Fractal Studies On Species Diversity And Spatial Pattern Of The Forest Community In The Lianhua Mountains

Developing researches using new mathematical models is an obvious trend in the process of development,from qualitative description to quantitative description,from simple linear description to exact nonlinear deacription in ecology modern. The studies on vegetation patterns and diversity by using fractal theory are generated under this background. Vegetation pattern possess fractal characteristics on each hierarchical levels,ie,from module,individual,population,community to landscape,they are all fitted to be described by the ideas and methods of fractal theory. The content of this dissertation was on the study of diversity and spatial pattern of the several typical forest community at the Lianhua Mountain National Nature Reserve in Gansu Province by using fractal theory,the results were: 1. The α-diversity on different spatial scales and the fractal power-law relationship between α-diversity and sample size in Picea-Abies forest and coniferous broad-leaved mixed forest at the Lianhua Mountain National Nature Reserve in Gansu Province were studied by using the data from two transects of 420m length and 10m width. The results showed that species richness indexes and Shannon indexes of Picea-Abies forest and coniferous broad-leaved mixed forest abruptly increased at the beginning and then kept steady while scale was enlarged,indicating the strong influence of scale on α-diversity in some scale ranges on the transects. The sampling scales of αdiversity on the transects were 60m for Picea-Abies forest and 90m for coniferous broad-leaved mixed forest,based on the scale changing curve of species richness indexes and Shannon indexes,indicating that the sampling scale of α-diversity on coniferous broad-leaved mixed forest transect was obviously greater than that on Picea-Abies forest transect. 2) The fractal power-low relationship of α-diversity index versus sample scales on coniferous broad-leaved mixed forest transect was more various than that on Picea-Abies forest transect. It showed that coniferous broad-leaved mixed forest community had greater spatial variation on community structure than Picea-Abies forest community. 2.On different selected sampling scales,the spatial variation characteristics of species diversity along elevation gradient on transect was analyzed using ?-diversity indexes and fractal geometry in cold temperate forest in Lianhua Mountain,The results showed that Cody index disclosed that canopy layer and shrub layer exhibited obvious species substitution properties along elevation gradient as sampling scales were 40m and 60 m respectively. The species turn-over of canopy layer and shrub layer had reversed properties at a same site. The spatial variation of community diversity was higher at the beginning at the end of the transect. The authors proposed a new index based on Cody index,the newindex,CSV(Community Structure Variation) index is between0~ 1,as the improvementofCody index,its description to the variation of community diversity along environmental gradient was closer to the actual circumstances. The result of CSV index analysis showed,that the obvious variation of community structure appeared at>60 m sampling scale incanopy layer,whereas the sampling scale was>40m in shrub layer. The sampling scalesfor detecting the variation properties of species richness and community structure were different on the transect. The results of fractal analysis showed that canopy layer had apparent scale dependence within60m scale,shrub layer exhibited scale dependence when observing scale<50 m. The scale dependent properties showed strong variation at small scales and low heterogeneity at large scale. 3. This study analyzed the scale variation on degrees of the spatial patterns of several typical community in Lianhua Mountain Gansu using box-counting dimension. The results showed that the spatial distribition of variety community were high(>1.5,near 2),which reflected the high spatial occupation degrees of the community. The order of spacial occupation degree was Abies faxoniana plantation(Q1)1.8505>Quercus liaotungensis.community(Q6)1.8156> Abies faxoniana community(Q3)1.7970>picea asperate-picea.purpurea community(Q2)1.7918> picea.purpurea -picea aspirate-Abies faxoniana community(Q7)1.7563>Abies faxoniana -Betula utilis -Betula platyphylla.community(Q4)1.7122> Pinus armandi community(Q5)1.5589, which reflected the variation of function and positions of main forest community at Lianhua mountain. The box-counting dimension of community pattern disclose the spatial occupation degree of community and its scale variation degree,the inflection pointing indicated the agreegating scale of individuals. 4. The spatial pattern of several typical community in Lianhua Mountain Gansu was studied by using information dimension. The results showed the order of the scale variation degree of pattern intensities was Abies faxoniana -Betula utilis -Betula platyphylla.community(Q4)1.8376> picea asperate-picea.purpurea community(Q2)1.5254> Abies faxoniana community(Q3)1.1394> picea.purpurea -picea aspirate-Abies faxoniana community(Q7)1.0959> Pinus armandi. community(Q5)0.9947> Abies faxoniana.plantation(Q1)0.8204> Quercus liaotungensis.community(Q6)0.4521.The variation of information dimension reflects the differences of community regeneration conditions. The information dimension of community patterns quantifies the scale variation degree of pattern intensity,and discloses the unevenness of individual distribution. 5. The spatial patterns of several typical community in Lianhua Mountain Gansu were studied. The results showed that correlation dimensions of variety community were high (near 2 ),spatial autocorrelation among individuals was high,and competition among individuals was serious. , The order of spatial autocorrelation degree was picea asperate-picea.purpurea community ( Q2 ) 1.7944 > Abies faxoniana -Betula utilis -Betula platyphylla.community(Q4)1.7798> Quercus liaotungensis.community(Q6)1.7442> Abies faxoniana community(Q3)1.6992> Abies faxoniana plantation(Q1)1.6656> Pinus armandi community(Q5)1.6404> picea.purpurea -picea aspirate-Abies faxoniana community(Q7)1.6342. It showedthe differences of scaling property of spatial autocorrelation among individuals of different communities. Correlation dimension of community pattern can disclose the scaling properties of spatial autocorrelation of individuals and the spatial occupation degree of community. The above studies on diversity and spatial pattern of the several typical forest community at the Lianhua Mountain National Nature Reserve in Gansu Province by using a same kind of ideas and methods of fractal theory,could avoid the shortcomings of former pattern analysis in methods,supplies new information of community diversty and vegetation patterns for us; whereas it organically combines the characteristics of community diversty and spatial pattern on multiscale and supplies a way to grasp the diversity and spatial pattern of the several typical forest community at the Lianhua Mountain National Nature Reserve in Gansu Province on different levels dimensionally and systematically for us. Generally this study is the extension and deepening of the theory of community diversty and pattern analysis,possessing important academic value on the one hand,and could supply theoretical instructions to the sustainable development of forestry on the other hand.