标题： 基于模型的方法评估天敌对害虫生物防治的效果 显示英文标题

标题： A model-based approach to assess the effectiveness of pest biocontrol by natural enemies

摘要： 主要目标：本文的目的是提出一种建模方法，用于评估在包含多种基于农药的管理策略的多样化农业景观中，天敌对害虫生物防治的有效性。 我们的方法结合了随机景观模型和种群动态的空间显式模型。 它使我们能够分析景观组成（半自然栖息地的比例、未处理作物、轻度处理作物和常规处理作物）对害虫生物防治有效性的影响。 有效性通过环境和农艺描述符进行衡量，分别衡量农药对环境的影响以及考虑到害虫造成的损失的整个景观的平均农艺生产力。 结论：农药的有效性、处理强度和害虫的内在增长率被发现是景观生产力的主要驱动因素。 由于减少农药使用而导致的生产力损失可以通过生物防治部分补偿。 然而，模型表明，即使有高效的天敌，也无法在减少农药使用的同时保持恒定的生产力水平。 半自然栖息地的破碎化和增加的作物轮作倾向于略微提高生物防治的有效性，但与天敌的捕食率相比，效果微乎其微。 本说明是在ANR项目PEERLESS的框架内撰写的 “预测生态工程对于景观生态系统服务和可持续性”(ANR-12-AGRO-0006)。 显示英文摘要

摘要： Main goal: The aim of this note is to propose a modeling approach for assessing the effectiveness of pest biocontrol by natural enemies in diversified agricultural landscapes including several pesticide-based management strategies. Our approach combines a stochastic landscape model with a spatially-explicit model of population dynamics. It enables us to analyze the effect of the landscape composition (proportion of semi-natural habitat, non-treated crops, slightly treated crops and conventionally treated crops) on the effectiveness of pest biocontrol. Effectiveness is measured through environmental and agronomical descriptors, measuring respectively the impact of the pesticides on the environment and the average agronomic productivity of the whole landscape taking into account losses caused by pests. Conclusions: The effectiveness of the pesticide, the intensity of the treatment and the pest intrinsic growth rate are found to be the main drivers of landscape productivity. The loss in productivity due to a reduced use of pesticide can be partly compensated by biocontrol. However, the model suggests that it is not possible to maintain a constant level of productivity while reducing the use of pesticides, even with highly efficient natural enemies. Fragmentation of the semi-natural habitats and increased crop rotation tend to slightly enhance the effectiveness of biocontrol but have a marginal effect compared to the predation rate by natural enemies. This note was written in the framework of the ANR project PEERLESS "Predictive Ecological Engineering for Landscape Ecosystem Services and Sustainability"(ANR-12-AGRO-0006).