Conflicting 14C scenarios in the Sopeña cave (northern Iberia): Dating the Middle-Upper Palaeolithic boundary by non-ultrafiltered versus ultrafiltered AMS 14C

Abstract

Sopeña is a Palaeolithic rock shelter with a pristine, horizontal and deep stratigraphic sequence. It was used almost continuously by human beings during all of the last glacial cycle, from at least 60,000 years ago till some 20,000 years ago. Sopeña yielded well stratified deposits corresponding to the Mousterian, Early Upper Palaeolithic and Gravettian. Several accelerator mass spectrometry (AMS) 14C dates for Sopeña have been published in various papers and those dates were seen by some scholars as contradictory. Here, we aim to set straight the record on the known dates for the Neanderthal-Homo sapiens boundary in Sopeña, by showing how these dates are consistent both with sample vertical depth in the strata and cultural adscription. The Sopeña sequence of AMS 14C dates suggests a scenario of a long Neanderthal local survival followed by a very quick replacement by modern humans, all around 35.5 ky BP. Despite the stratigraphic and cultural consistency of these dates, all made by laboratories with long years of experience and carried out within the last 12 years, the suggested scenario is significantly more recent than what has been proposed as a consensus by recent publications (Maroto et al., 2012, Higham et al., 2014) that rely increasingly on ultrafiltration techniques for the extraction of collagen prior to its AMS 14C dating. To better delimit the chronology of the disappearance of the Mousterian in Sopeña, we obtained new dates by both ultrafiltered (UF) and non-ultrafiltered (non-UF) AMS 14C methods on fresh samples from the same levels, locations and depths. The interpretation of the new dates and their reliability is not without issues. Overall, both non-UF and UF dates support an earlier sequence of events by some millennia than that earlier proposed, with a survival of the Mousterian arguably up to c. 40 ky BP which is still a few millennia later than c. 47 ky BP currently proposed for the region (Higham et al., 2014), and the earliest arrival of modern humans around 38 ky BP, matching the proposal for this region of Iberia by Maroto et al. (2012). We do value as more parsimonious, the older chronological scenario for Neanderthal disappearance and the arrival of modern humans in Sopeña as presented here. Although, some issues put forward by the new dates, both non-UF and UF, seem difficult to interpret and do not put away the alternative scenario of a quick replacement as suggested by the earlier set of non-UF AMS dates.