1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

8:07f=WðJul162004Þþmodel

AEA : 6845 Prod:Type:FTPpp:1212ðcol:fig::NILÞ

ED:VijayBNPAGN:sree SCAN:

ARTICLE IN PRESS

Please cite th

Yukon, Cana

1352-2310/$ - se

doi:10.1016/j.at

�CorrespondGeosciences, Id

tello, ID 8

+1208 282 441

E-mail addr

wake@unh.edu

siw@unh.edu (

Atmospheric Environment ] (]]]]) ]]]–]]]

www.elsevier.com/locate/atmosenv

F

Relationships between aerosol and snow chemistry at King Col,Mt. Logan Massif, Yukon, Canada

Kaplan Yalcin�, Cameron P. Wake, Jack E. Dibb, Sallie I. Whitlow

Climate Change Research Center, Institute for the Study of Earth, Oceans, and Space, University of New Hampshire, Durham, NH 03824,

USA

Received 9 February 2006; received in revised form 22 June 2006; accepted 25 June 2006

O

RECTED PROAbstract

Simultaneous samples of aerosol (n ¼ 48) and recent snow (n ¼ 193) chemistry were collected at King Col (4135m) in

the St. Elias Mountains, Yukon, between 17 May and 11 June 2001. Major ion concentrations in aerosol samples were low

with the total ionic burden averaging 5.52 neqm�3 at standard temperature and pressure (STP). Interspecies aerosol

relationships indicate the presence of (NH4)2SO4 aerosol at King Col. An aerosol Cl� deficiency relative to seawater

suggests volatilization of HCl by reaction with unneutralized SO2�4 that is present in half of the samples. Backwards

trajectories for select aerosol concentration peaks document the transport of Asian dust and anthropogenic emissions, the

eruption plume from the 22 May eruption of Sheveluch, Kamchatka, and sea salt from the marine boundary layer over the

Gulf of Alaska to King Col during the sampling period. Fresh snow chemistry generally mimics aerosol chemistry with

similar relative abundances and interspecies relationships except for large enrichments in snow Cl� and NO�3 relative to

aerosol due to snow scavenging of gas-phase HCl and HNO3. Although relatively strong correlations between aerosol and

snow concentrations were observed for species associated with accumulation mode aerosols, e.g. NHþ4 (r ¼ 0:56) and SO2�4

(r ¼ 0:43), only weak correlations were observed for dust and sea-salt species. These results are influenced by greater

variability in concentrations between replicate snow samples for species associated with coarse mode dust and sea-salt

particles and by snow scavenging of gas-phase HCl and HNO3.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: Aerosols; Air–snow transfer; Major ions; Mt. Logan; Yukon

R 53

55

57

CO1. Introduction

Ice cores retrieved from the polar ice sheets andsuitably located mountain glaciers arguably provide

UN59

61

63

is article as: Kaplan Yalcin et al., Relationships betwee

da, Atmospheric Environment (2006), doi:10.1016/j.atm

e front matter r 2006 Elsevier Ltd. All rights reserved

mosenv.2006.06.028

ing author. Present address: Department of

aho State University, Campus Box 8072, Poca-

3209, USA. Tel.: +1 208 282 2949; fax:

4.

esses: yalckapl@isu.edu (K. Yalcin), cameron.-

(C.P. Wake), jack.dibb@unh.edu (J.E. Dibb),

S.I. Whitlow).

the highest resolution and most direct view ofEarth’s paleoatmosphere (e.g., Legrand andMayewski, 1997). Recently, the St. Elias Mountainsin northwest North America have become the focusof an international ice coring effort (Fig. 1). Priorwork on St. Elias ice cores recovered from theNorthwest Col (5430m) of Mt. Logan in 1980 andEclipse Icefield (3017m) in 1996 demonstrated thatdifferences in elevation allow the sites to sampledifferent layers of the atmosphere, resulting in

65

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

.

REC

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

Fig. 1. Map of the St. Elias Mountains, southwest Yukon,

showing the location of the sampling site at King Col (4135m).

Nearby sites from which ice cores have been recovered (in

addition to King Col) include the Northwest Col (5340m) and

Prospector-Russell Col (5340m) on the summit plateau of Mt.

Logan, Bona-Churchill (4400m), Mt. Wrangell (4100m), and

Eclipse Icefield (3017m).

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]]2

UNCORcomplementary and distinct records (Holdsworth etal., 1988; Yalcin and Wake, 2001; Wake et al., 2002;Yalcin et al., 2003). During the 2001 and 2002 fieldseasons, a suite of new ice cores were recoveredfrom the St. Elias Mountains spanning an elevationrange from 3017 to 5340m. Ice cores were drilled atboth Prospector-Russell Col (5340m) and King Col(4135) on Mt. Logan (Goto-Azuma et al., 2003;Fisher et al., in press). Additional cores were drilledat the saddle (4400m) between Mt. Bona and Mt.Churchill (Thompson et al., 2004) and the summitice cap 4100m) of Mt. Wrangell (Kanamori et al.,2004), and two new ice cores were collected fromEclipse Icefield (3017m) (Yalcin et al., 2006a, b).Together, these records offer a unique opportunity

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

TED PROOF

to construct a three-dimensional view of the paleo-atmosphere.

However, interpreting the chemical records con-tained in ice cores requires knowledge of the site-specific relationships between snow chemistry andthat of the precipitating air mass that can beobtained by simultaneous collection of aerosol andsnow samples (e.g., Dibb and Jaffrezo, 1997). Suchinvestigations are important given the wealth ofpaleoenvironmental information potentially avail-able from ice cores and the uncertainty in relatingglaciochemical to atmospheric concentrations. Pro-grams such as the Dye 3 (Greenland) Gas andAerosol Sampling Program and the Summit, Green-land atmospheric research program represent inter-national collaborative efforts to better understandthe processes influencing chemical species eventuallypreserved in Greenland ice cores (Jaffrezo andDavidson, 1993; Dibb and Jaffrezo, 1997). Sam-pling campaigns have also characterized the aerosoland investigated air–snow chemical relationships atmountain glacier sites in the European Alps(Baltensperger et al., 1992) and Central Asia (Wakeet al., 1994; Sun et al., 1998; Shrestha et al., 2002).

In support of ice coring efforts on the summitplateau of Mt. Logan, an intermediate elevationclimbing camp and research station was establishedat King Col (60.351N, 140.361W, 4135m elevation)on the Mt. Logan massif during the 2001 fieldseason (Fig. 1). From 17 May to 11 June, twice dailyaerosol and three to five replicate fresh or surfacesnow samples were collected. The resulting databaseprovides the information needed to characterize thelate spring–early summer aerosol at King Col andexamine the relationships between aerosol and snowchemistry on an event basis.

2. Methods

2.1. Sample collection

Twice-daily aerosol samples were collected atKing Col from 17 May to 11 June using a 24V high-volume pump with inline flow meter powered by acombination of photovoltaic cells and batteries.Care was taken to minimize contamination. Aerosolsampling was conducted on the south side of theCol, 120m from camp in a direction perpendicularto the prevailing wind direction to minimizepotential local contamination from the camp. Anon-particulating suit, face mask, and polyethylenegloves were worn at all times during aerosol and

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]] 3

UNCORREC

snow sample collection. Each filter was pre-loadedonto a polyethylene cassette in a class 100 cleanroom at the University of New Hampshire andsealed inside a clean polyethylene bag. Just prior tosample collection, the filter was removed from thesealed bag and mounted face down inside a 0.20-mdiameter protective cylindrical Delrin housing 2mabove the snow surface. A polyethylene wind bafflehousing the aerosol filter holder minimized thepotential for collecting airborne snow and icecrystals.

Samples were collected on 1 mm pore size; 90mmdiameter Teflon filters (Fluropore filters, Millipore,Bedford, MA). Sample collection times ranged from8 to 15 h and averaged 12 h. Corrections for ambienttemperature and pressure allowed conversion of themeasured air volumes to cubic meters at standardtemperature and pressure (m3 STP). Sampledvolumes ranged from 17.19 to 42.63m3 STP andaveraged 29.89m3 STP. The mean flow rate was2.4m3 STP h�1 yielding a velocity at the face of thefilter of 10.5 cm s�1. This velocity is high enough forthis kind of filter to have collection efficiency greaterthan 90% for aerosol particles larger than 0.3 mm(Hinds, 1999). The cutoff for large particles, asgiven by their sedimentation velocity, is estimated atabout 20 mm (Hinds, 1999).

After sampling, the filters, still in their cassettes,were returned to their original clean polyethylenebags, double sealed in plastic bags, and stored in asnow cave. Ten procedural blanks were collected inthe course of the sampling campaign by placing aloaded filter cassette in the housing and immediatelyremoving and re-sealing. Filter blanks were ana-lyzed with the samples to quantify blank levels andcheck for contamination. Concurrent with aerosolsample collection, three to five replicate fresh oraged surface snow samples were collected about20m away from the aerosol sampling site byscraping the surface snow layer directly into pre-cleaned polypropylene cups. If measurable snow fellduring the sample collection time for a particularaerosol sample, then the surface snow layerconcurrently sampled was considered fresh snow.A total of 48 aerosol samples and 193 fresh andaged surface snow samples were collected during thesampling period.

101

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

TED PROOF

2.2. Analysis by ion chromatography, detection

limits, and precision

All samples were returned frozen to the ClimateChange Research Center at the University of NewHampshire for processing and analysis. Filters weresaturated with 0.5ml high purity methanol and thesoluble components extracted with three successive5ml aliquots of deionized Milli-Q water, andpreserved with chloroform. Filter extracts as wellas fresh and aged surface snow samples wereanalyzed by suppressed ion chromatography foreight inorganic ions (Na+, NHþ4 , K

+, Mg2+, Ca2+,Cl�, NO�3 , SO

2�4 ) and one organic ion ðC2O

2�4 Þ in a

dedicated laboratory at the University of NewHampshire Climate Change Research Center. Thecation system used a CS12A column with CSRS-ultra suppressor in auto suppression recycle modewith 20mM MSA eluent. The anion system used anAS11 column with an ASRS-ultra suppressor inauto suppression recycle mode with 6mM NaOHeluent. Precision of the results was monitored byrunning 10% of the samples in duplicate and foundto be 10% for aerosol Mg2+, less then 5% for otheraerosol cations, and less than 2% for aerosolanions. For snow samples, precision was 7% forK+, 4% for NHþ4 , 3% for Ca2+, and less then 2%for other species. Detection limits for snow sampleswere less than 2 ng g�1 for NO�3 , SO

2�4 , and C2O

2�4 ,

and less than 1 ng g�1 for all other species.Aerosol detection limits were defined as two

standard deviations of the blank values divided bythe mean sample volume (Shrestha et al., 2000). Thedetection limits for aerosol species were (in neqm�3

STP): Na+ (0.281), NHþ4 (0.008), K+ (0.034),Mg2+ (0.023), Ca2+ (0.025), Cl� (0.236), NO�3(0.004), SO2�

4 (0.049), C2O2�4 (0.002). The mean

concentration of the ten aerosol procedural blankswere subtracted from the sample concentrations,which occasionally resulted in below detection limitvalues for Na+ (6 samples), NHþ4 (2 samples), K+

(5 samples), Mg2+ (11 samples), Ca2+ (6 samples),and Cl� (12 samples). Sea-salt (ss) and non-sea-salt(nss) fractions for aerosol and snow samples wereestimated by determining the limiting sea-saltspecies (the species present in the lowest proportionrelative to seawater ratios) in each sample and usingthat species as the sea-salt indicator (Keene et al.,1986).

103

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]]4

3. Results and discussion

3.1. Characterization of King Col aerosol

Summary statistics of King Col aerosol chemicalcomposition measured between 17 May and 11 June2001 are presented in Table 1 and time series ofaerosol concentrations for major inorganic ions arepresented in Fig. 2. Aerosol concentrations are verylow, reflecting the remote location of the King Col

UNCORREC

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

Fig. 2. Concentrations of soluble ionic species in aerosol samples (solid

circles) snow samples collected at King Col between 17 May and 11 Ju

Table 1

Summary of aerosol composition (n ¼ 48) at King Col), 17 May–11 Ju

Na+ NHþ4 K+ Mg2+ Ca2+

Mean 1.26 1.34 0.17 0.15 0.52

Median 0.86 1.22 0.15 0.10 0.40

Std. dev. 1.04 0.89 0.11 0.12 0.45

Max 4.13 4.12 0.47 0.57 2.36

Min oDL oDL oDL oDL oD

DC: charge balance (sum of measured cations minus sum of measured

site relative to source regions. The total ionicburden (total cations plus total anions) ranged from1.30 to 13.21 neqm�3 STP and averaged5.52 neqm�3 STP. There is an excess of cations inall but three samples; with the charge balance (sumof measured cations minus sum measured anions)averaging 0.71 neqm�3 STP. This cation excesscould be balanced by carbonate ionsðHCO�3 =CO

2�3 Þ that were not measured in this

study. In fact, if the observed concentrations of

TED PROOF

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

line) and concurrent fresh (solid circles) and aged surface (open

ne 2001.

ne 2001 (unit: neqm�3 STP)

Cl� NO�3 SO2�4 C2O

2�4

DC

0.94 0.24 1.38 0.07 0.71

0.64 0.18 1.12 0.06 0.73

0.83 0.20 0.96 0.05 0.60

3.29 1.08 4.59 0.19 3.09

L oDL 0.05 0.17 0.01 �1.30

anions); DL: detection limit.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

Table 2

Correlation coefficients between soluble ionic species in aerosol

samples (n ¼ 48) collected at King Col, 17 May–11 June 2001

Na+ NHþ4 K+ Mg2+ Ca2+ Cl� NO�3 SO2�4

NHþ4 �0.18

K+0.92 �0.01

Mg2+ 0.38 0.55 0.39

Ca2+ �0.16 0.68 �0.15 0.76

Cl� 0.89 �0.14 0.84 0.39 �0.10

NO�3 �0.05 0.73 0.00 0.80 0.91 �0.01

SO2�4

0.03 0.90 0.13 0.71 0.75 0.05 0.83

C2O2�4�0.07 0.70 0.11 0.42 0.45 �0.07 0.53 0.71

Entries in bold denote correlations significant at the 99%

confidence level.

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]] 5

UNCORREC

Ca2+ and Mg2+ are assumed to be accompanied byan equal amount of carbonate ions (Toom-Sauntryand Barrie, 2002), the calculated charge balance isexcellent (averaging 0.14 neqm�3 STP). The remain-ing cation excess is likely balanced by smallamounts of methanosulfonate (MSA) or organicacids (e.g., formate, acetate) (Toom-Sauntry andBarrie, 2002).

Linear regression on all water-soluble species inthe aerosol (Table 2) showed that the sea-salt speciesNa+ and Cl� are highly correlated (r ¼ 0:89) andthe dust indicators Ca2+ and Mg2+ are wellcorrelated (r ¼ 0:76). Ammonium is highly corre-lated with SO2�

4 (r ¼ 0:90) and also correlates wellwith NO�3 (r ¼ 0:73), suggesting that these speciesare present in the same air masses. The highercorrelation between NHþ4 and SO2�

4 than betweenNHþ4 and NO�3 reflects the preferential reaction ofammonia with sulfuric rather than nitric acid(Bassett and Seinfeld, 1983). The mean NHþ4 /SO

2�4

molar ratio is 2.16, suggesting the principal form ofammonium sulfate particles at King Col is(NH4)2SO4 and not NH4HSO4. This result differsfrom Barrie and Barrie (1990) and Shrestha et al.(1997) who demonstrated that NH4HSO4 was theprincipal sulfate aerosol species in the high Arcticand Himalayas, respectively. Because sufficientNHþ4 exists in King Col aerosols to fully neutralizeSO2�

4 more than half the time, ammonium nitrateaerosol (NH4NO3) may be present at King Col, asimplied by the good correlation between NHþ4 andNO�3 . The collection of particulate nitrate asammonium nitrate may lead to a possible negativesampling artifact through individual evaporation asHNO3. However, nitrate loss from the filters byreaction with sulfate is expected to be important

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

TED PROOF

only when the NHþ4 /SO2�4 ratio is less than 1.5

(Pathak and Chan, 2005).Dust species, for example Ca2+, are highly

correlated with NO�3 (r ¼ 0:91) and well correlatedwith SO2�

4 (r ¼ 0:75), which suggests uptake ofsulfate and nitrate by dust particles (Jordan et al.,2003). However, we do not observe a strongrelationship between the equivalence ratios ofNHþ4 /SO

2�4 and Ca2+/SO2�

4 (r ¼ 0:20) as has beenreported from high elevations sites in central Asia(Sun et al., 1998). This is probably due to lower dustloadings in the King Col aerosol relative to CentralAsian aerosols, making (NH4)2SO4 a far moreimportant sulfate-bearing aerosol species thanCaSO4 at King Col. King Col aerosol Ca2+

averages 0.52 neqm�3 STP, while at a 5000m sitein the Himalaya aerosol Ca2+ averages1.24 neqm�3 STP during the monsoon season, whendust loadings are at their annual minimum, whileconcentrations are even higher at other times of theyear (Shrestha et al., 1997).

The Cl�/Na+ equivalent ratio on a per-samplebasis averages 0.69, considerably less than theseawater ratio of 1.16 (Keene et al., 1986), indicat-ing a Cl� deficiency in the King Col aerosol relativeto the Cl�/Na+ seawater ratio despite the highcorrelation between the two. Differences in Cl�/Na+ ratios relative to seawater may be attributableto acidification of sea-salt particles by reaction withH2SO4 (Legrand and Delmas, 1988). This reactionvolatilizes sea-salt chloride to produce gas phaseand highly soluble HCl, resulting in depletion ofaerosol Cl� and enrichment of Cl� in precipitation.Although the mean NHþ4 /SO

2�4 molar ratio in the

King Col aerosols is 2.16, suggesting (NH4)2SO4 isthe principal ammonium sulfate aerosol species,there is an excess of SO2�

4 charge equivalentsrelative to NHþ4 in 49% of the samples. A goodcorrelation between the Cl� deficit and unneutra-lized SO2�

4 (SO2�4 minus NHþ4 , in equivalence units)

in our samples (r ¼ 0:57) suggests this mechanism isindeed responsible for the Cl� deficiency in the KingCol aerosol. Acidification of sea-salt aerosol bynitric acid may be an additional volatilizationmechanism contributing to the aerosol Cl� defi-ciency relative to seawater.

Apportionment of the King Col aerosol into sea-salt and non-sea-salt fractions using Na+ andMg2+ as the sea-salt indicators (K+, Ca2+, andSO2�

4 were not found to be limiting in any of thesamples) suggests that nearly all (490%) of theaerosol Na+ at King Col is from sea salt. Mean-

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]]6

UNCORREC

while, nearly all (490%) of the aerosol K+, Ca2+,and SO2�

4 is from non-sea-salt sources. Only Mg2+

is found to have significant contributions from bothsea-salt and non-sea-salt sources, with 1/3 of theaerosol Mg2+ at King Col from sea salt.

3.2. Temporal variations in aerosol chemical

composition

Multi-day variations in aerosol concentrations atKing Col (Fig. 2) can be related to changes inaerosol source regions and transport and to thetiming of precipitation events. To investigate aero-sol source regions and transport pathways, back-ward trajectories using King Col as the end pointwere calculated for elevations of 3500, 4200 and5000m above sea level using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYS-PLIT) model (Draxler and Hess, 1997) with theNCEP global reanalysis data and model verticalvelocity (available online at http://www.arl.noaa.-gov/ready/hysplit4.html). While the complex terrainof the St. Elias Mountains cannot be captured, thebackward trajectories can indicate source regionsand large scale circulation patterns affecting theregion during the sampling campaign (e.g., Carricoet al., 2004). Knowledge of aerosol source regions(arid regions for dust aerosols, urban centers foranthropogenic aerosols) can be used to infer wheredifferent aerosols were added to an air mass. Tightlycoupled trajectory paths between 3500 and 5000melevation provide confidence in the model calcula-tions for trajectory paths reaching King Col, whiledivergent trajectory paths over this elevation rangesuggest a degree of uncertainty in the trajectorypaths. We discuss aerosol source regions as deducedusing backward trajectories for four aerosol con-centration peaks observed during the 17 May–11June sampling period, as well as a ‘‘clean air’’ periodfree of local precipitation (Fig. 3).

Events 1 and 2 are characterized by highconcentrations of Ca2+, Mg2+, NO�3 , NHþ4 , andSO2�

4 on 22 May and 26–27 May (Fig. 2). Modeledback trajectories for Event 1 diverge significantlyafter only 48 h. Consequently, we have less con-fidence in source region identification, although anAsian influence is suggested by the trajectoriesending at 4200 and 5000m. However, trajectorypaths for Event 2 are tightly coupled back throughtime and suggest an Asian source region incorpor-ating both crustal dust from Central Asia andanthropogenic pollution from northeast China and

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

TED PROOF

Korea (Fig. 3). Trajectories move around the baseof a low-pressure trough from central Siberia andacross the Gobi desert region, as would be expectedduring an outbreak of continental polar air con-comitant with dust storm generation (Sun et al.,2001). Trajectories continue across industrializedregions of Korea and northeast China where ananthropogenic SO2�

4 and NO�3 component couldhave been added to the air mass. The air then movesnorth around a high pressure ridge over the westernPacific where precipitation scavenging would beminimal before being entrained by the AleutianLow and advected into the St. Elias Mountains.

Precipitation plays an important role in short-term reductions of aerosol concentrations becausewet deposition is an important aerosol removalprocess (Shrestha et al., 2000). This is illustrated byabrupt decreases in aerosol concentrations at KingCol that are closely tied to the timing of localsnowfall events. For example, the King Col aerosolwas cleansed by snowfall events on 22–23 May and27 May, as reflected by abrupt decreases in aerosolconcentrations, ending the aerosol concentrationpeaks associated with Events 1 and 2.

Event 3 is a peak in aerosol SO2�4 and NHþ4 on

30–31 May without concurrent increases in aerosolspecies indicative of marine (Na+, Cl�) or crustal(Ca2+, Mg2+) sources (Fig. 2). A volcanic eruptionis one possible source of the SO2�

4 peak that wouldnot be associated with concurrent enhancements indust or sea-salt species. Three major eruptions ofSheveluch Volcano on the Kamchatka peninsulaoccurred between 19 and 22 May (Bulletin of theGlobal Volcanism Network, 2001). The eruption of22 May was the largest and reached an altitude of20 km. Back trajectories suggest the 22 Mayeruption plume did in fact reach the St. Elias on31 May within a narrow elevation range centered onKing Col. Trajectories arriving at lower (3500m) orhigher (5000m) elevations between 30–31 Mayoriginated south of the Aleutians or near the BeringStrait, respectively, and may not have carried thesignature of the Sheveluch eruption. Consequently,this moderately large eruption (VEI 4) may not beidentifiable in glaciochemical records from lower(Eclipse Icefield) or higher (Mt. Logan) sites,illustrating that differences in transport patternscan result in distinct paleovolcanic records atnearby sites with different elevations.

The volcanic SO2�4 peak at King Col is associated

with a concurrent enhancement in NHþ4 . This couldbe explained by oxidation of volcanic SO2 in the

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

NCORRECTED PROOF

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

Fig. 3. Air mass back-trajectories generated using the NOAA HYSPLIT model for four aerosol concentration events observed at King

Col during the 17 May–11 June sampling period: (A) high Ca2+–Mg2+–NHþ4 –SO2�4 –NO�3 event ending 26 May 14:00 UTC, (B) high

NHþ4 –SO2�4 event ending 31 May 05:00 UTC, (c) high Na+–Cl� event ending 5 June 04:00 UTC and (D) clean air period ending 2 June

16:00 UTC.

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]] 7

Uplume to produce sulfuric acid aerosols, which arethen neutralized by reaction with ammonium frombiological or combustive sources. If neutralizationof volcanic SO2�

4 by NHþ4 commonly occurs beforepreservation in glacial ice, then volcanic SO2�

4

deposition would be associated with concurrentenhancement in NHþ4 . Statistical techniques for

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

extracting paleovolcanic signatures from multivari-ate glaciochemical records, such as EOF analysis,are based on the assumption that volcanic SO2�

4

deposition should be independent of other iondeposition (Zielinski et al., 1996; Yalcin et al.,2003). This may not be the case when volcanic SO2�

4

has reacted with NHþ4 to produce ammonium

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]]8

C

sulfate aerosols. Although NHþ4 has been routinelyexcluded from such analyses, justification for thismethodology may be provided by our observationsof the Sheveluch volcanic plume at King Col.

Event 4 is a three-pronged peak in sea-salt species(Na+ and Cl�) with the peak aerosol concentrationsoccurring on 3, 5, and 7 June (Fig. 2). The daysfollowing the concentration peaks (4, 6, and 8 June)were days with snow, again illustrating the cleansingeffect of precipitation on aerosol concentrations.Back trajectories for all 3 days of high sea-saltconcentrations were similar, indicating air massesfrom the Gulf of Alaska moved along the southeastAlaska coast and into the St. Elias Mountains (Fig.3). Furthermore, the air masses sampled at KingCol during this period had been lofted from marineboundary layer within the prior 48 h. Thesetrajectories provide clear evidence that the Gulf ofAlaska is the source of sea-salt concentration peaksobserved at King Col. This sea-salt trajectory issimilar to trajectories associated with low concen-trations of all aerosol species, such as was observedon 2 June. However, the trajectories of 2 June hadbeen cleansed by orographic passage over the CoastMountains of British Columbia within the 48 hprior to reaching the St. Elias Mountains (Fig. 3).Hence, low aerosol concentrations at King Coloccur when air masses have recently been scavengedduring orographic lifting prior to reaching the St.Elias Mountains, or following local precipitationevents.

85

87

89

91Table 4

Correlation coefficients (n ¼ 193) between soluble species in fresh

ORRE

3.3. Characterization of fresh and surface snow

chemistry

Concentrations of water-soluble ions in fresh andsurface snow at King Col are presented in Table 3and Fig. 2. The average total ionic burden (totalcations plus total anions) ranged from 1.88 to

UNC

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

Table 3

Summary of fresh and surface snow composition (n ¼ 193) at

King Col (ueq l�1), 17 May–11 June 2001 (unit: ueq l�1)

Na+ NHþ4 K+ Mg2+ Ca2+ Cl� NO�3 SO2�4 C2O

2�4

DC

Mean 0.29 0.79 0.08 0.21 0.97 0.45 1.07 1.18 0.05 �0.39

Median 0.19 0.78 0.05 0.17 0.82 0.36 0.97 1.06 0.04 �0.21

Std. dev. 0.34 0.34 0.09 0.15 0.65 0.38 0.78 0.75 0.05 0.90

Max 3.04 1.74 0.80 0.65 3.81 4.02 3.44 3.91 0.30 1.90

Min 0.01 0.13 0.00 0.03 0.09 0.06 0.17 0.13 0.01 �3.16

DC: charge balance (sum of measured cations minus sum of

measured anions).

TED PROOF

11.79 ueq l�1 and averaged 5.07 ueq l�1. Ammoniumand sulfate represent a smaller fraction of the totalsoluble ionic impurities in the snow (40%) than inthe aerosol (49%). Meanwhile, both dust (Ca2+ andMg2+; 23%) and NO�3 (20%) represent greaterfractions of the total ionic burden in snow relativeto the aerosol (11% and 5%, respectively). Thecharge balance (sum of cations minus sum ofanions) of King Col fresh and surface snowaverages �0.39 ueq l�1, with an anion excess in62% of the samples that is presumably balanced bythe hydrogen ion. Samples with a cation excess arepresumably balanced by carbonate ions, in the caseof samples with high dust concentrations, and/ororganic acids (e.g., formate, acetate, methylsulfo-nate).

As with aerosol samples, linear regression ofspecies measured in snow samples (Table 4) showedthat the sea-salt species Na+ and Cl� are highlycorrelated (r ¼ 0:90), as are the dust species Ca2+

and Mg2+ (r ¼ 0:89). Calcium and magnesium arealso highly correlated with SO2�

4 (r ¼ 0:87 and 0.83,respectively), but not as well correlated with NO�3 insnow samples (r ¼ 0:63 and 0.58, respectively)compared to aerosol samples (r ¼ 0:90 and 0.80,respectively). The equivalence ratios NO�3 /NHþ4 andNO�3 /SO

2�4 suggest enrichment of NO�3 in snow

(1.36 and 1.09, respectively) relative to the aerosol(0.20 and 0.19, respectively). This could reflect drydeposition or snow scavenging of gas-phase HNO3

(Baltensperger et al., 1992; Shrestha et al., 1997;Toom-Sauntry and Barrie, 2002). Furthermore,NHþ4 is highly correlated with SO2�

4 (r ¼ 0:70) insnow, as with the aerosol, but not as well correlatedwith NO�3 (r ¼ 0:55), again suggesting snow scaven-ging of gas-phase HNO3. Although enrichment of

93

95

97

99

101

103

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

and surface snow collected at King Col, 17 May–11 June 2001

Na+ NHþ4 K+ Mg2+ Ca2+ Cl� NO�3 SO2�4

NHþ4 0.43

K+ 0.90 0.57

Mg2+ 0.09 0.63 0.23

Ca2+ 0.08 0.67 0.23 0.94

Cl� 0.81 0.61 0.90 0.32 0.37

NO�3 0.06 0.64 0.37 0.60 0.67 0.55

SO2�4

0.16 0.75 0.37 0.87 0.91 0.44 0.71

C2O2�4

0.16 0.54 0.51 0.33 0.33 0.50 0.72 0.50

Bold entries denote correlations significant at the 99% confidence

level. Italicized entries are significant at the 90% confidence level.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

Table 5

Scavenging ratios for water-soluble ions calculated from con-

current measurements of aerosol and fresh snow chemistry

(n ¼ 21) at King Col, 17 May–11 June 2001

Na+ NHþ4 K+ Mg2+ Ca2+ Cl� NO�3 SO2�4 C2O

2�4

Mean 462 1559 700 2697 4540 872 8253 1708 1421

Std. dev. 361 1504 479 1893 2972 779 5744 1072 817

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]] 9

UNCORREC

SO2�4 relative to NHþ4 is often observed in

precipitation due to heterogeneous SO2 oxidationin clouds (Calvert et al., 1985), SO2�

4 /NHþ4 ratios atKing Col suggest only slight SO2�

4 enrichment insnow (1.06) relative to aerosol (0.93). This resultimplies that much of the sulfur reaching King Colhas already been oxidized to SO2�

4 aerosol. Inter-estingly, all correlations are positive for the snowdata while the aerosol data show negative correla-tions between sea-salt (Na+, Cl�) and some non-sea-salt (Ca2+, NHþ4 , NO�3 ) species (Table 2). Thisreflects the greater importance of removal processes(wet and dry deposition, riming) in controllingconcentrations in snow, while air mass sourceregion and transport processes are of greaterimportance in controlling aerosol concentrations.

In addition to snow scavenging of gas-phaseHNO3, there is also evidence for snow scavenging ofgas-phase HCl at King Col. While the Cl�/Na+

ratio in aerosol samples showed a chloride deficitrelative to the seawater ratio, snow samples show achloride excess. The Cl�/Na+ equivalent ratio inour samples averages 2.12, greater than both theseawater ratio of 1.16 (Keene et al., 1986) and theratio of 0.69 observed in aerosol samples. Excesschloride in snow could be attributable to precipita-tion scavenging of HCl produced by acidification ofsea-salt particles (Legrand and Delmas, 1988;Toom-Sauntry and Barrie, 2002). With respect tothe snow, Na+ was the limiting sea-salt species in90% of the samples, Mg2+ in 8%, K+ in 1%, andSO2�

4 in 1%. As was the case for aerosols, nearly allof the Ca2+ (99%), SO2�

4 (96%), and K+ (90%) arefrom non-sea-salt sources while nearly all of theNa+ (96%) is from sea salt. Only Mg2+ has bothsignificant sea-salt (35%) and non-sea-salt (65%)sources.

3.4. Relationships between aerosol and snow

chemistry

To quantify the relationship between concentra-tions of soluble ionic impurities in King Col snowand aerosols, scavenging ratios (W) were calculatedaccording to

W ¼ pairCsnow=Cair,

where Csnow is the concentration in snow (ng g�1);Cair the concentration of aerosol (ngm�3 STP, andpair the density of air (1225 gm

�3 at STP) (Davidsonet al., 1993). Large values of W indicate efficientscavenging of that species by precipitation. Scaven-

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

TED PROOF

ging ratios were calculated from those aerosolsamples with concurrent fresh snow samples (21 of48 aerosol samples) and limited to aerosol measure-ments above detection limits.

Calculated scavenging ratios (Table 5) are con-sistent with values reported in the literature for sitesin Greenland (Davidson et al., 1985, 1993; Silventeand Legrand, 1993) and Central Asia (Sun et al.,1998; Shrestha et al., 2002), except for Ca2+, whichis found to be a factor of 2 higher at King Col.Meanwhile, scavenging ratios at King Col are afactor of two to four higher than those reported forJungfraujoch (3450m) in the Swiss Alps, except forNa+ and Cl� which show lower scavenging ratios atKing Col (Schwikowski et al., 1995). The higherscavenging ratios at King Col and other sitesrelative to Jungfraujoch might be explained by thefact that Jungfraujoch was in the cloud during allobserved precipitation events, hence the reportedscavenging ratios include only in-cloud processes,whereas below-cloud scavenging processes may alsobe important elsewhere. Event-based studies such asthat presented here report scavenging ratios asmuch as an order of magnitude higher thanscavenging ratios calculated from annual averageconcentrations because event-based studies reflectonly the conditions during precipitation events(Schwikowski et al., 1999).

Scavenging ratios are highly variable at King Colfor all species as seen by their standard deviations.This variability could be due to differences inprecipitation formation, particularly the extent ofsnowflake riming, but may also be influenced byspatial variability in snow chemistry, wind redis-tribution and mixing of fresh snow with older snow,and differences in aerosol compositions betweenprecipitating air masses. Higher scavenging ratiosare observed for species associated with coarse dustparticles (e.g., Ca2+ and Mg2+) compared tospecies associated with accumulation mode aerosols(e.g., NHþ4 and SO2�

4 ), reflecting more efficientprecipitation scavenging of coarse dust particles.

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]]10

UNCORREC

Accumulation mode aerosols (0.1–1.0 mm diameter)represent coagulation of ultra-fine nuclei modeparticles produced by gas-to-particle conversions,such as SO2�

4 from SO2 and NHþ4 from NH3 (Hinds,1999). The highest scavenging ratios are observedfor NO�3 , reflecting enrichment of NO�3 in snowrelative to aerosol by precipitation scavenging ofgas-phase HNO3 (e.g., Davidson et al., 1985;Shrestha et al., 2002). Scavenging ratios for Cl�

are twice those observed for Na+, indicating snowscavenging of gaseous HCl, leading to enrichmentof Cl� in snow relative to aerosol and relative toseawater ratios. It is possible that the high scaven-ging ratios for NO3 and Cl� could partly result fromnegative sampling artifacts through evaporation ofHNO3 and HCl from the filters. However, molarratios suggest full neutralization of sulfate by theammonium-rich King Col aerosol observed duringthe sampling period, suggesting volatilization on thefilters by reaction with sulfate acidity may benegligible. Furthermore, the dissociation constantfor ammonium nitrate decreases with decreasingtemperature and increasing relative humidity(Chang et al., 2000). Since the ambient temperatureat King Col did not exceed 0 1C during the samplingperiod and relative humidity is likely to approachsaturation during precipitation events, the calcu-lated scavenging ratios may not significantlyaffected by evaporative nitrate losses.

As evident from the time series of aerosol andfresh/surface snow chemical concentrations mea-sured at King Col (Fig. 2), increased concentrationsin the aerosol are generally reflected by increasedconcentrations in the snow, and vice versa. Tofurther investigate the relationship between aerosolconcentrations and concentrations in snow, linearregressions were performed between aerosol andfresh snow concentrations for each of the ninespecies measured. Aerosol samples were paired withconcurrent fresh snow samples collected immedi-ately after the aerosol sample was collected. Weused the mean concentration of the three to fivereplicate snow samples collected. Correlation coeffi-cients (r values) are low for species associated withsea salt (Na+ and Cl�) and dust particles (Ca2+,Mg2+, K+), ranging from 0.12 to 0.26. Correlationsbetween aerosol and snow concentrations are alsolow for NO�3 (r ¼ 0:20). Stronger correlationsbetween aerosol and snow concentrations areobserved for species present primarily as accumula-tion mode aerosols, such as NHþ4 (r ¼ 0:56) andSO2�

4 (0.43). The correlation between aerosol and

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

TED PROOF

snow concentrations is also high for C2O2�4

(r ¼ 0:63). The high correlation for C2O2�4 could

reflect the consistently low concentrations and lackof variability in both the aerosol and snow data.

A significant contribution from gas-phase HNO3

to snow NO�3 concentrations has already beendemonstrated; hence the weak correlation betweenaerosol NO�3 and snow NO�3 is expected. Snowscavenging of gaseous HCl could also explain theweak correlation between aerosol Cl� and snowCl�. Greater spatial variability in replicate snowsamples for species associated with dust and sea-saltparticles may contribute to weak correlationsbetween aerosol and snow concentrations. Thecoefficient of variation (standard deviation dividedby mean) for replicate snow samples is considerablyhigher for species present as coarse mode dust(Ca2+, Mg2+) and sea-salt (Na+, Cl�) particles(0.46–0.73) than for species associated with accu-mulation mode aerosols such as NHþ4 (0.23) andSO2�

4 (0.12). The greater spatial variability in freshsnow concentrations for coarse mode particlescould weaken the relationship between aerosol andsnow concentrations for Ca2+, Mg2+, K+, Na+,and Cl�. Interestingly, the coefficient of variation issignificantly lower for Cl� (0.46) compared to Na+

(0.73) since scavenging of gaseous HCl is animportant contributor to snow chloride concentra-tions. Likewise, a low coefficient of variation forreplicate snow samples was also observed for NO�3(0.17) since scavenging of gas-phase nitrate is animportant control on snow NO�3 concentrations.Greater spatial variability for species present ascoarse mode particles relative to species foundprimarily in the accumulation mode or gas phaseis a consistent result of snow chemistry surveys(Dibb and Jaffrezo, 1997; Yalcin et al., 2006a, b).

Differences between the observed aerosol andsnow concentrations could also be a result ofcompositional differences between the aerosolsampled at ground level and aerosol concentrationsat the heights where snow is forming (e.g., Dibb andJaffrezo, 1997), as well as differences in the timeintegrated by aerosol and snow samples. However,the high elevation of King Col often results inorographic precipitation due to upslope flow, andhence snowflakes are forming at altitudes similar toKing Col during such episodes. The degree ofsnowflake riming could also affect the relationshipbetween aerosol and snow concentrations becauseof different scavenging efficiencies for rimed snow,unrimed snow, and fog (Bergin et al., 1995). This

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

101

103

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]] 11

UNCORREC

idea is supported by two observations. First, fogwas a common occurrence at King Col during thesampling period. Second, replicate fresh snow andrime samples collected concurrently from a cleansurface on 4 June showed consistently higherconcentrations in rime relative to snow samples.Enrichment in rime relative to snow ranged from21% for NHþ4 and 46% for SO2�

4 to as much as80–95% for dust and sea-salt species and 102% forNO�3 . Hence, the extent of riming may affect therelationships between aerosol and snow chemistryto a greater degree for dust, sea salt, and NO�3relative to NHþ4 and SO2�

4 . However, because rimesamples were collected on just a single day, morework is needed to quantify the relative importanceof riming for air–snow relationships at King Col.

4. Conclusions

King Col, a remote high elevation site in the St.Elias Mountains, provides a unique platform tosample the aerosol chemistry in the remote freetroposphere. Ammonium and sulfate are highlycorrelated in the King Col aerosol (r ¼ 0:90) with amean NHþ4 /SO

2�4 molar ratio of 2.16 indicating

their presence as (NH4)2SO4. Dust species (e.g.,Ca2+) are well correlated with SO2�

4 (r ¼ 0:75) andNO�3 (r ¼ 0:91), implying uptake by dust particles.Nitrate is also well correlated with NHþ4 (r ¼ 0:73),which together with the high NHþ4 /SO

2�4 ratio

suggests NH4NO3 aerosol may also be present.However, there is an excess of SO2�

4 chargeequivalents relative to NHþ4 in 49% of the samples.Volatilization of HCl by reaction with unneutra-lized SO2�

4 may explain the Cl� deficiency in KingCol aerosol relative to seawater.

An understanding of the relationships betweenaerosol and snow chemistry in the St. EliasMountains is also needed because a suite of icecores have recently been collected from this region,including one from King Col Fresh snow chemistrygenerally mimics aerosol chemistry with similarrelative abundances and interspecies relationshipsexcept for large enrichments in snow Cl� and NO�3relative to aerosol due to snow scavenging of gas-phase HCl and HNO3. Although relatively strongcorrelations between aerosol and snow concentra-tions were observed for species associated withaccumulation mode aerosols, such as NHþ4(r ¼ 0:56) and SO2�

4 (r ¼ 0:43), only weak correla-tions were observed for dust and sea-salt species.These results may be influenced by greater varia-

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

bility in replicate snow samples for species asso-ciated with coarse mode dust and sea-salt particles.Snow scavenging of gas-phase HCl and HNO3

influences relationships between aerosol and snowchemistry at King Col. Snowflake riming and fogdeposition may also influence these relationships,although more work is needed to quantify theimportance at King Col.

ROOFAcknowledgments

We would like to thank G. Holdsworth (ArcticInstitute of North America at the University ofCalgary) and C. Zdanowicz (Geological Survey ofCanada) for field assistance, and D. Fisher of theGeological Survey of Canada for logistical support.We would also like to thank K. Kreutz, D. Fisher, J.Licciardi, and an anonymous reviewer for helpfulcomments that improved this manuscript. TheNational Science Foundation Office of PolarPrograms funded this research.

TED PReferences

Baltensperger, U., Schwikowski, M., Gaggeler, H.W., Jost, D.T.,

1992. The scavenging of atmospheric constituents by alpine

snow. In: Schwartz, S.E., Slinn, W.G.N. (Eds.), Precipitation

Scavenging and Atmosphere–Surface Exchange. Hemisphere,

Washington, DC, pp. 483–493.

Barrie, L.A., Barrie, M.J., 1990. Chemical components of lower

tropospheric aerosol in the high Arctic: six years of

observations. Journal of Atmospheric Chemistry 11, 211–226.

Bassett, M., Seinfeld, J.H., 1983. Atmospheric equilibrium model

of sulfate and nitrate aerosols. Atmospheric Environment 17,

2237–2252.

Bergin, M.H., Jaffrezo, J.L., Davidson, C.I., Dibb, J.E., Pandis,

S.N., Hillamo, R., Maenhaut, W., Kuhns, H.D., Makela, T.,

1995. The contribution of snow, fog, and dry deposition to

the summer flux of anions and cations at Summit, Greenland.

Journal of Geophysical Research 100, 16275–16288.

Bulletin of the Global Volcanism Network, 2001, p. 26.

Calvert, J.K., Lazrus, A., Kok, G.L., Heikes, B.G., Walega, J.H.,

Lind, J., Cantrell, C.A., 1985. Chemical mechanisms of acid

generation in the troposphere. Nature 317, 27–35.

Carrico, C.M., Bergin, M.H., Shrestha, A.B., Dibb, J.E., Gomes,

L., Harris, J.M., 2004. The importance of carbon and mineral

dust to seasonal aerosol properties in the Nepal Himalaya.

Atmospheric Environment 37, 2811–2824.

Chang, M.C., Sioutas, C., Kim, S., Gong, H., Linn, W.S., 2000.

Reduction of nitrate losses from filter and impactor samplers

by means of concentration enrichment. Atmospheric Envir-

onment 34, 85–98.

Davidson, C.I., Santhanam, S., Fortmann, R.C., Olson, M.P.,

1985. Atmospheric transport and deposition of trace elements

onto the Greenland Ice Sheet. Atmospheric Environment 19,

2065–2081.

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.

ARTICLE IN PRESS

AEA : 6845

1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

K. Yalcin et al. / Atmospheric Environment ] (]]]]) ]]]–]]]12

UNCORREC

Davidson, C.I., Jaffrezo, J.L., Mosher, B.W., Dibb, J.E., Borys,

R.D., Bodhaine, B.A., Rasmussen, R.A., Boutron, C.F.,

Ducroz, F.M., Cahier, M., Ducret, J., Colin, J.L., Heidam,

N.Z., Kemp, K., Hillamo, R., 1993. Chemical constituents in

the air and snow at Dye 3, Greenland II. Analysis of episodes

in April 1989. Atmospheric Environment 27, 2723–2737.

Dibb, J.E., Jaffrezo, J.L., 1997. Air–snow investigations at

Summit, Greenland: an overview. Journal of Geophysical

Research 102, 26795–26807.

Draxler, R.R., Hess, G.D., 1997. Description of the HYSPLIT-4

modeling system, NOAA Technical Memo, ERLARL-224,

24pp.

Fisher, D.A., Wake, C., Kreutz, K., Yalcin, K., Steig, E.,

Mayewski, P., Anderson, L., Zheng, J., Rupper, S., Zdano-

wicz, C., Demuth, M., Wasckiewicz, M., Dahl-Jensen, D.,

Goto- Azuma, K., Bourgeois, J., Koerner, R.M., Sekerka, J.,

Osterberg, E., Abbott, M., Finney, B.P., Burns, S.J. Stable

isotope records from Mount Logan and Eclipse ice cores and

nearby Jellybean Lake; water cycle of the North Pacific over

2000 years and over 5 vertical kilometers; sudden shifts and

tropical connections. Geographie Physique et Quaternaire, in

press.

Goto-Azuma, K., Shiraiwa, T., Matoba, S., Segawa, T.,

Kanamori, S., Fujii, Y., Yamasaki, T., 2003. An overview

of the Japanese glaciological studies on Mt. Logan, Yukon,

Canada in 2002. Bulletin of Glaciological Research 20, 65–72.

Hinds, W.C., 1999. Aerosol Technology. John Wiley, New York.

Holdsworth, G., Krouse, H.R., Peake, E., 1988. Trace acid

content of shallow snow and ice cores from mountain sites in

western Canada. Annals of Glaciology 13, 57–62.

Jaffrezo, J.L., Davidson, C.I., 1993. The Dye 3 Gas and Aerosol

Sampling Program: an overview. Atmospheric Environment

27A, 2703–2707.

Jordan, C.E., Dibb, J.E., Anderson, B.E., Fuelberg, H.E., 2003.

Uptake of nitrate and sulfate on dust aerosols during

TRACE-P. Journal of Geophysical Research 108, 8817.

Kanamori, S., Shiraiwa, T., Goto-Azuma, K., Benson, C.S.,

Naruse, R., 2004. Seasonal variations in density profiles and

densification processes at Mts. Logan and Wrangell. Eos

Transactions, American Geophysical Union, 85 (47), Fall

Meeting Supplement, Abstract PP21A-1367.

Keene, W.C., Pszenny, A.P., Galloway, J.N., Hawley, M.E.,

1986. Sea-salt corrections and interpretation of constituent

ratios in marine precipitation. Journal of Geophysical

Research 91, 6647–6658.

Legrand, M.R., Delmas, R.J., 1988. Formation of HCl in the

Antarctic atmosphere. Journal of Geophysical Research 93,

7153–7168.

Legrand, M., Mayewski, P.A., 1997. Glaciochemistry of polar ice

cores: a review. Reviews of Geophysics 35, 219–243.

Pathak, R.K., Chan, C.K., 2005. Inter-particle and gas–particle

interactions in sampling artifacts of P.M. 2.5 in filter-based

samplers. Atmospheric Environment 39, 1597–1607.

Schwikowski, M., Seibert, P., Baltensperger, U., Gaggeler, H.W.,

1995. A study of an outstanding Saharan dust event at the

high-alpine site Jungfraujoch, Switzerland. Atmospheric

Environment 29, 1829–1842.

Schwikowski, M., Brutsch, S., Gaggeler, H.W., Schotterer, U.,

1999. A high-resolution air chemistry record from an Alpine

Please cite this article as: Kaplan Yalcin et al., Relationships betwee

Yukon, Canada, Atmospheric Environment (2006), doi:10.1016/j.atm

TED PROOF

ice core: Fiescherhorn glacier, Swiss Alps. Journal of

Geophysical Research 104, 13709–13719.

Shrestha, A.B., Wake, C.P., Dibb, J.E., 1997. Chemical

composition of aerosol and snow in the high Himalaya

during the summer monsoon season. Atmospheric Environ-

ment 31, 2815–2826.

Shrestha, A.B., Wake, C.P., Dibb, J.E., Mayewski, P.A.,

Whitlow, S.I., Carmichael, G.R., Ferm, M., 2000. Seasonal

variations in aerosol concentrations and compositions in the

Nepal Himalaya. Atmospheric Environment 34, 3349–3363.

Shrestha, A.B., Wake, C.P., Dibb, J.E., Whitlow, S.I., 2002.

Aerosol and precipitation chemistry at a remote Himalayan

site in Nepal. Aerosol Science and Technology 36, 441–456.

Silvente, E., Legrand, M., 1993. Ammonium to sulfate ratio in

aerosol and snow of Greenland and Antarctic regions.

Geophysical Research Letters 20, 687–690.

Sun, J., Qin, D., Mayewski, P.A., Dibb, J.E., Whitlow, S.I., Li,

Z., Yang, Q., 1998. Soluble species in aerosol and snow and

their relationship at Glacier 1, Tien Shan, China. Journal of

Geophysical Research 103, 28021–28028.

Sun, J., Zhang, M., Tungsheng, L., 2001. Spatial and temporal

characteristics of dust storms in China and its surrounding

regions, 1960–1999: relations to source area and climate.

Journal of Geophysical Research 106, 18055–18066.

Thompson, L.G., Mosley-Thompson, E., Zagorodnov, V., Davis,

M.E., Mashiotta, T.A., Lin, P., 2004. 1500 years of annual

climate and environmental variability as recorded in Bona—

Churchill (Alaska) ice cores. Eos, Transactions, American

Geophysical Union 85(47), Fall Meeting Supplement, Ab-

stract PP23C-05.

Toom-Sauntry, D., Barrie, L.A., 2002. Chemical composition of

snowfall in the high Arctic: 1990–1994. Atmospheric Envir-

onment 36, 2683–2693.

Wake, C.P., Dibb, J.E., Mayewski, P.A., Zhongquin, L., Zichu,

X., 1994. The chemical composition of aerosols over the

eastern Himalayas and Tibetan Plateau during low dust

periods. Atmospheric Environment 28, 675–704.

Wake, C., Yalcin, K., Gundestrup, N., 2002. The climate signal

recorded in the oxygen isotope, accumulation, and major ion

time-series from the Eclipse Ice Core, Yukon. Annals of

Glaciology 35, 416–422.

Yalcin, K., Wake, C.P., 2001. Anthropogenic signals recorded in

an ice core from Eclipse Icefield, Yukon, Canada. Geophy-

sical Research Letters 28, 4487–4490.

Yalcin, K., Wake, C.P., Germani, M.S., 2003. A 100-year record

of North Pacific volcanism in an ice core from Eclipse Icefield,

Yukon, Canada. Journal of Geophysical Research 108.

Yalcin, K., Wake, C.P., Kang, S., Kreutz, K.J., Whitlow, S.I.,

2006a. Seasonal and spatial variability in snow chemistry at

Eclipse Icefield, Yukon, Canada. Annals of Glaciology 43, in

press.

Yalcin, K., Wake, C.P., Kreutz, K.J., Whitlow, S.I., 2006b. A

1000-year record of forest fire activity in three ice cores from

Eclipse Icefield, Yukon, Canada. The Holocene 16, in press.

Zielinski, G.A., Mayewski, P.A., Meeker, L.D., Whitlow, S.I.,

Twickler, M.S., 1996. A 110,000-yr record of explosive

volcanism from the GISP2 (Greenland) ice core. Quaternary

Research 45, 109–118.

101

n aerosol and snow chemistry at King Col, Mt. Logan Massif,

osenv.2006.06.028.