springtime carbon emission episodes at the gosan background site

Atmos Chem Phys 11 10911ndash10928 2011wwwatmos-chem-physnet11109112011doi105194acp-11-10911-2011copy Author(s) 2011 CC Attribution 30 License

AtmosphericChemistry

and Physics

Springtime carbon emission episodes at the Gosan background siterevealed by total carbon stable carbon isotopic composition andthermal characteristics of carbonaceous particles

J Jung and K Kawamura

Institute of Low Temperature Science Hokkaido University Sapporo 0600819 Japan

Received 30 March 2011 ndash Published in Atmos Chem Phys Discuss 6 May 2011Revised 29 August 2011 ndash Accepted 25 October 2011 ndash Published 3 November 2011

Abstract In order to investigate the emission of carbona-ceous aerosols at the Gosan background super-site (3317 N12610 E) in East Asia total suspended particles (TSP)were collected during spring of 2007 and 2008 and ana-lyzed for particulate organic carbon elemental carbon to-tal carbon (TC) total nitrogen (TN) and stable carbon iso-topic composition (δ13C) of TC The stable carbon isotopiccomposition of TC (δ13CTC) was found to be lowest duringpollen emission episodes (rangeminus262 permil tominus235 permil avgminus252plusmn 09 permil) approaching those of the airborne pollen(minus280 permil) collected at the Gosan site Based on a carbonisotope mass balance equation we found thatsim42 of TCin the TSP samples during the pollen episodes was attributedto airborne pollen from Japanese cedar trees planted aroundtangerine farms in Jeju Island A negative correlation be-tween the citric acid-carbonTC ratios andδ13CTC was ob-tained during the pollen episodes These results suggestthat citric acid emitted from tangerine fruit may be adsorbedon the airborne pollen and then transported to the Gosansite Thermal evolution patterns of organic carbon during thepollen episodes were characterized by high OC evolution inthe OC2 temperature step (450C) Since thermal evolutionpatterns of organic aerosols are highly influenced by theirmolecular weight they can be used as additional informa-tion on the formation of secondary organic aerosols and theeffect of aging of organic aerosols during the long-range at-mospheric transport and sources of organic aerosols

Correspondence toK Kawamura(kawamuralowtemhokudaiacjp)

1 Introduction

Carbonaceous aerosols that comprise elemental carbon (EC)and organic carbon (OC) have large impacts on humanhealth (Baltensperger et al 2008) visibility impairment(IMPROVE 2006) and radiation budget in the atmosphere(Forster et al 2007) Organic aerosols are primarily emit-ted from various sources and secondarily produced in the at-mosphere by oxidation of volatile organic compounds fol-lowed by condensation on pre-existing particles andor nu-cleation Novakov and Penner (1993) determined the rel-ative contributions of sulfate and organic aerosols to cloudcondensation nuclei (CCN) concentrations at a marine siteknown to be influenced by anthropogenic emissions andfound that organic aerosols account for the major part of boththe total aerosol number concentration and the CCN frac-tion Thus in regions that are affected by anthropogenicpollutants organic aerosols may play at least as importanta role as sulfate aerosols in determining the climate effectof clouds (Novakov and Penner 1993) Model simulation re-sults indicate that organic aerosols can enhance cloud dropletconcentrations and are therefore an important component ofthe aerosol-cloud-climate feedback system (OrsquoDowd et al2004)

Jeju Island Korea is located at the boundary of the Yel-low Sea and the East China Sea and is surrounded by main-land China the Korean peninsula and Kyushu Island JapanThe Gosan site is located on the western edge of Jeju Is-land facing the Asian continent and is isolated from resi-dential areas on the island (Kawamura et al 2004) In or-der to understand physicochemical and radiative propertiesof anthropogenic aerosols under Asian continental outflowseveral international experiments have been conducted atthe Gosan site such as ACE-Asia (Aerosol CharacterizationExperiment-Asia) (Huebert et al 2003) and ABC-EAREX

Published by Copernicus Publications on behalf of the European Geosciences Union

10912 J Jung and K Kawamura Carbon episodes at Gosan site

2005 (Atmospheric Brown Cloud-East Asia Regional Ex-periment 2005) (Nakajima et al 2007) To better under-stand the link between chemical and physical properties ofaerosols mainly transported from the Asian continent and re-gional climate change sources and formation mechanism ofsecondary aerosols should be investigated In addition con-tributions of local effects on the Gosan site aerosols shouldbe qualitatively and quantitatively evaluated

Pollen is one of the important sources of bioaerosols(Solomon 2002) These particles can cause serious allergicproblems to human health (Solomon 2002) and visibility im-pairment (Kim 2007) Most pollen emission events in Koreaoccur during the blooming season of plants from March toMay (Oh et al 1998) Because airborne pollen can be trans-ported very long distances (Porsbjerg et al 2003 Rousseauet al 2008) airborne pollen is not only a local problem butalso a regional and intercontinental problem However stud-ies regarding the impact of pollen on the aerosol chemicalcomposition at the Gosan site are rare (Jung and Kawamura2011)

Stable carbon isotopic compositions (δ13C) of total carbon(TC) are very useful for investigating sources and the long-range atmospheric transport of organic aerosols (Cachier etal 1986 Narukawa et al 2008 Miyazaki et al 2010) Onthe basis of theδ13C values and Na+TC ratios Narukawa etal (2008) estimated the contribution of marine organic mat-ter in TC in the high Arctic at Alert during spring Miyazakiet al (2010) estimated marine-derived carbon in TC overthe western North Pacific usingδ13C values Also Kawa-mura and Watanabe (2004) developed a novel method forcompound specific carbon isotope compositions of dicar-boxylic acids and related compounds using gas chromatogra-phyisotope ratio mass spectrometry (GCirMS) Since thenδ13C values of dicarboxylic acids and related compoundshave successfully been used to assess the extent of photo-chemical processing of aerosols during their long-range at-mospheric transport (Wang and Kawamura 2006 Aggarwaland Kawamura 2008)

In this study elevated concentrations of total carbon wereoften found in the total suspended particulate (TSP) sam-ples collected at the Gosan site during spring of 2007 and2008 Using remote monitoring and comprehensive in-situchemical andδ13C analyses we categorize the carbon emis-sion episodes to three types long-range transport anthro-pogenic pollution (LTP) from Asian continent Asian dust(AD) accompanying with the LTP and local airborne pollenepisodes We discuss the measuredδ13C values and the ther-mal evolution patterns obtained from heating of carbona-ceous particles Based on a carbon isotope mass balanceequation we quantify the TC fraction of local pollen Us-ing HCl fume treatment of the dust-enriched samples wequantify carbonate carbon in the TC and discuss removal ofcarbonate via the reaction with acidic gases during the long-range atmospheric transport

2 Samples and methods

TSP sampling was carried out at the Gosan supersite(3317 N 12610 E) on Jeju Island located approximately100 km south of the Korean peninsula (Fig 1) over suc-cessive periods that integrated 2ndash5 days from 23 March to1 June 2007 and from 16 to 24 April 2008 TSP samples werecollected on pre-combusted quartz fiber filters (20times 25 cm)using a high volume air sampler (Kimoto AS-810) at a flowrate of 50 m3 hrminus1 on the rooftop of a trailer (sim3 m above theground) Before and after sampling the filter samples werestored in clean glass jars (150 ml) with a Teflon-lined screwcap atminus20C prior to analysis Field blank filters were col-lected every month Hourly PM10 mass data was obtainedfrom the National Institute of Environment Research at theGosan Observatory Measurements of wind direction andwind speed at 10 m above ground level were obtained fromthe Korea Meteorological Administration at the Gosan Ob-servatory A total of 32 filter samples were analyzed in thisstudy

21 Airborne pollen and tangerine fruit samples

Three types of pollen samples were prepared and ana-lyzed in this study Two authentic standard pollen sam-ples from Japanese cedar (Pollencedar) and Japanese cy-press (Pollencypress) were obtained from the WAKO Chemi-cal Co (product No 168-20911 for Japanese cedar and 165-20921 for Japanese cypress) Additionally airborne pollen(PollenGosan) which mainly originated from Japanese cedartrees planted around tangerine farms on Jeju Island wereseparated from an aliquot of the TSP filter sample collectedduring a severe pollen episode (KOS751 16ndash21 April 2008)An aliquot (15 cm2) of the KOS751 filter sample was placedin a glass vial (50 ml) with a Teflon-lined screw cap and thepollen grains were separated by mild vibration using an au-tomatic vibrator (Iuchi HM-10) for 5 min The separatedpollen grains were then transferred to a pear shape flask forchemical analysis In order to track the sources of the di-carboxylic acids and related compounds and airborne pollentangerine fruit produced from Jeju Island was also preparedand analyzed for dicarboxylic acids and related compounds(Jung and Kawamura 2011) and for carbon isotopic compo-sitions of total carbon oxalic acid and citric acid In orderto prevent possible contamination over the surface of tanger-ine fruit the tangerine surface was mildly washed three timesusing ultra pure organic-free Milli-Q water

22 Organic and elemental carbon analysis

Organic carbon (OC) and elemental carbon (EC) were ana-lyzed with a Sunset carbon analyzer using the thermal-opticaltransmittance (TOT) protocol for pyrolysis correction (Birchand Cary 1996 Miyazaki et al 2009) A 214 cm2 punchof the quartz filter was placed in a quartz boat inside the

Atmos Chem Phys 11 10911ndash10928 2011 wwwatmos-chem-physnet11109112011

J Jung and K Kawamura Carbon episodes at Gosan site 10913

1

Figure 1

Fig 1 Image map of Gosan supersite (3317 N 12610 E) on JejuIsland located approximately 100 km south of the Korean peninsula

thermal desorption chamber of the analyzer The OC and ECwere respectively determined under prescribed temperatureprotocols in an inert atmosphere (100 He) and in an oxi-dizing atmosphere (He10 O2 mixture) The OC detectedin temperature steps of 300C 450C 600C and 650Care defined as OC1 OC2 OC3 and OC4 respectively TheEC detected in temperature steps of 550C 625C 700C775C 850C and 870C are defined as EC1 EC2 EC3EC4 EC5 and EC6 respectively The pyrolized organic car-bon (PC) which was converted from OC in the inert mode ofthe analysis was corrected by monitoring the transmittanceof a pulsed He-Ne diode laser beam through the quartz fiberfilter External calibration was performed before the anal-ysis using a known amount of sucrose The detection lim-its of OC and EC which are defined as three times of stan-dard deviation of field blanks were 026 and 001 microgC mminus3However these values are quite small compared to the instru-mentrsquos minimum quantifiable level of 05 microgC mminus3 given bythe manufacturer Therefore 05 microgC mminus3 was considered asthe detection limit of both OC and EC The analytical errorswhich are defined as the ratio of the standard deviation to theaverage value obtained from the triplicate analyses of the fil-ter sample of OC and EC measurements were 5 and 3 respectively

23 Determination of water-soluble inorganic ions

To measure water-soluble inorganic ions an aliquot(201 cm2) of the quartz filter was extracted with 10 mlof the Milli-Q water under ultrasonication (30 min) andthen passed through a disk filter (Millipore Millex-GV045 microm) The concentrations of cations and anions inthe water extracts were measured using an ion chro-matography (Metrohm 761) The sodium (Na+) cal-cium (Ca2+) and ammonium (NH+4 ) were determined us-

ing an Metrosep C2 column (150 mm) with 4 mM tar-taric acid1 mM 26-pyridinedicarboxylic acid as an eluent(flow rate 10 ml minminus1 sample loop volume 200 microl timeeluted 30 min) The nitrate (NOminus3 ) and sulfate (SO2minus

4 )were determined using a Shodex IC SI-90 4E column with18 mM Na2CO317 mM NaHCO3 as an eluent (flow rate12 ml minminus1 sample loop volume 200 microl time eluted15 min) The analytical errors of the water-soluble inorganicions were less than 14 based on the triplicate analysesof filter sample SO2minus

4 and Ca2+ concentrations were cor-rected for sea-salt fraction using Na+ as a sea-salt tracer inthis study It was found thatsim95 of SO2minus

4 andsim92 ofCa2+ were attributed to non-sea-salt SO2minus

4 (nss-SO2minus

4 ) andnon-sea-salt Ca2+ (nss-Ca2+) All the concentrations of OCEC and water-soluble inorganic ions reported here were cor-rected against the field blanks

24 Total carbon total nitrogen and carbonisotope analysis

Total carbon (TC) and total nitrogen (TN) were measured byan elemental analyzer (EA) (Carlo Erba NA 1500) whereasstable carbon isotope (δ13C) analyses were conducted usingthe same EA interfaced to an isotope ratio mass spectrometer(irMS) (Finnigan MAT Delta Plus) (Kawamura et al 2004)An aliquot of filter sample (201 cm2) was cut using a circu-lar cutter with a diameter of 16 cm and placed in a tin cupintroduced into the EA and then were oxidized in a combus-tion column packed with chromium (III) oxide at 1020CNitrogen oxides coming from the combustion column werereduced to molecular nitrogen (N2) in a reduction columnpacked with metallic copper at 650C The derived N2 andcarbon dioxide (CO2) were isolated using a gas chromato-graph (GC) installed inline with the EA and then measuredwith a thermal conductivity detector An aliquot of CO2gases was then introduced to the irMS through an interface(ThermoQuest ConFlo II) The carbon isotopic composition(δ13C) relative to the Pee Dee Belemnite (PDB) standard wascalculated using the standard isotopic conversion equation asfollows

δ13C(permil) =

[(13C12C)sample

(13C12C)standardminus1

]times1000

External calibration was conducted using known amountsof acetanilide in order to calculate mass concentrations ofTC and TN andδ13C of TC (δ13CTC) 5 standards rang-ing from 02 mg to 06 mg of acetanilide were prepared andanalyzed by the EA-irMS Acetanilide was purchased fromThermo Electron with aδ13CTC of minus2726 permil The mass con-centrations andδ13CTC values reported here were correctedagainst the field blanks using isotope mass balance equations(Turekian et al 2003) The blank levels of TC and TN massconcentrations were 20 and 13 of the measured con-centrations respectively The analytical errors for TC and

wwwatmos-chem-physnet11109112011 Atmos Chem Phys 11 10911ndash10928 2011


TN mass concentrations based on the triplicate analyses ofthe filter sample were 23 and 52 respectively Thestandard deviation ofδ13CTC based on the triplicate analysesof the filter sample were 003 permil and its analytical error was01 In order to remove carbonate carbon from the dust-enriched filter samples aliquots of the samples were treatedwith HCl fumes by the method described by Kawamura etal (2004) and Kundu et al (2010) Each filter cut from thedust-enriched samples was placed in a 50 ml glass vial andexposed to the HCl fumes overnight in a glass desiccator(10 l) The HCl treated filter samples were then analyzed forTC mass andδ13CTC as described above

Theδ13C values of the water-soluble fraction of the pollensamples were also analyzed in this study as follows Aliquotsof the pollen samples were extracted with the Milli-Q wa-ter under ultrasonication (5 mintimes 3 times) and then passedthrough a disk filter (Millipore Millex-GV 045 microm) to re-move water-insoluble suspended particles and filter debrisAfter concentrating the filtered water extracts using a rotaryevaporator the samples were applied with a micro glass sy-ringe to the prebaked quartz filters and then dried overnightusing silica gel in a glass desiccator Finally the preparedsamples were analyzed forδ13CTC using the same techniqueused for the bulk filter sample analysis The blank level of TCmass was less than 20 of the measured mass Theδ13CTCvalues reported here were corrected against the blank usingisotope mass balance equations (Turekian et al 2003)

25 Stable carbon isotopic compositions of the majordicarboxylic acids and related compounds

The δ13C values of the major dicarboxylic acids and citricacid were measured using the method developed by Kawa-mura and Watanabe (2004) Diacids and citric acid in theTSP samples were reacted with 14 BF3 in 1-butanol at100C for 60 min to form butyl esters (Kawamura 1993Jung and Kawamura 2011) After an appropriate amount ofinternal standard (n-alkane C13) was spiked to an aliquot ofthe derivatives13C12C ratios of the esters were measuredusing a GC (Hewlett-Packard HP6890) interfaced to theirMS Peak identification was performed by comparing theGC-irMS retention times with those of authentic standardsIdentification of the esters was also confirmed by mass spec-tra of the sample using a GC-mass spectrometry (ThermoTrace MS) system (Jung and Kawamura 2011) Theδ13Cvalues of the esters relative to the PDB standard were cal-culated from13C12C ratio of the C13 standard and itsδ13CThe internal standard was purchased from the WAKO Chem-ical Co and itsδ13C is minus2724 permil (Kawamura and Watan-abe 2004) Theδ13C values of free organic acids were thencalculated by an isotopic mass balance equation using themeasuredδ13C of the derivatives and the derivatizing agent(1-butanol) (Kawamura and Watanabe 2004) Each samplewas analyzed in duplicate and the averageδ13C values of thequantified compounds are reported Prior to actual sample

analysis we confirmed that theδ13C values of the workingstandards (a mixture of normal C16ndashC30 alkanes) were equiv-alent to the theoretical values within an analytical error of02 permil Around 055ndash283 ng of the working standards wereinjected to the GC-irMS The working standards were pur-chased from the biogeochemical laboratories at Indiana Uni-versity and theirδ13C ranged fromminus3324 permil tominus2849 permil(httpphpindianaedusimaschimmen-Alkaneshtml) In thispaper we reportδ13C values for oxalic and citric acids

3 Air mass backward trajectories and aerosol opticalthickness (AOT)

Air mass backward trajectories and satellite aerosol opti-cal thickness (AOT) can be utilized to characterize poten-tial source regions and transport pathways of air masses Airmass backward trajectories that ended at the measurementsite were computed for 500 m height above ground level us-ing the HYSPLIT (HYbrid Single-Particle Lagrangian Tra-jectory) backward trajectory analysis (Draxler and Rolph2011 Rolph 2011) All calculated backward trajectories ex-tended 96 h backward with a 1-h interval Up to 20 errorsof the traveled distance are typical for those trajectories com-puted from analyzed wind fields (Stohl 1998) Thus calcu-lated air mass pathways indicate the general airflow patternrather than the exact pathway of an air mass Because a long-range transported haze layer was frequently observed be-tweensim02 km and 3 km elevation over the Korean peninsula(Noh et al 2009 Yoon et al 2008) even though aerosolsampling was conducted atsim3 m height above the groundlevel backward trajectories that ended at 500 m height wereused in this study by assuming complete vertical mixing be-low a boundary layer height Because the Gosan site is lo-cated at the western edge of Jeju Island and there are nohigh mountains within several kilometers of the site localgeographical conditions rarely affect the backward trajectorycalculation

AOT values retrieved by the new version V52 of theNASA MODIS (Moderate Resolution Imaging Spectro-radiometer) algorithm called Collection 005 (C005) (Levyet al 2007a b) were used in this study AOT data that arepart of the MODIS TerraAqua Level-2 gridded atmosphericdata product are available on the MODIS web site (httpmodisgsfcnasagov) Remer et al (2005) reported the as-sociated errors of MODIS AOT withplusmn(005 + 015middot AOT)andplusmn(003 + 005middot AOT) over land and ocean respectively

AerosolAngstrom exponent (α) calculated from the AOTvalues at 440 and 870 nm measured by a sunphotometer wereobtained from the Gosan AERONET site (httpaeronetgsfcnasagov) Theα represents the wavelength (λ) dependenceof AOT (=minusd logAOTd logλ) A smallα indicates the pres-ence of particles with a large size and vice versa In orderto estimateα values during cloudy days of the sampling pe-riods at the Gosan siteα values were also obtained from



the Gwangju AERONET site (12650prime E 35134prime N) whichis locatedsim200 km north of the Gosan site During theepisodic periods excellent agreement of theα values wasobserved between the two sites with the regression slope of104 (R2 = 095) Thus we employed surrogateα values ob-tained from the Gwangju AERONET site for the cloudy daysof the sampling periods (KOS606 608 and 751) at the Gosansite (Table 1) Cloud-screened and quality-assured Level 20sunphotometer data determined by the AERONET algorithm(Dubovik and King 2000) were used in this study

4 Results and discussion

41 Categorization of the carbon emission episodes

The frequency distribution of TC mass concentrations at2 microgC mminus3 increments is shown in Fig 2 with a peak value inthe rage of 6ndash8 microgC mminus3 Gaussian fit of the frequency dis-tribution showed a peak center at 72 microgC mminus3 with the widthof 31 microgC mminus3 representing background TC mass distribu-tions during the entire sampling periods The Gaussian fitwas clearly separated from the total TC distribution with athreshold value of 10 microgC mminus3 (Fig 2) Thus this study de-fined the carbon episode as an average mass concentration ofTC gt10 microgC mminus3 Three carbon episodes such as long-rangetransported pollutants (LTP) Asian dust accompanying withLTP (AD + LTP) and local pollen episodes were observed asmarked in Fig 3 and summarized in Tables 1ndash2

Subdivision of the carbon episodes was conducted as fol-lows Pollen episodes were identified by the elevated con-centrations of citric acid and pollen in the TSP samples asdescribed by Jung and Kawamura (2011) Even though theTC value (75 microgC mminus3) for the KOS612 sample during thepollen episodes was lower than the threshold value we in-cluded this sample to the carbon episode for comparison ofdifferent episodes LTP episodes were identified by the el-evated concentrations of nitrate and sulfate during the car-bon episodes AD + LTP episodes were identified by theelevated concentrations of Ca2+ and low aerosolAngstromexponents as well as the elevated concentrations of nitrateand sulfate during the carbon episodes Mass concentra-tions of nitrate and sulfate during the LTP episodes weremore than 2 times higher than those during the pollen andnon-episodes (Table 2) indicating strong influences of an-thropogenic pollutants Air mass backward trajectories dur-ing the LTP episodes mainly originated from the eastern andnortheastern parts of China (Fig 4a) implying that pollu-tion aerosols emitted from these parts of China had an im-pact on the measurement site The MODIS AOT andα val-ues during the selected days (7 April 2007 KOS606 and25 April 2007 KOS614) of the LTP episodes showed thatsevere haze lingered over the eastern part of China extend-ing over the Yellow Sea and our measurement site (Fig 5andashd) A high AOTgt 10 and highα gt10 on 7 April 2007 and

2

Figure 2

0

2

4

6

8

10

12

0 10 20 30TC (μgC m-3)

Num

ber o

f dat

a

Fig 2 Frequency distribution of TC mass concentrations duringthe entire sampling period Number of data in each bin was cal-culated with 2 microgC mminus3 increment Gaussian fit of the frequencydistribution showed a peak center at 72 microgC mminus3 with the width of31 microgC mminus3

25 April 2007 indicated the presence of anthropogenic hazeaerosols Air mass backward trajectories for the KOS606and KOS614 samples during the LTP episodes are shown asyellow trajectories in Fig 4a implying that air masses forthe KOS606 and KOS614 samples mainly originated fromthe eastern part of China and migrated to the sample siteThe air mass backward trajectories generally showed similartransport pattern during each sampling period Theα valuesfor the KOS621 and KOS623 samples showed values withan average of 065ndash086 and lower than those for KOS606614 and 619 samples (avg 128ndash140) (Table 1) indicatingthat much larger particles had impacted on the KOS621 andKOS623 samples Air mass trajectories for the KOS621 andKOS623 samples shown in red trajectories in Fig 4a mainlyoriginated from the areas between Beijing and the northernpart of China

Two AD + LTP episodes were observed during 30 Marchndash2 April 2007 (KOS603) and 25ndash28 May 2007 (KOS627)A high AOTgt 10 and low α lt 04 during the selecteddays (31 March 2007 and 25 May 2007) of the AD + LTPepisodes clearly showed the presence of dust plumes overthe Yellow Sea (Fig 5endashh) Theα obtained from theAERONET AOT also showed low values (037plusmn 006)during the KOS627 AD + LTP episode (Table 1) indi-cating large size particles in the dust plumes Duringthe AD + LTP episodes average PM10 mass concentrationswere 362 and 157 microg mminus3 with peak values of 3704 and286 microg mminus3 for the KOS603 and KOS627 sampling peri-ods respectively The elevated concentrations of nss-Ca2+

(avg 75plusmn 02 microg mminus3) nitrate (avg 160plusmn 76 microg mminus3) andsulfate (avg 325plusmn 151 microg mminus3) in the KOS603 andKOS627 filter samples supported the presence of dust



Table 1 Specifics of sampling PM10 mass concentration non-sea-salt calcium ion (nss-Ca2+) andAngstrom exponent at the Gosan siteduring the carbon episodes in spring of 2007 and 2008

Sample ID Category1 Period PM10 mass2 (microg mminus3) nss-Ca2+ (microg mminus3) Angstrom exponent3

KOS603 AD + LTP 30 Marndash2 Apr 2007 362 (3704) 76 NAKOS606 LTP 6ndash9 Apr 2007 70 (158) 16 140plusmn 018KOS608 Pollen 11ndash13 Apr 2007 47 (86) 15 126plusmn 021KOS609 Pollen 13ndash16 Apr 2007 50 (120) 08 103plusmn 024KOS610 Pollen 16ndash18 Apr 2007 33 (52) 06 NAKOS611 Pollen 18ndash20 Apr 2007 38 (54) 05 137plusmn 037KOS612 Pollen 20ndash23 Apr 2007 37 (80) 05 NAKOS613 Pollen 23ndash25 Apr 2007 61 (94) 13 057plusmn 006KOS614 LTP 25ndash27 Apr 2007 129 (180) 55 128plusmn 015KOS619 LTP 7ndash9 May 2007 94 (168) 27 140plusmn 007KOS621 LTP 11ndash14 May 2007 74 (126) 31 086plusmn 046KOS623 LTP 17ndash18 May 2007 107 (148) 47 065plusmn 013KOS627 AD + LTP 25ndash28 May 2007 157 (286) 73 037plusmn 006KOS751 Pollen 16ndash21 Apr 2008 59 (184) 04 118plusmn 008KOS752 Pollen 21ndash24 Apr 2008 53 (82) 04 NA

1AD + LPT Asian dust plus long-range transported pollution episode Pollen pollen episode2Average (Maximum)3Angstrom exponents during the KOS606 608 and 751 sampling periods were obtained from the Gwangju AERONET site which is located atsim200 km north of the Gosan site

Table 2 Concentrations of chemical compositions and stable carbon isotopic composition (δ13C) in the total suspended particle (TSP)samples collected at the Gosan site during the carbon episodes

ParametersAD + LTP LTP Pollen Non-episodes

AVG plusmn STD (MINndashMAX)

δ13C (permil) minus218plusmn 20 (minus233ndashminus204) minus233plusmn 03 (minus235ndashminus230) minus252plusmn 09 (minus262ndashminus235) minus234plusmn 08 (minus249ndashminus217)Total carbon (TC) (microgC mminus3) 16plusmn 46 (12ndash19) 15plusmn 60 (11ndash24) 16plusmn 67 (75ndash28) 71plusmn 17 (37ndash98)Organic carbon (OC) (microgC mminus3) 13plusmn 45 (93ndash16) 12plusmn 47 (69ndash19) 14plusmn 82 (69ndash27) 63plusmn 20 (34ndash105)Elemental carbon (EC) (microgC mminus3) 38plusmn 02 (36ndash40) 37plusmn 15 (22ndash61) 25plusmn 15 (13ndash60) 20plusmn 083 (055ndash36)OCEC ratio 33plusmn 14 (23ndash43) 32plusmn 04 (27ndash38) 58plusmn 26 (38ndash12) 35plusmn 11 (21ndash62)Total nitrogen (TN) (microgN mminus3) 77plusmn 40 (48ndash11) 11plusmn 82 (49ndash25) 48plusmn 15 (23ndash70) 46plusmn 15 (15ndash76)TCTN ratio 25plusmn 19 (12ndash39) 16plusmn 058 (096ndash23) 35plusmn 12 (18ndash53) 17plusmn 058 (11ndash30)nss-Ca2+ (microg mminus3) 75plusmn 02 (73ndash76) 35plusmn 16 (16ndash55) 08plusmn 04 (04ndash15) 14plusmn 076 (016ndash25)NH+

4 (microg mminus3) 50plusmn 39 (22ndash77) 89plusmn 68 (31ndash197) 34plusmn 13 (16ndash51) 40plusmn 16 (12ndash61)NOminus

3 (microg mminus3) 160plusmn 76 (107ndash214) 181plusmn 139 (79ndash426) 64plusmn 22 (40ndash100) 69plusmn 33 (24ndash137)

nss-SO2minus

4 (microg mminus3) 325plusmn 151 (218ndash432) 362plusmn 219 (165ndash708) 132plusmn 46 (74ndash203) 175plusmn 67 (58ndash280)Citric acid (ng mminus3) 46plusmn 17 (34ndash58) 42plusmn 35 (064ndash92) 117plusmn 123 (20ndash320) 35plusmn 52 (017ndash18)

lowastCitric acid data was obtained from Jung and Kawamura (2011)

particles and anthropogenic pollutants (Tables 1ndash2) Airmass backward trajectories during the AD + LTP episodesclearly showed that dust particles mainly originated from theNei Mongol desert in China and were transported across theYellow Sea to the measurement site (Fig 4b)

Pollen episodes which were mainly caused by pollen fromJapanese cedar trees planted around tangerine farms wereobserved at the Gosan site during mid- to late April of 2007and 2008 (Jung and Kawamura 2011) Jung and Kawa-mura (2011) reported the enhanced concentrations of citricacid which may be directly emitted from tangerine fruitduring the pollen episodes likely adsorbed on the pollen

from Japanese cedar trees and then transported to the Gosansite Mass concentrations of citric acid during the pollenepisodes (range 20ndash320 microg mminus3) were several dozen timeshigher than the LTP (range 064ndash92 microg mminus3) and AD + LTPepisodes (range 34ndash58 microg mminus3) and non-episodes (range017ndash18 microg mminus3) Identification of pollen episodes was basedon daily human observation of pollen blowing and the micro-scopic image of pollens collected in the TSP samples (Jungand Kawamura 2011) A total of 8 pollen-enriched TSPsamples were collected during 11ndash23 April 2007 and 16ndash24 April 2008 (Table 1)



3

Figure 3

Mar 20 Apr 3 Apr 17 May 1 May 15 May 29 Apr 15 Apr 2502460

102030

0100200

1200

-26-24-22-20

0204060300320

TCT

N ra

tio

2007 2008

TC TN

TC a

nd T

N(m

g m

-3)

AD LTP

PM10

mas

s(m

g m

-3)

δ13C

(permil)

Citri

c ac

id(n

g m

-3)

Pollen

Fig 3 Temporal variations of mass concentrations of PM10 citric acid total carbon (TC) and total nitrogen (TN) carbon isotope ratios ofTC (δ13CTC) and TCTN ratios at the Gosan site during 23 March to 1 June 2007 and 16ndash24 April 2008 LTP represents the long-rangetransported pollution episode AD + LTP and pollen represent Asian dust companying LTP and airborne pollen episodes respectively Massconcentration of citric acid is obtained from Jung and Kawamura (2011)

4

Figure 4

Korea

JapanChina

a) LTP

Nei Mongol desert

d) Non-episode c) Pollen

b) AD+LTP

Fig 4 HYSPLIT air mass backward trajectories arriving at theGosan site during(a) the LTP (b) AD + LTP (c) pollen and(d)non-episodes Air mass backward trajectories that ended at the mea-surement site were computed for 500 m height above ground levelYellow trajectories represent the KOS606 and KOS614 samplingperiods while red ones represent the KOS621 and KOS623 sam-pling periods See Table 1 for specifics of each sampling period

Temporal variations of mass concentrations of citric acidand stable carbon isotopic composition during the pollenepisodes showed gradual transition from the non-episodesto the pollen episodes (Fig 3) Average mass concentra-tions of NOminus

3 and SO2minus

4 during the pollen episodes weresim2ndash3 times lower than those during the LTP and AD + LTPepisodes (Table 2) implying that a relatively low impactof anthropogenic emissions from the Asian continent dur-ing the pollen episodes Since it was cloudy during mostof the pollen episodes (Jung and Kawamura 2011) only afew MODIS images are available MODIS AOT andα val-ues during the selected day (19 April 2008) of the pollenepisodes showed relatively low aerosol loadings over Koreanpeninsula (Fig 5i j) Air mass backward trajectories dur-ing the pollen episodes mainly originated from the northernpart of China and the western North Pacific Ocean (Fig 4c)Dominant wind directions during the pollen episodes werenortherly and southeasterly Relatively low wind speed wasobserved for southeasterly winds with a median of 51 m sminus1

(range 03ndash134 m sminus1) compared to northerly winds with amedian of 71 m sminus1 (range 07ndash205 m sminus1)

Average mass concentrations of TC during the non-episodes (avg 71plusmn 17 microgC mminus3) weresim2 times lower thanthose during the carbon episodes (Table 2) Averagemass concentrations of nss-Ca2+ during the non-pollenepisodes (avg 14plusmn 076 microg mminus3) weresim5 times lower thanthose during the AD + LTP episodes (avg 75plusmn 02 microg mminus3)Mass concentrations of citric acid during the non-episodes(range 017ndash18 microg mminus3) were several dozen times lower than



5

Figure 5

(b) Aringngstroumlm exponent (LTP 7 Apr 2007)(a) AOT (LTP 7 Apr 2007)

(d) Aringngstroumlm exponent (LTP 25 Apr 2007)(c) AOT (LTP 25 Apr 2007)

(f) Aringngstroumlm exponent (AD+LTP 31 Mar 2007)(e) AOT (AD+LTP 31 Mar 2007)

6

(h) Aringngstroumlm exponent (AD+LTP 25 May 2007)(g) AOT (AD+LTP 25 May 2007)

(j) Aringngstroumlm exponent (Pollen 19 April 2008)(i) AOT (Pollen 19 April 2008)

Fig 5 MODIS aerosol optical thickness (AOT) andAngstrom exponent during the selected days (7 April 2007 and 25 April 2007) of theLTP episodes (Fig 5andashd) Values during the selected days of the AD + LTP (31 March 2007 and 25 May 2007) and pollen (19 April 2008)episodes are shown in Fig 5endashh and Fig 5indashj respectively



the pollen episodes (range 20ndash320 microg mminus3) and almost nopollen grains were observed from the microscopic image ofthe TSP samples These results indicated that airborne pol-lens and dust particles rarely had an impact on the TSP sam-ples during the non-episodes Air mass backward trajectoriesduring the non-episodes mainly originated either from thenorthern part of China or the western North Pacific Ocean(Fig 4d) indicating that anthropogenic pollutants emittedfrom the eastern part of China had rarely impacted on theTSP samples during the non-episodes

42 TC and TN mass concentrations during thecarbon episodes

Similar TC mass concentrations were obtained dur-ing the three carbon episodes with averages ranging15plusmn 60 microgC mminus3 to 16plusmn 67 microgC mminus3 (Table 2) Simi-lar amounts of OC and EC as well as TC during theLTP and AD + LTP episodes as shown in Table 2 in-dicated that the AD + LTP episodes in spring are fre-quently accompanied with anthropogenic pollutants emit-ted from the industrial regions of East China and North-east China However relatively high mass concen-trations of TN (avg 11plusmn 82 microgN mminus3) were obtainedduring the LTP episodes followed by the AD + LTPepisodes (avg 77plusmn 40 microgN mminus3) and the pollen episodes(avg 48plusmn 15 microgN mminus3) Much higher TCTN mass con-centration ratios (avg 35plusmn 12 range 18 to 53) were ob-served during the pollen episodes than in those during theLTP episodes (avg 16plusmn 058 range 096 to 23) mainly dueto the enhanced organic carbon mass of the airborne pollenThe highest OCEC ratios (avg 58plusmn 26 range 38 to 12)were also obtained during the pollen episodes confirmingthe enhanced organic carbon mass of the airborne pollen

Excellent correlation (R2 = 095) was obtained betweenmass concentrations of TN and TC during the LTP peri-ods (Fig 6a) implying similar sources of TN and TC Astrong correlation between mass concentrations of TN andTC was also obtained during the LTP plus non-episodic pe-riods (R2 = 082) (Fig 6b) suggesting that aerosols duringthe non-episodic periods in spring might be influenced bythe LTP aerosols from the Asian continent However poorcorrelations of TN versus TC mass concentrations were ob-tained during the pollen episodes (Fig 6a) Even though sim-ilar levels of TN were obtained during the pollen episodesTC values were highly variable ranging 75 to 28 microgC mminus3

mainly due to different strengths of pollen transport anddifferent meteorological conditions (Doskey and Ugoagwu1989 Puc and Wolski 2002 Palacios et al 2007) TC andTN mass concentrations obtained during the LTP plus non-episodic periods are compared to those from previous studiesin the Asian continent (Fig 6b) TC and TN concentrationswere obtained from the Hua mountain site in China in winter(size cut PM10 sampling flow rate 100 l minminus1 integra-tion time 10 h) (Li et al 2011) and three cities in China

7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)

Non-episodeLTPPollenAD

(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)

Asian continentNon-episode+LTP

(b)

Fig 6 (a) Scatter plot between mass concentrations of TN ver-sus TC during the entire sampling period Filled blue circle reddiamond and brown rectangular represent the LTP pollen andAD + LTP episodes respectively(b) Mass concentrations of TNand TC during the non-episode plus LTP episode are compared tothose from previous studies on the Asian continent TN and TCvalues on the Asian continent were obtained from urban areas ofChina Shanghai in winter and spring (size cut PM25 samplingflow rate 04 l minminus1 integration time weekly) (Ye et al 2003)Nanjing in winter (PM25 1110 l minminus1 12 h) (Yang et al 2005)Baoji in spring (PM10 100 l minminus1 8 h) (Wang et al 2010) as wellas the Hua mountain site in winter (PM10 PM10 100 l minminus1 10 h)(Li et al 2011) The analytical errors for TC and TN mass concen-trations at the Gosan site were 23 and 52 respectively

Shanghai in winter and spring (PM25 04 l minminus1 weekly)(Ye et al 2003) Nanjing in winter (PM25 1110 l minminus112 h) (Yang et al 2005) Baoji in spring (PM10 100 l minminus18 h) (Wang et al 2010) TN values were calculated from thenitrogen mass in particulate nitrate and ammonium Kunduet al (2010) reported thatsim97 of TN at the Gosan sitewas explained by nitrate and ammonium nitrogen Thus thecontribution of organic nitrogen to TN is negligible at theGosan site Interestingly the slope (283) of TN versus TCmass concentrations (avg TCTN mass ratio = 32plusmn 063) in



the Asian continent was much higher than that (083) forthe LTP plus non-episode samples (avg TCTN mass ra-tio = 17plusmn 057) at the Gosan site

Using the regression slope of TN versus TC mass concen-trations during the LTP plus non-episodic periods that mayrepresent normal atmospheric condition at the Gosan site inspring the contribution of the airborne pollen carbon in TCcan be roughly calculated as follows

TCpollen(microgC mminus3) = TC(microgC mminus3)

minus(083middotTN (microgN mminus3) + 379) (1)

The contribution of the airborne pollen carbon to TC wasroughly estimated to be 20 to 71 with an average of46plusmn 19

TCTN mass concentration ratio (39) in the strongAD + LTP episode sample (KOS603) was much higher thanthat (12) in the weaker AD + LTP episode sample (KOS627)(Fig 6a) The enhanced mass concentration of TC in theKOS603 sample may be in part attributed to the presenceof carbonate carbon in dust particles which wasnrsquot removedcompletely via the reaction with nitrogen dioxide (NO2) ni-tric acid (HNO3) and sulfur dioxide (SO2) (Zhang et al1994 Mamane and Gottlieb 1989 Underwood et al 2001)during the long-range atmospheric transport The enhancedcarbonate carbon in TC during the severe AD + LTP episodewill be discussed in detail in Sect 43

43 Stable carbon isotopic compositions of TC duringthe carbon episodes

Theδ13C values of TC (δ13CTC) during the carbon episodesare plotted as a function of TC mass concentrations in Fig 7aand are summarized in Table 2 Because TC mass weightedaverageδ13CTC values during the different episodes agreedwithin 3 permil to the arithmetic means ofδ13CTC we usedthe arithmetic means when comparing different episodesThe δ13CTC values during the LTP episodes ranged fromminus235 permil to minus230 permil with an average ofminus233plusmn 03 permilSimilar δ13CTC values were observed regardless of TC massconcentrations during the LTP episodes (Fig 7a) How-ever theδ13CTC values during the pollen episodes weremore negative ranging fromminus262 tominus235 permil with an aver-age ofminus252plusmn 09 permil Theδ13CTC values for the AD + LTPepisodes are relatively high rangingminus233 tominus204 permil withan average ofminus218plusmn 20 permil These results suggested thatδ13CTC can be utilized as an indicator of the possible sourcesof the carbon episodes at the Gosan site with relativelylow values during the pollen episodes compared to thoseduring the LTP and AD + LTP episodes Theδ13CTC dur-ing the strong AD + LTP episode (KOS603) was higher thanduring the weaker AD + LTP episode (KOS627) as shownin Fig 7a In order to quantify the effect of carbonate indust particle on the TC mass andδ13CTC measurementsthe HCl fume treated filter samples during the AD + LTPepisodes were also analyzed for TC mass andδ13CTC It

8

Figure 7

0

4

8

12

16

20

Total REM RMD Total REM RMDTC

(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)

Non-episode PollenLTP AD

(a)

KOS603

KOS627

Fig 7 (a) δ13CTC values as functions of TC mass concentra-tions and(b) TC mass concentrations andδ13CTC values in theaerosol samples collected during the AD + LTP episodes before andafter HCl fume treatment REM and RMD represent the remain-ing carbon and the removed carbon after the HCl fume treatmentrespectively KOS603 and KOS627 samples were collected dur-ing the strong AD + LTP (30 Marchndash2 April 2007) and less strongAD + LTP episodes (25ndash28 May 2007) respectively The standarddeviation of δ13CTC was sim003 permil and its analytical error wassim01

was clearly observed that the TC mass concentration andδ13CTC in the KOS603 sample decreased from 188 microgC mminus3

to 147 microgC mminus3 and fromminus204 permil tominus221permil after the HClfume treatment However those in the KOS627 sample wereinvariant before and after the HCl treatment (Fig 7b) Theremoved carbon (RMD-C) that was calculated from the dif-ference between TC mass concentrations before and after theHCl fume treatment was found to be 41 microgC mminus3 for theKOS603 sample

In order to characterize the carbonate carbon in the RMD-C the δ13C of the RMD-C (δ13CRMDminusC) was calculatedusing the isotopic mass balance equation as follows (Kawa-mura et al 2004)



δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C

TCmiddotδ13CRMD-C (2)

where REM-C represents the remaining carbon after theHCl fume treatment Isotope equilibrium exchange reactionswithin the inorganic carbon system ldquoatmospheric CO2 ndash dis-solved bicarbonate ndash solid carbonaterdquo lead to an enrichmentof 13C in carbonates (Hoefs 1997) Thus carbonate car-bon is isotopically heavy withδ13C values of around 0 permil(Hoefs 1997 Kawamura et al 2004) Theδ13CRMDminusC inthe KOS603 sample was calculated asminus141 permil This valuewas higher than that of the REM-C while much lower thanthose (minus13 tominus03 permil) of the standard Asian dust samplescollected from the Gunsu Province China (Kawamura et al2004) indicating that not only carbonate carbon but alsovolatile organic acids adsorbed on aerosol particles were re-moved by the HCl fume treatment Kawamura et al (2004)suggested that low molecular weight organic acids such asformic acetic and oxalic acids are possible candidates forthe removed volatile and semi-volatile organic acids By as-suming that theδ13C of the removed organic acids has thesameδ13C of the REM-C carbonate carbon in the KOS603sample was roughly estimated as 15 microgC mminus3 (8 in TC)using the isotopic mass balance Eq (2) and theδ13C of theremoved carbonate ofminus03 permil by Kawamura et al (2004)

Calcium carbonate in dust particle reacts with nitrogendioxide (NO2) nitric acid (HNO3) and sulfur dioxide (SO2)to produce calcium nitrate and calcium sulfate (Zhang et al1994 Mamane and Gottlieb 1989 Underwood et al 2001)The presence of carbonate in the KOS603 sample indicatedinsufficient amounts of gas phase NO2 HNO3 and SO2to remove all carbonate during the long-range atmospherictransport The much higher TCTN ratio during the strongAD + LTP episode (KOS603) was also attributed to the re-maining carbonate carbon However the negligible amountof carbonate in the KOS627 sample indicated that most of itscarbonate was removed via the reaction with NO2 HNO3and SO2 during its long-range atmospheric transport Therelatively high TN mass in the KOS627 sample also sup-ported the efficient removal of carbonate via reaction withHNO3

44 Impact of airborne pollen on TC measurements

441 δ13C values of TC in airborne pollenand tangerine fruit

A much lowerδ13CTC value was obtained in the PollenGosan(minus280 permil) than in the authentic standard pollensminus254 permilfor the Pollencedarandminus233 permil for the Pollencypress(Fig 8and Table 3) Theδ13CTC in the PollenGosan was slightlylower than the average value (minus268 permil) of 174 different

9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)

Fig 8 δ13CTC values in the airborne pollen collected at the Gosansite (PollenGosan) authentic standard pollens from Japanese cedarand Japanese cypress tangerine fruit and the filter sample collectedduring the severe pollen episode (KOS751) WS represents water-soluble fraction of the samples The error bar represents the stan-dard deviation of the duplicate analyses of the sample

species of pollen samples from C3 plants across the US(Jahren 2004) These differences may be attributed to thegeographical difference of the pollen samples Jahren (2004)reported that theδ13CTC values in the pollens from C3plants varied fromminus302 permil tominus245 permil depending on theirgeographical locations and the types of C3 plants ThePollenGosan was produced on Jeju Island Korea while theauthentic standard pollens were produced on Japan (personalcommunication to Wako Chemical Co) Thus the differ-ences ofδ13CTC values between the PollenGosanand the au-thentic standard pollens may be explained by the differentgeographical conditions Interestingly theδ13CTC in thetangerine peel (minus281 permil) was very similar to that in thePollenGosan(Fig 8 and Table 3) A slightly higherδ13CTCvalue (minus270 permil) was obtained in the tangerine juice thanin the tangerine peel Theδ13C values of the water-soluble fraction of the pollens were found to beminus247 permilminus261 permil andminus250 permil for the PollenGosan Pollencedar andPollencypress respectively (Fig 8 and Table 3) Theδ13CTCvalue for the water-soluble fraction of the PollenGosan washigher than that of the bulk PollenGosan However theδ13Cvalues of the water-soluble fraction in the Pollencedar andPollencypresswere slightly lower than the bulk pollen (Ta-ble 3) These results imply that the water-soluble fractionof the Pollencedarand Pollencypressmight originate from dif-ferent sources and be adsorbed to the pollens

In order to estimate the contribution of the airborne pollencarbon to aerosol TC during the pollen episodes the car-bon masses derived from the airborne pollen were deter-mined using the isotopic mass balance equation in Eq (2)Since most of aerosols at the Gosan site in spring were influ-enced by the long-range transport of anthropogenic aerosols



Table 3 Carbon isotope compositions of TC and oxalic and citricacids in selected samples during the pollen episodes pollens andtangerine fruit

Samplesδ13C (permil)

TC4 Oxalic acid5 Citric acid5

KOS611 minus250 minus166plusmn 003 minus247plusmn 006KOS751 minus262 minus211plusmn 03 minus258plusmn 03KOS752 minus261 minus210plusmn 03 minus259plusmn 03Pollen1Gosan minus280 ndash ndashWS2 PollenGosan minus247 minus166plusmn 008 minus263plusmn 02Pollen3cedar minus254 ndash ndashWS Pollencedar minus261 minus50plusmn 02 ndashPollen3cypress minus233 ndash ndashWS Pollencypress minus250 10plusmn 003 ndashWS Tangerine juice minus270 ndash minus293plusmn 04WS Tangerine peel minus281 minus195plusmn 05 minus274plusmn 04

1PollenGosanrepresent airborne pollens separated from the KOS751 sample2WS represents water-soluble fraction of sample3Pollencedar and Pollencypress represent authentic standard pollens from Japanesecedar and Japanese cypress respectively4The analytical error forδ13C of TC measurement was 23 5Averageplusmn Standard deviation Oxalic acid (COOH)2 Citric acid C3H5O(COOH)3

from the Asian continent as discussed in Sect 42 we as-sumed that carbonaceous particles during the pollen episodeswere mainly from the airborne pollen and the long-rangetransported organic pollutants Thus theδ13CTC in thePollenGosan(minus280 permil) and averageδ13CTC during the LTPepisodes (minus233 permil) were used as two end members to calcu-late the airborne pollen carbon mass (TCpollen) The TCpollenconcentrations were determined to be 04 to 167 microgC mminus3 (4to 62 in TC) with a median of 51 microgC mminus3 (42 ) duringthe pollen episodes (Fig 9) The median value of the TCpollenfraction in TC was quite similar to that obtained using theTN and TC regression approach in Eq (1) Thus it wasfound thatsim42 of TC in the TSP samples at the Gosan sitecould be attributed to the airborne pollens during the pollenepisodes The rest of TC can be explained by long-rangetransported anthropogenic OC and EC Local pollen can en-hance the mass of the organic aerosols and may overestimatethe relevant radiative forcing at the Gosan site These resultscan provide useful information for accurately qualifying andquantifying the impact of the long-range transported pollu-tants from the Asian continent in spring

442 Enhanced mass concentrations of citric acid andtheir sources during the pollen episodes

The cultivating area of tangerines in Jeju Island issim209 km2which accounts forsim11 of the total area of the island Inorder to dissipate the strong winds from the Pacific Oceanall tangerine farms are surrounded by Japanese cedar treesPollen in the air on Jeju Island in April was mainly from

10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total

Fig 9 Temporal variations of mass concentrations of TC and air-borne pollen carbon (TCpollen) determined using the isotope massbalance equation

the Japanese cedar trees (Agricultural Research Institute inJeju special self-governing province personal communica-tion 2011) Because tangerines are widely cultivated in thecoastal areas of Jeju Island the distance of the nearest tan-gerine farms to the sampling site is between several hundredmeters to several tens of kilometers depending on the winddirection Jung and Kawamura (2011) measured the elevatedmass concentrations of atmospheric citric acid (range 20ndash320 ng mminus3) in the TSP samples during the pollen episodesThey postulated that citric acid that may be directly emittedfrom tangerine fruits was likely adsorbed on pollens emit-ted from Japanese cedar trees planted around tangerine farmsand then transported to the Gosan site In order to track thesource and transport mechanism of citric acidδ13CTC val-ues during the pollen episodes were plotted as a function ofthe fraction of citric acid carbon (citric acid-C) in TC mass(Fig 10) It was clearly evident that theδ13CTC decreased ascitric acid-CTC mass ratios increased Because the airbornepollen showed much lowerδ13CTC values than the LTP parti-cles (Tables 2ndash3) the decrease ofδ13CTC with an increase ofcitric acid-CTC ratio demonstrates an increased contributionof airborne pollen to aerosol TC These results indicated thepositive correlation between citric acid and airborne pollenconcentration suggesting that citric acid emitted from tan-gerine fruit might be adsorbed on airborne pollen and thentransported to the Gosan site Divergence of theδ13CTC val-ues at a certain level of the citric acid-CTC ratios as shownin Fig 10 may be explained by different adsorption efficien-cies of citric acid on different pollen and different emissionfluxes of citric acid from tangerine fruit over time The di-vergence of theδ13CTC values also can be explained by thevariability of theδ13CTC of non-pollen carbon



11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05

Citric acid-CTC ratio ()

δ13C

(permil)

Fig 10 δ13CTC values as a function of citric acid carbon (citricacid-C) to TC mass ratios during the pollen episodes

The δ13C values of oxalic and citric acids in the selectedsamples during the pollen episodes authentic standard pol-lens and tangerine fruit are shown in Table 3 Theδ13Cof citric acid (minus274 permil) in the tangerine peel was simi-lar to theδ13CTC in the tangerine peel (minus281 permil) and thePollenGosan (minus280 permil) However theδ13C values of cit-ric acid in the PollenGosan(minus263 permil) and KOS751 samples(minus258 permil) were slightly higher than that in the tangerinepeel The pollen-enriched TSP samples were collected dur-ing spring of 2007 and 2008 whereas tangerine fruit was pro-duced during early winter of 2010 Thus these differencescan be partially explained by seasonal and annual variationsof δ13C of citric acid in the tangerine peel Theδ13C valuesof n-alkanes for individual lipids from the leaves of Quercuscastaneifolia showedsim25 and 52 permil differences for C29 andC31 n-alkanes respectively in autumn leaves compared withleaves sampled at the start of the growing season (Lockheartet al 1997) Thus the seasonal difference further supportsthe assumption that citric acid in the PollenGosanand KOS751samples originated from the tangerine peel (tangerine fruit)and then were transported to the Gosan site after adsorbingto the pollen from Japanese cedar trees planted around tan-gerine farms

The elevatedδ13C values of oxalic acid were obtainedfor the authentic standard pollen samplesminus50 permil for thePollencedarand 10 permil for the Pollencypress These highδ13Cvalues of oxalic acid can be explained by the adsorption ofaged oxalic acid on pollen before becoming air borne Theδ13C value of oxalic acid in the PollenGosanwas higher thanthose in the tangerine peel and the KOS751 sample Jung andKawamura (2011) reported similar amounts of oxalic and cit-ric acids in the tangerine peel suggesting that not only agedoxalic acid but also directly emitted oxalic acid from the tan-gerine peel may be adsorbed on the PollenGosanand trans-ported together

45 Thermal characteristics of carbonaceous particles

451 Thermal evolution pattern of OC during thecarbon episodes

Thermograms of OC and EC analysis for the carbon episodesamples are shown in Fig 11 and the results are summa-rized in Table 4 Unique evolution patterns of OC1 and OC2were obtained depending on the types of the carbon emis-sion episodes However OC3 and OC4 showed similar evo-lution patterns among the carbon episodes as shown in Table4 OC3 fractions in total OC = avg 14 to 20 and OC4fractions = avg 10 to 19 A similar temperature depen-dent EC evolution was obtained during the carbon episodesAround 93 of EC evolved at oven temperatureslt700C39 in EC1 36 in EC2 and 19 in EC3 temperaturesteps Even though sharp increase of the OC4 peak was notobserved in Fig 11 for all samples the OC4 fractions weresimilar to the OC3 fractions as seen in Table 4 This discrep-ancy was attributed to the broad evolution of OC in the OC4temperature step as shown in Fig 11

Thermal evolution patterns of OC during the LTP episodeswere subdivided to two groups relatively higher OC evo-lution in the OC1 temperature step (24plusmn 2 in total OC)than the OC2 temperature step for the KOS606 614 and 619samples and similar OC evolution in the OC1 (16plusmn 2 ) andOC2 temperature steps (21plusmn 2) for KOS621 and KOS623samples These subdivision coincided with the relativelyhighα values for KOS606 614 and 619 samples (avg 128ndash140) relative to those for the KOS621 and KOS623 sam-ples (avg 065ndash086) (Table 1) Air mass trajectories duringthe LTP episodes showed that air masses from East Chinaimpacted more on the KOS606 614 and KOS619 samples(yellow and white trajectories in Fig 4a)

OC2 (21 ) was much higher than OC1 (8 ) duringthe strong AD + LTP episode (KOS603) whereas similaramounts of OC1 (19 ) and OC2 (18 ) were obtained dur-ing the weaker AD + LTP episode (KOS627) Thermal evo-lution patterns of OC during the pollen episodes were clearlycharacterized by higher OC evolution in the OC2 tempera-ture step (29plusmn 6) (Fig 11b and Table 4) Mass concentra-tions of OC1 and OC2 in the TSP samples collected duringthe pollen episodes correlated well withR2 of 077 (Fig 12)

Miyazaki et al (2007) reported that medium molecularweight (gt200 g molminus1) water-insoluble organic species suchas nonacosane docosanol and hexadecanoic acid evolvedmostly in the OC2 temperature step with small fractionsevolving in the OC3 temperature step Sucrose has beenidentified as the major water-soluble organic compound in 15pollen species tested by Hoekstra et al (1992) 91 of theOC in sucrose compounds was evolved at the OC2 tempera-ture step (Miyazaki et al 2007) Thus it was suggested thatmajor fractions of the pollens collected at the Gosan site mayhave medium molecular level organic compounds Thermalevolution patterns of OC during the LTP episodes showed



Table 4 Average mass fractions of OC evolved at each temperature step in total OC and OC2OC1 mass ratios according to categorizedcarbon episodes

CategoryThe mass fraction in total OC [] OC2OC1 mass ratio

OC11 OC21 OC31 OC41 PC2

Pollen 15plusmn 3 29plusmn 6 18plusmn 2 12plusmn 6 26plusmn 7 19plusmn 06LTP case13 24plusmn 2 9plusmn 1 14plusmn 1 13plusmn 1 40plusmn 4 04plusmn 01LTP case24 16plusmn 2 21plusmn 2 20plusmn 7 19plusmn 7 24plusmn 15 13plusmn 03AD + LTP(KOS603) 8 21 18 10 43 27AD + LTP(KOS627) 19 18 16 16 30 10

1OC1 OC2 OC3 and OC4 represent the carbon evolved at a temperature of 300C 450C 600C and 650C respectively2PC represents the pyrolized organic carbon during the OC analysis mode in an inert atmosphere (pure Helium)3LTP case 1 includes KOS606 614 and 619 samples with relatively highAngstrom exponentgt124LTP case 2 includes KOS621 and KOS623 samples with relatively lowAngstrom exponentlt10

12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l

(d) LTP (KOS621 KOS623)

(c) LTP (KOS606 KOS614 KOS619)

(b) Pollen episode (KOS608-KOS613 KOS751 KOS752)

Time (sec)

(a) AD+LTP (KOS603 KOS627)

0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1

Fig 11 Thermograms of carbonaceous particles during the(a)AD + LTP (b) pollen (c) LTP (KOS606 614 and 619) and(d)LTP (KOS621 and KOS623) episodes The analytical errors of OCand EC measurements were 5 and 3 respectively Base line ofeach thermogram was shifted 05 toward the y-axis to better visual-ization of overlaying thermograms

relatively constant OC2 mass concentrations but stronglyvariable OC1 mass concentrations (Fig 12) InterestinglyPC fractions during the LTP episodes negatively correlatedwith OC2 fractions while positively correlated with OC1fractions (Table 4) The thermal evolution pattern of OC dur-ing the strong AD + LTP episode (KOS603) was quite differ-ent from that during weaker AD + LTP episodes (KOS627)The OC2OC1 mass ratio (27) during the strong AD + LTPepisode showed much a higher value than that during theweaker AD + LTP episodes (10) Additionally more PCwas formed in the sample for the strong AD + LTP episode(43 ) than that for the weaker AD + LTP episodes (30 )Miyazaki et al (2007) reported that lower molecular weight(ltsim180 g molminus1) water-soluble organic compounds evolvedmostly at the OC1 temperature step while a higher oven tem-

13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619

Fig 12 Scatter plot of mass concentrations of OC1 versus OC2The pollen LTP and AD + LTP episodes are shown as filled red di-amond blue circle and brown rectangular respectively OC1 andOC2 represent organic carbons evolved at 300C and 450C re-spectively

perature was needed for the higher molecular weight fractionof the organic aerosol resulting in increased OC2 and OC3fractions Lim et al (2010) suggested that organic acids aredominantly formed in cloud processing whereas large mul-tifunctional humic-like substances are dominantly formed inwet aerosols via radical-radical reactions Thus differentevolution patterns of OC obtained for the LTP and AD + LTPepisodes can be explained by different formation mecha-nisms of secondary organic aerosols and the effect of agingof organic aerosols during long-range atmospheric transportDifferent sources of organic aerosols from the Asian conti-nent may also contribute to the different thermal evolutionpatterns of OC



14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2

Fig 13 Scatter plot ofδ13CTC versus OC1 and OC2 fractions in total OC during the pollen episodes The error bar in x-axis represents theanalytical error ofδ13CTC measurement

452 Thermally resolved OC component versus stablecarbon isotopic composition

The relations between the OC fractions andδ13CTC dur-ing the LTP and AD + LTP episodes are also examinedIn contrast to the pollen episodes almost no correlationswere obtained between the OC fractions andδ13CTC ex-cept for the OC1 fraction that gave a moderate correlationof R2 = 059 (data are not shown) The moderate correla-tion between the OC1 fraction andδ13CTC was attributedto the elevatedδ13CTC during the strong AD + LTP episode(KOS603) Thus if we exclude theδ13CTC in the KOS603samples no correlation was obtained between the OC1 frac-tion and δ13CTC In order to investigate the dependenceof δ13CTC on thermally evolved OC fractions the fractionsof OC evolved at each temperature step in total OC duringthe pollen episodes are plotted as a function ofδ13CTC inFig 13a b The OC1 and OC2 fractions normalized bytotal OC mass showed good correlations withδ13CTC withR2 of 081 and 073 respectively during the pollen episodes(Fig 13a) whereas a moderate correlation was observed be-tween OC2 mass concentration andδ13CTC with R2 = 048(Fig 13b) However almost no correlation was observed be-tween the OC1 mass concentration andδ13CTC (Fig 13b)Because the PollenGosan had low δ13CTC of minus280 permil (Ta-ble 3) the negative correlations of the normalized OC2 massfraction and OC2 mass concentration withδ13CTC indicatedthat a large fraction of the pollen-enriched TSP samplesevolved at the OC2 temperature step (300ndash450C) Almostno correlation between OC1 mass concentrations andδ13CTCindicated that the positive correlation between the normal-ized OC1 fraction andδ13CTC was mainly attributed to a rel-ative decrease in OC1 fraction to total OC as OC2 fractionincreases during the pollen episodes A positive correlationbetween mass concentrations of OC1 and OC2 (Fig 12) anda negative correlation between the normalized OC1 and OC2

fractions (Fig 13) during the pollen episodes imply that asmall fraction of pollen also evolved in the OC1 temperaturestep but dominant fractions of them evolved in the OC2 tem-perature step

5 Summary and conclusion

Satellite remote sensing air mass backward trajectoriesand particulate chemical and stable carbon isotopic compo-sition analyses allowed us to categorize the carbon emis-sion episodes observed at the Gosan background super-site(3317 N 12610 E) in East Asia during spring of 2007and 2008 as long-range transported anthropogenic pollu-tants (LTP) from the Asian continent Asian dust accompa-nying LTP (AD + LTP) and local pollen episodes Carbonepisodes were defined in this study as mass concentration ofTCgt 10 microgC mminus1 Subdivision of the carbon episodes wasconducted as follows Pollen episodes were identified by theelevated concentrations of citric acid and pollen in the TSPsamples as described by Jung and Kawamura (2011) LTPepisodes were identified by the elevated concentrations ofnitrate and sulfate during the carbon episodes AD + LTPepisodes were identified by the elevated concentrations ofcalcium ion (Ca2+) and low aerosolAngstrom exponent aswell as the elevated concentrations of nitrate and sulfate dur-ing the carbon episodes

The carbon episodes caused by the pollen episodeswere characterized by the lowestδ13C of TC (δ13CTC)(avgminus252plusmn 09 permil) followed by the LTP episodes(avgminus233plusmn 03 permil) and the AD + LTP episodes(avgminus218plusmn 20 permil) Using the HCl fume treatmenton the dust-enriched samples we found thatsim8 in totalcarbon (TC) during the strong AD + LTP episode (KOS603sample) was attributed to carbonate carbon that was notremoved via the reaction with the acidic gases such as



nitrogen dioxide nitric acid and sulfur dioxide duringthe long-range atmospheric transport resulting in higherδ13CTC

Thermal evolution patterns of OC during the pollenepisodes were clearly characterized by higher OC evolutionin the OC2 temperature step (450C) Different evolutionpatterns of OC obtained for the LTP and AD + LTP episodescan be explained by different formation mechanisms of sec-ondary organic aerosols and the effect of aging of organicaerosols during the long-range atmospheric transport Dif-ferent sources of organic aerosols from the Asian continentmay contribute to the different thermal evolution patterns ofOC

Based on the carbon isotope mass balance equation wefound that during the pollen episodessim42 of TC was at-tributed to airborne pollen emitted from Japanese cedar treesplanted around tangerine farms in Jeju Island The nega-tive correlation between the citric acid-carbonTC ratios andδ13CTC and similarδ13C values of citric acid between theairborne pollens (minus263 permil) collected at the Gosan site andtangerine fruit (minus274 permil) imply that citric acid emitted fromtangerine fruit may be adsorbed on the airborne pollen andthen transported to the Gosan site

AcknowledgementsThis work was supported by a Grant-in-AidNo 2100923509 from the Japan Society for the Promotion ofScience (JSPS) and the Environment Research and TechnologyDevelopment Fund (B-0903) of the Ministry of the EnvironmentJapan We appreciate the financial support of a JSPS Fellowship toJ S Jung We acknowledge M H Lee for collecting samples at theGosan site The authors thank NOAA Air Resources Laboratory(ARL) for the provision of the HYSPLIT transport and dispersionmodel andor READY website (httpwwwarlnoaagovreadyphp)used in this publication We also thank Y J Kim and S C Yoonfor their effort in establishing and maintaining the Gwangju andGosan AERONET sites respectively This paper has been read byPhilip Meyers to improve its English We appreciate his comments

Edited by T Rockmann

References

Aggarwal S G and Kawamura K Molecular distribu-tions and stable carbon isotopic compositions of dicarboxylicacids and related compounds in aerosols from SapporoJapan implications for photochemical aging during long-range atmospheric transport J Geophys Res 113 D14301doi1010292007JD009365 2008

Baltensperger U Dommen J Alfarra R Duplissy J GaeggelerK Metzer A Facchini M C Decesari S Finessi E Rein-nig C Schott M Warnke J Hoffmann T Klatzer BPuxbaum H Geiser M Savi M Lang D Kalbere M andGeiser T Combined determination of the chemical composi-tion and of health effects of secondary organic aerosols ThePOYSOA project J Aerosol Med Pulm D 21 145ndash154 2008

Birch M E and Cary R A Elemental Carbon-Based Methodfor Monitoring Occupational Exposures to Particulate Diesel Ex-haust Aerosol Sci Technol 25 221ndash241 1996

Cachier H Buat-Menard M P Fontugne M and Chesselet RLong-range transport of continentally-derived particulate carbonin the marine atmosphere Evidence from stable carbon isotopestudies Tellus B 38 161ndash177 1986

Doskey P V and Ugoagwu B J Atmospheric deposition ofmacronutrients by pollen at a semi-remote site in northern Wis-consin Atmos Environ 23 2761ndash2766 1989

Draxler R R and Rolph G D HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) Model access viaNOAA ARL READY Website NOAA Air Resources Labora-tory Silver Spring MD available athttpwwwarlnoaagovHYSPLITphp 2011

Dubovik O and King M D A flexible inversion algorithm forretrieval of aerosol optical properties from Sun and sky radiancemeasurements J Geophys Res 105 20673ndash20696 2000

Forster P Ramaswamy V Artaxo P Berntsen T Betts R Fa-hey D W Haywood J Lean J Lowe D C Myhre GNganga J Prinn R Raga G Schulz M and Van DorlandR Changes in Atmospheric Constituents and in Radiative Forc-ing in Climate Change 2007 The Physical Science Basis Con-tribution of Working Group I to the Fourth Assessment Reportof the Intergovernmental Panel on Climate Change edited bySolomon S Qin D Manning M Chen Z Marquis M Av-eryt K B Tignor M and Miller H L Cambridge UniversityPress Cambridge United Kingdom and New York NY USA2007

Hoekstra F A Crowe J H Crowe L M Van Roekel T andVermeer E Do phospholipids and sucrose determine membranephase transitions in dehydrating pollen species Plant Cell Env-iron 15 601ndash606 1992

Hoefs J Stable isotope geochemistry Springer New York 1997Huebert B J Bates T Russell P B Shi G Kim Y J Kawa-

mura K Carmichael G and Nakajima T An overview ofACE-Asia Strategies for quantifying the relationships betweenAsian aerosols and their climatic impacts J Geophys Res 1088633doi1010292003JD003550 2003

IMPROVE Spatial and Seasonal Patterns and Temporal Variabilityof Haze and its Constituents in the United States Report IVavailable athttpvistaciracolostateeduimprovePublicationsReports20062006htm 2006

Jahren H The carbon stable isotope composition of pollen A RevPalaeobot Palyno 132 291ndash313 2004

Jung J and Kawamura K Enhanced concentrations of citric acidin spring aerosols collected at the Gosan background site in EastAsia Atmos Environ 45 5266ndash5272 2011

Kawamura K Identification of C2ndashC10 ω-oxocarboxylic acidspyruvic acid and C2ndashC3 α-dicarbonyls in wet precipitation andaerosol samples by capillary GC and GC-MS Anal Chem 653505ndash3511 1993

Kawamura K and Watanabe T Determination of stable car-bon isotopic compositions of low molecular weight dicarboxylicacids and ketocarboxylic acids in atmospheric aerosol and snowsamples Anal Chem 76 5762ndash5768 2004

Kawamura K Kobayashi M Tsubonuma N Mochida MWatanabe T and Lee M Organic and inorganic compositionsof marine aerosols from East Asia Seasonal variations of water



soluble dicarboxylic acids major ions total carbon and nitro-gen and stable C and N isotopic composition in GeochemicalInvestigation in Earth and Space Science A Tribute to Issac RKaplan The Geochemical Society 9 Elsevier 243ndash265 2004

Kim K W Physico-chemical characteristics of visibility impair-ment by airborne pollen in an urban area Atmos Environ 413565ndash3576 2007

Kundu S Kawamura K and Lee M Seasonal variation of theconcentrations of nitrogenous species and their nitrogen isotopicratios in aerosols at Gosan Jeju Island Implications for atmo-spheric processing and source changes of aerosols J GeophysRes 115 D20305doi1010292009JD013323 2010

Levy R C Remer L A and Dubovik O Global aerosol opticalproperties and application 15 to Moderate Resolution ImagingSpectroradiometer aerosol retrieval over land J Geophys Res112 D13210doi1010292006JD007815 2007a

Levy R C Remer L A Mattoo S Vermote E F and Kauf-man Y J Second-generation operational algorithm retrieval ofaerosol properties over land from inversion of moderate resolu-tion imaging spectroradiometer spectral reflectance J GeophysRes 112 D13211doi1010292006JD007811 2007b

Li J Wang G Zhou B Cheng C Cao J Shen Z and AnZ Chemical composition and size distribution of wintertimeaerosols in the atmosphere of Mt Hua in central China AtmosEnviron 45 1251ndash1258 2011

Lim Y B Tan Y Perri M J Seitzinger S P and TurpinB J Aqueous chemistry and its role in secondary organicaerosol (SOA) formation Atmos Chem Phys 10 10521ndash10539doi105194acp-10-10521-2010 2010

Lockheart M J Van Bergen P F and Evershed R P Vari-ations in the stable carbon isotope compositions of individuallipids from the leaves of modern angiosperms implications forthe study of higher land plant-derived sedimentary organic mat-ter Org Geochem 26 137ndash153 1997

Mamane Y and Gottlieb J Heterogeneous reactions of mineralswith sulfur and nitrogen oxides J Aerosol Sci 20 303ndash3111989

Miyazaki Y Kondo Y Han S Koike M Kodama D Ko-mazaki Y Tanimoto H and Matsueda H Chemical charac-teristics of water-soluble organic carbon in the Asian outflowJ Geophys Res 112 D22S30doi1010292007JD0091162007

Miyazaki Y Aggarwal S G Singh K Gupta P K and Kawa-mura K Dicarboxylic acids and water-soluble organic car-bon in aerosols in New Delhi India in winter Characteris-tics and formation processes J Geophys Res 114 D19206doi1010292009JD011790 2009

Miyazaki Y Kawamura K and Sawano M Size distributionsof organic nitrogen and carbon in remote marine aerosols Evi-dence of marine biological origin based on their isotopic ratiosGeophys Res Lett 37 L06803doi1010292010GL0424832010

Nakajima T Yoon S C Ramanathan V Shi G Y TakemuraT Higurashi A Takamura T Aoki K Sohn B J KimS W Tsuruta H Sugimoto N Shimizu A Tanimoto HSawa Y Lin N H Lee C T Goto D and SchutgensN Overview of the Atmospheric Brown Cloud East Asian Re-gional Experiment 2005 and a study of the aerosol direct ra-diative forcing in east Asia J Geophys Res 112 D24S91

doi1010292007JD009009 2007Narukawa M Kawamura K Li S-M and Bottenheim

J W Stable carbon isotopic ratios and ionic composi-tion of the high-Arctic aerosols An increase in d13C val-ues from winter to spring J Geophys Res 113 D02312doi1010292007JD008755 2008

Novakov T and Penner J E Large contribution of organicaerosols to cloud-condensation-nuclei concentrations Nature365 823ndash826 1993

Noh Y M Muller D Shin D H Lee H Jung J S Lee K HCribb M Li Z Kim Y J Optical and Microphysical Proper-ties of Severe Haze and Smoke Aerosol Measured by IntegratedRemote Sensing Techniques in Gwangju Korea Atmos Envi-ron 43 879ndash888 2009

OrsquoDowd C D Facchini M C Cavalli F C Ceburnis DMircea M Decesari S Fuzzi S Yoon Y J and Putaud JP Biogenically drived organic contribution to marine aerosolNature 431 676ndash679 2004

Oh J W Lee H B Lee H R Pyun B Y Ahn Y M Kim KE Lee S Y and Lee S I Aerobiological study of pollen andmold in Seoul Korea Allergol Int 47 263ndash270 1998

Palacios I S Molina R T and Rodriguez A F M The im-portance of interactions between meteorological conditions wheninterpreting their effect on the dispersal of pollen from homoge-neously distributed sources Aerobiologia 23 17ndash26 2007

Porsbjerg C Rasmussen A and Backer A Airborne pollen inNuuk Greenland and the importance of meteorological param-eters Aerobiologia 19 29ndash37 2003

Puc M and Wolski T Betula and Populus pollen counts and mete-orological conditions in Szczencin Poland Ann Agr Env Med9 65ndash69 2002

Remer L A Kaufman Y J Tanre D Mattoo S Chu D AMartins J V Li R-R Ichoku C Levy R C Kleidman RG Eck T F Vermonte E and Holben B N The MODISaerosol algorithm products and validation J Atmos Sci 62947ndash973 2005

Rolph G D Real-time Environmental Applications and DisplaysYstem (READY) Website NOAA Air Resources LaboratorySilver Spring MD available athttpwwwarlnoaagovreadyphp 2011

Rousseau D-D Schevin P Ferrier J Jolly D An-dreasen T Ascanius S E Hendriksen S-E and PoulsenU Long-distance pollen transport from North Americato Greenland in spring J Geophys Res 113 G02013doi1010292007JG000456 2008

Solomon W R Airborne pollen A brief life J Allergy ClinImmun 109 895ndash900 2002

Stohl A Computation accuracy and applications of trajectoriesA review and bibliography Atmos Environ 32 947ndash966 1998

Turekian V C Macko S A and Keene W C Concen-trations isotopic compositions and sources of size-resolvedparticulate organic carbon and oxalate in near-surface marineair at Bermuda during spring J Geophys Res 108 4157doi1010292002JD002053 2003

Underwood G M Song C Phandis M Carmichael G andGrassian V Heterogeneous reactions of NO2 and HNO3 onoxides and mineral dust A combined laboratory and modelingstudy J Geophys Res 106 18055ndash18066 2001

Wang H and Kawamura K Stable carbon isotopic composi-



tion of low-molecular weight dicarboxylic acids and ketoacidsin remote marine aerosols J Geophys Res 111 D07304doi1010292005JD006466 2006

Wang G Xie M Hu S Gao S Tachibana E and KawamuraK Dicarboxylic acids metals and isotopic compositions of Cand N in atmospheric aerosols from inland China implicationsfor dust and coal burning emission and secondary aerosol forma-tion Atmos Chem Phys 10 6087ndash6096doi105194acp-10-6087-2010 2010

Yang H Yu J Z Ho S S H Xu J Wu W S Wan C HWang X D Wang X R and Wang L The chemical compo-sition of inorganic and carbonaceous materials in PM25 in Nan-jing China Atmos Environ 39 3735ndash3749 2005

Ye B Ji X Yang H Yao X Chan C K Cadle S H ChanT and Mulawa P A Concentration and chemical compositionof PM25 in Shanghai for a 1-year period Atmos Environ 37499ndash510 2003

Yoon S C Kim S W and Kim M H Ground-based Mie-scattering lidar measurements of aerosol extinction profiles dur-ing ABC-EAREX 2005 Comparisons of instruments and inver-sion algorithms J Meteorol Soc JPN 86 377ndash396 2008

Zhang Y Sunwoo Y Kotamarthi V and Carmichael G Photo-chemical oxidant processes in the presence of dust An evalua-tion of the impact of dust on particulate nitrate and ozone forma-tion B Am Meteorol Soc 33 813ndash824 1994



2005 (Atmospheric Brown Cloud-East Asia Regional Ex-periment 2005) (Nakajima et al 2007) To better under-stand the link between chemical and physical properties ofaerosols mainly transported from the Asian continent and re-gional climate change sources and formation mechanism ofsecondary aerosols should be investigated In addition con-tributions of local effects on the Gosan site aerosols shouldbe qualitatively and quantitatively evaluated

Pollen is one of the important sources of bioaerosols(Solomon 2002) These particles can cause serious allergicproblems to human health (Solomon 2002) and visibility im-pairment (Kim 2007) Most pollen emission events in Koreaoccur during the blooming season of plants from March toMay (Oh et al 1998) Because airborne pollen can be trans-ported very long distances (Porsbjerg et al 2003 Rousseauet al 2008) airborne pollen is not only a local problem butalso a regional and intercontinental problem However stud-ies regarding the impact of pollen on the aerosol chemicalcomposition at the Gosan site are rare (Jung and Kawamura2011)

Stable carbon isotopic compositions (δ13C) of total carbon(TC) are very useful for investigating sources and the long-range atmospheric transport of organic aerosols (Cachier etal 1986 Narukawa et al 2008 Miyazaki et al 2010) Onthe basis of theδ13C values and Na+TC ratios Narukawa etal (2008) estimated the contribution of marine organic mat-ter in TC in the high Arctic at Alert during spring Miyazakiet al (2010) estimated marine-derived carbon in TC overthe western North Pacific usingδ13C values Also Kawa-mura and Watanabe (2004) developed a novel method forcompound specific carbon isotope compositions of dicar-boxylic acids and related compounds using gas chromatogra-phyisotope ratio mass spectrometry (GCirMS) Since thenδ13C values of dicarboxylic acids and related compoundshave successfully been used to assess the extent of photo-chemical processing of aerosols during their long-range at-mospheric transport (Wang and Kawamura 2006 Aggarwaland Kawamura 2008)

In this study elevated concentrations of total carbon wereoften found in the total suspended particulate (TSP) sam-ples collected at the Gosan site during spring of 2007 and2008 Using remote monitoring and comprehensive in-situchemical andδ13C analyses we categorize the carbon emis-sion episodes to three types long-range transport anthro-pogenic pollution (LTP) from Asian continent Asian dust(AD) accompanying with the LTP and local airborne pollenepisodes We discuss the measuredδ13C values and the ther-mal evolution patterns obtained from heating of carbona-ceous particles Based on a carbon isotope mass balanceequation we quantify the TC fraction of local pollen Us-ing HCl fume treatment of the dust-enriched samples wequantify carbonate carbon in the TC and discuss removal ofcarbonate via the reaction with acidic gases during the long-range atmospheric transport

2 Samples and methods

TSP sampling was carried out at the Gosan supersite(3317 N 12610 E) on Jeju Island located approximately100 km south of the Korean peninsula (Fig 1) over suc-cessive periods that integrated 2ndash5 days from 23 March to1 June 2007 and from 16 to 24 April 2008 TSP samples werecollected on pre-combusted quartz fiber filters (20times 25 cm)using a high volume air sampler (Kimoto AS-810) at a flowrate of 50 m3 hrminus1 on the rooftop of a trailer (sim3 m above theground) Before and after sampling the filter samples werestored in clean glass jars (150 ml) with a Teflon-lined screwcap atminus20C prior to analysis Field blank filters were col-lected every month Hourly PM10 mass data was obtainedfrom the National Institute of Environment Research at theGosan Observatory Measurements of wind direction andwind speed at 10 m above ground level were obtained fromthe Korea Meteorological Administration at the Gosan Ob-servatory A total of 32 filter samples were analyzed in thisstudy

21 Airborne pollen and tangerine fruit samples

Three types of pollen samples were prepared and ana-lyzed in this study Two authentic standard pollen sam-ples from Japanese cedar (Pollencedar) and Japanese cy-press (Pollencypress) were obtained from the WAKO Chemi-cal Co (product No 168-20911 for Japanese cedar and 165-20921 for Japanese cypress) Additionally airborne pollen(PollenGosan) which mainly originated from Japanese cedartrees planted around tangerine farms on Jeju Island wereseparated from an aliquot of the TSP filter sample collectedduring a severe pollen episode (KOS751 16ndash21 April 2008)An aliquot (15 cm2) of the KOS751 filter sample was placedin a glass vial (50 ml) with a Teflon-lined screw cap and thepollen grains were separated by mild vibration using an au-tomatic vibrator (Iuchi HM-10) for 5 min The separatedpollen grains were then transferred to a pear shape flask forchemical analysis In order to track the sources of the di-carboxylic acids and related compounds and airborne pollentangerine fruit produced from Jeju Island was also preparedand analyzed for dicarboxylic acids and related compounds(Jung and Kawamura 2011) and for carbon isotopic compo-sitions of total carbon oxalic acid and citric acid In orderto prevent possible contamination over the surface of tanger-ine fruit the tangerine surface was mildly washed three timesusing ultra pure organic-free Milli-Q water

22 Organic and elemental carbon analysis

Organic carbon (OC) and elemental carbon (EC) were ana-lyzed with a Sunset carbon analyzer using the thermal-opticaltransmittance (TOT) protocol for pyrolysis correction (Birchand Cary 1996 Miyazaki et al 2009) A 214 cm2 punchof the quartz filter was placed in a quartz boat inside the



1

Figure 1









4 (nss-SO2minus




δ13C(permil) =

[(13C12C)sample


]times1000




















2

Figure 2

0

2

4

6

8

10

12

0 10 20 30TC (μgC m-3)

Num

ber o

f dat

a

















nss-SO2minus








3

Figure 3


102030

0100200

1200

-26-24-22-20

0204060300320

TCT

N ra

tio

2007 2008

TC TN

TC a

nd T

N(m

g m

-3)

AD LTP

PM10

mas

s(m

g m

-3)

δ13C

(permil)

Citri

c ac

id(n

g m

-3)

Pollen


4

Figure 4

Korea

JapanChina

a) LTP

Nei Mongol desert


b) AD+LTP



3 and SO2minus






5

Figure 5




6











7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(b)













8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




























































1

Figure 1









4 (nss-SO2minus




δ13C(permil) =

[(13C12C)sample


]times1000




















2

Figure 2

0

2

4

6

8

10

12

0 10 20 30TC (μgC m-3)

Num

ber o

f dat

a

















nss-SO2minus








3

Figure 3


102030

0100200

1200

-26-24-22-20

0204060300320

TCT

N ra

tio

2007 2008

TC TN

TC a

nd T

N(m

g m

-3)

AD LTP

PM10

mas

s(m

g m

-3)

δ13C

(permil)

Citri

c ac

id(n

g m

-3)

Pollen


4

Figure 4

Korea

JapanChina

a) LTP

Nei Mongol desert


b) AD+LTP



3 and SO2minus






5

Figure 5




6











7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(b)













8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References












































































2

Figure 2

0

2

4

6

8

10

12

0 10 20 30TC (μgC m-3)

Num

ber o

f dat

a

















nss-SO2minus








3

Figure 3


102030

0100200

1200

-26-24-22-20

0204060300320

TCT

N ra

tio

2007 2008

TC TN

TC a

nd T

N(m

g m

-3)

AD LTP

PM10

mas

s(m

g m

-3)

δ13C

(permil)

Citri

c ac

id(n

g m

-3)

Pollen


4

Figure 4

Korea

JapanChina

a) LTP

Nei Mongol desert


b) AD+LTP



3 and SO2minus






5

Figure 5




6











7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(b)













8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References






































































nss-SO2minus








3

Figure 3


102030

0100200

1200

-26-24-22-20

0204060300320

TCT

N ra

tio

2007 2008

TC TN

TC a

nd T

N(m

g m

-3)

AD LTP

PM10

mas

s(m

g m

-3)

δ13C

(permil)

Citri

c ac

id(n

g m

-3)

Pollen


4

Figure 4

Korea

JapanChina

a) LTP

Nei Mongol desert


b) AD+LTP



3 and SO2minus






5

Figure 5




6











7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(b)













8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




























































3

Figure 3


102030

0100200

1200

-26-24-22-20

0204060300320

TCT

N ra

tio

2007 2008

TC TN

TC a

nd T

N(m

g m

-3)

AD LTP

PM10

mas

s(m

g m

-3)

δ13C

(permil)

Citri

c ac

id(n

g m

-3)

Pollen


4

Figure 4

Korea

JapanChina

a) LTP

Nei Mongol desert


b) AD+LTP



3 and SO2minus






5

Figure 5




6











7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(b)













8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




























































5

Figure 5




6











7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(b)













8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References

































































7

Figure 6

y = 066 x + 754R2 = 095

0

5

10

15

20

25

30

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(a)

KOS603

KOS627

y = 283 x + 141R2 = 098

y = 083 x + 379R2 = 082

0

20

40

60

80

0 5 10 15 20 25 30

TN (μgN m-3)

TC (μ

gC m

-3)


(b)













8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




































































8

Figure 7

0

4

8

12

16

20


(μgC

m-3

)

-25

-22

-19

-16

-13

-10

δ13C

(permil)

TC δ13C

KOS603 KOS627

(b)

δ13C

-28

-26

-24

-22

-20

-18

0 5 10 15 20 25 30

TC (μgC m-3)

δ13C

(permil)


(a)

KOS603

KOS627







δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




























































δ13CTC =REM-C

TCmiddotδ13CREM-C

+RMD-C







9

Figure 8

KOS751 sample

Pollen_Gosan

WS Pollen_Gosan

Pollen_cedar

WS Pollen_cedar

Pollen_cypress

WS Pollen_cypress

WS Tangerine juice

WS Tangerine peel

-30 -28 -26 -24 -22

d13C (permil)














10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




































































10

Figure 9

0

5

10

15

20

25

30

KO

S60

8

KO

S60

9

KO

S61

0

KO

S61

1

KO

S61

2

KO

S61

3

KO

S75

1

KO

S75

2

TC (micro

gC m

-3)

TC_pollen Total





11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




























































11

Figure 10

-27

-26

-25

-24

-23

0 01 02 03 04 05


δ13C

(permil)

















12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References

































































12

Figure 11

0002040608

0004081216

0002040608

0 200 400 600 800 10000002040608

Norm

alize

d ND

IR s

igna

l




Time (sec)


0300600900EC6EC5

EC4EC3

EC2EC1

OC4OC3OC2

Temp (C)

OC1



13

Figure 12

y = 270 x - 141R2 = 077

0

3

6

9

12

0 1 2 3 4 5 6

OC1 (μgC m-3)

OC

2 (μ

gC m

-3)

Pollen LTP AD+LTP

KOS603

KOS627

KOS606 614 619





14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References




























































14

Figure 13

(a) (b)

y = -029 x - 526R2 = 009

y = -208 x - 4811R2 = 048

0

2

4

6

8

10

12

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

(μgC

m-3

)

OC1 OC2

y = 252 x + 7864R2 = 081

y = -491 x - 9464R2 = 073

10

15

20

25

30

35

40

-27 -26 -25 -24 -23 -22δ13C (permil)

OC

frac

tion

()

OC1 OC2















References

































































References



























































springtime carbon emission episodes at the gosan background site

Documents