Modern pollen assemblages and their relationships to vegetation and climate in the Lhasa Valley, Tibetan Plateau, China. (22nd February 2018)

• Record Type: Journal Article
• Authors: Zhang, Yan-Jie; Duo, La; Pang, You-Zhi; Felde, Vivian A.; Birks, Hilary H.; Birks, H. John B.
• Is Part Of: Quaternary international. Volume 467(2018)Part B
• Page Start: 210
• Page End: 221
• Publication Date: 2018-02-22
• Subjects: Climatic variables -- Co-correspondence analysis -- Multivariate analysis -- Pollen equivalents -- Pollen ratios -- Procrustes analysis; Geology, Stratigraphic -- Quaternary -- Periodicals; Stratigraphie -- Quaternaire -- Périodiques; 551.79
• DOI: 10.1016/j.quaint.2018.01.040
• Languages: English
• ISSNs: 1040-6182

Abstract

Abstract: Forty-seven surface pollen samples and 141 vegetation quadrats were investigated in the Lhasa Valley in order to assess the relationships between local modern pollen assemblages and vegetation and climate. The regional climatic data for each sample are mean annual precipitation (Pann ), mean annual temperature (Tann ), summer temperature (Tsummer ), actual evapotranspiration (AET), potential evapotranspiration (PET), and moisture index (MI), estimated by co-kriging using meteorological data from 16 climate stations in or near the Lhasa Valley. Cluster analysis, principal components analysis (PCA), Procrustes analysis, and co-correspondence analysis (Co-CA) were used to evaluate the relationship between modern pollen assemblages and contemporary vegetation. Linear regression was used to examine the relationships between pollen ratios ( Artemisia /Chenopodiaceae (A/C), Artemisia /Cyperaceae (A/Cy), Artemisia + Chenopodiaceae/Cyperaceae (AC/Cy)), aridity pollen index, total arboreal pollen (AP), and the climatic variables. Cluster analysis and PCA results are generally consistent, and differentiate between pollen assemblages from arid conditions and those from more humid conditions. However, the PCA results distinguish pollen assemblages from shrubland slightly more clearly than the cluster analysis does. The PCA results show a general agreement between the modern pollen assemblages and contemporary vegetation types, but pollen assemblages from coniferous ( Juniperus ) woodland cannot be distinguished from shrub grassland, meadow, or shrub meadow pollen assemblages in the PCA results. Both Procrustes analysis and associated permutation tests and Co-CA show a strong statistically significant relationship between modern pollen and vegetation composition. Analysis of the relationship between the climatic variables and vegetation types suggests that humidity (Pann and MI) is the main variable related to the vegetation types within the restricted areas of the Lhasa Valley. Neither the A/C ratio nor the aridity pollen index is a reliable aridity indicator in the Lhasa Valley. The AP sum may be a weak precipitation indicator, whereas the A/Cy and AC/Cy ratios appear to be robust indicators of precipitation, aridity, and temperature at the scale of the Lhasa Valley.