JOURNAL
OF GEOPHYSICAL
RESEARCH,
VOL. 98, NO. C10, PAGES 18,269-18,276, OCTOBER
15, 1993
Responseof Lead Cyclingin the SurfaceSargassoSeato Changes
in TroposphericInput
ALAIN J. VERON,• THOMASM. CHURCH,:A. RUSSELl.FLEGAL,3 CLAIRC. PATTERSON,
4 AND YIGAL EREL4
Lead and its stableisotopeshave been analyzedin surfacewater samples(0-600 m) and trapped
pro'fides
collectedat theBermudaAtlanticTime SeriesStation(U.S. JointGlobalOceanFlux Study)in April
andNovember1989.Theseresultsare comparedwith wet atmospheric
leaddepositionas determinedfrom
precipitationcontinuouslycollectedin BermudasinceAugust 1988 as part of the Atmosphere-Ocean
ChemistryExperimentprogrin. Despitean expectedseasonalvariability,lead concentrations
in surface
watershaveclearlydecreased
by 30 to 40% since1979in response
to a corresponding
declineby a factorof
5 to 8 of the tropospheric
deposition.
This resultis corroborated
by stablelead isotopemeasurements
with
2ø6•Pb
ratioswhichare significamfiy
lessradiogenic
(1.18-1.20)in thefirst500 m of the 1989profile
thanthosemeasured
in 1984(1.20-1.21).Thisisotopicshiftreflectschanges
of leadore supplyin the United
Statesaswell as a relativeincreaseof the Eurafrieancontribution
to leadinputin the northwestAriantiethat
is likely due to the reductionof lead emissionsfrom gasolineconsumption
in North America.This fast
•onse of leadto interannual
variations
of thetroposphere
inputintosurface
waters
is related
to its
efficientbioreaefivity,asdemonstrated
with a sedimenttrapdeployedat 150 m in April 1989.Sedimenttrap
resultsshowthe rapidpenetration
of leadinto the first 200 m associated
with largeparticlesduringa period
of highplanktonactivity.Retrospective
isentropie
•ir masstrajectories
in threedimensions
coupledwiththe
precipitation
eventscollectedin Bermudain 1988-1989showthat30 to 40% of the annualleaddeposition
originatesfrom the tradeeasterlymeteorological
regime.This inputis clearlyevidencedwith lead isotopic
signalsobservedin surfacewaters(0-100 m) in April andNovember1989.We showthat lead accumulated
in the seasonal
mixedlayer(0-50 m) reflectsthisatmospheric
input.Takinginto accountthe isotopicsignal
measured
in thismixedlayer(1.178•0.001)aswell astherespective
contribution
of thetemperate
westerlies
and trade easterliesto atmosphericlead depositionto the SargassoSea, we calculatethat the isotopic
signaturefromthe Eurafficanregionsis 1.155(•0.004). Basedon theseresults,we calculatethatthe actual
206pb/•Pb averageratioin surfacewatersof theSargasso
Seais 1.188(•0.004).
INTRODUCTION
of the sourcesand transportprocessesof pollutanttracemetals
in the marine environment [Sheriand Boyle, 1988; Flegal et
al., 1989; Maring et al., 1989; Sturges and Barrie, 1987,
From the 1930s through the 1970s, the deposition of
anthropogenic
lead increasedby a factor of 3 to 5, primarily
1989a, b; Hamelin et al., 1989; Church et al., 1990; Vdron et
becauseof the consumptionof leadedgasolinein the northern
al., 1992].
hemisphere[Murozomiet al., 1969; Shen and Boyle, 1988]
Anthropogeniclead is very reactivein the ocean,causing30
causinglead to be the most atmosphericallyenrichedmetallic
to 40% of the contaminantlead depositedin the surfacewaters
pollutant from anthropogenicactivities. As a consequence,
to be readily accumulatedin surficialpelagicsediments[Vdron
contaminant lead constitutes more than 95% of the lead in the
et al., 1987; Hamelin et al., 1990] becauseof rapid vertical
troposphere and surface waters of the North Atlantic,
transportassociatedwith large marine particles generatedin
perturbating
naturalleadseawaterprofriesandcycling[Duceet
al., 1976; Ng and Patterson, 1981; Schaule and Patterson,
1983; Patterson, 1987; Patterson and Settle, 1987; Vdron et
al., 1987; Nriagu andPacyna,1988]. U.S. regulationson lead
concentrations
in gasolinehave resultedin a reductionof lead
emissions in North America
since the mid 1970s. The increase
the euphotic zone [Lambert et al., 1991] and to a lesser extent
with small suspended material repackaged at greater depth
[Sherrell et al., 1992]. One of the main uncertaintiesregarding
lead cycling in the ocean is due to the variations of the input
function.
Lead concentrations
in surface waters show seasonal
andsubsequent
decreasein lead depositionhasbeenmonitored variationsrelated to hydrographicfeatures,biologicalactivity,
using stableradiogenicisotopesof lead (M= 206, 207, 208) and aeolian deposition [Boyle et al., 1986], the relative
contributions
of which are difficult
to evaluate. In order to
measuredin corals[Sheriand Boyle, 1987], in lake sediments
tel'me
budget
calculations
and
better
understand
the responseof
[Shirahataet al., 1980] and in streamwaters[Erel et al., 1990].
Theseisotopesdisplaycharacteristicanthropogenicsignatures
from the different countries surrounding the North Atlantic
owing to the variety of lead oresusedto producealkyl lead or
nonferrousalloy [Chow et al., 1975; Church et al., 1990;
Hopperet al., 1991]. As such,lead isotopesare efficienttracers
surfaceseawaterto modificationsof the troposphericinput, we
:College
ofMarine
Studies,
University
ofDelaware,
Newark.
TheBATSstationhasbeenchosen
because
of its proximityto
have analyzedlead and its stableisotopesin surfacewaters(0600 m) and large particlesin April and November 1989 at the
JointGlobalOceanFlux Study(JGOFS)BermudaAtlanticTime
Series (BATS) station near Bermuda (31ø5N, 64ø1W). These
results are compared to the atmospheric deposition as
calculatedfrom lead concentrationsmeasuredin precipitation
•Laboratoire
desO6osciences
de l'Environnement,
Universit6
Aix- continuouslycollected at Bermuda since 1988 as part of the
Marseille
III, CNRS,Marseille,
France.
Atmosphere-Ocean
ChemistryExperiment(AEROCE) program.
qnstitute
ofMarine
Sciences,
University
ofCalifornia,
Santa
Cruz. theNorthAmerican
continent
andtoBermuda
which
represents
4Division
ofOeological
and
Planetary
Sciences,
California
Institute
of anidealplatform
to sample
theatmospheric
transport
of
Technology,
Pasadena.
anthropogenic
emissions
associated
withtheNorth
American
Copyright
1993
bythe
American
Geophysical
Union.
temperate
westerlies
[Church
etal.,1984,
1990;
Galloway
and
Whelpdale,1987]. Furthermore,SchauleandPatterson[1983],
Paper
number93IC01639.
Boyleet al. [1986],ShenandBoyle[1988],Sherrellet al.
0148-0227/93/93IC-01639505.00
[1992] and Sherrell and Boyle [1992] previouslyreportedlead
18,269
18,270
VERONET AL: LEADCYCLINGIN THESARG^SSO
SEA
Pb (pm01/cm2/m0nth)
40
,
2O
15
10
5
0
Sept Oct Nov Dec jaa Feb Mar Apr May jua
1988
Jul Aug Sept Oct Nov Dec
1989
Fig. 1. Monthly wet depositionof lead in Bermuda(Tudor Hill) during 1988 and 1989 as calculatedfrom rain events
continuouslycollectedas part of the Atmosphere-Ocean
ChemistryExperimentprogram.
the
analysisin the SargassoSea that can be comparedwith our data 0.002) and are used in the discussion.As a consequence,
in order to evaluatethe temporal dynamicsof lead transferin 2ø6pb/2ø7pbratios are referencedmore often than otherratios.
Lead in precipitationwas analyzedby direct injection of
surfacewatersin responseto the phasingout of leadedgasoline
in the western North Atlantic.
unfiltered samples acidified in 0.4% HC1 into the graphite
furnace (model 700) of a Perkin Elmer atomic absorption
Mm•oDs
spectrometer(model 1100b) [Tramontarioet al., 1987]. The
Rain sampleshave beencollectedin Bermuda(West Tower) overalluncertainty,includingblanks,waslessthan 10% of the
at Tudor Hill since August 1988 as part of the AEROCE measured concentrations.
A Conductivity/Temperature/depth
rosetteequippedwith Goprogram. The samples are obtained using special automatic
collectorswith low-densitypolyethylene(LDPE) buckets.The Flo bottles was deployed to collect samples for nutrients,
collector is deployed on a platform at the top of a 20 m oxygen,salinity, temperature,and chlorophyll.Hydrographic
anodized aluminium tower. Samples are attended daily and parameterswere provided by G.A. Cutter (Department of
Old Dominion University,Norfolk, Virginia,
collected after each precipitation event. The rain fall is Oceanography,
automatically gauged and comparedto the total depositionas private communication,1990).
recorded in a bucket.
RESULTS
Seawatersampleswere collectedon the R/V Cape Hatteras
using 30 L Teflon©-coated Go-Flo bottles specially cleaned
and mountedon a kevlar line. After sampling,the bottleswere
transportedto a portable clean area where up to 2 L of each
bottle were transferredto LDPE bottlesand immediatelyfrozen.
Eachsamplewas thawedand acidifiedto pH 1.5 with ullxapure
nitric acid a few hoursprior to analysis.
Soutartype sedimenttrapswere deployedas a floating array
at 150 m betweenApril 18 and 23 at the BATS station.The trap
AtmosphericTransportand Deposition
Air mass transportis characterizedusing retrospectiveair
mass trajectories based on isentropic analysis techniques.
Meteorologicaldata from the National MeteorologicalCenter
(Camp Springs,Maryland) are used to determinegeopotential
heightsand windsfrom which the isentropiclevels are deftned.
Assumingthat air parcelsretain their identity over the period
of analysis and consideringa dry adiabaticmotion of the air,
consistedof paired0.25 m2 gel-coatedfiberglassconeswith the trajectoriesfrom Bermudaare calculatedup to 1 week back
in time. The uncertaintiesdue to theseassumptionsare mainly
swimmer avoidance cod ends designed by Coale [1990].
Material in the cod ends was preserved with high-purity relatedto the paucityof wind data recordedover the openocean
buffered
formalin.
Recovered
sediment
was dried and all
and are largely discussedby Merrill [1989]. According to
measurements were salt corrected.
Stunder et al. [1990] and W. Ellis and J. Merrill (personal
Lead in seawaterwas extractedby a dithizone-chloroform communication,1991) trajectoriesare classifiedas oceanicor
method and determinedby isotope dilution thermal ionization continental based on the time spent over the ocean prior to
massspectrometry
usinga 2øSpbspike[Patterson
andSettle, collection (continental with less than 3 days over the ocean
1976]. The uncertaintyon each measurementswas determined and oceanicwith 3 days or more over the ocean).Trajectories
to be less than 2% of the total concentration.
A multicollector
from North America are generally defined as continental.
VG/Sector solid sourcemass spectrometerwas used to collect Among the oceanictrajectories,we sort trajectoriesassociated
multiplesetsof stableleadisotopes(2ø4pb,2øtpb,2ø7pb,and with the trade easterlies. We observe that 90% of the air masses
2øSpb)ratioson unspikedaliquots.Usingthe standardSRM transportedwith the trade easterliesare confinedfrom Juneto
981 from the National Institute of Standardsand Technology
(Gaitherburg,Md.), the precisionof the instrumentwas found
to be 0.1%.
Because of a small fractionation
factor and better
detectionlimits, ratios of 2ø6pb/2ø7pb
are more accurate(+
October.
Lead concentrations in precipitation have been volume
weighted by month in order to calculate monthly deposition
fluxes (Figure 1). Deposition is corrected for the marine
VERONET AL.:LEAD CYCLINGIN •
recycled component based on sodium concentrations in
precipitation and an enrichment factor of 5000 which
expresses the enrichment of Pb/salt in sea-salt aerosols
comparedto the Pb/salt in seawater [Patterson and Settle,
1987]. In this study, sea-saltcorrectionaccountsfor less than
5% of total lead concentrationsin precipitation.
By couplinglead depositionand the air mass trajectories
correspondingto each rain event collected in Bermuda, we
calculated
westerlies
the relative
contribution
and of the Eurafrican
of the North
trade
American
easterlies
to the
SARGASSOSEA
18,271
Lead Isotopes
Lead isotoperatios in surfaceseawaterare shownin Table
lb and axe comparedto continental sources(Table 2) and
atmosphericresultsobtainedby the combinationof air mass
trajectoriesand stablelead isotopesin Bermuda[Vdronet al.,
1992].In April, the2øtpb/2ø7pb
profiledisplays
twodifferent
trends with ratios ranging from 1.186 to 1.188 in surface
waters(0-90 m) and from 1.197 to 1.199 below (200-550 m).
The tropospheric
lead isotopicsignalshowsa clearEuroafrican
signature
duringthe timeof thesampling,
with 2ø6pb/2ø7pb
troposphericlead input into surfacewatersof the SargassoSea
(Figure 2). We estimate that input from the trade easterly
regimecouldrepresentasmuchas30 to 40% of the overalllead
input in 1989 in the northwestAtlantic.
ratios ranging from 1.13 to 1.18 (Table 2). Likewise, the
November 2ø6pb/2ø7pbprofile has two discrete trends
including(1) ratiosof 1.177 to 1.180 in the farst30 m and (2)
Lead Concentrations
radiogenicsignatureis confinedto the seasonalmixed layer
in Seawater
ratios of 1.188
and corroborates
Leadconcentrations
in seawaterarereportedin Table la. The
April profile extends to 550 m. Lead concentrationsshow a
minimum in surfacewaters (70-75 pM), increaserapidly to a
maximumof 120 pM at 200 m, and then slightly drop from 200
to 1.190 between
50 and 150 m. The less
the accumulation
of lead from Eurafrican
origin. The underlying water reflects a mixing of North
American and Euroafricansourcesas expectedfrom the April
to 550 m (a 5-10% decline). The minimum in surfacewaterscan
be explainedby high plankton activity during the time of the
samplingas shownby a chlorophyllpeak at 100 m and a large
oxygen consumption in the first 200 m (Figure 3a). The
November profile lies between 0 and 150 m. Lead
concentrationsdisplay a maximum in surfacewaters at 10 m
(89 pM) anddecreasesharplyto reach72 pM at 150 m. A mixed
layer (0-50 m) due to a seasonalthermoclineis evidencedin the
salinity and temperature profiles (Figure 3b). This
hydrographicstructureinhibits exchangeswith the underlying
waters and contributesto the accumulationof lead deposited
during summertimewhen plankton activity is low [Boyle et
al., 1986]. Therefore we can expect lead concentrationsto
increase in this seasonal mixed layer during summertime.
Despite of seasonalvariations, lead concentrationsin April
and November 1989 are 10 to 50% lower than the previously
reporteddatafrom the Sargasso
Sea (Figure4).
profile.The 2øtpb/2ø7pb
ratiosmeasured
in surfacewatersin
1989 are significantlylessradiogenicthan thoseobservedby
$hen and Boyle [1988] in 1984 (Figure 5), while they agree
with theSherrell et al. [1992] profile determinedin 1988 below
100 m. These resultsclearly reflect the impact of the phasing
out of leadedgasoline,as alreadysuggested
by the decreaseof
lead concentrations
in SargassoSea surfacewaters.
DISCUSSION
Impactof thePhasingoutof LeadedGasolinein theNorthwest
Atlantic
Lead inventory in the SargassoSea surface water. As
determined
fromprecipitation
samples
(AEROCEprogram),the
wet inputaccounts
for 126 pmolcm-2 of leadin Bermuda
in
1989 (Figure1). Assumingthatdry depositionrepresents
5 to
20% of the total lead flux over the ocean [Settle and Patterson,
1982; Dedeurwaerder et al., 1985; Jickells et al., 1987;
Church et al., 1984, 1991], we calculate that the total lead
Pb (pmol/cm2/month)
3o
2o
'
METEOROLOGICAL
REGIME
-.
i•! Eurafrlcan easterlies
[7]North
American
westerlles
.
15
10
5
.
o
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEPT
OCT
NOV
DEC
1989
Fig. 2. Relative contributionof North American(hatchedcolumns)and Eurafricanregions(shadedcolumns)to the
tropospheric
inputof leadin Bermuda
usingismtropictrajectories
coupledwitheachrainevent.
18,272
VERONET AL.:LEAD CYCLINGIN TI-IE$ARGASSOSEA
TABLE la. DissolvedLeadConcentrations
(pmolL-x) in Surface
Watersof the BermudaArianticTime Station(BATS)
in April andNovember1989
Depth,m
April 1989
0.5
10
15
30
70
90
150
200
350
550
November1989
nd
nd
75.5
73.8
nd
102
nd
120
112
115
83.2
89.0
nd
88.4
78.5
nd
72.1
nd
nd
nd
100 m (box 1) and 100-600 m (box 2). Box 1 correspondsto
the surfacemixed layer [Boyle et al., 1986] (Figure 3) and box
2 corresponds
to the subtropicalmode water (18øCwater) andto
the upperthermocline.The termsAA1/At andAA2/At represent
the responseof the 2 reservoirs (as expressedby the lead
inventory, A1 andA2, in each reservoir) from 1979 to 1989 as
a function of
f(t)
time dependenttroposphericlead input to the mixed
layer (box 1);
AoZ1 lead accumulatedin box 1 at t=o (1979) (1592 pmol
cm-2) [SchauleandPatterson,1983];
KP1 rate constantfor removal of lead from box 1 by large
particles(0.3 year'l) [Vdronet al., 1987; Hamelin et
al., 1990; Lambert et al., 1991 ];
Here nd means no determination.
I•2
rate constant for lead ventilation
from box 1 to box 2
(0.5 year-1)[Nozakiet al., 1976];
deposition
to theSargasso
Seain 1989is 150-3:30
pmolcm-2.
Similar estimates and calculations made by Schaule and
Patterson[1983] in the SargassoSea (about250 km northwest
of Bermuda)indicatetropospheric
leadfluxesof 950'3:120pmol
cm-2in 1979, suggestinga reductionof the lead input by a
factor of 5 to 8 between
1979 and 1989 to the western North
Atlantic Ocean. We can verify that the decline of lead
concentrations
observed
in surface waters between
1979 and
Ke rate constantfor lateral export of lead from box 1 (0.1
year-1)[Vdronet al., 1991];
Ki rate constantfor lateral inport of lead into box 1;
K21 rate constantfor water transfer from box 2 to box 1 (0.1
year-1);
AoZ• lead accumulatedin box 2 at t=o [1979] (8208 pmol
cm-2) [SchauleandPatterson,1983];
KP• rate constantfor removal of lead from box 2 by small
particles(0.013 year'l) [Lambertet al., 1981; Sherrell
1989 (Figure 4) is directly related to this reduction of the
et al., 1992; Sherrell and Boyle, 1992];
atmosphericinput. To do so, we consider the surface water
K23 rate constant for lead ventilation
from box 2 to the
inventoryof Pb in 1989 from (1) measuredlead concentrations
main and lower thermocline(0.06 year-1) [Jenkins,
in April and November 1989 seawaterprofiles (this study) and
1980].
from (2) a massbalancecalculationusingPb inventoryin 1979
[Schauleand Patterson, 1983]. We determine (2) using a 10This simple box model is presentedas that for the Lasaga
year atmosphericinput function and the oceanicvertical and [1980] expression.Inventories are integrated over a 10-year
advective in and out Pb transport in the Sargasso Sea as period (1979-1989) accordingto equations(1) and (2). The
describedby Boyle et al.. [1986], Shen and Boyle [1988], input function f(t) correspondsto the atmosphericdeposition
Lambert et al. [1991], Vdron et al. [1991, 1992], Sherrell et ascalculatedby Schauleand Patterson[ 1983] in 1979 and this
al. [1992], Sherrell and Boyle [1992]. This inventory (2) is study (1989). Assuminga linear input, we estimatean annual
calculatedaccordingto the following expressions:
depositionof 590 pmol cm-2 of lead to the SargassoSea
between 1979 and 1989. This assumption leads to an
AA1/At=
f(t)+(K21)A2
+(Ki)A1-(KP1)A1-(K12)A1-(Ke)A1
(1) atmospheric
inputof 630-•5pmolcm-2in 1983whichis in
AA2/At
=(K12)A1
- (K23)A2
- (KP2)A2
(2) good
agreement
with
Boyle
etal.'s[1986]
and
Jickells
etal.'s
[1987] calculationof total lead depositionto the SargassoSea
at t=0 (1979)Ai=AoZI andA2=AoZ2.
Expressions
(1)and(2) in 1983 (748 pmol cm-2).Lead exportedfrom the North
refer to a box modelfor lead cyclinginto surfacewatersof the Americanbasinis likely to partiallyreturnwith North Atlantic
SargassoSea wherethe sizesof the oceanicboxesare set at 0- Deep Waters at depth(when entrainednorthwardby the North
TABLE lb. IsotopicCompoai6on
of Dissolved
Leadin SurfaceWatersof theBATSStation
Near Bermudain April andNovember1989
,
April 1989
0.5
10
15
30
70
90
150
200
350
550
1.186
nd
1.186
1.187
nd
1.188
nd
1.197
1.199
1.198
Here nd means no determination.
18.50
nd
18.42
17.55
nd
18.29
nd
nd
18.61
18.69
November 1989
37.95
nd
37.91
36.12
nd
37.64
nd
nd
38.02
38.18
1.180
1.177
nd
1.180
1.188
nd
1.190
nd
nd
nd
18.45
18.25
nd
nd
18.62
nd
18.56
nd
nd
nd
38.11
37.75
nd
nd
38.26
nd
38.00
nd
nd
nd
VERONET AL.:LEADCYCLINGIN TI-IESARGASSO
SEA
Salinity (psu)
Chlorophyll a (gg/!)
0,,0
0,1
ß
0,2
I
ß
0,,4
0,3
I
ß
I
18,273
ß
35,6
35,8
36,0 36,2 36,4
0
lOO
100
!
Chl
2OO'
200
!
!
•Ox
3OO
300
I
i
i
I
I
I
400
i
i
i
400
i
i
i
I
500
500'
i
i
!
!
/
!
!
/
600'
600
!
!
(a)
7OO
ß
,
180
-
,
200
-
,
-
220
,
240
-
700
260
Oxygen(gM)
-
12
,
(b)
-
14
,
16
-
,
18
.
,
20
-
,
22
-
,
24
Temperature(oC)
Fig.3. (a) Chlorophyll
(solidline)andoxygen
(dashed
line)concentrations
in surface
waters
at theBermuda
Ariantic
Time
Series(BATS) stauonnear Bermudain April 1989. Co)Salinity(solidline) andtemperature(dashedline) in surfacewatersat
the BATS
station near Bermuda in November
1989.
Atlantic Current) or to be entrained within the Subtropical sedimenttrap deployedduring5 daysat 150 m in April 1989.
North Atlantic Gyre into the North African basin.Accordingto Leadfluxesmeasuredfrom thistrap (1.1 pmol cm-2 for 5 days)
Jickells et al. [ 1987] calculations, this latter recirculation and account for 20% of the previous 3 months of aeolian
upwelledmixing in the upper thermoclinedo not significantly deposition(5.9 pmol cm-2). Such an efficient transportis
contributeto the lead inventory in the SargassoSea surface related to high productivity in the first 100 m as shownby the
waters at the latitude of Bermuda. Therefore we consider Ki as
oxygen and chlorophyll profiles (Figure 3a). During this
negligible 'm our calculation.
particuliar
period,leadis rapidlyscavenged
fromsurfacewaters
According to expressions(1) and (2), we calculate that the associatedwith plankton and fecal pellets [Lambert et al.,
integratedlead inventorybetween0 and 600 m (box 1 and 2) 1991]. Stable isotopic ratios confirm this assumption, with
should be 5480+-1100pmol cm-2 in 1989. The measured large particlesdisplayinga 2ø6pb/2ø7pb
ratio of 1.187 at
inventoryfrom our two 1989 profiles at the BATS station is
150 m, similar to thoseobservedin surfacewatersof the April
ratio at 200 m
6500+.300pmol cm-2. Thesetwo determinations
are in good profile and differentfrom the 2ø6pb/2ø7pb
agreement, suggesting that the decline of atmospheric (1.197). These results suggest that lead is effectively
deposition
is likely to be mos•tlyresponsible
for the observed transportedthrough the surface water column in a few days
of
decreaseof lead inventoryin surfacewatersof the SargassoSea, duringa periodof highplanktonactivity.The homogcneity
signalin the first 100m (1.186-1.188)canbe
althoughthis result shouldbe consideredwith caution given the2ø6pb/2ø7pb
the uncertainties
on our calculation
and the observed seasonal
variationof lead concentrations
in the mixed layer [Boyle et
a/., 1986 and this study].
Evidence of a rapid transfer of lead in surface waters.
Comparisonof the 1979 (9800 pmol cm-2) and 1989 (6500
pmolcm-2) leadinventories
between0 and600 m confirmsthe
strong reactivity of lead in seawater. This reactivity is
evidenced by the observed seasonal change of isotopic
compositionin surface waters (0-100 m) between April and
November 1989 (Figure 5) without much change in
concentration.Biological processesare likely to generatethis
efficient scavengingof lead in surfacewatersas suggestedby a
explained by the mineralization of large sinking particles
(>10ixm)[Lambert
etal., 1991
].
•
Stable lead isotopicsignal. A changeof the U.S. isotopic
signaturefrom 1.21-1.23 to 1.19-1.20has been0b•ervedin
thewestern
NorthAtlantictroposphere
duringthepast'•
20 years
[Shenand Boyle, 1987; Pattersonand Settle 1987; Vdron et
al., 1992]. This variation has been mainly attributed to the
phasing
outof leadedgasoline
in NorthAmerica,involvinga
relative change of atmosphericpollutant lead sourcesand a
reduction
ofleadoresupply
fromtheveryradiogenic
MissoUri
ores. The atmospheric change of stable lead isotope
compositionis clearly evidencedin surface Watersof the
18,274
VERONET AD: LEADCYCLINGINTItE SARGASSO
SEA
Dissolved Pb (pmoi/!)
0
0
80
40
•
120
thermocline with no noticeable change (1.198:L-0.001). The
160
200 similaritybetweenthe profilesat 500-600rn suggests
thatthe
i
4/89•-•
downwardperturbationinducedby changesin the tropospheric
signal in the past 10-15 years has little or no effect on the
I 9/84.•.•'•ø-/
reservoir of dissolved lead in the main and lower thermocline.
•"-..•"x, 4/84
100'
SeasonalVariationsand EurafricanInput
.-'
,..--
200-
300-
I
Lead concentrationsshow significant variations between
the April and November 1989 profiles (Table 1). As shownby
Boyle et al. [ 1986] in the SargassoSea, lead concentrationsin
surfacewaterscan displayseasonalvariationsof 25% or more.
The low concentrationsobservedin the first 50 rn of the April
profile (74-75 pM) have been explainedby a rapid scavenging
of lead associatedwith large biogenicparticlesduring a period
of high plankton activity. The maximum measuredat 200 rn
(120 pM) could be due to a partial mineralization of these
aggregatesfrom which lead is released.Lead concentrations
in
the first 50 rn are 15 to 20% higher in the November profile
(83-89 pM) than in the April profile (74-75 pM). In addition,
the concentrations
in the first 50 m in November
are 10 to 20%
higher than in the underlyingwaters (72-78 pM). This relative
400'
increase in concentration
/
x/L.' /
500
can be attributed
to an accumulation
of lead as suggestedby the presenceof a seasonalthermocline
which generally inhibits vertical mixing from May to October
in the SargassoSea (Figure3) [Boyle et al., 1986].
'The November isotopic signal measured in the seasonal
mixedlayerat 0-50 rn (206pb/207pb
= 1.177-1.180)is clearly
lessradiogenic
thanin April (2ø6pb/2ø7pb
= 1.186-1.187)and
in the underlyingwaters (50-150 m) of the November profile
(2ø6pb/2ø7pb
= 1.188-1.190).The 2ø6pb/2ø7pb
ratios in this
6•84 •
600
mixed layer suggesta mixing of North Americanand Eurafrican
components (Table 2). While the eastern North American
isotopic signal is well defined in 1989 (2ø6Pb/2ø7pb=
1.204+0.004) [Church et al., 1990; Vdron et al., 1992], the
Euro-African signal, associatedwith the trade easterlies,is
700
Fig. 4. Dissolvedlead concentrations
in Sargasso
Sea surfacewatersin
1979 [Schauleand Patterson,1983] (open circles), 1984 [Boyleet al.,
1986] (3 profiles; opentriangles), 1987 [Sherrellet al., 1992] (open
square)and 1989 (this study)(2 profiles;solidcircles).Isotopedilution
massspectrometry
measurements
are represented
by solidlinesand
atomicabsorptionmeasurements
are indicatedby dashedlines.
Sargasso
Sea(JGOFSstation)with206pb/2ø7pb
ratiosvarying
more diffuse (2ø6pb/2ø7pb= 1.11-1.16) because of the
variation and extent of the different European sources.
According to the air mass trajectories, the Eurafrican
componentis a significant troposphericsource in Bermuda
duringsummertime(Figure 2) when lead is trappedby a water
column thermostratification.Based on Figure 2, we calculate
that lead from the Eurafrican region represents51% of the
(this study) (Figure 5). This differencepropagatesto 500 m.
Below 200 m, the isotopic signal in the April 1989 profile
troposphericinput into the SargassoSea from May to October
1989. Lead accumulatedin the seasonalmixed layer (90-•:10
pmol cm-2) accountsfor 80% of the atmospheric
deposition
reflects an older signal which extends down to the main
duringthatperiod(110-t:20
pmolcm-2).Assuming
thatleadin
from 1.203 in 1984 [Shen and Boyle, 1988] to 1.187 in 1989
TABLE 2. Recent•ø6•Po Signatures
in theTroposphere
of theRegions
Potentially
Contributing
to theAtmospheric
IsotopicSignalin theNorthwest
Adnntlc
Source
WesternEurope
Eastern United States
Trade easterlies
Mixed UnitedStates/F.
arafrican
••I•o
1.115-1.150
Date
1990
1.130-1.155
1988
Reference
Location
Irel•,
Suias,Fr., Ger. A.R. Flegalet al. (l•r•. comm.1992)
Sweden
Hopperet al. [1991]
1.198-1.203
1989
1.205-1.210
1988
We, stem N. Atlantic
Western N. Atlantic
V,•ron et al. [ 1992]
Churchet al. [1990]
1.150-1.160
1988
Southern N. Atlantic
Churchet al. [1990]
1.193
1.186
1.179
1.182
1.131
April 19, 1989
April 20, 1989
Apd121, 1989
April 21, 1989
April 22-23, 1989
JGOFS
JGOFS
JGOFS
JGOFS
.IGOFS
BATS
BATS
BATS
BATS
BATS
station
station
station
station
station
Vtron et al. [ 1992]
V•ron et al. [ 1992]
V•tronet al. [ 1992]
Wron et al. [ 1992]
Vtron et al. [1992]
VEIlONEl' AL.:LEADCYCLING
INTHESARGASSO
SEA
206Pb/207Pb
18,275
the phasing out of leaded gasoline in North America. By
couplingeach precipitationevent to the corresponding
1,17
1,18
1,19
1,20
0
1989
(m)
1984
1,21 retrospectiveisentropicair mass trajectories,we estimatethat
30 to 40% of lead depositionin the westernNorth Atlantic is
associatedwith the Eurafrican trade easterlies. Considering
these results, we calculate the average isotopic signal
(2ø6pb/2ø7pb)
in theSargasso
Seasurfacewaterandin theEuro100'
African trade easterlies to be 1.188 (+0.004) and 1.155
(•-0.004) respectively. These isotopic signaturesas well as the
1988
relative increaseof the Eurafricancontributionand its impact
on surfacewatersshouldbe takeninto accountwhenmodeling
the futureresponseof lead cyclingin relationto changesof the
tropospheric
input in the westernNorth AtlanticOcean.
200
Acknowledgments.This researchwas supportedby grantsfrom NSF
AtmosphericSciencesDivision (ATM 9040433 and 9013224) as part of
the AEROCE program. We are very grateful to G. Cutter for the ship
time providedas part of his NSF grant (OCE-8800371) and to J. Merrill
for meteorologicalanalysis(NSF ATM 9013192). The authorswish to
thank K. Brulandand W. Landing for their assistance
on boardthe R/V
Cape Hatteras as well as E.A. Boyle, T. lickells, and B. Hamelin for
300
useful reviews.
400
REFERENCES
Boyle, E.A., S.D. Chapnick, G.T. Shen, and M.P. Bacon, Temporal
variability of lead in the western North Atlantic Ocean, J.
Geophys.Res.,91, 8573-8593,1986.
Chow,TJ., C.B. Snyder,andJ.L. Earl, Isotoperatiosof lead as pollutant
sourceindicators,Proc. Spec.Meet. IAEA-SM-191/4, pp. 95-108,
Int. At. EnergyComm.Vienna, 1975.
500
Church, T.M., J.M. Tramontano, J.R. Skudlark, T.D. Jickells, J.J. Tokos
and A.H. Knap, The wet depositionof trace metalsto the western
Atlantic Ocean at the mid Atlantic coast and on Bermuda, Atmos.
Environ., 18, 2657-2664, 1984.
Church,T.M., A. Viron, A.R. Flegal, C.C. Patterson,D. Setfie, Y. Erel,
600'
H.B. Mating, and J.T. Merrill, Tracemetal sources,transport,and
scavengingin the North Atlantic atmosphere:Shipboardresultsof
precipitationand aerosols,Global Biogeochem.Cycles,4, 431-443,
1990.
Church,T.M., J.M. Tramontano,D.M. Whelpdale,M.O. Andteac,J.N.
Galloway, W.C. Keene, A.H. Knap, and J. Tokos, Atmospheric
and precipitation chemistry over the North Atlantic Ocean:
Shipboardresults from April-May, 1984, J. Geophys. Res., 96,
700
Fig. 5. Stableleadisotopes
(•0•Phti07pb)
in surfacewatersat the BATS
18,705-18,725, 1991.
stationnearBermudain 1984 [Shenand Boyle,1988] (opentriangles), Coale, K.H., Labyrinthof doom: A deviceto minimize the "swimmer"
1988[Sherrellet al., 1992](opensquares)
and 1989(thisstudy)(April,
componentin sedimenttrap collections,Limnol. Oceanogr,35(6),
closedcircles;November,solidsquares).
1376-1381, !990.
Dedeurwaerder,H.L., W.F. Baeyens,and F. Dehairs, Estimatesof dry
the mixed layer reflects summertime atmospheric input and
considering 2ø6pb/2ø7pb ratios in the mixed layer
(1.178+0.001)
and in the North American
westerlies
(1.204+0.004),we calculatethatthe integrated2ø6pb/2ø7pb
signalfrom Eurafricansourcesis 1.155+0.004 in the northwest
Atlantic troposphere.This result is in good agreement with
Church et al. [ 1990] measurementsin the trade easterlies which
displayed
2ø6pb/2ø7pb
ratiosrangingfrom 1.150 to 1.160 in
August 1988. Consideringthese isotopic signaturesand the
atmosphericsectoringat Bermuda,we calculatethat the average
206pb/2O7pb
ratio in the Sargasso
Seasurfacewatersshouldbe
1.188 (+0.004).
CONCLUSIONS
We use the combination of 3 years of atmospheric
depositionrecords with surface seawatercollections in 1979
and 1989 to determinethe impact of seasonaland long-term
changesof the troposphericinput functionon lead cycling in
the northwestAtlantic. We observedthat the troposphericlead
inputto the Sargasso
Seasurfacewatershasdeclinedby a factor
of 5 to 8 since1979.This significantdecreaseis mostlydueto
and wet depositionof severaltracemetalsin the southernbight of the
North Sea, in Proceedingso.fthe 5th International Conferenceon
HeavyMetalsin theEnvironment,vol. 1, editedby Lekkas,pp. 135137,CEP Consultants,
Edinburgh1985.
Duce, R.A., G.L. Hoffman, B.J. Ray, I.S. Fletcher,G.T. Wallace, J.L.
Fashing,S.R. Piotrowicz, P.R. Walsh, EJ. Hoffman, J.M. Miller, and
J.L. Heffter, Trace metalsin the marine.atmosphere:
sourcesand
fluxes, in Marine Pollutant Transfer, edited by H. Windom 'andR.
Duce,pp. 77-119,Heath,Lexington,Mass.,1976.
Erel Y., C.C. Patterson,
M.J. ScottandJJ. Morgan,Transportof industrial
lead in snowthroughsoil to streamwater and groundwater,Chem.
Geol., 85, 383-392, 1990.
Flegal,
A.R.,T.F.Duda,andS.Niemeyer,
Highgradients
ofleadisotopic
compositionin north-eastPacific upwellingfilaments,Nature, 339,
458-460, 1989.
Galloway,J.N., andD.M. Whelpdale,Watox 1986overviewandwestern
North Atlantic Ocean S and N budgets, Global Biogeochem.
Cycles,1,261-281, 1987.
Hamelin,B., F.E.Grousset,
P.E.Biscaye,
A. Zindler,andJ.M. Prospero,
Lead isotopesin trade wind aerosolsat Barbados:The influenceof
Europeanemissionsover the North Atlantic, J. Geophys.Res., 94,
16,243-16,259, 1989.
Hamelin, B., F. Grousset,and E.R. Sholkovitz,Pb isotopesin suficial
pelagic sedimentsfrom the North Atlantic, Geochirn.Cosrnochirn.
Acta, 54, 37-47, 1990.
Hopper,J.F.,H.B. Ross,W.T. Sturges,
andL.A. Barfie,Regionalsource
discrimination
of atmospheric
aerosolsin Europeusingthe isotopic
composition
of lead,Tellus,ser.B,43, 45-60, 1991.
Jenkins,WJ., TritiumandHe3 in the Sargasso
Sea,J. Mar. Res, 38,533569, 1980.
18,276
VERONET AL.:LEAD CYCLINGINTHE SARGAS$OSEA
Jickells, T.D., T.M. Church, and W.G. Deuser, A comparison of
atmospheric
inputsanddeep-ocean
particlefluxesfor the Sargasso
Sea,GlobalBiogeochem.
Cycles,1, 117-130,1987.
Lambert, C.E., C. Jehanno,N. Silverberg, J.C. Brun-Cottan,and R.
Chesselet,Log-normaldistributionsof suspended
particlesin the
Sherrell, R.M.ß and E.A. Boyle, The trace metal composition of
suspended
particlesin the oceanicwatercolumnnearBermuda,
Earth Planet. Sci. Lett., 111, 155-174, 1992.
Sherrell,R.M., E.A. Boyle, and B. Hamelin, Isotopicequilibration
betweendissolvedand suspended
particulatelead in the Atlantic
Ocean:Evidencefrom Pb-210 and stablePb isotopes,J. Geophys.
LarnO•nocean,J. Mar. Res,39(1),77-98, 1981.
Res., 97, 11,257-11,268,1992.
rt, C.E., A. V6ron,P. Buat-M6nard,M. Heymud,F. Grousset,
and
Shirahata,H., R.W. Elias,C.C. Patterson,
and M. Koide,Chronological
W. Simpson,The role of large biogenicparticlesin the transportof
variations in concentrations and isotopic compositions of
atmosphericpollutant Pb down to North Atlantic sedimentsß
an.thropogenic
atmospheric
leadin sediments
of a remotesubalpine
Oceanol. Actaß14, 67-76ß 1991.
pond,Geochim.Cosmochim.
Acta,44, 149-162,1980.
Lasaga,A.C., The kinetic treatmentof geochemicalcyclesßGeochim.
Stunder,B.J.B., R.S. Artz, G.D. Rolph, J.M. Harris,and J.T. Merrill,
Cosmochim.Actaß44, 815-828, 1980.
S.ummaryof meteorologicalconditionsoverthe North AtlanticOcean
Mating,H., C.C. Patterson,
andD.M. Settle,Atmospheric
inputfluxesof
duringGCE/CASE/WATOX,
GlobalBiogeochem.
Cycles,
4, •33-150,
industrial
and natural Pb from the westerlies to the mid-North
1990.
Pacific,in Chemical
Oceanography,
vol. 10,editedby J.P.Rileyand
Sturges,W.T., and L.A. Barfie, Lead 206/207 isotoperatiosin the
R. Chester,pp. 84-105, Academic,SanDiego,Calif., 1989.
atn?os.
phereof NorthAmerica:
Tracersof American
andCanadian
Merrill, J.T., Atmospheric
long rangetm.nsportto the PacificOcean,in
ermssxons,
Nature, 329, 144-146, 1987.
ChemicalOceanography,
vol. 10, editedby J.P.Riley andR. Chesterß
Sturges,
W.T., andL.A. Barfie,Theuseof stablelead206/207isotope
Academic,SanDiego,Calif., 1989.
Murozomi, M., T.I. Chow, and C.C. Patterson,Chemical concentrations
of pollutant lead aerosolsßterrestrial dusts and sea salts in
Greenland and Antartic snow strataß Geochim. Cosmochim. Actaß
33, 1247-1294, 1969.
Ng, A., and C.C. Patterson,Natural concentrations
of lead in ancient
arctic and Antarctic ice, Geochim. Cosmochim. Actaß 45, 21092121ß 1981.
Nozaki, Y., J. Thomson,and C.C. Turekian,The distributionof Pb-210
and Po-210 in the surface waters of the Pacific Ocean, Earth
Planet. Sci. Lett., 32, 304-312, 1976.
Nriagu, J.O., and J.M. Pacyna, Quantitative assessment
of worldwide
contaminationof air, water and soils by trace metals,Nature, 333,
134-139, 1988.
Patterson,C.C., Global pollution measuredby lead in mid-ocean
sediments,Nature, 326, 244-245, 1987.
Patterson,C.C., and D.M. Settle, The reductionof order of magnitude
errorsin lead analysesof biologicalmaterialsand natural watersby
evaluatingand controllingthe extent and sourcesof industriallead
contaminationintroducedduring samplecollectingand analysis,in
Accuracyin TraceAnalysis:Sampling,SampleHandling,Analysis,
ratiosandelementalcomposition
to discriminate
the originsof leadin
aerosols at a rural site in eastern Canada, Atrnos. Environ., 23,
1645-1657, 1989a.
Sturges,W.T., and L.A. Barfie, Stable lead isotope ratios in Artic
aerosols:Evidencefor the origin of artic air pollution, Atmos.
Environ.,23, 2513-2519,1989b.
Trarnontano, J.M., J.R. Schudlark, and T.M. Church, A method for the
collection,handlingand analysisof trace metalsin precipitation,
Environ. Sci. Technol, 21,749-753, 1987.
V6ron, A., C.E. Lambert,A. Isley, Linet P., and F. Grousset,Evidenceof
recentlead pollutionin deepnortheastAtlanticsedimentS,
Nature,
326, 278-281, 1987.
V6ron,A., T.M. Church,and A.R. Flegal,Impactof the subtropical
North AtlanticGyre circulationon the distribution
of dissolvedPb in
the easternNorth Atlantic Ocean, EOS Trans. AGU, 72(51), Ocean
Sciences
Meetingsuppl.,44, 1991.
V6ron, A., T.M. Church, C.C. Patterson, Y. Erel, and J.T. Merrill,
Continental origin and industrial sources of trace metals in
the northwestAtlantictroposphere,
J. Atmos.Chem.,14, 339-351,
1992.
edited by P. Lafieur, Natl. Bur. Stand.Spec. Publ. 422, 321-351,
1976:
Patterson, C.C., and D.M. Setfie, Review of data on eolian fluxes of
industrialandnaturalleadto thelandsandseasin remoteregionson
'•Sglobal
scale,
Mar.Chern,
22, 137,162,
1987.
Scheme
B.K.
and C.C. Patterson
Perturbations
of the naturallead
• ß'
.•
,
. ß
.
.
.
T.M. Church, College of Marine Studies,University of Delaware,
Newark, DE 19716.
Y. Erel and C.C. Patterson,Division of Geologicaland Planetary
Sciences,
Calffo•a Instituteof Technology,Pasadena,
CA 91125.
A.R. Flegal, Institute of Marine Sciences,University of California,
p•5•iiem theSargasso
Seaby industrial
leadß
m TraceElements
tn
'•bhter, editedby C.S.Wong,E,A.Boyleß
K. Binland,
D. Burton, SantaCruz, CA 95064.
•d E.D.Goldberg,
pp.487-503,Plenum,
NewYork,1983.
A.J. V6ron, Lab. des G6osciencesde l'Environnement, UniVersit6
Setti•!
D.M.,andC.C.Patterson,
Magnitude
andsources
ofprecipitationAix-Marseille
anddry depositionfluxesof industrialandnaturalleadsto the North
Pacificat Enewetak,J. Geophys.Res.,87, 8857-8859,1982.
Shen,G.T., andE.A. Boyle,Leadandcorals:Recons•cfion of historical
1TI, CNRS 132 CASE 431, 13397 Marseille cedex 20,
France.
industrial fluxes to the ocean surface,Earth Planet. Sci. Lett., 82, 289304, 1987.
Sheri,G.T., and E.A. Boyle, Thermoclineventilationof anthropogenic
lead in the westernNorth Atlantic, J. Geophys.Res., 93, 15,71515,732, 1988.
(ReceivedMay 29, 1992;
revised March 19, 1993;
acceptedJune4, 1993.)