Pollen productivity, dispersal, and correction factors for major tree taxa in the Swiss Alps based on pollen-trap results

Annual pollen traps in the period AD 1997-2005 were analysed to determine pollen-accumulation rates (PAR; grains deposited per unit-area and unit time) for major tree taxa in four regions of the Swiss Alps. The relation between vegetation cover and pollen deposition was then analysed. Different pollen-dispersal models and functions were applied and tested, including both empirical and mathematical weighting of the vegetation. Specifically, assumptions concerning the regional pollen component were elaborated. Absolute pollen-productivity (APP; grains released per unit vegetation cover and unit time) and correction factors (defined here as the linear relation between deposited pollen and vegetation given a fixed set of assumptions) were estimated for Larix, Picea, Pinus total, Pinus cembra, Pinus mugo, and Alnus alnobetula (=A. viridis). Pollen dispersal estimated for a realistic moderate wind speed (3-5 m s(-1)) and height of pollen release (1 m) results in an absolute pollen-productivity for Larix of c. 8000 grains cm(-2) yr(-1), Picea 12,000, and Pinus total 55,000. Heavy pollen grains such as Larix are less well represented at some distance from the source, both empirically and according to the dispersal functions. The results are fairly similar between the tested dispersal models and functions, suggesting rather robust values. Compared to relative pollen-productivity, absolute pollen pollen-productivity has the advantage that it is a physical taxon-specific parameter independent of other taxa. It therefore allows single-taxon vegetation reconstructions based on PAR data, which may be of special importance in areas with limited vegetation cover. 2008 Elsevier B.V. All rights reserved.