Modeling foraging ranges and spatial organization of Late Pleistocene
hunteregatherers in the southern Levant eA least-cost GIS approach
Brian F. Byrd
, Andrew N. Garrard
, Paul Brandy
Far Western Anthropological Research Group, 2727 Del Rio Place, Suite A, Davis, CA, 95618, USA
Institute of Archaeology, University College London, 31e34 Gordon Square, London WC1H 0PY, United Kingdom
Available online 14 August 2015
Foraging range modeling
This study takes a regional approach to understanding the nature of Near Eastern hunteregatherer
spatial organization near the height of the Last Glacial Maximum, circa 21,000 calibrated years ago. To do
so, we reconstructed the paleogeography and paleovegetation and then employed least-cost GIS analysis
to model foraging ranges and potential annual territorial extent associated with a selection of excavated
and dated sites throughout the southern Levant. Settlement trends in the region as a whole are explored
ﬁrst, followed by a case study of annual settlement scenarios in the arid Azraq Basin on the eastern edge
of the Levant, focusing on its distinctive large aggregation sites.
The results of the study reveal that potential maximum daily foraging ranges as well as habitats and
habitat zone heterogeneity within these foraging ranges differed greatly across the region. Due to
variance in potential plant and animal productivity, settlement patterns undoubtedly differed signiﬁ-
cantly across the southern Levant particularly with respect to the number of moves per year, the
importance of fusioneﬁssion strategies, the seasonality of relocation tactics, and the importance of group
territoriality. These variances in annual settlement options and emerging patterns within the southern
Levant at the height of the Last Glacial Maximum provide baseline conditions for understanding di-
vergences in adaptive trajectories within the wider region.
©2015 Elsevier Ltd and INQUA. All rights reserved.
This study employs least-cost GIS analysis to model foraging
ranges and the potential annual territorial extent of Near Eastern
hunteregatherers around the height of the Last Glacial Maximum,
circa 21,000 calibrated years ago (cal BP). After a discussion of how
the paleogeography and paleovegetation were reconstructed, this
modeling exercise consists of two parts. Initially, we take a
regional approach to the nature of spatial organization associated
with a selection of excavated sites dating to the Late Glacial
Maximum throughout the Southern Levant. In doing so, we sub-
divide the region into three areas and examine differences in
foraging ranges and the nature of associated habitats between
them. Predicted settlement pattern implications based on these
differences are also highlighted. In the second part of the study, we
use these foraging model insights as a starting point to explore in
more depth potential annual settlement scenarios in a single lo-
cality. This discussion centers on the arid Azraq Basin on the
eastern edge of the Levant, and its distinctive large aggregation
sites. In doing so, and to gain a nuanced understanding,
more archaeological variables are employed including site attri-
butes such as size, thickness, associated material culture, and ev-
idence of regional interaction including trade and exchange. The
results of this Azraq Basin case study highlight the potential of
least-cost modeling to provide insights into territoriality, travel
and trade corridors, and the orientation of annual settlement
Overall, we aim to make three main points. First, potential
maximum daily foraging ranges varied signiﬁcantly within the re-
gion. Second, habitats and habitat zone heterogeneity within these
foraging ranges also differed greatly across the region. Third, due to
variance in potential plant and animal productivity we predict that
organizational strategies differed signiﬁcantly across the southern
Levant particularly with respect to the number of moves per year,
E-mail addresses: email@example.com (B.F. Byrd), firstname.lastname@example.org
(A.N. Garrard), Paul@farwestern.com (P. Brandy).
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/quaint
1040-6182/©2015 Elsevier Ltd and INQUA. All rights reserved.
Quaternary International 396 (2016) 62e78
the importance of fusioneﬁssion strategies, the seasonality of
relocation tactics, and the importance of group territoriality. Finally
we conclude with a brief consideration of what these spatial pat-
terns within the southern Levant imply for the long-term trajectory
of adaptations that led to the emergence of sedentism and food
This study is focused on the time period from 24,000 to
18,000 cal BP, which temporally straddles the Last Glacial
Maximum of circa 21,000 calibrated years ago. This time span
is considered to be culturally transitional in the Levant, as it
encompasses the end of the late Upper Paleolithic (circa
30,000e21,300 cal BP) and start of the Epipaleolithic (circa 24,000/
21,300e11,600 cal BP; Garrard and Byrd, 2013; Goring-Morris and
Belfer-Cohen, 2003; Goring-Morris et al., 2009). The Upper Paleo-
lithic and Epipaleolithic are distinguished by a number of attri-
butes; the most fundamental is the emergence and dominance of
ﬂaked stone backed bladelets, which were used as small hunting
armatures in composite tools. It should be noted that the details of
the nature and timing of this transition in technology (and asso-
ciated settlement and subsistence strategies) is subject to consid-
erable discussion and debate (Goring-Morris and Belfer-Cohen,
The overall goal of this study is to gain further insight into
long-term trends in Epipaleolithic adaptations in order to
enhance our understanding of the causal factors leading to the
Natuﬁan in the Late Epipaleolthic (circa 14,600e11, 6 0 0 c al B P ) .
The Natuﬁan is widely recognized as the region's ﬁrst complex
sedentary hunteregatherers, and the Natuﬁan laid the founda-
tion for the Levantine early Neolithic, the world's earliest farmers
near the start of the Holocene circa 11,600 cal BP (Byrd, 2005;
Belfer-Cohen and Goring-Morris, 2011; Bar-Yosef and Valla,
As such, there is considerable interest in the underlying condi-
tions that led to the emergence of the Natuﬁan. For some time
consensus was that its Epipaleolithic precursors were uniformly
mobile hunteregatherers living in small groups. With increased
ﬁeldwork in the eastern Levant there is much greater appreciation
for the complexity of regional variation in artifact assemblages and
site characteristics of pre-Natuﬁan Epipaleolithic sites, and research
in the Azraq Basin, particularly on the large aggregation sites, has
been a seminal aspect of these developments (Garrard and Byrd,
1992, 2013; Maher et al., 2012a; Muheisen, 1988; Richter et al.,
2010, 2013). There is also a wide-spread recognition that most
traits of the Natuﬁan (architecture, mortars and pestles, on-site
burials, grave goods) had their origins earlier in the Epipaleolithic
(Maher et al., 2012b).
These insights highlight the need to acquire better under-
standing of regional patterns in settlement and subsistence
strategies prior to the Natuﬁan. In the Levant, such Epipaleolithic
studies have been limited, and notably include the early site
catchment work of Vita Finzi and Higgs (1970),reconstructing
seasonal settlement shifts in the Hisma of southern Jordan
(Henry, 1995:426e437), and hypothesizing potential spatial
orientation and extent of annual ranges focusing either on the
Mediterranean coast (Goring-Morris, 2009:85e86) or the Levant
as a whole (Bar-Yosef and Belfer-Cohen, 1989:451; Goring-Morris
et al., 2009). As an initial step, this study focusses on rigorously
developing background data to facilitate gaining new perspec-
tives into regional variation, trends in background conditions,
and potential settlement strategies at the start of the Epi-
paleolithic in the southern Levant (note that various terms have
been applied to this period of time; see Garrard and Byrd, 2013 for
Modeling environmental conditions at the height of the Last
Glacial Maximum is a necessary ﬁrst step in assessing how these
background conditions may have constrained and conditioned
hunteregatherer choices on how to distribute and organize
themselves across the landscape. A variety of ethnographic, eth-
noarchaeological, and experimental studies have demonstrated
that hunteregatherer settlement and subsistence strategies are
patterned in predictable ways with respect to a variety of factors
including terrain, environmental productivity, effective tempera-
ture, degree of resource homogeneity, and seasonality of resource
availability (Binford, 2001; Grove, 2009, 2010; Kelly, 1983,
2013:77e113). For example, Binford (1980) was one of the ﬁrst to
highlight the utility of effective temperature and primary produc-
tivity in assessing broad trends in both the relative contribution of
plants and animals in the diet and the degree of hunteregatherer
Optimal foraging theory has been important in generating
foraging efﬁciency expectations regarding subsistence choices
given variation in resource availability and density (Zeenah, 2004;
Kelly, 2013). Its application has provided insight into the organi-
zation of movement and foraging decisions in relationship to the
environment, while taking into account resource variability.
Generally, greater annual residential camp movement is antici-
pated with high resource homogeneity and year-round availability,
while fewer residential relocations (and increased use of a collector
strategy to acquire resources) are expected when resources are
patchy or highly seasonal. Central place foraging models, in
particular, have been a useful tool to gain insight into the factors
that condition resource acquisition and round-trip transport from
base camps (Orians and Pearson, 1979; Bettinger et al., 1997;
Zeenah, 2004; Morgan, 2009). The basic premise is that hunter-
egatherers aim to maximize their foraging efforts, and such models
provide optimal solutions to efforts to look for, acquire, and
transport resources. Notably, travel time becomes a key factor in
understanding foraging decisions tied to central places, high-
lighting the need to model the extent and nature of foraging areas
around base camps.
These studies provide important insight into how hunter-
egatherers respond in predictable ways with respect to various
environmental factors affecting resource availability and pro-
ductivity, and the need to take into account the effects of travel
time and transport on central place foraging. Overall, such factors
condition the nature and effectiveness of daily foraging (i.e., re-
turn rates), the degree of reliance on individual logistical forays
and their extent, the role of resource caching and ﬁeld processing,
and provide baseline information to make predictions regarding
the nature, frequency, and spatial extent of residential move-
ments within the annual cycle. The approach taken here is to
reconstruct the paleogeography and paleovegetation at the
height of the Last Glacial Maximum, model the extent and nature
of potential foraging territory around key sites, examine
patterning within the southern Levant, and then discuss the
3. Approach and methods
We focused attention on GIS modeling of prehistoric occupa-
tion circa 24,000e18,000 cal BP. To do so, we plotted in GIS the
locations of 25 prominent sites in the southern Levant that date to
this time period (Fig. 1;Ta bl e 1), and then developed GIS data sets
to facilitate our analysis. This included sites that have been
variously described as having either Upper Paleolithic or
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 63
Epipaleolithic assemblage traits. Our goal was not to delve into
these differences but rather to use these sites as sample data
points across the region to explore how potential background
conditions may have varied within the southern Levant. In order
to have a manageable data set, sites from the Negev and Sinai
were not included.
Fig. 1. Overview of modern Levant showing Last Glacial Maximum archaeological sites used in the study.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7864
Since the paleogeography was very different than today, we
reconstructed and mapped Last Glacial Maximum (circa
21,000 cal BP) sea level and inland lake level shorelines using, as a
starting point, the sources shown in Table 2. For comparative pur-
poses we also reconstructed paleoshorelines for the start of the
BollingeAllerod (circa 15,000 years ago) and the start of the Ho-
locene (some 11,600 years ago). We also deﬁned drainage catch-
ments and mapped stream channels within GIS (Greenbaum et al.,
2006; Lehner et al., 20 08).
For the building of a paleovegetation map we turned to the two
earlier reconstructions for the Late Pleistocene in the Levant, which
were by van Zeist and Bottema (1991:107e114) and Hillman (1996;
and in Moore et al., 2000:73e84). These were based on the inter-
pretation of a number of pollen cores, particularly from the Hula/
Huleh Valley in northern Israel (Baruch and Bottema, 1999; Tsudaka
in van Zeist and Bottema, 1991:104e105) and the Ghab Valley in
western Syria (Niklewski and van Zeist, 1970; Yasuda et al., 2000).
In reconstructing the palaeovegetation, the researchers took into
account: (1) the distribution and nature of present plant commu-
nities in the area; (2) the physiographic features, which include the
highland ranges which lie on either side of the Levantine Rift Valley
and run parallel to the Mediterranean coast, and the topography of
the inland plateau; (3) the impact of the prevailing westerly storm
tracks, with possible seasonal monsoonal inﬂuences from the
south; and (4) the current models of changing temperature and
moisture regimes through the Late Pleistocene.
For this publication, we have extrapolated from these earlier
reconstructions and taken into consideration more recent palae-
oenvironmental research from across the region (for summaries see
Enzel et al., 2008; Robinson et al., 2006) and particularly from the
Archaeological sites used in study, associated Epipaleolithic industry, and maximum foraging ranges.
Sites Reference Epipaleolithic
range (sq km)
in one day
foraging range (sq km)
in two days
Eastern Area (n ¼11) Ain al-Buhira
Coinman, 1993 not applicable 787 2795 3582
Ayn Qasiyya Richter et al., 2010 Kebaran and
973 3376 4348
Azraq 17 Garrard and Byrd, 2013 Not applicable 920 3324 4244
Jilat 6 Garrard and Byrd, 2013 Nebekian/Nizzanan
1115 3429 4544
Kharaneh IV Maher et al., 2012a Kebaran 1147 3413 4560
Tor al-Tareek (WHS-1065) Neeley et al., 1998 Nebekian 808 2818 3626
Tor Hemar Henry, 1995 Nebekian 588 1946 2534
Tor Sageer (WHNBS 242) Olszewski, 2011 Nebekian 832 2872 3704
Uwaynid 14 &18 Garrard and Byrd, 2013 Nebekian 1074 3465 4539
Yabrud Rust, 1950 Nebekian 773 2415 3188
Yutil al-Hasa (WHS 784) Olszewski et al., 1994 Nebekian 708 2699 3407
Mean (std dev) 884 ±178 2959 ±493 3843 ±661
Central Area (n ¼8) Ein Gev I, IV Bay Yosef, 1970, 1991a Kebaran 269 1659 1928
Madamagh Byrd, 2014 Nebekian 353 2046 2398
Ohalo II Nadel, 2003 not applicable 394 1735 2129
Tabaqat al Buma Banning et al., 1992 Kebaran 432 1203 1635
Urqan e-Rub Hovers and Marder, 1991 Kebaran 210 889 1100
Wadi Hammeh 26 Edwards et al., 1988 Kebaran 241 720 961
Wadi Hammeh 51 &52 Edwards et al., 1996 Kebaran 231 701 931
Wadi Hisban 2 Edwards et al., 1999 Nizzanan
351 1231 1581
Mean (std dev) 310 ±83 1273 ±499 1583 ±552
Western Area (n ¼6) Hayonim Bar-Yosef, 1970 Kebaran 759 1975 2734
Jiita II Hours, 1992 Kebaran 290 827 1117
Kebarah Cave Bar-Yosef, 1970 Kebaran 939 1985 2924
Ksar Akil Hours, 1992 Kebaran 372 762 1135
Moghr el Ahwal Garrard and Yazbeck, 2008 Kebaran 360 1054 1414
Nahel Hadera V Saxon et al., 1978 Kebaran 1034 2019 3053
Mean (std dev) 625 ±326 1437 ±617 2063 ±933
Nizzanan, closely related to Nebekian but dating later in the early Epipaleolithic.
Ancient shoreline elevations (meters above/below modern mean sea level) used for paleoenvironmental modeling.
Setting 21,000 cal BP 15,000 cal BP 11,500 cal BP Reference
Mediterranean and Red Sea 120 88 50 Fleming et al., 1998
Lake Lisan (Dead Sea), Jordan Valley 200 285 400 Bartov et al., 2002, 2003; Hazan et al., 2005;
Robinson et al., 2006
Lake Beit Shean, Jordan Valley na 250 None Bartov et al., 2002; Hazan et al., 2005 (Fig. 1)
Lake Kinneret (Tiberias/Galilee), Jordan Valley 200 210 210 Belitzky, 2002; Hazan et al., 2005
Lake Huleh (Hula), Upper Jordan Valley 73 Uncertain;
Ashkenazi, 2004; Feibel et al., 2009;
Lake Hasa, west-central Jordan 810 Uncertain;
Schuldenrein, 1998; Schuldenrein and Clark, 1994
Lake Azraq, northeastern Jordan 505 505 505 Garrard and Byrd, 2013; Jones and Richter, 2011
na: not applicable, as subsumed by Lake Lisan.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 65
results of isotopic analyses of cave speleothems (Bar-Matthews
et al., 1999, 2003). We have also drawn on palaeoenvironmental
reconstructions from personal ﬁeld research undertaken in eastern
Jordan and northwest Lebanon (Garrard and Yazbeck, 2008; Hunt
and Garrard, 2013).
On the basis of the gradient seen in present vegetation com-
munities across the region which relates closely to physiography
and rainfall (Hillman in Moore et al., 200 0:49e73;Zohary, 1973; see
also second column in Table 3), we have suggested a similar
gradient in the Last Glacial Maximum, although being very aware
that the detailed composition of the plant communities will have
changed. The geographical distribution of these vegetation zones
will also have altered in response to lower temperatures and lower
precipitation levels across the region, although the latter is partly
counterbalanced by the impact of reduced evaporation levels on
effective moisture regimes. In drawing the speculative boundaries
for each vegetation zone, we have made use of the “contour”lines
on the current isohyet map, but altered their values to account for
the factors outlined above (see third column in Table 3). It is
appreciated that there was much climatic variation within the Last
Glacial Maximum and what is shown in the map will only represent
one part of the spectrum. It is also understood that a number of
critical factors beyond physiography and rainfall are signiﬁcant in
the location of plant communities, but it is hoped that this will give
an approximation to the palaeovegetation of the time.
Next, we created cost surfaces to model foraging ranges and
travel routes. The cost surface is a grid whose values indicate the
effort required to traverse each cell which uses terrain to estimate
effort. This more accurately reﬂects ground conditions than the
Euclidean distance (van Leusen, 2002; Wood and Wood, 2006;
Howey, 2007; Morgan, 2009; White and Surface-Evans, 2012).
This cost surface can be subjected to further analysis to, for
example, estimate total travel time or optimized travel routes
(Kantner, 2012). Initially, this entailed creating a digital elevation
model (DEM) of the region's terrain using data from 90-m grid cells.
This DEM served as the basis for generating both a caloric and a
time-cost surface. Then we used cumulative time-cost distance to
estimate foraging ranges and generated least-cost paths on the
caloric surface to estimate travel routes.
For foraging ranges, we modeled two concentric polygons
around each site deﬁned using a least-cost GIS analysis that em-
ploys travel time as currency. The ﬁrst represents the distance an
individual walks in four hours from the site. We refer to this as the
maximum one-day foraging range, since it allows one to turn
around and return to the camp by the end of the day. This
maximum one-day foraging range is effectively the area within
which an individual could access any point from the site as a
starting point within four hours of walking (and assuming a four
hour return trip). Then we modeled what we refer to as the
maximum two-day foraging rangedthis represents an additional
area within which an individual could walk within eight hours from
a site. Both values are presented in square kilometers (sq km) and in
aggregate represent the maximum two-day foraging catchment of
a site. The maximum one-day foraging extent is particularly effec-
tive in characterizing the area of daily plant resource procurement,
and the area within which most members of the camp conduct
daily foraging (Kelly, 2013). Patch searching and intensive foraging
efforts, of course, would certainly reduce effective daily foraging. In
contrast, the maximum two-day foraging area effectively encom-
passes nearby settings where an individual would use encounter
strategies, such as hunting game, returning to snares, and similar
activities that regularly take place (Kelly, 1983). It should also be
noted that we have not chosen to present mean foraging radius for
each of these foraging area values. This is because least-cost GIS
modeling of travel distances and the resulting foraging area are
empirically based, the modeling can be altered by changing as-
sumptions, and this is inherently more accurate than general ap-
proximations of foraging radii. In fact, the quest for a uniform,
broadly applicable mean radius value also requires that one gloss
over variations in foraging range based on topography around the
starting point and other factors (of course mean radius can be
readily calculated for any of these sq km values).
Studies of energetic travel costs, travel times, and travel corri-
dors are based on the assumption that prehistoric site locations
were often located in areas with relatively easy access, and that
least-cost (in terms of energy) travel routes were preferred (van
Luesen, 2002; Wood and Wood, 2006; Byrd et al., 2008; Morgan,
2008). To model energetic travel costs, the calories per second
(cost) required to traverse each 295--295-foot (90--90-m)
elevation grid cell were calculated for the entire study area based
on a formula for metabolic rate for walking on a slope (Pandolf
et al., 1977):
Last Glacial Maximum (circa 21,000 cal BP) paleo-vegetation reconstruction in relationship to modern rainfall isohyets.
Vegetation zone Modern isohyet range for
each vegetation zone
Modern Isohyte “contour”
lines used to model boundaries
of 21,000 cal BP vegetation zones
Desert <100 0e75
Dry steppe 100e150 75e175
Moist steppe 150e200 175e300
Terebinth-almond woodland steppe 200e300 300e400
Oak-Rosaceae park woodland steppe 300e400 400e500
Dense deciduous Oak-Rosaceae woodland 400e600 500e700
Montane and Eu-Mediterranean forest >600 >700
Coastal sand dunes na
Based on Hillman in Moore et al. (2000:49e73).
Using data derived from Bar-Matthews et al. (1999, 2003); Hillman (1996);Hillman in Moore et al. (2000:73e84);Hunt and Garrard (2013:114e116); van Zeist and
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7866
For this model, we assume a 68-kg (150-pound) person carrying
no load, and a terrain coefﬁcient of 1. Per van Luesen (2002) we
assumed that people move across the landscape most efﬁciently
at 5% slope. Therefore, the slope factor (G) was adjusted by sub-
tracting 5% so estimate downhill metabolic and adding 5% to esti-
mate uphill metabolic rate. The overall caloric cost surface is the
sum of the downhill and uphill estimate. Therefore, we are
attempting to ﬁnd the least-cost path that accounts for both di-
rections of travel.
To estimate the velocity we started with the hiker function
(Tobler, 1993; Whitley and Hicks, 2001):
eeA real number derived from the exponential function of the
slope of the tangent line, commonly deﬁned as the base of the
natural logarithm that is a mathematical constantdalso called
However, since the Pandolf et al. (1977) model was designed to
describe metabolic rate for a speciﬁc range of speeds, we set a lower
limit on velocity. The metabolic rate was then converted to kilo-
calories per meter and multiplied by the velocity to derive the ki-
localories per second, which resulted in the estimation of cost-
surface with a continuous set of values ranging between about
0.10 kCal/s and 0.24 kCal/s. In general, the metabolic cost functions
as a transformation of surface slope, where a linear increase in
slope results in a much greater increase in metabolic cost. This cost-
surface was used for two calculations:
Cost distance to sea, or the cumulative metabolic travel cost
from each grid cell to the edge of the speciﬁc sections of coast
(northern Mediterranean, southern Mediterranean, Red Sea).
Least-cost path, or the shortest path between a site and the coast
using the cost distance surface. This model results in a line or
path from the site to the coast which was considered a potential
We recognize that the paleoreconstructions and resulting
foraging ranges are estimates; that they represent only one point in
time; and that environmental conditions certainly varied during
the time frame of discussion (circa 24,000e18,000 cal BP). The
least-cost travel routes could be further improved by estimating
relative travel rates per vegetation type (allowing nto vary in the
Pandolf equation). It is also important to keep in mind that
empirically we have calculated travel paths, assuming effort was
the only factor. Actual travelways, of course, are corridors with
breadth, and other factors invariably play a role in determining
routes. It is also acknowledged that we have examined only a
sample of the sites in the region assigned to this time frame, sub-
jectively selecting those from a variety of settings to provide suf-
ﬁcient geographic coverage. The objective was not to deﬁnitively
characterize precise moments in time and the full nuanced nature
of regional settlement, but rather to identify potential trends and
variations that provide a basis for generating hypotheses regarding
organizational options to be tested in the future.
4. Regional Last Glacial Maximum foraging results
4.1. Foraging extent
Fig. 2 graphically depicts the GIS analysis results of cumulative
travel time for one-day and two-day maximum daily foraging
ranges from the selected 25 sites in the southern Levant (see
Table 1 for data from each site). It should be noted that in locations
where several sites are situated, such as in the Eastern Area,
there is considerable overlap in foraging ranges. Overall,
the mean maximum one-day foraging range is 638 sq km,
increasing another 2054 sq km when the two-day foraging ranges
are added for a total 2-day maximum foraging catchment of
2692 sq km.
If we subdivide the sample by region within the southern Levant
edistinguishing sites west of the Levantine Rift/Jordan Valley
system from those within it, and from those east of the Levantine
Rift ethere are substantial differences in foraging ranges (Fig. 3;
Table 1). These differences are statistically signiﬁcant at one-day
maximum foraging ranges (analysis of variance [ANOVA]
p<0.000, df ¼2, F ¼19.041) and two-day maximum foraging
ranges (ANOVA p <0.000, df ¼2, F ¼27.145). However, the dif-
ferences in maximum one-day foraging area between regions have
the greatest implication for hunteregatherer adaptation. Not sur-
prisingly, the Central Area ethose sites within the Levantine Rift
system ehave the smallest foraging range (with a one-day mean
foraging area of 310 sq km, and a total two-day mean foraging area
of 1583 sq km), while those in the Eastern Area have the largest
foraging ranges (with a one-day mean foraging area of 884 sq km,
and a total two-day mean foraging area of 3843 sq km). West of the
Levantine Rift, intermediate values exist (with a one-day mean
foraging area of 625 sq km, and a total two-day mean foraging area
of 2063 sq km); moreover, this sample also has the greatest vari-
ance in results (see Table 1). These differences between regions are
tied to topography especially where sites are positioned in rela-
tionship to bodies of water. Notably, the much smaller foraging
areas in the Central Area are due to their being bounded by the
Jordan Valley freshwater paleolakes (notably Lake Lisan, but also
Lake Kinneret) and the Levantine Rift Valley's rugged and steeply
4.2. Foraging habitat
Glacial maximum vegetation associations and particularly the
distribution of forested and woodland areas were much more
restricted than today, with forested areas concentrated in
the western portion of the region, and distinctive park woodlands
present and pervasive in the Central Area (Fig. 4). Therefore
it's not surprising that the range of paleovegetation habitats
available within a foraging area differs by region in the southern
The extent of paleohabitats within the total two-day maximum
foraging range for the study sites is depicted in Fig. 5. Clear dif-
ferences can be discerned between regions. Foraging catchment
habitat heterogeneity (measured by taking the mean of all values)
is also signiﬁcantly different by region egreatest in the Central
Area, and least in the Western Area (Fig. 6). Moreover, this rela-
tionship remains the same when either one or two-day catchments
Fig. 7 summarizes these differences by presenting mean pale-
ohabitat representation by region within the maximum one-day
foraging range. The types of habitats and the relative dominance
of the two most common habitats vary considerably by region.
Notably, the Western Area is dominated by dense deciduous Oak-
Rosaceae Woodland and Montane and Eu-Mediterranean forest,
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 67
and together they represent 91% of the one-day foraging range
(with sand dunes comprising the only other habitat zone). In
contrast, the Central Area has the most diverse and more evenly
distributed range of habitats including woodlands, parklands, and
steppic areas. The two most common habitats are Oak-Rosaceae
park woodland steppe and Terebinth-almond woodland and
together they comprise 78% of the one-day foraging range. The
Eastern Area has intermediate values with respect to range of
habitats and the lowest percentage of the two most common
paleovegetation zones eMoist and Dry steppe (71%). However, the
next most common habitat is the Desert zone.
4.3. Regional summary and implications
In summarizing regional patterns within the southern Levant,
clear differences are evident in the size of daily foraging areas (see
Table 1). If one were to assume habitat productivity was the same
across the region, then larger daily foraging ranges means greater
daily access to resources, and fewer residential relocations
throughout the year as a result of declining return rates. However,
paleohabitat modeling reveals that both habitat heterogeneity and
potential productivity also vary greatly across the southern Levant
(see Fig. 6). Given the strong differences in the size of daily foraging
Fig. 2. Maximum one and two-day least-cost foraging areas for all sites by region using reconstructed sea and lake levels (circa 21,000 cal BP).
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7868
ranges and potential plant and animal productivity within these
areas shown in this analysis, we predict that Last Glacial Maximum
adaptive patterns should have been signiﬁcantly different across
the southern Levant. Such variation could have included the rela-
tive reliance on fusioneﬁssion strategies, the seasonality of relo-
cation tactics, the number of moves per year, and the importance of
The Central Area (falling within the Levantine Rift Valley) has
the smallest one-day foraging ranges, yet the greatest habitat
heterogeneity and undoubtedly the greatest potential productivity.
Hunteregatherer theoretical research would predict that owing to
the patchiness of resources, the inhabitants of this setting would
have been more likely to follow the dictums of central place theory
(Orians and Person, 1979; Zeenah, 2004; Morgan, 2009). This
would entail placing considerable reliance on logistical forays
(particularly to acquire resources within nearby productive habi-
tats) and/or employing seasonally based tactics to move residences
from one nearby habitat to another during an annual round (Kelly,
In contrast, the Western Area is characterized by larger one-day
foraging ranges (on average two times that of the Central Area) but
also by having the lowest habitat heterogeneity. In addition, the
densely forested vegetation zones that dominate these settings are
likely to have had much lower potential productivity for hunter-
egatherers since such habitats typically have lower densities of
seeds, nuts, and larger game. Therefore, it is most likely that resi-
dential mobility within the Western Area was higher than within
the Central Area, and that moves were typically short in distance
and within the same habitat. Seasonal relocations into different
habitats would have been more costly and less likely as they would
have required much longer moves.
Finally, the Eastern Area has much larger daily foraging ranges
(the one-day extent almost three times that of the Central Area, and
the two-day maximum extent almost two times that of the West-
ern Area). These open spaces had intermediate values for habitat
heterogeneity, but desert and dry steppe were most frequent, both
of which had lower hunteregatherer productivity than Central
Area settings. It is also likely that water sources played an impor-
tant role in tethering residents. This may have reduced the number
of residential moves per year eas groups were willing to tolerate
declining daily return rates for foraging eand increased the
incentive for logistical forays. At the same time, the larger area that
could be covered within the two-day maximum foraging range
beneﬁted logistical forays and increased the probability of
encountering large game, which more often was present in groups/
herds than in the Central and Western Areas.
5. The Azraq Basin ea southeastern Levantine example
5.1. Context and broader interaction
The Azraq Basin in the eastern portion of the southern Levant is
an ideal setting to explore the potential complexity of factors at
play in Late Glacial Maximum adaptations, particularly on account
of its rich archaeological record of settlement and regional inter-
action (see Fig. 4). This internal drainage basin is noteworthy in that
it has the two largest and most impressive early Epipaleolithic sites
in the southern Levant eKharaneh IV and Jilat 6 eand they greatly
contrast from the typical small size of all other sites of this time
period (Byrd and Garrard, 1990; Garrard and Byrd, 1992). They are
also characterized by thick middens, features, and structures eall
indicative of substantial and sustained occupation eand are often
referred to as aggregation sites focused on the exploitation of ga-
zelle herds (Muheisen, 1988; Martin et al., 2010; Maher et al.,
2012a; Garrard and Byrd, 2013; Richter et al., 2013).
Moreover, these two sites (and the basin as a whole) have quite
different early Epipaleolithic microlithic ﬂaked stone tool
manufacturing traditions. At Kharaneh IV these microliths have
been classiﬁed as Kebaran (Muheisen, 1988; Maher et al., 2012a;
Richter et al., 2013), while at Jilat 6 they are considered Nebekian
and Nizzanan, the latter considered to be closely related but
occurring slightly later in time (Garrard and Byrd, 2013). It is widely
perceived that these broadly contemporaneous lithic industries
(Kebaran vs. Nebekian and Nizzanan), which are mainly deﬁned by
backed bladelet manufacturing techniques and the size and shape
of the resulting microliths, represent either distinctive ethnic/social
groups or tool manufacturing traditions of long-standing duration
(Bar-Yosef, 1991b; Henry, 1995; Goring-Morris and Belfer-Cohen,
1997; Stutz and Estabrook, 2004; Goring-Morris et al., 2009; Ols-
zewski, 2011). This perception is reinforced by the fact that Kebaran
sites are documented mainly in the more mesic Western and
Central Areas of the Southern Levant, while Nebekian sites appear
to be primarily if not exclusively concentrated in the more arid
Eastern Area (see Fig. 4). In fact, the Azraq Basin is presently the
only place where both traditions are well documented (Garrard and
Byrd, 2013; Richter et al., 2010, 2011).
Given that these two sites may well have had some temporal
overlap, the nature of broader regional interaction and territoriality
also need to be considered, especially if these sites were inhabited
by distinctive hunteregatherer groups. It should be noted that it is
uncertain if speciﬁc occupation events at Kharaneh IV and Jilat 6
were absolutely contemporaneous owing to the precision of
radiocarbon dating results (Garrard and Byrd, 2013; Richter et al.,
2013). Yet it is probable that at some point during the early Epi-
paleolithic both Kebaran and Nebekian groups occupied the basin
at the same time, given that both traditions are also documented at
other sites in the Azraq Basin. As such, issues of territoriality and
the nature of regional interaction may have also played a role in
shaping adaptive patterns.
To understand the nature of broader regional interaction by
groups inhabiting the Azraq Basin, one line of evidence that can be
Fig. 3. Graph of mean Least-Cost one-day and two-day foraging range by region.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 69
examined is the relative degree of reliance on Red Sea versus
Mediterranean shells at sites in the region. Trends in the move-
ment of exotic material of this sort (regardless of precisely how
such long-distance goods were acquired) provide insight into the
nature, extent, and spatial organization of larger regional social
networks (Whallon, 2006; Pearce, 2014; Collar et al., 2015). To
generate some expectations regarding regional orientation of
marine shell bead procurement, we modeled least-cost travel
paths and cost trends from Jilat 6 and closely adjacent sites in Wadi
Jilat at three points in time during the Terminal Pleistocene (Fig. 8).
At the Last Glacial Maximum, the shortest and least-expensive
path for direct procurement was a northern route to the Medi-
terranean using the gap between Lake Lisan and Lake Kinneret
(near Ohalo II). This route is signiﬁcantly less costly than direct
procurement from the Red Sea or via a southern route to the
Over time, the two routes to the Mediterranean change, most
notably with an even shorter northern path; while the Red Sea
route remains effectively the same. Therefore, from a cost
perspective, one would predict that in the Azraq Basin during the
early Epipaleolithic direct marine shell procurement or down the
line trading would have been Mediterranean-oriented rather than
Red Sea-oriented. Although suitable marine shell source data from
the Basin are limited, they suggest a more complicated scenario.
Overall, in the Jilat area, very small samples from Nebekian sites
suggest a fairly even representation of Red Sea and Mediterranean
Sea species (Reese, 1991; Garrard et al., 1994). This is consistent
with results from Nebekian sites further south in the eastern Levant
(Reese, 1995, 2014).
In contrast, data from Kharaneh IV presented by Richter et al.
(2011:103e107) indicate that during the initial phase of site occu-
pation, Mediterranean species outnumber Red Sea species 11 to one
(see also Allcock, 2009). In the subsequent phase, species from both
locations are almost evenly represented. This pattern of Mediter-
ranean species initially dominating the Kebaran occupation of
Kharaneh IV lends credence to the expectation of different groups
Fig. 4. Last Glacial Maximum paleovegetation reconstruction in southern Levant in relationship to study sites and the Azraq Basin (circa 21,000 cal BP).
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7870
occupying the basin, with at least the initial occupation of Khar-
aneh IV represented by a group whose regional orientation inter-
action may have been more east-west focused; over time this
orientation shifts to north-south trade and exchangeelevel inter-
action more typical for Eastern Area sites. In contrast, hunter-
egatherers occupying the Jilat area had interaction consistent with
data from other Nebekian sites further to the south. These results
are indicative of different regional social networks/interaction
spheres and raise the likelihood that some forms of territoriality
may have existed (Whallon, 2006).
5.2. Daily foraging and annual settlement organization
With respect to a consideration of foraging ranges, both Jilat 6
and Kharaneh IV are located not in the center of the Azraq Basin
where the most extensive fresh water and riparian settings existed,
but rather in the western portion of the basin. With extensive daily
foraging ranges, the two sites have considerable one-day maximum
foraging overlap (Fig. 9). Two-day foraging ranges also encompass
almost the entire one-day foraging range of each other. The major
distinction is that the two-day maximum foraging range of Khar-
aneh IV extends to the springs and lake in the center of the basin,
while the two-day range for Jilat 6 does not, ending near the lower
Wadi Uwaynid (where other Nebekian sites have been
The one-day foraging habitat of each site is very homogenous,
consisting almost entirely of dry steppe. The two-day foraging
range is slightly more varied, but mostly with the inclusion of the
desert zone and only a small amount of the moist steppe. Although
the archaeological data from both sites are representative of
intensive occupation near local fresh water sources, it is unlikely
that either were occupied throughout each year. Instead, occupa-
tion undoubtedly took place for only a portion of the year, likely
representing an aggregation settlement and with a heavy emphasis
on targeting locally available gazelle herds in the dry steppe
(Garrard and Byrd, 1992; Martin et al., 2010).
Fig. 5. Reconstructed Last Glacial Maximum paleovegetation habitats within two-day maximum foraging range of study sites.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 71
So what was the nature of the rest of the annual settlement
system for the inhabitants of these two large early Epipaleolithic
sites? We predict the rest of the annual settlement system took
place either near or beyond their maximum two-day foraging range
in order to capitalize on other seasonally available resources. It is
also probable that other facets of the annual system were repre-
sented by dispersal into smaller residential groups, consistent with
the small size of sites documented elsewhere in the southern
The two most likely scenarios involve capitalizing on different
resource sets, and entail either repositioning to the east or to the
west (Fig. 10). Foraging further north or south within the dry steppe
is considered unlikely since seasonal constraints remain largely the
same, and on the south they would quickly be impinging on the
foraging range of early Epipaleolithic sites near Lake Hasa.
Dispersing and relocating eastward to the oasis at the center of
the basin in the desert zone is certainly a viable strategy since it
positions these foragers adjacent to resources associated with a
freshwater riparian habitat including migratory birds, marsh
plants, and large game. Although this locality would allow occu-
pation during the drier summer months owing to a perennial water
supply, many of the key resources here would have been most
plentiful during other seasons. However, this relocation scenario
entails only a modest 8 to 10 h of travel time, and is consistent with
the presence of a number of small early Epipaleolithic sites along
the marsh margins.
In contrast, shifting to the west into the wetter woodland steppe
or park woodlands would have afforded Jilat 6 and Kharaneh IV
occupants a wider and more varied range of resources, particularly
plants. Cereals and legumes would have been available in late-
spring and early summer (after any such resources may have
been collectable in the dry steppe), while nuts, such as pistachios
and almonds, were available in the fall. Minimum travel time es-
timates to the woodland steppe are around 11e12 h, and 13e15 h to
reach the park woodland (see Fig. 10).
Several relocation options were available to access woodland
steppe during the early Epipaleolithic including the uplands in the
Wadi el Hidan directly to the west, the Wadi Mujib uplands to the
southwest, and the Wadi Zarka uplands to the northwest. Each of
these drainage catchments are large and well-watered, and well-
suited for varied occupation events. In contrast, park woodland
habitat was much more restricted in extent. It was not present
directly to the west or to the southwest. Instead, the nearest
woodland was to the northwest along the edge of the Wadi Zarka
drainage basin or even further to the northwest near the uplands of
the Yarmouk River.
Territorial boundary issues may well have played a role in
constraining annual settlement at Jilat 6 and Kharaneh IV. If terri-
toriality existed and was deﬁned simply by the midpoint of travel
Fig. 6. Graph of mean one-day and two-day maximum foraging extent habitat het-
erogeneity by region.
Fig. 7. Graph of mean habitat relative frequency within maximum one-day foraging range for Last Glacial Maximum sites by region.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7872
distance, then the inhabitants of Jilat 6 would not have had access
to either the central basin or the park woodlands (as shown in
Fig. 10 by the line marking the midpoint of travel times). With
respect to the central basin, this assumption appears unlikely since
both Nebekian and Kebaran sites have been documented here
indicating at least some use by both groups (Rollefson et al., 2001;
Richter, 2011). A testable prediction of the role of territoriality in
such land-use patterns would be to ascertain if Kebaran sites
cluster on the north and west sides, closest to Kharaneh IV, and
Nebekian sites are more common on the south and east sides,
closest to Jilat 6.
Predicting where a western relocation took place is more chal-
lenging when considering the role of territorial ranges, in part
because there is almost no data on early Epipaleolithic sites in this
area. For a western relocation, the most likely scenario from a least-
cost perspective, and assuming some level of territoriality, would
entail Jilat 6 inhabitants dispersing northeast to the uplands of the
Wadi Hidan, and those from Kharaneh IV dispersing to the Wadi
Zarqa uplands. This would have afforded both groups access to
steppe-woodland and park-woodland habitats. The presence of a
Nizzanan site (closely related to Nebekian but dating later in the
early Epipaleolithic) in the lower Wadi Hisban (Edwards et al.,
1999), just north of the Wadi Hidan, provides some potential
support for such a northwest-to-southeast-oriented annual set-
tlement pattern reconstruction.
Whether the eastern- or western-oriented annual settlement
model is most viable hinges in part on the seasonality of these two
large aggregation sites. Martin et al. (2010) argue for winter and
early spring occupation when local grasslands attracted large ga-
zelle herds, Jones (2012) suggests spring/summer occupation pre-
dominated, and Maher et al. (2012b) suggest autumn and winter
occupation. It is also possible that exploitation of the center of the
basin was often logistical in nature, and characterized by periodic
exploitation events staged from Jilat 6 and Kharaneh IV. As such,
winter-early spring in the basin and summer-fall in the west is
considered the most likely annual settlement structure. This of
course would have entailed sustained westward residential re-
locations to capitalize on productive and seasonally available
Such an east-west settlement orientation is also applicable to
other settings along the eastern edge of the southern Levant, such
as the Wadi Hasa and Wadi Musa-Petra regions. In each of these
drainage catchments, seasonal movement to other habitats would
have enabled early Epipaleolithic hunteregatherers to access a
different set of resources, particularly with respect to plants. Such a
wide-ranging annual settlement structure also bears some
Fig. 8. Changes in Least cost travel paths between Jilat 6 and adjacent sites in the Azraq Basin and the sea between Last Glacial Maximum (21,000 cal BP), Early Natuﬁan
(15,000 cal BP) and Early Neolithic (11,600 cal BP).
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 73
similarity to reconstructions offered by Goring-Morris (2009;
Goring-Morris et al., 2009) in the western Levant.
6. Summary and discussion
The objective of this study has been to use least-cost GIS
modeling to provide some hypotheses regarding the potential
extent and nature of hunteregatherer annual settlement organi-
zation at the height of the Last Glacial Maximum in the southern
Levant. The results highlight three points. First, such modeling
must include paleogeographic and paleoenvironmental re-
constructions given the very different landscape of 21,000 years ago
as opposed to during the Natuﬁan or early Neolithic, let alone today.
A notable aspect of the Last Glacial Maximum landscape was how
Fig. 9. Maximum one and two-day least-cost foraging ranges for Early Epipaleolithic major aggregation sites Jilat 6 and Kharaneh IV in the Azraq Basin.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7874
Lake Lisan and Lake Kinneret impeded the east-west movement
and interaction within the region. Much more work reﬁning these
data sets, and further modeling how they changed across shorter
intervals of time is needed to generate more accurate insights into
diachronic shifts in background conditions during the Terminal
Second, hunteregatherer maximum one- and two-day foraging
ranges are quite a bit larger than have been suggested previously
for the Levant, and it follows logically that the annual extent of a
group's foraging range potentially could have been considerably
larger as well (Bar-Yosef and Belfer-Cohen, 1989:451; Goring-
Morris, 2009:85e87; Henry, 1989:174; Vita Finzi and Higgs, 1970).
Fig. 10. Modeled travel times and seasonal relocation reconstructions for Early Epipaleolithic major aggregation sites Jilat 6 and Kharaneh IV in the Azraq Basin.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 75
These values are, however, consistent with the larger body of
hunteregatherer ethnographic literature (Kelly, 2013: Figs. 4e8).
The implications of these results need to be woven into future
considerations of potential territorial ranges, population densities,
inter-group interaction, and trade and exchange within the Levant.
It is also important to keep in mind that these are maximum
foraging areas and that the distances are modeled based on how
long one walks (and still get back to camp at the end of one or two
days assuming a maximum of eight hours of walking per day). They
provide a useful tool especially for understanding travel distance if
going after targeted resources that require little search time (such
as places where game congregate, tree-based resources such as
nuts and fruits, food caches, or non-food resources). They are also
helpful for perceiving where settlement relocations are most likely
to occur (i.e., beyond the maximum foraging range) when central
place collector strategies are employed. Average daily foraging
ranges, of course, will always be smaller, especially if foraging has
considerable search, collection, or ﬁeld processing time.
Third, daily foraging extent, vegetation zones, and habitat het-
erogeneity differed signiﬁcantly within the southern Levant during
the Last Glacial Maximum. As a result, it is to be expected that
adaptive patterns varied as well, and may have included a number
of facets as highlighted below. Residential mobility (as measured by
the number of moves per year) and the distance per move is pre-
dicted to be the greatest in the Western Area owing to pervasive
habitat homogeneity and least in the Central Area (falling within
the Levantine Rift Valley) given the heterogeneity/patchiness of
resource distribution. Similarly, the use of logistical forays to ac-
quire resources is likely to have been most pervasive in the Central
Area, and also a commonly employed strategy in the Eastern Area. It
is likely that in the open spaces of the Eastern Area, such logistical
forays covered greater distances to ensure encounter strategies
with large game. It is further predicted that the relative reliance on
animal resources was greatest in the Eastern Area versus the other
two areas. It also follows logically, that annual territory was
greatest in the Eastern Area (owing to lower overall productivity
and the greater reliance on game) and the smallest in the Central
Area (given the terrain and the diversity of productive habitats
available within a relatively restricted area).
The relative importance of a ﬁssionefusion adaptive strategy
also undoubtedly varied spatially, and was most commonly
employed within the Eastern Area. In the Azraq Basin this appears
to have been driven by seasonal availability of large gazelle herds
and by tethering major settlement to infrequent well-watered lo-
calities. It also appears that territoriality may have been greatest in
the Eastern Area, consistent with the importance of logistical
hunting and aggregation events (Whallon, 2006). Such activities
may also reﬂect the need for additional mechanisms to enhance
social interaction within the mating groups of the southern Levant
that were the most spatially dispersed (Pearce, 2014). These attri-
butes of open space adaptation in the southeastern Levant suggest
that social complexity edistinguished by coalescing in larger
groups (presumably multiple bands) and employing cooperative
hunting procurement tactics ewas most evident within the
eastern Azraq Basin of the southern Levant. Moreover, the presence
of two large aggregation sites (each potentially occupied by
distinctive groups within the southern Levant) reveals the presence
and maintenance of larger social interaction spheres and regional
territoriality. The signiﬁcant differences in daily foraging range
within the southern Levant also highlight the unsuitability of uni-
formly applying a standard foraging radius (such as 5 km, 6 km, or
even 10 km) to assess potential productivity or other factors.
These variances in adaptive options and emerging patterns
within the southern Levant at the height of the Last Glacial
Maximum provide baseline conditions for divergences in adaptive
trajectories within the region. For example, Epipaleolithic groups
inhabiting the western Levantine woodlands may have been more
likely to shift to more costly resources over time owing to the ho-
mogeneous nature of the setting and declining foraging ranges,
owing in part to sea level rise. Similarly, if inhabitants in the Central
Area followed the dictums of central place theory eincreasingly
relying on logistical procurement tactics rather than seasonal re-
locations within different habitats ethen they may have been
preadapted to more sedentary conditions. In the long run, taking a
broader perspective on how these baseline conditions and corre-
sponding adaptations differed within the early Epipaleolithic
should enhance our understanding of the causal factors that un-
derlie subsequent shifts to sedentism and early food production.
We thank the guest editors for inviting us to participate in the
Society for American Archaeology symposium that formed the
basis for this issue, and Far Western Anthropological for providing
support to conduct this study. The graphs were prepared by
Kathleen Montgomery and the maps by Paul Brandy. We also thank
Adrian Whitaker, Nicole Birney, and two anonymous reviews for
their input on earlier versions of the manuscript.
Allcock, S., 2009. Beyond Trade and Subsistence: the Use of Shell Ornaments to Infer
Social Interaction and Increasing Complexity during the Early and Middle
Epipaleolithic, Jordan (unpublished MSc. dissertation). Institute of Archaeology,
University College, London.
Ashkenazi, S., 2004. Wetland drainage in the Levant (Late Hula, Amik G€
olü, and el-
Azraq Oasis): impact on avian fauna. In: Goren-Inbar, N., Speth, J.D. (Eds.),
Human Paleoecology in the Levantine Corridor. Oxbow Books, Oxford,
Banning, E.B., Dods, R.R., Field, J., Kuijt, I., McCorriston, J., Siggers, J., Taani, H.,
Triggs, J., 1992. Tabaqat al-Buma: 1990 excavations at a Kebaran and Late
Neolithic site in Wadi Ziqlab. Annual of the Department of Antiquities of Jordan
Bar-Matthews, M., Ayalon, A., Kaufman, A., Wasserburg, G.J., 1999. The eastern
Mediterranean paleoclimate as a reﬂection of regional events: Soreq Cave,
Israel. Earth and Planetary Science Letters 166, 85e95.
Bar-Matthews, A., Ayalon, A., Gilmour, M., Matthews, A., Hawkesworth, C.J., 2003.
Sea-land oxygen isotopic relationships from planktonic foraminifera and spe-
leothems in the Eastern Mediterranean region and their implication for pale-
orainfall during interglacial intervals. Geochimica et Cosmochimica Acta 67,
Bartov, Y., Goldstein, S.L., Stein, M., Enzel, Y., 2003. Catastrophic arid episodes in the
Eastern Mediterranean linked with the North Atlantic Heinrich events. Geology
Bartov, Y., Stein, M., Enzel, Y., Agnon, A., Reches, Z., 2002. Lake levels and sequence
stratigraphy of Lake Lisan, the late Pleistocene precursor of the Dead Sea.
Quaternary Research 57, 9e21.
Bar-Yosef, O., 1970. The Epi-Paleolithic Cultures of Palestine (unpublished Ph.D.
thesis). Hebrew University, Jerusalem.
Bar-Yosef, O., 1991a. Raw material exploitation in the Levantine Epi-Paleolithic. In:
Montet-White, A., Holen, S. (Eds.), Raw Material Economies Among Prehistoric
Hunter-Gatherers, University of Kansas Publications in Anthropology 19. Uni-
versity of Kansas, pp. 235e250.
Bar-Yosef, O., 1991b. The search for lithic variability among Levantine Epi-Paleolithic
industries. In: Meignen, L. (Ed.), 25 Ans d'
Etudes Technologiques en Pr
XIe Rencontre Internationaux d'Arch
eologie et Histoire d'Antibes. Editions
APDCA, Juan-les-Pins, pp. 319e335.
Bar-Yosef, O., Belfer-Cohen, A., 1989. The origins of sedentism and farming com-
munities. Journal of World Prehistory 3, 447e498.
Bar-Yosef, O., Valla, F.R., 2013. Natufian Foragers in the Levant: Terminal Pleistocene
Social Changes in Western Asia. In: International Monographs in Prehistory 19.
Ann Arbor MI.
Baruch, U., Bottema, S., 1999. A new pollen diagram from Lake Hula. Vegetational,
climatic and anthropogenic implications. In: Kawanabe, H., Coulter, G.W.,
Roosevelt, A.C. (Eds.), Ancient Lakes. Their Cultural and Biological Diversity.
Kenobi, Ghent, pp. 75e86.
Belfer-Cohen, A., Goring-Morris, A.N., 2011. Becoming farmers: the inside story.
Current Anthropology 52 (S4), S209eS220.
Belitzky, S., 2002. The evolution of environment and relief around the Ohalo II
prehistoric site. In: Nadel, D. (Ed.), Ohalo II. A 23,000-year-old Fisher-Hunter-
Gatherers' Camp on the Shore of the Sea of Galilee. Reuben and Edith Hecht
Museum, University of Haifa, Haifa, pp. 53e55.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7876
Bettinger, R.L., Malhi, R., McCarthy, H., 1997. Central place models of acorn and
mussel processing. Journal of Archaeological Science 24, 887e899.
Binford, Lewis R., 1980. Willow smoke and dogs' tails: Hunter-gatherer settlement
systems and archaeological site formation. American Antiquity 45, 4e20.
Binford, Lewis R., 2001. Constructing Frames of Reference: an Analytical Method for
Archaeological Theory Building Using Ethnographic and Environmental Data
Sets. University of California Press, Berkeley.
Byrd, B.F., 2005. Reassessing the emergence of village life in the Near East. Journal of
Archaeological Research 13, 231e290.
Byrd, B.F., 2014. The Late Pleistocene occupation of Madamagh rockshelter, south-
ern Jordan: new data and perspectives on an old excavation. In: Finlayson, B.,
Makarewicz, C. (Eds.), Settlement, Survey and Stone. Essays on Near Eastern
Prehistory in Honour of Gary Rollefson. ex oriente, Berlin, pp. 37e52.
Byrd, B.F., Brandy, P., Hildebrandt, W.R., Wee, S., Beason, M., Webb, T., 2008. Cultural
Resources Alternatives Assessment for the Shasta Lake Water Resources
Investigation, Shasta and Tehama Counties, California. Bureau of Reclamation,
Sacramento. Report on ﬁle. Northeast Information Center, Chico State,
Byrd, B.F., Garrard, A.N.,1990. The last Glacial Maximum in the Jordanian Desert. In:
Gamble, C., Soffer, O. (Eds.), The World at 18000 BP. Volume 2 Low Latitudes.
Unwin Hyman, London, pp. 78e96.
Coinman, N., 1993. WHS 618-Ain el-Buhira an Upper Paleolithic site in the Wadi
Hasa, West-Central Jordan. Pal
eorient 19 (2), 17e37.
Collar, A., Coward, F., Brughmans, T., Mills, B.J., 2015. Networks in archaeology:
phenomena, abstraction, representation. Journal of Method and Theory 22,
Edwards, P., Bourke, C., Colledge, S.J., Head, M.J., Macumber, P.G., 1988. Late Pleis-
tocene prehistory in the Wadi al-Hammeh, Jordan. Oxford. In: Garrard, A.N.,
Gebel, H.G. (Eds.), The Prehistory of Jordan. The State of Research in 1986,
British Archaeological Reports, International Series 396, pp. 525e565.
Edwards, P.C., Macumber, P.G., Head, M.J., 1996. The Early Epipaleolithic of Wadi al-
Hammeh. Levant 28, 115e130 .
Edwards, P.C., Head, M.J., Macumber, P.G., 1999. An Epipaleolithic sequence from
Wadi Hisban in the east Jordan Valley. Annual of the Department of Antiquities
of Jordan 43, 27e48.
Enzel, Y., Amit, R., Dayan, U., Crouvi, O., Kahana, R., Ziv, B., Sharon, D., 2008. The
climatic and physiographic controls of the eastern Mediterranean over the late
Pleistocene climates in the southern Levant and its neighbouring deserts.
Global and Planetary Change 60, 165e192.
Feibel, C.S., Goren-Inbar, N., Frumin, M., 2009. Historical perspectives on long-term
landscape evolution in the Hula Basin. In: Shea, J.J., Lieberman, D.E. (Eds.),
Transitions in Prehistory. Essays in Honor of Ofer Bar-Yosef. Oxbow Books,
Oxford, pp. 23e31.
Fleming, K., Johnston, P., Zwartz, D., Yokoyama, Y., Lambeck, K., Chappell, J., 1998.
Reﬁning the eustatic sea-level curve since the Last Glacial Maximum using far-
and intermediate ﬁeld sites. Earth and Planetary Science Letters 163, 327e342.
Garrard, A., Byrd, B., 1992. New dimensions to the Epipalaeolithic of the Wadi el-
Jilat in central Jordan. Pal
eorient 18 (1), 47e62.
Garrard, A.N., Byrd, B.F., 2013. Beyond the Fertile Crescent: Late Palaeolithic and
Neolithic Communities of the Jordanian Steppe. The Azraq Basin Project Vol-
ume 1: Project Background and the Late Palaeolithic (Geological Context and
Technology). In: Council for British Research in the Levant Supplemental Series,
vol. 13. Oxbow Books, Oxford.
Garrard, A., Yazbeck, C., 2008. Qadisha Valley prehistory Project, Lebanon. The
2004e2008 excavations at Moghr el-Ahwal. Bulletin d'Arch
eologie et d'Archi-
tecture Libanaises 12, 5e15.
Garrard, A., Baird, D., Byrd, B.F., 1994. The chronological basis and signiﬁcance of the
Late Palaeolithic and Neolithic sequence in the Azraq Basin, Jordan. In: Bar-
Yosef, O., Kra, R. (Eds.), Late Quaternary Chronology and Paleoclimates of the
Eastern Mediterranean. Radiocarbon, University Arizona, pp. 177e199 .
Goring-Morris, A.N., 2009. Two Kebaran Occupations near Nahal Soreq, and the
reconstruction of group ranges in the Early Epipalaeolithic of the Israeli littoral.
Eurasian Prehistory 6 (1e2), 75e93.
Goring-Morris, A.N., Belfer-Cohen, A., 1997. The articulation of cultural processes
and late Quaternary environmental changes in Cisjordan. Pal
Goring-Morris, A.N., Belfer-Cohen, A., 2003. More than Meets the Eye: Studies on
Upper Palaeolithic Diversity in the Near East. Oxbow Press, Oxford.
Goring-Morris, N., Hovers, E., Belfer-Cohen, A., 2009. The dynamics of Pleistocene
and Early Holocene settlement patterns and human adaptations in the Levant:
an overview. In: Shea, J.J., Lieberman, D.E. (Eds.), Transitions in Prehistory: Es-
says in Honor of Ofer Bar-Yosef. Oxbow, Oxford, pp. 185e252.
Greenbaum, N., Ben-Zvi, A., Haviv, I., Enzel, Y., 2006. The hydrology and paleohy-
drology of the Dead Sea tributaries. In: Enzel, Y., Agnon, A., Stein, M. (Eds.), New
Frontiers in Dead Sea Paleoenvironment Research, Special Paper, Geological
Society of America 401, pp. 63e93.
Grove, M., 2009. Hunter-gatherer movement patterns: causes and constraints.
Journal of Anthropological Archaeology 28, 222e233.
Grove, M., 2010. Logistical mobility reduces subsistence risk in hunting economies.
Journal of Archaeological Science 37, 1913e1921.
Hazan, N., Stein, M., Agnon, A., Marco, S., Nadel, D., Negendank, J., Schwab, M.,
Neev, D., 2005. The late Quaternary limnological history of Lake Kinneret (Sea of
Galilee), Israel. Quaternary Research 63, 60e77.
Henry, D.O., 1989. From Foraging to Agriculture: the Levant at the End of the Ice
Age. University of Pennsylvania Press, Philadelphia.
Henry, D.O. (Ed.), 1995. Prehistoric Cultural Ecology and Evolution: Insights from
Southern Jordan. Plenum Press, New York.
Hillman, G.C., 1996. Late Pleistocene changes in plant-foods available to hunter-
gatherers of the northern Fertile Crescent: possible preludes to cereal cultiva-
tion. In: Harris, D.R. (Ed.), The Origins and Spread of Agriculture and Pastoralism
in Eurasia. UCL Press, London, pp. 159e203.
Hours, F., 1992. Le Pal
eolithique et l'
eolithique de la Syrie et du Liban. Dar el-
Hovers, E., Marder, O., 1991. Typo-chronology and absolute dating of the Kebaran
complex: implications from the second season of excavations at Urkan e-Rub
IIA. Journal of the Israel Prehistoric Society 24, 34e58.
Howey, M.C.L., 2007. Using multi-criter ia cost surfa ce analysis to explore
past regiona l landscape s: a case study of ritual activ ity and social interaction
in Michigan , AD 1200e
1600. Journal of Archaeological Science 34,
183 0e18 46 .
Hunt, C.O., Garrard, A.N., 2013. Section B: the Late Palaeolithic eGeological Context.
In: Garrard, A.N., Byrd, B.F., 2013. Beyond the Fertile Crescent: Late Palaeolithic
and Neolithic Communities of the Jordanian Steppe. The Azraq Basin Project
Volume 1: Project Background and the Late Palaeolithic (Geological Context and
Technology). In: Council for British Research in the Levant Supplemental Series,
vol. 13. Oxbow Books, Oxford, pp. 53e135 .
Jones, J.R., 2012. Using gazelle dental cementum studies to explore seasonality and
mobility patterns of the Early-Middle Epipalaeolithic Azraq Basin, Jordan.
Quaternary International 252, 195e201.
Jones, M.D., Richter, T., 2011. Palaeoclimatic and archaeological implications of
Pleistocene and Holocene environments in Azraq, Jordan. Quaternary Research
76 (3), 63e72.
Kantner, J., 2012. Realism, reality, and routes: evaluating cost-surface and cost-path
algorithms. In: White, D.A., Surface-Evans, S.L. (Eds.), Least Cost Analysis of
Social Landscapes: Archaeological Case Studies. University of Utah Press, Salt
Lake City, pp. 225e238.
Kelly, R.L., 1983. Hunter-gatherer mobility strategies. Journal of Anthropological
Research 39, 277e306.
Kelly, R.L., 2013. The Foraging Spectrum: Diversity in HuntereGatherer Lifeways.
Cambridge University Press, Cambridge.
Lehner, B., Verdin, K., Jarvis, A., 2008. New global hydrography derived from
spaceborne elevation data. Eos, Transactions, AGU 89 (10), 93e94. http://www.
Maher, L.A., Richter, T., Macdonald, D., Jones, M.D., Martin, L., Stock, J.T., 2012a.
Twenty thousand-year-old huts at a hunter-gatherer settlement in eastern
Jordan. PLoS ONE 7 (2), e31447. http://dx.doi.org/10.1371/journal.pone.0031447.
Maher, L.A., Richter, T., Stock, J.T., 2012b. The pre-Natuﬁan Epipaleolithic: long-term
behavioral trends in the Levant. Evolutionary Anthropology 21, 69e81.
Martin, L., Edwards, Y., Garrard, A., 2010. Hunting practices at an eastern Jordanian
Epipalaeolithic aggregation site: the case of Kharaneh IV. Levant 42, 107e135.
Moore, A.M.T., Hillman, G.C., Legge, A.J., 2000. Village on the Euphrates. From
Foraging to Farming at Abu Hureyra. Oxford University Press, Oxford.
Morgan, C., 2008. Reconstructing prehistoric hunter-gatherer foraging radii: a case
study from California's southern Sierra Nevada. Journal of Archaeological Sci-
ence 35, 247e258.
Morgan, C., 2009. Optimal foraging patterns in the Sierra Nevada, Alta California.
California Archaeology 1, 205e226.
Muheisen, M., 1988. The Epipalaeolithic phases of Kharaneh IV. In: Garrard, A.N.,
Gebel, H.G. (Eds.), The Prehistory of Jordan. The State of Research in 1986,
British Archaeological Reports, International Series 396, pp. 353e367. Oxford.
Nadel, D., 2003. The Ohalo II ﬂint assemblages and the beginning of the Epi-
paleolithic. In: Goring-Morris, A.N., Belfer-Cohen, A. (Eds.), More than Meets the
Eye: Studies on Upper Palaeolithic Diversity in the Near East. Oxbow Books,
Oxford, pp. 216e229.
Neeley, M.P., Peterson, J.D., Clark, G.A., Fish, S.K., Glass, M., 1998. Investigations at
Tor al-Tareeq: an Epipaleolithic Site in the Wadi el-Hasa. Journal of Field
Archaeology 25, 295e317.
Niklewski, J., van Zeist, W., 1970. A late Quaternary pollen diagram from north-
western Syria. Acta Botanica Neerlandica 19, 737e754.
Olszewski, D.I., 2011. Lithic ‘culture’issues: insights from the Wadi al-Hasa Epi-
paleolithic. In: Healey, E., Campbell, S., Maeda, O. (Eds.), The State of the Stone:
Technologies, Continuities and Contexts in Near Eastern Lithics, Studies in Early
Near Eastern Production, Subsistence and Environment 13. ex oriente, Berlin,
Olszewski, D.I., Stevens, M., Glass, M., Beck, R.F., Cooper, J., Clark, G.A., 1994. The
1993 excavations at Yutil al-Hasa (WHS 784), an Upper/Epipaleolithic site in
west-central Jordan. Pal
eorient 20 (2), 129e141.
Orians, G.H., Pearson, N.E., 1979. On the theory of Central place foraging. In:
Horn, D.J., Stairs, G.R., Mitchell, R.D. (Eds.), Analysis of Ecological Systems. Ohio
State University, Columbus, pp. 155e177.
Pandolf, K.B., Givoni, B., Goldman, R.F., 1977. Predicting energy expenditure with
loads while standing or walking very slowly. Applied Physiology 43, 577e581.
Pearce, E., 2014. Modelling mechanisms of social network maintenance in hunter-
gatherers. Journal of Archaeological Science 50, 403e413.
Reese, D.S., 1991. Marine shells in the Levant: upper Palaeolithic, Epipalaeolithic and
Neolithic. In: Bar-Yosef, O., Valla, F. (Eds.), The Natuﬁan Culture in the Levant.
International Monographs in Prehistory, Ann Arbor, pp. 613e628.
Reese, D.S., 1995. Shells from the Wadi Hisma sites. In: Henry, D.O. (Ed.), Prehistoric
Cultural Ecology and Evolution: Insights from Southern Jordan. Plenum Press,
New York, pp. 385e390.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e78 77
Reese, D.S., 2014. Marine shell assemblage. (In Byrd, B.F., The Late Pleistocene
occupation of Madamagh Rockshelter, southern Jordan: new data and per-
spectives on an old excavation). In: Finlayson, B., Makarewicz, C. (Eds.), Set-
tlement, Survey and Stone. Essays on Near Eastern Prehistory in Honour of Gary
Rollefson. ex oriente, Berlin, pp. 49e50.
Richter, T., Allcock, S., Jones, M., Maher, L., Martin, L., Stock, J., Thorne, B., 2010. New
light on Final Pleistocene settlement diversity in the Azraq Basin (Jordan):
recent excavations at 'Ayn Qasiyya. Pal
eorient 35 (2), 49e68.
Richter, T., 2011. Nebekian, Qalkhan and Kebaran: variability, classiﬁcation and
interaction. New insights from the Azraq Oasis. In: Healey, E., Campbell, S.,
Maeda, O. (Eds.), The State of the Stone: Technologies, Continuities and Con-
texts in Near Eastern Lithics, Studies in Early Near Eastern Production, Subsis-
tence and Environment 13. ex oriente, Berlin, pp. 33e49.
Richter, T., Garrard, A.N., Allcock, S., Maher, L.A., 2011. Interaction before agriculture:
exchanging material and sharing knowledge in the Final Pleistocene Levant.
Cambridge Archaeological Journal 21, 95e114 .
Richter, T., Maher, L.A., Garrard, A.N., Edinborough, K., Jones, M.D., Stock, J.T., 2013.
Epipalaeolithic settlement dynamics in southwest Asia: new radiocarbon evi-
dence from the Azraq Basin. Journal of Quaternary Science 28, 467e479.
Robinson, S.A., Black, S., Sellwood, B.W., Valdes, P.J., 2006. A review of palae-
oclimates and palaeoenvironments in the Levant and Eastern Mediterranean
from 25,000 to 5,000 years BP: setting the environmental background for the
evolution of human civilisation. Quaternary Science Reviews 25, 1517e1541.
Rollefson, G.O., Quintero, L., Wilke, P., 2001. Azraq Wetlands Survey 2000. Pre-
liminary report. Annual Department Antiquities Jordan 45, 71e81.
Rust, A., 1950. Die H€
ohlenfunde von Jabrud (Syrien). Karl Wachholtz, Neumünster.
Saxon, E.C., Martin, G., Bar-Yosef, O., 1978. Nahal Hadera: an open-air site on the
Israeli littoral. Pal
eorient 4, 253e266.
Schuldenrein, J., 1998. Geomorphology and stratigraphy of prehistoric sites along
the Wadi al-Hasa. In: Coinman, N.R. (Ed.), The Archaeology of the Wadi al-Hasa,
West-Central Jordan, Surveys, Settlement Patterns and Paleoenvironments, vol.
1. Arizona State University, pp. 205e228. Anthropological Research Papers 50.
Schuldenrein, J., Clark, G.A., 1994. Landscape and prehistoric chronology of west-
central Jordan. Geoarcheology 9, 31e55.
Stutz, A.J., Estabrook, G.F., 2004. Computationality intensive multivariate statistics
and relative frequency distributions in archaeology (with an application to the
Early Epipaleolithic of the Levant). Journal of Archaeological Science 31,
Tobler, W.,1993. Three Presentations on Geographical Analysis and Modeling: Non-
isotropic Geographic Modeling Speculations on the Geometry of Geography
Global Spatial Analysis. Technical report. National Center for Geographic In-
formation and Analysis, 93 (1).
Van Leusen, M., 2002. Pattern to Process: Methodological Investigations into the
Formation and Interpretation of Spatial Patterns in Archaeological Landscapes.
Van Zeist, W., Bottema, S., 1991. Late Quaternary Vegetation of the Near East.
Wiesbaden. In: Tübinger Atlas des Vorderen Orients A18. Reichert.
Vita Finzi, C., Higgs, E., 1970. Prehistoric economy in the Mount Carmel area of
Palestine; site catchment analysis. Proceedings of the Prehistoric Society 36,
Weinstein-Evron, M., 1983. The paleoecology of the early Wurm in the Hula Basin,
eorient 9 (1), 5e19.
Whallon, R., 2006. Social networks and information: non-“utilitarian”mobility
among hunter-gatherers. Journal of Anthropological Archaeology 25, 259e270.
White, D.A., Surface-Evans, S.L., 2012. Least Cost Analysis of Social Landscapes:
Archaeological Case Studies. University of Utah Press, Salt Lake City.
Whitley, T., Hicks, L., 2001. Using a Geographic Information Systems (GIS) Approach
to Extract Potential Prehistoric and Historic Period Travel Corridors across a
Portion of North Georgia. http://www.brockington.org/papers/SEAC2001-
Wood, B.M., Wood, Z.J., 2006. Energetically optimal travel across Terrain: visuali-
zations and a New Metric of geographic distance with, archaeological appli-
cations. In: SPIE Proceedings, vol. 6060. San Jose, CA.
Yasuda, Y., Kitagawa, H., Nakagawa, T., 2000. The earliest record of major anthro-
pogenic deforestation in the Ghab Valley, northwest Syria: a palynological
study. Quaternary International 73/74, 127e136 .
Zeanah, David W., 2004. Sexual Division of Labor and Central place foraging: a
model for the Carson Desert of Western Nevada. Journal of Anthropological
Archaeology 23, 1e32.
Zohary, M., 1973. Geobotanical Foundations of the Middle East. Fischer, Stuttgart.
B.F. Byrd et al. / Quaternary International 396 (2016) 62e7878