reef water co system and carbon production of coral reefs

229

Reef Water CO2 System and Carbon Production of Coral Reefs:Topographic Control of System-Level Performance

Atsushi SUZUKI1 and Hodaka KAWAHATA1,2

1Institute for Marine Resources and Environment, National Institute of AdvancedIndustrial Science and Technology, Higashi 1-1, Tsukuba 305-8567, Japan

2Graduate School of Science, Tohoku University, Aoba, Sendai 908-8578, Japan

Abstract. The variations of seawater CO2 system and organic and inorganiccarbon production of coral reefs were investigated with respect to topographictypes and oceanographic settings. Because of dominant carbonate productionin coral reef ecosystems, most coral reefs are likely to act as a net or at least apotential CO2 source to the atmosphere. The comparison of the seawater CO2system parameters (pH, total alkalinity, dissolved inorganic carbon and partialpressure of CO2; pCO2) between a reef lagoon and the surrounding oceanallowed us to evaluate the system-level performance of the carbon cycle in theparticular reef system. Surface pCO2 in the lagoons of some atolls and barrierreefs in the western Pacific were consistently higher than those of their offshorewaters. The alkalinity decrease in the lagoon water was attributed to calcificationof reef organisms. Reef topography, especially residence time of lagoon water,affects the carbon budget of coral reefs to some extent. The offshore-lagoondifferences in pCO2 from several reefs showed a tendency to increase with thelonger residence time of reef water. Another important factor controllingcarbon turnover in coral reefs is proximity to land: terrestrial carbon andnutrient inputs were clearly observed in the northern Great Barrier Reef lagoonas well as a fringing reef of the Ryukyus. These coastal reefs serve as an activeCO2-releasing area due not only to calcification but also to degradation of land-derived carbon.

Keywords: coral reefs, carbon dioxide, carbon cycle, photosynthesis,calcification, metabolism, lagoon

1. INTRODUCTION

Coral reefs are recognized as major geological features of the earth’s surface,although they occupy only about 0.2% of the world’s ocean area (Smith, 1978).There are two important biogeochemical processes in a coral reef system:photosynthesis and calcification:

photosynthesis: CO2 + H

2O → CH

2O + O

2(1)

calcification: Ca2+ + 2HCO3 → CaCO

3 + H

2O + CO

2. (2)

Global Environmental Change in the Ocean and on Land, Eds., M. Shiyomi et al., pp. 229–248.© by TERRAPUB, 2004.

230 A. SUZUKI and H. KAWAHATA

The rapid calcification, which leads to reef growth and sediment formation, canbe attributed to a strong linkage to photosynthesis (Barnes and Chalker, 1990).

In recent decades, the role of coral reefs in the global biogeochemical cyclehas attracted much attention (Smith, 1978, 1981; Crossland et al., 1991; Gattusoand Buddemeier, 2000). According to the latest estimate of the global CaCO3budget, coral reefs contribute 32–43% of neritic CaCO3 production and 7–15%of global CaCO3 production (Table 1). The global contribution of coral reefsbecomes larger if we look at CaCO3 accumulation, because of the stability of reefcarbonates in the shallow marine environment. Sedimentary carbonates representthe largest reservoir of carbon on the earth and the fluctuation in global CaCO3budget influences atmospheric CO2 concentration. Carbonate shells of marinecalficifiers are expected to play a role of “primary neutralizers” of anthropogenicCO2.

In the early 1990s, there appeared to be some confusion in our understandingof the function of marine calcifying organisms, particularly in coral reef sciencecommunities (e.g. Kinsey and Hopley, 1991). Dissolution of carbonate (thereverse of the reaction in Eq. (2)), not precipitation, enhances the ocean’scapacity to absorb CO2 from the atmosphere. Ware et al. (1992) clearly stated thatcoral reefs generally work as sources of atmospheric CO2, based on the “0.6 rule”.Frankignoulle et al. (1994) reinforced the “0.6 rule” with examinations using aseawater CO2 system model and proposed a new index, Ψ (the released CO2/precipitated carbonate ratio).

Despite the comments of Ware et al. (1992), Kayanne et al. (1995) reportedthe diurnal change of pCO2 in Shiraho reef water of Ishighaki Island, theRyukyus, together with an estimate of high net community production (110mmol–1d–1) which can mask the CO2 released from carbonate production (110mmol–1d–1), and suggested the reef acts as a CO2 sink. While their result wascriticized for the great uncertainty in their estimation of the reef-offshoredifference of partial pressure of CO2 (pCO2), based on the light—pCO2 relationship

Table 1. Calcium carbonate flux estimates for different oceanic regions (after Iglesias-Rodriquezet al., 2002, with partial modification). Production was calculated from flux multiplied by area. Unitfor flux was converted from gC m–2y–1 in the original table to mmol m–2d–1.

Area(1012 m2)

Flux(mmol m–2d–1)

Production(PgC y–1)

Accumulation(PgC y–1)

NeriticCoral reefs 0.6 41.1 0.108 0.084Carbonate shelves 10 0.5–2.7 0.024–0.120 0.036Halimeda bioherms ? 82 0.02 0.02Bank/Bays 0.8 14 0.048 0.024Non-carbonate shelves 15 0.7 0.05 0.012Total 0.25–0.34 0.17

Slope 32 0.4 0.060 0.048Pelagic 283–300 0.3–0.6 0.41–1.1 0.1–0.144

Reef Water CO2 System and Carbon Production of Coral Reefs 231

obtained from relatively short-term observation (Gattuso et al., 1996a), theirfindings stimulated the subsequent debate on the sink/source problem of coralreefs (Kayanne, 1996). Suzuki et al. (1995) reported a relatively high P/R ratiofor a reef-flat community, on Shiraho reef again, supporting the conclusion ofKayanne et al. (1995). Kraines et al. (1996, 1997) also reported sink-typebehavior of a coral reef in Bora Bay, Miyako Island, which is adjacent to IshigakiIsland. In contrast to these results from the Ryukyu Islands of Japan, studies inMoorea, French Polynesia, and Yonge Reef of the northern Great Barrier Reef(GBR), showed that reef-flat communities are net sources of atmospheric CO2(Gattuso et al., 1993, 1995; Frankignoulle et al., 1996; Gattuso et al., 1996b).More recently, observations from mid-oceanic reefs (Majuro Atoll, Palau barrierreef and South Male Atoll), as well as the GBR, showed high pCO2 in reef lagoonscompared to that in the offshore area, confirming our current understanding thatmost coral reefs operate as sources of atmospheric CO2 (Kawahata et al., 1997,2000b; Suzuki et al., 1997; Suzuki and Kawahata, 1999). While measurements ofcommunity metabolism are considered significant in the coral reef CO2 source-sink debate (Gattuso et al., 1999), the problem has been recently examined fromsome different points of view, including a long-term geochemical perspectiveand the prediction of future response to global warming (Kleypas et al., 1999,2001).

Our knowledge of the carbon budget and cycling in coral reefs is still limitedand there remain many problems other than the CO2 sink/source issue. As thenumber of observations increases, differences in the mode of carbon cyclingamong coral reefs have become evident. The variations may be related totopographic types and oceanographic settings of individual reefs as well as theinfluence of human disturbance (Suzuki and Kawahata, 1999, 2003).

Here we re-examine published results on reef metabolisms obtained fromcoral reefs in the Indo-Pacific regions, including our own studies, in order tosurvey the latest progress on the CO2 sink/source issue and evaluate how and howmuch the topography and oceanographic settings influence the carbon turnoverof coral reefs. Special attention is paid to terrestrial influences on coastal reefs.

2. DATA SETS

2.1 Observations on marine CO2 system of coral reef waters

Coral reefs, which we re-examine in this study, include Shiraho Reef of theRyukyu Islands (Suzuki et al., 1995), Palau barrier reef (Kawahata et al., 1997),Majuro Atoll (Suzuki et al., 1997), South Male Atoll (Suzuki and Kawahata,1999), as well as the GBR (Kawahata et al., 2000b; Suzuki et al., 2001) (Fig. 1).In order to evaluate the CO2 system in reef water and to estimate organic andcarbonate carbon productions, temperature, salinity, pH, total alkalinity (AT) anddissolved inorganic carbon (DIC) were measured in discrete water samples.Details of measurements and calculations of pCO2 based on chemical equilibriumof CO2 system are described in the original papers mentioned above. In Palaureef, Majuro Atoll and the GBR, direct measurements of pCO2 were conducted


Tab

le 2

. S

umm

ary

of to

pogr

aphi

c fe

atur

es, c

arbo

n m

etab

olis

ms

and

air-

sea

CO

2 fl

ux f

or c

oral

ree

fsin

the

Ind

o-P

acif

ic r

egio

n. A

fter

Suz

uki

and

Kaw

ahat

a (2

003)

. N

et c

omm

unit

y ca

rbon

pro

duct

ion

(“N

et C

Pro

duct

ion”

in

the

tabl

e) i

s ex

pres

sed

in u

nits

of

mm

ol m

–2da

y–1 b

oth

for

orga

nic

carb

on(O

rg-C

) and

inor

gani

c ca

rbon

(CaC

O3)

pro

duct

ions

. Sym

bols

P/R

and

RO

I rep

rese

nt th

e ra

tio

of g

ross

com

mun

ity

prod

ucti

on to

res

pira

tion

and

the

rati

o of

net

org

anic

pro

duct

ion

to n

et in

orga

nic

carb

onpr

oduc

tion

, res

pect

ivel

y. S

ymbo

l δpC

O2

repr

esen

t the

dif

fere

nce

in s

eaw

ater

pC

O2

betw

een

lago

onw

ater

and

the

offs

hore

wat

er a

nd d

efin

itio

n is

giv

en b

y E

q. (3

) in

the

text

. Inf

luen

ce o

n δ p

CO

2 va

lues

caus

ed b

y th

e of

fsho

re-l

agoo

n di

ffer

ence

in te

mpe

ratu

re a

nd s

alin

ity

wer

e ne

glig

ible

for m

ost c

ases

.A

ster

isk

(*)

deno

tes

that

the

pro

duct

ion

rate

s w

ere

esti

mat

ed a

t sy

stem

lev

el i

nclu

ding

act

ive

reef

flat

com

mun

ity

in a

lago

on. P

rodu

ctio

n es

tim

ates

for

Bor

a B

ay w

as a

rea-

wei

ghed

ave

rage

bas

ed o

nth

e va

lues

for

ree

f fl

at a

nd l

agoo

n co

mm

unit

ies.

Ree

fS

cale

Are

aM

ean

dept

hR

esid

ence

tim

eN

et C

pro

duct

ion

P/R

RO

IpC

O2

(µat

m)

δCO

2Fl

uxR

efer

ence

of

pCO

2 va

lues

(km

)(k

m2 )

(m)

Org

-CC

aCO

3In

side

Out

side

(µat

m)

Ato

llFa

nnin

g A

toll*

15 ×

15

125

532

d−0

.427

1−0

.01

400

370

+30

sour

ceSm

ith

and

Pesr

et (

1974

)

Can

ton

Ato

ll*11

× 1

111

06.

250

d3

14

0.01

290−

400

330

+15

sour

ceSm

ith a

nd J

okie

l (19

78)

Chr

istm

as I

s.*

15 ×

20

180

310

−15

d6

2.5

>12.

422

031

0−8

0si

nkSm

ith e

t al.

(198

4)

Maj

uro

Ato

ll*39

× 1

032

939

15 d

827

0.

2936

834

5+2

3so

urce

Suzu

ki e

t al.

(199

7)

Sout

h M

ale

Ato

ll32

× 1

8

68

368

362

+6

sour

ceS

uzuk

i an

d K

awah

ata

(199

9)

Ruk

an-s

ho*

1.7

× 1

∼

3

7210

81.

50.

67

neut

ral

Ohd

e an

d va

n W

oesi

k (1

999)

Bar

rier

ree

fT

iahu

ra B

arri

er R

.∼1

km

wid

e

6

h75

186

1.2

0.4

so

urce

Gat

tuso

et a

l. (1

993)

Yon

ge R

eef

∼2 k

m w

ide

33

253

1.1

0.1

so

urce

Gat

tuso

et a

l. (1

996b

)P

alau

Bar

rier

Ree

f13

km

wid

e

19

414

366

+46

sour

ceK

awah

ata

et a

l. (1

997)

Nor

ther

n G

BR

∼80

km w

ide

∼

40

33

931

1+2

9so

urce

Suzu

ki e

t al.

(200

1)So

uthe

rn G

BR

>100

km

wid

e

∼40

326

314

+12

sour

ceSu

zuki

et a

l. (2

001)

Fri

ngin

g re

ef (

Ryu

kyu

Isla

nds,

Jap

an)

Shi

raho

Ree

f∼1

km

wid

e

1.1

4 h

110

100

1.3

1.1

157−

521

322

−7si

nkK

ayan

ne e

t al

. (1

995)

Shi

raho

Ree

f4−

8 h

3612

71.

10.

3

H

ata

et a

l. (2

002)

Bor

a B

ay*

1 ×

0.6

0.35

∼32−

4 h

164

132

1.4

1.2

<0

sink

Kra

ines

et

al.

(199

7)


using a shipboard nondispersive infrared (NDIR) gas analyzer system with ashower-type air-sea equilibrator onboard R/V Hakurai-Maru (Kawahata et al.,1997; Kawahata et al., 2000b; Suzuki et al., 2001).

Other than the results described above, information on reef metabolism at thecommunity and/or system level and pCO2 values from some coral reefs in theIndo-Pacific regions was cited from published reports (Table 2). Additionalgeomorphological data on each coral reef are also shown in Table 2. Coral reefsin the table can be divided into three categories with respect to reef morphology:(1) atolls, (2) barrier reefs, and (3) fringing reefs (Fig. 2).

2.2 Topographic characteristics of coral reefs examined

2.2.1 AtollsMajuro Atoll (7°N, 171°E; Fig. 3A) in the Marshall Islands, central Pacific,

has a semi-enclosed lagoon with a maximum depth of 67 m. The lagoon has onlyone deep passage with a maximum depth of 46 m. The topography effectivelyisolates the wide interior lagoon from the outside ocean. South Male Atoll (4°00′N, 73°25′ E; Fig. 3B) of the Maldive Islands, northern Indian Ocean, is an oval-shaped atoll. The lagoon water is connected with the offshore water through manydeep channels between shallow reef flats.

Fanning Atoll (Smith and Pesret, 1974) and Canton Atoll (Smith and Jokiel,1978) are also oceanic atolls in the central Pacific. The coral reef of ChristmasIsland, central Pacific, has a land-locked shallow lagoon (2–4 m deep) and Smithet al. (1984) conducted the survey of this lagoon. Rukan-sho (Rukan reef) is asmall-scale atoll off Okinawa Island of the Ryukyus (Ohde and van Woesik,1999).

2.2.2 Barrier reefsIn the Palau (Belau) Archipelago of the Caroline Islands (7°N, 134°E; Fig.

3C), a wide lagoon (Southern Lagoon; Maragos and Cook, 1995) with a maximumdepth of 55 m exists between the islands and a barrier reef platform (hereafterPalau barrier reef lagoon). A narrow channel at the northern end of the lagoon is

Fig. 1. Locality of coral reefs examined in this study. Although Rucan reef is classified as an atollhere, more appropriately it can be categorized as a platform reef (Ohde and van Woesik, 1999).


the only deep passage (maximum depth 65 m) along the reef platform extendingfor approximately 86 km.

The GBR extends approximately 2600 km along the eastern coast of Australia,covering a wide range in latitude (15 degrees; Fig. 1). The major portion of totalshelf area lies between 30 m and 40 m isobaths. Despite the name of “barrier”, theouter reefs are not continuous and the continuity of outer reefs is defined as“linear density” of reefs along the shelf break (Pickard et al., 1977). There is astriking contrast between the northern and southern GBR. The northern GBR (9–16°S) is a continuous stretch with the linear density reaching 90%, while thesouthern GBR (>16°S) exhibits apparently low values around 10%. In thenorthern GBR region, the wet-tropical catchments supply a large proportion oftotal runoff to the GBR throughout the year (Mitchell and Furnas, 1996). On theother hand, two large dry-catchment rivers (Fitzroy and Burdekin Rivers) showclear seasonality in discharge volumes. During the research cruise conducted inthe dry season of 1996, freshwater input was recognized as a salinity decrease inthe inshore area of the northern GBR lagoon, while no influence of freshwaterinput was detected in the southern GBR region (Suzuki et al., 2001). In this studywe examine the data from the northern and southern parts of the GBR separately.

2.2.3 Fringing reefShiraho Reef, located on the east coast of Ishigaki Island, southern Ryukyu

Islands, Japan, is a well-developed fringing-type reef without a deep lagoon. Thereef flat is about 800 m in width and is composed of four topographic sub-unitsincluding outer reef slope, reef crest, inner reef flat, and moat (Nakamori et al.,1992). The moat is a trough-like depression (shallow lagoon) along the coast witha depth that never exceeds 3 m (Fig. 2). Suzuki et al. (1995) measured communitymetabolism on the inner reef flat while Kayanne et al. (1995) reported diurnal

Fig. 2. Schematic diagram of morphological types of coral reefs including fringing reefs, barrierreefs and atolls (A). Topographic zonation of fringing reefs (B) and barrier reefs and atolls (C) areshown in the lower panels. A shallow depression of a fringing reef is usually referred as a moat(shallow lagoon), of depth shallower than 2–3 m.

Fringing Reef Barrier Reef Atoll

Channel Lagoon MotuMoatFore-reef

slope

LagoonMoat

(A)

(B) (C) Reef flat

Reef flatMotu

Fringing Reef Barrier Reef and Atoll

Reef crestReef crest


pCO2 variations at a deeper station near the boundary between the moat and theinner reef flat. More recently, Hata et al. (2002) conducted a comprehensive studyof carbon fluxes at Shiraho Reef in Ishigaki Island in September 1998.

3. RESULTS AND DISCUSSION

3.1 A definition of the coral reef sink-source problem

We need to start by clarifying the definition of the problem. Our concern forthe CO2 performance of coral reefs has several components. The mode change ofcoral reef metabolisms due to anthropogenic CO2 increase in the atmosphere willbe answered by the comparison of reef performance between pre- and post-

Fig. 3. Spatial distributions of pCO2 in surface waters around oceanic atolls and barrier reefs.Average value of the offshore-lagoon difference in pCO2 (δpCO2) at in-situ temperature and salinityobserved in each reef is shown in each panel. Majuro Atoll (A) and South Male Atoll (B) representoceanic atolls while barrier reefs include Palau barrier reef (C) and the northern and southern GBR(D and E). Three mid-oceanic atolls and barrier reefs (A, B and C) have lagoons of similar size, butwhich are different in the degree of closure to the surrounding sea.


industrial revolution periods. The future response of carbon and carbonateproduction in coral reefs to global warming and the following sea-level rise hasattracted much attention (Kinsey and Hopley, 1991; Kleypas et al., 1999, 2001).The role of coral reefs over a geological time scale has also been discussedelsewhere (Kleypas, 1997; Gattuso et al., 1999). Although these topics arecertainly interesting, here we focus on the contribution of the coral reef ecosystemto the global carbon cycle in a current/natural state within a relatively short timescale. Some previously published reports employed a similar definition of theproblem to that adopted here (Kinsey and Hopley, 1991; Ware et al., 1992;Gattuso et al., 1999).

Among studies focusing on the influence of coral reef performance on theglobal carbon cycle in a present state within a relatively short time scale, twodifferent approaches have been employed. Ware et al. (1992), as well as Kinseyand Hopley (1991), focused on carbonate production of coral reefs, ignoringorganic carbon production because net organic carbon production is generallybelieved to be small and the P/R ratio at community level is almost 1 (Crosslandet al., 1991). On the other hand, Gattuso et al. (1999) stressed the significance ofcommunity metabolism measurements in the coral reef CO2 source-sink debate.We also consider the importance of organic carbon metabolisms, as well ascarbonate production, based on the following two viewpoints. First, the organiccarbon production process in reef systems is somewhat different from theoffshore processes, especially with respect to nutrient budgets. The high C:N:Pratio of benthic plants in coral reefs (550:30:1; Atkinson and Smith, 1983)relative to plankton (106:16:1; Redfield et al., 1963) allows reefs to produce moreorganic matter per unit delivery of nutrients than can be produced by adjacentplankton communities (Crossland et al., 1991). The high biomass/productionratio of coral reefs compared to the offshore ecosystems is another importantcharacteristic of coral reef systems with respect to carbon storage (Smith, 1981).Second, our knowledge of the fate of organic carbon produced in coral reefs is stilllimited in terms both of the export to the surrounding ocean and sedimentaryburial in a lagoon bottom (see Hata et al., 1998, 2002), which may not allow usto neglect the contribution of organic carbon production as trivial. According tothe viewpoints described above, we start the examination by taking organiccarbon production into account.

3.2 ROI: an index for sink/source behavior

The sink/source behavior of coral reefs is primarily controlled by the balancebetween photosynthesis and calcification, which shift the chemical equilibriumof oceanic CO2 system in opposite directions (Eqs. (1) and (2)). Photosynthesisdecreases pCO2 while calcium carbonate production raises pCO2. Suzuki (1998)examined seawater pCO2 change caused by photosynthesis and calcification andproposed the molar ratio of organic to inorganic production (ROI) as a criterion ofthe CO2 sink/source behavior of marine metabolism. In this study, we apply theindex ROI to evaluate the sink/source performance of each coral reef in the datasets.


Fig

. 4.

Met

abol

ic e

ffe c

ts o

n si

nk/s

ourc

e po

tent

ial o

f se

a wa t

e r w

ith

resp

e ct t

o th

e a t

mos

phe r

ic C

O2.

(A)

Cha

nge s

in

pCO

2 ( δ

pCO

2) c

a use

d by

the

va r

ious

ra t

e s o

f ph

otos

ynth

e sis

and

ca l

c ifi

c ati

on.

Ca l

c ula

tion

s a r

e c o

nduc

ted

for

sali

nity

35

a nd

25°C

sup

posi

ng in

itia

l con

diti

on o

f se

a wa t

e r is

234

6µm

ol k

g–1 in

tota

l alk

a lin

ity,

200

0 µm

ol k

g–1 in

DIC

, and

345

µa t

m in

pC

O2.

Sha

ded

a re a

indi

c ate

sth

e re

gion

whe

re p

CO

2 in

c re a

sed

by m

e ta b

olis

m. (

B)

Diu

rna l

me t

a bol

ism

cyc

les

a nd

the

c rit

e ria

for

sink

/sou

rce

beha

vior

. (C

) S

e dim

ent

c om

posi

tion

as

a n i

ndex

for

sin

k/so

urc e

be h

a vio

r of

ove

rlyi

ngw

a te r

wit

h re

spe c

t to

CO

2 exc

hang

e be

twe e

n th

e se

a wa t

e r a

nd th

e a t

mos

phe r

e . S

hade

d a r

e a in

dic a

tes

sedi

men

t c o

mpo

siti

ons

whe

re p

CO

2 is

exp

e cte

d to

inc

rea s

e by

se d

imen

tary

pa r

tic l

e fo

rma t

ion.


Figure 4A displays changes in pCO2 (δpCO2) caused by various combinationsof net organic production (=photosynthesis – respiration) and net inorganicproduction (=calcification – dissolution) in seawater with a salinity of 35 at 25°C.This figure indicates that pCO2 decreases (δpCO2 < 0, therefore, sink-typebehavior) when net organic production is larger than approximately 0.6 times netinorganic carbon production (ROI > 0.6). In the present study, the symbol ROI

0

denotes the critical ratio for δpCO2 = 0.The critical value of approximately 0.6 for ROI

0 is almost identical to the Ψvalue (molar ratios of released CO2 to precipitated CaCO2) proposed byFrankignoulle et al. (1994). Therefore, ROI can be recognized as a differentexpression of Ψ, suggesting that the degassing and absorption processes of CO2in the air-sea interface are identical to photosynthetic CO2 uptake from seawaterand respiratory CO2 release of aquatic organisms, respectively, with respect tothe effects on the chemical equilibrium in seawater.

A similar analogy can be applied to the chemical compositions of reefsediments, if the sediments store the products of past reef metabolism. Accordingto Buddemeier (1996), a reef that is an atmospheric CO2 sink has to deposit morethan 12 weight % organic matter (6 weight % organic carbon) in reef sediments.This criterion can be extended to a three-component system shown in Fig. 4C.

3.3 Temporal and spatial scale of the problem

In order to apply the index ROI to coral reef metabolism, we need to examinethe scales of discussion in time and space. Usually we discuss the sink/sourcebehavior of coral reef organisms and/or communities on the daily or yearly basis.The concept and terminology of daily metabolism are shown in Fig. 4B, whichindicates the importance of estimating net organic and inorganic production fora certain period in order to determine the CO2 sink/source behavior of organismsand/or communities.

The sink/source debate in the early 1990s mostly focused on the “coral reefflat”. There are two methods for measuring metabolisms of reef communities:flow respirometry and the slack-water method. Flow respirometry has beenemployed on a reef flat community under unidirectional flow by many researchers,including the pioneering work by Odum and Odum (1955). The slack-watermethod was proposed by Kinsey (1978) and has been widely used in enclosedconditions on reef flats or moats during low tide.

Gattuso et al. (1993, 1995) reported that the reef flat community of thebarrier reef of Moorea is a net source of atmospheric CO2, based on daily netorganic and inorganic production measurements using flow respirometry. Bycontrast, Kayanne et al. (1995) and Suzuki et al. (1995) employed the slack-watermethod to measure reef-flat metabolism in Shiraho Reef, concluding that the reefflat community showed a sink-like behavior.

Contradictory results are, at least partly, a reflection of the large errors inmeasuring the community metabolism rate of coral reefs. Difficulties in productionestimates are caused by the relatively short residence time of water in fringing and


barrier reefs. As an example, the diurnal pattern of pH in seawater on a fringingreef is shown in Fig. 5A. Dual modulation on variations of pH and other CO2parameters due to tide and light cycles makes the measurements of communitymetabolism difficult. The daily budget estimate cannot escape a large uncertaintycaused by the integration of short-term metabolism rates.

On the other hand, deep lagoons of oceanic atolls and barrier reefs have arelatively long residence time of several weeks to months (Table 2). Consequently,the diurnal changes of pH in both the lagoon and the offshore waters are usuallysmall (Fig. 5B). In this case, a more reliable approach to examining the CO2 sink/source behavior of coral reefs can be successfully applied to atolls and barrierreefs (Kawahata et al., 1997; Suzuki et al., 1997). These studies compared thepCO2 differences (δpCO2) between the lagoon water (pCO2,L) and the offshorewaters (pCO2,O) of mid-oceanic reefs, such as Majuro Atoll and Palau barrierreef:

Fig. 5. Temporal variations of pH in the surface seawater on Shiraho fringing reef (A) and MajuroAtoll lagoon (B). Panel (A) represents the diurnal pH variation of reef water at a station located onthe inner reef flat (Suzuki et al., 1995). Hatched areas indicate the periods of slack water during lowtide. Abrupt changes of reef water pH were caused by the influx of the offshore water while pHincrease and decrease during the daytime and nighttime slack-water periods, respectively, weredriven by reef metabolisms. In the panel (B), all pH values at many stations were plotted against timeincluding repeated pH measurements at fixed stations R1 and R2 in the lagoon and the offshore,respectively. The pH at St. R1 in the lagoon was 8.289 on September 10 and 8.290 on September 12.The difference between the pH values on September 12 and 13 at St. R2 offshore was again only0.001. Repeat measurements indicate that day-to-day pH variations of the lagoon and offshore watermasses were small during the observation period.

7

.

5

0

8.00

8.50

9.00

p H

a

t

2

5

C

601

81

2 D e

p

t

h

(

c

m

)

Sep. 20 Sep. 21

D

a

y N

i

g

h

t D

a

y

D

e

p

t

h

(A) Shiraho fringing reef

0

1

0

0

200

12

Slack-w ater period

S

u

p

p

o

s

e

d

o

f

f

s

h

o

r

e

l

e

v

e

l

pH

p H

a

t

2

5

C

S

e

p

.

1

0 Sep. 11 Sep. 12 S

e

p

.

1

3 Sep. 14

Offshore mean = 8.300

L

a

g

o

o

n

East

West

Offshore

8.22

8

.

2

4

8

.

2

6

8.28

8

.

3

0

8.32

8.289

8.290

8.299 8.298

St . R1

S

t

.

R2

(B) Majuro Atoll DayNight


δpCO2 = pCO

2,L – pCO

2,O. (3)

Since water and dissolved materials typically enter atolls across the reef flat fromthe ocean, the composition of lagoon water reflects metabolic activity, not onlyin the lagoon but also on the reef flats and knolls. This condition favors theevaluation of the net effect of reef metabolisms on reef water-atmosphere CO2exchange. Smith (1988) demonstrated the ability of a similar approach toexamine nutrient budgets of confined aquatic ecosystems. This strategy can alsobe applied to marine CO2 systems to examine the sink/source behavior of coralreefs at system level.

3.4 Sink/source behavior of coral reefs

In this section we test the suitability of the index ROI for evaluating the sink/source performance using the data sets (Table 2). Can sink/source behavior of theparticular coral reef be predicted based on a system-level ROI value?

The first example is Majuro Atoll (Suzuki et al., 1997), the net organic andinorganic carbon production of which was re-calculated as 8 mmol m–2d–1 and 27mmol m–2d–1, respectively (Fig. 6), supposing a residence time of 15 days, asreported by Kraines et al. (1999). Because the ROI value is 0.29, which is less thanthe critical value of 0.6, it is predicted that the atoll shows positive δpCO2 (CO2source). The average δpCO2 values observed in Majuro Atoll was around +48µatm and the prediction agrees with the result of pCO2 observations (Fig. 3A).

Net organic and inorganic carbon production reported at community- andsystem-levels, together with observed δpCO2 values, are plotted in Fig. 7. Otherthan Majuro Atoll, Fanning Atoll and Canton Atoll in the category of atolls inTable 2 showed considerably smaller ROI values than the critical ratio. HighδpCO2 was evident for those atolls, indicating consistency with ROI-basedprediction.

Fig. 6. The DIC and total alkalinity budgets for Majuro Atoll (modified from Suzuki et al., 1997).Concentrations of AT and DIC are normalized at offshore salinity (S = 33.69) and represented by theunit of µmol L–1. Production and inputs are shown in mol day–1 corresponding to a residence timeof 15 days reported by Kraines et al. (1999). Net organic and inorganic carbon production wasestimated to be 8 mmol m–2d–1 and 27 mmol m–2d–1, respectively, supposing the total area of the innerreef flat and the lagoon (329 km–2).


While most community- and system-level estimates are plotted in the“source” region (ROI < 0.6), some exceptions are found. Christmas Island lagoonindicated higher ROI values than the critical ratio and a negative δpCO2 value wasindeed observed (CO2 sink). The phosphorus input from guano deposits wasexpected to stimulate active organic production in the lagoon, while low salinitydue to fresh water input may have partly contributed to the extremely low pCO2in lagoon water (Smith et al., 1984). A metabolic study of Rukan reef reported 0.7for ROI, which is close to the critical value (Ohde and van Woesik, 1999).Therefore, the community metabolism is predicted to be almost neutral withrespect to air-sea CO2 flux. This prediction does not contradict the conclusion ofthe original authors.

Two fringing reefs, Shiraho Reef (Kayanne et al., 1995) and Bora Bay(Kraines et al., 1997) in the Ryukyu Islands, indicate higher ROI values than thecritical ratio (ROI > 0.6) and a negative δpCO2 value was indeed reported fromShiraho Reef (CO2 sink) by the authors. The relationship between ROI andobserved δpCO2 of these reefs is consistent with the hypothesis, even when thosereefs showed a CO2-sink behavior. Is CO2-sink-like performance common for thefringing type of reefs? We cannot exclude the possibility that some fringing reefshave higher ROI values than the critical ratio due to large contribution from anactive reef flat community, where high ROI values can be expected. On the otherhand, the influence on reef metabolism of fertilizer used on farmland has been

R O

I

=

0

.

6

sink

3002001 0

00-

1

0

0

-

1

0

0

0

100

2

0

0

3

0

0

I

n

o

r

g

a

n

i

c

c

a

r

b

o

n

p

r

o

d

u

c

t

i

o

n(

m

m

o

l

m-

2

d-

1 )

O

r

g

a

n

i

c

c

a

r

b

o

n

p

r

o

d

u

c

t

i

o

n(

m

m

o

l

m-

2

d-

1)

Shiraho1 (-7)

C

h

r

i

s

t

m

a

s

(

-

8

0

)Canton (+15)

Fanning (+30)

s o

u

r

c

e

M

a

j

u

r

o

(

+

2

3

)

T

i

a

h

u

r

a

(

+

)

Bora Bay (-)

RukanS

h

i

r

a

h

o2

S

y

s

t

e

m

-

l

e

v

e

lR

e

e

f

f

l

a

t

c

o

m

m

u

n

i

t

y

L

e

g

e

n

d

Y

o

n

g

e

(

+

)

Fig. 7. Community- and system-level estimates on net organic and inorganic carbon productionlisted in Table 1. Values in parenthesis represent the offshore-lagoon difference in pCO2 (δpCO2;µatm) at in-situ temperature and salinity. Reefs with only signs of minus and plus in parenthesisrepresent the trend in observed δpCO2 although exact values are not available. A line, whichrepresents ROI = 0.6, is shown in the diagram. The line is a critical boundary for constant pCO2 fornormal surface seawater with salinity of 35 at 25°C (Suzuki, 1998). Labels Shiraho1 and Shiraho2

denote the results of Kayanne et al. (1995) and Hata et al. (2002), respectively. Note that estimateson Shiraho Reef (reef flat community; Kayanne et al., 1995), Bora Bay reef (total system) andChristmas Island (shallow lagoon community) are plotted in “sink” area of the diagram.


pointed out as a possible cause of the relatively high organic production in the reefflat community of Bora Bay as well as Shiraho Reef (Kraines et al., 1997).

More recently, Hata et al. (2002) conducted a comprehensive study ofcarbon fluxes, including community metabolism and carbon export in the form ofdissolved and particulate matter at Shiraho Reef in September 1998. Accordingto their estimates, ROI can be calculated to take a relatively small value of 0.3 forthe reef flat community, suggesting a CO2-source-like behavior. Because ShirahoReef is the site where negative δpCO2 observation was reported previously(Kayanne et al., 1995), the implication of the results for the CO2 sink/sourcedebate is important. The study, however, was done during the recovery stage fromthe 1998 mass bleaching event in the region. Special caution may thus be neededwhen discussing this event.

In conclusion, the observations of δpCO2 listed in Table 2 agree well with thetheoretical prediction that ROI can be used as an index of sink/source behavior(Fig. 7). Most community- and system-level estimates appeared to support a CO2-source (ROI < 0.6). Although some coral reefs showed sink-like metabolisms (ROI> 0.6), some regional effects, which enhance organic carbon production, havebeen pointed out as a likely cause of their CO2-sink type behavior.

3.5 Topographic control on CO2 system in coral reefs

The observed δpCO2 values varied from reef to reef (Fig. 3). Although theratio ROI is a primary factor determining the sink-source behavior of a particularreef system, there must be another factor controlling the magnitude of δpCO2values. In addition, there appears to be a slightly increasing trend in δpCO2 withlonger reef water residence time (Table 2). Open atolls and fringing reefs withshort water residence times tend to show small or even negative δpCO2 values.In other words, the slower the renewal of the water is, the higher is pCO2 in reefwater.

Difference of carbon turnover in coral reefs due to residence time variationcan be seen in AT-DIC diagrams for South Male Atoll and Palau Barrier Reef (Fig.8). The graphic approach to carbon cycle study of coral reefs using an AT-DICdiagram was illustrated by Suzuki and Kawahata (2003). Most of the lagoonsamples of both reefs are plotted around a “calcification line” in the AT-DICdiagrams, suggesting predominant carbonate production in these reefs withnegligible excess organic carbon production. In detail, the distance between theoffshore and lagoon clusters in the diagram along the calcification line is muchlarger for Palau barrier reef than South Male Atoll, reflecting the largercompositional change of Palau lagoon water from the surrounding ocean.

The lagoon of South Male Atoll has an open nature, being connected to thesurrounding ocean by numerous deep channels (30–70 m deep), which are muchdeeper than typical Pacific atoll channels. In contrast to scanty reef developmentin typical Pacific atoll lagoons, the flourishing of branching corals on knolls inthe lagoon of Maldivian atolls (Kohn, 1964) is probably related to the large flowof oceanic water through the broad, deep channels. Thus, the small changes inCO2 system parameters including AT, DIC and pCO2 values between the offshore


and the lagoon can be attributed to dilution due to the high flushing rate of thelagoon. On the other hand, Palau barrier reef lagoon shows a higher degree ofclosure than South Male Atoll, having only one deep entrance channel. As theresidence time increases, more time becomes available for calcification. Theobserved decreases in AT and DIC in the lagoon compared to the offshore waterwere greater in Palau lagoon than in South Male Atoll lagoon. This evidence isconsistent with the observed difference of δpCO2 between the two reefs.

3.6 Terrestrial influence on coastal coral reefs

While oceanic reefs, such as South Male and Majuro atolls, have noterrestrial influence from land, the GBR and Shiraho Reef are closely locatedalong the coast. Alteration of the mode of carbon cycling can be expected in thesecoastal reef systems. Indeed, we found significant differences between the AT-DIC diagrams of the northern and southern GBR lagoons (Fig. 9), with aremarkable difference between them in terms of water, sediment and nutrientinput from land.

In the AT-DIC diagram for the southern GBR lagoon (Fig. 9A), most of thelagoon samples of both reefs are plotted around a “calcification line”, suggestingpredominant carbonate production with negligible excess organic carbonproduction. By contrast, overall displacement of the lagoon-water calcificationline from the offshore compositions can be seen along the increasing DIC trend(Fig. 9B). As described in Suzuki et al. (2001), this indicates the presence of netexternal carbon inputs to the lagoon other than surface oceanic exchange.Possible processes which increase DIC of the system are oxidation of organic

Fig. 8. Total alkalinity-DIC plots for two oceanic reef lagoons including South Male Atoll (A) andPalau barrier reef (B). Normalized values at constant salinities are plotted in order to remove theinfluences from rainwater input and evaporation. Major metabolic and biogeochemical processes areshown in each panel. Most dominant process in coral reefs is calcification and a calcification pathis represented on the AT-DIC diagram by a line of slope 2.

2 0

2

01

9

7

01

9

2

01

8

7

01

8

2

0

C a l

c

i

f

i

c

a

t

i

o

n

2 0

0

01

9

5

01

9

0

01

8

5

01

8

0

0

2 1

0

0

2 1

5

0

2 2

0

0

2 2

5

0

2 3

0

0

( B

)

P

a

l

a

u

b

a

r

r

i

e

r

r

e

e

f

N

o

r

m

a

l

i

z

e

d

a

t

S

=

3

4

.

5

O

f

f

s

h

o

r

eL

a

g

o

o

n

C a l

c

i

f

i

c

a

t

i

o

n

P

h

o

t

o

s

y

n

t

h

e

s

i

s

C a l

c

i

f

i

c

a

t

i

o

n

+ D e

g

a

s

s

i

n

g

D

e

g

a

s

s

i

n

g

R

e

s

p

i

r

a

t

i

o

n

( A

)

S

o

u

t

h

M

a

l

e

A

t

o

l

l

N

o

r

m

a

l

i

z

e

d

D

I

C

(

m

o

l

k

g-

1)

N

o

r

m

a

l

i

z

e

d

a

t

S

=

3

5

.

02

1

7

0

2

2

2

0

2

2

7

0

2

3

2

0

2

3

7

0

O

f

f

s

h

o

r

e

L

a

g

o

o

n

N

e

a

r

r

e

e

f

Nor

mal

ized

AT (

mol

kg-

1 )

C a l

c

i

f

i

c

a

t

i

o

n

l

i

n

e

R

e

s

i

d

e

n

c

e

t

i

m

e

W

e

l

l

f

l

u

s

h

e

d S

t

a

g

n

a

n

t


matter and DIC supply from external sources such as atmospheric input, rainfalland river discharge. Suzuki et al. (2001) concluded that the possible externalsource of carbon into the GBR lagoon is terrestrial input, including directdischarge of DIC in freshwater and oxidation of land-derived organic matter.Calcification, together with DIC inputs from terrestrial sources, serves a majorrole in the net carbon cycle of the northern GBR lagoon system. No features ofexcess organic carbon production can be seen.

Do nutrient inputs from land stimulate organic carbon production in a reefarea? Suzuki et al. (2001) estimated the riverine C:N:P molar ratio in the GBRregion as 1120:30:1, based on their carbon budget calculation and annual riverinenitrogen and phosphorous input estimated by Furnas et al. (1995). These ratiosare much higher than the Redfield ratio for plankton (106:16:1; Redfield et al.,1963), and the ratio of benthic plants in coral reefs (550:30:1; Atkinson andSmith, 1983). Therefore, even if all discharged nutrients are utilized by biologicalprocesses of both planktonic organisms and benthic communities, surplus carbonmust remain in the reef water as oxidized DIC. Although nutrient inputs maystimulate organic carbon production in a reef area, net oxidation of organic matterrather than net organic carbon fixation would be expected in the GBR lagoon.

A similar influence was also recognized in the Shiraho fringing reef of theRyukyu Islands (Kawahata et al., 2000a). Land derived freshwaters, includingriver water and groundwater, make a relatively large contribution to the reef’scirculation system. These terrestrial waters exhibit extremely high pCO2, up to6,400 µatm, reflecting enrichments in AT and DIC due to dissolution of carbonaterocks and decomposition of organic matter in the soil of subtropical island. Theseterrestrial influences may enhance CO2 degassing from the coastal zone. While

Fig. 9. Normalized total alkalinity-DIC plots with corresponding pCO2 (µatm) contours for thesouthern and northern GBR lagoons (A and B, respectively). Overall displacement from the idealcalcification line from the offshore composition towards DIC high in panel (B) can be attributed toexternal carbon inputs to the northern GBR lagoon other than oceanic exchange.

C a l

c

i

f

i

c

a

t

i

o

n

C a l

c

i

f

i

c

a

t

i

o

n

2 5 0

3 0 0

3 5 0

4 0 0

2 5 0

3 0 0

3 5 0

4 00

n

o

r

m

a

l

i

z

e

d

a

t

S

=

3

5

1 9

8

01

9

6

01

9

4

01

9

2

01

9

0

01

8

8

0

2

2

2

0

2

2

4

0

2

2

6

0

2

2

8

0

2

3

0

0

2

3

2

0( B

)

N

o

r

t

h

e

r

n

G

B

R

l

a

g

o

o

n

1 9

8

01

9

6

01

9

4

01

9

2

01

9

0

01

8

8

0

2

2

2

0

2

2

4

0

2

2

6

0

2

2

8

0

2

3

0

0

2

3

2

0

O

f

f

s

h

o

r

eL

a

g

o

o

nO

f

f

s

h

o

r

eL

a

g

o

o

n

n

o

r

m

a

l

i

z

e

d

a

t

S

=

3

5

( A

)

S

o

u

t

h

e

r

n

G

B

R

l

a

g

o

o

n

O

x

i

d

a

t

i

o

nD

I

C

i

n

p

ut

T

e

r

r

e

s

t

r

i

a

l

i

n

f

l

u

e

n

c

e

L

i

t

t

l

e Ev

i

d

e

n

t

Nor

mal

ized

AT (

mol

kg-

1 )

N

o

r

m

a

l

i

z

e

d

D

I

C

(

m

o

l

k

g-

1)


sink-type behavior has been reported for reef flats of fringing reefs, the influenceof terrestrial inputs on carbon turnover of the entire fringing reef system remainsa subject for future discussion.

The future change of coral reef calcification is currently a subject ofdiscussion with respect to global warming and the accompanying changes ofcarbonate saturation state in seawater (Kleypas et al., 1999, 2001). The recentdegradation of coastal reef ecosystems due to anthropogenic impacts is also verysevere. In addition, recurrent mass bleaching events, which can be at least partlyattributed to the global warming, may result in a decrease in organic productionand fishery yield in coral reefs (Hughes et al., 2003). A further understanding ofcarbon dynamics in coral reefs is required not only for analyzing the role of coralreefs in the global carbon cycle but also for predicting the future change of coralreefs under rapidly changing environment.

4. SUMMARY

A system-level net organic-to-inorganic carbon production ratio is a masterparameter for controlling sink/source behavior of coral reefs with respect toatmospheric CO2. Due to dominant carbonate production with negligible excessorganic carbon production in coral reef ecosystems, most coral reefs act as a net,or at least CO2 source to the atmosphere. This rule is supported by observationsfrom atolls and barrier reefs in the western Pacific: higher pCO2 and lower totalalkalinity in the lagoon waters were evident compared to their offshore waters.Reef topography, especially residence time of lagoon water, affects the carbonbudget of coral reefs to some extent. There is a relatively weak, but stillrecognizable positive relationship between the lagoon-offshore difference inpCO2 and residence time of reef water. Another important factor controllingcarbon turnover in coral reefs is proximity to land: terrestrial carbon and nutrientinputs were recognized in most fringing and barrier types of reefs. Terrestrialinputs of carbon and nutrient may enhance the CO2 release from coral reefs to theatmosphere.

Acknowledgements—We express our appreciation to T. Ayukai of the Australian Instituteof Marine Science and F. Nishibori and K. Goto of the Kansai Environmental EngineeringCenter Co. Ltd. (KANSO). The manuscript was improved by valuable comments fromtwo anonymous reviewers and L. P. Gupta of AIST. This study was supported by“Northwest Pacific Carbon Cycle Study” consigned to the KANSO by the NEDO.Additional supports were from “GCMAPS” funded by the Ministry of Education, Culture,Sports, Science and Technology and “Study on the selection of biodiversity conservationarea of the coral reef” funded by the Ministry of the Environment of Japan.

REFERENCES

Atkinson, M. J. and S. V. Smith (1983): C:N:P ratios of benthic marine plants. Limnol. Oceanogr.,28, 568–574.

Barnes D. J. and B. E. Chalker (1990): Calcification and photosynthesis in reef-building corals andalgae. p. 109–131. In Coral Reefs, Ecosystem of the World 25, ed. by Z. Dubinsky, Elsevier,Amsterdam.


Buddemeiwe, R. W. (1996): Coral reefs and carbon dioxide. Science, 271, 1298–1299.Crossland, C. J., B. J. Hatcher and S. V. Smith (1991): Role of coral reefs in global ocean production.

Coral Reefs, 10, 55–64.Frankignoulle, M., C. Canon and J.-P. Gattuso (1994): Marine calcification as a source of carbon

dioxide: Positive feedback of increasing atmospheric CO2. Limnol. Oceanogr., 39, 458–462.Frankignoulle, M., J.-P. Gattuso, R. Biondo, I. Bourge, G. Copin-Montegut and M. Pichon (1996):

Carbon fluxes in coral reefs. II. Eulerian study of inorganic carbon dynamics and measurementsof air-sea CO2 exchange. Mar. Ecol. Prog. Ser., 145, 123–132.

Furnas, M. J., A. W. Mitchell and M. Skuza (1995): Nitrogen and phosphorous budgets for the centralGreat Barrier Reef Shelf. Great Barrier Reef Marine Park Authority Research Report No. 36, 194pp., Great Barrier Reef Marine Park Authority, Townsville.

Gattuso, J.-P. and R. W. Buddemeier (2000): Ocean biogeochemistry: calcification and CO2.Science, 407, 313.

Gattuso, J.-P., M. Pichon, B. Delesalle and M. Frankignoulle (1993): Community metabolism andair-sea CO2 fluxes in a coral reef ecosystem (Moorea, French Polynesia). Mar. Ecol. Prog. Ser.,96, 259–267.

Gattuso, J.-P., M. Pichon and M. Frankignoulle (1995): Biological control of air-sea CO2 flux: effectof photosynthetic and calcifying marine organisms and ecosystem. Mar. Ecol. Prog. Ser., 129,307–312.

Gattuso, J.-P., M. Frankignoulle, S. V. Smith, J. R. Ware and R. Wollast (1996a): Coral reefs andcarbon dioxide. Science, 271, 1298.

Gattuso, J.-P., M. R. Pichon, B. Delesalle, C. Canon, G. Copin-Montegut and M. Pichon and M.Frankignoulle (1996b): Carbon fluxes in coral reefs. I. Lagrangian measurements of communitymetabolism and resulting air-sea CO2 disequilibrium. Mar. Ecol. Prog. Ser., 145, 109–121.

Gattuso, J.-P., M. Frankignoulle and S. V. Smith (1999): Measurement of community metabolismand significance in the coral reef CO2 source-sink debate. Proc. Natl. Acad. Sci. USA, 96,13017–13022.

Hata, H., A. Suzuki, T. Maruyama, N. Kurano, S. Miyachi, Y. Ikeda and H. Kayanne (1998): Carbonflux by suspended and sinking particles around the barrier reef of Palau, western Pacific. Limnol.Oceanogr., 43, 1883–1893.

Hata, H., S. Kudo, H. Yammano, Y. Kurano, N. Kurano and H. Kayanne (2002): Organic carbon fluxin Shiraho coral reef (Ishigaki Island, Japan). Mar. Ecol. Prog. Ser., 232, 129–140.

Hughes, T. P., A. H. Baird, D. R. Bellwood, M. Card, S. R. Connolly, C. Folke, R. Grosberg, O.Hoegh-Guldberg, J. B. C. Jackson, J. Kleypas, J. M. Lough, P. Marshall, M. Nyström, S. R.Palumbi, J. M. Pandolfi, B. Rosen and J. Roughgarden (2003): Climate Change, Human Impacts,and the Resilience of Coral Reefs. Science, 301, 929–933.

Iglesias-Rodriquez, M. D., R. Armstrong, R. Feely, R. Hood, J. Kleypas, J. Milliman, C. Sabine andJ. Sarmiento (2002): Progress made in study of ocean’s calcium carbonate budget. Eos Trans.AGU, 83, 365–375.

Kawahata, H., A. Suzuki and K. Goto (1997): Coral reef ecosystems as a source of atmospheric CO2:evidence from pCO2 measurements of surface waters. Coral Reefs, 16, 261–266.

Kawahata, H., I. Yukino and A. Suzuki (2000a): Terrestrial influence on the Shiraho fringing reef,Ishigaki Island, Japan: high carbon input relative to phosphate. Coral reefs, 19, 172–178.

Kawahata, H., A. Suzuki, T. Ayukai and K. Goto (2000b): Distribution of the fugacity of carbondioxide in the surface seawater of the Great Barrier Reef. Mar. Chem., 72, 257–272.

Kayanne, H. (1996): Coral reefs and carbon dioxide. Science, 271, 1298.Kayanne, H., A. Suzuki and H. Saito (1995): Diurnal changes in the partial pressure of carbon

dioxide in coral reef water. Science, 269, 214–216.Kinsey, D. W. (1978): Productivity and calcification estimates using slack-water periods and field

enclosures. p. 439–468. In Coral Reefs: Research Methods, ed. by D. R. Stoddart and R. E.Johannes, UNESCO, Paris.

Kinsey, D. W. and D. Hopley (1991): The significance of coral reef as global carbon sinks—responseto Greenhouse. Palaeogeogr., Palaeoclimatol., Palaeoecol., Global Planet. Change Sect., 89,363–377.


Kleypas, J. A. (1997): Modeled estimates of global reef habitat and carbonate production since thelast glacial maximum. Paleoceanogr., 12, 533–545.

Kleypas, J. A., R. W. Buddemeier, D. Archer, J.-P. Gattuso, C. Langdon and B. N. Opdyke (1999):Geochemical consequences of increased atmospheric carbon dioxide on coral reefs. Science,284, 118–120.

Kleypas, J. A., R. W. Buddemeier and J.-P. Gattuso (2001): The future of coral reefs in an age ofglobal change. Int. J. Earth Sci., 90, 426–437.

Kohn, A. J. (1964): Notes on reef habitats and gastropod molluscs of a lagoon island at North MaleAtoll, Maldives. Atoll Res. Bull., 102, 1–5.

Kraines, S., Y. Suzuki, K. Yamada and H. Komiyama (1996): Separating biological and physicalchanges in dissolved oxygen concentration in a coral reef. Limnol. Oceanogr., 41, 1790–1799.

Kraines, S., Y. Suzuki, T. Omori, K. Shitashima, S. Kanahara and H. Komiyama (1997): Carbonatedynamics of the coral reef systems at Bora Bay, Miyako Island. Mar. Ecol. Prog. Ser., 156, 1–16.

Kraines, S. B., A. Suzuki, T. Yanagi, M. Isobe, X. Guo and H. Komiyama (1999): Rapid waterexchange between the lagoon and open ocean at Majuro Atoll due to wind, waves and tide. J.Geophys. Res., 104, 15635–15653.

Maragos, J. E. and C. W. Cook, Jr. (1995): The 1991–1992 rapid ecological assessment of Palau’scoral reefs. Coral Reefs, 14, 237–252.

Mitchell, A. W. and M. J. Furnas (1996): Terrestrial inputs of nutrients and suspended sediments tothe GBR lagoon. p. 59–71. In Proceedings of the Great Barrier Reef Science, Use andManagement; A National Conference, Great Barrier Reef Marine Park Authority, Townsville.

Nakamori, T., A. Suzuki and Y. Iryu (1992): Water circulation and carbon flux on Shiraho coral reefof the Ryukyu Islands, Japan. Continent. Shelf Res., 12, 951–970.

Odum, H. T. and E. P. Odum (1955): Trophic structure and productivity of a windward coral reefcommunity on Eniwetok Atoll. Ecol. Monogr., 25, 291–320.

Ohde, S. and R. van Woesik (1999): Carbon dioxide flux and metabolic processes of a coral reef,Okinawa. Bull. Mar. Sci., 65, 559–576.

Pickard, G. L., J. R. Donguy, C. Henin and F. Rougerie (1977): A review of the physicaloceanography of the Great Barrier Reef and western Coral Sea. Aust. Inst. Mar. Sci., Monogr.Ser., 2, Australian Institute of Marine Science, Townsville, 134 pp.

Redfield, A. C., B. H. Ketchum and F. A. Richards (1963): The influence of organisms on thecomposition of sea-water. p. 26–77. In The Sea, Vol. 2, ed. by M. N. Hill, Wiley.

Smith, S. V. (1978): Coral-reef area and the contributions of reefs to processes and resources of theworld’s oceans. Nature, 273, 225–226.

Smith, S. V. (1981): Marine macrophytes as a global carbon sink. Science, 211, 838–840.Smith, S. V. (1988): Mass balance in coral reef-dominated area. p. 209–226. In Coastal-Offshore

Ecosystem Interactions, Lecture Notes on Coastal and Estuarine Studies, Vol. 22, ed. by B.-O.Jansson, Springer, Berlin.

Smith, S. V. and P. L. Jokiel (1978): Water composition and biogeochemical gradients in the CantonAtoll lagoon. Atoll Res. Bull., 221, 15–53.

Smith, S. V. and F. Pesret (1974): Processes of carbon dioxide flux in the Fanning Island lagoon.Pacific Sci., 28, 225–245.

Smith, S. V., S. Chandra, L. Kwitko, R. C. Schneider, J. Schoonmaker, J. Seeto, T. Tebano and G.W. Tribble (1984): Chemical stoichiometry of lagoonal metabolism. U. Hawaii/U. South PacificInternat. Sea Grant Prog. Tech. Rep. UNIHI-SEAGRANT-CR-84-02.

Suzuki, A. (1998): Combined effects of photosynthesis and calcification on the partial pressure ofcarbon dioxide in seawater. J. Oceanogr., 54, 1–7.

Suzuki, A. and H. Kawahata (1999): Partial pressure of carbon dioxide in coral reef lagoon waters:comparative study of atolls and barrier reefs in the Indo-Pacific Oceans. J. Oceanogr., 55, 731–745.

Suzuki, A. and H. Kawahata (2003): Carbon budget of coral reef systems: an overview ofobservations in fringing reefs, barrier reefs and atolls in the Indo-pacific region. Tellus, 55B,428–444.


Suzuki, A., T. Nakamori and H. Kayanne (1995): The Mechanism of production enhancement incoral reef carbonate systems: model and empirical results. Sediment. Geol., 99, 259–280.

Suzuki, A., H. Kawahata and K. Goto (1997): Reef water CO2 system and carbon cycle in MajuroAtoll, the Marshall Islands in the central Pacific. Proc. 8th Int. Coral Reef Symp., 1, 971–976.

Suzuki, A., H. Kawahata, T. Ayukai and K. Goto (2001): The oceanic CO2 system and carbon budgetin the Great Barrier Reef, Australia. Geophys. Res. Lett., 28, 1243–1246.

Ware, J. R., S. V. Smith and M. L. Reaka-Kudla (1992): Coral reefs: sources or sinks of atmosphericCO2? Coral Reefs, 11, 127–130.

A. Suzuki (e-mail: [email protected]) and H. Kawahata

reef water co system and carbon production of coral reefs

Documents