Coral Reefs (2011) 30:1101–1105 DOI 10.1007/s00338-011-0810-y
NOTE
Disappearance of Acropora from the Marquesas (French Polynesia) during the last deglacial period G. Cabioch • C. C. Wallace • M. T. McCulloch • H. Zibrowius • P. Laboute • B. Richer de Forges
Received: 22 April 2011 / Accepted: 28 July 2011 / Published online: 27 August 2011 Ó Springer-Verlag 2011
Abstract The major reef-building coral genus Acropora has never been recorded, living or fossil, from the Marquesas Islands in the central Pacific Ocean, which are characterized by limited modern reef formations. During the ‘‘Musorstom 9’’ cruise in 1997, investigations of marine platforms representing drowned reef systems revealed for the first time the presence of two Acropora species as fossils at seven Marquesas islands. The predominant species was Acropora valida, which was widespread in the archipelago and dated between 7.4 and 48.6 ka, providing evidence of an earlier Pacific distribution pattern broader
Communicated by Geology Editor Prof. Bernhard Riegl G. Cabioch (&) IRD (Institut de Recherche pour le De´veloppement), UMR LOCEAN/IPSL, 32 Avenue Henri Varagnat, 93143 Bondy Cedex, France e-mail:
[email protected] C. C. Wallace Museum of Tropical Queensland, 70–102 Flinders Street, Townsville, QLD 4810, Australia M. T. McCulloch School of Earth and Environment, The University of Western Australia and ARC Centre of Excellence in Coral Reef Studies, 35 Stirling Highway, Crawley, WA 6009, Australia H. Zibrowius Station Marine d’Endoume, rue de la Batterie des Lions, 13007 Marseille, France P. Laboute 24 rue Bon, 98800 Noume´a Cedex, New Caledonia B. R. de Forges IRD, UMR 7138, B.P. A5, 98848 Noume´a Cedex, New Caledonia
than previously observed. It is proposed that disappearance of Acropora after 7.4 ka was linked to climatic events probably ENSO events controlling the distribution of corals and coral reefs in the eastern Pacific without excluding alternatively the effects of an increase in sea-level rise. Keywords Marquesas Acropora Fossil Climate Biogeography
Introduction The reef formations around the Marquesas Islands in the central Pacific Ocean are limited in extent (Crossland 1927; Brousse et al. 1978; Chevalier 1978) and dominated by boulders of Porites spp. Most remarkable in the islands is the absence of the widespread coral genus Acropora, which elsewhere plays a major role in reef development and coral species diversity (Chevalier 1978; Wallace 1999). The poor development of reefs in the Marquesas (only 27 species of Scleractinia with a few species of non-scleractinian corals have been recorded: Crossland 1927, Chevalier 1978, Richer de Forges et al. 1999) has been variously ascribed to one or several factors including the role of very high terrigenous inputs (Crossland 1927; Brousse et al. 1978), geographic isolation of the archipelago (Davis 1928), and upwelling of cold waters (Crossland 1927). Bathymetric data reveal evidence of wide and narrow drowned reef features surrounding the islands. Rougerie et al. (1992) interpreted the wide 80- to 90-m depth platform as a former barrier reef, drowned during the rapid increase of sea level known as the Melt Water Pulse 1-A (MWP 1-A sensu Fairbanks 1989). In September 1997, during the cruise ‘‘Musorstom 9’’ onboard the R/V IRD ‘‘Alis’’ devoted to the study of the
123
1102
fauna and flora of the Marquesas, dredging of sub-marine platforms revealed the occurrence of tiny fragments of fossil Acropora specimens and attested to the occurrence of submerged reefal platforms (Richer de Forges et al. 1999). Subsequently in 2002, during the cruise ‘‘ReMarq’’ likewise onboard the R/V IRD ‘‘Alis’’, detailed multibeam bathymetric maps were obtained and precise dredging performed on all major marine terraces in the Marquesas. These dredges provided coral heads and reef limestone attesting to occurrence of fossil reefs, more extensive and with greater species diversity than the present reefs. Identification of bathymetric features coupled with coral dating revealed the occurrence of successive drowned reefs covering the whole of the last deglacial sea-level rise (Cabioch et al. 2008). Four reef generations (RG) were identified from 26 to 9 ka. The 3 first RG are characterized by the presence of more or less wide platforms, clearly showing reef zones (i.e., fore-reef, reef flat, back-reef): RG0, between 26 and 19 ka, RG1, between 18 and 15 ka, and RG2, between 14 and 11.5 ka, ranging from 125–115, 110–90, and 80–68 m depth, respectively. The RG3, younger than 11.5 ka, was observed at depths from 60 to 50 m, characterized by wide submarine platforms with scattered fossil coral colonies of Porites. This paper focuses on the record of fossil Acropora in the Marquesas archipelago, their specific identification, and the possible cause of their disappearance.
Methods The Marquesas archipelago consists of 12 islands extending from 8° to 11°S and 141° to 138°W (Fig. 1), with a general trend of 150°N obliquely to the movement of the Pacific Plate. Variations in alignment of the islands over a period from 1 to 5 MYA are attributed to the activity of a hot spot (Clouard 2000). The archipelago is located close to the Equator, in an area subjected to upwellings when trade winds are active and downwellings when trade winds disappear. This zone is directly within the influence of the El Nin˜o Southern Oscillation (ENSO) phenomenon. Along the slopes of the islands of Eiao, Nuku Hiva, Hiva Oa, four reef generations (RG) were identified by Cabioch et al. (2008), indicating episodic reef development during the deglacial sea-level rise: RG 0, 26–19 ka (125–115 m deep); RG 1, 18–15 ka (110–95 m), RG 2, 14–11.5 ka (80–68 m); and RG 3, younger than 11.5 ka (around 55–60 m) (Fig. 2). For this study, we examined 13 fragments of fossil Acropora obtained from seven of the islands at 13 stations 50–1,000 m deep during the cruise ‘‘Musorstom 9’’ in 1997 (Figs 1, 3). Specimens were identified using conventional techniques for the study of coral skeletons,
123
Coral Reefs (2011) 30:1101–1105 140°W
48.6±1 1278 1275 1286 1287
Eiao
138°W
a
1281
Motu One
170°E 180°
170°W 160°W 150°W 140°W 0
Marquesas Islands
10°S
French Polynesia
20°S
Vanuatu
8°S
Fiji New Caledonia
30°S
Nuku Hiva
1182
1288
1304 9°S
Ua Huka 1148
12.2±2.2
7.4±2.2
Ua Pou
Hiva Oa 1218 1201
1210 10°S
12.2±1 1244
Fatu Hiva b 11°S
Fig. 1 The Marquesas archipelago, a location in the central Pacific. b Distribution of the fossil Acropora. Numbered stars indicate site numbers sampled during the Musorstom 9 expedition; arrowed sites indicate location of dated Acropora valida samples; numbers in gray boxes indicate sample age in 1,000 years (ka)
using a Wild M5 binocular microscope with an eyepiece measuring graticule, and in comparison with examples of Acropora species in the Worldwide Acropora Collection at the Museum of Tropical Queensland, Australia. Some specimens were also examined and photographed using the scanning electron microscope (SEM) at the Advanced Analytical Centre at James Cook University, Australia, in order to obtain a permanent record of their appearance (the dating process used on four specimens was destructive). Three of the four specimens for dating were selected from samples collected in the upper terraces and presumed to be little reworked. The fourth sample was collected from the sediment base at 1,000 m, probably collapsed from the upper terraces. Because the fragments were tiny and the method destructive, only 4 corals were dated by U/Th thermal ionization mass spectrometry (TIMS) at the Research School of Earth Sciences at the Australian National University, Canberra. Among the criteria used to validate the U/TH dates, it was necessary to compare the initial d234U of coral samples with the modern marine value, i.e., 138–150% (Bard et al. 1996). The initial d234U of the Acropora fossils was between 144.2 and 146.1%, except in sample G55114, which was 155.3% (Table 1). Thus, ages were corrected for inherited 230Thnr according to the equation from McCulloch et al. (2010).
Coral Reefs (2011) 30:1101–1105
b
12.2 ka
kyears 15
7.4 ka
10
5 20
Younger Dryas
40 Depth range of DR1304 (7.4 ka)
60
MWP-1A
80
Depth range of DW1210 (12.2 ka)
100
120 Depth range of DW1218 (12.2 ka)
Sea level (m) below modern sea level
Fig. 2 a Idealized scheme of the succession of reef generations in the Marquesas during the last deglacial sealevel rise (adapted after Cabioch et al. 2008) and location of the dated Acropora. Porites dating is from Cabioch et al. (2008). b Theoretical location of the dated Acropora (gray arrows) on the 20-kyear sea-level curve (after Fairbanks 1989) versus bathymetric range of the dredging
1103
Succession of reef generations in the Marquesas during the last deglacial period
a
RG 2
Reworked corals
Ac: 7.4±2.2 ka (reworked)
RG 3
P: 9.100 ka P: 9.590 ka
P: 12.420 ka
60
80
RG 1 RG 0
P: 11.990 ka (reworked) Ac: 12.2±1.0 ka (reworked) P: 15.100 ka P: 15.460 ka
P: 26.500 ka Ac: 12.2±2.2 ka (reworked)
500 m
100
120
Porites sp. (P) Acropora sp. (Ac) RG : Reef Generation
Results and discussion The fossil Acropora fragments were up to 40 mm maximum length and 10 mm diameter. Of the 13 specimens, 10 were identified as Acropora valida (Dana, 1846) (Fig. 3), one as Acropora cerealis (Dana, 1846), and two could not be identified reliably to species. Colonies of A. valida are generally small, corymbose forms or thick tables and are found in a wide range of reef environments including outer reef flat, reef edge and tops of submerged reefs (Wallace 1999). Colonies of A. cerealis (Dana 1846) have a similar growth form and are generally common in upper reef slopes (Wallace 1999). The dated samples could be grouped into two sets. The first (three samples) were dated at 7.4–11.2 ka and the second (one sample) at 48.6 ka (Table 1). This last age attests to the presence of reefs during the Marine Isotopic Stage (MIS) 3, but at an unknown depth, because the sample was not in original position. Group one fossils were from three reef platforms: Nuku Hiva (50–58 m), Hiva Oa (98–100 m) and Hiva Oa (125–135 m); the last sample was from Eioa (see Figs. 1, 2). Taking into account the previously derived chronology of the drowned reefs (Cabioch et al. 2008), the living period of each of these Acropora can be attributed to a given reef generation, RG (Fig. 2). The 12.2-ka fossil from 125 to 135 m and the 12.2 ka from 98 to 100 m are reworked and have grown during RG 2 (Fig. 2). The 7.4-ka sample from 50 to 58 m probably developed during a
140
Depth (m)
growth stage posterior to RG 3 and would also be reworked. According to these results, no Acropora fossils were found after 7.4 ka. It is noteworthy that RG 3, at less than 60 m, was interpreted as the starting point of the settlement of non-reefal coral communities in the Marquesas as observed today (Cabioch et al. 2008). Although the period around 11.5–12 ka seems to be a period marked by drastic changes, the age of 7.4 ka shows that the Acropora continued to grow over time until very recently. Acropora is absent from several sites from the East Pacific so that the presence of fossils in the Marquesas gives us additional data to examine the biogeography of this group in this region of the Pacific (Wallace 1999). Glynn and Ault (2000) surveyed the presence and abundance of coral species in 14 sites of the eastern and central Pacific and noted the disappearance of A. valida and Porites rus in the East Pacific in 1982–1983 after the ENSO warming event. This observation clearly shows the link between the presence of coral species and the climate. The Marquesas archipelago is under the influence of ENSO events and is probably strongly influenced by the increase in frequency and intensity of ENSO following the last deglacial sea-level rise (Gagan et al. 2004). The disappearance of Acropora in the Marquesas, especially A. valida, would have occurred after 7.4 ka, a period marked by significant climatic variations in the Pacific as noted by Gagan et al. (2004), the period from 7 to 5 ka corresponding to the transition to the modern ENSO periodicities. Several studies also show between 7.5 and 8.5 ka increases in the sea-level
123
1.0
1.0 12.2
48.6 1
1.2 146.1
155.3 0.5
0.35 14.2 31
76
0.0034
0.003
0.1400 1.15 140.4
134.5 4.57 G55114
31.51
3.7 G55111
8.35
1.0
0.4260
50.7
2.2 7.4
12.2 1
1.9 145.2
144.2 0.100
0.120 11.840
16.650
12
16
0.0011
0.001
0.1179
0.1619 0.9
1.92 140.4
137.5 4.19 G55108
6.24 3.3 G55107
10.95
Corrected age (kyear) ±Initial d234U Initial d234U (%) ±Age (kyear) Age (kyear) [230Th/232Th] ±[230Th/238U] [230Th/238U] ±d234U d234U (%) Th (ppt)
230
U (ppm) Coral sample
Table 1 TIMS U/Th dates of Acropora
123
2.2
Coral Reefs (2011) 30:1101–1105 Error (kyear)
1104
Fig. 3 a Research vessel R.V. ‘‘Alis’’ moored in Baie des Vierges, Marquesas (photograph, P. Laboute/IRD). b, c Electron micrograph of Acropora valida MTQ G55109 (by P. Muir/Museum of Tropical Queensland) showing b radial corallites and c whole sample. d Ua Pou Island, Marquesas (photograph, P. Laboute/IRD)
rise, attributed by some authors to a meltwater pulse (Blanchon 2011). The disappearance of Acropora in the Marquesas would be attributed to the effects of this event. Nevertheless, we can observe that this demise in the Marquesas is preceded by the replacement of typical reefs by coral banks, probably consistent with a change in the ENSO frequency and intensity. We cannot exclude alternative hypotheses and even the combination of several events in the Acropora demise from the Marquesas. Of all the living Indo-Pacific Acropora species, A. valida has the broadest range, extending from the far western Pacific to occasional presence on the Colombian coast (Von Prahl and Mejia 1985), and from 33°N latitude to 31°S (Wallace 1999). Its earliest fossil record to date is
Coral Reefs (2011) 30:1101–1105
from 105,000 years on Stradbroke Island, eastern Australia (Pickett et al. 1985; Wallace 1999): however, the fossil record at species level for Pacific Acropora has not been explored in detail. The appearance and disappearance of this hardy Indo-Pacific species over at least a 50-ka period in the recent fossil history of the Marquesas highlights the role of climatic changes in the distribution of coral species in the central Pacific and shows that further understanding of the fossil history of key species such as A. valida could enhance our ability to predict the future prospects of coral reefs in the context of global climate change. It is noteworthy that in the Caribbean, the initial demise of A. palmata is thought to be linked with severe storm damage related to climatic events (MacIntyre et al. 2007). This highlights the major role of climatic events in the distribution and demise of some of coral species and their future at the global scale. Acknowledgments The authors wish to thank the director of IRDTahiti and Captain R. Proner and crew of the IRD R/V ‘‘Alis’’ for their help and assistance during the cruise. Paul Muir and Antoine Jorre are thanked for assistance with photography and SEM of the fossils and Paul Muir with figures. We greatly appreciate the assistance provided by Dr Graham Mortimer in the U-series age determinations undertaken at the Research School of Earth Sciences at the Australian National University. Our thanks are extended to the Topic Editor of Coral Reefs, B. Riegl, and two anonymous reviewers.
References Bard E, Hamelin B, Arnold M, Montaggioni LF, Cabioch G, Faure G, Rougerie F (1996) Deglacial sea level record from Tahiti corals and the timing of global meltwater discharge. Nature 382:241–244 Blanchon P (2011) Meltwater pulses. In: Hopley D (ed) Encyclopedia of modern coral reefs, Structure, form and process. Series Encyclopedia of Earth Sciences, Springer , 1st edn, pp 683–690 Brousse R, Chevalier JP, Denizot M, Salvat B (1978) Etude ge´omorphologique des Iles Marquises. Cahiers du Pacifique 21:9–74 Cabioch G, Montaggioni L, Frank N, Seard C, Salle´ E, Payri C, Pelletier B, Paterne M (2008) Successive reef depositional
1105 events along the Marquesas foreslopes (French Polynesia) since 26 ka. Mar Geol 254:18–34 Chevalier JP (1978) Les coraux des Iles Marquises. Cahiers du Pacifique 21:243–283 Clouard V (2000) Etude ge´odynamique et structurale du volcanisme de la Polyne´sie franc¸aise de 84 MA a` l’actuel. The`se de l’Universite´ de Polyne´sie franc¸aise, p 262 Crossland C (1927) Marine ecology and coral formations in the Panama region, Galapagos and Marquesas Islands and the atoll of Napuka. Trans R Soc Edinburgh 55:531–554 Davis WM (1928) The coral reef problem. Am Geogr Soc, Spec Publ 9:1–596 Fairbanks RG (1989) A 17,000-year glacio-eustatic sea level record:influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature 342:637–642 Gagan MK, Hendy EJ, Haberle SG, Hantoro WS (2004) Post-glacial evolution of the Indo-Pacific warm pool and El Nin˜o-Southern Oscillation. Quaternary Int 118–119:127–143 Glynn PW, Ault JS (2000) A biogeographic analysis and review of the far eastern Pacific coral reef region. Coral Reefs 19:1–23 Macintyre IG, Glynn PW, Toscano MA (2007) The demise of a major Acropora palmata bank-barrier reef off the southeast coast of Barbados, West Indies. Coral Reefs 26:765–773 McCulloch M, Taviani M, Montagna P, Lopez Correa M, Remia A, Mortimer G (2010) Proliferation and demise of deep-sea corals in the Mediterranean during the Younger Dryas. Earth Planet Sci Lett 298:143–152 Pickett JW, Thompson CH, Kelley RA, Roman D (1985) Evidence of higher sea level during isotopic stage 5C in Queensland, Australia. Quaternary Res 24:103–114 Richer de Forges B, Poupin J, Laboute P (1999) La campagne MUSORSTOM 9 dans l’archipel des ˆıles Marquises (Polyne´sie franc¸aise). Compte-rendu et liste des stations. In: Crosnier A (ed) Re´sultats des campagnes MUSORSTOM, vol. 20. Me´moires du Muse´um national d’Histoire naturelle 180:9–29 Rougerie F, Wauthy B, Rancher J (1992) Le re´cif barrie`re ennoye´ des Iles Marquises et l’effet d’ıˆle par endo-upwelling. CompteRendu Acad Sci Paris II 315:677–682 Von Prahl H, Mejia A (1985) Primer informede un coral acroporido, Acropora valida (Dana, 1846) (Scleractinia; Astrocoeniina; Acroporidae) para el Pacifico Americano. Rev Biol Trop 33: 39–43 Wallace CC (1999) Staghorn corals of the world—a revision of the genus Acropora (Scleractinia; Astrocoeniina; Acroporidae) worldwide, with emphasis on morphology, phylogeny and biogeography. CSIRO Publ, Collingwood. Australia
123
