中国科学院南京地质古生物研究所机构知识库

The basis for the interpretation of fossil-pollen assemblages in terms of vegetation and climate is the present-day relationship of vegetation and climate to pollen rain. Detailed modern pollen spectra from the Tibetan Plateau are described here to explore the relationship between modern pollen rain and vegetation. Two hundred and thirty four (234) pollen surface samples were collected from moss polsters, top soil, and lake surface sediments from forests, shrublands, shrub meadows, meadows, steppes, and deserts in the Tibetan Plateau. Pollen assemblages from each vegetation type are detailed described using pollen percentage data, and compared descriptively and numerically using cluster analysis. Pollen spectra from forests are characterized by high percentages of tree pollen types including Pinus, Abies, Picea, Quercus, and Betula. Pollen spectra from shrublands have highest amounts of shrub pollen. The dominants of shrublands, such as Rhododendron, Juniperus, Salix, and shrub Quercus, are well-represented in most of these pollen spectra. Pollen spectra from shrub meadows have less shrub pollen than those from shrublands, but more than those from meadows, steppes and deserts. The most frequent shrub pollen in this vegetation type is Rosaceae. Most of pollen spectra from shrub meadows are dominated by Cyperaceae pollen. Pollen spectra from meadows are characterized by the very high percentages of Cyperaceae pollen. The highest amounts of Cyperaceae pollen occur in pollen spectra from alpine-marshy meadows. Pollen spectra from Stipa steppes are characterized by the highest percentages of Poaceae pollen, and high Cyperaceae pollen percentages, whereas pollen spectra from Artemisia steppes have the highest percentages of Artemisia pollen. Pollen spectra from arid deserts are dominated by Chenopodiaceae. Main vegetation types can be distinguished by their modern pollen rain, i.e., modern pollen spectra do reflect the modern vegetation at local and regional scale in the Tibetan Plateau. This modern pollen database can thus be used to explore the pollen/vegetation and pollen/climate relationships by a variety of numerical methods.