REPORTS
Pleistocene Extinction of Genyornis newtoni: Human Impact on Australian Megafauna Gifford H. Miller,* John W. Magee, Beverly J. Johnson,† Marilyn L. Fogel, Nigel A. Spooner, Malcolm T. McCulloch, Linda K. Ayliffe‡ More than 85 percent of Australian terrestrial genera with a body mass exceeding 44 kilograms became extinct in the Late Pleistocene. Although most were marsupials, the list includes the large, flightless mihirung Genyornis newtoni. More than 700 dates on Genyornis eggshells from three different climate regions document the continuous presence of Genyornis from more than 100,000 years ago until their sudden disappearance 50,000 years ago, about the same time that humans arrived in Australia. Simultaneous extinction of Genyornis at all sites during an interval of modest climate change implies that human impact, not climate, was responsible. Australia suffered a major loss of its largeand medium-sized land mammals in the Late Pleistocene. All marsupials exceeding 100 kg (19 species) and 22 of 38 species between 10 and 100 kg became extinct (1), along with three large reptiles and the ostrich-sized Genyornis newtoni (2); two other large flightless birds, the emu (Dromaius novaehollandiae) and the cassowary (Casuarius casuarius), survived. Dwarfing and range restriction occurred in many other species (1). Collectively, the lost species are often referred to as the Australian megafauna, although most are of modest body mass. For more than a century the cause of this exceptional extinction has been debated without a clear consensus, largely because of the difficulty in dating faunal remains close to the limit of radiocarbon dating (3). The cause of megafauna extinction initially focused on climate change (4) or human predation (5), G. H. Miller and B. J. Johnson, Center for Geochronical Research, Institute of Arctic and Alpine Research (INSTAAR), and Department of Geological Sciences, University of Colorado, Boulder, CO 80309 – 0450, USA. J. W. Magee, Department of Geology, Australian National University, Canberra ACT 0200, Australia. M. L. Fogel, Geophysical Laboratory, Carnegie Institution of Washington, 5251 Broad Branch Road, NW, Washington, DC 20015, USA. N. A. Spooner, M. T. McCulloch, L. K. Ayliffe, Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia. *To whom correspondence should be addressed. Email: gmiller@colorado.edu †Present address: School of Oceanography, Box 357940, University of Washington, Seattle, WA 98195–7940, USA. ‡Present address: Laboratoire des Sciences du Climatet et de l’Environnement (LSCE), Domaine du CRNS– Bt 12, Ave. de la Terrasse, 91198 Gif-sur-Yvette, Cedex, France.
but more recently has included indirect consequences of human activity, particularly ecosystem change resulting from burning practices (6–8). Resolving the debate requires secure dates on the extinction events, on the arrival of humans in Australia, and on major climate and environmental changes. Here we present new dates that constrain the ages of these key events based on amino acid racemization, accelerator mass spectrometry (AMS), 14C and thermal ionization mass spectometry (TIMS) U-series analyses on eggshells, and luminescence dates on associated sediment. The abundance of fossil eggshell reflects, in order of importance, its preservation potential, subsequent exposure potential, and the number of nesting birds. To be preserved, mobile sediment, commonly eolian sand, must be available to bury the eggshell quickly. Eggshell is most frequently collected from natural exposures or deflation hollows. Without recent erosion, eggshell is rarely encountered because it survives only briefly at the surface. Bird populations are controlled primarily by climate, nutrient availability, and predator pressure (including humans). Eggshells of large, flightless birds offer distinct advantages over bone for dating purposes. They are well preserved and surprisingly common across Australia, their dense calcite matrix is resistant to diagenesis, and, from a single fragment, 14C activity, amino acid racemization (AAR), and stable isotope ratios can be measured. By dating eggshells of Genyornis and Dromaius collected from deposits formed during wetter and drier intervals, environmental change may be placed in a secure time frame, and the youngest Genyornis eggshells can hence provide a date
on the bird’s extinction within each region. By dating the extinction event in several regions spanning a range of climates, continental extinction may be evaluated. Most birds incorporate about 3% complex organic compounds in their eggshells as an intracrystalline organic web that is highly resistant to diffusional loss. Epimerization of the common protein amino acid isoleucine, preserved in the proteinaceous residues, follows first-order, reversible kinetics (9, 10). Expressed as the proportion of the non-protein D-alloisoleucine to its protein source, L-isoleucine, D/L increases from essentially zero in modern eggshell to an equilibrium ratio of 1.30 (11). The temperature dependence of isoleucine epimerization in Dromaius eggshell is well known (12). Eggshells of both Genyornis and Dromaius commonly occur together in all regions (13), suggesting that they coexisted and nested in close proximity, contrary to a recent assertion (14). We measured D/L ratios in eggshells of both taxa from three regions (15): (i) an extensive series (1200 analyses) from dune and beach deposits within a 50-km radius of Madigan Gulf (MG) in Lake Eyre, the terminal playa of the Lake Eyre Basin; and modest collections (ii) from a 50-km radius centered on the north end of Lake Frome (LF) and (iii) from a 150-km radius centered on the confluence of the Darling and Murray rivers (DM) (Fig. 1). The MG region is currently arid [21°C, ;150 mm precipitation (16)]; LF is semiarid (19°C, ;400 mm precipitation) and has denser plant cover and thick riparian vegetation; and the DM region is semiarid (17°C, ;400 mm precipitation) with more reliable stream flow. The Darling River drains the summer-dominant precipitation regime of the north, whereas the Murray drains the winter-westerly– dominated precipitation regime of southern Australia. Consequently, the DM sites have one of the most dependable moisture supplies in the interior, and should have been a refugium for fauna during periods of climate stress. In all three regions, precipitation can occur in any season, but summer precipitation dominates in the MG region. Before European settlement, it supported mixed C3 and C4 vegetation. LF and DM support primarily C3 plants. In the MG region, the D/L record is calibrated by 90 paired AAR and calibrated AMS 14 C dates (17) on Dromaius eggshells, 8 luminescence dates between 40 and 120 thousand years ago (ka) on eolian sand from which eggshell has been collected, and 5 TIMS U-series dates on Genyornis eggshell. An age model developed from this calibration set allows calendar ages to be assigned to each D/L ratio for all samples ,130,000 years old (Fig. 2). Frequency histograms for the two taxa derived from the age model document the
www.sciencemag.org SCIENCE VOL 283 8 JANUARY 1999
205
REPORTS time dependence of the collection series over the past 130 thousand years (ky) (Fig. 3); the frequency distribution of the 90 14C-dated Dromaius eggshells (Fig. 3) mirrors that of the larger data set of model-derived ages. MG was continually occupied by breeding populations of Dromaius for at least the past 130 ky, even through the stage 2 arid phase. Intervals of greatest Dromaius eggshell abundance reflect peak sediment mobility (preservation) during early stage 5, early stage 3, and at about the stage 2/1 transition. Although the distribution of Genyornis eggshells is similar for the older part of the record, eggshells abruptly disappear about 50,000 years ago. The continuous presence of Dromaius eggshells confirms that the absence of younger Genyornis eggshell is not due to a lack of preservation or exposure potential, but must indicate the actual extinction of the birds in this area. Insufficient age calibration is available for the other two regions to plot frequency distributions in the time domain, but we can evaluate them in the D/L domain (Fig. 4). D/L is a time-dependent quantity, but the epimerization rate exhibits a strong temperature dependency, and Pleistocene temperatures in central Australia were up to 9째C lower than in the Holocene (12); consequently, time is not linearly expressed. Furthermore, because the LF and DM regions are cooler than the MG region, the age for a specific D/L ratio is higher. Within these constraints, D/L histograms for all three regions exhibit similar trends. All three reveal near continuous occupation by Dromaius. The lack of old material at DM simply reflects the poor exposure there of older sediment. In sharp contrast to the continuity of Dromaius histograms, Genyornis displays a distinct termination in all three areas. Because of the temperature differences, the age of the extinction event must be evaluated at each site. To do this, we radiocarbon dated fragments of both Genyornis and Dromaius that straddle the extinction event,
Fig. 1. Locality map showing the Lake Eyre Basin (hatched area), Madigan Gulf sites (MG), Lake Frome sites (LF), and the Darling-Murray sites (DM) at the confluence of the Darling (D) and Murray (M) rivers.
206
dated Genyornis eggshells directly by TIMS U-series (18), and dated sediment enclosing the youngest Genyornis eggshell by luminescence (19). At MG, Genyornis became extinct between D/L ratios of 0.46 and 0.50. AMS 14C dates on 11 Dromaius eggshells within that range are all .38,700 14C years before the present (yr B.P.), essentially at the limit of the method. Five luminescence dates from the corresponding eolian unit are between 63 6 5 ka and 44 6 5 ka (20). U-series dates on terminal Genyornis eggshells from two different sites in MG are 47 and 65 ka. Radiocarbon ages of Genyornis eggshell that received only an acid-leaching pretreatment suggest that many of these eggshells are contaminated by younger carbon. For example, three fragments from levels stratigraphically below luminescence dates of 50 to 60 ka produced finite apparent radiocarbon ages between 36,700 and 39,500 14C yr B.P. Using a more rigorous pretreatment (90% acid leach; drilling out all pores), we obtained radiocarbon ages older than 42,000 14C yr B.P. for two fragments of Genyornis that gave the lowest D/L in the MG collections and that previously gave finite radiocarbon ages younger than 40,000 14C yr B.P. with less rigorous pretreatment. We conclude that the extinction event occurred at or beyond the limits of 14C dating. In the DM region Genyornis became extinct at a D/L ratio between 0.28 and 0.32, substantially lower than at MG because average temperatures at DM are 4째C less (21). One Dromaius eggshell slightly younger than the extinction event (D/L 5 0.27) has a radiocarbon
Fig. 2. Age model used to convert D/L ratios to calendar ages based on 90 paired and calibrated (17) AMS 14C dates and D/L ratios, 5 paired TIMS U-series dates and D/L ratios, and 8 luminescence dates on sediment from which eggshell D/L ratios were measured. The two initial linear segments are based on earlier assessments (12), whereas the older two linear segments are simple regressions fit through the data. Other configurations are plausible, but yield similar conclusions. LGM deposits in MG are thin, and the shallow burial, coupled with the low glacial-age temperature, results in poor age resolution for the interval between 16 and 60 ka, but the similar patterns for 14C- and AAR-dated samples (Fig. 3) suggest that any biases in the histograms are minimal.
age of 37,600 14C yr B.P., and three eggshells from within the extinction window range between 38,100 and .45,100 14C yr B.P. Two Genyornis eggshells that slightly predate the extinction, based on their D/L ratios, yielded ages of 42,600 and .44,400 14C yr B.P., and five fragments from a single stratigraphic level, all with D/L ratios of 0.34 and therefore slightly older than the extinction, have an average TIMS U-series age of 53 6 3 ka (22). No independent dates are available from LF. The temperature averages 2째C lower there than at MG (21); consequently, the D/L ratio expected of a correlative extinction event is 0.40 6 0.03, midway between those of extinction events at MG and LF. The D/L ratios of the youngest Genyornis at LF are higher (0.46 6 0.02); hence, extinction may have occurred slightly earlier, although the small sample size may mean we failed to locate the youngest Genyornis there. Our dates bracket the timing of Genyornis extinction. Luminescence and U-series dates demonstrate that Genyornis was common as recently as 60 to 50 ka. The radiocarbon evidence from eggshells of both taxa spanning the extinction event collectively demonstrate that it occurred at or beyond the effective limit of 14C dating of carbonates (at least 42,000 14C yr B.P., corresponding to a calendar age of .46 ka). Thus, the extinction
Fig. 3. Age-frequency histograms for Dromaius (above) and Genyornis (below) for the past 130 ky from the MG region, derived from D/L ratios on individual eggshell fragments collected from widely scattered sites that have been converted to calendar ages from the age model shown in Fig. 2. Solid bars in the upper panel represent the AMS 14C-dated eggshells superimposed on the AAR-dated series. The sudden disappearance of Genyornis ;50 ka defines the extinction event in this region. Marine isotope stage boundaries are shown.
8 JANUARY 1999 VOL 283 SCIENCE www.sciencemag.org
REPORTS occurred 50 6 5 ka. The similar pattern and timing of Genyornis extinction across three different regions, including the DM region that would be expected to serve as a refugium during times of drought, and the continuous occurrence of Dromaius through the extinction event and on to the present at all three sites suggest that extinction was taxonomically selective and on a continental scale. To evaluate whether a change in climate may have led to the extinction, we examined paleoclimate reconstructions that extend through the extinction event in MG (past 150 ky) (20, 23) and DM (past 70 ky) (24), concentrating on changes in available moisture, the limiting variable for vegetation across most of Australia. During arid times, when precipitation is reduced over the Lake Eyre Basin, playa conditions dominate in MG, and groundwater salinity rises because of evaporative concentration. As the watertable is lowered, salts that crystallize at the groundwater capillary fringe within the sediment matrix disrupt the surface and promote deflation, removing sediment and salts from the basin. While disruption and deflation keep pace with watertable lowering, the playa floor is lowered, until the groundwater level reaches equilibrium with the evaporative conditions, and a halite crust develops. Thus, the maximum depth of deflation is a secure proxy for the intensity of an arid phase. Arid-phase deflation excavated the playa floor below its current level (15 m below sea level) about 140 ka, 60 to 50 ka, and between about 30 and 16 ka. Of these, the most extensive excavation occurred at 140 ka, when the playa was eroded to bedrock, as much as 19 m below sea level. Lake Eyre last carried a permanent waterbody ;60 ka; slow drying followed, but the extent of playa deflation, and by direct inference the intensity of aridity, was less
than at 140 ka, and probably less than the arid phase 30 to 16 ka. Eolian sands were deposited adjacent to the lake during the highstand 60 ka and as it subsequently dried. Genyornis eggshells occur throughout the eolian unit, but not in younger deposits, indicating that extinction occurred at the end of a modest drying event or during an immediately subsequent wetter interval. The most complete record of moisture change in the DM region comes from the Willandra Lakes (24). A long period of aridity terminated about 60 ka, with increased moisture supporting abundant vegetation and filling the basin with permanent water. Genyornis and Dromaius eggshells are present in deposits of this age. Beginning at ;40 ka, the climate began to become drier; Dromaius was present, but no Genyornis have been reported in this or any younger units. After an interval of variable conditions, final desiccation occurred shortly after 20 ka, and Lake Mungo has remained dry since. In this region, Genyornis apparently became extinct before the onset of aridity. It thus seems that Genyornis was able to survive through the range of natural environmental changes caused by Pleistocene climate oscillations. At the time of Genyornis extinction, 50 6 5 ka, modest aridity (MG) to slightly wetter-than-present conditions (DM) prevailed. Consequently, climate change as an explanation for Genyornis extinction is unlikely. Genyornis is considered to be an element of the Australian megafauna (1), and its bones are associated with some of the youngest stratigraphic levels containing megafaunal remains. Thus, we argue that extinction of the Australian megafauna in general is likely to be coincident with Genyornis extinction, and that this occurred ;50 ka. Dating the arrival of the first humans to Australia has proven difficult. The most reli-
Fig. 4. D/L-frequency histograms for Dromaius and Genyornis from all three regions. Dromaius D/L ratios have been converted to equivalent Genyornis D/L ratios by dividing by 1.18, their average offset in paired collections of the two taxa, so that within each region the two taxa can be directly compared. Some key radiocarbon dates are indicated at their associated D/L ratio, showing that extinction occurred at or beyond the limit of 14C dating of eggshell.
able dates are luminescence ages of quartz grains enclosing human artifacts. The oldest widely accepted dates are 55 6 5 ka (25, 26), near the time of the extinction; older dates from northwest Australia (27) have been questioned (28, 29), and indirect evidence that has been interpreted as humanly induced, such as changes in fire regime (30, 31), remains tenuous. Humans are known to have had an impact on large birds elsewhere. The extinction of all species of moa (Dinornis spp.) from New Zealand within a few centuries of human colonization is documented to have been a consequence of hunting and fires (32). Genyornis carried about twice the body mass of Dromaius and was presumably less agile. It is plausible that because it was less mobile, or possibly because of a more vulnerable breeding strategy, Genyornis was selectively predated by early hunters. However, evidence of direct predation on Genyornis by humans is limited to a single site (33), and kill sites for other elements of the megafauna are equally rare [for example, (1, 34–36)]. Alternatively, ecosystem disruption as a consequence of early human activity may have impacted the fauna differentially, possibly on the basis of dietary preferences. The gross feeding habits of Genyornis can be reconstructed from the isotopic composition of its eggshells, and its skeletal morphology. The 13C/12C ratio of carbon in a bird’s food source is preserved in the calcite matrix of, and organic compounds within, its eggshells. In modern eggshell of Dromaius, Struthio (ostrich), and Coturnix (quail), the d13C of the organic fraction is enriched by 2 6 1 per mil relative to the diet of the parent; the offset between the organic and inorganic fractions is 9 6 1, 15 6 1, and 14 6 1 per mil, respectively (37, 38). The organic-inorganic offsets are indistinguishable in Late Quaternary eggshell of Dromaius (10 6 2 per mil; n 5 45) and Struthio (14 6 1 per mil; n 5 49). The average isotopic offset between the two fractions for fossil Genyornis (14 6 2 per mil; n 5 74) is similar to that of Struthio and Coturnix. If we assume that the same 2 per mil enrichment of the eggshell organic fraction over dietary d13C measured in our modern controls applies to fossil samples, it is possible to reconstruct dietary d13C for the extinct Genyornis. Dietary d13C reconstructed from 65 Genyornis and 49 Dromaius eggshells that were laid between 60 and 45 ka in MG averages 222 and 218 per mil, respectively. Twenty of 49 Dromaius eggshells are between 213 and 215 per mil, but some individuals have values as light as 226 and as heavy as 29 per mil. Genyornis group more tightly (54 of 65 are between 219 and 225 per mil; none are heavier than 214 per mil), suggesting that Genyornis ate primarily C3 browse, whereas Dromaius had wider dietary tolerances including a signif-
www.sciencemag.org SCIENCE VOL 283 8 JANUARY 1999
207
REPORTS icant component of C4 grasses. In contrast, 25 MG Dromaius eggshells from the last glacial maximum (AMS 14C dates for each are between 28,000 and 14,000 yr B.P.) have a dietary d13C clustered between 219 and 222 per mil, indicating a shift to regional dominance by C3 plants and a corresponding adjustment in the birds’ diet. Genyornis from the DM region apparently ate exclusively C3 plants (d13C 5 226 6 1 per mil; n 5 9). The isotopic data indicate that Dromaius had broader dietary tolerances than did Genyornis and was able to adapt its feeding strategies to a wider variety of ecological conditions. This ability may have contributed to its survival through the extinction event. The cranial morphology of the dromornithids, of which Genyornis is a member, indicates that the beak was used primarily in a shearing motion, a characteristic of browsing animals (39). Consequently, we conclude that Genyornis was primarily a browser and likely to be dependent on extensive shrubland. Fire has exerted a major control on Australian vegetation throughout the Quaternary, but the adaptations to fire are to ignition during a restricted portion of the year and only after sufficient fuel loading had occurred. One possibility is that burning practices of the earliest human immigrants differed enough from that of the natural fire cycle to disrupt ecosystems across the semiarid zone. The ecosystem may have been particularly susceptible as a result of the continent’s geological quiescence; soils in this low-relief landscape were depleted of most nutrients, resulting in a lack of ecosystem resiliency. We postulate that human burning at times of the year and at frequencies to which the vegetation was not pre-adapted resulted in a dramatic decrease in tree and shrub vegetation across the continental interior, which in turn placed unprecedented stress on the dependent fauna (6, 7, 40). This stress, possibly coupled with modest drying that occurred simultaneously and perhaps some direct human predation, led to megafauna extinction. Most of the extinct megafauna were browsers (1). Vegetation over much of their former range is currently desert scrub and spinifex grassland, providing circumstantial evidence in support of major vegetation change. Those taxa with restricted dietary requirements, such as Genyornis, were at greatest risk, whereas more cosmopolitan, opportunistic feeders, such as Dromaius, were able to survive as a new equilibrium was established. Top-level predators were unable to sustain themselves in the face of reduced prey. References and Notes
1. T. F. Flannery, Archaeol. Oceania 25, 45 (1990). 2. P. Vickers-Rich, in The Antipodean Ark: Creatures from Prehistoric Australia, S. Hand and M. Archer, Eds. (Angus and Robertson, Sydney, 1987), pp. 48 –50.
208
3. Direct dating of megafaunal bone by radiocarbon remains hampered by the poor preservation of collagen, and most dates on faunal remains are close to the radiocarbon limit. Although collagen constitutes 25% by weight in modern bone, under alternately wet and dry conditions in warm regions like Australia it is rapidly denatured into soluble gelatin and leached from the bone (41). Baynes (42) reviewed 91 published radiocarbon dates on megafaunal remains or sediments containing them; ages as young as 6000 14C yr B.P. have been accepted by some authors. After applying rigorous criteria to assess the reliability of a radiocarbon date and its association with faunal remains (43), he rejected most dates, including all those younger than 28,000 14C yr B.P. The few remaining sites with apparent finite 14C ages between 28,000 and 40,000 yr B.P. were not directly on megafaunal remains, and either the association of the megafauna with the dated level remains debated or there are uncertainties in the source of the actual carbon that was dated, leaving the timing of extinction unresolved. 4. R. Tate, Trans. Proc. Philos. Soc. S. Aust. 2, 13 (1879). 5. R. Owen, Researches on the Fossil Remains of the Extinct Mammals of Australia (Erxleben, London, 1877). 6. D. Merrilees, J. R. Soc. West. Aust. 51, 1 (1968). 7. R. Jones, Aust. Nat. Hist. 16, 224 (1969). , Nature 246, 278 (1973). 8. 9. A. S. Brooks et al., Science 248, 60 (1990). 10. G. H. Miller, P. B. Beaumont, A. J. T. Jull, B. J. Johnson, Philos. Trans. R. Soc. London Ser. B 337, 149 (1992). 11. Epimerization rates exhibit systematic offsets between taxa as a result of differences in the amino acid sequences in the protein each bird uses. D/L ratios measured in Dromaius and Genyornis eggshells from the same stratigraphic horizon indicate that isoleucine epimerizes 1.18 times faster in Dromaius; Dromaius D/L ratios have been adjusted by this amount to allow plots comparing the two taxa (Figs. 3 and 4) on a common scale. 12. G. H. Miller, J. W. Magee, A. J. T. Jull, Nature 385, 241 (1997). 13. Dromaius and Genyornis apparently nested in similar habitats; eggshells of both taxa are commonly found at the same stratigraphic level in many eolian deposits, although Genyornis is usually the dominant taxon. Both are occasionally found as crushed whole or partial eggs, more often as concentrations of eggshell fragments. Eggshells of the two genera can be readily distinguished by characteristic macro- and microstructures (44). 14. J. H. Field and W. E. Boles, Alcheringa 22, 177 (1998). 15. The Genyornis analyzed from MG come from 185 different sites, from LF 28 sites, and from DM 17 sites. Dromaius from MG were collected from 198 sites, from LF 30 sites, and from DM 36 sites. 16. Climate data are mean annual temperature and mean annual precipitation calculated with programs described by McMahon et al. (21). 17. In Fig. 2 and in general discussions we have converted radiocarbon ages to calendar ages using the CALIB 3.0 Program (45) for samples ,10,000 14C yr B.P., and the equation CalibAge 5 25.85 3 1026(Conv14CAge)2 1 1.39(Conv14CAge) 2 1807 (E. Bard, unpublished data based on paired U-series and 14C dates). However, when citing specific dates, we leave these in radiocarbon years because most are at or beyond the limits of radiocarbon dating, for which there is no generally accepted terminology or calibration curve. All luminescence and U-series dates are given in calendar years; the U-series dates have 2s uncertainties of about 2%, whereas the luminescence dates have 1s uncertainties of 10 to 15%. 18. We evaluated the potential of TIMS U-series dating using 20 Genyornis eggshell fragments of known AAR. U-series dates are shown to be reproducible among different fragments from the same unit, despite a wide range in U concentrations, but some samples have enough common thorium (232Th) to require correction for unsupported 230Th. Eggshell has discrete pores that allow the developing embryo to exchange gases with the atmosphere, and in some
iiii
19. 20. 21. 22.
23. 24. 25. 26. 27. 28. 29. 30.
31. 32.
33. 34. 35. 36. 37. 38. 39. 40. 41. 42.
43.
44.
45. 46.
fossil samples these have been infilled with clay, a carrier of thorium. We thus restrict our discussion to samples with 230Th/232Th activity ratios .30. Supplementary material for 14C, TIMS U-series, and OSL data is available at www.sciencemag.org/ feature/data/985798.shl. J. W. Magee, thesis, Australian National University, Canberra (1997). J. P. McMahon, M. F. Hutchinson, H. A. Nix, K. D. Ord, Centre for Resource and Environmental Studies, Australian National University, Canberra (1995). Although the 230Th/232Th activity ratios of these samples are below our preferred threshold, the reproducibility between many samples from a single level with large differences in U concentrations suggests that the ages are reliable. J. W. Magee, J. M. Bowler, G. H. Miller, D. L. G. Williams, Palaeogeogr. Palaeoclimatol. Palaeoecol. 113, 3 (1995). J. M. Bowler, Archaeol. Oceania 33, 120 (1998). R. G. Roberts, R. Jones, M. A. Smith, Nature 345, 153 (1990). R. G. Roberts et al., Quat. Geochronol. 13, 575 (1994). R. L. K. Fullagar, D. M. Price, L. M. Head, Antiquity 70, 751 (1996). N. A. Spooner, ibid. 72, 173 (1998). R. G. Roberts et al., Nature 393, 358 (1998). G. Singh, A. P. Kershaw, R. Clark, in Fire and Australian Biota, A. M. Gill, R. A. Groves, I. R. Noble, Eds. (Australia Academy of Science, Canberra, 1981), pp. 23– 54. A. P. Kershaw, G. M McKenzie, A. McMinn, Proc. Ocean Drilling Prog. Sci. Results 133, 107 (1993). M. M. Trotter and B. McCulloch, in Quaternary Extinctions: A Prehistoric Revolution, P. S. Martin and R. G. Klein, Eds. (Univ. of Arizona Press, Tucson, 1984), pp. 708 –727. M. Smith, G. Miller, G. Van Tets, Rec. S. Aust. Mus. 27, 228 (1994). D. R. Horton, Archaeol. Phys. Anthropol. Oceania 15, 86 (1980). T. F. Flannery, The Future Eaters (Reed Books, Melbourne, 1994). J. L. Kohen, Aboriginal Environmental Impacts (Univ. of New South Wales Press, Sydney, 1995). B. J. Johnson, thesis, University of Colorado (1995). , M. L. Fogel, G. H. Miller, Geochim. Cosmochim. Acta 62, 2451 (1998). P. Murray and D. Megirian, Rec. S. Austr. Mus., in press. A. P. Kershaw, Nature 322, 47 (1986). T. W. J. Stafford, P. E. Hare, L. Currie, A. J. T. Jull, D. Donahue, J. Archaeol. Sci. 189, 35 (1991). A. Baynes, in Taphonomy Symposium 95, Canberra, 28 –29 April 1995 (Department of Anthropology and Archaeology, Australian National University, Canberra, 1995), pp. 12–14. D. J. Meltzer and J. I. Mead, in Environments and Extinctions: Man in Late Glacial North America, J. I. Mead and D. J. Meltzer, Eds. (University of Maine Center for the Study of Early Man., Orono, 1985), pp. 145–173. D. L. G. Williams and P. V. Rich, in Vertebrate Paleontology of Australasia, P. V. Rich, J. M. Monaghan, R. F. Baird, T. H. Rich, Eds. (Pioneer Design Studio, Lilydale, Victoria, Australia, 1992), pp. 871– 891. M. Stuiver and P. Reimer, Radiocarbon 35, 215 (1993). Supported by the U.S. National Science Foundation, Division of Atmospheric Sciences (G.H.M., B.J.J., and M.L.F.), the Australian National University (J.W.M., M.T.M., L.K.A., and G.H.M.), and the University of Colorado Faculty Fellowship Program (G.H.M.). We thank J. Bowler for initiating our study of the Lake Eyre Basin; A. Baynes, J. Chappell, A. Chivas, P. Clark, P. DeDekker, T. Flannery, P. Latz, P. Murray, M. Smith, and S. Webb for discussions; C. Hart for lab assistance; G. Atkin, O. Miller, and M. Miller for field assistance; and the station owners and aboriginal community leaders in our field areas for land access and logistical assistance.
iiii
9 October 1998; accepted 24 November 1998
8 JANUARY 1999 VOL 283 SCIENCE www.sciencemag.org
Pleistocene Extinction of Genyornis newtoni: Human Impact on Australian Megafauna Gifford H. Miller, John W. Magee, Beverly J. Johnson, Marilyn L. Fogel, Nigel A. Spooner, Malcolm T. McCulloch, and Linda K. Ayliffe
Supplementary Material Table 1.14C Data Table Madigan Gulf region (MG) Genus
D/L
AAL Lab-ID
14
C age
14
C Lab-ID
Dromaius 0.024
7608
Fm:1.160±0.007
AA-20890
Dromaius 0.018
7176
Fm:1.186±0.007
AA-12620
Dromaius 0.042
8236
1145±45
AA-22892
Dromaius 0.074
7117
1405±65
AA-12611
Dromaius 0.122
7130
1565±50
AA-12618
Dromaius 0.067
7828
1620±50
AA-19012
Dromaius 0.107
7130
1725±50
AA-13042
Dromaius 0.102
7122
1770±60
AA-12616
Dromaius 0.248
7118
2985±85
AA-12612
Dromaius 0.209
7479
3585±60
AA-15138
Dromaius 0.233
7190
3730±55
AA-13046
Dromaius 0.218
7129
4420±75
AA-15668
Dromaius 0.430
7512
5600±80
AA-15665
Dromaius 0.360
8119
6010±65
AA-25248
Dromaius 0.337
7128
6175±75
AA-12617
Dromaius 0.352
8111
7240±55
AA-25249
Dromaius 0.314
7120
7525±80
AA-12613
Dromaius 0.335
7511
7725±120
AA-15664
Dromaius 0.342
7355
8175±70
AA-15408
Dromaius 0.342
7355
8510±70
AA-15134
Dromaius 0.278
6914
9475±75
AA-15144
Dromaius 0.284
6914
9520±80
AA-10239
Dromaius 0.384
7452
9635 ± 65
AA-16397
Dromaius 0.407
6878
10,070±90
AA-10234
Dromaius 0.464
8000
10,370±85
AA-20897
Dromaius 0.541
8233
10,620±70
AA-22891
Dromaius 0.510
7931
11,300±85
AA-20894
Dromaius 0.478
7341
11,370±75
AA-14032
Dromaius
0.570 7341
11,390±105
AA-17269
Dromaius
0.500 8144
11,400±70
Dromaius
0.350 8230
11810±110
AA-25257
Dromaius
0.338 7605
11,850±120
AA-17274
Dromaius
0.429 6877
11,875±115
AA-10232
Dromaius
0.375 7214
11,905±90
AA-15141
Dromaius
0.403 8231
12,055±75
AA-25254
Dromaius
0.486 7186
12,220±155
AA-25258
Dromaius
0.468 7930
12,275±150
AA-20887
Dromaius
0.490 8313
12,350±125
AA-25253
Dromaius
0.504 7187
12,425±95
AA-13045
Dromaius
0.381 7951
12,440±90
AA-20893
Dromaius
0.461 7827
12,480±110
AA-19013
Dromaius
0.540 7186
12,555±80
AA-25259
Dromaius
0.246 7219
12,680±125
AA-15145
Dromaius
0.560 7120
13,880±190
AA-12614
Dromaius
0.753 7224
16,490±120
AA-13049
Dromaius
0.541 7344
17,310±125
AA-15135
Dromaius
0.529 7583
18,405±115
AA-22899
Dromaius
0.558 7121
18,450±165
AA-12615
Dromaius
0.502 8268
19,815±190
AA-25251
Dromaius
0.510 8268
19,845±140
AA-25250
Dromaius
0.590 5533
20,230±150
AA-16396
Dromaius
0.559 7803
20,660±265
AA-19005
Dromaius
0.554 8233
20,915±155
AA-22890
Dromaius
0.555 8280
22,520±180
AA-25255
Dromaius
0.597 7514
23,275±325
AA-15666
Dromaius
0.524 8237
23,315±180
AA-22895
Dromaius
0.556 8038
23,430±215
AA-20892
Dromaius
0.556 8157
23,585±190
AA-22905
Dromaius
0.494 8140
23,615±210
AA-22887
Dromaius
0.565 8140
23,720±245
AA-22888
Dromaius
0.541 7123
24,030±245
AA-20896
Dromaius
0.544 7123
24,055±245
AA-13041
Dromaius
0.514 7694
24,835±240
AA-17276
Dromaius
0.539 8000
25,055±280
AA-20881
Dromaius
0.494 8235
25,680±230
AA-22894
Dromaius
0.518 8232
25,705±350
AA-22889
Dromaius
0.510 8260
26,195±245
AA-22893
Dromaius
0.533 7806
26,750±475
AA-19006
Dromaius
0.739 7843
27,345±530
AA-19014
AA-22903
Dromaius
0.515
8280
27,500±290
AA-25256
Dromaius
0.517
8149
27,625±325
AA-23256
Dromaius
0.606
7830
29,440±640
AA-19011
Dromaius
0.545
8417
29,970±380
AA-25316
Dromaius
0.562
7683
30,605±435
AA-17275
Dromaius
0.546
7592
31,720±490
AA-17272
Dromaius
0.549
7137
32,355±570
AA-13044
Dromaius
0.541
8045
32,990±550
AA-20884
Dromaius
0.552
7592
33,035±705
AA-17273
Dromaius
0.584
7137
33,455±580
AA-20889
Dromaius
0.551
8302
34,035±590
AA-25252
Dromaius
0.540
8045
34,600±1900
AA-20885
Dromaius
0.543
8160
34,815±765
AA-22886
Dromaius
0.605
8048
38,760±2320
AA-20886
Dromaius
0.527
8160
39,585±1190
AA-22885
Dromaius
0.569
8285
41,700±1450
AA-25247
Dromaius
0.588
7524
42,600±1650
AA-20891
Dromaius
0.535
8145
42,970±1800
AA-22904
Dromaius
0.596
7513
43,400±1900
AA-01900
Dromaius
0.591
7808
43,500±3800
AA-19007
Dromaius
0.546
8043
>37,240
AA-20883
Dromaius
0.608
6920
>39,700
AA-10237
Dromaius
0.592
7694
41,650
AA-20882
Genyornis 0.460
8161
42,830±880
CAMS-43987
Genyornis 0.460
8161
42,930±720
CAMS-43986
Genyornis 0.460
8161
31,070±450
AA-25239
Genyornis 0.471
8161
>37,000
AA-25240
Genyornis 0.483
7517
34,275±715
AA-16398
Genyornis 0.489
7136
39,740±1400
AA-13043
Genyornis 0.493
7516
34,800±755
AA-15667
Genyornis 0.494
8284
39,285±1200
AA-25242
Genyornis 0.496
7519
35,130±830
AA-25245
Genyornis 0.504
7518
41,300±1500
AA-25244
Genyornis 0.549
7525
45,440±1200
CAMS-43988
Genyornis 0.549
7525
36,700±960
AA-17271
Genyornis 0.550
5534
40,740±2100
AA-04248
Genyornis 0.588
7615
37,700±930
AA-25243
Genyornis 0.588
7615
38,500±1200
AA-25243
Genyornis 0.590
6875
39,500±1100
AA-25246
Genyornis 0.594
8141
41,100±1400
AA-25241
Lake Frome region (LF) Genus
D/L
AAL Lab-ID
Dromaius
0.034
5531
455±50
AA-04247
Dromaius
0.115
5530
2025±60
AA-04246
Dromaius
0.260
7942
6490±70
AA-19692
Dromaius
0.427
7943
6575±80
AA-19688
Dromaius
0.364
7943
7000±75
AA-19694
Dromaius
0.369
6889
10,790±105
AA-10236
Dromaius
0.449
7811
14,545±100
AA-19008
Dromaius
0.450
7936
19,780±190
AA-19693
Dromaius
0.507
6894
23,005±200
AA-15407
14
C age
14
C Lab-ID
Darling-Murray Rivers confluence (DM) Genus
D/L
AAL Lab-ID
Dromaius
0.023
6807
Fm:1.2663
AA-19015
Dromaius
0.232
7509
11,160±115
AA-15671
Dromaius
0.288
7197
13,045±85
AA-13047
Dromaius
0.204
8326
13,680±100
AA-25260
Dromaius
0.272
6919
13,835±110
AA-10240
Dromaius
0.307
6806
17170±130
AA-25262
Dromaius
0.24
6755
17,995±205
AA-08722
Dromaius
0.254
6755
18,050±240
AA-08725
Dromaius
0.309
6750
19,255±200
AA-08721
Dromaius
0.321
6779
21,140±190
CAMS-1929
Dromaius
0.347
7508
21,360±235
AA-15670
Dromaius
0.319
5604
37,600±1000
AA-04251
Dromaius
0.346
5608
38,100±1100
AA-04252
Dromaius
0.350
8318
40,520±1280
AA-25261
Dromaius
0.376
6887
42,100±1700
AA-10235
Dromaius
0.358
6761
>45,100
AA-08723
Genyornis 0.348
5609
42,600±1900
AA-04253
Genyornis 0.396
7200
>44,400
AA-13048
14
C age
14
C Lab-ID
Table 2. TIMS U-series data Table. TIMS U-series data 230
D/L Uppm
Th/232Th activity
230
Th/238U activity
234
U
Age (ka)
Age (ka) before correction for unsupported Th
Sample
Locality
1A
Bullysandhill
0.91
0.29
120.0
0.72696
199.18 97.9±1.2
1E
Bullysandhill
0.92
5.48
1360.0
0.67553
190.25 88.6±1.1
2D
Williams Point
0.5
0.15
31.9
0.56906
236.00 65.6±1.6
3B
Kutjitara West Bluff
0.6
0.08
43.14
0.59395
235.00 69.6±0.8
7E
Eric Island
0.57
0.12
31.43
0.45507
257.48 48.0±0.4
A1*
Perry Sand Hills
0.34
0.23
18.35
0.50340
203.00
56±2
58.0±1.1
A2-a*
Perry Sand Hills
0.34
0.75
2.52
0.61340
287.00
52±2
68.5±2.6
A2-b*
Perry Sand Hills
0.34
0.90
2.23
0.61320
266.00
51±15
70.1±5.2
A2-c*
Perry Sand Hills
0.34
0.91
3.95
0.53000
264.00
50±8
57.9±4.3
A2-d*
Perry Sand Hills
0.34
0.74
3.15
0.57990
187.00
58±9
71.4±2.1
All samples are eggshell of Genyornis newtoni; samples 1, 2, 3, and 7 are from the Madigan Gulf Region, whereas samples A1 and A2 are from the Perry Sand Hills at the confluence of the Darling and Murray rivers. Errors given are ±2 sigma. *Samples A2 (a-d) are from different fragments of the same egg; Sample A1 is from a different egg at the same stratigraphic level.
Table 3.OSL Data Table. OSL Data Palaeodose (Gy)
Dose rate (Gy/ka)
Age (ka)
D/L
WM Pt 11
84.9 ± 4.2
1.348 ± 0.077
63.0 ± 4.7
0.57
WmPt 13
64.8 ± 3.4
1.319 ± 0.075
49.1 ± 3.8
0.49
LDom 8
94.6 ± 5.4
0.832 ± 0.037 113.7 ± 8.3 1.00
Dul 1
31.1 ± 1.4
0.658 ± 0.043
47.3 ± 3.8
0.50
TBM 92/1 1
63.2 ± 4.2
0.750 ± 0.045
84.3 ± 7.6
0.81
TBM 92/1 2
58.2 ± 5.1
0.755 ± 0.039
77.1 ± 7.9
0.81
M95 TL2
121.0 ± 6.3
1.043 ± 0.038 116.0 ± 7.4 0.98
M95 TL5
30.4 ± 1.7
0.691 ± 0.042
Sample
44.0 ± 5.1
0.51
OSL DataTable. Optically-stimulated luminescence (OSL) dates on quartz sand from Magee (1997) that were used to calibrate the D/L-age relationship for the Madigan Gulf region of the Lake Eyre Basin, Australia. D/L is the proportion of D-alloisoleucine to L-isoleucine in eggshell from the same or correlative deposits of Genyornis or Dromaius (converted to equivalent Genyornis ratios). Uncertainty is one sigma.