Sprecher
Beschreibung
Archaeology uses palynology to study past human-landscape interactions. Truly quantitative vegetation reconstructions of archaeological sites and landscapes are crucial for this research, but these are notoriously difficult to obtain from fossil pollen records. Recent advancements in palynological modelling methods like the Landscape Reconstruction Algorithm (LRA) provide an important step forward. Here I present an exceptional case study of the Middle Palaeolithic site Neumark-Nord 2 (Sachsen-Anhalt, Germany). At this site, Neanderthals were present semi-permanently for ca. 1500 years in the early temperate Eemian, 125 ka ago. By applying the LRA with different input parameter values of relative Pollen Productivity Estimates to Neumark-Nord 2 and three neighbouring sites, the plausible range of past vegetation composition was reconstructed in detail on the regional (50 km) and local (200 m) scale during the period of Neanderthal habitation. The regional vegetation consisted mostly of light forest, with small elements of open land and closed-canopy forest, similar to reconstructions based on pollen percentages alone. The local vegetation was almost completely open with only some trees, contrary to the pollen percentages that suggest dominance of light forest. The quantitative vegetation reconstructions are partially different from qualitative reconstructions based on other palaeoecological proxies, thus showing the importance of a combination of quantitative modelling methods and qualitative approaches to better interpret palaeoecological and archaeological datasets. The approach used here shows that the LRA can be applied to non-Holocene archaeological sites to provide new insights into hominin-landscape interactions at high spatial and temporal resolution.