Introduction

A population,the most basic unit of community structure and function,is one of the most important levels used for analysis in ecology.Population structure and spatial distribution patterns,the two main topics of concern among ecologists,have become the focus of ecological research in recent years(Somanathan and Borges 2000;Didier and Porter 2003;Miyadokoro et al.2003).Plant population structure is created by the integrated effects of individual plant viability and the external environment.The study of plant population structure provides more than an understanding of the development and evolutionary trends of a population;it can also allow the analysis of the relationship between the population and the external environment while determining the status of a population in a community(Stewart 1989).

The spatial distribution pattern of a plant population refers to spatial distribution for the individual,which is caused by the mutually integrated effects of its own characteristics,interspeci fi c relationships and environmental conditions(Greig-Smith 1983;Dale 1999).To a certain extent,distribution re fl ects the relationship among populations and habitats as well as the status and role of an individual within the community(Arista 1995).Knowledge of distribution provides important and signi fi cant information related to ecological processes of plant populations as well as intra-and interspeci fi c relationships(Getzin et al.2008;Wiegand et al.2007).The study of the spatial patterns of plant populations not only includes a quantitative description of population structure,but also allows the prediction of the dynamic development trends of both populations and communities.In addition,a knowledge of plant distribution will allow researchers to analyze the type of spatial distribution pattern of a population and reasons for its formation;it will also allow the development of appropriate protective measures designed to promote the ecological restoration of a population,such as provide guidance for supplemental seedling or selective cutting(Borchsenius et al.2004).Therefore,the study of plant population structure and spatial distribution patterns has very important scienti fi c signi fi cance in illuminating ecological and biological characteristics of plant populations;this knowledge also helps in the understanding of the laws controlling variations in population dynamics and the relationship of plant distribution to the environment,as well as to the factors controlling population formation,stability,and ecological succession(Johnson 1997;Harper 1977).

取空白辅料适量，置于100 mL量瓶中，采用上述样品专属性试验方法，对空白辅料用强光照射、高温、酸水解、碱水解和氧化的方法进行加速破坏，处理后的样品取续滤液进行高效液相色谱（HPLC）分析，以研究辅料是否干扰样品有关物质的检查。

三种不同类型的关联企业有着不同的特点，在对其进行司法处置时也应当有着不同的标准，每种类型由于其与涉黑组织的关系有着不同程度的关联与交叉，在规范关联性企业的处置模式之前，需要对其概念进行统一的界定，本文称之为“涉黑关联企业”，但最核心的对黑社会组织起重要支撑作用的是“关联性企业”，也可以作为扫黑除恶打击的重点。

Quercus aquifolioides is endemic and widely distributed in the southeastern part of the Tibetan and Yunnan-Guizhou Plateaus.This evergreen tree reaches heights of 20 m and mainly grows on sunny hillsides at elevations of 2000–4500 m within mountain pine(Pinus densata)forest.Q.aquifolioides,a dominant and widely distributed species in this region,plays important ecological roles by enriching biodiversity and helping to conserve soil and water.Therefore,previous research related to the populations of Q.aquifolioides,have mainly focused on topics such as forest soil(Yang and Zhao 2013),physiological and reproductive characteristics of the species(Wang et al.2009;Li et al.2009),and the morphological structure of this species and its relationship with the environment(Li et al.2006;Zhou et al.2015).Some scholars have also conducted a related study on Q.aquifolioides populations on the Tibetan Plateau(Song et al.2004;Duan et al.2008;Zhu et al.2014;Chen et al.2015).However,no known comprehensive reports have analyzed the variations in the population structure and spatial distribution patterns of Q.aquifolioides on Sejila Mountain.Knowing the spatial distribution of Q.aquifolioides and its spatial correlation at different stages of growth would establish a scienti fi c basis for the sustainable use and recovery of local forest resources.The goals of our study were(1)to analyze the structural dynamics of Q.aquifolioides populations;(2)to study the spatial pattern dynamics of the populations at different ecological stages and to reveal their related spatial association;and(3)to discuss the in fl uence of the biological characteristics of the species and related environmental factors on the population.

Materials and methods

Study site

产生这种问题的原因一是在于人民警察中个别人员对于继续盘问制度本身的错误认识和理解，不能够正确认识和理解继续盘问制度与传唤、先行拘留、拘传等强制措施的区别及适用条件。二是在于人民警察中个别人员对继续盘问工作时限的适用具有更大的自由度。根据《公安机关适用继续盘问规定》第十一条规定了继续盘问的时限一般为十二小时，某些情形可以延长至二十四小时，特殊情形下可以延长至四十八小时。因此，适用继续盘问可以在时间限制上有更大的自由度。

where A is plot size;n is the number of plants in the plot,uij is the distance between the ith and jth subject trees,and r is the spatial scale.When uij≤r,Ir=1,or Ir=0,i is the center of the circle,uijis the radius of the circle,the value of the area A is the ratio to the entire circle,and could be a correction factor of boundary effects.In practical application,we simpli fi ed Formula(1)as Formula(2),using（ r）to replace（ r）:

Fig.1 Location of the study site

Plot establishment and surveys

where a value of（ r）=0 indicates a random distribution,a value of（ r）＞ 0 indicates an aggregated distribution and a value of（ r）＜0 indicates a uniform distribution.

Data analysis

Classi fi cation of the population by DBH

Multivariate point pattern analysis provides a descriptive analysis of different relationships among different populations(Formula 3):

Table 1 Number of trees,tree height,diameter at breast height(DBH)of saplings,mediumand large-sized trees at the sampling sites

Type Number of trees Average tree height Average DBH Maximum DBH(plants/ha) (m) (cm) (cm)Saplings 121 5.6 6.2 9.9 Medium trees 226 11.9 18.2 29.9 Large trees 21 16.0 34.0 43.0

Preparation of a static life table

We developed a static life table for Q.aquifolioides populations on Sejila Mountain.This was done using a combination of comprehensive survey plots data and the number of Q.aquifolioides individuals in different age classes,using the static life table method(Harcombe 1987;Cui et al.2011).

The survival curve

The population survival rate could be shown using a survival curve showing variations in age to re fl ect the population dynamics(Armesto et al.1992).Deevey(1947)divided survival into three types:type I survival curves are convex,indicating that only a few individuals died before reaching the natural age of physiological longevity,showing a low level of early mortality;type II survival curves are diagonal,with the same mortality rate for each age group;type III survival curves are concave,representing higher mortality of young individuals while mortality decreased gradually over time.A survival curve is a valid interpretation of a static life table,which is one of the main methods of studying population dynamics,and the analysis of these curves could provide a deeper understanding of population dynamics(Stewart 1989;Díaz et al.2000).The survival curve of Q.aquifolioides populations on Sejila Mountain was based on diameter class on the abscissa and the natural logarithm of survival amount on the ordinate axis.

Spatial pattern analysis

We used the Ripley’s L Function to analyze the plant populations(Ripley 1977).The value^K（ r）re fl ects the distribution pattern of plant populations in space(Formula 1):

The study area,located on Sejila Mountain in eastern Bayi District,Tibet Autonomous Region(Tibet),China,is part of the Nyainqentanglha Mountains.This watershed lies between the Nyang River Basin and Parlung Zangbo River.This region features steep mountains with a mean slope of 28°and a mean elevation of 3540 m.Vegetation changes from low to high elevation transitioning mainly from deciduous broad-leaved forest to mountainous temperate coniferous forest,subalpine dark coniferous boreal forest,boreal alpine shrub,and ultimately to grass meadow.The study site(29°38′29′′N,94°42′02′′E;Fig.1)features a humid and semi-humid subalpine cool climate with distinct wet and dry seasons.Rain falls frequently in spring and summer with a rainy season generally from April to October.Autumn and winter are often windy and dry.The mean annual precipitation,relative humidity,and temperature are about 800 mm,64%,and-1.8°C,respectively.The mean temperatures in January and July as well as the minimum and maximum temperatures during the year are-7.8,9.8,-31.6,and 24.0°C,respectively.The mainly brown and acid brown soil has an average thickness of 60 cm with obvious humi fi cation(Wang and Zhang 2011;Wang et al.2012).The dominant trees in the study area are Q.aquifolioides,Betula utilis,Abies georgei var.smithii,B.platyphylla,and Picea likiangensis var.linzhiensis.The shrub layer is composed mainly of species such as Rhododendron nyingchiense,Sorbus rehderiana,Rosa sericea var.omeiensis,Lonicera lanceolata,Lonicera inconspicua,and Rhododendron hirtipes.The herbaceous layer mainly includes species such as Parasenecio quinquelobus,Sinopodophyllum hexandrum,Ainsliaea latifolia,Vicia tibetica,Fragaria moupinensis,Oxalis acetosella,and Circaea alpina.

In June to August,2015,we established a single 1 ha(100 m×100 m) fi xed monitoring sample plot.The southwest corner of plot was set as the origin.Data tallying information related to Q.aquifolioides in the quadrat were collected,Use 205-type tree diameter tape(Jiejia Co.Ltd.Shenyang,China),IMPULSE-200LR type laser altimeter,diastimeter(Leaves Qiao Technology Co.Ltd.Beijing,China)or steel tape and tape(Kanghong measuring Co.Ltd.Shangqiu,China)respectively for measuring and recording the tree diameter at breast height(DBH),tree heightand plants geographicalcoordinates (x,y)(Table 1).

where n1,n2represent individuals at different growth stages,I and j are individual 1 and individual 2,respectively.Other abbreviations follow formula(1).We simpli fi ed Formula(3)as Formula(4):

师：嗯，你提出了一个很有价值的问题，就是“角“有大小之别，那角的大小用什么来度量？这节课，我们一起来学习角的度量。板书课题：角的度量

Analysis of spatial correlation

Under the same environmental conditions,trees of the same age class and DBH responded consistently to those conditions(Frost and Rydin 2000);therefore,the DBH of trees was used during data analysis to replace the age classes.Based on a combination of life history characteristics of Q.aquifolioides and fi eld survey data,we divided the Q.aquifolioides trees in the study area into the following DBH classes:Class I,DBH＜5 cm;Class II,DBH 5 to＜10 cm;Class III,DBH,10 to＜15 cm;…Class IX,DBH 40 to＜45 cm,creating nine size classes(I–IX).The data related to the number of trees for each diameter class were measured in each plot.According to the diameter class characteristics and data,we divided the Q.aquifolioides populations on Sejila Mountain into three age groups,namely:saplings(Class I–II),medium-(Class III–VI),and large-sized trees(Class VII–IX).

In order to improve the accuracy of spatial distribution pattern analysis,we used a stochastic Monte Carlo simulation with 1000 replicates to obtain a 99%con fi dence envelope.The maximum and minimum values of（ r）were in the upper and lower envelope curve of the coordinate values.With r as the abscissa,the lower envelope of the ordinate was plotted.We used data from the practical distribution of the population to obtain values of（ r）at different scales.An aggregate distribution would be plotted above the envelope curve,a uniform distribution would be plotted below the envelope curve and a random distribution is within the envelope curve(Wiegand and Moloney 2004).

When the value of1,2（r）＞0 there was a positive association.When the value of1,2（r）＜0 there was a negative association.When the value of1,2（r）=0,there was no association.If1,2（r）was within the envelope line,individuals at different growth stages showed no association.If1,2（r）was above the envelope line,individuals at different growth stages showed a positive association.If1,2（r）was below the envelope,individuals at different growth stages showed a negative association.We used ADE4 software to process data(Thioulouse et al.1997).The spatial scale was 25 m and step size was 1 m.We obtained a 99%con fi dence envelope through 1000 Monte-Carlo simulations.We used Excel(Microsoft,Redmond,OR,USA)and Origin 9.0 mapping software(OriginLab,Northhampton,MA,USA)for data analysis.

Results

Population structure

The individuals of the Q.aquifolioides population were mainly aggregated into Classes I–VI,which included 94.3% of all individuals,with many fewer individuals in Classes VII–IX(Fig.2).Although the number of individuals in Classes I and II was less than that of Class III,the total number of individuals in Classes I and II was greater than in Classes IV–IX.The diameter class structure of the studied population was an atypical‘pyramid’type,with the population classi fi ed as an expanding population.However,growth was limited and the population had a tendency to stabilize.An analysis of the tree height structure indicated that relatively few individuals were less than 5 m tall,while individuals with heights up to 10 m were most common(206 individuals,accounting for 56% of the population).Few individuals were taller than 20 m,accounting for only 2.4% of the population.That is,the number of saplings was limited with larger quantities of medium-sized trees and a small number of large-sized trees in the plots;this fi nding was consistent with the results of the analysis of diameter class structure.Q.aquifolioides showed a growth-oriented pattern in the study area,but growth was limited;a tendency was observed for the population to stabilize,so regeneration of the population could be promoted by the planting of supplemental seedlings.

本文基于粗粒度策略对数据进行划分，设样本数据集合M={m1,m2,…,ml}包含l个样本对象，任意数据之间的距离(指欧氏距离)为dist(di，dj)，粒度变量为Gv，表示数据原点的融合半径，计算公式为：

Fig.2 Diameter class and height distribution of Q.aquifolioides populations.Size class de fi nitions based on diameter at breast height(DBH):Class I,DBH＜5 cm;Class II,DBH 5 to＜10 cm;Class III,DBH,10 to＜15 cm;… Class IX,DBH 40 to＜45 cm,creating nine size classes(I–IX)

Static life table of Q.aquifolioides

When checking fi eld survey data of Q.aquifolioides populations,we found that the data for age classes I and II changed;the measurement of survival for those classes was smaller than that of age classes II and III.At the same time,because the number of individuals in age Class V exceeded that of Class I,the smoothing zone for the age classes I–V was determined.Based on the assumptions of the speci fi c life table,the age composition of the studied population was stable.The proportion of different ages was stable and unchanged,af fi rming age classes I–V,calculating section survival numbers:the average(class mid-value)=T/n=319/5≈64.According to the difference between the maximum and minimum survival numbers,namely,(79–45=34),the count by was 4 and the difference between the survival number of adjacent age groups was 34/4≈9.Thus,after smoothing correction to give ax,we prepared a speci fi c life table of Q.aquifolioides populations(Table 2).

As can be seen from the static life table of Q.aquifolioides populations on Sejila Mountain,the mortality rate increased with increasing age.This indicates the biological characteristics of the population as well as intraspeci fi c competition were major factors affecting their growth and development.The lowest mortality rate was in Class I,only 0.11,and might have occurred because this age class of individuals had less competition than other individuals.With increasing age,intraspeci fi c competition increased and mortality rates rose;these results could show there were some de fi ciencies in the availability of the seedling bank for age Class I.Planting seedlings would be necessary to ensure the sustainable development of the population.Age classes V–VII had higher mortality because of the intraspeci fi c competition for the resources(space,sunlight,etc.)over time;Age Class VIII experienced the highest mortality because Q.aquifolioides populations enter into the age where physiological death frequently occurs.Life expectancy of the Q.aquifolioides populations continued to show a constant decreasing trend with increasing tree age,which depended mainly on the biological characteristics of their own populations.This indicated that natural growth and regeneration of Q.aquifolioides populations on Sejila Mountain was relatively stable.

Survival curve of Q.aquifolioides populations

Figure 4 shows the distribution of individuals in the Q.aquifolioides populations on SejilaMountain in the 100 m×100 m fi xed standard plot.Figure 4 shows that the density and spatial distribution of Q.aquifolioides populations in the different growth stages had large differences.The medium-sized trees were most frequent and had the densest populations with 221 individuals ha-1.Samplings were the second most dense with 121 individuals ha-1.The distribution of large-sized trees was far less dense than the density of saplings and medium-sized trees with about 21 individuals ha-1.

银行规模与向小企业贷款力度呈反向关系，广西贫困县县域经济主要以中小企业为主。对县域中小企业提供足够资金支持，就会对县域经济第二产业发展产生显著的促进作用。发展本地小商业银行，有利于向本地县域中小企业贷款，服务本地经济。

Table 2 Static life table of Q.aquifolioides populations

DBH diameter at breast height;X age-class;Axindividual number of actual survivor for x age class;ax number of surviving individuals after smoothing for x age class;lxstandard number of surviving individuals starting from x age class(with 1000 as the base),lx=ax/a0×1000;dxstandardized mortality,number of individuals from age class x to x+1,dx=lx–lx+1;qxmortality from age class x to x+1 during the interval,qx=dx/lx;Lxaverage number of surviving individuals from age class x to x+1 during the interval,Lx=(lx+lx+1)/2;Txtotal number of surviving individuals from age class x to individuals exceeding age class x,Tx=∑Lx;exindividual life expectancy which enters into the age class x,ex=Tx/lx;Kxdisappearance rate,namely,loss degree,Kx=lnlx–lnlx+1

DBH class Ax ax lx lnlx dx qx Lx Tx ex Kx I 45 82 1000.00 6.91 109.76 0.11 945.12 4000.02 4.00 0.12 II 76 73 890.24 6.79 109.75 0.12 835.37 3054.90 3.43 0.13 III 79 64 780.49 6.66 109.76 0.14 725.61 2219.53 2.84 0.15 IV 62 55 670.73 6.51 109.75 0.16 615.86 1493.92 2.23 0.18 V 57 46 560.98 6.33 219.52 0.39 451.22 878.06 1.57 0.50 VI 28 28 341.46 5.83 182.92 0.54 250.00 426.84 1.25 0.76 VII 13 13 158.54 5.07 85.37 0.54 115.86 176.84 1.12 0.78 VIII 6 6 73.17 4.29 48.78 0.67 48.78 60.98 0.83 1.10 IX 2 2 24.393.19– – 12.20 12.200.50–

Spatial distribution of Q.aquifolioides populations

The survival curve of Q.aquifolioides populations on Sejila Mountain was close to Deevey type I;age classes I–V had a higher survival rate,and then the survival rate decreased with increasing age(Fig.3).Among them,theself-thinning effect caused by intraspeci fi c competition was the main reason leading to a decline in the survival rate for age classes VI–IX.

Fig.3 Survival curve of the Q.aquifolioides populations.Size class de fi nitions based on diameter at breast height(DBH):Class I,DBH＜5 cm;Class II,DBH 5 to＜10 cm;Class III,DBH,10 to＜15 cm;… Class IX,DBH 40 to＜45 cm,creating nine size classes(I–IX)

Spatial distribution pattern analysis

Figure 5 shows Q.aquifolioides presented an aggregated distribution with a trend for a random distribution among saplings,medium-to large-sized trees.Saplings on a scale of 0–33 m showed an aggregated distribution,and on a scale of0–7 m,the aggregation intensity gradually increased;however,on a scale of 8–33 m,the aggregation intensity gradually weakened.On a scale of 34–50 m,the saplings showed a random distribution;on a scale of 0–29 m,medium-sized trees showed an aggregated distribution and aggregation intensity gradually increased;however,on a scale of 30–50 m medium-sized trees showed a random distribution,while on a scale of 10–29 m the aggregation intensity of medium-sized trees was signi fi cantly stronger than that of saplings;on a scale of 20–30 m large-sized trees showed an aggregated distribution,but at other scales they showed a random distribution.

Spatial association analysis

From the spatial association analysis of Q.aquifolioides populations on Sejila Mountain at the different age classes(Fig.6),we found no correlations among saplings,medium-,and large-sized trees at scales smaller than 0–5 m,while a signi fi cant and negative association was observed on scales of 6–50 m.Here,strong competition existed between saplings,medium-trees,and large-sized trees,while there were no obvious associations between mediumtrees and large-sized trees in any of the scales.

2.5.14 术后白细胞尿 出现白细胞尿是前列腺等离子电切术后常见症状。有临床研究发现白细胞尿与尿路感染并无相关性。术后1周患者尿液中的白细胞浓度升高，但在术后4周时，尿液中的平均白细胞计数降低。因此，白细胞尿可能与前列腺手术创面炎性细胞的渗出有关，白细胞尿不能反映术后菌尿的可能性。随着手术创面的愈合，白细胞尿可自然好转，不需要进行临床干预。

Fig.4 Spatial distribution points of Q.aquifolioides individuals in a population.P entire population,S saplings,diameter at breast height(DBH)＜10 cm;M medium-sized trees,DBH 10 to＜30 cm;L large-sized trees,DBH 30 to＜45 cm

Fig.5 Spatial distribution pattern of Q.aquifolioides at different developmental stages.S saplings,diameter at breast height(DBH)＜10 cm;M medium-sized trees,DBH 10 to＜30 cm;L large-sized trees,DBH 30 to＜45 cm.Thin dashed lines indicate function value(L(r));Black lines indicate the upper and lower limits of the 99% con fi dence interval.Points above the upper limit of the interval stand for a clumped distribution,points within the interval stand for a random distribution,and points below the lower limit of the interval stand for a regular distribution

Discussion

The study of population structure is conducive to understanding regeneration types within a community and allows further speculation on the development trend of a community(Condit et al.2000);diameter class structure may affect the spatial patterns of plant populations.To a certain extent,diameter class structure re fl ected the living conditions of the population,while at the same time the height of the individuals within the plant population may re fl ect the status of the individual in the vertical structure of the community(Frost and Rydin 2000;Ma and Zu 2000;Lan et al.2012).In the present study,the individual Q.aquifolioidespopulationsweremainly aggregated in Classes I–VI that accounted for 94.3% of all individuals;fewer individuals were found in Classes VII–IX.However,the number of individuals in classes I and II were less than that of Class III,the total number of individuals in classes I and II was also greater than in classes IV–IX.The population was expanding so that the diameter class structure formed an atypical‘pyramid’type.The population growth was limited and exhibited a tendency to become stable.Based on tree height structure,there was a limited number of saplings,larger quantities of medium-sized trees,and a small number of large-sized trees;this fi nding was consistent with the results of diameter class structure.In terms of diameter class structure,the Q.aquifolioides population was growth-oriented,but growth was limited.The ability for natural regeneration and diffusion was rather weak.There was a tendency for the population to become stable,so we could promote regeneration of the population by planting supplemental seedlings.In addition,seedling survival could be stimulated through the use of selective cutting of medium-and larger-sized trees.This would allow for increased regeneration of the population.

The life table and survival curve characteristics of the population represent the combined effects of a species’own characteristics,interspeci fi c relationships,and the effects of environmental conditions(Crawley 1986;Skoglund and Verwijst 1989).Mortality increased constantly with an increaseinageforQ.aquifolioidespopulations,sothatClass I had the lowest mortality rate.This occurred because seedlings grew in the community substratum with less competition among individuals;seedlings mutually helped each other to use resources so that the community improved the ability of individuals to survive.At the same time,the numberof seedling present inthe community was quite low;additional seedlings were needed to ensure the sustainable development of the population,even though the entire population was experiencing stable growth.Mortality increased gradually with increasing age,mainly because intraspeci fi c competition caused an enhanced self-thinning effect.The natural age of physiological death was the main cause of mortality of large-sized Q.aquifolioides trees.Life expectancy of Q.aquifolioides populations exhibited a constant decreasing trend with increasing age that depended on the biological characteristics of each population.That is,mortality was low for young tree and high forolder trees making the survival curve of Q.aquifolioides populations on Sejila MountainclosetoDeevey typeI.Thehighdeath ratelaterin lifeoccurred becauseindividualswere affected by intra-and interspeci fi c competition for light,moisture,nutrients,and space;so that selection intensity increased,leading to an increasing death rate(Cheng 2010).

（5）缸筒表面外观质量 激光淬火后，缸筒淬火区无明显的氧化脱碳现象，表面粗糙度值较低，经磁粉无损检测后表面无微小裂纹现象。并经专业的检测手段检测缸筒淬火区没有变形缺陷产生。

Fig.6 Spatial correlation of Q.aquifolioides at different developmental stages.S saplings,diameter at breast height(DBH)＜10 cm;M medium-sized trees,DBH 10 to＜30 cm;L large-sized trees,DBH 30 to＜45 cm.Thin dashed lines indicate the function value(L1,2(r));Black lines indicate the upper and lower limits of the 99%con fi dence interval.Points above the upper limit of the interval indicate a positive association,points within the interval indicate no spatial association,and points below the lower interval indicate a negative association

The spatial distribution patterns of plant populations are extremely important to revealing the dynamics of a population and predicting the development of populations and communities(Wu et al.2002).Some researchers believe that young trees have a tendency to form an aggregated distribution in order to improve their survival rates(Condit et al.2000).As the ages of trees increases,their individual ability to withstand external stressors increases,causing them to tend to take on a uniform or random distribution(Condit et al.2000).The results of the present study were basically consistent with this concept.As saplings grew to become large-sized trees,the spatial pattern of the population was basically transformed from an aggregated to a random distribution.Saplings and medium-sized trees exhibited an intense aggregated distribution at small scales,while large-sized trees showed a random distribution.The aggregated distribution of saplings was caused by uneven seed germination and limited early seedling survival(Grubb 1977;Harms et al.2000).Canopy gaps(Nicotra et al.2008),soil(Hall et al.2004),and terrain(He et al.1977)may lead to an aggregated distribution of saplings.Saplings took advantage of favorable resources and formed clusters of trees that improved their chances of survival;meanwhile,large-sized trees were randomly distributed based on intraspeci fi c competition that mainly resulted in self-thinning of the population(Condit et al.2000).

The spatial association of different age classes of the population should be analyzed to understand the reproduction and spatial diffusion characteristics of the population(Zhang et al.2013).Intraspeci fi c links refer to the mutual association of the spatial distribution within a single species in different age classes,which could re fl ect the mutualistic relationship among individuals within populations(Wang et al.2010).Some authors believe that among individual plants with a similar body size,a stronger positive association among individuals would be conducive to the stable development of the population.Conversely,a stronger negative association with more intense intraspeci fi c competition would not be conducive to the development of smaller individuals;these factors would affect the natural regeneration of an entire population(Zhang et al.2005;You et al.2010).In the present study,strongly negative associations were found among saplings,medium-,and large-sized trees of Q.aquifolioides at larger scales because the presence of medium-and large-sized trees resulted in an intense inhibition of regeneration for saplings.With the ability of saplings to deal with external stressors increasing and the demands for space,soil,moisture,and nutrients;competition among different age classes will increase constantly over time(Song et al.2010).Medium-and large-sized trees exhibited essentially no association because the medium-and large-sized trees outcompeted saplings leading to some death of saplings.This infraspeci fi c competition also increased the distance between medium-and large-sized trees,reducing competition among them.This may also occur because the medium-and large-sized trees had an independent ability to withstand environmental stressors,so that their mutual dependence was greatly reduced,leading to decreasing association with the change of aggregation intensity.This was bene fi cial to the sound development of entire Q.aquifolioides populations.

Q.aquifolioides is an important component of vegetation on Sejila Mountain,Tibet,China and plays extremely important roles in protecting biodiversity and in conserving water and soil.The present study addresses the population structure and spatial distribution patterns of this species;correctly understanding the population structure change and the relationship between the population and the environment is important.In the future,the mutualistic association between Q. aquifolioides and other plant populations should be studied in more depth.The goal would be to reveal the mechanisms involved in the maintenance of the plant population and the relationships between Q.aquifolioides and other species as well as the entire community.

Acknowledgements We would like to thank all those who provided helpful suggestions and critical comments on the manuscript.We also thank San Gao for providing a critical revision of the English.Zhiqiang Shen performed research and wrote the paper.Xiaoqin Tang,Jie Lu and Zhiqiang Shen designed the experiments.Zhiqiang Shen,Min Hua,Xingle Qu,Jiangping Fang and Jingli Xue collected and analyzed the data.We also thank LetPub(www.letpub.com)for its linguistic assistance during the preparation of this manuscript.

为了进一步说明优化的分类模型在红枣品种分类中的优势性，对比了两个模型在训练集样本与测试集样本一致的情况下的相关系数(R2,[0,1])和均方根误差(RMSE)，结果如表4所示。从表中结果可知GA-SVM分类模型的训练集和测试集相关系数均在95%以上，同时均方根误差也较低。相比于传统SVM分类方法，GA-SVM分类模型在干制红枣品种分类方面有着显著优越性。

References

Arista M(1995)The structure and dynamics of an Abies pinsapo forest in southern Spain.For Ecol Manage 74:81–89

Armesto JJ,Casassa I,Dollenz O(1992)Age structure and dynamics of Patagonian beech forests in Torres del Paine National Park,Chile.Vegetatio 98:13–22

Borchsenius F,Nielsen PK,Lawesson JE(2004)Vegetation structure and diversity of an ancient temperate deciduous forest in SW Denmark.Plant Ecol 175:121–135

Chen LT,Flynn DFB,Zhang XW,Gao XL,Lin L,Luo J,Zhao CM(2015)Divergent patterns of foliarδ13C andδ15N in Quercus aquifolioides with an altitudinal transect on the Tibetan Plateau:an integrated study based on multiple key leaf functional traits.J Plant Ecol 8:303–312

Cheng HM(2010)Life table and survival analysis of Tilia breviradiata population in Dashu Mountain.J Zhejiang Univ(Agric Life Sci)36:341–347(in Chinese)

Condit R,Ashton PS,Baker P,Bunyavejchewin S,Gunatilleke S,Gunatilleke N,Hubbell SP,Foster RB,Itoh A,LaFrankie JV,Lee HS,Losos E,Manokaran N,Sukumar R,Yamakura T(2000)Spatial patterns in the distribution of tropical tree species.Science 288:1414–1418

Crawley MJ(1986)Plant ecology.Blackwell,London,pp 97–185

Cui CM,Wang XA,Guo H,Li W(2011)Structure and dynamic change of natural Pinus tabulaeformis populations in the Ziwuling forest area.Arid Zone Res 28:111–117(in Chinese)

Dale MRT (1999)Spatial pattern analysis in plant ecology.Cambridge University Press,Cambridge,p 119

Deevey ES(1947)Life tables for natural populations of animals.Q Rev Biol 22:283–314

Díaz S,Mercado C,Alvarez-Cardenas S(2000)Structure and population dynamics of Pinus lagunae M.-F,Passini.For Ecol Manage 134:249–256

Didier KA,Porter WF(2003)Relating spatial patterns of sugar maple reproductive success and relative deer density in northern New York State.For Ecol Manage 181:253–266

Duan DY,Ouyang H,Song MH,Hu QW(2008)Water sources of dominant species in three alpine ecosystems on the Tibetan Plateau,China.J Integr Plant Biol 50:257–264

Frost I,Rydin H(2000)Spatial pattern and size distribution of the animal-dispersed tree Quercus robur in two spruce-dominated forests.Écoscience 7:38–44

Getzin S,Wiegand T,Wiegand K,He F(2008)Heterogeneity in fl uences spatial patterns and demographics in forest stands.J Ecol 96:807–820

Greig-Smith P(1983)Quantitative plant ecology.Blackwell,London,pp 54–60

Grubb PJ(1977)The maintenance of species-richness in plant communities:the importance of the regeneration niche.Biol Rev 52:107–145

Hall JS,McKenna JJ,Ashton PMS,Gregoire TG(2004)Habitat characterizations underestimate the role of edaphic factors controlling the distribution ofEntandrophragma.Ecology 85:2171–2183

Harcombe PA(1987)Tree life tables.Bioscience 37:557–568

Harms KE,Wright SJ,Calderón O,Hernández A,Herre EA(2000)Pervasive density-dependentrecruitmentenhances seedling diversity in a tropical forest.Nature 404:493–495

Harper JL(1977)Population biology of plants.Academic Press,London,p 96

He FL,Legendre P,LaFrankie JV(1977)Distribution patterns of tree species in a Malaysian tropical rain forest.J Veg Sci 8:105–114 Johnson JB(1997)Stand structure and vegetation dynamics of a subalpine wooded fen in Rocky Mountain National Park,Colorado.J Veg Sci 8:337–342

Lan G,Getzin S,Wiegand T,Hu Y,Xie G,Zhu H,Cao M(2012)Spatial distribution and interspeci fi c associations of tree species in a tropical seasonal rain forest of China.PLoS ONE 7:e46074 Li CY,Zhang XJ,Liu XL,Luukkanen A,Berninger F(2006)Leaf morphological and physiological responses of Quercus aquifolioides along an altitudinal gradient.Silva Fenn 40:5–13

Li C,Wu C,Duan B,Korpelainen H,Luukkanen O(2009)Agerelated nutrient content and carbon isotope composition in the leaves and branches of Quercus aquifolioides along an altitudinal gradient.Trees 23:1109–1121

Ma KM,Zu YG(2000)Fractal properties of vegetation pattern.Acta Phytoecol Sin 24:111–117(in Chinese)

Miyadokoro T,Nishimura N,Yamamoto S(2003)Population structure and spatial patterns of major trees in a subalpine oldgrowth coniferous forest,central Japan.For Ecol Manage 182:259–272

Nicotra AB,Chazdon RL,Iriarte SVB(2008)Spatial heterogeneity of light and woody seedling regeneration in tropical wet forests.Ecology 80:1908–1926

Ripley BD (1977)Modelling spatial patterns.J R Stat Soc 39:172–212

Skoglund J,Verwijst T(1989)Age structure of woody species populations in relation to seed rain,germination and establishment along the river Dala lven,Sweden.Vegetatio 82:25–34

Somanathan H,Borges RM(2000)In fl uence of exploitation on population structure,spatial distribution and reproductive success of dioecious species in a fragmented cloud forest in India.Biol Cons 94:243–256

Song MH,Zhou CP,Ouyang H(2004)Distributions of dominant tree species on the Tibetan Plateau under current and future climate scenarios.Mt Res Dev 24:166–173

Song YY,Li YY,Zhang WH(2010)Analysis of spatial pattern and spatial association of Haloxylon ammodendron population in different developmental stages.Acta Ecol Sin 30:4317–4327(in Chinese)

Stewart GH(1989)The dynamics of old-growth Pseudotsuga forests in the western Cascade Range,Oregon,USA.Vegetatio 82:79–94

Thioulouse J,Chessel D,Doledec S,Olivier JM(1997)Ade-4:a multivariate analysis and graphical display software.Stat Comput 7:75–83

Wang XC,Zhang QB(2011)Evidence of solar signals in tree rings of Smith fi r from Sygera Mountain in southeast Tibet.J Atmos Solar Terr Phys 73:1959–1966

Wang ZS,Liu H,Wei Na XuWX,An SQ,Liu SR(2009)Effects of stand regeneration management regimes and age on genetic structure of Quercus aquifolioides(Sclerophyllous Oak)in Southwestern China.For Sci 55:142–148

Wang L,Sun QW,Hao CY,Tian SN,Zhang SS,Chen YK,Zhang XP(2010)Point pattern analysis of different age class Taxus chinensis var.mairei individuals in mountainous area of southern Anhui Province.Chin J Appl Ecol 21:272–278(in Chinese)

Wang YF,čufar K,Eckstein D,Liang EY(2012)Variation of maximum tree height and annual shoot growth of Smith fir at various elevations in the Sygera Mountains,southeastern Tibetan Plateau.PLoS ONE 7:e31725

Wiegand T,Moloney KA(2004)Rings,circles,and null-models for point pattern analysis in ecology.Oikos 104:209–229

Wiegand T,Gunatilleke S,Gunatilleke N(2007)Species associations in a heterogeneous Sri Lankan dipterocarp forest.Am Nat 170:E77–E95

Wu XP,Zheng Y,Ma KP(2002)Population distribution and dynamics of Quercus liaotungensis,Fraxinus rhynchophylla and Acer mono in Dongling Mountain,Beijing.Acta Bot Sin 44:212–223(in Chinese)

Yang QS,Zhao Y(2013)Concentrations and pollution assessment of six heavy metals in Quercus aquifolioides forest soils.Adv Mater Res 807–809:184–189

You HZ,Liu XL,Miao N,He F,Ma QY(2010)Individual association and scale effect of spatial pattern of Quercus aquifolioides populations along the elevation gradients.Acta Ecol Sin 30:4004–4011(in Chinese)

Zhang YD,Liu SR,Zhao CM(2005)Spatial pattern of sub-alpine forestrestoration in westSichuan.Chin J ApplEcol 16:1706–1710(in Chinese)

Zhang XW,Zhang XP,Guo CY,Zhang Q(2013)Point pattern analysis of Pteroceltis tatarinowii population at its different development stages in limestone mountain area of north Anhui,East China.Chin J Ecol 32:542–550(in Chinese)

Zhou HR,Xu M,Pan HL,Yu XB(2015)Leaf-age effects on temperature responses of photosynthesis and respiration of an alpine oak,Quercus aquifolioides,in southwestern China.Tree Physiol 35:1236–1248

Zhu WZ,Wang SG,Yu DZ,Jiang Y,Li MH(2014)Elevational patterns of endogenous hormones and their relation to resprouting ability of Quercus aquifolioides plants on the eastern edge of the Tibetan Plateau.Trees 28:359–372