Rapid range shifts and megafaunal extinctions associated with late Pleistocene climate change

The cause of the late Quaternary extinctions across the globe has been debated extensively. While numerous hypotheses exist, including disease and extraterrestrial impacts, the leading theories center on human hunting and climate change as primary drivers. In North America, the near-simultaneous occurrence of major human migrations and climate shifts complicates efforts to isolate their individual contributions. This challenge is amplified by the scarcity of well-dated archaeological and paleontological sites spanning this extinction period. Archaeological evidence indicates that Paleo-Indian groups hunted at least six of the 36 megafaunal genera in North America, supporting the human hunting hypothesis. However, despite widespread human presence across the Americas by at least 13,000 cal BP, their population density remained low. The extent to which hunting impacted megafauna populations thus remains uncertain. This ambiguity has led to the "one-two punch" hypothesis, suggesting that the combined effects of climate change and human impacts caused the extinction of North American megafauna around 13.0-12.5 ka cal BP. Unlike Australia and New Zealand, where human arrival and major climate changes were temporally separated by millennia, North America experienced these events concurrently. By 14,700 cal BP, rising temperatures during the Bølling-Allerød warming event led to the retreat of continental ice sheets, ending the Last Glacial Maximum. Two millennia later, this warming trend reversed abruptly during the Younger Dryas Cooling Event (12.9-11.7 ka cal BP), marked by rapid temperature drops in the Northern Hemisphere. Understanding the biodiversity response to these climatic fluctuations at a local scale could provide valuable insights into the causes of this global mass extinction event. However, current temperature fluctuation estimates for the late Quaternary rely on Greenland ice core data, which aren't easily transferable to central North America. The impact of the Younger Dryas on the Great Plains, for instance, remains debated, with descriptions ranging from "near glacial conditions" to mean annual temperatures only ~5°C cooler than present. Moreover, traditional mean annual temperature proxies don't account for seasonality during the Younger Dryas; the apparent cooling could reflect increased seasonality with cold winters but relatively warm summers. To investigate the ecosystem's response to Pleistocene-Holocene climatic fluctuations, the researchers sequenced ancient DNA from vertebrate fossils and sediment from Hall's Cave, a Texas limestone cavern with a well-dated sedimentary record spanning from the Last Glacial Maximum to the present. The cave's finely stratified sediments deposited during the Younger Dryas make it an ideal study site.

The paper reviews existing literature on the late Quaternary extinctions, highlighting the debate between human hunting and climate change as the primary causes. It cites studies supporting the involvement of Paleo-Indians in megafauna hunting but acknowledges the uncertainty regarding the impact of their hunting pressure given their relatively small population size. The "one-two punch" hypothesis, attributing extinctions to the combined effects of climate change and human impacts, is also discussed. The authors contrast the North American situation, with its concurrent human arrival and climate change, with Australia and New Zealand, where these events were temporally distinct. The literature also highlights the challenges in accurately estimating temperature fluctuations in central North America during the late Quaternary due to reliance on Greenland ice core data, emphasizing the limitations of these proxies in capturing regional variations and seasonality.

The study employed ancient DNA (aDNA) sequencing from vertebrate fossils and sediments excavated from Hall's Cave, Texas. Two aDNA approaches were used: bulk bone metabarcoding (BBM) and sedimentary ancient DNA (sedaDNA). For BBM, bulk-bone samples from strata dating from the Last Glacial Maximum to the early Holocene were analyzed to characterize the faunal assemblage. Short mitochondrial barcoding regions (12S rRNA and 16S rRNA) were targeted, generating millions of reads. SedaDNA analysis involved sequencing two short chloroplast loci (trnL and rbcL) from sediment samples to characterize the floral assemblage. Again, millions of reads were generated. The researchers explored aDNA preservation throughout the chronological sequence, finding no evidence of systematic changes and concluding that the aDNA assemblages reflected temporal shifts in community composition rather than preservation bias. Samples were grouped into four climate intervals: Last Glacial Maximum (LGM), Bølling-Allerød, Younger Dryas (YD), and Early Holocene. The onset of the Younger Dryas was defined based on a visible "rancholabrean mat" in the cave profile, dating to 12.6 ka cal BP. The age-depth model used in the study included previously published radiocarbon dates from the site, allowing for age estimation of all samples. Both morphological and bulk-bone records were analyzed and compared with sedimentary aDNA record to assess the consistency and differences between the methods. Statistical analyses, including Non-metric multidimensional scaling (NMDS) and alpha-diversity calculations, were performed to analyze changes in community composition and diversity over time. Habitat modeling, based on geographic ranges and climate data from WorldClim, was used to infer past temperatures at Hall's Cave based on the presence of species with known climatic niches. The aDNA plant data were also compared with existing palynological data from the site to assess consistency and potential discrepancies between the two proxies.

The study revealed high vertebrate diversity at Hall's Cave, with at least 100 species identified using BBM. The faunal assemblage was largely consistent with previous morphological analyses, but aDNA analysis added seven previously unidentified mammal genera and numerous bird, amphibian, reptile, and fish species. Sedimentary aDNA showed a discrepancy in felid detection, suggesting additional DNA sources beyond bone for these species in the cave sediments. Analysis of plant sedaDNA revealed a diverse floral record, although the abundance of certain taxa, like hackberry, didn't necessarily reflect their dominance in the local flora. The results largely agreed with previous palynological studies, except for *Pinus*, whose absence in the aDNA data suggested its absence in the local vicinity, contrasting with its presence in the pollen record due to the latter's ability to travel long distances. NMDS analysis showed significant shifts in both plant and animal assemblages across the Late Pleistocene-Early Holocene transition. Vertebrate diversity significantly declined from the LGM and Bølling-Allerød to the YD and Early Holocene, a pattern observed across various taxonomic groups except small mammals. Plant diversity also declined during the YD but recovered in the Early Holocene. Small mammal diversity closely mirrored Greenland temperature fluctuations, demonstrating range shifts in response to temperature changes; cold-adapted species were present during the LGM and YD, while warm-adapted species appeared during warmer periods. The presence of certain species in different strata allowed inference of past temperatures at Hall's Cave; the LGM was significantly colder than present, while the Early Holocene was warmer. The disappearance of burrowing mammals indicated significant decreases in soil depth, suggesting a regional drying trend. The aDNA record supports a vegetation change from mesic grassland to open woodland by the end of the LGM. The disappearance of prairie chickens, grassland species, and the appearance of woodland species like bobwhite and wild turkey, aligns with this transition. This change is consistent with the sedaDNA data, which shows a progressive shift to drier vegetation with decreased tree cover during the YD. The Holocene is marked by increased juniper and a drier woodland habitat. The large mammal extinctions coincided with the habitat shift and decreased plant diversity. The presence of non-analogous faunas in Pleistocene strata further suggests an ecosystem collapse, with species pairings not observed in present-day North American faunas.

The study's findings provide multiple lines of evidence for dramatic ecological changes in central Texas between the LGM and the Holocene. The LGM featured a mesic grassland with thick soils and diverse burrowing mammals, maintained by large grazers. The Bølling-Allerød witnessed a transition to open woodland, coinciding with the loss of several large grazers. The YD brought a decline in both plant and animal diversity and the extinction of large mammals. The Early Holocene saw a recovery in plant diversity and a shift to live oak-juniper woodland. However, the decimated faunal diversity, especially megafauna, did not recover. The contrasting responses of plants and vertebrates to the YD suggest that factors beyond climate, potentially human hunting, contributed to the permanent loss of large mammal diversity. This hypothesis is supported by the rapid responses of rodents, mirroring plant responses to climate changes via migration, unlike the large mammals which faced both climate change and hunting pressure. The high-resolution aDNA data from Hall's Cave contrast with other sites, such as Rancho La Brea, highlighting the exceptional preservation at this site. The authors suggest multidisciplinary follow-up studies to further quantify past climates, biodiversity, and biotic shifts.

This study, using ancient DNA from Hall's Cave, reveals dramatic ecological shifts in central Texas during the Pleistocene-Holocene transition. The differing responses of plants and large mammals to Younger Dryas climate change strongly suggest a role for human hunting in the megafaunal extinctions, in addition to climate change. The exceptional preservation of aDNA at Hall's Cave provides a unique opportunity for future multidisciplinary research to further refine our understanding of past ecosystems and extinction events.

The study focuses on a single cave site in central Texas, limiting the generalizability of findings to other regions. While the researchers addressed aDNA preservation bias, taphonomic processes could still influence the fossil record, although they attempt to minimize this through detailed methodological analysis. The reliance on existing radiocarbon dates for the age-depth model introduces some uncertainties into age estimations of samples analyzed. The analysis focuses primarily on taxonomic resolution at the genus level, limiting the ability to fully track fine-scale changes in biodiversity. Finally, while the study suggests human hunting played a role, it does not provide direct evidence of hunting at Hall's Cave.