Research Article Source Allocation of Long-Range Asian ...downloads.hindawi.com/journals/amete/2014/268037.pdf · Backward Trajectory Calculation. In order to trace the transportation

Research ArticleSource Allocation of Long-Range Asian DustsTransportation across the Taiwan Strait by InnovativeChemical-Assisted Identification Methods

Yi-Hsiu Jen1 Yi-Chi Liu1 Iau-Ren Ie1 Chung-Shin Yuan1 and Chung-Hsuang Hung2

1 Institute of Environmental Engineering National Sun Yat-Sen University Kaohsiung 80424 Taiwan2Department of Safety Health and Environmental Engineering National Kaohsiung First University Kaohsiung 824 Taiwan

Correspondence should be addressed to Chung-Shin Yuan ycsngimailnsysuedutw

Received 13 June 2014 Accepted 8 August 2014 Published 27 August 2014

Academic Editor Hesham El-Askary

Copyright copy 2014 Yi-Hsiu Jen et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

This study used the backward trajectory calculation to obtain the transportation routes of Asian dusts and further combined thechemical composition with the enrichment factor (EF) and the grey relational analysis (GR) to identify the potential sources ofeighteen Asian dust storm (ADS) events The results showed that the chemical compositions of atmospheric particles sampled atthe Pescadores Islands were very similar to source soils fugitively emitted from Inner Mongolia which could assist in identifyingthe source regions of Asian dusts This study further compared the source allocation of Asian dusts obtained from EF GR andbackward trajectory which showed that the source regions of Asian dusts obtained from these three methods were quite similarThe similarity of backward trajectory and GR reached as high as 833 Moreover the similarity of backward trajectory calculationand EF or GR was up to 778 while that of the GR and EF was up to 833 Overall these three methods can successfully allocatethe source regions of Asian dusts by 667 Moreover these innovative chemical-assisted methods can be successfully applied toidentify the source regions of Asian dusts for 18 ADS events

1 Introduction

At the beginning (FebruaryndashMay) of each year Asian dustsare frequently blown fromhyperarid arid and semiarid landslocated at Northwest and North China South Mongolia andSouth Siberia [1ndash3] For the past half-century the frequencyof Asian dust storms (ADS) occurring in China increasedsignificantly from 05 (1950rsquo) to 14 (2000) times per yearespecially the strong andor super ADSThis phenomenon iscorrelated mainly to climate change both locally and globally[4 5] It is estimated that approximately 800 million metrictons of sands were emitted to the atmosphere by ADS whichaccounts for almost half of natural fugitive sands globally[6 7]

ADS could transport in the upper troposphere easterlyto Korea Japan and even North America and may alsotransport southerly to Okinawa Taiwan Hong Kong andPhilippine due to anticyclone of atmospheric circulation [8ndash11] Although many sands might be settled down to theocean and provided nutrients (eg Fe2+) to the oceanic

biosphere which actually benefits the circulation of globalgeochemistry ADS could deteriorate ambient air qualityreduce atmospheric visibility cause adverse effects on humanhealth and result in poor scenery and public welfare [12ndash14]Additionally great losses of national economy due to ADSare constantly reported in theWest Pacific regions includingcrop damage soil productivity losses livestock losses massmigration human health impacts and impacts on climate[15ndash18]

In general ADS invades Taiwan 2-3 times per year andlasts for about 1-2 days for each invasion The concentrationof suspended particles (PM

10) in the events of ADS increases

approximately 2-3-fold when compared to normal days [7 1920] During the ADS periods significant increases of atmo-spheric aerosols particularly PM

10 have been frequently

observed in Taiwan [7 8 19 21] Particularly yellow rainepisodes have been reported both in Japan and Taiwan at theend of March 2000 [22]The morphology and apparent colorof dust particles filtered from the yellow rain were similar toyellow sands at their source [22] Furthermore the results

Hindawi Publishing CorporationAdvances in MeteorologyVolume 2014 Article ID 268037 10 pageshttpdxdoiorg1011552014268037

2 Advances in Meteorology

obtained from the investigation on source apportioningatmospheric aerosols at anAsian dust episode inmetroKaoh-siung indicated that the percentage of fugitive dusts increasedsignificantly from85 (regular periods) to 262 (Asian dustperiods) [23] Moreover during the transportation of ADSacross continents the physicochemical characteristics ofAsian dusts might be changed by mixing with anthropogenicparticles emitted from typical stationary and mobile sources[24 25] Yuan et al [19] reported that the size distributionof atmospheric aerosols sampled at the Pescadores Islandslocated in the middle of the Taiwan Strait during the periodof ADS switched from bimode to single mode with a peakin coarse particles Besides the sulfate (SO

4

2minus) and ammonia(NH4

+) of coarse particles (PM25minus10

) increase significantlyduring the period of ADS when compared to non-ADSperiods [7 26 27]

In order to investigate the feasibility of chemical-assistedADS source identification methods which can be used totrace back the soil source regions of ADS this study com-pared the transportation routes of ADS event from the back-ward trajectory calculation to the chemical composition ofatmospheric particles and soil sources Top soils collected inthree soil source regions of InnerMongolia were resuspendedand sampled to analyze their chemical composition forfurther conducting the enrichment factor analysis (EF) andgrey relational analysis (GR) Ultimately this study revealedand discussed the similarity percentages from the backwardtrajectory calculation EF and GR and further obtained thesource regions of Asian dusts for 18 ADS events

2 Methodologies

21 Sampling Sites of Atmospheric Particles and Soil DustsLocated at the middle of Taiwan Strait between southeasternChina and Taiwan Island the Pescadores Islands have anarea of 12797 km2 which has the subtropical weather beingmainly influenced by East Asian monsoons During the 18ADS events the atmospheric particleswere sampled atHsiao-men site (23∘38101584047110158401015840N 119∘30101584031610158401015840 E) which is located atthe northwestern coastline of the Pescadores Islands and isapproximately 12m above the ground and 500m and 50mfar from the coastline and themajor roads respectively At thesampling sites this study used to simultaneously collect PM

10

(including PM25minus10

and PM25) with dichotomous samplers

(Anderson Model Series 241) The sampling flow rate of thesampler was 167 Lmin for a total of 24 hours and the filtersused in this study were 37mm quartz filters

The soil samples were collected from three ADS sourceregions in the western (38∘13101584019 to 39∘06101584045N 115∘21101584047to 116∘53101584038 E) central (40∘06101584049 to 41∘3910158400N 110∘28101584035 to111∘32101584020 E) and eastern (43∘03101584027 to 43∘28101584037N 116∘41101584041 to119∘02101584009 E) InnerMongolia China [7] Among them strongwinds and dry climate result in serious soil erosion and thuscause heavyADS frequently (InnerMongoliaWeb Site 2003)A total of thirty soil samples were taken at the top 10ndash15 cmsurface of open sand lands and were then separately stored inlabeled sampling bagsThe soil sampleswere then sent back totheAir Pollution Laboratory at the Institute of Environmental

20 4 (km)

Sampling site

Central

EastWest

Pescadores Islands

Inner Mongolia

Soil sampling region

Inner Mongolia

N

Figure 1 Location of atmospheric particles (995333) at the PescadoresIslands and soil source regions (I) in the Inner Mongolia

Engineering in National Sun Yat-Sen University Taiwan forfurther chemical analyses The locations of all soil samplingsites are illustrated in Figure 1

22 Backward Trajectory Calculation In order to trace thetransportation of air parcel backward trajectories from aspecific receptor site were commonly used in this studywhich could be used to identify the air mass transportationroutes from specific source regions A hybrid single-particleLagrangian integrated trajectory (HYSPLIT) has been awidely used model that plots the trajectory of a single airparcel from a specific location and height above ground overa period of time [28 29] Backward trajectory calculationused the interpolated measured or modeled meteorologicalfields to estimate the most likely route over the geographical

Advances in Meteorology 3

Power atomizer

Dichotomous sampler

15 m

10 m

05 m

02 m

Figure 2 Schematic diagram of the resuspension chamber

areas that provided air parcels from its source to a receptorat a given time The method essentially follows a parcel of airbackward in hourly steps for a specified length of time [1030] Moreover HYSPLIT uses archived 3-dimensional mete-orological fields generated from observations and short-termforecasts HYSPLIT can be run to generate either forward orbackward trajectories using several available meteorologicaldata archives which is developed by the National Oceanicand Atmospheric Administrationrsquos (NOAA) Air ResourcesLaboratory (ARL) All backward trajectories were drawn atan arrival height of 50m above the sea level in this studyBy using the NOAA-HYSPLIT model the dates with highparticle concentration at the Pescadores Islands were deter-mined and the transportation routes of air parcels toward thePescadores Islands during the sampling periods were thensimulated This information was further applied to examinethe possible long-range transportation routes conveying ADSfrom several long-range soil source regions to the PescadoresIslands

23 Enrichment Factor Analysis Soil samples collected fromthe eastern central and western parts of the Inner Mongoliawere sent back to the Air Pollution Laboratory at NationalSun Yat-Sen University for further soil resuspension andchemical analyses The soil samples were resuspended witha dry powder atomizer at the top of a resuspension chamber(2mtimes 1m2) which is illustrated in Figure 2The resuspendeddusts were then sampled as PM

25minus10and PM

25for a total

of 05 hours by two dichotomous samplers with a totalvolumetric flow rate of 167 Lmin which was located atthe bottom of the resuspension chamber Particulate matterwith diameters larger than 10 120583m tended to settle downat the bottom of the resuspension chamber Besides theresuspended soil dusts were further characterized chemically

to apply for the identification of Asian dusts at the downwindsites by using an enrichment factor (EF) The enrichmentfactor (EF) defined by (1) was used to correlate the downwindatmospheric particles sampled at the Pescadores IslandsTaiwan to ADS originated from Inner Mongolia China

EF = (Tr Ref)PM(Tr Ref)crust

(1)

where EF is the enrichment factor Tr is the trace elementRef is the reference element and subscripts PM and crustrepresents atmospheric particles and crustal soils Amongthem the reference element has to be initially determinedPrevious studies reported that the mass ratio of Mg K andFe (or Al) to Al (or Fe) showed a relatively narrow rangefor coarse particles but not for fine particles which can beused as the fingerprints of Asian dusts originated from theInner Mongolia [7] Generally the enrichment factor close tounity (EF asymp 1) indicates that the elements consider very highpossibility from the specific sources It is also defined that theelements might come from the specific source when the EFranged from 01 to 10

24 Grey Relational Analysis The technique of grey relationalanalysis is suitable for unascertained problems with poorinformation [31] It provides an efficient solution to theuncertainty multi-input and discrete data problem Thiskind of interaction is mainly through the connection amongparameters and some conditions that are already knownAlso it will indicate the relational degree between twosequences with the help of the grey relational analysisMoreover the grey relational grade could utilize the discretemeasurement method to judge the distance

When the range of the sequence is too large or thestandard value is too enormous it will cause the influenceof some factors to be neglected Also in the sequence if thefactorsrsquo goals and directions are different the grey relationalanalysis might also produce unexpected incorrect resultsTherefore preprocessing of all the data is highly requiredThis process is called the grey relational generating An idealsequence is 119909

0(119896) (119896 = 1 2 3) for three responses The

definition of the grey relational grade in the grey relationalanalysis is to show the relational degree between 119899 sequences(1199090(119896) and 119909

119894(119896) 119894 = 1 2 119899 119896 = 1 2 3 119899) The

grey relational coefficient 120585119894(119896) could be calculated by the

following equation

120585119894(119896) =Δmin + 120577ΔmaxΔ0119894(119896) + 120577Δmax

(2)

where Δ0119894(119896) = |119909

0(119896) minus 119909

119894(119896)| is the difference of absolute

value between 1199090(119896) and 119909

119894(119896) 120585 is the distinguishing coeffi-

cient (0 1) Δmin is the smallest value of Δ0119894(119896) and Δmax is

the largest value of Δ0119894(119896) After averaging the grey relational

coefficients the grey relational grade 119884119894can be obtained as

119884119894=

119898

sum

119896=1

120573119896120585119894(119896) (3)

where 119898 is the number of process responses and 120573119896is the

weight value


Central

Central Central

West West West

East EastEast

ETR ETR ETR

STC STC STC

SST SST SST

Figure 3 Backward trajectories for three types of ADS transportation routes from the Inner Mongolia to the Pescadores Islands

3 Results and Discussion

31 Correlation of Chemical Composition with Backward Tra-jectory Calculation A total of 18 ADS events were observedat the Pescadores Islands fromNovember 2002 to April 2006Results obtained from the backward trajectory calculationindicated that three main types of routes dominated theAsian dusts transported from the Inner Mongolia to thePescadores Islands which included the eastward transporta-tion and retraced route (ETR) southeastern transportationand circumrotated route (STC) and straight southeasterntransportation route (SST) as illustrated in Figure 3

Table 1 summarizes that the ERT-type ADS eventsaccounted for approximately 333 of the total of 18 ADSevents which were mostly accompanied with strong north-eastern monsoons occurring from late February to AprilWhen the ADS transported toward the east the trans-portation routes would detour toward the southwest due toatmospheric circulation which usually passed through thePescadores Islands The STC-type ADS events were mainly

affected by the strength of the northeasternmonsoonsmostlyobserved fromNovember to next April accounting for about222 of the total of 18 ADS events Weakening the north-eastern monsoons would result in a smooth curvature of thebackward trajectories The SST-type ADS events occurredmostly from December to next April in which air parcelstransported almost directly from northern China to thePescadores Islands Taiwan

According to the probability of transportation routes itshowed that the SST-type ADS was dominant in all ADStransportation routes accounting for approximately 445of the 18 ADS events The ERT- and STC-type ADS wereordered the second and third routes respectively Moreoverthis study further aggregated the routes of 18 ADS events andtried to trace them back to their major source regions All oftheADS routeswere transported along the coastline or passedthrough the major mega cities of North andor East Chinawhich could alter the chemical composition of atmosphericdusts while transporting from the Inner Mongolia to thePescadores Islands During the transportation processes


Table 1 The transportation types and soil sources of 18 ADS events in this study

Transportation route types Events Routes probability Predilection periods Asian dust source regionETR

119899 = 18

333 Late FebsimApril Central (83)STC 222 Novsimnext April West (75)SST 445 Decsimnext April East (88)(1) ETR is the eastward transportation and retraced route(2) STC is the southeastern transportation and circumrotated route(3) SST is the straight southeastern transportation route

Asian dusts would mix with particulate matter emittedfrom local anthropogenic industrial and mobile sourcesThis study further compared the chemical composition ofresuspended source soils and atmospheric dusts for similartransportation route type as illustrated in Figure 4

The results indicated that the most abundant chemicalcomposition of source soils were EC SO

4

2minus Clminus NO3

minus OCand Ca2+ in the eastern source region of Inner Mongolia ECFe Al SO

4

2minus Ca and Ca2+ were in the central source regionEC SO

4

2minus OCNO3

minus Clminus andCawere in thewestern sourceregion respectively On the other hand the most abundantchemical composition for different transportation routeswere SO

4

2minus Ca2+ NO3

minus Clminus Na+ and EC for the ERT-typeADS SO

4

2minus Clminus Na+ NO3

minus Al and Fe for the STC-typeADS and SO

4

2minus EC NO3

minus Na+ Clminus and Ca2+ for the SST-type ADS respectively The chemical composition betweenthe soils collected in the eastern and western soil sources ofInner Mongolia and Asian dusts sampled at the PescadoresIslands were similar except the water-soluble Ca2+ andmetalCa The chemical composition of soils collected at each soilsource regionwas similar but not identical which could causethe chemical composition variation of atmospheric dustsThus the chemical composition from each soil source wassimilar to the chemical composition of atmospheric dust ontheir corresponding transportation route Among them theair masses traveled through several major industrial cities ofChina and transported along the coastal regions by the back-ward trajectories Previous studies [29 30 32 33] showedthat the chemical proportion on the atmospheric dust waschanged by the anthropogenic pollutions or nature sourcesemission such as SO

4

2minus and NO3

minus originated from directemissions of major industries (such as petrochemical powerplant and incinerator) or mobile sources (including onshoreand offshore) Calcium and aluminum were dominated bystone processing construction and the cement industry ECwas originated primarily from direct emissions of fuel andcoal combustion

This study further compared the chemical composition ofthe source soils and atmospheric dusts sampled for differenttransportation routes The results showed that the chemicalfingerprints of the most soil sources and atmospheric dustswere coincided except some chemical species were differentActually the chemical composition of atmospheric dustscould be influenced by the anthropogenic emission sourcesfor different ADS transportation routes thus resulting in theincrease of the portion of SO

4

2minus NO3

minus and EC MoreoverAsian dusts could be long-range transported across the

ocean to the Pescadores Islands which could increase theproportion of Na+ and Clminus due to oceanic spray As a wholethe source soils were not exactly the same in their chemicalcomposition between the eastern central or western InnerMongolia but the chemical species of the source soils inthe Inner Mongolia and atmospheric dusts in the PescadoresIslands were similar in the same transportation routes Itcould be explained that both backward trajectory calculationand chemical composition can be used to decide the ADStransportation routes concurrently

32 Correlation of Chemical Composition with EnrichmentFactors and Grey Relational Analysis The enrichment factorwas further applied to identify the ADS source region in theInner Mongolia toward the Pescadores Islands in this studyDuring the transportation processes water-soluble ions andcarbonaceous components of Asian dusts could be interferedby the anthropogenic emissions However the characteristicsof metallic elements are less likely varied in the ambient airConsequently this study selected Al or Fe as trace metalswhich are fairly representative in the earth crust and can beused to calculate the EF of 15 metals of Asian dusts originatedfrom the Inner Mongolia during the 18 ADS events as showninTable 2TheEF values of 15metal compositions in the rangeof 01 to 10 were shown to be highly relevant as illustratedin Figure 5 Source identification results showed that 72of ADS out of 18 events were originated from the centralsoil source region of the Inner Mongolia However threeADS events (numbers 3 7 and 8) were not consistentlyidentified for their source regions by using Al and Fe asreference elementsThe source regions identified by EF valueswere either the western or the central Inner Mongolia Itmight be attributed that Asian dusts could be mixed withother anthropogenic particles during the ADS transportationprocesses

This study further applied the grey relational analysis toallocate the source regions of 18 ADS events as summarizedin Table 3 The results indicated that approximately 27 23and 50 of Asian dusts were originated from the easternwestern and central Inner Mongolia for the 18ADS eventsrespectivelyMoreover if the results of grey relational analysiswere close to 1 the soil source of ADS events could beidentified with relatively high correlation Unlike the resultsof EF analysis the source regions would be determineddecisively by the grey relational analysis However if thegrey relational grades of three soil source regions wereclose to each other it could be difficult to allocate their


0

10

20

30

40

50()

()

60

70

80

90

100

Na Ca AlFe Mg KZn Cr TiMn Ba Sr

Ni Pb CuEC OC

0

10

20

30

40

50

60

70

80

90

100

East Central West

SST ETR STC

Source soils

Atmospheric dusts

Fminus

Clminus

Na+ K+

NOminus

3 SO 2minus4

NH +

4

Mg2+

Ca2+


Ni Pb CuEC OC Fminus

Clminus

Na+ K+

NO minus

3 SO 2minus4

NH +

4

Mg2+

Ca2+

Figure 4 Chemical fingerprints of resuspended source soils and atmospheric dusts during the ADS events

sources in this specific case For example number 13 ADSevent was identified from the central region of the InnerMongolia by backward trajectory calculation Its GR gradeswere shown as 078 091 and 089 for the eastern centraland western soil source regions respectively Similar resultswere obtained from EF analysis indicating that the chemicalcomposition ofAsian dusts played a crucial role in identifyingthe soil source regions of Asian dusts Overall the chemical-based enrichment factor and the grey relational analysisboth showed that the soil source region of number 13 ADS

event was highly likely emitted from the central InnerMongolia

33 Comparison of Three Source Identification Methods Fig-ure 6 illustrates and compares three methods for identifyingthe source regions of 18 ADS events Through the backwardtrajectory calculation enrichment factor and grey relationalanalysis could be applied to identify the ADS events numbers1 2 6 8 9 11 and 13ndash18 which accounted for 667 of the


Table 2 Source allocation of 18 ADS events in the Inner Mongolia by enrichment factor analysis

ADS events Dates of ADS events Reference elements ADS source region Identified sourcesEast Central West

1 Mar 8-9 2002 Al 14 9 12 EastFe 13 9 12

2 Mar 18ndash20 2002 Al 14 11 13 EastFe 14 13 13

3lowast Mar 31 2002 Al 11 12 13 West centralFe 11 12 11

4 Apr 13 2002 Al 7 9 8 CentralFe 7 9 7

5 Apr 17ndash19 2002 Al 11 13 12 EastFe 11 8 9

6 Feb 25 2003 Al 12 15 12 CentralFe 13 14 12

7lowast Mar 7-8 2003 Al 12 15 15 West centralFe 13 13 13


9 Apr 26-27 2003 Al 12 14 14 CentralFe 12 13 12

10 Feb 6 2004 Al 10 12 10 CentralFe 10 12 9

11 Feb 15-16 2004 Al 10 13 10 CentralFe 11 11 11

12 Apr 2 2004 Al 11 13 10 CentralFe 10 13 11

13 Apr 13ndash15 2004 Al 11 14 14 CentralFe 12 14 12

14 Nov 29 2005 Al 12 15 14 CentralFe 12 15 13

15 Dec 21ndash26 2005 Al 13 14 15 WestFe 13 11 13

16 Mar 14ndash17 2006 Al 9 13 9 CentralFe 10 12 12

17 Mar 19 2006 Al 8 13 8 CentralFe 8 13 12

18 Apr 24-25 2006 Al 10 15 10 EastFe 12 15 12

lowastDifferent source regions determined by EF values using Al and Fe as reference elements

18 ADS events Comparing backward trajectory calculationwith enrichment factor the similarity percentage was 778for 18ADS events including ADS events numbers 1 2 5ndash9 11 12 and 14ndash18 in this study In terms of enrichmentfactor and grey relational analysis it was shown that theywere coincident for 833 of 18 ADS events including ADSevents numbers 1 2 4ndash6 8ndash11 12 and 14ndash18 In terms ofgrey relational analysis and backward trajectory calculationit was shown that they were coincident for 778 of 18 ADSevents including ADS events numbers 1 2 4ndash6 8ndash11 and 14ndash18 Generally speaking the source allocationmethods of ADStransported from Inner Mongolia to the Pescadores Islands

were relatively accurate especially for simultaneous applica-tion of enrichment factor and grey relational analysis Moreimportantly accurate chemical fingerprints of source soilsand atmospheric particles are highly required for chemicallyallocating the source regions of the ADS events

In order to further confirm the above three optimalmeth-ods feasible for allocating the mostly possible sources of ADSevents this study used the past general method to comparewith the above three methods which calculated the massratios of Al and Sr (AlSr) which had the highest differencesin the 15 metals for 18 ADS events Table 4 lists the mass ratioof AlSr for resuspended soils collected in the InnerMongolia


Na Ca Al Fe Mg K Zn Cr Ti Mn Ba Sr Ni Pb Cu


Na Ca Al Fe Mg K Zn Cr Ti Mn Ba Sr Ni Pb CuNa Ca Al Fe Mg K Zn Cr Ti Mn Ba Sr Ni Pb Cu



00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)

Figure 5 EF values of 15 metals with Al or Fe as reference elements at the Pescadores Islands For example (a1)sim(a3) are the EF of easterncentral and western source regions calculated by using aluminum as reference element (b1)sim(b3) are the EF of eastern central and westernsource regions calculated by using iron as reference element

Table 3 Source allocation of 18 ADS events in the Inner Mongolia by grey relational calculation

ADS events Dates of ADS events ADS source regions Identified sourcesEast Central West

1 Mar 8-9 2002 090 087 082 East2 Mar 18ndash20 2002 081 076 074 East3 Mar 31 2002 079 084 087 West4 Apr 13 2002 081 074 079 East5 Apr 17ndash19 2002 079 078 071 East6 Feb 25 2003 085 092 084 Central7 Mar 7-8 2003 087 091 089 Central8 Mar 26-27 2003 078 091 083 Central9 Apr 26-27 2003 065 087 083 Central10 Feb 6 2004 077 083 085 West11 Feb 15-16 2004 070 084 078 Central12 Apr 2 2004 083 086 093 West13 Apr 13ndash15 2004 078 091 089 Central14 Nov 29 2005 076 087 083 Central15 Dec 21ndash26 2005 072 078 089 West16 Mar 14ndash17 2006 063 087 084 Central17 Mar 19 2006 069 079 080 Central18 Apr 24-25 2006 092 089 083 East


(778)

(667)

Grey relationalanalysis

Backwardtrajectory

calculation

Enrichment factor

(778) (833)

Figure 6 Comparison of the correlation percentages among back-ward trajectory calculation enrichment factor and grey relationalanalysis

Table 4 Comparison of the AlSr ratios for source soils andatmospheric particles sampled in the Inner Mongolia and at thePescadores Islands

Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents

East 556 plusmn 43 330 plusmn 933 1 2 4 5 18Central 3250 plusmn 1180 1611 plusmn 651 6ndash9 11ndash14 16 17West 25 plusmn 08 1329 plusmn 426 7 10 12 15

and atmospheric particles sampled at the Pescadores IslandsThe results showed that the resuspended soil samples inthe central Inner Mongolia had relatively higher AlSr ratio(3250 plusmn 1180) and observed simultaneously with the higherAlSr ratio (1611 plusmn 651) of atmospheric particles at thePescadores Islands Furthermore the resuspended soils andAsian dusts in the eastern Inner Mongolia both had high rel-ative standard deviation (RSD) of 77 and 283 respectivelyThe AlSr ratios of the resuspended soils and Asian dusts inthe eastern and central Inner Mongolia had their significantcharacteristics which could be used to identify the soil sourceregions of ADS events while the AlSr ratios in the westernInner Mongolia were relatively broad Overall although theAlSr ratio could assist in identifying the sources of ADSits confidence level was relatively lower than the backwardtrajectory calculation enrichment factor and grey relationalanalysis Finally this study has selected the mostly possibleanalysismethods to identify the sources of ADS events whichcould be used to trace the source areas of ADS event in thefuture

4 Conclusions

This study successfully combined backward trajectory calcu-lation enrichment factor and grey relational analysis to allo-cate the potential source(s) of 18 ADS events and identified

the transportation routes of Asian dusts transported fromthe Inner Mongolia to the Pescadores Islands Three majortransportation routes observed for 18ADS events includedETR STC and SSTThevariations of chemical components ofthe atmospheric dusts or source soils were correlated to theirtransportation routes in which the chemical characteristicsof Asian dusts could be influenced by anthropogenic particlesemitted from local stationary and mobile sources during theADS transportation processes However most of chemicalcomponents between source soils and atmospheric dusts forthe same transportation routes were coincident Moreoversource identification results showed that 50ndash72 out of18 ADS events were originated from the central region ofInner Mongolia Comparison of the three methods showedthat the chemical-assisted methods demonstrated the accu-racy of 667 for identifying the source regions of ADSevents Particularly simultaneous application of enrichmentfactor and grey relational analysis raised the accuracy up to833 Overall these three chemical-assisted tracing meth-ods were feasible for identifying the soil source regions ofADC events

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors gratefully acknowledge the financial supportfrom the National Science Council and kind assistancefrom the Institute of Environmental Engineering at NationalSun Yat-Sen University The authors would like to expresstheir sincere appreciation for its financial and technologicalsupport to accomplish this study

References

[1] H Bian X Tie J Cao Z Ying S Han and Y Xue ldquoAnalysis ofa severe dust storm event over China application of the WRF-dust modelrdquo Aerosol and Air Quality Research vol 11 no 4 pp419ndash428 2011

[2] J J Cao J C Chow J G Watson et al ldquoSize-differentiatedsource profiles for fugitive dust in the Chinese Loess PlateaurdquoAtmospheric Environment vol 42 no 10 pp 2261ndash2275 2008

[3] M T Cheng W C Chou C P Chio et al ldquoCompositions andsource apportionments of atmospheric aerosol during Asiandust storm and local pollution in central Taiwanrdquo Journal ofAtmospheric Chemistry vol 61 no 2 pp 155ndash173 2008

[4] R A Duce C K Unni B J Ray JM Prospero and J TMerrillldquoLong-range atmospheric transport of soil dust fromAsia to thetropical North Pacific temporal variabilityrdquo Science vol 209no 4464 pp 1522ndash1524 1980

[5] X Y Zhang S L Gong Z X Shen et al ldquoCharacterization ofsoil dust aerosol in China and its transport and distributionduring 2001 ACE-Asia 1 Network observationsrdquo Journal ofGeophysical Research D Atmospheres vol 108 no 9 pp 4261ndash4274 2003

[6] X Y Zhang R Arimoto and Z S An ldquoDust emission fromChinese desert sources linked to variations in atmospheric


circulationrdquo Journal of Geophysical Research D vol 102 no 23pp 28041ndash28047 1997

[7] C S Yuan C X Hai and M Zhao ldquoSource profiles andfingerprints of fine and coarse sands resuspended from the soilssampled in the central inner mongoliardquo Particuology vol 4 pp304ndash311 2006

[8] Y I Tsai and C L Chen ldquoCharacterization of Asian dust stormand non-Asian dust storm PM

25aerosol in southern Taiwanrdquo

Atmospheric Environment vol 40 no 25 pp 4734ndash4750 2006[9] I Mori M Nishikawa T Tanimura and H Quan ldquoChange

in size distribution and chemical composition of kosa (Asiandust) aerosol during long-range transportrdquo Atmospheric Envi-ronment vol 37 no 30 pp 4253ndash4263 2003

[10] S Tan G Shi and H Wang ldquoLong-range transport of springdust storms in Inner Mongolia and impact on the China seasrdquoAtmospheric Environment vol 46 pp 299ndash308 2012

[11] F Tsai G T J Chen T H Liu W D Lin and J Y TuldquoCharacterizing the transport pathways of Asian dustrdquo Journalof Geophysical Research vol 113 Article ID D17311 2008

[12] R D Brook ldquoIs air pollution a cause of cardiovascular diseaseUpdated review and controversiesrdquo Reviews on EnvironmentalHealth vol 22 no 2 pp 115ndash137 2007

[13] C Kang W Kim H Ko and S Hong ldquoAsian Dust effects onTotal Suspended Particulate (TSP) compositions at Gosan inJeju Island KoreardquoAtmospheric Research vol 94 no 2 pp 345ndash355 2009

[14] S Wang S Tsay N Lin et al ldquoFirst detailed observations oflong-range transported dust over the northern South ChinaSeardquo Atmospheric Environment vol 45 no 27 pp 4804ndash48082011

[15] M V K Sivakumar ldquoImpacts of sand stormsdust stormson agriculturerdquo in Natural Disasters and Extreme Events inAgriculture pp 159ndash177 2005

[16] D Y Jeong ldquoSocio-economic costs from yellow dust damages inSouth Koreardquo Transborder Environmental and Natural ResourceManagement pp 185ndash198 2008

[17] H Mu S Otani M Shinoda et al ldquoLong-term effects oflivestock loss caused by dust storm onmongolian inhabitants asurvey 1 year after the dust stormrdquo Yonago Acta Medica vol 56no 1 pp 39ndash42 2013

[18] H Lee Y Honda Y H Lim Y L Guo M Hashizume and HKim ldquoEffect of Asian dust storms on mortality in three Asiancitiesrdquo Atmospheric Environment vol 89 pp 309ndash317 2014

[19] C S Yuan C C SauM C Chen et al ldquoMass concentration andsize-resolved chemical composition of atmospheric aerosolssampled at the Pescadores Islands during Asian dust stormperiods in the years of 2001 and 2002rdquo Terrestrial Atmosphericand Oceanic Sciences vol 15 no 5 pp 857ndash879 2004

[20] B Lee H K Lee andN Jun ldquoAnalysis of regional and temporalcharacteristics of PM

10during an Asian dust episode in Koreardquo

Chemosphere vol 63 no 7 pp 1106ndash1115 2006[21] S Chen L Hsieh M Kao W Lin K Huang and C Lin

ldquoCharacteristics of particles sampled in southern Taiwan duringthe Asian dust storm periods in 2000 and 2001rdquo AtmosphericEnvironment vol 38 no 35 pp 5925ndash5934 2004

[22] C J Ma M Kasahara S Tohno R Holler and T KamiyaldquoChemical composition of single raindrops fallen in yellowrainfall episode in Japanrdquo inProceedings of the 2ndAsianAerosolConference pp 185ndash186 Pusan Republic of Korea 2001

[23] S H Liu and C S Yuan Source apportionment and opticalproperties of atmospheric aerosols in metro Kaohsiung [MS

thesis] Institute of Environmental Engineering National SunYat-sen University Kaohsiung Taiwan 2000 (Chinese)

[24] G McTainsh Y Chan H McGowan J Leys and K Tews ldquoThe23rd October 2002 dust storm in eastern Australia characteris-tics and meteorological conditionsrdquo Atmospheric Environmentvol 39 no 7 pp 1227ndash1236 2005

[25] Z X Shen J J Cao R Arimoto et al ldquoIonic composition ofTSP andPM25 during dust storms and air pollution episodes atXian Chinardquo Atmospheric Environment vol 43 pp 2911ndash29182009

[26] T Nakamura K Matsumoto and M Uematsu ldquoChemicalcharacteristics of aerosols transported from Asia to the EastChina Sea An evaluation of anthropogenic combined nitrogendeposition in autumnrdquo Atmospheric Environment vol 39 no 9pp 1749ndash1758 2005

[27] R Zhang Z Shen T Cheng M Zhang and Y Liu ldquoThe ele-mental composition of atmospheric particles at Beijing duringAsian dust events in spring 2004rdquo Aerosol and Air QualityResearch vol 10 no 1 pp 67ndash75 2010

[28] S F Kong B Han Z P Bai L Chen J W Shi and ZXu ldquoReceptor modeling of PM

25 PM10

and TSP in differentseasons and long-range transport analysis at a coastal site ofTianjin Chinardquo Science of the Total Environment vol 408 no20 pp 4681ndash4694 2010

[29] L Zhu X Huang H Shi X Cai and Y Song ldquoTransportpathways and potential sources of PM

10in BeijingrdquoAtmospheric

Environment vol 45 no 3 pp 594ndash604 2011[30] HH Tsai C S Yuan CHHung andY C Lin ldquoPhysicochem-

ical properties of PM25

and PM25minus10

at inland and offshore sitesover Southeastern coastal region of Taiwan straitrdquo Aerosol andAir Quality Research vol 11 no 6 pp 664ndash678 2011

[31] J L Deng A Course on Grey System Theory HUST PressWuhan China 1990

[32] G Dongarra E Manno D Varrica M Lombardo and MVultaggio ldquoStudy on ambient concentrations of PM

10 PM10minus25

PM25

and gaseous pollutants trace elements and chemical spe-ciation of atmospheric particulatesrdquo Atmospheric Environmentvol 44 no 39 pp 5244ndash5257 2010

[33] T C Li W H Chen C S Yuan S P Wu and X H WangldquoPhysicochemical characteristics and source apportionment ofatmospheric aerosol particles in Kinmen-Xiamen AirshedrdquoAerosol andAirQuality Research vol 13 no 1 pp 308ndash323 2013

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ClimatologyJournal of

EcologyInternational Journal of


EarthquakesJournal of


Hindawi Publishing Corporationhttpwwwhindawicom

Applied ampEnvironmentalSoil Science

Volume 2014

Mining


Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal of

Geophysics

OceanographyInternational Journal of


Journal of Computational Environmental SciencesHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal ofPetroleum Engineering


GeochemistryHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Atmospheric SciencesInternational Journal of


OceanographyHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in


MineralogyInternational Journal of


MeteorologyAdvances in

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Paleontology JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Geological ResearchJournal of


Geology Advances in


obtained from the investigation on source apportioningatmospheric aerosols at anAsian dust episode inmetroKaoh-siung indicated that the percentage of fugitive dusts increasedsignificantly from85 (regular periods) to 262 (Asian dustperiods) [23] Moreover during the transportation of ADSacross continents the physicochemical characteristics ofAsian dusts might be changed by mixing with anthropogenicparticles emitted from typical stationary and mobile sources[24 25] Yuan et al [19] reported that the size distributionof atmospheric aerosols sampled at the Pescadores Islandslocated in the middle of the Taiwan Strait during the periodof ADS switched from bimode to single mode with a peakin coarse particles Besides the sulfate (SO

4

2minus) and ammonia(NH4

+) of coarse particles (PM25minus10

) increase significantlyduring the period of ADS when compared to non-ADSperiods [7 26 27]

In order to investigate the feasibility of chemical-assistedADS source identification methods which can be used totrace back the soil source regions of ADS this study com-pared the transportation routes of ADS event from the back-ward trajectory calculation to the chemical composition ofatmospheric particles and soil sources Top soils collected inthree soil source regions of InnerMongolia were resuspendedand sampled to analyze their chemical composition forfurther conducting the enrichment factor analysis (EF) andgrey relational analysis (GR) Ultimately this study revealedand discussed the similarity percentages from the backwardtrajectory calculation EF and GR and further obtained thesource regions of Asian dusts for 18 ADS events

2 Methodologies

21 Sampling Sites of Atmospheric Particles and Soil DustsLocated at the middle of Taiwan Strait between southeasternChina and Taiwan Island the Pescadores Islands have anarea of 12797 km2 which has the subtropical weather beingmainly influenced by East Asian monsoons During the 18ADS events the atmospheric particleswere sampled atHsiao-men site (23∘38101584047110158401015840N 119∘30101584031610158401015840 E) which is located atthe northwestern coastline of the Pescadores Islands and isapproximately 12m above the ground and 500m and 50mfar from the coastline and themajor roads respectively At thesampling sites this study used to simultaneously collect PM

10

(including PM25minus10

and PM25) with dichotomous samplers

(Anderson Model Series 241) The sampling flow rate of thesampler was 167 Lmin for a total of 24 hours and the filtersused in this study were 37mm quartz filters

The soil samples were collected from three ADS sourceregions in the western (38∘13101584019 to 39∘06101584045N 115∘21101584047to 116∘53101584038 E) central (40∘06101584049 to 41∘3910158400N 110∘28101584035 to111∘32101584020 E) and eastern (43∘03101584027 to 43∘28101584037N 116∘41101584041 to119∘02101584009 E) InnerMongolia China [7] Among them strongwinds and dry climate result in serious soil erosion and thuscause heavyADS frequently (InnerMongoliaWeb Site 2003)A total of thirty soil samples were taken at the top 10ndash15 cmsurface of open sand lands and were then separately stored inlabeled sampling bagsThe soil sampleswere then sent back totheAir Pollution Laboratory at the Institute of Environmental

20 4 (km)

Sampling site

Central

EastWest

Pescadores Islands

Inner Mongolia

Soil sampling region

Inner Mongolia

N

Figure 1 Location of atmospheric particles (995333) at the PescadoresIslands and soil source regions (I) in the Inner Mongolia

Engineering in National Sun Yat-Sen University Taiwan forfurther chemical analyses The locations of all soil samplingsites are illustrated in Figure 1

22 Backward Trajectory Calculation In order to trace thetransportation of air parcel backward trajectories from aspecific receptor site were commonly used in this studywhich could be used to identify the air mass transportationroutes from specific source regions A hybrid single-particleLagrangian integrated trajectory (HYSPLIT) has been awidely used model that plots the trajectory of a single airparcel from a specific location and height above ground overa period of time [28 29] Backward trajectory calculationused the interpolated measured or modeled meteorologicalfields to estimate the most likely route over the geographical


Power atomizer

Dichotomous sampler

15 m

10 m

05 m

02 m




25minus10and PM

25for a total




(1)






119894(119896) 119894 = 1 2 119899 119896 = 1 2 3 119899) The


following equation

120585119894(119896) =Δmin + 120577ΔmaxΔ0119894(119896) + 120577Δmax

(2)

where Δ0119894(119896) = |119909

0(119896) minus 119909







119884119894=

119898

sum

119896=1

120573119896120585119894(119896) (3)


weight value


Central

Central Central

West West West

East EastEast

ETR ETR ETR

STC STC STC

SST SST SST










119899 = 18




4

2minus Clminus NO3


4


4

2minus OCNO3


4

2minus Ca2+ NO3


4



4

2minus EC NO3


4

2minus and NO3



4

2minus NO3






0

10

20

30

40

50()

()

60

70

80

90

100


Ni Pb CuEC OC

0

10

20

30

40

50

60

70

80

90

100

East Central West

SST ETR STC

Source soils

Atmospheric dusts

Fminus

Clminus

Na+ K+

NOminus

3 SO 2minus4

NH +

4

Mg2+

Ca2+



Clminus

Na+ K+

NO minus

3 SO 2minus4

NH +

4

Mg2+

Ca2+








1 Mar 8-9 2002 Al 14 9 12 EastFe 13 9 12



4 Apr 13 2002 Al 7 9 8 CentralFe 7 9 7


6 Feb 25 2003 Al 12 15 12 CentralFe 13 14 12



9 Apr 26-27 2003 Al 12 14 14 CentralFe 12 13 12

10 Feb 6 2004 Al 10 12 10 CentralFe 10 12 9

11 Feb 15-16 2004 Al 10 13 10 CentralFe 11 11 11

12 Apr 2 2004 Al 11 13 10 CentralFe 10 13 11


14 Nov 29 2005 Al 12 15 14 CentralFe 12 15 13



17 Mar 19 2006 Al 8 13 8 CentralFe 8 13 12

18 Apr 24-25 2006 Al 10 15 10 EastFe 12 15 12











00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)






(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


Power atomizer

Dichotomous sampler

15 m

10 m

05 m

02 m




25minus10and PM

25for a total




(1)






119894(119896) 119894 = 1 2 119899 119896 = 1 2 3 119899) The


following equation

120585119894(119896) =Δmin + 120577ΔmaxΔ0119894(119896) + 120577Δmax

(2)

where Δ0119894(119896) = |119909

0(119896) minus 119909







119884119894=

119898

sum

119896=1

120573119896120585119894(119896) (3)


weight value


Central

Central Central

West West West

East EastEast

ETR ETR ETR

STC STC STC

SST SST SST










119899 = 18




4

2minus Clminus NO3


4


4

2minus OCNO3


4

2minus Ca2+ NO3


4



4

2minus EC NO3


4

2minus and NO3



4

2minus NO3






0

10

20

30

40

50()

()

60

70

80

90

100


Ni Pb CuEC OC

0

10

20

30

40

50

60

70

80

90

100

East Central West

SST ETR STC

Source soils

Atmospheric dusts

Fminus

Clminus

Na+ K+

NOminus

3 SO 2minus4

NH +

4

Mg2+

Ca2+



Clminus

Na+ K+

NO minus

3 SO 2minus4

NH +

4

Mg2+

Ca2+








1 Mar 8-9 2002 Al 14 9 12 EastFe 13 9 12



4 Apr 13 2002 Al 7 9 8 CentralFe 7 9 7


6 Feb 25 2003 Al 12 15 12 CentralFe 13 14 12



9 Apr 26-27 2003 Al 12 14 14 CentralFe 12 13 12

10 Feb 6 2004 Al 10 12 10 CentralFe 10 12 9

11 Feb 15-16 2004 Al 10 13 10 CentralFe 11 11 11

12 Apr 2 2004 Al 11 13 10 CentralFe 10 13 11


14 Nov 29 2005 Al 12 15 14 CentralFe 12 15 13



17 Mar 19 2006 Al 8 13 8 CentralFe 8 13 12

18 Apr 24-25 2006 Al 10 15 10 EastFe 12 15 12











00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)






(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


Central

Central Central

West West West

East EastEast

ETR ETR ETR

STC STC STC

SST SST SST










119899 = 18




4

2minus Clminus NO3


4


4

2minus OCNO3


4

2minus Ca2+ NO3


4



4

2minus EC NO3


4

2minus and NO3



4

2minus NO3






0

10

20

30

40

50()

()

60

70

80

90

100


Ni Pb CuEC OC

0

10

20

30

40

50

60

70

80

90

100

East Central West

SST ETR STC

Source soils

Atmospheric dusts

Fminus

Clminus

Na+ K+

NOminus

3 SO 2minus4

NH +

4

Mg2+

Ca2+



Clminus

Na+ K+

NO minus

3 SO 2minus4

NH +

4

Mg2+

Ca2+








1 Mar 8-9 2002 Al 14 9 12 EastFe 13 9 12



4 Apr 13 2002 Al 7 9 8 CentralFe 7 9 7


6 Feb 25 2003 Al 12 15 12 CentralFe 13 14 12



9 Apr 26-27 2003 Al 12 14 14 CentralFe 12 13 12

10 Feb 6 2004 Al 10 12 10 CentralFe 10 12 9

11 Feb 15-16 2004 Al 10 13 10 CentralFe 11 11 11

12 Apr 2 2004 Al 11 13 10 CentralFe 10 13 11


14 Nov 29 2005 Al 12 15 14 CentralFe 12 15 13



17 Mar 19 2006 Al 8 13 8 CentralFe 8 13 12

18 Apr 24-25 2006 Al 10 15 10 EastFe 12 15 12











00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)






(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in




119899 = 18




4

2minus Clminus NO3


4


4

2minus OCNO3


4

2minus Ca2+ NO3


4



4

2minus EC NO3


4

2minus and NO3



4

2minus NO3






0

10

20

30

40

50()

()

60

70

80

90

100


Ni Pb CuEC OC

0

10

20

30

40

50

60

70

80

90

100

East Central West

SST ETR STC

Source soils

Atmospheric dusts

Fminus

Clminus

Na+ K+

NOminus

3 SO 2minus4

NH +

4

Mg2+

Ca2+



Clminus

Na+ K+

NO minus

3 SO 2minus4

NH +

4

Mg2+

Ca2+








1 Mar 8-9 2002 Al 14 9 12 EastFe 13 9 12



4 Apr 13 2002 Al 7 9 8 CentralFe 7 9 7


6 Feb 25 2003 Al 12 15 12 CentralFe 13 14 12



9 Apr 26-27 2003 Al 12 14 14 CentralFe 12 13 12

10 Feb 6 2004 Al 10 12 10 CentralFe 10 12 9

11 Feb 15-16 2004 Al 10 13 10 CentralFe 11 11 11

12 Apr 2 2004 Al 11 13 10 CentralFe 10 13 11


14 Nov 29 2005 Al 12 15 14 CentralFe 12 15 13



17 Mar 19 2006 Al 8 13 8 CentralFe 8 13 12

18 Apr 24-25 2006 Al 10 15 10 EastFe 12 15 12











00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)






(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


0

10

20

30

40

50()

()

60

70

80

90

100


Ni Pb CuEC OC

0

10

20

30

40

50

60

70

80

90

100

East Central West

SST ETR STC

Source soils

Atmospheric dusts

Fminus

Clminus

Na+ K+

NOminus

3 SO 2minus4

NH +

4

Mg2+

Ca2+



Clminus

Na+ K+

NO minus

3 SO 2minus4

NH +

4

Mg2+

Ca2+








1 Mar 8-9 2002 Al 14 9 12 EastFe 13 9 12



4 Apr 13 2002 Al 7 9 8 CentralFe 7 9 7


6 Feb 25 2003 Al 12 15 12 CentralFe 13 14 12



9 Apr 26-27 2003 Al 12 14 14 CentralFe 12 13 12

10 Feb 6 2004 Al 10 12 10 CentralFe 10 12 9

11 Feb 15-16 2004 Al 10 13 10 CentralFe 11 11 11

12 Apr 2 2004 Al 11 13 10 CentralFe 10 13 11


14 Nov 29 2005 Al 12 15 14 CentralFe 12 15 13



17 Mar 19 2006 Al 8 13 8 CentralFe 8 13 12

18 Apr 24-25 2006 Al 10 15 10 EastFe 12 15 12











00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)






(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in




1 Mar 8-9 2002 Al 14 9 12 EastFe 13 9 12



4 Apr 13 2002 Al 7 9 8 CentralFe 7 9 7


6 Feb 25 2003 Al 12 15 12 CentralFe 13 14 12



9 Apr 26-27 2003 Al 12 14 14 CentralFe 12 13 12

10 Feb 6 2004 Al 10 12 10 CentralFe 10 12 9

11 Feb 15-16 2004 Al 10 13 10 CentralFe 11 11 11

12 Apr 2 2004 Al 11 13 10 CentralFe 10 13 11


14 Nov 29 2005 Al 12 15 14 CentralFe 12 15 13



17 Mar 19 2006 Al 8 13 8 CentralFe 8 13 12

18 Apr 24-25 2006 Al 10 15 10 EastFe 12 15 12











00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)






(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in







00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

00101

110

00101

110

00101

110

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

Enric

hmen

t fac

tors

MeanMaxMin

MeanMaxMin

(a1)

(a2)

(a3)

(b1)

(b2)

(b3)






(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


(778)

(667)


Backwardtrajectory

calculation

Enrichment factor

(778) (833)



Sourceregions

Soil samplesAlSr

Dust samplesAlSr

Number of ADSevents



4 Conclusions





Acknowledgments


References




































25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in






























25 PM10






and PM25minus10




10 PM10minus25

PM25












Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in










Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in

Documents

Research Article Source Allocation of Long-Range Asian ...downloads.hindawi.com/journals/amete/2014/268037.pdf · Backward Trajectory Calculation. In order to trace the transportation