A large-scale erosion anomaly (2nd c. BC- 4th c. AD) in NW Alps: a locally-defined onset of the Anthropocene

International audienceThe research program Pygmalion consisted in a 4-yearslong unprecedented scientific effort to investigate complexhuman-climate-environment interactions in NorthernFrench Alps. Thanks to a wide geographic cover, it ledto evidence an erosion anomaly that began in 200 BC,at the time of the Allobroges Gallic tribe, just prior theirembedment into the Roman Empire and ended at the endof the Roman period (around 400 AD). Indeed during thewhole Holocene period, a strong link has been evidencedbetween climate and erosion fluxes at a regional scale.This link was obviously broken at the end of Iron Age.At that time, huge erosion fluxes were recorded fromsedimentary records taken in lakes with small catchmentareas, both in high and low elevation sites. Even regionaldetrital fluxes in Lake Bourget, representative of a 4000km2 catchment area, present an anomaly, whereas noglacier advance nor temperature drop were recorded inthe area. Moreover paleovegetation and fire regime datapoint major changes in vegetation cover that pinpointsa drastic change in land-use practices at regional scalesuggesting a reinforcement of pasturing activities. In LakeAnterne high-altitude (2060 m asl) catchment area, thisis confirmed by DNA barcoding data acquired on fromlake sediment evidencing the unprecedented intensivepresence of both sheeps and cows as well as by thediscovery of a ruined shepherd caban, the first occupationof which being dated 2nd c. BC. Finally, based on themeasurement of molecular biomarker, we evidenced thisperiod as peak in broomcorn millet cultivation aroundLake Bourget and Lake Paladru.Based on above-mentioned field evidences we argue thatchanges in land-use practices between 200 BC and 400 ADhad such an important impact that it significantly affectedthe erosion budget at a regional scale. From our datasetcovering the whole Holocene, this was the first time, in theconsidered area, that human activities became intensiveenough to significantly alter an important geodynamicprocess such as erosion. Following various authors wehence wonder if this could not define a local onset of theAnthropocene period characterised by the emergence ofhumans as a major geologic agent