Studies of vegetation–environment relationships and vegetation dynamics in Chinese subtropical forests
Appears in the following Collection
- Naturhistorisk museum 
AbstractThe principal aim of this study was to investigate vegetation-environment relationships and examine patterns of changes in forest understorey vegetation in five Chinese subtropical forests. Furthermore, an Ecological Field Theory (EFT) model for single-tree influence on understorey vegetation was applied to assess the relative contributions of tree influence to variation in understorey species composition. In focus are five study areas, located in the southern and southwestern parts of China and consisting of subtropical mixed coniferous and broadleaved forests. In each study area, fifty sample plots 1 m2in size were randomly placed within each of ten 10×10 m macro-plots. Plant species composition was recorded in all 250 1-m2 plots, using frequency in sub-plots as a measure of species abundance. A total of 33 environmental variables were recorded for the 1-m2 plots as well as the 10×10 m macro-plots. All trees in all macro-plots were mapped and measured. The three ordination methods – DCA, LNMDS and GNMDS – were used in parallel to find corresponding (consensus) ordination axes, which were likely to represent true gradients in species composition. Three dimensions (axes, gradients) were needed to describe the variation in vegetation in two areas, two dimensions in the other three areas. GNMDS was finally chosen for interpretation and presentation of vegetation–environment relationships.Environmental interpretation of ordinations was made by split-plot GLM and non-parametric correlation analysis. Four major underlying complex environmental gradients were correlated with the species composition gradients: (1) A litter-related compositional gradient, reflected in favourability for bryophytes, was found in four areas (TSP, LCG, CJT and LXH). (2) A topography-related compositional gradient, reflected in variation both in vascular plant and bryophyte species composition, was found in four areas (TSP, LGS, CJT and LXH). Relationship with inclination was found in three areas (TSP, CJT and LXH) and with aspect favourability and heat index in two areas (LGS and CJT). (3) A soil acidity/soil mineral nutrients-related compositional gradient, reflected in variation in vascular plant species composition, was found in three areas (LCG, LGS and CJT). This was related to soil mineral nutrients concentrations in one of these areas (LGS). Finally, (4) a tree density-related compositional gradient with variation mainly at the macro-plot scale was observed as the first axis (GNMDS 1) in two areas (LCG and LXH). These four gradients will be referred to as the litter-layer depth, topography (inclination and aspect favourability), soil acidity/soil mineral nutrient concentrations, and tree density ecoclines (i.e. gradients in environmental conditions and species composition). Four out of twelve consensus ordination axes could not be interpreted ecologically by the environmental variables available. Changes in understorey vegetation (single-species abundances, species number and species composition) in the 1-m2 sample plots were studied in four areas during a first two-year, a consecutive three-year and a full five-year period. The results showed that: (1) a larger number of vascular plant species than expected by chance decreased and increased significantly in abundance in two and two areas, respectively; (2) the number of vascular plant species per plot increased significantly in two areas; (3) a larger number of bryophyte species than expected by chance decreased and increased significantly in abundance in three and two areas, respectively; (4) the number of bryophyte species number per plot decreased significantly in two areas but increased significantly in one other area. Finally, (5) significant plot displacement along gradients in species composition (interpreted GNMDS ordination axes) was observed in two areas for the main gradient (GNMDS 1) and one area for the second gradient (GNMDS 2). The patterns of change observed for bryophytes are attributed to climatic fluctuations. The increase in the abundances and numbers of vascular plant species is most likely due to seasonal variation and more favourable climatic growth conditions during specific years. No clear indications were found of changes for vascular plant species that may be linked to soil acidification or direct effects of air pollutants. An EFT model for single-tree influence on ground species composition was developed for each of the five study areas. Optimal model parameters were found by maximizing the eigenvalue of one constrained ordination axis (RDA), obtained by use of the tree influence index as the only constraining variable. Results showed that: (1) the eigenvalue of the first RDA axes varied between the five study areas, generally accounting for only a small part of the variation in species composition; and (2) the relative amount of compositional turnover attributable to tree influence differed between study areas and between species groups, but was generally low. We concluded that in Chinese subtropical forests, trees influence the understorey more in a collective manner than through the effect of single trees.
List of papers. The thesis is based on the following papers (I, II and III), which will be referred to in the text by their Roman numerals:
I.H.Y. Liu, T. Økland, R.H. Økland, J.X. Gao, Q.R. Liu, O. Eilertsen and H. Bratli. Gradient analyses of forests ground vegetation and its relationships to environmental variables in five subtropical forest areas, S and SW China. Sommerfeltia, submitted.
II.H.Y. Liu, T. Økland, R.H. Økland, J.X. Gao, QR. Liu, O. Eilertsen and H. Bratli. Changes in forest understorey vegetation in four subtropical forest areas, S and SW China. Plant Ecology, submitted.
III.H.Y. Liu, R.H. Økland and T. Økland. Single-tree influence on understorey vegetation in five Chinese subtropical forests. Journal of Integrative Plant Biology, submitted.