Sea-level during the early deglaciation period in the Great Barrier Reef, Australia

Transcription

1 Global and Planetary Change 53 (2006) Sea-level during the early deglaciation period in the Great Barrier Reef, Australia Yusuke Yokoyama a,b,, Anthony Purcell c, John F. Marshall c, Kurt Lambeck c a Department of Earth and Planetary Sciences, University of Tokyo, Tokyo , Japan b Institute for Research on Earth Evolution, Japan Agency for Marine-Earth Science and Technology, Japan c Research School of Earth Sciences, The Australian National University, Canberra, ACT0200, Australia Received 31 August 2004; accepted 20 January 2006 Available online 22 June 2006 Abstract Ooids samples recovered from the Capricorn Channel in the southern Great Barrier Reef (GBR) were analyzed to study paleo sea-level during the late glacial period. Their ages and depth distribution were compared with numerical predictions obtained using a detailed glacio-hydro isostatic model to examine the reliability of the ice model. A step-wise dissolution technique was used to remove secondary carbon contamination and determine the exact timing of formation. The dating results indicate that ooid formation took place around 16,800 cal. yr BP, immediately prior to a period of accelerated melting of the global ice sheets, the Meltwater pulse 1a event. Sea-level predictions obtained using rheological parameters optimized for the Australian coast are consistent with the radiocarbon-derived ooid depth age data, which suggest that at 16.8 cal. ka sea-level in the southern GBR was 100 m below its present height Elsevier B.V. All rights reserved. Keywords: sea-level; Great Barrier Reef; ooids; last deglaciation; radiocarbon dating 1. Introduction Sea-level during the Last Glacial Maximum (LGM) has been observed as much as ca. 130 m below its present value (Yokoyama et al., 2000a, 2001a,b) for sites remote from former ice sheets (far-field sites). However, sea-level lowstand observations for far-field sites are sparse, and more data are required to accurately constrain global ice volumes as a function of time. Corresponding author. Department of Earth and Planetary Sciences, University of Tokyo, Tokyo , Japan. Tel.: ; fax: address: yokoyama@eps.s.u-tokyo.ac.jp (Y. Yokoyama). Investigations of Last-Interglacial (LIG) shorelines around the Australian continent permit identification of several distinct zones each with its own tectonic setting (Murray-Wallace and Belperio, 1991). Shoreline data of this type can be used to determine the level of tectonic activity during the last ca. 130 ka (eg., Marshall and Thom, 1976; Schwebel, 1984; Marshall and Davis, 1984; Belperio et al., 1995). These studies consistently suggest that since ca. 130 ka, there has been little tectonic activity along the eastern coast of Australia, which makes this region very well-suited to the study of sea-level change. The Australian coastline has been extensively studied for evidence of palaeo-sea-level change, but most of the observations so far obtained are for the Holocene and LIG periods (eg., Stirling et al., 1995), with very few /$ - see front matter 2006 Elsevier B.V. All rights reserved. doi: /j.gloplacha

2 148 Y. Yokoyama et al. / Global and Planetary Change 53 (2006) observations available during the Last Glacial Maximum (LGM) and Late Glacial (Ferland et al., 1995). Global observations indicate that eustatic sea-level (esl) during the LIG was ca. 5 7 m higher than the present (eg., Stirling et al., 1995) and consistent shoreline elevations have been reported around Australia(Murray- Wallace and Belperio, 1991). Similarly, coral data from around the world have been used to study far-field Holocene sea-levels and observations from the Australian coast are found to be in good agreement with the global trend for this period (eg. Woodroffe, 2002). It is however more difficult to obtain far-field observations of sea-level change during the glacial period because these sites were inundated by rising sea-levels during the early Holocene. In particular, almost no data for LGM sea levels exists for the GBR, with the exception of a solitary observation by Veeh and Veevers (1970). In this paper we report radiocarbon dating results from ooid samples recovered from the Capricorn Channel in the southern GBR and use the results of detailed glacio-hydro-isostatic modeling to verify the credibility of the ice model to predict sea-level history during the last deglaciation. 2. Ooids Ooids have been widely studied in modern and geological settings, they are spherical, inorganically precipitated calcium carbonate grains with diameters in the range ( mm) that are formed in shallow marine or marginal marine environments (usually shallower than 5 m). The Atlantic continental shelf off the southern United States was examined where we could find carbonate sediments (Pilkey et al., 1966, 1969). They found the ooids formation off the Florida and concluded that the formation was taken place at palaeo shoreline environment during the sea-level lowstand and the existing of the ooids suggests low sedimentation rates in the region during the last 20,000 years (Pilkey et al., 1969). Each ooid consists of a nucleus, usually a sand-size quartz grain or carbonate particle, surrounded by a cortex of precipitated CaCO 3 whose internal architecture (as seen in thin section) shows either concentric or radial structure, or sometimes both. The mineralogy of modern ooids is aragonite or Mg-calcite, whereas older ooids have been converted to calcite (eg., Taylor and Illing, 1969; Purser and Loreau, 1973). Most modern examples are aragonite and exhibit a concentric structure; this type is commonly referred to as Bahamian ooids. The less common Mgcalcite ooids tend to be radial. The concentric structure is believed to be produced as a result of nearly continuous bottom agitation, whereas the radial structure is believed Table 1 Earth response parameters used in this study H l (km) η um ( Pa s) to result from growth in more protected, but still relatively high energy environments. 3. Glacio-hydro-isostasy η lm ( Pa s) The average change in water depth across the area of the ocean basins is called the change in ice volume equivalent sea-level or eustatic sea-level (esl) which is related to the change in volume of the ice sheets as follows: Df esl ð/; tþ ¼ q i =q o DV i ðtþ=a o ðtþ ð1þ where ΔV i (t) is the change in ice sheet volume between the present and time t, A o (t) is the area of the ocean at time t, and ρ i and ρ o represent density of ice and ocean-water respectively. The observed change in relative sea-level (rsl) at a particular site is the difference between past and present sea-level elevation at that location. In the absence of tectonic and geological signals, this quantity is in turn comprised of the change in esl and the isostatic component calculated for that position (Lambeck et al., 2003) and may be expressed mathematically as a sum of corresponding components: Df rsl ð/; tþ ¼Df esl ð/; tþþdf iso ð/; tþ ð2þ where Δζ rsl (ϕ, t) represents the change in rsl at location ϕ between time t and the present, while Δζ iso (ϕ, t)represents the isostatic component of sea-level change which includes the effects of both ice and water loading. Since relative sea-level change is a function of both position and time, it is necessary to extract the isostatic component as well as any tectonic component from rsl observations in order to deduce the value of the esl component (Yokoyama et al., 2000a, 2001a; Lambeck et al., 2002). That is: Df esl ð/; tþ ¼Df obs rsl ð/; tþ Df isoð/; tþ ð3þ For the purposes of numerical modeling a 3-layer Maxwell visco-elastic earth model is employed, consisting of an elastic lithosphere, an upper mantle (assumed to extend from the base of the lithosphere to the 670 km seismic discontinuity), a lower mantle (lying between 670 km and the core mantle boundary), and a liquid core. The viscosity values for the optimal earth model for the Australian coast are listed in Table 1 and were determined from comparison between observed and calculated sea-

3 Y. Yokoyama et al. / Global and Planetary Change 53 (2006) Fig. 1. Location of the Capricorn Channel and transects for the sea-level model predictions. Contour intervals for top panel and the lower panels are 200 m and 100 m respectively. Site A1 is located at the present day shoreline whereas A5 locates most offshore site. Other locations (A2 A4) are in between the two sites with equally distributed. B1 is the site on the B side of the line and B3 is at the B edge. The site B2 is the middle point of the B B line. Shaded area shows the distribution of the ooids sand (Marshall, 1977). levels for the Holocene (Nakada and Lambeck, 1989; Lambeck and Nakada, 1990; Lambeck et al., 2002). 4. Location and samples The Capricorn Channel is located at the southern end of the Great Barrier Reef (Fig. 1). The surrounding area is considered to have been tectonically stable for the last 150 ka although no Last Interglacial sea-level indicators have been found above present day sea-level (Marshall and Davis, 1984). This suggests that either the shoreline has not been preserved or some long term subsidence has taken place. The heights of sea-level indicators formed during the middle of the Holocene sea-level high stand (ca cal yr BP) accord with the results predicted by glacio-hydro-isostatic models and with data from other localities in Australia. This indicates that the region has undergone little tectonic activity through the late Holocene. However, the possibility of a small amount of subsidence (ca. 5 7 m/135 ka) cannot be excluded. Several investigations conducted using bathymetric and seismic survey results have established possible positions of various paleo-shorelines in the region (Carter and Johnson, 1986). No radiometric ages are available for these observations, but their surveys revealed several features, including terraces at depths of 103, 114 and 133 m below present sea-level. Veeh and Veevers (1970) used a submersible to recover a coral from a depth of 175 m on the foreslope of the southern GBR. They dated the coral using the uranium series method and obtained an age of 17+/ 1 cal ka using α- spectromentry (Veeh and Veevers, 1970). This age was confirmed by Yokoyama et al. (2000b) using TIMS, which produced an age of 16.9+/ 0.1 cal ka. However, the habitable depth range of this species of coral is between m and the coral was not in growth

4 150 Y. Yokoyama et al. / Global and Planetary Change 53 (2006) position, making this observation less than entirely reliable as an indicator of sea-level. It remains the case therefore, that no reliable sea-level data for the post LGM Holocene period has so far been reported from this region. Ooid samples were recovered by the Australian Bureau of Mineral Resources as part of a marine geological survey in this region (Marshall, 1977). The samples were collected using small pipe dredges. X-ray diffraction results and SEM/EDAX analysis showed that the ooids are all Mgcalcite (15 mol% MgCO 3 )(Marshall and Davis, 1975). 5. Dating Radiocarbon dating was conducted in order to obtain the age of the ooids. Because the nuclei consist of older carbonate material, a bulk radiocarbon measurement of the ooids tends to bias the determination towards an older age than the actual formation of the ooid cortex (Sarnthein, 1972; Kump and Hine, 1986). In order to overcome this, a step-wise dissolution technique was employed to produce the target carbon for AMS (Accelerator Mass Spectrometry) measurements (Burr et al., 1992; Yokoyama et al., 2000b). The ultrasonically cleaned samples were immersed in phosphoric acid in a glass apparatus designed for this experiment (Yokoyama et al., 2000b). The apparatus is then attached to a vacuum line and evacuated. Once the vacuum reached a level of 10 5 Torr the sample was dissolved in the phosphoric acid. Sample CO 2 was collected cryogenically and the pressure monitored by a transducer. The sample gas aliquot was then placed into a glass tube with iron powder and reduced to graphite under a hydrogen atmosphere in a high temperature furnace. The measured radiocarbon ages were converted to calendar ages using CALIB4.4 (Stuiver et al., 1998). 6. Results 6.1. Glacio-hydro-isostasy across the GBR Since the GBR is far away from former ice sheets, the most significant contribution to observed sea-level change is the esl function. The isostatic component is still not negligible however, particularly the water loading term, so that in order to use our observed values of rsl to obtain constraints on esl (using Eq. (3)) it is necessary to numerically model the magnitude of the isostatic component and determine its spatial and temporal variability. The dependence of sea-level model predictions on the rheological properties of the earth has been examined in some detail for this region (eg. Lambeck, Fig. 2. Predicted relative sea-level for both A A and B B transects in the Capricorn Channel area. At all sites and for all times, relative sea-level is shallower than eustatic sea-level due to the contribution of the isostatic component. Relative sea-level values for the transect parallel to the present day coastline (b) show little spatial variability whereas the sites in the perpendicular transect (a) exhibit large differences in rsl due to the effects of differential loading of the seafloor. 2002) and the earth model employed in our analysis was the optimal set of rheological parameters deduced for the GBR region. The rsl curves for sites along the transect lines A A and B B (Fig. 1) are given in Fig. 2. For all sites the rsl curve lies above (shallower than) the esl curve throughout the period of interest. Because of the water loading effect, the magnitude of the isostatic component reaches as much as 25 m during the LGM even though the GBR is a far-

5 Y. Yokoyama et al. / Global and Planetary Change 53 (2006) field site. Spatial variability for sites located along the transect A A is large, rsl values differing in some cases by as much as 15 m because of the variation in distance of the site from the coast. This reflects the effect of crustal tilting due to the spatial variability of the water load, with sites closer to the modern coastline experiencing later inundation and a smaller water load, and thus having a shallower rsl. This is confirmed in Fig. 2b which shows the rsl curve for 3 sites along the B B transect which is roughly parallel to the present day coastline. The spatial variability of the rsl signal is significantly reduced for these sites Observed sea-level from the ooids and implication of esl Radiocarbon dating results are shown in Table 2.The difference between the ages obtained for the outer- and inner-most layers is to be expected, since SEM and thin section observation reveal older nuclei enveloped by the younger outer coatings (Marshall and Davis, 1975). Given that the cortex of the ooids is precipitated in a relatively short time interval, the radiocarbon dating results obtained from the cortex would be expected to show nearly identical ages. However, the results from the sequential dissolution shows that the obtained ages range from ca ka 14 C BP. We aligned the size distribution of the ooids ( mm) so that progressive dissolution recovered, more or less the same portion of the ooids cortex from each grain. The age results obtained from the outermost section of the ooids shows a much younger age than the others, suggesting some contamination by younger carbon. Petrographic and SEM examination of the outer surface shows evidence of boring by endoliths and some secondary infill. On the other hand, the dating result obtained from the final split of sample CO 2 suggests much older carbon contamination. This is consistent with the petrographic observation that the nucleus commonly consists of detrital foraminifera, coral, and other biogenic calcite (Marshall and Davis, 1975). It seems that some carbon contribution from the nucleus has produced the apparently older radiocarbon age. On the other hand, the radiocarbon ages from the two intermediate portions of the Table 2 Radiocarbon dating results of ooids sample Sample Dissolved percent Radiocarbon age (yr BP) Calendar age (cal yr BP) CPR ,350+/ ,830 CPR ,340+/ ,605 CPR ,770+/ ,100 CPR ,940+/ ,050 Fig. 3. Comparison between predicted sea-level and the observed depth age of the ooids recovered from the GBR (box). The close agreement between the model predictions for this site and time, and the observed value of sea-level change derived from the ooid sample would seem to indicate that ooid samples are potentially a very precise sea-level indicator and would repay further systematic study. cortex are identical within analytical uncertainties (Table 2). The study using conventional β-ray counting method on the Bahamian bank oolitic sands also reported the influence of nuclei on the radiocarbon measurements (Martin and Ginsburg, 1965). Therefore radiocarbon dating results on the intermediate portion of the cortex is often employed as the reliable number since it is thought to be free from both younger and older carbon contaminations (eg., Pilkey et al., 1969; Locker et al., 1996; Wiedicke et al., 1999). The ooids were recovered from a depth of 123 m as illustrated in Figs. 1 and 3. A detailed survey of the region conducted by the Australian Bureau of Mineral Resources determined the distribution pattern of ooids in the Southern GBR (Marshall, 1977). Since ooids are found throughout this area, but only at depths between 100 to 120 m, we should consider the possibility of secondary transportation of the ooids from shallower sites following their original deposition (eg., Kumar et al., 1977; Kump and Hine, 1986). A source depth shallower than 100 m is not a candidate for the origin of the ooids sands since surface sediments from these depths contain no ooids and the physical oceanography of the area would not have permitted re-deposition (Marshall, 1977). We conclude that the formation of the ooids took place at approximately 100 m water depth when the site was just below sea-level at 16.8 cal ka. The timing of the ooids formation is consistent with previously published results. Series of oolitic sand dune submerged off the Florida were radiocarbon dated using

6 152 Y. Yokoyama et al. / Global and Planetary Change 53 (2006) AMS and they found 3 distinct brief sea-level still stands at 14.5, 14.0, and Ckarespectively(Locker et al., 1996). The timing of the first sand dune formation is consistent with the GBR ooids formation but the Florida's ooids sand dune was found at 80 m depth of sea-level. The apparent discrepancy in terms of the depth between the two locations is expected since Florida is much closer to the former ice sheet (Laurentide ice sheet) and hence the ice loading component is much larger than that for the GBR (eg., Potter and Lambeck, 2003). Therefore the correction of the glacio-hydro-isostasy on the observed palaeo-sea-level indicators is essential, and the data can be compared each other once it would be done (Lambeck et al., 2002). The observed sea-levels are consistent with predicted sea-level off Florida (Lambeck et al., 2002) so that our currently employed ice model is also validated. Ooid formation is not observed in the present day GBR. The observed age of formation of the ooid samples considered in this study coincides with a period when eustatic sea-level rise slowed following the first meltwater pulse at 19 cal ka at the end of the LGM (Yokoyama et al., 2000a; Clark et al., 2004). Conditions were apparently favourable for ooid growth at this time as the region was inundated but formation ceased when the rise in eustatic sea-level again accelerated at about 16 cal ka during MWP 1a (Fairbanks, 1989; Lambeck et al., 2002). Ooid growth occurred when the rate of eustatic sea-level rise was about 3.3 mm/year and stopped when it accelerated by a factor of 5 (ca mm/year). Ooid formation is dependent on both the chemistry of the ocean water and the physical oceanographic conditions prevailing at the particular site. They seem to require strong wave action as well as water chemically saturated in calcium carbonate. The degree to which ocean chemistry varies between glacial and inter-glacial periods is still being debated (Opdyke and Walker, 1992), yet apparent synchronism of the ooids formation at both GBR and off Florida may support this possibility. Scrutiny of the ooids' structure could be a potential tool to address this problem and further samples with a wider geographical and temporal distribution could potentially make ooids an important paleo-environmental proxy. 7. Conclusions We report here the first sea-level data from the GBR as old as 16 cal ka other than a single report of a loose coral. Ooids are found in the Capricorn Channel only in the depth range from 100 to 120 m and yield an age of formation of 16.8 cal ka under stepwise dissolution radiocarbon dating. This is the first time ooids have been radiocarbon dated using this technique and the results obtained clearly show older and younger carbon contamination for the inner- and outer-most sample portions respectively, though the close agreement between the results obtained for the two central portions of the ooids gives us reasonable confidence in the accuracy of the dates obtained for these layers. The results of glaciohydro-isostatic modeling of sea-level change in the GBR were compared with the data obtained from the ooids. The agreement between the observations and the model results suggests the reliability of the ice model we constructed to model glacio-hydro-isostasy. There may also be a potential use for ooid samples in studies of the history of sea-water chemistry though there is still a need for more detailed geochemical investigation to understand the exact relationship between ooid formation and sea-water chemistry. Any such study would require more observational data and a detailed model of ooid development. Acknowledgements We thank two anonymous referees and Y. Rosenthal for their constructive comments. The work presented here was partly supported by Grants-in-Aid from the JSPS and COE project at the University of Tokyo. This paper is also a contribution to the INQUA working group on Coastal and Marine Processes. References Belperio, A.P., Murray-Wallace, C.V., Cann, J.H., The Last Interglaial shoreline in southern Australia: morphostratigraphic variations in a temperate carbonate setting. Quaternary International 26, Burr, G.S.E., Edwards, R.L., Donahue, D.J., Druffel, E.R.M., Taylor, F.W., Mass spectrometric 14 CandU Th measurements in coral. Radiocarbon 34 (3), Carter, R.M., Johnson, D.P., Sea-level controls on the postglacial development of the Great Barrier Reef, Queensland. Marine Geology 71, Clark, P.U., McCabe, A.M., Mix, A.C., Weaver, A.J., Rapid rise of sea-level 19,000 years ago and its global implications. Science 304, Fairbanks, R.G., 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, Ferland, M.A., Roy, P.S., Murray-Wallace, C.V., Glacial lowstand deposits on the outer continental shelf of southeastern Australia. Quaternary Research 44 (2), Kumar, N., Damuth, J.E., Gorini, M.A., Relict magnesian calcite oolite and subsidence of the Amazon shelf: discussion. Sedimentology 24, Kump, L.R., Hine, A.C., Ooids as sea-level indicators. In: Van de Plassche, O. (Ed.), Sea-Level Research: a Manual for the Collection and Evaluation of Data. Geo Books, Norwich, pp Lambeck, K., Sea-level change from Mid-Holocene to Recent time: an Australian example with global implications. In: Mitrovica, J.X., Vermeersen, B.L.A. (Eds.), Glacial Isostatic Adjustment and the

7 Y. Yokoyama et al. / Global and Planetary Change 53 (2006) Earth System. American Geophysical Union, Washington DC, pp Lambeck, K., Nakada, M., Late Pleistocene and Holocene sealevel change along the Australian coast. Palaeogeography, Palaeoclimatology, Palaeoecology(Global and Planetary Change Section) 89, Lambeck, K., Yokoyama, Y., Purcell, A., Into and out of Last Glacial Maximum: sea-level change during the oxygen isotope stage 3 and 2. Quaternary Science Reviews 21, Lambeck, K., Purcell, A., Johnston, P., Nakada, M., Yokoyama, Y., Water-load definition in the glacio-hydro-isostatic sea-level equation. Quaternary Science Reviews 22, Locker, S.D., Hine, A.C., Tedesco, L.P., Shinn, E.A., Magnitude and timing of episodic sea-level rise during the last deglaciation. Geology 24, Marshall, J.F., Marine Geology of the Capricorn Channel Area, vol Bureau of Mineral Resources. Marshall, J.F., Davis, P.J., High-magnesium calcite ooids from the Great Barrier Reef. Journal of Sedimentary Petrology 45 (1), Marshall, J.F., Davis, P.J., Last interglacial reef growth beneath modern reefs in the southern Great Barrier Reef. Nature 307, Marshall, J.F., Thom, B.G., The sea-level in the last interglacial. Nature 263, Martin, E.L., Ginsburg, R.N., Radiocarbon ages of oolitic sands on Great Bahama Bank. In: Pullman (Ed.), The Sixth International Conference on Radiocarbon and Tritium Dating, Washington, pp Murray-Wallace, C.V., Belperio, A.P., The last interglacial shoreline in Australia a review. Quaternary Science Reviews 10, Nakada, M., Lambeck, K., Late Pleistocene and Holocene sealevel change in the Australian region and mantle rheology. Geophysical Journal International 96, Opdyke, B., Walker, J.C.G., Return to the coral reef hypothesis: basin to shelf partitioning of CaCO 3 and its effect on atmospheric CO 2. Geology 20, Pilkey, O.H., Schnitker, D., Pevear, D.R., Oolites on the Georgia continental shelf edge. Journal of Sedimentary Petrology 36, Pilkey, O.H., Blackwelder, B.W., Doyle, L.J., Estes, E., Terlecky, P.M., Aspects of carbonate sedimentation on the Atlantic continental shelf off the southern United States. Journal of Sedimentary Petrology 39 (2), Potter, E.-K., Lambeck, K., Reconciliation of sea-level observations in the western North Atlantic during the last glacial cycle. Earth and Planetary Science Letters 217, Purser, B.H., Loreau, J.P., Aragonitic, supratidal encrustations on the Trucial coast, Persian Gulf. In: Purser, B.H. (Ed.), The Persian Gulf. Springer, pp Sarnthein, M., Sediments and history of the postglacial transgression in the Persian Gulf and northwest Gulf of Oman. Marine Geology 12, Schwebel, D.A., Quaternary stratigraphy and sea-level variation in thesoutheastofsouthaustralia.in:thom,b.g.(ed.),coastal Geomorphology in Australia. Academic Press, Sydney, pp Stirling, C.H., Esat, T.M., McCulloch, M.T., Lambeck, K., Highprecision U-series dating of corals from western Australia and implications for the timing and duration of the Last Interglacial. Earth and Planetary Science Letters 135, Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B., McCormac, G., van der Plicht, J., Spurk, M., INTCAL 98 radiocarbon age calibration, 24,000 0 cal BP. Radiocarbon 40 (3), Taylor, J.C.M., Illing, L.V., Recent intertidal calcium carbonate sedimentation at Qatar, Persian Gulf. Sedimentology 12, Veeh, H.H., Veevers, J.J., Sea-level at 175 m off the Great Barrier Reef 13,600 to 17,000 years ago. Nature 226, Wiedicke, M., Kudrass, H.-R., Hubscher, C., Oolitic beach barriers of the last Glacial sea-level lowstand at the outer Bengal shelf. Marine Geology 157, Woodroffe, C.D., Coasts. Cambridge University Press, Cambridge. 623 pp. Yokoyama, Y., Lamberk, K., De Deckker, P., Johnston, P., Fifield, L.K., 2000a. Timing of the Last Glacial Maximum from observed sea-level minima. Nature 406, Yokoyama, Y., Esat, T.M., Lambeck, K., Fifield, L.K., 2000b. Last Ice Age millennial scale climates changes recorded in Huon Peninsula corals. Radiocarbon 42, Yokoyama, Y., De Deckker, P., Lambeck, K., Johnston, P., Fifield, L.K., 2001a. Sea-level at the Last Glacial Maximum: evidence from northwestern Australia to constrain ice volumes for oxygen isotope stage 2. Palaeogeography, Palaeoclimatology, Palaeoecology 165, Yokoyama, Y., Purcell, A., Lambeck, K., Johnston, P., 2001b. Shore-line reconstruction around Australia during the Last Glacial Maximum and Late Glacial Stage. Quaternary International 83 85, 9 18.