Climate and humans set the place and time of Proboscidean extinction in late Quaternary of South America

• ENMs were used to model the Proboscidean distribution throughout the last 126 ka.
• Climate changes drove a wide contraction and fragmentation of geographical ranges.
• The strongest range shifts occurred just where humans not coexisted with megafauna.
• The refugee populations would be extirpated in only 550 years under human hunting.
• Climate and humans set then the place and time of Proboscidean extinction.

The late Quaternary extinctions have been widely debated for a long time, but the varying magnitude of human vs. climate change impacts across time and space is still an unresolved question. Here we assess the geographic range shifts in response to climate change based on Ecological Niche Models (ENMs) and modeled the timing for extinction under human hunting scenario, and both variables were used to explain the extinction dynamics of Proboscideans during a full interglacial/glacial cycle (from 126 ka to 6 ka) in South America. We found a large contraction in the geographic range size of two Proboscidean species studied (Cuvieronius hyodon and Notiomastodon platensis) across time. The largest contractions of their geographical ranges occurred in the northern part of South America, where we previously reported no evidence of coexistence among earliest humans and non-sloth megafauna, including Proboscideans. Our results herein support a strong effect of climatic changes on geographical range dynamics of Proboscideans throughout late Quaternary, although this does not fully support climate change as the single cause of their extinctions. We show that both Proboscideans were narrowly distributed on scattered patches of suitable habitats (i.e., refugia) around 11 ka, period in which the earliest humans potentially arrived in South America, increasing the population density thereafter. Under this overall unsuitable climatic condition at 11 ka, both Proboscideans would be extinct after around 550 years of human hunting, but if climatic conditions were suitable like in Last Glacial Maximum (LGM), the time-for-extinction would be at least 3 times longer under the same human hunting pressures. Thus, our findings support the “Broken Zig-Zag” model and show that South American Proboscideans might have been completely extinct due to human impacts during periods of climate crisis. We conclude, in agreement with an increasing body of evidence in the recent literature, that the late Quaternary megafaunal extinction event was the result of additive effects from different stressors, and that the relative magnitude of these impacts vary across space and time. Indeed, climate changes set the place where the Proboscideans were extinct in South America, whereas the humans set the time of these extinctions.

Figure optionsDownload high-quality image (292 K)Download as PowerPoint slide