Spatial Heterogeneity Of Shrub On Small Scale In Natural Secondary Shrub In Linan

By using geostatistical methods and GIS, this paper studied the spatial heterogeneity and distribution patterns of shrub species diversity, herb species diversity and vegetation coverage in natural secondary shrub in Taiyang town, Linan City,the effects of terrain factors(slope, aspect and convexity)and fire severity on the vegetation parameters were quantified based on RDA ordination and p RDA analysis.Aimed at the thorough analysis the relationship between influence factors and vegetation distribution. It is a great significance that the study to learn about the ecology of natural secondary shrub landscape, providing basis and reference for the ecological restoration of secondary shrub and the operation management. The main results were as follows:(1)There were 83 species of higher plants in the plots, which belong to 67 genera in 38 families mainly comprised by subtropical families and genera, and there are 49 kinds of tree species in shrub layer, four tree species that important value more than ten are 8.2% in total, 11 tree species that important value between 1~5 are 22.4% in total; There are 34 kinds of tree species in herb layer, important value between 5~10 are two between 1~5 are ten, tree species that important value between 1~5 characterizing deciduous secondary shrub that degraded from evergreen broadleaved forest.(2)Biomass of natural secondary shrub is 0.2~32.86 t·hm-2, spatial variation is larger. Brush biomass after logarithmic transformation to be normal distribution, gaussian model is the optimal model of semivariogram, which has moderate spatial correlation in the range of 81.05 m and showed banded spatial distribution with evident Pattern.(3)In the study area, shrub species diversity(H1), herb species diversity(H2) and vegetation coverage(C) were in normal distribution, and the coefficient of variation(CV) ranged from 25.83%~43.14%; H1 and C were accorded with exponential models, H2 was accorded with spherical models; The spatial dependence of the vegetation parameters was at a scale of 6.6~193.6 m, H1 and H2 had moderate spatial autocorrelation, the spatial autocorrelation of C was strong. Fractal dimensions were higher for C, so the spatial structure of C had a strong scale dependence with a complex spatial pattern. H1 and H2 showed banded spatial distribution, while C presented scattered type.(4)Among the environment factors, slope had the strongest effects on H1, and the aspect could explain 3.8% of the spatial variability of H1, respectively. Fire severity had the most powerful effects on W、H2 and C, the elevation could explain 29.7、16.8% and 17.2% of the spatial variability of H2 and C. The spatial variability of the vegetation parameters was influenced by environment factors, existing differences in the effects of different environment factors on the spatial variability of the vegetation parameters. Structual factors which included slope、aspect、convexity and fire severity had produced important influences on the spatial heterogeneity of vegetation. An important influence on the spatial heterogeneity of vegetation.