4tmospherw Emtronment Vol 26A No 11 pp 2001 2017 1992 0004 6981 92 $ 5 0 0 + 0 0 0 Pnnted in Great Britain ( 1992 Pergamon Press Ltd

REGIONAL ANALYSIS OF C L O U D CHEMISTRY AT HIGH ELEVATIONS IN THE EASTERN U N I T E D STATES

ZHENG LI and VINEY P ANEJA

Department of Marine. Earth and Atmospheric Scmnces. North Carohna State Umverslty Raleigh NC 27695-8208 U S A

(First received 15 February 1991 and m final form 13 January 1992)

Abstract--Results from the collection and chemical analysis of cloudwater samples collected from May to October 1986-1988 from the five high-elevation ( >950m MSL) Mountain Cloud Chemistry Program (MCCP) sites (Whiteface Mountain, NY, Mt Moosdauke, NH, Shenandoah Park, VA, Whitetop Mountain, VA, Mt Mitchell, NC) in the eastern United States are summarized The resulting database documents the regional chemical climatology of high-elevation forest ecosystems in the eastern U S Clouds occurred at these sites on 32-77% of the days during the sample collection period More than 90% of cloud samples were acidic (pH < 5 0) The lowest cloudwater pH (2 29 integrated 1-h collection penod) was recorded at Mt Mitchell, NC At all sites sulfate and nitrate were the dominant anions and hydrogen and ammonium were the dominant cations in cloudwater samples Mount Mitchell recewed the most acidic clouds and highest chemical exposures, while the Whiteface summit site received the least acidic and lowest chemical exposures compared to other MCCP high-elevation sites Cloud pH and major chemical components exhibited a seasonal trend with the maxima during the summer months, and correlated well with temperature and ozone concentrations The mean c.qmvalent ratios of SO~- to NO 3 were found to be 1 9-3 9 at these sites It is noted that SO 2- correlated highly with hydrogen ion, suggesting that contribution to cloud acidity by sulfate and/or Its precursors may be slgmficant

Ke~, word index Cloud chemistry, regional analysis, ozone, chemical exposure

1 INTRODUCTION

High-elevation forests in the eastern United States have shown signs of injury and dechne during the past two decades In recent years, there has been increasing concern with the possible impact of atmospheric acid- ity on forested ecosystems (Klein and Perkins, 1988, Cowling, 1989. Bruck et al 1989, Jacobson et al, 1990b, Hertel et al, 1990) It is now believed that acidic cloud deposition may contribute to observed forest decline at high-elevation locations where mountain slopes are frequently immersed in clouds (Jacobson et al, 1990a, Saxena and Lin, 1990. Aneja et al, 1990a, 1992, Cowhng et al, 1991)

Acidic clouds and fogs have been characterized over several decades in the U S and Europe, both in urban and rural areas Table 1 shows examples of cloud aodl ty measurements at some of those locations The cloud pH values reported by these investigators ranged from 2 2 to 7 6 Most authors focused on individual case studies, except for Weathers et al (1986), who studied a widespread acid cloud event at six non-urban sites in the eastern U S However. their study was based on one single event and limited to moderate elevation No research has been done for high-elevation regional cloud chemistry chmato- logical analysis from north to south tn the eastern U S to compare the acidity exposures between different sites Possible forest decline in high-elevation ecosys-

tems across the eastern U S necessitated docu- mentation of the chemical exposure and distribution of clouds at high elevation on a regional scale

Beginning in 1986, a series of measurements, includ- Ing major cations and anions in cloud water and precipitation, gas-phase measurements of ozone, sulfur dioxide and nitrogen oxides, and meteorologi- cal parameters (Aneja et al, 1992) were made at six remote sites in the eastern U S These measurements were made as part of the Mountain Cloud Chemistry Project (MCCP) sponsored by the U S Environ- mental Protectton Agency Five high-elevation (~> 950 m MSL) sites were selected from 35 to 45°N to be representatives of the geographic and meteorological variability in this large region One low-elevation site (Howland, ME, 65 m MSL) was instrumented to allow evaluation of the impact of elevattonal gradient In these ecosystems, red spruce (Pwea rubens Sarg ) and Fraser fir (Abws fraserl l-Pursh ] Poir ) are the domi- nant tree species, they have shown signs of dechne above the cloud base, which is frequently observed around 800-1200 m (Mohnen et al, 1990a)

The objectives of this research are to (1) characterize the exposure of montane forested ecosystems to chem- icals in cloud water at high elevations in the eastern U S, (n) determine nor th-south gradients of cloud chemistry based on observations made at five high- elevation M C C P sites, (Ul) study the regional chemical chmatology at high elevations in the eastern U S

2001

bo

I,o

Tab

le 1

T

he r

ange

s of

clo

udw

ater

aca

d~ty

obs

erve

d at

oth

er l

ocat

ions

Yea

r of

T

ype

of

Ran

ge o

f R

efer

ence

st

udy

Loc

atio

n co

llec

tor

pH

Clo

ud w

ater

rep

orte

d fr

om l

ow e

leva

tion

H

ough

ton

(195

5)

1954

N

orth

east

U S

A

stai

nles

s st

eel o

r m

ckel

scr

een

4 5-

7 2

Mro

se (

1966

) 19

57

Ger

man

y pr

evio

us c

loth

3

8-5

1 L

azru

s et

al

(197

0)

1967

P

uert

o R

aco

alum

anum

scr

een

4 9-

5 4

Wal

dman

et

al

(198

2)

1981

C

ahfo

rma,

U S

A

Cal

teeh

rot

atin

g ar

m

2 2-

-4 0

M

unge

r et

al

(198

3)

1981

-198

2 C

ahfo

rma,

U S

A

Cal

teeh

Tef

lon

stnn

gs

2 2-

5 8

Fuzz

a et

al

(198

4)

1982

N

ew Y

ork,

U S

A

scre

en m

apac

tor

4 3-

6 4

Jaco

b et

al

(198

5)

1982

-198

3 C

ahfo

mla

, U

S A

C

alte

ch r

otat

ing

arm

2

2-6

3 M

mr

et a

l (1

986)

19

85-1

986

Mid

wes

tem

U S

A

Cal

tech

rot

atin

g ar

m

2 9

-4 1

W

eath

ers

et a

l (1

986)

19

84

Eas

tern

U S

A

Tef

lon

stnn

gs

2 9-

3 0

Clo

ud w

ater

col

lect

ed b

y ai

rcra

ft

Odd

le (

1962

) 19

60

Um

ted

king

dom

gl

ass

tube

4

4-7

2

Pet

renc

hunk

and

D

rozd

ova

(196

6)

1961

-196

4 U

S S

R

inte

grat

ed s

ampl

e 3

4-5

9

Sco

tt (

1978

) 19

76

Aus

trah

a ce

ntri

fuge

4

6-7

5 S

cott

and

Lau

lam

en (

1979

) 19

77

Mic

higa

n, U

S A

ny

lon

wan

d 3

7-4

0 D

aum

et

al

(198

4)

1981

-198

3 E

aste

rn U

S A

sl

otte

d-ro

d lm

pact

or

3 1-

6 1

Sax

ena

et a

l (1

985)

19

82-1

983

McM

urdo

, A

ntar

ctic

a T

eflo

n pr

obe

4 9

-6 2

K

hem

am e

t al

(1

987)

19

83-1

985

Pun

e, I

ndm

st

mnl

ess

stee

l sh

eets

6

3-7

6 H

egg

and

Hob

bs (

1981

) 19

79

Nor

thw

este

rn U

S A

ce

ntri

fuge

4

2-6

5

Clo

ud w

ater

col

lect

ed a

t m

ount

ain

site

s O

ktta

(19

68)

1963

Ja

pan

copp

er s

cree

n 3

5-6

5 C

astd

lo (

1979

) 19

76

Whi

tefa

ce M

tn,

NY

, U

S A

ro

tati

ng s

tmnl

ess

stee

l tu

be

3 4

-4 2

F

alco

ner

and

Fal

cone

r (1

980)

19

77-1

979

Whi

tefa

ce M

tn,

NY

, U

S A

A

SR

C,

Tef

lon

stn

ng

2

7-4

7 W

eath

ers

et a

l (1

986)

19

84

Eas

tern

U S

A

Tef

lon

stri

ngs

2 8-

3 1

Moh

nen

and

Kad

lece

k (1

989)

19

82-1

987

Whi

tefa

ce M

tn,

NY

, U

S A

A

SR

C,

Tef

lon

stn

ng

2

5-4

8 A

neja

et

al

(199

0a)

1987

M

t M

itch

ell,

NC

, U

S A

C

alte

ch, T

eflo

n st

nn

g

2 9-

4 6

Ane

ja e

t al

(1

9901

))

1986

-198

8 M

t M

itch

ell,

NC

, U

S A

A

SR

C,

Tef

lon

stn

ng

2

2-5

5 S

axen

a an

d L

m (

1990

) 19

86-1

987

Mt

Mit

chel

l, N

C,

U S

A

AS

RC

, Tef

lon

stn

ng

2

2-5

4

N

go

e, >

High-elevation cloud chemistry 2003

2 MCCP SITE DESCRIPTION AND MEASUREMENT METHODS

The Mounta in Cloud Chemistry Program (MCCP) con- SlStS of five h~gh-elevatlon samphng sites in the eastern U S Whiteface Mountain, NY, Mt Moosflauke, NH, Shenandoah Park, VA, Whltetop Mountain, VA, Mt M~tchell, NC, and one low-elevation sampling site, Howland, ME Figure 1 illustrates the location of the M C C P sites

At Whiteface, Whltetop and Mt Mitchell, the main sites are located on the summit of mountains, while the main sites at Shenandoah and Moosdauke are along a rldgehne Sub- sites, such as those at Whiteface, Shenandoah and Mt Mitchell are located along the slope of the same mountain

The nor thernmost high-elevation site In the network is Whiteface Mounta in (WF) (44°2YN, 73~59'W), located m the northeastern Adirondack Mountains in New York, at an elevation of 1483 m The summit is above the tree hne, providing access to regional air flow (sub-site 1) The White- face Mounta in-Lake Placid Turn sampling site (sub-site 2) is located at 1245 m, adjacent to a balsam fir canopy

Mt Moosllauke (MS), NH (43°59'N, 71 °48'W), is one of the most southern peaks of the White Mountains It is located about 50 km southwest of Mt Washington (1917 m) and about 10 km northeast of the United States Forest Service (USFS) Hubbard Brook Experimental Forest and Water- shed The forest composmon ranges from mixed hardwoods at lower elevations to spruce-fir (about 10% spruce) at mid- elevations, and pure balsam fir at high elevations The meteorological s t a t lon-MCCP site at Mt Moosdauke is at 962 m and is partmlly shielded from the prevadmg westerly winds

The Shenandoah (SH), VA, site (38~72'N, 78°20'W) is in the Shaver Hollow Watershed, located in the north-central sector of the Shenandoah National Park The meteorological tower was erected m the watershed at an elevation of 1040 m The tower location ~s representative of the surrounding

deciduous forest canopy In ad&tlon to this site, two other sub-sites have been estabhshed

The Whltetop Mounta in (WT) site (36°38'N, 81°36'W) is located m the Mt Rogers Nauona l Recreation Area of the Jefferson National Forest m southwestern V~rgtma, 6 km southwest of Mt Rogers, the highest peak m Vlrgima The TVA Whltetop Mountain summit research station (at 1689 m) straddles the main ndgehne of the Appalachian range, strategically located to intercept air from several directions In addition to the summit station, other sub-sites have been established

The southernmost M C C P site is located m Mt Mitchell State Park (MM), NC (35°44'N, 82°16'W) The site is at Mt Gibbs (1950 m MSL)~ 2 5 k southwest of Mt Mitchell, which is the highest peak m the eastern U S (2038 m MSL) The summit is covered with Fraser fir, and the region from 1500 to 1800 m is an ecosystem composed mainly of mixed fir and spruce

The low elevation site is in the Howland Forest (HF), ME (45°IYN, 68J43'W) It is located at 65 m elevation near Howland, ME, 35 miles north of Bangor The forest is spruce with some of balsam fir, hemlock and white pine

Each of the locations had a meteorological walk-up tower to prowde measurements above the forest canopy The meteorological sensors were mounted on the top of the tower (Mohnen et al, 1990a) With regard to the cloudwater collection, which was performed manually at all sites, an ASRC (Atmospheric Science Research Center, State Umver- slty of New York at Albany) passive cloud collector (Fal- coner and Falconer, 1980) was mounted on the top platform of towers above the surrounding canopy at most sites The only exceptions to this are Whiteface Mountain, where the collector was placed on the roof of the summit research lab, and Whltetop Mountain, where it was on a platform built over the research trader The manual cloudwater collection commenced hourly during the cloud events The cloud event is signaled when a stationary object at a distance of 1 km

f

Wl

Whiteface (WF)

~d Forest (HF)

doosdauke (Ms)

N

T Whltetop

TN

GA

PA

VA

Mitchell

Washington DC

¢J ¢J

~3

o %

x~

% ,to

Fig 1 Map of eastern Umted States showing the locations of M C C P sites

2004 Z LI and V P ANnA

from the observation point becomes obscured by cloud and stays consistently out of view for more than 15 nun Once a cloud event has begun, a 500 ml polyethylene bottle was attached to the cloud collector with a clean tube Collection bottles were changed hourly At the end of each event, the cloud collectors were cleaned by rinsing with delomzed water untd the conductwlty of the nnse water was within + 10/zScm- 1 of that of the delonmed water The collectors were then covered with a plasuc bag untd the start of the next event Wash water was &scarded

Llqmd-water content of each cloud was measured with a gravtmetnc sampler (Valente et al, 1989) only at the main site The accuracy and precls~on of the instrument are + 0 1 and + 0 05 g m- 3, respectively, with the time basis of the sample measurement being 1-h integrated samphng The pH- meter was cahbrated before every measurement, using stand- ard buffer solutions at pH 4 00 and 7 00 The pH and total volume collected were measured ~mmedmtely after collec- tion Thereafter, these samples were refrigerated at ~4°C and were shipped for pH, conductwlty and major-Ion chenucal analysis to either the site speczfic wet chenucal laboratory or to llhnols State Water Survey (ISWS) laboratory The lomc chenucal composmon of the samples was determined using ion chromatography (IC)

The mamtenance and cahbratlon for these instruments were performed routinely under the MCCP protocol and Quahty Assurance/Quahty Control plan m order to ensure data quahty (Mohnen, 1990b) EPA audits at the sites and m chenucal laboratories were made dunng the first month of each field season The Central Analytical Laboratory (CAL) at the Ilhnols State Water Survey also prowded the QA/QC for analytical measurement for the MCCP Sample collec- tion, analys~s, and quahty assurance were cntlcal elements of this program At CAL, both non-precipitating 0 e clouds) and precipitating 0 e ramfali) samples from the sites were analysed for pH, conducUwty and concentrations of NH~, Na ÷, K + Ca 2+, Mg 2+ SO~-, NO~ and CI-

3 CHARACTERIZATION OF CLOUD CHEMISTRY

3 1 Cloud mterceptzonfrequency at M C C P sttes

The frequency of cloud Immersion detected at the five M C C P summit sites for 1986-1988 field seasons Is reported m Table 2 At the two northern sites, clouds occurred 37 (WF1) and 19% (MS1) of the time dunng the 3-year field seasons The three southern sites experienced cloud Immersion 11 (SH1), 30 (WT1) and

29% (MM1) of the time durmg the same penod Table 2 suggests that cloud frequency decreased from 1986 to 1988, reflecting a shift from wet weather m 1986 to drought condmons m 1988 over most of the eastern U S Mount Moosdauke and Shenandoah Park sites experienced fewer cloudy periods due to their lower elevations In comparison to the long-term chmate data, Mohnen et al (1990a) also reported that the northern M C C P sites experienced above-normal cloudiness m both 1986 and 1987 and below-normal cloudiness m 1988 However, the southern sites had below-normal cloudiness during all the three field seasons

Dmrnal patterns of cloud frequencies reported by Mohnen et al (1990a) indicate that a preference for cloud lmpactlon was during mght and morning hours (7 p m - 1 0 a m ) Whiteface Mountain, Whltetop Mountain and Mt Mitchell sites experienced more than twice as much cloudiness dunng the early morn- mg as m the afternoon Th~s phenomenon at summit sites may be hnked to orographlc mechanisms as well as to the planetary boundary layer lowenng below the level of the mountain top after sunset However, no such dmrnal trends were observed at Mt Moosdauke and Shenandoah Park

3 2 Cloud ac~dtty

The natural pH value of ram Is thought to be ~ 5 6, which represents the acidity of pure water m equthb- hum with the mean atmospheric concentration of CO2 (~330 ppm) This value Is often used as an important reference for acl&ty of cloud water (Robin- son, 1984, Lacaux et al, 1987) However, the acl&ty of natural cloud vanes m &fferent parts of the world (Table 1) For example, when there is a lack of the common basic compounds such as N H 3 or CaCO3, Charlson and Rodhe (1982) showed that the pH of ram water, influenced by sulfur compounds, can be expected to be about 4 5-5 0 This result was later confirmed by Noller et al (1986) Therefore, the pH of cloud water below ~ 5 0 is assumed to be influenced by anthropogenlc pollution

Table 2 Cloud frequency at MCCP sites, June to September 1986-1988

Elevation Site (m) 1986 1987 1988 Mean

Cloud frequency ( % o f hoursm cloud) WF1 1483 45 40 25 37 MSI 962 25 21 6 19 SH1 1040 18 7 6 11 WTI 1689 38 28 26 30 MM1 1950 35 28 23 29

Percentage oftotal days expenencmg some cloud WFI 1483 79 80 73 77 MS1 962 52 51 22 42 SH1 1040 46 25 25 32 WT1 1689 76 67 62 68 MM1 1950 84 75 68 76

H i g h - e l e v a t i o n cloud chemistry 2005

Figure 2 shows the frequency distribution of pH for cloudwater samples collected at the MCCP sites dur- ing the 3-year (1986-1988) field season For com- parison, the per cent occurrences of pH values for rain water at the same locations are also included The pH recorded at MCCP sites varied from 2 29 to 7 3 for cloud samples, and from 2 75 to 5 68 for precipitation, which was due m part to differences m liquid-water content (Aneja et al, 1992) The mean and minimum pH are given in Table 3 The mean pH was computed from the standard transformation

where [ H + ] is the hydrogen ion concentration in eq - 1 and n is the total number of samples The mean pH value for cloud samples (4 01) at the

Whiteface Mountain summit (site 1) was higher than at the Whiteface slope site 3 91, site 2) This result may reflect the vertical gradient of cloud acidity because the slope site is near the cloud base Several studies has shown that cloud water is more acidic at elevations near the cloud base (Kms et al , 1988) This vertical (spatial) variation in cloudwater concentra- tion may be explained if one assumes that LWC increases with height above cloud base (Pruppacher and Klett, 1980) thus providing increased dilution of pre-existing aerosol, vertically LWC data at the slope

20

Z l 0 ,

LU 5

0 ¸

20

~ 1 5 ¸

). tO z 10, w

O I.LI n- 5 M.

20

E 15

>- tO ,,=, 10

0 w ~ s

(a.oJg (a)

RAIN A

2 3 4 5 6 7 pH

30

25

20 ,b ,z,, 15

u 5

Ct.QUD RAIN

.k i

2 3

A ,.

4 5 6 pH

(b)

OJ:~UD (C) RAIN h

i 2 3 4 5 6

pH

30'

25-

~20 • 0 Z15" ul

010. u.

5"

Ct.OJO (d) RAIN

i 1 2 3 4 5 6

pH

2O

RAICt'~N ~l (e) . RAINCtDI'~ i t ( f )

-,&:' '-" fT,,'

pH pH

Fig 2 Frequency distributions of pH for both cloudwater and rainwater s a m p l e s co l l ec t ed from May to October 1986-1988 at (a) Whiteface Mountain, summit, (b) Whiteface Mountain, site 2, (c) Mt Moosllauke,

(d) Shenandoah Park, (e) Whltetop Mountain, (f)Mt Mitchell

2006 Z LI and V P ANF.JA

Table 3 Frequency (number of hours) of pH<40, 35, 30 and 25 at MCCP sttes, May to October 1986-1988

No of Site samples pH < 4 0 pH < 3 5 pH < 3 0 pH < 2 5 Mean Mm

WFI 634 58 8 (373) 22 1 (140) 1 26 (8) 0 4 01 2 75 WF2 66 63 6 (42) 42 4 (28) 9 09 (6) 0 3 91 2 74 MSI 200 715(143) 395(79) 155(31) 0 365 26 SHI 55 85 5 (47) 40 (22) 1 82 (I) 0 3 65 2 92 WTI 601 71 6 (430) 37 1 (223) 6 82 (41) 0 3 77 2 59 MMI 477 83 9 (400) 50 1 (239) 9 85 (47) 0 21 (I) 3 57 2 29

6

5-

4-

3

• • a . ( )

• " ~ . : •

• • • ~ee • •

• e

R'2 = 0 54g 2

' 0 ' ' ~ 6 . 0 ' ' 0 ' ' ' IO 2 30 40 5 70 8 go 100110120

OZONE (ppbv)

6

5

~4

3 ,

2 )0

(b)

• •e •• •

• . : ; ; . .

R'2 = 0451

2'0 3'0 4'0 5; 6; ~o

OZONE (ppbv)

5

~4

3

2

• ( c )

• • • e • • •

R~2 = 0 373 i ) i i

0 20 30 40 50 60

OZONE (ppbv)

7

6

5

4

3

2

\ " • (d)

%._.. • e • e e eeee •

, , '0 '0 , , , 0 20 30 4 5 O0 70 80 go

OZONE (ppbv)

5

~4

3-

( e )

ee • 'it' # •

R~2 = 0461 2

,0 2o 3~ 4o 5o 6'0 7'o 8'0 ~o ,~0, ;°,20 OZONE (ppbv)

Fig 3 Cloud pH values vs ozone concentrations measured from May to October 1986-1988 at (a) Whiteface Mountain, summit, (b) Mt Moosllauke, (c) Shenandoah Park, (d) Whltetop Mountain, (e) Mt

Mitchell

Hlgh-elevatmn cloud chemtstry 2007

it ÷+ + WF1, 1987 A >_,

A 4- ° 6

~ 4 0 4

2 2

0 5 10

( a ) TEMPERATURE (*C)

M S 1 , 1 9 8 7

. . . . i . . . . i . . . . I . . . . !

15 2 0 0 5 t 0 15 20

TEMPERATURE(°C)

..J 5 Z O O

4 O ..J " • ~" 3 - ¢k

(c)

SH1, 1987

+ 4-+ + + *= 4- 4-4-++, 4-

. . . . i . . . . ! . . . . | . . . . 5 10 15

TEMPERATURE (* C)

20

7

it WT1,1987

0 ÷ ÷

0 5 10 15 20

( d ) TEMPERATUREIOC )

6

5

4

~ 3-

F~g 4

MM1, 1987 4-

. . . . i . . . . | . . . . | . . . .

5 10 15 2 0

TEMPERATURE (°C)

Cloud pH values vs ambmnt temperature measured during 1987 at (a) Wh]teface Mountam, summit, (b) Mt Moosflauke, (c) Shenandoah Park, (d) Whltetop Mountam, (e) Mt Mitchell

site (site 2) was not measured and thus, we are proposing a hypothesis to provide insight m the vertical changes m cloudwater pH

The minimum cloud pH was 2 75 at Whiteface site 1, 2 74 at Whiteface site 2, 2 60 at Mt Moosllauke, 2 65 at Shenandoah Park, 2 59 at Whltetop Mountain and 2 29 at Mt Mitchell Falconer and Falconer (1980) reported that 90% of all cloudwater samples ranged m pH from 2 66 to 4 66 for the periods of August- September 1977 and 1979 at Whiteface summit Weathers et al (1986) observed a single acldm cloud event at several s~tes in the eastern U S m 1984 and found that the pH ranged from 2 8 to 3 09 Several other mvesngators studying urban fogwater ac~dtty m Cahfornla found that the pH was as low as ~2 2 (Waldman et al , 1982, Munger et al , 1983, Jacob et al ,

1985) The sample with pH value of 2 29 measured at Mt Mitchell in 1986 was more acl&c than that re- ported by Weathers (1986) for other sites in the eastern U S and by Mmr et al (1986) in an urban mldwestern site It was, In fact, close to the level of urban fog water m Cahforma

It is documented that cloud or fog with a pH less than 4 0 is believed to cause measurable damage to fohage and yield of some plants, such as red oak, white pines and spruce (Jacobson, 1984, Evans 1984, Jacob- son et al , 1990a) It is also known that even one exposure of pH less than 2 5 can damage certain crop species (Grantt et al, 1984) The number of hourly cloud samples, as well as percentage of samphng time, w]th pH <4 0, 3 5, 3 0 and 2 5 are also g~ven m Table 3 It appears that Mt Mitchell received the most acl&c

2008 Z L[ and V P ANEJA

cloud exposure during the samphng period, while the Whiteface summit site received the least acidic expos- ure Mount Moosflauke and Shenandoah Park sites are lower in elevation, with lower frequency of cloud lmmersmn, but close to the cloud base, thus the percentage of total samphng hours with acidic cloud (pH <4 0 and 3 5) Is higher than Whiteface summit and Whltetop Mountam

Based on an empmcal analysis, it is found that cloud pH value is a function of ambient temperature and ambient ozone concentrauon (F]gs 3 and 4) The best curve fit between pH and ozone is

[p i l l = A/[O3] B ,

where A and B are constants and greater than zero (for each case, first- and second-order relatmnshlps were also considered) The coefficients of determmatmn (r 2) were above ~ 0 4, while a linear relationship between cloud pH and ambient temperature (cloud acidity increasing w~th increasing ambmnt temperature) was obtained, suggesting a seasonal dependence

A seasonal variation of p H m cloud samples can also be noted (Fig 5) Mean pH values calculated by monthly bas~s are shown m F~g 5 for the five s~tes during the 3 years The results imply that strong acidic cloud waters are found in warm and humid summer months Earher studies by Aneja et al (1990a), who analysed the cloud pH collected m 1987 at Mt Mitch- ell, utlhzlng the cloud and ram ac~dlty/conductwlty (CRAC) real-t~me automated sampler, show a slmdar seasonal trend It IS observed that the high-pressure system and mr stagnation events frequently occurred during the summer months m the eastern U S, whtch allows a greater build-up of those acidic aerosols and atmospheric photochemical oxidants (Aneja et al, 1990b, 1991, Clmborn and Aneja, 1991) Enhanced

ozone at the locale by cloud passage appears to contribute to increased cloud acidity as well as to the concentrations of acidic species in the cloud water It is also found that cloudwater pH seasonal variation was consistent with the trend of atmospheric oxidants, such as 03 (Fig 6) The oxldaUon of SO2 might be a hmmng factor for the generatmn of atmospheric acid- ity (Calvert et al, 1985) The substantml 03 concentra- tion at high elevation in the eastern U S durmg the warm season slgmfies an oxidizing condmon con- duclve to the formation of HNO 3 and HzSO 4 from atmospheric NOx and SO2 When the photochemical production of H2Oz and 03 are relatively low m the cold seasons, the amount of oxidants may be hmmng for converting SOz and NOx to H2SO 4 and H N O 3 ,

which will reduce atmospheric acidity The meteoro- log]cal condltmns, such as temperature and cloud hqmd-water content, may also affect the varmtlon of pH in the cloud

3 3 Cloud chemtcal composition

Mean morgamc composmons of cloudwater sam- ples collected at five summit sites and one sub-site at Whiteface Mountam from May to October 1986-1988 are presented m Table 4 We are explonng inter- relatmnshlps with simple normal linear statistics We are appealing to the Central Llmlt Theorem and thus mean, regression coefficient and correlatmn, etc, are approximately normal, even ff the original dlstnbu- tmn may not be normal (Dickey, 1991) The mean inorganic mn balances for all sites were withm 6%, lndtcatmg that the concentratmns of orgamc acids, such as formic and acetic acid, m cloud water at the sites are small As may be expected, the aver- age concentrations noted at MCCP s~tes seem to be lower than those reported for urban fog water in

A

=J Z 0

0

50

46

42

38'

34'

30

WF1

1986-88 : MS1

SH1

WT1

/~ MM1 f MOWrH

Fig 5 Monthly averaged pH values measured at MCCP sites from May to October 1986-1988

Tab

le 4

S

tati

stic

al s

um

mar

y o

f ch

emxc

al s

pec

ies

m c

lou

d w

ater

(m

#eq

f -

1)

coll

ecte

d a

t M

CC

P s

ites

for

198

6-19

88

Sit

e S

O, 2

N

O~

C

1 -

H ÷

N

H~

N

a +

K

+

Ca

2 +

M

g 2

÷

Am

on

C

atio

n

An

ion

/cat

ion

WF

1

X

245

91

6 49

20

3 11

9 3

65

2 7

16

4 1

343

349

0 98

S

281

116

10

3

229

136

96

4

24

2

37

3

87

8

%

35

4

13 1

9 0

94

2

93

2

17

28

0

53

0

39

2

32

0

59

m

ax

2224

13

44

115

1778

92

0 14

8 61

0 52

6 74

WF

2

X

525

133

9 37

38

0 22

5 4

93

2 59

26

7

6 69

66

8 64

6 1

03

S 57

3 14

2 9

3 4

44

21

2 7

01

2 19

35

8

59

%

40

10

07

1

28

9

17

14

0

37

0

2

20

3

05

1

max

24

32

625

36 7

18

20

878

47

12

135

32 2

MS

I X

44

8 23

1 2

37

4

40

17

9 1

79

3

5

13

9

73

8

703

662

10

6

S 58

2 32

5 47

1

521

215

48

4 3

21 6

11

1

%

32

8

17

17

3

23

1

31

1

31

0

26

1

01

0

54

m

ax

2860

22

90

499

2511

92

9 55

6 38

7

152

77

SH

1

X

367

180

19 8

32

5 20

6 11

9

3 57

17

4

5 64

56

6 56

9 0

99

S 33

9 13

8 24

8

260

191

29 5

3

54

20 2

7

9 %

3

23

1

58

1

8

28

5

18

2

10

5

03

1

15

3

04

9

max

12

38

541

144

1202

84

0 17

4 14

8

83 9

46

2

WT

1 X

40

0 16

9 19

1

344

170

8 36

3

04

29 1

6

32

589

561

1 05

S

422

184

19 1

37

6 16

9 12

3

12

45 3

6

43

%

34

8

14 7

1

7 2

99

1

48

1

37

02

6

2 53

0

55

m

ax

2690

14

89

158

2570

10

70

158

31

442

50 1

MM

I X

57

6 20

5 32

9

462

218

19 4

7

78

55 1

13

9

809

776

1 04

S

608

239

58 8

48

3 24

8 4

0

45 7

12

3 24

6

%

36

1

12

9

21

2

91

1

37

1

22

0

49

3

47

0

88

m

ax

3700

21

18

691

5128

20

39

587

910

1189

22

5

m

¢11

,¢ o Q

t2t.

¢,0

X =

mea

n v

alu

es,

S =

sta

nd

ard

dev

iati

on

s, %

= p

er c

ent

of t

ota

l m

easu

red

io

nic

spe

cies

, m

ax =

max

imu

m v

alu

e m

easu

red

du

rin

g t

he

sam

pli

ng

per

iod

A

nio

n=

SO

]-

+N

O 3

+C

1-

Ca

tlo

n=

H +

+N

H,~

+ N

a t

+K

+ +

Ca

2+

+M

g 2

+

2010 Z LI and V P ANEJA

4 0 .

3 9 .

3 8 .

3 7 ,

3 6 .

3 5 .

3 4 -

3 3 -

3 1

3 O

MM1, 1986-88

- - - O - - ozone

" J d N " " s b

' 8 0

• 7 0

• 6 0

• 5 0

4 0

A > ,,s

v

W Z O N O

MONTH

Fig 6 Monthly averaged ozone concentrations and pH values measured at Mt Mitchell from May to October 1986-1988

1800

A 1600 O"

1400 =L

1200 Z 0

1000'

I'- 800 ' Z W (3 600' Z 0 ¢.3 400 '

200 '

0 WF1 WF2 MS1 SH1 WT1 MM1

[ ] Mg++

[ ] Ca++

[ ] K+ [ ] Na+

[ ] NH4+

[ ] H+ [ ] Cl [ ] NO0 [ ] ,SO4

SITES

Fig 7 Mean 1on concentraUons for the MCCP sites, from May to October 1986-1988

Los Angeles and the Midwest (Munger et al, 1983, Mmr et al, 1986), however, the maximum sulfate concentration (3700 #eq ( - 1) measured at Mt Mitchell was close to or h~gher than those at some of the urban locations mentioned above The sulfate and mtrate concentrations were generally higher than fog water reported at Albany, NY (Fuzz1 et al, 1984), and cloud water at high elevation m the Sierra Nevada (Collett et al, 1990) The Ca 2 + concentration, which is derived from soil dust, was the h~ghest among the trace metals at most s~tes, with one exceptmn at Mt Moosdauke, where the mean Na + concentration, thought to be of marine origin, was h~gher than the Ca 2 + concentra-

tlon Significant concentrations of metals and non-sea salt Ca 2 ÷ at most sites show that the air masses are continental in ongm (Klm and Aneja, 1992) On some occasions, extreme h~gh concentrations of these minor constituents in the cloud were observed, for example, CI - concentration reached 499 #eq f - 1 at Mt Moosll- auke and 6 9 1 ~ e q f -1 at Mt Mitchell, and Ca 2+ reached 526 #eq f - ~ at Whiteface Mountam site 1 and 1189/~eq f - ~ at Mt Mitchell However, the high trace- metal content m cloud may catalyse SO z oxldatmn when atmospheric H 2 0 z and 03 are relatwely low

Figure 7 illustrated the mean total concentraUons for all sites The exposure gradient (only m terms

H~gh-elevatmn cloud chemistry 2011

40 30

~ 2 0

0 0 oJ c0 ~- to ¢~o ~-~ 00 ca o o ~ oJ ~ ~ - m ~J~ r~ to c~ o o

o o o o o o o o o ~ ~ o o o o o o o o o ~

Al AI

o o o o o o o o o o o o o o o o o o o

SO4110"3 paq/I) SO4110"3 p.eq/i)

z

E

F~g 8

'°I ' 1 40-~- ()b I e} , 0 7 2 0 1 1 l 2 0

,,=, ~ 10

10 ta.

u. 0 0 o o o o o o o o ~

¢~ ¢~ ~o to t o h . 0~ ca o o AI o o o o o o Q o ~ o ~ c~ ¢o , ~ Lo ~D r~ to ca

AI O o o o o o o o o o o ~ cu co ~ ~ co r-. t o o~

o o o o o o o o o SO4 (10"3 i~eq/I) S04 (10"3 l~qll) 30

,0~ i0) I ~ ~0 30 ~ ¢~

20 ~ 10

I1. 10 0 0 ^l

od ~ ~ to to t~ to ca o o o o o o o o o o o ~ ~ o ~ eu ¢0 ~ t to ~ r-- t o ca

o o o o o o o o o o Al

o ~ t'-,J ~ ~ to co r~ to ca

o o o o o o o o o o ~ ( 1 0 " 3 I~l/I) S04 (10"3 ~t~l/I)

Frequency distributions, M a y - O c t o b e r 1986-1988, of SO~- at (a) Whiteface Mounta in , summit, (b) Whiteface Mounta in , s~te 2, (c) M t Moosdauke, (d) Shenandoah Park, (e) Whl te top Mounta in , ( f ) M t M~tchell

of concentration) along north to south Appalachians can be observed The figure indicates that the nor- thernmost site, Whiteface Mountain site 1, had the lowest chemical concentration exposure m the cloud, while the southernmost site, Mt Mitchell, experienced highest concentration exposure The lower elevauon Mt Moosllauke site had more concentrated cloud than Shenandoah Park and Whltetop Mountain

The frequency dlstnbutlons of the pnnopa l ions, S O ~ - , N O ~ and N H +, at the five sites are dlustrated m Figs 8-10 The frequency distnbutions of equivalent concentration of ,ons are based on hourly samples Our intent is to present extremes m concentrations within events based on 1-hour samphng of clouds Thus volume-weighted averages for events are not

uuhzed At the northern sites, over 60% SO~- concen- trations were less than 200 # e q / - ~ at Whiteface site 1, 40% at Whiteface site 2 and ~ 5 0 % at Mt Moosd- auke At the southern sites, over 40% SO 2- were less than 200 # e q / - ' at Shenandoah Park and Whltetop Mountain, and 28 5% at Mt Mitchell

3 4 Ratios of wn concentrations

The raUos of mean major ion concentrations at the high elevation sites are listed in Table 5 The eqmval- ent ratios of mean SO 2- to N O ~ were found to be between ~ 2 and 3 9 at the M C C P sites Nearly three times higher inputs of S O l - than N O ~ were noted at Whiteface Mountain site 2 and two times higher at Mt Mitchell, suggesting that the contnbutlon of sulfate to

~(A) ~:II-0

2012 Z L~ and V P ANEJA

80 50 ~ L ~

g 6°. ~ 4o

0 Z 40' i . :~ ~ 20 g ~

0 0 t~ o~ .q- to ~ r-- Q0 ¢n o o

o o o o o o o o ~

$ . . . . . . . o o o ; . . . . . . . . . . . . . NO3 116"3 peq/I)

NO3 (10"3 l.aD¢lll )

60 0

50 40

O ~" 40 t~ 30 " 1 ~ [

m 30 ~ 20

0 m 2 0 ,.= ~ lo

10

0 0

o o o o o o o o o ~ ~ AI

o~ ~ eq t~ ~6- to co t~ co en o o o o o o o o o o o o o o o o o

NO3 (10"3 i,Leq/I) NO3 (10"3 peq/I)

50'

so I (c) I ~" 4o (0

~ ~ 30 ':lu.._2 8 ~ ~ 20 20

u. 1

0 0

o o o o o o ° ~'~ ~ AI

o o o o o o o o o o

NO300**3 p.e¢l/I) NO3 (10"3 ite~:l/I)

Fig 9 Frequency distributions, May-October 1986-1988, of NO 3 at (a) Whtteface Mountain, summit, (b) Whiteface Mountain, s~te 2, (c) Mt Moosdauke, (d) Shenandoah Park, (e) Wh~tetop Mountain, (f) Mt M~tchell

acidity may be considerable at high elevations m the eastern U S This is m contrast to the observations m the western U S 0 e southern Cahforma), for which mtrate ls the dominant anion due to the high NOx emission m those areas (Waldman et al, 1982, Jacob et al, 1985, 1986) The mean ratios of SO 2- to H ÷ were greater than 1 0, and mean ratms of N O f to H ÷ were between 0 35 and 0 55 at all the high-elevation sites

Figure 11 shows the relation between sulfate con- centratlons and hydrogen 1on concentraUons m the cloud samples collected at M C C P sRes during the 1986-1988 field seasons It 1s seen that SO~- is highly correlated with hydrogen mn at all the hlgh-elevatmn

M C C P sites with a coefficient of determination, R 2 > 0 8 The sum of SO~- and N O 3 vs H ÷ are Illustrated m Fig 12, and m general, higher correlation coetficlents are observed at all those sites One sample collected at Mt Machell was noted for very high concentration of H ÷ ( ~ 5000 peq : - 1 w]th moderate S O ~ - + N O ~ ( ~ 3500 #eq f - 1), ln&catmg that other species, such as orgamc acid or HCI, may contribute to cloud aclday under certain biological and meteoro- logical condmons

In general, marme-ongln species, CI- and Na ÷, have to balance each other If no other sources are present (Saxena and Lm, 1990) The overall eqmvalent

Hlgh-elevat~on cloud chemistry 2013

60 40

Z 30 ' ~ 20

° . , 20

u. 10 u_

0

o o o o o o o o o ~ ~ o o o o o o o o o ~

Ai At

o O O o o O O O o O O o o O o O o o O o

NH4 (10"3 geq/I) NH4 (10"3 g~l/I)

. I ~.II IEII . ,z, (e)

(b) ~ 40

30

~ 30

~o ~ ~o

u.

0 . . . . . . . . . imml, , ~ t-J ~ ~ Lo ~ f-- co 0~ o o

o o o o o o o o o ~

o o o o o o o o o ~ ~ AI

o o o o o o o o o o o ~ oJ co ~ to co m- 00 o~

o o o o o o o o o o

N H 4 ( 1 0 " ' 3 g e q / I ) N H 4 ( 1 0 " 3 i~eq/ I )

60 40 / (f) 50 ~ 30

~ 4o

20 O 30 7 ~

O 20 1

~ 10 U.

0 ~ 04 e~ ~ m to r-- co o~ o o

o o o o o o o o o ~

o o o o o o o o o ~ ~ AI

o o o o o o o o o o

o o o o o o o o o o

N H 4 ( 1 0 " ' 3 p .eq/ I I

N H 4 (10"'3 geq/I)

F~g 10 Frequency distributions, May-Oc tobe r 1986-1988, of N H 2 at (at Whiteface Mountain, summit, (b) Whiteface Mountain, s~te 2 (c) Mt Moosllauke, (d) Shenandoah Park, (e) Wh~tetop Mountain, (f) Mt Mitchell

C1 /Na + ratio m our samples is 1 32-2 29, mdmatmg that a large portion of C1- m cloud were from sources other than sea salt and the roads m the northern part of the country, especially at Whltetop Mountain This result implies that C1- may contribute to the cloud acidity in the form of HCI from Industrial areas (Petrenchuk and Drozdova, 1966) However, the ratio C1-/(SO24 - + N O ~ +CI- ) , on an equivalent basis, averaged 002 at Whiteface Mountain, 003 at Mt Moosdauke and 0 04 at the three southern sites This ln&cates that, m general, CI can be neghglble for contributing acl&ty to clouds

There are two known processes, scavenging of particulate sulfate and m-cloud SO 2 oxidation, which are responsible for cloud SO~- concentration The coefficient of determination between SO~- and gas- eous SO 2 measured at Mt Mitchell during the cloud events was not good (r2=0 15), m&catmg that the particulate sulfate scavenging was responsible for a majority of the sulfate in the cloud This finding is consistent with the results of Relsmger and Imhoff (1989), which showed the cloudwater sulfate concen- traUon to be less than free-air aerosol concentration The gaseous SO2 concentrations were also very low, m

2014 Z LI and V P ANEJA

3000 , 4000

o t oo I o: v ~ 2000

- o 1-10 -ooo 0 0 , • , • 0 • , -

0 500 1000 1500 2000 0 1000 2000 3000 4000

H+ (l.teq/I) H+ (l.teq/I)

1500

o"1000 '

O 500'

0 • i • i • / • i - i i

0 200 400 600 800 1000 1200 1400

f

Y = - 22 244 + 1 2O83x R 2̂ = 0 907~ tc. =

• , ,4" °

H+ (lalq/l)

4000 ,

t y = 4 8 6 9 6 + 1 0 0 2 1 x R ^ 2 = 0 8 3 6 (d)

3ooo I

2ooo.I ** : > / • , 4 . ~ o o 1 o o o - I ~ -

, ,

0 soo 1000 IS00 2000 2600 3000

H+ (1.~i¢1/I)

5000 .j Y - 58819+ 1 0533x R".2.080, . ( (e)

4000 t ~ | o o

~O~3000 t o o o o

2000 ]

¢fJ lOOO -

o , • ! • ) • ! • | • i

0 1000 2000 3000 4000 5000 6000

H+ O.mq/I)

Fig 11 The relatzon between sulfate concentratzons and hydrogen zon concentratmns zn the cloud samples collected May-October 1986-1988 at (a) Whiteface Mountain, summ]t, (b) Whiteface Mountain, site 2, (c) Mt

Moosdauke. (d) Shenandoah Park, (e) WhRetop Mountain, if) Mt Mitchell

general, at Mt Mitchell It is recogmzed that aerosol particles tend to accumulate m the lower t roposphere and can be t ransported over large distances For th~s reason they have the potential of makmg a slgmficant con tnbu tmn to a o d deposmon at elevated sites that are remote from source regions Bradow and Aneja

(1988) found that cloudwater composmon ls very slmdar to aerosol composmon at Mt Mitchell These results provzde the evidence that m-cloud ox]datzon contr ibuted httle to the chemical composmon of cloud water, whde aerosol nucleation may have been the primary source of sulfate m cloud water, at least at the

Table 5 Ratios of mean Ion concentration m cloud samples at MCCP sites for 1986-1988

SO 2- SO~- SO 2- SO~- NO3 NO~- N O ; Cl- K + Ca 2+ Mg 2+ NH~"

S]te NO3 C1- H + NH2 H + NH2 CI- Na + Na + Na z+ Na + H +

WF1 269 377 1 21 205 045 076 1404 1 78 074 439 1 11 059 WF2 3 94 56 1 38 2 33 0 35 0 59 14 2 1 9 0 52 5 41 1 36 0 59 MS1 194 1892 102 25 053 129 978 132 0,2 077 041 041 SH1 204 185 1 13 178 055 087 905 167 03 147 048 064 WT1 236 209 1 16 235 049 099 884 229 036 348 076 05 MMI 278 1737 124 262 044 094 624 17 04 285 072 047

High-elevation cloud chemtstry 2015

4000

• 3000 -.,I

2000 o Z

+ 1000

o ffl ~ 0

'=21636+15537x R^2=0881

0 500 1000

H+ ~eq/I)

<(a)

÷ ÷

1500

A

Z

2000

4000 / ...... A.- | y = -4474g+1557gx R ^ 2 = 0 9 6 3

3000 +

ooo t

1000 t + ~ + ÷

0 I • i • i • i • i ' - - , i

0

(b)

500 1000 1500 2000 2500 3000

H+ 0teq/I)

A 2000

15oo

O 1000 Z

÷ ~' s00 o Or}

0

y = 12 6s4 . ~ + (c)

i • i I • i - i • i

200 400 600 800 1000 1200 1400

H÷ (1~1/I)

4000 ~'=~LO'3000'ty=68507+14325X R^2=09+~+

2000

.=.

1000 ~' + 0

0 0 500 1 000 1500 2000 2500 3000

H+ (l.m~l)

A =,- -,i

Z

O ¢.0

6000 I ÷ / / y = 81 405 + 1 4290x R^2 = 0 811 (e) 5000 ÷ ÷ 4000 + ,~. +

+ 3000 + + +

2000

1000

! 0 , ' o o ' ' ' 0 10 0 20 0 3000 4000 5000 6000

H+ (l.teq/I)

Fig 12 The relation between [SO 2-] + [NO ~ ] and [H +] m the cloud samples collected May-October 1986--1988 at (a) Whiteface Mountain, summit, (b) Whiteface Mountain, site 2, (c) Mt Moosllauke, (d)

Shenandoah Park, (e) Whltetop Mountain, (f) Mt Mitchell

Mt Mitchell h]gh-elevatmn site However we cannot exclude the poss]bdlty that the aerosol itself may have resulted from an earher m-cloud oxidation

4 S U M M A R Y A N D C O N C L U S I O N S

The purpose of this study was to characterize cloud chemistry exposures at h]gh elevatmns over the eastern U S H]gh-elevauon locations m the eastern U S were frequently immersed in clouds Cloud fre- quency ranged from 32 to 77% of the days at these sites More than 90% of cloud samples at those sites were acldm (pH < 5 0) One extreme value of pH 2 29 was recorded at Mt Mitchell The observaUons we report here document the presence of chemical pollu- tants that can exert a more subtle stress on these h~gh-

elevation ecosystems (Bormann, 1988), m parttcular, ozone and acid deposmon (predominantly as acid cloud water) (Hertel et al , 1990)

The ]omc composmon of the cloud water from each field season and for each site was dominated by sulfate and mtrate amons (48-50%), and hydrogen and am- monium catmns (43-47%) The southernmost site, Mt Mitchell, recewed, in general, the most acidic clouds (mean pH 3 57) and major chemical exposures (total averaged ion concentration 1586 #eqf -~) The Whiteface summit, the northernmost s~te, received the least acidic cloud (mean pH 4 01) and chemical expos- ures (total averaged ion concentrauon 691/~eqf-1) compared to the other MCCP sites

Cloud pH and major chemical components exhib- Ited a seasonal trend w]th the maxima during the summer months, and correlated with temperature and

2016 Z LI and V P ANEJA

ozone concentra t ions The seasonal var ia t ion may be due to (1) high ozone concent ra t ion dur ing the summer whmh provides the cond lhon of convers ion of SO 2 and NO2 to sulfate and mtra te m the gas phase and cloud droplets, (u) increased react ion rate dependent on temperature, (ui) o ther meteorological conditions, such as increased high-pressure systems and reduced hqmd-wate r content

The mean eqmvalent ratios of SO 2- to N O ~ were found to be between 1 9 and 3 9 at these sites It is observed that SO 2- was highly correlated with hydro- gen 1on for all events sampled, indicat ing tha t the cont r ibu t ion to cloud acidity by sulfate may be sigm- ficant

Acknowledoements--Thls research has been funded through a cooperative agreement with the U S Environmental Pro- tectlon Agency (813934-01-2) as part of the Mountain Cloud Chemistry Program We express sincere appreciation to Prof V Mohnen, Principal Investigator, MCCP, for providing the data, and Dr W Robarge and Dr E Cowling for review and comments on the manuscnpt We are indebted to our MCCP colleagues B Bailey, S M Goltz, J Healey, J Hornlng, J A Kadlecek, J Meagher and J T Sigmon and their many graduate students who collected the data used m the analysis at various MCCP sites as per the protocols Thanks to Mrs P Aneja and Ms B Batts in the preparation of the manu- script

Disclaimer--The contents of this document do not neces- sarily reflect the views and pohcies of the U S Environmental Protection Agency, nor the vmws of all members of the Mountain Cloud Chemistry Consortm, nor does mention of trade names or commercial or non-commercial products constitute endorsement or recommendation for use

REFERENCES

Aneja V P, Clalborn C S, Bradow R, Baumgardner R and Paur R (1990a) Dynamic chemical characterization of Montane clouds Atmosphertc Enwronment 24A, 563-572

Aneja V P, Cahhorn C S, LI Z and Murthy A (1990b) Exceedanees of the National Ambient Air Quality Stand- ard for ozone occurring at a 'pristine' area site J A~r Waste Man Ass 40, 217-220

Aneja V P, Busmger S, LI Z, Clalborn C and Murthy A (1991) Ozone climatology at high elevations in the south- ern Appalachmns J geophys Res 96, 1007-1021

Aneja V P, Robarge W P. Claiborn C S, Murthy A, Kim D S, LI Z and Cowhng E B (1992) Chemical climatology of high-elevation spruce-fir forests In Southern Appala- chian Mountains Enwr Pollut 75, 89-96

Bormann F H (1988) Air pollution and temperature forests creeping degradation 9 In Symp Bwnc Impoverishment (edited by Woodwell (3 M ), pp 1-24 Cambridge Univer- sity Press, Cambridge

Bradow R L and Aneja V P (1988) Aerosol compositional effects on mountain clouds attributing sources of acid deposition In Atmospherzc Aerosols and Nucleation (edited by Wagner P E and Vah G ), Vol 309, pp 40-43

Bruck R, Robarge W P and McDaniel A (1989) Forest dechne in the boreal montane ecosystems of the southern Appalachian mountains IVat Azr Soil Pollut 48, 161-180

Calvert J G , Lazrus A, Kok G L, Helkes B (3, Walega J G , Lind J and Cantrell C A (1985) Chemical mech- anisms of acid generation m the troposphere Nature 317, 27

Castlllo R (1979) An investigation of the acidity of stratus cloud water and its relationship to droplet distnbuuon, pH of rain, and weather parameters Ph D thesis, Department of Atmospheric Sciences, State Umverslty of New York, Albany

Charlson R J and Rodhe H (1982) Factors controlling the acidity of natural rainwater Nature 295, 683-685

Claiborn C S and Aneja V P (1991) Measurements of atmospheric hydrogen peroxide in the gas phase and in cloud water at Mt Mitchell, North Carolina J #eophys Res 96, 18711-18787

Collett J L Jr, Daube B C Jr and Hoffmann M R (1990) The chemical composition of intercepted cloudwater in the Sierra Nevada Atmospheric Environment 24A, 959-972

Cowling E B (1989) Recent changes in chemical climate and related effects on forests m North America and Europe Ambw 18, 167-171

Cowling E B, Shriner D S, Barnard J F Lucler A A, Johnson A H and Klester A R (1991) Airborne chem- icals and forest health m the United States In Proc lnt Umon Forest Research Orgamzat~ons, Montreal, Canada (m press)

Daum P H, Schwartz S E and Newman L (1984a) Acidic and related constituents in hqmd water stratlform clouds J geophys Res 89, 1447-1458

Daum P H, Kelly T J , Schwartz S E and Newman L (1984b) Measurement of the chemical composition of stratlform clouds Atraosphertc Enwronment 18, 2671- 2684

Dickey D (1991) Department of Statishcs, North Carolina State University, Raleigh, NC 27695, pets comm

Evans L S (1984) Botanical aspects of acidic precipitation Bot Rev 50, 449

Falconer R E and Falconer P D (1980) Determination of cloud water acidity at a mountain observatory in the Adirondack Mountains of New York State J geophvs Res 85, 7465-7470

Fuzzi S, Castillo R A, Jiusto J E and Lala G G (1984) Chemical composition of radiation fog water at Albany, New York. and its relationship to fog mlcrophysms J geophys Res 89, 7159-7164

Fuzzi S, Ors1G and Mariotti M (1983) Radiation fog liquid water aodlty at a field statzon m the Po Valley J Aerosol Scl 14, 135-138

Garner J H B, Pagano T and Cowling E B (1989) An evaluation of the role of ozone, aod deposition, and other mrborne pollutants in the forests of eastern North Amer- ica U S Forest Service. Southeastern Forest Experiment Station, Asheville. NC, General Techmcal Report SE-59, p 172

Grantt A L and Musselman R C (1984) Simulated acldm fog injures lettuce Atmospheric Enwronment 18, 887

Hegg D A and Hobbs P V (1981) Cloud water chemistry and the production of sulfates in clouds Atraospher,c Enwronment 15, 1597-1604

Hertel G D, Adams M B, Andrus S, Barnard J E, Brandt C J, Eagar C, Kiester A R, Martin B E, Medlarz S A, McFadden M W, Palmer C and Peterson C E (1990) The effects of acid deposition and ozone on forest tree species results of the Forest Response Program 83rd Annual Meeting and Exhibmon, Air and Waste Manage- ment Association, Pittsburgh, PA, 24-29 June 1990, paper no 90-187 1

Houghton H G (1955) On the chemical composlhon of fog and cloud water J Met 12, 355-357

Isaac G A and Daum P H (1987) A winter study of air, cloud and precipitation chemistry in Ontario, Canada Atmospheric Environment 21, 1587-1600

Jacob D J, Waldman J M, Munger J W and Hoffmann M R (1985) Chemical composition of fogwater collected along the California coast Enwr Scs Teehnol 19, 730-736

Jacob D J , Munger J W, Waldman J M and Hoffmann M R (1986) The H2SO4-HNOs-NH 3 system at high

High-elevation cloud chemistry 2017

humidities and m fogs 1 Spatial and temporal patterns m the San Joaqum Valley of Cahforma J geophys Res 91, 1073-1088

Jacobson J S (1984) Effects of acidic aerosol fog, mist and rain on crops and trees Phil Trans R Soc Lond 305, 327

Jacobson J S, Bethard T, Heller L I and Lassole J P (1990a) Response of Pwea rubens seedling to lntenmttent mist varying m acidity, and m concentrations of sulfur-, and nitrogen-containing pollutants Ph)t Plant 78, 595-601

Jacobson J S, Heller L I, Yamado K E, Osmeloskl J F, Bethard T and Lassole J P (1990b) Fohar injury and growth response of red spruce to sulfate and mtrate acidic mist Can J For Res 20, 58-65

Khemanl L T, Momm G A, Nalk M S, Prakasa Rao P S, Safal P D and Murthy A S R (1987)Influence ofalkahne particulates on pH of cloud and rain water m India Atmospherw Enmronment 21, 1137-1145

Klm D S and Aneja V P (1992) Chemical composition of clouds at Mt Mitchell, North Carohna, U S A Tellus 44B, 41-53

Kms L, Mueller K O and Melxner F X (1988) Experiment Alsdan (1985) some prehmmary results of height resolved measurements of trace gases, aerosol composltmn, cloud- and precipitation water In Acid Deposttwn at High Elev- ations, NATO ASI Ser, Vol 252 (edited by Unsworth M H and Fowler D), pp 375-394 Kulwer, Boston, MA

Klein R M and Perkins T D (1988) Primary and secondary causes and consequences of contemporary forest decline Bot Rev 54, 1-43

Lacaux J P, Servant J and Baudet J G (1987) Acid rain in the tropical forests of the Ivory Coast Atmospherw Enwr- onment 21, 2643-2647

Lazrus A L, Baynton H W and Lodge J P Jr (1970) Trace constituents m oceamc cloud water and their ongm Tellus 22, 106-113

Mohnen V A, and Kadlecek J A (1989) Cloud chemistry research at Whiteface Mountain Tellus 41B, 79-81

Mohnen V A, Aneja V P, Bailey B, Cowling E V, Glotz S M, Healey J, Hormg J, Kadlecek J A, Meagher J, Mueller S F and Slgmon J T (1990a) An assessment of atmospheric exposure and deposition to high elevatmn forests m the eastern United States EPA/600/3-90/058, U S Environmental Protection Agency, Research Trmngle Park, NC

Mohnen V A (1990b) A data users grade to the Mountain Cloud Chemistry ProJect U S Enwronmental Protectmn Agency, Research Triangle Park, NC

Mrose H (1966) Measurements ofpH and chemical analyses of ram-, snow-, and fog-water Tellus 8, 266-270

MulrP S WadeK A,Car te rB H,ArmentanoT V and

Pnbush R A (1986) Fog chemistry at an urban mldwestern site J Air Pollut Control Ass 36, 1359-1361

Munger J W, Jacob D T, Waldman J M and Hoffmann M R (1983) Fogwater chemistry m an urban atmosphere J aeophys Res 88, 5109-5121

Noller B N, Currey N A, Ayers G P and Glllet R (1986j Naturally acidic rainwater at a site m northern Australia In 7th lnt Clean Aw Conf , Sydney, August. pp 189-190

Od&e B C V (1962) The chemical composition of pre- cipitation at cloud levels J Met Soc 80, 535-538

Oklta T (1968) Concentration of sulfate and other Inorganic materials m fog and cloudwater and m aerosol J Met So~ Japan 46, 120-126

Petrenchuk O P and Drozdova V M (1966) On the chemical composition of cloud water Tellus 18, 280-286

Pruppacher M R and Klett H R (1980) Mtcrophysics of Clouds and Precipltatwn, p 18 D Reldel, Dordrecht

Relsmger L M and Imhoff R E (1989) Analysis of sum- mertlme cloudwater measurements made m a southern Appalachian spruce forest Wat Air Sozl Pollut 45, 15

Robinson E (1984) Natural emissions source The acl&c deposition phenomenon and tts effects Critical Assess- ment Review Papers, Vol 1 In Atmospheric Scwnce. Chap 2

Romer F G, Vdjeer J W, Van den Beld L, Slangowal H J and Veldkamp A A (1985) The chemical composition of cloud and rainwater Results ofprehmlnary measurements from an aircraft Atmospheric Environment 19, 1847-1858

Saxena V K, Parungo F P and Rugglero F H (1985) Airborne measurements of the Antarctic cloud water acidity Ant J U S 19, 201-203

Saxena V K and Lm N -H (1990) Cloud chemistry measure- ments and estimates of acidic deposition on and above cloudbase coniferous forest Atmosphertc Enwronment 24A, 329-352

Scott B C and Laulamen N S (1979) Concentration of sulfate m precipltatmn J appl Met 18, 138-147

Scott W D (1978) The pH of cloud water and the production of sulfate Atmospherw Environment 12, 917-921

Valente R J. Mallant R K A, Mclaren S E and Schemen- auer R S (1989) Field mtercompanson of ground-based clouds physics instruments at Whltetop Mountain. Vlr- gima J Atmos Oceamc Technol 6, 396-406

Waldman J M, Munger J W, Jacob D J. Flagan R C. Morgan J J and Hoffmann M R (1982) Chemical com- position of acid fog Sctence 218, 677-679

Weathers K C, Likens G E, Bormann F H. Eaton J S. Bowden W B, Anderson J L, Cass D A Galloway J N, Keene W C, Kimball K D, Huth P and Smfley D (1986) A regional acl&c cloud/fog water event m the eastern United States Nature 319, 657-658