Click
He e
o
Full
A
icle
Signi icance o non‐sinking pa icula e o ganic ca bon and da k
CO
2
ixa ion o he e o ophic ca bon demand in he mesopelagic
no heas A lan ic
Fede ico Bal a ,
1
Ja ie A ís egui,
1
E a Sin es,
2,4
Josep M. Gasol,
3
Thomas Rein hale ,
2,4
and Ge ha d J. He ndl
2,4
Recei ed 2 Ma ch 2010; accep ed 1 Ap il 2010; published 5 May 2010.
[1] I is gene ally assumed ha sinking pa icula e o ganic
ca bon (POC) cons i u es he main sou ce o o ganic ca bon
supply o he deep ocean’s ood webs. Howe e , a majo
disc epancy be ween he a es o sinking POC supply
(collec ed wi h sedimen aps) and he p oka yo ic o ganic
ca bon demand ( he o al amoun o ca bon equi ed o
sus ain he he e o ophic me abolism o he p oka yo es;
i.e., p oduc ion plus espi a ion, PCD) o deep‐wa e
communi ies has been consis en ly epo ed o he da k
ealm o he global ocean. While he amoun o sinking POC
lux declines exponen ially wi h dep h, he concen a ion o
suspended, buoyan non‐sinking POC (nsPOC; ob ained
wi h oceanog aphic bo les) exhibi s only small a ia ions
wi h dep h in he (sub) opical No heas A lan ic. Based on
a ailable da a o he No h A lan ic we show he e ha he
sinking POC lux would con ibu e only 4–12% o he PCD
in he mesopelagic ealm (depending on he p ima y
p oduc ion a e in su ace wa e s). The amoun o nsPOC
po en ially a ailable o he e o ophic p oka yo es in he
mesopelagic ealm can be pa ly eplenished by da k
dissol ed ino ganic ca bon ixa ion con ibu ing be ween
12% o 72% o he PCD daily. Taken oge he , he e is
e idence ha he mesopelagic mic ohe e o ophic bio a is
mo e dependen on he nsPOC pool han on he sinking POC
supply. Hence, he enigma ic majo misma ch be ween he
o ganic ca bon demand o he deep‐wa e he e o ophic
mic obio a and he POC supply a es migh be subs an ially
smalle by including he po en ially a ailable nsPOC and i s
au och honous p oduc ion in oceanic ca bon cycling models.
Ci a ion: Bal a ,F.,J.A ís egui,E.Sin es,J.M.Gasol,T.Rein hale ,
and G. J. He ndl (2010), Signi icance o non‐sinking pa icula e o -
ganic ca bon and da k CO
2
ixa ion o he e o ophic ca bon demand
in he mesopelagic no heas A lan ic, Geophys. Res. Le .,37,
L09602, doi:10.1029/2010GL043105.
1. In oduc ion
[2] I is gene ally accep ed ha he deep‐wa e he e o-
ophic ood web elies on he o ganic ma e (OM) gene -
a ed by p ima y p oduc ion in he sun‐li su ace wa e s. On
a global a e age, abou 30% o he su ace wa e ’s p ima y
p oduc ion is expo ed in o he da k ocean as sedimen ing
pa icles [A ís egui e al., 2005b; Buessele and Boyd,
2009]. Al hough he concen a ion o dissol ed o ganic
ca bon (DOC) is gene ally abou 10 imes highe han ha o
pa icula e o ganic ca bon (POC), he su ace‐expo ed
DOC pool makes i g ea es con ibu ion o oxygen con-
sump ion in he uppe ocean [Hansell e al., 2009]) and i
leas a g ea e dep hs (abou 10% o oxygen consump ion
[A ís egui e al., 2002]). Hence, passi ely sinking pa icles
o igina ing om he eupho ic zone ha e been conside ed as
he p incipal sou ce o o ganic ca bon a ailable o he
he e o ophic ood web in he ocean’s in e io [Buessele e
al., 2007]. The majo i y o he POC expo ed om he
eupho ic zone is emine alized in he mesopelagic laye
(be ween 100–1000 m dep h), leading o a ypical exponen ial
a enua ion o he sinking POC concen a ion wi h dep h.
[3] Assuming mass balance, he supply a e o o ganic
ca bon (gene ally calcula ed om sinking POC collec ed by
sedimen aps) eaching he ocean’s in e io should ma ch
he ca bon demand o he he e o ophic bio a inhabi ing he
meso‐and ba hypelagic ealm. The deep‐wa e bio a is
as ly domina ed by p oka yo es, in e ms o abundance and
biomass. Howe e , he p oka yo ic o ganic ca bon demand
( he o al amoun o o ganic ca bon equi ed o sus ain he
he e o ophic me abolism o p oka yo es, de e mined as
espi a ion plus p oduc ion, PCD) has been shown o con-
inuously exceed he sinking POC lux in o he da k ocean
[Bu d e al., 2010]. Recen es ima es epo a misma ch
be ween deep‐wa e PCD and sinking POC lux by up o 2–
3 o de s o magni ude [Rein hale e al., 2006; S einbe g e
al., 2008; Bal a e al., 2009]. Mo eo e , a global budge ing
exe cise indica ed ha e en he highes es ima e o sinking
POC lux om su ace wa e s would only explain abou
50% o he measu ed oxygen consump ion in he da k ocean
[del Gio gio and Dua e, 2002]. This pa adoxical imbalance
has been shown o a y spa ially. PCD was 3–4 imes and
a ound 10 imes g ea e han he sinking POC lux in he
sub opical and suba c ic Paci ic, espec i ely [S einbe g e
al., 2008]. This s ong imbalance be ween POC supply
and demand ep esen s one o he g ea challenges in con-
empo a y biological oceanog aphy and ma ine biogeo-
chemis y, and indica es majo gaps in ou unde s anding o
he deep ocean ca bon lux. I is appa en ha some majo
componen s and aspec s o o ganic ca bon s ocks and luxes
ha e no been aken in o accoun adequa ely.
[4] One o he majo POC pools in he ocean’s in e io no
adequa ely aken in o accoun in he oceanic ca bon budge s
1
Facul ad de Ciencias del Ma , Uni e sidad de Las Palmas de G an
Cana ia, Las Palmas de G an Cana ia, Spain.
2
Facul y Cen e o Ecology, Depa men o Ma ine Biology,
Uni e si y o Vienna, Vienna, Aus ia.
3
Depa amen de Biologia Ma ina i Oceanog a ia, Ins i u de
Ciències del Ma , CSIC, Ba celona, Spain.
4
Depa men o Biological Oceanog aphy, Royal Ne he lands
Ins i u e o Sea Resea ch, Den Bu g, Ne he lands.
Copy igh 2010 by he Ame ican Geophysical Union.
0094‐8276/10/2010GL043105
GEOPHYSICAL RESEARCH LETTERS, VOL. 37, L09602, doi:10.1029/2010GL043105, 2010
L09602 1o 6
is slow‐sinking, o almos suspended non‐sinking POC
(nsPOC) collec ed wi h oceanog aphic bo les bu , due o
hei buoyan na u e, no accumula ing in sedimen aps
used o de e mine sinking POC lux [A ís egui e al., 2009].
In a p e ious s udy, Bal a e al. [2009] ound a signi ican
co ela ion be ween nsPOC and po en ial espi a ion ( he
main pa ame e de e mining PCD) in he deep wa e s o he
sub opical No heas A lan ic, suppo ing he iew ha
nsPOC plays a key ole in da k ocean me abolism, and
hence in supplying bioa ailable subs a es o suppo he
PCD. Ne e heless, he quan i a i e o igin o his nsPOC
s ill emains la gely enigma ic. The bulk meso‐and ba hy-
pelagic nsPOC may come om di e en sou ces: i) om
sel ‐assembly o dissol ed o ganic ma e ial yielding po ous
mic ogels ha can be eadily colonized by mic oo ganisms
[Chin e al., 1998], ii) om sinking pa icles disagg ega ed
by physical o ces [Bu d and Jackson, 2009] o he
ac i i y o mic obes and zooplank on [She idan e al.,
2002], iii) om la e ally ad ec ed nsPOC, o i ) om
in si u p oduc ion o o ganic ma e ial due o p oka yo ic
chemoau o ophs.
[5] He e we quan i y he con ibu ion o da k ocean
chemosyn hesis o PCD in he sub opical No heas
A lan ic and compa e i wi h he con ibu ion o he sinking
POC lux. We use da a colla ed om he meso‐and
ba hypelagic No h A lan ic on sedimen ing pa icula e o -
ganic ma e (POM) lux, POM s anding s ock ( hus in-
cluding he esiden nsPOC) and PCD o e‐e alua e he
appa en disc epancy be ween p oka yo ic o ganic ma e
demand and supply in he ocean’s in e io .
2. Me hods
2.1. S udy Si e and Sampling
[6] Mo e han 9000 km we e co e ed du ing he AR-
CHIMEDES‐I (No embe –Decembe 2005) c uise on boa d
RV Pelagia o esol e he a eal a iabili y in o ganic ma e
and p oka yo ic ac i i y in he meso‐and ba hypelagic
wa e s o he eas e n No h A lan ic (Figu e S1 in Tex S1 o
he auxilia y ma e ial).
1
Samples we e aken om se en
dep hs: he su ace mixed laye (a e age dep h 50 m) o
pa icula e o ganic ma e (POM) and elec on anspo
sys em (ETS) only, he base o he mixed laye (100 m
laye ), he mesopelagic (250, 500 and 900 m), he No h
Eas A lan ic Deep Wa e (NEADW; a e age dep h 2750 m)
and he Lowe Deep Wa e (LDW; a e age dep h 4000 m).
Samples om he dis inc wa e masses we e collec ed wi h
12 L NOEX (no oxygen exchange) bo les moun ed on a
CTD (conduc i i y, empe a u e, dep h) ame o de e mine
o al p oka yo ic abundance, leucine inco po a ion, POM,
ETS and da k DIC ixa ion as desc ibed below.
2.2. P oka yo ic Abundance, He e o ophic
Me abolism and Pa icula e O ganic Ma e
Concen a ions
[7] P oka yo ic abundance, leucine inco po a ion, pa ic-
ula e o ganic ca bon and ni ogen, espi a o y ac i i y o he
elec on anspo sys em (ETS) and non‐sinking pa icula e
o ganic ca bon (nsPOC) and ni ogen (nsPON) we e ana-
lyzed as explained in a p e ious s udy [Bal a e al., 2009]
(see also auxilia y ma e ial).
2.3. Dissol ed Ino ganic Ca bon (DIC) Fixa ion
[8] DIC ixa ion was measu ed ia he inco po a ion o
[
14
C]‐bica bona e (3.7 × 10
6
Bq, Ame sham) in 50 ml
seawa e samples. T iplica e samples and o maldehyde‐
ixed blanks we e incuba ed in he da k a in si u empe a-
u es o 60–72 h. Incuba ions we e e mina ed by he
addi ion o o maldehyde (2% inal concen a ion) o he
samples, il a ion on o 0.2‐mm polyca bona e il e s and
insing wi h 10 ml o ul a‐ il e ed seawa e (<30 kDa).
A e wa d, he il e s we e exposed o a ume o concen-
a ed HCl o 12 h, ans e ed in o scin illa ion ials and
a e adding 8 ml scin illa ion cock ail (Canbe a‐Packa d,
Fil e Coun ), coun ed in he scin illa ion coun e o 10 min.
The esul ing mean disin eg a ions pe minu e (DPM) o he
samples we e co ec ed o he mean DPM o he blanks and
con e ed in o DIC ixed o e ime and co ec ed o he
na u al DIC concen a ions as measu ed by con inuous low
analysis [S oll e al., 2001].
3. Resul s and Discussion
[9] Unlike he gene al exponen ial decline in sinking POC
wi h dep h [Ma in e al., 1987], he a e age concen a ions
o nsPOC and nsPON emained ai ly cons an wi h dep h
down o ba hypelagic wa e s (4000 m dep h) o he (sub)
opical No heas A lan ic (Figu es 1a and 1b). This pa e n
con as s wi h he epo ed dec ease wi h dep h o nsPOC a
he Be muda A lan ic Time‐se ies (BATS) s a ion, in he
cen e o he No h A lan ic sub opical Gy e, bu is simila
o he dis ibu ion obse ed by Alonso‐González e al.
[2009] in he Cana y Cu en (CanC), a he no heas e n
side o ou sampling egion. O e all, ou nsPOC alues
spans he ange o he obse ed CanC and BATS con-
cen a ions (Figu e S2 in Tex S1 o he auxilia y ma e ial),
sugges ing a ansi ional con inen al shel ‐open ocean g a-
dien in nsPOM concen a ions. Gene ally, he p oka yo ic
con ibu ion o he nsPOC and nsPON dec eased wi h dep h
(Figu es 1c and 1d). Sub ac ing he p oka yo ic con ibu-
ion om he bulk nsPOC and nsPON pool, we ob ain he
amoun po en ially a ailable o he e o ophic u iliza ion o
nsPOC (nsPOC_A ) and nsPON (nsPON_A ), which did
no exhibi a gene al dep h‐ ela ed end (Figu es 1e and 1 ).
The a io nsPOC_A : nsPON_A was also ai ly cons an
h oughou he wa e column sugges ing ha he e a e no
majo shi s in he C:N a io o nsPOM in deep‐wa e s
(Figu e 1g), in con as o he epo ed inc easing C:N a ios
wi h dep h o sedimen ing POM and DOM [Schneide e al.,
2003; Hopkinson and Vallino, 2005]. This s abili y in he
elemen al composi ion o suspended POM, in combina ion
wi h i s ai ly cons an concen a ion h oughou he wa e
column o ou egion o s udy sugges s ha i is ei he no
u ilized bio ically o ha i s u iliza ion by he deep‐sea bio a
is ma ched by in si u p oduc ion o ex e nal impo o
nsPOM. Alonso‐González e al. [2009] obse ed lowe a -
e age C:N a ios in he CanC han in ou s udy, al hough
hei a ios clea ly inc eased in hei wes e n s a ions o he
sampling box, acco ding o he measu ed high espi a ion o
nsPOC h ough i s wes wa d anspo .
[10] P oka yo ic espi a ion (R) is he main pa ame e
a ec ing he PCD in he deep A lan ic, as i is ypically 1–
1
Auxilia y ma e ials a e a ailable in he HTML. doi:10.1029/
2010GL043105.
BALTAR ET AL.: DARK CO
2
FIXATION AND C DEMAND L09602L09602
2o 6
2 o de s o magni ude highe han p oka yo ic he e o ophic
p oduc ion [Rein hale e al., 2006; Bal a e al., 2009].
The e o e, ou PCD es ima es g ea ly ely on he con e sion
o measu emen s o he elec on anspo sys em (ETS) o
R, i.e., he R:ETS a io used (see auxilia y ma e ial). Ou
PCD es ima es we e b acke ed based on an R:ETS a io o
0.086, de i ed om bac e ial cul u es in senescen phase
[Ch is ensen e al., 1980], and o 0.6, ob ained om he
Figu e 1. Box‐Whiske plo o he e ical dis ibu ion o (a) bulk non‐sinking pa icula e o ganic ca bon (nsPOC) and
(b) ni ogen (nsPON), con ibu ion o he p oka yo ic biomass o he bulk (c) nsPOC and (d) nsPON, po en ially a ailable
(e) nsPOC and ( ) nsPON a e sub ac ing he p oka yo ic C‐and N‐biomass, espec i ely, and (g) he a io o po en ially
a ailable nsPOC o nsPON. All o ganic ma e concen a ions a e in mmol l
−1
.
BALTAR ET AL.: DARK CO
2
FIXATION AND C DEMAND L09602L09602
3o 6
same cul u es in exponen ial g ow h phase [Ch is ensen e
al., 1980] and in si u measu emen s o ac i e mesopelagic
p oka yo es in he CanC egion [A ís egui e al., 2005a].
The PCD dec eased by one o de o magni ude om he
base o he eupho ic zone owa ds he ba hypelagic zone
independen o he R:ETS a io used (Table 1). The oxygen
u iliza ion a es (OUR) epo ed o he cen e s o he No h
and Sou h A lan ic sub opical Gy es [Jenkins, 1982;
Jenkins and Wallace, 1992; B ea, 2008] a e wi hin he ange
o ou PCD es ima es (Table 1). Assuming ha hese bio-
geochemical es ima es ep esen a e age alues om ou
egion o s udy, he R:ETS a io should be close o 0.2.
Ne e heless, i is mo e plausible o assume ha he R:ETS
a io a ies be ween 0.2 and 0.6, since ou egion o s udy
spans a ansi ion zone wi h mo e nsPOM han obse ed in
he cen e s o he sub opical Gy es.
[11] Dissol ed ino ganic ca bon (DIC) ixa ion by che-
moau o ophic mic obes occu s h oughou he deep wa e s
o he A lan ic [He ndl e al., 2005], amoun ing on a e age
12 ± 5 mmol C m
−3
d
−1
in he mesopelagic ealm o he (sub)
opical No heas A lan ic (Table 1). This eshly p oduced
o ganic ca bon ep esen s an nsPOC sou ce in he meso‐
and ba hypelagic wa e s, as mic obial cells a e oo small o
sedimen . The au och honously p oduced nsPOC po en ially
accoun s o 72 ± 53% o 12 ± 9% (conside ing an R:ETS o
0.086 and 0.6, espec i ely) o he daily he e o ophic p o-
ka yo ic ca bon demand in he mesopelagic ealm (Table 2).
[12] Fo compa ison, he sinking POC lux based upon a
model om a compila ion o sedimen ap da a o he No h
A lan ic [An ia e al., 2001] was used o es ima e he po-
en ial con ibu ion o he sinking POC o PCD. We calcu-
la ed he sinking POC using he su ace p ima y p oduc ion
(PP) es ima es om he NE A lan ic gy e [Longhu s e al.,
1995] (28 mmol C m
−2
d
−1
), om he wes e n gy e
[S einbe g e al., 2001] (35 mmol C m
−2
d
−1
) and as an
uppe limi 50 mmol C m
−2
d
−1
( o he en i e No h A lan ic
[Longhu s e al., 1995] including empe a e egions whe e
PP is highe ). The po en ial con ibu ion o he sinking POC
amoun s o 4–6% o he PCD pe day when using a su ace
PP o 28 and 35 mmol C m
−2
d
−1
, espec i ely (Table 2).
Only when using he un ealis ically high su ace PP o 50
mmol C m
−2
d
−1
o his (sub) opical mid‐oceanic egion,
he con ibu ion o he sinking POC o he mesopelagic PCD
equals he con ibu ion (12%) o he da k CO
2
ixa ion
(Table 2). Despi e he a he low DIC ixa ion a es in he
ba hypelagic ealm, e en in his laye he con ibu ion o he
sinking POC o he PCD does no exceed he con ibu ion
de i ed om DIC ixa ion (Table 2). Thus, a signi ican
ac ion (a leas simila o he sinking POC con ibu ion) o
he o ganic ca bon equi ed by he meso‐and ba hypelagic
mic obial communi y could be supplied by chemoau o o-
phic CO
2
ixa ion in he (sub) opical No heas A lan ic’s
in e io .
[13] As shown p e iously, he majo ene gy sou ce o
p oka yo ic DIC ixa ion is likely de i ed om ammonia
oxida ion by C ena chaeo a, as a chaeal amoA genes a e by
a mo e abundan han bac e ial amoA genes in he meso-
pelagic ealm o he No h A lan ic [Agogué e al., 2008].
Table 1. Compa ison o he Va ia ion in he Di e en Dep h
Laye s o he P oka yo ic Ca bon Demand
a
Dep h
(m)
PCD
(R:ETS = 0.086)
PCD
(R:ETS = 0.6)
OUR
NA l
OUR
SA l AP
100 58.9 353.4 41.3
250 18.3 109.5 36.5 26.4 32.4
500 11.9 71.5 14.7 16.8 3.7
900 4.2 25.2 6.5 12.3 0.3
2750 3.3 19.9 0.2
4000 4.7 28.5 0.3
a
PCD, p oka yo ic ca bon demand. Assuming an R:ETS a io o 0.086
and 0.6, oxygen u iliza ion a es (OUR) epo ed o he No h A lan ic
Sub opical Gy e (OUR NA l) [Jenkins, 1982], and o he Sou h A lan-
ic Sub opical Gy e (OUR SA l) [B ea, 2008], and da k DIC ixa ion a es
(AP, au o ophic p oduc ion). Me abolic a es in mmol C m
−3
d
−1
.
Table 2. Po en ial Con ibu ion (in %) o Da k CO
2
Fixa ion (AP) and o he A ailable Sinking POC o he PCD
a
Dep h AP Sinking POC (PP = 28) Sinking POC (PP = 35) Sinking POC (PP = 50)
% Con ibu ion o PCD Assuming R:ETS =0.086
100 70 55 82 154
250 177 38 57 106
500 31 18 27 51
900 7 19 29 54
2750 5 4 6 10
4000 7 1 2 4
mesop a g 72 ± 53 25 ± 6 37 ± 10 70 ± 18
ba hy a g 6 ± 1 3 ± 1 4 ± 2 7 ± 3
Mesop AP: Sinking a io 2.8 1.9 1.0
Ba hyp AP: Sinking a io 2.4 1.6 0.9
% Con ibu ion o PCD Assuming R:ETS =0.6
100 12 9 14 26
250 30 6 9 18
500 5 3 5 8
900 1 3 5 9
2750 1 1 1 2
4000 1 0 0 1
mesop a g 12 ± 9 4 ± 1 6 ± 1 12 ± 3
ba hy a g 1 0 1 1
Mesop AP: Sinking a io 2.8 1.9 1.0
Ba hyp AP: Sinking a io 2.4 1.6 0.9
a
Assuming an R:ETS o 0.086 and 0.6. The con ibu ion om he sinking POC was in e ed om a POC lux model [An ia e al., 2001] assuming a
su ace p ima y p oduc ion (PP) o 28 mmol C m
−2
d
−1
(NE A lan ic gy e [Longhu s e al., 1995]), 35 mmol C m
−2
d
−1
(wes e n gy e [S einbe g e al.,
2001]) and o 50 mmol C m
−2
d
−1
(N A lan ic [Longhu s e al., 1995]).
BALTAR ET AL.: DARK CO
2
FIXATION AND C DEMAND L09602L09602
4o 6
Based on he adioca bon signa u e o a chaeal lipids, i has
been shown ecen ly ha 83% o he a chaeal ca bon is
de i ed by au o ophy in he sub opical mesopelagic No h
Paci ic (a 670 m dep h) [Ingalls e al., 2006]. The adio-
ca bon signa u e o DNA collec ed om mesopelagic Pa-
ci ic wa e s (670–915 m) allowed o di e en ia e be ween
he h ee majo ca bon pools ha a e po en ially a ailable o
p oka yo es: esh DOC eleased om POC (D
14
C>
+50‰), ambien DIC (D
14
C∼−200 o −100 ‰), and aged
bulk DOC (D
14
C=−525‰)[Hansman e al., 2009]. These
au ho s concluded ha bo h DIC and esh DOC (p esum-
ably eleased om sinking POC) a e u ilized subs an ially,
while ambien DOC is no a majo subs a e o mesopelagic
p oka yo es [Hansman e al., 2009]. The e is also e idence
o conside able chemoli ho ophic p oduc ion on sinking
POC collec ed by sedimen aps deployed in he mesope-
lagic (100–750 m dep h) No h Paci ic, whe e p oka yo es
con ibu es be ween 7–90% o he o al mic obial p oduc-
ion [Ka l e al., 1984].
[14] Taken oge he , he cu en pe cep ion on he de-
pendence o he da k ocean’s he e o ophic mic obial ac-
i i y on sedimen ing POC, and he p e iously epo ed
misma ch be ween o ganic ca bon supply and demand need
e ision [Rein hale e al., 2006; S einbe g e al., 2008;
Bal a e al., 2009; Bu d e al., 2010]. The la ge and ai ly
cons an s ock o buoyan , nsPOC needs o be conside ed as
a po en ially a ailable dynamic pool o POC o he e o o-
phic deep‐wa e mic obes. We ha e shown ha abou 12 ±
9–72 ± 53% o he amoun o PCD equi ed by mic o-
he e o ophs in he mesopelagic ocean is po en ially sup-
plied by deep‐wa e chemoli ho ophs ia DIC ixa ion.
The e o e, he appa en gap be ween he e o ophic o ganic
ma e demand and supply in he mesopelagic ealm migh
be subs an ially smalle han p e iously hough . Ne e he-
less, he ex e nal sou ces o his nsPOM s ill needs o be
cons ained. Figu e S3 (in Tex S1 o he auxilia y ma e ial)
illus a es he o e all con ibu ion o au o ophic p oduc ion
(AP) and sinking POC o PCD assuming a PP o 28 mmol C
m
−2
d
−1
and a R:ETS ange o 0.4–0.2, which based on he
abo e in o ma ion p obably ep esen s ealis ic a e age
alues o ou egion o s udy. The iew eme ging om
Figu e S3 is ha , in spi e o he signi ican con ibu ion o
AP o PCD a conside able ac ion o he PCD s ill needs o
be accoun ed by o he sou ces. Acco ding o ou obse a-
ions, we sugges ha con inuous la e al ad ec ion o
nsPOM om he con inen al ma gins may ep esen an
impo an ac ion o he missing ca bon espi ed in ou
egion o s udy. Fu u e esea ch should ocus on e ining he
nu i i e quali y o buoyan nsPOC e sus sedimen ing
POC. The e ined pa hways o o ganic ma e supply and
demand in he da k ocean migh allow us o a i e a a
mechanis ic unde s anding o deep ocean ood web s uc u e
and ac i i y and ul ima ely, should lead o imp o ed models
on he da k ocean’s ole in global ca bon cycling.
[15]Acknowledgmen s. We hank he cap ain and c ew o R/V Pelagia
o hei help du ing wo k a sea. A. Smi pe o med he leucine inco po a-
ion measu emen s. M. F. Mon e o, and M. Espino helped wi h he Elec on
T anspo Sys em analyses and I. J. Alonso‐González wi h he pa icula e
o ganic ma e measu emen s. Wa e masses we e iden i ied by H. M.
an Aken. This esea ch was suppo ed by a p edoc o al Fellowship o
he Spanish Minis y o Educa ion and Science (AP2005‐3932) o F.B., a
g an o he Ea h and Li e Science Di ision o he Du ch Science Founda-
ion (ALW‐NWO; ARCHIMEDES p ojec , 835.20.023) o G.J.H. and a
g an o he Spanish Minis y o Educa ion and Science o J. A. (Remolinos
Oceánicos y Deposiciones A mos é icas (RODA) p ojec ; CTM 2004‐
06842‐C03/MAR). The wo k was ca ied ou wi hin he ame o he EU
‘Ne wo ks o Excellence’Ma Be and Eu Oceans.
Re e ences
Agogué, H., M. B ink, J. Dinasque , and G. J. He ndl (2008), Majo g adi-
en s in pu a i ely ni i ying and non‐ni i ying A chaea in he deep No h
A lan ic, Na u e,456, 788–791, doi:10.1038/na u e07535.
Alonso‐González, I., J. A ís egui, J. C. Vilas, and A. He nández‐Gue a
(2009), Suspended POC anspo and consump ion in su ace and deep
wa e s o he Cana y Cu en egion: A box model s udy, Global Biogeo-
chem. Cycles,23, GB2007, doi:10.1029/2008GB003185.
An ia, A. N., e al. (2001), Basin‐wide pa icula e ca bon lux in he A lan-
ic Ocean: Regional expo pa e ns and po en ial o a mosphe ic CO
2
seques a ion, Global Biogeochem. Cycles,15, 845–862, doi:10.1029/
2000GB001376.
A ís egui, J., C. M. Dua e, S. Agus i, M. Do al, X. A. Al a ez‐Sagado,
and D. A. Hansell (2002), Dissol ed o ganic ca bon suppo o espi a-
ion in he da k ocean, Science,298, 1967, doi:10.1126/science.1076746.
A ís egui, J., C. M. Dua e, J. M. Gasol, and L. Alonso‐Sáez (2005a), Ac-
i e mesopelagic p oka yo es suppo high espi a ion in he sub opical
no heas A lan ic Ocean, Geophys. Res. Le .,32, L03608, doi:10.1029/
2004GL021863.
A ís egui, J., S. Agus í, J. J. Middelbu g, and C. M. Dua e (2005b),
Respi a ion in he mesopelagic and ba hypelagic zones o he ocean, in
Respi a ion in Aqua ic Ecosys ems,edi edbyP.A.delGio gioand
P. J. L. B. Williams, pp. 181–205, Ox o d Uni . P ess, Ox o d, U. K.
A ís egui, J., J. M. Gasol, C. M. Dua e, and G. J. He ndl (2009), Mic obial
oceanog aphy o he da k ocean’s pelagic ealm, Limnol. Oceanog .,
54(5), 1501–1529.
Bal a , F., J. A ís egui, J. M. Gasol, E. Sin es, and G. J. He ndl (2009),
E idence o p oka yo ic me abolism on suspended pa icula e o ganic
ma e in he da k wa e s o he sub opical No h A lan ic, Limnol.
Oceanog .,54(1), 182–193.
B ea, S. (2008), Ci culación, mezcla, en ilación y mine alización de las
aguas del A lán ico Su , Ph.D. hesis, Uni . o Vigo, Vigo, Spain.
Buessele , K. O., and P. W. Boyd (2009), Shedding ligh on p ocesses ha
con ol pa icle expo and lux a enua ion in he wiligh zone o he
open ocean, Limnol. Oceanog .,54(4), 1210–1232.
Buessele , K. O., e al. (2007), Re isi ing ca bon lux h ough he ocean’s
wiligh zone, Science,316, 567–570, doi:10.1126/science.1137959.
Bu d, A. B., and G. A. Jackson (2009), Pa icle agg ega ion, Annu. Re .
Ma . Sci.,1,65–90.
Bu d, A. B., e al. (2010), Assesing he appa en imbalance be ween geo-
chemical and biochemical indica o s o meso‐and ba hypelagic biolog-
ical ac i i y: Wha he @$#! is w ong wi h he p esen calcula ions o
ca bon budge ?, Deep Sea Res., Pa II, in p ess.
Chin, W‐L., M. V. O ellana, and P. Ve dugo (1998), Spon aneous assem-
bly o ma ine dissol ed o ganic ma e in o polyme gels, Na u e,391,
568–572.
Ch is ensen, J. P., T. G. Owens, A. H. De ol, and T. T. Packa d (1980),
Respi a ion and physiological s a e in ma ine bac e ia, Ma . Biol. Be lin,
55, 267–276, doi:10.1007/BF00393779.
del Gio gio, P. A., and C. M. Dua e (2002), Respi a ion in he open ocean,
Na u e,420, 379–384, doi:10.1038/na u e01165.
Hansell, D. A., C. A. Ca lson, D. J. Repe a, and R. Schli ze (2009),
Dissol ed o ganic ma e in he ocean, Oceanog aphy,22(4), 202–211.
Hansman, R. L., S. G i in, J. T. Wa son, E. R. M. D u el, and A. E. Ingalls
(2009), The adioca bon signa u e o mic oo ganisms in he mesopelagic
ocean, P oc. Na l. Acad. Sci. U. S. A.,106(16), 6513–6518, doi:10.1073/
pnas.0810871106.
He ndl, G. J., T. Rein hale , E. Tei a, H. . Aken, C. Ve h, A. Pe n hale ,
and J. Pe n hale (2005), Con ibu ion o A chaea o o al p oka yo ic
p oduc ion in he deep A lan ic Ocean, Appl. En i on. Mic obiol.,71,
2303–2309, doi:10.1128/AEM.71.5.2303-2309.2005.
Hopkinson, C. S., J ., and J. J. Vallino (2005), E icien expo o ca bon o
he deep ocean h ough dissol ed o ganic ma e , Na u e,433, 142–145,
doi:10.1038/na u e03191.
Ingalls, A. E., S. R. Shah, R. L. Hansman, L. I. Aluwiha e, G. M. San os,
E. R. M. D u el, and A. Pea son (2006), Quan i ying a chaeal communi y
au o ophy in he mesopelagic ocean using na u al adioca bon, P oc.
Na l. Acad. Sci. U. S. A.,103, 6442–6447, doi:10.1073/pnas.0510157103.
Jenkins, W. J. (1982), Oxygen u iliza ion a es in he No h A lan ic Sub-
opical Gy e and p ima y p oduc ion in oligo ophic sys ems, Na u e,
300, 246–248, doi:10.1038/300246a0.
Jenkins, W. J., and D. W. R. Wallace (1992), T ace Based In e ences o
New P ima y P oduc ion in he Sea, pp. 299–316, Plenum, New Yo k.
BALTAR ET AL.: DARK CO
2
FIXATION AND C DEMAND L09602L09602
5o 6

Ka l, D. M., G. A. Knaue , J. H. Ma in, and B. B. Wa d (1984), Bac e ial
chemoli ho ophy in he ocean is associa ed wi h sinking pa icles,
Na u e,309,54–56, doi:10.1038/309054a0.
Longhu s , A., S. Sa hyend ana h, T. Pla , and C. Ca e hill (1995), An
es ima e o global p ima y p oduc ion in he ocean om sa elli e adi-
ome e da a, J. Plank on Res.,17,1245–1271, doi:10.1093/plank /
17.6.1245.
Ma in, J. H., G. A. Knaue , D. M. Ka l, and W. W. B oenkow (1987),
VERTEX: Ca bon cycling in he no heas Paci ic, Deep Sea Res., Pa
A,34, 267–285, doi:10.1016/0198-0149(87)90086-0.
Rein hale , T., H. . Aken, C. Ve h, P. J. l. Williams, J. A is egui, C. Robinson,
P. Leba on, and G. J. He ndl (2006), P oka yo ic espi a ion and p oduc-
ion in he meso‐and ba hypelagic ealm o he eas e n and wes e n No h
A lan ic basin, Limnol. Oceanog .,51, 1262–1273.
Schneide , B., R. Schli ze , G. Fische , and E. M. Nö hing (2003), Dep h‐
dependen elemen al composi ions o pa icula e o ganic ma e (POM)
in he ocean, Global Biogeochem. Cycles,17(2), 1032, doi:10.1029/
2002GB001871.
She idan, C. C., C. Lee, S. G. Wakeham, and J. K. B. Bishop (2002),
Suspended pa icle o ganic composi ion and cycling in su ace and
midwa e s o he equa o ial Paci ic Ocean, Deep Sea Res., Pa I,
49, 1983–2008, doi:10.1016/S0967-0637(02)00118-8.
S einbe g, D. K., C. A. Ca lson, N. R. Ba es, R. J. Johnson, A. F. Michaels,
and A. H. Knap (2001), O e iew o he US JGOFS Be muda A lan ic
Time‐se ies S udy (BATS): A decade‐scale look a ocean biology and
biogeochemis y, Deep Sea Res., Pa II,48,1405–1447, doi:10.1016/
S0967-0645(00)00148-X.
S einbe g, D. K., B. A. Van Mooy, K. Buessele , P. W. Boyd, T. Koba i,
and D. M. Ka l (2008), Bac e ial s. zooplank on con ol o sinking
pa icle lux in he ocean’s wiligh zone, Limnol. Oceanog .,53(4),
1327–1338.
S oll, M. H. C., K. Bakke , G. H. Nobbe, and R. R. Haese (2001),
Con inuous‐ low analysis o dissol ed ino ganic ca bon con en in
seawa e , Anal. Chem.,73(17), 4111–4116, doi:10.1021/ac010303 .
J. A ís egui and F. Bal a , Facul ad de Ciencias del Ma , Uni e sidad de
Las Palmas de G an Cana ia, Campus Uni e si a io de Ta i a, E‐35017 Las
Palmas de G an Cana ia, Spain. ([email p o ec ed])
J. M. Gasol, Depa amen de Biologia Ma ina i Oceanog a ia, Ins i u de
Ciències del Ma , CSIC, Pg. Ma í im de la Ba celone a 37‐49, E‐08003
Ba celona, Spain.
G. J. He ndl, Depa men o Biological Oceanog aphy, Royal
Ne he lands Ins i u e o Sea Resea ch, NL‐1790 AB Den Bu g,
Ne he lands.
T. Rein hale and E. Sin es, Facul y Cen e o Ecology, Depa men o
Ma ine Biology, Uni e si y o Vienna, Al hans . 14, A‐1090 Vienna,
Aus ia.
BALTAR ET AL.: DARK CO
2
FIXATION AND C DEMAND L09602L09602
6o 6