Do moss samples, pollen traps and modern lake sediments all collect pollen in the same way? A comparison from the forest limit area of northernmost Europe

• Published in: Vegetation history and archaeobotany Vol. 21; no. 3; pp. 187 - 199
• Main Authors: Lisitsyna, Olga V, Hicks, Sheila, Huusko, Antti
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer-Verlag 01-05-2012; Springer; Springer Nature B.V
• Subjects: Anthropology; Archaeology; Betula pubescens; Betula pubescens subsp. tortuosa; Biogeosciences; Birch trees; Boreal forests; climate; Climate Change; Cluster analysis; Coniferous forests; Cyperaceae; Earth and Environmental Science; Earth Sciences; Earth, ocean, space; Evergreen trees; Exact sciences and technology; Forests; Freshwater; Hardwood trees; herbs; Lake sediments; Lakes; Marine and continental quaternary; Mosses; mosses and liverworts; Original Article; Paleobotany; Paleontology; Picea abies; Pine trees; Pinus; Pinus sylvestris; Pollen; Principal components analysis; Sediment samplers; Sediments; Shrubs; Surficial geology; Traps; Trees; Vegetation; Woodlands; Scandinavia; Russia; Fennoscandia; Europe; Finland; PN, Arctic; Europe, Fennoscandia; Pollen trap; Northern forest limit; Northern Fennoscandia; Modern pollen deposition; Moss sample; Lake sediment sample; Quaternary; upper Quaternary; paleoclimate; lake sediments; pollen diagrams; Cenozoic; Phanerozoic; cluster analysis; pollen analysis; biostratigraphy; Holocene
• ISSN: 0939-6314; 1617-6278
• DOI: 10.1007/s00334-011-0335-x

Moss polsters, pollen traps and lake surface sediment samples are commonly used as climate calibration data or as modern analogues for reconstructing vegetation from fossil profiles, but the differences in pollen content between these media have received little attention. This study aims to analyse how the three media differ in reflecting individual vegetation types and spatial differences in vegetation. 119 modern samples (64 moss polsters, 37 lake surface sediment samples and 18 pollen traps from which a collection was made annually) were taken from northern Fennoscandia and the Kola Peninsula as a broad transect crossing the northernmost forest limits of Betula pubescens ssp. czerepanovii (mountain birch), Pinus sylvestris (Scots pine) and Picea abies (Norway spruce). The pollen assemblages from these samples were compared with the surrounding vegetation visually and via PCA (principle components analysis) and cluster analysis. Both comparisons allow a correct distinction between pollen assemblages of arctic/alpine heath, mountain birch dominated areas, and boreal coniferous forests. The differences between the vegetation zones are stronger than the differences between the sampling media. Nevertheless, lake sediment samples from the mountain birch woodland zone tend to overestimate pine and underestimate birch. Pollen traps are biased towards lower tree pollen percentages and higher values of shrubs, herbs and Cyperaceae. This bias is especially strong in traps that have missing years in the data. Irrespective of the vegetation zone, pollen traps tend to have lower Pinus pollen percentages than in the adjacent moss polsters.