Quaternary Science Reviews 344 (2024) 108979

Contents lists available at ScienceDirect

Quaternary Science Reviews

journal homepage: www.elsevier.com/locate/quascirev

Seasonality and desertification drove the global extinction of megafauna in

the late Quaternary
Maurício E. Graipel a,*,1, Matheus S. Lima-Ribeiro b,1, Jorge J. Cherem a ,
Paul R. Momsen Miller c , José A.F. Diniz-Filho d , Nilton C. Cáceres e
a
Departamento de Ecologia e Zoologia, CCB, Universidade Federal de Santa Catarina, 88.040-970, Florianópolis, Santa Catarina, Brazil
b
National Institute for Science and Technology (INCT) in Ecology, Evolution and Biodiversity Conservation, 74.690-900, Goiânia, Goiás, Brazil
c
Departamento de Engenharia Rural, CCA, Universidade Federal de Santa Catarina, 88.034-001, Florianópolis, Santa Catarina, Brazil
d
Departamento de Ecologia, ICB, Universidade Federal de Goiás, 74.690-900, Goiânia, Goiás, Brazil
e
Departamento de Ecologia e Evolução, CCNE, Universidade Federal de Santa Maria, 97.105-900, Santa Maria, Rio Grande do Sul, Brazil

A R T I C L E I N F O A B S T R A C T

Handling Editor: Donatella Magri The causes of megafauna extinction in the late Quaternary have long been a controversial subject, which may be
related to the limitations of adequate information to evaluate global hypotheses related to climate change and
Keywords: the impacts of modern human dispersal. We propose a new global spatio-temporal approach using variations of
Climate change Earth’s orbital parameters and atmospheric CO2 levels as forcing factors for environmental and climate change
Orbital cycles
with worldwide distribution. We analysed the overlap between 142 times-of-extinction of megafauna, and high
Overkill hypothesis
obliquity (as a forcing factor for seasonality), low atmospheric CO2 levels (as a forcing factor for desertification),
Predation
and the arrival of modern humans. We found that critical periods of seasonality and desertification intensified in
the last 800 ka BP, and made the last 50 ka BP exceptionally severe in relation to the entire Quaternary. These
critical periods significantly overlapped with 87% of extinctions in continental and connected islands, compared
with 32.1% overlap of extinctions with the arrival of modern humans. In contrast the arrival of modern humans
on isolated islands overlapped with 90.9% of the extinctions. The arrival of modern humans in continental re­
gions had 81.3% of overlap with critical periods of climate change, suggesting that the synchrony observed
between extinctions and the dispersal of modern humans in continental regions was driven by climate.

1. Introduction resources and increasing competition and predation (Broughton et al.,

2008; Guthrie, 1984; Kiltie et al., 1984; Wolverton et al., 2009).
Terrestrial megafauna (animals >44 kg) suffered global extinction Adverse climate seasonality on population dynamics are one of the
during the late Quaternary as both the climate changed during the last factors that most impacts juvenile individuals of megafauna (Gaillard
glaciation and modern humans colonized every continent (Koch and et al., 1998, 2000; Gaillard and Yoccoz, 2003; Loveridge et al., 2006).
Barnosky, 2006). This critical period of their lives leaves them vulnerable to unfavorable
A previous study conducted by some of the present authors (Araujo seasons which extend for more time than expected (Kiltie et al., 1984;
et al., 2017) found a poor relationship between megafauna extinctions Koch et al., 1989; McCullagh, 1969; Wolverton et al., 2009), increasing
and proxies for mean annual temperature variations when compared to the predation by large predators (Loveridge et al., 2006; Van Val­
the time of the arrival of modern humans. However, the traditional kenburgh et al., 2016), and hunting by humans. Modern populations of
proxies for mean annual temperature do not take into account the effects megaherbivores (>800 kg) are thought to be largely immune to the ef­
of seasonality. Increased seasonality can mean extremely cold winters, fects of predation and this has molded perception of Pleistocene herbi­
which is associated with abrupt climate change (Denton et al., 2005; vores. But young African elephants fall prey to predators principally at
Seersholm et al., 2020). These adverse periods affect species by reducing the end of the dry season, in spite of maternal protection (Van

* Corresponding author.
E-mail addresses: graipel.me@gmail.com (M.E. Graipel), mslima.ribeiro@gmail.com (M.S. Lima-Ribeiro), jjcherem@gmail.com (J.J. Cherem), r.miller@ufsc.br
(P.R. Momsen Miller), jafdinizfilho@gmail.com (J.A.F. Diniz-Filho), niltoncaceres@gmail.com (N.C. Cáceres).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.quascirev.2024.108979
Received 18 March 2024; Received in revised form 20 August 2024; Accepted 18 September 2024
Available online 26 September 2024
0277-3791/© 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

Valkenburgh et al., 2016). Furthermore, when African elephants need to youngest reliable paleontological records (or last appearance dates) and
make larger movements in search of resources during dry years, an oldest reliable archaeological records (or first appearance dates),
unusually large number of young calves become vulnerable to predation respectively. As identification of fossil remains to the species level is not
by large carnivores when they become separated from or are no longer always possible or is often controversial, we established the
able to follow their family groups (Loveridge et al., 2006). Considering time-of-extinction for megafauna at the genus rather than species level.
that the species richness of big carnivores was greater in the Pleistocene To represent the time transgressive nature of both megafauna extinc­
and many of them were significantly larger than their modern coun­ tions and human colonization across continents and islands, we followed
terparts (Van Valkenburgh et al., 2016), harsher seasonal conditions the reasoning of Araujo et al. (2017) and estimated regionalized
make juveniles more susceptible to predators, and have the potential to time-of-extinctions and MHA in 21 regions worldwide (see Table 1; the
lead species to extinction in a few centuries (Brook and Johnson, 2006). complete dataset is available in supplementary material - File A.1).
To capture the effects of extreme seasonality on the extinction of Extinctions that occurred before 90 ka BP were not considered in our
megafauna, this study explores a new approach using variations of analyses due to the large error in the estimated dates. Since the ex­
Earth’s orbital parameters and atmospheric CO2 levels as forcing factors tinctions are categorized by region, we considered the extinction of a
for environmental and climate change with worldwide distribution. The taxa in that region even if it survives elsewhere; if a genera is present as
obliquity was used as a driver for seasonality and the low atmospheric extant and extinct species (e.g., Equus, Bison), it was considered in our
CO2 levels as a driver for desertification. analyses only when the extinct taxa were reliably identified at the spe­
Primary global parameters determining the harshness of climate cies level in original studies (otherwise, living and uncertain identified
seasonality make it possible to compare with equally global megafauna taxa were omitted) (File A.1.). The Caribbean islands used by Araujo
extinction data. Orbital variations determine seasonality, primarily due et al. (2017) were not considered in our analyses due to the lack of
to the tilt of the earth’s axis. Cyclical variations of this tilt lead to harsher reliable data for megafauna extinction and MHA dates, nor were South
or milder seasons, with greater tilt leading to more intense summers and and Southeast Asia considered, due to the lack of reliable megafauna
winters, primarily at higher latitudes in both northern and southern extinction dates (Turvey et al., 2013). Haplomastodon and Notiomastodon
hemispheres (Troost et al., 2009; Zachos et al., 2001). And, when greater were considered distinct genera by Araujo et al. (2017) in northern
tilt is associated with higher total insolation, generated by orbital vari­ South America, but here we consider them to be the same genus
ations, including eccentricity and precession, adverse conditions can be (Lima-Ribeiro et al., 2013), and the oldest record was therefore excluded
intensified (e.g., vegetation fire: Kappenberg et al., 2019; monsoonal from the analysis, because both taxa were being considered for the same
climate: Webster, 1987; harsher seasons: Zachos et al., 2001). region.
Low levels of atmospheric CO2 reduce growth of C3 plants, which All radiocarbon dates for both time-of-extinction and MHA were
become starved from above (Cerling et al., 1998; Ecker et al., 2020), and calibrated using the software Calib Rev 8.1.0, with the IntCal20 curve
when there is no uptake of nutrients in cold soils, the plants starved from for the Northern Hemisphere and SHCal20 curve for the Southern
below (Beerling et al., 2012; Raberg et al., 2024), leading to desertifi­ Hemisphere (Reimer et al., 2020), and considering 95% confidence in­
cation in arid plains and high altitudes and latitudes (Ellis and Palmer, tervals (2σ) around the mean (File A.1).
2016; Johnson et al., 2016) (see File A4). In this scenario, the search for
more suitable foraging regions would be less effective, contributing to
the global demise of megafauna, which were the most vulnerable species 2.2. Orbital dynamics and atmospheric composition
to movement constraints in the late Quaternary (Lima-Ribeiro et al.,
2014). We specifically sought critical primary climatic factors such as high
Here, we investigated if unusual periods of orbital dynamics and CO2
levels generate abnormally high-intensity ecological stressors that Table 1
correlate with megafauna extinction in the late Quaternary on different Summary ages of megafauna extinction and human colonization in the 21
continents and islands. The following stressors were selected: 1) orbital continental and island regions applied in this study. The extinction time interval
represent the dates in which the first and last megafauna taxa became extinct in
data (Laskar et al., 2004, 2011), including high obliquity as a primary
each region. The complete dataset is available in supplementary material (File
driver that generates critical periods of seasonality (Zachos et al., 2001)
A.1).
and environmental changes (Kutzbach, 1981); and 2) low CO2 concen­
tration (Ellis and Palmer, 2016; Johnson et al., 2016) driving critical Continent Region No. of Extinction Time Human
Extinct Interval (cal yr BP) Arrival Date
periods of desertification. To contrast the climatic correlates, we also Taxa (cal yr BP)
considered human predictors in our analyses: 3) modern human arrival
Oceania Australia 11 89,000–44,000 65,000
across the world acting as “super predators”, i.e. the Overkill Hypothesis
Africa East 7 32,920–12,510 174,500
(Martin and Klein, 1984; Saltré et al., 2019). We compared the timing of North 6 37,000–6,130 160,000
fluctuations in seasonality and desertification, by themselves or in as­ South 5 62,900–8,190 97,500
sociation with the timing of modern human arrival, to find the best North Bering 5 17,430–10,990 15,890
explanation for the extinction of megafauna in continental and insular America Central 7 13,090–10,420 13,660
East 12 13,700–9,260 15,260
regions worldwide. West 13 13,600–9,870 23,130
Europe North 7 34,670–11,180 45,660
2. Methods South 5 31,420–16,150 44,310
Asia Central 5 15,790–9,970 45,910
Russia
2.1. Paleontological and archaeological data West Siberia 6 36,190–7,640 17,670
South Andes 11 25,400–9,690 15,010
We gathered paleontological and archaeological records represent­ America Central 8 14,310–7,790 12,920
ing events of megafauna extinctions (from 84 genera) and ‘arrival of North 11 44,130–9,360 14,540
Patagonia 5 13,470–11,710 13,440
modern human’ (MHA) for most continents and several islands from
Connected Japan 5 43,990–16,990 35,780
Araujo et al. (2017), MHA in Australia (Clarkson et al., 2017), and Island Tasmania 7 56,000–34,820 39,940
time-of-extinction and MHA in Africa (time-of-extinction: Faith, 2014; Wrangel 1 4,040 3,610
MHA: Douze et al., 2015; Smith et al., 2007; Tryon and Faith, 2013) (File Isolated Madagascar 6 2,230-140 1,870
A.1). These studies originally filtered high-quality dates, from which we Island New 5 12,290-530 610
Zealand
estimated the time-of-extinction of megafauna and MHA from the

2
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

obliquity, which creates strong seasons, with greatest effects on sea­ the time period between 11,399 and 11,300).
sonality at high latitudes (Ellis and Palmer, 2016; Troost et al., 2009;
Zachos et al., 2001), in the prairies, tundra, savannah and others open 2.3. Testing overlap of variables
habitats where most species of extinct megafauna inhabited (Johnson,
2002) (File A.1; Table A.1). Low CO2 (<200 ppm) contributes to We used a simple statistics, the Z-score, to test the chronological
desertification in marginal environments (Cerling et al., 1998; Ecker overlapping of the time-of-extinction with cpGCC and MHA in an
et al., 2020; Ellis and Palmer, 2016) (Table A.2). The relationship be­ equivalent way (Fig. 1). We also tested the chronological overlapping
tween low atmospheric CO2 levels and reduced net primary productivity between MHA in each of the 21 biogeographic regions and cpGCC in
(NPP) of vegetation is well-established (Prentice and Harrison, 2009; order to know if humans dispersed out of Africa and colonized all con­
Zhang et al., 2024), with cold-adapted C3 plants being most susceptible tinents and large islands during periods associated with orbital cycles
(Ecker et al., 2020). (Chan et al., 2019; Rito et al., 2019) (File A.1).
The variations in obliquity and terrestrial Milankovitch insolation at Z-scores were obtained from the formula: Z = (x – μ)/(σ) for each of
65◦ during the summer solstice in the northern hemisphere were ob­ the three scenarios below:
tained for the entire Quaternary (Laskar et al., 2004), since the northern
hemisphere is considered to be the driving hemisphere for global climate 1. The dates of time-of-extinction as population means (μ) with
change during the Quaternary (De Vleeschouwer et al., 2017). Carbon respective calculated errors (σ) and the closest century of cpGCC as
dioxide (CO2) was recorded from the EPICA (European Project for Ice sample value (x). The null hypothesis predicts no temporal relation
Coring in Antarctica) Dome C ice core covering the last 800 ka BP, where between extinction and cpGCC. The alternative hypothesis predicts
present refers to 1950 AD (Bereiter et al., 2015; Lüthi et al., 2008). extinction occurs during cpGCC.
Prior to the mid-Pleistocene transition, glacial cycles were modu­ 2. The dates of time-of-extinction as population means (μ) with
lated by obliquity on an approximately 41,000 years cycle. However, respective calculated errors (σ) and the closest century of MHA, 2σ
temperature and CO2 during the glacial cycles over the last 800,000 from the mean, as the sample value (x). The null hypothesis predicts
years began to exhibit a quasi-100,000 year’s cycle that mirrors the no temporal relation between extinction and MHA. The alternative
Earth’s approximately 100,000 year’s orbital eccentricity cycle, fluctu­ hypothesis predicts extinctions overlaps with MHA.
ating as a function of orbital insolation changes (Ellis and Palmer, 2016). 3. The dates of MHA as population means (μ) with respective calculated
Seasonality as a function of high obliquity and desertification as a errors (σ) and the closest century of cpGCC as the sample value (x).
function of low CO2 were treated as independent variables, with the 5% The null hypothesis predicts no temporal overlaps between MHA and
of centuries with the highest obliquity and the 5% of centuries with the cpGCC. The alternative hypothesis predicts MHA overlap with
lowest CO2 selected as ‘critical periods of global climate change’ cpGCC.
(cpGCC), being tested considering one-tailed distribution. We selected
time periods within the Quaternary with at least one of the following We used two standard deviations (2σ) for all calibrated radiocarbon
characteristics: 5% of the centuries with the highest values of the dates estimating time-of-extinction or MHA: Z = (x – μ)/(2σ/2); for the
obliquity (>0.422244 radians or 24.19251◦ ; Table A.1); 5% of the other dating methods (U, ESL, OSL) with only 1σ available we used the
centuries with the lowest values of CO2 (<189.55 ppm; Table A.2). We formula: Z = (x – μ)/(σ) for standardizing the analyses (File A.1). Time-
used standardized values (zero to one) for obliquity and CO2. of-extinctions happened in 68.3% of cases before the peaks of cpGCC
In complementary analyses the 5% of centuries with either the (9.5; 25; 48.5; 92 ka BP). Considering that calendar time scale offsets of
combined highest obliquity and highest insolation, and combined lowest up to 1,000 years between climatic event and radiocarbon-calibrated
obliquity and lowest CO2 were tested considering two-tailed distribution megafaunal transitions were also found by Cooper et al. (2015) we
(File A.4). added 1,000 years to the Extinction Error: Z = (x – μ)/(2σ/2 + 1,000) or
We chose to identify the 5% of centuries with the most extreme Z = (x – μ)/(σ +1,000) (Files A.1 and A.2; Table A.3).
values of the entire Quaternary, rather than choosing only those cen­ Besides accounting for possible uncertainties around the relation­
turies whose values are in the lowest or highest 5% of the standardized ships between extinctions and cpGCC (e.g., delayed extinctions), this
scale from zero to one. For example, using 5% of the standardized range procedure also tentatively corrects the Signor-Lipps effects in our ana­
for high obliquity, less than 1% of the centuries of the entire Quaternary lyses (i.e. we are considering that the times-of-extinction may have
would be included as critical values, and using 5% of the standardized extended up to 1000 years after every last appearance date). We
range of low CO2, only 0.5% would be included as critical values. considered the existence of significant temporal overlap between
Centuries were identified by the last year in a given century (11,300 is response and predictor variables when the Z-score was less than 2σ.

Fig. 1. Example of the analysis of the overlap between the time-of-extinction for Diprotodon in Australia and its possible predictors (seasonality,
desertification, and modern human arrival - MHA). Normal distribution for MHA (black curve) and time-of-extinction (blue curve). Horizontal shading on x-axis
represents the 5% of centuries with highest seasonality (red line) and the 5% with highest desertification (orange line). Seasonality and desertification show Z-score
values lower than 2σ and MHA shows a value higher than 2σ. Only seasonality and desertification can therefore be considered eligible as predictors of the extinction
timing of Diprotodon, with seasonality being closer to the mean of time-of-extinction than desertification.

3
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

Thus, the probability of a spurious overlap is p < 0.05 (Fig. 1). 3. Results

2.4. Meta-analysis 3.1. Megafauna, seasonality, desertification and humans

We used a formal meta-analytical approach to test the existence of Our analyses show that ‘critical periods of global climate change’
significant differences between the odds ratios of the number of ex­ (cpGCC) had a major contribution to megafauna extinctions (Fig. 2a) in
tinctions that showed temporal overlap with cpGCC and/or MHA (p < all continental regions (89.1 % of 119 cases) and in connected islands
0.05 in the Z-score test) in continents, grouping the regions of each (66.7% of 12 cases) (Fig. 3a) corresponding to 87% of extinctions in
continent (n = 119 extinctions), and islands: the connected islands (n = these regions (Table A.3), whereas the MHA showed higher contribution
12 extinctions) and the isolated islands (n = 11 extinctions) (File A.1). in isolated islands (90.9% of 11 cases) (Fig. 3b; Table A.3). MHA had low
An island was considered isolated if the sea depth separating it from the overlap with time-of-extinctions in continental regions and connected
mainland was more than 110 m (Faurby and Svenning, 2016). The islands (32.1% of 131 cases) (Fig. 2b and 3b; Table A.3). Our results
heterogeneity test was significant (I2 = 78%; τ2 = 7.2828; p < 0.01), support significant overlap of the time-of-extinction of the most mega­
justifying the use of the random effects model (Everitt and Hothorn, fauna with cpGCC (Fig. 2a and 3a; Table A.3), and show a weak overlap
2010). The statistical meta-analysis was performed using “meta” pack­ between time-of-extinction and MHA globally (Fig. 2b and 3b;
age in R version 3.6.1 (2024-08-08). Table A.3).
Most extinctions on the continents occurred during periods of high
seasonality (68.9%), mainly in Australia, North America, and South
America. Desertification periods played a minor role in the number of

Fig. 2. Timing of climatic predictors (a) and modern human arrival (b) related to megafauna extinction worldwide in the last 200 ka. Orbital variation of
obliquity (blue line), high seasonality (blue vertical bars representing 5% most critical centuries of high obliquity); CO2 levels (orange line) and desertification
(yellow vertical bars representing the 5% most critical centuries of low CO2); mean time of modern human arrival (filled circles overlap critical climatic predictors;
open circles do not overlap climatic predictors) and confidence intervals (black horizontal bars; 2σ); The complete dataset is available in supplementary material
(Files A.1 and A.2).

4
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

Fig. 3. Proportional contribution of climatic predictors (a) and modern human arrival (b) to megafauna extinction by regions in the last 65 ka. a) Pro­
portions of extinctions coeval with seasonality and desertification together (shading of continents and islands) or separately (pie charts); b) Proportions of extinctions
attributed to modern human arrival (MHA) (shading of continents and islands) or separately (pie charts); The complete dataset is available in supplementary material
(Files A.1 and A.2).

Fig. 4. Forest plot synthesizing the odds ratio between chronological association of megafauna extinction time with critical periods of global climate
change (cpGCC; Climatic Hypothesis) or the arrival of modern humans (MHA; Overkill Hypothesis), across continental and insular regions in the world
(Odds Ratio ¼ chance cpGCC/MHA). The vertical full line indicates null effect, size of gray squares represents the regional effect sizes (weight), horizontal lines
through the squares represent 95% confidence intervals (CI) and the diamond represents the summary effect size. The summary effect significantly favors (z = 2.30;
<0.0215) the hypothesis of megafauna extinction driven by Climatic Hypothesis in continental regions and connected islands and absence of synergy (CI does not
overlaps the null effect line); the modern human arrival is favored only on isolated islands.

5
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

continental extinctions (21.8%), being an important factor in Africa and 4. Discussion

Europe. On connected islands, both seasonality and desertification
contributed with 33.3%: high seasonality mainly in Tasmania, and The evolutionary amplification of phenotypes that generated
desertification mainly in Japan (Fig. 2a and 3a; Table A.3). hypermorphic specializations (e.g., large body size and antlers/horns),
The time-of-extinctions at continental scale were synchronous with especially in species with low reproductive rates, are characteristics that
cpGCC and MHA in 26.9% of 119 cases. In only seven cases, MHA have been associated with the extinctions of megafauna in the late
overlapped with extinctions in continental regions, without climate Quaternary, whose species were characterized by Wolverton et al.
change as a contributing cause. Regionally these concurrent contribu­ (2009) as hypermorphic K-strategists. Our analyses clearly show that
tions had a reasonable overlap with time-of-extinctions only in North continental extinctions of hypermorphic K-strategist megafauna and
America (44.1%) and South America (42.9%), with low synchrony in human dispersal are both associated with successive critical periods of
Africa, Asia, Australia, Europe and connected islands (3.6% of 62 ex­ Earth’s orbital variation during the late Quaternary. The traditional
tinctions) (Fig. 3b; Table A.3). proxies for mean annual temperature and precipitation are important
regional predictors to analyses of extinction of megafauna (Řičánková
3.2. Extinctions across continental and insular regions et al., 2015; Román-Palacios and Wiens, 2020; Stewart et al., 2021).
However, they do not take into account the effects of increasing of
The meta-analysis of random effects models provides strong evidence seasonality (Denton et al., 2005; Seersholm et al., 2020).
that global extinctions occurred during cpGCC (Odds Ratio = 11.28; In the late Quaternary there were three waves of megafauna ex­
95% CI = 1.43–88.93) and absence of synergy between cpGCC and tinctions under such conditions. Two waves are marked by the increase
MHA. The Odds Ratio was significantly favorable to climatic factors as in obliquity and Milankovitch insolation of the planet (Fig. 2a and 5),
explanation for extinctions in all continental regions and connected which are associated with intra-annual seasonality (Zachos et al., 2001),
islands, but with a higher number of extinctions during MHA in isolated and drastic regional environmental changes (Gasse, 2000; Schneider
islands (Fig. 4). et al., 2014; Tierney et al., 2017). One occurred between 53.1 and 44.2
ka BP, when the mean insolation remained high (501.3 W/m2) for a long
3.3. Dispersion of modern humans in time and space period of 9,000 years, while the other occurred between 11.3 and 7.6 ka
BP, for a period of 3,800 years. During this second wave of high obliq­
The MHA occurred during cpGCC in continental regions in 81.3% of uity the mean insolation remained even higher (525.8 W/m2), and both
cases, and 43.8% of these events are associated with high seasonality, factors forced the beginning of the interglacial period, the Holocene
mainly in Europe, and 37.5% with desertification, mainly in North (Ellis and Palmer, 2016) (Fig. 2a; Table A.1). Between these two
America and Africa (File A.2). MHA dates occurred closer to cpGCC obliquity waves, a third wave occurred during an exceptional period
dates (n = 21 regions; average ± SD = 2,866 ± 2,829 years) than to the from 31.4 to 18 ka BP with 82 centuries being identified as critically low
time-of-extinctions (19,661 ± 34,484 years, including Africa: n = 142 CO2 (Table A.2). This very short period of time of 45.6 ka (between 53.1
extinction; 7,488 ± 8,694 years, excluding Africa: n = 124 extinctions). and 7.6 ka BP) concentrated 20.9% of all the critical values of the 800 ka
This proximity between MHA dates and cpGCC dates was similar to BP, the period analysed for low CO2 (Fig. 5; Table A.4).
proximity observed between the time-of-extinctions and the closest Together, these three waves, two of high seasonality (high obliquity)
cpGCC dates (1,578 ± 2,007 years, including isolated islands: n = 142 and one of desertification (low CO2) (Fig. 2a and 5; Table A.4), explain
extinctions; 1,187 ± 1,454 years, excluding isolated islands: n = 131 the timing of most extinctions of the late Quaternary in continental re­
extinctions) (File A.1). gions and connected islands (87% of 131 cases; Table A.3).
During the last 800 ka BP, the successive waves of extreme season­
ality became broader and more frequent than the previous periods of the

Fig. 5. Critical periods of seasonality (blue vertical bars) and desertification (yellow vertical bars) during the last million years. Obliquity (blue line) for the
last 1,000 ka BP; atmospheric CO2 levels (orange line) for the last 800 ka BP; arrows indicate the only two periods with three critical waves juxtaposed in close
succession (first wave: high seasonality; second wave: desertification; third wave: high seasonality; source: Table A4; see also Fig. A.1); note increasing over time
(gray line), with the most critical period of climate change occurring during the last three waves.

6
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

Quaternary (Fig. 5; Fig. A.1). The greater proportion of cpGCC (p = The increasing insolation in periods with greater seasonality, mainly
0.46) observed through the juxtaposition of a number of centuries of low between 11.3 and 7.6 ka BP, when mean insolation remained even
CO2 between two very high seasonality in the periods between 53.1 and higher (525.8 W/m2) (see also File A.4), can also be associated with the
7.6 ka BP made this period exceptionally climatically severe throughout increase in extreme climatic conditions such as El Niño-Southern
the last 800 ka BP (Fig. 5 and A.1; Table A.4). Oscillation (ENSO), which usually occurs at intervals of at least three
The period between 330.5 and 375.1 ka BP had fewer critical cen­ years (Tudhope et al., 2001), and could also help to explain the high
turies proportional to the time interval of their duration (p = 0.30). proportion of extinctions among hypermorphic K-strategists who also
Furthermore, the interglacial warming at the end of these periods were need three years or more to become adults (Wolverton et al., 2009).
not as long (~5,000 years) as the Holocene (>10,000 years) (Ellis and Seasonal variations are the main causes of intra-annual mortality of the
Palmer, 2016; Zachos et al., 2001), and the number of low CO2 centuries most vulnerable individuals, especially those who have never reached
was only 36% (29 regarding 82 centuries) of that observed between 53.1 reproductive age, as juveniles, affecting reproductive success and
and 7.6 ka BP (Fig. 5; Table A.4), but also deserve attention in relation to consequently population growth (Gaillard et al., 1998; Wolverton et al.,
extinctions, since they partially reproduced the conditions associated to 2009). For example, North American ungulates that survived the
large part of extinct megafauna observed in the last 50 ka BP. The pe­ extinction period have relatively short gestation periods (none longer
riods of megafauna extinction prior to this 50 ka limit, such as those than 10 months) and early ages at first reproduction (none older than
described for intensively sampled regions of Kings Creek Catchment, three years of age). Furthermore, body mass is related to gestation time
Australia (Wroe et al., 2013), also overlapped with periods of high for living and extinct taxa of ungulates, including shorter gestation times
seasonality (Fig. A.2). However, the greater amplitude of errors associ­ and lower body mass as characteristics of surviving taxa (Wolverton
ated with the estimate of time-of-extinctions, due to limitations of the et al., 2009).
dating technique, makes analyses prior to 50 ka difficult (Stuart, 2015) For living species in Australia and Tasmania, there is also an
(Fig. A.2). observed reduction in the risk of extinction for species with higher
In continental regions and connected islands only 17 genera went reproductive output, such as those that have larger than expected litters
extinct without overlapping with cpGCC. Five extinctions occurred relative to their body size (Cardillo, 2003). There are no studies that
during periods with CO2 below 200 ppm, but above the limit defined by evaluate the effects of ENSO on the age structure of populations leading
our analysis, CO2 < 189.55 ppm (average time below 200 ppm in species to extinction. However, massive mortality of pinniped puppies
extinction interval = 36% of centuries, ranging 9%–58% of centuries), and juveniles, unable to reach their preys, has been reported during El
including Damaliscus in southern Africa, Ursus in westhern Siberia, Niño events along the coast of South America due to the migration of
Palorchestes and Protemnodon in Tasmania, and Coelodonta in northern their preys (Jaksic, 2001). During ENSO-related droughts, strong dis­
Europe. Extinctions that occurred during periods with CO2 less than 200 turbances in the population dynamics of African buffalo (Syncerus caffer)
ppm (n = 5), may be related to the progressive effects of low atmo­ were observed in a semi-arid, mid-latitude savanna (Marshal et al.,
spheric CO2, far above the limit defined by the level of significance of 5% 2011). ENSO may also be related to the extinction of one of the most
of critical centuries (CO2 < 189.55 ppm). In these cases we suggest that emblematic species of mammals, the thylacines (Tasmanian tiger or
desertification be investigated as a cause of extinction or a combination wolf, Thylacinus cynocephalus). ENSO increased duration and intensity of
of forcing factors (File A4), as also observed regarding to MHA in droughts across Australia at the time that thylacines and Tasmanian
Australia to 65 ka BP. devils (Sarcophilus harrisii) became extinct (~3,200 cal yr BP), sug­
Panthera become extinct at 2,853 ± 209 years before of cpGCC in the gesting synergy between climate change and other drivers (such as
central region of Russia, northern Europe and the Bering region in North human intensification or dingo competition/predation) (White et al.,
America. These extinctions represent 60% of records analysed to this 2018).
genus in this study (File A.1). Other five genera also become extinct in The importance of orbital cycles in increasing adverse environmental
six regions before of cpGCC, but more closed of critical periods, at 2,418 conditions during the time that extinctions occurred is also reinforced by
± 209 years, Syncerus in southern Africa, Mylohyus and Palaeolama in unprecedented fire activity observed in the Pleistocene-Holocene tran­
eastern North America, Glossotherium in Andes and central region of sition. (Kappenberg et al. (2019) suggest that fire has been controlled by
South America and Nothrotherium in northern South America (File A.1.). orbital forces and intensified during warmer interglacial periods in 100
These nine extinctions that occurred less than 3,000 years before cpGCC ka intervals during the 600 ka BP, with high seasonality being respon­
could be explained by Signor-Lipps effects, especially in relation to the sible for the increase in fire in terrestrial systems. Climate change and
genus Panthera, once the last fossil records of carnivores tend to occur increased human density could have contributed to fire-regime changes
much earlier than the final extinction of a species (Martin, 1989; at individual sites or at specific times (Marlon et al., 2009; O’Keefe et al.,
Ahlering and Carrel, 2001). 2023).
Three post-cpGCC extinct genera were not explained by our analyses, In addition to the importance of orbital cycles evidenced in this
Zygomaturus in Tasmania, Notiomastodon in northern South America and study, for a better understanding of megafauna extinctions, it also helps
Mammuthus in Wrangel Island. These genera went extinct after cpGCC, to better explain the eventual relationship observed between time-of-
which could be explained, for exemple, by regional environmental al­ extinctions and the MHA. Of 42 extinctions occurring during the pe­
terations after cpGCC, which did not allow adaptation to new condi­ riods of MHA on continents and connected island, 32 also occurred
tions; or may result from Signor-Lipps effects related to posterior periods during cpGCC (Table A.1). This is because MHA occurred during cpGCC
of cpGCC. in 81.3% of the continental regions (File A.2). Considering that climatic
Major contributions to megafauna extinctions found in this study conditions caused human migrations in and out of Africa (Chan et al.,
were high seasonality in Australia, North America, and South America, 2019; Rito et al., 2019) (Fig. 3b), it is plausible to suggest that both MHA
and desertification in Africa, Europe, and Japan (Fig. 3a; Table A.3). and time-of-extinctions were driven by the same underlying climatic
Increasing seasonality was the dominant explanation in the 1980′s for parameters. cpGCC contributed to human dispersal and settlement
the late Quaternary extinctions (Kiltie et al., 1984; Koch et al., 1989). around the globe by creating corridors suitable for migration within
Our analyses also were concentrated in the last glacial period (100 ka Africa about 120 ka BP, or made regions unsuitable for permanence,
BP), which include 31% of the most critical centuries of climate change which would have increased pressure to seek out more climatically
in the last 800 ka (e.g., 30.4% of high seasonality; Table A.1, and 46.1% favorable regions (Chan et al., 2019); by promoting a reduction in sea
of low CO2; Table A.2), mainly affecting the forms most vulnerable to level around 65 ka BP, that created land bridges and reduced distances
extinction (such as the hypermorphic K-strategist taxa) (Geist, 1999; for humans to invade Sahul across South Asia (Clarkson et al., 2017);
Guthrie, 1984; Wolverton et al., 2009). and by generated a warming wave around 13 ka BP, that melted the ice

7
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

and opened a coastal migration route for a wave of Clovis hunters to energy conserved in each level is minimal (Whittaker, 1970), so pred­
invade North America from northeast Asia (Brook and Barnosky, 2012). ators should be rare or less abundant than their prey in community as­
We obtained low temporal overlaps between time-of-extinctions and semblages (Ahlering and Carrel, 2001). Therefore, their chances of
MHA on most continents. However, the synergy between impacts from dying in a fossilizing environment are not very good (Martin, 1989).
changing environments and the hunting pressure that humans must On the other hand, species that evolved on isolated islands, free of
have exerted some time after their arrival and during the increase in predators, would be more sensitive to hunting by humans (Diamond,
population density on different continents cannot be ignored. MHA was 1984; Faurby and Svenning, 2016), especially given the common
more related with extinctions which occurred in North America and occurrence of dwarfism in large mammals living on islands (Raia et al.,
South America, usually in synchrony and possibly in synergy with 2003). However, the hypothesis of extinction due to MHA on islands
cpGCC (Fig. 3b; Table A.3). This is because climate change promoted around the world is not supported by the significant and detrimental
changes in species distributions, potentially resulting in regional turn­ impact of these humans on insular biotas following island colonization
over, including dispersions and extinctions (Cooper et al., 2015). prior to the Holocene, suggesting that models using island extinctions as
Although studies indicate that climate has caused declines in North evidence in support to anthropogenic megafauna overhunting need to be
America’s megafauna populations (Stewart et al., 2021), the synchrony reconsidered (Louys et al., 2021).
and possible synergy with hunting by modern human should be inves­ The cpGCC that drove the megafauna to extinction in continental
tigated case by case (Broughton and Weitzel, 2018). regions must have been similar to those observed on connected islands,
Africa has been considered a separate case, with a few extinctions in such as Japan, where 100% of extinctions occurred during cpGCC,
the late Quaternary, being associated to co-evolution of megafauna with which were previously explained as changes of the ecosystem driven by
humans (Kelt and Meyer, 2009). However, the number of extinct taxa in climatic changes (Iwase et al., 2012). The timing of the extinctions in
Africa was not as low as suggested (24 large mammals >5 kg) (Faith, Tasmania overlapped with both cpGCC (50%) and MHA (33.3%)
2014). In subtropical and tropical regions, as is the case of large (Table A.3). This time coinciding with the establishment of a connection
extension of Africa, extinctions would not necessarily be subjected to the between Australia and Tasmania via the Bassian landbridge between 43
increased seasonality (Kiltie et al., 1984; Wolverton et al., 2009). Our and 37 ka BP (Turney et al., 2008), with climate being the most plausible
results suggest desertification (61.1% of extinctions) as the main causal driver of megafauna extinction (Lima-Ribeiro and Diniz-Filho, 2014;
factors of extinctions in Africa (Fig. 3a; Table A.3), especially in East Saltré et al., 2019).
Africa (85.7%) when dry conditions prevailed in both hemispheres The megafauna resilience to extinction may have been put to a strong
during the Last Glacial Maximum (Gasse, 2000), but also in North Africa test spatially and temporally during extended interglacial periods, such
(66.7%). High seasonality explained 33.3% of extinctions in north Af­ as the Holocene that started 11.65 ka BP (Walker et al., 2009; Zachos
rica, at the threshold of the Green Sahara event (deMenocal et al., 2000; et al., 2001). This mechanism, associated with orbital cycles (Ellis and
Tierney et al., 2017), and 40% in southern Africa, furthest from tropical Palmer, 2016), reduces the physical size of island environments due to
zone (File A.1). In fact, although the number of extinct species is not as marine transgressions (Fernández-Palacios et al., 2016). During inter­
low (Faith, 2014) as previously commented (Klein, 1984), Africa should glacial periods, the reduction in area of the islands (connected with the
be considered a “fortunate anomaly” (Koch and Barnosky, 2006) continent during the glacial period) may have contributed to the ex­
because of the large number of surviving megafauna species present tinctions of large predators (Faurby and Svenning, 2016) and would
during the last glacial period, 75% (Faith, 2014). This “fortunate represent refuge for large remaining herbivores. In such cases, such
anomaly” can be explained as a consequence of having large areas of islands would be sources for reverse colonization to the continents
sub-tropical and tropical savannas in both hemispheres, allowing (Bellemain and Ricklefs, 2008), by species impacted for the long periods
migration or dispersion towards large areas in one or other hemisphere of adverse climate. For example, the Wrangel woolly mammoth, which
during alternating unfavorable seasons (or usually out of phase; see experienced genetic decline and fixation of deleterious mutations due to
seesaw effect) (Stocker, 1998; Stocker, 1998). Therefore, a lower num­ isolation and small population size (Fry et al., 2020) and reduction in
ber of extinctions is expected in Africa during cpGCC, except during long water quality (Arppe et al., 2019), only disappeared less than 4 ka (El
global periods of desertification and low seasonality in both hemi­ Adli et al., 2017). It could still be coexisting with us if a normal inter­
spheres (18–31.4 ka BP), as happened during the period when many glacial period (~5 ka) had come to an end and a new glacial period had
African extinctions occurred (Fig. 3a; Table A.3). connected the island with the continent.
Furthermore, it is plausible that large predators played an important The megafauna extinctions in isolated islands evaluated in our study
role when associated with cpGCC (Loveridge et al., 2006; Van Val­ occurred in 90.9% of the cases during the Holocene (Table A.3).
kenburgh et al., 2016). The increase in predation pressure on juveniles Nevertheless, these extinctions did not necessarily happen due to direct
by large predators may also have been different on each continent (Kiltie action of hunters/gatherers. In Madagascar, for example, the “Subsis­
et al., 1984; Loveridge et al., 2006; Van Valkenburgh et al., 2016; tence Shift Hypothesis” (Godfrey et al., 2019) suggests that hunting/­
Wolverton et al., 2009) during cpGCC. The largest extant African foraging was not responsible for extinctions, as humans coexisted for a
predator, Panthera leo (~160 kg), is relatively small in size when long time with the megafauna. However, the extinctions occurred very
compared to extinct cats from other continents (Smith et al., 2003), yet quickly after humans changed their subsistence activity to herding/­
has a large effect on juvenile hyperherbivores during long periods of farming, which promoted great environmental changes and increased
extended drought (Loveridge et al., 2006). This difference in size is most human population on the island (Louys et al., 2021). Furthermore, in the
evident in South America, where the largest number of extinct predator South Island of New Zealand, a similar shift occurred after the intro­
species was recorded (Nogués-Bravo et al., 2010), being associated with duction of the sweet potato by trans-Pacific carriage, expanding existing
the occurrence of the biggest predators in the late Pleistocene: Smilodon horticultural systems to colder and drier environments on the South
fatalis ~400 kg, P. atrox ~430 kg and S. populator ~440 kg (Bocherens Island (Godfrey et al., 2019). Several high severity fire events were
et al., 2016; Chimento and Agnolin, 2017; Smith et al., 2003). associated with the first two centuries of Māori horticultural expansion
It is worth considering that, in some cases, the disappearance of (700–800 cal yr BP) accompanied by a transformation of the vegetation
predators in the fossil record at the same time as their prey (O’Keefe (McWethy et al., 2010) close to the time of some large-bird extinctions
et al., 2023) should not be considered as a factor that rules out the (Anderson and Petchey, 2020) (File A.1). In addition to this scenario, the
predator’s contribution to the extinction of its prey, as previously dis­ high levels of consumption of eggs and adult males of moas (who were
cussed for extinction of Panthera in Europe, Asia and North America. A responsible for hatching the eggs) by hunters were inferred in archae­
principle of animal ecology is that the number of individuals at a given ological deposits in the South Island of New Zealand. Thus, the high
trophic level will decrease with increasing trophic rank because the level of egg exploitation recorded may have placed the moas under an

8
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

extreme and unsustainable reproductive stress (Oskam et al., 2012), a since fictions can lead to misinterpretations, as Sam Magruder would
scenario that would be comparable to the effect of predation by large discover if he had gone to the future, I decided to study the population
predators under critical conditions of climate change acting on juvenile dynamics of actual hyper herbivores and the conditions that could
megafauna mammals in continental regions. extirpate populations. The predator-prey interaction clearly demon­
We did not find a significant spatiotemporal interaction between strates how young individuals of hyper herbivores become vulnerable to
MHA and megafauna extinctions in continental regions and connected African lions during long periods of extreme drought. And when the
islands (Fig. 3b), also in accordance with several studies (Carotenuto same age class is suppressed seasonally, year after year, populations can
et al., 2018; Lima-Ribeiro and Felizola Diniz-Filho, 2013; Louys et al., be drastically reduced by not incorporating reproductively active in­
2021; Saltré et al., 2019; Stewart et al., 2021). Even in the cradle of the dividuals. So, for years I looked for data that represented the seasonality
Overkill Hypothesis, in North America, we found significantly more and drought conditions at the time the extinctions occurred. That’s why
temporal overlap between time-of-extinctions and cpGCC (91.2%) than I also thank Michael Palmer who, in a forum, proposed to help Ralph
between time-of-extinctions and MHA (52.9%), usually in synchrony Ellis publish "the best explanation about what triggers glaciations"; with
with cpGCC (Table A.1). Therefore, our study made it clear that the this study I learned a lot about the importance of orbital cycles I use in
impact of MHA on megafauna cannot be reduced to a simple chrono­ this study. I also thank my brother, Rafael Graipel, and several re­
logical overlap with the timing of extinctions. searchers with whom I had the opportunity to discuss the extinction of
However, studies suggest that synergy between the factors must have megafauna.
occurred and may have been the coup de grâce in some cases
(Lima-Ribeiro et al., 2013; Mondanaro et al., 2019). This indicates that Appendix A. Supplementary data
these spatio-temporal interaction effects are sometimes more difficult to
capture in general analyses, requiring case-by-case analysis, especially Supplementary data to this article can be found online at https://doi.
considering the increase in human population density and reduction of org/10.1016/j.quascirev.2024.108979.
essential resources for certain megafauna species. This may have been
the case for the suppression of suitable habitats due to the shift in the References
subsistence mode from hunting/gathering to herding/farming in iso­
lated islands (Godfrey et al., 2019) or the occupation of caves by humans Ahlering, M.A., Carrel, J.E., 2001. Predators are rare even when they are small. Oikos 95,
471–475. https://doi.org/10.1034/j.1600-0706.2001.950312.x.
formerly used by Ursus in Europe (Mondanaro et al., 2019).
Anderson, A., Petchey, F., 2020. The transfer of kumara (Ipomoea batatas) from east to
By linking Earth’s orbital cycles and atmospheric CO2 levels with south polynesia and its dispersal in New Zealand. J. Polyn. Soc. 129, 351–382.
seasonality and desertification, and with the demise of megafauna, our https://doi.org/10.15286/JPS.129.4.351-382.
study suggests that the late Quaternary had the most acute series of Araujo, B.B.A., Oliveira-Santos, L.G.R., Lima-Ribeiro, M.S., Diniz-Filho, J.A.F.,
Fernandez, F.A.S., 2017. Bigger kill than chill: the uneven roles of humans and
successive critical periods, which exerted great selective pressure on the climate on late Quaternary megafaunal extinctions. Quat. Int. 431, 216–222. https://
megafauna, being primarily responsible for the extinctions as well as doi.org/10.1016/j.quaint.2015.10.045.
human dispersal. Arppe, L., Karhu, J.A., Vartanyan, S., Drucker, D.G., Etu-Sihvola, H., Bocherens, H.,
2019. Thriving or surviving? The isotopic record of the Wrangel Island woolly
mammoth population. Quat. Sci. Rev. 222, 105884. https://doi.org/10.1016/j.
CRediT authorship contribution statement quascirev.2019.105884.
Beerling, D.J., Taylor, L.L., Bradshaw, C.D.C., Lunt, D.J., Valdes, P.J., Banwart, S.A.,
Pagani, M., Leake, J.R., 2012. Ecosystem CO2 starvation and terrestrial silicate
Maurício E. Graipel: Conceptualization, Data curation, Formal weathering: mechanisms and global-scale quantification during the late Miocene.
analysis, Investigation, Methodology, Writing – original draft, Writing – J. Ecol. 100, 31–41. https://doi.org/10.1111/j.1365-2745.2011.01905.x.
review & editing. Matheus S. Lima-Ribeiro: Data curation, Investiga­ Bellemain, E., Ricklefs, R., 2008. Are islands the end of the colonization road? Trends
Ecol. Evol. 23, 461–468. https://doi.org/10.1016/j.tree.2008.05.001.
tion, Validation, Writing – original draft, Writing – review & editing. Bereiter, B., Eggleston, S., Schmitt, J., Nehrbass-Ahles, C., Stocker, T.F., Fischer, H.,
Jorge J. Cherem: Writing – original draft, Writing – review & editing. Kipfstuhl, S., Chappellaz, J., 2015. Revision of the EPICA Dome C CO 2 record from
Paul R. Momsen Miller: Validation, Writing – original draft, Writing – 800 to 600 kyr before present. Geophys. Res. Lett. 42, 542–549. https://doi.org/
10.1002/2014GL061957.
review & editing. José A.F. Diniz-Filho: Validation, Writing – original
Bocherens, H., Cotte, M., Bonini, R., Scian, D., Straccia, P., Soibelzon, L., Prevosti, F.J.,
draft, Writing – review & editing. Nilton C. Cáceres: Validation, 2016. Paleobiology of sabretooth cat Smilodon populator in the pampean region
Writing – original draft, Writing – review & editing. (Buenos Aires Province, Argentina) around the last glacial maximum: insights from
carbon and nitrogen stable isotopes in bone collagen. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 449, 463–474. https://doi.org/10.1016/j.palaeo.2016.02.017.
Declaration of competing interest Brook, B.W., Barnosky, A.D., 2012. Quaternary extinctions and their link to climate
change. In: Saving a Million Species. Island Press/Center for Resource Economics,
The authors declare that they have no known competing financial Washington, DC, pp. 179–198. https://doi.org/10.5822/978-1-61091-182-5_11.
Brook, B.W., Johnson, C.N., 2006. Selective hunting of juveniles as a cause of the
interests or personal relationships that could have appeared to influence imperceptible overkill of the Australian Pleistocene megafauna. Alcheringa an
the work reported in this paper. Australas. J. Palaeontol. 30, 39–48. https://doi.org/10.1080/03115510609506854.
Broughton, J.M., Byers, D.A., Bryson, R.A., Eckerle, W., Madsen, D.B., 2008. Did climatic
seasonality control late Quaternary artiodactyl densities in western North America?
Data availability Quat. Sci. Rev. 27, 1916–1937. https://doi.org/10.1016/j.quascirev.2008.07.005.
Broughton, J.M., Weitzel, E.M., 2018. Population reconstructions for humans and
I have shared the link to my data at the attached file step megafauna suggest mixed causes for North American Pleistocene extinctions. Nat.
Commun. 9, 5441. https://doi.org/10.1038/s41467-018-07897-1.
Cardillo, M., 2003. Biological determinants of extinction risk: why are smaller species
Acknowledgements less vulnerable? Anim. Conserv. 6, 63–69. https://doi.org/10.1017/
S1367943003003093.
Carotenuto, F., Di Febbraro, M., Melchionna, M., Mondanaro, A., Castiglione, S.,
We are grateful for the relevant comments of three anonymous re­ Serio, C., Rook, L., Loy, A., Lima-Ribeiro, M.S., Diniz-Filho, J.A.F., Raia, P., 2018.
viewers and the Editor. I (M.E.G.) would like to thank George Gaylord The well-behaved killer: late Pleistocene humans in Eurasia were significantly
Simpson. At the end of the 1980s, when I was starting biology college, I associated with living megafauna only. Palaeogeogr. Palaeoclimatol. Palaeoecol.
500, 24–32. https://doi.org/10.1016/j.palaeo.2018.03.036.
read a curious story about mammals that were isolated in South America
Cerling, T.E., Ehleringer, J.R., Harris, J.M., 1998. Carbon dioxide starvation, the
(Splendid isolation: the curious history of South American mammals, development of C4 ecosystems, and mammalian evolution. Philos. Trans. R. Soc.
1980). What led to the disappearance of these splendid mammals has London. Ser. B Biol. Sci. 353, 159–171. https://doi.org/10.1098/rstb.1998.0198.
remained in my reading and thinking for more than 30 years. I realized Chan, E.K.F., Timmermann, A., Baldi, B.F., Moore, A.E., Lyons, R.J., Lee, S.S.,
Kalsbeek, A.M.F., Petersen, D.C., Rautenbach, H., Förtsch, H.E.A., Bornman, M.S.R.,
in 2012 that a time machine was not necessary to understand what the Hayes, V.M., 2019. Human origins in a southern African palaeo-wetland and first
fossils were saying (The Dechronization of Sam Magruder, 1996). And migrations. Nature 575, 185–189. https://doi.org/10.1038/s41586-019-1714-1.

9
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

Chimento, N.R., Agnolin, F.L., 2017. The fossil American lion (Panthera atrox) in South Johnson, C.N., Rule, S., Haberle, S.G., Kershaw, A.P., McKenzie, G.M., Brook, B.W.,
America: palaeobiogeographical implications. Comptes Rendus Palevol 16, 850–864. 2016. Geographic variation in the ecological effects of extinction of Australia’s
https://doi.org/10.1016/j.crpv.2017.06.009. Pleistocene megafauna. Ecography 39, 109–116. https://doi.org/10.1111/
Clarkson, C., Jacobs, Z., Marwick, B., Fullagar, R., Wallis, L., Smith, M., Roberts, R.G., ecog.01612.
Hayes, E., Lowe, K., Carah, X., Florin, S.A., McNeil, J., Cox, D., Arnold, L.J., Hua, Q., Kappenberg, A., Lehndorff, E., Pickarski, N., Litt, T., Amelung, W., 2019. Solar controls of
Huntley, J., Brand, H.E.A., Manne, T., Fairbairn, A., Shulmeister, J., Lyle, L., fire events during the past 600,000 years. Quat. Sci. Rev. 208, 97–104. https://doi.
Salinas, M., Page, M., Connell, K., Park, G., Norman, K., Murphy, T., Pardoe, C., org/10.1016/j.quascirev.2019.02.008.
2017. Human occupation of northern Australia by 65,000 years ago. Nature 547, Kelt, D.A., Meyer, M.D., 2009. Body size frequency distributions in African mammals are
306–310. https://doi.org/10.1038/nature22968. bimodal at all spatial scales. Glob. Ecol. Biogeogr. 18, 19–29. https://doi.org/
Cooper, A., Turney, C., Hughen, K.A., Brook, B.W., McDonald, H.G., Bradshaw, C.J.A., 10.1111/j.1466-8238.2008.00422.x.
2015. Abrupt warming events drove Late Pleistocene Holarctic megafaunal turnover. Kiltie, R.A., Martin, P.S., Klein, R.G., 1984. Seasonality, gestation time, and large
Science 349, 602–606. https://doi.org/10.1126/science.aac4315. mammal extinctions. In: Martin, P.S., Klein, R.G. (Eds.), Quaternary Extinctions: A
De Vleeschouwer, D., Vahlenkamp, M., Crucifix, M., Pälike, H., 2017. Alternating Prehistoric Revolution. University of Arizona Press, Tucson, pp. 299–314.
Southern and Northern Hemisphere climate response to astronomical forcing during Klein, R.G., 1984. Mammalian extinctions and stone age people in Africa. In: Martin, P.S.,
the past 35 m.y. Geology 45, 375–378. https://doi.org/10.1130/G38663.1. Klein, R.G. (Eds.), Quaternary Extinctions: A Prehistoric Revolution. University of
deMenocal, P., Ortiz, J., Guilderson, T., Adkins, J., Sarnthein, M., Baker, L., Arizona Press, Tucson, pp. 553–573.
Yarusinsky, M., 2000. Abrupt onset and termination of the African humid period. Koch, P.L., Barnosky, A.D., 2006. Late quaternary extinctions: state of the debate. Annu.
Quat. Sci. Rev. 19, 347–361. https://doi.org/10.1016/S0277-3791(99)00081-5. Rev. Ecol. Evol. Syst. 37, 215–250. https://doi.org/10.1146/annurev.
Denton, G.H., Alley, R.B., Comer, G.C., Broecker, W.S., 2005. The role of seasonality in ecolsys.34.011802.132415.
abrupt climate change. Quat. Sci. Rev. 24, 1159–1182. https://doi.org/10.1016/j. Koch, P.L., Fisher, D.C., Dettman, D., 1989. Oxygen isotope variation in the tusks of
quascirev.2004.12.002. extinct proboscideans: a measure of season of death and seasonality. Geology 17,
Diamond, J.M., 1984. Historic extinction: a Rosetta stone for understanding prehistoric 515–519. https://doi.org/10.1130/0091-7613(1989)017<0515:OIVITT>2.3.CO;2.
extinctions. In: Martin, P.S., Klein, R.G. (Eds.), Quaternary Extinctions: A Prehistoric Kutzbach, J.E., 1981. Monsoon climate of the early Holocene: climate experiment with
Revolution. University of Arizona Press, Tucson, pp. 824–862. the Earth’s orbital parameters for 9000 Years ago. Science 214, 59–61. https://doi.
Douze, K., Wurz, S., Henshilwood, C.S., 2015. Techno-cultural characterization of the org/10.1126/science.214.4516.59.
MIS 5 (c. 105 – 90 Ka) lithic industries at blombos cave, Southern cape, South Africa. Laskar, J., Fienga, A., Gastineau, M., Manche, H., 2011. La2010: a new orbital solution
PLoS One 10, e0142151. https://doi.org/10.1371/journal.pone.0142151. for the long-term motion of the Earth. Astron. Astrophys. 532, A89. https://doi.org/
Ecker, M., Kelley, D., Sato, H., 2020. Modelling the effects of CO2 on C3 and C4 grass 10.1051/0004-6361/201116836.
competition during the mid-Pleistocene transition in South Africa. Sci. Rep. 10, Laskar, J., Robutel, P., Joutel, F., Gastineau, M., Correia, A.C.M., Levrard, B., 2004.
16234. https://doi.org/10.1038/s41598-020-72614-2. A long-term numerical solution for the insolation quantities of the Earth. Astron.
El Adli, J.J., Fisher, D.C., Vartanyan, S.L., Tikhonov, A.N., 2017. Final years of life and Astrophys. 428, 261–285. https://doi.org/10.1051/0004-6361:20041335.
seasons of death of woolly mammoths from Wrangel Island and mainland Chukotka, Lima-Ribeiro, M.S., Diniz-Filho, J.A.F., 2014. Obstinate Overkill in Tasmania? The
Russian Federation. Quat. Int. 445, 135–145. https://doi.org/10.1016/j. closest gaps do not probabilistically support human involvement in megafaunal
quaint.2016.07.017. extinctions. Earth-Science Rev. 135, 59–64. https://doi.org/10.1016/j.
Ellis, R., Palmer, M., 2016. Modulation of ice ages via precession and dust-albedo earscirev.2014.04.005.
feedbacks. Geosci. Front. 7, 891–909. https://doi.org/10.1016/j.gsf.2016.04.004. Lima-Ribeiro, M.S., Felizola Diniz-Filho, J.A., 2013. American megafaunal extinctions
Everitt, B., Hothorn, T., 2010. Meta-analysis. In: Everitt, B., Hothorn, T. (Eds.), and human arrival: improved evaluation using a meta-analytical approach. Quat. Int.
A Handbook of Statistical Analyses Using R. Taylor and Francis Group, LLC, 299, 38–52. https://doi.org/10.1016/j.quaint.2013.03.007.
pp. 267–279. Lima-Ribeiro, M.S., Hortal, J., Varela, S., Diniz-Filho, J.A.F., 2014. Constraint envelope
Faith, J.T., 2014. Late Pleistocene and Holocene mammal extinctions on continental analyses of macroecological patterns reveal climatic effects on Pleistocene mammal
Africa. Earth-Science Rev. 128, 105–121. https://doi.org/10.1016/j. extinctions. Quat. Res. 82, 260–269. https://doi.org/10.1016/j.yqres.2014.02.003.
earscirev.2013.10.009. Lima-Ribeiro, M.S., Nogués-Bravo, D., Terribile, L.C., Batra, P., Diniz-Filho, J.A.F., 2013.
Faurby, S., Svenning, J.-C., 2016. Resurrection of the island rule: human-driven Climate and humans set the place and time of Proboscidean extinction in late
extinctions have obscured a basic evolutionary pattern. Am. Nat. 187, 812–820. Quaternary of South America. Palaeogeogr. Palaeoclimatol. Palaeoecol. 392,
https://doi.org/10.1086/686268. 546–556. https://doi.org/10.1016/j.palaeo.2013.10.008.
Fernández-Palacios, J.M., Rijsdijk, K.F., Norder, S.J., Otto, R., Nascimento, L., Louys, J., Braje, T.J., Chang, C.-H., Cosgrove, R., Fitzpatrick, S.M., Fujita, M.,
Fernández-Lugo, S., Tjørve, E., Whittaker, R.J., Santos, A., 2016. Towards a glacial- Hawkins, S., Ingicco, T., Kawamura, A., MacPhee, R.D.E., McDowell, M.C.,
sensitive model of island biogeography. Glob. Ecol. Biogeogr. 25, 817–830. https:// Meijer, H.J.M., Piper, P.J., Roberts, P., Simmons, A.H., van den Bergh, G., van der
doi.org/10.1111/geb.12320. Geer, A., Kealy, S., O’Connor, S., 2021. No evidence for widespread island
Fry, E., Kim, S.K., Chigurapti, S., Mika, K.M., Ratan, A., Dammermann, A., Mitchell, B.J., extinctions after Pleistocene hominin arrival. Proc. Natl. Acad. Sci. 118. https://doi.
Miller, W., Lynch, V.J., 2020. Functional architecture of deleterious genetic variants org/10.1073/pnas.2023005118.
in the genome of a Wrangel island mammoth. Genome Biol. Evol. 12, 48–58. https:// Loveridge, A.J., Hunt, J.E., Murindagomo, F., Macdonald, D.W., 2006. Influence of
doi.org/10.1093/gbe/evz279. drought on predation of elephant (Loxodonta africana) calves by lions (Panthera leo)
Gaillard, J.-M., Festa-Bianchet, M., Yoccoz, N.G., 1998. Population dynamics of large in an African wooded savannah. J. Zool. 270, 523–530. https://doi.org/10.1111/
herbivores: variable recruitment with constant adult survival. Trends Ecol. Evol. 13, j.1469-7998.2006.00181.x.
58–63. https://doi.org/10.1016/S0169-5347(97)01237-8. Lüthi, D., Le Floch, M., Bereiter, B., Blunier, T., Barnola, J.-M., Siegenthaler, U.,
Gaillard, J.-M., Festa-Bianchet, M., Yoccoz, N.G., Loison, A., Toïgo, C.T., 2000. Temporal Raynaud, D., Jouzel, J., Fischer, H., Kawamura, K., Stocker, T.F., 2008. High-
variation in fitness components and population dynamics of large herbivores. Annu. resolution carbon dioxide concentration record 650,000–800,000 years before
Rev. Ecol. Systemat. 31, 367–393. https://doi.org/10.1146/annurev. present. Nature 453, 379–382. https://doi.org/10.1038/nature06949.
ecolsys.31.1.367. Marlon, J.R., Bartlein, P.J., Walsh, M.K., Harrison, S.P., Brown, K.J., Edwards, M.E.,
Gaillard, J.-M., Yoccoz, N.G., 2003. Temporal variation in survival of mammals: a case of Higuera, P.E., Power, M.J., Anderson, R.S., Briles, C., Brunelle, A., Carcaillet, C.,
environmental canalization? Ecology 84, 3294–3306. https://doi.org/10.1890/02- Daniels, M., Hu, F.S., Lavoie, M., Long, C., Minckley, T., Richard, P.J.H., Scott, A.C.,
0409. Shafer, D.S., Tinner, W., Umbanhowar, C.E., Whitlock, C., 2009. Wildfire responses
Gasse, F., 2000. Hydrological changes in the African tropics since the last glacial to abrupt climate change in North America. Proc. Natl. Acad. Sci. 106, 2519–2524.
maximum. Quat. Sci. Rev. 19, 189–211. https://doi.org/10.1016/S0277-3791(99) https://doi.org/10.1073/pnas.0808212106.
00061-X. Marshal, J.P., Owen-Smith, N., Whyte, I.J., Stenseth, N.C., 2011. The role of El
Geist, V., 1999. Periglacial ecology, large mammals, and their significance to human Niño–Southern Oscillation in the dynamics of a savanna large herbivore population.
biology. In: Bonnichsen, R., Turnmire, K.L. (Eds.), Ice Age Peoples of North America: Oikos 120, 1175–1182. https://doi.org/10.1111/j.1600-0706.2010.19155.x.
Environments, Origins, and Adaptations of the First Americans. Oregon State Martin, L.D., 1989. Fossil history of the terrestrial Carnivora. In: Carnivore Behavior,
University Press, Corvallis, pp. 78–94. Ecology, and Evolution. Springer, US, Boston, MA, pp. 536–568. https://doi.org/
Godfrey, L.R., Scroxton, N., Crowley, B.E., Burns, S.J., Sutherland, M.R., Pérez, V.R., 10.1007/978-1-4757-4716-4_20.
Faina, P., McGee, D., Ranivoharimanana, L., 2019. A new interpretation of Martin, P.S., Klein, R.G., 1984. Prehistoric overkill: the global model. In: Martin, P.S.,
Madagascar’s megafaunal decline: the “Subsistence Shift Hypothesis.”. J. Hum. Evol. Klein, R.G. (Eds.), Quaternary Extinctions: A Prehistoric Revolution. University of
130, 126–140. https://doi.org/10.1016/j.jhevol.2019.03.002. Arizona Press, Tucson, pp. 354–403.
Guthrie, R.D., 1984. Alaskan megabucks, megabulls, and megarams: the issue of McCullagh, K., 1969. The growth and nutrition of the african elephant. Afr. J. Ecol. 7,
Pleistocene gigantism. Carnegie Museum Nat. Hist. Spec. Rep. 8, 482–510. 85–90. https://doi.org/10.1111/j.1365-2028.1969.tb01196.x.
Iwase, A., Hashizume, J., Izuho, M., Takahashi, K., Sato, H., 2012. Timing of megafaunal McWethy, D.B., Whitlock, C., Wilmshurst, J.M., McGlone, M.S., Fromont, M., Li, X.,
extinction in the late late Pleistocene on the Japanese archipelago. Quat. Int. 255, Dieffenbacher-Krall, A., Hobbs, W.O., Fritz, S.C., Cook, E.R., 2010. Rapid landscape
114–124. https://doi.org/10.1016/j.quaint.2011.03.029. transformation in South Island, New Zealand, following initial Polynesian
Jaksic, F.M., 2001. Ecological effects of El Niño in terrestrial ecosystems of western South settlement. Proc. Natl. Acad. Sci. 107, 21343–21348. https://doi.org/10.1073/
America. Ecography 24, 241–250. https://doi.org/10.1111/j.1600-0587.2001. pnas.1011801107.
tb00196.x. Mondanaro, A., Di Febbraro, M., Melchionna, M., Carotenuto, F., Castiglione, S.,
Johnson, C.N., 2002. Determinants of loss of mammal species during the Late Quaternary Serio, C., Danisi, S., Rook, L., Diniz-Filho, J.A.F., Raia, P., 2019. Additive effects of
“megafauna” extinctions: life history and ecology, but not body size. Proc. R. Soc. B climate change and human hunting explain population decline and extinction in
Biol. Sci. 269, 2221–2227. https://doi.org/10.1098/rspb.2002.2130. cave bears. Boreas 48, 605–615. https://doi.org/10.1111/bor.12380.

10
M.E. Graipel et al. Quaternary Science Reviews 344 (2024) 108979

Nogués-Bravo, D., Ohlemüller, R., Batra, P., Araújo, M.B., 2010. Climate predictors of sapiens. Proc. Natl. Acad. Sci. 104, 6128–6133. https://doi.org/10.1073/
late quaternary extinctions. Evolution (N. Y) 64, 2442–2449. https://doi.org/ pnas.0700747104.
10.1111/j.1558-5646.2010.01009.x. Stewart, M., Carleton, W.C., Groucutt, H.S., 2021. Climate change, not human population
O’Keefe, F.R., Dunn, R.E., Weitzel, E.M., Waters, M.R., Martinez, L.N., Binder, W.J., growth, correlates with Late Quaternary megafauna declines in North America. Nat.
Southon, J.R., Cohen, J.E., Meachen, J.A., DeSantis, L.R.G., Kirby, M.E., Ghezzo, E., Commun. 12, 965. https://doi.org/10.1038/s41467-021-21201-8.
Coltrain, J.B., Fuller, B.T., Farrell, A.B., Takeuchi, G.T., MacDonald, G., Davis, E.B., Stocker, T.F., 1998. The seesaw effect. Science 282 (61), 61–62. https://doi.org/
Lindsey, E.L., 2023. Pre–Younger Dryas megafaunal extirpation at Rancho La Brea 10.1126/science.282.5386.
linked to fire-driven state shift. Science 381. https://doi.org/10.1126/science. Stuart, A.J., 2015. Late Quaternary megafaunal extinctions on the continents: a short
abo3594. review. Geol. J. 50, 338–363. https://doi.org/10.1002/gj.2633.
Oskam, C.L., Allentoft, M.E., Walter, R., Scofield, R.P., Haile, J., Holdaway, R.N., Tierney, J.E., Pausata, F.S.R., deMenocal, P.B., 2017. Rainfall regimes of the green
Bunce, M., Jacomb, C., 2012. Ancient DNA analyses of early archaeological sites in Sahara. Sci. Adv. 3. https://doi.org/10.1126/sciadv.1601503.
New Zealand reveal extreme exploitation of moa (Aves: Dinornithiformes) at all life Troost, T.A., van Dam, J.A., Kooi, B.W., Tuenter, E., 2009. Seasonality, climate cycles
stages. Quat. Sci. Rev. 52, 41–48. https://doi.org/10.1016/j.quascirev.2012.07.007. and body size evolution. Math. Model Nat. Phenom. 4, 135–155. https://doi.org/
Prentice, I.C., Harrison, S.P., 2009. Ecosystem effects of CO2 concentration: evidence 10.1051/mmnp/20094605.
from past climates. Clim. Past 5, 297–307. https://doi.org/10.5194/cp-5-297-2009. Tryon, C.A., Faith, J.T., 2013. Variability in the middle stone age of eastern Africa. Curr.
Raberg, J.H., Crump, S.E., de Wet, G., Harning, D.J., Miller, G.H., Geirsdóttir, Á., Anthropol. 54, S234–S254. https://doi.org/10.1086/673752.
Sepúlveda, J., 2024. BrGDGT lipids in cold regions reflect summer soil temperature Tudhope, A.W., Chilcott, C.P., McCulloch, M.T., Cook, E.R., Chappell, J., Ellam, R.M.,
and seasonal soil water chemistry. Geochim. Cosmochim. Acta 369, 111–125. Lea, D.W., Lough, J.M., Shimmield, G.B., 2001. Variability in the El niño-southern
https://doi.org/10.1016/j.gca.2024.01.034. oscillation through a glacial-interglacial cycle. Science 291, 1511–1517. https://doi.
Raia, P., Barbera, C., Conte, M., 2003. The fast life of a dwarfed giant. Evol. Ecol. 17, org/10.1126/science.1057969.
293–312. https://doi.org/10.1023/A:1025577414005. Turney, C.S.M., Flannery, T.F., Roberts, R.G., Reid, C., Fifield, L.K., Higham, T.F.G.,
Reimer, P.J., Austin, W.E.N., Bard, E., Bayliss, A., Blackwell, P.G., Bronk Ramsey, C., Jacobs, Z., Kemp, N., Colhoun, E.A., Kalin, R.M., Ogle, N., 2008. Late-surviving
Butzin, M., Cheng, H., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P., megafauna in Tasmania, Australia, implicate human involvement in their extinction.
Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.A., Kromer, B., Manning, S.W., Proc. Natl. Acad. Sci. 105, 12150–12153. https://doi.org/10.1073/
Muscheler, R., Palmer, J.G., Pearson, C., van der Plicht, J., Reimer, R.W., pnas.0801360105.
Richards, D.A., Scott, E.M., Southon, J.R., Turney, C.S.M., Wacker, L., Adolphi, F., Turvey, S.T., Tong, H., Stuart, A.J., Lister, A.M., 2013. Holocene survival of Late
Büntgen, U., Capano, M., Fahrni, S.M., Fogtmann-Schulz, A., Friedrich, R., Pleistocene megafauna in China: a critical review of the evidence. Quat. Sci. Rev. 76,
Köhler, P., Kudsk, S., Miyake, F., Olsen, J., Reinig, F., Sakamoto, M., Sookdeo, A., 156–166. https://doi.org/10.1016/j.quascirev.2013.06.030.
Talamo, S., 2020. The IntCal20 northern hemisphere radiocarbon age calibration Van Valkenburgh, B., Hayward, M.W., Ripple, W.J., Meloro, C., Roth, V.L., 2016. The
curve (0–55 cal kBP). Radiocarbon 62, 725–757. https://doi.org/10.1017/ impact of large terrestrial carnivores on Pleistocene ecosystems. Proc. Natl. Acad.
RDC.2020.41. Sci. 113, 862–867. https://doi.org/10.1073/pnas.1502554112.
Řičánková, V.P., Robovský, J., Riegert, J., Zrzavý, J., 2015. Regional patterns of Walker, M., Johnsen, S., Rasmussen, S.O., Popp, T., Steffensen, J.-P., Gibbard, P.,
postglacial changes in the Palearctic mammalian diversity indicate retreat to Hoek, W., Lowe, J., Andrews, J., Björck, S., Cwynar, L.C., Hughen, K., Kershaw, P.,
Siberian steppes rather than extinction. Sci. Rep. 5, 12682. https://doi.org/10.1038/ Kromer, B., Litt, T., Lowe, D.J., Nakagawa, T., Newnham, R., Schwander, J., 2009.
srep12682. Formal definition and dating of the GSSP (Global Stratotype Section and Point) for
Rito, T., Vieira, D., Silva, M., Conde-Sousa, E., Pereira, L., Mellars, P., Richards, M.B., the base of the Holocene using the Greenland NGRIP ice core, and selected auxiliary
Soares, P., 2019. A dispersal of Homo sapiens from southern to eastern Africa records. J. Quat. Sci. 24, 3–17. https://doi.org/10.1002/jqs.1227.
immediately preceded the out-of-Africa migration. Sci. Rep. 9, 4728. https://doi. Webster, P.J., 1987. The variable and interactive monsoon. In: Fein, J.S., Stephens, P.L.
org/10.1038/s41598-019-41176-3. (Eds.), Monsoons. Wiley, New York, pp. 269–330.
Román-Palacios, C., Wiens, J.J., 2020. Recent responses to climate change reveal the White, L.C., Mitchell, K.J., Austin, J.J., 2018. Ancient mitochondrial genomes reveal the
drivers of species extinction and survival. Proc. Natl. Acad. Sci. 117, 4211–4217. demographic history and phylogeography of the extinct, enigmatic thylacine
https://doi.org/10.1073/pnas.1913007117. (Thylacinus cynocephalus). J. Biogeogr. 45, 1–13. https://doi.org/10.1111/jbi.13101.
Saltré, F., Chadoeuf, J., Peters, K.J., McDowell, M.C., Friedrich, T., Timmermann, A., Whittaker, R.H., 1970. Communities and Ecosystems. Macmillan Company, London.
Ulm, S., Bradshaw, C.J.A., 2019. Climate-human interaction associated with Wolverton, S., Lyman, R.L., Kennedy, J.H., La Point, T.W., 2009. The terminal
southeast Australian megafauna extinction patterns. Nat. Commun. 10, 5311. Pleistocene extinctions in North America, hypermorphic evolution, and the dynamic
https://doi.org/10.1038/s41467-019-13277-0. equilibrium model. J. Ethnobiol. 29, 28–63. https://doi.org/10.2993/0278-0771-
Schneider, T., Bischoff, T., Haug, G.H., 2014. Migrations and dynamics of the 29.1.28.
intertropical convergence zone. Nature 513, 45–53. https://doi.org/10.1038/ Wroe, S., Field, J.H., Archer, M., Grayson, D.K., Price, G.J., Louys, J., Faith, J.T.,
nature13636. Webb, G.E., Davidson, I., Mooney, S.D., 2013. Climate change frames debate over
Seersholm, F.V., Werndly, D.J., Grealy, A., Johnson, T., Keenan Early, E.M., Lundelius, E. the extinction of megafauna in Sahul (Pleistocene Australia-New Guinea). Proc. Natl.
L., Winsborough, B., Farr, G.E., Toomey, R., Hansen, A.J., Shapiro, B., Waters, M.R., Acad. Sci. 110, 8777–8781. https://doi.org/10.1073/pnas.1302698110.
McDonald, G., Linderholm, A., Stafford, T.W., Bunce, M., 2020. Rapid range shifts Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and
and megafaunal extinctions associated with late Pleistocene climate change. Nat. aberrations in global climate 65 Ma to present. Science 292, 686–693. https://doi.
Commun. 11, 2770. https://doi.org/10.1038/s41467-020-16502-3. org/10.1126/science.1059412.
Smith, F.A., Lyons, S.K., Ernest, S.K.M., Jones, K.E., Kaufman, D.M., Dayan, T., Zhang, X., Evans, J.P., Burrell, A.L., 2024. Less than 4% of dryland areas are projected to
Marquet, P.A., Brown, J.H., Haskell, J.P., 2003. Body mass of late quaternary desertify despite increased aridity under climate change. Commun. Earth Environ. 5,
mammals. Ecology 84, 3403. https://doi.org/10.1890/02-9003. 300. https://doi.org/10.1038/s43247-024-01463-y.
Smith, T.M., Tafforeau, P., Reid, D.J., Grün, R., Eggins, S., Boutakiout, M., Hublin, J.-J.,
2007. Earliest evidence of modern human life history in North African early Homo

11