Journal of Archaeological Science 58 (2015) 103e112
Contents lists available at ScienceDirect
Journal of Archaeological Science
journal homepage: http://www.elsevier.com/locate/jas
New estimations of habitable land area and human population size at
the Last Glacial Maximum
Joanna R. Gautney a, *, Trenton W. Holliday a, b
a
b
Department of Anthropology, Tulane University, 101 Dinwiddie Hall, 6823 St. Charles Avenue, New Orleans, LA 70118, USA
Evolutionary Studies Institute, University of the Witwatersrand, Private Bag 3, Wits 2050, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 November 2014
Received in revised form
20 March 2015
Accepted 21 March 2015
Available online 4 April 2015
The estimation of human population size during the Pleistocene is complex, and one which has been
dealt with extensively in the literature. However, because many of these previous estimations are based
in part on archaeological site distributions, they are more a reflection of present-day geography than of
what the Earth looked like in the past. We address this issue by calculating an estimation of habitable
land area during the Last Glacial Maximum (between 22 and 19 kya) when sea level was 120 m lower
than today using the polygon creation function in Google Earth. We then subtract areas of land that were
likely uninhabitable during the LGM e either due to glacier cover, extreme aridity, elevation, or areas at
high latitudes. From this, the combined habitable land areas of Eurasia, Africa and the Australian landmass are estimated as 76,959,712.4 km2. This estimation is then coupled with population density data for
medium-to large-bodied carnivores, and ethnographic population density data for hunteregatherers
culled from the literature. Total human census population size in the Old World during the Last Glacial
Maximum is estimated at 2,117,000e2,955,000 based on carnivore densities and 3,046,000e8,307,000
for hunteregatherer densities.
© 2015 Elsevier Ltd. All rights reserved.
Keywords:
Last Glacial Maximum
Human population size
Human biogeography
Paleogeography
Sea level change
1. Introduction
Prehistoric human population size is a fundamentally important
question in paleoanthropology. Evolutionarily speaking, it is populations that evolve, and since smaller populations are more likely
to go extinct than larger ones, population size is a critically
important variable in evolutionary biology (Harmon and Braude,
2010). With regard to human evolution, population size and density played a role in the origins and spread of modern human
behavior (Powell et al., 2009; Stringer, 2012), and it is a central
question in the debate over the geographic origins and early migrations of anatomically modern humans. One of the strongest
arguments against the multiregional hypothesis for the evolution
of modern humans is that it requires a large enough population size
at any given time to have maintained gene flow across the Old
World (Holliday et al., 2014).
Issues such as effective population size versus census population size of Pleistocene human populations are used by proponents
* Corresponding author. Tel.: þ1 831 210 5751.
E-mail address: [email protected] (J.R. Gautney).
http://dx.doi.org/10.1016/j.jas.2015.03.028
0305-4403/© 2015 Elsevier Ltd. All rights reserved.
of both the Recent African Origin [RAO] and Multiregional Evolution [MRE] models of modern human origins (Eller et al., 2009; Ray
et al., 2005; Rogers and Harpending, 1992; Wolpoff et al., 2000).
Effective population size (Ne), originally defined by Sewall Wright
(1931), refers to the breeding population, including both males
and females, of a species. Ne is a subset of individuals within the
census population number (N). The relationship between Ne and N
is a complicated one that can be affected by habitat variables as well
as population expansion and contraction, and recent studies have
indicated that there is likely no simple relationship between the
two (Belmar-Lucero et al., 2012; Palstra and Fraser, 2012). These
concepts play an integral role in any discussion of the origin and
spread of modern humans, and N/Ne estimates have been used to
support sometimes contradictory models. For example, a small Ne
size during the Pleistocene has been cited as evidence against MRE,
as it would have been unlikely that the sufficient level of gene flow
MRE requires could have been maintained across the Old World
with so few humans spread over such a large area (Harpending
et al., 1993, 1998). Supporters of MRE have responded that population expansions and contractions, along with local population
extinction, were the mechanism of gene flow during the Pleistocene, preventing regional speciation events (Eller et al., 2009).
104
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
The issue of effective and census population size is significant
from the perspective of gene flow, and also similarly in questions of
the transmission of culture. In his book The Selfish Gene, Richard
Dawkins (1976) first popularized the concept of units of human
culture (“memes”) being analogous to the role genes play in
Darwinian evolution, and over the past 30 years, using an evolutionary framework to explain the evolution of human culture has
drawn much attention from social scientists (Atran, 2001; Bentley
et al., 2004; Cavalli-Sforza and Feldman, 1981; Eerkens and Lipo,
2005; Henrich, 2001; Mesoudi et al., 2004; Shennan, 2000). If early
human culture may be considered as an inheritance system of
adaptive information (Shennan, 2002), population size and density
play as important a role in the evolution of culture as they do in the
strictly biological processes involved in the evolution and spread of
anatomically modern humans. Dual inheritance theory postulates
that culture and genes have coevolved in modern humans by natural selection, and operate in similar ways (Richerson and Boyd,
1978). Selection operates on variation, both genetic and memetic,
and variation is increased as population size and density are
increased. Cultural transmission of ideas can occur more quickly,
and new information can be better retained, in a large population
(Eerkens and Lipo, 2005). Therefore, estimations of Pleistocene
population size are important and integral to a wide range of
paleoanthropological debates.
These issues have been covered in great detail in the literature,
but the glaring omission, however, is that to date, very little
attention has been paid to calculating population densities and
population size based on actual land area estimates, and in
particular land area estimates that take into account the lower sea
levels associated with glacial periods. Without accurate Pleistocene
habitable land area estimates, population density and size estimates will simply not be reliable.
The estimation of prehistoric population size is likewise problematic in numerous ways. In the past, researchers have relied on
archaeological site distributions and/or estimations of hunteregatherer population densities to estimate population size
(Hassan, 1981; Binford, 2001; Bocquet-Appel et al., 2005). The
“dates as data” method for regions around the world, championed
thirty-five years ago by Rick (1987) and many others since
(Holdaway and Porch, 1995; Kuzmin and Keates, 2005; Peros et al.,
2010; Shennan and Edinborough, 2007), relies on the complex
relationship between chronometric dates and human occupation of
a given region. The number of dates is directly related to the extent
and frequency of occupation, and thus represents human activity at
that moment in time (Rick, 1987). These data may theoretically be
used to make inferences about migration and settlement patterns,
demography, and population density.
While this method has experienced a surge in popularity in the
past thirty-five years, it is not without limitations. Surovell and
Brantingham (2007) argue that this method does not highlight
demographic trends, but rather taphonomic bias. Taphonomic
processes such as the weathering and erosion of older material lead
to an overrepresentation of more recent material. Therefore, any
reconstruction of prehistoric populations through time will suggest
relatively more activity in a given region in recent times than in the
more distant past, whether that is an accurate reflection of the
record or not (Surovell and Brantingham, 2007; Surovell et al.,
2009). Williams (2012) highlights other methodological problems
including the issue of inadequate sample sizes, sampling errors
resulting in skewed results, and the influence of radiocarbon calibration on analyses.
This approach also carries with it several problems we wish to
address more specifically with the current study. First, there were
likely areas of occupation that left no archaeological record, the
significance of which is only just now being recognized (Bailey and
Flemming, 2008; Bicho and Haws, 2008). Second, large potentially
habitable land areas that were exposed during glacial maxima are
largely unavailable for study due to Holocene sea level increases.
Evidence indicates that the exploitation of marine resources likely
played a significant role in the population and geographic expansion of anatomically modern humans after 150 kya (Erlandson,
2001). Additionally, the early colonization of Australia by 50 kya
(Thorne et al., 1999) is suggestive of a seafaring ability and implies
regular exploitation of marine resources (Allen and O'Connell,
2008). Theoretical biases in modern archaeology have led to an
underestimation of the importance of marine resources and coastal
environments during the Middle and Upper Paleolithic. These
biases are based on the relative rarity of coastal Pleistocene
archaeological sites in Europe, which was interpreted by archaeologists in the late 20th century as evidence that humans did not
regularly exploit marine resources until Early Holocene (Bicho and
Haws, 2008). However, if these coastal areas were occupied during
glacial maxima, there would be a clear bias in the archaeological
record, as those sites would have been submerged when sea level
rose. Those now-submerged coastal regions would potentially have
been more productive during glacial periods than arid inland
landscapes, and thus more attractive as settlement locations (Bailey
and Flemming, 2008). By ignoring these now-submerged areas, we
would ignore some of the most logical areas of human occupation
during glacial periods. Third, it is likely that many Pleistocene
humans lacked the technological sophistication to live at population densities comparable to those of hunteregatherers in the
ethnographic present (Churchill, 2014; Marlowe, 2005; Richerson
et al., 2009). Population size and density are inextricably linked
in the archaeological record. Increased population density requires
technological innovation to more efficiently extract adequate resources from the environment, and the maintenance of that technological innovation requires a certain level of population density.
Neither of these conditions likely applied to most Pleistocene
groups, as they lacked both technological sophistication and high
population density. Therefore, Pleistocene human population sizes
based on hunteregatherer data from the ethnohistoric present are
almost certainly gross overestimates, at least prior to the Upper
Paleolithic (Holliday et al., 2014).
In order to address these issues, we present a new method for
estimating habitable land area during the Last Glacial Maximum
(LGM). The period known as the Last Glacial Maximum began when
global ice volumes reached their maximum between 22 and 19 kya
(Clark et al., 2009; Yokoyama et al., 2000). These ice volumes
remained fairly constant through 19,000 years ago, when the
climate began to warm and sea level began to rise. The method and
data we present include areas that were exposed during glacial
maxima but are below current sea level. We estimated the areas of
Eurasia, Africa and Australia when sea level was 120 m lower using
the polygon creation function in Google Earth, which calculates the
area within the polygon. We couple these data with population
density data for wide-ranging medium-to large-sized carnivores
and recent human foragers to estimate Pleistocene population size.
2. Material and methods
2.1. Land area estimation
We previously used this method to calculate habitable land area
in Africa and Eurasia proper during glacial maxima (Gautney and
Holliday, 2013; Holliday et al., 2014). We reported estimates of
Eurasia and Africa when sea level was at its highest during the LGM
using the polygon creation function in Google Earth. This function
calculates the land area contained within the polygon. Google Earth
is a free geographical information and virtual map program
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
available for private and commercial use, which maps the Earth
using satellite data provided by the National Oceanic and Atmospheric Administration, the United States Navy, the National
Geospatial-Intelligence Agency, among other agencies. The 3D
virtual map includes topographical data both above and below
current sea level using elevation data derived from NASA's Shuttle
Radar Topography Mission (STRM). Google Earth provides users
with a suite of tools for creating their own maps. Included in these
is the polygon creation function, in which users may draw shapes
on a map by plotting points. Elevation data for each point are given.
The software can then calculate the land area contained within the
polygon.
Our previously reported estimate of habitable land area for the
Pleistocene Old World was 66,637,563.4 km2 (Gautney and
Holliday, 2013; Holliday et al., 2014). Using the same method, we
now report revised data from Africa and Eurasia, and in addition,
we report results including the Australian landmass during the
LGM. Thus, here we estimate the areas of Eurasia, Africa and
Australia when sea level was at its lowest during the Late Pleistocene. Areas of land that were likely uninhabitable during the LGM
were then subtracted from that total area. These areas of land,
described below, were measured using the same method. We
consider uninhabitable land areas to be regions covered by ice
sheets in northern Europe, and latitudes, altitudes, and areas of
aridity at which human occupation was unlikely. Those areas
omitted in our estimate include the area of northern Europe
covered by the Scandinavian Ice Sheet, latitudes at which permafrost was uniformly present, altitudes higher than 3,000 m, and
areas designated as “tropical extreme deserts” (Ray and Adams,
2001).
105
Due to the presence of these large ice sheets, sea level was
drastically different during the LGM than in present day. Lambeck
et al. (2002) estimated that sea level stood 120e130 m lower
than present day throughout the Old World (Lambeck et al., 2002).
Comprehensive studies of ice sheet extent, ice sheet modeling,
records of sea level change, and geochemical models of glacial
isostatic adjustment are consistent with that estimate and indicate
a net decrease in sea level of 120e135 m at the LGM (Clark and Mix,
2002). For the purposes of this study, the polygons were drawn
120 m lower than present day sea level, with one landmark plotted
approximately every mile along all coastlines.
2.1.1. Eurasia
A map of Eurasian habitable land area at the LGM looks quite
different from a present-day map (Figs. 1 and 2). By far the largest
land area exposed when sea level dropped was the Sunda Shelf,
extending the southeastern region of Asia further southward.
While Eurasia and Australia have never been connected by a land
bridge, the distance between the two coastlines was much narrower during this period, and this exposed a large area of habitable
land. However, as stated above, areas in other regions during this
period were likely uninhabitable for a variety of reasons. Those
areas were also calculated and removed from the above estimates.
The Scandinavian Ice Sheet covered large land areas in northern
Europe that would otherwise have been exposed at a sea level
120 m lower than present day (Svendsen et al., 2004). As the name
implies, the ice sheet covered all of present-day Scandinavia, along
with portions of the Baltic region of northern Europe. Unlike
northern Europe, for the most part, Siberia remained free of glacier
cover. Ice sheets were not present in Siberia during this period
Fig. 1. Reconstruction of European coastlines and the extent of the Scandinavian Ice Sheet at the LGM.
106
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
Fig. 2. Reconstruction of Asian coastlines, the Himalayan Plateau, and land area north of 65" North latitude.
(Mangerud et al., 2002), but thick and continuous permafrost
covered all of Siberia and the Russian Far East during the LGM.
Archaeological materials dating to the LGM are in fact present in
parts of Siberia. However, no known archaeological site north of 63"
North latitude in Siberia has ever been reliably dated to the Last
Glacial Maximum (Kuzman, 2008). The Paleolithic Siberian site of
Yana is located at 71" North latitude and has been dated to
approximately 27 kya, corresponding to what would traditionally
have been known as Würm III, or the end of the Last Pleniglacial
(Gamble, 1986), and just before the ice sheets expanded to
maximum positions (Clark et al., 2009; Pitulko et al., 2004). For this
reason, Siberia north of 65" North latitude is excluded from the
estimate (Fig. 1).
Likewise, the Himalayan Plateau was largely uninhabitable
during this or any other period in the Pleistocene. Some indigenous
Himalayan and Andean groups today have developed physiological
adaptations to living at high altitudes (Jansen and Basnyat, 2011;
Young and Reeves, 2002). However, the prevalence of hypoxiarelated illnesses faced by all human populations have limited our
ability to live at altitudes higher than 3,000 m, as few cultural/
technological innovations exist to ameliorate these problems, even
today (Beall and Steegmann, 2000). Recent archaeological discoveries at the Pucuncho Basin in the Peruvian Andes do demonstrate
that hunteregatherers were present at an elevation 4480 m above
sea level in the Terminal Pleistocene (~12 kya), when temperatures
were relatively milder (Rademaker et al., 2014). However, to date
this is the only evidence of such a settlement, and no evidence
indicates that settlements at similar elevations in the Himalayan
Plateau during the LGM would have been feasible. For this reason,
the portion of the Himalayan Plateau at elevations higher than
3,000 m was removed from the Eurasian estimate.
During Pleistocene glacial maxima, the Arabian Desert would
have been uninhabitable, as it is classified as “tropical extreme
desert”. This is defined as a region almost entirely barren with less
than 2% vegetation cover (Ray and Adams, 2001). With the exception of portions of the Levant that were more temperate due to
proximity to the Mediterranean, the entire area occupied by the
Arabian Desert was subtracted from the Eurasian habitable land
estimate. Though the Persian Gulf was dry due to lower sea levels
during the LGM, the Tigris and Euphrates Rivers likely extended
through this area to the Strait of Hormuz (Glennie, 1998; Lambeck,
1996; Teller et al., 2000). The area contained between the rivers was
probably a marshy, lake-dotted environment certainly suited to
human habitation (Ionides, 1937). The area designated as the
Arabian Desert borders the Tigris and Euphrates River Valleys to the
west and excludes that marshland (Fig. 3).
2.1.2. Africa
The mass of arid land that comprised the Arabian Desert
extended west into the similarly arid Sahara. During this and other
glacial periods, the Sahara was more extensive than it is today, and
throughout the rest of the African continent, forested areas were
drastically reduced in size. The portion of the Sahara also classified
as “tropical extreme desert” dramatically expanded (Ray and
Adams, 2001). This likely rendered large areas of northern Africa
uninhabitable. Jolly et al. (1998) combined African pollen and lake
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
107
Fig. 3. Reconstruction of the African coastline and the extent of the Sahara and Arabian Desert at the LGM.
data with climate and vegetation simulations to create regional
climate models for different parts of Africa. Under their simulation,
parts of northern Africa along the Mediterranean coast would likely
have been habitable. Areas that would have been more temperate
north of the Sahara were included in the habitable land area estimation (Fig. 3).
Today, the Nile River Valley is a long, fertile strip of land cutting
through a portion of the Sahara in the east. The LGM, however, was
a dry period across the Nile Basin. Reduced precipitation in East
Africa (Goudie, 1992) left Lake Victoria reduced in size and an
isolated basin unconnected to the White Nile (Beuning et al., 1997).
The White and Blue Nile Rivers, sources of the main Nile River,
were drastically reduced in flow during this period. The White
Nile, unconnected as it was from Lake Victoria, became a muchreduced seasonal stream, likely drying out completely during all
or parts of the year (Adamson et al., 1980; Williams et al., 2000).
The seasonal discharge of the Blue Nile was also reduced (Said,
1993). As a result, the main Nile River was a low-discharge, seasonal river and would not have been a consistently habitable
location for human populations to live year-round. We therefore
exclude the Nile River Basin from our estimation of habitable land
area (Fig. 3).
2.1.3. Sahulland
In a manner similar to the Sunda Shelf discussed above, the
Sahul Shelf was exposed as dry land during the LGM. New Guinea
and Tasmania were connected via dry land to mainland Australia as
a large landmass known as Sahulland. Multiple independent
studies of different locations around the Sahul Shelf have found
that Australian LGM shorelines were 120e135 m lower than present levels, corresponding with estimates reported by Lambeck
et al. (2002) (Ferland et al., 1995; Van Andel and Veevers, 1967;
Yokoyama et al., 2000) (Fig. 4).
A good deal of Australian archaeological research over the past
30 years has focused on late Pleistocene human response to climate
change. During the LGM, the interior areas of the Australian landmass were substantially more arid than at present, and the arid
zones were much larger (Smith, 1989). The glacial climate of the
arid zones would likely have made human occupation difficult if
not impossible. Several studies have argued that during the LGM,
human populations underwent a drastic decrease in population
size and a reduction in settlement range (Hiscock, 1988; Lampert
and Hughes, 1987; Veth, 1989). Veth (1989) proposed the “Islands
in the Interior” model for the colonization of Australia's arid zone,
arguing that limited areas with reliable water sources in the interior became refugia for human populations during climatic extremes. In a recent analysis of the 25e12 kya Australian
archaeological record, Williams et al. (2013) argued that Australia
at the LGM was indeed characterized by large areas devoid of human activity (barriers), corridors and human population refugia.
Their geospatial analysis of LGM bioregions and the archaeological
record suggest a reduction in occupied territory of nearly 80%. To be
conservative, we removed the areas designated by Veth (1989) as
“barriers”, which were the arid Great Victorian Desert, the Simpson
Desert, and the Great and Little Sandy Deserts (Fig. 4).
2.2. Density and population size estimates
Many attempts have been made to estimate global effective and
census human population size during various points in the Pleistocene (see Tables 1 and 2 in Holliday et al., 2014). Effective population estimates drawn from genetic data range from 700 to
108
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
Fig. 4. Reconstruction of the Sahulland coastline, the Great Victorian Desert, the Simpson Desert, and the Great and Little Sandy Deserts at the LGM.
100,000, and census estimates range from 40,000 to 1,000,000
(Ayala, 1995; Dennell et al., 2011; Haigh and Maynard Smith, 1972;
Hassan, 1981; Zhivotovsky et al., 2003). As previously stated, many
of the census estimations were calculated by using land area data
based on archaeological site distribution paired with hunteregatherer population density data. This method of estimation is
problematic because it excludes areas that likely were habitable,
but are archaeologically invisible due to sea level increases after the
LGM.
We paired the above LGM habitable land area estimates with
average hunteregatherer population density data groups taken
from Binford (2001), and with a range of population density estimates for large-bodied carnivores. We use carnivores of comparable size to humans as a means of establishing a baseline for an
average density the environment could support without cultural
and technological innovations humans possessed by the LGM.
Some groups from Binford's data set of n ¼ 339 were excluded. We
eliminated data from complex hunteregatherer groups who rely
extensively on anadromous fish, as their exploitation was unlikely
before the PleistoceneeHolocene boundary (Binford, 1990;
Holliday, 1998). We also excluded the groups Binford designated
as “suspect”. These “suspect” groups are those that were likely
hunteregatherers in the past, but at the time of observation may no
longer have been strictly observing traditional subsistence patterns
and social structures (Binford, 1990). Our data set includes 219
hunteregatherer groups, with a mean density of 0.1223 individuals
per km2 and a median density of 0.0444 individuals per
km2 (Table 1). Since we are estimating population density across
the entire Old World, which admittedly has a wide range of habitats, environments, and carrying capacities, we use mean and
median in our model to produce a range estimate. Because models
are inherently approximations of reality, use of the mean and
median allows us to avoid making even more assumptions that
might be necessary were we to fine-tune the model more.
Binford's (2001) dataset includes population densities for
Australian Aboriginal groups (n ¼ 56 groups, 7 of which were
“suspect” and thus removed). The mean population density for
these 49 groups is 0.1676 individuals per km2. The mean population
density for coastal groups is 0.2428 individuals per km2, and for
inland groups it is 0.0489 individuals per km2. These densities are
not dramatically different from the mean of our selected dataset of
Binford's groups (0.1223 individuals per km2). However, Australia
was home to unique conditions during the LGM that almost
certainly caused population size to be significantly smaller than it
was in the rest of the Old World during this period, and much
smaller than estimates of contemporary hunteregatherer densities.
As discussed above, large portions of the continent are thought to
have been uninhabitable due to extreme aridity (Smith, 1989),
which meant a drastic reduction in habitable land area. As a
consequence, human population size likely declined by approximately 61% at the height of the glacial period (Williams, 2013).
Human population density in Australia just before the LGM has
109
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
Table 1
Hunteregatherer population densities taken from Binford (2001), excluding those who exploit anadromous fish, and those categorized as “suspect”.
Group
Density
Group
Density
Group
Density
Shompen
Onge
Jarwa
North Island
Semang
Orogens
Ket
Gilyak
Yukaghir
Nganasan
Siberian Eskimo
Paraujano
Akuriyo
Nukak
Calusa
Heta
Tehuelche
Chono
Alacaluf
Ona
Yahgan
Hukwe
Hadza
!Kung
G/Wi
!Ko
Auni-khomani
Xam
Larikia
Anbara
Murngin-Yolngu
Jeidji-Forestriver
Wikmunkan
Kakadu
Nunggubuyu
Yintjingga
Yir-yoront
Tiwi
Kuku-Yalanji
Groote Eylandt
Walmbaria
Mulluk
Worora
Lungga
Lardil
Kaiadilt
Karadjeri
Kariera
Warunggu
Djaru
Walbiri
Mardudjara
Ildawongga
Pintubi
Undanbi
Jinibarra
Karuwali
Alyawara
Ngatatjara
Badjalang
Pitjandjara
Dieri
Arenda-southern
Jankundjara
Arenda-northern
Ualaria
Nakako
Barkindji
Karuna
Wongaibon
Jaralde
Mineng
Tjapwurong
39.54
40.1
44.65
33.38
17.57
4.3
1.64
19.31
0.61
0.46
4.7
35
7.04
9.34
38.73
9.6
1.89
13.64
14.98
7.27
28.42
2.9
24
6.6
2.93
1.03
0.64
2.43
40
43.7
11.76
17
19.31
8.8
23
31
8
37.5
50
22.9
58
45
11
4.5
30
66
3.75
9.5
16.28
3.98
1.16
0.75
0.45
1.5
21.74
16
2
1.21
0.4
13.4
0.6
1.93
1.1
1
2.66
9
0.87
15.43
18
5.12
40
7
35
Bunurong
Kurnai
Tasmanians-eastern
Tasmanians-western
Seri
Kiliwa
Diegueno
Lake Yokuts
Serrano
Luiseno
Wukchumi
Tubatulabal
Nomlaki
North Foothill Yokuts
Patwin
Gabrielino
Monache
Chumash
Chimariko
Salinan
Pomo-northern
Yana
Atsugewi
Maidu-Mountain
Achumawi
Modoc
Klamath
Guaicura
Chichimec
Death Valley
Karankawa
Coahuilenos
Panamint Shoshoni
Koso Mountain Shoshoni
Kawaiisu Shoshoni
Saline Valley Shoshoni
Antarianunts-Southern Paiute
Owens Ranch Paiute
Kawich Mountain Paiute
Kaibab Southern Paiute
Mono Lake Paiute
Deep Spring Paiute
Salmon-Eater Shoshoni
Pyramid Lake Paiute
Ute-Timanongas
Cattail Paiute
Fish Lake Paiute
Honey Lake Paiute
Hukunduka Shoshoni
Gosiute Shoshoni
Spring Valley Shoshoni
White Knife Shoshoni
Rainroad Valley Shoshoni
Reese River Shoshoni
North Fork Paiute
Grouse Creek Shoshoni
Ute-Wimonantci
Bear Creek Paiute
Antelope Valley Shoshoni
Washo
Surprise Valley Paiute
Wind River Shoshoni
Ruby Valley Shoshoni
Bohogue North Shoshoni
Uintah Ute
Harney Valley Paiute
Sheep-eater Shoshoni
Little Smoky Shoshoni
Uncompahgre Ute
Comanche
Kiowa
Kiowa Apache
Cheyenne
25.04
17.7
8.17
13.35
25.48
12.25
18.1
38.1
17.58
67.9
24.21
17.2
35
38.29
82
64.9
28.7
118.2
50
37.4
108.4
31.3
17.93
23.5
17.25
22.89
13.36
6
9
1.29
21
1.68
2.12
8.57
11.9
2.32
3.45
38.04
1.99
3.71
5.9
3.54
6.9
18.53
3.47
22
3.89
10.6
2.96
1.67
6.09
11.71
4.28
16.7
16.04
1.64
2.6
1.1
1.13
14.9
13.59
1.87
13.79
1.04
7.48
1.24
6.24
1.82
4.29
2.33
1.4
4.14
4.82
Arapahoe
Crow
Teton Lakota
Kutenai
Bannock
Gros-Ventre
Plains Ojibwa
Piegan
Blackfoot
Assiniboine
Plains Cree
Blood
Sarsi
Coeur d'Alene
Ojibwa-Kitchibuan
Kitikitegon
Micmac
Flathead
Rainy River Ojibwa
North Saulteaux
Shuswap
Pekangekum Ojibwa
Round Lake Ojibwa
Alcatcho
Nipigon Ojibwa
Mistassini Cree
Ojibwa-Northern Albany
Waswanip Cree
Weagamon Ojibwa
Montagnais
Sekani
Beaver
Slave
Kaska
Tahltan
Chilcotin
Carrier
Mountain
Han
Hare
Attawapiskat Cree
Koyukon
Chippewyan
Kutchin
Ingalik
Satudene
Nabesna
Rupert House Cree
Dogrib
Tutchone
Holikachuk
Naskapi
Norton Sound Inuit
Kobuk Inuit
Kotzebue Sound Inuit
Labrador Inuit
Great Whale Inuit
Caribou Inuit
Noatak Inuit
Nunamiut Inuit
Mackenzie Inuit
Sivokamiut Inuit
Point Hope Inuit
Copper Inuit
Utkuhikhaling-miut
Aivilingmiut Inuit
Ingulik Inuit
West Greenland
Bafffin Island Inuit
Netsilik Inuit
Angmakaslik
Tareumiut Inuit
Polar Inuit
7.5
5.81
8.77
2.01
2.31
3.37
2.79
2.54
3.46
3.21
2.73
4.44
1.75
1.5
5
3.09
4.32
1.5
1.21
1.2
12.4
3.08
1.75
7.5
0.87
0.58
1.43
0.41
0.51
0.41
0.82
0.51
1
0.9
1.16
11.52
7.59
0.78
1.8
0.33
1.43
1.09
0.46
1.7
2.71
0.55
0.77
0.9
0.88
0.92
1.52
0.42
7.61
2.67
6.63
2.78
1.86
0.3
2.2
0.96
3.84
15
4.2
0.43
0.38
0.32
0.54
4.73
1.26
0.25
7.72
3.86
0.41
110
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
been estimated as 0.005 individuals per km2, and this estimate was
used to calculate population size (Williams et al., 2013). This estimation is based on archaeological data, the problems with which
we have highlighted, but we believe the extreme aridity of the
continent justifies the use of Williams et al.'s (2013) density estimation. However, for comparative purposes, we have also calculated Australian LGM population size using our estimated carnivore
and hunteregatherer densities as well.
estimate a population size for the combined areas of Africa and
Eurasia of 2,998,820e8,260,262. Using data reported by Grant et al.
(1992), we also calculated a population estimate based on the
population densities of medium-to large-sized carnivores with
range sizes greater than or equal to 50 km2. If mean density for
these carnivores is 0.0384 individuals per km2, and median density
is 0.0275, we calculate a population range of approximately
2,120,000e2,950,000.
If habitable land area of Sahulland was 9,418,730.8 km2, we
estimate human population size during this period to be approximately 47,000 individuals, using Williams et al.'s (2013) population
density estimate of 0.005 individuals per km2. Because population
size in Australia did drop to its lowest at the height of the LGM (and
Williams' density estimate is for just prior to the LGM), and because
this estimate assumes that all areas not designated as barriers were
populated, which is unlikely, this figure should be considered a
population ceiling. Using the medium-to large-bodied carnivore
density data, the population estimate would be considerably larger
at 259,000e362,000. If we apply Binford's densities in this case, we
arrive at an estimate of 418,192e1,151,912 individuals. This range, as
well as that based on carnivore data, is significantly larger than
what the carrying capacity of the environment would likely have
been based on paleoclimatic data (Williams et al., 2013).
Ultimately, then, we estimate total human census population
size worldwide during the Last Glacial Maximum of ca.
2,117,000e2,955,000 based on medium-to large-bodied carnivore
density estimates, and 3,046,000e8,307,000 individuals based on
Binford's (2001) hunteregatherer densities for Africa and Eurasia
combined with Williams et al.'s (2013) LGM density estimate for
Sahulland (Table 3).
3. Results
4. Discussion
We estimate habitable land area across Africa, Eurasia, and
Australia when sea level was 120 m lower than present day during
the LGM to have been 76,959,712.4 km2 (Table 2).
Because population densities of ethnographic groups in
Australia are likely very different from those seen during the LGM,
and because of the unique climatic and geographic conditions in
Sahulland during that time, we calculated population size estimates for Africa/Eurasia and Sahulland separately. Our separate
land area estimate for Africa and Eurasia is 67,540,981.6 km2. This is
higher than our previously reported estimate, because we added
land areas from North Africa along the Mediterranean Sea, and also
the Levant that would likely have been temperate. With a median
population density for Binford's (2001) data set of 0.0444 individuals per km2, and the mean of 0.1223 individuals per km2, we
We present these habitable land area estimations to address a
gap in our knowledge of human biogeography at the Last Glacial
Maximum. It is impossible to paint an accurate and complete picture of human demographics during this period while overlooking
population distribution, especially when marine resources would
likely have become even more important as inland conditions
deteriorated with glacial advance.
The new census population estimates we present more accurately reflect actual habitable land areas during this period than
many past estimates. Our census population size estimate differs
considerably from many of the previously published data on the
subject. Some of these estimates are unrealistically low, ranging
from a mere 40,000 globally (Haigh and Maynard Smith, 1972) to
300,000 (Eller, 2002), but many estimates fall within a range of
Table 2
Land areas.
Location
Land area in km2
Africa
Eurasia
Sahulland
Scandinavian Ice Sheet
Himalayan Plateau
Siberia north of 65" N
Arabian Peninsula
Sahara Desert
North Africa/Levant
Arid Sahulland
Eurasia Uninhabitable
Eurasia Habitable
Africa Uninhabitable
Africa Habitable
Sahulland Uninhabitable
Sahulland Habitable
Eurasia þ Africa Habitable
Total old world land area
Habitable Total
Uninhabitable Total
30,493,900.3
62,280,164.2
11,021,024.1
3,689,822.0
2,887,550.1
2,178,189.0
3,630,924.7
13,735,600.0
889,002.9
1,602,293.3
12,386,485.8
49,893,678.4
12,846,597.1
17,647,303.2
1,602,293.3
9,418,730.8
67,540,981.6
103,795,088.6
76,959,712.4
26,835,376.2
Table 3
Density estimations.
Hunteregatherers
Median
Mean
Carnivores
Median
Mean
Hunteregatherers
Williams et al. (2013)
Median
Mean
Carnivores
Median
Mean
Africa & Eurasia land area estimate
Density (individuals per km2)
Population size estimate
67,540,981.60
67,540,981.60
0.0444
0.1223
2,998,820
8,260,262
67,540,981.60
67,540,981.60
0.027513228
0.0384
1,858,271
2,593,574
Sahulland land area estimate
Density (individuals per km2)
Population size estimate
9,418,730.80
9,418,730.80
9,418,730.80
0.005
0.0444
0.1223
47,094
418,192
1,151,911
9,418,730.80
9,418,730.80
0.027513228
0.0384
259,140
361,679
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
1,000,000e7,000,000 (Birdsell, 1972; Hassan, 1981; Hawks, 1999;
Weiss, 1984). Biraben (1979) estimated that at the time of the
PleistoceneeHolocene boundary, global population size may have
been around 6,000,000. The upper end of our estimate is larger
than all of these estimates, but most closely matches that presented
by Hawks (1999), who provided estimates of population sizes of ca.
2,000,000e7,000,000 based on archaeological and genetic data.
Human niches have always been largely defined by the technology the groups possess, and during the LGM, human technology,
and consequently the ability to negotiate deteriorating environmental conditions, limited population density far more than what
we may observe ethnographically in hunteregatherer groups
(Marlowe, 2005). Furthermore, these estimates assume that all
habitable land areas available during the LGM were occupied,
which is unlikely to be the case. For these reasons, our estimations
should therefore be considered to be upper end estimates.
Acknowledgments
The authors would like to thank Craig Matthews and Zachary
Gautney for technical help with this project, as well as Lukas Friedl
(University of West Bohemia, Czech Republic), Haley Holt Mehta
(Tulane University, New Orleans, LA), Brian Pierson (Pierce College,
Woodland Hills, CA), and Whitney Karriger (Tulane University, New
Orleans, LA) for their support, suggestions, and advice. We also
thank the two anonymous reviewers who provided valuable and
helpful criticism of this work.
References
Adamson, D.A., Gasse, F., Street, F.A., Williams, M.A.J., 1980. Late Quaternary history
of the Nile. Nature 288, 50e55.
Allen, J., O'Connell, J.F., 2008. Getting from Sunda to Sahul. In: Clark, G., Leach, F.,
O'Connor, S. (Eds.), Islands of Inquiry: Colonization, Seafaring, and the
Archaeology of Maritime Landscapes. ANU E Press, Canberra, Australia,
pp. 31e46.
Atran, S., 2001. The trouble with memes. Hum. Nat. 12, 351e381.
Ayala, F.J., 1995. The myth of Eve: molecular biology and human origins. Science
270, 1930e1936.
Bailey, G.N., Flemming, N.C., 2008. Archaeology of the continental shelf: marine
resources, submerged landscapes and underwater archaeology. Quat. Sci. Rev.
27, 2153e2165.
Beall, C.M., Steegmann, A.T., 2000. Human adaptation to climate: temperature, ultraviolet radiation, and altitude. In: Stinson, S., Bogin, B., Huss-Ashmore, R.,
O'Rourke, D. (Eds.), Human Biology: an Evolutionary and Biocultural Approach.
Wiley-Liss, New York, pp. 163e224.
Belmar-Lucero, S., Wood, J.L.A., Scott, S., Harbicht, A.B., Hutchings, J.A., Fraser, D.J.,
2012. Concurrent habitat and life history influences on effective/census population size ratios in stream-dwelling trout. Ecol. Evol. 2, 562e573.
Bentley, R.A., Hahn, S.J., Shennan, S.J., 2004. Random drift and culture change. Proc.
R. Soc. Lond. Ser. B Biol. Sci. 271, 1443e1450.
Beuning, K.R.M., Kelts, K., Ito, E., Johnson, T.C., 1997. Paleohydrology of Lake Victoria,
East Africa, inferred from 18O/16O ratios in sediment cellulose. Geology 25,
1083e1086.
Bicho, N.F., Haws, J.A., 2008. At the land's end: marine resources and the importance
of fluctuations in the coastline in the prehistoric hunter-gatherer economy of
Portugal. Quat. Sci. Rev. 27, 2166e2175.
Binford, L.R., 1990. Mobility, housing, and the environment: a comparative study.
J. Anthropol. Res. 46, 119e152.
Binford, L.R., 2001. Constructing Frames of Reference. University of California Press,
Berkeley.
!volution du nombre des hommes. Population 34,
Biraben, J.-N., 1979. Essai sur l'e
13e25.
Birdsell, J.B., 1972. Human Evolution. Rand McNally, Chicago.
Bocquet-Appel, J.P., Demars, P.Y., Noiret, L., Dobrowsky, D., 2005. Estimates of Upper
Palaeolithic meta-population size in Europe from archaeological data.
J. Archaeol. Sci. 32, 1656e1668.
Cavalli-Sforza, L.L., Feldman, M.W., 1981. Cultural Transmission and Evolution: a
Quantitative Approach. Princeton University Press, Princeton.
Churchill, S.E., 2014. Thin on the Ground: Neandertal Biology, Archaeology and
Ecology. Wiley-Blackwell, New Jersey.
Clark, P.U., Mix, A.C., 2002. Ice sheets and sea level of the Last Glacial Maximum.
Quat. Sci. Rev. 21, 1e7.
111
Clark, P.U., Dyke, A.S., Shakun, J.D., Carlson, A.E., Clark, J., Wohlfarth, B.,
Mitrovica, J.X., Hostetler, S.W., McCabe, A.M., 2009. The Last Glacial Maximum.
Science 325, 710e714.
Dawkins, R., 1976. The Selfish Gene. Oxford University Press, Oxford.
Dennell, R.W., Martinon-Torres, M., Bermudez de Castro, J.M., 2011. Hominin variability, climatic instability and population demography in Middle Pleistocene
Europe. Quat. Sci. Rev. 30, 1511e1524.
Eerkens, J.W., Lipo, C.P., 2005. Cultural transmission, copying errors, and the generation of variation in material culture and the archaeological record.
J. Anthropol. Archaeol. 24, 316e334.
Eller, E., 2002. Population extinction and recolonization in human demographic
history. Math. Biosci. 177e178, 1e10.
Eller, E., Hawks, J., Relethford, J.H., 2009. Local extinction and recolonization, species
effective population size, and modern human origins. Hum. Biol. 81, 805e824.
Erlandson, J.M., 2001. The archaeology of aquatic adaptations: paradigms for a new
millennium. J. Archaeol. Res. 9, 287e350.
Ferland, M.A., Roy, P.S., Murray-Wallace, C.V., 1995. Glacial lowstand deposits on the
outer continental shelf of southeastern Australia. Quat. Res. 44, 294e299.
Gamble, C., 1986. The Palaeolithic Settlement of Europe. Cambridge University
Press, Cambridge.
Gautney, J.R., Holliday, T.W., 2013. New calculation of habitable land area during
glacial periods and its implications for Pleistocene hominin population size.
Am. J. Phys. Anthropol. (Suppl. 56), 127 (abstract).
Glennie, K.W., 1998. The desert of Southeast Arabia: a product of quaternary climatic
change. In: Alsharhan, A.S., Glennie, K.W., Whittle, G.L., Kendall, C.G.St.C. (Eds.),
Quaternary Deserts and Climatic Change. Balkema, Rotterdam, pp. 279e291.
Goudie, A., 1992. Environmental Change. Clarendon, Oxford.
Grant, J.W.A., Chapman, C.A., Richardson, K.S., 1992. Defended versus undefended
home range size of carnivores, ungulates and primates. Behav. Ecol. Sociobiol.
31, 149e161.
Haigh, J., Maynard Smith, J., 1972. Population size and protein variation in man.
Genet. Res. 19, 73e89.
Harmon, L.J., Braude, S., 2010. Conservation of small populations: effective population sizes, inbreeding, and the 50/500 rule. In: Braude, S., Low, B.S. (Eds.), An
Introduction to Methods and Models in Ecology, Evolution, and Conservation
Biology. Princeton University Press, Princeton, pp. 125e138.
Harpending, H.C., Sherry, S.T., Rogers, A.R., Stoneking, M., 1993. The genetic structure of ancient human populations. Curr. Anthropol. 34, 483e496.
Harpending, H.C., Batzer, M.A., Gurven, M., Jordell, L.B., Rogers, A.R., Sherry, S.T., 1998.
Genetic traces of ancient demography. Proc. Natl. Acad. Sci. 95, 1961e1967.
Hassan, F.A., 1981. Demographic Archaeology. Academic Press, New York.
Hawks, J.D., 1999. The Evolution of Human Population Size: a Synthesis of Genetic
and Paleoanthropological Data (Ph.D. dissertation). University of Michigan, Ann
Arbor.
Henrich, J., 2001. Cultural transmission and the diffusion of innovations: adoption
dynamics indicate that biased cultural transmission is the predominate force in
behavioral change and much of sociocultural evolution. Am. Anthropol. 103,
992e1013.
Hiscock, P., 1988. Prehistoric Settlement Patterns and Artefact Manufacture at Lawn
Hill, Northwest Queensland (Ph.D. thesis). University of Queensland, Brisbane,
Australia.
Holdaway, S., Porch, N., 1995. Cyclical patterns in the Pleistocene human occupation
of southwest Tasmania. Archaeol. Ocean. 30, 74e82.
Holliday, T.W., 1998. The ecological context of trapping among recent huntergatherers: implications for subsistence in terminal Pleistocene Europe. Curr.
Anthropol. 39, 711e720.
Holliday, T.W., Gautney, J.R., Friedl, L., 2014. Right for the wrong reasons: reflections
on modern human origins in the post-Neanderthal Genome era. Curr.
Anthropol. 55, 696e724.
Ionides, M.G., 1937. The Regimes of the Rivers Tigris and Euphrates. Spon, London,
England.
Jansen, G.F.A., Basnyat, B., 2011. Brain blood flow in Andean and Himalayan highaltitude populations: evidence of different traits for the same environmental
constraint. J. Cereb. Blood Flow Metab. 31, 706e714.
Jolly, D., Harrison, S.P., Damnati, B., Bonnefile, R., 1998. Simulated climate and biomes of Africa during the late quaternary: comparison with pollen and lake
status data. Quat. Sci. Rev. 17, 629e657.
Kuzman, Y.V., 2008. Siberia at the Last Glacial Maximum: environment and
archaeology. J. Archaeol. Res. 16, 163e221.
Kuzmin, Y.V., Keates, S.G., 2005. Dates are not just data: palaeolithic settlement
patterns in Siberia derived from radiocarbon records. Am. Antiq. 70, 773e789.
Lambeck, K., 1996. Shoreline reconstructions for the Persian Gulf since the last
glacial maximum. Earth Planet. Sci. Lett. 142, 43e57.
Lambeck, K., Yokoyama, Y., Purcell, T., 2002. Into and out of the Last Glacial
Maximum: sea level change during Oxygen Isotope Stages 3 and 2. Quat. Sci.
Rev. 21, 343e360.
Lampert, R.J., Hughes, P.J., 1987. The Flinders ranges: a Pleistocene outpost in the
arid zone? Rec. South Aust. Mus. 20, 29e34.
Mangerud, J., Astakhov, V., Svendsen, J., 2002. The extent of the Barents-Kara ice
sheet during the Last Glacial Maximum. Quat. Sci. Rev. 21, 111e119.
Marlowe, F.R., 2005. Hunter-gatherers and human evolution. Evol. Anthropol. 14,
54e67.
Mesoudi, A., Whiten, A., Laland, K.N., 2004. Perspective: is human cultural evolution
Darwinian? Evidence reviewed from the perspective of the origins of species.
Evolution 58, 1e11.
112
J.R. Gautney, T.W. Holliday / Journal of Archaeological Science 58 (2015) 103e112
Palstra, F.P., Fraser, D.J., 2012. Effective/census population size ratio estimation: a
compendium and appraisal. Ecol. Evol. 2, 2357e2365.
Peros, M.C., Munoz, S.E., Gajewski, K., Viau, A.E., 2010. Prehistoric demography of
North America inferred from radiocarbon data. J. Archaeol. Sci. 37, 656e664.
Pitulko, V.V., Nikolsky, P.A., Girya, E.Y., Basilyan, A.E., Tumskoy, V.E., Koulakov, S.A.,
Astakov, S.N., Pavlova, E.Y., Anisimov, M.A., 2004. The Yana RHS site: humans in
the Arctic before the Last Glacial Maximum. Science 303, 52e56.
Powell, A., Shennan, S., Thomas, M.G., 2009. Late Pleistocene demography and the
appearance of modern human behavior. Science 324, 1298e1301.
Rademaker, K., Hodgins, G., Moore, K., Zarillo, S., Miller, C., Bromley, G.R.M.,
Leach, P., Reid, D.A., Alvarez, W.Y., Sandweiss, D.H., 2014. Paleoindian settlement
of the high-altitude Peruvian Andes. Science 346, 466e469.
Ray, N., Adams, J.M., 2001. A GIS-based vegetation map of the world at the Last
Glacial Maximum (25,000-15,000 BP). Internet Archaeol. 11. http://intarch.ac.
uk/jou3rnal/issue11/rayadams_toc.html.
Ray, N., Currat, M., Berthier, P., Excoffier, L., 2005. Recovering the geographic origin
of early modern humans by realistic and spatially explicit simulations. Genome
Res. 15, 1161e1167.
Richerson, P.J., Boyd, R., 1978. A dual inheritance model of the human evolution
process I: basic postulates and a simple model. J. Soc. Biol. Struct. 1, 127e154.
Richerson, P.J., Boyd, R., Bettinger, R.L., 2009. Cultural innovations and demographic
change. Hum. Biol. 81, 211e235.
Rick, J.W., 1987. Dates as data: an examination of the Peruvian pre-ceramic radiocarbon record. Am. Antiq. 52, 55e73.
Rogers, A.R., Harpending, H.C., 1992. Population growth makes waves in the distribution of pairwise genetic differences. Mol. Biol. Evol. 9, 552e569.
Said, R., 1993. The River Nile: Geology, Hydrology and Utilization. Pergamon Press,
Oxford, England.
Shennan, S.J., 2000. Population, culture history and the dynamics of culture change.
Curr. Anthropol. 41, 811e835.
Shennan, S.J., 2002. Genes, Memes and Human History: Darwinian Archaeology and
Cultural Evolution. Thames and Hudson, London.
Shennan, S., Edinborough, K., 2007. Prehistoric population history: from the late
glacial to the late Neolithic in central and northern Europe. J. Archaeol. Sci. 34,
1339e1345.
Smith, M.A., 1989. The case for a resident human population in the central
Australian ranges during glacial aridity. Archaeol. Ocean. 24, 93e105.
Stringer, C., 2012. Lone Survivors: How We Became the only Humans on Earth. Holt,
New York.
Surovell, T.A., Brantingham, P.J., 2007. A note on the use of temporal frequency
distributions in studies of prehistoric demography. J. Archaeol. Sci. 34,
1868e1877.
Surovell, T.A., Byrd Finley, J., Smith, G.M., Brantingham, P.J., Kelly, R., 2009. Correcting temporal frequency distributions for taphonomic bias. J. Archaeol. Sci.
36, 1715e1724.
Svendsen, J.I., Alexanderson, H., Astakhov, V.I., Demidov, I., Dowdeswell, J.A.,
Funder, S., Gataullin, V., Henriksen, M., Hjort, C., Houmark-Nielsen, M.,
! Jakobsson, M., Kjær, K.H., Larsen, E.,
!lfsson, O.,
Hubberten, H.W., Ingo
Lokrantz, H., Lunkka, J.P., Lyså, A., Mangerud, J., Matiouchkov, A., Murray, A.,
€ ller, P., Niessen, F., Nikolskaya, O., Polyak, L., Saarnisto, M., Siegert, C.,
Mo
Sievert, M.J., Spielhagen, R.F., Stein, R., 2004. Late Quaternary ice sheet history of
northern Eurasia. Quat. Sci. Rev. 23, 1229e1271.
Teller, J.T., Glennie, K.W., Lancaster, N., Singhvi, A.K., 2000. Calcareous dunes of the
United Arab Emirates and Noah's Flood: the postglacial reflooding of the Persian (Arabian) Gulf. Quat. Int. 68e71, 297e308.
Thorne, A., Grün, R., Mortimer, G., Spooner, N.A., Simpson, J.J., McCulloch, M.,
Taylor, L., Curnoe, D., 1999. Australia's oldest human remains: age of the Lake
Mungo 3 skeleton. J. Hum. Evol. 36, 591e612.
Van Andel, T.H., Veevers, J.J., 1967. Morphology and Sediments of the Timor Sea.
Department of National Development, Bureau of Mineral Resources, Australia.
Veth, P.M., 1989. Islands in the interior: a model for the colonisation of Australia's
arid zone. Archaeol. Ocean. 24, 81e92.
Weiss, K.M., 1984. On the number of member of the genus Homo who have ever
lived, and some evolutionary implications. Hum. Biol. 56, 637e649.
Williams, A.N., 2012. The use of summed radiocarbon probability distributions in
archaeology: a review of methods. J. Archaeol. Sci. 39, 578e589.
Williams, A.N., 2013. A new population curve for prehistoric Australia. Proc. R. Soc.
280 (1761), 20130486.
Williams, A.N., Ulm, S., Cook, A.R., Langley, M.C., Collard, M., 2013. Human refugia in
Australia during the Last Glacial Maximum and Terminal Pleistocene: a geospatial analysis of the 25-12 Ka Australian archaeological record. J. Archaeol. Sci.
40, 4612e4625.
Williams, M.A.J., Adamson, D., Cock, B., McEvedy, T., 2000. Late Quaternary environments in the White Nile region, Sudan. Glob. Planet. Change 26, 305e316.
Wolpoff, M.H., Hawks, J., Caspari, R., 2000. Multiregional, not multiple origins.
J. Phys. Anthropol. 112, 129e136.
Wright, S., 1931. Evolution in Mendelian populations. Genetics 16, 97e159.
Yokoyama, Y., Lambeck, K., De Deckker, P., Johnston, P., Fifield, L.K., 2000. Timing of
the last glacial maximum from observed sea-level minima. Nature 406,
713e716.
Young, A.J., Reeves, J.T., 2002. Human adaptation to high terrestrial altitude. In:
Lounsbury, D.E., Bellamy, R.F., Zajtchuk, R. (Eds.), Medical Aspects of Harsh
Environments. Office of the Surgeon General, Borden Institute, Washington,
D.C, pp. 647e691.
Zhivotovsky, L.A., Rosenberg, N.A., Feldman, M.W., 2003. Features of evolution and
expansion of modern humans, inferred from genome-wide microsatellite
markers. Am. J. Hum. Genet. 72 (5), 1171e1186.