Date(s) - 05/11/2018
4:00 pm - 5:30 pm
HS XVIIb Main Building UoC
by Dr. Martin Theuerkauf
Institute for Botany and Landscape Ecology, University of Greifswald
Vegetation reconstruction from pollen data, although dealing with counts and percentage values, has long been a mere semi-quantitative field. Pollen percentages do not directly represent past vegetation composition, because plants species produce pollen in very different amounts and with different dispersal patterns. This bias in pollen data is well known since the inception of the field 100 years ago, but correction for a long time relied on ad-hoc informed guesswork.
Today, a suite of methods is available that enable true quantitative interpretation of pollen data. The methods cover different spatial scales: REVEALS for example aims to translate pollen deposition from large lakes into regional vegetation composition. LOVE and Marco Polo reconstruct stand-scale vegetation composition using pollen data from very small sites. The extended downscaling approach and the multiple scenario approach explore vegetation patterns in landscapes using multiple pollen records.
So far these methods are still rarely applied, however, for a number of reasons. Most importantly, the underlying parameters pollen productivity and pollen dispersal require elaborate calibration and are hence so far available for some regions only.
The talk will give an overview of the main current approaches and illustrate their strength and weaknesses in a number of examples. It will then introduce the new ROPES approach. This method does not require pollen productivity as a parameter, and so may overcome major limitations of the present methods. It is suited to extend quantitative vegetation reconstruction into new regions and pre-Holocene periods, and allows analysis of long pollen records that cover several glacial/interglacial cycles.