Spatial Characteristics Of Three Kinds Of Typical Natural Forests In Xiaolong Mountains,Gansu Province

Through typical sample investigation, univariate and bivariate distribution, FSSI and FSSD were used to analyse the spatial characteristics of different communities,dominant populations, DBH classes trees in Pinus tabulaeformis forest(communityⅠ), P. tabulaeformis-Quercus aliena var. acuteserrata mixed forest(community Ⅱ),Q. aliena var. acuteserrata forest(community Ⅲ) in Xiaolong ?ountains, Gansu Province, and to discuss the species and size selective preference, structural characteristics of cutting individuals and thinning priority of different main species in community Ⅱ with structure-based management so that we can provide reference for the analysis of microstructure and accurate structure adjustment on the natural communities in Xiaolong ?ountains. The results showed that:(1) The ? of community Ⅰ, community Ⅱ, community Ⅲ respectively were0.511, 0.508, 0.508, randomly distributed, and their ? were 0.507, 0.503, 0.485, in a moderate growth state. The community Ⅰ were poorly mixed, while the other two were highly mixed; ?ost trees in different communities with same dominance or mingling degree were randomly distributed, and the same distribution pattern or mingling degree trees were approximately equal in different dominance levels, and the main difference among them were that the same dominance degree or distribution pattern trees were mixed by the same or another kind of species in community Ⅰ,while they were surrounded by 3-4 kinds of other species in communityⅡ, Ⅲ. The FSSI of different communities were 0.576, 0.698, 0.739, and the FSSD were 0.788,0.588, 0.526, so the spatial structure of community Ⅲ was better than the other two’s.FSSI and FSSD had strong consistency in terms of spatial structure expression. The results of 3D scatter diagram, being seen as the bivariate distribution of sample level,was consistent with individual bivariate distribution.(2) The distribution pattern of P. tabulaeformis population in community Ⅰ was uniform(? =0.255), with a stronger competition potential but badly mixed; The same dominance or mingling degree trees were mostly randomly distributed, and mostindividuals with the same distribution pattern or mixed class were almost equal in every dominance degree, but the same dominant degree or pattern trees were badly mixed. P. tabulaeformis, Q. aliena var. acuteserrata populations in community Ⅱboth distributed randomly, close to the middle state and well mixed; The same dominant or mingling degree trees were mostly in random distribution, and the same pattern individuals were almost equally distributed in every dominance degree, but surrounded by 2-3 species; The main differences were: The same dominant trees in P.tabulaeformis population were mainly mixed with 2-3 others species, while they, in Q.aliena var. acuteserrata population, were surrounded by 3-4 species. The individuals of Q. aliena var. acuteserrata population in community Ⅲ randomly distributed,were close to the middle state and well mixed; ?ost of the same dominance or mingling trees were in random distribution, and most trees with same pattern were in dominant and subdominant, but 19.8% trees were still randomly distributed and in moderate state, vast trees in same dominance or pattern were surrounded by 2-3 others species, 66.7% trees were in dominant and mixed well; The bivariate distribution characteristics of different communities and their dominant populations were basically identical, in other words, the dominant populations obviously controlled their community structure, also reflecting the contribution value of different dominant populations in community was inequitable.(3) Small trees in community Ⅰ and intermediate individuals in community Ⅱwere in clumped distribution, while the rest formed unique nested random distribution pattern with the overall communities. ?ost small trees in these communities were suppressed, the intermediate trees were in moderate state or subdominant position and large ones were in dominant position. ?ith the increase of DBH, trees in communityⅠwere increasingly well-mixed, but the large in communityⅡ, Ⅲ were well-mixed,but the intermediate mixed worst. ?ost trees with same dominance or mingling degree of different diameter classes in each communities were randomly distributed,and most large trees with same mingling degree or distribution pattern were dominant or subdominant, while the small were weak or absolutely inferiority, the intermediate in communityⅠ were weak or in medium state, and in community Ⅱ,Ⅲ were in dominant or subdominant. Different diameter classes trees in communityⅡ,Ⅲ with same distribution pattern or differentiated degree generally mixed with 3-4 other species, while in community Ⅰ the small were completely surrounded by others species, while the rest were mixed with same or another species.(4) In community Ⅱ with structure-based forest management, Carpinus turczaninowii and Betula platyphylla had the largest species preference value, while Pinus armandii had the largest size preference value. The Spp of main constructive and dominant species, Q. aliena var. acuteserrata and P. tabulaeformis, were 0.25,0.24, while the NG were 1.34, 1.40. The main cutting wood were trees with 6-14 cm DBH. 50.4% of cutting individuals were randomly distributed, while 83.0% mixed well and mostly suppressed. 41.9% of cutting individuals were randomly distributed and weakly grew; 65.0% were in random distribution or uniform distribution but well mixed; 60.7% didn’t take obvious growth advantages but mixed well. Six different populations greatly influenced by cutting interference had specially prominent thinning priority index values, but each population had specific thinning priority value.