Development Stage Dominant Height And Competitive Relationship Of Natural Spruce-fir-broadleaf Mixed Forest

The growth of natural uneven-aged and mixed forest has always been the focus of forest researchers,and it is affected by the development stage,tree size,competition of the surrounding trees and site conditions.However,there was not unified methods for the methodology about the division of development stage,top height estimation and competitive relationship for mixed forests which caused great uncertainties.In this study,permanent sample plots from natural spruce-fir-broadleaved mixed forests in Jilin province were used to classify development stages,estimate stand dominant height,evaluate inter-and intra-specific competition and to predict the growth of individual trees.(1)The development stage division of spruce-fir-broadleaved mixed forest.A method for the development stage division of mixed forest was proposed,which combined the two-way indicator species analysis(TWINSPAN)with the stand structure,tree species diversity and growth.One hundred and seventy-two permanent sample plots of spruce-fir-broadleaved mixed forest in Jilin Province were selected for initial classification based on the TWINSPAN method.The growth,species diversity and stand structure were compared and analyzed to determine the development stage.The result indicated that the spruce-fir-broadleaved mixed forests could be divided into three groups(group 1:spruce-birch mixed forest;group 2:fir-softwood-broadleaved mixed forest;group 3:Korean pine-spruce-hardwood-broadleaved mixed forest)by TWINSPAN.In general,the diameter of all stands showed an inversed J-shaped distribution.The height structure development of stand showed a trend related with tree species characteristics.There were differences in stand factors and periodic annual increment among these groups.The stand average diameter,mean height,dominant height,dominant diameter and volume all ranked by ascending order as group 2,group 1 and group 3;the periodic annual increment ranked by ascending order as group 3,group 2 and group 1.The species richness ranked by descending order as group 2,group 3 and group 1.The species dominance of group 2 and 3 were lower than that of the group 1.Hence,according to the above classification and characteristics analysis,the development stage of spruce-fir-broadleaved mixed forest was determined as fir-softwood-broadleaved mixed forest,spruce-birch mixed forest,Korean pine-spruce-hard-broadleaved mixed forest.(2)The dominant height estimation method for spruce-fir-broadleaved mixed forest.The dominant height estimation method of even-aged forest was extended to mixed forest,and then the reasonable threshold of sample plot area and dominant height estimation method of mixed forest were put forward.Based on the data from 12 one-hectare permanent sample plots(DBH?5cm)in 2013,the sample plots,with an area ranging from 0.01ha to 0.50ha with interval of0.01ha,were randomly selected and repeated 200 times with Monte Carlo method.In total,there were 120,000 subplots generated from these 12 plots.Considering the difference among tree species and the basis for selecting dominant trees(whether the maximum DBH is the criterion or not)and combing with four basic methods of determining dominant trees(traditional methods,CE;adjusted largest trees,ALT;U-estimator,UE;Sub-plot estimation,SUB),there were 14methods to estimate top height,in total.The paired t-test and correlation coefficient were used to analyze and compare the correlation among each top height estimation method and between top height and forest attributes.The uncertainty of the influence of sample size and forest density on the top height estimation method was measured.The results showed that the top height of natural spruce-fir-broadleaved mixed forest was dependent on the estimation method of dominant height,sample size and stand density,to a great extent.There are significant differences among these 14 top height estimation methods(0.02m?9.49m).The influence of sample area on CE method was the strongest and that on UE method was the weakest.Stand density had no or positive influence on the dominant height estimation.The top height estimation had lightly positive correlation with the factors describing stand productivity.Combining with the result of uncertainty analysis and production practice,we recommended the adjusted largest trees irrespective of tree species as the optimum method for estimating top height of natural spruce-fir-broadleaved mixed forest.(3)Modified distance-dependent competition index,i.e.the area potential available index(APA)considering tree species.Based on the data from 300 resample plots with their area equal to 0.50ha,the combination of Mingling and wood density,representing the difference among tree species and species characteristics,with diameter at breast height,height,crown width,representing the size of trees,was defined as the weight to produce weighted Voronoi diagram.The individual basal area growth mixed effect models including tree species dummy variable were used to compare and analyze the effect of tree characteristic on the generation of weighted Voronoi diagram.The individual tree growth model with the combination of crown width and Mingling as the weight for APA had significant dummy variable parameters and maximum adjusted determination coefficient(R²_adj)value.The modified APA index with consideration of tree species was better than that from traditional methods and was suitable for mixed forests.(4)The selection of optimal competition index and competitor determination method for spruce-fir-broadleaved mixed forest.Based on the continuous survey data from 48 resample plots,with their area equal to 0.0625ha,nine distance-dependent competition indices were calculated by 9 different competitor selection methods,respectively,and then 165 combination of competitor selection methods and competition indices were gained.The performance of individual basal area growth mixed effect models including competition and tree species dummy variables was compared to determine the optimal competitor selection method.The results showed that the competitor selection method was,to great extent,dependent on distance-dependent competition indices.The contribution of competition indices calculated by different competitor selection methods to individual growth models had slight differences.According to the importance and consistency of the competitor determination method to calculation of the same competition index and sensitivity of determining the response of tree growth of each tree species to competitive effect,the study derived that the optimal competition selection methods of different competition index were different.Among them,the crown overlapping method was suitable for these two competition indices proposed by Hegyi(1974)and Sun(1977);the fixed radius method was suitable for the competition index proposed by Martin&Ek(1984);the vertical search cone method was suitable for the competition index proposed by Alemdag(1978).In addition,the performances of 19 competition indices(9 distance-independent competition indices and 10 distance-dependent competition indices)in the individual tree growth models of different tree species in spruce-fir-broadleaved mixed forest were analyzed and compared.The result showed that the distance-dependent competition index,proposed by Bella(1971)had best performance in assessing the effect of competition on the individual basal area growth and quantifying the competition relationship among trees in natural spruce-fir-broadleaved mixed forests.(5)The intra-and inter-specific competitive relationship of spruce-fir-broadleaved mixed forest.The competition intensity among tree species was quantified using the above-mentioned competition index.The vertical structure of the stand was divided by the crown light competition factor method.The individual basal area mixed effect model including random effect at sample plot level and the intra-and inter-specific competition indices was built.Fixed effect parameter estimates of models was compared to evaluate the intra-specific and inter-specific competition intensity for tree species in different layers.The result showed that the competition source of main tree species with exception of spruce was intra-specific competition and the intensity of inter-and intra-specific competition of each tree species in different layer had slight difference.Among them,the influences of competition on the growth of Korean pine in the first and second layers were all from inter-specific competition.Meanwhile,the effects on the growth of Korean pine in the third and regeneration layers were all from intra-specific competition.The influences of competition on the growth of fir in different layers with exception of the first layer were all from intra-specific competition.The influences of competition on the growth of spruce in the first and third layers were all from inter-specific competition and meanwhile,that on the growth of spruce in the second and regeneration layers were all from intra-specific competition.Korean pine,spruce and fir are the main target species of spruce-fir-broadleaved mixed forest,their main competition sources were white birch,ribbed birch,poplar,elm and amur linden.In addition,comparing the influence of management measures on the competition of target tree species,the management measures could modify the competition intensity and source of target tree species.The study proposed the method to divide the development stage and top height estimated,modified the distance-dependent competition index and quantified the intra-and inter-specific competition for natural spruce-fir-broadleaved mixed forest.The results would provide technical support and theoretical basis for growth and yield prediction and management of the forests.