Calibrating vegetation cover and pollen assemblages in the Flint Hills of Kansas, U.S.A.

Abstract

The quantitative relationship between pollen assemblages in sediment and vegetation cover is largely unknown because many factors influence this relationship. This lack of quantitative relationship is particularly acute in grassland regions, where both past and future climate change have the potential to determine grassland composition and cover. The tool used to reconstruct past grassland cover is the relative abundance of distinct fossil pollen types preserved in sediment. However, the interpretation of grassland pollen assemblages as grassland vegetation types needs to be refined to improve these reconstructions. Using pollen found in the surface sediments from 24 artificially-constructed ponds in the Flint Hills ecoregion of Kansas, USA, I examined relationships between pollen and vegetation in the tallgrass prairie biome, which includes woody components. By comparing the pollen data to field-surveyed vegetation data and land cover classifications taken from Kansas Gap Analysis Program data, I correlated pollen and vegetation in this ecoregion. Pollen productivity estimates for Artemisia, Ambrosia, Asteraceae, Chenopodiaceae, Cornus, Fabaceae, Juniperus, Maclura, Poaceae, Populus, Quercus, and Salix were calculated via the Extended R-Value Model. Common pollen types identified in sediments are mostly herbaceous grassland plant species such as Poaceae, Artemisia, and Ambrosia, but woody plants such as Populus, Quercus, and Juniperus are also represented. PPEs have been calculated for four of these taxa in Europe, and values from the Flint Hills are higher. These are the first PPEs reported for eight of these taxa. This research will further advance quantitative vegetation reconstructions in the Great Plains of North America and refine interpretations of how climate change affects grasslands.