Rapid Flooding of the Sunda Shelf: A Late-Glacial Sea-Level Record

Fairbanks R. G., Nature 342, 637 (1989);

; Paleoceanography 5, 937 (1990).

Bard E., Hamelin B., Fairbanks R. G., Zindler A., Nature 345, 405 (1990).

Bard E., et al., Nature 382, 241 (1996).

Montaggioni L., et al., Geology 25, 555 (1997).

Chappell J., Polach H.A., Nature 349, 147 (1991).

Edwards R. L., et al., Science 260, 962 (1993).

Ages in calendar years before present (1950 A.D.) are expressed in thousands of years (ka). ¹⁴C ages have been calibrated to calendar years by using the INTCAL98 calibration set and the calibration program CALIB 4.0 [

Stuiver M., et al., Radiocarbon 40, 1041 (1998);

Bard E., Arnold M., Hamelin B., Tisnerat-Laborde N., Cabioch G., Radiocarbon 40, 1085 (1998);

], which is distinctly different from the earlier version.

Blanchon P., Shaw J., Geology 23, 4 (1995).

Harris P. T., Sediment. Geol. 125, 235 (1999).

H. D. Tjia and K. K. Liew, in Tectonic Evolution of Southeast Asia, R. Hall and D. Blundell, Eds. (Geology Society Special Publication 106, Geological Society, London, 1996), p. 291; T. O. Sinandjuntak and A. J. Barber, in Tectonic Evolution of Southeast Asia, R. Hall and D. Blundell Eds. (Geology Society Special Publication 106, Geological Society, London, 1996), p. 185.

Q. Zhao, Q. Li, P. Wang, Eds. WESTPAC Paleoceanographic Maps, (Unesco/IOC Publication, Marine Geology Laboratory, Tongji University, Shanghai, 1995).

Tjia H. D., Z. Geomorph. N.F. 24, 405 (1980).

Stattegger K., et al., Ber. Rep. Geol. Paläontol. Inst. Univ. Kiel 86, 211 (1997);

. The inner shelf is dominated by low sedimentation rates and erosion after the LGM, the middle shelf by highly differentiated coastal deposits belonging to a transgressive system, and the shelf margin by large depositional hiatuses. A major fault (Lupar line) running from western Borneo, easterly Natuna Island, to the continental slope off Vietnam seems to have been inactive during the last tens of thousands of years (10). Additionally, a system of small-scale faults observed in our seismic survey results in gravitational mass movement near the shelf break.

T. Hanebuth and K. Stattegger, Mar. Geol., in press; T. Hanebuth, unpublished data; A. Richter, thesis, University of Kiel, Kiel, Germany (1998); A. Schimanski, thesis, University of Kiel, Kiel, Germany (1999).

P. A. Pirazzoli, World Atlas of Holocene Sea-Level Changes (Elsevier, Amsterdam, 1991), p. 117; Singapore Tide Gauge (Horsburgh Lighthouse).

AMS ¹⁴C ages were measured with a 3-MV high-voltage engineering Europa Tandetron 4130 AMS system to a precision (counting statistics and machine statistics) of 0.3% for modern samples. This resulted in typical ¹⁴C age uncertainties (1σ) of 50 to 80 years.

Nadeau M.-J., et al., Nucl. Instrum. Methods B 123, 22 (1997);

; M.-J. Nadeau et al., Radiocarbon 40(1), 239 (1998).

P. B. Tomlinson, The Botany of Mangroves (Cambridge Univ. Press, Cambridge, 1986). The mangrove, as a highly sensitive sea-level indicator, belongs to the tidal range between mean sea level and mean high water datum. The detailed biozonation of mangrove habitat is negligible in this case because of the microtidal character in this area.

The ash content of carefully cleaned organic material from the tidal flat environment indicates very rapid burial of litter after death in the mangrove swamp without any significant decay [see

Gastaldo R. A., Allen G. P., Huc A. Y., Geol. Soc. Am. Spec. Pap. 286, 107 (1993);

]. Measurement of δ¹³C_org and the age difference of humic acids to leach residues adds further information to help recognize fresh material.

Supplementary figure is available at www.sciencemag.org/feature/data/1046133.shl.

Supplementary table is available at www.sciencemag.org/feature/data/1046133.shl.

A number of primary and secondary natural effects on control of true ages have been considered—for example, tidal range, material mixing, changes in ¹⁴C production, and subsidence and uplift. There is no evidence for compaction of these young sediments. Bioturbation is also negligible because of very high sedimentation rates locally. If we take into account that the equatorial ocean siphoning [J. X. Mitrovica and W. R. Peltier, J. Geophys. Res. 96(B12), 20053 (1991)] secondarily decreased the water depth by several meters, our value corresponds well with the generally estimated depth of about −120 m during LGM and therefore does not indicate a significant effect of hydroisostasy [

Fleming K., et al., Earth Planet. Sci. Lett. 163, 327 (1998);

]. Another aspect that is probably important is the effect of gravity field and centrifugal force on the global distribution of newly generated water masses in a scale of meters, as mentioned by W. R. Peltier [Science265, 195 (1994)].

If we assume that there is a minor chance of sampling the first coral individuals that started to colonize the substratum and that coral settling started 200 to 300 years after inundation (4), we find an almost perfect match with the record of terrestrial organic material from the Sunda Shelf, which provides direct age control within the time interval of missing coral ages. If we further consider a significant rise in atmospheric CO₂ caused by adding large amounts of CO₂, probably from the ocean (i.e., ¹⁴C depleted), we expect a drop in atmospheric ¹⁴C. This we see as a plateau. If the corals are in direct exchange with the atmosphere, then they should not have the 400-year reservoir age subtracted and end up with an age similar to our organics. If they do have a 400-year reservoir age, then the drop in atmospheric ¹⁴C will reach the corals with (some) delay, thus reducing their apparent reservoir age during the ¹⁴C plateau. This works together with the coral startup time and sampling to explain the observed discrepancies.

Stuiver M., Grootes P. M., Braziunas T. F., Q. Res. 44, 341 (1995);

Grootes P. M., Stuiver M., White J. W. C., Johnsen S., Jouzel J., Nature 366, 552 (1993).

Project 03G01115A was supported by the German Ministry of Education and Research. T.H. acknowledges a graduate school stipend funded by the Deutsche Forschungsgemeinschaft. We thank the team from the Leibniz Laboratory in Kiel for support and assistance. We gratefully acknowledge the positive comments by C. D. Woodroffe and two anonymous reviewers.