JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 106,NO. D12, PAGES 12,579-12,588,JUNE 27, 2001
Vertical profiles of cloud condensationnuclei above Wyoming
DavidJ.Delene
• andTerryDeshler
Departmentof AtmosphericScience,Universityof Wyoming,Laramie,Wyoming
Abstract. High-resolution(125 m) profilesof cloudcondensation
nucleus(CCN) number
concentrationat 1.0% supersaturation
were measured12 timesat Laramie, Wyoming (41øN),
and twice at Lauder,New Zealand(45øS),extendingfrom the surfaceto 200 mbar. The
balloon-borneinstrumentsincludeda condensationnucleus(CN) counterto measure
particleswith diametergreaterthan-10 nm and an opticalparticlecounterto measure
particleswith diameter>0.3 •[m (D0.3). Within verticalprofiles,variationsin CCN
concentrationwere typicallypositivelycorrelatedwith changesin CN and D0.3concentrations
andoftencorresponded
with changesin relativehumidity,typicallypositively,but
occasionallynegativelycorrelated.The aerosolprofilesgenerallyshowseveraldistinct
layersthatcan be definedby equivalentpotentialtemperatureandrelativehumidity. These
layersareusedto summarizethe 14 verticalprofilesby classifyingaerosolmeasurements
into
five distinctatmosphericlayers:surface,lower tropospheric,
uppertropospheric,
stratospheric
andregionsof high humidity. Laramiesummerandwinter profilesshowthat
the meanCCN concentration
decreases
betweenthe lower anduppertroposphericlayers(450
to130crn
-3insummer,
150to65cm-3in winter).Theaverage
summer
CCN/CNratioin
Wyoming showsan increasefrom 0.09 in the lower troposphereto 0.17 in the upper
troposphere.The summerCCN concentrations
at Lauder,New Zealand,were abouttwice
the summerCCN concentrations
measuredat Laramie,Wyoming,while the CN andD0.3
concentrations
were approximatelythe same.
1. Introduction
determinechangesin aerosolnumberconcentration
due to
anthropogenic
sulfateand three differentparameterizations
The number concentration and activity of cloud
[Twomey,1977; Joneset al., 1994; Ghan et al., 1993] to
condensationnuclei (CCN) influence cloud droplet spectra relatechangesin aerosolnumberconcentration
to changesin
and henceprecipitationprocessesand cloud albedo [Hobbs, clouddropletconcentration.
Theyconcluded
thatrefiningthe
1993; Jennings, 1993]. Increases in CCN concentration, input parameters
might be more importantthan improving
resultingfrom increasedSO2 emissions,have been suggested
as possibly offsetting global temperaturechange expected
from increasedCO2 concentrations[Wigley, 1989; Twomey,
1991]. Aerosolscan affect climatedirectlyby scatteringand
absorbingradiation and indirectly by altering the scattering
characteristicsof clouds [Charlson et al., 1992]. The direct
and indirectclimate forcing of aerosolsis being consideredin
global climate models [Boucher and Anderson, 1995;
Haywood and Shine, 1995]; however, our limited
understandingin a numberof areashas so far precludeda full
inclusion of either affect into these models [Penner et al.,
1994].
Chuang et al. [1997] addressed indirect
anthropogenic forcing with a coupled climate/chemistry
model to predictchangesin the aerosolnumberspectrumdue
to anthropogenicsulfate. Using a microphysicsmodel to
improve the parameterizationof Ghan et al. [1993], changes
in cloud droplet concentrationare related to the aerosol
spectrum. Pan et al. [1998] investigatedindirect forcing of
aerosolsusing the method of Charlson et al. [1992] to
modelsto minimize uncertainties.Vertical profilesof aerosol
size distribution
and CCN
fraction
over continents would
provide someof the informationrequired. Here we present
high-resolution
verticalprofilesof CN, CCN, andparticles>
0.3 gm over Wyoming, a remote midcontinentalNorth
American
site.
Several studies have measured continental CCN
number
concentrationnear the Earth's surface [Hudson and Squires,
1978; Hudson and Frisbie, 1991; Philippin and Betterton,
1997], and field campaignshave used aircraft to measure
continental CCN concentrations aloft [Hobbs et al., 1985;
Hudson and Xie, 1998]. Figure 1 presentsexamplesof
verticalCCN profilesobtainedwith aircraftusinga varietyof
instruments.The profilesconsistof four to six measurements
rangingfrom the surfaceto a maximumaltitudeof-5 km.
The profiles show a generaldeceasein CCN concentration
with increasing
heightabovethe surfacebut havelow vertical
resolution. The CCN measurements
do not showa systematic
decreasein concentrationwith decreasingsupersaturation.
Addingto the difficultyin interpreting
CCN variationsarethe
•Nowat theCooperative
Institute
forResearch
in Environmentalinherent difficulties in making good CCN measurements
Sciences,
Universityof Colorado/NOAA,Boulder,Colorado.
Copyright2001 by theAmericanGeophysical
Union.
Papernumber2000JD900800.
0148-0227/01/2000JD900800509.00
[Jiusto et al., 1981]. Aircraft measurementsof CCN have
been confined almost exclusivelyto the lower troposphere.
Recentuppertropospheric
CCN measurements
conductedby
Hudson and Xie [1998] found constantCCN concentrations
with increasingaltitudeabovethe lowertroposphere.
12,579
12,580
DELENE AND DESHLER: CLOUD CONDENSATION NUCLEI ABOVE WYOMING
Continental
Profiles
•'•"-Miles
City, Montana(1.0%)
- ß-
Miles City, Montana(1.0%)
"'
SouthwestArizona (0.35%)
A heater on the common inlet of the aerosol instruments
raisesthe sampleair temperatureto 40øC duringballoon
ascent. After passingthe heated inlet, the air sample
temperaturedecreasesas it flows through 0.8 m of
uninsulated stainless steel tubing to each instrument;
however,the air enteringeach instrumentremainsabovethe
- [] ' Yukon Valley, Alaska(0.7%)
ambienttemperature.Thus the humidityis estimatedto be
<30% at the instrumentinlets. Heating the airstreamcould
reduce the size of volatile aerosolsby forcing water in the
ambientaerosolto evaporate.Changingthe ambientaerosol
Southern
Florida
(0.7%)
SouthernGermany (0.5%)
size will affect the measurements to the extent that it reduces
the aerosolparticlesbelow a thresholdsize. This will
predominantlyaffect the D0.3 measurements.If particle
composition
wereknown,it wouldbe possibleto theoretically
correctthe D0.3 measurements
to replacethe water;however,
since compositioninformationis not available for these
measurements,
no correctionsare applied. In both the CN
and CCN
\
measurements the aerosols are reintroduced to a
supersaturated
environment
priorto measurement.
Thusthese
measurements
will only be affected if heating causesthe
aerosol to shrink below a critical activation size. This was not
I
200
400
600
800
•
I
I
I
•
I
1000 1200 1400
CCN Concentration
[cm-3]
the case for CN.
Several flights with two instrumentsto
comparesamplingfrom a heatedandunheatedinlet indicate
thatheatingthe air samplestreamhasminimalimpacton CN
concentrations.
Calculations of diffusional loss of aerosol in
the heater and samplinginlet indicate<10% particle loss
betweenthe surfaceand 200 mbarfor particles>30 nm. Over
Figure 1. Continental cloud condensationnucleus (CCN)
profiles obtained from aircraft measurements. The the same pressurerange, diffusional lossesfor 20-nm
measurementlocation and percent supersaturation
are given particlesincrease
from 10 to 15%, whilefor 10-nmparticles,
for eachprofile. The Miles City, Montana,profilesare from diffusionallossesare abouttwice as high. Sincethe true size
Hobbs et al. [1985]; the Arizona, Alaska, and southern of the CN particlesare not known,no corrections
are made
Florida profiles are from Hoppel et al. [1973]; and the
southern Germany profile is from Schafer and Georgii
[1994].
loss.
The balloon-borne
CCN
counter uses a static thermal-
gradient diffusion chamber to create a supersaturated
To improvethe verticalresolutionand altituderangeof
existingCCN profiles,a balloon-borne
CCN counterhasbeen
usedto obtain14 CCN profilesextendingfrom the surfaceto
200 mbar.
for diffusional
The CCN counter, operated at a constant
environmentwhere CCN activate and grow. The CCN
concentrationis deduced from the amount of laser light
scattered by droplets within the diffusion chamber.
Measurement uncertainties for
CCN
concentrations (at
ambient
pressure)
of50to500cm-3are36%to11%owing
to
The minimumdetectableconcentration
(at
supersaturation
of 1%, wasincludedon 12 flightsat Laramie, countingstatistics.
ambient
pressure)
is
-20
cm
-3.
Delene
et
al.
[1998]
provided
Wyoming (41øN), and 2 flights at Lauder, New Zealand
(45øS). The 12 flights above Wyoming representa short a descriptionof the balloon-borneCCN counter,described
on NaC1 aerosols,and
climatologyof CCN concentrationat a midlatitudeNorthern calibrationat 1% supersaturation
presented
some
preliminary
CCN
profiles.
A morecomplete
Hemispherelocation,while the 2 flightsaboveNew Zealand
description
of
the
instrument
and
its
calibration
is givenby
illustratethe variabilityof CCN concentrations
with location.
2. Instrumentation
Delene and Deshler [2000]. For the measurements
presented
here,the CCN counterwas operatedwith a platetemperature
differencethat would createa 1% supersaturation.
This plate
Typically,
balloon
flightsstartatdawnwithclearskiesand
light winds becauseof operationalconstraints. The
instrument
package
consists
of threeaerosol
counters:
a CCN
counteroperatedat a constantløk supersaturation,
a
temperaturedifferencewas carefullychecked;however,the
supersaturation
distributionin the chamberhas not been
measured.Recentmeasurements
of the particlesizefor 50%
activation at 1% supersaturation
suggestthat the actual
condensation
nucleus(CN) counterthat measuresaerosols supersaturation
may be lower than indicatedby the plate
largerthan -10 nm in diameter,and an opticalparticle temperaturedifference. Work is continuingto further
counter(OPC)thatmeasures
aerosols
withdiameter
>0.3 [xm characterizethe chamber'ssupersaturation.
(D0.3).Althoughtheinstrument
package
ascends
to a pressure The balloon-borne OPC and CN counter
are
also
of 10 mbarto measurestratospheric
aerosol,theCCN counter Universityof Wyominginstruments.They are basedon the
is switchedoff at 200 mbar to avoid overheatingresulting measurementof scatteredwhite light by individual particles
fromreducedheattransferat low pressures.
It takes-30 min passingthroughthe instrumentfollowingthe designof Rosen
to obtainan ascentprofileto 200 mbar. CCN measurements [1964]. Both countersare continuousflow instrumentsthat
are obtainedevery30 s, providinga verticalresolutionof countindividualparticles. The flow rate for eachinstrument
-125 m. The CN and D0.3measurements
are recordedevery was measuredbefore and after each balloon flight. Typical
flowratesare13 cm3 s-1for theCN counter
and167cm3 s'l
10 s, whichgivesa verticalresolutionof-40 m.
DELENE AND DESHLER: CLOUD CONDENSATION NUCLEI ABOVE WYOMING
12,581
abovethe surface,is nearlyconstantor slightlydecreases
until
5 km abovemean sea level, and then increasesin the upper
troposphere.The relative humidityis nearly constantin the
lower troposphere,decreasesabruptly at ~5 km, and is
relativelyconstantand <20% in the uppertroposphere.The
aerosolconcentrations
are relativelyconstantfrom the surface
refraction between 1.30 and 1.50 is -10%.
For aerosols with
to -5 km, decreasesharply at ~5 km, and are relatively
diameter>0.3 •tm, the uncertaintiesin concentrationare 75% constantin the upper troposphere.Figures2b-2f differ from
to2%forconcentrations
of0.001to1.0cm-3respectively. Figure 2a (and the eight profilesnot presented)in showing
The balloon-borne CN counter is similar to the OPC,
variationsin aerosolconcentration
that correspondto relative
except the particles are exposed to a supersaturated humidity changes,in addition to the typical decreasein
environmentof ethylene glycol vapor to grow particles to aerosol concentration between the lower and upper
(at-5 km for Figure2a).
optically detectable sizes [Rosen and Hofmann, 1977]. troposphere
Laboratorytests indicate a countingefficiency of >50% for
The thermodynamicparametersof equivalentpotential
particles> 10 nm with little pressuredependenceat pressures temperatureand relative humiditycan be usedto groupthe
between 200 and 75 mbar [Rozier, 1993]. To measure aerosolmeasurements
into five atmosphericlayers: surface,
troposphericCN concentrations,the tropospheric air is lower troposphere,upper troposphere,stratosphere,and
diluted by a factor of 100 before it entersthe CN counterto humidity layers. The altitude boundariesof each layer were
avoid particle coincidence[Rosenand Hofmann, 1977]. In determined independently for each sounding using a
the stratosphere
the dilutionis removed. Particlecoincidence consistent thermodynamic definition. Changes in the
should be considered for undiluted CN concentrations >100
boundariesof these layers reflect the influence of many
cm-3. The dilution valve is set suchthat concentrations
this atmospheric processes. Aerosol concentrations above
high are avoided throughoutthe flight. For tropospheric Laramie, Wyoming, are summarized by averaging
(diluted) CN
measurements the uncertainties for measurements
over the lower and upper troposphericlayers.
concentrations
of 100to 1000cm'3are8% to 3% owingto These layer averages are then compared with two
for the OPC. In-flight flow rate measurements
are planned
for future balloon flights. For the OPC the aerosol size is
determinedfrom light intensitymeasurements
at 40ø from the
forward direction using Mie theory and assumingspherical
particleswith an index of refractionof 1.45 [Hofmannand
Deshler, 1991]. The sizingerror for particleswith indicesof
measurements above Lauder, New Zealand.
countingstatistics.
Careful
evaluation
of
each
aerosol
instrument
Measurements
under
within the surface,stratospheric,
and humiditylayersare not
troposphericflight conditions has resulted in some data extensiveenoughto providemeaningfulaverages;however,it
modificationssince the preliminary results of Delene et al. is useful to define these layers so that measurementswithin
[1998]. Connectingthe inlet heaterto the OPC with 5-mm themcan be excludedfrom the lower anduppertropospheric
(inside diameter)stainlesssteel tubing adds a pressuredrop layers.
that decreasesthe laboratory measuredflow rate by 5%,
Nighttimecoolingof air near the Earth's surfaceproduces
resultingin a 5% increasein the aerosolconcentration. The a temperatureinversion resultingin a shallow surfacelayer.
addedpressuredrop is negligiblefor the low flow rate CN The equivalentpotential temperaturestronglyincreasesfrom
counter. Recent laboratorytesting indicatesthat the actual the bottomto the top of the surfacelayer;an increaseof-4 K
dilution ratio is greater than used by Delene et al. [1998], per 200 m is apparentin Figures2a, 2c, and 2d. The top of
leadingto a correctionof the tropospheric
CN measurements the surface layer is located where the equivalentpotential
reported by Delene et al. [1998] by a factor of 1.6. temperaturestartsto increaserelatively slowly or is constant
Laboratorycalibrationsof the balloon-borneCCN counter with height. The first horizontalsolid line abovethe surface
using a higher resolution,greatermagnification,lower noise, on the right-handside of the equivalentpotentialtemperature
more light sensitivevideo camera, and comparisonsof the profiles in Figure 2 indicatesthe top of the surfacelayer.
CCN counter to a commercially built CN counter (TSI Figure 2e shows that the aerosol concentration decreases
Incorporated, model 3010 condensationparticle counter) through the surface layer; however, Figure 2a shows no
resulted in a more accurate absolute calibration of the CCN
changewith heightthroughthe surfacelayer. By examining
counter. Thusthe CCN concentrations
reportedhere are 18% all of the aerosolprofiles, it is evident that concentration
lower than those reported by Delene et al. [1998]. changesin the surfacelayer are mostapparentin the smaller
Calibrations with the new video camera have shown that the
sized aerosol. The CN concentrationtypically decreases
CCN counterhas an accuracyof ~10% at 1% supersaturation sharplywhile the D0.3concentrationdecreasesvery slightly.
[Delene and Deshler, 2000].
This may be due to a strongtemperatureinversiontrapping
newly producedaerosols. These aerosolscould be from the
nearbycity of Laramie (population26,000). Hudson[1991]
3. Atmospheric Aerosol Layers
points out that even a small coastal town produces a
Figure 2 presents measurementsof D0.3, CCN, CN, significantincreasein CCN concentration.There seemsto be
temperature,and relative humidity for five balloon ascent a relationshipbetween the strengthof the temperature
profilesin Wyoming and one profile in New Zealand. Seven inversion and the degree to which the CN concentration
additionalWyomingprofilesandoneNew Zealandprofile are changes from the surface to the bottom of the lower
not shown. The aerosolprofile shownin Figure 2a is typical troposphericlayer.
of the additionalWyomingprofilesnot shownhere. These
Thelowertropospheric
layergenerally
coverstheregionof
profiles show several distinct layers defined by the convectiveinstabilityand significantcloud formationin the
Thelowertropospheric
layerextendsupwardfor
thermodynamic parameters of equivalent potential atmosphere.
temperatureand relative humidity. In these profiles the severalkilometersabovethe surfacelayerandis cappedby
equivalentpotentialtemperature
increasesin the first 300 m thefirstsharpincrease
in equivalent
potentialtemperature,
an
12,:582
DELENE AND DESHLER: CLOUD CONDENSATIONNUCLEI ABOVE WYOM•G
July
25,1997
Concentration
atSTP
[cm
-3]
Relative
Humidity
[%]
100 10• 10z 103 104 105 0
50
b
100
October
22,1997
Concentration
atSTP
[cm
-3]
February03, 1998
Relative
Humidily
[%]
ld• 100 10• 10z 103 104 10s 0
50
Concentration
atSTP[cm
-3]
100
100 10t
!
•
11
11
•
lO
lO
10z
Relative
Humidily
[%]
103 104 105 0
i
•
50
,
,
100
,
13
11
o
lO
100 10•
10•
103 104 105275
Concentration
atSTP
[cm
-3]
325
375
June
10,1996
Concentration
atSTP
[cm
-3]
Relative
Humidity
[%]
100 10• 102 103 104 105 0
i
,
lff• 10o 10•
Equivalent
Theta
[K]
[
50
100
102 103 104 105275
Concentration
atSTP
[cm
-3]
e
325
375
August
02,1996
Concentration
atSTP
[cm
-3]
Relative
Humidity
[%]
16• 100 10• 102 103 104 105 0
i i i 13
50
100
tl
10z 103 104 10s 275
325
375
Equivalent
Theta
[K]
I(Y• 100 10•
10z 103 104 10s 275
10•
104 105 275
325
325
375
Equivalent
Theta
[K]
375
Equivalent
Theta
[K]
September
05,1996
Concentration
atSTP
[cm
-3]
Relative
Humidity
[%]
1(•• 100 10• 10z 103 104 105 0
i
i
50
lqO
i i i 13
11
D
Concentration
atSTP
[cm
-3]
10•
11
ß
100 10•
[
i i i 113
11 ao•
•
Concentration
atSTP
[cm
-3]
100 10•
Concentration
atSTP[cm
-3]
Equivalent
Theta
[K]
030
I(Y• 100 10•
102 103 104 105275
Concentration
atSTP
[cm
-3]
325
375
Equivalent
Theta
[K]
Figure 2. Vertical profiles of particleswith diameter> 0.3 gm (D>0.3 gm, thin line), CCN concentration
(!% supersaturation,circles), and condensationnucleus (CN) concentration(D>0.0! gm, thick line) at (a)
July 25, !997; (b) October22, !997; (c) February3, 1998; (d) June !0, 1996; (e) August 2, 1996; and (,f)
September5, 1996. Open circlesrepresentmeasurements
below the detectionlimit of the CCN counter. The
concentration
measuredby eachaerosolinstrumenthasbeencorrectedto standardtemperatureand pressure
(STP). Equivalentpotentialtemperature(thick line) and relativehumidity (thin line) are shownat fight. The
left axis denotesthe measuredatmosphericpressure,and the far fight axis denotesthe altitudeabovemean
sealevel. The surfaceat Laramie,Wyoming,is at-2.2 km abovemeansealevel. The horizontallines on the
far fight of the equivalentpotentialtemperature
plotsdenotethe lower(solidlines)andupper(dashedlines)
troposphericlayers. All profilesshownwere obtainedshortlyafter sunrise,usuallywith clear skiesand light
surfacewinds. The February3, !998, profile is from Lauder,New Zealand.The otherfive profilesare from
Laramie, Wyoming.
DELENE AND DESHLER: CLOUD CONDENSATION NUCLEI ABOVE WYOM•G
increaseof-2 K per 200 m. When a sharpincreasein
equivalent
potentialtemperature
is notpresent,thefirstsharp
decreasein relativehumidityis usedto definethe top of the
lower tropospheric
layer. For the examplesin Figure 2,
relativehumiditydecreases
areusedto aid in determiningthe
top of the lower troposphericlayer, the secondsolid
horizontal line above the surfaceon the right-hand axes in
Figure 2. The lower troposphericlayer is similar to a
thermodynamically
well-mixedlayer, but not as narrowly
defined. A lowertropospheric
layercan alwaysbe identified
in our profiles. Figure2a showsan exampleof the lower
tropospheric
layer,whichrunsfromnearthe surfaceto -5.1
km. The equivalentpotentialtemperatureslowly increases
within the layer, followed by a relatively suddenincrease
above the layer. The relative humidityis nearly constant
throughout
the layerand decreases
sharplyabovethe layer.
The lowertropospheric
layerhasslightlydecreasing
aerosol
concentrations
with a sharpdecrease
abovethe layer. Aerosol
concentrations
do not always decreasewith height in the
lower tropospheric
layeras is evidentby the slightincreases
12,583
a 2 orderof magnitudedecreasein D0.3as comparedwith a 1
orderof magnitudedecrease
in CN. Thesecorrelations
could
be relatedto cloudprocesses.A traceamountof precipitation
wasreportedupwindof the flight shownin Figure2b. The
high relative humidity observedin Figure 2c at 10 km
suggests
thatthe air waswithinor neara cirruscloud. Figure
2d at 6.8 km, Figure2e at 6.8 km, and Figure2f at 6.5 km
show changesin aerosolconcentrations
that are positively
correlatedwith relativehumidity. The correlationof D0.3 with
relative humidity is probably not a reflection of particle
growth in high humidity layers since the D0.3 particlesare
measuredafter dryingin the inlet heater. Rather,the positive
correlationmay reflect processingin clouds or humidity
layersthatmay increasegasto particleconversion.Kleinman
and Daum [1991] observed a correlation between aerosol
concentration
and water vaporin verticalprofilesobtainedon
12 aircraftflights,alongwith somerapidchangesin aerosol
concentration
that corresponded
with changesin water vapor.
Positive correlation of aerosol and relative humidity may
result from an interleaving of air masseswith different
origins; humidity may also be an important factor in
controllingthe formationand/orgrowthof the aerosols[Salk
troposphericlayerin Figure2c.
d and Hobbs, 1994]. Clouds
The upper troposphericlayer representsa stableregion et al., 1986; Hegg, 1990; Perta
whereequivalent
potentialtemperature
increases
steadilywith may alsobe responsiblefor elevatingthe CCN concentration
1994;Radke
height, humidity is low, and there is little variation in withinhumiditylayers[Saxenaand Grovenstein,
humidity. Generally,the uppertropospheric
layer extends and Hobbs, 1991].
downwardfrom the top of the sounding(200 mbar), or the
stratosphere,until the equivalent potential temperature 4. Summary of Laramie, Wyoming, Aerosol
becomesrelativelyconstant.The uppertropospheric
layeris Profiles
markedin Figure2 by the two horizontaldashedlinesalong
The aerosol measurementsabove Laramie, Wyoming, are
the right-handaxesandis evidentin Figure2a from -6 to 12
layersandarealso
km.
The D0.3 and CCN concentrations are constant groupedinto upperandlowertropospheric
throughoutthis layer,while the CN concentration
decreases divided into summer (June, July, August, and September;
slightly. In the winter,when the tropopause
is low, the seven balloon flights) and winter (November and January;
measurements
occasionally
extendedinto the stratosphere.
In four balloon flights) categories. The October 22, 1997,
thesecasesthe top of the uppertropospheric
layeris defined balloon flight (Figure 2b) falls between the summer and
in CCN
and CN
concentrations observed in the lower
by increases
in equivalentpotentialtemperature
and ozone, winter seasonand is not included in the summary. Figure 3
which mark the stratosphere.No stratospheric
layers are shows the variability of the lower and upper tropospheric
shownin Figure2. Flightsthat extendinto the stratosphere measurementsand gives aerosol seasonal averages and
typically show increasesin CCN concentrations
in the standarddeviationsfor the sevensummerballoon flights and
stratosphere;
however, the number of stratospheric
CCN the four winter balloonflights. The summeraveragesindicate
layersis too limitedto warrantfurtherdiscussion
here. Future that betweenthe lower and upper troposphericlayersthe D0.3
concentrationdecreasesby more than an order of magnitude
measurements
areplannedto investigatestratospheric
CCN.
decreaseby factorsof 3In addition to the surface,lower, and upper tropospheric and the CN and CCN concentrations
layers, which can be identified on all profiles, there are 6. The winter averages,comparedwith the summeraverages,
layers
sometimes layers where the relative humidity changes indicatethatbetweenthe lower anduppertropospheric
abruptly.Typically,whenthesehumiditylayersarepresentin the D0.3 and CCN concentration decrease is not as
whiletheCN concentration
decrease
is similar.
a balloonascentprofile(Figures2b-2f), theyarenot observed pronounced,
The CCN concentrationspresented in Figure 3 are
during balloon descent. This indicatesthat the humidity
with previousmeasurements.
Pruppacherand
layers have a small horizontalextent (<100 km). CCN consistent
concentrationchangeswithin humidity layers are usually Klett's [1997] summaryof continentalCCN measurements
of600to5000cm-3.Themost
extensive
datato
mirroredby similar changesin D0.3and CN concentrations. givearange
withourmeasurements
arethosecollected
byHobbs
Figure 2a presentsno humiditylayers,while Figures2b-2f compare
presentall the examplesof humiditylayersfoundin the data et al. [1985] over the High Plains of the United States.
set. For dataanalysispurposes,measurements
nearor within Frequency distributionsof the High Plains CCN
indicated two modes: a relatively low CCN
humidity layers have been groupedinto a layer category measurements
concentration
mode
withmeanconcentration
of 310cm-3and
separate
from theloweranduppertropospheric
layers.
modewithmeanconcentration
of
Humidity layerscan be separated
into caseswith negative a higherCCNconcentration
areat 1% supersaturation
at
correlationor positivecorrelationbetweenrelativehumidity 2200cm-3. Theseconcentrations
and pressure
and CCN concentration.Figure 2b at 4 km and Figure 2c at ambientpressure,not at standardtemperature
theseconcentrations
to STPwouldcausea
10 km show negative correlationcases where a relative (STP). Converting
change
in
concentration
of
less
than
a
factorof 2 assuming
humidityincrease,to a maximumof-70%, is associated
with
a decreasein aerosolconcentration.Figure 2b at 4 km shows
lower tropospheric
measurements.Hobbset al. [1985]
12,584
DELENE
AND DESHLER:
CLOUD
Summer
CONDENSATION
NUCLEI
ABOVE
WYOM]NG
Flights at Laramie
•o•A:
; [ Lower
I
I Troposphere
I
['
[
[ .:
I o•B:
:: I Upper
[
I Troposphere
[
I
[
[ -=
ß _•
t• 4 -
!•.o•
_ B.1
_
-
'• 02
õ1
r CCN = 445 __157
ß•. '•)
•'•101
- Do.3=16+_8
:_
--
.
_
10-1
-
-
-
-
_
-
950613
950926
960610
960802
g60905
970612
10-1
34
............ ...-'....
• 10ø
(.•
970725
CCN=126__
950613
950926
960610
960802
0'-', ,
960905
970612
970725
Flight Date
Flight Date
Winter Flights at Laramie
C'
n
Lower Troposphere
105 -__[
[
E'04 CN = 3183
•/____10
3
[
[ :_
105
• ' 04
+ 1122
Upper Troposphere
CN=
437+_
192
_
CCN
____10
3
=
64
_
36
:
•_
:
-/
-
CCN=146_
20
:*'•
10•
_.101
• 10ø
-Do.3 = 4.4 _ 2.2
•
_----
..--
_
:
-
-,•
'i
-
ß
_
.
.
10-•
• 10ø
Do3 = 0.7_ 0.5
.
-
--
I
951121
I
960124
I
961127
Flight Date
•
970122
10-1
I
951121
960124
I
961127
I
970122
Flight Date
Figure 3. Summer(a) lower and (b) upper troposphere
and winter (c) lower and (d) upper troposphere
aerosol concentrationsfrom measurementsat Laramie, Wyoming, with a balloon-borneoptical particle
counter (D>0.3 pm, circles), CCN counter (1% supersaturation,
squares),and CN counter (D>0.01 pm,
asterisks). The error bars represent1 standarddeviationfor the measuredconcentrations
within the layer.
The aerosolconcentrations
havebeencorrectedto standardtemperatureandpressure(STP). The averageand
standarddeviationfor the balloonflight data are displayed. Only measurements
abovethe detectionlimit of
the CCN counterare included, so the upper troposphereCCN concentrationmay be biased to higher
concentration,
especiallyfor wintermeasurements.
Becauseof instrumentproblems,somedataaremissing.
postulatethat the low-concentrationmode is the result of a the lower troposphere
(Figures4a and 4b) and is relatively
backgroundsourceof CCN and that the high-concentration consistentbetweenthe summerand winter flights. The lower
mode is due to additional natural and/or anthropogenic CCN/CN ratio in the lower troposphere
is consistentwith
sources
of CCN. Thelowertropospt/eric
CCNconcentrationssmaller particles at low altitudes, pointing to a lower
presentedin Figure 3 agreewith the low "background"CCN atmospheric
sourceof CN. The CCN/CN ratiosin theupper
concentrations.
troposphere
may be low for an aged aerosol,where the
The CCN/CN ratio and the D0.3/CCN ratio for summerand median size may be near the activation size for 1%
winterfightsat Laramie,Wyoming,are presented
in Figure4. supersaturation;
however,the WyomingCCN/CN ratiosare
The CCN/CN ratio is largerin the uppertroposphere
than in consistentwith three studiessummarizedby Pruppacherand
DELENE AND DESHLER: CLOUD CONDENSATION NUCLEI ABOVE WYOMING
Balloon
0.5
FI ights at
A- Summer Flights
•
0.4
I
I
I
I
I
0.09
Upper-
0.17 _+0.09
Laramie
B- Winter Flights
0.,5
I
Lower-
12,585
•
__+0.07
0.4
o
•
I
•
Lower-
0.05
_+ 0.02
Upper-
0.16 _+0.07
o
o3 0.3
z
z02 ß
0.2
(D
0
0.1
0.1
I
950613
950926
960610
960802
960905
970612
970725
951121
Flight Date
0.14
0.12
o
I
I
I
/ /,
961127
970122
Flight Date
D'
Summer Flights
I
960124
I
0.14
I
Lower-
0.04
Upper-
0.01 _+0.01
+_ 0.03
o
n- 0.08
I
Lower
0.12
0.10
Winter Flights
I
I
I
-
- O. 02 _+ O. 01
Upper - 0.01 _+0.01 -
0.10
n- 0.08
z
z
O
.-00•0.06
•0•0.06
.
0.04
0.04
0.02
0.02
o.oo 950613 950926 960610 960802 960905 970612 970725
o.oo
Flight Date
•3
951121
960124
/,o/?
961127
970122
,.
Flight Date
Figure4. Aerosolratios,CCN/CNfor (a) Summer
and(b) winterandD0.3/CCN
for (c) summer
and(d)
winterflights,at Laramie,Wyoming.Theaverage
andstandard
deviation
for thelowertropospheric
and
uppertropospheric
layersaredisplayed
in thelegends.Theaerosol
ratiosarecomputed
by averaging
aerosol
measurements
withina layerandthentakingtheratioof theaerosolaverages.
Klett [1997], where the continental CCN/CN ratio was from
0.004 to 0.21. In New Zealandthe CCN/CN ratio washigher,
approaching1.0 in somecases. The D0.3/CCNratio is larger
in the lower tropospherethan in the uppertroposphere,again
pointing to differences in aerosol size distribution and/or
solubility.This differencedecreases
in the winterflights.
Figure 5 presents the percentage vertical change in
concentrationper kilometer(aerosolvertical gradient)for the
lower and upper troposphericlayers. Usually, the CN
gradientis the largest.The aerosolgradients
for CN, CCN,
and D0.3are similar in the upper summerand lower winter
troposphere
but vary in the lower summerandupperwinter
troposphere,perhapsreflecting changesin atmospheric
mixingsummerto winter;however,theseobservations
arenot
extensiveenoughto makeany generalizations
regardingthe
relationship between aerosol concentrations and
meteorologicalprocesses.
12,586
DELENE AND DESHLER: CLOUD CONDENSATION NUCLEI ABOVE WYOMING
Summer Flights at Laramie
A: Lower Troposphere
100
•
ß ß 75
•o•'
c:
•
o
I
DO
.3
I
B' Upper Troposphere
•
= 9 _+20
* CN
100
25
ß
ß
o --
O
-25 -• []
.•
!
I
I
•
75
o
50
* CN -14 _+ 9
'•
25
[]
-o
0
o
ß
•
o
[]
o
o
o
•
-50
DO
.3
= -1• +_'15
•
nCCN
-7_410
-25
[] õ
o õ
•
B *
-5o
<• -75
<[[ -75
I
-100
•
o
•
•
•
oCCN
-8
18
19+_
+_
12 •
50
•
•
950613
I
950926
I
960610
I
960602
I
I
960905
I
970612
-100 •,• •,
970725
FlightDate
•o•,o•o• •
•o•,••o•
FlightDate
Winter Flights at Laramie
i
LowerTroposphere
lOO
I
I
I
oD0.3 - -25 _441I
lOO
75
5o
*
CN = -35
+_ 19
. .
,
D' UpperTroposphere
I
I
o D0.3
I
4 + 20
•CCN
=24_+
16
*
CN
=
-16
_+ 10
25
25-
o
0-
_
[]
-25
-
•
_
-25
[]
o
o
-5o
-50
-100 • 951121
I
_<•
8
-75
I
960124
I
961127
•
970122
FlightDate
-75
-1 00 951121
960124
961127
970122
FlightDate
Figure5. Summer(a) lowerand(b) uppertroposphere
andwinter(c) lowerand(d) uppertroposphere
verticalaerosolgradients
for D0.3(circles),CCN (squares),
andCN (asterisks).Negativegradients
indicatea
percentage
decrease
in concentration
with increasing
heightabovethe surface.The averages
andstandard
deviations
of thelayerdataaredisplayed
in thelegends.
of-3.0 km. The measurementsin Wyoming are obtained
-50-100 km downwind of the Snowy Range with peak
elevationsof 3.6 km. Althoughtwo flightsare not enoughto
they do providean interestingcontrast
As a point of comparison,two balloon flights were form any conclusions,
conducted at Lauder, New Zealand (45øS), which has a to the Wyoming measurements. Table 1 comparesthe
with the two New
landscapeand climatesimilarto Laramie,Wyoming(41øN). averagesummerWyomingmeasurements
Lauder is -100 km downwind of the SouthernAlps that run Zealand measurements.The most strikingdifferenceis that
are approximatelytwice as high in New
alongthewestcoastof theSouthIslandwithpeakelevationsCCN concentrations
5. ComparisonBetweenWyoming and New
Zealand
Profiles
DELENE AND DESHLER: CLOUD CONDENSATION NUCLEI ABOVE WYOMING
12,587
Table1. MeanandStandard
Deviations
of Aerosol
Concentrations,
Ratios,andGradients
in theLowerandUpperTroposphere
for
Summer
BalloonFlightsatLaramie,
Wyoming
(SevenFlights),andLauder,NewZealand(TwoFlights)
Laramie,
Wyomin$
Lauder, New Zealand
Lower
Troposphere
Upper
Troposphere
Lower
Troposphere
Upper
Troposphere
Do.
3,cm'3
16+ 8
1.3+ 0.8
7.0+ 1.8
0.9+ 0.4
CCN,cm'3
CN,cm'3
Do.3/CCN
445+ 157
6837+ 3842
0.041+ 0.026
126+ 34
966+ 541
0.011+ 0.006
964+ 17
5662+ 4860
0.007_+0.002
246+ 49
445+ 206
0.004_+0.001
0.59 _+0.16
CCN/CN
0.09 _+0.07
0.17 _+0.09
0.27 + 0.24
D0.
3gradient,
%km'•
9 + 20
-13+ 15
-30+ 3
18+ 30
CCNgradient,
% km'•
CNgradient,
% km'•
-8 5 18
-195 12
-7 + 10
-145 9
-3 5 25
-345 46
2 5 0.3
-3 5 8
Definitionof theloweranduppertroposphere
is givenin thetextandis usedconsistently
for boththeWyomingandNewZealand
measurements.D0.3denotesparticleswith diameter>0.3 gm, CCN is cloud condensation
nucleus,and CN is condensation
nucleus.
Theaerosol
ratios(D0.3/CCN
andCCN/CN)arecomputed
by averaging
measurements
withina layerandthencomputing
theratioof the
aerosol
averages.Aerosolgradients
(D0.3,CCN, andCN) givethepercentage
change
perkilometerin concentration.
Negativeaerosol
gradients
indicatea decrease
in concentration
withincreasing
heightabovethesurface.
Zealand as in Wyoming, whereasthe concentrationsof D0.3
were measured with a balloon-borne
and CN
higher vertical resolutionand a greater altitude range than
previously available profiles. The CCN profiles were
are similar
or lower.
This results in New
Zealand
CCN/CN ratio nearing 1.0 in the upper troposphereon one
New Zealand flight. The higherCCN concentrations
in New
Zealand are not the result of enhancedlocal pollution since
the measuredCN and ozone (not shown) concentrationswere
below the concentrationsmeasuredin Wyoming. The larger
CCN concentrationmay be the result of obtaining New
Zealand measurementsat the height of the seasonalcycle in
CCN concentration [Hobbs et al., 1985]; however, none of
the Wyoming profiles has CCN concentrationsnear those
obtainedin New Zealand. The higher CCN concentrationis
likely due to differencesbetween the ambient particle size
distribution and/or chemical compositionin New Zealand
comparedto Wyoming.
The upper troposphericCCN measurementsabove New
Zealand agree well with late spring/early summer upper
tropospheric
measurements
madeby Hudsonet al. [1998] of
instrument
and have a
typically obtainednear dawn, with clear skies and light
surface winds.
Concurrent aerosol measurements were also
madeat smaller(CN) andlarger(D>0.3 gm) sizes. The high
verticalresolutionof the balloonprofilesshowsthat aerosol
measurementscan be classified into distinct atmospheric
layersbasedon equivalentpotentialtemperatureand relative
humidity. The profiles reveal that changes in CCN
concentrationare correlated with changes in the aerosol
concentration
at othersizesand are associated
with humidity
changes. A typicalprofile consistsof a relativelyconstant
CCN concentrationwithin the lower troposphericlayer
typically from 0.3 to 2.5 km above the surface, a decrease
abovethe lower troposphericlayer, and a relativelyconstant
CCN concentration
in the upper troposphere.This typical
profile indicatesthat verticalmixing playsan importantrole
-200-300cm'3 overthe Southern
Ocean. Thisagreementin the distributionof CCN. Differencesfrom this typical
suggeststhat there is no major changein CCN concentration profile occurin the presenceof humiditylayers,which were
changes
when upper troposphericair passesover New Zealand; observedin 5 of the 14 CCN profiles. Concentration
however, the CCN concentrations reported here are associatedwith humidity layers are evident in the three
CN, CCN, andD0.3.
substantiallyhigher in the lower troposphereover New differentaerosolsizemeasurements,
The averagesummerlower and upper troposphericCCN
Zealand comparedto the marine boundarylayer over the
concentrationat Laramie, Wyoming (445 + 157 and 126 + 34,
respectively), shows little variability between flights
the CN concentration.The uppertroposphericNew Zealand conductedunder similar meteorologicalconditions. The
CN measurements agree with upper tropospheric average summer CCN/CN ratio in Wyoming shows an
measurementsof-600
cm'3 over the Southern Ocean increasefrom 0.09 in the lower troposphereto 0.17 in the
[Hudsonet al., 1998]. In the lower troposphere,however,the uppertroposphere.The CCN/CN ratio increasesbetweenthe
New Zealand CN concentrationsare substantiallylarger than lower and upper troposphere,indicatingdifferencesin the
measurements
of <500 cm-3 in the marineboundary
layer size distributionand/orcompositionof the aerosol. Aerosol
[Hudson et al., 1998].
The contrast in CN and CCN
gradients
withinthe loweranduppertropospheric
layersshow
concentration between marine air and continental air has been
relatively small percentagechangeswithin the layers and
observedin previousstudies[Pruppacherand Klett, 1997]. indicate that the smaller-sizedaerosol have the largest
Hudson [1991] observed that surface CN concentrations are percentagedecreasewith increasingheightabovethe surface.
much higher 100 km inland, while CCN concentrationsare The decrease in CCN concentration between the lower and
similar.
upper troposphereand the typical negativegradientin CCN
concentrationwithin the lower tropospheresuggesta CCN
sourcenearthe surface. Two measurements
of CCN profiles
6. Conclusions
Southern Ocean [Hudson et al., 1998].
A similar contrast
between marine air and air over New Zealand is observed in
above
New
Zealand
indicate
concentrations
that
are
Fourteen,high vertical resolution,midlatitude, continental approximatelytwice as high, in both the lower and upper
CCN profiles have been summarized. These CCN profiles troposphere,
as CCN concentrations
in Wyoming. Although
12,588
DELENE AND DESHLER: CLOUD CONDENSATION NUCLEI ABOVE WYOMING
suggestive
of a difference
in CCN concentration
between Hudson,J.G., Y. Xie, andS.S. Yum, Verticaldistributionsof cloud
these two sites, further measurements
in the Southern
Hemisphere
arerequired
toestablish
thisdifference.
condensation
nucleispectraoverthesummertime
Southern
Ocean,
J. Geophys.
Res.,103, 16,609-16,624,1998.
Jennings,
S. G., AerosolEffectson Climate,pp. 275-297,Univ. of
Ariz. Press,Tucson, 1993.
Acknowledgments.
Lyle WomackandJasonGonzalesprovided Jiusto,J. E., R. E. Ruskin,andA. Gagin,CCN comparisons
of static
engineering
supportin conducting
balloonflights. Jefferson
Snider,
diffusionchambers,J. Rech.Atmos., 15, 291-302, 1981.
Perry Wechsler, and Gabor Vali provided valuable support, Jones,A., D. L. Roberts,and A. Slingo,A climatemodelstudyof
suggestions,
andcomments.Assistance
with the balloonflightsby
indirect radiative forcing by anthropogenicsulphateaerosols,
thepersonnel
of NIWA at Lauder,in particular,
BrianMcNamara,is
Nature, 370, 450-453, 1994.
gratefullyacknowledged.This researchwas supportedby grants Kleinman,L. I., and P. H. Daum, Vertical distributionof aerosol
from the National Aeronauticsand Space Administrationand the
particles,watervapor,andinsolubletracegasesin convectively
National Science Foundation.
mixedair,J. Geophys.
Res.,96, 991-1005,1991.
References
Boucher, O., and T. L. Anderson, General circulation model
assessmentof the sensitivity of direct climate forcing by
anthropogenic
sulfateaerosolsto aerosolsize and chemistry,J.
Geophys.
Res.,100, 26,117-26,134,1995.
Charlson,R. J., S. E. Schwartz, J. M. Hales, R. D. Cess, J. A.
Pan, W., M. A. Tatang, G. J. McRae, and R. G. Prinn, Uncertainty
analysisof indirect radiative forcing by anthropogenicsulfate
aerosols,J. Geophys.Res.,103, 3815-3823, 1998.
Penner, J. E., R. J. Charlson, J. M. Hales, N. S. Laulainen, R. Leifer,
T. Novakov, J. Ogren, L. F. Radke, S. E. Schwartz,andL. Travis,
Quantifying and minimizing uncertaintyof climate forcing by
anthropogenicaerosols,Bull. Am. Meteorol. Soc., 75, 375-400,
1994.
CoakleyJr.,J. E. Hansen,andD. J. Hofmann,Climateforcingby Perry, K. D., and P. V. Hobbs, Further evidence for particle
nucleation in clear air adjacent to marine cumulus clouds, J.
anthropogenic
aerosols,
Science,255, 423-430, 1992.
Geophys.Res.,99, 22,803-22,818, 1994.
Chuang,C. C., J. E. Penner,K. E. Taylor,A. S. Grossman,
andJ. J.
Walton, An assessment
of the radiativeeffectsof anthropogenic Philippin, S., and E. A. Betterton, Cloud condensationnuclei
concentrationsin Southern Arizona: Instrumentationand early
sulfate,J. Geophys.Res.,102, 3761-3778, 1997.
observations,Atmos. Res., 43, 263-275, 1997.
Delene, D. J., and T. Deshler, Calibrationof a photometriccloud
Pruppacher,H. R., and J. D. Klett, Microphysicsof Clouds and
condensationnucleus counter designed for deployment on a
Precipitation,pp. 287-289, Kluwer Acad. Norwell, Mass., 1997.
balloonpackage,
J. Atmos.OceanicTechnol.,17, 459-467,2000.
Delene,D. J., T. Deshler,P. Wechsler,and G. A. Vali, A balloon- Radke, L. F., and P. V. Hobbs, Humidity and particlefields around
some small cumulus clouds, J. Atmos. Sci., 48, 1190-1193, 1991.
bornecloud condensation
nuclei counter,J. Geophys.Res., 103,
Rosen,J. M., The verticaldistributionof dustto 30 km, J. Geophys.
8927-8934, 1998.
Ghan,S. J., C. C. Chuang,andJ. E. Penner,A parameterization
of
clouddropletnucleation,Singleaerosoltype,Atmos.Res.,30,
197-221, 1993.
Res., 69, 4673-4676, 1964.
Rosen, J. M., and D. J. Hofmann, Balloonborne measurements of
condensation
nuclei,J. Appl. Meteorol., 16, 56-62, 1977.
Haywood,
J. M., andK. P. Shine,Theeffectof anthropogenic
sulfate Rozier, W. R., Analysis of a balloonborne, continuous flow
and soot aerosol on the clear sky planetaryradiation budget,
condensation
nuclei growthchamber,M. S. thesis,85 pp., Univ.
of Wyo., Laramie, December1993.
Geophys.
Res.Lett.,22,603-606, 1995.
Hegg,D. A., Heterogeneous
production
of cloudcondensation
nuclei Salk (Suck) S. H., R. E. Thurman, and C. H. Kim, Growth of
ultrafineparticlesby browniancoagulation,Atmos.Environ., 20,
in the marineatmosphere,
Geophys.Res.Lett., 17, 2165-2168,
1990.
Hobbs, P. V., Aerosol--Cloud--ClimateInteractions,pp. 33-73,
Academic,SanDiego, Calif., 1993.
Hobbs,P. V., D. A. Bowdle, andL. F. Radke,Particlesin the lower
773-777, 1986.
Saxena, V. K., and J. D. Grovenstein, The role of clouds in the
enhancementof cloud condensationnuclei concentrations,Atmos.
Res., 31, 71-89, 1994.
troposphere
overthe High Plainsof the UnitedStates,II, Cloud Schafer, B., and H. W. Georgii, Airborne measurementsof
condensation nuclei and cloud condensation nuclei above the
condensation
nuclei, J. Clim. Appl. Meteorol., 24, 1358-1369,
Alpine Foothills,Contrib.Atmos.Phys.,67, 225-340, 1994.
1985.
Hofmann,D. J., and T. Deshler, Stratosphericcloud observations Twomey,S., AtmosphericAerosols,Elsevier,New York, 1977.
duringformationof theAntarcticozoneholein 1989,J. Geophys. Twomey, S., Aerosols,clouds and radiation,Atmos. Environ., Part
Res., 96, 2897-2912, 1991.
Hoppel,W. A., J. E. Dinger,andR. E. Ruskin,VerticalProfilesof
CCN at VariousGeographical
Locations,J. Geophys.Res.,30,
A, 25A, 2435-2442, 1991.
Wigley, T. M. L., Possibleclimate changedue to SO2-derivedcloud
condensationnuclei, Nature, 339, 365-367, 1989.
1410-1420, 1973.
D. DeleneScientificComputingResearchAssociate,Universityof
Hudson,J. G., Observationsof anthropogenic
cloud condensation
nuclei,Atmos.Environ.,Part A, 25A, 2449-2455, 1991.
NorthDakota,JohnD. OdegardSchoolof AertzpaceSciences,
University
Hudson,J. G., andP. R. Frisbie,Surfacecloudcondensation
nuclei Avenue & Tulana
Drive, Grand Forks, ND 58202-9007.
and condensation nuclei measurementsat Reno, Nevada, Atmos.
([email protected])
Environ., Part A, 25A, 2285-2299, 1991.
Hudson,J. G., andP. Squires,Continental
surfacemeasurements
of
T. Deshler,Departemntof AtmosphericScience,Universityof
Wyoming,Laramie,WY 82071.
CCN flux, J. Atmos.Sci., 35, 1289-1295, 1978.
Hudson,J. G., and Y. Xie, Cloud condensation
nucleimeasurements
in thehightroposphere
andin jet aircraftexhaust.,Geophys,
Res. (ReceivedFebruary7, 2000;revisedOctober4, 2000; accepted
Lett., 25, 1395-1398, 1998.
October4, 2000.)