Extinction chronology of the woolly rhinoceros Coelodonta antiquitatis in the context of late Quaternary megafaunal extinctions in northern Eurasia
Megafaunal extinctions in northern Eurasia (excluding Mediterranean islands) since the Last Interglacial claimed about 37% of species with body weights >44 kg. Here we review the dating evidence for the timings of these extinctions, which were staggered over tens of millennia. Moreover, individual species disappeared at different times in different geographical areas. For example, cave bear probably disappeared ca. 30.5-28.5 ka, at approximately the onset of GS-3 (beginning of 'LGM'), whereas cave lion survived until the Lateglacial ca 14 ka. Others survived into the Holocene: woolly mammoth until ca 10.7 ka in the New Siberian Islands and ca 4 ka on Wrangel Island, giant deer to at least 7.7 ka in western Siberia and European Russia. It is evident that climatic and vegetational changes had major impacts on species' ranges, and moreover the contrasting chronologies and geographical range contractions are consistent with environmental drivers relating to their differing ecologies. However, the possible role of humans in this process has still to be satisfactorily explored. We present a detailed review of the radiocarbon record and extinction chronology of an exclusively northern Eurasian species, the woolly rhinoceros Coelodonta antiquitatis, an animal adapted to feeding on low-growing herbaceous vegetation and a dry climate with minimal snowfall. During the Last Glacial the overall (time-averaged) range of C. antiquitatis extended across most of northern Eurasia, but the species failed to reach North America. On the basis of 233 ultrafiltered AMS radiocarbon dates, together with 50 others which pass our auditing criteria, we reconstruct a detailed chronology for this species. C. antiquitatis was widespread over most of the time span covered by radiocarbon dating, but from ca 35 ka (calibrated) it apparently contracted towards the east, culminating in its probable extinction ca 14 ka, with the latest dates from north-eastern Siberia. It disappeared from Britain, at the western, more oceanic end of its range, at ca 35 ka BP apparently corresponding with the onset of Interstadial GI-7. In north-eastern Siberia C. antiquitatis survived through much of GI-1d,e and probably went extinct at or close to the onset of GI-1c (Allerød) ca 13.9 ka BP. Its final extinction probably relates to Lateglacial interstadial warming and increased precipitation (especially snowfall) accompanied by the replacement of low-growing herbaceous vegetation by shrubs and trees. The survival of woolly rhinoceros in north-eastern Siberia when it had disappeared further west may relate to the later persistence of open vegetation in that region. The timing of its final extinction could relate to the fact that the Lateglacial interstadial was significantly warmer than any other event in the previous 50 millennia. There are indications of gaps in the pattern of dates in certain areas that may have significance in terms of temporary absences of C. antiquitatis, some of which may correspond to interstadial phases. The radiocarbon record of C. antiquitatis is still poor in some potentially important areas, and further work will corroborate or reject the hypotheses of regional 'gaps' in occurrence, and might perhaps reveal currently unsuspected later survival.