Pollen Assemblages In Surface Sediments And Their Relationships With Vegetation And Climate In Southwestern Yunnan

The relationships between modern pollen rain and vegetation/climate are the key to accurately interpret palaeovegetation and palaeoclimate from fossil pollen data.A diversity of vegetation types and a number of lakes?reservoirs?in southwestern Yunnan provide us an excellent opportunity to study these relationships.In this study,we collected lake surface sediment samples from 178 lakes?reservoirs?using gravity sampler in southwestern Yunnan.Pollen analysis of these samples together with available vegetation data and observational climate data from 1961 to 2011qualitatively and quantitatively reveals the spatial distribution characteristics of modern pollen rains,and their relationships with vegetation and climate in southwestern Yunnan.The study region is mainly divided into three vegetation areas,i.e.evergreen broadleaved forest area,monsoon evergreen broadleaved forest area,and seasonal rainforest and semi-evergreen rainforest area.Pollen analysis of 178 surface samples shows that the spatial distribution of modern pollen rains is related to main vegetation types in different vegetation areas.Pinus,Alnus,and Artemisia pollen is mainly distributed in evergreen broadleaved forest and monsoon evergreen broadleaved forest areas;evergreen Quercus,Lithocarpus/Castanopsis,Cyclobalanopsis,Betula,Rosaceae pollen is generally found in monsoon evergreen broadleaved forest,and seasonal rainforest and semi-evergreen rainforest areas;Moraceae,Euphorbiaceae,Asteraceae and Poaceae pollen often occurs in seasonal rainforest and semi-evergreen rainforest area.The spatial distribution of those major pollen types in modern pollen rains is generally in agreement with that of dominant and constructive plant species in different vegetation areas.The result of principal component analysis?PCA?on pollen percentages of arboreal pollen types shows that the first and second principal component axes with eigenvalues of 0.417 and 0.163 explain 41.7%and 16.3%total variation of pollen data,respectively.Wet-loving Moraceae,Lithocarpus/Castanopsis,Castanea,and Rosaceae with positive loadings and dry-loving Pinus and Alnus with negative loadings indicate that the first principal component is a proxy of humidity;on the second principal component axis,cold-loving Alnus,Myrica,Juglans and Tsuga are located in its negative part,warm-loving evergreen Quercus,Castanea,Rhamnaceae,and Euphorbiaceae are located at its positive part,reflecting the second principal component is a temperature proxy.The result of Redundancy analysis?RDA?on pollen and climate data shows that,mean annual precipitation?MAP?is positively correlated with Lithocarpus/Castanopsis,Betula,Moraceae and Theaceae,and negatively related to Pinus,Juglans,and Alnus;mean temperature in July(T_july)is positively correlated with evergreen Quercus,deciduous Quercus,Euphorbiaceae and Rhamnaceae,and negatively correlated with Oleaceae,Myrica,and Myrtaceae;and mean annual temperature?MAT?and mean temperature in January(T_january)are positively correlated with Corylus/Carpinus,Salix,and Platycarya,and negatively correlated with Tsuga,Ulmus,Myrtaceae.The RDA result indicates that MAP,MAT,T_july and T_january are four climatic parameters controlling the spatial distribution of modern pollen rain in southwestern Yunnan.Therefore,modern pollen rain from lake surface sediments provides a basis for quantitative reconstruction of palaeoclimate using modern and fossil pollen data from lakes in southwestern Yunnan.