Geomorphology xxx (2014) xxx–xxx

GEOMOR-04680; No of Pages 7

Contents lists available at ScienceDirect

Geomorphology

j ourna l homepage: www.e lsev ie r .com/ locate /geomorph

Late Holocene sea-level change and reef-island evolution in New Caledonia

Hiroya Yamano a,b,⁎, Guy Cabioch b, Christophe Chevillon b,c, Jean-Lambert Join b,d

a National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 305-8506, Japanb Institut de Recherche pour le Développement, BP A5, 98848 Nouméa cedex, New Caledoniac Direction de l'Environnement, 19 av. Foch, 98846 Nouméa cedex, New Caledoniad UMR IPGP 7154, Laboratoire GéoSciences Réunion, Université de la Réunion, Réunion, France

⁎ Corresponding author at: National Institute for EnvironTsukuba, Ibaraki 305-8506, Japan. Fax: +81 29 850 2219.

E-mail address: hyamano@nies.go.jp (H. Yamano).

http://dx.doi.org/10.1016/j.geomorph.2014.03.0020169-555X/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Yamano, H., et al., Lhttp://dx.doi.org/10.1016/j.geomorph.2014.0

a b s t r a c t

a r t i c l e i n f o

Article history:Received 31 August 2012Received in revised form 30 October 2013Accepted 2 March 2014Available online xxxx

Keywords:Sea-level changeReef islandForaminiferaNew Caledonia

In New Caledonia, numerous cays are distributed on platform reefs in the southwest lagoon behind the barrierreef. At Mba Island, a vegetated sand cay in this area, we examined Holocene sea-level change, and reef develop-ment and evolution. The late Holocene sea-level curve for the area was updated using newly found fossilmicroatolls. Component-specific dating of foraminifera tests in the island sediment provided reliable ages ofisland formation. Mba Island initially formed around ~4500 cal yr B.P., when sea level was ~1.1 m higher thanat present. The island formed on the leeward side of the platform reef, where the base had been constructed oflagoon- and reef-derived sediment of the platform reef. Sea-level fall since 2800 cal yr B.P. has likely causedreef-flat emergence at low tide and created a suitable habitat for foraminifera. Such a transition in ecology dueto sea-level fall has been observed in other Pacific reefs, suggesting that the interaction between reef-flatformation, sediment production and sea-level fall is one of the critical factors determining island development.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Reef islands are low-lying and composed of unconsolidated sedi-ments derived from calcareous reef-building organisms. Such islandscan be divided into two major groups: motus and cays. Motus developon coral atolls, and seaward beaches comprise cobbles and gravelsthat were carried from the reef front by high-energy events. Con-versely, cays are generally composed of sands, and are consideredto form under mean energy, rather than stormy, conditions. Bothtypes of reef islands provide habitable land for indigenous communi-ties, and appear highly vulnerable to environmental changes, such assea-level rise (e.g., Woodroffe, 2008).

The interactions among Holocene sea-level change, reef growth,sediment production, and sediment transport exert a fundamental con-trol on the initiation and development of reef islands (McLean andWoodroffe, 1994; Perry et al., 2011). However, in comparison withreef growth studies, detailed chronological studies of reef-island devel-opment have been limited since the pioneeringwork at Cocos (Keeling)Island in the western Indian Ocean (Woodroffe et al., 1999).

Circular platform reefs often support cays, formed by the wave-focused accumulation of calcareous sediments produced on the reefplatform (Gourlay, 1988;Mandlier andKench, 2012). Although platform

mental Studies, 16-2Onogawa,

ate Holocene sea-level change3.002

reefs appear simple in comparison with atolls, there are significantvariations in the structure and development of the associated cays(Kench et al., 2005; Woodroffe et al., 2007; McKoy et al., 2010; Kenchet al., 2012). Understanding these variations requires further case stud-ies to relate Holocene sea-level change to reef-flat formation and islandevolution.

In this study, we examined the relationships among Holocenesea-level change, reef-flat development, and sand cay evolution inNew Caledonia. Previous geological research has provided detaileddata on sea-level change and reef development (e.g., Cabioch, 2003;Andréfouët et al., 2009). In New Caledonia, there are numerous cayslocated on platform reefs in the large southwest lagoon, behind thebarrier reef (Guilcher, 1988, p. 133). However, unlike similar cays inAustralia or the Maldives, their characteristics and development havenot been extensively studied. This study thus provides the first detailedexamination of reef-island evolution in New Caledonia, and provides anexcellent dataset for understanding paleo sea-level changes and associ-ated reef development.

2. Setting

New Caledonia is situated between 15° and 25° south, and between160° and 170° east (Fig. 1). The main island of New Caledonia, GrandeTerre, has a 500-km-long coastline fringed by reefs and mangroves,which constitute one of the longest barrier reefs in the world(Andréfouët et al., 2009). Subsidence is the dominant geologic processaround Grande Terre (Fig. 1), except in the southeastern area, which

and reef-island evolution in New Caledonia, Geomorphology (2014),

Amédée Is.(-0.14)

Ricaudy (-0.07)

Mba Is.Signal Is.

Dumbéa

50 km

Baie de Pirogues

Le Prédour

Mara

Tenia Is.(-0.11)

New CaledoniaAustralia

Fig. 1.Map of NewCaledonia showing the study sites (Mba Island and Signal Island) and other sites referred to in this study. Numbers in parentheses are subsidence rates (mm/yr) basedon Cabioch (2003).

2 H. Yamano et al. / Geomorphology xxx (2014) xxx–xxx

has been subjected to uplift. These spatial variations in vertical motions(subsidence and uplift) explain the modern reef geomorphology. Thehighest rates of subsidence were found offshore, which correlates tothe formation of the large southwest lagoon (Andréfouët et al., 2009).

Numerous cays occur on platform reefs in the large southwestlagoon behind the barrier reef (Guilcher, 1988, p. 133). A sand cay anda shingle cay are located on the leeward and windward sides of a plat-form reef, respectively. The location of these cays is attributable towave focusing on the platform reef (Gourlay, 1988; Mandlier andKench, 2012). The most frequently occurring wind regime is the south-easterly trade wind, with an average speed of 8 m/s and direction of110°, which comprises more than 2/3 of yearly wind occurrence(Ouillon et al., 2004). During trade wind episodes, the significant waveheight and period in the lagoon are typically 0.8–1.4 m and around3 s, respectively (Jouon et al., 2009). The tidal range is 1.5 m.

Holocene sea-level change in New Caledonia has been examinedthrough peat and reef cores, along with surface fossil coral and oyster

Please cite this article as: Yamano, H., et al., Late Holocene sea-level changhttp://dx.doi.org/10.1016/j.geomorph.2014.03.002

samples (Baltzer, 1970; Coudray and Delibrias, 1972; Lecolle andCabioch, 1988). Cabioch et al. (1989) synthesized the results of severalstudies and proposed that sea level reached ~1.1 m higher than thepresent level at ~5500 14C yr B.P. (~6500 cal yr B.P.), then fell smoothlyto the present level. However, due to a lack of reliable sea-level indica-tors in the late Holocene, the duration of higher sea levels and timing ofrelative sea-level fall, which affect reef-island evolution, remain unclear.

Reef cores described in previous studies have provided informationon Holocene reef development. Modern coral reefs in New Caledoniawere initiated at ~8000 cal yr B.P. on a Pleistocene reef foundation(Cabioch et al., 1995; Cabioch, 2003; Frank et al., 2006). The reefs exhib-ited a catch-up growth pattern in which coral growth postdates thestabilization of sea level (e.g., Cabioch et al., 1995).

Island evolution was examined for Mba Island, a vegetated sand caylocated in the southwest lagoon (Fig. 1). The island is ellipsoidal inshape (1000 m long and 350 m wide). The total land area is approxi-mately 0.33 km2, occupying half of the platform reef surface. Cores

e and reef-island evolution in New Caledonia, Geomorphology (2014),

?4420

3930

3200

3400

3470

CN S

CW E

4370

4140

2720

2680

1840

1810

2020

3380

3840

3630

4200

2200

1370

2060

4460

4250

3520

3470

Trade wind

1 m

100 m

2860

4420

3930

3200

3400

3470

Holocene-Pleistocene boundary(Castellaro, 1999)

Modern(assumed) Modern

(assumed)

Modern(assumed)

Modern(assumed)

MSL

LLW

MSL

LLW

PW

MWPMCW MCE

PE

MEP

PN

MNP MCNMCS

MSP

PS

b) Transect 3

c) Transect 2Transect 3

Transect 2

PN

C

PS

PE

PW

200 m

Tradewind

a N

Platform reef rim

ScaleSand

Coral fragment

Legend

Fig. 2. aMba Island transects and sampling points. Locations of the long cores that penetratedHolocene facies (Castellaro, 1999) are close to the points PN, C, and PS. The island is subjectedto tradewinds (direction=110°). b Transect 3, parallel to the long axis ofMba Island. The Holocene–Pleistocene boundary is indicated (Castellaro, 1999). c Transect 2, parallel to the shortaxis of Mba Island. Also indicated are the locations of sediment samples and internal facies based on sediment texture. Numbers show radiocarbon ages (cal yr B.P.) of fresh foraminiferatests from the sediment samples. Further details of radiocarbon ages are given in Table 1. MSL — mean sea level, LLW — lowest low water level.

3H. Yamano et al. / Geomorphology xxx (2014) xxx–xxx

taken at three points (Fig. 2), each greater than 20m in length, indicatedthat Mba Island is composed of modern and Holocene unconsolidatedcarbonate sediments with coral fragments, lying on a Pleistocene reefunit (Castellaro, 1999). The thickness of the sediment layer rangesfrom 4 to 8 m. The Holocene–Pleistocene boundary is situated at 2.3m below mean sea level (MSL) in the southern part of the island, and5 m below MSL in the central and northern parts of the island (Fig. 2).

3. Methods

Emergent fossil microatolls occur on the leeward reef flat of SignalIsland, 16 km southeast of Mba Island (Fig. 1). The reef flat is open tothe lagoon and does not have a reef crest, indicating that no pondingoccurs during low tides. To reconstruct paleo sea-levels, surface eleva-tions of both the fossil microatolls and modern microatolls were mea-sured. Specimens were collected at the edge of two fossil microatolls(Porites sp. and Faviidae sp.) using a hammer and chisel.

We surveyed the topography of Mba Island and reef flat along twotransects, parallel to the long and short axes of the island (Fig. 2). Topo-graphic surveys and elevations of all samples were reduced to MSL. Tocollect internal sediments from Mba Island, we augured holes at thecenter of the island and at additional two points along each transect.In addition to the core samples, sediment samples were collected from~30 cm below the surface at 12 points (Fig. 2). We sieved sedimentsamples through a screen mesh at 1.0ϕ intervals, and determined theskeletal composition of sediments coarser than sand-sized using pointcounts under a binocular microscope.

We used component-specific dating, which targets selected sedi-ment components (Woodroffe et al., 2007), to reconstruct island forma-tion history. From the sediment samples, larger benthic foraminifera(Baculogypsina sphaerulata) tests that still had spicules attached werecollected (Grade A type of Yamanouchi, 1988). Ford and Kench (2012)

Please cite this article as: Yamano, H., et al., Late Holocene sea-level changehttp://dx.doi.org/10.1016/j.geomorph.2014.03.002

showed less durability of skeletal remains of calcareous algae andforaminifera (including B. sphaerulata) based on tumbling barrel exper-iments. Consequently, fresh B. sphaerulata tests would be suitablefor dating to infer the age of formation of the facies inwhich they are in-cluded, because the foraminifera could have been transported fromtheir original habitat by currents soon after their death (Yamano et al.,2001; Weisler et al., 2012). The total weight of the foraminifera spec-imens was about 0.2 g. To check the reliability of the foraminiferadates, we collected coral (Acropora sp.) specimens from the samesediment samples at locations C and S (Table 1), in addition to the fora-minifera specimens. Branch orientations and preservation of surfacestructure suggest these corals are largely in situ and equivalent to thecoral thickets commonly observed on lagoon seafloor.

Carbon-14 dating was conducted on the two fossil microatoll speci-mens (Porites sp. and Faviidae sp.), the fresh foraminifera tests, and thetwo in situ Acropora sp. specimens (Table 1). X-ray diffraction analysisshowed no evidence that any of the coral specimens had undergonediagenetic alteration. All the age determinations were carried out byARTEMIS (Accélérateur pour la Recherche en sciences de la Terre,Environnement, Muséologie Installé à Saclay; Richardin et al., 2010).Carbon-14 age data were corrected for isotopic fractionation and cali-brated to calendar years before present using the software Calib ver. 6(Stuiver and Reimer, 1993), the Marine09 dataset (Reimer et al.,2009), and a local reservoir effect (ΔR) of−3± 9 yr for New Caledonia(Petchey et al., 2008). The ages of fresh foraminifera tests from thebeaches (locations PN, PS, PW and PE in Fig. 2) were assumed to bemodern (b60 cal yr B.P.), as Dawson et al. (2012) demonstrated atRaine Island, Great Barrier Reef of Australia.

The ages and elevations of the microatolls were integrated withpreviously reported data (Baltzer, 1970; Coudray and Delibrias, 1972;Lecolle and Cabioch, 1988; Cabioch et al., 1989, 1995; Frank et al.,2006) to generate an updated Holocene sea-level curve for southwest

and reef-island evolution in New Caledonia, Geomorphology (2014),

Table 1Radiocarbon ages from Signal Island fossil microatolls and Mba Island sediment (foraminifera and coral). Ages are rounded to the nearest 10.

Lab code Sample code Island Samplelocationa

Material Elevation relativeto MSL (m)

1σ range of conventionalage (yr B.P.)

1σ range of calibratedage (cal yr B.P.)

Calibrated age with medianprobability (cal yr B.P.)

SacA 3212 Signal_1_2005 Signal N/A Coral microatoll (Porites sp.) 0.04 3040 ± 40 2750–2850 2810SacA 3213 Signal_2_2005 Signal N/A Coral microatoll (Faviidae sp.) −0.06 3940 ± 40 3860–3990 3930SacA 11989 NC-Mba 01 Mba C-050 Foraminifera (Baculogypsina sphaerulata) 2.21 3580 ± 30 3430–3530 3470SacA 11990 NC-Mba 02 Mba C-090 Foraminifera (Baculogypsina sphaerulata) 1.83 3330 ± 30 3140–3250 3200SacA 11991 NC-Mba 03 Mba C-167 Foraminifera (Baculogypsina sphaerulata) 1.29 3510 ± 30 3360–3430 3400SacA 11992 NC-Mba 04 Mba C-256 Foraminifera (Baculogypsina sphaerulata) 0.4 3940 ± 30 3870–3980 3930SacA 11993 NC-Mba 05 Mba C-333 Coral (Acropora sp.) −0.37 4250 ± 30 4310–4410 4360SacA 11994 NC-Mba 06 Mba C-333 Foraminifera (Baculogypsina sphaerulata) −0.37 4290 ± 30 4380–4480 4420SacA 11995 NC-Mba 07 Mba S-067 Foraminifera (Baculogypsina sphaerulata) 1.15 4090 ± 35 4080–4200 4140SacA 11996 NC-Mba 08 Mba S-173 Foraminifera (Baculogypsina sphaerulata) 0.09 4255 ± 30 4310–4420 4370SacA 11997 NC-Mba 09 Mba E-078 Foraminifera (Baculogypsina sphaerulata) 3.3 2885 ± 30 2610–2630 2680SacA 11998 NC-Mba 10 Mba E-151 Foraminifera (Baculogypsina sphaerulata) 2.57 3090 ± 30 2810–2920 2860SacA 11999 NC-Mba 11 Mba E-201 Foraminifera (Baculogypsina sphaerulata) 2.07 2935 ± 30 2700–2750 2720SacA 12000 NC-Mba 12 Mba S-064 Foraminifera (Baculogypsina sphaerulata) 1.31 2385 ± 30 1970–2070 2020SacA 12001 NC-Mba 13 Mba S-135 Foraminifera (Baculogypsina sphaerulata) 0.6 2205 ± 30 1760–1860 1810SacA 12002 NC-Mba 14 Mba S-192 Coral (Acropora sp.) 0.03 2280 ± 30 1850–1940 1890SacA 12003 NC-Mba 15 Mba S-192 Foraminifera (Baculogypsina sphaerulata) 0.03 2235 ± 30 1800–1890 1840SacA 12004 NC-Mba 16 Mba W-047 Foraminifera (Baculogypsina sphaerulata) 3.38 3710 ± 30 3580–3680 3630SacA 12005 NC-Mba 17 Mba W-174 Foraminifera (Baculogypsina sphaerulata) 2.11 3870 ± 30 3790–3900 3840SacA 12006 NC-Mba 18 Mba W-262 Foraminifera (Baculogypsina sphaerulata) 1.23 3495 ± 30 3340–3420 3380SacA 12007 NC-Mba 19 Mba MCN-030 Foraminifera (Baculogypsina sphaerulata) 2.31 4130 ± 30 4140–4250 4200SacA 12008 NC-Mba 20 Mba MCE-030 Foraminifera (Baculogypsina sphaerulata) 3.13 3575 ± 30 3420–3520 3470SacA 12009 NC-Mba 21 Mba MCS-030 Foraminifera (Baculogypsina sphaerulata) 2.8 3620 ± 30 3470–3560 3520SacA 12010 NC-Mba 22 Mba MCW-030 Foraminifera (Baculogypsina sphaerulata) 2.8 4165 ± 30 4180–4310 4250SacA 12011 NC-Mba 23 Mba MNP-030 Foraminifera (Baculogypsina sphaerulata) 2.8 4325 ± 30 4410–4500 4460SacA 12012 NC-Mba 24 Mba MEP-030 Foraminifera (Baculogypsina sphaerulata) 4.81 2420 ± 30 2010–2110 2060SacA 12013 NC-Mba 25 Mba MSP-030 Foraminifera (Baculogypsina sphaerulata) 1.63 1830 ± 30 1320–1400 1370SacA 12014 NC-Mba 26 Mba MWP-030 Foraminifera (Baculogypsina sphaerulata) 4.03 2520 ± 30 2140–2260 2200

a Valid for Mba Island samples only. Sample location indicates the point name (Fig. 2) and depth from surface (cm).

4H.Yam

anoetal./G

eomorphology

xxx(2014)

xxx–xxx

Pleasecite

thisarticle

as:Yamano,H

.,etal.,LateHolocene

sea-levelchangeand

reef-islandevolution

inNew

Caledonia,Geom

orphology(2014),

http://dx.doi.org/10.1016/j.geomorph.2014.03.002

5H. Yamano et al. / Geomorphology xxx (2014) xxx–xxx

New Caledonia. The elevations of the samples used as sea-level indica-tors were corrected for subsidence, based on previously publishedrates (Cabioch, 2003) (Fig. 1).

As reef-island evolution is closely related to reef-flat formation(e.g., McLean and Woodroffe, 1994; Perry et al., 2011) and in somecases subsequent infilling of lagoon (Perry et al., 2013), the history ofreef-flat formation in southwestern New Caledonia was synthesizedfrom the age of coral in cores taken at Amédée Island (Frank et al.,2006) and Ricaudy Reef (Lecolle and Cabioch, 1988; Cabioch et al.,1995; Fig. 1). For consistency, the 14C ages were calibrated to calendaryears before present, based on the same procedure as describedabove. In cases where δ13C values were not available for past samples,we assumed −1‰ and −24‰ for marine carbonate and charcoal,respectively (Stuiver and Polach, 1977). Although the ages are expressedas 1σ error ranges (Table 1), radiometric ages are discussed as the valueswith median probability.

4. Results

4.1. Updated late Holocene sea-level curve

The emergent microatolls are located 0.06m belowMSL and 0.04mabove MSL, and samples yielded ages of 3930 and 2810 cal yr B.P.,respectively (Table 1). The highest living microatolls are 0.60 m belowMSL. This is consistent with the observation of Scoffin and Stoddart(1978), that the upper surfaces of microatolls are usually constrainedclose to the mean low water spring level. In order to reconstruct paleosea levels based on fossil microatolls, subsidence must be considered.Because no age data was available for Pleistocene facies in Mba Islandto estimate the subsidence rate, we used subsidence rate of AmédéeIsland to reconstruct elevations of the microatolls. As the subsidenceinferred from Amédée Island was 0.14 mm/yr (Cabioch, 2003), sealevel was estimated to have been 1.09 m higher than present at3930 cal yr B.P., and 1.03mhigher than present at 2810 cal yr B.P. Incor-porating these new data into a previous sea-level curve presented byCabioch et al. (1989), it is apparent that a sea-level highstand, 1.1 m

Present-day MSL

cal

met

ers

rela

tive

to p

rese

nt-d

ay M

SL

Reef-flat formation

Reef-island formation

Sea-level curve

-10

-8

-6

-4

-2

0

2

6000800010000

M

C

C

M

O

Fig. 3. Updated Holocene sea-level curve for southwest New Caledonia. Microatoll data in thisand Delibrias, 1972). Horizontal bars indicate 1σ error ranges of calibrated ages. The tops of theindicate the past range of MSL, because mangroves occur between mean low water springs an(Hopley et al., 2007). The temporal evolution of reef flats is interpreted on the basis of Ricau2006) and the evolution of reef islands is based on Mba Island (Table 1 and Fig. 2). MSL — mea

Please cite this article as: Yamano, H., et al., Late Holocene sea-level changehttp://dx.doi.org/10.1016/j.geomorph.2014.03.002

above the present level, occurred from ~6500 to 2800 cal yr B.P. Thiswas followed by sea-level fall after 2800 cal yr B.P (Fig. 3).

4.2. Island topography and sediment

The changes in elevation across the island indicate that there areseveral ridges encircling a central depression (Fig. 2). The highest pointis ~6 m above MSL. The reef flat is subaerially exposed during lowtides. Although live corals are scarce on the reef flat (most likely due tothe subaerial exposure), abundant foraminifera (B. sphaerulata) liveamidst the algal turf.

Medium to coarse sand (0.0 to 2.0Φ) dominate the island sedimentsamples, and comprise up to 90% of the total weight of the sediments.However, in the lower samples from cores taken at locations C and S,and at the south and east beaches (PS and PE, respectively), granule-to pebble-sized fractions (−1.0 to−3.0Φ) comprised up to 20% of thetotal weight. The gravel in the base of the core at locations C and Sincluded fresh, probably in situ coral (Acropora sp.) fragments. Coral,foraminifera, Halimeda, and molluscs were the dominant constituentsof the sediment samples (Fig. 4), with coral and foraminifera comprisingmore than 60% of the total weight. The foraminifera component waspredominantly (N95%) B. sphaerulata. Along Transect 3, the foraminiferacomponent was greater on the windward side, exceeding 25% of thesediment. Along Transect 2, more than 25% of the total foraminiferafractionwas found at sample locationsWand E. In all the beach samples(PN, PS, PW, and PE), foraminifera comprisedmore than 30% of the sed-iment (Fig. 4).

Paired dating of foraminifera and coral in two samples showed con-sistent ages, inwhich the ages overlapped in the 1σ range (Table 1). Thisindicates that foraminiferawith fresh spicules give reliable estimates forthe date of formation of Mba Island. Along Transect 3, the earliest datesare concentrated on the leeward side of the island (MPN, N, and MNC),as well as at the base of the central core (C), ranging from 4460 to4140 cal yr B.P. (Fig. 2). The ages become younger toward the south,with the youngest (1370 cal yr B.P.) measured at location MSP. Theoldest sediment on Transect 2, measured at MWC, shows a similar age

yr B.P.

Formation of island

Formation ofplatform reef flat

020004000

icroatoll (Signal; this study)

oral (Ricaudy; Lecolle and Cabioch, 1988;Cabioch et al., 1995)

oral (Amedee; Frank et al., 2006)

angrove peat (Mara and Dumbea; Baltzer, 1970)

yster (Le Predour and Baie des Pirogues;Coudray and Delibrias, 1972)

study were integrated with past sea-level indicators (peat: Baltzer, 1970; oyster: Coudrayvertical bars for microatolls indicate past MSL. Vertical bars for mangrove peat and oysterd mean high water springs and oysters occur between MSL and mean high water neapsdy and Amédée reef cores (Lecolle and Cabioch, 1988; Cabioch et al., 1995; Frank et al.,n sea level.

and reef-island evolution in New Caledonia, Geomorphology (2014),

0

20

40

60

80

100

ForaminiferaCoralsCoralline algaeHalimedaMolluscsOthersIndeterminate

MN

P-0

30C

-333

N-1

73M

CW

-030

MC

N-0

30N

-067

C-2

56W

-174

W-0

47M

CS

-030

MC

E-0

30C

-050

C-1

67W

-262

C-0

90E

-151

E-2

01E

-078

MW

P-0

30M

EP

-030

S-0

64S

-192

S-1

35M

SP

-030

PE

-030

PW

-030

PS

-030

PN

-030

4460

4420

4370

4250

4200

4140

3930

3840

3630

3520

3470

3470

3400

3380

3200

2860

2720

2680

2200

2060

2020

1840

1810

1370

Mod

ern

Mod

ern

Mod

ern

Mod

ern

cal yr B.P.

Sample location

Wei

ght %

Legend

Fig. 4. Sediment constituents in Mba Island. Sample location indicates the point name and depth from surface (cm) (Table 1 and Fig. 2). Samples are arranged according to radiocarbondates.

6 H. Yamano et al. / Geomorphology xxx (2014) xxx–xxx

(4250 cal yr B.P.) to those in the leeward core of Transect 3. The agesbecome younger toward both the western and eastern beaches (PWand PE).

5. Discussion

Fig. 3 shows the updated sea-level curve for southwest New Caledo-nia, and age-depth data derived fromprevious studies. Sea level reachedits present position at ~7500–7000 cal yr B.P. This is consistent withstudies from oceanic islands in the western Pacific (Palau: Kayanneet al., 2002; Vanuatu: Andréfouët et al., 2009; Papua New Guinea:Chappell and Polach, 1991). Sea level reached as high as 1.1 m abovethe present-day level and stabilized from ~6500 to 2800 cal yr B.P.Relative sea-level fall occurred after 2800 cal yr B.P.

Although direct evidence of the timing of reef-flat formation at MbaIsland was not available, it can be inferred from other reefs in the sameregion (Ricaudy and Amédée reefs; Fig. 1). Ricaudy and Amédée reefsboth caught up with sea level at 7000–6500 cal yr B.P. (Fig. 3). TheRicaudy reef expanded seaward, and the seaward margin yields agedates of 6510 and 5650 cal yr B.P. (Lecolle and Cabioch, 1988). Themost recent age from the Amédée reef flat was measured as4970 cal yr B.P. (Frank et al., 2006). On the platform reef of Signal Island,fossil microatolls on the leeward reef flat yield ages of 3930 and2810 cal yr B.P. Collectively, these results suggest that the main bodyof the reef flats in the area formed from 7000 to 5000 cal yr B.P., andthe leeward reef flats had formed by 2800 cal yr B.P.

The island formed directly over a sediment-infilled lagoon. Theoriginal composition of the lagoon is likely to have been a mixture ofin situ and reef-derived materials, as the gravels consisted of fresh,in situ corals (Table 1; Castellaro, 1999), and sands were dominatedby B. sphaerulata tests that would have been transported from shal-low reef flats (Yamano et al., 2001; Dawson et al., 2012). The forma-tion of Mba Island was initiated on the leeward side of the platformat ~4500 cal yr B.P., likely followed by the formation of the main body

Please cite this article as: Yamano, H., et al., Late Holocene sea-level changhttp://dx.doi.org/10.1016/j.geomorph.2014.03.002

of the reef flat and infilling of the lagoon, probably driven by the focus-ing effect of wave processes on platform reefs (Gourlay, 1988; Mandlierand Kench, 2012). Following initial formation, the island grew by incre-mental accretion (Fig. 2).

Ecological changes appear to have occurred as a result of relativesea-level fall after 2800 cal yr B.P. The fall could have been fatal tosome species on the reef flat, but provided a suitable habitat forB. sphaerulata, leading to increased proportion of B. sphaerulata tests inthe island sediment (Fig. 4). As foraminifera are easily transported,such a transition may directly contribute to island formation (Yamanoet al., 2000; Perry et al., 2011; Dawson et al., 2012).

In conclusion, there are clear relationships among Holocene sea-level change, reef development, and sand cay formation in New Caledo-nia. Following the formation of the reef flat, Mba Island initially formedon an infilled lagoon at the leeward side at a location determined bywave focusing. The island developed from lagoon- and reef-derived sed-iment under sea levels higher than those of the present. This indicatesthat post-highstand sea-level fall is not a prerequisite for island initia-tion as presented by Woodroffe et al. (1999) and Kench et al. (2005).Continued growth of the island is likely to have been driven by sea-level fall, which created a suitable habitat for foraminifera and conse-quently provided a continuous sediment supply. The occurrence of fora-minifera on reef flats accompanying sea-level fall has been observedelsewhere in the Pacific (Yamano et al., 2001), and has been shown tolikely enhance reef-island development elsewhere (Woodroffe andMorrison, 2001; Kayanne et al., 2011; Weisler et al., 2012). Sea-levelfall, however, could also cause a cessation of island development. AtBewick Cay on the Great Barrier Reef of Australia, rapid island growthcan be attributed to ecological transitions associated with reef-flatemergence, resulting from continuous sea-level fall since 6400 cal yrB.P; however, since 4000 cal yr B.P. the island has experienced veryfew morphological changes (Kench et al., 2012). This suggests thatthere is a critical water depth at which sediment production and subse-quent transport to reef islands are enhanced (Kench et al., 2012).

e and reef-island evolution in New Caledonia, Geomorphology (2014),

7H. Yamano et al. / Geomorphology xxx (2014) xxx–xxx

Investigation of sea-level change and island formation based on reliablepaleo sea-level indicators (e.g., microatolls) and precise component-specific dating of island sedimentwould providemore accurate insightsinto the timing and triggering mechanisms of island evolution.

Acknowledgments

We thank the staff at Institut de Recherche pour le Développment(IRD) for the field and logistics support. This study was supported bythe L'Agence National de la Recherche (ANR) INTERFACE Program andby a Japan Society for the Promotion of Science (JSPS) PostdoctoralFellowship for Research Abroad (to HY). HY thanks Serge Andréfouëtand UR CoRéUs for accepting his stay at IRD.

References

Andréfouët, S., Cabioch, G., Flamand, B., Pelletier, B., 2009. A reappraisal of the diversity ofgeomorphological and genetic processes of New Caledonian coral reefs: a synthesisfrom optical remote sensing, coring and acoustic multibeam observations. CoralReefs 28, 691–707.

Baltzer, F., 1970. Datation absolue de la transgression honocene sur la cote ouest deNouvelle-Caledonie sur des échantillons de tourbes a palétuviers. Interprétationnéotectonique. C. R. Acad. Sci. Paris Série D 271, 2251–2254.

Cabioch, G., 2003. Postglacial reef development in the South-West Pacific: case studiesfrom New Caledonia and Vanuatu. Sediment. Geol. 159, 43–59.

Cabioch, G., Montaggioni, L.F., Faure, G., 1995. Holocene initiation and development ofNew Caledonian fringing reefs, SW Pacific. Coral Reefs 14, 131–140.

Cabioch, G., Thomassin, B.A., Lecolle, J.F., 1989. Age d'émersion des récifs frageantsholocenes autour de la bbGrande TerreNN de Nouvelle-Calédonie (SO Pacifique);nouvelle interprétation de la courbe des niveaux marins depuis 8000 ans B.P. C. R.Acad. Sci. Paris Série II 308, 419–425.

Castellaro, C., 1999. Reconstructionspaléoclimatiques et paléoenvironmentales a l'Holoceneet au Pléistocene terminal en Nouvelle-Calédonie et aux Seychelles (région indo-pacifique): l'entregistrement par les coraux. PhD Thesis University of Province,Marseille (212 pp.).

Chappell, J., Polach, H., 1991. Post-glacial sea-level rise from a coral record at Huon Penin-sula, Papua New Guinea. Nature 349, 147–149.

Coudray, J., Delibrias, G., 1972. Variations du niveau marin au-dessus de l'actuel enNouvelle-Calédonie depuis 6000 ans. C. R. Acad. Sci. Paris Série D 275, 2623–2626.

Dawson, J.L., Hua, Q., Smithers, S.G., 2012. Benthic foraminifera: their importance to futurereef island resilience. Proceedings of the 12th International Coral Reef Symposium.

Ford, M.R., Kench, P.S., 2012. The durability of bioclastic sediments and implications forcoral reef deposit formation. Sedimentology 59, 830–842.

Frank, N., Turpin, L., Cabioch, G., Blamart, D., Tressens-Fedou,M., Colin, C., Jean-Baptiste, P.,2006. Open system U-series ages of corals from a subsiding reef in New Caledonia:implications for sea level changes, and subsidence rate. Earth Planet. Sci. Lett. 249,274–289.

Gourlay, M.R., 1988. Coral cays: products of wave action and geological processes in a bio-genic environment. Proceedings of the 6th International Coral Reef Symposium, 2,pp. 491–496.

Guilcher, A., 1988. Coral Reef Geomorphology. John Wiley & Sons, New York (228 pp.).Hopley, D., Smithers, S.G., Parnell, K.E., 2007. The geomorphology of the Great Barrier

Reef. Development, Diversity, and Change. Cambridge University Press, Cambridge(546 pp.).

Jouon, A., Lefebvre, J.P., Douillet, P., Ouillon, S., Schmied, L., 2009. Wind wave measurementsand modelling in a fetch-limited semi-enclosed lagoon. Coast. Eng. 56, 599–608.

Kayanne, H., Yamano, H., Randall, R.H., 2002. Holocene sea level changes and barrier reefformation on an oceanic island, Palau Islands, western Pacific. Sediment. Geol. 150,47–60.

Please cite this article as: Yamano, H., et al., Late Holocene sea-level changehttp://dx.doi.org/10.1016/j.geomorph.2014.03.002

Kayanne, H., Yasukochi, T., Yamaguchi, T., Yamano, H., Yoneda, M., 2011. Rapid settlementof Majuro Atoll, central Pacific, following its emergence at 2000 years Cal BP.Geophys. Res. Lett. 38, L20405.

Kench, P.S., McLean, R.F., Nichol, S.L., 2005. Newmodel of reef-island evolution: Maldives,Indian Ocean. Geology 33, 145–148.

Kench, P.S., Smithers, S.G., McLean, L.F., 2012. Rapid reef island formation and stabilityover and emerging reef flat: Bewick Cay, northern Great Barrier Reef, Australia.Geology 40, 347–350.

Lecolle, J.F., Cabioch, G., 1988. La limite holocene-pleistocene dans le recif frangeantRicaudy (Nouvelle-Caledonie). Geochronogie, facies et diagenese. Implicationseustatiques et neotectoniques. Mar. Geol. 81, 241–260.

Mandlier, P.G., Kench, P.S., 2012. Analytical modelling of wave refraction and convergenceon coral reef platforms: implications for island formation and stability. Geomorphology159–160, 84–92.

McKoy, H., Kennedy, D.M., Kench, P.S., 2010. Sand cay evolution on reef platforms,Mamanuca Islands, Fiji. Mar. Geol. 269, 61–73.

McLean, R.F., Woodroffe, C.D., 1994. Coral atolls. In: Carter, R.W.G., Woodroffe, C.D. (Eds.),Coastal Evolution: Late Quaternary Shoreline Morphodynamics. Cambridge UniversityPress, New York, USA, pp. 267–302.

Ouillon, S., Douillet, P., Andréfouët, S., 2004. Coupling satellite data with in situ measure-ments and numerical modeling to study fine suspended-sediment transport: a studyfor the lagoon of New Caledonia. Coral Reefs 23, 109–122.

Perry, C.T., Kench, P.S., Smithers, S.G., Riegl, B., Yamano, H., O'Leary, M.J., 2011. Implica-tions of reef ecosystem change for the stability and maintenance of coral reef islands.Glob. Chang. Biol. 17, 3679–3696.

Perry, C.T., Kench, P.S., Smithers, S.G., Yamano, H., O'Leary, M.J., Gulliver, P., 2013. Timescales and modes of reef lagoon infilling in the Maldives and controls on the onsetof reef island formation. Geology 41, 1111–1114.

Petchey, F., Anderson, A., Zondervan, A., Ulm, S., Hogg, A., 2008. Newmarine ΔR values forthe South Pacific subtropical gyre region. Radiocarbon 50, 373–397.

Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C.,Buck, C.E., Burr, G.S., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P.,Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., McCormac,F.G., Manning, S.W., Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S., Turney, C.S.M., van der Plicht, J., Weyhenmeyer, C.E., 2009. INTCAL09 and MARINE09 radiocar-bon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51, 1111–1150.

Richardin, P., Gandolfo, N., Moignard, B., Lavier, C., Moreau, C., Cottereau, E., 2010. Centreof Research and Restoration of the Museums of France: AMS radiocarbon dates list 1.Radiocarbon 52, 1689–1700.

Scoffin, T.P., Stoddart, D.R., 1978. The nature and significance of microatolls. Philos. Trans.R. Soc. Lond. B 284, 99–122.

Stuiver, M., Polach, H.A., 1977. Discussion: reporting of 14C data. Radiocarbon 19,355–363.

Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised CALIB radiocarbon pro-gram. Radiocarbon 35, 215–230.

Weisler, M.I., Yamano, H., Hua, Q., 2012. A multidisciplinary approach for dating humancolonization of Pacific atolls. J. Island Coast. Archaeol. 7, 102–125.

Woodroffe, C.D., 2008. Reef-island topography and the vulnerability of atolls to sea-levelrise. Glob. Planet. Chang. 62, 77–96.

Woodroffe, C.D., McLean, R.F., Smithers, S.G., Lawson, E.M., 1999. Atoll reef-island forma-tion and response to sea-level change: West Island, Cocos (Keeling) Islands. Mar.Geol. 160, 85–104.

Woodroffe, C.D., Morrison, R.J., 2001. Reef-island accretion and soil development onMakin, Kiribati, central Pacific. Catena 44, 245–261.

Woodroffe, C.D., Samosorn, B., Hua, Q., Hart, D.E., 2007. Incremental accretion of a sandyreef island over the past 3000 years indicated by component-specific radiocarbondating. Geophys. Res. Lett. 34, L03602.

Yamano, H., Kayanne, H., Yonekura, N., 2001. Anatomy of a modern coral reef flat: arecorder of storms and uplift in the late Holocene. J. Sediment. Res. 71, 295–304.

Yamano, H., Miyajima, T., Koike, I., 2000. Importance of foraminifera for the formation andmaintenance of a coral sand cay: Green Island, Australia. Coral Reefs 19, 51–58.

Yamanouchi, H., 1988. The distribution of sandy sediments around coral reefs and theeffect of topography. Proceedings of the 6th International Coral Reef Symposium, 3,pp. 497–502.

and reef-island evolution in New Caledonia, Geomorphology (2014),