Application of straight-chain lipids as biomarkers in the reconstruction of the landscape in a driftsand from the Netherlands.

Abstract

Historical land use systems as shifting cultivation and plaggen agriculture on nutrient poor sandy soils caused serious degradation of soil and vegetation, resulting in Late Holocene sand drifting. Locally the coversand landscape was transformed into a driftsand landscape. On sheltered deposition sites polycyclic driftsand profiles developed. They are valuable geo-ecological records of alternating instable (sand drifting) and stable (soil formation) phases in landscape development. Previous research has been done on the reconstruction of the landscape genesis, based on pollen analysis and radiocarbon dating of polycyclic driftsand deposits. Recently, the application of Optically Stimulated Luminescence (OSL) dating improved the geochronology. The principles of pollen transport and deposition does not allow to select the plant species from pollen spectra, responsible for the on-site production of soil organic matter during soil formation in stable periods. A possible proxy for this, that could add information, is the application of straight-chain lipids analysis. In this pilot study we introduced the biomarker method to separate local and regional species, found in pollen spectra. Different species often have their own specific patterns of alcohols and alkanes of different carbon chainlengths. Therefore, comparing the straight-chain lipid patterns of both (expected) vegetation species and selected soil samples is a means to assess the landscape history. In order to do this the recently developed VERHIB model was used in combination with a cluster analysis. The results obtained from VERHIB and the cluster analysis suggest a dynamic sequence of dominant vegetation species, which in majority support the existing theories about the landscape history of polycyclic deposits in the Netherlands. Some results (e.g. the distribution of Pinus throughout the profile) are much harder to link to earlier research and are likely to be incorrect. Although a couple of problems arose during the analytical process, the application of straight-chain lipids is found to be certainly a very promising technique. Nevertheless, a lot more research has to be carried out to draw more solid conclusions on the strength of the biomarker technique in sandy environments and its impact on the ideas about the cultural-ecological history of polycyclic soils in the Netherlands.