and emission inventory
in 2001 of particulatematters in China. Progress in Natural Science
16 (2), 223e231 (in Chinese).
Zhang, Q., Streets, D.G., He, K.B., Wang, Y.X., Richter, A.,
Burrows, J.P., Uno, I.,Jang, C.J., Chen, D., Yao, Z.L., Lei, Y.,
2007. NOx emission trend for China, 1995e2004: the view from the
ground and the view from space. Journal ofGeophysical Research-Part
D-Atmospheres 112, D22306.
Zhang, Y.H., Su, H., Zhong, L.J., Cheng, Y.F., Zeng, L.M., Wang,
X.S., Xiang, Y.R.,Wang, J.L., Gao, D.F., Shao, M., Fan, S.J., Liu,
S.C., 2008. Regional ozone pollutionand observation-based approach
for analyzing ozoneeprecursor relationshipduring the PRIDE-PRD2004
campaign. Atmospheric Environment 42 (25),6203e6218.
Zhang, Q., Streets, D.G., Carmichael, G.R., He, K.B., Huo, H.,
Kannari, A., Klimont, Z.,Park, I.S., Reddy, S., Fu, J.S., Chen, D.,
Duan, L., Lei, Y., Wang, L.T., Yao, Z.L., 2009.Asian emissions in
2006 for the NASA INTEX-B mission. Atmospheric Chemistryand Physics
9, 5131e5153.
Zhang, Q., 2005. Study on Regional Fine PM Emissions and
Modeling in China. Ph.D.thesis, Tsinghua University, China, Beijing
(in Chinese).
Zhao, Y., Wang, S.X., Duan, L., Lei, Y., Cao, P.F., Hao, J.M.,
2008. Primary air pollutantemissions of coal-fired power plants in
China: current status and futureprediction. Atmospheric Environment
42 (36), 8442e8452.
Zheng, J.Y., Zhang, L.J., Che, W.W., Zheng, Z.Y., Yin, S.S.,
2009. A highly resolvedtemporal and spatial air pollutant emission
inventory for the Pearl River Deltaregion, China and its
uncertainty assessment. Atmospheric Environment 43(32),
5112e5122.
Zheng, J.Y., Zhong, L.J., Wang, T., Louie, P.K.K., Li, Z.C.,
2010. Ground-level ozone in thePearl River Delta region: analysis
of data from a recently established regional airquality monitoring
network. Atmospheric Environment 44 (6), 814e823.
http://www.gdep.gov.cn/zlkz/http://www.gdep.gov.cn/hjjce/hjxt/201010/t20101021_114994.htmlhttp://www.legco.gov.hk/yr07-08/chinese/panels/ea/papers/ea0128cb1-666-4-ec.pdfhttp://www.legco.gov.hk/yr07-08/chinese/panels/ea/papers/ea0128cb1-666-4-ec.pdfhttp://www.gzepb.gov.cn/zwgk/gs/fzgh/200810/t20081010_54234.htmhttp://www.gzepb.gov.cn/zwgk/gs/fzgh/200810/t20081010_54234.htmhttp://www.gdep.gov.cn/zcfg/zxfg/201008/t20100809_82410.htmlhttp://www.gdep.gov.cn/zcfg/zxfg/201008/t20100809_82410.htmlhttp://www.gdepb.gov.cn/flfg/dfxfg/201007/t20100713_76038.htmlhttp://www.gdepb.gov.cn/wrfz/dqfz/xgwj/201003/P020100308608490907835.pdfhttp://www.gdepb.gov.cn/wrfz/dqfz/xgwj/201003/P020100308608490907835.pdfhttp://www.issrc.org/ive/
Emission trends and source characteristics of SO2, NOx, PM10 and
VOCs in the Pearl River Delta region from 2000 to 20091.
Introduction2. Data and methods2.1. Methods for estimating
emissions2.2. Activity data and emission factors2.2.1. Activity
data processing2.2.2. Determination of emission factors and control
efficiencies
3. Results and discussion3.1. Emission trends in the PRD
region3.2. Variations in source characteristics3.2.1. SO2
emissions3.2.2. NOx emissions3.2.3. PM10 emissions3.2.4. VOCs
emissions
3.3. Comparison with satellite data and ground observations3.4.
Implication for air pollution control policy
4. Summary and conclusionsAcknowledgmentsAppendix A.
Supplementary materialReferences