Modern Pollen-Vegetation/Climate Relationships In Southwestern Yunnan

Located in the southwestern part of the Yunnan-Guizhou Plateau,southwestern Yunnan is affected by warm and humid air currents from the Indian Ocean brought by the southwest monsoon,and thus is one region suitable for the study of global monsoon change.With meeting mountains and rivers,vertical and horizontal ravines,well preserved original vegetation types,and abundant forest resources,this low latitude region is of great significance in the study of natural environment evolution and vegetation change.Correctly understanding the relationships between modern pollen-vegetation/climate provides a basis for better reconstructing past vegetation and climate using fossil pollen data.Therefore,78 surface samples were collected from lakes and reservoirs for pollen/charcoal analysis in Baoshan,Dehong,Lincang,and Puer of southwestern Yunnan.Pollen/charcoal,vegetation,and climate data were then analyzed using numerical methods to study modern pollen-vegetation /climate relationships,and modern spatial patterns of charcoals in southwestern Yunnan and their relationships with forest fires and human activities.Seventy-eight sampling lakes and reservoirs are located in three vegetation zones,i.e.the seasonal rainforest and semi-evergreen seasonal rainforest zone in the semi-humid evergreen broadleaved forest zone in the Hengduan Mountains of the western Yunnan(zone A),the monsoonal evergreen broadleaved forest zone in the middle mountains and river valleys of Southwestern Yunnan(zone B)and the mountains and inter-mountain basins of southern and southwestern Yunnan(zone C).Pollen spectra from surface sediments of lakes and reservoirs are dominated by arboreal pollen with mean pollen percentages of 66.6 %(31.0 ~85.8 %)and then herbaceous pollen.In the arboreal pollen,Alnus,Pinus,and Lithocarpus/Castanopsis have high percentages,followed by Betula,Cyclobalanopsis and Moraceae;Juglans,Corylus/Carpinus,Castanea and evergreen Quercus pollen is common.These arboreal pollen taxa belong to four ecological groups,i.e.coniferous forest,temperate deciduous broadleaved forest,subtropical evergreen broadleaved forest,and tropical broadleaved forest.In the herbaceous pollen,dominant upland herbs include Poaceae,Asteraceae,Artemisia,Cyperaceae,Polygonaceae,Cruciferae,and Rubiaceae.Abundant spores were found in these pollen spectra,probably due to hot and humid climatic conditions in the study region.A comparison of modern pollen and vegetation data indicates that the spatial patterns of dominant pollen taxa in pollen spectra from lake surface samples are consistent with the spatial distribution of main modern vegetation types,and thus pollen spectra generally reflect modern vegetation in southwestern Yunnan.Canonical correspondence analysis(CCA)on pollen and climate data was employed to study pollen-climate relationships quantitatively.The climate data at sampling sites were obtained by linear interpolation of longitude,latitude and altitude using monthly temperature and precipitation data from 27 meteorological stations in southwestern Yunnan.Pollen data are pollen percentages of selected 18 pollen taxa with strong representativeness,large coverage,average pollen percentage of more than 0.4 %,and the presence at least in more than half of pollen spectra.The results of CCA show that mean annual precipitation(MAP),mean annual temperature(MAT),and spring precipitation(P-spring)are three climatic parameters controlling the spatial distribution of modern pollen rains in southwestern Yunnan.The correlation analysis between these climatic parameters and pollen percentages of 18 pollen taxa exhibits their linear or non-linear correlations at different degrees.These numerical analyses provide a reference for the selection of mathematical models in future quantitative reconstructions of paleoclimate using fossil pollen data from lakes and reservoir in southwestern Yunnan.The spatial patterns of charcoal percentages and concentrations at different particle sizes(small <50,medium 50-100,and large >100 ?m)was analyzed to reveal the relationships between charcoals and forest fires,vegetation types,as well as human activities in southwestern Yunnan.The similarity in spatial patterns of charcoal concentrations from lake surface samples and 2001~2013 fire activities derived from satellite data indicates that charcoal data reflects regional forest fires.Positive correlations were found between medium to large charcoal percentages and pollen percentages of coniferous taxa and xerophytic herbaceous taxa,suggesting that the frequent of forest fires might be associated with vegetation types.Additionally,our results show that forest fires are also related to human activities.