A pollen-based pseudobiomisation approach to anthropogenic land-cover change

摘要

The degree of anthropogenic modification of land cover through the mid to late Holocene is of significant interest for archaeologists, climate modellers and conservation ecologists, amongst others. Spatially extensive pollen data provide an appropriate resource for the reconstruction of land-cover change; however, traditional simple methods (e.g. the AP/NAP ratio) do not provide the degree of detail required for these user groups. A range of more sophisticated methods have been developed (e.g. mechanistic model-based methods), but they require a high degree of understanding of the pollen-vegetation relationship, which is resource- and time-consuming.This paper proposes a method of intermediate complexity by which major land-cover changes can be assessed for individual sites and regions. A pseudobiomisation approach is used, in which pollen taxa are assigned to different land-cover classes (LCC), and the sum of adjusted pollen proportions for each class used to determine an LCC affinity score for individual pollen samples within stratigraphic pollen sequences. In a pilot study, data from pollen core sites from northwest Scotland and southwest England for the last 8000 cal. BP have been classified into ten land-cover classes, including predominantly wooded, semi-open and open types. Results highlight how land-cover change was punctuated rather than gradual through the mid to late Holocene, and show how habitat diversity increased after ～6000 cal. BP with the arrival of agriculture, before reaching a minimum in early modern times.This pollen-based method is potentially applicable to the reconstruction of long-term land-cover change within Europe and other temperate-zone regions.