<<

. 30
( 43 .)



>>


extinction. For the cold-adapted Moderns model the population size at 25-kyr
varied between 29 780 and 56 744 and never became extinct. The model
assumed that the bioclimatic differences between the four stages were main-
tained relative to each other as climatic conditions ¬‚uctuated. The criteria for
Modern Human populations differed. These always started at a level of 50 in the
Euro-Siberian zone where they originated and where it was assumed they had
not reached the environmental carrying capacity. In no simulation were Modern
Human populations established in the other bioclimatic stages at 40 kyr.
The time interval chosen for the simulations was 0.5 kyr. Temperatures were
estimated into four relative categories for each interval from the 18 O Green-
land Ice Cores (GRIP, 1993): warm (10.6% of the intervals); moderately warm
(22.7%); moderately cold (24.2%); and cold (42.4%). Thus, cold events domi-
nated OIS 3 (66.6%; n = 67). The proportion of cold intervals increased towards
the end of OIS 3: 58“50-kyr, 47.06% cold, n = 17; 50“40-kyr, 60% cold, n =
20; 40’30-kyr, 80% cold, n = 20; 30“25-kyr, 80% cold, n = 10. If a population
178 Neanderthals and Modern Humans

was considered to be warm-adapted then it was allocated a growth rate per cell
of 10 population units for each warm period and of 5 for each moderately warm
period, a decline rate of 5 for each moderately cold period and of 10 for each
cold period. The reverse applied to a cold-adapted population. A warm-tolerant
population would maintain the same population level during warm periods but
would decline as above in cold events. The reverse applied to a cold-tolerant
population. Growth rates were the same across all bioclimatic stages. A growth
rate of 10 units per time interval would, if the units were equated to people
(see below) mean a growth by 2 per 100 years, a measure that is considered
conservative for an expanding population. A population reaching a population
level in a cell below 0 population units was considered extinct. Recovery into
that cell would only occur where an adjacent cell still retained a population of
the same species and only if climatic conditions were favourable for growth.
Modern populations could only spread into adjacent cells under favourable con-
ditions. Movement from one cell to another only occurred from populated cells
to adjacent cells during favourable climatic conditions and between consecutive
bioclimatic stages and at 0.5-kyr intervals. In cases where adjacent cells were
more than one bioclimatic stage apart, colonisation was only permitted after two
consecutive favourable time intervals. Thus, for example, if a populated cell be-
longed to the supra-Mediterranean stage, then an adjacent meso-Mediterranean
cell could be colonised at the next time interval by an advancing cold-adapted
Modern population provided that conditions in that interval were cold. If the
adjacent cell was thermo-Mediterranean instead, then the colonisation would
require two successive cold intervals.
Four scenarios were tested for Neanderthals and two for Moderns. These were
considered to provide contrasting situations and clearly other models could be
generated to test more extreme situations. It is considered that the scenarios
tested serve to highlight the main testable predictions of the models. The four
Neanderthal scenarios tested the populations to be cold-adapted, cold-tolerant,
warm-tolerant and warm-adapted (Figure 7.11). For Moderns cold-adapted and
warm-adapted scenarios were tested. These models were intra-speci¬c density-
dependent models. In other words the populations of the two forms ¬‚uctuated
in response to environmental conditions and their own starting population lev-
els but were unaffected by the presence of the other form. Population sizes
per cell were capped at twice the maximum number of starting units (i.e. 200)
irrespective of bioclimatic stage to re¬‚ect a measure of environmental carrying
capacity. If we were to equate population units to people in the landscape, then
100 units in a 50 — 50-km square would represent a density of 0.04/km2 or
4 per 100 km2 ; 200 units would represent 0.08/km2 . In Africa, Hadza popu-
lation density in an area of 2500 km2 (therefore equivalent to one cell in the
Iberian models) was between 600 and 800, that is 0.24-0.32/km2 (O™Connell &
Modern Human colonisation and Neanderthal extinction 179

60000



50000



40000


Cold Adapted
30000


Warm Tolerant
20000
Population Size




Cold Tolerant
10000

Warm Adapted
0
-60 -50 -40 -30 -20

Time (kyr BP)

Figure 7.11. Contrasted fates of the simulated Neanderthal population under four
different scenarios. See text for details.



Hawkes, 1988), and !Kung San density in Dobe in the Kalahari was 466 in
9000 km2 , that is 0.05/km2 (Lee, 1979). Given that the density of most contem-
porary hunter“gatherer populations ranges between 0.01 and 0.4/km2 (see, for
example, Diamond, 1991) these model estimates are conservative and within
expected limits. In order to test the effect of interspeci¬c competition a scenario
was generated in which the presence of Modern Humans in the same cell as
Neanderthals further depressed the Neanderthal population by an additional 5
population units per time period of contact. This is a signi¬cant competitive
coef¬cient being equivalent to between 0.5 and 1.0 of the population growth
rate in favourable conditions and exceeding the growth rate in unfavourable
conditions.
Finally, in order to test longer time-scale implications, models spanning the
time period 130“0 kyr were generated for Neanderthal populations. In these
cases the time interval was set at 2 kyr but the rate of population change was
kept as in the earlier models. The climate estimates were based on Imbrie et al.
(1992).
180 Neanderthals and Modern Humans

The underlying conclusion of all the models was invariant of the degree
of climate-induced population ¬‚uctuation modelled or of the starting popula-
tion size: given the nature of the climatic oscillations of OIS 3, warm-adapted
populations always tended towards cumulative decline, and cold-adapted ones
always grew in size and never became extinct, supporting Finlayson et al.™s
(2000a) global model of human colonisation and Neanderthal extinction.
Of the four climatic models generated for Iberian Neanderthal populations the
warm-adapted, cold-intolerant model best ¬tted the empirical evidence (Fig-
ure 7.11). The cold-adapted, warm-intolerant model generated an expanding
Neanderthal population in Iberia with time and the population was still grow-
ing at the cut-off point of 25 kyr. The warm-tolerant and cold-tolerant models
produced population extinctions at 44 and 42 kyr respectively, far too rapid to
¬t the empirical data. Figure 7.11 illustrates the evolution of the warm-adapted,
cold-intolerant, Iberian Neanderthal population from 58 kyr to its extinction
at 31 kyr. For Moderns, the warm-adapted, cold-intolerant, model generated
the extinction of the Iberian population at 36 kyr and was thus unrealistic. The
cold-adapted, warm-intolerant model for Moderns on the other hand produced
an expanding population which eventually colonised the entire Iberian Penin-
sula, a pattern consistent with more generalised models of human population
growth (Ambrose, 1998). The evolution of the warm-adapted Neanderthal and
cold-adapted Modern Human Iberian populations from 40 kyr to 25 kyr is il-
lustrated in Figure 7.12. According to this model there is a protracted period
of 9 kyr (40“31 kyr) during which both populations occur within the Iberian
Peninsula.
Five cells were selected for analysis of the evolution of local populations
within Iberia and these were chosen to correspond to bioclimatically distinct
units within which empirical archaeological evidence existed. I shall call these
cells by the archaeological site which they represent: Gibraltar, Carihuela,
Caldeirao, Ermitons and El Castillo (Figure 7.13). The ¬ve sites show distinctive
patterns of population evolution (Figure 7.13). The Gibraltar population persists
longest and actually grows on two occasions during the period, at 34“35 kyr
and 31.5 kyr, the latter just before its ¬nal extinction at 31 kyr (Figure 7.13a).
At 32 kyr the population is brie¬‚y regionally extinct and there are no human
populations in Gibraltar. Moderns do not arrive until after the Neanderthals have
become extinct but the population grows rapidly from 30 kyr (Figure 7.13a).
This model therefore predicts no overlap of the two populations in Gibraltar.
The situation in the mountains to the north-east, at Carihuela, is similar (Figure
7.13b). Here the Neanderthal population does not exhibit the temporary recov-
ery of the Gibraltar population and the population becomes ¬nally extinct at
33.5 kyr, 2.5 kyr before the Gibraltar population. There is also an earlier period
of regional extinction, between 35.5 and 34.5 kyr, during which time there are no
Modern Human colonisation and Neanderthal extinction 181

50000



40000



30000



20000
Population Size




Moderns
10000

Neanderthals
0
-42 -40 -38 -36 -34 -32 -30 -28 -26 -24

Time (kyr BP)

Figure 7.12. Simulated evolution of the Neanderthal and Modern Iberian populations.



human populations in Carihuela (Figure 7.13b). Moderns arrive at 32 kyr, 2 kyr
before they reach Gibraltar, but there is no overlap with the Neanderthals (Figure
7.13b). The pattern in Caldeirao (Portugal) to the north-west is again different.
The Neanderthal population has a similar recovery capacity to the Gibraltar
population with increases at 35“34 kyr and 31.5 kyr, when Moderns are already
in the area (Figure 7.13c). The Neanderthal population is extinct at 32.5 kyr but
there is a re-entry into the area at 31.5 kyr with the ¬nal extinction at 31 kyr, at the
same time as at Gibraltar that is further south and due to the bioclimatic situation
of this site (Figure 7.9). The Moderns, however, arrive much earlier (by 35.5 kyr)
so there is a period of 3 kyr during which the two populations overlap region-
ally, the highest of all (Figure 7.13c). It is noteworthy that Bocquet-Appel &
Demars (2000b), using a different modelling procedure, reached a similar con-
clusion. In the north, the site at Ermitons is within the Mediterranean coastal
region and the Neanderthal population survives until 36 kyr (Figure 7.13d).
The model predicts a re-entry of Neanderthal elements into the area brie¬‚y
at 34 kyr, by which time the Moderns are well established, and the ¬nal ex-
tinction takes place at 33.5 kyr. The Moderns arrive early (at 38.5 kyr) so
we observe here a long period of regional overlap of the two populations,
predicted at 2.5 kyr (Figure 7.13d). Finally, the pattern at El Castillo, within
182 Neanderthals and Modern Humans

(a)




Castillo
Ermitons

Caldeirao


Carihuela

Gibraltar

100
(b)



80




60




40
Population Size




Neanderthals
20


Moderns
0
-42 -40 -38 -36 -34 -32 -30 -28 -26 -24


Time (kyr BP)

Figure 7.13. (a) Regions selected for analysis based on known sites; (b) simulated
evolution of Neanderthal and Modern populations in Gibraltar; (c) Carihuela;
(d) Caldeirao; (e) Ermitons; and (f) El Castillo.
Modern Human colonisation and Neanderthal extinction 183

(c) 120



100



<<

. 30
( 43 .)



>>