Global late Quaternary megafauna extinctions linked to humans, not climate change

Abstract

The late Quaternary megafauna extinction was a severe global-scale event. Two factors, climate change and modern humans, have received broad support as the primary drivers, but their absolute and relative importance remains controversial. To date, focus has been on the extinction chronology of individual or small groups of species, specific geographical regions or macroscale studies at very coarse geographical and taxonomic resolution, limiting the possibility of adequately testing the proposed hypotheses. We present, to our knowledge, the first global analysis of this extinction based on comprehensive country-level data on the geographical distribution of all large mammal species (more than or equal to 10 kg) that have gone globally or continentally extinct between the beginning of the Last Interglacial at 132,000 years BP and the late Holocene 1000 years BP, testing the relative roles played by glacial-interglacial climate change and humans. We show that the severity of extinction is strongly tied to hominin palaeobiogeography, with at most a weak, Eurasia-specific link to climate change. This first species-level macroscale analysis at relatively high geographical resolution provides strong support for modern humans as the primary driver of the worldwide megafauna losses during the late Quaternary.

Keywords: climate change; macroecology; megafauna extinction; overkill; palaeoecology.

Figure 1.

Global maps of late Quaternary large mammal extinction severity, hominin palaeobiogeography, temperature anomaly and precipitation velocity. (a) The proportion of extinct large mammal species (more than or equal to 10 kg) in each TDWG country during the last 132 000 years, only counting extinctions earlier than 1000 years BP. (b) The cumulative number of extinct large mammal species occurring in each TDWG country. (c) Hominin palaeobiogeography (see the text for further explanation). (d) Mean anomaly in mean annual temperature between the LGM and today. (e) Mean velocity in annual precipitation between the LGM and today. TDWG countries shaded in dark grey were excluded from analyses. The climate change variables were standardized to range between 0 and 1.

Figure 2.

Proportion of extinct large mammals occurring in each TDWG country against mean annual temperature anomaly. Red crosses are countries occurring in the region of ‘H. sapiens-only’. Yellow crosses indicate countries occurring within the region ‘Archaic-combined’. Blue crosses indicate countries within the region ‘Homo-origin’. Lines of matching colours are predicted values based on a SAR model where hominin history and temperature anomaly with an interaction effect describe the proportion of extinct large mammals. Shaded areas represent 95% CIs. The statistical details are available in table 1.

Figure 3.

Box plot of proportion of large mammals lost per TWDG country in hominin palaeobiogeographic regions. Middle line and box represent the median and first to third quartiles, respectively, and whiskers extend to the furthest data point that is no more than 1.5 times the interquartile range.