N t l i f tili ti b th Natural iron fertilization above the Kerguelen Plateau (Southern Ocean):

Impact on carbon remineralization by the Impact on carbon remineralization by the microbial food web

Ingrid Obernosterer

Laboratoire d’Océanographie MicrobienneObservatoire Océanologique de Banyuls

CNRS UPMCFrance

Thanks to

Urania Christaki @ Université du Littoral Côte d’Opale, Wimereuxp ,

Andrea Malits @ Laboratoire d’Océanographie de VillefrancheVillefranche

Dominique Léfèvre @ Laboratoire de Microbiologie, Gé l i E l i M i M illGéologie et Ecologie Marine, Marseille

Philippe Catala @ Laboratoire d’Océanographie pp g pMicrobienne, Banyuls

Questions adressed

How does natural iron fertilization affect different components of the microbial food web ?components of the microbial food web ?

Wh h h f f What are the consequences on the transfer of carbon through the microbial food web ?

The natural iron fertilization experiment KEOPS (Kerguelen Ocean and Plateau compared Study) (Kerguelen Ocean and Plateau compared Study)

F B d t l 2008From Boyd et al. 2008

HNLCStationBloom

Station

The Kerguelen Bloom

Blain et al. (2007)

Marked response of heterotrophic bacteria

Bloom HNLC

01 2 3 4 5 6

01 2 3 4 5 6Bacterial Abundance (x 108 L-1) Bacterial Abundance (x 108 L-1)

50

100

50

100) 100

150

100

150Dep

th (

m)

200

250

200

250

Bacterial AbundanceChla

Chlorophyll (µg L-1) Chlorophyll (µg L-1)

Obernosterer et al. (2008)

Marked response of heterotrophic bacteria

40 50 60 70 80 900

Bl HNLC

High Nucleic Acid cells (%)

50Bacterial Production 0.25 0.042

(µmol C L-1 d-1)

100

150Dep

th (

m)

Bacterial Respiration 0.94 0.25(µmol C L-1 d-1)

B t i l G th150

200

D

Bloom All values are mean of upper 100 m

Bacterial Growth Efficiency 21 14

(%)

250

BloomHNLC

All values are mean of upper 100 m.

Obernosterer et al. (2008)

Comparison among iron fertilization experiments in the Southern Ocean

Bacterial heterotrophic production(µg C L-1d-1)

HNLC HNLC + Fe0.5 3KEOPS X 6

CROZEX 0 4 3 6 X 9

0.2-0.5 0.2-1.1SOIREE, EisenEx, SOFEX X 2-3

CROZEX 0.4 3.6 X 9Zubkov et al. (2007)

Wh t ti l t b t i l h t t hi ti it ?

Hall and Safi (2001), Arrieta et al. (2004),

Oliver et al. (2004)

What stimulates bacterial heterotrophic activity?

Direct or indirect response to iron addition?

not Fe, but organic carbon first limiting factor

Similar diatom biomass in Fe-fertilization studies

b d ff h d f h bl

From Armand et al. (2008)

….but differences in the duration of the blooms : Mesoscale experiments: ≈ 1 to 40 daysKerguelen bloom : ≈ 60 to 80 days

The duration of the bloom and the mode of fertilization allows adaptation of the bacterial community.

40

45

-7

Strikingly different bacterial diversity at the two stations

25

30

35

40

NA

C11

-

SAR

92

Agg

58

10

15

20

25

A21

M43 SA

R86

IIA

rctic

96B

-16

96B

-1rc

tic 9

6BD

-19

c 97

A-1

7

97A

-13

ence

s

RO

S R

CA Bloom

0

5

10

Ros

eo.

eo.

SAR

11- A OM

Arc

tic 9

I

Ar

Arc

tic

4A

rctic

9

er o

f seq

u

5

10

15 cter

NA

C11

-6

SAR

116

III

Ros

e

SAR

86 II

I SAR

86 I

Arc

tic 9

7A-1

4

Num

b

HNLC5

20

25

Ros

eoba

c

acte

r

HNLC

35Alpha

proteobacteria beta Gamma

proteobacteria Bacteroidetes

SAR1

1

30

Pola

riba

West et al. (2009)

Distinct bacterial groups contribute to C-cyclingwithin the Kerguelen bloom within the Kerguelen bloom

HNLCBloomSAR86

SAR92 Agg58

SAR11

NAC11-7RCA

Obernosterer et al. (in rev.)

Temporal changes of SAR92

Decline-BloomRelative Abundance (%)

End-BloomRelative Abundance (%)

Mid-BloomRelative Abundance (%)

chla

a aa

SAR92chla

chla (µg L 1) chla (µg L )chla (µg L )

Obernosterer et al. (in rev.)

Pronounced response of heterotrophic bacteria to natural iron fertilization, in terms of activity

d it itiand community composition.

D f b h l l f Does it matter for biogeochemical cycling of elements?

Transfer of carbon through the microbial gfood web

Heterotrophic nanoflagellates - the first trophic link

0 0 5 1 1 5 2

Heterotrophic Nanoflagellates (x106 L-1)

0

20

0 0.5 1 1.5 2

40

60th (

m) 35 95Grazing loss of

bacteria (in %, d-1)60

80

Dep

t

100

120

BloomHNLC

Christaki et al. (2008)

Viruses - a sink of bacterial heterotrophic production

00 5 10 15 20

Viral Particles (x109 L-1)

20

4040

60

80Dep

th (

m)

1.2 0.3Lytic Viral Production

(x 106 ml-1 h-1)

80

100

D

Frequency of infected cells (%) 50 30

120

140

BloomHNLC

Malits et al. (in prep.)

Ciliates - the link between the microbial food web and the classical food web

100-400 organisms L-1

100

12050-100 µm 30-50 µm 20-30 µm 10-20 µm

Tintinnid60

80

(mg

C m

-2)

40

60

Bio

mas

s

0

20

Bloom HNLC

Aloricated Ciliate

Aloricated CiliateBloom HNLC

Christaki et al. (2008)

Carbon flow through the microbial food web

Bloom HNLC

CO2 CO2

Viralloop

HNANHeterotrophic bacterial

carbon demand

Viralloop

Heterotrophic bacterial

carbon demand

Ciliates

40% 90%

Mesozooplankton

S di t ti

60%GCP

100 mmol C m-2 d-1

GCP

40 mmol C m-2 d-1

10%

Sedimentation100 mmol C m d 40 mmol C m d

Can the functioning of the microbial food web affect carbon export ?carbon export ?

Export efficiency*

Bloom HNLCBloom HNLC≈ 28% ≈ 58%

*Export Production (234Thorium, POC, PON) : Primary production (C-and N-uptake rates)

Savoye et al. (2008)

Conclusions

d f h h -Pronounced response of heterotrophic bacteria to natural iron fertilization, in terms of activity and community composition.

- Rapid mineralization of organic carbon due to

mmu y mp .

p gmicrobial food web processes.

Merci de Merci de votre

attention!

Banyuls sur mer, March 2010