Calibrating Vegetation Cover and Grassland Pollen Assemblages in the Flint Hills of Kansas, USA

Abstract

Grassland cover and composition respond to climate and\rhave undoubtedly changed during the Holocene, but quantitative\rreconstructions from fossil pollen have been vague about spatial scale and\rtaxon-specific cover. Here, we estimate\rthe relevant source area of pollen for sedimentary basins approximately 50 m in\rradius, and we report pollen productivity estimates for 12 plant taxa in the\rtallgrass prairies of central North America. Both relevant source area of pollen\rand pollen productivity estimates were calculated via the Extended R-Value\rModel. To obtain these estimates, we collected and quantified the pollen found\rin surface sediment samples from 24 ponds across the study area. Vegetation was\rsurveyed in the field in a 100 m radius around each pond, and vegetation\rmaps from the Kansas Gap Analysis Project (GAP) were used to a radius of 2 km.\rPollen fall speeds were calculated according to Stoke’s Law. Pollen assemblages\rfrom basins approximately 50 m in radius have a relevant source area of 1060 m\rin this grassland landscape. Pollen productivity estimates range from 0.02 to over\r30 among the 12 taxa: Artemisia, Ambrosia, Asteraceae, Chenopodiaceae, Cornus, Fabaceae, Juniperus, Maclura, Poaceae, Populus, Quercus, and Salix.\rWoody taxa generally have higher pollen productivity than herbaceous taxa\r(except for Chenopodiaceae and Ambrosia).