 | ||
| IODP publications Expeditions Data & samples Policies News & photos Search | ||
|
doi:10.2204/iodp.pr.325.2010 BackgroundSea level change as a global climate indicatorPrior to Integrated Ocean Drilling Program (IODP) Expedition 310 (Tahiti Sea Level), only a few sea level curves based on coral reef records had been accurately dated for the last deglaciation: in Barbados between 19,000 and 8,000 calendar years before present (cal. y BP; A.D. 1950) (Fairbanks, 1989; Bard et al., 1990a, 1990b; Peltier and Fairbanks, 2006), in New Guinea between 13,000 and 6,000 cal. y BP (Chappell and Polach, 1991; Edwards et al., 1993), in Vanuatu between 23,000 and 6 cal. y BP (Cabioch et al., 2003), and in Tahiti between 13,750 cal. y BP and 2,380 radiocarbon years before present (14C y BP) (Bard et al., 1996) (Fig. F1A). Until recently, the Barbados and Vanuatu curves were the only ones to encompass the entire deglaciation. However, these sites, like New Guinea, are located above active subduction zones where tectonic movements can be large and discontinuous. Therefore, the reconstructed sea levels may be biased by variations in the rate of tectonic uplift and/or abrupt coseismic vertical motions. Also, Barbados is under the influence of glacial isostatic adjustment because of the waxing and waning of the North American ice sheet (Lambeck et al., 2002; Milne et al., 2009). Hence, there is a clear need to study past sea level changes in tectonically stable regions (or in areas where vertical crustal deformation is slow and/or regular) located far away from former ice-covered regions (far-field). The expeditions linked to IODP Proposal 519 (Expedition 310, Tahiti Sea Level, and Expedition 325, Great Barrier Reef Environmental Changes [GBREC]) aim to provide the most comprehensive deglaciation curves from tectonically stable regions by conducting offshore drilling of fossil coral reefs now preserved at 40–120 m below present sea level. Expedition 310 was successfully completed in 2005 (offshore phase) and 2006 (Onshore Science Party) (Camoin, Iryu, McInroy, et al., 2007). The Barbados record suggested that the last deglaciation was characterized by three brief periods of accelerated melting superimposed on a smooth and continuous rise of sea level with no reversals (Fig. F1A). These so-called meltwater pulse (MWP) events—19ka-MWP (Yokoyama et al., 2000; Clark et al., 2004; DeDeckker and Yokoyama 2009; Hanebuth et al., 2009), MWP-1A, and MWP-1B (~13,800 and 11,300 cal. y BP; Fairbanks, 1989; Bard et al., 1990)—were interpreted to be a consequence of massive inputs of continental ice (~40–50 mm/y in sea level rise that is roughly equivalent to annual discharge rates of 16,000 km3 for MWP-1A). Originally, MWP-1A was thought to correspond to a short and intense cooling between 14,100 and 13,900 cal. y BP in the Greenland ice core records (Johnsen et al., 1992; Grootes et al., 1993) and therefore to postdate the initiation of the Bölling-Alleröd warm period at ~14,800 cal. y BP (Broecker, 1992). However, cumulative evidence from far-field sites suggests that the timing of MWP-1A was slightly older than originally proposed (e.g., ~14,600–14,700 cal. y BP; Hanebuth et al., 2000; Webster et al., 2004); this has been confirmed by the Expedition 310 Tahiti sea level record (Deschamps et al., unpubl. data). The apparent sea level jump evident in the New Guinea Huon Peninsula coral record at ~11,000 cal. y BP (Edwards et al., 1993) lags MWP-1B by a few centuries compared to the coral record observed at Barbados (Fig. F1A). Two of the inferred meltwater pulses (MWP-1A and MWP-1B) may have induced reef-drowning events (Blanchon and Shaw, 1995). Two "give-up" reef levels have been reported at 90–100 and 55–65 meters below sea level (mbsl) on the Mayotte foreslopes (Comoro Islands) and have been related to MWP-1A and MWP-1B, respectively (Dullo et al., 1998), with similar features recorded in the Caribbean (MacIntyre et al., 1991; Grammer and Ginsburg, 1992). In contrast, the continuous coral record obtained from onshore Tahiti suggests that there are no major changes in the rate of sea level rise during the time of the inferred post–Younger Dryas (YD) meltwater pulse (MWP-1B) (Bard et al., 2010). A third Acropora reef-drowning event at ~7600 cal. y BP was reported by Blanchon and Shaw (1995). Thus, there are still uncertainties about the general pattern of sea level rise during the last deglaciation, including the duration and amplitude of the maximum lowstand during the Last Glacial Maximum (LGM) and potential links between increased glacial meltwater and accelerated sea level rise (Broecker, 1990). Furthermore, sawtooth sea level fluctuations between 19,000 and 15,280 cal. y BP (Locker et al., 1996; Yokoyama et al., 2000, 2001c, 2006b); the precise timing, rate, and amplitude of MWP-1A; and a sea level fall coeval with climatic change around 11,000 cal. y BP are still controversial topics (Lambeck et al., 2002). Obtaining direct sea level information based on coral reef records older than the last glacial is difficult, and few studies have been reported. The Expedition 310 Tahiti coral record extends back past the last interglacial and provides important insights into the climate system during the penultimate deglaciation (Thomas et al., 2009; Fujita et al., 2010). Offshore drilling at the shelf edge of the Great Barrier Reef (GBR) will provide paleoclimate and sea level records extending into marine isotope Stage 3 and beyond. Sea level compilations indicate that local sea level histories varied considerably around the world in relation to both the postglacial redistribution of water masses and to a combination of local processes (Lambeck, 1993; Peltier, 1994; Yokoyama et al., 2001b, 2001c; Lambeck et al., 2003, 2006; Milne et al., 2009), although significant deviations between model predictions and field data have been noted for several regions (Camoin et al., 1997). Post-LGM sea level changes at far-field sites provide basic information regarding the melting history of continental ice sheets and the rheological structure of Earth. The effect of hydro-isostasy on local sea level will depend on the size of the adjacent landmass: near small islands, the addition of meltwater will produce a small differential response between the island and the seafloor, whereas the meltwater load will produce significant differential vertical movement between larger islands (or continental margins) and the seafloor (Nakada, 1986; Yokoyama et al., 1996). Thus there is a need to establish the relative magnitudes of hydro-isostatic effects at two ideal sites, one on an oceanic island and another on a continental margin, located at a considerable distance from the major former ice sheets. It is essential that both sites are tectonically stable throughout the time period proposed for the investigation so that the proposed Northern and Southern Hemisphere deglaciation curves from Barbados and New Guinea can be rigorously tested. Tahiti (Expedition 310, completed in 2006) and the GBR (Expedition 325, completed in 2010) are ideal locations in which to perform these tests. Climatic and oceanographic changes during the last deglaciationThe Quaternary period of Earth's history is marked by major cyclical changes in global climate reflected in the growth and decay of high-latitude ice sheets. We are currently in a relatively warm interglacial following the LGM, which occurred around 21,000 cal. y BP (Yokoyama et al., 2000; Mix et al., 2001). These glacial–interglacial climate oscillations are related to cyclical changes in the distribution of incoming solar radiation due to variations in Earth's orbit around the sun (so-called "Milankovich cycles"). However, it is also clear that strong feedbacks within the earth system operate to amplify and modify the initial forced changes. Understanding the nature of these feedbacks and the mechanisms through which they influence the timing, rates, and magnitude of climate change remain outstanding issues in climate science. In this context, the tropical oceans play a crucial role in modulating global climate on glacial–interglacial to interannual (i.e., El Niño Southern Oscillation [ENSO]) timescales. One of the key objectives of Expedition 325 is to elucidate the timing, magnitude, and mechanisms of tropical climate change across a major climate transition—namely, from the peak of the last glaciation (the LGM) to the relative warmth of the early Holocene. Paleodata indicate that the LGM mean global surface temperature was cooler than it is at present by several degrees (Mix et al., 2001). However, there were large latitudinal differences in the magnitude of this cooling, with the tropics in general showing less difference (with respect to preindustrial temperatures) than the high latitudes. Estimates of tropical SSTs based on proxies in deep-sea sediment cores now indicate a mean LGM cooling of ~1.7°C compared to the present, with significant regional variations (MARGO Project Members, 2009; Otto-Bliesner et al., 2009) that include a ~3°C cooling in the western sector of the Western Pacific Warm Pool (WPWP) (e.g., Linsley et al., in press) (Fig. F1B). At a few sites, late-glacial age corals (although not LGM) have been used to estimate cooling (Guilderson et al., 1994; McCulloch et al., 1996; Beck et al., 1997; Gagan et al., 2000). Some of these estimates are similar to those derived from deep-sea sediments, whereas others indicate larger differences (e.g., up to 4°–6°C cooling). Resolving the inconsistencies between proxy-based reconstructions remains an important priority; subtle diagenesis of some coral samples (e.g., Allison et al., 2005), possible changes in oceanic Sr/Ca affecting coral SST reconstructions (e.g., Stoll and Schrag, 1998; Martin et al., 1999), possible seasonal biases in climate reconstruction from deep-sea sediments, and real differences in regional climate are all possible contributing factors. The main transition from glacial to interglacial climate occurred in the interval 19,000–9,000 cal. y BP. However, the rise in temperature (and sea level) was not simple and smooth, at least regionally. Greenland ice cores, as well as North Atlantic deep-sea sediment records, suggest that there was a severe climate reversal during the course of the last deglaciation around 12,000–13,000 cal. y BP during the YD (Fig. F1A). Outside the North Atlantic region, diverse paleoclimate evidence suggests synchronous climate events were widespread in the Northern Hemisphere (e.g., Wang et al., 2001; Yuan et al., 2004; Yokoyama et al., 2006a). However, until recently, there was sparse evidence for the YD in the tropics. A coral Sr/Ca-based SST reconstruction from Vanuatu was used to suggest that SST during the YD was ~4°C cooler (Corrège et al., 2004). Fossil coral records from Expedition 310 in Tahiti also captured ~3°C cooling (Asami et al., 2009) and oceanic environmental change during the YD (Inoue et al., 2010). More recently, Griffiths et al. (2010) made paired measurements of calcite δ18O and fluid inclusion δ18O in stalagmites from southern Indonesia to estimate air temperatures ~6°C cooler during the YD. The same stalagmite δ18O records indicate that the Indonesian-Australian monsoon was stronger during the YD, in contrast to the weaker YD monsoon recorded by stalagmite δ18O records from China (e.g., Yuan et al., 2004). It is important to acknowledge that some of these stalagmite- and coral-based reconstructions of tropical temperatures show YD cooling that is significantly greater than that estimated from deep-sea sediment cores. More coral data are required to help resolve this issue. Recent studies in the tropical western Pacific have documented Holocene climatic variations, including ~0.5°–1°C warmer SSTs in the GBR and WPWP during the early Holocene, based on analysis of Sr/Ca in corals and Mg/Ca in planktonic foraminifers (Gagan et al., 2004; Stott et al., 2004; Linsley et al., in press), and parallel analyses of δ18O indicate that the surface ocean in the WPWP freshened through the Holocene. Coral, lake, and geoarchaeological evidence suggest that ENSO variability was substantially reduced in the early to middle Holocene (e.g., Sandweiss et al., 1996; Rodbell et al., 1999; Tudhope et al., 2001; Moy et al., 2002; McGregor and Gagan, 2004), a finding that challenges state-of-the-art ocean-atmosphere general circulation models, which, for the most part, reconstruct more modest changes in ENSO for this interval (e.g., Liu et al., 2000; Otto-Bliesner et al., 2003; Brown et al., 2008). New speleothem δ18O records from southern Indonesia also suggest that monsoonal rainfall was weaker during the early Holocene (Griffiths et al., 2009), and a coral record from Expedition 310 in Tahiti suggests local SSTs may have been cooler than they are at present (DeLong et al., 2010). Given the continued uncertainties in constraining the full range of tropical western Pacific climate and the importance of this region to global climate, additional paleodata are required. Some of the most debated points are
New approaches to these long-standing challenges will be provided by IODP Expedition 325 at the GBR. The Great Barrier Reef: its suitability, previous results, and promiseThe GBR is the largest epicontinental reef system on Earth, extending 2000 km in a northwest–southeast direction along the northeast coast of Queensland (Australia) (Davies et al., 1989) (Fig. F2). The origin of this morphologically and biologically important sedimentary system is poorly constrained, with ages of <500 k.y. before present assigned to the initiation of the northern GBR system (McKenzie, Davies, Palmer-Julson, et al., 1993; Davies and Peerdeman, 1998; International Consortium, 2001; Webster and Davies, 2003; Braithwaite et al., 2004) and, more recently, ages between 560 and 670 k.y. before present assigned to the southern GBR (Dubois et al., 2008). The northern, central, and southern GBR define ideal sites for the evaluation of sea level changes during the period from 20,000 to 8,000 cal. y BP. The reefs on the shelf edge east of Cooktown (Australia) form the semicontinuous outer barrier of the northern GBR. In this area, as well as in the far northern GBR, the reef is narrow, with ribbon reefs on its eastern edge and extensive coastal fringing reefs and patch reefs. In the south, the GBR broadens, with patch reefs separated by open water or narrow channels. On the outer shelf east-northeast of Townsville (Australia), modern reefs form a line of pinnacles seaward of the main reef edge with lateral growth on the windward margin. South of 15°30′S, the reefs are generally ≥30 km offshore and reach 100 km offshore at 22°30′S. Farther south, the shelf widens considerably to >200 km. East of Mackay (Australia), the modern reefs form a complex series of flood-tide deltaic reefs (i.e., Pompey Complex) (Hopley, 2006). The coastal lagoon between the main GBR reef tract and the mainland has a maximum depth of 145 m but rarely exceeds 60 m (Wolanski, 1982). Studies of the GBR (McKenzie et al., 1993; Davies and Peerdeman, 1998) focused on the areas southeast of Townsville and east of Cooktown and defined the morphologic shape of the outer reef and upper continental slope, as well as the geological origin of the GBR itself. Based on high-resolution seismic profiles in the fore reef section in front of the GBR, Feary et al. (1993; in McKenzie et al., 1993) recognized three seismic mega-sequences that define a clearly aggradational upper sequence (0–490 ms), a transitional middle sequence (490–555 ms), and a progradational lower sequence (below 555 ms). In 1991, indirect evidence that the GBR is very young, having initiated during marine isotope Stages 9–11 (McKenzie et al., 1993; Davies and Peerdeman, 1998) was recovered during Ocean Drilling Program (ODP) Leg 133. In 1995, a new phase of drilling at Boulder Reef (15°23.944′S, 145°26.182′E; 86 m core depth below seafloor [CSF-A]) and Ribbon Reef 5 (15°22.40′S, 145°47.149′E; 210 m CSF-A), using a reef-mounted jack-up platform, further enhanced this story, proving that the northern GBR is ~100 m thick and rests on a subreef subtropical red coralline algal facies that, in turn, overlies a deepwater temperate grainstone facies (Davies and Peerdeman, 1998; International Consortium, 2001). Strontium isotope and magnetostratigraphic data from the base of the Pleistocene coral reef sequence have confirmed that the origin of the GBR is very young, perhaps <500,000 y (International Consortium, 2001; Webster and Davies, 2003; Braithwaite et al., 2004). Detailed stratigraphic and sedimentary facies analysis of the 210 m long Ribbon Reef 5 drill core shows that the upper part of the platform is composed of cycles of transgressional cool-water coralline-dominated carbonates topped by shallow-water highstand coral reefs (Webster and Davies, 2003; Braga and Aguirre, 2004). However, the Holocene reef does not show this cyclic sedimentary couplet, as it is coral dominated from its inception at 8000 cal. y BP. Previous sedimentological and geophysical studies on the shelf edge have identified a succession of subsea morphologic structures interpreted as drowned reefs at 100, 90, 60–50, and 35–40 mbsl (Carter and Johnson, 1986; Harris and Davies, 1989; Larcombe et al., 1995; Hopley, 2006; Beaman et al., 2008), especially in the following four areas:
For example, a series of drowned linear reefs and lagoons occupy specific depths over at least a 30 km stretch on the outer continental shelf in the vicinity of Hydrographer's Passage in the southern GBR region. Based on the R/V Southern Surveyor cruise in September–October 2007, Webster et al. (2008a, 2008b) identified five primary drill site transects from three of these key regions on the Cooktown, Cairns, and Mackay shelf edges (Fig. F2). Proposed drill sitesAvailable site survey data at the Great Barrier ReefThe proposed drill sites on the GBR are distributed in three distinct regions (Fig. F2): Offshore Cooktown (Ribbon Reef 5 and 3), Offshore Cairns (Noggin Pass), and McKay shelf (Hydrographer's Passage). From previous site survey data (described in detail in the June 2007 preliminary report to the Environmental Protection and Safety Panel [EPSP] and recently synthesized by Beaman et al., 2008), it was clear that a succession of barrier reefs occupy the outer shelf between 40 and 100 mbsl with terrace features at ~80–110 mbsl along much of the GBR. These features have not previously been investigated in detail. For example, with the exception of the Ribbon Reef 5 region, only limited systematic high-resolution swath bathymetry mapping, imaging, or sampling had ever been attempted. However, it is clear that these submerged reef structures have the potential to provide unique and critical information about the nature of sea level and climatic change offshore eastern Australia and important information about their role as habitats and substrates for present-day biological communities. Prior to Expedition 325, proponents led a site survey cruise to gather the most comprehensive data set ever collected from the GBR shelf edge (Webster et al., 2008b). The cruise on the Southern Surveyor acquired the remaining site survey information needed for IODP drilling operations in the GBR. Four study sites (Ribbon Reef, Noggin Pass, Viper Reef, and Hydrographer's Passage) were mapped along the Queensland margin where the approximate location of submerged reefs is known. The data types acquired and submitted to the IODP Site Survey Database (SSDB) were:
These data were used to define specific drill targets for Expedition 325 drilling operations (see "Operational strategy" for proposed transect locations and available site survey information). Summary of 2007 Great Barrier Reef site survey data for the proposed sitesMackay shelf, Hydrographer's Passage (HYD-01C and HYD-02A)EM300 swath mapping of the Hydrographer's Pass survey area covers 810.68 km2. Based on a detailed examination of all available site survey data, we proposed to drill two transects of holes across the best developed fossil reef features, one in the northwest (HYD-01C) and the other in the southeast (HYD-02A). See Figure F2 for the general location, Figures F3 and F4 for detailed maps of HYD-01C, and Figures F5 and F6 for detailed maps of HYD-02A. Site survey data from the northwest of Hydrographer's Pass illustrate the succession of morphological features that define the HYD-01C transect:
Site survey data from the southeast of Hydrographer's Pass illustrate the succession of morphological features that define the HYD-02A transect:
Offshore Cooktown, Ribbon Reef 5 and 3 (RIB-01C and RIB-02A)EM300 swath mapping of the Ribbon Reef survey area covers 1609.87 km2. The Ribbon Reef 5 area was also surveyed by Webster and colleagues in 2005 using a Reson 8101 (240 kHz) swath mapping system and Datasonics CAP-6600 Chirp 3.5 kHz subbottom profiler (Beaman et al., 2008). Based on a detailed examination of all available site survey data (multibeam, backscatter, seismic profiles, AUV imagery, and bottom samples), the initial plan was to drill two transects of holes during Expedition 325, across the best developed fossil reef features, one off Ribbon Reef 5 (RIB-01C) and another off Ribbon Reef 3 (RIB-02A). However, only one transect (RIB-02A) was drilled, and so only the details regarding that particular transect are included in this report. See Figure F2 for the general location and Figures F7 and F8 for detailed maps. Site survey data seaward of modern Ribbon Reef 3 illustrate the succession of morphological features that define the RIB-02A transect:
Offshore Cairns, Noggin Pass (NOG-01B)EM300 swath mapping of the Noggin Pass survey area covered 1243.27 km2. See Figure F2 for the general location and Figures F9 and F10 for detailed maps. Available site survey data seaward of modern Noggin Reef illustrate the succession of morphological features that define the NOG-01B transect:
| ||
