【作者】 毕守东；

【导师】 邹运鼎；

【作者基本信息】 安徽农业大学 ， 森林保护学， 2007， 博士

【摘要】 本文在系统综述群落生态学有关理论与研究进展的基础上,较全面地阐述了害虫管理中的现代数学方法,这些方法涉及多元统计分析、灰色系统理论、模糊数学与地学统计学等,并着重将它们应用于桃园、李园、枣园等的优势种天敌评价中。其主要研究结果如下:1.根据桃园节肢动物群落物种的组成和数量,按照营养和取食关系,将群落划分为植食类、捕食类、寄生类和中性类亚群落,按照系统分科的方法,将每个亚群落分成不同的功能类群,采用群落分析的多个参数比较,分析了群落的结构和季节动态。结果表明,桃园共调查出节肢动物55种,分属于37科,其中植食类28种,分属于18科;捕食类22种,分属于14科;寄生类3种,分属于3科;中性类2种,分属于2科。植食类的相对丰盛度为0.953367,季节变动幅度为0.38～0.97,捕食类相对丰盛度为0.046188,季节变动幅度为0.01～0.54,两者变化趋势一致。各亚群落的优势集中性表现为植食类(0.980648)>捕食类(0.886292)>寄生类(0.420000)。H’多样性指数为捕食类节肢动物亚群落>节肢动物总群落>植食类节肢动物亚群落;Hill多样性指数(N1)为捕食类节肢动物亚群落>节肢动物总群落>植食类节肢动物亚群落;Hill多样性指数(N2)为捕食类节肢动物亚群落>节肢动物总群落>植食类节肢动物亚群落。捕食类节肢动物群落的个体数量高峰期与植食类群落个体数量高峰值期相吻合,表明捕食类节肢动物是桃树生长期植食类害虫的重要天敌。2.应用主成分分析法、最优分割法和系统聚类分析方法对桃园害虫天敌群落进行分析。结果表明,群落具有明显的主导因素和时间格局;对桃园节肢动物群落55种节肢动物的数量变化进行系统聚类分析,选置信水平λ=0.47,将其分为4个时期,利用最优分割法,将害虫天敌的发生量分为3段,利用主成分分析法,明确了桃园主要害虫及其优势天敌种。3.采用模糊相似优先比与模糊综合评判方法研究桃园害虫与其主要天敌种群数量的相互联系,模糊相似优先比分析显示:桃一点叶蝉的主要天敌是草间小黑蛛;桃蚜的主要天敌是蟹蛛;叶甲科害虫的主要天敌是草蛉;卷叶蛾科害虫的主要天敌是草蛉。模糊综合评判显示:桃一点叶蝉的优势种天敌仍是草间小黑蛛。4.应用灰色系统理论研究了李园中主要害虫桃蚜的各种天敌对桃蚜的控制作用大小,对桃蚜与其各种天敌之间关联度进行分析,结果是,天敌的日捕食总量与桃蚜种群数量之间关联度高的天敌依次是大草蛉(0.7449)、龟纹瓢虫(0.7395)、八斑球腹蛛(0.7354)、肖蛸蛛(0.7198)等,天敌的日捕食总量与理想优势种天敌的日捕食总量之间关联度较高的依次是,草间小黑蛛(0.9692)、龟纹瓢虫(0.6706)、肖蛸蛛(0.6505);天敌数量与桃蚜种群数量之间关联度较高的依次是,大草蛉(0.7449)、龟纹瓢虫(0.7395)、八斑球腹蛛(0.7354)和肖蛸蛛(0.7198);天敌数量与理想优势种天敌的数量之间关联度较高的依次是草间小黑蛛(0.9703)、龟纹瓢虫(0.6725)和肖蛸蛛(0.6482)。综合分析得出,对桃蚜种群数量控制作用大的优势种天敌依次是草间小黑蛛、大草蛉、龟纹瓢虫、八斑球腹蛛、肖蛸蛛。5.利用灰色局势决策理论,对李园桃蚜的优势种天敌作了灰色评判。结果表明,从捕食强度来看,整个调查期内,李园桃蚜的优势种天敌前三位的是:大草蛉(0.3646)、龟纹瓢虫(0.3202)、八斑球腹蛛(0.0445)。6.对枣园节肢动物进行了系统调查,共查到52种,分属47科。蜘蛛类的相对丰盛度最高,为0.3465,瘿蚊科的相对丰盛度为0.2309,二者在群落中所占的比例较高。6月8日和9月21日是枣园节肢动物群落个体数量的两个峰值期。对不同时期的整个群落及各亚群落进行模糊聚类,确定阈值T=0.92,各次调查的整个群落分为四类:3月22日、4月6日和11月22日各自归为一类,其余归为一类;当阈值T=0.94时,不同时间的天敌亚群落聚为3类:3月22日、4月6日各自归为一类,其余归为一类;当阈值T=0.94时,非天敌亚群落聚为4类:3月22日、4月6日和6月8日各自归为一类,其余归为一类。聚类结果反映了群落特征的季节差异,聚类交替现象则反映了群落结构在较长时间内的复杂性。7.利用地统计学原理和方法对枣园中枣瘿蚊及其优势种天敌草间小黑蛛种群的空间结构和空间依赖性进行了研究。结果表明,6月8日、8月27日、10月19日等枣树主要生育期阶段,枣瘿蚊种群的半变异函数模型为球形。在林间基本上呈聚集分布,其样点空间依赖性距离分别为26.3m、7.8m和22.0m,其空间依赖性程度分别为88.05%、85.77%和87.58%。草间小黑蛛种群的半变异函数也为球形,呈聚集分布,空间依赖距离分别为28.3m、22.5m和22.2m,且在该取样尺度下,总变异中分别有90.09%、87.96%和85.72%是由空间依赖性产生的。两者空间依赖关系说明草间小黑蛛对枣瘿蚊有较强的追随作用。8.应用地统计学中的交叉变差函数法对枣园中枣瘿蚊及其优势种天敌草间小黑蛛种群的空间相关性进行了研究。结果表明,5月21日至10月19日近5个月的枣树主要生育期阶段,除8月27日与9月21日草间小黑蛛与枣瘿蚊种群没有明显的空间相关性以外,其他各次尤其是5月30日与10月19日都充分表现出草间小黑蛛对枣瘿蚊在不同距离下的数量和空间的追随关系,再次证明草间小黑蛛是枣瘿蚊的优势种天敌。9.利用地学统计学原理和方法对草坪中粟茎跳甲及其优势种天敌草间小黑蛛种群的空间结构和空间相关性进行了研究。结果表明,粟茎跳甲种群的半变异函数模型为球形,在草坪上呈聚集分布,其样点的空间依赖性距离为18.82m,在该取样尺度下,空间依赖性程度为49.39%;交叉验证的估计误差呈正态分布,偏差均值KRE=0.0638,一致性系数KRMSE=0.9900,粟茎跳甲种群在调查区域内有5个聚集斑块。草间小黑蛛种群的半变异函数也为球形,亦呈聚集分布,空间依赖性距离为22.45m,且在该取样尺度下,总变异中有33.8%是由空间依赖性产生的,交叉验证的估计误差亦呈正态分布,偏差均值KRE=0.0185,一致性系数KRMSE=0.9902,草间小黑蛛在调查区域内也有5个聚集斑块,并且与粟茎跳甲的聚集斑块位置基本相同。两者变程、空间分布格局及空间依赖关系的一致性说明了草间小黑蛛种群对粟茎跳甲种群在空间位置上的较强的追随关系,充分证明了草间小黑蛛是粟茎跳甲的优势种天敌。更多还原

【Abstract】 Based on a systemic review on the advances of community ecology theories,this paper demonstrated the application of modern mathematic methods,such as multivariable statistics,grey system theory,fuzzy,and geostatistics,and so on,in the pest management system.In the practice,these mathematic methods were used to evaluate the dominant natural enemies in peach orchards,plum orchards and jujube orchards.And the corresponding findings were showed as follows:1.The arthropod community was divided into four sub-communities:phytophages, predators,parasitorids and neutralities,according to the trophic and taxonomy relationship.Several community parameters were employed to analyse the structures and dynamics of arthropod community and the sub-communities.The results showed that there were 55 species of arthropod belonging to 37 families,including 28 species of phytophages representing 18 families;22 species of predators representing 14 families;3 species of parasitords representing 3 families;2 species of neutralities representing 2 families in peach orchards.In the whole,the relative abundance of phytophages and predators were 0.953367 and 0.046188 respectively,which indicated that the phytophages dominated the arthropod community in peach orchards.Whereas,these abundances exhibited a similar seasonal fluctuate which ranged from 0.38 to0.97 for the phytophages, and from 0.01 to 0.54 for the predators.The dominant concentration of phytophages was 0.980648,larger than that of predators（0.886292）and parasitoids（0.420000）.The other parameters such as diversitv index（H）and diversity index（Hill）including N₁,N₂,revealed the same trends,and showed the following order:predators sub-community＞arthropod community＞phytophages sub-community.The dynamics of the relative abundance of predators was similar to that of the phytophages.Therefore,it could be found that the predators were dominant natural enemies which influenced phytophages in the peach growth stage.2.Investigations on the structure of arthropod community in peach orchards were conducted with the principal component analysis,the optimal sorting method and system clustering analysis.The results showed that more obvious main factors and temporal patterns exist in the community.Dealing with the population dynamics of 55 species of arthropod by system clustering analysis and takingλ=0.47,we divided the community into four stages.The number of pests and their natural enemies was grouped into three parts by the optimal sorting method.The dominant species of pests and their natural enemies were demonstrated by the principal component analysis.3.The correlation of the major pest insects and natural enemies in peach orchards was analysed,by the method of fuzzy-similitude-priority（FSP）and fuzzy-judgement（FJ）.The results of FSP showed that the dominant natural enemy of Erythroneura sudra,,Myzus persicae,Chrysomelidae and Tortricidae were Thomisidae,Erigonidium graminicola, Chrysopidae,Chrysopidae respectively.And,the results of FJ showed that the dominant natural enemy of Erythroneura sudra was Erigonidium graminicola. 4.In order to evaluate the controlling rate of the various natural enemies on the dominant pest Myzus persicae in plum field,the relational grade between M.persicae and their various natural enemies was analyzed with the grey system theory.The results indicated that the relational grade between daily total predation of natural enemies and M.persicae population size was given as follows,from highly to lowly in proper order: Chrysopa septempunctata（0.7449）,Propylaca japonica（0.7395）,Therdion octomaculatum （0.7354）,Tetragnathidae（0.7198）etc,and the one between daily total predation of ideal dominant natural enemies and natural enemies was Enigonidium graminicola（0.9672）, Propylaca japonica（0.6706）,Tetragnathidae（0.6505）respectively.The higher relational grade between the population size of natural enemies and that of Myzus persicae was Chrysopa septempunctata（0.7449）,Propylaca japonica（0.7395）,Therdion octomaculatum （0.7354）and Tetragnathidae（0.7198）respectively.Also the one between the population size of ideal dominant natural enemies and that of natural enemies was Enigonidium graminicola（0.9703）,Propylaca japonica（0.6725）and Tetragnathidae（0.6482） respectively.Based on the results above,more influential natural enemies on M.persicae population were E.graminicola,Chrysopa septempunctata,Propylaca japonica, T.octomaculatum and Tetragnathidae from highly to lowly successively.5.The grey policy decision theory was used to evaluate the dominant natural enemies of Myzus persicae.Three major natural enemies of Myzus persicae were in order of Chrysopa septempunctata（0.3646）、Propylaea japonica（0.3202）and Theridonn octomacutatum（0.0445）,by the predatory ability,in the whole sampling stage.6.Based on the principle of community ecology and by the method of fuzzy clustering,this paper studied the temporal and quantitative dynamics of the arthropod community and sub-community in jujube orchard（Feidong 2003）,aimed to utilize natural enemy resources for pest control.There were 52 species of arthropod belonging to 47 families in the orchard.The relative abundances of arachnida and cecidomyidae were higher than other families in total community,i.e.,0.3465 and 0.2309 respectively.（The number of individuals of the arthropod community came to the peak on 8 June and 21 September.）The individuals of the arthropod community were the highest on 8 June and 21 September.The total community and pests -neutral insect sub -community could be clustered into 4 clusters,respectively.For total community,the clusters（T=0.92）were that on 22 March,on 6 April,on 22 November,and on other datas,and for pests -neutral insect sub -community,they（T=0.94）were that on 22 March,on 6 April,on 8 June,and on other datas.The natural enemy sub-community could be clustered into 3 clusters（T=0.94）,i.e., that on 22 March,on 6 April,and on other datas.The results of cluster analysis partly reflected the seasonal differences of total community and sub-community,while the temporal overlaps of cluster results mirrored the complexity of community structure.7.Investigations on the spatial structure and distribution of Contarinia sp.and its natural enemy,Erigonidium graminicola,in Feidong at different period were conducted with geostatistics.The results indicated that the semivariogram of Contarinia sp.in 8 June, 27 August and 19 October could be described by the spherical models,showing an aggregated spatial distribution with the distances of spatial dependence which were 26.3m, 7.8m and 22.0m respectively.The degrees of its spatial dependence were 88.05%,85.77% and 87.58%respectively.As for Erigonidium graminicola,the corrrsponding models were also spherical ones,showing an aggregated distribution with the distances of spatial dependence which were 28.3m,22.5m and 22.2m respectively.The degrees of spatial dependence were 90.09%,87.96%,85.72%which revealed that Erigonidium graminicola was dominant natural enemies against population of Contarinia sp.in the Jujube orchard examined.8.Investigations on the spatial correlation of Contarinia sp.and its natural enemy, Erigonidium graminicola,in Feidong at different period were conducted with Cross-variogram.The results indicated that there is a significant spatial dependence between Erigonidium graminicola and Contarinia sp.from 21 May to 19 October except 27 August and 21 September,which showed strongly the following relationship of Erigonidium graminicola to Contarinia sp.in number and spatial especially on May 30 and October 19.It is demonstrated that Erigonidium graminicola is the dominant natural enemy again.9.Investigations on the spatial structure and correlation of Chactocnema ingenua and its natural enemy,Erigonidium graminicola,were conducted with geostatistics.The results indicated that the semivariogram of Chactocnema ingenua displayed as a spherical models, showing an aggregated spatial distribution with the distance of spatial dependence which was 18.82.The degrees of its spatial dependence were 49.39%.The estimating error of cross validation present normal distribution.The KRE equals to 0.0638,while the KRMSE is 0.9900.There are five aggregated patchs for Chactocnema ingenua in the survey area.As for Erigonidium graminicola,the corrrsponding models were also spherical ones, showing an aggregated distribution with the distances ranges of spatial dependence which was 22.45m.The degrees of spatial dependence were 33.8%.The KRE equals to 0.0185, while the KRMSE is 0.9902.In agree with the Chactocnema ingénue,Erigonidium graminicola also demonstrated five aggregated patchs.The consistence of Chactocnema ingénue with Erigonidium graminicola in the range,the spatial distribution and the degrees of spatial dependence supported the spatial following relationship of Erigonidium graminicola to Chactocnema ingénue.All of the results,gave the strong evidence that Erigonidium graminicola was the dominant natural enemy in the lawn.更多还原