扩散对破碎化景观上宿主-寄生种群动态的影响

景观破碎化和扩散是空间种群模型的重要因素，对生物入侵存在着深远的影响。本章将基于偶对近似模型，探讨由局部和全局宿主-寄生相互作用共同决定的扩散模式对破坏性景观上疾病入侵与传播的影响。其中，生境破坏由生境丧失量与生境破碎化程度来描述。模拟结果显示，宿主和病毒的全局扩散对疾病的入侵与种群密度产生不对称效应：病毒的全局扩散对系统产生的影响较宿主的全局扩散更为显著。不同扩散模式下，生境丧失越高或破碎化程度越低，均将越有害于寄生病毒的入侵；同时，生境的破坏程度也显著地影响了入侵阈值对扩散模式的响应机制。本文研究结果暗示，景观破碎化的空间分布格局以及病毒扩散的限制均可作为物种保护与管理中有效的疾病控制策略。该研究结果在一定意义上丰富和发展了寄生感染理论，为物种保护提供了生态学理论依据。

The effect of dispersal on the population dynamics of a host-parasite system in fragmented landscape

Landscape structure and dispersal strategy are two important factors affecting population dynamics and thus entail profound effects on biological invasions. However, insufficient attention has been paid to the combined effects of landscape structure and dispersal behavior on the epidemic transmission in spatially explicit landscapes. In this paper, pair approximation (a method from statistical physics for deriving ordinary differential equations that approximate the spatial dynamics of populations in a lattice environment) was applied to examine the impact of dispersal strategy, determined by the interplay between local and global host-parasite interactions, to the invasion and spread of epidemics on the fragmented landscape. The intensity of habitat destruction in the model, instead of undergoing a random loss, was quantified by the level of habitat loss and fragmentation. By incorporating the detailed information on dispersal strategy in the spatially structured landscape, we are able to address the following questions: (1) How does the dispersal pattern affect the invasion and transmission of the parasitic disease? (2) How does the impact of dispersal patterns depend on the landscape structure? By the use of invasion analysis and numerical simulations, the effect of global dispersal in hosts and parasites was found asymmetric in terms of the epidemic invasion and population dynamics. First, the global dispersal of parasites (epidemics) was shown to pose a much stronger impact than the global dispersal of hosts on the invasion threshold of the epidemic disease. Second, the average host density at the equilibrium state enhanced slightly with the increasing proportion of global dispersal in hosts, but declined dramatically with the increasing proportion of global dispersal in parasites. Finally, both the population density of hosts and parasites at equilibrium, measured by the relative occupancy, increased with the increasing proportion of global dispersal, but the impact of global dispersal on the population density of the parasites was more dramatic. Under different dispersal scenarios, high level of habitat loss and low level of habitat fragmentation were shown to be detrimental to the invasion of parasitic disease. The invasion threshold of parasitic disease was increased gradually with the increase of habitat loss, but declined monotonously with the increase of the clustering degree of lost patches. Meanwhile, the invasion threshold of parasitic disease reached the peak when all dispersal events were completely local; this indicates that local dispersal was preferred for controlling the invasion and spreading of disease. Results also showed that the mechanistic response of the invasion threshold to the proportion of local dispersal in all dispersal events could be altered under different levels of habitat destruction. When habitat destruction is severe (i.e. high proportion of habitat loss and low correlation between adjacent unsuitable habitat), the influence of global dispersal has been proved to be more important on disease transmission. However, when habitat destruction is subtle, the global dispersal plays only a weak effect on disease transmission. We thus suggest that it is possible to control disease transmission by modifying the spatial structure of the landscape and the dispersal capacity of epidemics. This work highlights the importance of encompassing more realistic spatial patterns of landscape and dispersal in conservation planning and has extended and enriched the research of population and community epidemiology.