Surviving Savanna, Missing Megafauna: Extinction In The Tropical Plains of South America

Written By Tristan Rapp

This is a guest article, kindly contributed by Tony (writing under a pseudonym). He is a lifelong nature enthusiast with a Bachelor’s degree in Biology. He writes about paleoecology, paleontology, and paleoanthropology at his blog http://prehistoricpassage.com

Introduction

The causes of the Late Quaternary megafaunal extinctions in the Americas remain a subject of intense debate. Most attention has focused on North America, where scientists blame either humans, climate change, or a mix of both(1)(2)(3). As North America undoubtedly experienced profound environmental change concurrent with the extinctions, it is not surprising that many promote full or partial climatic explanations in its case. However, right next door lies a continent that experienced even more severe extinctions at about the same time: South America (4)(5). This is in spite of very different climatic pressures. Detailed examinations of extinction dynamics on this continent will therefore provide critical insights into extinction events elsewhere.

South America contains a great variety of ecosystems and plant and animal life. In fact, it is the most biodiverse continent (6)(7). Yet, something is missing in this apparent Eden. While there is an unparalleled number of total plant and animal species on this landmass, there is a striking lack of megafauna- animals larger than 45 kg. However, it was not always like this. A great number of megafaunal taxa once called South America home, including ones that would seem completely out of place there today. In fact, by the Late Pleistocene, the average body mass of mammals in South America was greater than on any other continent (8).

Extinctions in the Pampas, Patagonia, and Andes have received heavy focus from researchers (9)(10)(11), since these areas (along with the Pacific coast) have decent to great conditions for fossil and artifact preservation. These regions would make a great topic for a possible future piece, but today’s focus will be on the extinctions in the rich but often overlooked savannas and grasslands of the South American tropics.

Join us as we explore and evaluate the climatic and human explanations for the loss of the incredible and eccentric beasts in this part of the world.

Neotropical Paradise

When it comes to biodiversity, South America outshines all other continents. The massive Amazon rainforest formed during the Eocene tens of millions of years ago and is famous for the sheer number of plant and animal species it contains. However, the less well-known tropical savanna and grassland regions to the south, east, and north of the Amazon contain exceptional biodiversity as well.

Today, the Cerrado of Brazil is considered the richest savanna on the planet with an exceptionally high number of plant and animal species(12). The biodiversity is a function of its central location between rainforest, wetland, and dry forest biomes. A section of the Brazilian highlands runs through this region, creating striking landscapes. Habitats in the Cerrado range from typical grassy savannas to marshes and gallery forests. There is a high rate of endemism here (species found nowhere else).

Figure 1. Cerrado landscape in the Chapada dos Veadeiros National Park, Brazil.

Terms of use: This image is licensed under an Attribution-NonCommercial-NoDerivs 2.0 Generic license. It is attributed to Luís Felipe Figueiredo. The image is unedited and can be found here.

South of the Cerrado lies the Humid Chaco or Chaco Oriental, located largely in Paraguay and northeastern Argentina. This region spans both tropical and subtropical latitudes and climates. It contains a mosaic of woodlands and savannas (13). The endangered marsh deer can be found here, along with typical South American fauna such as the black howler monkey (Alouatta caraya), giant anteater (Myrmecophaga tridactyla), jaguar (Panthera onca), cougar (Puma concolor), and maned wolf (Chrysocyon brachyurus).

The open plains north of the Amazon rainforest include the Llanos and the smaller Guianian savanna. The Llanos ecoregion contains a large area of nutrient poor soils on uplands while low-lying areas are seasonally flooded with much richer soils(13). It is home to the endangered Orinoco crocodile (Crocodylus intermedius). The Guianian savanna is located east of the Llanos and contains three distinct patches, the largest being known as the Gran sabana. The Beni savanna to the southwest of the Amazon is home to its own highly diverse fauna.

The Pleistocene ecosystems of many areas have frequently been likened to the plains of eastern and southern Africa today, so it may seem cliche to make that comparison once again. Yet of all the world, the grassy habitats of tropical South America would have had a particularly strong resemblance to Africa (perhaps only behind southern and southeastern Asia), which is surprising considering that the two landmasses are separated by a large ocean. Not only were the climate and landscape broadly similar, the still-extant monkeys and crocodilians would have shared their habitat with extinct proboscideans known as gomphotheres, horses, and toxodontids- a lineage unique to the New World but which resembled something in between rhinos and hippos.

However, South America had its own unique flavor too. It spent a large part of the Cenozoic as an island before becoming connected to North America with the formation of the Isthmus of Panama 3 million years ago(14). During that isolation, it developed unique lineages. The aforementioned toxodontids-represented by Mixotoxodon in the north and Toxodon in the south-would be among those, but there were also glyptodonts (giant relatives of the armadillo), ground sloths, and macraucheniids- large herbivores with a peculiar morphology that defies easy description. All but the macraucheniids eventually made it into North America.

Figure 2. A mixed forest-savanna landscape in Pleistocene Amazonia with extinct and extant animals. Extinct genera include Xenorhinotherium, Eremotherium, Paleolama, Notiomastodon, Toxodon, and Glyptodon.

Art with permission from Júlia d'Oliveira

South American llamas, horses, deer, felids, canids, tapirs, proboscideans, and tremarctine (short-faced) bears all arrived from North America (15)(16). Most famous among the felids of Pleistocene South America was the sabre-toothed Smilodon populator, perhaps the largest cat to ever live(17). Many of these taxa managed to survive into the present day, but even these are nowhere near as widespread or diverse as they were during the Pleistocene. In fact, aside from the giant anteater and capybaras, all current South American mammals larger than 45 kg are North American migrants from the Great American Biotic Interchange (GABI)(18).

So what happened to this Neotropical Serengeti? 83% of South America genera vanished in a geologic flash (4)(5). The Earth was leaving the ice age when extinctions began, so climate explanations are tempting. We will see if they withstand scrutiny.

The Climate Context

Like in most parts of the globe, the climate in tropical South America was cooler during the glacial periods than at present(19)(20). Many of the regions discussed in this article would have, at times, experienced climates that may not strictly classify as tropical by today’s standard but since these currently lie either fully or partially in tropical latitudes, they will be referred to here as such for the sake of simplicity. 

South America can be considered a tough nut to crack for those devising climatic explanations for megafaunal extinctions. Most of South America is located in the less climatically turbulent southern hemisphere, with the remainder being fully within the northern hemisphere tropics. This spared it from the extreme temperature fluctuations in sections of the northern hemisphere over the past few million years(21). Changes in moisture regimes would still have had important effects on plants and animals, but the temperature changes were relatively modest. The exceptional biodiversity still present in South America and the maintenance of large body sizes by the extinct megafauna (6)(8) strongly hint against intense environmental stress during the Pleistocene.

All of this makes it difficult to propose a climatic mechanism that could be solely or mainly responsible for extinctions in this continent; like elsewhere, the last glacial-interglacial transition was certainly a time of change but South American megafauna survived numerous such transitions. As a result, an interesting explanation has been proffered which combines both human and climate factors. This hypothesis is known as the “broken zig-zag” and it was proposed by paleontologist Alberto Cione(18)(22). It posits that humans were central to South American extinctions through overhunting but that this occurred in an ecologically sensitive period.

According to this concept, the abundance of megafauna is naturally much higher during glacials- which are dry with widespread grassy habitats-than in interglacials, which are wet and more heavily forested. This is because most of the extinct fauna are said to have been adapted to open habitats. Rainforests and other closed forests would have covered a small area during glacial periods but then greatly expanded at the expense of savanna and grassland by the start of interglacial periods due to warmer and wetter conditions. 

As a result, these open-habitat megafauna would have been under ecological stress and some even at minimal viable numbers until the next glacial period commenced and relieved pressure. The population of megafauna during glacials and interglacials is therefore represented by the upward and downward lines in a zig-zag, respectively. The key difference between the start of the current interglacial and previous ones is that humans were present. The “breaking” of the zig-zag occurred when humans decapitated South American megafauna when they were in a highly depleted and vulnerable state, triggering an extreme extinction event. This is what gives the hypothesis its name.

The implication is that without the reduction of grasslands, extinctions would either not have occurred or would have been far more mild. Essentially, climate was not the sole cause of the extinctions, but it was a necessary factor along with humans. This appears to be a fairly sound theory based on the mechanism, especially in lower latitudes of South America which have experienced forest expansion since the Last Glacial Maximum (23)(24). Further, the model’s reliance on a natural, typical environmental shift from glacial to interglacial conditions as opposed to sudden climatic shocks explains how South American non-megafaunal biodiversity was retained. Unsurprisingly, the broken zig-zag hypothesis has strongly influenced South American research on the extinctions (25)(26)(27)(28). 

It is certainly a commendable attempt to explain what happened in the continent, but is it really correct? Proper scrutiny has unfortunately not been applied to this model despite a slew of climatological and paleoenvironmental evidence that would call it into question. This piece will seek to address its flaws. 

The Restrained Rainforest

Perhaps the biggest issue with the broken zig-zag hypothesis is that it does not appreciate the nuances of climate and vegetation dynamics during the Quaternary in South America, which leads it to overestimate the extent to which rainforest and other closed-canopy forests actually expanded by the time the megafauna went extinct. It is tempting to assume that the hypothesis has some validity if one assumes that the great size of the Amazon and Atlantic rainforests in recent times (pre-deforestation) represents the interglacial norm, and accepts traditional notions of widespread glacial aridity. However, both of these assumptions are flawed. Starting with the first, we must recognize that climatic and environmental conditions during the extinction window were not the same as those in the late Holocene.

Figure 3. Aerial photo of the Peruvian Amazon.

Terms of use: This image is licensed under an Attribution-ShareAlike 3.0 Unported license. It is attributed to Martin St-Amant. The image is unedited and the original can be found here.

Extinctions in South America started slightly later than those in North America (29). They occured between 12.9 and 10.9 thousand years ago, spanning the terminal Pleistocene (specifically Younger Dryas) and early Holocene, with most of the decline concentrated in the first half of this period (30). Late survival dates for South American megafauna in the Pampas have strongly been questioned after improved dating showed alleged Holocene fossils to be from the end-Pleistocene (31). A study published earlier this year claims megafauna in parts of Brazil survived as late as 3500 BP (27), but these extremely anomalous dates have not been corroborated and probably also suffer from dating issues (G.G. Politis, personal communication, April 28, 2025). 

In any case, it seems rather unlikely that megafauna in non-remote parts of South America would survive so much longer than their counterparts in the rest of the mainland Americas, so we can say with high certainty that all or nearly all extinct South American megafauna did not make it past the early Holocene. This means that any conversation about the climatic and environmental changes that potentially led to the demise of these animals must be centered specifically on the last glacial-interglacial transition. So what do we know about that timeframe?

The terminal Pleistocene, or late glacial, was a turbulent era with the Younger Dryas being its last and most famous act. The Younger Dryas featured a weakening of the Atlantic Meridional Overturning Current (AMOC) resulting in a southwards migration of the Intertropical Convergence Zone (ITCZ), the latter being responsible for bringing rain to equatorial regions (32)(33). While not affecting all of South America, this caused much wetter conditions in northeast Brazil (now dominated by arid vegetation) and drier conditions in northwestern South America such as Colombia and Venezuela. There was a temporary surge in lowland and montane rainforest plants in northeast Brazil (33)(34). Meanwhile, the Cerrado exhibited a heterogeneous response with regard to hydrology and vegetation at this time(35)(36).

The early Holocene, by contrast, appears to have been a largely dry period; the South American monsoon was actually fairly weak for much of the current interglacial (37)(38). Pollen analysis from various sites indicates that many parts of tropical South America only experienced savanna to forest transitions around the late Holocene as opposed to the early Holocene (23)(24). A paleovegetation study by Maksic et al. (2018) corroborated this through climate simulations by showing that most of the Holocene was less forested than the last few thousand years (39). In truth, the Amazon and Atlantic rainforests apparently only reached their maximum interglacial extent fairly recently.

The Pleistocene-Holocene transition was not characterized by massively reduced open habitat, but how does it stack up to the height of the ice age- the Last Glacial Maximum (LGM)- around 20 thousand years ago? The overall extent of the Amazon rainforest during the LGM has long been a subject of intense debate, made more complicated by the relatively sparse paleoenvironmental sites in the Amazon basin. Some argue that the Amazon rainforest had contracted greatly during this period on the basis of (alleged) widespread LGM aridity or low atmospheric CO₂ (40)(41), while others have contested this (24)(42). However, marine core data from near the mouth of the Amazon river records basin-wide trends and shows substantial forest cover in the basin during the LGM, with greater-than-present savanna encroachment occurring only at the edges (24)(43). There is, however, a potential caveat in that it is unclear how much of this forest cover was dry forest vs. rainforest. Meanwhile, the Atlantic rainforest appears to have been robust at this time based on the demographic history of its small rainforest mammals (44). 

Regardless, the LGM was only a few thousand years in length and should not be used as the sole benchmark for the last glacial period as a whole, which lasted from 115 to 11.7 thousand years ago and contained considerable climatic and environmental variability. Studies on different Brazilian sites in the northeast, Cerrado, and Atlantic rainforest indicate that climate and vegetative shifts during the Late Quaternary were incredibly diverse across geography and time. Periods of rainforest expansion in northeast Brazil occurred multiple times over several tens of thousands of years and were often linked to cold Heinrich stadials(34); the Younger Dryas was not exceptional or severe in this regard. Other regions of Brazil experienced localized changes through history- periods of high moisture and tree cover were frequently out of sync between different sites (35)(36)(45)(46). In each case, episodes of strong woody growth were in no way exclusively or strongly correlated with the onset of the current interglacial; often, the opposite seems to be true.

All in all, the data on vegetative changes over time is far more complex than implied by the broken zig-zag model. The early Holocene was drier and less forested than the late Holocene. The Younger Dryas may ironically have brought wet conditions to some regions but its overall effects were neither homogeneous nor unusual in the context of the last glacial period. The Last Glacial Maximum-stereotyped as very open- was a period when widespread rainforest still flourished. Contrary to the simplistic dichotomy of “dry and open” glacials vs. “wet and heavily forested” interglacials, there is a considerable degree of variability in moisture and vegetation within both glacials and interglacials in tropical South America. Consequently, it would be wrong to frame the last glacial-interglacial transition as an exceptionally difficult time for the open-habitat animals living there as they already would have been well-accustomed to such changes.

Problems With Bioclimatic Modeling

Some studies have tried to model the ranges of extinct megafauna over time in what is known as bioclimatic modeling. These studies look at Late Pleistocene fossil distributions to determine habitat suitability based on temperature and precipitation parameters (often using the LGM as the reference point), and then infer changes in suitability for various species over time. This is in contrast to reconstructing regional paleoenvironments directly through proxies like pollen data. They consistently show that animals would have experienced moderate to severe reductions in suitable habitat going into and during the Holocene (26)(28)(47)(48), which would seem to provide support for the broken zig-zag model.

However, there are major problems with this method; one potential issue is that the ranges of the extinct species may not be properly represented by the fossil record. The distribution of fossils is influenced by taphonomic factors, and most of tropical South America is quite bad for preservation as a result of acidic soils. The maps below show that soil pH correlates rather well with megafaunal fossil distribution; for example, the vast but acidic Cerrado in central Brazil is suspiciously fossil-poor and the arid but pH-neutral west of Peru is fairly rich. If fossil deposition is heavily influenced by soil acidity (perhaps alongside other factors like climate), then true past abundance by region would be difficult to infer.

With tropical regions suffering from a dearth of fossils for reasons beyond megafaunal abundance and the researchers’ questionable assumption that the cold LGM represents optimal conditions for megafauna, it is quite possible that the warmth-tolerance of extinct South American animals is not being fully captured by the models. Thus, interglacial climates may appear less suitable than they really are (especially in certain regions). Further, animals can and do shift the climatic space that they occupy over time. For example, a study analyzing the distributions of European ungulates over the Late Quaternary shows that animals started to occupy areas of greater warmth and lower rainfall by the Holocene (49). Also, temperatures in equatorial Africa were 3-5 ℃ cooler during the LGM compared to the present (50) but animals there clearly adjusted well to interglacial warmth and continued to thrive at all latitudes.

To illustrate how these problems can produce biased results in bioclimatic modeling, we can look at the suggested mid-Holocene distribution for the giant ground sloth Eremotherium in one of these papers(47). It shows that suitable habitat for this animal would have been massively reduced compared to the LGM, with the best habitat restricted to small areas on the fringes of its former range. Those familiar with the high ecological plasticity and broad historical range of Eremotherium (51) would recognize that this alleged mid-Holocene distribution is highly implausible. This suggests that something is seriously flawed with this type of approach.

Another potential issue with modeling suitable conditions for animals using peak-glacial distributions is that density may change over time. Cione suggests that animal density in open habitats may increase during interglacials due to higher vegetative productivity (and hence food) resulting from greater warmth and humidity, but overall abundance would decrease as the losses in habitat would greatly override productivity increase (18). Yet, we have already shown that losses in open areas could not have been that large when extinctions took place, so it is tempting to wonder if increased productivity actually did make up for this. 

Compared to the LGM, the extinction period may not have been wetter but it was certainly warmer and had higher CO₂ levels (between 55-85 ppm higher) (52). CO₂ is a major factor in plant productivity; studies have shown that going from glacial to pre-industrial levels of CO₂ can greatly increase plant biomass but this applies primarily to C3 plants as opposed to C4 plants (53)(54), the latter of which comprise most grasses in tropical South America. However, C4 plants still benefit from higher temperatures (54), so research on how primary productivity might have changed in grassy biomes since the LGM would be welcome.

Regardless, it is not logical that South American open-habitat megafauna would have experienced such huge contractions in range or population as a result of climate change. In Africa today, savanna animals flourish wherever suitable habitat exists across a wide swathe of climates. Why would South American savanna animals- who clearly still had plenty of open habitat available to them- not have been able to do the same?

Figure 4. African megafauna thrive in the interglacial world, in many types of climates.

Terms of use: This image is licensed under an Attribution-NonCommercial-ShareAlike 2.0 Generic license. It is attributed to Daniel Giménez. The image is unedited and the original can be found here.

Ecosystem Engineers

The Gran Chaco, also known as the Dry Chaco, is a vast region east of the Andes and west of the Humid Chaco spanning tropical and subtropical zones. The Caatinga is a highly biodiverse arid region situated in the northeast of Brazil. Both are dry forests and not savanna or grassland environments per se, but they are worth discussing as there is a potential that these regions might have been different under the right circumstances. It is possible that they might have been more savanna-like if the resident megafauna had not gone extinct, which would suggest that there is even more space available for open-habitat megafauna than meets the eye.

A study by Doughty et al. (2015) speculates about how extinct megafauna may have reduced woody growth in savanna and dry forest regions of South America (55). We are becoming increasingly aware of how large animals shaped and continue to shape their environments in various past and present environments around the world, and one way is by preventing an excess of woody growth (56)(57). It is therefore logical to expect that the extinct South American megafauna would have exerted strong top down controls on woody vegetation in dry forests where trees do not grow as tall, thick, or fast as in rainforests.

Dantas and Pausas (2022) show that plants across much of South America have various defenses to herbivory which suggest that they were heavily consumed by megafauna in the past (26). The study delineates arid, nutrient-rich savannas where trees have physical defenses from moist, nutrient-poor savannas where trees rely more on chemical ones. Doughty et al. (2015) point out that savannas in South America are much woodier than African ones at any given soil moisture level, suggesting this is due to megafaunal presence in the latter but not the former (55). 

Both studies postulate that the dry forests like the Chiquitano, Gran Chaco, and Caatinga would be more savanna-like today if the megafauna were still present, and that present savannas like the Cerrado would be less woody as well. One can imagine gomphotheres, toxodonts, glyptodonts, and ground sloths in the semi-arid Gran Chaco creating a mosaic of woodland and savanna akin to the southern African bushveld by trampling and destroying saplings and trees. It is also intriguing to speculate on whether now-disjointed savanna ecoregions might even be connected through corridors between forests if these animals were still present.

Figure 5. Arid Caatinga vegetation in northeastern Brazil.

Terms of use: This image is licensed under an Attribution-NonCommercial-ShareAlike 2.0 Generic license. It is attributed to Otávio Nogueira. The image is unedited and the original can be found here.

Further, it is quite possible that increases in woody cover in some regions between 13,000 and 10,000 years ago (38) could have been influenced partly by their extinction too. Disentangling this from the effects of shifting precipitation and CO₂ levels is difficult since the combined influence of abiotic factors on vegetative composition overrides that of herbivores, but research suggests that megafaunal extinction may have contributed partially in other parts of the world (56)(57). Megafauna were especially large in South America; tropical parts of the continent are currently underexplored with regard to herbivory-vegetation dynamics but would be ripe for future research.

In sum, when considering the role of large animals in shaping landscapes, we should recognize that even in recent times, there is potentially more suitable habitat for savanna-dwelling animals than appears at first glance, and there is a chance that the degree of woodiness in certain parts of South America shortly following extinctions may not have been fully natural either.

Relevance of Regional Dynamics

Considering environmental change during the extinction period being relatively modest, the climatic flexibility of megafauna, and the ability of megafauna to shape their environments to be more favorable to them, there does not seem to be much room for climatic arguments. However, there are still those who favor the importance of regional dynamics in explaining the extinctions for parts of South America (9)(11)(28). It may seem intuitive to believe that the relative importance of factors contributing to extinction varied across geography. However, we must keep in mind that for any cause to truly be meaningful, it should be impactful enough that it is possible to envision a different outcome in its absence.

Let us now shift to the human element. Prates and Perez (2021) reveal the striking correlation between the rise of Fishtail Projectile Points- a kind of fluted point similar to those used by the Clovis in North America- and the extinction of megafauna (30). As humans, known in archeological contexts as “Paleo-Indians”, started to use these projectiles to hunt megafauna, their populations grew rapidly while megafaunal populations crashed. The study focuses on the three regions that have the greatest evidence of human-megafauna interaction (the Andes, Patagonia, and Pampas) but Fishtail Points have also been found in tropical regions such as southern Brazil, Venezuela, and Guyana (59). 

The paucity of Fishtail tips outside the southern cone and Andes should not be assumed to imply a lack of use by Paleo-Indians in other parts of South America and may instead reflect the poor preservation conditions mentioned earlier, along with perhaps less archeological research in certain areas (60). It can therefore be assumed that similar patterns applied to tropical sections of the continent. However, further exploration of human-megafauna interactions in less-studied regions would certainly be welcome.

Figure 6. Fishtail projectile points from Brazil, redrawn from their original illustrations.

Terms of use: This image is licensed under an Creative Commons Attribution 4.0 International license. It is attributed to Daniel Loponte, Mirian Carbonera & Romina Silvestre. The image is unedited and the original can be found here.

Now recall how Prates and Perez (2021) indicate extinctions took place between 12.9 and 10.9 thousand years ago with the great majority of the decline concentrated within the first half of this period (30). The abrupt onset, short duration, wide geographical scope, and high magnitude of the extinctions combined with their close association with new hunting technology makes it challenging to argue for the importance of regional environmental dynamics. It is for this reason that those who espouse theories involving a climatic contribution such as the broken zig-zag favor a protracted chronology of extinction wherein megafauna survived into the tail end of the early Holocene or even later (4)(18)(25)(27).

To better illustrate the insignificance of regional climate dynamics, we can look at changes in northeast Brazil during the terminal Pleistocene. As mentioned earlier, there was a temporary increase in precipitation in northeast Brazil during the Younger Dryas that coincided with an increased abundance of rainforest taxa (33). In one study, researchers use this fact to argue that climate may have been decisive in extinctions in this part of South America by reducing available habitat for megafauna (28). Looking at the broader geographic context makes it difficult to support this conclusion.

There is an antiphased pattern of moisture in northern South America during weak AMOC periods- as precipitation increases in northeast Brazil, it decreases in northwestern South America (Venezuela, Colombia) (43). Indeed, this is what paleoclimate data indicates happened during the Younger Dryas (32)(33)(58). This decrease in precipitation in the northwest should have caused a decrease in rainforest cover there or at least prevented its expansion (43), theoretically benefitting local grassland animals. Yet, the animals living in that area did not fare any better than those living in northeastern Brazil.

If environmentally stable regions failed to preserve animal communities any better than environmentally unstable ones, then localized climate dynamics cannot be considered “decisive” in any meaningful sense. At most, climate change causing forest expansion in certain areas may have only hastened an inevitable demise at the hands of widely dispersed and innovative humans. Cione, who invented the broken zig-zag theory, even points out that almost all surviving South American megafaunal mammals are Holarctic migrants from the Great American Biotic Interchange, are often clustered in remote environments, and none are particularly large(18). What would have stopped Paleo-Indians from eliminating hulking, slow-moving prey such as glyptodonts, toxodonts, or ground sloths even if those taxa had larger areas to roam?

The Last Glacial Maximum is frequently assumed to be some sort of ideal period for megafauna by South American researchers (26)(28)(47)(48). But we can try to imagine what would happen in a world where LGM-like conditions persisted for several thousand more years, and where there were indeed more widespread plains supporting a larger animal biomass despite lower productivity. In this hypothetical scenario, could it be that this larger prey base would have only further augmented human numbers following the advent of Fishtail technology? If so, do we have reason to think there would have been an appreciable difference in extinction outcomes aside from perhaps a slight delay? This is of course speculation- we do not know exactly what would have transpired in that scenario. Still, looking at it from this angle shows that we must exercise caution and not lavishly assume that losses would have been far less severe under supposedly ideal conditions.

Conclusion

Given the difficulty of proposing wholly or primarily climatic explanations for extinctions in South America, it is not surprising that some have chosen a middle ground hypothesis to bridge climate and human factors. The broken zig-zag model, which involves humans eliminating open-habitat adapted megafauna when they were at their greatest extinction risk, is creative yet flawed. We would expect to see the best evidence for this hypothesis in tropical parts of South America where rainforest is today quite widespread, but actual paleoecological support for it is weak for a number of reasons.

First, climatic and paleoenvironmental research casts serious doubt on the idea that tropical South America was significantly more forested during the extinction window than during the glacial period as a whole. Second, the results from bioclimatic modeling showing substantial declines in suitable habitat are not reliable as they fail to account for a variety of taphonomic and ecological factors. Third, we cannot assume that the balance of forest to savanna in South America now, as well as during and immediately after the extinctions, is or was natural considering that the loss of keystone megafauna may itself have accentuated woody growth. Lastly, looking at the speed, ubiquity, and intensity of the extinctions, we must reconsider whether or not regional environmental dynamics were truly relevant to overall outcomes.

Ultimately, the points presented in this piece serve as a cautionary tale about how even popular and seemingly compelling hypotheses like the broken zig-zag may end up being problematic upon further inspection.

References

  1. Martin, P. S. (1973). The Discovery of America: The first Americans may have swept the Western Hemisphere and decimated its fauna within 1000 years. Science, 179(4077), 969–974. https://doi.org/10.1126/science.179.4077.969

  2. Grayson, D. K., & Meltzer, D. J. (2003). A requiem for North American overkill. Journal of Archaeological Science, 30(5), 585–593. https://doi.org/10.1016/s0305-4403(02)00205-4

  3. Broughton, J. M., & Weitzel, E. M. (2018). Population reconstructions for humans and megafauna suggest mixed causes for North American Pleistocene extinctions. Nature Communications, 9(1). https://doi.org/10.1038/s41467-018-07897-1

  4. Barnosky, A. D., & Lindsey, E. L. (2010). Timing of Quaternary megafaunal extinction in South America in relation to human arrival and climate change. Quaternary International, 217(1-2), 10–29. https://doi.org/10.1016/j.quaint.2009.11.017

  5. Koch, P. L., & Barnosky, A. D. (2006). Late Quaternary Extinctions: State of the Debate. Annual Review of Ecology, Evolution, and Systematics, 37(1), 215–250. https://doi.org/10.1146/annurev.ecolsys.34.011802.132415

  6. IUCN. (n.d.). The IUCN Red List of Threatened Species. IUCN Red List; International Union for Conservation of Nature and Natural Resources. Retrieved May 11, 2025, from https://www.iucnredlist.org/

  7. Raven, P. H., Gereau, R. E., Phillipson, P. B., Chatelain, C., Jenkins, C. N., & Ulloa, C. U. (2020). The distribution of biodiversity richness in the tropics. Science Advances, 6(37), eabc6228. https://doi.org/10.1126/sciadv.abc6228

  8. Lyons, S. K., Smith, F. A., & Brown, J. H. (2004). Of mice, mastodons and men: human-mediated extinctions on four continents. Evolutionary Ecology Research, 6(3), 339-358. 

  9. Prado, J. L., Martinez-Maza, C., & Alberdi, M. T. (2015). Megafauna extinction in South America: A new chronology for the Argentine Pampas. Palaeogeography, Palaeoclimatology, Palaeoecology, 425, 41–49. https://doi.org/10.1016/j.palaeo.2015.02.026

  10. Raczka, M. F., Mosblech, N. A., Giosan, L., Valencia, B. G., Folcik, A. M., Kingston, M., Baskin, S., & Bush, M. B. (2019). A human role in Andean megafaunal extinction? Quaternary Science Reviews, 205, 154–165. https://doi.org/10.1016/j.quascirev.2018.12.005

  11. Metcalf, J. L., Turney, C., Barnett, R., Martin, F., Bray, S. C., Vilstrup, J. T., Orlando, L., Salas-Gismondi, R., Loponte, D., Medina, M., De Nigris, M., Civalero, T., Fernández, P. M., Gasco, A., Duran, V., Seymour, K. L., Otaola, C., Gil, A., Paunero, R., & Prevosti, F. J. (2016). Synergistic roles of climate warming and human occupation in Patagonian megafaunal extinctions during the Last Deglaciation. Science Advances, 2(6), e1501682. https://doi.org/10.1126/sciadv.1501682

  12. Simon, M. F., Grether, R., de Queiroz, L. P., Skema, C., Pennington, R. T., & Hughes, C. E. (2009). Recent assembly of the Cerrado, a neotropical plant diversity hotspot, by in situ evolution of adaptations to fire. Proceedings of the National Academy of Sciences, 106(48), 20359–20364. https://doi.org/10.1073/pnas.0903410106

  13. Schipper, J. (n.d.). One Earth. Retrieved May 11, 2025, from https://www.oneearth.org/ecoregions/navigator

  14. O’Dea, A., Lessios, H. A., Coates, A. G., Eytan, R. I., Restrepo-Moreno, S. A., Cione, A. L., Collins, L. S., de Queiroz, A., Farris, D. W., Norris, R. D., Stallard, R. F., Woodburne, M. O., Aguilera, O., Aubry, M.-P., Berggren, W. A., Budd, A. F., Cozzuol, M. A., Coppard, S. E., Duque-Caro, H., & Finnegan, S. (2016). Formation of the Isthmus of Panama. Science Advances, 2(8), e1600883. https://doi.org/10.1126/sciadv.1600883

  15. Stehli, F. G. (1985). The Great American biotic interchange. Plenum Pr.

  16. Webb, S. D. (1991). Ecogeography and the Great American Interchange. Paleobiology, 17(3), 266–280. https://doi.org/10.1017/s0094837300010605

  17. Christiansen, P., & Harris, J. M. (2005). Body size ofSmilodon (Mammalia: Felidae). Journal of Morphology, 266(3), 369–384. https://doi.org/10.1002/jmor.10384

  18. Cione, A. L., Tonni, E. P., & Soibelzon, L. (2009). Did humans cause the Late Pleistocene-Early Holocene mammalian extinctions in South America in a context of shrinking open areas?. American megafaunal extinctions at the end of the Pleistocene, 125-144. 

  19. Schneider von Deimling, T., Ganopolski, A., Held, H., & Rahmstorf, S. (2006). How cold was the Last Glacial Maximum? Geophysical Research Letters, 33(14). https://doi.org/10.1029/2006gl026484

  20. Annan, J. D., Hargreaves, J. C., & Mauritsen, T. (2022). A new global surface temperature reconstruction for the Last Glacial Maximum. Climate of the Past, 18(8), 1883–1896. https://doi.org/10.5194/cp-18-1883-2022

  21. Herrando-Moraira, S., Nualart, N., Galbany-Casals, M., Garcia-Jacas, N., Ohashi, H., Matsui, T., Susanna, A., Tang, C. Q., & López-Pujol, J. (2022). Climate Stability Index maps, a global high resolution cartography of climate stability from Pliocene to 2100. Scientific Data, 9(1), 48. https://doi.org/10.1038/s41597-022-01144-5

  22. Tonni, E. P., Cione, A. L., & Soibelzon, L. H. (2003). The Broken Zig-Zag: Late Cenozoic large mammal and tortoise extinction in South America. Revista Del Museo Argentino de Ciencias Naturales, vol. 5, no. 1. http://sedici.unlp.edu.ar/handle/10915/5371

  23. Dantas, V. L., & Pausas, J. G. (2022). The legacy of the extinct Neotropical megafauna on plants and biomes. Nature Communications, 13(1), 129. https://doi.org/10.1038/s41467-021-27749-9

  24. Mayle, F. E., Beerling, D. J., Gosling, W. D., & Bush, M. B. (2004). Responses of Amazonian ecosystems to climatic and atmospheric carbon dioxide changes since the last glacial maximum. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 359(1443), 499–514. https://doi.org/10.1098/rstb.2003.1434

  25. Miotti, L., Tonni, E., & Marchionni, L. (2018). What happened when the Pleistocene megafauna became extinct? Quaternary International, 473, 173–189. https://doi.org/10.1016/j.quaint.2018.01.004

  26. Lima-Ribeiro, M. S., Nogués-Bravo, D., Terribile, L. C., Batra, P., & Diniz-Filho, J. A. F. (2013). Climate and humans set the place and time of Proboscidean extinction in late Quaternary of South America. Palaeogeography, Palaeoclimatology, Palaeoecology, 392, 546–556. https://doi.org/10.1016/j.palaeo.2013.10.008

  27. Faria, F. H. C., Carvalho, I. de S., Araújo-Júnior, H. I. de, Ximenes, C. L., & Facincani, E. M. (2025). 3,500 years BP: The last survival of the mammal megafauna in the Americas. Journal of South American Earth Sciences, 153, 105367. https://doi.org/10.1016/j.jsames.2025.105367

  28. Machado CCC, de Lucena VA, Canassa NF, de Oliveira DH, de Araujo HFP (2024). Effects of paleoclimatic variables on suitable open habitats for Pleistocene–Holocene megafauna in South America. Quaternary Research 1–11. https:// doi.org/10.1017/qua.2024.33 

  29. Surovell, T. A., Pelton, S. R., Anderson-Sprecher, R., & Myers, A. D. (2015). Test of Martin’s overkill hypothesis using radiocarbon dates on extinct megafauna. Proceedings of the National Academy of Sciences, 113(4), 886–891. https://doi.org/10.1073/pnas.1504020112

  30. Prates, L., & Perez, S. I. (2021). Late Pleistocene South American megafaunal extinctions associated with rise of Fishtail points and human population. Nature Communications, 12(1). https://doi.org/10.1038/s41467-021-22506-4

  31. Politis, G. G., Messineo, P. G., Stafford, T. W., & Lindsey, E. L. (2019). Campo Laborde: A Late Pleistocene giant ground sloth kill and butchering site in the Pampas. Science Advances, 5(3), eaau4546. https://doi.org/10.1126/sciadv.aau4546

  32. Zhang, H., Cheng, H., Christoph Spötl, Zhang, X., Cruz, F. W., Sinha, A., Auler, A. S., Stríkis, N. M., Wang, X., Gayatri Kathayat, Li, X., Li, H., Pérez‐Mejías, C., Cai, Y., Ning, Y., & Richard Lawrence Edwards. (2021). Gradual South‐North Climate Transition in the Atlantic Realm Within the Younger Dryas. Geophysical Research Letters, 48(8). https://doi.org/10.1029/2021gl092620

  33. Ilham Bouimetarhan, Cristiano Mazur Chiessi, González, C., Dupont, L., Voigt, I., Prange, M., & Zonneveld, K. A. F. (2018). Intermittent development of forest corridors in northeastern Brazil during the last deglaciation: Climatic and ecologic evidence. Quaternary Science Reviews, 192, 86–96. https://doi.org/10.1016/j.quascirev.2018.05.026

  34. Piacsek, P., Behling, H., Ballalai, J. M., Nogueira, J., Venancio, I. M., & Albuquerque, A. L. S. (2021). Reconstruction of vegetation and low latitude ocean-atmosphere dynamics of the past 130 kyr, based on South American montane pollen types. Global and Planetary Change, 201, 103477. https://doi.org/10.1016/j.gloplacha.2021.103477

  35. Cassino, R. F., Ledru, M.-P., Santos, R. de A., & Favier, C. (2020). Vegetation and fire variability in the central Cerrados (Brazil) during the Pleistocene-Holocene transition was influenced by oscillations in the SASM boundary belt. Quaternary Science Reviews, 232, 106209. https://doi.org/10.1016/j.quascirev.2020.106209

  36. Horák‐Terra, I., Antonio Martínez Cortizas, Fernandes, C., Alexandre Christófaro Silva, Mighall, T., Barbosa, P., Carlos Victor Mendonça‐Filho, De, E., Francisco Willian Cruz, & Vidal‐Torrado, P. (2020). Late Quaternary vegetation and climate dynamics in central‐eastern Brazil: insights from a ~35k cal a bp peat record in the Cerrado biome. JQS. Journal of Quaternary Science/Journal of Quaternary Science, 35(5), 664–676. https://doi.org/10.1002/jqs.3209

  37. Novello, V. F., Cruz, F. W., Vuille, M., Stríkis, N. M., Edwards, R. L., Cheng, H., Emerick, S., de Paula, M. S., Li, X., Barreto, E. de S., Karmann, I., & Santos, R. V. (2017). A high-resolution history of the South American Monsoon from Last Glacial Maximum to the Holocene. Scientific Reports, 7(1), 44267. https://doi.org/10.1038/srep44267

  38. Novello, V. F., Cruz, F. W., McGlue, M. M., Wong, C. I., Ward, B. M., Vuille, M., Santos, R. A., Jaqueto, P., Pessenda, L. C. R., Atorre, T., Ribeiro, L. M. A. L., Karmann, I., Barreto, E. S., Cheng, H., Edwards, R. L., Paula, M. S., & Scholz, D. (2019). Vegetation and environmental changes in tropical South America from the last glacial to the Holocene documented by multiple cave sediment proxies. Earth and Planetary Science Letters, 524, 115717. https://doi.org/10.1016/j.epsl.2019.115717

  39. Maksic, J., Shimizu, M. H., de, M., Igor Martins Venancio, Cardoso, M., & Ferreira, F. S. (2018). Simulation of the Holocene climate over South America and impacts on the vegetation. The Holocene, 29(2), 287–299. https://doi.org/10.1177/0959683618810406

  40. Clapperton, C. M. (1993). Nature of environmental changes in South America at the Last Glacial Maximum. Palaeogeography, Palaeoclimatology, Palaeoecology, 101(3-4), 189–208. https://doi.org/10.1016/0031-0182(93)90012-8

  41. Sato, H., Kelley, D. I., Mayor, S. J., Martin Calvo, M., Cowling, S. A., & Prentice, I. C. (2021). Dry corridors opened by fire and low CO2 in Amazonian rainforest during the Last Glacial Maximum. Nature Geoscience, 14(8), 578–585. https://doi.org/10.1038/s41561-021-00777-2

  42. Colinvaux, P. A., De Oliveira, P. E., & Bush, M. B. (2000). Amazonian and neotropical plant communities on glacial time-scales: The failure of the aridity and refuge hypotheses. Quaternary Science Reviews, 19(1), 141–169. https://doi.org/10.1016/S0277-3791(99)00059-1

  43. Akabane, T. K., Chiessi, C. M., Hirota, M., I. Bouimetarhan, Prange, M., S. Mulitza, Jr, B., C. Häggi, Staal, A., Lohmann, G., Boers, N., Daniau, A. L., Oliveira, R. S., Campos, M. C., Shi, X., & De, E. (2024). Weaker Atlantic overturning circulation increases the vulnerability of northern Amazon forests. Nature Geoscience. https://doi.org/10.1038/s41561-024-01578-z

  44. Leite, Y. L. R., Costa, L. P., Loss, A. C., Rocha, R. G., Batalha-Filho, H., Bastos, A. C., Quaresma, V. S., Fagundes, V., Paresque, R., Passamani, M., & Pardini, R. (2016). Neotropical forest expansion during the last glacial period challenges refuge hypothesis. Proceedings of the National Academy of Sciences, 113(4), 1008–1013. https://doi.org/10.1073/pnas.1513062113

  45. De Oliveira, P., Raczka, M. F., McMichael, C., Pinaya, J., & Bush, M. B. (2019). Climate change and biogeographic connectivity across the Brazilian cerrado. Journal of Biogeography, 47(2), 396–407. https://doi.org/10.1111/jbi.13732

  46. Ledru, M. P. (2002). Late Quaternary history and evolution of the cerrados as revealed by palynological records. In The Cerrados of Brazil: ecology and natural history of a Neotropical savanna (pp. 33-50). Columbia University Press. 

  47. Lima-Ribeiro, M. S., Varela, S., Nogués-Bravo, D., & Diniz-Filho, J. A. F. (2012). Potential suitable areas of giant ground sloths dropped before its extinction in South America: the evidences from bioclimatic envelope modeling. Natureza & Conservação, 10(2), 145–151. https://doi.org/10.4322/natcon.2012.022

  48. Araújo, T., Machado, H., Mothé, D., & dos Santos Avilla, L. (2021). Species distribution modeling reveals the ecological niche of extinct megafauna from South America. Quaternary Research, 1–8. https://doi.org/10.1017/qua.2021.24

  49. Leonardi, M., Boschin, F., Boscato, P., & Manica, A. (2022). Following the niche: the differential impact of the last glacial maximum on four European ungulates. Communications Biology, 5(1). https://doi.org/10.1038/s42003-022-03993-7

  50. Kelly, M. A., Russell, J. M., Baber, M. B., Howley, J. A., Loomis, S. E., Zimmerman, S., Nakileza, B., & Lukaye, J. (2014). Expanded glaciers during a dry and cold Last Glacial Maximum in equatorial East Africa. Geology, 42(6), 519–522. https://doi.org/10.1130/g35421.1

  51. H. Gregory McDonald. (2023). A Tale of Two Continents (and a Few Islands): Ecology and Distribution of Late Pleistocene Sloths. Land, 12(6), 1192–1192. https://doi.org/10.3390/land12061192

  52. Shakun, J. D., Clark, P. U., He, F., Marcott, S. A., Mix, A. C., Liu, Z., Otto-Bliesner, B., Schmittner, A., & Bard, E. (2012). Global warming preceded by increasing carbon dioxide concentrations during the last deglaciation. Nature, 484(7392), 49–54. https://doi.org/10.1038/nature10915

  53. Gerhart, L. M., & Ward, J. K. (2010). Plant responses to low [CO2] of the past. New Phytologist, 188(3), 674–695. https://doi.org/10.1111/j.1469-8137.2010.03441.x

  54. Ward, J. K., Myers, D. A., & Thomas, R. B. (2008). Physiological and Growth Responses of C3 and C4 Plants to Reduced Temperature When Grown at Low CO2 of the Last Ice Age. Journal of Integrative Plant Biology, 50(11), 1388–1395. https://doi.org/10.1111/j.1744-7909.2008.00753.x

  55. Doughty, C. E., Faurby, S., & Svenning, J.-C. (2015). The impact of the megafauna extinctions on savanna woody cover in South America. Ecography, 39(2), 213–222. https://doi.org/10.1111/ecog.01593

  56. Owen-Smith, N. (1987). Pleistocene extinctions: the pivotal role of megaherbivores. Paleobiology, 13(3), 351–362. https://doi.org/10.1017/s0094837300008927

  57. Malhi, Y., Doughty, C. E., Galetti, M., Smith, F. A., Svenning, J.-C., & Terborgh, J. W. (2016). Megafauna and ecosystem function from the Pleistocene to the Anthropocene. Proceedings of the National Academy of Sciences, 113(4), 838–846. https://doi.org/10.1073/pnas.1502540113

  58. Venancio, I. M., Shimizu, M. H., Santos, T. P., Lessa, D. O., Portilho-Ramos, R. C., Chiessi, C. M., Crivellari, S., Mulitza, S., Kuhnert, H., Tiedemann, R., Vahlenkamp, M., Bickert, T., Sampaio, G., Albuquerque, A. L. S., Veiga, S., Nobre, P., & Nobre, C. (2020). Changes in surface hydrography at the western tropical Atlantic during the Younger Dryas. Global and Planetary Change, 184, 103047. https://doi.org/10.1016/j.gloplacha.2019.103047

  59. Loponte, D., Okumura, M., & Carbonera, M. (2015). New records of fishtail projectile points from Brazil and its implications for its peopling. Journal of Lithic Studies, 3(1). https://doi.org/10.2218/jls.v3i1.1312

  60. Bird, D., Miranda, L., Vander Linden, M., Robinson, E., Bocinsky, R. K., Nicholson, C., Capriles, J. M., Finley, J. B., Gayo, E. M., Gil, A., d’Alpoim Guedes, J., Hoggarth, J. A., Kay, A., Loftus, E., Lombardo, U., Mackie, M., Palmisano, A., Solheim, S., Kelly, R. L., & Freeman, J. (2022). p3k14c, a synthetic global database of archaeological radiocarbon dates. Scientific Data, 9(1), 27. https://doi.org/10.1038/s41597-022-01118-7


Tristan Rapp
Next

Victim to the Woods: Souslik and Steppe Dynamics in the European Holocene