Spatial scaling of pollen-plant diversity relationship in landscapes with contrasting diversity patterns

Abstract

Abstract Mitigating the effects of global change on biodiversity requires its understanding in the past. The main proxy of plant diversity, fossil pollen record, has a complex relationship to surrounding vegetation and unknown spatial scale. We explored both using modern pollen spectra in species-rich and species-poor regions in temperate Central Europe. We also considered the biasing effects of the trees by using sites in forests and open habitats in each region. Pollen samples were collected from moss polsters at 60 sites and plant species were recorded along two 1 km-transects at each site. We found a significant positive correlation between pollen and plant richness (alpha diversity) in both complete datasets and for both subsets from open habitats. Pollen richness in forest datasets is not significantly related to floristic data due to canopy interception of pollen rather than to pollen productivity. Variances (beta diversity) of the six pollen and floristic datasets are strongly correlated. The source area of pollen richness is determined by the number of species appearing with increasing distance, which aggregates information on diversity of individual patches within the landscape mosaic and on their compositional similarity. Our results validate pollen as a reconstruction tool for plant diversity in the past.

Plant phenotyping relevance

現代の花粉データと植物種多様性を比較し、花粉を過去の植物多様性再構築に用いる測定・推定手法を明示的に検証しているため、単なる生態学的なルーチン測定ではない。

Code and data availability

The paper's pollen and vegetation datasets are deposited on Zenodo and the analysis code is on GitHub, both with explicit availability statements.