ArticlePDF Available

New Ages for the Last Australian Megafauna: Continent-Wide Extinction About 46,000 Years Ago



All Australian land mammals, reptiles, and birds weighing more than 100 kilograms, and six of the seven genera with a body mass of 45 to 100 kilograms, perished in the late Quaternary. The timing and causes of these extinctions remain uncertain. We report burial ages for megafauna from 28 sites and infer extinction across the continent around 46,400 years ago (95% confidence interval, 51,200 to 39,800 years ago). Our results rule out extreme aridity at the Last Glacial Maximum as the cause of extinction, but not other climatic impacts; a “blitzkrieg” model of human-induced extinction; or an extended period of anthropogenic ecosystem disruption.
(42) suggest that preindustrial levels were 1.9 ppt.
Assuming a steady state and a 4.9-year lifetime, these
levels imply natural emissions of 10 Gg year
41. J. H. Butler et al., Nature 399, 749 (1999).
42. P. J. Fraser, unpublished data.
43. We estimate emissions of 1 Gg year
in Europe in
1999, 0.06 Gg year
in Australia in 1998–99, and 1
Gg year
in the western United States in 1999 (5).
44. S. Karlsdottir, I. S. A. Isaksen, Geophys. Res. Lett. 27,
93 (2000).
45. M. Mayer, C. Wang, M. Webster, R. Prinn, J. Geophys.
Res. 105, 22869 (2000).
46. A. M. Thompson, Science 256, 1157 (1992).
47. Y. Wang, D. Jacob, J. Geophys. Res. 103, 31123
P. J. Crutzen, P. H. Zimmerman, Tellus 43AB, 136 (1991).
The ALE, GAGE, and AGAGE projects involved substantial
efforts by many people beyond the authors of this paper
(5). In its latest phase (AGAGE), support came (and
comes) primarily from NASA, with important contribu-
tions from the Department of Environment, Transport
and the Regions (United Kingdom); Commonwealth Sci-
entific and Industrial Research Organization (Australia);
Bureau of Meteorology (Australia); and NOAA, among
others (5).
28 December 2000; accepted 17 April 2001
Published online 3 May 2001;
Include this information when citing this paper.
New Ages for the Last
Australian Megafauna:
Continent-Wide Extinction
About 46,000 Years Ago
Richard G. Roberts,
* Timothy F. Flannery,
Linda K. Ayliffe,
Hiroyuki Yoshida,
Jon M. Olley,
Gavin J. Prideaux,
Geoff M. Laslett,
Alexander Baynes,
M. A. Smith,
Rhys Jones,
Barton L. Smith
All Australian land mammals, reptiles, and birds weighing more than 100
kilograms, and six of the seven genera with a body mass of 45 to 100 kilograms,
perished in the late Quaternary. The timing and causes of these extinctions
remain uncertain. We report burial ages for megafauna from 28 sites and infer
extinction across the continent around 46,400 years ago (95% confidence
interval, 51,200 to 39,800 years ago). Our results rule out extreme aridity at
the Last Glacial Maximum as the cause of extinction, but not other climatic
impacts; a “blitzkrieg” model of human-induced extinction; or an extended
period of anthropogenic ecosystem disruption.
Twenty-three of the 24 genera of Australian
land animals weighing more than 45 kg
(which, along with a few smaller species,
constituted the “megafauna”) were extinct by
the late Quaternary (1–3). The timing and
causes of this environmental catastrophe have
been debated for more than a century (4, 5),
with megafaunal extirpation being attributed
to the impact of the first human colonizers (1,
5– 8), who arrived 56 4 thousand years ago
(ka) (9 –13), or climate change (4 ) [in partic-
ular, increased aridity at the Last Glacial
Maximum (19 to 23 ka) (14 )]. A resolution to
this debate has been thwarted by the lack of
reliable ages for megafaunal remains and for
the deposits containing these fossils. The dis-
appearance of one species of giant bird (Ge-
nyornis newtoni) from the arid and semi-arid
regions of southeastern Australia has been
dated to 50 5 ka, on the basis of 700
samples of eggshell (8), but no secure ages
for extinction have been reported for the giant
marsupials or reptiles, which constitute 22 of
the 23 extinct genera of megafauna weighing
45 kg. Here we present burial ages, ob-
tained using optical and
U dating
methods, for the remains of several megafau-
nal taxa (mostly giant marsupials; see Table
1) discovered at sites located in the humid
coastal fringe and drier continental interior of
Australia and in the montane forest of West
Papua (Fig. 1), which was joined to Australia
by a land bridge at times of lowered global
sea level.
Most major biogeographic and climatic
regions, and all five main groups of fossil
sites (14 ), are represented in our survey.
Most of the sites in southwestern Australia
are caves that have acted as pitfall traps,
School of Earth Sciences, University of Melbourne,
Melbourne, Victoria 3010, Australia.
South Austra-
lian Museum, Adelaide, South Australia 5000, Austra-
Laboratoire des Sciences du Climat et de
l’Environnement, 91198 Gif-sur-Yvette, France.
Commonwealth Scientific and Industrial Research
Organization (CSIRO) Land and Water, Canberra, ACT
2601, Australia.
Department of Earth Sciences, Uni-
versity of California, Riverside, CA 92521, USA.
CSIRO Mathematical and Information Sciences, Mel-
bourne, Victoria 3168, Australia.
Western Australian
Museum, Perth, Western Australia 6000, Australia.
National Museum of Australia, Canberra, ACT 2601,
Department of Archaeology and Natural
History, Research School of Pacific and Asian Studies,
Australian National University, Canberra, ACT 0200,
Department of Earth Sciences, La Trobe
University, Melbourne, Victoria 3086, Australia.
*To whom correspondence should be addressed. E-
Present address: Department of Geology and Geo-
physics, University of Utah, Salt Lake City, UT 84112,
Fig. 1. Map of the Australian region
showing the megafauna sites dated in
this study. Site numbers: 1, Ned’s Gully;
2, Mooki River; 3, Cox’s Creek (Bando); 4,
Cox’s Creek (Kenloi); 5, Tambar Springs; 6,
Cuddie Springs; 7, Lake Menindee (Sunset
Strip); 8, Willow Point; 9, Lake Victoria
(site 50); 10, Lake Victoria (site 51); 11,
Lake Victoria (site 73); 12, Montford’s
Beach; 13, Lake Weering; 14, Lake Cor-
angamite; 15, Lake Weeranganuk; 16,
Lake Colongulac; 17, Warrnambool; 18,
Victoria Fossil Cave (Grant Hall); 19, Vic-
toria Fossil Cave (Fossil Chamber); 20,
Wood Point; 21, Lake Callabonna; 22,
Devil’s Lair; 23, Kudjal Yolgah Cave; 24,
Mammoth Cave; 25, Moondyne Cave; 26,
Tight Entrance Cave; 27, Du Boulay Creek;
28, Kelangurr Cave. The bold dashed line
crossing the continent indicates the ap-
proximate present-day boundary be-
tween the zones dominated by summer
rainfall from monsoonal activity (north of
the line) and winter rainfall from westerly
storm tracks (south of the line). The stippled area indicates the zone that receives less than 500 mm
rainfall per year and where potential evapotranspiration exceeds mean monthly evapotranspiration
year-round with negligible runoff. Climatic data are from (24, 38) and references therein.
8 JUNE 2001 VOL 292 SCIENCE www.sciencemag.org1888
Table 1. Megafaunal taxa represented at the study sites. The names of the numbered sites are
given in Table 2 and Fig. 1. Taxa represented by articulated remains are indicated by X and Cf.,
whereas x and cf. denote taxa represented by disarticulated remains or remains for which
articulation is uncertain. Parentheses indicate that Genyornis newtoni is represented by a footprint
at Warrnambool (site 17) and by eggshell at Wood Point (site 20). The extant Macropus
giganteus, M. fuliginosus, and Sarcophilus harrisii are included as they are represented by
individuals up to 30% larger in dental dimensions than the living forms. The gigantic form of M.
giganteus is referred to here as M. g. titan, and that of S. harrisii as S. h. laniarius. Vombatus
hacketti and Wallabia kitcheneri belong to genera extant in eastern Australia but extinct in
Western Australia.
123456*6 7 8 9, 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26J 26H 26D 27 28
Articulated remains represented X . XX .. .XX X X X X . XXX . X . X . XXX .. .X .
Meiolania sp. indet. . . . . . . x . . . . . ............. . . . ..
Megalania prisca ......x.. . ............... . . . ..
Wonambi naracoortensis ......... . .............x. . . . ..
Pallimnarchus sp. indet. . . . . x x . . . . . ............. . . . ..
Quinkana sp. indet. . . . . . . x . . . . . ............. . . . ..
Genyornis newtoni .....xx.. . ......(X) . . (x) X ..... . . ..
Progura naracoortensis ......... . ........x...... . . . ..
Megalibgwilia ramsayi ......... . ........x....xx. . . ..
Zaglossus hacketti ......... . .............x. . . . ..
Sarcophilus harrisii laniarius x........ . x..x.x..x...... . . x ..
Diprotodon optatum X x XX .xx.. x .....x....X .... . . .X .
Diprotodon sp. indet. . . . . x . . . . . . . . x . . . . x . . .... . . . ..
Maokopia ronaldi ......... . ............... . . . .x
Zygomaturus trilobus ......... . .......xX ....XX ..x..
Zygomaturus sp. indet. . . . . . . . . . . . . . . . x . . . . . .... . . . ..
Palorchestes azael .x....x.. . ...... . .x...... . . . ..
Phascolonus gigas xx... . x .. XX..xX .....X ..... . . ..
Vombatus hacketti ......... . . . x ..
Thylacoleo carnifex xx....... x . . .x.x.xX . .x.x. . . . ..
Propleopus oscillans ......... . ........x...... . . . ..
Borungaboodie hatcheri ......... . ............... . . x ..
Macropus ferragus ...... ... X ............... . . . ..
Macropus fuliginosus ......... . ...........xxXX x. ...
Macropus giganteus titan X x...xx.. . x . X x XX ..x.X .... . . . ..
Macropus sp. indet. . . . . x . . . . . . . .....xx...... . . . ..
Procoptodon goliah x..... .X . X x.......X ...... . . . ..
Procoptodon sp. indet. . . . . x . . . . . . . . . . x . . . . . .... . . . ..
Protemnodon anak ...... ... x . . . .X x......... . . . ..
Protemnodon brehus ...... ... X . Cf. ..x. . . x ..
Protemnodon hopei ......... . ............... . . . .x
Protemnodon roechus X ........ . ........x...... . . . ..
Protemnodon sp. indet. . x . . . . x . X . ...........x... . . . ..
Simosthenurus baileyi ......... . ........x...... . . . ..
Simosthenurus brownei ......... . ........X ...XXX ..x..
Simosthenurus gilli ......... . .......xX ...... . . . ..
Simosthenurus maddocki ......... . ........x...... . . . ..
Simosthenurus newtonae ......... . .......xx...... . . x ..
Simosthenurus occidentalis ......... . ........X ...xx.x x x..
Simosthenurus pales ......... . ........x...... . . x ..
Simosthenurus sp. indet. . . . . . . . . . . . . . . . x . . . . . x . . . . . . . .
Sthenurus andersoni x ........x.X .... . . . ..
Sthenurus atlas ......... x ....... ........ . . . ..
Sthenurus stirlingi ......... . ..........X .... . . . ..
Sthenurus tindalei ......... x ....... ...X .... . . . ..
Sthenurus sp. indet. . . . . . . x . x . . . . . x . . . . . . .... . . . ..
Wallabia kitcheneri ......... . .............x. . . x ..
*Site 6, units 5, 6a, and 6b. Site 6, units 7 to 12.
whereas the sites in eastern Australia consist
mainly of aeolian deposits along the edges of
former or present lake basins, river or swamp
deposits, and coastal dune deposits. To max-
imize our prospects of encountering fossils
close in age to the terminal extinction event,
we chose sites that geomorphological and
stratigraphic evidence indicated were rela-
tively young. The most recent megafaunal
site may not be included in our survey, but we
consider that a sufficient number of sites (n
28) have been dated to discern a clear pattern
in the distribution of burial ages.
A review (15) of 91 radiocarbon (
ages obtained for Australian megafauna be-
fore 1995 rejected the vast majority of ages as
being unreliable (16), including all those
younger than 28 ka before the present (B.P.).
The remaining
C ages were close to or
beyond the practical limits of the technique,
or were on materials that had ambiguous
associations with the megafaunal remains.
Radiocarbon dating of bone and charcoal old-
er than 35 ka is problematic using conven-
tional sample pretreatments (17–19). Conse-
C ages were used in this study only
for comparison with ages of 50 ka obtained
Table 2. Optical ages for burial sediments, supporting data, and sample contexts.
Site* Sample context
Grain size
Dose rate
(Gy ka
Optical age
1. Ned’s Gully Megafaunal unit, sample 1 180–212 0.76 0.09 35 247 6
Megafaunal unit, sample 2 90–125 0.78 0.09 36 346 6
New South Wales
2. Mooki River Megafaunal unit 90–125 1.77 0.16 74 442 4
3. Cox’s Creek (Bando) Megafaunal unit, sample 1 90–125 1.43 0.14# 75 353 5
Megafaunal unit, sample 1 180–212 1.38 0.14# 75 354 6
Megafaunal unit, sample 2 90–125 1.43 0.14# 72 650 6
4. Cox’s Creek (Kenloi) 5 cm above megafaunal unit 90–125 0.93 0.06 47 251 4
30 cm below megafaunal unit 90–125 0.97 0.06 51 253 4
30 cm below megafaunal unit 180–212 0.94 0.06 54 358 4
5. Tambar Springs Megafaunal unit (spit 4) 90–125 1.43 0.09 2.9 0.2 2.0 0.2
6. Cuddie Springs Above main megafaunal unit (unit 4) 90 –125 2.47 0.15 41.3 1.2 16.7 1.2
Megafaunal unit (unit 5) 90–125 2.22 0.14 59 2 27 2
Megafaunal unit (unit 6a) 90–125 1.95 0.12 59 3 30 2
Megafaunal unit (unit 6b) 90–125 2.72 0.17 99 5 36 3
7. Lake Menindee (Sunset Strip) Megafaunal unit 180–212 1.69 0.09 113 867 6
8. Willow Point Attached to MV specimen 90–125 0.86 0.08 47 255 6
9. Lake Victoria (site 50) Attached to MV specimen 180–212 0.59 0.07§ 30 352 8
10. Lake Victoria (site 51) Attached to MV specimen 90–125 0.71 0.08 38 254 7
11. Lake Victoria (site 73) Attached to MV specimen 90–125 1.71 0.18 165 597 11
12. Montford’s Beach Attached to MV specimen 90–125 0.83 0.10 49.7 1.2 60 7
13. Lake Weering Attached to MV specimen 90–125 1.42 0.15# 117 382 9
14. Lake Corangamite Attached to MV specimen, sample 1 90–125 1.50 0.18 79 452 7
Attached to MV specimen, sample 1 180–212 1.47 0.18 78 453 7
Attached to MV specimen, sample 2 180–212 1.46 0.17 70 448 6
15. Lake Weeranganuk Attached to MV specimen 180 –212 6.3 0.7# 437 18 70 8
16. Lake Colongulac Attached to MV specimen 180 –212 1.59 0.16§ 131 10 82 10
17. Warrnambool From MV sediment slab with footprint 180–212 0.61 0.08 37 360 9
South Australia
18. Victoria Fossil Cave (Grant Hall) 20 cm below top of megafaunal unit 90–125 1.28 0.08 107 984 8
19. Victoria Fossil Cave (Fossil Chamber) 50 cm below top of megafaunal unit 90 –125 0.67 0.04 115 6 171 14
50 cm below top of megafaunal unit 180–212 0.65 0.04 102 8 157 16
20. Wood Point Unit containing eggshell 90–125 1.60 0.14# 88 355 5
21. Lake Callabonna Attached to MV specimen 90 –125 0.62 0.07 46 275 9
Western Australia
22. Devil’s Lair Above main megafaunal unit (layer 28) 90–125 1.22 0.05 51 242 2
Megafaunal unit (layer 32) 90–125 1.71 0.07# 79 347 2
Megafaunal unit (layer 39) 90–125 1.35 0.06 65 248 3
23. Kudjal Yolgah Cave Megafaunal unit (pit 2) 90–125 1.11 0.05 51 246 2
Attached to WAM specimen 125–250 1.22 0.14 56 246 6
24. Mammoth Cave Upper megafaunal unit 90–125 0.72 0.09 40 455 9
Attached to WAM specimen, sample 1 90–125 1.04 0.15 66 263 9
Attached to WAM specimen, sample 2 90–125 0.90 0.12 67 374 10
25. Moondyne Cave Megafaunal unit 90–125 0.68 0.05# 89 7 131 14
26. Tight Entrance Cave Megafaunal unit (unit J) 90–125 0.72 0.04 24 333 4
Megafaunal unit (unit H) 90–125 0.52 0.03 23 345 6
Megafaunal unit (unit D) 90–125 0.73 0.04 103 14 141 21
27. Du Boulay Creek Attached to WAM specimen 90–125 2.7 0.3# 215 22 80 12
West Papua
28. Kelangurr Cave Megafaunal unit 90–125 1.07 0.10§ 17.6 1.0 16 2
*Sites with taxa represented by articulated remains are marked by . MV and WAM indicate sediment removed from megafaunal collections at the Museum of Victoria and
Western Australian Museum, respectively. Mean 1uncertainty. Samples with a significant deficit of
U compared to
Ra are marked by #, and those with a significant
excess of
U over
Ra are marked by §. Paleodose values include 2% uncertainty associated with laboratory beta-source calibration. The Cuddie Springs (site 6) samples have
multiple paleodose populations, of which the highest are shown (29). The Montford’s Beach (site 12) and Du Boulay Creek (site 27) samples have paleodose distributions consistent
with partial bleaching (37), so the minimum paleodose values and age estimates [obtained using a minimum age model (35)] are shown. The high paleodose of the Lake Weeranganuk
(site 15) sample was obtained from aliquots with saturating exponential plus linear growth curves of luminescence intensity versus dose (36). This sample also has an unusually high
dose rate, which is due to concentrations of 20 ppm of all radionuclides in the
U decay series. If these concentrations were lower in the past, then the optical age would be older.
8 JUNE 2001 VOL 292 SCIENCE www.sciencemag.org1890
from optical dating of megafauna-bearing
sediments and
U dating of flow-
stones formed above and below megafaunal
remains. Optical dating is a luminescence-
based method that indicates the time elapsed
since the sediment grains were last exposed
to sunlight (20–22). The optical age corre-
sponds to the burial age of megafaunal re-
mains in primary deposition, whereas
U dating gives the crystallization age of
the flowstone, and thus a constraining age for
remains above or below the flowstone. Sup-
port for the optical ages reported here (Table
2) is provided by their consistency with the
C and
U ages obtained at
megafaunal sites where comparisons have
been made ( Table 3) (19, 23–25). All three
methods yield concordant ages within the
time range of
C dating, and beyond this
limit the optical and
U ages are in
good agreement and correct stratigraphic
Optical and
U dating were con-
ducted primarily on deposits containing the
remains of megafauna in articulated ana-
tomical position (Table 1) to avoid uncer-
tainties introduced by post-depositional
disturbance and reworking of fossils. This
conservative approach is vital because the
remains must be in primary depositional
context to estimate the time of death from
optical dating of the burial sediments or
U dating of the enclosing flow-
stones. We also dated some deposits with
disarticulated remains, but we recognize
that these ages will be too young if the
remains have been derived from older
units. A sandstone slab bearing the impres-
sion of a Genyornis footprint and dune
sands containing burnt fragments of Geny-
ornis eggshell were also dated. Sediment
samples for optical dating were collected
on site from stratigraphic units that were
clearly related to the megafaunal remains;
in addition, lumps of sediment attached to
megafaunal remains in museum collections
were removed for dating (Table 2). We
adopted a conservative approach to dating
of museum samples (20), owing to their
small size and the lack of an in situ dose
rate measurement. Confidence in the age
estimates for the museum specimens is giv-
en by the close agreement between the ages
of the museum and field-collected samples
from Kudjal Yolgah Cave (site 23; see
Table 2). Our main conclusions, however,
are based on field-collected samples, which
yield the most reliable and precise ages.
Calcite flowstones were prepared for
U dating using standard methods and
were analyzed by thermal ionization mass
spectrometry (19, 24, 25), and the ages
(Table 3) have been corrected for detrital
Th contamination (26 ).
The youngest optical ages obtained for
deposits with articulated megafaunal remains
(Table 2) (27 ) are 47 4 ka for Ned’s Gully
(site 1) in Queensland and 46 2kafor
Kudjal Yolgah Cave (site 23) in Western
Australia. This result implies broadly syn-
chronous extinction across the continent.
Claims have also been made (28) for articu-
lated remains of Simosthenurus occidentalis
of similar age from Tight Entrance Cave (site
26, unit H or below) in Western Australia,
and several sites (3, 4, 8, 9, and 10) in New
South Wales produced slightly older ages (50
to 55 ka) for articulated megafauna. In con-
trast, much younger apparent burial ages
were obtained for some sites containing dis-
articulated remains (Table 2) (27); the
youngest such age is 2.0 0.2 ka for frag-
mented remains at Tambar Springs (site 5).
Optical dating of individual grains from the
Cuddie Springs deposit [site 6 (23)] indicates
that some sediment mixing has occurred (29).
We interpret the young ages obtained for
disarticulated remains and the indication of
sediment mixing at Cuddie Springs as evi-
dence that the remains are not in their prima-
ry depositional setting, but have been eroded
from older units and redeposited in younger
units with contemporaneous sediment and
The youngest measured burial age for
articulated remains may be older than the
terminal extinction event, unless the most
recent burial site is fortuitously included in
our survey. But each optical age has a
relative standard error of 5 to 15%, so the
measured age could by chance be less than
the true extinction age at some sites. Ac-
cordingly, we built a statistical model of
the data under the assumption that the true
burial ages are a realization of a Poisson
process of constant intensity up to the time
of extinction. That is, we assumed that the
true burial ages are distributed randomly
through time, with equal numbers per unit
time, on average. The optical age is the true
burial age plus a Gaussian error with a
mean of zero and a standard deviation equal
to the reported standard error. We estimat-
ed the time of extinction by maximum like-
lihood, confining attention to articulated
remains with optical ages of 55 ka (30).
This avoids a potential difficulty caused by
the undersampling of sites much older than
the extinction event. Using this model, the
maximum likelihood estimate of the extinc-
tion time is 46.4 ka, with 68% and 95%
confidence intervals of 48.9 to 43.6 ka and
51.2 to 39.8 ka, respectively.
Our data show little evidence for faunal
attenuation. Twelve of the 20 genera of
megafauna recorded from Pleistocene depos-
its in temperate Australia (1, 2) survived to at
least 80 ka, including the most common and
widespread taxa, and six of these genera
(Diprotodon, Phascolonus, Thylacoleo, Pro-
coptodon, Protemnodon, and Simosthenurus)
are represented at the two sites dated to
around 46 ka. These data indicate that a
relatively diverse group of megafauna sur-
vived until close to the time of extinction.
Further sites are needed to test this proposi-
tion and to identify the cause(s) of megafau-
nal extinction.
Table 3.
U ages for Western Australian flowstones, supporting data,
and sample contexts. The subscripts (t) and (0) denote the present and initial
values of
U, respectively. All errors are 2. Ages for flowstones at
Devil’s Lair (site 22), Tight Entrance Cave (site 26), and Victoria Fossil Cave
(Grant Hall, site 18, and Fossil Chamber, site 19) are reported elsewhere (19,
24, 25, 28).
Site* Sample context
Detrital Th
activity ratio
activity ratio
age (ka)
23. Kudjal Yolgah Cave Above megafaunal unit,
sample 1
U 0.008 0.302 0.006 102 12 112 14 13.9 0.2 34.7 0.9
C 0.294 0.008 102 31 112 34 33.6 1.6
Above megafaunal unit,
sample 2
U 0.008 0.331 0.003 105 7 117 7 5.37 0.05 38.5 0.6
C 0.308 0.005 105 18 116 19 35.4 1.0
24. Mammoth Cave Above upper megafaunal unit U 0.025 0.375 0.003 49 357 3 5.95 0.05 47.9 0.6
C 0.353 0.006 49 16 56 18 44.4 1.3
Below upper megafaunal unit U 0.056 0.457 0.005 73 4865 4.49 0.05 60.0 1.0
C 0.429 0.009 72 22 84 25 55.2 2.2
25. Moondyne Cave Above megafaunal unit U 0.061 0.341 0.004 40 445 4 17.7 0.3 43.1 0.7
C 0.335 0.008 41 24 46 27 42.2 1.8
*All three sites have taxa represented by articulated remains. Data corrected (C) and uncorrected (U) for detrital
Th contamination (26). The detritally corrected ages are
considered more reliable.
The burial ages for the last known
megafaunal occurrence suggest that extinc-
tion occurred simultaneously in eastern and
western Australia, and thus probably conti-
nent-wide, between 51 and 40 ka (95% con-
fidence interval), at least 20 ka before the
height of the Last Glacial Maximum. We
estimate that the megafauna had vanished
within 10 5 ka of human arrival [56 4ka
(9 –13)] across a wide range of habitats and
climatic zones. Megafaunal extinction in
Australia occurred tens of millennia before
similar events in North and South America,
Madagascar, and New Zealand, each of
which was preceded by the arrival of humans
(31). A prediction of the “blitzkrieg” model
of human-induced extinction [as proposed
first for North America (32) and later for
New Zealand (33)] is that megafaunal extinc-
tion should occur soon after human coloniza-
tion, and that extinction is followed by wide-
spread ecosystem disruption (1). Alternative-
ly, human arrival may first have triggered
ecosystem disruption, as a result of which the
megafauna became extinct (8). The latter se-
quence of events allows for a substantial time
interval between human colonization and
megafaunal extinction, so that climatic fac-
tors may also be involved (34). There is
sufficient uncertainty in the ages for both
human colonization and megafaunal extinc-
tion that we cannot distinguish between these
possibilities, but our data are consistent with
a human role in extinction. Resolving this
debate would require more precise ages for
human colonization and megafaunal extinc-
tion, as well as an improved understanding of
human interactions with the Australian land-
scape and biota during the earliest period of
human occupation.
References and Notes
1. T. F. Flannery, Archaeol. Oceania 25, 45 (1990).
2. P. Murray, in Vertebrate Palaeontology of Australasia,
P. Vickers-Rich, J. M. Monaghan, R. F. Baird, T. H. Rich,
Eds. (Pioneer Design Studio, Melbourne, 1991), pp.
3. T. F. Flannery, R. G. Roberts, in Extinctions in Near
Time: Causes, Contexts, and Consequences,R.D.E.
MacPhee, Ed. (Kluwer Academic/Plenum, New York,
1999), pp. 239–255.
4. C. S. Wilkinson, Proc. Linn. Soc. New South Wales 9,
1207 (1884).
5. R. Owen, Researches on the Fossil Remains of the
Extinct Mammals of Australia (Erxleben, London,
6. D. Merrilees, J. R. Soc. West. Austr. 51, 1 (1968).
7. R. Jones, Archaeol. Phys. Anthropol. Oceania 3, 186
8. G. H. Miller et al., Science 283, 205 (1999).
9. R. G. Roberts, R. Jones, M. A. Smith, Nature 345, 153
10. R. G. Roberts et al., Quat. Sci. Rev. 13, 575 (1994).
㛬㛬㛬㛬 , Ancient TL 16, 19 (1998).
12. J. M. Bowler, D. M. Price, Archaeol. Oceania 33, 156
13. A. Thorne et al., J. Hum. Evol. 36, 591 (1999).
14. D. R. Horton, in Quaternary Extinctions: A Prehistoric
Revolution, P. S. Martin, R. G. Klein, Eds. (Univ. of
Arizona Press, Tucson, AZ, 1984), pp. 639680.
15. A. Baynes, Rec. West. Austr. Mus. (suppl. 57), 391
16. D. J. Meltzer, J. I. Mead, in Environments and Extinc-
tions: Man in Late Glacial North America, J. I. Mead,
D. J. Meltzer, Eds. (Center for the Study of Early Man,
Univ. of Maine, Orono, ME, 1985), pp. 145–173.
17. J. P. White, T. F. Flannery, Austr. Archaeol. 40,13
18. S. Van Huet, R. Gru¨n, C. V. Murray-Wallace, N. Red-
vers-Newton, J. P. White, Austr. Archaeol. 46,5
19. C. S. M. Turney et al., Quat. Res. 55, 3 (2001).
20. Optical ages were calculated from the burial dose
(paleodose), measured using the photon-stimulated
luminescence (PSL) signal, divided by the dose rate
due to ionizing radiation (21, 22). The portion of each
sample exposed to daylight was first removed under
dim red illumination and discarded. Quartz grains in
three ranges of diameter (90 to 125 m, 180 to 212
m, and 125 to 250 m) were extracted from the
remaining sample using standard procedures (22) and
etched in 40% hydrofluoric acid for 45 min. As a test
of internal consistency, some stratigraphic units were
dated using more than one sample or grain-size
fraction. Paleodoses were obtained using single-ali-
quot regenerative-dose protocols, statistical models,
and experimental apparatus as described (35–37).
Each aliquot was illuminated for 100 to 125 s at
125°C, and paleodoses were calculated from the first
3to5sofPSLarising from the burial, regenerative,
and test doses, using the final 20 s as background.
Each sample was given a preheat plateau test (22)
using aliquots composed of 100 grains, and a re-
peat regenerative dose was given to verify that the
protocol yielded the correct (known) dose (35, 36).
The paleodoses in Table 2 were obtained from ali-
quots typically composed of 10 grains to permit
detection of insufficient bleaching before burial from
examination of the paleodose distribution (37). For
samples with clearly asymmetrical (positively
skewed) distributions, mean paleodoses were calcu-
lated using the minimum age model (35); the central
age model (35) was used for other samples. For some
samples, paleodoses were also obtained using the
standard multiple-aliquot additive-dose method (22).
The dose rates due to
U and
Th (and their
daughter products) and due to
K were calculated
from a combination of high-resolution gamma and
alpha spectrometry (to check for disequilibria in the
U and
Th decay series), thick-source alpha
counting, x-ray fluorescence of powdered samples,
and field measurements of the gamma dose rate.
Cosmic-ray dose rates were estimated from pub-
lished data [J. R. Prescott, J. T. Hutton, Radiat. Meas.
23, 497 (1994)] and an effective internal alpha dose
rate of 0.03 Gy ka
was assumed for all samples.
Gamma and beta dose rates were corrected for the
estimated long-term water content of each sample
and for beta-dose attenuation [V. Mejdahl, Archae-
ometry 21, 61 (1979)]. We used the sediment far-
thest from the bone to date the museum specimens
to minimize any dose rate heterogeneity in the sed-
iments adjacent to the bone. The gamma dose rates
for the museum specimens were estimated from
the attached lumps of sediment and from sedi-
ment-bone mixtures, using an uncertainty of
20% to accommodate any spatial inhomogeneity
in the gamma radiation field. This uncertainty was
also applied to field samples collected without
measuring the in situ gamma dose rate; in situ
measurements had uncertainties of less than 5%.
Some samples had a significant deficit or excess of
U with respect to
Ra (see Table 2). The
optical ages for these samples were calculated
using the measured radionuclide concentrations,
but any error due to post-burial uranium migration
should be accommodated within the age uncer-
21. D. J. Huntley, D. I. Godfrey-Smith, M. L. W. Thewalt,
Nature 313, 105 (1985).
22. M. J. Aitken, An Introduction to Optical Dating: The
Dating of Quaternary Sediments by the Use of Pho-
ton-Stimulated Luminescence (Oxford Univ. Press,
Oxford, 1998).
23. J. Field, J. Dodson, Proc. Prehist. Soc. 65, 275 (1999).
24. L. K. Ayliffe et al., Geology 26, 147 (1998).
25. K. C. Moriarty, M. T. McCulloch, R. T. Wells, M. C.
McDowell, Palaeogeogr. Palaeoclimatol. Palaeoecol.
159, 113 (2000).
26. Ages corrected for detrital
Th contamination were
obtained by determining the thorium and uranium
isotope compositions for different splits of the same
flowstone (each split containing different proportions
of the detrital end member and the pure authigenic
calcite phase). Mixing line plots of
Th versus
Th, and of
Th versus
provide estimates of the detritally corrected
U and
U ratios as well as the isotope
ratios of the detrital end member phases:
Th 0.37 0.04,
Th 0.00 0.03
(Kudjal Yolgah Cave);
Th 0.36 0.02,
Th 0.00 0.02 (Mammoth Cave, upper
Th 0.28 0.03,
Th 0.01 0.03 (Mammoth Cave, lower flow-
stone); and
Th 0.29 0.04,
Th 0.00 0.01 (Moondyne Cave, using the
average detrital end member isotope ratios from five
nearby sites). The half-lives of
U and
Th used in
the age calculation are 244,600 490 years and
75,381 590 years, respectively.
27. See the supplementary figure at Science Online
28. G. J. Prideaux, G. A. Gully, L. K. Ayliffe, M. I. Bird, R. G.
Roberts, J. Vertebr. Paleontol. 20 (suppl. to no. 3),
62A (2000).
29. Single-grain optical dating and finite mixture models
[R. G. Roberts, R. F. Galbraith, H. Yoshida, G. M.
Laslett, J. M. Olley, Radiat. Meas. 32, 459 (2000)]
were used to distinguish paleodose (and hence age)
populations in the sediment samples. Multiple dis-
crete populations were identified, which we attribute
to the mixing of grains with different burial histories.
The populations with the highest paleodoses ( Table
2) yielded optical ages consistent with the
C ages
obtained from pieces of charcoal (23).
30. We cannot be certain that articulated remains were
recovered from the upper megafaunal unit at Mam-
moth Cave (site 24), so the optical age of 55 9ka
was not included in the data set.
31. P. S. Martin, D. W. Steadman, in Extinctions in Near
Time: Causes, Contexts, and Consequences, R. D. E.
MacPhee, Ed. (Kluwer Academic/Plenum, New York,
1999), pp. 17–55.
32. P. S. Martin, Science 179, 969 (1973).
33. R. N. Holdaway, C. Jacomb, Science 287, 2250
34. Much of the 60- to 40-ka interval was marked by
generally wetter conditions than at present in both
eastern Australia (24, 38) [G. C. Nanson, D. M. Price,
S. A. Short, Geology 20, 791 (1992); J. M. Bowler,
Archaeol. Oceania 33, 120 (1998)] and southwestern
Australia [J. Balme, D. Merrilees, J. K. Porter, J. R. Soc.
West. Austr. 61, 33 (1978), using the revised chro-
nology for Devil’s Lair (19)]. But monsoonal activity
may have been variable with short-lived climatic
oscillations (38), in keeping with evidence from deep-
sea cores of climate instability [J. P. Sachs, S. J.
Lehman, Science 286, 756 (1999); S. L. Kanfoush et
al., Science 288, 1815 (2000)].
35. R. F. Galbraith, R. G. Roberts, G. M. Laslett, H. Yoshida,
J. M. Olley, Archaeometry 41, 339 (1999).
36. H. Yoshida, R. G. Roberts, J. M. Olley, G. M. Laslett,
R. F. Galbraith, Radiat. Meas. 32, 439 (2000).
37. J. M. Olley, G. G. Caitcheon, R. G. Roberts, Radiat.
Meas. 30, 207 (1999).
38. B. J. Johnson et al., Science 284, 1150 (1999).
39. We thank S. Eberhard, J. Field, G. Gully, L. Hatcher, R.
McBeath, D. Merrilees, G. Miller, R. Molnar, K. Mori-
arty, A. Ritchie, I. Sobbe, the late G. van Tets, J.
Wilkinson, D. Witter, T. Worthy, and R. Wright for
sample collection, field assistance, and discussions;
the Western Australian Museum and Museum of
Victoria for permission to access their collections; M.
Olley for preparing the gamma spectrometry sam-
ples; R. Galbraith for mixture modeling; and R.
Gillespie and O. Lian for comments. Supported by a
Large Grant and a Queen Elizabeth II Fellowship from
the Australian Research Council (R.G.R.).
27 February 2001; accepted 25 April 2001
8 JUNE 2001 VOL 292 SCIENCE www.sciencemag.org1892
... The age of a single bone specimen (Zygomaturus trilobus) reported by Westaway et al. (2017) from the Willandra Lakes in western New South Wales implies a megafaunal presence in Australia at ~36-32 ka; however, the ages of several specimens reported by Roberts et al. (2001) across the continent suggest the disappearance of the large animals around 46 ka. The ~36-32 ka age is yet to be confirmed in other sites in Australia. ...
... Red crosses in CHAR curve are significant charcoal peaks. diversity decline of the animals around 46 ka (Roberts et al., 2001). ...
... The CFS trend indicates this marked decline was brief and decline in megafaunal presence in the basin was generally gradual from this time, as previously reported in some other parts of the world (e.g., interior Alaska) where megafauna extinction was also not an abrupt event (Conroy et al., 2020). While our recorded timing of ~43.3 ka is comparable to previous age estimates of megafaunal extinction ranging from ~50 to 40 ka on the Australian mainland (Roberts et al., 2001;Miller et al., 2005;Gillespie et al., 2006;van der Kaars et al., 2017;Hocknull et al., 2020;David et al., 2021), the robust chronological control and relatively higher sampling resolution of the core MD03-2607 (supplementary Fig. 4) provide a more precise age estimate of the timing of megafaunal decline in the Murray Darling Basin. ...
Full-text available
The timing and cause of megafaunal extinctions are an enduring focus of research interest and debate. Despite the developments in the analysis of coprophilous fungal spores (CFS), the proxy for reconstructing past megaherbivore changes, the environmental consequences of this fauna loss remain understudied. This is partly due to the general obscurity of such a signal in pollen records, as well as limitations in disentangling human and extinction ecological impact, and the lack of spatial information of megafauna changes in site-level sedimentary records. In Australia, the debate centres on the possibility that habitat loss through climate change, vegetation-fire change, human intervention, or a combination of these factors led to the extinction of some large animals during the Late Pleistocene. Pollen and plant isotope studies have also demonstrated that vegetation-fire responses following the Late Pleistocene megafaunal extinctions were characterized by increased vegetation density and fire activity due to reduced grazing/browsing pressure. Here, we use a well-dated marine sedimentary core record from the Murray Darling Basin in southern Australia and apply palynological and functional palaeoecological approaches to reconstruct the Late Pleistocene megafaunal abundance changes, the timing and potential cause of extinction across the basin and investigate if extinction was associated with any signal of trait-based vegetation changes. We infer megafaunal abundance changes from the abundance of CFS and compare this with climatic proxies from the same core. We then link modern observations of fruit, seed and fire response traits of plant genera within the basin to the fossil pollen record to reconstruct palaeo vegetation community traits and determine if extinction was associated with any changes in plant community trait composition. Closely-spaced 14C dates obtained from planktonic foraminifera and δ18O tie points place a major decline in CFS, and thus the timing of extinction, within the basin at ∼43.3 ka. While climate-driven environmental changes largely controlled megafaunal presence, human arrival and frequent landscape burning are considered the most likely primary cause of extinction or, at the very least, megafauna decline in the Murray Darling Basin. We also found that the proposed period of megafaunal decline was also accompanied and followed by a decline in the prevalence of plants with larger seeds and fruits that were likely to have been once dispersed by megaherbivores. Our study supports the idea of a human-driven megafaunal extinction in mainland Australia and that the extinction caused changes in vegetation due to reduced plant dispersal and herbivory. However, high fire activity primarily linked to these vegetation changes was not observed, as humans were already practicing landscape burning before the period of megafaunal extinction and likely continued to do so afterward.
... Shortfaced kangaroos [1], giant dromornithid birds [2] and browsing diprotodontid megaherbivores [3] contributed to a very different ecosystem to that of modern Australia. These recently extinct Quaternary species show that current mammal and bird diversity is less rich than it was in the Late Pleistocene [4]. However, the relatively few described Australian Quaternary squamate reptiles (lizards and snakes) largely mirror extant morphologies [5,6]. ...
Full-text available
There are more species of lizards and snakes (squamates) alive today than any other order of land vertebrates, yet their fossil record has been poorly documented compared with other groups. Here, we describe a gigantic Pleistocene skink from Australia based on extensive material that includes much of the skull and postcranial skeleton, and spans ontogenetic stages from neonate to adult. Tiliqua frangens substantially expands the known ecomorphological diversity of squamates. At approximately 2.4 kg, it was more than double the mass of any living skink, with an exceptionally broad, deep skull, squat limbs and heavy, ornamented body armour. It probably filled the armoured herbivore niche that land tortoises (testudinids), absent from Australia, occupy on other continents. Tiliqua frangens and other giant Plio-Pleistocene skinks suggest that small-bodied groups that dominate vertebrate biodiversity might have lost their largest and often most morphologically extreme representatives in the Late Pleistocene, expanding the scope of these extinctions.
... Defining megafauna is challenging. In the literature, authors usually choose arbitrary thresholds based on body mass, length, or taxonomic group (Atwood et al. 2020;Barnosky 2009;Boulanger and Lyman 2013;Malhi et al. 2016;Moleón et al. 2020;Ripple et al. 2016;Roberts et al. 2001;Villavicencio et al. 2016;Wroe et al. 2013). My study based the megafauna definition on a Gaussian approach of body mass using an extensive database and integrating trophic level. ...
Full-text available
Megafauna are amongst the most threatened species on our planet. Understanding threats to megafauna is important to plan effective protection. Previous studies identified threats to these species, but did not define the level of risk. In this paper, I have investigated the megafauna threat level and the associated conservation measures. First, I have used a semi-quantitative approach to define megafauna species based on body mass, trophic level and system (e.g. marine, land). My initial data includes more than 24,076 species, accounting for 96.2% of the IUCN Red List data for mammals, reptiles and birds (including sub-species and sub-populations). Second, I estimated the IUCN Threat Impact Score (TIS) on the species defined as megafauna (n = 404). Finally, based on the TIS and other parameters, I investigated the most threatened megafauna to pinpoint conservation priorities and strategies for their respective functional groups. Results show that megafauna conservation priorities are category-specific. For instance, each category of species (e.g. terrestrial predator) is sensitive to specific threats and requires various levels of protection depending on a variety of parameters (e.g. spatial scale, number of threats). I propose category-specific conservation priorities and a global strategy based on six pillars of action to improve megafauna protection: education and awareness; sanctions; regulated eco-tourism, coordination; data; and climate change.
... If the CWR taxa are so prone to decline today (e.g., , Fisher et al. 2014, then could they not have been predisposed to extinctions in the past as well? Despite mounting evidence demonstrating extinctions of small-bodied vertebrates during the Quaternary (e.g., Archer & Baynes 1972, Archer 1976, 1981, Archer & Bartholomai 1978, Godthelp 1997 , there still appears to be an over-emphasis placed on understanding the fate of 'megafaunal' taxa to the exclusion of small-bodied, less 'charismatic' taxa (e.g., Flannery 1990, Miller et al. 1999, Roberts et al. 2001, Barnosky et al. 2004, Johnson 2006, van der Kaars et al. 2017). The general under-representation of these CWR mammals in extinction studies ignores the long-term trends that have been responsible for shaping extant populations and distributions (Archer et al. 2019). ...
... In contrast to the cool steppe-tundra lands of the northern hemisphere, cool Australian grasslands did not support large herds of substantial animals weighing hundreds of kilos (Dawson 2006(Dawson , 2020. Australian megafauna species were much less massive than those of Africa, Eurasia, or the Americas, and by the LGM, Australia's moderately-mega megafauna had gone (Miller 2005;Field and Boles 1998;Roberts et al. 2001;Bradshaw et al. 2021). So, the highest value land targets were macropods of very modest size, compared to the mammoths, horses, buffalo, reindeer, elk or wildebeest of their contemporaries elsewhere. ...
... In addition, megafauna extinction during the Pleistocene and human agency therein is cited in support of an Early Anthropocene start date. Evidence is collected from 28 sites on the Australian Continent (Roberts, 2001(Roberts, , 1888; and in North America from "random hunting, and low maximum hunting" (Alroy, 2001(Alroy, , 1893. Comparable dates to the "Early Anthropocene" hypothesis include the "Anthropocene soil" by (Certini and Scalenghe, 2011), which suggests using widespread changes in the pedosphere is "the best indicator of the rise to dominance of human impacts on the total environment" (ibid., 1269). ...
The Anthropocene hypothesisthe Anthropocene Epoch Hypothesis continues to generate global debates, both in academia and among practitioners. The basic claim of the concept is that human activities have transformed planet Earth and its atmosphere. Subsequently, humans have evolved into a geological force. The transformation implies planetary-scale changes, evidence of which is provided in the scale and scope of environmental challenges that have significantly broadened and deepened and continue to threaten the very processes—from a stable climate to biodiversity—on which life on earth in general and human development in particular depend. These developments provide pause for thought on the important nexus of human activities and their subsequent impacts on planet Earth; however, a number of important questions remain unanswered.
Full-text available
(Chinese Title: 中国东部灵长类及其他常见兽类的分布变迁:1573~1949——基于地方志与 GIS 技术的量化分析. The final version was submitted to Sun Yat-sen University Library and The Ministry of Education of the People's Republic of China on December 22, 2022.) Background: Primatology is an important branch of biological anthropology, and it is a cross-disciplinary area with biology, psychology, etc.. Humans are causing the 6th Mass Extinction since several hundred years ago. The long-term co-existence between people and wild animals exerted a long-term and formative influence on the distribution of animals. And mammals, especially primates and other large/medium-sized mammals, had a particularly close relationship with people and were affected severely, thus become important indicators of ecological change. The human population of China, especially in eastern China, increased rapidly from the late Ming Dynasty to the Republic of China (ROC), and the distribution of wild animals including primates was greatly affected. Numerous local gazetteers in Ming dynasty, Qing Dynasty and ROC are well preserved today, and their records of local products are valuable resources for researching the historical biodiversity. However, previous studies only focused on the records of present animal and ignored the records of absent animal, resulting in obvious biases, and there is still a lack of quantitative studies. In order to understand the changes of the distribution of mammals and the influence from the population increasing in China better, I made full use of the records in local gazetteers to reconstruct the distribution of mammals and analyzed the distribution history of large and medium-sized mammals quantitatively. Methods: I established the Database of Wild Mammal Records in Chinese Local Gazetteers. And innovatively, I fixed the biases in previous researches, i.e. I analyzed the historical changes of biodiversity by using the data of both presence and absence records of mammals. In eastern China from 1573 to 1949 (sorted into 4 periods), I reconstructed the distribution of 14 kinds of mammals which sorted in 5 functional groups with ecological values by using ArcGIS 10.3 software. The mammals are: (1) primates: “yuan” - gibbons & Colobinae(including langur monkeys and snub-nosed monkeys, different to modern taxonomy in Chinese) and “hou” - macaque monkeys; (2)large carnivores: tigers, leopards and bears; (3)medium-sized carnivores: wolves, foxes, “li”(civets, including Felinae and Viverridae, different to modern taxonomy in Chinese), dholes and mustelids; (4)large deer: large deer as a whole (including moose); (5)medium-sized deer: “Zhang-She”(including water deer and musk deer), muntjac deer and roe deer. I used statistical software, e.g. SPSS, Fragstats and SmartPLS, to analyze the changes of distribution(area, altitude, slope gradient and fragmentation index) and the impact from the increase of human population. And analyzed the indexes of functional group richness and species(kinds) richness combined with the relevant events of human population history, to show ecological environmental changes in each provincial-level administrative regions. Results: (1)In general in all periods, with the increase of human population, the distribution area of primates, large carnivores, large-sized deer and medium-sized deer retracted, the mean altitude and mean slope gradient increase and the distribution changed from distributed widely to be confined in mountainous areas with high altitudes and high slopes, as the “refuge effect”. (2)The distribution of medium-sized carnivores expanded in general, confirming to be a typical ecological decline phenomenon - mesopredator release. The mean altitude and mean slope gradient also increase, but meaning expand from plains to mountains. (3)The ecological environment in the research area deteriorated with the increase of human population, but some areas in some periods recovered temporarily after specific events e.g. wars in Sichuan in late Ming–early Qing, Taipingtianguo rebellion, muslin anti-Qing revolts in Tongzhi reign. Conclusion: (1)Based on the reconstruction, I provide directive evidences of the human interference on the historical distribution of mammals, and show the specifics. Thus, quantitatively, I prove that the increase of local human population played a significant role in this process. The stereotype “wild large/medium-sized mammal live in hilly areas” is not a natural status but a man-made phenomenon in long term. The results provide a support for further researches. (2)Pioneeringly, I discovered that the environment in eastern China experienced the mesopredator release phenomenon in recent centuries, providing a base for further researches by researchers. (3)And my innovative reorganization of local gazetteers and the application of quantitative methods also provide examples for similar further researches. Keywords: Primate; Local Gazetteer; Historical Zoogeography; Quantitative History; 6th Mass Extinction 背景:灵长类学是生物人类学的重要分支,也是人类学与生物学、心理学及其他学科的交叉领域。数百年来,人类正逐步造成第六次物种大灭绝。人类与野生动物长期相处,对动物分布格局产生了长期且塑造性的影响,而兽类(即哺乳纲动物)尤其是包括灵长类动物在内的大中型兽类与人类关系密切,受影响亦尤为明显,是生态变化的重要指示性物种。明后期至民国是中国,尤其是中国东部人口急速增长的时期,灵长类等大中型兽类分布受影响明显,且该时期地方志资源丰富,明代、清代及民国时期的地方志大量保存至今,其对各地物产的记载是研究历史上生态多样性的宝贵资源。但已有相关研究只关注动物在当地分布(Presence)的记录,而忽略了动物在当地不分布(Absence)的记录,造成明显偏差,且量化研究尚十分欠缺。为更清楚地了解灵长类等兽类分布变化及其受到中国人口剧增的影响,笔者充分利用地方志中的记载,重拟各类曾广泛分布的灵长类等大中型兽类分布情况并量化分析其分布变迁历史。 方法:笔者建立了中国地方志兽类记录数据库,创新性地修正已有研究的偏差,在新方法中同时使用兽类分布与不分布的数据插值制图重拟其分布的历史变迁。以1573年(万历元年)至1949年间(分四阶段)的中国东部大陆地区为时空范围,笔者使用地理信息系统软件ArcGIS 10.3重拟14类兽类动物的分布,并根据生态意义将这些兽类划分为5个生态功能群,即(1)灵长类功能群:猿类(长臂猿以及含叶猴与金丝猴在内的疣猴,与现代科学分类体系所指猿类不同)、猴类(猕猴属);(2)大型食肉兽类功能群:虎、豹类、熊类;(3)中型食肉目兽类功能群:狼、狐类、狸类(猫亚科与灵猫科,与现代科学分类体系所指狸类不同)、豺、鼬类;(4)大型鹿类功能群:大型鹿类整体(含麋);(5)中型鹿类功能群:獐麝类(獐与麝类)、麂类、狍。笔者通过SPSS、Fragstats、SmartPLS等统计软件量化分析分布的面积、海拔、坡度与破碎化程度等变化情况及分析人口剧增对动物分布的影响;并根据功能群数与大中型兽类种类数两种指标对各省区的大中型兽类多样性变化进行梳理,结合人口史资料研究当地生态变迁。 结果:(1)总体上,随着人口剧增,灵长类、大型食肉目兽类、大型鹿类及中型鹿类的分布面积均有不同程度的缩减,分布平均海拔与平均坡度均有所提升,其分布从较广泛分布缩减至主要在高海拔大坡度的山地分布,产生“避难所效应”。(2)而中型食肉目兽类的分布总体上实现了扩张,经证实为中型捕食者释放效应这一典型的生态衰退现象,其分布平均海拔与平均坡度亦有所提升,但主要为从平原地区向山地扩散。(3)量化证实研究区域的大中型兽类多样性总体上随着人口增长而恶化,但明末清初四川战乱、太平天国运动、同治年间回民反清斗争等事件后部分地区的多样性在特定阶段有所恢复。 结论:(1)基于重拟,笔者给出了历史上人类干扰灵长类等大中型兽类分布的直接证据并展示了具体变化过程,量化证实中国东部人口增长在大中型兽类分布缩减过程中起了明显的作用,“大中型兽类主要在山地分布”这一刻板印象并非自然状态,而是人类长期塑造的结果,为学界提供了进一步研究的支持。(2)笔者首次发现了近世中国东部经历了中型捕食者释放效应现象,为学界提供了进一步研究的基础。(3)同时笔者对地方志动物记载的创新性整理及量化处理,也将为将来类似研究提供范例。 关键词:灵长类,地方志,历史动物地理学,量化历史学,第六次大灭绝
The use of phylogenetic methods for the of study macroarchaeological processes of cultural evolution has long been advocated for, especially for the study of artefact shape. Due to technical limitations, the use of discrete character traits in these efforts prevails, which comes along with a non-trivial set of challenges. However, recent methodological advances in palaeobiology (Parins-Fukuchi 2017; Zhang, Drummond, and Mendes 2021) allow us now to infer Bayesian phylogenies using continuous characters. In this article, we present an exploratory analysis of cultural evolution in the European Final Palaeolithic and earliest Mesolithic (~15.000-11.000 BP) using a time-scaled Bayesian phylogeny based on the outline shape of lithic tools obtained from legacy data via 2D outline based geometric morphometrics. We estimate the rates of diversification and trait evolution in artefact shape, and test whether the recovered patterns correlate with the climatic and environmental changes within this timeframe. While common in cultural evolutionary analyses of language, the extension of Bayesian phylogenetic methods to archaeology represents a major methodological breakthrough.
Full-text available
Islands have long been recognized as distinctive evolutionary arenas leading to morphologically divergent species, such as dwarfs and giants. We assessed how body size evolution in island mammals may have exacerbated their vulnerability, as well as how human arrival has contributed to their past and ongoing extinctions, by integrating data on 1231 extant and 350 extinct species from islands and paleo islands worldwide spanning the past 23 million years. We found that the likelihood of extinction and of endangerment are highest in the most extreme island dwarfs and giants. Extinction risk of insular mammals was compounded by the arrival of modern humans, which accelerated extinction rates more than 10-fold, resulting in an almost complete demise of these iconic marvels of island evolution.
The Lancefield megafauna site is located on the southwest edge of the small town of Lancefield, 70 km NNE of Melbourne. The site is located in a swamp, a depression (possibly formed by a collapsed lava tunnel) which is almost surrounded by weathered Pliocene basalts which have formed a laterite cap. A natural spring flow under this cap emerges at the swamp and the water then drains into Deep Creek, a tributary of the Maribyrnong River. Three fossil megafaunal assemblages occur at Lancefield. The original discovery of 1843, now known as the Mayne Site, was the focus of investigations in the nineteenth-century and again in 1991 (van Huet 1993). The South Site, found in 1983, has been excavated three times since the early 1980s, most recently by van Huet in 1991 (van Huet 1994). The Classic Site was discovered in 1973 and was the focus of major investigations in 1975-76 (Horton 1976; Ladd 1976; Gillespie et al. 1978; Horton and Samuel 1978). Â
For at least the last two decades, since the publication of Martin's forceful articles (1967, 1973, 1984) discussions of late Pleistocene extinctions on the two continental landmasses of the Americas and Sahul have been basically polarised between his 'blitzkrieg' view and the opposing theory of climatic change. In Australia, Flannery (1990) has been the most recent strong proponent of the blitzkrieg model, and the target of critical comments from several other researchers (Horton 1990; Grayson 1 990; Bowdler 1990). Those who reject the model point to its supporters' inability to demonstrate that extinctions occurred very shortly after the arrival of humans. Cogently phrased as ‘Where are the kill sites?', critics have drawn attention to two major unfulfilled archaeological requirements: the clear association of extinct animals with any human activity and the dating of extinct animals to within the period of human settlement. While some Australian sites appear to provide such data - Seton (Hope et al. 1977), Lancefield (Gillespie et al. 1978), Lime Springs (Gorecki et al. 1984), Trinkey (Wright 1986) and Cuddie Springs (Dodson et al. 1993) - and are often accepted as so doing, strict examination of the evidence reveals the flimsy basis to such claims (Grayson 1990; Baynes 1992). While a detailed re-examination of all claimed dates and associations, such as has been carried out by Mead and Meltzer (1984) and Grayson (1991) for the American data, has not been published for Australia, greater precision in regard to these at any site will contribute to the larger picture. At Spring Creek (Flannery and Gott 1984) eight extinct species have been claimed to survive to well within the period of human settlement, though with only minimal evidence of human association. However, the association of the date of 20,000 yr BP with extinct fauna is not very strong, and our research has been aimed at either strengthening or discarding it. In particular, we determined to try to obtain dates directly on the bones themselves as well as re-examining the circumstances under which they were deposited at the site.
A unifying, predictive hypothesis is developed that explains many facets of late Quaternary biotic change in Australia. Pleistocene faunal extinction (commonly called megafaunal extinction) is envisaged as the precipitating mechanism for much environmental change. The extinction is thought to be of the “blitzkrieg” type as presented by Martin (1973), and was followed by suppression through human hunting of the remaining large herbivores. Much vegetation was left uneaten because of these events, resulting in an increased standing crop of fuel. Increased fuel continuity and mass ensured that fires could become much larger and hotter than was usually possible previously. The changed fire regime eventually eliminated fire sensitive Gondwanan rainforest elements from almost all parts of the continent that were not protected. Aborigines responded to the changed fire regime with firestick farming. This maintained a high diversity as it ameliorated the worst effects of fires on medium-sized mammals and possibly some plants. Twentieth century Australian mammal extinctions are the result of a trophic cascade that followed the cessation of firestick farming.
A thermoluminescence dating program has sampled 3 sites on the shores of Lake Mungo, the burial sites Mungo I and Mungo III and the site of earlier palaeomagnetic studies. Two sites on Outer Arumpo lake at Long Water Hole gully and the Outer Arumpo lunette, provide a basis for comparison with ages from Lake Mungo and between lake basin and lakeshore sediments within the Outer Arumpo sequence. Most results show reasonable agreement, both with previous TL results and with available radiocarbon ages. The Golgol unit is confirmed as older than 100 ka. Ages beyond 40 ka from the upper levels of the Lower Mungo unit at Lake Mungo and from the equivalent sands on the Outer Arumpo lunette provide upper limits to ages of burials inserted into the Lower Mungo sands (Mungo I and III). Some problems arise in TL signals from lacustrine sediments which, in samples from Long Water Hole gully, appear too young compared to apparently more reliable results from aeolian components.
The story of the Willandra Lakes is also the story of those ancient people who lived there. The landforms, sediments and soils provide the environmental framework within which the patterns of human occupation must be interpreted. The original stratigraphic system involved just two units, the Mungo and Zanci. Two additional units are now defined; one incorporating complexities between Mungo and Zanci, the Arumpo Unit, and a second to acknowledge the reality of a lake full phase near 18ka cal., postdating Zanci aridity of the last glacial maximum. This new unit is defined from Lake Mulurulu as the Mulurulu Unit.
Geological extinction of a continental megafauna of Holarctic mammoths, American ground sloths, and Australian diprotodonts, to name a few mammalian examples, rivals pulsing ice sheets and fluctuating sea levels in being a hallmark of the Quaternary. To these more familiar examples of late Quaternary extinction (LQE), younger fossils recently recovered from oceanic islands, including bird and land snail taxa from the Pacific and various endemic terrestrial vertebrates from the Caribbean, add many thousands of species and endemic populations to the extinction list. Apart from loss of a few pinnipeds and sirenians (large coastal or estuarine mammals), the LQE was strictly a terrestrial accident. Unlike the case on islands, on the continents virtually all small vertebrates (other than commensals or parasites of the large mammals) escaped extinction. Looking toward the continents from deep-water islands promises to aid in our understanding of what happened, when it did, and what forced the change.
Carbon isotopes in fossil emu (Dromaius novaehollandiae) eggshell from Lake Eyre, South Australia, demonstrate that the relative abundance of C4 grasses varied substantially during the past 65,000 years. Currently, C4 grasses are more abundant in regions that are increasingly affected by warm-season precipitation. Thus, an expansion of C4 grasses likely reflects an increase in the relative effectiveness of the Australian summer monsoon, which controls summer precipitation over Lake Eyre. The data imply that the Australian monsoon was most effective between 45,000 and 65,000 years ago, least effective during the Last Glacial Maximum, and moderately effective during the Holocene.