Extinction of megafauna: How could the
research get so wrong?
Ron W Nielsen aka Jan Nurzynski1
Environmental Futures Centre, Gold Coast Campus, Griffith University, Qld, 4222, Australia
October, 2013
The published evidence for the human-mediated extinction of megafauna is
examined and is found to be unsubstantiated.
It is shown that the claimed
evidence is not based on the data describing the growth of human population but
on invented data describing the growth of a non-existent, phantom genus we could
call Phasmapithecus. Furthermore, it is also shown that contrary to the published
claim even these invented data do not support the concept of the human-mediated
extinction of megafauna.
Introduction
The problem of the extinction of many species of megafauna is hard to solve because it
requires an accumulation of massive amount of data. Many contributing factors have to be
also considered and only one of them is the possible human mediation.
1
[email protected]; [email protected]; http://home.iprimus.com.au/nielsens/ronnielsen.html
Suggested citation:
Nielsen, R. W. aka Nurzynski, J. (2013). Extinction of megafauna: How could the research get so wrong?
http://arxiv.org/ftp/arxiv/papers/1309/1309.1804.pdf
1
Claimed empirical evidence
A good way of showing that extinction of megafauna might have been caused by humans is
to demonstrate a clear correlation between the rates of the extinction and the growth of
human population. The only such study was published by Barnosky (2008). His results,
reproduced in Fig. 1, are now also available as a slide for teaching (National Academy of
Sciences, 2008). They are claimed to show a close correlation between the rapid decline in
the number of species of megafauna and the growth of human population. “The numbers of
megafauna species lost were modest until the human growth curve began its rapid
exponential rise between 15.5 and 11.5 kyr B.P.” (Barnosky, 2008, p. 11544. Italics added.).
Thus, the extinction pulse (the sudden and relatively short-lasting decline in the number of
species) is explained as being correlated with the beginning of the rapid increase in the size of
human population.
Population "Data" (in 100,000)
Number of Megafauna Species
Trend 1
Trend 2
Extinction Window
600
500
400
300
200
100
100000
10000
0
1000
Year BP
Fig. 1. The claimed correlation (Barnosky, 2008, p. 11544) between the extinction of
megafauna and the assumed growth of human population.
2
The immediate question is why the extinction spree lasted for only such a relatively short
time. Why was the process of extinction discontinued even though the claimed size of human
population continued to increase, the hunting ability was improving and the number of
species was still high, reduced from around 300 just before the claimed extinction pulse to
only around 180 towards its end?
1000
100
10
25 ka
100000
Population "data" (in 100,000)
Number of Megafauna Species
Trend 1
Trend 2
Extinction Window
10000
1
1000
Year BP
Fig. 2. Evidence claimed to be in support of the anthropologically driven extinction of
megafauna (Barnosky 2008; National Academy of Sciences, 2008), shown earlier in Fig. 1, is
now displayed using logarithmic scales for both axes.
The claimed close correlation between the extinction of megafauna and the growth of
population is also strongly questionable for another reason. The extinction is represented by a
3
step-wise distribution but the growth of the population follows a continuously increasing
trajectory. The claimed correlation between the two trajectories is, therefore, unconvincing.
It is interesting how much one can learn sometimes by simply looking at the data from a
different perspective. Fig. 2 presents precisely the same data as published by Barnosky
(2008) and as shown in Fig. 1 but now displayed using logarithmic scales for both axes. In
this representation, population data follow approximately two straight lines, both being
clearly uncorrelated with the extinction data. The claimed correlation does not exist.
There was a beginning of a rapid increase in the growth of population around 25 ka
(thousands of years ago). However, there was no change in the population growth pattern just
before the claimed onset of the extinction pulse at around15.5 ka. A 10,000 year gap between
the beginning of the rapid increase around 25 ka and the beginning of the extinction pulse
around 15.5 ka is too large to claim any meaningful correlation between these two events.
Results published by Barnosky (2008) are interesting but for the unintended reason. They
demonstrate that there is no correlation between the growth of the population and the
extinction of megafauna. The hypothesis of the human-assisted extinction is not supported by
the population data.
However, the published claim is even more intriguing because the population data used to
support the hypothesis of the human-mediated extinction of megafauna do not even describe
the growth of human population. These “data” were created by Hern (1999) and are
reproduced in Table 1. Even a quick glance at this table shows that the published set of
numbers cannot be interpreted as describing the growth of human population.
First, it is well-known that genus Homo did not even exist 3 Ma (million years ago). The set
of numbers created by Hern (1999) cannot be linked with Australopithecus or with genus
Homo because Australopithecus ceased to exist approximately 1.2 Ma and genus Homo did
4
not emerge until around 2.4 Ma or even later if the questionable classification of H. habilis is
excluded (see Fig. 3). This set of numbers describes, therefore, the growth of a non-existing
genus, which for convenience of this discussion we might call Phasmapithecus, from Latin
phasma or Ancient Greek φάσμα (ghost, spectre, apparition) and the Greek word πίθηκος
(ape).
Table 1. Population “data” created by Hern (1999) and used by Barnosky (2008) in support
of the hypothesis of the anthropologically driven extinction of megafauna.
Year [BP]
3,000,000
2,500,000
2,000,000
1,750,000
1,500,000
1,250,000
1,000,000
900,000
800,000
700,000
600,000
500,000
Population
20=1
21=2
22=4
23=8
24=16
25=32
26=64
27=128
28=256
29=512
210=1,024
211=2,048
T2
500,000
500,000
250,000
250,000
125,000
125,000
100,000
100,000
100,000
100,000
100,000
100,000
400,000
300,000
250,000
200,000
150,000
100,000
50,000
25,000
20,000
15,000
12,500
212=4,096
213=8,192
214=16,384
215=32,768
216=65,536
17
2 =131,072
218=262,144
219=524,228
20
2 =1,048,576
221=2,097,152
222=4,194,304
100,000
50,000
50,000
50,000
50,000
50,000
25,000
5,000
5,000
2,500
2,500
Year [BP]
10,000
8,000
6,000
4,000
3,000
2,000
340
140
68
22
0=AD 1998
Population
223=8,388,608
224=16,777,216
225=33,554,432
226=67,108,864
27
2 =134,217,720
2 28  268 ,434 ,000
2 29  536 ,868 ,000
2 30  1,073 ,736 ,000
2 31  2,147 ,472 ,000
2 32  4 , 294 ,944 ,000
T2
2,000
2,000
2,000
1,000
700
1650
200
80
46
37
2 32.5  6,000 ,000 ,000
Projections
AD 2053
2013
2053
2103
2173
2243
2313
2383
2453
Population
8,589,888,000
16,777,216,000
34,359,552,000
68,719,104,000
137,438,208,000
274,876,416,000,
549,752,832,000
1,000,000,000,000
T2
40
50
70
70
70
70
70
70
T2 – doubling time
Second, the created genus is endowed with bizarre reproductive properties and with the
exceptional longevity, because it had an ability of self-replicating after 500,000 years. Hern
appears to be aware of this problem but he promptly points out that he could have started his
construction from “Adam and Eve” (Hern, 1999, p. 63) confirming that the numbers he listed
5
were of his own creation and that they had nothing to do with the paleontological or
archaeological research. However, even if we start with two representatives of this phantom
genus we still have the problem of longevity and of the peculiar procreation abilities because
the initial two representatives increased to four after 500,000 years. The same problem
applies also to other early stages of growth of this invented genus.
Fig. 3. The sequential representation of human evolution based on the data presented by
Cremin (2007), Gibbons (2009), Heile-Selassie (2010), Thanukos (2010) and Zagura (2008)
showing also the phantom genus, Phasmapithecus, created by Hern (1999) and used by
Barnosky (2008) in support of the postulate of the human-assisted extinction of megafauna.
Third, the set of numbers listed in Table 1 represents a crudely constructed patchwork of
exponential growth sections characterised by drastically different doubling times and
consequently by drastically different growth rates, showing no signs of smooth transitions
between adjacent sections. Thus for instance, around 2 Ma, the doubling time changed
6
abruptly from 500,000 years to 250,000 years, decreasing neatly by a factor of two,
suggesting immediately the man-made construction rather than natural causes. This pattern of
abrupt transitions, often involving the factor of two or producing round numbers for the
doubling times, continues until 3 ka when the doubling times stopped following the earlier
regular pattern probably to force the created numbers to agree more closely with the data for
the growth of human population during the AD era. However, for no apparent reason, from
2103 until 2453 the population of this phantom genus is assumed to follow an exponential
trajectory corresponding to the growth rate of 1%.
Fourth, the projected population of Phasmapithecus in 2053 is about twice as high as any
reasonably expected size of human population around that time. Furthermore, the projected
population in 2453 is about ten times higher than the expected maximum of human
population.
All these features should have served as warning signs that the numbers constructed and
published by Hern (1999) do not describe the growth of human population. Indeed, it was
never the intention of Hern to claim them as “data.” “I did not intend in any way to offer my
table of the number of doublings as ‘data’ or a prediction. All of the numbers - and I don't
think there are any exceptions - are imprecise to wildly wrong and based on estimates ranging
from considered evaluation of available data (?) to total speculation” (Hern, 2013). His sole
purpose was to use these numbers to argue that in analogy with the growth of cancer cells, the
size of human population might have already doubled a critical number of times (Hern,
2013). Whether his claim is correct is another issue but he did not have to construct such an
elaborate set of numbers to argue his point.
The calculation of the number of doubling times depends only on just two parameters, the
initial and the final size of the population or more generally on the initial and final size of any
7
growing entity. It does not matter whether the growth is exponential or not, or made of
exponential sections as used by Hern (1999) who has a strong preference for using
exponential functions (Hern, 2013). In his calculations, Hern started with the initial size of
only one representative of his phantom genus, supposed to represent the genus Homo, so his
claim of the critical number of doubling times can be questioned but this issue has nothing to
do with the topic of the extinction of megafauna. The important point is that his created
numbers should have been never used in support of the hypothesis of the human-assisted
extinction of megafauna because this set does not describe the growth of human population.
However, as already pointed out earlier, even these numbers do not show any correlation with
the extinction pulse.
It is interesting that Hern (1999) constructed seven rapid increases in the growth of his
phantom genus Phasmapithecus before 15.5 ka but all of them were ignored in the discussion
of the extinction of megafauna. It is also interesting that Hern did not create a single rapid
increase anywhere near the extinction pulse but a rapid increase was claimed to support the
concept of the human-driven extinction. Could it be that the desire to see a support for a
cherished hypothesis might be so strong that it interferes with the scientific process of
investigation?
Another “evidence” used by Barnosky (2008) in support of the human-mediated extinction of
megafauna is the claimed close trade-off between the biomass of megafauna and the biomass
of humans. Barnosky’s calculations (Barnosky, 2008, p. 11455) are reproduced in Fig. 4.
“The results suggest that biomass loss from the early megafauna extinctions in Australia and
the first pulse of extinctions in Eurasia and Beringia were almost exactly balanced by the
gain in human biomass” (Barnosky, 2008, p. 11544. Italics added.). This claim was endorsed
and reinforced by Avise, Hubbell and Ayala (2008). However, this “evidence” cannot be
8
accepted not only because it is also based on the meaningless, and made-up “data” created by
Hern (1999) describing the growth of Phasmapithecus but also because it is impossible to
calculate and verify the biomass transfer. It is impossible to support the claim of the “inverse
relationship between human biomass and nonhuman megafaunal biomass” (Avise, Hubbell &
Ayala, 2008, p. 168).
Fig. 4. The biomass tradeoff between megafauna and Phasmapithecus calculated by
Barnosky (2008) and used in support of the hypothesis of the human-assisted extinction of
megafauna.
The biomass transfer between megafauna and humans can be only achieved by ingestion but
it is obvious that only a minute fraction of food we consume is converted into our body mass.
Barnosky’s calculations are based on the following three implicit and unacceptable
assumptions: (1) humans [or in his case the representatives of Phasmapithecus created by
Hern (1999)] including children ate mainly, if not exclusively, megafauna; (2) they ate
everything, including bones, skin and waste; (3) everything they ate was converted to their
body mass and retained as their body mass until they died.
9
A single individual living 40 years, weighing on average 50 kg and consuming 1 kg of food
per day would have consumed approximately 15,000 kg of food during his lifetime but
converted only about 0.3% of the consumed food into his body mass. Considering that only a
fraction of the daily diet might have consisted of megafauna and that only a fraction of the
mass of killed megafauna was consumed as food, our data about human consumption, the
size of human population and the number of killed animals would have to be unrealistically
precise to calculate the biomass transfer between megafauna and humans. We would also
have to account for the loss of megafauna by natural causes, unrelated to human
consumption.
Summary and conclusions
We have shown that the claimed correlation (Barnosky, 2008) between the growth of human
population and the extinction of megafauna was based on invented “data,” the “data” which
do not describe the growth of human population but the growth of a phantom genus
Phasmapithecus, the set of numbers created by Hern (1999) who never expected that they
would be ever used as human population “data” (Hern, 2013). Furthermore, we have also
shown that even these invented “data” do not support the hypothesis of the human-assisted
extinction of megafauna because contrary to the published claim the essential correlation
does not exist.
Finally, the disturbing questions are: How could the research get so wrong? How was it
possible to publish fiction in the peer-reviewed scientific literature? How was it possible to
publish invented data, while it was so glaringly obvious that these data were invented and that
they had nothing to do with the growth of human population? How was it possible for
Barnosky to use these so obviously incorrect set of numbers? How was it possible to accept
for publication Barnosky’s unsubstantiated evidence, the evidence based on so-obviously
10
invented data? How was it possible to accept for publication the claim of a correlation that
did not exist even for this invented data? How was it possible to accept for publication the so
obviously incorrect claim of a balance between biomass loss and biomass gain? How was it
possible to repeat this incorrect claim in another publication? Are these just isolated cases or
do they point to a more serious issue of a bias or of an oversight in peer-reviewed journals?
The examples we have used here do not represent a well-structured research based on solid
evidence and on logical progression, research leading to results that could be debated or used
to stimulate further investigation. The examples used here represent unscientific approach to
research, the approach based on inventing evidence and on using the invented evidence to
claim new invented evidence. This is no longer science but a creative storytelling, which
might be interesting but does it have a place in scientific publications?
We can find many examples of invented stories in numerous publications discussing the
growth of population2 but it appears that this problem in also present in other fields of
research.
Extensive correspondence with Tony Barnosky and Warren Hern is gratefully acknowledged.
References
Avise, J. C., Hubbell, S. P., & Ayala, F. J. (Eds) (2008). In the light of evolution: Vol. II,
Biodiversity and extinction. Washington, DC: National Academy of Sciences.
Barnosky, A. (2008). Megafauna biomass tradeoff as a driver of Quaternary and future
extinctions. Proceedings of the National Academy of Sciences, 105 (Suppl 1), 11543–
11548.
Cremin, A. (2007). Archaeologica: The World’s Most Significant Sites and Cultural
Treasures. Sydney, Australia: Random House Australia Pty Ltd.
This issue is discussed in the forthcoming book: The growth of human population and economic progress
explained.
2
11
Gibbons, A. (2009). A new kind of ancestor: Ardipithecus Unveiled, Science, 326, 36-43.
Haile-Selassie, Y. (2010). Phylogeny of early Australopithecus: new fossil evidence from the
Woranso-Mille (central Afar, Ethiopia). Phil. Trans. R. Soc. B, 365, 3323–3331.
Hern, W. M. (1999). How many times has the human population doubled? Comparisons with
cancer. Population and Environment, 21(1), 59-80.
Hern, W. M. (2013). Private communication.
National Academy of Sciences (2008). Megafauna biomass tradeoff as a driver of
Quaternary and future extinctions. PowerPoint Slide for Teaching. Proceedings of the
National Academy of Sciences of the United States of America,
http://www.pnas.org/content/105/suppl.1/11543/F2.expansion.html
Thanukos, A. (2010). Names: They are a-changing. Edo Evu Outreach, 3, 410-415.
Zagura, S. (2008). Current topics in human evolutionary genetic. In J. D. Bengsten, (Ed.), In
Hot Pursuit of Language in Prehistory: Essays in the Four Fields of Anthropology - in
Honor of Harold Crane Fleming (pp. 343-358). Amsterdam, The Netherlands: John
Benjamins Publishing Co.
12