Distribution and vegetation representation of pollen assemblages from surface sediments of Nam Co, a large alpine lake in the central Tibetan Plateau

Lacustrine fossil pollen records have been widely used to reconstruct palaeovegetation and palaeoclimate changes on the Tibetan Plateau (TP). However, little is known about the vegetation representation of modern lacustrine pollen assemblages there. This paper presents the results of modern pollen investigation on 63 surface sediments from the lake basin and 37 topsoil samples from the drainage area of a large lake, Nam Co, located in the central TP. It aims to assess quantitatively the influences on lacustrine pollen assemblages of the pollen sources and sedimentary processes, and to establish vegetation representations for modern lacustrine pollen assemblages. Modern pollen assemblages from topsoils of different vegetation had diagnostic features in terms of their composition and pollen percentage. The spatial variabilities and results of principal component analysis suggested that lacustrine pollen assemblages were influenced by both the regional/local source vegetation and sedimentary processes. The lacustrine pollen assemblages were mainly homogeneous due to in-lake sedimentary processes (mixing and redistribution). An accumulation zone for lacustrine pollen assemblages was found in the deep lake basin (depth > 60 m) due to sediment focusing. The results of boosted regression tree analysis further confirmed that source vegetation was the predominant factor (85.8%) responsible for the vegetation representation of lacustrine pollen assemblages, while sedimentary processes accounted for only 14.2%. The results of discriminant analysis indicated that most lacustrine pollen assemblages (90.5%) were representative for the regional vegetation of alpine steppe in the Nam Co catchment and central TP, while only 9.5% were representative for the local meadow vegetation. Therefore, it is recommended that lacustrine pollen assemblages from deep lake basin of accumulation zone in large lakes of the TP can be used to retrieve efficiently the signals from regional vegetation and climate changes.