On the measurements of cloud condensation nuclei at palmer station, Antarctica

Pergamon

Atmospheric Environment Vol. 31, No. 23, pp. 4039-4044, 1997 PII:

S1352-2310(97)00250-1

,C 1997 Elsevier Science Ltd All rights reserved. Printed in Great Britain 1352-2310,/97 $17.00 + 0.00

SHORT COMMUNICATION O N THE M E A S U R E M E N T S O F C L O U D C O N D E N S A T I O N N U C L E I AT P A L M E R STATION, A N T A R C T I C A T. P. DEFELICE,* V. K. SAXENA* and SHAOCAI YU* Department of Geosciences, Atmospheric Sciences Group, Univ. of Wisconsin at Milwaukee, 3209 N. Maryland Ave., Milwaukee, WI 53211, U.S.A.; and * Department of Marine, Earth and Atmospheric Science, North Carolina State University, P. O. Box 8208, Raleigh, NC 27695-8208, U.S.A. (First received 16 February 1997 and in final form 19 M a y 1997. Published September 1997)

K e y word index: Cloud condensation nuclei (CCN), Antarctic region, meteorology and CCN.

INTRODUCTION Cloud condensation nuclei (CCN) spectral measurements are among the recent priorities for the study of aerosol/climate interactions (e.g. W M O / W M P No. 19, August 1992; Hobbs and Huebert, 1996). For example, the potential link between CCN formation, the oxidation of dimethylsulfide (DMS), and particle nucleation in the marine atmosphere represents an important component in the DMS-cloud-climate hypothesis (Charlson et al., 1987; Shaw, 1983). However, there are large uncertainties existing with respect to such a link due to poor understanding of the intricate mechanisms that control the relationship among the DMS sea-toair flux, atmospheric DMS chemistry and CCN formation in the marine atmosphere (Bates et al., 1989). Palmer Station, Antarctica offers a unique and unusual opportunity to investigate the CCN-DMS relationship due to the known and limited sources of aerosol in the vicinity of Palmer Station, Antarctica (Robinson et al., 1984; Hogan et al., 1990). However, CCN spectral measurements are especially rare in the Antarctic region. Consequently, we present and discuss the predominant characteristics associated with a first dataset of daily daylight period (i.e. ~ 13-15 h long) averaged CCN spectral measurements at a remote region of the globe, namely Palmer Station, Antarctica. Palmer Station is located on the Antarctic Peninsula and sits on a glacial moraine between the Piedmont glacier and the Bellings-Hausen Sea in the midst of natural Antarctic sources (Hogan, 1975; Shaw, 1979), which could include sea-toair emission of marine organic nuclei. The predominant local pollution source is a diesel power plant approximately 100 m WNW of the sampling platform. Other anatural pollution sources during January and February are minor and include; a total of 6 tourist ships, the Polar Duke, occasional outdoor paint work and the use of on/offload vehicles. Daily daylight period averages are chosen since (i) the day is dominated by daylight during our sampling period, (ii) the C C N measurement frequency is a maximum during the daylight period, (iii) DeFelice (1996) has related temporal CCN concentra-

* Author to whom correspondence should be addressed.

tions to smaller-scale phenomena (i.e. mesoscale and shorter) at the same remote region, and (iv) they are sufficient for most climate modelling efforts. The Palmer data represent a natural Southern Hemisphere remote marine arctic climate, especially after all significant local pollution sources are filtered out.

BACKGROUND First-time measurements of CCN spectra were made at Palmer Station, Antarctica (64°46'S, 64°05'W) during January 1994 (Saxena, 1996) and February 1994 (DeFelice, 1996), using the CCN-spectrometer (e.g. DeFelice and Saxena, 1994) during the daily observing periods ( ~ 07302200 h local time) of 11 January-26 February 1994. The CCN spectra may be described by N = Cs k

(1)

where N is the number of CCN activated at a given supersaturation, s. C and k are constants of the distribution. A small k parameter (i.e. k < 1.0) indicates that the observed cloud droplet spectra is determined by the CCN population. A value of k > I indicates that the observed cloud droplet spectra is determined by the inherent cloud dynamics. A large k value may also mean that sampling takes place in the vicinity of a non-steady-state phenomenon (i.e. in the vicinity of evaporating rain droplets, or a frontal boundary), or there is a source of small CCN. Small CCN require high supersaturations, consequently, an increase in small CCN pushes the high supersaturation end of the distribution upward relative to the lower end, resulting in higher k values. Spectra may have multiple k values if sampling within polluted air masses, inhomogeneously mixed air masses, or non-steady-state phenomena, and in this case it means that the k values were varying in time. More than one value of k parameter for the same air sample might also suggest that the measurements do not conform to equation (1) form although the measurements may be fitted to this form in small supersaturation intervals. It may be recalled that Twomey (1977) used equation (1) for mathematical convenience, and it may not be a valid representation of the CCN activity spectrum during periods of non-steady-state phenomena.

4039

4040

Short communication

The instantaneous temporal CCN concentrations at 0.3 and 1.0% supersaturation ranged between 79 and 158 cm-3 and between 110 and ~ 2300 cm-3, respectively, during the period of 11 January-7 February. The instantaneous CCN concentrations at 0.3 and ~ 1.1% supersaturation ranged between 0 and 200 cm- 3 and between < 1 and 692 cm- 3, respectively, during the period of 13-26 February. The lower CCN concentrations measured during the 13-26 February period occurred under foggy, precipitating and non-foggy periods following widespread precipitation. CCN concentrations may be greater than 200 cm- 3 at Palmer, if sampling takes place in the vicinity of virga, high windspeeds, dissipating clouds under certain conditions, local pollution and dimethylsulfide phytoplankton bloom episodes. The instantaneous k values were generally below 1 during the period of 11 January-7 February, and they were generally between 0 and ~ 3 during 13-26 February. Many of the spectra during the latter period had multiple k values (DeFelice, 1996). The windspeeds were most frequently between ~ 3 and ~ 23 m s- 1 during the daily period represented in this study. Saxena (1996) observed CCN concentrations beyond those which could be traditionally accounted for when the cloudbase descended to the surface and dissipated on 17, 19, 20 January and 7 February 1994 under typical meteorological conditions for his sampling period. The aforementioned non-traditional amounts of CCN were termed CCN bursts by Saxena (1996). However, DeFelice (1996) provides an example of a cloudbase passage on 16 February without an equivalent increase in CCN, and discusses some of the possible reasons behind these observations. Drizzle, rain and/or snow fell during ~ 50% of the days on the average between 11 January and 4 February 1994, and ~ 85% of the days on the average during 13-25 February 1994. The occurrence of the precipitation was most frequently between 2200 and 0700h local time during 13-25 February. The 21 January daily period had light precipitation falling during its last -~4h. Samples associated with winds ~< 3 m s ~, and obvious contamination from local anthropogenic pollution sources and the aforementioned CCN bursts (Saxena, 1996) are not included below. Samples associated with wind speeds ~< 3 m s-1 were excluded because of the increased likelihood of including local unnatural pollution. The result of the latter is a dataset (Table 1) that is representative of the natural environment of Palmer Station, Antarctica between 21 January and 4 February, arbitrarily termed the January period, and between 13 and 25 February, arbitrarily termed the February period. Saxena (1996) and DeFelice (1996) provide some additional details of their CCN spectral measurements.

RESULTS The 21 January-4 February 1994 period, or the January period, is characterized by; (i) relatively constant temporal CCN concentrations (Table 1), (ii) relatively constant equivalent potential temperatures, Theta E (Fig. la), (iii) partly cloudy skies with respect to wind direction during the daytime (Fig. lb), and (iv) winds predominantly from the eastsoutheast through the southwest (i.e. ~ 100 and ~ 240 ° true). The equivalent potential temperature is a good tracer of air masses since it is conserved during both dry and moist adiabatic processes. The 13-25 February 1994 period, or the February period, is characterized by: (i) relatively variable temporal CCN concentrations (Table 1); (ii) relatively variable Theta E (Fig. la); (iii) variable cloudy to overcast skies with respect to wind direction during the daytime (Fig. lb) and (iv) winds predominantly from the northeast, southwest and the northwest (i.e ~ 30 and 60 °, -~ 210 and 240 °, and ~ 315 and 345 ° true) with none between southeast and southwest.

Table 1. Daily average CCN spectral parameters during 21 January 26 February 1994 at Palmer Station, Antarctica Date (1994)

Julian day

21 January 22 January 23 January 24 January 25 January 26 January 27 January 28 January 29 January 30 January 31 January 1 February 2 February 3 February 4 February 5 February 6 February 7 February Average-January period

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

12 February 13 February 14 February 15 February 16 February 17 February 18 February 19 February 20 February 21 February 22 February 23 February 24February 25 February 26 February Average-February period

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

CCN (1%) (cm-3) @ 1% @ 0.3% s supersaturation, s 137 154 149 155 161 135 138 148 147 146 154 168 162 150 142 161 142 160

105 125 141 111 97 119 105 117 122 146 147 153 124 120 106 146 106 160

0.22 0.17 0.18 0.28 0.42 0.1 0.23 0.2 0.15 0 0.04 0.08 0.22 0.19 0.25 0.22 0.24 0

151

125

0.2

63 33 33 8 4 I0 7 35 6 6 4 5 13 27 61

35 5 9 2 0.3 I 0.5 9 1 3 3 4 11 17 52

0.49 1.5 1.1 1.2 2.1 1.9 2.2 1.1 1.3 0.64 0.22 0.18 0.16 0.38 0.14

21

10

1.0

There are no generally significant correlations between daily averaged CCN and air temperature, cloud amount or windspeed, except for a possible exception between dailyaveraged CCN and windspeed values obtained during the February period (Fig. 2).

DISCUSSION AND SUMMARY The data suggest that this site was generally under the influence of an air mass from the Antarctic Plateau during the January period based on the relatively constant wind directions, equivalent potential temperatures (Theta E; Fig. la), CCN (Fig. la) and the CCN spectra (Table l). According to Hogan et aL (1990) relatively cold air temperatures and nearly constant aerosol concentrations in the boundary layer are characteristic of air from the Antarctic Plateau. The mean surface air temperature was slightly cooler during January than during the February period (Fig. lb). In contrast, the site was under the influence of a transitionary weather period during the February period

Short communication

4041

January Period, 1994 250

i

"

I

l

"

i

•

•

i

i

I

l

280

l

240

.....

m

230 Z

279 •-4---

.-. 220 o~

I Them E

\

~. 21o

278

QQ

200

190 ~,

180 @

I

277

Upper limit of maximum

170 276 ~

~ 160 L.°°°.o..~,,

150

t..

roo..,~ 'o

!

275

140

<

130 ' 60

120 30

!

90

120

150

180

210

' 270

240

' 300

274 330

Daily Average Wind Direction (Deg.) February Period, 1994

288

140, q

120'

Z

....r"O ....... CCN [ . " ...... T h e t a E I \

!

~_

__ Upper limit of maximum

.,,~

I

I 286

'q[~ctrorfor Them E

@

o J

' 284""

100 '

-282~

80

' @ o~

E

- 28o

60 40

.278...~

20

.276

",~

1.

<

0 30 (a)

274

,

60

90

120

150

180

210

240

270

300

330

Daily Average Wind Direction (Deg.) Fig 1. (a) The daily average CCN (active at 1% supersaturation) concentrations and the equivalent potential temperature (Theta-E), as well as (b) cloud cover (tenths) and air temperature vs wind direction (WD) plotted as daily averages during 21 January-25 February 1994. The daily averages represent the time between ~ 07 : 30 (___00 : 30) and ~ 22 : 00 ( + 00 : 30) LDT. The mean WD is ~ 210 and ~ 175 ° during the January and February periods, respectively.

since the aerosol concentrations were highly variable and associated with different air-mass origins as indicated by k (Table 1) and Theta E (Fig. la). Highly variable CCN concentrations would be measured as the result of sampling both sides of the meandering boundary, between warmer, aerosol-enhanced maritime polar air and colder, constant aerosol-laden air from the Antarctic Plateau, and are typi-

cally found during February in this region (Hogan et al., 1990). Note to ignore the other possible sources of C C N variability as suggested in the background section. For instance, it has been shown that ~ 8 0 - 9 0 % of the aerosol mass in the Antarctic troposphere deposited to the ice sheet is composed of fine particle ( < 1 #m diameter) non-sea-salt (NSS) sulfate (Shaw, 1988), and that

4042

Short communication

January Period, 1994 ~- 10 l

I i'~l AirTemperatureI [] CloudCover

L

MeanCloudCover

g

~

,s•~,

:.;.;.;.

h= o~

MeanAir Temlmature

,

, , , ,, ,

97

•

... s

~.............. . < ,',~,,~,

120 153 180 220 Daily Average Wind Direction (Deg. True)

230

February Period, 1994 MeanCloudCover [] AirTemperature [] CloudCover

0

f

~9

•••~

,,•,,,, ¢•,, •

"•FS

..;.;.;

MeanAir Temperature

%~%

•ss~

,'•','•

.,;,,,;

,•g

,,,,,.,

6

s s ~

s s J i s ¢ ¢ i s ¢ s ~

""'",,, [.,,

~i:

.~_ •~

~ii~: , ' , ' , ' ,

2

.

:,;,;,. ......... .'.;,;,: ............. .7..7.~

,•,,s ~s "•~,d

..;-;.;:

',

,., .~.;.~.:

,,, ...-...

t l t l

x / s ~ ~,

~ ~:.!.~ ,,,~,',

~17Iii~!~ " " "

30

s s ~ •

:::~ !iiii!~i#i , • ~ • % ~

ii iiii,

. . . .

--iiiii!i

-i!~:,i~ i','-.',." 63 88 228. 321 Daily Average Wind Direction (Deg. True)

s t t ~

330

(b)

Fig. 1. (continued).

atmospheric conversion processes involving the marine biogenic sulfur gas DMS are a major source of Antarctic NSS sulfate aerosols (Delmas, 1982). The CCN at Palmer Station generally have significant amounts of NSS sulfate. However, the recent study of Kawamura et al. (1996) indicates that there are water soluble dicarboxylic acids in Antarctic aerosols. They found that oxalic (C2) or succinic (C4) was the dominant diacid species, followed by azelaic (C9) and malomic (C3) acids. The concentrations of oxacilic, succinic and pyruvic acids in the summer Antarctic aerosols were 10.29, 61.53 and 0.78 ng m - 3, respectively (Kawamura et al., 1996). They concluded that organic aerosols over the coast of Antarctica result from the sea-to-air emission of marine

organics and subsequent photochemical transformation. Since the organic acids in aerosol particles may also be one of the primary sources of CCN in the atmosphere, the organic acids in Antarctic aerosols can also make a contribution to the formation of CCN in the Antarctic atmosphere. In summary, a first time daily daylight averaged CCN spectral database covering January and February 1994 at Palmer Station, Antarctica is presented in response to the recent research need for the area of aerosol/climate research. It is suggested that significantly different temporal characteristics in the daily daylight period averages of CCN concentrations could be attributed to differences in the naturally

Short communication

.[

180

?

e

~D

i

I

160 o

140

i

i

in

i

[]

I

I

I

-[]

i

I

I

D

[]

[]

i

120

o

4043

I00

Januaa'yPeriod ] FebruaryPeriod

n

•

80 60 40 •<

20 0

"~

,

0

Ill

,

4

.m

!

6

I

,

8

,

10

I

,

12

l

,

I

14

,

,

'

18

20

22

,

16

24

Z Daily A v e r a g e W i n d Speed (Knots)

e~

,.-.,

@ ",,.,

180

!

I

!

l

!

I

!

!

160 a

i

~

140

~

120

[]

g

~ 100 s0

•

~

60

~

40

January Period

I

a February period

I

20

~

~

0

I

I

I

I

I

I

~1

I

I

1

2

3

4

5

6

7

8

9

Daily A v e r a g e

.o

,-,

,t~

180

u

er~

¢~ ' ~q

o

Cloud

160

o o

140

o

Amounts

o

(10=overcast,

u O

O

10

0=clear)

!

i

!

O

oo

I o JanuaryPeriod o

120

•

FebruaryPeriod

loo

o

~

so

•

"~ =

--~

Z

,e 0

,

0

I

I

1

2

i •

I

3

•

i

I

4

i

I

i

5

I

i

6

I

7

Daily A v e r a g e Air T e m p e r a t u r e (°C) Fig. 2. The relationships between the concurrent daily average CCN (active at 1% supersaturation) concentration and air temperature, cloud amount and wind speed during the January and February periods.

inherent meteorological conditions, once the data have been quality assured.

Acknowledgement--This work was supported by the UWM graduate school, and the Division of Polar Programs, National Science Foundation under OPP-9218538.

REFERENCES Bates, T. S., Clarke, A. D., Kaustin, V. N., Johnson, J. E. and Charlson, R. J. (1989) Oceanic dimethylsulfide and marine aerosols: difficulties associated with assessing their covariance. Global Biogeochemical Cycles 3, 299-304.

4044

Short communication

Charlson, R. J., Lovelock, J. E., Andrea, M. O. and Warren, S. G. (1987) Oceanic phytoplankton, atmospheric sulfur, cloud albedo and climate. Nature 326, 655-661. DeFelice, T. P. (1996) Variations in cloud condensation nuclei at Palmer Station Antarctica during February 1994. Atmospheric Research 41, 229-248. DeFelice, T. P. and Saxena, V. K. (1994) On the variation of cloud condensation nuclei in association with cloud systems at a mountain-top location. Atmospheric Research 31, 13-39. Delmas, R. J. (1982) Antarctic sulfate budget, Nature 299, 677-678. Hobbs, P. V. and Huebert, B. J., eds (1996) Atmospheric Aerosols. A new focus of the International Global Atmospheric Chemistry (IGAC) Project, 40 pp, + 6 pp appendix. [Available from the International Global Atmospheric Chemistry (IGAC) project office, Building 24-409, MIT, Cambridge, Massachusetts, 02139-4307, U.S.A.] Hogan, A. W. (1975)Antarctic aerosols. Journal of Applied Meterology 14, 550-559. Hogan, A. W., Egan, W. G., Samson, J. A., Barnard, S. C., Riley, D. M. and Murphy, B. B. (1990) Seasonal variation of some constituents of Antarctic tropospheric air. Geophysical Research Letters 17, 2365-2368. Kawamura, K., Semere, R., Imai, Y., Fujii, Y. and Hayashi, M. (1996) Water soluble dicarboxylic acids and related

compounds in Antarctic aerosols. Journal of Geophysical Research 101, 18,721-18,728. Robinson, E., Bamesberger, W. L., Menzia, F. A., Waylett, A. S. and Waylett, S. F. (1984) Atmospheric trace gases measurements at Palmer Station, Antarctica, Journal of Atmospheric Chemistry 2, 65-81. Saxena, V. K. (1996) Bursts of cloud condensation nuclei (CCN) by dissipating clouds at Palmer Station, Antarctica. Geophysical Research Letters 23, 69-72. Shaw, G. E. (1979) Considerations on the origin and properties of the Antarctic aerosols. Review of Geophysics 17, 1983-1988. Shaw, G. E. (1983) Antarctic aerosols: A review. Review qf Geophysics 26, 89-112. Twomey, S. (1977) Atmospheric Aerosols, 302 pp. Elsevier, New York. Twomey, S. and Wojciechowski, T. A. (1969) Observations of the geographical variation of cloud nuclei. Journal qf Atmospheric Science 26, 684-688. World Meteorological Organization, WMO, WMP (1992) Proceedings of the WMO workshop on cloud microphysics and applications to global change, 10-14 August 1992, 406 pp. Toronto, Canada. (World Meteorological Organization Programme on physics and chemistry of clouds and weather modification research. WMP report # 19. WMO, 41 Giuseppe-Motta, Case postale No. 2300, CH-1211 Gen6ve 2.)