Old Dominion University
ODU Digital Commons
OEAS Faculty Publications
Ocean, Earth & Atmospheric Sciences
2001
Control of Phytoplankton Growth by Iron Supply
and Irradiance in the Subantarctic Southern Ocean:
Experimental Results From the SAZ Project
P. W. Boyd
A. C. Crossley
G. R. DiTullio
F. B. Griffiths
D. A. Hutchins
See next page for additional authors
Follow this and additional works at: http://digitalcommons.odu.edu/oeas_fac_pubs
Part of the Biogeochemistry Commons, Marine Biology Commons, and the Oceanography
Commons
Repository Citation
Boyd, P. W.; Crossley, A. C.; DiTullio, G. R.; Griffiths, F. B.; Hutchins, D. A.; Queguiner, B.; Sedwick, Peter N.; and Trull, T. W.,
"Control of Phytoplankton Growth by Iron Supply and Irradiance in the Subantarctic Southern Ocean: Experimental Results From the
SAZ Project" (2001). OEAS Faculty Publications. Paper 83.
http://digitalcommons.odu.edu/oeas_fac_pubs/83
Original Publication Citation
Boyd, P.W., Crossley, A.C., DiTullio, G.R., Griffiths, F.B., Hutchins, D.A., Queguiner, B., . . . Trull, T.W. (2001). Control of
phytoplankton growth by iron supply and irradiance in the subantarctic Southern Ocean: Experimental results from the SAZ project.
Journal of Geophysical Research-Oceans, 106(C12), 31573-31583. doi: 10.1029/2000JC000348
This Article is brought to you for free and open access by the Ocean, Earth & Atmospheric Sciences at ODU Digital Commons. It has been accepted for
inclusion in OEAS Faculty Publications by an authorized administrator of ODU Digital Commons. For more information, please contact
[email protected].
Authors
P. W. Boyd, A. C. Crossley, G. R. DiTullio, F. B. Griffiths, D. A. Hutchins, B. Queguiner, Peter N. Sedwick, and
T. W. Trull
This article is available at ODU Digital Commons: http://digitalcommons.odu.edu/oeas_fac_pubs/83
JOURNAL
OF GEOPHYSICAL
RESEARCH,
VOL. 106, NO. C12, PAGES 31,573-31,583, DECEMBER
2001
Control of phytoplankton growth by iron supply and irradiance
in the subantarctic
Southern
Ocean:
Experimental results from the SAZ Project
P. W. Boyd,• A. C. Crossley,
2 G. R. DiTullio, 3 F. B. Griffiths,4 D. A. Hutchins?
B. Queguiner,6 P. N. Sedwick,? and T. W. Trull ?
Abstract. The influenceof irradianceand iron (Fe) supplyon phytoplanktonprocesses
wasinvestigated,north (47øS,142øE)and south(54øS,142øE)of the SubantarcticFront in
australautumn(March 1998).At both sites,residentcellsexhibitednutrient stress(Fv/
F m < 0.3). Shipboardperturbationexperimentsexaminedtwo light (mean in situ and
elevated)and two Fe (nominally0.5 and 3 nM) treatmentsunder silicicacid-replete
conditions.Mean in situ light levels(derived from incidentirradiances,mixed layer depths
(MLDs), wind stress,and a publishedverticalmixingmodel) differed at the two sites,25%
of incidentirradianceI 0 at 47øSand 9% I 0 at 54øSbecauseof MLDs of 40 (47øS)and
90 rn (54øS),when thesestationswere occupied.The greaterMLD at 54øSis reflectedby
tenfold higher cellular chlorophylla levelsin the residentphytoplankton.In the 47øS
experiment,
chlorophyll
a levelsincreased
to > 1 /•g L-• onlyin thehigh-Fetreatments,
regardlessof irradiance levels,suggestingFe limitation. This trend was also noted for cell
abundances,silicaproduction,and carbon fixation rates. In contrast,in the 54øS
experiment
therewereincreases
in chlorophyll
a (to >2/•g L-•), cellabundances,
silica
production,and carbonfixation only in the high-lighttreatmentsto which Fe had been
added, suggestingthat Fe and irradiance limit algal growth rates. Irradiance by altering
algal Fe quotasis a key determinantof algal growthrate at 54øS(when silicicacid levels
are nonlimiting);however,becauseof the integralnature of Fe/light colimitationand the
restrictednature of the current data set, it was not possibleto ascertainthe relative
contributionsof Fe and irradianceto the control of phytoplanktongrowth. On the basisof
a climatologyof summermean MLD for subantarctic(SA) waterssouthof Australia the
47ø and 54øSsitesappear to represent minimum and maximum MLDs, where Fe and Fe/
irradiance,respectively,may limit/colimit algal growth.The implicationsfor changesin the
factorslimiting algal growthwith seasonin SA waters are discussed.
1.
significant
increases
in chlorophyll
(i.e.,>1.5 p,gL -1) in con-
Introduction
The role of iron (Fe) in controllingphytoplanktongrowth
rate has been demonstratedin the high-nitrate, low-chloro-
phyll (HNLC) regionsof the equatorialPacific[Kolberet al.,
1994; Coale et al., 1996a, 1996b] and NE subarctic Pacific
Ocean [Martin et al., 1989;Coale,1991;Boydet al., 1996] from
in situ and shipboard Fe enrichments,respectively.In the
Southern Ocean, shipboardFe enrichmentshave also exhibited Fe-elevatedalgal biomass,but in many studiesthere were
•Centrefor ChemicalandPhysical
Oceanography,
NationalInstitute of Water and AtmosphericResearch,Department of Chemistry,
University,of Otago, Dunedin, New Zealand.
2Instituteof AntarcticandSouthernOceanStudies,Universityof
Tasmania, Hobart, Tasmania, Australia.
3GriceMarine Laboratory,Universityof Charleston,
Charleston,
South Carolina, USA.
4CSIRODivisionof Marine Research,Hobart,Tasmania,Australia.
5College
of MarineStudies,
University
of Delaware,Lewes,Delaware, USA.
6Laboratoired'Oceanographieet de Biogeochimie,CenTre
d'Oceanologiede Marseille, UMR CNRS 6535, Marseille, France.
7AntarcticCRC, Hobart, Tasmania,Australia.
Copyright2001 by the American GeophysicalUnion.
Paper number 2000JC000348.
0148-0227/01/2000J C000348509.00
trol treatments[seede Baar and Boyd,2000]. Indirect evidence
of the role of iron supplywasprovidedby de Baar et al. [1995],
who demonstrateda positiverelationshipbetween the magnitude of in situ phytoplanktonstocksand ambient Fe levels in
the vicinity of the Polar Front (PF) in the South Atlantic.
Furthermore, Timmermanset al. [1998] and Boyd et al. [1999]
have presentedphysiologicalevidenceof Fe-stressedresident
phytoplanktonin the Atlantic PF region and in subantarctic
(SA) waters,respectively.Recently,unequivocalevidencefor
the role of Fe in controllingalgal growthrates in the Southern
Ocean was provided by the Southern Ocean Iron Release
Experiment (SOIREE) in situ mesoscaleiron fertilization
[Boydet al., 2000].
Despite this confirmationof the important role of Fe in the
open SouthernOcean [Boydet al., 2000] the existenceof deep
mixed layers in this region [Mitchellet al., 1991; Nelson and
Smith, 1991] and the antagonisticrelationshipbetween algal
light limitation and Fe limitation [Raven, 1990; Sunda and
Huntsman,1997] meansthat light limitation or Fe/iight colimitation may control phytoplanktongrowth at times and in various regions in Southern Ocean waters. Raven [1990] used
predicted cellular Fe contentsand Fe acquisitionrates of cultured phytoplanktonto estimate that the Fe requirementsof
cells growingunder light limitation were 50-fold higher than
31,573
31,574
BOYD ET AL.: SUBPOLAR PHYTOPLANKTON,
IRON, AND LIGHT
Table 1. A Comparisonof the Main Chemical,Physical,and BiologicalPropertiesUnder
Ambient Conditions(SampledFrom 20 m Depth) at the 47ø and 54øSSitesa
Parameter
Units
47øS
54øS
Mixed layer depth
m
40
90
Attenuance,
Kd
DailyincidentPAR
m-•
mol quantam-2 d- •
0.055
24
0.041
19
Mean
24
knots
18
Temperature
wind stress
C
11
4
Chla
Nitrate
Silicicacid
/xgL- •
/xmolL- •
•mol L -•
0.2
>8
0.9
0.15
>25
2.5
DFe
nmol L- •
0.07
0.07b
Cyanobacterial
abundance
x 103cellsmL- •
Fv/Frn
dimensionless
CellularChl a
C fixation
Pmax
Pbmax
Biogenicsilica
Silicaproduction
Fe uptake
Fe:C uptake ratio
C:Si uptake ratio
30
1
0.3
0.25
fg cell-•
/xmolC L -• d-•
•mol C L- • h- •
/xgC/xg Chla - • h- •
•mol L -•
nmolSi L- • d- •
fmolFe mL-• h- •
6.6
0.84
0.18
9.2
0.07
18.0
2.5
50.1
0.34
0.04
2.1
0.53
9.0
2.3
pmol:/xmol
Mol:mol
52.0
46.7
78.0
37.8
apAR, photosynthetically
activeradiation;Chl a, chlorophylla.
bWatersampled
from45 m.
under light-saturatedgrowth.These estimateswere confirmed
for algal lab culturesby Sunda and Huntsman [1997], who
demonstratedthat in addition to irradiance, algal cell size
playeda role in determiningcellularFe requirements.Maldonado et al. [1999] demonstratedFe/light colimitationof algal
growth duringwinter deckboardexperimentsin the NE subarctic Pacific. Investigatorshave jointly consideredthe influence of Fe/light on phytoplankton processesin Southern
Oceanwatersbut report conflictingresults.Deckboardstudies
[vanLeeuwe,1996;Boydet al., 1999, 2000] observedFe/light
colimitation,whereasTakeda [1998] reported no significant
differencein iron-elevatedgrowth rate when cellswere incubatedunderhigh or low irradiances(seesection4.3).
The presentstudytook place in SA waterssouthof Australia, a region characterized by subnanomolardissolvedFe
(DFe) [Sedwick
et al., 1997],low silicicacidlevels[Zentaraand
Kamykowski,1981], and deep surfacemixedlayers[Rintoulet
al., 1997]. The objectiveof this studywas to determinewhat
factor(s)controlsalgal growthrates in SA waters.However,
becauseof the potentiallyconfoundinginfluencesof concurrent Fe and light/silicicacid limitation on algal processesin
australautumn,two setsof complementaryshipboardperturbation experimentswere performedin which either silicicacid/Fe or Fe/irradiancelevelswasaltered.Thispaperreportson
the Fe/irradianceexperiments,while resultsfrom the silicic
acid/Fe experimentsare summarizedby Sedwicket al. [1999]
and presentedin detail by Hutchinset al. [thisissue].
boyswere rinsedand then filled with seawater.Sampleswere
takenfrom thissupplyfor time zero measurements.
Five treatments were prepared: treatments 1-3 were incubatedunder
meanin situirradiance:unamendedseawater(control),low-Fe
enrichment(low Fe, low light), high-Feenrichment(high Fe,
low light);while treatments4-5 were incubatedundergreater
thanmeanin situirradiance,low-Feenrichment(low Fe, high
light), andhigh-Feenrichment(highFe, highlight).
Low- (0.5 nM Fe) andhigh-Fe(3 nM Fe) enrichments
were
addedasFeC12chelatedwith ethylenediaminetetra aceticacid
(EDTA) in a 1:1.5ratio [Coale,1991].In the "low-light"treatment,cellswere exposedto a lightlevelcorresponding
to the in
situmeanirradiancereceiveddailyastheyareverticallymixedin
theupperocean.In the"high-light"
treatment,cellswereexposed
to a higherthan meanin situirradiance.Mean in situirradiance
levelswereestimatedusingshipboard
dataon mixedlayerdepth
(MLD), meanwind stress,water columnlight attenuance,and
incidentirradianceTo,in conjunction
with equationsfor the verticaldisplacement
of phytoplankton
byturbulentmixing[Denman
and Gargett,1983].At the relativelyhighwind stresses
recorded
(Table 1), it was assumedthat cellscirculatedthroughoutthe
surfacemixedlayer,ratherthancirculating
withineddiesof length
scaleslessthan the MLD [seeMacTntyre,
1998].
The estimated
in situirradiances,
expressed
asa percentage
of
dailyincidentirradianceTo,for the low- (and high-)light treatmentsin the47øand54øSexperiments
were25%To (50%To)and
9% To(25% To),respectively.
Carboyswere enclosedin neutral
densityscreening
equivalentto the calculated% To.A single25 L
2.
Methods
carboywasusedper treatment(asopposedto multiple4 L botartifactssuchaswalleffects[Berget
Fe/irradianceperturbationexperimentsof 7 and 10 day du- tles)to minimizecontainment
reration were performed on board RSV Aurora Australisin the al., 1999]duringthe long-term(up to 10 days)incubations
waters.Thusthereare no true replicatetreatvicinityof 47øand54øS(142øE),respectively
(seemap) [Rintoul quiredin subpolar
for subsamples
and Trull, this issue].Water was sampledfrom 20 m depth ments. However, there are pseudoreplicates
•4C,and32Si),
andthetrends
observed
fromindependent
usinga "clean"pump with the subsurfaceintake flushedfor 1 (55Fe,
(chlorophyll
a, •4C,and32Si)maybecompared.
hour prior to sampling[Hutchinsand Bruland,1998].Pumped techniques
In March 1998, surface ambient silicic acid levels were low,
seawaterwascollectedwithin a cleanenvironment(submicron
filtered air, laminar flow), and 25L clean polycarbonatecar- in the range of concentrations
known to limit diatom growth
BOYD ET AL.: SUBPOLAR PHYTOPLANKTON,
(<1 /xM (47øS)and 2.4/xM (54øS))[Paasche,1973] and may
have confoundedthe studyof Fe/irradianceon algal growth.
Therefore silicic acid (chelex-100purified) [Sedwicket al.,
1999] was added to all carboys(final concentration3.5 /xM,
comparableto winterreserve[Boydet al., 1999]).After spiking
with Fe/silicicacidthe neckand capof eachcarboywassealed
with parafilm. Carboyswere incubatedon deck in pumped
surfaceseawater,and incubatortemperatureswere monitored.
Carboyswere subsampled
(class100 clean air conditions)at
noonon t = 2.25, 5.25, 7.25 days(47øS)andt = 2.25, 6.25,
and 10.25days(54øS).
Seawater subsampleswere analyzed using active fluorescence (fast repetition rate fluorometry(t = 0 only [after
Kolberand Falkowski,1993]), size-fractionatedchlorophylla
(>20, 5-20, 2-5, and 0.2-2/.rm fractions)followingJointand
?oreroy [1983], flow cytometry(see below), biogenic silica
[Qu•guiner,this issue],and DFe [Sedwicket al., 1997]. Phosphate, silicicacid, and nitrate + nitrite levelswere analyzed
usingan Alpkem Flow SolutionAnalyser.Rate measurements
IRON, AND LIGHT
31,575
presenceof a transientmixed layer, chlorophylla was 0.2
L-1 from0 to 75 m (seesection
4.1,Figurelc). Thecells<2
/xm that dominatedchlorophylla were cyanobacteriasuchas
Synechococcus
(Table 1). The algalcommunityhad suboptimal
valuesof F•,/Fm (Table 1), suggesting
impairedphotosynthetic
competencedue to nutrient stress.Estimatedcellular chloro-
phylla was6.6, 3.8, and 20 fg cell-1 for cells>0.2, 0.2-2
(mainumprokaryotes),and >2/•m (eukaryotes),respectively.
Southof the SA Front at 54øS,temperatureswere 4.4øCin a
90 m MLD. Surfacenitratelevelswere>25 and2.5/xmolL-1
for silicicacid (Table 1). DFe levelswere 0.07 and 0.11 nmol
L-1 at 45and75m depth,respectively.
Chlorophyll
a was0.15
/xgL-1, witha subsurface
chlorophyll
a maximum
(SCM)just
belowthe mixedlayer (Figure ld). Cells <2 •m (cyanobacteria) and >20 •m (diatoms) dominatedchlorophylla in the
upper 90 m, while diatomswere dominantin the SCM. F v/Fm
in surfacewaters (0.25) was indicativeof a resource-limited
algalassemblage
(Table 1). Cellular chlorophylla were 50, 62,
and46 fg cell-• for cells>0.2, 0.2-2,and>2/.rm,respectively.
included•4Cuptake[Jointetal., 1993],55Feuptake[Schmidt Note, thiscellularchlorophylla contentfor picophytoplankton
and Hutchins, 1999], algal silica productionusing 32Si corresponds
to around6% of their dry matter. Despite similar
[Brzezinskiand Phillips, 1997], and photosynthesis:irradi- chlorophylla levelsbetween sites,biogenicsilica levelswere
higherat the 54øSsite(Table 1). Ratesof carbon
ance (?:I) characteristics[Mackeye! al., 1995]. Subsamples considerably
for rate measurements
were incubated for 24 hours on deck
fixationand silicaproductionwere higherat 47øScomparedto
at the same % I 0 as used for the corresponding carboy 54øS,whereasFe uptake was similar at both sites.
(except for the 1 hour ?:I incubations).
Cell abundanceswere analyzedusing a Becton Dickinson
FACScanflow cytometerwith a 15 mW argon ion laser. Flow
rate and size calibrationwere determinedusing Fluoresbrite
beadsand lab-culturedcellssuchas Synechococcus
[Olsonet
al., 1993].Two-dimensionalscatterplotsof FL2 (orange fluorescence)and FL3 (red fluorescence)providedestimatesof
the medium/largestronglyred autofluorescent
cells(e.g., diatoms) and cyanobacteria[Wrightet al., 1991, 1996; Crossley,
1998]. Although flow cytometrydescribesfluorescenceintensityrather than direct measurementof cell size,sizecalibration
experimentsindicatethat 2 •m diameteris the cutoff between
small and medium/largecells.These categorieswere used,in
conjunctionwith size-fractionatedchlorophylla to estimate
cellular chlorophylla for prokaryotesand eukaryotes.
Algal net growth rates were estimated using the instantaneouschangein chlorophylla or in cell abundance.Estimates
of diatomnet growthrateswere obtainedfrom silicaturnover
(silica production/biogenic
silica).Two general experimental
artefactsmust be noted. First, DFe levelsfrom a vertical profile at 54øS[Sedwicke! al., 1999]were considerablylower than
thosein carboysat 54øS,suggesting
inadvertentFe contamination of carboyseawater.Thus, for 54øSthere was no "1ow-Fe"
treatment, and this experimentmust be viewed as an investigation of the effectsof alteringirradiancelevelson (initially
Fe-stressed)cellsunder Fe-repleteconditions.Second,there
was no high-lightcontrol carboyat either site; this may have
implicationsat the 54øSsite (see section4.2).
3.
3.1.
Results
Physical, Chemical, and Biological Properties
The 47•Ssite at the time of samplinghad a seasonalMLD of
80 m with evidenceof a transientmixedlayer (0-40 m, Figure
la). Surfacetemperaturewas 11.5øC,and noon incidentirra-
3.2.
The 47øSPerturbation Experiment
The high-Fe treatments exhibited the greatest increase in
chlorophylla levels at 47øSwith little change in the other
treatmentsover 7 days(Figure 2a). This trend was supported
by flow cytometric and rate measurements.Cyanobacterial
abundancesdoubledin the high-Fe treatmentsby day 4 (Figure 2c), and diatomabundances
increased>sixfold over 7 days
(Figure2e). Nitratelevelsdecreased
by 2.5-4.5 •mol L-• in
the high-Fe treatmentsbut changedlittle in the others (data
not shown).DFe levelsin the controlwere 0.3 nmol L -•,
suggestingminor contamination, whereas the low-Fe and
high-Fetreatmentscontained0.7 and2.7 nmolL -• DFe on
day 2, respectively(Table 2).
While chlorophylla levels increasedover 7 days for both
high-Fe treatments,rates of carbon fixation leveled on day 4
after initially quintupling(Figure 3a). Again, there was little
changein production rates in the other treatments. Cells >5
•m were responsiblefor 40% of productionin both high-Fe
treatments(data not shown).The initial uptake of Fe was
greatestin both high-Fe treatments,but the temporal trends
were lessclear (Figure 3c) than for carbonfixation or silica
production. Fe:C uptake ratios generally decreasedseveral
fold after time zero in all treatmentsfor all cells (data not
shown).Silicaproductionrateswere pronouncedonly in the
high-Fe treatments,increasingslightlybetween days0 and 4,
then markedlyafter 4 days(Figure 3e). This uptake pattern
differs from that for C fixation.
Fe enrichment(high-Fe treatments)resultedin a doubling
of cellularchlorophylla (>0.2 •m) after 7 days,whereaslevels
changedlittle for the control/low-Fetreatments(Figure 4a).
Cellular chlorophylla (0.2-2 •m) doubledin all treatments,
while levelsremained constantor declinedfor >2 •m cells in
carboysover 7 days. There were fourfold increasesin the
light-saturatedphotosynthesisrate (Pmax) in both high-Fe
diances
were350 •mol quantam-2 s-• (Figurelb). Surface treatmentsover 7 days(not shown);however,trendsin Pbmax
nitrate+ nitritelevelswere>8 •mol L-•, silicicacidwas<1 (normalizedto chlorophylla) were lessclear (Figure 4c).
•mol L -•, andDFewas<0.1 nmolL -• (Table1). Despitethe max(normalizedto algalabundance)exhibitedsimilartrendsto
31,576
BOYD ET AL.: SUBPOLAR PHYTOPLANKTON,
26.4
26.6
27.0
26.8
27.2
IRON, AND LIGHT
100
0
200
.......
ß
300
ß•.•.
400
-. -. -, • •
ß
B
A
ß
2O
ß
500
.
ß
ß
4O
ß
ß
ß
ß
ß
6O
ß
ß
ß
ß
ß
ß
ß
8O
ß
lOO
ß
ß
12o
14o
16o
0.0
i
i
0.1
0.2
0.3
0.4
0.5
0.0
0.1
i
c
2O
40
6O
60
8O
8(
100
120
total
> 20 •m
5-20 •m
0.3
0.4
i
i
0.5
D
20
4O
0.2
10(
12(
2-5 •m
140
160
0.2-2 •m
14(
16(
....
Figure 1. Vertical profilesof (a) sigmatheta at the 47øS(March 9, 1998,solidline) and54øSsites(March
18, 1998, dotted line); (b) photosynthetically
active radiation (PAR) at the 47ø and 54ø sites;and the
partitioningof chlorophylla into 4 sizefractionsat (c) 47øand (d) 54øS.No replicates.PAR wasprofiledwith
a BiosphericalInstrumentsPRR-600 reflectanceradiometerat local noon,with the vesselorientatedtoward
the Sun to minimize ship shadow.
Pmaxfor both high Fe treatments(Figure4e). Photoinhibition high-Fe/high-lighttreatment.As in the 47øSexperiment,cells
was not observedduring the 47øSexperiment.
>5 t•m were responsiblefor 40% of carbon fixation in the
Algal (net) growthrates(from chlorophylla) were 0.7 dou- high-Fe/high-light
carboy(data not shown).
blingsd- • by days2 and4 in thehigh-Fe/high-light
andhighInitial ratesof Fe uptakeweresimilarat bothsites(Table 1);
Fe/low-lighttreatments,respectively,and decreasedthereafter however,uptake patternsdiffered betweensites(Figures3c
(Table 3a). Net growthrateswere generallylow for the other and 3d). Both high-Fetreatmentsexhibitedthe highestinitial
treatments.In contrast,net growthrates(from cell abundance) increasein Fe uptake at 54øS.Despite higher biogenicsilica
were ---0.2-0.3doublings
d-• after48 hoursfor the high-Fe levels at the 54øSsite, silica productionrates were initially
treatmentsand lower for the others(Table 3b).
higherat 47øSthan at 54øS(Table 1) and exhibitedthe highest
increasesin the high-lighttreatments(Figure 3f). Cellular
3.3. The 54øSPerturbation Experiment
chlorophylla quadrupledin both high-lighttreatmentsby day
In contrastto the 47øSexperiment,chlorophylla levelsin- 10 at 54øS(Figure4b). Cellularchlorophylla declinedin cells
creasedmarkedlyonlyin both high-lighttreatments(either > 1 <2 t•m over 10 daysbut increasedin cells>2 t•m, particularly
or 3 nmolL- • Fe wasadded,Table2), andthelagpriorto an for the high-lighttreatments(data not shown).
increasein chlorophylla was4 dayslonger(Figure 2b). Total
In contrastto 47øS,Pmaxat 54øSwasinitially fivefoldlower
cell abundanceincreasedfourfoldin bothhigh-lighttreatments (Table 1). As for other measurements
at 54øS,Pmaxincreased
becauseof fivefoldand twofoldincreasesin cyanobacterialand onlyin thehigh-lighttreatments(datanot shown).Pbmax
(chlodiatom abundances,respectively(Figures 2d and 2f). This rophylla) exhibitedinitial increases
in all carboys,followedby
trend was supportedby other measurements,with rates of a progressive
decrease,with little differencein the magnitude
carbonfixationincreasingmarkedlyonly in the high-lightcar- of Pbmaxbetweentreatments(Figure 4d). P•maxincreased
boys (Figure 3b). Initial productionrateswere half those at greaterthan fivefoldonly in both high-lighttreatmentsat 54øS
47øSbut were comparablein magnitudeby day 10 in the (Figure4f). Therewasphotoinhibition
in incubatedsamplesat
BOYD ET AL.: SUBPOLAR PHYTOPLANKTON,
IRON, AND LIGHT
31,577
1.4
--e--
1.2
A
Control
--D- lowFe lowLight
•
/
--•. - h•ghFe low Light
--•-- lowFe highLight
•
highFe h•ghL•ght
I
I
I
/
I
I
I
I
I
0.6
I
I
I
0.4
I
I
0.2
0.0
6
8O
C
m
D
5
60
m
0
•
4
3
40
•
2
>,• •- 20
1
o
0
4o
10
.
F
8
6
4
2
0
0
2
4
6
8
0
2
4
6
8
10
12
Days
Figure 2. Changesin (a) chlorophylla levels(>0.2 txm), (c) cyanobacterial
abundances,
and (e) diatom
abundances
with time during the Fe/irradianceperturbationexperimentsat the 47ø site. Changesin these
respectivepropertiesat the 54øSsiteare presentedin Figures2b, 2d, and 2f. Time zero for experiments
was
March 9, 1998(47øS),andMarch 18, 1998(54øS).Carboyswerealsosampledat 1400hourson March 11, 14,
and 16 (47øS),and1400hourson March20, 24, and28 (54øS).Diatomabundances
wereobtainedfromcounts
of medium/largesinglered fluorescentcells(SRFCs), which are mainlycharacterizedby diatoms.
54øS;
onday2,Pbmax
wasdepressed
at >200txmolquantam
-2
s-• for all treatments.
Growthrates (chl a specific,net) increasedprimarilyin the
high-light
treatments,
attaining
0.5-0.8doublings
d- • byday6
and>1 doublings
d-• after10days(Table3a).In contrast,
net
specific
growth
rates(cellabundance)
were---0.2doublings
d-•
in the high-lighttreatmentsbetweendays6 and 10 (Table 3b).
Diatomgrowthratesweregenerally
0.1 doublings
d- • or less
Table 2. DissolvedFe Levelsat t - 0 (From 20 m Depth) and in the ExperimentalCarboysDuring the 47ø and 54øS
Experimentsa
47øS
Treatment
Control
Low-Fe/low-light
High-Fe/low-light
Low-Fe/high-light
High-Fe/high-light
t = 0
t = 2
0.07
54øS
t = 4
t = 7
t = 0
0.30
0.35
0.30
0.97'
0.63
2.70
0.68
2.81
0.64
2.02
0.45
2.02
0.51
1.88
0.96
1.88
t = 2
t = 6
t =
10
1.56
1.02
0.72
0.72
3.14
1.39
3.77
0.46
2.15
0.57
1.35
0.40
1.93
0.45
0.67
aFelevelsare in nmolL-•. An asterisk
denotes
(initial)Fe contamination
of the carboy.
Notethesearethe maximum
DFe levelsfor each
carboy,asthe DFe analysis
wasperformedon subsamples
takenfrom eachcarboy.Coale[1991]hasreportedcontamination
occurring
during
suchtransferprocedures.
31,578
BOYD ET AL.' SUBPOLAR PHYTOPLANKTON, IRON, AND LIGHT
B
A
c
ß E
2
0
8OO
8OO
=
6OO
ß• ,•
--•---•---.-•
Control
F
low Fe low light
high Fe low light
low Fe highlight
high Fe highlight
•
17
6OO
ii
400
ii
4OO
ii
/
/
•
._oE 200
111
/
o
2OO
/
/
/*
0
8
0
2
4
6
8
10
12
DAYS
Figure 3. Changesin community
ratesof (a) carbonfixation,(c) Fe uptake,and(e) silicaproductionduring
the Fe/irradianceperturbationexperiments
at 47ø.Changesin theserespectivepropertiesat the 54øSsite
are presentedin Figures3b, 3d, and 3f. Variationbetweenpseudoreplicates
(in all panels)was<10% in
all cases.
for all treatments
but increased
to 0.3 doublings
d-• in the
high-lighttreatmentsby day 10 (data not shown).
terial abundance in waters between Australia
and Antarctica.
The reducedtemperaturesat 54øSwereprobablyalsoresponsible,in part, for the reducedphysiological
rates(suchascarbonfixation)recordedat thissite,relativeto the 47øSsite[see
4.
Discussion
Banse,1991].
4.1. Comparison Between Conditions at the 47ø and 54øS
There are no strong latitudinal trends in summer mean
Sites
MLD in SA waters(Plate 1). Thusthe MLDs at 47ø and 54øS
SA waters comprisearound half of the areal extent of ice- whenthe stationswere occupiedmaybe due to localforcing.
free watersof the SouthernOcean[Boydet al., 1999].The 47ø Indeed,at 47øS,similarchlorophylla levelsfrom 0 to 75 m and
and 54øS sites were located in the water masses north and
differencesin both temperatureand salinitybetweenthe 0-40
south of the SA Front, respectively[Rintouland Trull, this m and40-80 m depthhorizons(datanot shown)areindicative
issue]andthereforeprovidesnapshots
of the propertiesof the of horizontal advection of a less dense water mass into this
differentwater massesin the subpolarSouthernOcean.The region.While this feature at 47øSmay have been transient,
it wascharacterized
by cellsgeneticallyadapted
main physicaldifferencesbetweensiteswere water tempera- nevertheless,
at 54øS,
ture and MLD (Table 1). The 7øC decreasein temperature to a higherlightregimecomparedwith phytoplankton
betweensiteswaslikely responsible
for the thirtyfolddecrease whichhad ninefoldhighercellularchlorophylla (higherfor
in cyanobacterial
abundances
at 54øS.Marchantet al. [1987] both size fractions),indicativeof increasedlight-harvesting
reportedtemperatureto be the maindeterminantof cyanobac- requirementsfor the residentcellsat 54øS.Thus the 47øSsite
BOYD ET AL.' SUBPOLAR PHYTOPLANKTON,
31,579
' 5O
2O
A
B
--e--control
low Fe low light
•
--•--
hi Fe low light
00
--•' - lowFe highlight
•,
•o
IRON, AND LIGHT
h•ghFehighlight
/
5O
12
/
/
/
00
/
•
5O
0
18
D
c
.=
15
•'
12
9
•0
6
E
o.
•
,o
3
0.09
0.08
0.07
0.06
0.05
/
0.04
/
/
0.03
0.02
0.01
0.00
0
2
4
,
i
6
8
0
2
4
6
8
10
12
DAYS
Figure 4. Changesin cellularchlorophyll
a of the algalcommunityat (a) 47øand (b) 54øSsites.Changesin the
light-saturated
photosynthetic
rate of the phytoplankton
assemblage
with time duringthe Fe/irradiance
perturbation experiments
for Pbmax
(normalizedto chlorophyll)at the (c) 47øand (d) 54% siteandfor P(,max
(normalized
to cellabundance)
at the (e) 47øSand (f) 54øSsite.Note, differentscalesare usedfor P•maxbecauseof differences
in cell abundanceat eachsite.The standarderror of the estimationof Pbmax
P(,max
wasgenerally<5%.
provided a contrasting"inoculum" of cells acclimated to a
different light environmentthan were cellsfrom the 54øSsite
for the Fe/light experiments.
In March 1998eachsitewasprobablycharacterizedby highnitrate, low-silicicacid, low-chlorophyll(HNLSLC) [Dugdale
et al., 1995] conditions,as predictedfor SA waterslate in the
growth season[Boydet al., 1999]. ObservedDFe levels at 47
and 54øSwere low and comparableto those previouslyreportedin this region[Sedwicket al., 1997]and in SA watersSE
of New Zealand [Boydet al., 1999]. There was evidenceof
nutrient-stressed
residentcellsat both sites(Fv/F m -- 0.25),
tenfoldlessthanthethreshold
forFestress
(1.0nmolFekg-•), as
indicatedby flavodoxinexpression,
in SA waters[seeBoydet al.,
1999,Figure5d]. Thusthereis evidenceof silicicacid/Festressof
the residentcellsat the 47øSsite and of both Fe stressand light
limitation(enhancedlight-harvesting
abilities,lowerratesof carbon fixationandPmax)for residentcellsat the 54øSsite.
4.2.
Comparison of the Fe/Light Perturbation Experiments
There
were
clear
differences
in the outcomes
of the two
experiments,
withonlyhigh-Fesupply(>2 nmolL-2_,regard-
less of mean light levels) mediating increasesboth in algal
stocksand in the magnitude of rate processesin the 47øS
levels[seeLippemeieret al., 1999].Hutchin et al. [this issue] experiment.Significantincreasesin algal stocks/rateprocesses
report evidence from deckboard experimentsof iron/silicic were observedonly in the high-light treatments in the 54øS
acid colimitationof algal growthrates of the residentcells at experiment;noting that, becauseof contamination,all treat47øS.As no data are availablefrom Fe/Si deckboardpertur- mentswere probablyFe-replete at 54øS.On the basisof the
bationsor on the kineticsof silicaproductionat 54øS,it is not observedphysiologicalpropertiesof the resident cells at the
knownwhethersilicicacid supplylimited algalgrowthrates at two sites(see Table 1), the mean in situ light levelsat the two
the 54øSsite. However, it is likely that the residentcellswere siteswaslikely the key determinantof the different experimenFe-stressedat both sitesgiven that ambient DFe levelswere tal outcomes.
which was due to either low-Fe levels and/or low-silicic acid
31,580
BOYD ET AL.: SUBPOLAR PHYTOPLANKTON,
IRON, AND LIGHT
Table3a. Phytoplankton
GrowthRates(Doublings
d-•) Estimated
FromInstantaneous
Changein Chlorophyll
a Levels
(Net Growth)a
47øS
Treatment
t = 2
t = 4
54øS
t = 7
Control
0.13
0.02
0.12
Low-Fe/low-light
High-Fe/low-light
Low-Fe/high-light
High-Fe/high-light
0.19
0.22
0.67
0.63
0.29
0.79
-0.15
0.30
-0.07
0.22
0.16
0.22
t = 2
t =
6
t =
10
-0.16
0.11
0.36
-0.19
-0.18
-0.10
-0.18
0.24
0.29
0.51
0.83
0.32
0.09
1.13
1.04
aFor assumptions,see section2.
At 47øS,Fe-mediated changesin chlorophylla and other in the SouthernOcean [seede Baar and Boyd,2000] the addipropertiesappearedto be independentof meanlight levelsin tion of Fe to carboysresultsin elevatedchlorophylla levels.
the high- or low-light treatments.If the residentpopulation This wasnot the casein the treatments(1 and 3 nM Fe, final
wasFe-limited
at 47øS(ambient
DFewas<0.1nmolL-•), it is concentration)incubatedunder conditionsof mean in situ
puzzlingthat there were only slightincreasesin chlorophylla irradiances.In order to resolvethe observedlack of a response
or cyanobacterial
abundance
in thecontrol(0.3nmolL-• Fe) by Fe-stressedphytoplanktonto elevatedFe supplythe effects
and/orin the low-Fe(0.5nmolL -• Fe) treatments.
It should of low light levelsmustbe considered.Raven[1990]estimated
be noted that theseDFe levelsin the carboysare the maximum that the Fe requirementsof lab-culturedcells growingunder
possiblelevels(see Table 2 legend).Moreover,while small light limitationwere fiftyfoldhigherthan under light-saturated
eukaryoteshave been shownto have lower algal Fe require- growth. In the present experimentsa 3 nM Fe enrichment
ments than large cells [Sundaand Huntsman,1997], there is representsa greater than thirtyfold increasein DFe levels
evidenceof high Fe requirementsfor cyanobacteriafrom lab- relative to ambient levels.Thus, at the mean light levelssimculturedstudies[Brand,1991],althoughfield data are lacking. ulatinga 90 m MLD (low-lightFe-repletetreatments)suchFe
Raven[1990]and, more recently,BehrenfeldandKolber[1999] enrichmentwill likelybe insufficientto alleviatealgalFe stress
reportthat thismaybe dueto highPSI:PSII ratiosin cyanobac- (this calculationis problematicdue to uncertaintiesaboutthe
teria.
nature of bioavailableFe [Wellset al., 1995]).
On the basisof experimentswith algal cultures,Sundaand
In contrast,in the high-light Fe-replete treatments there
Huntsman [1997] report that iron limitation and light limitawere large increasesin cellular chlorophylla. Such a trend
tion of algal growth are integrallylinked. At the 54øSsite, in
suggestsalleviation of Fe stress[Greeneet al., 1991] since
the absenceof a high-lightcontroltreatmentandwith evidence
cellular chlorophylla contentincreasedeven thoughcellsreof Fe contaminationin all carboysis it possibleto commenton
ceivedhigherthan mean in situ irradiances.The latter should
the Fe statusof the residentcells?Phytoplanktonunder amlead
to decreasedlight-harvestingrequirementsand would
bient conditionsat 54øSdisplayedevidenceof nutrient stress
tend
to
decreasecellularchlorophylla [Van Leeuwe,1996,and
(Fv/F m = 0.25) consistentwith low-Fe levels.Furthermore,
references
therein].These findingspoint to the simultaneous
the expressionof the molecularmarker for algal Fe stress,
limitation
of
algal growth at the 54øSsite by light climate/Fe
flavodoxin[LaRocheet al., 1996],washighestalongthe 142øE
meridian at 54øSwhere it was tenfold greater than was ob- supply.Low mean irradiances,becauseof the deeper MLD,
elevatedthe algalFe requirementsof the resident
servedin the SubtropicalFrontal Zone (J. LaRoche,personal considerably
cells
in
these
low Fe waters and hence, by setting algal Fe
communication,2000). Thus there is evidence of an Fequotas
in
this
HNLSLC region, were a key determinantof
stressedalgalcommunityat 54øS.DespitethisFe stress,which
shouldresult in depressedcellular chlorophylla content[see phytoplanktongrowthat 54øS(under silicicacid-repleteconGreeneet al., 1991;Vassilievet al., 1995],thiswasnot observed ditions).Becauseof the integralnatureof Fe/lightcolimitation
at 54øSwhere the resident cells had tenfold higher cellular and the limitednatureof the presentdata set,it is not possible
chlorophylla levels, indicative of increasedlight-harvesting to ascertain the relative contributions of irradiance and Fe
supplyto the control of algal growth. This subjectrequires
capability,than at 47øS.
In other deckboardFe enrichmentexperimentsconducted further research in SA waters.
Table 3b. Phytoplankton
GrowthRates(Doublings
d-•) Estimated
FromInstantaneous
Changes
in CellAbundances
(Net Growth)a
47øS
Treatment
t = 2
Control
-0.05
Low-Fe/low-light
High-Fe/low-light
Low-Fe/high-light
High-Fe/high-light
- 0.04
0.29
0.02
0.21
aFor assumptions,
see section2.
t = 4
0.00
0.02
0.11
-0.05
0.25
54øS
t = 7
t = 2
t -
6
t =
10
0.10
-0.05
0.10
0.06
0.04
0.02
0.154
0.08
0.01
- 0.03
0.02
- 0.02
0.10
0.02
0.20
0.27
0.05
0.02
0.22
0.16
BOYD ET AL.: SUBPOLAR
PHYTOPLANKTON,
IRON, AND LIGHT
31,581
45S
140
120
50S
50S
100
55S
55S
80
60S
60S
60 :•
40
65S
20
70S
70S
130E
140E
150E
160E
170E
180E
170W
160W
150W
Plate 1. A plot of mean summermixed layer depths(MLDs) for the Australasian-Pacific
sectorof the
SouthernOcean (130øE-150øW,40ø-75øS)collatedfrom the WorldOceanAtlas. Suppliedby M. Hadfield
(National Instituteof Water and AtmosphericResearch,Wellington).
4.3.
Comparison With Other Iron/Light Perturbation
Field
Studies
As far asis known,onlyfive other Fe/light experimentshave
been conducted in the field, four in the Southern Ocean and
one in the NE subarcticPacific,whereMaldonadoet al. [1999]
demonstratedFe/light colimitationin a deckboardsimulation
of an 80 m MLD. They reported the highestchlorophylla
levels in a high-Fe/high-lighttreatment and the lowest in a
low-Fe/low-lighttreatment. This is consistentwith the 54øS
experiment(although there was no true low-Fe treatment at
54øS).Maldonadoet al. [1999]reportedlittle changein Pbm•x
over time in their experiments,which they attributed to the
confoundinginfluenceon Pbmaxof light limitation [Richardson
et al., 1983] and Fe limitation [Greeneet al., 1991]. In the
presentexperiments,up to tenfold increasesin cellularchlorophylla were observed,illustratingthe pitfalls of usingPbmax
to expressthe maximumrate of photosynthesis
[seeHenleyand
Yin, 1998] and the need to normalizePmaxto cell abundance.
Boydet al. [1999]carriedout a deckboardFe enrichmentin
SA watersSE of New Zealandin australspring(MLD 80 m) at
10%-50% Io and reportedlittle changein chlorophylla (0.3
in cellularpigmentcontentdue to Fe limitation,whereaslight
limitation induced an increase in cellular pigment content
(whichwaslesspronouncedfor Fe-depletecells).During the
SOIREE in situ mesoscaleFe enrichment,Boydet al. [2000]
also conducted
deckboard
Fe enrichments
in which the mean
lightlevelswerevaried.They reportedthat therewasrelatively
little increase in chlorophyll concentrationsin a treatment
mimicking a 100 m mixed layer, relative to the observed
greaterthanfifteenfoldincreases
in treatmentsrepresenting65
and 40 m MLDs. These findingsare in contrastto Takeda
[1998],who reportedno differencein growthrates (instantaneouschangein chlorophylla levels) betweenFe-enriched
deckboard
carboys
(>1 nmolL -• Fe added)incubated
at <5
and 40% Io in polar waterssouthof the PF. The reasonfor
thesedifferentoutcomesin deckboardFe/lightexperimentsin
the SouthernOcean is not presentlyknown.
4.4. Estimating A!gal Growth Rates in the Southern Ocean
In deckboard Fe enrichments in the Ross Sea, Martin et al.
[1990]reportedincreasedalgal(net) growthrates(chlorophyll
a specific)
from0.3 to 0.7 doublings
d-•. In the 54øSexperi/•g L-•) at 10%I o after5 daysbutincreases
to 0.7/•g and>1 ment,chlorophylla specificgrowthratesincreasedto > 1 douthusexceeding
the theoret/•g L -• in the 30 and 50% Io carboys,respectively.
These blingd-• in highlighttreatments,
trends are similar to those observedfor the 54øSexperiment. ical maximumgrowthrate at 4øC [Banse,1991]. However, as
Van Leeuwe[1996] examinedthe effectsof Fe/light limitation cellularchlorophylla increasedby up to tenfold in the 54øS
on lab culturesof the Antarctic flagellatePyraminomonassp. experiment,growth rates cannotbe reliably estimatedusing
from waters south of the PF. Van Leeuwe observed a decline
proxiessuchaschlorophylla in regionsof the SouthernOcean
31,582
BOYD ET AL.: SUBPOLAR PHYTOPLANKTON,
IRON, AND LIGHT
where Fe/light colimitationof algal growth may exist. Esti- Behrenfeld,M. J., and Z. S. Kolber, Widespreadiron limitation of
phytoplanktonin the South Pacific Ocean, Science,283, 840-843,
matesof net algal growthrate (cellular abundanceand silica
1999.
turnover)at 54øSsuggestthat growthratesunderFe-replete Berg, G. M., P.M. Glibert, and C.-C. Chen, Dimension effectsof
conditions
were0.25doublings
d-•.
4.5.
Seasonalityin Factors Limiting Production
in the Southern
Ocean
In the presentFe/lightexperimentsall treatmentswere also
enriched with silicic acid. The interactions between algal
growth,Fe, andsilicicacidare exploredbySedwicket al. [1999]
and Hutchinset al. [this issue],who report evidenceof Fe/Si
limitation at 47øSin March 1998. Qu•guiner[this issue]also
exploresthe possibility
of silicicacidlimitation.Thesefindings,
together,are indicativeof seasonality
in the factorslimitingthe
growthof eukaroticphytoplanktonin SA waters.In contrast,
prokaryoticphytoplanktonhave higher iron quotas [Brand,
1991], have no requirementfor silicicacid, and appearto be
particularlysensitiveto temperature[Marchantet al., 1987];
thustheir growthrateswill be limitedby differentenvironmental factorsthan thosefor eukaryotes.
As discussed
by Boydet al. [1999], the seasonalprogression
of the factorscontrollingthe growthof eukaryoticphytoplankton in SA watersmay be irradiancein winter, Fe/light in early
springand autumnwhenwater columnlightlevelsare elevated
becauseof shallowerMLDs (causedby reducedwind stress),
and higherincidentirradiances[Bishopand Rossow,1991].If
silicicacid levelsare not limiting in spring,then Fe limitation
will control growth,whereasin summer/earlyautumn, if the
MLDs remainrelativelyshallow,Fe and silicicacidavailability
will probablycontrolgrowthrates.At 54øSin March 1998,Fe,
light, and silicicacid may have togetherlimited algal growth.
Although the effectsof light limitation on Fe uptake [Sunda
and Huntsman,1997] and on silicicacid uptake [Nelsonand
Brzezinski,1997;Brzezinskiet al., 2001], and that of Fe limitation on silicicacid uptake [Hutchinsand Bruland,1998] have
been demonstrated,the interrelationshipsbetweenFe, light,
and silicic acid limitation are less clear and require further
study.
Acknowledgments. We thank the officersand crew of R/V Aurora
Australisfor their skillsat sea.We are gratefulfor assistance
from the
supportstaff and workshopsat CSIRO (Hobart) and AntarcticDivision (Hobart). Field supportfor this work was providedby ANARE
(via ASAC proposal 1156). We acknowledgethe efforts of Lisette
Robertsonand Michael Grose (Antarctic CRC, Hobart) for the shipboard analysisof samplesfor size-fractionatedchlorophylland Rick
Van den Enden (Antarctic CRC, Hobart) for phytoplanktonspecies
identification.Vertical irradianceprofiles, CTD, and macronutrient
datawerekindlyprovidedby Don MacKenzie(CSIRO, Hobart), Mark
Rosenberg(Antarctic CRC, Hobart), and Neil Johnson(Antarctic
CRC, Hobart). We are gratefulto Mark Hadfield(NIWA, Wellington)
for the provisionof a plot of meansummermixedlayerdepthscollated
from the WorldOceanAtlas and Julie LaRoche (Institut fur Meeresekunde,Kiel, Germany) for a personalcommunication.The comments of an anonymousreviewer and the editor greatly assistedin
improvingthismanuscript.This researchwassupportedby Antarctic
CRC, Hobart, ANARE, Hobart, Australia. PB acknowledgesthe
supportof the New Zealand GovernmentFRST programmevia the
Ocean Ecosystemproject. DH acknowledgesthe support of NSF
INT 9802132 for participation in this study. BQ the acknowledges
French S.O. JGOFS (ANTARES program) and the Comit6 Franqais de Coop6rationFranco-Australienneen Scienceset TechnologiesMarines (IFREMER and INSU/CNRS).
References
Banse,K., Ratesof phytoplanktoncell divisionin the field and in iron
enrichmentexperiments,Limnol. Oceanogr.,36, 1866-1898, 1991.
enclosureson ecologicalprocessesin pelagic systems,Limnol.
Oceanogr.,44, 1331-1340, 1999.
Bishop,J. K. B., and W. R. Rossow,Spatialand temporalvariabilityof
globalsurfacesolarirradiance,J. Geophys.
Res.,96, 16,839-16,858,
1991.
Boyd,P. W., D. Muggli,D. Varela, R. H. Goldblatt,R. Chretien,K. J.
Orians,and P. J. Harrison, In vitro iron enrichmentexperimentsin
the NE subarcticPacific,Mar. Ecol. Prog.Ser.,136, 179-196, 1996.
Boyd,P. W., J. LaRoche,M. Gall, R. Frew, and R. M. L. McKay, The
role of iron, light, and silicatein controllingalgal biomassin subAntarcticwatersSE of New Zealand,J. Geophys.Res.,104, 13,39513,408, 1999.
Boyd, P. W., et al., A mesoscalephytoplanktonbloom in the polar
SouthernOcean stimulatedby iron fertilization,Nature, 407, 695702, 2000.
Brand, L. E., Minimum iron requirementsof marine phytoplankton
and the implicationsfor the biogeochemical
controlof new production, Limnol. Oceanogr.,36, 1756-1771, 1991.
Brzezinski,
M. A., andD. R. Philips,
Evaluation
of 32Siasa tracerfor
measuringsilicaproductionratesin marinewaters,Limnol. Oceanogr.,42, 856- 865, 1997.
Brzezinski,M. A., T. A. Villareal, and F. Lipschultz,Silicaproduction
and the contributionof diatomsto new and primary productionin
the centralNorth Pacific,Mar. Ecol. Prog.Ser.,in press,2001.
Coale,K. H., Effectsof iron, manganese,copperand zinc enrichments
on productivityandbiomassin the subarcticPacific,Limnol. Oceanogr.,36, 1851-1864, 1991.
Coale, K. H., et al., A massivephytoplanktonbloom inducedby ecosysterfi-scale
iron fertilization experimentin the equatorialPacific
Ocean, Nature, 383, 495-501, 1996a.
Coale, K. H., S. E. Fitzwater, R. M. Gordon, K. S. Johnson,and R. T.
Barber, Control of communitygrowth and export productionby
upwellediron in the equatorialPacificOcean,Nature,379, 621-624,
1996b.
Crossley,A. C., Shipboardflow cytometricanalysisof protistpopulation structurein the SouthernOcean,paper presentedat VII SCAR
InternationalBiologicalSymposium,Sci. Counc.for AntarcticRes.,
Christchurch,N. Z., 1998.
de Baar, H. J. W., and P. W. Boyd,The role of iron in planktonecology
and carbon dioxide exchangeof polar and subpolaroceans,in
JGOFS SymposiumVolume,IGBP Books Set., vol. 5, pp. 61-140,
CambridgeUniv. Press,New York, 2000.
de Baar, H. J. W., J. T. M. de Jong,D.C. E. Bakker, B. M. L6scher,
C. Veth, U. Bathmann,and V. Smetacek,Importance of iron for
plankton blooms and carbon dioxide drawdownin the Southern
Ocean, Nature, 373, 412-415, 1995.
Denman,K., andA. Gargett,Time and spacescalesof verticalmixing
and advectionof phytoplankton
in the upperocean,Limnol. Oceanogr.,28, 801-815, 1983.
Dugdale,R. C., F. P. Wilkerson,andH. J. Minas,The role of a silicate
pump in drivingnew production,Deep SeaRes.,Part I, 42, 697-719,
1995.
Greene, R. M., R. J. Geider, and P. G. Falkowski, Effect of iron
limitationon photosynthesis
in a marinediatom,Limnol.Oceanogr.,
36, 1772-1782, 1991.
Henley, W. J., and Y. Yin, Growth and photosynthesis
of marine
Synechococcus
(Cyanophyceae)
underiron stress,
J. Phycol.,34, 94103, 1998.
Hutchins,D. A., and K. W. Bruland, Iron limited diatom growth and
Si:N uptake ratios in a coastalupwellingregime,Nature, 393, 561564, 1998.
Hutchins, D. A., P. N. Sedwick, G. DiTullio, P. W. Boyd, B.
Qu6guiner,F. B. Griffiths, and A. C. Crossley,Control of phytoplanktongrowthby iron and silicicacid availabilityin the subantarcticSouthernOcean:Experimentalresultsfrom the SAZ Project,
J. Geophys.Res.,this issue.
Joint, I. R., and A. J. Pomroy,Productionof picoplanktonand small
nanoplanktonin the Celtic Sea,Mar. Biol., 77, 19-27, 1983.
Joint, I., A. Pomroy, G. Savidge,and P. Boyd, Size-fractionatedprimary productivityof the NE Atlantic in May-June 1989,Deep Sea
Res., Part II, 40, 423-440, 1993.
Kolber, Z., and P. G. Falkowski,Use of active fluorescenceto estimate
BOYD ET AL.: SUBPOLAR PHYTOPLANKTON,
IRON, AND LIGHT
31,583
phytoplanktonphotosynthesis
in situ,Limnol. Oceanogr.,38, 1646-
Schmidt,M. A., and D. A. Hutchins,Size-fractionatedbiologicaliron
and carbonuptake along a coastalto offshoretransectin the NE
Pacific,Deep SeaRes.,Part II, 46, 2487-2503, 1999.
K. S. Johnson,S. Lindley, and P. G. Falkowski,Iron limitation of Sedwick,P. N., P. R. Edwards,D. J. Mackey,F. B. Griffiths,and J. S.
Parslow,Iron and manganesein surfacewaters of the Australian
phytoplankton
photosynthesis
in the equatorialPacificOcean,Nature, 371, 145-149, 1994.
subantarcticregion,Deep SeaRes.,Part I, 44, 1239-1253, 1997.
LaRoche, J., P. W. Boyd, R. M. L. McKay, and R. J. Geider, Fla- Sedwick,P. N., G. R. DiTullio, D. A. Hutchins,P. W. Boyd, F. B.
vodoxinasan in situmarkerfor iron stressin phytoplankton,
Nature,
Griffiths, T. W. Trull, and B. Queguiner,Limitation of algal pro382, 802-805, 1996.
ductionby iron deficiencyin the Australiansectorof the subantarcLippemeier,S., P. Hartig, and F. Colijn, Direct impact of silicateon
tic SouthernOcean, Geophys.Res.Lett., 26, 2865-2868, 1999.
the photosyntheticperformanceof the diatom Thalassiosira
weiss- Sunda,W. G., and S. A. Huntsman,Interrelated influenceof iron, light
flogii assessed
by on- and off-line PAM fluorescencemeasurements,
andcell sizeon marinephytoplanktongrowth,Nature,390, 389-392,
1665, 1993.
Kolber, Z. S., R. T. Barber, K. H. Coale, S. E. Fitzwater, R. M. Greene,
J. Plankton Res., 21,269-283,
1999.
1997.
MacIntyre,S., Turbulentmixingandresourcesupplyto phytoplankton, Takeda,S., Influenceof iron availabilityon nutrientconsumption
ratio
of diatoms in oceanicwaters, Nature, 393, 774-777, 1998.
in PhysicalProcesses
in Lakes and Oceans,CoastalEstuarineStud.,
vol. 54, edited by J. Imberger, pp. 561-590, AGU, Washington, Timmermans,K. R., et al., Iron stressin the Pacific region of the
D.C., 1998.
SouthernOcean:Evidencefrom enrichmentassays,
Mar. Ecol. Prog.
Set., 166, 27-41, 1998.
Mackey, D. J., J. Parslow,H. W. Higgins, F. B. Griffiths, and J. E.
O'Sullivan,Planktonproductivityand biomassin the westernequa- van Leeuwe, M. A., A Barren Ocean, iron and light interactionswith
torial Pacific:Biologicaland physicalcontrols,Deep SeaRes.,Part I,
phytoplanktongrowthin the SouthernOcean,Ph.D. thesis,189 pp.,
42, 499-533, 1995.
Univ. of Groningen,Groningen,Holland, 1996.
Marchant, H. J., A. T. Davidson,and S. W. Wright, The distribution Vassiliev,I. R., Z. Kolber, K. D. Wyman, D. Mauzerall, V. K. Shukla,
and abundance of chroococcoidcyanobacteriain the Southern
and P. G. Falkowski,Effects of iron limitation on photosystem,II,
Ocean,Proc.NIPR Syrup.Polar Biol., 1, 1-9, 1987.
Composition and light utilisation in Dunaliella tertiolecta,Plant
Martin, J. H., R. M. Gordon, S. E. Fitzwater, and W. W. Broenkow,
Physiol.,109, 963-972, 1995.
VERTEX: Phytoplankton/ironstudiesin the Gulf of Alaska,Deep Wells, M. L., N.M. Price, and K. W. Bruland, Iron chemistryin
Sea Res., Part A, 36, 649-680, 1989.
seawaterand its relationshipto phytoplankton:A workshopreport,
Mar. Chem., 48, 157-182, 1995.
Martin, J. H., S. E. Fitzwater, and R. M. Gordon, Iron deficiencylimits
phytoplanktongrowth in Antarctic waters, Global Biogeochem.Cy- Wright, S. W., S. W. Jeffrey, R. F. C. Mantoura, C. A. Llewellyn, T.
cles,4, 5-12, 1990.
Bjornland, D. Repeta, and N. A. Welschmeyer,Improved HPLC
Maldonado, M., P. W. Boyd, N.M. Price, and P. J. Harrison, Comethodfor the analysisof chlorophylisand carotenoidsfrom marine
limitation of phytoplanktongrowthby light and Fe duringwinter in
phytoplankton,Mar. Ecol. Prog.Set., 77, 183-196, 1991.
the subarcticPacific Ocean, Deep Sea Res.,Part II, 46, 2475-2486, Wright, S. W., R. L. Waters, R. L. van den Enden, P. Scott, P. G.
1999.
Strutton,A. C. Crossley,H. J. Marchant, F. B. Griffiths, and P. I.
Mitchell, B. G., E. A. Brody, O. Holm-Hansen, C. McClain, and J.
Bonham,Phytoplanktonpopulationsin the SouthernOcean:DistriBishop,Light limitationof phytoplanktonbiomassand macronutribution, abundance,productivityand compositionduring Jan-Mar,
ent utilizationin the SouthernOcean,Limnol. Oceanogr.,36, 16621996, paper presentedat Annual Conference,AustralianMar. Sci.
1677, 1991.
Assoc., Hobart, Tasmania, 1996.
Nelson,D. M., and M. A. Brzezinski,Diatom growthand productivity Zentara, S. J., and D. Kamykowski,Geographicvariationin the relain an oligotrophicmidoceangyre:A 3-yr recordfrom the Sargasso
tionshipbetweensilicicacid and nitrate in the SouthPacificOcean,
Sea near Bermuda, Limnol. Oceanogr.,42, 473-486, 1997.
Deep Sea Res.,Part A, 28, 455-465, 1981.
Nelson,D. M., andW. O. SmithJr., Sverdruprevisited:Critical depths,
maximum chlorophyll levels, and the control of Southern Ocean
P. W. Boyd, Centre for Chemicaland PhysicalOceanography,Naproductivityby the irradiance-mixingregime,Limnol. Oceanogr.,36, tional Institute of Water and AtmosphericResearch,Department of
1650-1661, 1991.
Chemistry, University of Otago, Dunedin, 9001 New Zealand.
Olson, R. J., E. R. Zettler, and M.D. DuRand, PhytoplanktonAnal([email protected])
A. C. Crossley,Institute of Antarctic and SouthernOcean Studies,
ysisUsingFlow cytometry,in Handbookof Methodsin AquaticMicrobialEcology,editedby P. F. Kemp et al., pp. 175-186,A. F. Lewis, University of Tasmania, P.O. Box 252-77, Hobart, Tasmania 7001,
New York, 1993.
Paasche,E., Silicon and the ecologyof marine planktonicdiatoms,1,
Thalassiosira
pseudonana
(Cyclotella
nana) grownin chemostats
with
silicateas the limiting nutrient, Mar. Biol., 19, 117-126, 1973.
Qu6guiner,B., Biogenicsilicaproductionin the Australian sectorof
the Subantarctic Zone of the Southern Ocean in late summer 1998,
J. Geophys.Res., this issue.
Raven,J. A., Predictionsof Mn and Fe useefficienciesof phototrophic
growthas a functionof light availabilityfor growthand C assimilation pathway,New Phytol.,116, 1-17, 1990.
Richardson,K., J. Beardall, and J. A. Raven,Adaptation of unicellular
algaeto irradiance:An analysisof strategies,New Phytol.,93, 157191, 1983.
Rintoul, S. R., and T. W. Trull, Seasonalevolutionof the mixedlayer in
the SubantarcticZone southof Australia,J. Geophys.
Res.,this issue.
Rintoul, S. R., J. R. Donguy,and D. H. Roemmich,Seasonalevolution
of upperoceanthermalstructurebetweenTasmaniaand Antarctica,
Deep SeaRes.,Part I, 44, 1185-1202, 1997.
Australia.
G. R. DiTullio, Grice Marine Laboratory,Universityof Charleston,
205 Fort Johnson, Charleston, SC 29412, USA.
F. B. Griffiths, CSIRO Division of Marine Research, GPO Box
1538, Hobart, Tasmania 7001, Australia.
D. A. Hutchins,Collegeof Marine Studies,Universityof Delaware,
700 Pilottown Road, Lewes, DE 19958, USA.
B. Queguiner,Laboratoire d'Oceanographieet de Biogeochimie,
Centre d'Oceanographiede Marseille,UMR CNRS 6535, Parc Scientifique et Technologiquede Luminy, Case 901, F-13288 Marseille
Cedex 09, France.
P. N. Sedwick and T. W. Trull, Antarctic CRC, GPO Box 252-80,
Hobart, Tasmania 7001, Australia.
(ReceivedMarch 30, 2000; revisedMay 8, 2001;
acceptedMay 21, 2001.)