【摘要】 在害虫生物防治研究中,选择适合的天敌种类或类型是能否有效控制害虫的关键.虽然传统的功能反应试验能够为评价天敌的捕食行为与猎物(或寄主)密度的关系提供有益的依据,但不能根据这些结果对天敌捕食能力进行统计比较.本文在Michaelis-Menten模型的基础上,提出多级层次结构的功能反应模型,使试验习子(如天敌类型、猎物种类等)对捕食行为的影响,表达为模型参数的改变.并根据Ibralim和Rahman(1997)发表的12组功能反应试验结果,应用这一多层次结构模型统计比较不同天敌类型的捕食行为、及猎物种类、猎物生长期对试验结果的影响.取得如下结论:1.根据模型预测值与试验观测结果的吻合程度及模型误差的分布,说明以Michaelis-Menton为基础的多层次结构的功能反应模型能够用于统计比较捕食螨种类间的捕食行为,以及猎物类型、猎物生长阶段和其它环境因子对天敌捕食行为的影响作用.2.这一模型的分析结果不仅从统计角度证实了Ibahim和Rahman(1997)的推测,即红色叶螨是更适于这种捕食螨的猎物.同时为捕食螨种类的选择及田间释放技术提供统计依据.更多还原

【Abstract】 In pest biological control,selecting the most proper natural enemy species or strain plays an important role related to the success of pest management.Although traditional functional response experiment provides useful information in this decision making procedure, but no statistical sound conclusion could be drawn.Considering the parameter estimation precision, Michaelis-Menton model,a re-parameterization of Holling’s disc equation,was used instead for typeⅡfunction response in this paper.With the function response experiment published by Ibrahim and Rahman(1997),hierarchical structure model based on Michaelis-Menton model is used to statistically compare the predatory behavior of phytoseiid predator on different hosts through linking two model parameters(predation potential and half saturated host density) with the experiment design factors.Following conclusions are drawn with the analytical results of this hierarchical structure function response model.(1) The consistency between model prediction and experimental observation,and normality distribution pattern of model residue show that this model could be applied in comparing predatory behavior between predator species or strains,and also in evaluating the effectiveness of host species,host life stages,and physical or biological environment factors on predatory behavior of predators.(2) Analytical results from this model also proved the guess that the red leaf mite is a more suitable prey made by Ibrahim and Rahman in their publication.And these results also provided beneficial evidence in natural enemy species or strain selection,and in designing field natural enemy mass release schedule.更多还原